I...

 

 

   
  
    

 

 
 
  
 
  
     
  
 
  
     
  

 

 

   

W— .
”0.0-”0“.'--«-..*‘-.-‘mfiw“-C"‘mw.~‘--:—.-D. O.-- - ’o - oo-oo~oaom~oo”-oo«a¢“Wm-W "v.“‘v-"vv—V'Y— ..“‘- -_ ‘. ' _ _ ‘\ ~ ‘ _ ‘ V-
o
r
.
C I I I . .
. ‘ _ - .
.
. V . ( E.R IIIE
H . ‘ O
. I
I
. ‘
l
3
.
.
.
.
o .
.‘.
..‘
. v ‘
o
‘
.
~ 9
.
U-‘
u
.
<0.
. .
_ ‘ -
V . . .
.
V ’ -
I
‘- .
.. .
.‘r'
,.
.
A
. - I
.I. u
.v
- I
. .' -
-.o .
'.
. . r’
‘ - n o
-c ' .— . - - - . .
;, -
‘ a ‘A . ‘
"‘ A .-
a . r .
. .
. I ’ .-
u . b l D '
- v . . . ~ - . - u - -. .
. . ' . ' . ,v t g . V
. l
a o. ~ ' . - - , u-o .
. l . .
o ' ' . 0 9 ..- 0- - - ‘ .
F I o h ‘ .7. ' ‘I O p
. o ‘ n . _,'., . 0.v- y .- l’,.'.. v... -0 o. a: a o.
p
' I .l r - I I .0 - ' _ v
I . . .. . . . . . ‘ - . ,. . - - -
' - ' - . o . o
‘ . , / .-. .. . ., _-
l "' . .
a " - ' u .-— - o. - ’ ~ .
.4 . ..- , l — -o . . .
. v.1. - v- v ’ — . w
‘ O . . "vv ' - . .
' ~ . - . ... . .. .9 . . I a . v..
- . .’ ‘ “" o a' ’ I .I” r r g
. n It - ...( .' .--". .,. , . . -
_ -'o o ’ 1 -,- .or . ‘ p ' . o v ' r
, . . - . o o n v ~O4¢ » a. fl 0
’ o ' . - .'--O -’ ,..... .. ' .-
' ' " l- o ._ .. g. ,-.A—. g .
A v . - n v. " -N ." .03 O . .g _ ’ v. _.. -. ., --, '-
° ' " ' . ~ ’ onolaa C' ,r,. to. . ¢
9 ’ ' ‘ . . O O l .i n J ' ’ -' "' -
, o . 01- I' . . '--'-p ~ .-'- -" - .,
. o - to . ... .. ¢ ' y .. ' V .-' -' ' ..‘..
' ’ '° . . ° ' ' ' .0 - a . ' . '4" .' D' '—
. . - ' - ¢ I - ,3 4- v, p . . , . ~ .' v v op- C'J- ° ' r ‘ -
l . o c- “— '40 .-o a.» n . - ',. A . c’o . o . a. ..,-.’o -.O of. {n O. .--a b .
' ' " ' " ' ' "" ' ' ' "' '1 "' ' - ~- ' .n "r :- -¢o- - v .- --'-’ ‘ -.- . n
' ' " ' ' ' ' - ‘ ' "" ' "' "‘ ' 'V'l a- "l'-ao . - v‘a -. . ..r.. - - v -r;v-;’."ooo . , . . ‘ - «
- . - . o 0 a I I a.» 9 O “"'-/r 0 ~- . .. ,0. n .r'- -‘ - ~ .. .r. , ,. 4 .o'.‘ -'-... 0" o ... .. 4-f- v r
-. ‘0. lo. . -0 - to! -" III “ - 'l‘ G 7 It... ’ .. .0 . '7..l , o . '--. 'Huoy...4'l“...r -- .v ,. o.
. ' - . " "- ." ‘-’-t ~'l¢'r'IQ'-v- D- . ‘ ' .' .mr " '- ..I ' ‘ . r- ’00 .. -. .. 'A- .0 -‘v.0 "o
' " "‘ ' " "" "‘ """"" """ """""H-" ."‘."l.' 04- ~ ',-.1 , '0' I .-;:-.-.r -'o(erI'o""I0v.-r 6" _'--l '-
' u' F ‘ ' _l"f"- ’ 0‘ a 0" I o‘. " ‘ . 4. A '_-.,. —".’l ;‘U_ '. — v.1; .'_-.
‘ "' ‘ ’ ‘ ' ' ’ ‘ ' "" -""M._'.” vr-v- v,v’l'~ 1’ Afr " “O/--'o“-. 4;. p.-- .l'- ' - I
., ' .r . . ,. - ;.o¢. - "--i-- l " . ’g ‘ .’.-,-',-'f‘.l'¢“'l‘,’_", ’, ~.t _.. ..-v .
" ' ' ' " " ' ‘ ' ""'.." ' ""“""‘."" ‘P;""" ---OI';.1 -'r . v ‘ -";d:090'_ol."—O-o .8- “‘o —'
, .-. .. I .- 1,. . ... .....I;,',ndlnvul v ‘m 'IIIOOF- ."'_'....., ' ~ 1 .',.’.’.
' ‘ . “"0‘ "I"' "’ ‘ " ”"""‘-"' """","'-Oac.t "':c"' ‘ D. l. u r- a'l"? ‘ I. ‘ -¢’-'r.’.. I. . ’1 I-o'v - .(0'- ":‘ ‘-o.-
. ’r' , ' . -- ' - ’ -' ~o - ».3 ' ""' mt)”. c v; a ’1 'l’ -'J -r--« pl!" w ‘. . r‘ -,-Od-'loor14-r .o,"--o4v. drz-.-v‘vo’.‘-v-’~-:
o ,1 n.-- .. ... cl . 0 o .0: I. . 'O"' " ""l-O'IM.‘Oto-!:'”".--' . u I . y I. r flrrrfo.'-c,.a”. . 9' I’d vr-gf’ [K-‘v .
, . _ .' . . . I-r- . . I .. I . . - ’ f'v .— ,. .9. ',.-, 9.“ ,,o, .." "fl-u. r v- ' 4-».0- --'
I _..- -O.l.o .v '-Ool .I ' .. . iD'Cfl OHIID.._-I-!.." .‘1-l,‘4‘..:' _. ., fl ."”,'_r'.'/”;;. (1" ‘ ,,._,",.'.':~‘, ,'...
' " "'- ' ' " "‘ ' " " " -' ' C"".l""‘.. ""'r'rU-(.r.l (:0. I. ’v - 0".c, 1‘de .4.O- —‘ ‘ frofv
-' 4.». .. q 'I . o . O - l'-~-’ 'Irr..'.o 9'..'.v’,..-' .;n — ’, .J, ,,‘ ... ‘0', I ' OY'W - . .¢,.‘. ' -'.,
. .." . v ‘ - ,. .., - 4 ..¢o~-—-ooo .o-.-o¢. - .“ 1 ..- - ,g..'. .".l',,..,(, . ,_,. ¢~,.. 10", .‘U_'. .JM V-op— ’..v, .fl.‘1.'..-ofi'ou.d.. '4" N, ._
x. r "‘ ' ‘ " """ ' 0 op -.O.‘.? - 14.; ._-"'"’ ' '!;"4- -,,d-.‘r ,1 ~84 r-n- -.--.
1r- -' ' ' ' pJJ-t'vp-a l o. , I- L.’ - .1... 9t. 0”. 0:1h‘mg,"ov-'»:M 0"¢." r—
17.“ .. . I'v- - > ‘ -I '-’/" r- ‘ 'l’f” - I10 ,,-.- --'r’—' t v -..aza(f'rv-vv
.,... - o 0 - -. . ‘ ., . g. ' v, a‘r' ~,,-' 1 ,. ' _. . - _""‘.1...A‘.>. ‘ . ':”‘, ' .U'.’r'.u , n o:
_ ’_ o . '- ' IO ' . ."‘ - -- _ , .0 .— o . , . - .a ‘,.l' Jr-,J.’¢n- " .--."¢"A.. ’.. f. . "f;’ a'- - .1 ' v ¢ :'?M.v,".“..
fl“, . / -« a v t ’6‘! .- \ .nv '.O.-¢ u ‘ q ~ hi‘l....¢,-y-.‘ I. fwfi .;- . v.,'-.g "-4nJ"’O-f la: (”a -r" ’2’; -f'
"’ ,- ,-.o- o '1 ,r o '- ' ' ' -- - a a tvd’vr‘ '1 0' I' ,q..-..)a .uw -.pcovl('-~-/ ‘O' ' 'L}('- ‘m v' 11.9’/:.'¢‘. ’(fpp.
‘40- -O-. ‘ " ‘V ‘ "" ‘ '- o/l t’0‘1-;-.4¢' o: ,p- a, 9 ~(« 0 o! (I an... (3" rrl/r_000v.-- ....
"-. , a l. 9 o.‘.... ...... f. .-
‘g...u O ' "t " "-00 " - . ”I". .40!,. ,¢.
1 ‘ ~' ' "‘,°I" ' ' 'oV'.("fill_l’r'o . o.-‘-' . -.-‘ ‘I..' '1
U - " ' '” """ ”.".;' or on - .Yvr'o . "o"-- .’ '<U' P ' noWlscv-a. o
g". {-1 “ " 0f ' o- ‘0 It If a 4. o- .""‘vlo.oo~r -
. .o' -0..‘o¢o O‘VAhc
.‘ -’ C"CVJ’ I .ar.
. I“ "‘"07 whit" ’4“.
.4'.’ "Cv0¢o¢dv zor
' p.’-- [.4’ .9-.. .
._'-0.' . QOOI' oll- ’-0 -oo'glv-’-
‘0‘. 'f" '.g}~o—v
' "' m -. 7/ '
uw‘”. I. ' 0" .v’ )“fi'.
-“. ’ 4 l . a rh'f‘Ovav rr 4
9.5... u 0 . . fl. , ' C. 4 . ., '.'4 - - ’ . " I "’ O rt.v 'lf""‘v'::f.‘ ’6‘"!
. 0 a... c1”. ... ‘ a . I ’ , .r . -. _ _ f . , '

.0 '. '. ":0. 921...”: D -.0)':: “l" 0' , ' H ‘PI-oal/J': ”('9‘ ‘4’}. 0" o: 4,90: p 'w ,I’o opvn:rvd
1'. .¢-—‘ ”.0. I o O h oth' o a O},o’ ..a JV .-o/ r 4” "V,.‘ pr ' . " ’," '
“’-o--“"'- -' v": 0 H — - 'n ' o“! 1-“! 0 H '2‘} L": If‘r/ I h "o coo . —r D" I! re r I-
. cc . r4 - . . .. ~ . . p ‘ ‘l l a -/ u‘. ‘ .4 _' I . ‘,_."."' ;r' ' ', ’
~‘ .‘.r. . ....ooa" _ I‘.'. ' ' I'lr" "

 

flllllllllllllllflIllIlllflllflflflfljfl‘llwlfllflflll _ LIB , AR y

3 1293 1

Michig .1 State
University

 

 

 

ABSTRACT
ELECTROPHORETIC AND PARTIAL CHEMICAL
CHARACTERIZATION OF BOVINE MILK FAT
GLOBULE MEMBRANE PROTEINS
By

Michael E. Mangino

The lipid fractions of milk fat globule membrane (MFGM) have been
characterized by a number of researchers, while the protein fractions have
received little attention as to their physical characteristics. This is
due, in part, to the resistance of these protein fractions to solublization
by buffers usually employed in protein chemistry. .

