$33‘ ‘3‘??? OF RSGH-Afififlli‘ifi GLYCEMDES
ERG-M 3H2 MLLK :3 3-GLOBULE MEMBRANE

331211333 3's" {“516 Day we of .‘u. S,
M36331E'fifi1i‘e’; 33’. ”E 'JN“ ”F5151?“
1331:: '3’. W’ 133’?
“53532

In

 

J “““L

LIBRARY

Michigan Sta to
University

 
 
 
 

§
%

  

ABSTRACT
STRUCTURE OF HIGHéMELTING GLYCERIDES
FROM THE MILK FAT-GLOBULE MEMBRANE

by Don P. Wolf

This study was undertaken to determine the structure of high-
melting glycerides associated with the milk fat-globule membrane.

The procedure involved digestion of the triglyceride with pancreatic
lipase, chromatographic separation of the fatty acids and glycerides
and gas-liquid chromatography of the fatty acid esters from the
separated fractions.

Definition of location of individual fatty acids within the
membrane high~melting glycerides indicated that the beta positions
of these glycerides were occupied primarily by a saturated fatty acid
of l4, 16 or 18 carbon atoms.

The trisaturated glyceride content of membrane highdmelting
glyceride was found to be 71.2% while, those isomeric forms of
disaturated~monounsaturated and monosaturated—diunsaturated glycerides
which contained a saturated fatty acid in the beta position were found
to predominate.

Calculation of triglyceride types and isomeric forms indicated
that, while random distribution was found on the basis of saturated
and unsaturated, the individual fatty acids were not randomly distri-

buted in the triglycerides under study.

STRUCTURE OF HIGHéMELTING GLYCERIDES

FROM THE MILK FAT-GLOBULE MEMBRANE

By

Don P. Wolf

A THESIS

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

MASTER OF SCIENCE

Department of Food Science

1962

ACKNOWLEDGMENTS

Sincere gratitude is extended to Dr. L. R. Dugan, Jr. for his help
and guidance throughout the course of this study.

The author also expresses his appreciation to Dr. J. R. Brunner
and Dr. V. R. Harwalkar for their helpful suggestions and counsel.

Recognition is extended to the graduate students and staff members

for their encouragement and assistance.

ii

TABDE OF CONTENTS

page

INTRODUCTION 0 I O O O O O O O O O O O O O O O O O O O O O 1

EVE-w OF LITERATUE O O O O O O O O O O O O O O O O O O O 2
High%e1ting Glyceride O O O O O O O O O O O O O O O O 2
Pancreatic Lipase . . . . . . . . . . . . . . . . . . 4

Thin-Layer Chromatography . . . . . . . . . . . . . . 5

EXPERIMENTAL PROCEDURE . . . . . . . . . . . . . . . . . . 7
Pancreatic Lipase Digestion . . . . . . . . . . . . . 7
Column Chromatography .. ... . ... . . . . . . . . . . 9
Thin-Layer Chromatography . . . . . . . . . . . . . . 9

Gas-Liquid Chromatography . . . . . . . . . . . . . . ll
EXPERIMENTAL RESULTS . . . . . . . . . . . . . . . . . . . 12

DISCUSSION . . . . . . . . . . . . . . . . . . . . . . . . 23
Isolation of HighyMelting Glyceride . . . . . . . . . 23
Pancreatic Lipase Digestion . . . . . . . . . . . . . 23
Separation of the Digestion.Mixture . . . . . . . . . 24
Glyceride Structure . . . . . . . . . . . . . . . . . 26

Triglyceride Types and Isomeric Forms . . . . . . . . 29
SUMMARY AND CONCLUSIONS . . . . . . . . . . . . . . . . . 32

LITE RATURE C ITE D O O O O O O O O O O O O O O O O O O O O O 3 3

iii

II.

III.

IV.

VII.

VIII.

TABLES
page
Fatty acid compositions of butteroil triglycerides and the
diglycerides, monoglycerides and free fatty acids resulting

from lipolytic action . . . . . . . . . . . . . . . . . . . . 13

Fatty acid compositions of butteroil high-melting tri-
glycerides and the free fatty acids resulting from lipolytic
aetion O 0 .~ 0 O O O I I O O O 0 O O I O C O O O O I O O O O 14

Fatty acid compositions of membrane HMG and the products of
lipolysis from HMG preparations 1, 2 and 3 . . . . . . . . . 18

Fatty acid compositions of membrane HMG and the products of
lipolysis from.HMG preparations 4, 5 and 6 . . . . . . . . . 19

Fatty acid compositions of membrane high~melting triglycerides
and the diglycerides and free fatty acids resulting from
lipOlytiC aetion I O O O O O O O O O O O O O O C O O O I O O 20
Triglyceride types and isomeric forms of the fats under study 21

Triglyceride types of the fats under study . . . . . . . . . 22

A comparison of the fatty acid composition of the fat-globule
membrane HMG and HMC isolated from milk fat . . . . . . . . . 28

iv

FIGURES
page

Schematic isolation procedure for obtaining the high~melting
glyceride fraction from the milk fat-globule membrane . . . . 8

Thin-layer adsorption chromatography of lipid classes on
Silica Ge]- 0 O O O O O O C O I O C O O O O O I O O O O O O O 10

Gas chromatogram of fatty acid methyl esters obtained from
membrane HMG preparation No. 5. . . . . . . . . . . . . . . 16

Gas chromatogram.of fatty acid methyl esters obtained from
butteroil HMG preparation No. 2. . . . . . . . . . . . . . . l7

INTRODUCTION

Considerable research in the last few years has been devoted to
the nature of the fat globule membrane existing in whole milk. However,
the natures of the proteins and lipids in the complex structure compos-
ing this membrane are not well characterized. Interest has centered
around a high-melting glyceride fraction isolated from several sources
butpresumed to be associated with the milk fat membrane. Present
theories suggest that this fraction may serve as a link between the fat
globule and the membrane.

The purpose of this study was to determine the structure of this
high-melting glyceride, which is insoluble in 95% ethanol at room
temperature, in order to contribute to the elucidation of the glycerides

associated with the fat-globule membrane.

