? it 2 .
1.; K. x.
_ .1 .2.
. “:17.
ml:

1. .22.).

wt

 

 

 

 

. n...
3 .5

 

.. m...
as. ea.

. as
f...

1.:
Via

L... a
5.... .9 .
11:.

 

 

 

I. .37....7 J.

IE-.. 3...”. .. . . ., . ., ,. :335awaaaﬁmmﬁd

.xs.....§§...s, . , . . . . _

 

 

THESIS.

:70!

This is to certify that the

dissertation entitled
CHEMICAL AND BIOLOGICAL MODIFICATION OFANHYDROUS MILKFAT

presented by
AZIZ CHAFIC AWAD

has been accepted towards fulﬁllment
of the requirements for

Ph.D. Food Science

 

 

degree in

i Jév (94%
/ MaJoVrofesscg

Date /0 “/2 —OO

 

MS U is an Afﬁrmative Action/Equal Opportunity Institution 0-12771

 

UBRARY

Michigan State
University

 

 

 

PLACE iN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before dat

 

e due.
MAY BE RECALLED with earlier due date if requested.
DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

11/00 C'JCIRCJDatoompes-p.“

CHEMICAL AND BIOLOGICAL MODIFICATION
OF ANHYDROUS MILKFAT
By

Aziz Chaﬁc Awad

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

DOCTOR OF PHILOSOPHY

Department of Food Science and Human Nutrition

2000

ABSTRACT
CHEMICAL AND BIOLOGICAL MODIFICATION
OF ANHYDROUS MILKFAT
By

Aziz Chaﬁc Awad

Multiple lines of evidence indicate that saturated fatty acids and cholesterol
raise plasma cholesterol levels. Several methods for improving the nutritional
properties of milkfat have been developed over the past few years. In this study, two
processes were developed to reduce the hypercholesterolemic effect of milkfat. The
ﬁrst process was based on converting the hypercholesterolemic myristic acid to
myristoleic acid using the A9-desaturase system of hen hepatic microsomes.
Desaturation of myristoyl-CoA to myristoleoyl-COA was maximal at pH 7.4, linear
with time up to 10min, and proportional to protein concentrations. The initial velocity
was linear with substrate concentrations between 13 and to 200uM. The desaturation
of myristoyl-CoA was stimulated by bovine serum albumin and reduced by
cytoplasmic proteins. Direct desaturation of free myristic acid by hen liver
microsomal Ag-desaturase without prior activation to myristoyl-CoA by the addition
of adenosine triphospimte (ATP) and CoA was observed when the incubation medium
was mixed at mixing speeds greater than 250 rpm in the presence of fatty acid-

binding proteins (FABP). Direct desaturation was linear with time, proportional to the

microsomal protein concentrations, and maximal at pH 7.9. The greatest direct
desaturation rate was observed at a mixing speed of 500 rpm in the presence of
FABP. As the A9-desaturase is inactive on triacylglycerol substrates, the application
of the desaturation process on milkfat necessitates a lipase- catalyzed hydrolysis of
milkfat triacylglycerols to liberate the ﬁ'ee fatty acids (FF As) prior to their
desaturation, followed by re-esteriﬁcation the FFAs. They would not all return to
their original positions in the triacylglycerol moiety, hence the reaction of the ﬁnal
material to milkfat would be somewhat tenuous.

The second process is based on reducing cholesterol and FFAs in anhydrous
milkfat through fat blending after the cholesterol was removed. The removal of
cholesterol is based on the speciﬁc aﬁinity of B-cyclodextrin for ﬁ'ee cholesterol and
cholesterol esters. The process, optimized by response surface methodology, consists
of a sequence of steps: reﬁning, dilution with distilled water, heating, mixing, and
centrifugation. The process is efﬁcient since the cholesterol was almost completely
(>98%) removed. Free ﬁitty acid reduction was greater than
92.40 %, which was more than 45 % greater than the prior-art processes. Gerbil
feeding studies showed that modiﬁed milkfat (i.e., cholesterol-reduced milkfat)
reduced serum total cholesterol by 28% and triacylglycerols by 47%.

Also, modiﬁed milk fat could be used for whole milk reconstitution andconsequently

the whole spectrum of dairy products such as ice cream, yogurt, and cheese.

To my parents who

taught me the things that matter

iv

ACKNOWLEDGEMENTS

I would like to express my sincere gratitude to my wise and wonderful major
professor Dr. J. Ian Gray who helped support and inspire me in this research project.

He has been an excellent role model and I hope to live up to his standard of intellectual
integrity. He always shares credits and never passes blame.

I would like to extend my appreciation to the other members of my graduate
committee, Drs. Jump, Romsos, Steﬁ'e, and Zeikus, who believed in this project ﬁ'om the
start and provided essential guidance along the way. I would like to acknowledge Dr.
Dale Romsos who offered a clear vision of what this dissertation should be and expertly
helped nurture it. Special thanks to Dr. Robert Templeman for his guidance in performing
the statistical analySes.

I also owe a large debt of gratitude to Dr. John Partridge and my ﬁiend John
Engstrom for their help in the dairy plant.

Most especially, I wish to thank my brothers Elie and Tony for their support,
encouragement, and love, which made my achievement possible.

TABLE OF CONTENTS

LIST OF TABLES ................................................................................. xi

INTRODUCTION ................................................................................... 1
CHAPTER 1

LITERATURE REVIEW .......................................................................... 4

1.1 EFFECT OF DIETARY FATTY ACIDS ON BLOOD CHOLESTEROL . . . . . ......4

1.1.1 Stearic Acid ............................................................................ 5

1.1.2 Palmitic Acid ........................................................................... 7

1.1.3 Myristic Acid .......................................................................... 7

1.1.4 Laurie Acid ............................................................................ 8

1.1.5 Medium Chain Fatty Acids ........................................................... 8

1.1.6 Trans Unsaturated Fatty Acids ................................................. . ..... 9

1.1.7 Polyunsaturated Fatty Acids (W-3,L W-6) ......................................... 9

1.1.7.1 Linoleic Acid (LA) ...................................................... 10

1.1.7.2 W-3 Polyunsaturated Fatty Acids ...................................... 10

1.1.8 Monounsaturated Fatty Acids ...................................................... 11

1.1.9 Cholesterol ............................................................................ 11

1.2 ANIMAL FATS ........ - ........................................................................ 1 2

1.2.1 Methods to Modify Milkfat ........................................................ 14

1.2.1.1 Biological Processes ..................................................... 14

1.2.1.1.] Diet Alteration ................................................ 14

1.2.1.2 Chemical Processes ...................................................... 15

1.2.1.2.] Interesteriﬁcation .............................................. 15

1.2.1.2.2 Complex Formation .......................................... 16

1.2.1.3 Physical Processes ....................................................... 18

1 .2.1.3.1 Short-Path Distillation (SPD) ........................... .... 1 8

1213.2 Melt Crystallization (MC) ................................... l9

1.2.1.3.3 Supercritical Fluid Extraction (SFE) ........................ 20

1.3 ENZYMOLOGY OF DESATURATION .................................................. 21

1.3.1 Biosynthesis of Unsaturated Fatty Acids ....................................... 21

1.3.2 Electron Transport Components ................................................. 23

1.3.3 Role of Membrane Lipid in the Desaturation Reaction .................... .. ..25

1.3.4 Role of Metal Ions in Ag-Desaturation ........................................... 25

1.3.5 Cytoplasmic Proteins and Desaturation ......................................... 26

CHAPTER 2
DESATURATION OF MYRISTOYL-CoA TO MYRISTOLEOYL-COA

BY HEN LIVER MICROSOMAL A’-DESATURASE .................................... 28
2.1 ABSTRACT ............................................................................... 28
2.2 INTRODUCTION ........................................................................ 29
2.3 MATERIALS AND METHODS ........................................................ 31

2.3.1 Materials ............................................................................. 31
2.3.2 Preparation of Hen Liver Microsomes .......................................... 31
2.3.3 Enzyme Assays ................................................................. , ....32
2.3.4 Lipid Extraction and Fatty Acid Identiﬁcation ............................. ....32
2.3.5 Gas Chromatographic (GC) Analysis ........................................... 34
2.3.6 Trypsin Digestion .................................................................. 35
2.3.7 Statistical Analysis ............................................................. ....35

2.4 RESULTS AND DISACUSSIONS .................................................... 36
2.4.1 Factors that Modify The Yield of Desaturation of Myristoyl-CoA. . . . . . ....36
2.4.1.1 Effect of Incubation Time .............................................. 37

2.4.1.2 Effect of Microsomal Protein Concentration ........................ 37

2.4.1.3 Effect of Substrate Concentration ..................................... 40

2.4.1.4 Eﬁ‘ect of Reducing Agent .............................................. 42

2.4.1.5 Eﬁem of pH .............................................................. 44

2.4.1.6 Eﬁ'ect of Competitive Inhibitors ....................................... 44

2.4.1.7 Effect of Bovine Serum Albumin and Cytosolic Proteins .......... 49

2.5 CONCLUSION ............................................................................ 54

CHAPTER 3

DIRECT DESATURATION OF FREE MYRISTIC ACID BY HEN LIVER
MICROSOMAL A’-DESATURASE WITHOUT PRIOR ACTIVATION TO

MYRISTOYL-CoA DERIVATIVE ............................................................ 55
3.1 ABSTRACT ..... ~ ........................................................................ 5 5
3.2 INTRODUCTION ....................................................................... 56
3. 3 MATERIALS AND METHODS ...................................................... 58

3..31 Materials.........................'. ................................................. 58
3. 3. 2 Preparation of Hen Liver Microsomes ......................................... 58
3. 3. 3 Enzyme Assays .................................................................... 59
3.3.4 Lipid Extraction and Fatty Acid Identiﬁcation ............................. ....59
3.3.5 Gas Chromatographic (GC) Analysis ........................................... 61
3.3.6 Trypsin Digestion ................................................................... 62
3.3.7 Statistical Analysis ............................................................. ....62
3.4 RESULTS AND DISACUSSIONS ................................................... 62
3.4.1 Factors That Aﬂect Desaturation of Free Myristic Acid ...................... 63
3.4.1.1 Eﬁ‘ect of Incubation Time .............................................. 63
3.4.1.2 Effect of Microsomal Protein Concentration ........................ 65
3.4.1.3 Effect of pH ................................................................ 67
3.4.1.4 Effect of Temperature ................................................... 67
3.4.1.5 Effect of Reducing Agent .............................................. 67
3.4.1.6 Effect of Bovine Serum Albumin (BSA) ............................. 70

3.4.1.7 Effect of Microsomal Supernatant ...................................... 72

3.4.1.8 Effect of a—Bromohexadecanoate ....................................... 79
3.4.1.9 Effect of Mixing Rate ..................................................... 83
3.4.2 Substrate Speciﬁcity Studies ...................................................... 88
3.4.3 Desaturation of Milkfat ............................................................ 94
3.5 CONCLUSION ............................................................................. 94
CHAPTER 4
REDUCTION OF THE HYPERCHOLESTEROLEMIC EFFECT
OF ANHYDROUS MILKFAT
4.1 ABSTRACT ................................................................................ 97
4.2 INTRODUCTION ......................................................................... 98
4.3 MATERIALS AND METHODS ....................................................... 100
4.3.1 Materials ............................................................................ 100
4.3.2 Preliminary Process ................................................................ 101
4.3.3 Experimental Design ............................................................. 101
4.3.4 Extraction and Determination of Total Cholesterol .......................... 102
4.3.5 Quantitation of Free Fatty Acids (FFAs) ................................... .. ..105
4.3.6 Determination of the Hypercholesterolemic Eﬂ‘ect of Anhydrous
Milkfat .......................................... 105
4.3.7 Statistical Analysis ................................................................ 106
4.4 RESULTS AND DISCUSSION ....................................................... 106
4.4.1 Process Development and Optimization ...................................... 106
4.4.2 Optimization Based on Y; (Percentage Cholesterol Reduction) ........... 108
4.4.2.1 KOH:F FA Mole Ratio ................................................ 108
4.4.2.2 Mixing Time ........................................................... 113
4.4.2.3 B-CyclodextrinzCholesterol Mole Ratio ............................ 119
4.4.2.4 Mixing Rate ............................................................ 119
4.4.3 Optimization Based on Y2 (percent FFA reduction) ........................ 119
4.4.3.1 KOHzFFA Mole Ratio ................................................ 121
4.4.3.2 Calcium Chloride:FFA Mole Ratio .................................. 121
4.4.3.3 Mixing Rate ............................................................ 122
4.4.4 Process Evaluation .............................................................. 123
4.4.5 Effect of Anhydrous Milkfat on Blood Lipid levels ......................... 123
4.5 CONCLUSION ......................................................................... 127
SUMMARY AND CONCLUSIONS .......................................................... 130
FUTURE RESEARCH ........................................................................... 133
REFERENCES ..................................................................................... 134

LIST OF FIGURES
Chapter 2:

Figure 1 Process for converting myristoyl-CoA to myristoleoyl-CoA using the

A9-desaturase system of hen liver microsomes ..................................... 33

Figure 2 The desaturation of myristoyl-CoA as a ﬁmction of time by the

Ag-desaturase system of hen liver microsomes .................................... 38

Figure 3 The desaturation of myristoyl-CoA as a function of microsomal protein
concentration by the Ag-desatttrase system of hen liver microsomes..

Figure 4 The relationship of initial velocity to substrate concentration for the hen
microsomal A9-desaturase catalysed reaction ...................................

Figure 5 Eﬁ‘ect of increasing concentrations of reduced pyridine nucleotide
(NADH) on the desaturation of myristoyl CoA by the A9-desaturase

system in hen liver microsomes ...................................................

Figure 6 Effect of pH on the desaturation of myristoyl-CoA by the Ag-desaturase

system of hen liver microsomes ...................................................

Figure 7 Effect of competitive inhibitors on the desaturation of myristoyl-CoA

by the A9-desaturase system of hen liver microsomes ..........................

Figure 8 Effect of palmitoy1-, stearoyl-, and oleoyl-CoA, and free CoA on the

activity of the A9-desaturase of hen liver microsomes ..........................

Figure 9 Eﬁ‘ect of bovine serum albumin (BSA) and microsonml supernatant
OMS) on the desaturation of myristoyl-CoA under the experimental

conditions described in Figure 1 ..................................................

Figure 10 Design of study to establish the effect of heat treatment and trypsin
digestion of the microsomal supernatant containing the cytosolic

proteins (CP) on the desaturation reaction .......................................

Figure 11 Eﬂ‘ect of heat and trypsin digestion of the microsomal supernatant
on the desaturation of myristoyl-CoA using the A9-desaturase system

of hen hepatic microsomes ..........................................................

Chapter 3:

Figure 1 Preliminary assay for the direct desaturation of ﬂee myristic acid to
myristoleic acid in the absence of CoA and ATP, by the A9-desaturase

....40

...42

...46

...47

...48

....51

....52

...53

system of hen liver microsomes ...................................................... 61

Figure 2 Direct desaturation of ﬁee myristic acid as a function of time by the A9-

desaturase system of hen hepatic microsomes ...................................