MFGM material was sequentially extracted with 0.6 M KCl and
centrifuged to yield five 100 S supernatant and pellet fractions. These
fractions were characterized both chemically and physically by the use of
polyacrylamide gel electrophoresis in a SDS system.

The original extraction and centrifugations resulted in a nearly
equal distribution of lipid and protein material between the supernatant
and pellet fractions. Subsequent extractions of the pellet fractions
resulted in a much greater loss of lipid than protein material to the
supernatant fractions. The protein to lipid ratio increased from 0.83
for fraction l-P to 1.65 for fraction 5-? while the phospholipid to
protein ratio decreased from 0.31 for fraction l-P to 0.24 for fraction S-P.

Concomitant with these compositional changes the pellet fractions
decreased in solubility with each extraction. Examination of the
electrophoretograms obtained for these fractions indicated that the

decreasing solubility was due to the aggregation of low molecular weight

polypeptides into larger, less soluble ones.

It was concluded that the removal of lipoidal material from the
pellet fractions caused an irreversible aggregation of the protein
moieties and that this aggregation was the main cause for the

decreased solubilities of these fractions.

ELECTROPHORETIC AND PARTIAL CHEMICAL
CHARACTERIZATION OF BOVINE MILK FAT

GLOBULE MEMBRANE PROTEINS

BY

Michael E. Mangino

A THESIS

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

MASTER OF SCIENCE

Department of Food Science and Human Nutrition

1973

l0
WA
MO ACKNOWLEDGMENTS

The author wishes to express his sincere gratitude to Dr. J. R.
Brunner, Professor of Food Science and Human Nutrition, for his patience,
counsel and inspiration during the course of this study.

Appreciation is also extended to Dr. C. M. Stine and Dr. H. A.
Lillevik for aid in preparation of the manuscript.

Acknowledgment is due to Ursula for aid and counsel on numerous
technical matters.

The author also expresses his gratitude to his wife, Lynn, for her

aid and encouragement during the course of this study.

iii

TABLE OF CONTENTS

Page
ACKNOWLEDGMENTS . . . . . . . . . . . . . . . . . . . . . . . . iii
LIST OF TABLES . . . . . . . . . . . . . . . . . . . . . . . . vi

LIST OF FIGURES . . . . . . . . . . . . . . . . . . . . . . . . vii

INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . 1
REVIEW OF THE LITERATURE . . . . . . . . . . . . . . . . . . . 2
Historical Survey of the Fat Globule Membrane . . . . . . . . 2
Isolation and Characterization of the Fat Globule Membrane. . 4
Gel Electrophoresis of Membrane Proteins . . . . . . . . . . 6
EXPERIMENTAL . . . . . . . . . . . . . . . . . . . . . . . . 12
Chemical Methods . . . . . . . . . . . . . . . . . . . . . . 12
Nitrogen . . . . . . . . . . . . . . . . . . . . . . . . 13
Hexose . . . . . . . . . . . . . . . . . . . . . . . . 12
Hexosamine . . . . . . . . . . . . . . . . . . . . . . . . 13
Sialic Acid . . . . . . . . . . . . . . . . . . . . . . . 14

Total Lipid . . . . . . . . . . . . . . . . . . . . . . . 15
Phospholipid . . . . . . . . . . . . . . . . . . . . . . . 16

Lowry Protein . . . . . . . . . . . . . . . . . . . . . . 17
Physical Methods . . . . . . . . . . . . . . . . . . . . . . 17
Preparation and Fractionation of Membrane Proteins . . . . 17
Polyacrylamide Disc-Gel Electrophoresis . . . . . . . . . 19

Acid Gels . . . . . . . . . . . . . . . . . . . . . . . . 19

iv

10% SDS-Gels
5% SDS-Gels
RESULTS

Distribution of Protein, Lipid and Phospholipid
Between Membrane Fractions . . . . .

Chemical Composition of Membrane Fractions .
Molecular Weights Determined by Gel Electrophoresis
Gel ElectrOphoretic Patterns .

Acid System
SDS Systems

DISCUSSION

Distribution of Protein, Lipid and Phospholipid
Between Membrane Fractions . . . . .

Composition of Membrane Fractions
Pellets .

Supernatants
Molecular Weights of Membrane Proteins .

Gel Electrophoretic Patterns of Membrane Proteins
Acid Gels .
SDS Gels

CONCLUSIONS
BIBLIOGRAPHY

APPENDIX

Page
20

20

23

23
23
24
24

24
25

41

41.

42
42
44

45

48
48
48

57
58

64

Table

II.

III.

IV.

VI.

VII.

LIST OF TABLES

Recovery of fat globule membrane compnents
in sequential supernatant and pellet

fractions

Distribution of fat globule membrane components in
sequential supernatant and pellet fractions
(expressed as percentage)

Distribution of fat globule membrane components in
sequential supernatant and pellet fractions
(expressed as component ratios)

Molecular weights
membrane pellet
electrophoresis

Molecular weights
membrane pellet
electrophoresis

Molecular weights

of the polypeptides of fat globule
fractions as determined by SDS
in 10% polyacrylamide gels

of the polypeptides of fat globule
fractions as determined by SDS
in 5% polyacrylamide gels

of the polypeptides of fat globule

membrane supernatant fractions as determined by SDS

electrOphoresis

Molecular weights

in 10% polyacrylamide gels

of the polypeptides of fat globule membrane

supernatant fractions as determined by SDS electrophoresis
in 5% polyacrylamide gels .

vi

Page

26

28

29

30

31

32

Figure

10.

11.

LIST OF FIGURES

Diagram for the sequential extraction of membrane
material with 0.6 M KCl

Electrophoretograms of five membrane fractions
after solublization and electrophoresis
according to the method of Takayama, £3.31:

10% polyacrylamide-SDS gel electrophoretograms
of the membrane pellet fractions stained for
protein

5% polyacrylamide-SDS gel electrophoretograms
of the membrane pellet fractions stained for
protein

10% polyacrylamide-SDS gel electrophoretograms
of the membrane pellet fractions stained for
carbohydrate

10% polyacrylamide-SDS gel electrophoretograms of
the membrane supernatant fractions stained for
protein

5% polyacrylamide-SDS gel electrophoretograms of the

membrane supernatant fractions stained for protein .

10% polyacrylamide-SDS gel electrophoretograms of the
membrane supernatant fractions stained for
carbohydrate

10% polyacrylamide-SDS gel electrophoretograms of
pellet fractions l—P and S—P stained for protein

Recovery of milk fat globule membrane components
in the pellet fractions

Composition of milk fat globule membrane components
in the pellet fractions

vii

Page

22

34

34

36

38

38

4O

40

55

56

INTRODUCTION

The fat globules that occur in milk are secreted from the mammary
gland by a method described as reverse pinocytosis. At the time of
secretion, these fat droplets are surrounded by a membrane layer about
90A thick.

Recently, workers employing the electron microscope, comparative
chemical analysis and enzymatic studies have presented strong evidence
in support of the theory that the milk fat globule membrane arises from
the plasma membrane of the mammary gland. Studies of the milk fat
globule membrane should give insight into the composition and structural
organization of other, more difficult to isolate, membrane systems.

The composition of the lipid moieties of the milk fat globule
membrane has been well elucidated while the protein components have long
resisted characterization, partially due to their inherent insolubility.

Several workers have fractionated membrane proteins on the basis of
solubility in different solvent systems. Other than chemical analysis,
little data are available to explain these solubility differences.

It was the purpose of this study to separate the milk fat globule
membrane proteins on the basis of solubility and to characterize the

resulting fractions chemically and electrophoretically.

REVIEW OF THE LITERATURE

Historical Survey of the Fat

 

Globule Membrane

 

Ascherson (1840) was the first to report the existence of a thin
membrane surrounding the fat globules in milk which he labeled the
"haptogen membrane." He believed that it arose from the condensation
of albumin at the fat/plasma interface.

Babcock (1885) examined fat globules before and after churning and
could detect no change in their appearance, nor could he detect fragments.
in the serum. He therefore concluded that no membrane existed. Later
he (Babcock, 1889) reported the existence of a protein that he felt was
responsible for the clustering of fat globules during the creaming
process, which could also serve as a membrane. He likened the creaming
of milk to the coagulation of blood and called this protein lacto-fibrin.

Storch (1897) realized that it would be possible for many milk
proteins to adhere to the fat globules and devised a method of washing the
fat globules. Ether extraction of these washed globules yielded a residual
mucin-like protein which gave a positive test for carbohydrate.

Hattori (1925) found that the "hapteins" were different than other
milk proteins on the basis of their physical and chemical characteristics.
He described these proteins as keratin-like and noted their extreme lack
of solubility.

Titus (1928) and co-workers isolated fat globules without any
previous washing. Examination of the isolated protein led them to

2

conclude that the main protein of the fat globule membrane was casein.

Palmer (1924, 1933, 1936) and his co-workers in a series of experi-
ments were the first to demonstrate that the membrane protein was com-
plexed with phospholipid. The result of these experiments was the theory
that the membrane was composed of a single globulin—like protein linked
to various amounts of phospholipid.

Brunner (1953) and co-workers showed that the protein material of
the membrane was altered during homogenization. The unhomogenized
membrane protein exhibited sedimentation characteristics similar to
lactoglobulins in the analytical ultracentrifuge. 0n the basis of this
observation, the protein was provisionally classified as globulin-like
in nature.

Herald and Brunner (1957) fractionated the membrane into two com-
ponents by centrifugation at 25,000 x g in 0.02 M sodium chloride for
one hour. The pellet obtained was insoluble in phosphate and veronal
buffers and was termed "insoluble." The mucoidal supernatant was
designated as the "soluble" fraction. This represented the first
demonstration that the membrane complex contained more than one protein
entity. The insoluble fraction was tentatively classified as a
pseudokeratin.

For a more comprehensive review of the membrane proteins including
accepted methods of preparation, the enzymatic content and the possible
origin and structure of the fat globule membrane, the reader is referred

to the review of Brunner (1973).

Isolation and Characterization

 

Of the Insoluble Membrane Fraction

 

After their classification of the protein that was not soluble in
0.02 M sodium chloride as the insoluble fraction, Brunner and Herald
(1957) reported the heterogeneous nature of this fraction. Solubliza-
tion of this material by various dissociating agents and subsequent
analysis by moving-boundry electrophoresis showed it to be made of more
than one species.

Thompson and Brunner (1959) substantiated the earlier observation
of Storch that the membrane proteins were glycoproteins. Thompson (1960)
further examined the insoluble membrane protein and on the basis of
results obtained from low angle x-ray diffraction described this protein
as fibrous in nature.

Alexander and Lusena (1961) fractionated membrane material into
five pellets and a soluble fraction by treating washed cream that was
previously stored at -20 C with 2% deoxycholate and employing differential
centrifugation on the resulting mixture. They found that the lipid to
protein and phospholipid to total lipid ratios were much higher in the
supernatant fraction than in any of the pellets. Physical character-
ization of the pelleted material was not attempted because of its
insolubility in normal dispersing agents.

Harwalker and Brunner (1965) studied the effects of dissociating
agents on the membrane fractions and concluded that the insoluble

fraction was a heterogeneous mixture of lipoproteins. Analytical

4

5
ultracentrifugal examination of the solublized material showed a large
number of boundries indicating heterogenity. They concluded from the
effects of these bond-specific dissociating agents on the insoluble
membrane fraction that hydrophobic bonding was the principal stabilizing
factor in the membrane complex with significant contributions from
covalent disulfide bonds.