LITERATURE REVIEW

HighéMelting Glyceride

The lipid-protein nature of the membrane present at the fat-plasma
interface in whole milk has been recognized for years. Palmer and Weise
(1933) were the first to isolate high-melting glycerides (HMG) from the
membrane. Rimpula and Palmer (1935) isolated HMGS from the membrane of
artificial as well as natural creams. They postulated that the fatty
'acid residues of the membrane phospholipids were attached to this HMG
with almost chemical affinity. Jenness and Palmer (1945) isolated
high-melting triglycerides of similar properties from butterfat (4.5%
of total lipid), washed-cream buttermilk extracts and washed-cream
serum extracts (37.4%) by crystallization from ethanol. They reported
the following physical properties for these fractions: iodine value
5.0-7.1, saponification equivalent l98.8h204.0 and melting point 52-
53° C. Jenness and Palmer (1945) found that the characteristics of
the HMG obtained from butteroil by crystallizing from ethanol were inde-
pendent of the composition of the butteroil. In order to explain the
high concentrations of HMG found in butter serum, Jenness and Palmer
(1945) proposed that during churning some of the proteinephospholipid
and phospholipid-HMG linkages are broken. The protein-phospholipid
complex was then found to be relatively richer in protein than the
original membrane while the complex of the butter serum was relatively
richer in phospholipid. Upon melting the butter, the phospholipids
continue to cling to some of the HMG molecules pulling them into the

serum thus accounting for the higher proportions of HMG in butter serum.

3

The hypothesis of Rimpula and Palmer (1935) was substantiated by
Jenness and Palmer (1945) when they found that ether extracted only
small amounts of HMS from unconcentrated butter serum indicating that
this fraction was still "bound" by the membrane. Patton and Keeney
(1958) reported the isolation of an acetone-insoluble HMG from the
membrane. The characteristics of this fraction and the ethanol-
insoluble glyceride fraction obtained by Thompson, Brunner and Stine
(1959) were similar in that stearic and palmitic acids were the
principal fatty acid components. The iodine values reported by these
workers showed no agreement. The absence of glycerides in the ethanol
soluble fraction led Thompson g£_al, (1959) to suggest that the entire
membrane triglyceride is HMG. Keeney (1961) reported the presence of
distearyl triglycerides in an acetone-insoluble high-melting glyceride
from milk fat. However, he found that none of the triglycerides from
this fraction contained more than 50% stearic acid. Since any tristearin
present would surely be a constituent of the HMG fraction, it was
deduced that milk fat contains no tristearin.

Several good reviews have been made on the fatty acid composition
of milk fat: Jack and Smith (1956) and Shorland and Hansen (1957).
The work on fatty acid composition of milk fat that has been accomplished
in the past few years was reviewed by Herb, Magidman, Luddy and
Riemenschneider (1962). From their own research they reported the
identification of at least 60 fatty acids in milk fat. This list
included several acids (odd-numbered, monoethanoid from C15-23) not
previously reported. Twenty-seven minor fatty acids were found each

in a concentration of less than 0.1%.

4
Boatman, Decoteau and Hammond (1961) and Eshelman, Manzo, Marcus,
Decoteau and Hammond (1960) using the mercaptoacetic acid method found
milk fat to contain from 21.5to 32.0% by weight trisaturated glyceride.
Their results showed no preferential selection or exclusion of any of
the major fatty acids from the trisaturated glycerides and the percent-

ages agreed with the amount calculated by random distribution.

Pancreatic Lipase

Pancreatic lipase provides a useful method for determining the
structure of glycerides because of its specificity for hydrolyzing
the ester linkages at the one and three positions of triglycerides.
Studies on the specificity of this enzyme have been made by a number
of workers, notably by Mattson and Beck in America (1955, 1956) and
Savary and Desnuelle (1955, 1956) in France. Enzymatic hydrolysis of
synthetic triglycerides of known structure has verified the specificity
of this enzyme.

Desnuelle (1961) reported that shorter, saturated fatty acids
are liberated more rapidly than longer ones and that saturated acids
from-C18 down to C12 and the C18 monounsaturated fatty acids are split
off at similar rates. He also found that pancreatic lipase acts
exclusively on emulsified esters for he observed an increase in activity
when emulsions began to form and a decrease or the absence of activity
when the substrate was in solution. All workers agreed that the hydrolysis
of triglycerides by pancreatic lipase proceeds in a stepwise fashion
from triglyceride to 1,2-diglyceride to 2-monoglyceride.

Pancreatic lipase has been used extensively in the determination

of the structure of triglycerides. Youngs (1961) described a method

5
for the quantitative determination of the six types of glycerides
found in fats. VanderWal (1960), Mattson and Lutton (1958) and
Patton, Evans and McCarthy (1960) have done similar work on the
structure of milk fat employing gas-liquid chromatography for analyz-
ing the hydrolyzed products. Patton gg_§1, (1960) found a higher
concentration of C10, C12 and 614 saturated and C14 and C16 mono-
unsaturated acids and a lower concentration of C18 acids in the
monoglycerides after digestion with pancreatic lipase. Ast and
VanderWal (l96l),from a study of the distribution of Saturated and
unsaturated fatty acids of butterfat, concluded that butterfat is
another of the group of fats in which saturated and unsaturated
acids become associated as S3, SZU, SU2 and U3 in proportions which
can be specified, at least approximately, by application of the laws
of probability operating freely or with some restriction.

Jensen, Sampugna and Gander (1961) used milk lipases, which are
also specific for the one and three positions of glycerol, to determine
the composition of butteroil. Their studies revealed lower concentra-
tions of C4, 018 saturated and C18 monounsaturated acids and higher
amounts of C14 acids in the monoglycerides after hydrolysis. Kumar,
Pynadath and Lalka (1960) used pancreatic lipase to establish that

butyric acid is located at the alpha positions in milk fat.

Thin-Layer Chromatography
Thin-layer adsorption chromatography (TLC) on silicic acid has
been used for the resolution of a wide variety of lipids. The method
as applied to lipids remained in obscurity until Stahl, Schroter, Kraft

and Renz (1956) described equipment and procedures for the preparation

6

of chromatoplates and demonstrated the potential usefulness of TLC
in the fractionation of substances other than terpenes. Several
reviews of the literature on the general application of TLC have
been made: Demole (1958), Stahl (1958) and Mangold (1961).

Thin-layer chromatography has been applied to mixtures of mono-,
di- and triglycerides by several workers. Jensen §£_al, (1961), who
hydrolyzed milk fat with milk lipases, used TLC in separating glycerides
from a silica gel column. Privett and Blank (1961) developed a micro-
method for the determination of component mono-, di- and triglycerides.
Their method included the ozonization of dOuble bonds and the catalytic
reduction of the ozonides followed by separation and quantification
of the glyceryl residues by TLC. ‘Malins and Mangold (1960, 1960) also
reported'on the separation of mono-, di and triglycerides on TLC.
In addition, they separated the methyl esters of a C18 fraction
derived from.menhaden oil by reversed-phase-partition chromatography

on siliconized silicic acid plates.

EXPERIMENTAL PROCEDURE

The cream used throughout this study was supplied by the Michigan
State University Creamery from a mixed herd source.

Figure 1 represents schematically the procedure employed in
isolating the HMG from the milk fat-globule membrane.

The steps followed in this experiment to determine glyceride
structure were: (1) digestion of the triglyceride with pancreatic
lipase, (2) isolation of the digestion products, and (3) determination

of the fatty acid composition of these products.