Figure 3 Direct desaturation of free myristic acid as a ﬁmction of microsomal
protein concentration by the A9-desaturase system of hen liver

microsomes .........................................................................

Figure 4 Effect of pH on the direct desaturation of free myristic acid by the A9-

.. 66

...68

desaturase system of hen liver microsomes ........................................ 69

Figure 5
Figure6

Figure7

Figure8

Figure9

Figure 10

Figurell

Figure 12

Figure 13

Figm'e 14

Figure 15

Figure 16

Figure 17

Figure 18

Chapter 4:

Figure 1

Figure2

Effect of temperature on the direct desaturation of ﬁ-ee myristic
acid by the A9-desaturase system of hen liver microsomes ..................... 71

Eﬂ‘ect of pyridine nucleotide on the direct desaturation of tree

myristic acid by the Ag-desaturase system of hen liver micro somes ............ 73

Assay for the eﬂ‘ect of bovine serum albumin and microsomal
supernatant (MS) on the direct desaturation of ﬁ'ee myristic acid by the

A9-desaturase system of hen liver microsomes .................................... 74
Eﬂ‘ect of bovine serum albumin and microsomal supernatant on the
direct desaturation of myristic acid by the A9-desaturase system

of hen liver microsomes ................................................................ 75
Qualitative scheme for the movement of C14:0—protein complexes
in the incubation medium .............................................................. 77

Heat treatment and trypsin digestion of the microsomal supernatant

used in the direct desaturation of free C14:0 by the A9-desatttrase system

of hen hepatic microsomes ........................................................... 78
Effect of heat treatment and trypsin digestion on the microsomal

supernatant on the direct desaturation of free C14:0 by the A9-desaturase
system of hen hepatic microsomes .................................................. 79
Assay for the effect of a-bromohexadecanoate on the direct

desaturation of free myristic acid by the Ag-desaturase system of hen

hepatic microsomes .................................................................. 81
Effect of a-bromohexadecanoate: substrate mole ratios on the direct
desaturation of myristic acid by the A9-desaturase system of hen hepatic
microsomes ........................................................................... 82
Final assay for the direct desaturation of myristic aciszABP

complexes by the A9-desaturase system of hen liver microsomes in the
absence of CoA and ATP ............................................................. 86
Effect of mixing rate on the direct desaturation of myristic

aciszABP complexes by the A9-desaturase system of hen hepatic

microsomes ............................................................................ 87
Effect of substrate concentration on the direct desaturation myristic
aciszABP complexes by the A9-desaturase system of hen liver

microsomes ............................................................................. 88
An enzyme-substrate complex illustrating both the geometrical

and the physical complementarity btween active site (E1) and

reaction side (81) ...................................................................... 91
Effect of different substrates on the speciﬁc activity of the hen
microsomal A9-desaturase system ................................................... 97

Response surface methodology for percent reduction of

cholesterol (Y1) in anhydrous milkfat at a mixing rate of 1000 rpm

and a B-cyclodextrinzcholesterol mole ratio of 5.58 ........................... 115
Response surface methodology for percent reduction of cholesterol

(Y1) in anhydrous milkfat using a mixing time of 10 min and a

Figure 3
Figure 4

Figure 5
Figure 6

Figurc7
Figure8

Figure 9

B-cyclodextrinzcholesterol mole ratio of 5.58 .................................... 116
Response surface methodology for percent reduction of cholesterol

(Y1) in anhydrous milkfat using a mixing time of10 min and a

mixingrate of1000 rpm ............................................................ 117
Response surface methodology for percent reduction of vcholesterol

(Y1) in anhydrous milkfat using a KOHzFFA mole ratio of 1 and

amixingrate of1000 rpm. .......................................................... 119
Response surface methodology for percent reduction of cholesterol

(Y1) in anhydrous milkfat using a KOHzFFA mole ratio ofl and a
B-cyclodextrin:cholesterol mole ratio of 5.58 ............... . ..................... 120
Response surface methodology for percent reduction of cholesterol (Y1)

in anhydrous milkfat using a KOHzFFA mole ratio of 1 and a mixing

time of 10 min ........................................................................ 121
Response surface methodology for FFA reduction (Y2) in anhydrous

milkfat .................................................................................. 124
Optimized process to remove cholesterol and FFA in

anhydrous milkfat .................................................................... 127
Blood lipid response of gerbils to modiﬁed milkfat .............................. 129

Table 1

Table 2

Table 3

Table 4

Table 5

Table 6

Table 7

LIST OF TABLES

Variables and their levels for central composite design used in

optimizing the process for reducing cholesterol and free fatty acids

in anhydrous milkfat ............................................................... .104
Central composite design arrangement and responses used in

optimizing the process for reducing cholesterol and free fatty acids in
anhydrous milkfat ................................................................... 106
Parameter estimates and analysis of variance of the second order

polynomial regression for ﬁve factors and two responses

(cholesterol (Y1) and FFA (Y2) reduction) ...................................... 109
Parameter estimates and analysis of variance of the second order

polynomial regression for four factors and one response (cholesterol
reduction (Y1)) ....................................................................... 110
Parameter estimates and analysis of variance of the second order

polynomial regression for 3 factors and one response

(FFA reduction (Y2)) .................................................................. 1 1 1
Cholesterol reduction in anhydrous milkfat as predicted by the

second order polynomial regression for 4 factors ................................. 112
Reduction of ﬁee fatty acids as predicted by the second order polynomial
regression for 3 factors ............................................................... 1 13

LIST OF APPENDICES

Appendix 1 Mass spectra of authentic standard and sample cis-9-Cl4zl .......... 135

INTRODUCTION

Cardiovascular disease kills almost one million Americans annually. Another two
million die from strokes. Forty million suﬂ‘er ﬁ'om some forms of cardiovascular disease.
Atherosclerosis aﬁects 95 percent of all Americans over the age of 50. According to
several studies during the last 20 years, scientists and doctors have pieced together much
of the puzzle surrounding heart diseases. In 1994, an expert panel of clinicians and
researchers convened by the National Heart, Lung and Blood Institute of the American
National Institutes of Health sent out a clear message: high blood cholesterol is bad, and

lowering it can reduce the risk of heart disease (Expert Panel, 1994).

Dietary management is the initial step in lowering blood cholesterol. Multiple
lines of evidence indicate that dietary satmated fatty acids and cholesterol increase blood
cholesterol and therefore increase the risk of heart disease. Better food habits can help
reduce one of the major risk factors for heart attack — high blood cholesterol. The eating
plan of the American Heart Association (AHA) and the National Cholesterol Education
Program (NCEP, 1993) describes the latest advice of medical and nutrition experts. The
best way to help lower blood cholesterol levels is to eat less saturated fat, cholesterol,
total fat, and maintain a healthy body weight. The composition of this plan consists of

Step I and Step II diets.

To achieve the recommendations of Step I diet, the saturated ﬁt and cholesterol
contents of a typical American diet must be reduced by approximately 25%. A Step II
diet requires greater decreases in saturated ﬁt and cholesterol: Both must be decreased by
almost 50%. For both diets, total ﬁt should provide 5 30% of calories, representing a

reduction of approximately 15%.

Recognizing these concerns, the ﬁt and oil industry is looking at new techniques
to alter the composition of ﬁts. The biggest challenge is to reconcile ﬁmctional needs
with nutritional concerns. With ﬁts and oils, one cannot make a choice based on ﬁtty
acid composition alone. Taste is a very important ﬁctor.

The ﬁ'yingandthe baking industrymayhavejumped fromthe “the ﬁyingpaninto
the ﬁre” by switching to partially hydrogenated vegetable oils (PHVO) to maintain the
textural and stability qualities of the saturated animal ﬁts that they have replaced in
industrial applications. Trans ﬁtty acids generated during the processing of PHVO now
appear as detrimental, if not worse, than saturated ﬁtty acids (Ascherio et al., 1994).

The foremost element Of this study is to produce a healthy natural animal ﬁt that
has a desirable ﬂavor and functionality. This study addresses alternatives to the

hypercholesterolemic anirml ﬁts and trans-rich PHVO by:

1- a) Developing an enzymatic process to convert the hypercholesterolemic myristic

acid to myristoleic acid using the Ag-desattrrase system of hen hepatic
microsomes;
b) Elucidating at the biochemical level the underlying mechanism of the

desaturation reaction vis-a-vis substrate speciﬁcity and the interaction between

substrate and ﬁtty acid binding proteins and bovine serum albumin.

The primary aim of this part is to test the following hypothesis:

When a ﬂu fatty acid is bound to fatty acid-binding proteins (FABPs) in the
form of protein-monomer complexes, its activation to the CoA derivative is not .
necessary in order to be desaturated by the Ag-desaturase system when using high

mixing rates. FABP-fatty acid complexes are substrates for the A9- desaturase

in the presence of high mixing rates.

2- Reducing the hypercholesterolemic effect of animal ﬁts by lowering their

cholesterol contents by chemical means.

This study will provide basic knowledge for acceptable, cost- eﬂective improvement of

food lipids, particularly animal ﬁts, to ensure the health of consumers and to ﬁnd new

uses of animal ﬁts in food production

CHAPTER 1
LITERATURE REVIEW

1.1 EFFECTS OF DIETARY FATTY ACIDS ON BLOOD CHOLESTEROL
Multiple lines of evidence indicate that saturated fatty acids raise plasma
cholesterol levels. Both animal (Hegsted and Gallagher, 1967) and human studies (Keys
et aL, 195 7) have demonstrated elevation in serum cholesterol levels following increased

saturated ﬁt intake. These observations, coupled with epidemiological evidence of
increased risk of coronary heart disease (CHD) in populations consuming excessive
amounts of saturated ﬁts, explain the emphasis on the reduction of saturated ﬁt by most
health agencies.

Generally, saturated ﬁtty acids increase plasma low density lipoproteins (LDLs)
which are very atherogenic, partly by reducing receptor-mediated uptake (Grundy and
Vega, 1988a). The high density lipoproteins (I-IDLs) provide protection against
atherosclerosis by transporting cholesterol ﬁom tissue to liver for conversion to bile acids
and excretion. Although saturated ﬁtty acids maintain relatively high levels of HDL
(Grundy and Vega, 1988b), it cannot be said that they have unique HDL-raising activity.

Dietary saturated ﬁtty acids include caprylic, caproic, lauric, myristic, palrnitic,
and stearic acids. A key question is whether all saturated ﬁtty acids raise the LDL-
cholesterol level. If not, then it is inappropriate to consider all saturated ﬁtty acids

together when making dietary recommendations. Multiple evidence suggests that all

saturated ﬁtty acids do not have the same effect on LDL-cholesterol concentrations. The
focus has shifted to understanding the role that individual dietary ﬁtty acids play in the
health of the consumer. Thus, to examine the actions of saturated ﬁtty acids on LDL-
cholesterol concentrations, it is necessary to consider each fatty acid separately. The
satruated ﬁtty acids, lauric and palrnitic acids increase plasma cholesterol and
triacyglycerol levels (Denkee and Grundy, 1992), whereas polyunsaturated ﬁtty acids
(PUFAs) have the opposite effect (Mensink and Katan, 1992). Myristic acid is the most

hypercholesterolemic long-chain saturated ﬁtty acid, followed by lauric acid

(Hayes, 1995).

1.1.1 Stearic Acid

It has been suggested that stearic acid does not raise the serum cholesterol level.
Bonanom and Grundy (1988) produced a synthetic fat with approximately 45% stearic
acid by chemical randomization between a completely saturated soybean oil and a high
oleic-saﬂlower oil. Oleic acid was the other predominant ﬁtty acid (50%), while the
palrnitic acid content was approximately 5%. This synthetic ﬁt was compared with palm
oil which has about 45% palrnitic acid and with high oleic saﬁlower oil which is mainly
oleic acid. The study showed that stearic acid did not raise total cholesterol (or LDL-
cholesterol) compared to oleic acid, while palmitic acid markedly increased cholesterol
levels. No other lipoproteins were changed by stearic acid.

The inﬂuence of butterﬁt, beef ﬁt, cocoa butter, and olive oil on serum
cholesterol and lipoproteins concentrations was studied by Denke and Grundy (1991 ).
Olive oil served as a baseline against which the three highly saturated ﬁts were

compared. Of particular interest to this study was whether beef ﬁt (tallow) and cocoa

butter which are rich in stearic acid, are as hypercholesterolemic as butterfat. The study
was carried out in 10 middle-aged men under metabolic ward conditions. Liquid formula
diets containing 40% of the calories ﬁ'om the test ﬁt were fed in random order for three
weeks each, with lipoprotein concentrations measured on the last ﬁve days. Butterﬁt
raised LDL concentrations the most (4.23 :1: 0.15 mmol/L; x :t SE). Beef tallow produced
lower concentrations of LDL (4.03 :i: 0.18 mmol/L), while cocoa butter resulted in even
lower concentrations (3.82 i 0.15 mmol/L). The lowest concentrations of LDL were
observed with olive oil (3.62 :t 0.18 mmol/L). Because intakes of total cholesterol were
balanced during this comparison, the different effects of these three ﬁts (butterﬁt, cocoa
butter, and beef tallow) on plasma cholesterol concentrations undoubtedly were related to
differences in their ﬁtty acid compositions. Cocoa butter was less hypercholesterolemic
than butterﬁt and beef tallow because it contains a higher level of stearic acid and
relatively low levels of cholesterol-raising saturated ﬁtty acids such as myristic acid.

This study provides additional support for the concept that stearic acid does not raise
serum total cholesterol.

Why does stearic acid not raise the cholesterol level? There are at least two
propositions that have been developed. First, the absorption of stearic acid might be
incomplete as suggested by Apgar et a1. (1987) or, second, stearic acid might be rapidly
converted to oleic acid in the body (Elovson, 1965). The ﬁrst proposition was not
supported by Bonanome and Grundy (1989) who compared the amounts of ﬁtty acids
excreted in the bum stool with the quantity ingested daily. Fecal excretion data

indicated that oleic acid was 99% absorbed, palrnitic acid was 96-97% absorbed, and

stearic acid was 90-94% absorbed. Therefore, the ﬁct that stearic acid does not raise
cholesterol levels cannot be attributed to a lack of absorption

The second proposition is that stearic acid is rapidly converted to oleic acid in the body
(Elovson, 1965). However, these investigators did not present any evidence for the
accumulation of stearic acid in plasrm lipids on feeding a high stearate diet, even though
large quantities of stearic acid were absorbed. Instead, the oleic acid content of the

plasma lipids was increased, thus suggesting the conversion of stearic acid to oleic acid.

1.1.2 Palmitic Acid

Palmitic acid is the predominant saturated acid in beef tallow, lard, butterﬁt, and
even in certain vegetable ﬁts and oils, e.g., palm oil and cottonseed oil. According to
Keys et al. (1965), palmitic acid raises the serum total cholesterol by 2.7 mg/dl for every
1% in dietary calories. Studies with monkeys showed that palmitic acid is
hypercholesterolemic compared to dietary oleic and linoleic acids (Pramod and Hayes,
1992). More recently, Kho'sla and Hayes (1993) and Pronczuck et al. (1994) reported that
excessive intake of dietary cholesterol exerts a synergistic effect on the metabolism of
palrnitic acid-rich ﬁts, causing them to be hypercholesterolemic. In the absence of
dietary cholesterol and in individuals with normal lipoprotein proﬁles, palrnitic acid does
not ordinarily raise plasma total cholesterol concentrations or LDLs (Sundram et al.,

1994,1995; Hayes, 1995).