In an experiment similar to that of Alexander and Lusena, Hayashi
and Smith (1965) treated washed cream with 1% deoxycholate. After
centrifugation, 45% of the membrane protein was in the supernatant along
with 67% of the total membrane phospholipid. The pellet protein
fractions contained a much higher protein to phospholipid ratio. These
workers hypothesized that the membrane was composed of an insoluble
protein region covered by a more soluble outer layer.

In a subsequent experiment, Hayashi (1965) and co-workers showed
the deoxycholate solublized proteins to be heterogeneous by separating
them into three density classes. No attempts were made to classify the
insoluble protein fractions.

Chien and Richardson (1967) used a combination of physical stirring
and centrifugation to separate the membrane proteins into five fractions.
The easily removed material accounted for approximately half of the
original protein and they concluded that this came from an easily
soluble outer layer. They thought that the remainder of the protein
came from an insoluble inner layer of the membrane.

Swope and Brunner (1969) used differential sedimentation to
separate the undissociated membrane proteins into a soluble supernatant
containing approximately 6% of the total protein and three pellet

fractions. They found that as the density of these pellets increased,

6
the protein to lipid ratio increased and the carbohydrate to protein
ratio decreased.

It should be noted that all the methods thus far described for
the characterization of membrane proteins have yielded data pertaining
to the chemical composition of the various fractions obtained.

Attempts to physically characterize the membrane system have met
with very limited success and only indicated that the membrane is com-
posed of a heterogeneous mixture of glyco—lipoproteins. The lack of
data pertaining to the physical composition of the membrane proteins can

be attributed to their resistance to solublization.

Gel Electrophoresis of Membrane Proteins

 

Most membrane proteins have long resisted characterization by gel
electrophoresis because of their inherent insolubility. Harwalker and
Brunner (1965) found that the bulk of fat globule membrane protein
applied to gels did not enter the running gel even after treatment with
various dissociating agents. This prompted Brunner (1969) to conclude
that gel electrophoresis, under normal conditions, was not a satisfactory
technique for studying the membrane complex.

Takayama, EE.El: (1966) developed a system containing phenol,
acetic acid, water (2:1:1, w/v/v), 2 M urea that solublized the proteins
of the mitochondrial transport system. The solublized proteins were
electrophoretically separated on acid gels (7.5% acrylamide, 35% acetic
acid and 5 M urea) with reproducible patterns.

This method was modified by Ray and Marinette (1971) by the
addition of 5% mercaptoethanol to the solublizing solution. The

acrylamide content of the gels were lowered to 6% and the urea to 4 M.

This modification when used on rat liver plasma membranes gave more and
sharper zones than did the original system.

Keenan, gt_a1,, (1970) employed the original system of Takayama
gt_al,, (1966) to compare the proteins of the fat globule membrane to
those of the mammary gland plasma membrane. They found that the two
membrane systems were almost identical in protein distribution and that
the major component of both systems was the same.

The introduction of a method to estimate the molecular weights of
proteins and their subunits on polyacrylamide gels by Shapiro, et_a1:,
(1967) added a new dimension to disc gel electrophoresis. These authors
reported that the binding of sodium dodecyl sulfate (SDS) to proteins
negates their native charge and causes them to separate on the basis of
their molecular weights only. They compared the relative mobility (i.e.,
the ratio of the mobility of the protein to that of the tracking dye) of
ten proteins of known molecular weights and achieved an essentially
linear relationship. This system incorporated 1% mercaptoethanol into
the sample buffer to break disulfide bonds and insure that the protein
standards were in monomeric form. The combination of SDS and mercapto—
ethanol proved to be a very powerful solublizing agent.

Weber and Osborn (1969) further investigated this system by com—
paring the relative mobilities of 40 proteins of known molecular weights.
They found that by altering the gel concentration or the degree of
cross-linkage within the gel that molecular weights within the range
of 11,000 to 220,000 could be obtained with an accuracy of :_10%.

Weber and Kuter (1971) explained this phenomenon by showing that
all proteins examined bound 1.4 g of SDS per gram of protein regardless

of the size or shape of the protein molecule. The binding of a specific

amount of SDS per weight of protein gave all proteins the same net
charge and converted them into rod shaped molecules whose length was
proportional to their molecular weights.

Rodbard and Chramback (1971) in a theoretical treatment of molecular
size estimations by use of polyacrylamide gels noted points of caution
that should be taken. They point out that if the assumptions of Weber
and Kuter are true, that there is no mathematical basis for being able
to determine the molecular weight of a molecule by electrophoresis under
any conditions if only one gel concentration is used. They cautioned
that when comparisons are made between different systems, the retarda-
tion coefficient of the molecules should be compared and not some
derived function of it, viz., its molecular weight. They further
stated that claims of accuracy in the SDS gel system of :_10% may not be
valid because rigorous statistical treatment of the rather limited data
has not been performed. Their final point was that at least two gel
concentrations should be used if molecular weight estimates are to be
made.

Heriman and Phillips (1970) applied the system of Shapiro to the
membranes of rough and smooth endoplasmic reticulum isolated from rat
liver. They found the system gave reproducible results as long as pre-
cautions were taken to insure that the membranes were always prepared
in the same way. This led to the conclusion that meaningful comparisons
of the protein profiles of a given system are valid only as long as the
nmmhods used to isolate the proteins are identical.

A lipoprotein fraction obtained from erythrocytes was studied by
.Iuliano and Rothstein (1971). They found this previously insoluble
fraction to be totally soluble in a solution containing SDS and mercapto-

ethanol. They reported that the solution used to solublize this fraction

completely dissociated the lipid-protein complex.

The outer membrane of Escherichia coli was solublized with SDS

 

into lipid and protein components by Bragg and Hou (1972). Removal of
the SDS by dialysis caused the reassociation of these components into
membrane-like structures. When these solublized membranes were electro-
phoresed according to the method of Weber and Osborn, a reproducible
pattern was obtained. When the pH of the cathodic buffer was changed by
the addition of NaOH from pH 7.2 to pH 11.0,a reproducible pattern was
also obtained. The patterns obtained between the two systems, however,
were not identical. Thus, another variable must be considered when the
protein patterns of the same substance are compared and the electrophor-
esis is performed under slightly different conditions.

Allen and Crumpton (1972) used 5% SDS to solublize the membranes
of human thymocytes. They stained gels both with coomassie blue and with
periodate-Schiff reagent, obtaining 20 coomassie blue positive zones and
six that were positive for carbohydrate. The carbohydrate-positive bands
corresponded to six of the protein bands, indicating that these were
glycoproteins. A carbohydrate-positive, protein—negative zone was found
near the ion front of each gel. Butanol extraction of the membrane
removed 98% of the lipid and 62% of the neutral sugar. When the butanol
extracted protein was electrOphoresed, the fast migrating band was no
longer present, and was, therefore, considered to be a glycolipid.

In studies on the outer mitochondrial membrane of beef heart,
Hayashi and Roderick (1972) used the method of Shapiro. Carbohydrate
staining indicated only one positive zone migrating ahead of all the
protein zones and was considered to be a glycolipid. These authors

concluded that this portion of the membrane, while being reported to

10

contain carbohydrate, actually contained no glycoproteins.

Keenan and Huang (1972) employed a pH 7.0 phosphate buffer contain-
ing 1% mercaptoethanol and 1% SDS to solublize the fat globule membrane
proteins. This mixture was incubated with stirring at 37 C for four
hours. The remaining insoluble material was removed by centrifugation
and the samples applied to the gel system of Weber and Osborn (1969).
They reported that the fat globule membrane complex contained 14 poly-
peptide units ranging in molecular weights of from 11,500 to 102,000
daltons. The main zone had a molecular weight of 68,000 daltons.

Koblyka and Carraway (1972) used almost identical conditions to
separate the polypeptides of the fat globule membrane. The major
differences between these two reports as to methods employed were:

1. Keenan and Huang used freshly drawn warm milk; Koblyka

and Carraway used chilled milk.

2. Keenan's group used a 10% acrylamide gel, whereas Koblyka
and Carraway used a gel concentration of 5%.

3. Koblyka and Carraway assumed that all polypeptides that
showed molecular weights of less than 50,000 daltons were
contaminants from the milk serum, while Keenan and Huang
assumed that their preparation was free of contaminants
and all bands should be considered part of the membrane
complex.

4. Koblyka and Carraway used both a protein and a carbohydrate
stain, while Keenan and Huang used only a protein stain.

Koblyka and Carraway reported the presence of six coomassie blue
positive zones ranging in molecular weights from 53,000 to 240,000

daltons. They also reported the presence of six carbohydrate containing

11

polypeptides, three of which were definitely different from the coomassie
blue positive zones. The main component in their system had a molecular
weight of 66,000 daltons.

These two reports are significant because they contain the first
published data concerning the actual number of polypeptides in the fat
globule membrane complex and give molecular weight estimates for these
proteins. The disparity in the results, however, indicates the need for

further investigation of the membrane system.

EXPERIMENTAL

Chemical Analysis

 

Nitrogen

Duplicate nitrogen analyses were performed in a semi micro—Kjeldahl
apparatus. Ten to twenty mg of dried sample was mixed with 4 ml of
digestion mixture consisting of 5.0 g CuSO4.5H20 and 5.0 g SeO in 500

ml of concentrated H2804. Digestion was carried on for one hour or
until the solution was clear. The contents of the digestion flasks
were allowed to cool for 30 min and 1 m1 of 30% H202 was added. Diges- -
tion was continued for another hour and allowed to cool for 30 min. The
flasks were then rinsed with 10 ml of deionized water and allowed to
cool for an additional 30 min. The mixture was neutralized with 25 ml
of a 40% NaOH solution and the released ammonia was steam distilled into
15 ml of a 4% boric acid solution containing 5 drops of indicator
solution. The indicator solution was made by dissolving 400 mg of
bromocresol green and 40 mg of methyl red in 100 m1 of 95% ethanol. The
distillation was continued until a final volume of 75 ml was reached.
The ammonia-borate complex was titrated with 0.020 N HCl previously
standardized with tris hydroxymethyl aminomethane. A reagent blank was
determined and subtracted from the sample values. The average recovery
of tryptophan was 97.2%.
Hexose

The phenol-sulfuric acid method of Dubois, et_al, (1965) was used

to determine hexose. Five hundred to one thousand ug of sample in 1 ml

12

13

of water was mixed with 1 m1 of 5% mixture of redistilled phenol and water
in a 25 m1 Erlenmeyer flask. Five m1 of concentrated H2304 was added
rapidly while the flask was agitated to insure maximum mixing and heat
development. The samples were allowed to stand at room temperature for

10 min and were then kept in a 25 C water bath for an additional 20 min.
Transmittance was read at 490nm. A standard curve was constructed to
cover a range of 0 to 50 ug of a galactose-mannose (1:1) mixture. A

reagent blank was prepared by omitting the protein from the reaction

mixture.

Hexosamine

 

The method of Johnson E£.§l: (1960) was used to determine the
hexosamine content of the samples. One to two mg samples were weighed
directly into 5 m1 ampoules. One ml of 4 N HCl was added and the samples
were frozen in a mixture of dry ice and ethanol. The ampoules were
evaporated, refrozen and sealed under a vacuum. Standard mixtures of
glucosamine-galactosamine (1:1) were treated in the same manner as the
samples to serve as recovery standards. The samples were placed in a hot
air oven at 100 C and hydrolyzed for six hours to release hexosamine. The
hydrolyzed samples were transferred to a micro-Kjeldahl flask and the
ampoules were rinsed with 1 ml of 4 N NaOH and two 1 m1 rinses of
deionized water.