Pancreatic Lipase Digestion

The conditions of digestion were selected from several procedures
previously reported (Mattson and Volpenhein, 1961 and Ast and VanderWal,
1961). I

V The digestion mixture consisted of 0.2-0.5 g. of triglyceride,

'20 ml. of distilled water, 0.5 ml. of 45% aqueous solution of CaClZ,
0.2 ml. of a 1% aqueous solution of bile salts (sodium taurocholate)
and 100 mg. of pancreatic lipase (Mann Research Laboratories, pork
pancreas-crude). Digestion was carried out at 40° C. with continuous
agitation. The pH was maintained at 8 by periodic additions of 0.1 N
NaOH and the digestion was allowed to proceed for five minutes.

At the end of the hydrolysis period, 5 ml. of 6N HCl and 15 ml.
of ethanol were added. The lipids were then recovered by extraction
with petroleum ether. The petroleum ether solution was washed with

water, dried with sodium sulfate, and the solvent removed under vacuum.

0
0

35 ,7. CREAM
WASHED SIX TIMES WITH-
mhtt venues or vanes AT 40°C;
WASHE CREAM

CHURVED

MILK FAT mo seam
1041MB To a 5°C. SEPARMED wmc
LABORATORY SEPARATOR

 

r 1
mm FAT mane-001mm seam
FREEZE MY
om.

3 '5 ETHANOL IN cm
FOR I5 MIN" FILTER

 

 

 

 

 

l ,
mama LPbI I)
[man 21: mm emu. FILTER (11.121.131m
5 muss “SET“
radii: @012) m .0“...
”BINED WITH I I) FILTER
Mao (11.531 um
WASH 8X fig! mamas
WITH ETHEI
AT 311 0c. $30M. Train-OE
1“ CRYSTALLIIE AT ZI-
“tfi 23,-...
commas mm 1 1) ‘ nun
Mm" misfits
WALLIIE
4 X WITHEQGI
AT 21-23' c..
FILTER
SIPERN'ATANTS I PufifHED
HMGF

Fig. 1. Schematic isolation procedure for obtaining the high-
melting glyceride fraction from the milk fat-globule
membrane.

-Column Chromatography m

‘ The neutral glycerides and fatty acids liberated during digestion
were separated by adsorption column chromatography on Florisil accord-
ing to the method of Carroll (1961). The free fatty acids, tri-, di-
and monoglycerides were eluted in a stepwise fashion employing ethyl
ether, hexane, methanol and acetic acid as solvents. The column
eluate was collected in 25~ml. portions, the solvent removed and the
purity of each fraction was ascertained on thin layers of silicic acid
as described below. Figure 2 represents a typical separation of
_ glycerides by thin-layer chromatography. The acetic acid employed
during elution of the free fatty acids was removed from the sample

by washing with water.

Thin-Layer Chromatography

The plates for thin-layer chromatography were prepared as outlined
by Mangold (1961).

TLC was employed directly to separate the digested lipid mixture.
Samples were dissolved in a minimal amount of chloroform and streaked
on the thin-layer silicic acid plates. Solvent systems of 50:50:l
and 10:90:l, ethyl etherzpetroleum ether:acetic acid were employed and
the plates were developed for 30-45 minutes. The resulting glyceride
bands were_made visible under ultraviolet light by spraying with
dichlorofluorescein. The outlined bands, defining the separate
fractions, were recovered by scraping off the glass plate and extract-

ing with petroleum ether.

10

 

 

 

ILS cu.

 

0

<2)
C2:

 

 

 

3 o 2-.
0. . . G e f.) 9 .
A B C D E F G H I J K

Fig. 2. Thin-layer adsorption chromatography of lipid classes

on Silica Gel. Solvent: Petroleum ether-ethyl ether-
acetic acid, 90/10/1, v/v/v. Development time: 40 min.
Indicator: Dichlorofluorescein. Amounts: approx. 20k
each. a) diglyceride from Florisil column, b) cholesterol
c) tripalmitin, d) triglyceride from Florisil column,

e) monoglyceride from Florisil column, f) monoglyceride,
g) fatty acids from Florisil column, h) acetic acid,

i) cholesterol acetate, j) myristic acid, k) Capric acid.

ll
Gas-Liquid Chromatography

Methyl esters of the neutral glyceride fatty acids were prepared
according to the base-catalyzed interestification described by Smith
and Jack (1954). The free fatty acids were esterified by refluXing
in methanolic HCl (Youngs, 1961) or in methanol and sulfuric acid
(Trammel and Janzen, 1961).

An Aerograph Model A-90-C Gas Chromatograph equipped with
tungsten hot wire thermal conductivity detectors was employed in
conjunction with a Leeds and Northrup Type C Speedomax recorder
having a 5 mv. scale span and chart speed of 2 min./inch. Ten foot,
one-quarter inch (O.D.), copper gas chromatographic columns packed
with diethylene glycol succinate were employed for the resolution of
fatty acid methyl esters C4-018.

Operating conditions were as follows: Oven temperature-180° C.,
injector temperature-240° C., helium flow rate-75-100 cc./minute and
detector current-265 ma.

Quantitative analysis was achieved by measuring the areas of
the peaks with a planimeter. Unresolved peaks were separated for
measurement by drawing the shortest possible perpendicular line from
the recorder tracing to the baseline of the chromatogram. The area
percentage for the major fatty acids, based on the total area of the
major fatty acids, was computed. Unidentifiable areas and minor fatty

acid areas were not included in this calculation.

EXPERIMENTAL RESULTS

The fatty acid compositions listed ianables I-V were expressed
quantitatively as percentages of the major peak areas of the gas.
chromatogramS. The fatty acids were identified by 18:1 or similar
designation, the first number referring to the number of carbons in
the chain and the second the number of double bonds.

The results of the fatty acid analyses of butteroil triglycerides
and the diglycerides, monoglycerides and free fatty acids resulting
‘from lipolytic action are presented in Table I. Butteroil, as used
in this text, refers to the milk lipid not associated with the fat~
globule membrane. The data indicate that 10:0, 12:0, 14:0 and 16:0'
existed in higher concentrations in both the diglycerides and mono-
glycerides, except for 10:0 in the diglycerides. In contraSt, 18:0
and 18:1 were found in lower concentrations in the partial glycerides.
These results were substantiated by the free fatty acid analyses
which indicated a higher than random concentration of 18:0 and 18:1
esterified at the alpha position of butteroil.

The data presented in Table I agree well with those of McCarthy,
Patton and Evans (1960), Patton g£_§l, (1960) and Jensen 35 a1, (1961).

These results from the lipolysis of the HMG isolated from butteroil
are expressed in Table II. Increased quantities of 10:0, 12:0, 18:1
and 18:2 and decreased quantities of 16:0 and 18:0 were found in the
free fatty acid fraction. The value for 14:0 while showing an overall

increase, was not as consistent.