1.1.3 Myristic Acid

Myristic acid is the most potent hypercholesterolemic long chain fatty acid,

followed by palrnitic and lauric acid (Hayes, 1995). look et al. (1994) compared myristic
acid with pahnitic acid and oleic acid (all at 10% energy) and found that it signiﬁcantly
increased both total and LDL-cholesterol concentrations. Myristic acid also demonstrated
a much greater cholesterol-raising effect than palmitic acid when substituted for oleic

acid.

1.1.4 Laurie Acid

The cholesterolemic effects of lauric acid were evaluated in a study involving 14
subjects (Denke and Grundy, 1992). Compared with palrnitic acid, lauric acid lowered
total and LDL-cholesterol concentrations by 4 and 6 %, respectively. In contrast, when
compared with oleic acid, lauric acid was hypercholesterolemic, raising both total and

LDL-cholesterol concentrations by 11-12%.

1.1.5 Medium Chain Fatty Acids

Medium chain ﬁtty acids such as caprylic acid and caproic acid are present in
moderate amounts in butterﬁt. Currently, it is believed that they do not raise serum
cholesterol levels, but apparently they can increase triacylglycerol concentrations when
fed in large amounts (Larry et al. 1990). Nonetheless, the medium chain ﬁtty acids which
act like carbohydrates do have the potential as replacements of long chain, cholesterol-

raising ﬁtty acids in the diet (Kennedy, 1991)

1.1.6 Trans Unsaturated Fatty Acids

When vegetable oils undergo hydrogenation, a portion of the linoleic acid is
completely saturated to produce stearic acid. Another portion is converted to mono
or di-unsaturated ﬁtty acids, cis and trans. Since double bonds can migrate during
hydrogenation, several different cis and trans isomers are produced (cis-9, trans-13;
trans-9, trans-12; trans-8, cis 12; trans-8, cis-13; or mono unsaturated trans-9 elaidic
acid) (Ratnayke and Pelletier, 1992). Trans isomers have been implicated in increasing
LDL-cholesterol levels. lock and Katan (1992) studied the effects of elaidic
(trans C18:1 (n-9)) on plasma total cholesterol. Compared to linoleic acid, they observed
an increase of LDL-cholesterol of 9.3 mg/dl and a decrease of HDL-cholesterol by 2.3
mg/dl No signiﬁcant difference in serum triacylglycerol content was observed between
the trans diet and the linoleate diet. They also observed a linear dose response
relationship, but indicated that the metabolic effects of trans ﬁtty acid as function of
quantity ingested were not fully understood.

Margarine, the major source of trans ﬁtty acids in the US diet, has increasingly
replaced butter over the last 20 years (USDA, 1997). Ascherio et al. (1994) reported that
30,000 coronary deaths per year in the United States could be caused by consumption of
hydrogenated ﬁts and called for the elimination or a major reduction of these ﬁts in

foods.

1.]. 7 Polyunsaturated Fatty Acids (W-3, W-6)
The major W-3 and W-6 ﬁtty acids in foods are alpha-linolenic and linoleic acids,

respectively. Humans can obtain W-3 and W-6 ﬁtty acids only by ingesting them. They

are called essential ﬁtty acids, because when a htunan cell builds a ﬁtty acid, it cannot

create double bonds in these sites.

1.1. 7.1 Linoleic Acid (LA)

The plasma cholesterol-lowering effect of LA is one of the most consistent
ﬁndings in nutrition. Studies carried out under strict metabolic ward conditions
consistently show that the LDL-cholesterol concentrations are low when diets
supplemented with LA are consumed. Iacono and Dougherty (1991) determined the
effect of LA on plasma lipoproteins in a study carried out in 11 middle aged, healthy
male subjects for two 40 day periods. The diets were designed to maintain saturated and
monounsaturated ﬁtty acids at 10 %, and LA at 3.8 and 10.8 %. LDL-cholesterol
concentrations decreased 18% with the lower LA diet, and 22% with the higher LA diet.
HDL-cholesterol concentrations were decreased by 2.3% on the lower LA diet and by
12.5% on the higher LA diet. The mechanism by which LA lowers the cholesterol
contents of plasma LDL and HDL is not known. There has been concern expressed about
the HDL- lowering effect of LA because HDL-cholesterol concentrations are inversely
related to the risk of CHD (Mahley, 1985). Mensink and Katan (1989) showed that
reasonable levels of intake of W-6 polyunsaturated ﬁtty acids do not signiﬁcantly lower

HDL cholesterol concentrations.

1.1. 7.2 W-3 Polyunsaturated Fatty Acids

The W—3 ﬁtty acids are especially concentrated in some ﬁsh oils, canola, and

walnut oil (USDA, 1991). It has been shown tlmt ﬁsh oils may be equally or more

10

hypolipidemic than linoleic acid (Kromhout et. al. 1985). Recent epidemiological
observations of low plasma lipids in populations consuming seafood reﬂect the
effectiveness of dietary ﬁsh oils in reducing plasma lipids, especially triacylglycerols and
their association with a decreased incidence of atherosclerosis, thrombosis, and coronary

inﬁrctions (Leaf and Webber, 1988).

1.1.8 Monounsaturated Fatty Acids

Oleic acid is the major monounsaturated ﬁtty acid in foods. It is especially
concentrated in canola oil and olive oil. Because of the concern about the HDL-lowering
effect of linoleic acid, researchers re-examined the potential for monounsaturated fatty
acid, especially oleic acid, as a replacement for saturated ﬁtty acids. Mattson and Grundy
(1985) compared oleic acid and linoleic acid as replacements for saturated ﬁtty acids.
Diets rich in the three types of ﬁtty acids were fed in randomized order to 12 men. Oleic
acid was as effective as linoleic acid in lowering LDL cholesterol, but without having the
l-IDL- lowering effects of linoleic acid.

Furthermore, on the basis of epidemiological and metabolic ward studies, there
is increasing evidence that olive oil may have a useﬁrl role in the prevention of CHD
(Willet, 1994). This effect also has been observed with other monounsaturated ﬁtty
acids, so oils with high levels of monounsaturated ﬁtty acids will be ﬁvored by

COHSUIDCI‘S.

1.1.9 Cholesterol
Although the consumption of cholesterol does not seem to be a major ﬁctor

in CHD, Khosla and Hayes (1993) and Pronczuck et al. (1994) reported that excessive

ll

intake of dietary cholesterol exerts a synergistic effect on the metabolism of palrnitic
acid-rich ﬁts, causing them to be hypercholesterolemic. In the absence of dietary
cholesterol and in individuals with normal lipoprotein proﬁles, palmitic acid does not
ordinarly raise total plasrm cholesterol concentrations or LDL (Sundram et al., 1994,
1995; Hayes, 1995). Epidemiological studies further suggested that dietary cholesterol
increases the risk of coronary heart disease beyond its serum cholesterol-raising effect
(Stamler and Shekelle, 1988). Mechanisms for this latter effect are unknown.

Dietary cholesterol is prone to oxidation Increased attention has focused on
speciﬁc eﬂ‘ects associated with certain cholesterol oxide products and the need to
eliminate or prevent their formation in the human diet. Peng et al. (1979) reported that
oxides of cholesterol are toxic and cause degeneration of aortic smooth muscle cells in
tissue culture, and may lead to the development of atherosclerosis (Peng et al., 1991).

The overall conclusions from both experimental and epidemiological studies are
consistent in showing that dietary saturated ﬁtty acids elevate serum cholesterol
concentrations. Three saturated ﬁtty acids - palmitic acid, lauric acid, and myristic acid,
undoubtedly increase LDL cholesterol concentrations. Thus, when dietary
recommendations are made, the primary attention should be given to removing these

three fatty acids from the diet.

1.2 ANIMAL FATS

Animal ﬁts are stable to oxidation and possess rmiquely pleasing ﬂavors not found
in other ﬁts. Milkﬁt has received the most attention because of its commercial
importance. It confers distinctive properties on dairy products that aﬂect processing.

Milkﬁt is a good source of essential ﬁtty acids and it contains a high proportion of short

12

chain ﬁtty acids which contributes to its ease of digestibility (Kennedy, 1991).
Moreover, milkﬁt contains conjugated linoleic acids (CLAs) which are recognized for
their potential ability to inhibit cancer (Clement et al., 1994). CLAs are unusual because
they are abundant in products from ruminant animals. They are formed by
biohydrogenation of PUFAs in the rumen of cows and subsequently ﬁnd their way into
milk (Gmr, 1994).
One epidemiological study compared dietary habits in rural Finland which has

one-quarter the incidences of colon cancer with those in urban Copenhagen, Denmark
(Maclennan et al., 1978). The community with the low incidence of colon cancer
consumed more potatoes and whole milk than the higher incidence group which
consumed less white bread and meat.

Of the natural ﬁts, milkﬁt is the most varied in its chemical characteristics and
functional properties. The melting point of milkﬁt increases with increasing saturation
and chain-length of its constituent ﬁtty acids (Walstra and Jenness, 1984). The melting
point of milkﬁt is also aﬁ‘e‘cted by the positioning of the ﬁtty acids on the glycerol

molecule (Walstra et al., 1994). In its native form, milkﬁt does not suit various food

formulations. For example, the wide melting range of milkﬁt, -40 to 40°C (Walstra et
al., 1994) makes it diﬁicult to produce spreadable butter at reﬁigeration temperature.
Consequently, new ﬁelds of application remain constrained due to the limited
functionality (pourability and spreadability) of milkﬁt. An optimum ﬁt cannot be always
obtained from nature. Animal ﬁts, when viewed in their native state, have limited use.
But they can become an economic asset when viewed as raw materials to produce

modiﬁed ﬁts with desirable characteristics.

13

1.2.1 Methods to Modify Milkfat

The ﬁt and oil industry is looking continuously at new techniques to alter the fat
molecules. The biggest challenge is to reconcile the functional needs with the nutritional
concerns. Several methods for improving the nutritional and physical properties of
milkfat have been developed over the past few years. These approaches can be
conveniently divided into biological (diet alteration), chemical (interesteriﬁcation), or

physical processes (distillation, crystallization, supercritical ﬂuid extraction).

1.2.1.1 Biological Processes
I. 2. I. 1.1 Diet Alteration

Changing the ﬁtty composition of milkﬁt by altering the diet of cows is an old
concept which has received some recent attention (Banks et al., 1990). The inclusion of
ﬁts and oils in the diet of ruminants may exert the following effects, thereby inﬂuencing,
either directly or indirectly, lipogenic pathways leading to milkﬁt secretion (Faulkner
and Pollock, 1989; Banks 'et al., 1990).

1. The synthesis de novo of fatty acids such as palmitic acid and stearic acid by

microorganims in the rumen may be reduced, leading to reductions in blood

lipids, mammary gland uptake, and yield of long-chain ﬁtty acids in the milk.
2. The amounts of long-chain ﬁtty acids released in the rumen by lipolysis may

be increased, leading to increases in the blood lipids, mammary gland uptake

and yield of long-chain ﬁtty acids in milk.

14

3. The intramarmnary synthesis of ﬁtty acids may be reduced, thus decreasing the

yields of short -and medium -chain ﬁtty acids in milk.

These effects appear to be inﬂuenced by ﬁctors such as:(a) stage of lactation; (b)
roughage-to-concentrate ratio and the ﬁt content of the basal diet; and (c) amount,
composition and physical form of ﬁt added to the basal diet (Grurmner, 1991).

A major technological breakthrough was the development of protected ﬁt supplements
obtained by crosslinking the ﬁt with protein and treating with formaldehyde. The
feeding of protected lipids was introduced as a means of producing ﬁt with elevated
linoleic acid levels. Feeding protected soybean oil to cows resulted in a 3.6 fold increase
in linoleic acid (Banks et al., 1990) when compared to the basal diet. The degree of
success in altering the ﬁtty acid composition of milkfat through the feeding of protected

lipid supplements has been inconsistent among different laboratories (Grummer, 1991 ).

1.2.1.2 Chemical ProceSses
l. 2. 1.2.1 Interesteriﬁcation

lnteresteriﬁcation of milkﬁt alters the distribution of ﬁtty acids in the
triacylglycerol molecules and thus changes its physical properties such as melting

behavior, crystallimtion, and plasticity. lnteresteriﬁcation can be accomplished by

heating the ﬁt or a mixture of ﬁts and oils at a relatively high temperature (~ 200°C) for
a long period of time. However, catalysts are commonly used that allow the reaction to
be completed in a shorter time at lower temperatures. Alkali metals and alkali metal

alkylates are eﬁ‘ective low-temperature catalysts, sodium methoxide being the most

15

popular (Nawar, 1985).

Interesteriﬁcation has not yet been applied to milkﬁt as its feasibility is restricted
by the ﬁct that it often requires two techniques that are deleterious to milkﬁt ﬂavor,
reﬁning and deodorization, to remove the free ﬁtty acids (0.2 to 0.4 %). These acids
consume the catalyst or inactivate the active catalyst once it is formed (Sreenivasan, ‘
1978).

Recently, the use of lipases (triacylglycerol acylhydrolase, EC 3.1.1.3) to catalyze
the interesteriﬁcation reactions has received considerable attention. Triacylglycerol
lipases can be used in several ways to modify triacylglycerols. In an aqueous medium
with an emulsiﬁed triacylglycerol substrate, hydrolysis is the dominant reaction, but in
organic media (heptane and hexane) esteriﬁcation and interesteriﬁcation become
predominant (Bloomer, 1990). Enzymatic catalysts oﬁ‘er certain advantages over
chemical catalysts including a controllable positional speciﬁcity and the ability to run the

reaction at lower temperatures (Chang et al., 1990)

1.2. 1.2.2 Complex Formation

This technique is used to reduce cholesterol concentrations in milk and dairy
products by complexing cholesterol and its esters with a complexing agent such as
B—cyclodextrin. Cyclodextrins are cyclic oligosaccharides obtained by enzymatic
degradation of starch. They consist of six, seven, or eight glucose monomers arranged in
a donut shaped ring. These compounds are denoted alpha, beta or garmna cyclodextrin,
respectively. B-Cyclodextrin is not hygroscopic and contains 13.6% moisture at 30°C

and 86% relative humidity (Szejtli, 1982a). Cyclodextrins are water-soluble because of

16

the location of free hydroxyl groups on the external rim of the molecule (Szejti, 1982a).
Solubility is a function of temperature, the higher the temperature the higher the
solubility. The solubility of B-cyclodextrin increases from 0.8% at 0.5 °C to 39.7% at
90 °C. The internal cavity which is hydrophobic allows the cyclodextrins to complex
molecules such as aromatic alcohols, ﬁtty acids and their esters, and cholesterol.
Churcntly, cyclodextrins are used (1) to control volatility of agricultural compounds
designed to control pathogens, insects, and weeds; (2) to stabilize pharmaceutical
products (drugs, vitamins), ﬁagrances, and skin care lotions through encapsulation; and
(3) to enhance color, odor, and ﬂavor stability in beverages and processed foods (Szejtli,
1982b). The most important parameters that determine whether a given molecule can
form complexes with B—cyclodextrin are its hydrophobicity and its relative size and
geometry in relation to the cyclodextrin cavity (Szejtli, 1982b). When dissolved in water,
the cyclodextrin molecules are able to accommodate smaller guest molecules, or
ﬁrnctional groups of molecules less hydrophilic than water in their internal cavities
(Szejtli, 1982b). In aqueous solution, the slightly apolar cyclodextrin cavity is occupied
by water molecules, an energetically unﬁvored process (polar-apolar interaction). These
water molecules are therefore readily substituted by appropriate guest molecules such as
cholesterol which are less polar than water (Szjetli, 1982b).