Acetylacetone reagent was prepared by adding 1 ml of distilled
acetylacetone to 25 ml of l M Na2C03 solution and 20 m1 of H20. The pH
was adjusted to 9.8 and the solution made up to 50 ml with deionized
water. The acetylacetone reagent must be used within 30 min of its prepa-

ration. Ehrlich's reagent was prepared by dissolving 2 g of p-dimethylamino-

benzaldehyde into 250 ml of a 3.5% solution of HCl in ethanol.

14

Five and five-tenths m1 of the acetylacetone solution was added to
each digestion mixture. The final pH of this mixture should be
between 9.5 and 10.0. The flasks were stoppered and placed in a boiling
water bath for 20 min. After cooling, the flasks were attached to a
micro-distillation apparatus and the chromogen was distilled into a 10 ml
volumetric flask containing 7.5 m1 of Ehrlich's reagent. Transmittance
was read after 30 min at 548 nm.

A standard curve containing 0.0 to 20 ug was constructed from a
mixture of glucosamine-galactosamine (1:1). The average recovery was 97%
of the standard mixture. A reagent blank contained 1 ml of 4 N NaOH,

1 ml of 4 N HCl and 2 m1 of H 0 that was treated in the same manner as

2

the samples.

Sialic Acid

 

Warren's (1959) thiobarbituric acid method as modified by Marier,
§£_al, (1963) was used to determine the sialic acid content of the pro—
teins. Sialic acid was released from the samples by hydrolyzing 5 mg
samples in 5 ml of 0.1 N H2804 for 40 min at 80 C. Aliquots of from
0.1 to 0.5 ml were made to a total volume of 0.5 ml with deionized
water in a conically-shaped test tube. One-tenth ml of a sodium

metaperiodate solution (0.2 M in 9M H P04) was added. The samples were

3
mixed and allowed to stand at room temperature for 20 min. Because of
the small volumes employed, adequate mixing is very important. One ml

of a sodium arsenite solution (10% in 0.5 M Na2S04 and 0.1 N H2S04) was
added and the tubes were shaken until the yellow brown-color disappeared.
Three m1 of a barbituric acid solution (0.6% recrystallized barbituric

acid in 0.5 M Na2804) was added. The tubes were agitated vigorously to

extract the chromophore into the cyclohexanone layer. The emulsion

15

was broken by centrifugation and transmittance of the upper layer was
monitored at 549 nm. A standard curve was prepared using N—acetyl
neuraminic acid as a standard in concentration of from 0 to 20 ug.
Deionized water was used instead of sample in determinations serving

as reagent blanks.

Total Lipid

 

Total lipid was determined by a micro-modification of the method
of Monjonnier and Troy (1925). Duplicate samples of from 10 to 25 mg
were weighed directly into conically-shaped centrifuge tubes. One and
five-tenths m1 of a 2% KCl solution was added to each tube along with
2 drops of concentrated NH4OH. The tubes were capped and allowed to
stand at room temperature for 2 hr to solublized the samples. One and
five-tenths m1 of 95% ethanol was added to each tube and the tubes were
capped and agitated for l min. Two and five-tenths ml of ethyl ether
was added to each tube; the tubes were capped and agitated for 30 sec.
The pressure was released and agitation was continued for an additional
30 sec. Two and five-tenths ml of petroleum ether was added to each tube
followed by the agitation procedure described above.

The tubes were centrifuged to separate the phases and the upper
solvent phase was removed and added to a previously tared aluminum dish
that was held at 45 C on an electric hot plate. Each tube was rinsed
with 1 m1 of a mixture of ethyl ether and petroleum ether (1:1) and the
extraction procedure was repeated twice. The resulting solvent layers
were added to the dish and evaporated to dryness. The dishes were then

placed in a vacuum oven at 100 C. After 30 min, they were transferred

to a dessicator, allowed to cool, and weighed. The solvent blanks were

16

treated in the same manner and their average weight was subtracted from
the weight of each sample. Results were expressed in percentages of

total sample weight.

Phospholipid

 

The limited amount of lipid material recovered by the lipid
extraction procedure negated the usually employed lipid-phosphorus pro-
cedure. Thus, it was assumed that all phosphorous measured in a given
membrane fraction was a constituent of the phospholipid moiety. Swope
and Brunner (1968) demonstrated that the amount of phosphorous found in
the protein portion of three membrane fractions was fairly low and
nearly constant between fractions. They believed that this phosphorous -
was attributable to residual phospholipid material.

A modification of the method of Ames (1966) was used to determine
the phosphorous content. The method as described is capable of deter-
mining 0.01 uM of phosphate.

Duplicate aqueous samples, containing 200 to 250 pg of membrane and
0.06 ml of a 10% solution of Mg(N03)2. 6H20 in 95% ethanol, were heated
over a Bunsen burner in a Pyrex test tube until the brown fumes disappeared.
Six-tenths ml of 0.5 N HCl in deionized water was added and the tightly
stoppered tubes were heated in a boiling water bath for 15 min and
allowed to cool.

To each tube was added 1.4 ml of a freshly prepared solution con-
taining 1 part of a 10% ascorbic acid solution and 6 parts of a 0.42%

ammonium molybdate solution in l N H 804. The tubes were incubated at

2
45 C for 20 min and the contents volumetrically diluted to 5.0 ml with

deionized water. Transmittance was monitored at 660 nm.

17

Standard solutions containing from 0 to 3.72 pg phOSphorus along
with a reagent blank were treated in a manner identical to that of the
samples to construct a standard curve. The amount (ug) of phosphorus

in the samples was multiplied by 25 to give the phospholipid equivalents

in pg.

Lowry Protein

 

The method of Lowry, et_a1: (1951) was used for analysis of the
protein content of non-lyophilized membrane fractions.

Five-tenths ml of each sample was placed in a test tube with 0.5 ml
of 0.2 M NaOH (final concentration of 0.1 M NaOH in 1 m1) and was allowed
to stand at room temperature for 30 min.

Five m1 of alkaline copper solution (made freshly by the dilution
of 1 ml of a solution of 0.5% cupric sulfate and 1% sodium tartrate to
50 ml with a 2% sodium carbonate solution) was added to each sample and
the mixture allowed to incubate at room temperature for 10 min.

After incubation, 0.5 ml of Folin reagent diluted 1:1 with deion—
ized water was added with rapid agitation. Transmittance was read at
750 nm after 30 min. A standard curve was constructed with crystallized
bovine serum albumin in the range of from 0 to 250 pg. The standard
protein and a reagent blank were treated in the same manner as the

membrane fractions.

Physical Methods

 

Preparation and Fractionation of Membrane Proteins
The milk used in this work was obtained from the Michigan State
University dairy herd which consisted mainly of Holstein cows. The

milk was collected warm, at the time of milking and separated within

18

one hour at 37 C. The separated cream was diluted with three volumes of
distilled water kept at 37 C and reseparated. This procedure was re-
peated three times to insure adequate removal of skim milk proteins
(Swope, 1968). The washed cream was stored at 4 C overnight and churned
in Erlenmeyer flasks on a rotary shaker at room temperature. After the
emulsion was broken, the aqueous portion was filtered through four
layers of cheese cloth. The unchurned butter granules were removed by
centrifugation at 1000 x g for 30 min. The resulting aqueous suspension
was designated as the original membrane material (0).

Figure 1 represents the scheme used to separate the membrane
material into five pellet (l-P through 5-P) and five supernatant (l-S
through 5-S) fractions. In all cases the pellets were resuspended in a ’
volume of 0.6 M KCl equal to the original volume of the 0 fraction and
allowed to stir overnight at 4 C. Samples of the original membrane
material, the supernatants and the pellets were dialyzed exhaustively
against several changes of distilled water at 4 C. Portions of the
dialyzed fractions were saved for physical analysis and the remainder
lyophilized for subsequent chemical analysis.

The sedimentation values for the pellets were determined by a

rearrangement of the equation of Trautman (1961):
13

 

 

S = 10 1n (ra/rb) ;
w 2T
. 13
to yield ST = 10 1n (ra/rb) ;
w 2
where S = Svedberg units
T = time in hours
ra = the radius from the center of rotation (cm) to the bottom

of the liquid

19

rb = the radius from the center of rotation (cm) to the
air-liquid meniscus, and
w = the angular velocity in rad/min = 2n (rpm).

For the 30 rotor:

r = 10.5 cm
a
rb = 5.0 cm
rpm = 30,000
would yield
13

ST = 10 In (10.5/5.0) 208

 

(2n)2 (30,000)2
The samples were centrifuged at 30,000 rpm for 2 hr which yields:

5(2)

208

S 104 S pellet.
Polyacrylamide Disc-Gel Electrgphoresis

Disc electrophoresis was performed in the acid system of Takayama
(1966) and the sodium dodecyl sulfate (SDS) system of Weber and Osborn
(1969). Electrophoresis was conducted in a 12 sample Buckler apparatus.
Gels were stained for protein for at least 2 hr in a solution of coomassie
blue as described by Weber and Osborn (1969) and destained by diffusion
in a Bio Red destainer. Gels were stained for carbohydrate by the method
of Zacharias (1969).
Acid Gels

The acid system employed gels formed with 7.5% acrylamide, 35%
acetic acid and contained 5 M urea. The samples were solublized in a
system of phenolacetic acid—water (2:1:1, w/v/v) with the aid of

ultrasonification. Samples were applied to the gels followed by a layer

20

of the solublizaing solution. The buffer tanks were filled with 10%
acetic acid with the lower tank serving as the cathode. Electrophoresis
was carried out at room temperature for 2.5 hr with a constant current
of 5 mA/ tube. The gels were stained and destained as previously

described.

10% SDS-Gels

 

The 10% acrylamide, sodium dodecyl sulfate (SDS) gels were run
according to the method of Weber and Osborn (1969) with two exceptions:
1. The tank buffer was diluted 1:1 with deionized water,
2. The samples and standards were dissolved and applied
to the gels in 0.05 M phosphate buffer (pH 7.1) containing
1% SDS and 1% mercaptoethanol.
The samples were layered over the gels by increasing their densities
with sucrose. Electrophoresis was carried out at room temperature with
a constant current of 8 mA/ tube for 4 hr. The lower electrode served
as the anode. Gels were stained for both protein and carbohydrate as

previously described.

5% SDS-Gels

 

In order to obtain more reliable estimates of the molecular weights
of large polypeptides, % acrylamide gels were used. For these gels
commercial Cyanogum 41 was used in place of acrylamide and bisacrylamide.
The samples were dissolved and electrophoresed in exactly the same
manner as described above for the 10% SDS-gels. The time of electrOphoresis
was reduced from 4 to 3 hr.