12

13

.monEmm omnsu mo mwmum>< o

.monEmm “sow mo mwmuo>< 0

.mwcoe mfinsow mo “means msu 0Coomm wen 0cm muoeumo mo Hogans mzu moumcwflmmw munwflm umuww one o

 

 

0.0m 0.0H 0.0N 0.0m D A<HOH
0.00 0.00 N.mm N.0~ m A<HOH
N.N m.H q.N 0.0 Nuwa
N.0N 0.5a «.mm 0.0m Huwa
m.0H 0.5 N.0 m.m ouwa
0.00 0.0m ¢.mm 0.0m 0H0H
0.0a N.NN 0.0H «.ma ouqa
q.m 0.m «.0 0.0 0HNH
N.N 0.0 0.N 0.0 0"0H
0.N 0.0 0.0 H.H ouw
ll'UI'II-I.III.II.‘II.IIIIIIII."II.I.I|I.I..II"IIIINN0H<II'IIUI'III'SI'I'III'I'U'E'Inuns-I'll..-

omwfiom

xuumm moan oovfiuoohawocoz. nouwumomeflm nmcfihmozawflua m0wo<

 

cowuom uwu%aomfia Eoum wowuanmou mcflom
muumm moum 0cm mokumozfiwocoe «movwhoozawfiv men can mowfiumoxawfiuu Hwououusn mo mcowuwmomsoo caom huumm

H mam<H

1.111.114

_..T‘vl1r.,

v1llzvlt.

.11!!!)

.Ajllvltllll.

.IL

111-114

IIIIIIIIIIII1178

c t u . o .
III] ! t I I a s l

.‘Jl‘IIl)-IIII.'.IJ!I‘J]-Iuov

r.

..I‘-Ill

..‘I\ll‘)

l4

moamfimm mounu mo mmmuo>¢ m

 

 

0.5a m.MH . D HmHOH
mam mam m :38.
N.N ¢.H NuwH
w.qH H.NH Huwa
D.HH N.¢H ouwH
m.aq .n.m¢ ou0H
0.NN O.HN ouda
0.0 0.m ouma
H.N n.H OHOH
.ll.‘n-ull-"l'.l.ll."'l."0'II'IIEUIIE'IANV WNH<I-Iu-"'l.'ollu"""ul'l'lu'lI-E"
mwflom huumm comm mwwumohawwuH . 0wo<

 

muowuoo owumaomHH Eouw
wcfiufiomou mwfiom muumm menu ecu 0cm mowwumomamfiuu wcfiuaoelewfls Hwououuon mo mcowuwmomaoo 0Hom huumm

HH mnm¢H

 

I p

.11...

.JO.\IrVoVAII'wIIl’u-I':1I.

\I'l‘lIlllIIIIJ-

{IIIJIIIIII'I

{cl-1'0

‘IIIII‘II‘V.’

<1]I.Io

15

Table V represents an average of six individual determinations of
glyceride structure (Tables III-IV) of HMG from the fat-globule membrane.
In most resPects the data were similar to those exhibited by the HMG of
butteroil indicating a higher than random concentration of 16:0 and 18:0
in the beta position of the triglyceride. In contrast to the results
with butteroil HMG, the concentration of 14:0 decreased in the free
fatty acid fraction of lipolyzed membrane HMG.

The compositions of butteroil HMG and membrane HMG show striking
dissimilarities in the saturated fatty acids myristic and stearic. The
melting point for butteroil HMG was found to be from 5-8° C. below the
52-53° C. reported by Thompson gt 31. (1959) for membrane HMG (see
Figures 3 and 4).

Using data obtained from pancreatic lipase hydrolysis, the tri-
glyceride types and isomeric forms were computed for the fats studied,
by the method of VanderWal (1960). The results are shown in Table VI.
Comparison of the calculated values for triglyceride types with those
predicted by random distribution indicate that the HMGs, as well as
butteroil, are among that group of fats in which saturated (S) and
unsaturated (U) become associated as S3, SZU’ SUZ and U3 in proportions
which can be specified, at least approximately, by application of the

laws of probability, see Table VII.

PEAK HEIGHT

16

 

 

 

::==?16:0

 

 

   
    

 

 

O
3'.
O
Q
H
3 '0 '3
3 '3
a
g .
a ‘5 ..
c: 'T c: 1: :5
H to
c 9 on o. c w 00 on H N
N v If) “ H E Q 0.
p. 3 F4 I r4

    

 

   

2 4 8 10 12
RETENTION TIME (MIN.)

14 16 18 20

Fig. 3. Gas chromatogram of fatty acid methyl esters obtained
from membrane HMG preparation No. 5.

  

PEAK HEIGHT

17

 

—:, 14:0
16:0

    

 

12:0

1

' 14:
1.

15 0
5-0 Branched
18:0 Branched

18 0

   

16 1
17:0
18:2

 

  

 

 

 

 

 

 

0 2 4 6 8 10 12 14 16 18 20 22 24
RETENTION TIME (NIN.)

Fig. 4. Gas chromatogram of fatty acid methyl esters obtained
from butter oil HMG preparation No. 2.

 

«.ma 5.5 0.0N H.0H H.0N N.HH D Hmuoa

 

 

0.Nw N.Nm N.mm w.0w 0.05 m.ww m HmHOH
N.N 0.H O.N w.H m.N 0.H NumH
H.mH 0.0 w.wH O.NH mo¢H H.0H HumH
N.mN m.wN w.wH ¢.mN w.NN H.0N ouwa
m.¢¢ w.w¢ m.H¢ m.¢¢ w.o¢ m.m¢ Ou0H
m.¢H n.mH 0o¢H m.mH N.MH HoHH Ou¢H
m.H H.H m.0 w.N a.m 0.H OnNH
0.0 m.o H.N H.H m.N H.H OuOH
8 III'I'Iol'nl"E'I'IEII'EI'I'I'I‘IIN mmH<|"'I-"""'I'E'IOEII.'88'E'E'88888'888'I
1 o
<mh UH <mm UH <mh UH QHU<
m .02 N .02 H .02
coaumumomum 02D doHuonmoHo 02D coaumwmmoum 02D

 

 

0 0mm N «H meHumemon 02D Eowm mwmhaomfla mo muosuoum men 0cm 02D oCmHDEmE mo mcowuwmomaoo 000m huumm

HHH HHQ<H

.‘l‘a‘l‘.

udIainln’zll‘-

' II .I
0 o I

11‘111.

o
q I v 4 V
I l ‘llu

1.111185!

.II‘1I.1-\IU

111.1}

..‘l].\.¢.