Several investigators have used B—cyclodextrin to reduce the cholesterol levels in
animal products. Courregelongue et al. (1989) used 10 % B-cyclodextrin to remove 41 %
of the sterols ﬁom anhydrous milkﬁt. Bayol et al. (1989) reported the removal of 80 %
Al-cholesten-3-one from anhydrous milkﬁt, while Roderbourg et al. (1993) demonstrated

that at least 37% of the cholesterol could be removed from anhydrous milkﬁt in one

17

single step. Graille et al. (1991) reported the simultaneous removal of 50% cholesterol
and 52 % FFAs ﬁom cream with B-cyclodextrin. In Australia, Okenﬁrll et al. (1991)
invented a new process called SIDOAK to reduce cholesterol in dairy products. The
process consisted of adding 6- cyclodextrin to milk and mixing below 10 °C. The
insoluble complexes of cholesterol and [3- cyclodextrin were removed by centrifugation.
The maximum cholesterol reduction obtained was 80-90 %. Yen and Tsai (1995), using
10% [3- cyclodextrin, indicated the removal of 95 % cholesterol simultaneously with 50
% FFAs ﬁom lard. Awad et al. (1996) used B- cyclodextrin to reduce the cholesterol
level in egg yolk by 96%. The process consisted of mixing the diluted yolk with [3-
cyclodextrin at 50 °C for 10 min. The B— cyclodextrin -cholesterol complexes were

removed ﬁom the medium by centrifugation.

1.2.1.3 Physical Processes
Milkﬁt ﬁ‘actionation using physical processes such as short-path distillation, melt
crystallization, and supercritical ﬂuid extraction has been applied in conjunction with

cholesterol extraction.

1.2.1.3. I Short-Path Distillation (SPD)

SPD is a relatively well known process and is based on the evaporation of
molecules into substantially gas-free space, i.e., vacuum. The controlling ﬁctor is the
rate at which the molecules escape from the heated surﬁce of the distilling liquid and are
received by the cooled condenser surﬁce. Arul et al. (1988) ﬁactionated milkﬁt by SPD

into four ﬁactions at temperatures of 245 and 265 °C and pressures of 220 and 100 um

18

Hg. Two ﬁactions were liquid, one was semi-solid, and one was solid at room
temperature. The solid ﬁ'action contained cholesterol at a concentration of 0.2 mg/g ﬁt
compared to 2.6 mg/g ﬁt in native milkﬁt or to 16.6 mg/g ﬁt in the liquid ﬁaction. SPD
of milkﬁt offers an opportunity to obtain fractions with different chemical and physical
properties. However, there are some major drawbacks to this technique: (1) the ratio of
saturated ﬁtty acids (C16:0, C14:0, C12:0) to the unsaturated ﬁtty acids in the liquid and
semi-solid ﬁactions is 1.46 to 2.7 fold greater, respectively, than native milk ﬁt; (2) the
use of high temperatures might decompose or polymerize the triacylglycerols,

particularly those with high unsaturation, even when distilled under vacuum; and (3)

capital investment costs are high (Arul et al., 1988).

1.2.1.3.2 Melt Crystallization (MC)

This technique is based on a slow cooling of the melted milkﬁt, a short duration
of stabilization at the fractionation temperature and the separation of the crystals from the
liquid phase by vacuum ﬁltration. Makhlouf et al. (1987) ﬁactionated milkﬁt into seven
fractions by MC at temperatures ranging from 26 to 9°C. Butter made from a blend of
different ﬁactions showed two to four fold improvements in spreadability at 7°C
compared to regular butter. After crystallization, the concentration of cholesterol in the
liquid ﬁaction (3.8-mg/ g ﬁt) was nearly twice that in the solid ﬁaction (2.0-mg/ g ﬁt).
Despite its simplicity, MC is a tedious technique and requires many ﬁactionation steps to
obtain low melting point ﬁactions. The major drawback of this technique is the

signiﬁcant overlapping of molecular weights of triacylglycerols (Makhlouf et al., 1987).

19

1.2.1.3.3 Supercritical Fluid Extraction (SFE)

In this process, a product is treated with a gas (usually carbon dioxide) of high
density, low viscosity, and reduced surface tension under high pressure and temperature.
The procedure has the advantage of absence of potentially toxic solvents and by-products
(Friedrich and Pryde, 1984). Am] et al. (1987) ﬁactionated milk ﬁt into eight different
ﬁactions at 10 - 35 MPa and 50 to 70 °C. They found that cholesterol tended to
concentrate in the low and intermediate melting fractions. Lirn et al. (1991) acheived an
overall cholesterol reduction of 92.6 % with a process yield of 88.5 %. The extraction
was done at 40 °C and 24.1 - 27.5 MPa.

Recently, Bhaskar et al. (1993) fractionated milk ﬁt into ﬁve different ﬁactions
(SI — SS) with diﬁ'ering cholesterol levels at 3.5 - 24.1 MPa and 40 - 60 °C. The
saturated to unsaturated ﬁtty acid ratio along with cholesterol levels increased ﬁ'om
ﬁaction $2 — SS compared to the native anhydrous milkﬁt. However, in ﬁaction Sl , the
cholesterol level and the ratio of saturated-to-unsaturated ﬁtty acids decreased by 51 and
24 %, respectively, with respect to the control. The decrease in the saturated-to-
unsaturated ﬁtty acid ratio in S1 was due to the decrease in short chain ﬁtty acids C420
to C10:0 which may be viewed as undesirable due to their potential hypocholesterolemic
eﬁ‘ect (Grurnmer, 1991).

It can be concluded that all the approaches developed to modify the properties
of milkﬁt do not improve the nutritional properties. None of these approaches alter the
nature of the saturated ﬁtty acids. Instead, the saturated ﬁtty acids remain intact and are
separated into diﬂerent ﬁactions. The fractionation of milkfat by physical processes has

been eﬁ‘ective to only a modest extent in improving its melting property. Moreover,

20

fractionation may have a negative impact on the aroma of milkﬁt.

The reported literature has recognized that modifying milkﬁt to reduce its
hypercholesterolemic effects in order to ﬁt under the current dietary recommendation
guidelines will not be competitive in the manufacture of ﬁt products, at least until a
process is developed that will simultaneously remove cholesterol, free ﬁtty acids, and

reduce the myristic acid content.

1.3 ENZYMOLOGY OF DESATURATION
1.3.1 Biosynthesis of Unsaturated Fatty Acids
Saturated ﬁtty acids are known to be the precursors of unsaturated ﬁtty acids in

higher organisms (Masoro, 1968). Many studies have been carried out with
A9-desaturase , which is responsible for the biosynthesis of monounsaturated ﬁtty acids.
What has been found to be true for the A9-desaturase is, in general applicable, to the other
ﬁtty acyl CoA desaturases (A,‘5 and A5) (Brenner , 1974). In contrast to the A9 ﬁtty acyl
CoA desaturase which synthesizes monounsaturated ﬁtty acids, the A6 and A5 ﬁtty acyl
CoA desaturases synthesize polyunsaturated ﬁtty acids (Jeﬁ‘coat, 1979).

Early experiments using liver homogenates established that the desaturation of ﬁtty

acids involves two enzymatic reactions, viz., the activation of the ﬁtty acid to the acyl-

CoA derivative by thiokinases in the presence of CoA, ATP and MgH, followed by the

desaturation of the derivative by the acyl-CoA desaturase system (Masoro, 1968).

21

(1) Activation:
0
H
H Thiokinases, M;
CH3 - (CH2)x - CH2 - CH2 - (CH2)y- C - OH + CoA + ATP 4

 

 

0

H
CH3 - (CH2)x - CH2 - CH2 - (CH2)y - C- SCOA + AMP + ppi

(2) Desaturation:
O
H Desaturase

.
CH3 -(CH2)X- CH2 - CH2 -(CH2)y- C-SCoA + NADH + H + o2 _,

o
n +
CH3 -(CH2)X- CH=CH - (CH2); C - SCoA + 2H20 + NAD

Acyl-Co A is the substrate and reduced nicotinamide adenine dinucleotide (NADH) is the
reducing agent. Oxygen is a necessary requirement for the reaction.

Holloway (1971) reported that marnmlian A9 ﬁtty acyl CoA desaturases are bound
to the endoplasmic reticulum and have an obligatory requirement for NADH, molecular
oxygen, cytochrome b5, and NADH cytochrome b5 reductase (EC 1.6.2.2). These data
coupled to those of Raju and Reiser (1972) who demonstrated the lack of desattuation of
lipid bound stearate, established ﬁtty acyl CoA as the true substrate for stearoyl CoA
desaturase. The same has since been demonstrated to be true for the A6 and A5 ﬁtty acyl
CoA desaturases (Brennner, 1971).

The rules for mammalian desaturation, as described by Jeﬁcoat (1979), are as follows:
(1) All desaturases require molecular oxygen and a reduced pyridine nucleotide to
catalyse the direct oxidative desaturation of preformed ﬁtty acids usually in the

form of coenzyme A esters;

22

(2) Double bonds are always introduced into the methylene chain at a ﬁxed position
ﬁ‘om the carboxyl group;

(3) When the substrate is a saturated ﬁtty acid, the ﬁrst double bond is
inserted between carbon atoms 9 and 10. Unlike plants, animals cannot
introduce double bonds ﬁrther away than nine carbons ﬁ'om the carboxyl group;

(4) When the substrate is already unsaturated, subsequent double bonds are inserted
between the double bond nearest the carboxyl group and the carboxyl group itself,
in such a way as to usually maintain the methylene interrupted distribution of
double bonds;

(5) Unsaturated ﬁtty acids with trans rather than cis double bonds are generally

treated by the enzymes as saturated ﬁtty acids.

1 . 3. 2 Electron Transport Components
The requirment for three protein components in the stearoyl CoA desaturase has

already been established. '

NADH —> fpl
\ 02, stearoyl CoA

Ascorbate —> Cytochrome b5 ——-> CSF
/' Oleoyl CoA +2H20
NADPH —§ fp;

where 11) l: NADH - cytochrome b; reductase; fpzz NADPH - cytochrome c reductase.

(adapted ﬁom Oshino et al., 1971)
CSF: cyanide sensitive ﬁctor

23

These components have been isolated and characterized by Spartz and Strittmatter
(1973) and Strittmatter et al. (1974). The puriﬁed cytochrome b5 reductase (EC 1.6.2.2)
and cytochrome b5 consist of a hydrolytic catalytic segment exposed to the aqueous
environment and a hydrophobic portion involved in binding to the microsomal
membrane. The terminal stearoyl CoA desaturase is a more nonpolar polypeptide of
53000 daltons (62 % hydrophobic amino acid residues) (Strittmatter et al., 1974). It is
completely submerged in the hydrocarbon region of the membrane with the catalytic
donmin exposed to the aqueous environment to permit reduction of its non— heme iron by
cytochrome b, and interaction with the hydrophilic portions of substrate molecules

(Enock et al., 1976).

Fatty acyl-CoA
NADH 02

0
Phospholipid

bilayer .

 

Diagrammatic representation of the A9 ﬁtty acyl CoA desaturase complex. The location of the three
proteins, NADH-cytochrome b5 reductase (I), cytochrome b5 (11), and the desaturase (III) are shown in the

phosphilipid bilayer of the endoplasmic reticulum. (Diagram adapted from Jeﬂ’coat, 1977).

The function of the electron transport proteins is to transfer reducing power ﬁom the
reducing agent (NADH) to the desaturase protein. The heme iron of cytochrome b5
accepts electrons ﬁom the reduced ﬂavin nucleotide and passes them on the non-heme

iron of the desaturase.

24

NADH + I-F

’CZS l-COA +H20
NAD): fpl FAD: ><F cyto. b5 x A9-desat.

\FADHz C18: O-CoA + 1/2 O2
(Adapted from Jeﬁ‘coat, 1979).

I. 3.3 Role of Membrane Lipid in the Desaturation Reaction

Studies by Rogers and Strittmatter (1974) and Strittmatter and Rogers (1975) have
demonstrated a random distribution and translational movement of cytochrome b5 and
NADH-cytochrome b5 reductase in the phospholipid bilayers during electron transfer
ﬁom the ﬂavoprotein to the heme protein. They also suggested that the most likely
ﬁrnction of the lipid is to orient the proteins with respect to each other and therefore to act
indirectly on the activity of the desaturase. Studies by Jones et al. (1969) showed that
extraction of liver microsomes with 90 % (v/v) acetone resulted in 100% loss of stearoyl
CoA activity, which could be restored to normal levels by the addition of the total

extracted lipid ﬁaction.

1.3.4 Role of Metal Ions in A9-Desaturation

Strittmatter et al. (1974) showed that the puriﬁed rat liver stearoyl CoA desaturase
is a non-heme iron-containing enzyme, as 80% of its activity was inhibited upon the
addition of 10 mM bathophenathroline sulfonate (ion chelator). The stearoyl CoA

desaturase iron participates in the catalytic process by mediating oxidation-reduction

H
reactions through reversible changes in its oxidation state. Mg is not required for the

25

desaturation reaction per se. However, it is an obligatory component in the activation

reaction of the ﬁtty acid (Masaro, 1968).

I. 3. 5 Cytoplasmic Proteins and Desaturation

The biosynthesis of unsaturated fatty acids involves an interplay between soluble
ﬁtty acid synthetase and membrane-bound enzymes, the desaturases. Free ﬁtty acids can
be highly toxic within the cell, hence there are particular species of cytoplasmic proteins
which bind to them to protect cell membranes and membrane-bound enzymes ﬁom their
cytotoxic effect. This is analogous to the albumin in the circulation system (Bass, 1985).
The functional role of these proteins is not clearly understood. In crude microsomal
preparations, full activity of the, A9—, A6—, A5- fatty acyl-CoA desaturase system is
expressed. Upon repeated washing, the activities of the three enzymes are lost, but can
be restored by the addition of cytoplasmic proteins (Catala et al., 1975).
Fractionation studies of cytoplasmic proteins of the liver showed that A9-desaturation is
stimulated by a 46,000 mOlecular weight protein (Jeffcoat et al., 1976) in the way that
bovine serum albumin affects enzyme activity. Brophy and Vance (1976) and Jeﬁ‘coat et
al. (1977) suggested that this protein acts as a ﬁtty acid binding protein which regulates
the availability of ﬁtty or ﬁtty acyl-CoA for the lipid-metabolizing enzymes.
Jeﬁ‘coat (1977), working with puriﬁed A9-desaturase, demonstrated that cytoplasmic
proteins are not obligatory for the desaturation reaction per se and suggested that the role
of bovine serum albumin is to protect the acyl-CoA substrate from the effect of
acylthiolester hydrolases. Initially, it was believed that acyl-CoA (ACBP) and ﬁtty acid

binding proteins (FABP) were one and the same. However, more recent studies

26

(Rasmussen et aL, 1990) have shown these ﬁrnctions reside in separate proteins having

different molecular weights (FABP (Mr =12 - 15 x103) and ACBP (M, = 9932) and no

sequence similarities.