Molecular weights of protein subunits were estimated from a plot

of the relative mobility versus the log of the molecular weights of

21

protein standards run on both concentrations of SDS gels. The standards
used were B-lactOglobulin (18,400), ovalbumin (43,000), catalase (60,000),
and phosphatase A (94,000). Relative mobilities were calculated by
measurement of the protein zone migration distance,dye migration distance
and the length of the gels from the following relationships:

Relative mobility = distance prptein migrated x
length of gel after destaining

 

length of gel before destaining,
distance dye migrated

 

Buttermilk (0)

made to 0.6 M KC1
centrifuged at 4 C
to 100 S pellet

 

 

 

Flotation layer lst KCl Pellet (1-P) lst KCl Supernatant(l-S)

(discard) made to 0.6 M KCl (save)
centrifuged at 40C
to 100 S pellet

 

 

 

Flotation layer 2nd KCl Pellet (2-P) 2nd KCl Supernatant(2-S)

(discard) made to 0.6 M KCl (save)
centrifuged at 4°C
to 100 S pellet

 

 

3rd KCl Pellet (3-P) 3rd KCl Supernatant(3-S)

made to 0.6 M KCl (save)
centrifuged at 40C
to 100 S pellet

4th KCl Pellet (4-P) 4th KCl Supernatant(4-S)

made to 0.6 M KC1 (save)
centrifuged at 4 C
to 100 S pellet

 

 

 

5th KCl Pellet (S-P) 5th KCl Supernatant(5-S)
(save) (save)
Figure 1. Diagram for the sequential extraction of membrane

material with 0.6 M KCl
22

RESULTS

Distribution of

 

Protein, Lipid, and Phospholipids

 

Between Membrane Fractions

 

The recoveries of protein, lipid, and phospholipid for each membrane
fraction are presented in Table I. The first column gives the actual
protein recovery as the percentage of the original material. There was
a 7.7% loss of the toal protein after the first centrifugation, probably
due to the discarding of the flotation layer. To better represent the
manner in which the recovered protein was distributed between the various
fractions, the data in the second column represent the protein recovery
for each fraction when the sum of the protein contents of l—S and l-P
are taken as 100%. The third and fourth columns of Table I show the
distribution of lipid and phospholipid between the membrane fractions.

No corrections were made for losses or gains of these components. The
recovery of material in the supernatant fractions 3-S through S-S was

too low to allow for accurate lipid or phospholipid determinations.

Chemical Composition

 

0f Membrane Fractions

 

Table 11 represents the percentage composition of the membrane
fractions in terms of protein, lipid, hexose, sialic acid, and protein
+ total carbohydrate.

The average nitrogen values of membrane protein fractions reported

by Swope and Brunner (1970) ranged from 14.01% to 14.14% as compared to

23

24

approximately 16.0% for normal milk proteins. The difference presumably
being due to the fact that membrane proteins are actually glycoproteins
(Thompson and Brunner, 1959). For purposes of calculating the total
protein content of the membrane fractions, it was assumed that all the
carbohydrate detected was conjugated to the proteins. In all subsequent
comparisons of components to protein, the protein content will be taken
as the sum of the protein and carbohydrate moieties. Calculations
derived from the data of Table 11 show, in fact, that if the nitrogen
contents of fractions l-S and l-P are divided by the protein + carbohy—
drate content of these fractions, the resulting percentages of nitrogen
are 14.5% and 14.2% respectively. These values are in agreement with the
aforementioned data of Swope and Brunner and strengthens the assumption'
that the total protein content of the membrane fractions is best repre-
sented by the sum of the values for protein and carbohydrate.

Table III expresses the same data of Table II as component ratios
rather than percent composition for the chemical components of the

membrane fractions.

Molecular Weights Determined

 

By Gel Electrophoresis

 

The molecular weights obtained for the polypeptides of the membrane
fractions are tabulated in Tables IV, V, VI, and VII. Tables IV and V
give the values for the pellet fractions in 10% and 5% gels respectively.
Tables VI and VII give the values for the supernatant fractions in the
same gels.

Gel Electrophoretic Patterns

 

Acid System

 

The electrophoretic patterns obtained for fractions 0, 1-S, 2-S,

25
l-P and 5—P, where the samples were solublized with phenol- acetic acid-
water, (2:1:1, w/v/v) and applied to the gel system of Takayama, gt_a1:,

(1966), are shown in Figure 2.

SDS Systems

 

Figures 3 and 4 show the electropherograms obtained when the pellet
fractions were applied to 10% and 5% SDS gels and stained with coomassie
blue to show proteins. Figure 5 is the pattern obtained when the pellet
fractions are stained for carbohydrate by the periodate—Schiff method
after being assayed on 10% gels.

Figures 6 and 7 show the patterns obtained when the supernatant
fractions were stained for protein after being applied to 10% and 5%
gels. Figure 8 is the pattern obtained when the supernatant fractions
are applied to 10% gels and stained for carbohydrate.

Figure 9 compares the patterns obtained for fractions l-P and 5—P

on a 10% gel stained for protein.

TABLE 1. Recovery of fat globule membrane components in sequential

supernatant and pellet fractions

 

(a) (b)

 

 

Fraction Protein Protein Lipid Phospholipid
Original(c) 100.0 ---- 100.0 100.0

1 S 44.5 48.2 40.0 57.4

2 S 11.9 12.9 12.5 16.5

3 S 1.0 1 0 * *

4 S 1 0 1.0 * *

5 S 1.0 1 0 * *

1 P 47.8 51.8 38.8 48.4

2 P 35.8 38.8 20.9 30.9

3 P 34.9 37.8 19.7 29.9

4 P 34.0 36.9 14.0 25.0

5 P 33.1 35.9 13.5 26.1

(a) Percent recovery of protein with the original sample as 100.0%
(b) Percent recovery when the 7.7% loss of protein due to the fluff

is disregarded
(c) Original material is freshly prepared buttermilk
(*) Insufficient sample for meaningful assay

26

 

 

TABLE 11. Distribution of fat globule membrane components in sequential
supernatant and pellet fractions (expressed as percentage)
Fraction Protein Lipid Hexose iiiéic Hexosamine ngtgin Total
%
0 37.02 53.00 1.54 0.55 1.29 40.37 93.40
1- 33.35 57.25 1.33 0.63 1.42 36.73 93.98
- 18.81 72.88 0.94 0.38 1.72 21.85 94.73
1-P 38.09 51.88 2.01 0.76 1.98 42.84 94.72
2-P 46.02 43.95 1.75 0.76 1.92 50.43 94.40
- 47.03 42.96 1.51 0.77 1.94 51.25 94.21
- 54.26 35.73 1.60 0.73 1.78 58.37 94.10
- 54.62 35.36 1.63 0.70 1.55 58.50 93.86

 

27

 

 

 

 

 

28

0Hq.0 000.H mqm.0 000.0 mm0.0 NH0.0 0N0.0 000.0 mum
mmm.0 qm0.a NNN.0 050.0 Hm0.0 mH0.0 nm0.0 000.0 mlq
nmm.0 mmH.H NON.0 N00.0 mm0.0 0H0.0 00m.0 mam.0 mlm
nHm.0 an.H 00N.0 000.0 0m0.0 0H0.0 000.0 «00.0 min
m0~.0 0Nw.0 Ham.0 HHH.0 000.0 0H0.0 500.0 w~0.0 mIH
00H.0 00m.0 0N0.0 mmH.0 mn0.0 ma0.0 m00.0 mHN.0 mum
mam.0 ~00.0 mom.0 N00.0 m00.0 NH0.0 0m0.0 n0m.0 mIH
mHN.0 N0n.0 00m.0 m00.0 «no.0 «H0.0 wm0.0 000.0 0
wfinfiq wfiawq cHeuonm afieuoum cwououm awwuoum cfiwuowm HmuOH mafiuemwm
wfimfiaonmmosm samuoum vfimwwonmmonm 0mm mcfiemmoxwm 0wum oHHmfim emoxom cweuoum

AmOHumu unecanoo mm 00mmmuaxev maoauumum
ueHHma 0cm ucmumcuwasm Hmaucmzvem ca mucmaoaaoo mcmunams madnon umw mo coausnwuuwfio .HHH mqm<e

TABLE IV. Molecular weights of the polypeptides of fat globule membrane
pellet fractions as determined by SDS electrophoresis in 10%
polyacrylamide gelsa

 

 

 

Fraction
l-P 2-P 3—P 4—P 5—P
138,000 138,000 138,000 138,000 138,00
123,000 123,000 123,000 123,000
118,000 118,000 118,000
112,000 112,000
110,000 110,000 110,000
107,000 107,000 107,000 107,000
90,000 90,000 90,000 88,000 88,000
86,000
82,000 82,000
74,000 74,000 74,000 74,000
70,000 70,000 70,000
66,000 66,000 63,000 66,000 63,000
58,000
54,000 54,000
46,000 48,000 48,000 46,000 48,000
39,000 39,000
36,000
28,000
23,000 23,000 23,000 23,000 23,000
15,500 15,500 15,500 15,500 15,500
13,500 13,500 13,500

 

average of five determinations

29

TABLE V. Molecular weights of the polypeptides of fat globule membrane
pellet fractions as getermined by SDS electrOphoresis in 5%
polyacrylamide gels

 

 

 

Fraction
l-P 2-P 3-P 4-p s-p
208,000 208,000 208,000 208,000 208,000
174,000
151,000 153,000 153,000 153,000
140,000 140,000 140,000 140,000
133,000 133,000
123,000 123,000
108,000 108,000 108,000 108,000 108,000
90,000 90,000 90,000
86,000 86,000
82,000 82,000 82,000 82,000 82,000
60,000 64,000 63,000 63,000 60,000
46,000 46,000 48,000 48,000 46,000
39,000 39,000 39,000 39,000 39,000
30,000 30,000 30,000 30,000 30,000
22,000 23,000 23,000 23,000 23,000

 

a . . .
average of five determinations

30

TABLE VI. Molecular weights of the polypeptides of fat globule membrane
supernatant fractions as determined by SDS electrophoresis in
10% polyacrylamide gels

 

 

 

Fraction
0 l-S 2—S 3-S 4—S S-S

123,000 123,000 123,000
118,000 118,000 118,000 118,000 118,000 118,000
90,000 90,000 90,000 96,000 96,000 96,000
80,000 80,000 80,000
76,000 74,000 74,000
63,000 60,000 60,000
56,000

48,000 48,000
44,000 43,000 43,000
40,000 40,000
35,000 35,000
32,000 32,000
29,000
26,000
23,000 23,000
17,000 18,000
14,000 15,000

 

average of five determinations

31

TABLE VII. Molecular weights of the polypeptides of fat globule membrane
supernatant fractions as determined by SDS electrophoresis in
5% polyacrylamide gels

 

 

 

Fraction

0 1-8 2-S 3-S 4—S 5-S
153,000 153,000 153,000
140,000 140,000 140,000 140,000 140,000 140,000
108,000 108,000 108,000 108,000 108,000 108,000
90,000 90,000 90,000
81,000 82,000 80,000
75,000
60,000 60,000 60,000
57,000
48,000 48,000 48,000
39,000 37,000
32,000 32,000
23,000 23,000

 

a . . .
average of five determinations

32

 

.
-‘~Ous\-sh‘.o
§\\\d\. .\Q

Q“ .s
3.....1 .

.sksv

q-

n
'
.

h \u,\

 

E .H- .-
2:33
a'
:I

o o 0
6 4 2

120—
100.—- -
80—

Tfi u SEE 3039.

5P

4P

3P

2P

1P

34

Figure 4.-—S% polyacrylamide-SDS gel electrophoretograms
of the membrane pellet fractions stained for protein.

Figure 5.--10% polyacrylamide—SDS gel electrophoretograms
of the membrane pellet fractions stained for carbohydrate.

35

MOLECULAR WEIGHT x 10'3

INT—-
150-—'

12W“ “

90—

60—

3.0 ll
3:.