. n v o I I t
n I! l l I v n 0
3‘1 1 ‘ I O u 1'
I. a I III II“. 0 '7‘

._llvlt"lllIJvl'-D‘I

uJ‘YvI‘IJ‘IIIOo‘l.)

p 1 I I
'1II 14

93)}:A

‘IIDI

0.0a 5.0 0.0H 0.0a 0.5 0.5 5.HN 0.0 HJ¢H D HmuOH

 

 

5.H0 0.00 «.00 N.00 N.N0 0.N0 0.05 5.00 0.00 m HmHOH
¢.N 0.H 0.H 0.0 0.0 N.H 0.0 0.0 N.H Nu0H
0.0H 5.0 0.0 0.0a 0.5 0.0 d.0H 0.0 0.NH Hu0H
H.0N 0.0m 0.0N 0.0H 0.0m ¢.¢N ¢.0H 0.0a 0.00 0u0H
0.00 0.00 H.5q 0.00 H.0q 5.00 0.00 N.5¢ n.0q 0”0H
0.0a 0.0a 0.0a N.qH 5.5a H.0H 0.0a 0.0a 0.0a 0"¢H
0.0 0.0 0.N 0.0 H.0. 0.0 5.0 0.0 5.0 ouNH
N.N 0.H H.H 0.H 0.0 0.H 0.0 0.H «.N 0u0H
9

l '888'88'888 on-"II'II'EEEI'E'I'IlI-E"'IUOIAN. mmaH<I'l'IEUUIUOEII-"EI-E-ll'l'EIEIEEIUEIEOoI-U"

SE on 8. 4% on on. «E on on Boa
0 .OZ 0 .OZ . q .02
coflumummoum 02D nowumuwmoum 02D . soaumuwmoum 020

 

0 0cm 0 «q mcofiumummonm.uzm scum mam0Home mo muosvoum osu 0cm 02D mcmunaoa mo maowuflmomaoo 0wom huumm

>H 50.3”.

’11111Ii'

vlud“1\n§:'l

[-IrLIIOJYbIAI.

1’1.'VIV.

Ilp‘ll‘.vll.l1

null'd'l‘l‘l"jr2jl’ljtlil‘lltl'1

1")YII‘I..

1:111!!!)

20

mofimfimm mmusu mo mwmno><

0
mmHmEmm xwm mo m0mum>< m

 

 

0.0a 0.0 5.0a D HmuOH
0.00 H.H0 0.00 m HmuOH
0.0 0.0 d.H Nu0H
H.0H 0.0 ¢.0 HuwH
0.NN 0.0N 0.¢N 0u0H
0.0% «.00 0.0¢ 0u0H
0.0a N.0H N.¢H ouqa
0.6 0.H 0.N 0uNH
0.N 0.0 N.H 0u0H
.l-I'8'...I.In"'8'---"Iul""lu.l"'lu"0\o mm“¢I..l'I'E'I'E'I'E'i'ol'l"---IIIIIEIIII'I-
mvfiom
huumm moan novfluoozamflm mowflumozawfluH 0004

 

 

> 0H0<H

dowuom owuhaomwa scum wcHuHSmoH mwflom
kuumm mouw 0cm mouwuooxawflw map can momwumomawfiuu wawufioelswfl; odmunEQE mo meowufimomEoo 000m muumm

.111.‘11¢"11ll‘1,3u.

VIII . o
. o n I
Q II" I
1. ‘1. I] cl I.

.111121.

.IIIVI.

v.1l11llallt.

'09-.

~81]:

nIV'I

113

1‘

21

ms“ CH manouw Dhom mnu waoEm m N men on Hmnvo ma pH

Hmzhmwcm> mo conuma men on wcfiwnooom wouwasoamu

.mwmkfimCm 50 unnom

«mowanoozawocoarm
.maowuwmomum use fig maoowm Hkom men a“ m N n 0

 

 

 

 

mHm0Hmum 00 vasom mm mamfimm macs? onu ca manonw Hmom men waoEm m N u m
0.00 n 0
0.0 0.0 0.0m N.H 0.0 0.N 0.0m N.H5 0.00 n m
womanama_aoum oz:
. 5.00 n 0
0.H 5.0 «.0N 0.0 N.0 0.0 5.00 0.00 0.00 n m
HflOHmuusn Scum 02D
0.0 0.0 0.00 0.0 ¢.N 5.0a 0.00 H.00 0.00 n 0
0.05 n m
Hunchmuufim
25 mm: :mm . mam m8 Npmo ammo 98 use

 

AN usmwosv mumEOmH "cofluwmomaoo

AN uzwfimzv mommfl “cowuwmomfioo

 

 

omwsum Hows: mumm men 00 mahom owHoEOmH 05m mommu mwwnoomfiwflua

H> mfl0¢H

 

22

 

 

 

H.o 0.0 m.o N.o o.~. a.~ p

H.m m.N a.a m.q a.wa “.ma Nam

s.mm o.©N m.om a.om m.ma w.qa :Nm

N.Ha N.Ha a.qs o.qs o.am H.am mm

Eonnmm .ono sowawm .onU Eowcmm .onU momma okuoumawwufi
02D odmunfimz 02D Hwououunm Hwonouudm

 

 

HH> NHO<H

mwdum Hows: mumm men 00 momma wwwuookawfluH

DISCUSSION

Isolation of HighaMelting Glyceride

The yields of HMG were generally so low that difficulty was
encountered in obtaining quantities sufficient for gas chromatographic
analysis after digestion and separation of the lipolyzed lipid mixture.
The yield of HMG from one gallon of 35% raw cream was between 0.5 and
1.0 g. Increased yields resulted when the membrane was concentrated
by 1yophilization instead of "salting out" with ammonium sulfate as
employed by Thompson gt_al, (1959). In the salting-out procedure,
non-complexed lipid was removed by washing with ethyl ether, however,
due to disruption of the lipid-protein complex this non-complexed
lipid could not be selectively removed after 1yophilization. Because
more non-complexed, alcohol soluble lipid was present in those
samples that were concentrated by 1yophilization, increased amounts
of alcohol were required during crystallization.

Thin-layer chromatography indicated that the HMG fraction isolated
by precipitation from alcohol contained cholesterol, cholesterol esters
and possibly phospholipids (see Figure 2). Inability to quantitatively
precipitate phospholipids with acetone would account for their presence
in the HMG fraction (Smith and Jack, 1959). Because these contaminants
are present in such low concentrations any fatty acids contributed by

them would be essentially negligible in the glyceride structure analysis.

Pancreatic Lipase Digestion
The digestion of a fat with pancreatic lipase was allowed to proceed
for a maximum of five minutes. Digestion times previously reported range

from 15-45 minutes depending on the degree of hydrolysis desired. However,

23

24

since early results employing intervals up to 45 minutes were
inconsistent, the digestion time was decreased in accord with Jensen
(1962). Discordant data from longer periods of digestion could be
a result of the migration of fatty acids in the beta position to an
alpha carbon or to limited hydrolysis of the fatty acids esterified
at the beta position. The specific cause was not established in
this study.