27

CHAPTER 2

DESATURATION OF MYRISTOYL—COA TO MYRISTOLEOYL-COA

BY HEN LIVER MICROSOMAL A9-DESATURASE.

2.1 ABSTRACT

Desaturation of myristoyl-CoA to myristoleoyl-COA was measured in
microsomal preparations of hen liver. Desaturation was maximal at pH 7.4.
Enzymatic activity was linear with time up to 10 min and proportional to microsomal
protein concentrations. The initial velocity was linear with substrate concentrations

between 13 and to 200uM. Decrease in desaturation activity was observed at

substrate concentrations greater than 266uM. There was an absolute requirement for
reduced pyridine nucleotide (N ADH), while maximum activity was observed at a
myristoyl-CoAzNADH mole ratio of 1. Competitive inhibition studies of myristoyl-
CoA desaturation suggest that the inhibitors stearyl and oleyl-CoA were more
erreetive than palmitoyl-CoA. Free CoA did not inhibit the A9-desaturase system. The
desaturation of myristoyl-CoA was stimulated by bovine serum albumin and reduced
by cytoplasmic proteins. The effect of cytoplasmic proteins in the enzymatic reaction
was completely abolished by trypsin digestion and boiling for 30 min. On the basis of
the present data, it was concluded that 9,10-desaturation of acyl-CoA derivatives

containing 12to 18carbonﬁttyacylchainsarecatalyzedbythesameenzyme.

28

2.2 INTRODUCTION

Enzymatic desaturation of saturated ﬁtty acids to form
monounsaturated ﬁtty acids is known to occur in various aerobic organisms including
yeast and several animal species (Brett et al., 1971). Studies with liver homogenates
have established that the desaturation of ﬁtty acids involves two enzymatic reactions,

viz., the activation of the ﬁtty acid to the acyl-CoA derivative by thiokirrases in the

presence of CoA, ATP and MgH, followed by the desaturation of the derivative by

the acyl-CoA desaturase system (Masoro, 1968).

(1) Activation:
0
H
H Thiokinases, IV?
CH3 - (CH2)x - CH2 - CH2 - (CH2)y- C - OH + CoA + ATP 4

 

 

0

II
CH3 - (CH2)x- CH2 - CH2 - (CH2)y -C- SCoA + AMP + ppi

(2) Desaturation:

0
ll Desaturase

CH3 -(CH2)x- CH2 - CH2- (CH2)y-C -SCoA + NADH + H ++ 02 ——>
0

II
CH3 -(CH2)x- CH=CH- (CH2)y- C - SCoA + 21120 + NAD.

Acyl-Co A is the substrate and reduced nicotinamide adenine dinucleotide (N ADI-I) is
the reducing agent. Oxygen is a necessary requirement for the reaction.

Holloway (1971) reported that mammalian A9 ﬁtty acyl-CoA desaturases are

bound to the endoplasmic reticulum and have an obligatory requirement for NADH,

29

molecular oxygen, cytochrome b5, and NADH cytochrome b, reductase (EC 1.6.2.2).
These components have been characterized by Spartz and Strittmatter (1973) and
Strittmatter et al. (1974). The puriﬁed cytochrome b5 reductase (EC 1.6.2.2) and
cytochrome b5 consist of a hydrolytic catalytic segment exposed to the aqueous
environment and a hydrophobic portion that is involved in binding to the microsonml.
membrane. The terminal stearoyl CoA desaturase is a more nonpolar polypeptide of
53000 daltons and contains 62 % hydrophobic amino acid residues (Strittmatter et al.,
1974). It is completely submerged in the hydrocarbon region of the membrane with
the catalytic domain exposed to the aqueous environment to permit reduction of its
non- heme iron by cytochrome b5 and interaction with the hydrophilic portion of
substrate molecules (Enock et al., 1976).

Using puriﬁed stearoyl-CoA desaturase, Enoch et al. (1976) reported that
9,10—desaturation of ﬁtty acyl-CoA derivatives with chain lengths of 12 to 19 carbon
atoms is catalyzed by the same enzyme, with stearoyl-CoA being the best substrate.
These studies, coupled with those of Raju and Reiser (1972) who demonstrated the
lack of desaturation of lipid-bound stearate, established ﬁtty acyl-CoA as the true
substrate for stearoyl CoA desaturase. Similar conclusions have been established for
the A6 and A5 fatty acyl CoA desaturases (Brennner, 1971).

As ﬁee fatty acyl—CoA derivatives can be highly toxic within the cell, there
are particular species of cytoplasmic proteins which bind to them to protect cell
membranes and membrane bound enzymes from their cytotoxic effect. This is
analogous to the presence of albumin in the blood circulatory system (Bass, 1985).

The functional role of these proteins is not clearly understood. Brophy and Vance

30

(1976) and Jeﬁ'coat et al. (1976, 1977, 1978) suggested that these proteins act as ﬁtty
acid-binding proteins which regulate the availability of ﬁtty acids or ﬁtty acyl-CoA
for the lipid-metabolizing enzymes. The latter investigators, working with puriﬁed
A9-desaturase, demonstrated that bovine serum albumin (BSA) is not obligatory for
the desattuation reaction per se, suggesting its role to be the protection of the ﬁtty
acyl-CoA substrate from the effect of acylthiolester hydrolases.

In this study, the initial objective was to develop an assay system to monitor
the desaturation of myristoyl-CoA to myristoleoyl-CoA by A9-desaturase system in
hen liver microsomes. This assay procedure was subsequently used to study several
properties of the enzyme system including speciﬁc activity, substrate speciﬁcity and

the interaction of the substrate with bovine serum albumin and cytosolic proteins.

2.3 MATERIALS AND METHODS
2. 3. 1 Materials

Myristoleoyl-, myristoy1-, palrnitoyl-, stearoyl-, oleoyl-CoA (lithium salt),
tetradecanoic, cis-9-tetradecenoic and pentadecanoic acids. BSA, trypsin, trypsin
inhibitor, and reduced nicotenamide adenine dinucleotide (N ADH) were obtained
ﬁom Sigma Chemical Co. (St Louis, MO). The standards were of greater than 98 %
chemical purity according to the manuﬁcturers. Magnesium chloride and potassium
hydroxide were obtained ﬁ'om Aldrich Chemical Co. (Milwaukee, WI). Laying hens

(White Leghorn) were obtained from the MSU Poultry Farm.

2. 3. 2 Preparation of hen liver microsomes

31

For each set of experiments, three laying white Leghorn hens were sacriﬁced.
The whole livers were excised and cooled immediately over ice. The livers were
minced and mixed together to decrease possible variations in desaturation yield ﬁom
one single liver preparation to another. A 20g sample of the ground livers was
immediately homogenized in a Waring blender for 30 sec with 60 ml of chilled 0.1 M
potassium phosphate buffer (pH 7.4) containing 0.3 M sucrose. The temperature of
the buﬂer solution was 0 to 2°C. The resulting homogenate was centrifuged at 15,000
x g for 30 min at 2 °C. The sedirnented cellular debris and mitochondria were
discarded. Microsomes containing the Ag-desaturase system were sedirnented ﬁ'om
the postmitochondrial supernatant fraction by centrifuging at 105,000 x g for 1 hr at
2 °C. The resulting microsomal pellet was either suspended in the homogenizing
buffer or in the microsomal supernatant (containing the cytoplasmic proteins) to
provide a ﬁnal concentration of 2.66 mg of microsomal protein/ml. The protein
concentration was determined by the method of Lowry et aL (1951). The remaining
ground livers were stored at -20 °C for no longer than 10 hours. Independent studies

indicated that there was no loss in enzymatic activity over the duration of the holding

period.

2. 3.3 Enzyme assays
The process to convert myristoyl-CoA to myristoleoyl-CoA using the A9-
desaturase system of hen liver microsomes is outlined in Figure 1. The rationale for

each subsequent step in the process is described below.

32

60 mg BSA dissolved in 2.4 mL distilled (d)H20
.1
1,000 nmoles C14:0 CoA dissolved in 200 pl dH20
1
Incubate for 5 min at room temperature (25 :t 2 °C)
1

6.2 mg NADH dissolved in 1 mL dHZO
(NADH:C14:0 CoA mole ratio of 9)

l.
0.94 mg MgClz dissolved in 0.1 mL deO
.1
Adjust pH to 7i4 with 1N HCl

V

Add 3 ml hen liver microsomes suspended in homogenizing buffer
(2.66 mg microsomal prote'in/mL homogenizing buffer)

V

Incubate the digest at 37 °C for 5 min
(total volume of the digest was adjusted to 7.5 mL with dHZO )

Figure 1- Process for converting myristoyl-CoA to myristoleoyl-CoA using the
rig-desaturase system of hen liver microsomes. The total volume of the digest was
adjusted to 7.5 mL by the addition of 0.8 mL distilled water. The rrricrosomes were
suspended in the homogenizing buffer with a vortex mixer.

33

2. 3. 4 Lipid extraction and fatty acid methyl esterzﬁcation

The enzymatic reaction was stopped by adding 4 ml of 10 % methanolic KOH
to saponify the lipids. The methanolic KOH solutions were heated at 65° C for 30 min
and then acidiﬁed with 7 N HCL. Samples were extracted three times with a mixture
of hexane: isopropanol (3:2, v/v). The combined solvent extracts were evaporated
under nitrogen. The internal ﬁtty acid standard C15:0 (0.02 mg) was added to each of
the samples which were then derivatized to their ﬁtty acid methyl esters by the boron

triﬂuoride/methanol esteriﬁcation procedure of Morrison and Smith (1964).

2. 3. 5 Gas chromatographic (GC) and GC-MS analyses

Gas chromatographic analyses were carried out on a Hewlett-Packard 5890 A
gas chromatograph (Avondale, PA) equipped with a ﬂame ionization detector (FID),
split-mode(1/ 12), and a fused silica capillary column (30 m x 0.25 mm) coated with a
0.25 pm ﬁhn of DB-225 (J & W Scientiﬁc, Folsom, CA). Operating conditions of the
gas chromatograph were as follows: temperatures — injector 275 °C, detector 300 °C,
colunm 165 °C; hold 5 min; increase over temperature 0.5 °C/min to 180 °C, hold 20
min, increase 10 °C to 215 °C, hold 10 min. Fatty acid methyl esters were identiﬁed
by comparing their relative gas chromatographic retention times with those of known
standards. Standard curves were generated from known concentrations of pure methyl
ester standards of C14:0 and cis-9-C14zl ranging ﬁ'om 7.8 to 62.5 pg and containing
0.01 pg ofC15:0. The areas ofthe C14:0, C14:1, and C:15:0 peaks were determined
using a digital integrator. The C14:0 and C14:1 peak areas were divided by the

internal standard area to obtain the response ratios for both standards and samples.

34

The standard response ratios for C14:0 and cis-9- C14:1 were plotted against their
corresponding concentrations. Standard response ratio plots should bracket the
sample response ratio. The C1420 and cis-9-C14:l concentrations in the sample
extracts were quantiﬁed from the standard response ratio plots and corrected for the
dilution ﬁctor (df = 2). Initial analyses indicated that there was no endogenous cis-9r
C14:1 in the medium. Thus, the cis-9-C14zl observed was derived as a result of the
desaturation reaction per se.

The identity of cis-9-C14:1 was conﬁrmed by comparing the mass spectra of
authentic cis-9-C14:l standard and that derived ﬁ'om the desaturation of C14:0
(Appendix 1). The GC-MS analyses were carried out on a JEOL A x 505H double
focusing mass spectrometer equipped with a Hewlett-Packerd 5890 J GC. The ﬁtty
acid methyl esters, dissolved in hexane, were injected using a splitless injection
technique onto a 30m x 0.25 mm i.d., 25pm coated DB 225 capillary column
interﬁced directly into the ion source. The GC oven temperature was programmed
ﬁ'om 100°C to 165°C at 5°C per min and held for 5 min then to 172°C at 0.5 °C per
min and ﬁnally to 230 °C at 10°C per min. The injector and transfer line temperatures
were maintained at 230 °C and 230°C, respectively. Fatty acids were analyzed in the
m/z range of 45 — 600 for the methyl ester derivatives in positive electron ionization
mode (E1). The analyses indicated that the ﬁtty acid resulting from the desaturation

of Cl4:0 by the hen liver desaturase system was indeed cis-9-Cl4: 1.

2. 3. 6 Trypsin digestion
To a 6 mL aliquot of the microsomal supernatant containing 31.5mg/mL of

cytosolic proteins, 340 mg of trypsin were added, and the mixture was incubated for 1

35

hour at 37 °C. Proteolysis was stopped with the addition of 680 mg trypsin inhibitor.

2. 3. 7 Statistical analysis

All treatments were performed in duplicate (n=2). Statistical analyses were
based on the polynomial regression and analysis of variance (ANOVA) procedures of
SYSTAT 7.0 (SPSS Inc., 1997). ANOVA was also used to analyze the data and
included NADH, competitive inhibitors, heat treatment, trypsin digestion, BSA and
microsomal supernatant (MS) as main effects and BSA x MS as interaction. The
Tukey multiple comparison procedure was used for mean separation (Berk, 1998). If
levels within a ﬁctor were signiﬁcantly different, they were examined further using
the student t-test for differences among individual means. Results were judged to be

statistically signiﬁcant based on type I error rate of 5%.

2.4 RESULTS AND DISCUSSIONS

Mammalian desaturases have been known for many years to be bound to the
endoplasmic reticulum and require oxygen and NADH (Oshino et al., 1966; 1971). In
mammalian tissues, the formation of monoenes ﬁom free ﬁtty acids is a two-stage
reaction: activation of the saturated ﬁtty acid to ﬁtty acyl-CoA by acyl-CoA
synthetase, followed by desatuartion of the acyl-CoA derivative (Masoro, 1968). In
the activation reaction, the energy is generally supplied by adenosine triphosphate
(ATP) conversion to adenosine monophosphate (AMP) and pyrophosplmte (Pande
and Mead, 1968). Previous studies of desaturation generally used ﬁee ﬁtty acids as

the substrate in the presence of CoA, Mg”, and ATP, and thus the rates of both

36

reactions were measured. In the present study with A9-desatlrrase of hen liver
microsomes, the desaturation was studied as a single reaction using the activated form
of the substrate (myristoyl-CoA) in order to ensure that the results were not due to
inhibition of the initial activation of free myristic acid to myristoyl-CoA. When the
ﬁtty acyl-CoA is used as the substrate, the activation reaction of the substrate does '

not take place and therefore it is not a rate limiting step in the desaturation reaction.