30—-

IP 29 3? 4p

HMGHMI

 

1P 2P 3? 4p
FRACTION

36

5F

 

‘

5P

Figure 6.--10% polyacrylamide-SDS gel electrOphoretograms
of the membrane supernatant fractions stained for protein.

Figure 7.--5% polyacrylamide—SDS gel electrophoretograms
of the membrane supernatant fractions stained for protein.

37

120—
100—

80"—

60"-

20. ‘-

mmcuun WEIGHT 1 10'3
i

180'-
15‘!—

ancuun menu I: 10'3

F

O

 

 

 

b..-

25

3S

FRACTION

 

3's
FRACTION

38

4S

{'5

55

55

Figure 8.--10% polyacrylamide-SDS gel electrophoretograms
of the membrane supernatant fractions stained for
carbohydrate.

Figure 9.-—10% polyacrylamide-SDS gel electrophoretograms
of pellet fractions l-P and S-P stained for protein.

 

 

I P s P
FRACTION

40

DISCUSSION

Distribution of Protein, Lipid and Phospholipid

 

Between Membrane Fractions

 

Centrifugation of the original buttermilk material made to 0.6
M KCl to yield fractions 1-S and l—P resulted in an almost even distri-
bution of lipid, protein, and phospholipid between the two fractions.
There was a 7.7% loss of protein and 21.1% loss of lipid following this
preparative step which is attributed to the discarding of the flotation
layer formed during centrifugation. A similar layer was reported by
Swope (1968) who speculated that this material, high in lipid and low
in protein, might be analogous to the chylomicron fraction of plasma
lipoproteins.

Resuspension of the pelleted material with subsequent recentrifugation
resulted in 25.10% of the protein, 46.13% of the lipid and 36.16% of the
phospholipid of fraction l-P being recovered in fraction 2—S. The amount
of material lost to the flotation layer was 0.21% of the protein, 13.92%
of the lipid, and 2.07% of the phospholipid.

It is doubtful that resuspension of fraction l-P with 0.6 M KCl
resulted in the solubilization of all the material that was found in
fraction 2-S. Mere probably a large portion of this material was
actually low-density lipoprotein that was pelleted in the first centri-
fugation, being trapped by the higher density material.

All further resuspensions and recentrifugations of the pellet

fractions resulted in a solublization of less than 1% of the original

41

42

membrane protein which lends credence to the theory that the large amount
of material present in fraction 2-S was not solely a result of solubliza-
tion. The low yields of fractions 3-S through S-S negated meaningful
chemicals analysis of these fractions.

Data in Table 11 show that the loss of lipid from each pellet
fraction is much greater than the loss of protein. This observation
indicates that the increasing insolubility of sequential pellet fractions

results from the loss of lipid material.

Composition of Membrane Fractions

 

Pellets

The data in Table 111 represents the chemical composition of the
supernatant and pellet fractions of the membrane material expressed as
component ratios.

The protein and lipid contents of the five pellet fractions are
inversely related to each other. Whereas the protein content in pellets
l-P to S-P increases from 42.84% to 58.50%, the protein to lipid ratio
increases from 0.83 to 1.65.

Closer examination of the data pertaining to lipid concentrations
reveals that the ratio of phospholipid to total lipid increased from 0.27
for pellet l-P to 0.47 for pellet 5-P, indicating that phospholipids are
being selectively retained by these fractions. That some phospholipids
are being lost can be seen by the decrease in the phospholipid to protein
ratio between fractions 1-P and S-P, viz., from 0.31 to 0.24.

The net effect is a gradual erosion of protein, a greater erosion of
phospholipid and a much greater erosion of lipid from the pellet fractions
as the extraction with KCl progresses. These data are shown graphically

in Figures 10 and 11.

43

The hexose content of the pellets gradually declined from 2.01%
of the total weight for pellet 1-P to 1.6% of the total weight for pellet
4-P. The loss of hexose from pellet l-P to pellet S-P represents a
change of 18.91% of the total hexose. The change in the hexose to pro—
tein ratio, however, is much more dramatic. The ratio for pellet l-P is
0.047 and 0.028 for pellet S-P; a net change of 40.43%. In this case,
as in all comparisons of carbohydrate to protein content, the actual
loss of hexose is much less than is the decrease in the ratio of hexose
to protein. This can be explained by the observation that the loss of
lipid material is greater at each step than is the loss of carbohydrate.
The loss of lipid increases the relative concentration of protein so
that as the carbohydrate is decreasing by about 20% the protein concen-
tration in the pellet is increasing by approximately 37%. The net
effect is a dramatic decrease in the hexose to protein ratio.

The above relationship is most readily seen when the data for
sialic acid are examined. The total percentage loss is from 0.76% for
pellet l-P to 0.70% for pellet S-P or a loss of 8%. The ratio of sialic
acid to protein decreases from 0.018 for pellet l-P to 0.012 for pellet S-P;
a loss of 33%. The almost constant content of sialic acid between pellet
fractions when compared to the ratio of sialic acid to protein for these
pellets indicates that sialic acid containing proteins are lost from the
pellet fractions.

The loss of hexosamine from the pellet fractions is noteworthy.

The absolute amount of hexosamine remains almost constant for pellets
l-P, Z-P, and 3-P. There is a large loss between pellets 3-P and 4-P,
and between 4-P and 5-P. The overall result is that from pellet l-P to

pellet 5-P the total loss of hexosamine and the ratios of hexosamine

44

to protein give values that are within experimental error of those
obtained for hexose.

In terms of toal carbohydrate the change in percent of the total
pellet from pellet l-P to pellet 5-P is from 4.75 to 3.88%; a loss of
18%. The ratio of carbohydrate to protein changes from 0.111 to 0.066
for a loss of 40.5%. Again the change in toal carbohydrate to protein
ratio is affected much more by the relative increase in protein concen-
tration in sequential pellet fractions than by the actual loss of car-
bohydrate material. Restated, the loss of lipid from each pellet
fraction is much greater than is the loss of carbohydrate. The loss of
carbohydrate from pellet 2—P through pellet S—P is linear within experi-
mental error while the loss of lipid is stepwise.

The change in the ratio of carbohydrate to protein is stepwise in
a manner closely following the decrease in lipid, indicating further that
the major change in the composition of the pellet fractions is the loss

of lipid material.

Sppernatants

 

The first supernatant material is similar to the first pellet
fraction, being lower in protein and carbohydrate than the first pellet
and higher in lipid material. The main difference is the increase in
the ratio of phospholipid to lipid.

The second supernatant material is characteristically different
from other fractions, being high in lipid and low in protein with a
protein to lipid ratio of 0.300. The hexose to protein and sialic acid
to protein ratios are similar to fraction 1-P. The greatest difference
is in the hexosamine to protein ratio of 0.079 compared to 0.046 for

pellet l—P, the fraction with the next highest ratio.

45

The phospholipid to lipid ratio of fraction 2—S of 0.199 is the
lowest of all fractions examined while the phospholipid to protein ratio
of 0.426 is the highest of any fraction. The total carbohydrate content
of this fraction is 13.9%, a value comparable to the 15.0% estimated
for a pure glycoprotein isolated from a soluble membrane protein by Swope,
§t_al: (1968).

In general, from these data, certain trends can be seen in the
composition of the membrane fractions as related to their insolubility.
The most significant variables are the ratios of protein to lipid and
phospholipid to protein.

In all cases, the more soluble fractions have a lower protein to
lipid ratio and a higher phospholipid to protein ratio than do the more

insoluble fractions.

Molecular Weights of Membrane Proteins

 

The molecular weights obtained for the various membrane fractions
are listed in Tables III, IV, V, VI. The values listed represent an
average of five determinations. If a component failed to appear on at
least three to five replicates, it was not included in the total.

The tables list the molecular weights obtained in decreasing size
for each fraction at a particular gel concentration. An empty space
indicates that a band at that molecular weight failed to appear on at
least three of the electrOpherograms obtained for that fraction.

While the molecular weight data obtained are informative, caution
must be exercised in the interpretation of these results. It should be
noted that the molecular weights derived from the two gel concentrations
are in reasonable agreement between the range of approximately 55,000

to 110,000. Values outside of this range may vary by as much as 50%

46

for the same band at different gel concentrations and cannot be assured
to be accurate until agreement on their molecular weight is achieved at
two different gel concentrations (Rodband and Chramback, 1971). Even to
consider the values within the region of agreement between the two gel
concentrations as unequivocally accurate values can not be justified.
The basic assumption for the accuracy of molecular weight determinations
on SDS gels is that every protein binds the same amount of SDS and that
separation is due to size alone (Weber and Kuter, 1971). That all
proteins do not bind the same amount of SDS has been shown by Kogen,
35.31: (1972). They found that under optimum conditions elastin binds
only 0.5 g of SDS per g of protein as compared to the assumed value of
1.5 g/g. Some, if not all, membrane proteins may vary from the ideal as
was noted by Carraway and Koblyka (1970) who worked with erythrocyte
membrane proteins. They stated that their results suggested that very
hydrophobic proteins may show non-ideal behavior in SDS gels when compared
to molecular weight standards. Another problem associated with making
reliable molecular weight estimates for proteins in a complex system is
that the relative mobility of a given band is governed by its concentra-
tion (Fairbanks, §t_a1,, 1971). When large amounts of sample are
applied in order to visualize minor components the higher concentration
of the major components increases their mobility as well as the mobilities
of minor components that travel just ahead of them. Thus, a broad major
compnent band may cover an area on a gel corresponding to a molecular
weight of 20,000 or more. Always measuring the leading edge of the

band does not completely standardize the method if more than one concen-
tration of the sample is used to measure the relative mobilities. For

these reasons, to consider the values listed in Tables IV through VII

47

the actual molecular weights of the membrane proteins would be open to
severe criticism. True molecular weight values cannot be determined
until each fraction is isolated in a pure enough form to allow for a more
rigorous analysis.

The values obtained, however, are fairly reproducible for a given
fraction at a given gel concentration. For this reason they may justi-
fiably be used as points of reference to observe changes in the electro-
phoretic patterns obtained between fractions. Thus, a given band will be
referred to by its apparent molecular weight in the indicated gel system.

Molecular weight estimates are not included for the periodate-
Schiff reagent positive bands for two reasons:

1. Glycoproteins have been shown to bind varying degrees of

SDS and their molecular weights vary greatly between
different gel concentrations (Butscher, 1971).

2. The periodate-Schiff reagent positive bands were, in most
cases, very diffuse which made the determination of a
reproducible value for their relative mobilities precarious.

The diffuse characteristic of these bands is attributalbe to the
employment of a dialysis step prior to staining to prevent adventitious
dye binding to the SDS- protein complexes (Glossman and Neville, 1971).
This requires prolonged soaking of the gels and probably allows for some

diffusion of the glycoproteins.

Gel Electrophoretic Patterns

 

Of Membrane Proteins

 

Acid Gels

Figure 2 shows the electrophoretic patterns of fractions 0, l-S,
2-S, l-P, and S-P on the acid gel system of Takayama, §t_313 (1966).

An amount of each sample containing 5 mg of protein was dissolved
in 1 ml of the phenol, acetic acid, water system as previously described.
Twenty five 1 of each sample was applied to each gel. The system works
well for the soluble fractions. However, the more insoluble fractions
form a thick gel rather than a solution and further addition of solvent
does not disperse the gel. This characteristic was noted by Osborn (1966)

when a similar phenol containing system was used to solublize the membrane

and enve10pe of various strains of Salmonella. This gelation character-

istic makes representative sampling of the mixture uncertain and tends

 

to impede a portion of the material from entering the gel. This phenom-
enon is shown in Figure 2 by the increasing amount of material that does
not enter the gel and by the decreasing intensity of the bands that do
enter when fractions of increasing insolubility were examined. For these
reasons this system was considered unsatisfactory for comparing the

membrane fractions and a different solublizing system was utilized.