The degree of hydrolysis achieved, in a given time, was found
to vary with the fats studied. The HMG fractions containing longer,
more saturated fatty acids required longer digestion times to achieve
the same degree of hydrolysis than did a sample of butteroil or corn
oil. Since the HMG is solid at the digestion temperature some question
was raised concerning the ability of pancreatic lipase to hydrolyze
the triglycerides. However, digestion of a sample of pure tripalmitin
(melting point 65.1° C.) was accomplished under the same experimental
conditions. Separation of the digestion products on TLC indicated that
solid fats as well as liquid fats are hydrolyzed by pancreatic lipase.
An explanation for the difference in digestion times lies in the
hypothesis of Desnuelle (1961) that pancreatic lipase acts on emulsified
esters only. The triglycerides composed of short chain and/or highly
unsaturated fatty acids, which are liquid at the temperature employed

during digestion, tend to emulsify more readily than the higher melting

triglycerides.

Separation of the Digestion.Mixture
Since pancreatic lipase is Specific for the fatty acids esterified
at the one and three positions of a triglyceride, the glyceride structure

can be determined by analyzing the triglyceride and the free fatty acids

25

liberated during lipase hydrolysis. The anion exchange technique
described by Hornstein, Alford, Elliott and Crowe (1960) for the
recovery of free fatty acids was attempted, but, due to failure to
suitably recover the esters of free fatty acids formed directly on
the resin, the method was abandoned. McCarthy and Duthie (1962)
have recently reported that Amberlite IRA-400 would remove the free
fatty acids from a lipid mixture but that quantitative recovery of
these acids was virtually impossible.

According to Quinlin and Weiser (1958), glycerides and free
fatty acids can be resolved on a column of silica gel. This procedure
failed to produZe satisfactory separations when used in this laboratory.

Chromatography on Florisil can be used to separate classes of
neutral lipids in much the same way as chromatography on silicic acid
(Carroll, 1961). Accurate quantitative analysis of a complex lipid
mixture by column chromatography, with either adsorbent, is complicated
by several factors. One disadvantage is incomplete resolution of
components having similar elution characteristics. TLC indicated that
the tri-, di- and monoglyceride fractions were not pure in many cases
(see Figure 2). Moreover, peak identification by reference to elution
volumes of authenic lipids is dependent on exact reproduction of the
conditions of analysis. The greatest limitation in achieving reproducible
results employing Florisil is in obtaining the proper degree of hydration.
This factor has been shown to affect the chromatographic properties of
Florisil markedly, changing its adsorption affinity more for some lipids
than for others (Carroll, 1961). In addition to column chromatography

on Florisil, preparative TLC was employed to separate the digestion

'1-

26
mixture. Difficulty was encountered in esterifying the glyceride
fractions scraped from the thin-layer plates that had been made

visible by spraying with dichlorofluorescein.

Glyceride Structure

Ideally the fatty acid composition should be expressed as
mole percentage. The thermal conductivity detector employed in the
gas Chromatograph responds on a weight rather than a molar basis.
Since the differences between weight percent and mole percent are
very small for fatty acids having more than ten carbons and because
comparisons of the relative fatty acid compositions of the samples
to determine glyceride structure was the objective of this study,
the expression of fatty acid data as area percentage was considered
adequate. Although errors up to 2% in individual fatty acids are
introduced by excluding the minor fatty acids in the calculation
of either area percentage or mole percentage, only the chromatographic
peak area of the seven major fatty acids were used to compute area
percentages.

. .

Butteroil, being readily available, was used to establish the
techniques employed in this study. Because no effort was made to
recover the short chain fatty acids, the area percentages of the
remaining fatty acids were 0-3% higher than those reported by Jensen
gt; 3;. (1961).

Because of the difficulties experienced in resolving the lipid
mixture after lipolysis, fatty acid compositions of monoglycerides

and in some cases diglycerides, were not obtained. These data were

not crucial, however, as glyceride structure could be determined from

 

27
triglyceride and liberated free fatty aciddata alone.
Pancreatic lipase analysis on membrane HMG indicate that the
beta positions of these glycerides are largely esterified with saturated
fatty acids containing 14, 16 or 18 carbon atoms. The averages of the

O
values for the liberated fatty acids and triglyceride composition data

permit calculation of the composition of the monoglyceride with the
following formula assuming 1,3 random, 2-random distribution: C2=3x
(C1,2,3)-2x(C1,3) (VanderWal, 1_962).‘ The symbols c1,2,3, c1,3 and c2
in the formula indicate the glyCerol carbons to which the corresponding
acyl groups are attached.

The monglyceride composition of membrane HMG was calculated to
consist of 15% myristic acid, 55% palmitic acid and 30% stearic acid
within an error of 2-3%.' These results indicate that the beta position
of membrane HMG contains only long chain, saturated fatty acids.
Obviously, then, any unsaturated fatty acids found in the diglycerides
must be esterified at the alpha positibns.

The monoglycerides from butteroil HMG was similarly calculated to
contain 18% myristic acid, 55% palmitic acid, 19% stearic acid, and 7%
oleic acid. The concentrations of myristic acid and stearic acid differ
widely from the values obtained by Patton and Keeney (1958), who isolated
HMGs from butteroil and membrane by precipitating from acetone (see
Table VIII).

The fatty acid compositions of the monoglycerides isolated from
hydrolyzed butteroil differed considerably from those calculated by

VanderWal's formula. This was not surprising for thin-layer chromatography

indicated that many of the di- and monoglyceride fractions were contaminated.

28

.0oumunummon 0H0 HmuOHo
.A000H0 hucomm 0am couumm mo mumnw
.moamawm omusu mo mwmnm><o

.moamawm me mo owmuo><0

_ . . .ozm
Hwououunn 0am 02D odmueaoa soon you mamumoumaouso mmw Hmowunova confimuno A000Hv aflm.mw nonmaosaw

 

 

 

 

0.0 0.0 0.0 0.0 wmdwwmmmfiD
s.H s.H s.o . mama
.H.s H.NH «.0 o.m Dawn
0.0N N.¢H 0.qN 0.0H 0u0H
0.00 5.0% 0.0q 0.0m 0u0H
HJNH 0.HN N.¢H 0.HH Dada
¢.H 0.0 0.0 0.0 OuNH
em.o oanfi BN.H .H.o ouoH
III'IU'J'IIIUE'I'E'IIOII'IIN.mmu¢r"""""'IIE'O'II'I'I
0.5m“ name .
umw xfiwa Eoum 02D oH500Hwnumm ms“ Eoum QED 0Hom huumm
use xflaa

Scum UmumHOmfl 02D 0cm 02D ocmhnaoa wasnoawlumm map mo coauwmomfioo 000m huumm m£u mo GOmDHmmEoo <

HHH> mum<H

(VIII

III 191‘ Ill

I‘lri‘d‘l..].l'|l’1'l?lf|no

11“,!1'!‘ ' 1’-

‘ A l I. 10 I

I n s . a h 0

III III III. ‘1! 1
. I 1‘ Q. I '1 .