2.4.1 Factors that modib' the yield of desaturation of myristoyl-CoA

Hen liver microsomes, like rrricrosomes from other sources (Brett et al.
1971), catalyzed the oxygen-and reduced pyridine nucleotide-dependent desaturation
of myristoyl-CoA to myristoleoyl-CoA. In order to test the reliability of the assay
conditions, the effects of time, microsomal protein concentration, substrate
concentrations, reducing agent, pH, competitive inhibitors, BSA and microsomal

supernatant on enzyme activity were investigated.

2.4.1.] Effect of incubation time

Under the assay conditions, the amount of myristoleate formed was directly
related to the incubation times. When the desaturation reaction was stopped after 5
min, a linear relationship between desaturation and time was observed (Figure 2). A
similar relationship was observed when the reaction was terminated at 10 min. The
linear relationship between desaturation rate and incubation times (up to 10 min) is
indicated by the equation Y = 3.7643X (R2 = 0.93). Monoene production declined

sharply after 30 min of incubation which may be due to the decrease in substrate

37

 

45.00

33.75 L

22.50 —

 

11.25 - , —
Y = -0.0968X + 4.175X

R2= 0.962

nmol C14:1/mg microsomal protein

 

 

 

000 l l l I l l
0 5 1015 20 25 30 35

Time (min)

Figure 2- The desaturation of myristoyl-CoA as a ﬁrnction of time by the
A9-desaturase of hen liver microsomes. The desaturation activity was measured at
37°C under conditions described in Figure l with diﬂemnt incubation times. The
complete system contained 8 mg hen liver microsomal protein, 40 mM potassium
phosphate (pH 7.4), 120 mM sucrose, 133 11M C14:0CoA, 60mg BSA, 1.16 mM
NADH, 1.31 mM MgC12, pH 7.4, in a total volume of 7.5 mL.

38

concentration as a result of its utilization in other enzymatic reactions. In the liver
cell, any saturated acyl-CoA can suffer a variety of fates, such as being hydrolyzed to
the unesteriﬁed fatty acid, converted into lipid, or undergoing B-oxidation, chain

elongation or desaturation (V oet, 1990).

2. 4. 1. 2 Eﬂect of microsomal protein concentration

The rate of desaturation of myristoyl-CoA was shown to be directly
proportional to the amount of the microsomal protein over the range of 2 to 8 mg per
7.5 ml of ﬁnal incubation volume (Figure 3). The linear relationship between the
desaturation rate and the microsomal protein concentration (up to 8 mg per 7.5 ml of
ﬁnal incubation volume) is indicated by the equation Y = 5.1419X (R2 = 0.99).
Enzyme activity was slowed at protein concentrations greater than 10 mg per 7.5 ml
of incubation medium. A relatively high microsomal protein concentration of 8 mg
per 7.5 ml incubation volume was purposely employed in all the assays, unless
otherwise indicated, to assure that there was sufﬁcient production of
myristoleate over the incubation time of the assay in order to be quantiﬁable by the
gas chromatographic procedure. At constant levels of substrate, monoene formation
was slowed at high protein concentrations and slowed by prolonged incubation. These
latter two ﬁndings were also observed in rat brain desaturation systems (Cook and
Spence, 1973) and could be due, in part, to the addition or accumulation of

desaturation product.

39

 

   

 

 

50 I I I I I
40 — -—
.E
E
g 30 -— —
v
v-
0
'5 20 - —
E
C
10 - Y = -o.2872x2 + 7.0225x-i
R2= 0.9944
I I I I

 

O 2 4 6 8 1O 12
mg microsomal protein

Figure 3- The desaturation of myristoyl CoA as a function of microsomal protein
concentration by the A9 desaturase of hen liver microsomes. The desaturation activity
was measured under the conditions described in Figure 1 with different microsomal
protein concentrations. The complete system contained 40 mM potassium phosphate
(pH 7.4), 120 mM sucrose, 133 pM C14:0CoA, 60mg BSA, 1.16 mM NADH,

1.31 mM MgC12, pH 7.4, in a total volume of 7.5 mL. Incubations were at 37°C for
5min.

4o

2.4.1.3 Effect of substrate concentration

Figure 4 reveals the relationship of initial velocity to substrate concentration
for the hen microsomal A9-desaturase catalyzes reaction. The plot was linear between
13 and 133 uM. Inhibitory effects of substrate concentration occurred at
concentrations in excess of 300 uM. The assays of initial rates could not be measured
accurately at substrate concentrations lower than 13 uM, because it was diﬂicult to
have suﬁcient product to be accurately determined by gas chromatography. Because
of the inhibition at high substrate concentrations and variation at the very low
concentrations (i.e., less than 13 uM), it was diﬁicult to determine K... for the
desaturation reactions. Unless otherwise indicated, most assays were carried out at a
substrate concentration of 133 uM, allowing recovery of product signiﬁcantly above
background values (if any) and avoiding the possibility of substrate inhibition.
Further, the use of this constant substrate concentration decreased the likelihood of
non-ﬁrst order reaction conditions of reduced substrate availability, by incorporation
into lipids by acyltransferase. These data, together with those in Figures 2 and 3,
clearly show that the assay described here gives a true measure of the initial velocity

under pseudo ﬁrst order reaction conditions.

2.4.1.4 Eﬂect of reducing agent

The reducing equivalent in the desaturation reaction is used to reduce the

[Ag-desaturase system through an electron transport chain (Oshino et al., 1971).

41

 

\l

0)

01

.h

(a)

N

 

Y = -71~:-5x2 + 0.0422X_
R2= 0.8391

A
j

 

 

nmol C14:1/minlmg microsomal protein

1 l l l L

0 100 200 300 400 500 600
SUbstrate Concentration (pM)

D

Figure 4- The relationship of initial velocity to substrate concentration for the hen
microsorml A9-desaturase catalyzed reaction. The desaturation activity was measured
under conditions described in Figure 1 with diﬂ‘erent substrate concentrations. The
complete system contained 8 mg hen liver microsomal protein, 40 mM potassium
phosphate (pH 7.4), 120 mM sucrose, 60mg BSA, 1.16 mM NADH, 1.31 mM MgC12,
pH 7.4, in a total volume of 7.5 mL. Incubations were at 37°C for 5min.

42

The microsomal nonphosphorylating electron transport chain involved in the fatty
desaturation includes a ﬂavoprotein, the desaturase enzyme attributable to an hemo-
protein and cytochrome b5 (Oshino et al., 1971). The reducing equivalent required for
the desaturation site is supplied by the NADH via cytochrome b5. The rate of
cytochrome b; reduction by NADH in rat liver microsomes is in the order of 1,000
nmoles per min per mg protein. In contrast, the activity to desaturate stearyl CoA is,
at most, less than 10 nmoles of oleate formed per min per mg protein in even highly
“induced microsomes” (Oshino et al., 1971). Thus, the capacity of the microsomal
system to supply reducing equivalents to cytochrome b5 is greater than that required
for the desaturation of fatty acyl CoAs.

The effect of NADH concentrations on the desaturation of myristoyl-CoA is
shown in Figure 5. A wide range of NADHzmyristoyl-CoA mole ratios was assessed
under the experimental conditions described in Figure 1. In the absence of NADH,
no product formation was observed suggesting the absence of endogenous reducing
equivalents. At NADH: myristoyl-CoA mole ratios ranging ﬁ'om 1 to 9, the
desaturation was maximal. Decrease in desaturation was observed at NADH:
myristoyl-CoA mole ratio of 12. A NADH: myristoyl-CoA mole ratio of 9 was
employed for all the assays, unless otherwise indicated, to ensure that the electron

transport system was not rate-limiting in the desaturation reaction.

2.4. 1.5 Effect of pH

The velocity of enzyme-catalyzed reactions depends on pH. The importance of

pH in the desaturation reaction is due to its ability to: (1) change the ionization state

43

 

 

 

nmol C14:1/min/mg microsomal protein

 

0 1 3 6 9 12
NADH/C1420 CoA (mole ratio)

Figure 5- Eﬂ'ect of increasing concentrations of reduced pyridine nucleotide (NADH)
on the desaturation of myristoyl-CoA by the A9—desaturase system in hen liver
microsomes. The assay was performed as descn'bed in Figure l with different
NADHzmyristoyl-COA mole ratios. The complete system contained 8 mg hen liver
microsomal protein, 133 uM C14:0 CoA, 40 mM potassium phosphate (pH 7.4),
120 mM sucrose, 60mg BSA, 1.31 mM Mng, pH 7.4, in a total volume of 7.5 mL.
Incubations were at 37°C for 5min

"° Means with no common superscript are signiﬁcantly diﬁ‘crent (p<0.05).

of the A9-desaturase system, (2) change the ionization state of the substrate which in
turn aﬁ‘ects the reactivity of the reaction, and (3) change the solubility of the substrate
solubilizing protein (V oet, 1990). The A9-desaturase system gave a bell-shape curve
of velocity versus pH (Figure 6) which is typical of most enzymes. Optimal pH for
the formation of monoenes ﬁom myristoyl-CoA in hen liver microsomes was 7.4. At

physiologically extreme pH values, the desaturation reaction was completely

inhibited.

2. 4. 1. 6 Eﬂect of competitive inhibitors

Paulsrud et al. (1970) reported that desaturase activity in rat liver microsomes
increases with increasing fatty acid carbon number from 10 through 18. Brett et al.
(1971), who examined four other desaturation systems, viz., microsomal preparations
of goat mammary gland, hen liver, and whole cells of T orulopsis bombicola and
ChloreIIa vulgaris, concluded that a widely distributed single enzyme whose activity
is maxirml with stearic acid and high with palmitic acid also, inserts the double bond
between C9 and Cu). Working with puriﬁed stearyl-CoA desaturase, Enoch et al.
(1976) reported that 9,10-desaturation of fatty acyl-CoA derivatives of varying chain
lengths (12 to 19 carbon atoms) by rat liver microsomes are catalyzed by this enzyme,
with stearyl-CoA being the best substrate. This is consistent with the observation that
oleate is the major monounsaturated fatty acid of rat liver lipids. Similar observations
are presented in Figure 8 when several acyl-CoA derivatives were used as inhibitors.
Stearoyl-, oleoyl-, and palmitoyl-CoA derivatives were effective competitive

inhibitors for the [Ag-desaturase of hen liver microsomes.

45

 

.h

I I I I
Y = 0.089x3 — 2.581xJ + 23.703x — -66.57
R2 = 0.9535

(JO

nmol C14:1/min/mg microsomal protein
... N

 

 

 

 

mo
Q
\l
(I)

9 101112

Figure 6- Effect of pH on the desaturation of myristoyl-CoA by the
A9—desaturase of hen liver microsomes. The assay was performed as described
in Figure 1 with diﬁ‘erent pH values. The complete system contained 8 mg hen
liver microsomal protein, 133 uM C14:O CoA, 40 mM potassium phosphate
(pH 7.4), 120 mM sucrose, 60mg BSA, 1.16 mM NADH, 1.31 mM MgC12, in
a total volume of 7.5 mL. Incubations were at 37°C for 5min.

46

60 mg BSA dissohied in 2.4 mL deo

V

1,000 mnoles C14:0 CoAIdissolved in 200 pl dH20

v

Incubate for 5 min at room temperature

V

1000 nmoles competitive inhil'aitor dissolved in 200 pl deO

v

6.2 mg NADH dissolved in 1 mL dH20
(NADH:C14:0 CriA mole ratio of 9)

V

0.94 mg MgClz disso|lved in 0.1 mL dHZO

V

Adjust pH to 714 with 1N HC]

V

Add 3 ml hen liver microsomes suspended in homogenizing buffer
(2.66 mg microsomal protelin/mL homogenizing buﬂ‘er)

V

Incubate the digest at 37 °C for 5 min
(total volume of the digest was adjusted to 7.5 mL with dH20 )

Figure 7- Eﬂ‘ect of competitive inhibitors on the desaturation of myristoyl-CoA by the
A9-desaturase system of hen liver microsomes. The total volume of the digest was
adjusted to 7.5 mL by the addition of 0.6 mL distilled water. The total number of
acyl-CoA derivatives (substrate and competitive inhibitor) in the incubation medium
did not exceed 2000 nmoles as over this concentration the substrate itself was
inhibitory.

47

110 l I l l T

 

 

 

 

§
9 100
E
o 90
I.—
O
“E, 80
E
3 70
(U
8
'0 60
Q)
.2
l! 50
d)
0:
40 \
6°00} 6°00“ %Noo?‘ 00% 00660
0" o" 0"

Competitive inhibitors

Figure 8- Eﬂ‘ect of palmitoyl—, stearoyl-, and oleoyl-CoA, and ﬁee CoA on the
activity of the A9-desaturase of hen liver microsomes (normalized to the control =
100%) using myristoyl-CoA as a substrate. Incubations were carried out under
conditions of initial velocity described in Figure 7. The concentration of the
myristoyl-CoA in the incubation medium (133.3pM) was equal to the concentration
of the competitive inhibitor. The total number of acyl-CoA derivatives (substrate and
competitive inhibitor) in the incubation medium should not exceed 2000 nmoles
(266.66 uM) since over this concentration the substrate itselfwas detrimental to the
enzyme activity.

"° Means with no common superscript are signiﬁcantly different (p<0.05).

48

The desaturase enzyme was not inhibited by ﬁ'ee CoA. Again, this is consistent with
the results reported by Enoch et al., (1976). The inhibition is competitive with
stearoyl-, oleoyl-, and palrnitoyl-CoA . Stearoyl- and oleoyl-CoA were more eﬂective
in inhibiting the desaturase activity than palmitoyl-CoA. In the presence of stearoyl-
and oleoyl-CoA, the desaturation rate decreased by 52.3 and 54.1%, respectively. In -
the presence of palmitoyl-CoA the decrease in the desaturation rate was 43.4%
(Figure 8).

The fact that inhibition of enzymatic desaturation has been shown to be
dependent on the chain length of the acyl-CoA derivatives indicates that the number
of methylene groups of the fatty acid may regulate the desattuase activity via
London-Van der Waal’s forces that bind the fatty acid to the enzyme. As the
hydrocarbon tail of the fatty acid is essentially hydrophobic, London-Van der Waals
dispersion forces would be able to provoke binding with the enzyme. Because the
dispersion forces vary as 1/97, D being the distance between the interacting
molecules, the substrate and the lipoproteic enzyme must be close enough to make the
binding eﬂeaive. These forces increase with the number of carbon units of the
molecule and in consequence may be quite large. Salem (1962) calculated the effect
of each -CH2- group at an interaction distance of 5 A° to provoke a London-Van der
Waals dispersion (Wdisp.) energy of: w disp./N= 0.4 Kcal/mole, where N is the
number of identical units. This strengthening of the London-Van der Waals forces
with the number of -CH2- may explain the high aﬂinity of the enzyme toward

stearoyl- and oleoyl-CoA.