SDS Gels
Figure 3 shows the electr0phoretic patterns obtained from fractions
l-P through S-P applied to a 10% SDS gel and stained with coomassie blue.

The amount of material applied to each gel was varied in an attempt to

48

49

show as many of the bands as possible. Attempts to employ densitometer
tracings of the electropherograms failed because many of the bands
migrated in close proximity, thus, negating base line readings between
zones. Longer gels offered no improvement in separations. The high
ionic strength of the buffer system was reflected in slow migration rates
of the proteins. Increased gel length requires greater running time,
obstructing any increase in resolution becasue of increased diffusion.
For these reasons, comparisons of the relative contribution that each zone
makes to the total pattern of a given fraction must be estimated visually.
Examination of the electropherograms shown in Figure 3 indicates
that changes are occuring in the protein distribution of successive
membrane pellet fractions. One zone in the first three pellet fractions
possessed an estimated weight of 90,000. With increasing insolubility
of the pellet fraction, this band decreased in intensity until in
fractions 4-P and S-P it represents a minor component. Similarly, the
major bands in fraction l-P, with molecular weights of 66,000 and 46,000,
decreased in intensity with successive extractions. Also, there seems
to be a decrease in the zone representing a molecular weight of 15,500,
and the zone with a molecular weight of 13,500 is on longer visible
beyond fraction 3—P. These alterations in the pellet composition appear
to be balanced by an increase in the intensitives of the zones attribut-
able to molecular weights of 74,000, 112,000 and 138,000 in fraction S-P.
Fraction 5-P is the only specimen containing appreciable amounts of
material unable to enter the gel. The electropherograms in Figure 4
indicate that pellet fractions separated on 5% gels show similar changes
in protein distribution. The principal difference being in the values

estimated for the molecular weights obtained for the zones at both

50

extremities of the gel.

The extreme change in the protein patterns between fractions l-P
and S—P are apparent in Figure 9 which compares these two fractions on
10% gels.

There are two possible explanations for the changes that occur in
the electrophoretic patterns of the pellet fractions:

1. That extraxtion with KCl is selectively removing some

protein components and therefore causing an increase
in the residual components.

2. That extraction with KCl is promoting a reorganization of the

proteins in the system.

Analytical data overwhelmingly support the second possibility. The
amount of protein lost to the supernatant fraction from fraction l-P is
low and in subsequent extractions the loss is neglibible. Concurrently,
there is a notable change in the protein to lipid and phospholipid to
protein ratios of these fractions.

As lipid is lost from successive pellets, the residual protein tends
to aggregate as is evidenced by the decrease in intensity of low molecular
weight and a concurrent increase in the intensity of the large molecular
weight species comprising fractions 1-P to S-P. This apparent aggregation
suggests that the hydrophobic groups of the protein molecules are involved.

Figure 5 represents the electrophoretic patterns of the pellet
fractions applied to 10% SDS gels and stained for carbohydrate. The
patterns indicate that the large molecular weight proteins contain carbo—
hydrate while the smaller ones do not. As was previously noted, it was
impossible to correlate the carbohydrate-staining polypeptides with the

coomassie blue-staining species. Comparisons of the patterns in Figures

51

3 and 5, however, suggest that the major zones of the pellet fractions
contain carbohydrate. In the first pellet fraction (1-P), the most
intensely stained zone corresponds to the 90,000 molecular weight zone.
In the final pellet, (S-P), the molecular weight of the most intense
zones were estimated at 74,000 and 112,000. The very diffuse zone at
the bottom of each gel possesses characteristics of a glycolipid
(Leonard, 1970).

The reason that the lower molecular weight species did not stain
for carbohydrate may be that they were removed from the gels during the
long soaking period required to remove the SDS. Whatever the cause,
these gels add qualitative proof to the contention that at least some of
the membrane proteins are glycoproteins.

The electrophoretic patterns of the original material and of
fractions 1-S through 5-S on 10% SDS gels are shown in Figure 6. In
general, the zones from the supernatant fractions are more distinct and
show less distortion than do the zones of the pellet fractions. This is
an indication of greater homogenity of the components and of less assoc-
iation between different proteins (Nichol, §t_§13, 1964). Fraction 3-S
through S-S show a very faint zone with a molecular weight of 118,000
and, when larger amounts of material are placed on the gels, another
fainter zone with a molecular weight of 96,000. Thus, it appears that
the small amount of protein that is being solublized by KCl extraction
following the second pellet is the 118,000 molecular weight component.
It is of interest to note that one of the major components of fraction
S-P, the most insoluble fraction, showed a molecular weight of 112,000.

The differences in patterns between fractions 0 and l-S are note-

worthy. Fraction 0 is richest in the lower molecular weight species,

52

showing two main zones of molecular weights of 35,000 and 32,000.
Fraction l-S is rich in the low molecular weight componenent, viz.,
15,000, but has three main zones with molecular weights of 48,000,

60,000 and 90,000. Immediately apparent from these patterns is the

shift in the amounts of the higher molecular weight materials between
fractions 0 and l-S. A re-examination of the pattern of fraction l—P
indicates that it is also richer in the higher molecular weight components,
indicating that the original treatment of the membrane material caused

a shift towards the higher molecular weight species. This condition may
have resulted from hydrophobic aggregation occuring concomitantly with
the loss of lipid material. This phenomenon is common in other lipo-
protein systems much as blood plasma (Scanu, 1967), egg yolk (Cook and
Martin, 1969), and mitochondrial structured protein (Green and Fleischer,
1963).

The 2—S fraction seems to consist of two main zones that are also
the main zones of fraction l—S. They possess molecular weights of 60,000
and 90,000.

Figure 7, showing electropherograms of the supernatant fractions
run on 5% gels, illustrates the same shift towards higher molecular weight
species between fractions 0 and 1-8.

The carbohydrate-containing proteins in the supernatant fractions
are shown in Figure 8. Again, the high molecular weight species stain
more intensely for carbohydrate than do the lower molecular weight species.
A notable exception is the low molecular weight zone in fraction 1-S
found just before the glycolipid smear. It should also be noted that in
fraction 2-S the top two zones which are barely visible in Figures 6 and

7 have become more prominant, indicating that these proteins are rich in

carbohydrate.

53

The funnel shaped smear that is seen to extend over the top one-
fourth of the gel for fraction 0 was always present when this fraction
was electrophoresed. Before staining it could be seen as an opalescent
area and is probably related to the higher lipid content of this
fraction.

It should also be noted that fraction 3—S through 5-S, when
stained for carbohydrate, did not contain the glycolipid smear that was
so prominant in the previous fractions.

At first, consideration was given to the fact that the densities of
the fractions were increased to aid in application to the gels by the
addition of sucrose and that this smear might represent an artifact
caused by such additions. Sucrose was added to fractions 3-S through
S-S, however, and the absence of any smear indicated that the soaking
period had effectively removed any free sucrose that might be present.

The electrophoretic patterns raise many questions that can not at
this time be answered. The principal problem being to determine to what
extent the distribution of proteins seen on the gels is related to their
distribution in the native system. It remains to be determined how many
of the zones are single proteins and how many, if any, are the result of
aggregation of similar or different protein species. Also to be determined
is the effect that high concentrations of mercaptoethanol incorporated into
the solublizing system has on the protein patterns observed. Normally, it
would be expected that inter-chain disulfide bonds would be broken, yield-
ing more subunits than in the native state. Leonard (1970) states, how—
ever, that while the use of mercaptoethanol is necessary to achieve com-
plete solublization of erythrocyte membranes, it does not function by

breaking disulfide bonds. It is possible that the interaction between

54

the double bonds of the lipid hydrocarbon chains and sulfhydryl groups
of proteins may be broken by the mercaptoethanol (Robinson, 1966). It
has not been determined if these situations apply to the proteins of the
fat globule membrane.

Elucidation of these questions can only be achieved by studies
with the isolated proteins of the membrane system. Then, their true
molecular weights and amino acid composition can be compared. Until

such data are obtained, the questions raised will remain unanswered.

.mcofiuomhm ueHHem esp
:fi mpcecoasoo oceanEeE efisnofiw Hem xHME mo xhe>oeem .0H ehzwflm

 

 

 

 

 

 

 

5:08...
man 04 a-» d..~ d...
d u d + 4 o
. 0.
a, In
22.. /
0/9 1 ON
o|.I|I-||I||||I|O
22.2320 /o’ 1.. - on
a, ll»!
£380 in, I»
a 1 ov
I 00
. 00
32,33.

:8ch

55

.mcofluompm ueHHem ecu cw mu:0:00£00 ecmanEeE easnofim new xHflE mo :oflpflmomeou .HH enamH0

 

 

 

 

 

c0209.“.
Elm QIV 11m QIN ll—
oo. x 2:280 + 2020:0350 I
a
01111111111 Illlllllllo aHHHUVAHHHHW
3:00 .20... S 203$ .11 In. a I
u n
. .\. l
22.. .23 8 22.2320 .8 .1111“
\ I
u

 

0' a

8:00 .20... .8 £280 oxo

 

0.

ON

Om

O¢

00

00
23.60

56

CONCLUSIONS

Sequential extraction of milk fat globule membrane material with
0.6 M KCl and subsequent centrifugation of the extracted material
yielded five supernatant and five pellet fractions of varying

solubilities.

As the fractions decrease in solubility, the protein to lipid ratio

increases and the phospholipid to protein ratio decreases.

The total carbohydrate content and the ratio of carbohydrate to

protein decrease with decreasing solubility of the protein species.

The most soluble fraction, 2-S, has the highest ratios of phospho-
lipid to protein, hexosamine to protein and the lowest protein to

lipid ratio.

The removal of lipid by extraction with KCl seems to be the main

cause of decreased solubility.

There seems to be an aggregation of low molecular weight proteins
to form higher molecular weight species as the degree of insolubility

increases.

The larger molecular weight proteins stain more intensely for

carbohydrate than do the lower molecular weight proteins.

57

BIBLIOGRAPHY

Alexander, K. M., and Lusena, C.V.
1961. Fractionation of the lipoproteins of the fat globule membrane
from cream. J. Dairy Sci., 44;1414.

Allan, D. and Crumpton, M. J.
1972. Isolation and composition of human thymocyte plasma membrane.
Biochem. Biophys. Acta, 274:22.

Ames, B. N.
1969. Assay of inorganic phosphate, total phosphate and phosphatase.

Methods of Enzymology VIII (Neufeld, E. F. and Ginsburg, V.,
eds.) Acad. Press, New York.

Ascherson, F. M. _
1840. On the physiological utility of the fats and a new theory of
cell formation based on their c00peration and suggested by
several new facts. A translation by Emil Hatschek in
The Foundations of Colloid Chemistry, E. Benn, Ltd., London,
1925.

Babcock, S. M.

1885. A study of the fat globules of milk. N.Y. (Geneva) Agr. Expt.
Sta., 4th Ann. Rpt. 293.

Babcock, S. M.
1889. Fibrin in milk.
Wis. Agr. Expt.Sta., 6th Ann. Rpt., 63.