1’131'1‘111‘1‘

...|Illl‘a'.

u l III
a o 5 I .0
I 10 I ‘1' 0

V I
u L v
1'.

I11 1‘

I v c I 1|:
0 c u . n
o I II} 01 | I G
I I ll.“ 11101111.
0
I

33' I! III."

7:!

'11] v

29

The fact that butteroil HMG exhibits a lower melting point and
a higher unsaturated fatty acid content than membrane HMG indicates
that variable triglyceride fractions may be precipitated from alcohol.
The variations observed in the degree of unsaturation among butteroil
HMG samples appeared to be inversely related to the number of
recrystallizations. Therefore, to obtain a pure HMG from butteroil,
more than the five crystallizations used for membrane HMG should be
employed. The crystallizations should be continued until no glycerides
are found in the alcohol supernatant. The difference in the unsaturated
fatty acid content of butteroil HMG found in this study and that
reported by Patton and Keeney (1958) is due in part to the amount of

contamination of the samples analyzed.

Triglyceride Types and Isomeric Forms
The percentage of trisaturated glyceride (G83) in butteroil was
calculated to be 34.1 by the method of VanderWal (1960), which utilizes
the specificity of pancreatic lipase. This value lies within the range

reported by Hilditch (1956) of 15-41% GS for butteroil. Eshelman §£_§l,

3
(1960) employing the mercaptoacetic acid method obtained values of 27.2
and 28.0. Attempts to reproduce these results by Eshelman's procedure
were unsuccessful.

The data in Table VII Show clearly that the calculated values for
triglyceride types in the fats studied are similar to those predicted
by random distribution. It is also obvious that the fatty acyl groups
do not assume positions within the molecules completely at random but
become segregated in the 2- and 1-3, positions. Determining triglyceride

types by classifying the fatty acyl groups only as saturated and unsaturated

is not ideal, for it obscures patterns in the placement of the individual

.30

fatty acids. Forty percent of the stearic and palmitic acids occurring
in the beta position of membrane HMG is not compatible with the 33%
predicted by random distribution. This leads to the observation that
while random distribution based on S and U may occur, the individual
fatty acids are not randomly distributed in the triglycerides studied.

The results indicate that the membrane HMG contains 71.2% tri-
saturated glyceride and no triunsaturated (GU3). Twenty-six percent

U but only 1.2% of this fraction is the

of the triglycerides are G82

isomeric form SUS, which contains an unsaturated fatty acid in the
beta position. The remaining 2.8% of the membrane high-melting -
triglyderides are GSUZ of which 2.4% comprise the USU isomer. The
butteroil and the butteroil high~melting glycerides are comparable

in isomer form to those of membrane HMG. However, butteroil contains
only 34.1% G83 while butteroil HMG contains 64.6%.

In butteroil 17.2% of the triglycerides contain an unsaturated
fatty acid in the beta position. 'This figure is reduced to 6.1% in
butteroil HMG and 1.6% in membrane HMG based on pancreatic lipase
hydrolysis data. Butteroil and the HMG fractions isolated from butteroil
and membrane lipid are similar to pig fat, in that the predominant
glyceride forms are those which contain a saturated fatty acid in the
beta position. A

While the isomeric forms calculated for membrane HMG indicate that
the beta position of these glycerides contains some unsaturated fatty

aCid, it appears from.monoglyceride calcdlation that this position in

membrane HMG is esterified only with saturated fatty acids.

31

The significance of the membrane HMG containing beta saturated
fatty acids of 14, 16 or 18 carbon atoms is not realized at the
present time. Are these HMG molecules selectively adsorbed by the
phospholipids or do they, solely on the basis of their physical
properties, migrate to the interface where they are complexed with
phospholipids?

So little is known concerning lipid-lipid interaction that it
is beyond the scope of this study to speculate regarding the nature
of the complex, if any, formed between the membrane HMG and the

phospholipids associated with the milk fat-globule membrane.

SUMMARY AND CONCLUSIONS

Recently some interest has centered around a highqmelting
glyceride isolated from the lipid associated with the membrane
surrounding the milk fat-globule. The purpose of this study was
to determine the structure of this ethanol-insoluble glyceride.

The procedure for the determination of glyceride structure
was accomplished in three steps: (1) digestion of the triglyceride,
with pancreatic lipase, (2) adsorption chromatographic separation
of fatty acids and glycerides on Florisil columns, and (3) gas-
1iquid chromatography of the esters of the fatty acids from the
separated fractions.

Structure analysis of the HMG isolated from membrane lipid
indicate that the beta position of these glycerides is primarily,
if not entirely, occupied by a saturated fatty acid of 14, 16 or
18 carbon atoms.

The results indicate that membrane HMG was composed of 71.2%
trisaturated glyceride and no triunsaturated glyceride. Those
isomeric forms of GSZU and GSUz'which contained a saturated fatty
‘ acid in the beta position were found to predominate. The butteroil
and butteroil high-melting glycerides, while containing smaller
-amounts of G83, were found to contain the same predominant isomeric
forms as the membrane HMG.

I Calculation of triglyceride types and isomeric forms led to the
Iconclusion that, while random distribution was indicated on the basis
of the distribution of saturated and unsaturated, the individual fatty

acids were not randomly distributed in the triglycerides under study.

32

.(1)

(2)

(3)

(4)

(5)

‘(61

(7)

(8)

(9)

(10)

(11)

(12)

LITERATURE CITED

Ast, H.J., and VanderWal, R.T.
1961. The structural components of milk triglycerides.
J. Am. Oil Chemist's Soc., 2567.

Boatman, Carolyn, Decoteau, A.E., and Hammond, E.G.
1961. Trisaturated glycerides of milk fat. J. Dairy Sci.,
44:644.

Carroll, K. K.
1961. Separation of lipid classes by chromatography on
Florisil. J. Lipid Research, 2: 135.

Demole, E.J.
1958. Applications de la microchromatographie d'adsorption
>sur couches mences. J. Chrom., 1:24.

Desnuelle, P. -
1961. Advances inyEnzymology. Vol. 24. Interscience
Publishers, Inc., New York.