49

2. 4. I . 7 Eﬁrect of bovine serum albumin and cytosolic proteins

BSA has been shown to bind fatty acyl-CoA derivatives and protect them
ﬁom acylthiolester hydrolases (Jeﬁcoat et al., 1976). The effect of BSA on the
desaturase system was investigated and results clearly show a stimulation of enzyme
activity under the experimental conditions used (Figure 9). The speciﬁc enzyme
activity, expressed as nmol C14:1 produced per min per mg protein, was found to be
3.49 and 1.93 in the absence and presence of the microsomal supernatant,
respectively. When BSA was not added to the incubation medium, the speciﬁc
activity dropped to 3.09 and 1.73 in the absence and presence of microsomal
supernatant, respectively. Figure 9 also reveals that the microsomal supernatant has
the capacity to reduce the speciﬁc activity of the desaturase system whether BSA was
present or not.

The effect of heat and trypsin digestion on the capacity of the microsomal
supernatant to decrease the desaturation activity was investigated. The microsomal
supernatant obtained as described earlier is heat labile. The activity of the microsomal
supernatant which contain the cytosolic proteins, was completely lost after 30 min of
boiling using the assay conditions described in Figure 10. Boiling induced
precipitation of the proteins. The increase in the desaturase speciﬁc activity due to the
heat precipitation of cytosolic proteins was 1 nmol per min per mg microsOmal
protein compared to the control (Figure 11).

The eﬂ‘ect of trypsin digestion on the microsomal supernatant is shown in
Figure 11. Desaturase speciﬁc activity was increased by 1.04 nmol per min per mg

microsomal protein compared to the control. The control experiment was performed

50

 

 

 

nmol C14:1/min/mg microsomal protein

 

"“('5 P‘ £5 “‘55 % £3)
.99“ 85° 99‘” .69

iSqubilizing proteins

Figure 9- Effect of bovine serum albumin (BSA) and microsomal supernatant (MS)
on the desaturation of myristoyl-CoA under the experimental conditions described in
Figure 1. The complete system contained 8 mg hen liver microsomal protein
suspended in the homogenizing buﬂ'er, 40 mM potassium phosphate (pH 7.4),

120 mM sucrose, 133 pM C14:0 CoA, 60mg BSA, 1.16 mM NADH,1.31 mM
Mng, pH 7.4, in a total volume of 7.5 mL. Incubations were at 37°C for 5min.
When MS was added, the 8 mg microsomal protein were suspended in MS instead of
the homogenizing buﬁ‘er. + BSA indicates the presence of bovine serum albumin; -
BSA indicates the absence of bovine serum albumin; +MS indicates the presence of
microsomal supernatant; -MS indicates the absence of microsomal supernatant.

"" Means with no common superscript are signiﬁcantly different (p<0.05).

51

3 mL microsomal supernatani containing 31.5 mg CP / mL

V

1,000 mnoles C14:0 CoAldissolved in 200 pl dH20

v

Incubate for 5 min at room temperature

V

6.2 mg NADH dissolved in 1 mL deO
(NADH:C 14:0 ColA mole ratio of 9)

V

0.94 mg MgClz dissollved in 0.1 mL deO

V

Adjust pH to 7‘4 with 1N HCl

V

Add 3 ml hen liver microsomes suspended in homogenizing buffer
(2.66 mg microsomal protein/mL homogenizing buﬁer)

V

Incubate the digest at 37 °C for 5 min
(total volume of the digest was adjusted to 7.5 mL with 0.2 deH20 )

Figure 10- Design of study to establish the effects of heat treatment and trypsin
digestion of the microsomal supernatant containing the cytosolic proteins (CP) on the
desaturation reaction. The microsomal supernatant was trypsin digested in one set of
experiments and boiled for 30 min in another set.

52

 

:5 3.0 l I I
o

a

To

E

8

o 2.5
.§

E

a)

E

.E

g 2.0
E

0

723

c 1.5

 

 

 

Control Heat Trypsin

Figure 11- Effect of heat and trypsin digestion of the microsomal supernatant on the
desaturation of myristoyl-CoA using the A9-desaturase system of hen hepatic
microsomes. The experimental conditions are descrrhed in Figure 10.

"b Means with no common superscript are signiﬁcantly different (p<0.05).

53

with the native microsomal supernatant. The marked decrease in desaturase activity in
the presence of cytosolic proteins could be due to: (l) the incorporation of myristoyl-
CoA into complex lipids by the acyltransferase, (2) the cleavage of myristoyl-CoA to
free myristic acid by acylthiolester hydrolases, and (3) elongation of myristoyl-CoA

to palmitoyl-CoA by the fatty acid synthetases. Cook and Spence (1973) reported that
the rate of incorporation of stearoyl-CoA into phosphoglycerides used as a measure of
acyltransferase activity greatly exceeded its rate of desaturation. A comparison of the
desaturation rate in the presence and absence of BSA (Figure 9) indicates that the
albumin in some way protects the acyl-CoA from the action of endogenous

competing enzymes such as acylthiolester hydrolases and acyltransferase, but not

ﬁ'om the desaturase. Cytosolic proteins and BSA showed no effect on the desaturation

rate when partially puriﬁed stearoyl-CoA desaturase was used (Jeﬂ‘coat et al., 1976).

2. 5 Conclusion

The current research eﬂon is directed at an understanding of the regulation of
myristoyl-CoA desaturation by the A9-desaturase system of hen liver microsomes.
Experimental evidence has been presented to show that the assay described here gives
a true measure of the initial velocity of the reaction under pseudo ﬁrst order reaction
conditions. The inhrhition of the desaturation of myristoyl-CoA by oleoyl-, stearoyl-,
and palrnitoyl—CoA suggests that all these acyl-CoA derivatives are probably
catalde by the same enzyme, stearoyl-CoA desaturase.

The involvement of cytosolic proteins which reduce the desaturation of

myristoyl-CoA has been established. Heat treatment and trypsin digestion were

54

effective means to abolish the capacity of cytosolic proteins to reduce the desaturation
of myristoyl-CoA.

Although all the available evidence implicates the CoA ester of the fatty
acid as the substrate for desaturation, direct desaturation of the fatty acid without
prior activation to the CoA derivative is possible using the experimental conditions

described in Chapter 3.

55

CHAPTER 3

DIRECT DESATURATION OF FREE MYRISTIC ACID BY HEN LIVER
MICROSOMAL A’-DESATURASE WITHOUT PRIOR ACTIVATION TO '
MYRISTOYL-CoA DERIVATIVE.
3.] ABSTRACT
Direct desaturation of free myristic acid by hen liver microsomal

A9-desaturase without prior activation to myristoyl-CoA by the addition of adenosine
triphosphate (ATP) and CoA was observed when the incubation medium was mixed
at mixing speeds greater than 250 rpm in the presence of fatty acid-binding proteins
(FABP). Desaturation was linear with time and proportional to the microsomal
protein concentration Desaturation was maximal at pH 7.9. The greatest desaturation
rate was observed at a mixing speed of 500 rpm in the presence of FABP.
Desaturation decreased at mixing speeds greater than 500 rpm. Data suggest that
when ﬁee myristic acid is bound to FABP in the form of protein-monomer
complexes, its activation to the CoA derivative is not necessary for it to be
desaturated by the A9-desaturase when using mixing rates greater than 250 rpm.
Myristic acid:FABP complexes serve as substrates for the A9-desaturase at mixing
rates greater than 250 rpm. Desaturation was reduced by bovine serum albumin
(BSA) and a-bromohexadecanoate, and no desaturation was observed in the absence
of FABP. These ﬁndings suggest that F ABP may regulate the accessibility of fatty

acids in the desaturation reaction to the active site of the desaturase rather than just

56

. protecting the membrane-bound desaturase ﬁ'om the cytotoxic effect of ﬁ'ee fatty

acids.

3.2 INTRODUCTION

Enzymatic desaturation of saturated fatty acids to form monounsaturated fatty
acids is known to occur in various aerobic organisms including yeast and several
animal species (Brett et al., 1971). Early experiments using liver homogenates
established tlmt the desaturation of fatty acids involves two enzymatic reactions, the

catalyzed activation of the fatty acid to the acyl-CoA derivative by thiokinases in the

presence of CoA, ATP, and MgH, followed by desaturation of the derivative by the
acyl—CoA desaturase system (Masoro, 1968).

It has been reported by Holloway (1971) that mammalian A9- fatty acyl-CoA
desaturases are bound to the endoplasmic reticulum and have an obligatory
requirement for reduced nicotinamide adenine dinucleotide (N ADH), molecular
oxygen, cytochrome b5, and NADH cytochrome b5 reductase (EC 1.6.2.2). These
components have been isolated and characterized by Spartz and Strittmatter (1973)
and Strittmatter et al. (1974). The puriﬁed cytochrome b5 reductase (EC 1.6.2.2) and
cytochrome b5 consist of a hydrolytic catalytic segment exposed to the aqueous
environment and a hydrophobic portion involved in binding to the microsomal
membrane. Stearoyl CoA desaturase is completely submerged in the hydrocarbon
region of the membrane with the catalytic domain exposed to the aqueous
environment to permit reduction of its non-heme iron by cytochrome b5 and

interaction with the hydrophilic portion of substrate molecules (Enock et al., 1976).

57

Using puriﬁed stearoyl-CoA desaturase, the latter investigators reported that 9,10-
desaturation of fatty acyl-CoA derivatives with chain lengths of 12 to 19 carbon
atoms is catalyzed by the same enzyme, with stearoyl-CoA being the best substrate.
These experiments, coupled with those of Raju and Reiser (1972) who demonstrated
the lack of desaturation of lipid bound stearate, established fatty acyl-CoA as the true
substrate for stearoyl CoA desaturase. The same has since been demonstrated to be
true for the A6 and A5 fatty acyl CoA desaturases (Brennner, 1971).

The ﬁrst evidence suggesting the possibility of a direct desaturation of oleoyl
phosphatidycholine to linoleoyl phosphatidylcholine by a membrane -bound enzyme
system was presented by Gurr et al. (1969). Pugh and Kates (1977) presented
evidence for direct As desaturation of eicosatrienoyl lecithin to arachidonoyl lecithin
by rat liver microsomes.

We have undertaken studies to show whether ﬁ‘ee myristic acid could be
directly desaturated without ﬁrst being activated to the CoA derivative. This question
has previously been considered by other investigators (Masoro, 1968; Raju and
Reiser, 1972; Cook and Spence, 1973; Pugh and Kates, 1973; Brenner, 1974; Catala
et al., 1975; Cook, 1978; Voet, 1990), but no evidence for the direct desaturation of
free fatty acids without being ﬁrst activated to the CoA derivative was presented. The
data to be presented here will clearly show direct desaturation of ﬁee myristic acid in
the absence of CoA and ATP, when the incubation medium is mixed at speeds greater
than 250 rpm in the presence of fatty acid-binding protein (F ABP). As described in
the previous chapter, we developed an enzyme assay to quantitate the direct

desaturation of free myristic acid by Ag-desaturase system in hen liver microsomes.

58

This procedure is now being used to study several properties of the enzyme system
including speciﬁc activities, substrate speciﬁcity, and interaction of substrate with

BSA and FABP.

3.3 MATERIALS AND METHODS
3.3.1 Materials

Tetradecanoic, cis-9-tetradecenoic, pentadecanoic acids, bovine serum albumin
(BSA), trypsin, trypsin inhibitor, sodium lactate, NADH, and oxidized nicotimmide
adenine dinucleotide (NAD) were obtained ﬁom Sigma Chemical Co. (St Louis,
MO). The standards were of greater than 98 % chemical purity according to the
manufacturers. Magnesium chloride and potassium hydroxide were obtained from
Aldrich Chemical Co. (Milwaukee, W1). A Lightnin Lab Master (SI) mixer (model
L1U03 and an impeller type A100 (dia. 1”)) were obtained from Matoon and Lee
Equipment, (Farmington, MI). Laying hens (White Leghorn) were obtained from the

MSU Poultry Farm.

3.3.2 Preparation of hen liver microsomes

For each set of experiments, three laying white Leghorn hens were sacriﬁced.
The whole livers were excised and cooled immediately over ice. The livers were
minced and mixed together to decrease possible variations in desaturation yield from
one single liver preparation to another. A 20g sample of the ground livers was
immediately homogenized in a Waring blender for 30 see with 60 ml of chilled 0.1 M

potassium phosphate buffer (pH 7.4) containing 0.3 M sucrose. The temperature of

59

the buffer solution was 0 to 2 °C. The resulting homogenate was centrifuged at
15,000 x g for 30 min at 2 °C. The sedimented cellular debris and mitochondria were
discarded. Microsomes containing the A9-desaturase system were either used directly
in the enzyme assay as mitochondrial supernatant or sedirnented from the
postrnitochondrial supernatant ﬁaction by centrifuging at 105,000 x g for 1 hr at 2 °C.
The resulting microsomal pellet was either suspended in the homogenizing buﬂ‘er or
in the microsomal supernatant (containing the cytoplasmic proteins) to provide a ﬁnal
concentration of 2.66 mg of microsomal protein/ml. The protein concentration was
determined by the method of Lowry et al. (1951). The remaining ground livers were
stored at -20 °C but for no longer than 10 hours. Independent studies indicated that

there was no loss in enzymatic activity over the duration of the holding period.

3.3.3 Enzyme assays
The preliminary assay to convert ﬁ'ee myristic acid to myristoleic acid using
the A9-desaturase system of hen liver microsomes is outlined in Figure l. The

rationale for each subsequent step in the process is described below.

3. 3 4 Lipid extraction and fatty acid methyl esterification

The enzymatic reaction was stopped by adding 20 ml of hexane: isopropanol

(3:2, v/v). Samples were extracted three times with a mixture of hexane: isopropanol
(3 :2, v/v) and the combined solvent extracts evaporated under nitrogen. The internal
fatty acid standard C15:0 (0.02 mg) was added to each of the samples which were
then derivatized to their fatty acid methyl esters by the boron triﬂuoridezmethanol

method of Morrison and Smith (1964).

60

60 mg bovine albumin dissolved in 2.4 ml dH20
1.
1,000 nmoles ﬁ'ee C14:0 dissolved in 10 pl ethanol
.1
Incubate for 5 min at room temperature
1

6.2 mg NADH dissolved in 1 ml dHZO
(NADH:C14:0 mole ratio of 9)

l
0.94 mg MgClz dissolved in 0.1 ml dH20
1
Adjust pH to 7.4 with 1N HCl
1

Add 3 ml hen liver mitochondrial supernatant containing
2.66 mg microsomal protein arlrd 31.5 mg cytosolic protein/ml

V

Mix the digest at 500 rpm for 30 min at 37 °C in a 50 mL glass beaker (dia.1.5”)
(using a Lightnin Lab Master (S1) Mixer model L1U03
and an impeller type A100(dia 1”))

Figure 1- Preliminary assay for the direct desaturation of free myristic acid to
myristoleic acid, in the absence of CoA and ATP, by the Ag—desaturase system of
hen liver microsomes. The total volume of the digest was adjusted to 7.5 mL by the
addition of 0.99 mL distilled water.