Bragg, P. D., and Hou, C.
1972. Organization of proteins in the native and reformed outer
membrane of Escherichia coli. Biochem., Biophys. Acta.,
2145478.

 

Brunner, J. R.
1969. Milk lipoproteins. Structural and Functional Aspects of
Lipoproteins in Living Systems, (Tria, E., and Scanu, A.M.,
eds.) Acad. Press, London and New York.

Brunner, J. R.
1973. Physical equilibria in milk: The lipid phase. In Fundamentals
(in of Dairy Chemistry, 2nd ed., (B. H. Webb and A. H. Johnson,
press) eds.) Avi Publishing Co. Inc., Westport, Conn.

58

59

Brunner, J. R., Duncan, C. W., and Trout, G. M.
1953a. The fat-globule membrane of unhomogenized and homogenized milk.
I. The isolation and amino acid composition of the fat-
membrane proteins. Food Res., 185454.

Brunner, J. R., Duncan, C. W. and Trout, G. M.
1953c. The fat-globule membrane of unhomogenized and homogenized
milk. 111. Differences in the sedimentation diagrams of the
fat-membrane proteins. Food Res., 185469.

Brunner, J. R., and Herald, C.T.
1958. Some electrophoretic characteristics of two protein components
of the fat-globule membrane of normal milk. J. Dairy Sci., 41:1489.

Brunner, J. R., Lillevik, HA. A., Trout, G. M., and Duncan, C. W.
1953b. The fat-globule membrane of unhomogenized and homogenized milk.
II. Differences in the electrophoretic patterns of the fat-
membrane proteins. Food Res., 18:463.

Carraway, K. L., and Koblyka, D.
1970. Comparative study of erythrocyte membranes by gel electro-
phoresis. Biochem. Biophys. Acta, 219:238.

Chien, H. C., and Richardson, T.
1967a. Gross structure of the fat globule membrane of cows milk.
J. Dairy Sci., 595451.

Chien, H. C., and Richardson, T.
1967b. Gross structure of the fat globule membrane of cows milk. 11.
Distribution of amino acids, hexosamine, ribonucleic acid,
copper and iron. J. Dairy Sci., 5931868.

Cook, W. H., and Martin, W. G.
1969. Egg lipoproteins. In Structural and Functional Aspects of
Lipoproteins in Living Systems, (E. Tria and A. M. Scanu, eds.)
Acad. Press, London.

 

 

Dowben, R. M., Brunner, J. R. and
1967. Studies on milk fat globule membranes. Biochim. Biophys.
Acta, 135:1.

Dubois, M., Gillis, K. A., Hamilton, J. R., Rebers, P.A., and Smith, F.
1956. Colormetric method for determination of sugars and related
substances. Anal. Chem., 2§f350.

Fairbanks, G., Stick, T. L., and Wallach, D.F. H.
1971. Electrophoretic analysis of the major polypeptides of the
human erythrocyte membrane. Biochemistry, 1251971.

Glossman, H. and Neville, D. M.
1971. Glycoproteins of cell surfaces. A comparative study of three
different cell surfaces of the rat. J. Biol. Chem., 246:6339.

60

Green, D. E., and Fleischer, S.
1963. The role of lipids in mitochondrial electron transport and
oxidative phosphorylation. Biochim. Biophys. Acta, 195554:

Harwalker, V. R. and Brunner, J. R.
1965. Effects of dissociating agents on physical properties of fat
globule membrane fractions. J. Dairy Sci., £831139.

Hattori, K.
1925. Membrane of the fat globule in milk. J. Pharm. Soc., Japan,
No. 516:123. (Original not seen. Cited by C. T. Herald,
Ph.D. thesis. Michigan State University. 1956).

Hayashi, H., and Roderick, C. A.
1972. The proteins of the outer membrane of beef heart mitochondria.
Biochim. Biophys. Acta, 282:166.

Hayashi, S., Erickson, D. R., and Smith, L. M.
1965. Membranous material of bovine milk fat globules. 11. Some
physical and enzymatic properties of the deoxycholate-
released lipoproteins. Biochemistry, 432557.
Hayashi, S. and Smith, L. M. ,
1965. Membranous material of bovine milk fat globules. I. Comparison
of membranous fractions released by deoxycholate and churning.
Biochemistry, 4g2550.

Heriman, N. D., and Phillips, A. H.
1970. Similarity and limited multiplicity of membrane proteins from
rough and smooth endoplasmic reticulum. Science, 170:1222.

Herald, C. T. and Brunner, J. R.
1957. The fat-globule membrane of normal cows milk. I. The isolation
and characteristics of two membrane—protein fractions. J.
Dairy Sci., 49; 948.

Johansen, P. C., Marshall, R. D. and Neuberger, A.
1960. Carbohydrates in proteins. 2. The hexose, hexosamine,acety1
and amide-nitrogen content of hen's egg albumin. Biochem. J.,
77:239.

Juliano, R. L., and Rothstein, A.
1971. Properties of an erythrocyte membrane lipoprotein fraction.
Biochim. Biophys. Acta, 249:227.

Keenan, T. W., Morre, D. J., Olson, D. E. Yunghans, W. N. and Patten, S.
1970. Biochemical and morphological comparison of plasma membrane
from bovine mammary gland. J. Cell Biol., 44380.

Keenan, T. W., and Huang, C. M.
1972. Membranes of mammary gland. VI. Lipid and protein composition
of golgi apparatus and rough endoplasmic reticulum from bovine
mammary gland. J. Dairy Sci., 5331586.

61

Koblyka, D., and Carraway, K. L.

1972.

Kogen, H.
1972.

Proteins and glycoproteins of the milk fat globule membrane.
Biochim. Biophys. Acta, 288:282.

M., Crombie, G. D., Jordon, R. E., Lewis, W., Franzblau, C.
Proteolysis of elastin-ligand complexes. Stimulation of
elastase digestion of insoluble elastin by sodium dodecyl
sulfate. Biochemistry, 1153412.

Leonard, J.

1970.

Lmny,0.
1951.

Marier, J.
1963.

Protein and glycolipid components of human erythrocyte membranes.
Biochemistry, 9:1129.

H., Rosenbrough, N. J., Farr, L. A., and Randall, R. J.
Protein measurement with the Folin phenol reagent. J. Biol.
Chem., 193:265.

R., Tessier, H., and Rose, D.
Sialic acid as an index of the K—casein of bovine milk. J.
Dairy Sci., 393373:

Monjonnier, T., and Troy, H. C.

1925.

Nichol, L.
1964.

Osborn, M.
1966.

Palmer, L.
1924.

Palmer, L.
1936.

Palmer, L.
1933.

Technical Control of Dairy Products, 2nd ed.
Menjonnier Bros., Inc., Chicago, Ill.

 

W. Bethune, J. L., Kegels, G., and Hess, E. L.
Interacting protein systems. In The Proteins 11, (H. Neurath,
ed.) Acad. Press, New York.

 

J.

Preparation of lipopolysaccharide from mutant strains of
Salmonella in Methods of Enzymology VIII, (S. P. Calowick and
N. 0. Kaplan, eds. ), Acad. Press, New York.

 

 

S. and Samuelsson, E.
The nature of the substances absorbed on the surface of fat
globules in cows milk. Proc. Soc. Expt'l. Biol. Med., 213537.

S. and Tarassuk, N. P.
The effect of the adsorption "membrane" around the fat
globules on curd tension of cow's milk. J. Dairy Sci., 125323.
S., and Wiese, H. F.
Substances adsorbed on the fat globules in cream and their
rela tion to churning. II. The isola tion and identification
of adsorbed substances. J. Dairy Sci., 16341.

Ray, T.. K., and Marinetti, G. V.

1971.

The separation of membrane proteins by polyacrylamide gel
electrophoresis., Biochim. BiOphys. Acta, 233:787.

62

Robinson, J. D. - .
1966. Interaction between protein sulfhydryl groups and lipid
double bonds in biological membranes. Nature, 212:199.

Rodbard, D., and Chrambach, A.
1971. Estimation of molecular aradious, free mobility, and valence
using polyacrylamide gel electrophoresis. Analyt. Biochem.,
40:95.

Scanu, A.
1967. Binding of human high density lipoprotein apoprotein with
aqueous dispersions of phospholipids., J. Biol. Chem., 242:711.

Shapiro, A. L., Vinuela, E., and Maizel, J. V.
1967. The molecular weight estimation of polypeptide chains by
electrophoresis in SDS-polyacrylamide gels. Biochem. Biophys.
Research Comm., 285815.

Storch, V.
1897. On the structure of "fat globules" in cow's milk. (Translated
and communicated by H. Faber). Analyst., 33}197.

Swope, F. C.
1968. Ph.D. Thesis, Michigan State University.

Swope, F. C., and Brunner, J. R.
1968. The fat globule membrane of cow's milk: A reassessment of

isolation procedures and mineral composition., Milchwissenschaft,
23:470.

Swope, F. C. and Brunner, J. R.
1970. Characteristics of the fat globule membrane of cow's milk., J.
Dairy Sci., 535691.

SwOpe, F. C., Rhee, K. C., and Brunner, J. R.
1968. The isolation and partial characterization of a fat globule
membrane protein, Milchwissenschaft, 23f744.

Thompson, M. P.
1960. Ph.D. Thesis, Michigan State University.

Thompson, M. P., and Brunner, J. R.
1959. The carbohydrates of some glycoproteins of bovine milk. J.
Dairy Sci., 423369.

Takayama, K., MacLenna, D. H., Tzagoloff, A., and Stoner, C. P.
1966. Studies on the electron transfer system. LXVII. Polyacrylamide
gel electrophoresis of the mitochondrial electron transfer
complexes. Arch. Biochem. Biophys., 1133223.

63

Trautman, R.
1963. Computational methods of eltracentrifugation. In Ultra—
centrifugal Analysis in Theory and Experiments., (J. W. Williams,
ed.), Acad. Press, New York.

 

Titus, R. W., Sommer, H. H., and Hart, E. B.
1928. The nature of the proteins surrounding the fat globules in
milk. J. Biol. Chem., 26; 237.

Warren, L.
1959. The thiobarbituric acid assay of sialic acids., J. Biol. Chem.,
234:1971.

Weber, K., and Kuter, D. J.
1971. Reversible denaturation of enzymes by sodium dodecyl sulfate.,
J. Biol. Chem., 246:4504.

Weber, K. and Osborn, M.
1969. The reliability of molecular weight determinations by

dodecyl sulfate polyacrylamide gel electrophoresis., J. Biol.
Chem., 244:4406.

Zacharius, R. M., Tatiana, Z. E., Morrison, J. H., and Woodlock, J. J.
1969. Glycoprotein staining following electrophoresis on acrylamide
gels., Analyt. Biochemistry, 39;148.

Tabel Al. Some important chemicals used in this study and their sources.

 

 

Chemical Source
Schiff Reagent . Fisher
Mannose
Acetylacetone

2-mercapotethanol

Urea

SDS

2-thiobarbituric acid Eastman Organic Chemicals
N,N,N',N',-tetramethy1ethy1enediamine (TEMED) '
Acrylamide

N,N'methylenebisacrylamide

Sodium azide

TRIS—hydroxymethyl aminomethane Sigma

(Sigma 121, Trisma base and Sigma 7-9)

Galactosamine Nutritional Biochemicals

N-acetylneuramic acid Calbiochem

Galactose Phanstiehl Laboraiories,
nc.

 

 

64

11111111111111!!! 11117117171“ +

 

27820