Eshelman, L.R., Manzo, E.Y;, Marcus, S.J., Decoteau, A.E., and
Hammond, E.G.
1960.‘- Determination of trisaturated glycerides in milk fats with
mercaptoacetic acid. Anal. Chem.,.32:1431

Herb, S.F., Magidman, P., Luddy, F.E., and Riemenschneider, R.W.
1962. Fatty acids of cows' milk. B. Composition by gas-liquid
chromatography aided by other methods of fractionation.
J. Am. Oil Chemist's Soc., 325143.
Hilditch, T.P.
1956. The Chemical Constitution_2f_Natura1 Fats. 3rd edition,
. John Wiley and Sons, Inc. New York.

 

Hornstein, 1., Alford, J. A., Elliott, L .E., and Crowe, P.F.
1960. Determination of free fatty acids in fat. Anal. Chem.,
32:540.

Jack, E.L., and Smith, L;M.
1956. Chemistry of milk fat: A review. J. Dairy Sci., _2:1.
Jenness, R., and Palmer, L.S.

1945. Substances adsorbed on the fat globules in cream and
their relation to churning. VI. Relation of the high-
melting triglyceride fraction to butterfat and the

»"membrane". J. Dairy Sci., 285653.

Jensen, R.G.
1962. Private communication.

33

(13).

(14)

(15)

(16)

(17)

(18)

(19)

(20)

(21)

(22)

(231

(24)

34

Jensen, R.G., Sampugna, J., and Gander, G.W.
1961. Fatty acid composition of the diglycerides from
lipolyzed milk fat. J. Dairy Sci., 4451983.

Keeney, P.G. _
1961. Establishing the presence of distearyl triglycerides
in a high-melting fraction of milk fat using silicic
acid chromatography. J. Dairy Sci., 4451933.

Khmar, S., Pynadath, T.I., and Lalka, K.
1960. Location of butyric acid in milk triglycerides.
Biochim. et Biophys. Acta, 42:373.

Malins, D.C., and Mangold, H.K.
1960. Analysis of complex lipid mixtures by thin-layer
chromatography and complimentary methods. J. Am.
Oil Chemistb Soc., 31:576.

Mangold, H.K. . '
1961. Thin-layer chromatography of lipids. J. Am. Oil
Chemist's Soc., 385708.

Mangold, H.K., and Malins, D.C.
1960. Fractionation of fats, oils and waxes on thin-layers of
silicic acid. J. Am. Oil Chemist's Soc., 315383.

Mattson, F.H., and Beck, L.K.
1955. Hydrolysis of triglycerides by pancreatic lipase.
J. Biol. Chem., 214:115.

Mattson, F.H., and Beck, L.K.
1956. Specificity of pancreatic lipase for the primary
hydroxyl groups of glycerides. J. Biol. Chem., 2425735.
Mattson, F.H., and Lutton, E.S.
1958. The specific distribution of fatty acids in the
glycerides of animal and vegetable fats. J. Biol.
Chem. .233:868.

Mattson, F.H., and Volpenhein, R.A.
1961. The use of pancreatic lipase for determining the
distribution of fatty acids in partial and complete .
glycerides. J. Lipid Research, 4:58.

McCarthy, R.D., and Duthie, A.H.

1962. A rapid quantitative method for the separation of the
free fatty acids from other lipids. J. Lipid Research,
3:117.

McCarthy, R.D., Patton, S., and Evans, Laura
1960. Structure and synthesis of milk fat. II. ‘Fatty acid
distribution in the triglycerides of milk and other

animal fats. J. Dairy Sci., 43:1196.

35

(25) Palmer, L.S., and Weise, H.F.'
1933. Substances adsorbed on the fat globule in cream and
their relation to churning. II. The isolation and
identification of adsorbed substances. J. Dairy Sci.,

4351288.

(26) Patton, S., Evans, Laura, and McCarthy, R.D.

1960. The action of pancreatic fipase on milk fat. J.
Dairy Sci., 43:95.

(27) Patton, S., and Keeney, P.G.
1958. The high-melting glyceride fraction from milk fat. J.
Dairy Sci., 43:1288.

(28) Privett, 0.5., and Blank, M.L.
1961. A new method for the analysis of component mono-,
di- and triglycerides. J. Lipid Research. ‘3z37.

(29) Quinlin, Patricia, and Weiser, H.
1958. Separation and determination of mono-, di and tri-
glycerides in monoglyceride concentrates. J. Am. Oil
Chemist's Soc., 335325.

(30) Rimpula, C.E., and Palmer, L.S.
1935. Substances adsorbed on thefat globules in cream'and‘
their relation to churning. IV. Factors influencing
the composition of the adsorption "membrane". J.
Dairy Sci., 33:827.

(31) Savary, P and Desnuelle, P.

’)

1955. Chromatographic study of the action of pancreatic
‘1ipase on mixed triglycerides. Compt. rend., 240:
2571.

(32) Savary, P., and Desnuelle, P.
1956. Use of pancreatic lipasefor the study of the structure
of some naturally occurring fats. Biochim. et Biophys.
Acta, 335349

(33) Shorland, F.B;, and Hansen, R.P.
1957. The minor fatty acid constituents of butterfat (A
Review). Dairy Sci. Abstr., 325168.

(34) Smith, L.M., and Jack, E.L.

1954. The unsaturated fatty acids of milk fat. 1. Methyl
ester fractionation and isolation of monoethenoid
constituents. J. Dairy Sci., 31:380.

(35) Smith, L.M., and Jack, E.L.
' 1959. Isolation of milk phospholipids and determination of
their polyunsaturated fatty acids. J. Dairy Sci.,
42:767.

(36)

(37)

(38)

(39)

<40)

(41)

(42)

Stahl, E.,

1956.

Stahl, E.

1958.

36 .. I

Schroter, G., Kraft, G. and Renz, R.

Thin- -1ayer chromatography (the method affecting
factors and a few examples of application).
Pharmazie,133:633.

“Thin-layer chromatography. II. Standardization,

detection, documentation and application. Chemiker
Ztg., §3_:323.

Thompson., M..P., Brunner, J. R., and Stine, C.M.

1959.

Characteristics of high~me1ting triglyceride fractions
from the fat-globule-membrane and butter oil of bovine
milk. J. Dairy Sci., 4351651.

Trammel, J. A., and Janzen, J. J.

1961.

A method for determining the "free fatty acids"
milk using gas chromatography. Paper M981. Fifty-
sixth Annual Meeting, ADSA, Madison, Wisconsin.

VanderWal, R.J.

1960.

Calculation of the distribution of the saturated and
unsaturated acyl groups in fats, from pancreatic lipase
data. J. Am. Oil Chemist's Soc., 33:18.

VanderWal, R.J.

1962.

Youngs,
1961.

Unpublished data.

C.G.

Determination of the glyceride structure of fats. J.
Am. Oil Chemist's Soc., 33:62.

 

”'TITI'IMHILW[11101114111111111“

3 78 4238