61

3. 3.5 Gas chromatographic (GC) and GC-MS analyses

Gas chromatographic analyses were carried out on a Hewlett-Packard 5890 A
gas chromatograph (Avondale, PA) equipped with a ﬂame ionization detector (FID),
split-mode(1/ 12) and a fused silica capillary column (30 m x 0.25 mm) coated with
0.25 pm ﬁlm of DB-225 (J & W Scientiﬁc, Folsom, CA). Operating conditions of the
gas chronmtograph were as follows: temperatures - injector 275° C, detector 300° C,
column 165° C; hold 5 min; increase oven temperature 0.5° C/min to 180° C, hold 20
min, increase 10° C to 215° C, hold 10 min. Fatty acid methyl esters were identiﬁed
by comparing their relative gas chromatographic retention times with those of known
standards. Standard curves were generated from known concentrations of pure methyl
ester standards of C14:0, cis-9-Cl4:1 ranging from 7.8 to 62.5 pg and containing 0.0]
pg of C15:0. The peak area corresponding to C14:0, cis-9-C14zl, and C:15:O were
determined using a‘ digital integrator. The C14:0 and cis-9-C14zl peak areas were
divided by the interim] standard area to obtain the response ratios for both standards
and samples. The standard response ratios for Cl4:0 and cis-9-Cl4zl were plotted
against their corresponding concentrations. Standard response ratio plots should
bracket the sample response ratio. The C14:0 and cis-9-C14zl concentrations in the
sample extracts were quantiﬁed from the standard response ratio plots and corrected
for the dilution factor (df = 2). Initial analysis indicated that there was no endogenous
cis-9-C14:1 in the medium. Thus, all the cis-9-C14zl observed was derived as a result
of the desaturation reaction per se. The identity of cis-9-Cl4zl was conﬁrmed by
comparing the mass spectra of authentic cis-9-Cl4zl standard and that derived ﬁ'om

the desaturation of Cl4:0 (Appendix 1). The GC-MS analyses were carried out on a

62

JEOL A x 505H double focusing mass spectrometer equipped with a Hewlett-
Packerd 5890 J GC. The fatty acid methyl esters, dissolved in hexane, were injected
using a splitless injection technique onto a 30m x 0.25 rmn i.d., 25pm coated DB 225
capillary column interfaced directly into the ion source. The GC oven temperature
was programmed from 100°C to 165 °C at 5°C per min and held for 5 min then to 172
°C at 0.5 °C per min and ﬁnally to 230°C at 10°C per min. The injector and transfer
line temperatures were maintained at 230 °C and 230 °C, respectively. Fatty acids
were analyzed in the m/z range of 45 — 600 for the methyl ester derivatives in positive

electron ionization mode (El).

3. 3.6 Trypsin digestion
To each 6 mL of microsomal supernatant containing 31.5mg/mL of cytosolic
proteins, 340 mg of trypsin were added, and the mixture was incubated for 1 hr at

37°C. Proteolysis was stopped with the addition of 680 mg trypsin inhibitor.

3. 3. 7 Statistical analysis

All treatments were performed in duplicate (n=2). The statistical analyses were
based on the linear and polynomial regressions and analysis of variance (AN OVA)
procedures of SYSTAT 7.0 (SPSS Inc., 1997). ANOVA was also used to analyze the
data and included NADH, competitive inhibitors, heat treatment, trypsin digestion,
BSA and microsomal supernatant (MS) as main effects and BSA x MS as interaction.

The Tukey multiple comparison procedure was used for mean separation

63

(Berk, 1998). If levels within a factor were signiﬁcantly different, they were
examined further using the student t-test for differences among individual means.

Results were judged to be statistically signiﬁcant based on type I error rate of 5%.

3.4 RESULTS AND DISCUSSION

Mammalian desaturases have been known for many years to be bound to the
endoplasmic reticulum and require oxygen and NADH (Oshino et al., 1966; 1971). In
mammalian tissues, the formation of monoenes from free fatty acids is a two-stage
reaction: activation of the saturated fatty acid to fatty acyl-CoA by acyl-CoA
synthetase, followed by desaturation of the acyl-CoA derivative (Masoro, 1968). In
the activation reaction, the energy is generally supplied by adenosine triphosphate
(ATP) conversion to adenosine monophosphate (AMP) and pyrophosphate (Pande
and Mead, 1968). In the present study with A9-desaturase in hen liver microsomes, the
results indicate that direct desaturation of free myristic acid without prior activation to
myristoyl-CoA by the addition of ATP and CoA is possible when the incubation
medium is mixed at mixing speeds greater than 250 rpm in the presence of fatty acid -
binding proteins.

3. 4.] Factors that affect the desaturation free myristic acid

In Chapter 2, it was demonstrated that hen liver microsomes catalyzed the
oxygen-and reduced pyridine nucleotide-dependent desaturation of myristoyl-CoA to
myristoleoyl-CoA. In the current study, we used essentially the same enzyme assay as

for myristoyl-CoA with two exceptions: (1) ﬁ'ee myristic acid was used as the

substrate instead of myristoyl-CoA and (2) the reaction mixture was mixed at 500
rpm to provoke direct collision between substrate and the Ag-desaturase. In
preliminary experiments (Figure 1), the reliability of the assay conditions was tested
by measuring the effect of incubation time, microsomal protein concentration, pH,
temperature, reducing agent, substrate speciﬁcity, inhibition, eﬁ‘ect of BSA, and

microsomal supernatant on the enzyme activity.

3.4.1.1 Eﬂect of incubation time

Under the assay conditions, the amount of myristoleate formed was directly
proportional to the incubation times. Incubations were stopped after 30 min, over
which time the response of the enzyme was linear (Figure 2). Linearity of response
did appear to be consistent for 60 min. Longer incubation times were used with free
myristic acid than with myristoyl-CoA in order to have sufﬁcient monoene to be
accurately determined. Desaturation of myristoyl-CoA was also linear with time but
slowed after 30 min (Chapter 2). The decrease of monoenes formation with time
when myristoyl-CoA was used as a substrate, may be due to the decrease in substrate
concentration as a result of its utilization in other enzymatic reactions (Chapter 2). In
contrast, the direct desaturation of {ice myristic acid was not aﬁ‘ected by competing

enzymic reactions.
3.4.1.2 Eﬂect of microsomal protein concentration

The rate of direct desaturation of ﬁ'ee myristic acid was linear with microsomal

protein concentrations over the range of 2 to 8 mg per 7.5 ml of ﬁnal incubation

65

 

     

 

 

 

3.5 I I I I I I
.E
0:
'§ 3.0— 0 —
o.
g 2.5- -
8 L
o _
,§ 2.0
E
or 1.5 r —
E
if 1'07 Y=0.0551X ‘
O R’=0.9199
'5 0.5 -
E
C

0.0 l l l l l l

0 10 20 30 40 50 60 70
Time (min)

Figure 2- Direct desaturation of ﬁ’ee myristic acid as a function of time by the A9-
desaturase system of hen liver microsomes. The desaturation activity was measured at
37°C under conditions described in Figure l with different incubation times. The
complete system contained 8 mg hen liver microsomal protein, 94.5 mg cytosolic
protein, 40 mM potassium phosphate (pH 7.4), 120 mM sucrose, 133 pM ﬁee Cl4:0,
60mg BSA, 1.16 mM NADH, 1.31 mM MgC12, pH 7.4, in a total volume of 7.5 mL.
The digest was mixed at 500 rpm.

66

volume (Figure 3). Inhibition occurred at protein concentrations somewhat greater
than 8 mg. There was a marked decrease at the concentration of 10 mg per 7.5 ml of
ﬁnal incubation volume. A relatively high microsomal protein concentration of 8 mg
per 7.5 ml incubation volume was purposely employed in all the assays, unless
otherwise indicated, to assure there is enough production of myristoleate during the >
incubation time of the assay. At constant levels of substrate, monoene formation was
inhibited at high protein concentrations. This ﬁnding was also observed when
myristoyl-CoA was used as a substrate. The decrease in the desaturation rate with
high protein concentrations could be due in part to the addition or accumulation of

desaturation products.

3.4.1.3 Eﬂect of pH

The pH of the reaction system inﬂuences the velocity of all enzyme-catalyzed
reactions. The importance of pH in the desaturation reaction is due to its ability to:
(1) change the ionization state of the A9-desaturase system, (2) change the ionization
state of the substrate which in turn affects the reactivity of the reaction, (3) and
change the solubility of the substrate solubilizing protein. The Ag-desaturase was
active within a pH range 6.5 to 8.5 (Figure 4). The optimal pH for the formation of
monoenes from the direct desaturation of free myristic acid in hen liver microSomes

was 7.9.

3.4.1.4 Eﬂect of temperature

Temperature is a key factor in the desaturation reaction due to its ability to:

67

 

  

 

 

 

1-5 I I I I T
O
c
-— 1.0— a
E
\
‘.'.
‘-
‘—
O
“o
g 0.5 — -
Y = -0.01155x2 + 0.2899X
R2= 0.9611
0.0 l I l l

O 2 4 6 8 10 12
mg microsomal protein

Figure 3- Direct desaturation of free myristic acid as a function of microsomal protein
concentration by the A9-desaturase system of hen liver microsomes. The desaturation
activity was measured at 37°C under conditions described in Figure l with different
microsomal protein concentrations. The complete system contained 94.5 mg cytosolic
protein, 40 mM potassium phosphate (pH 7.4), 120 mM sucrose, 133 pM free C14:0,
60mg BSA, 1.16 mM NADH, 1.31 mM mg MgClz, pH 7.4, in a total volume of 7.5
mL. The digest was mixed at 500 rpm for 30 min.

68

 

In

- 0.24 r I
0.23
0.22
0.21
0.20
0.19
0.18

0.17

 

 

nmol C14:1/min/mg microsomal prote

0.16
6

 

Figure 4- Effect of pH on the direct desaturation of ﬁ'ee myristic acid by the
A9—desaturase system of hen liver microsomes. The desaturation activity was
measured at 37°C under conditions described in Figure 1 at different pH values. The
complete system contained 8 mg microsomal protein, 94.5 mg cytosolic protein, 40
mM potassium phosphate (pH 7.4), 120 mM sucrose, 133 pM free C14:0, 60mg
BSA, 1.16 mM NADH, 1.31 mM MgC12, in a total volume of 7.5 mL. The digest was
mixed at 500 rpm for 30 min.

"° Means with no cormnon superscript are signiﬁcantly different (p<0.05).

69

(1) change the rate of the reaction, (2) denature the desaturase system and the
cytoplasmic proteins, and (3) change the ﬂuidity and the crystalline-liquid

crystalline phase transition of the membrane (Sackman and Trauble,l97l).

The optimal temperature for the direct desaturation of free myristic acid by hen liver
microsomes was 37 °C (Figure 5). At lower and higher temperatures, rates of the
desaturation reaction were aﬂected. No C14:1 formation was observed at 25 °C for
the duration of the experiment (30min). Mixing time has no effect on the temperature

as the temperature remained unchanged during the whole incubation period.

3.4.1.5 Effect of reducing agent
The generation of NADH in situ by adding NAD+ and sodium lactate and
taking advantage of the presence of lactate dehydrogemise (LDH) was investigated

using the following coupled enzyme assay.

 

 

- Lactate dehydrogenase
Laaate 53.“ m, Pymvate
NAD NADH
Myristoleic acid (014: 1)‘ Myristic acid (C1420)
A9- desaturase

LDH is present in the mitochondrial supernatant of hen liver (V oet, 1990)). The effect

70

 

.0
N
rs

0.18L

0.12-

0.06 -

 

 

nmol C14:1/min/mg microsomal protein

 

 

.9
No
0

0 27 34 41 48 55
Temperature (°C)

FigureS- Effect of temperature on the direct desaturation of free myristic by the
A9-desaturase system of hen liver microsomes. The desaturation activity was
measured under conditions descn'bed in Figure 1 at different temperatures. The
complete system contained 8 mg microsomal protein, 94.5 mg cytosolic protein, 40
mM potassium phosphate (pH 7.4), 120 mM sucrose, 133 pM ﬁ’ee C14:0, 60mg
BSA, 1.16 mM NADH, 1.31 mM mg MgClz, pH 7.4, in a total volume of 7.5 mL.
Thedigestwasmixedat 500rpmfor30min.

"° Means with no common superscript are signiﬁcantly diﬂ‘erent (p<0.05).

71

of NADH and generated in situ NADH on the direct desaturation of myristic acid is
shown in Figure 6. The difference between added NADH and that generated in situ
one was not statistically signiﬁcant. In subsequent experiments, NADH was

generated in situ.

3.4.1.6 Eﬂect of bovine serum albumin (BSA)

Free fatty acids (FFAs) are potent non-speciﬁc detergents that inhibit most
enzymatic reactions (Bass, 1985). The latter reported that BSA protects many of the
enzymes that are affected by FFAs, either partially or completely, by complexing the
FFAs and abolishing their detergent properties. The capacity of BSA for tight binding
of long chain fatty acids is well known. Studies of the structural interactions between
FFA and BSA using "c NMR spectroscopy revealed multiple binding sites,
hydrophobic interaction with the hydrocarbon chain, and electrostatic interaction with
the carboxylate anion (Parks et al., 1983; Hamilton et al., 1984; Cistola et al., 1987 a).
Brecher et al. (1984) reported that methyl esteriﬁcation of the carboxyl group of long
chain fatty acids decreased the binding of fatty acids to BSA by 40 to 50 % at pH 7.4.
Results of this study show that the direct desaturation of myristic acid is decreased in
the presence of BSA (Figure 8) when evaluated as described in Figure 7. Speciﬁc
activity, expressed as nmol cis-9-Cl4:1 produced per min per mg protein, decreased
by 12.9% in the presence of BSA. In the absence of the microsomal supernatant,
direct desaturation of myristic acid did not occur whether BSA was present or not
(Figure 8). These observations might be attributed to the tight binding BSA-C14:0

complexes (multiple binding sites) that make myristic acid sterically inaccessible to

72

 

.0
N
or

.0

N

o
I
I

0.15-

0.10-

p

O

01
l

 

 

nmol C14:1/min/mg microsomal protein

 

.0
o

 

 

NADH1 NADH2
Pyridine nucleotide

Figure 6- Eﬂ‘ect of pyridine nucleotide on the direct desaturation of ﬁ'ee myristic acid
by the A9-desaturase system of hen liver microsomes. The desaturation activity was
measured at 37°C under conditions described in Figure l.The complete system
contained 8 mg microsomal protein, 94.5 mg cytosolic protein, 40 mM potassium
phosphate (pH 7.4), 120 mM sucrose, 133 pM ﬁ'ee Cl4:0, 60mg BSA, 1.24 mM
NAD, 3.09 mM sodium lactate, 1.31 mM MgC12, pH 7.4, in a total volume of 7.5 mL.
The digest was mixed at 500 rpm for 30 min. NADHI was generated in situ and
NADH2 was added directly. When NADH was added directly, NAD and sodium
lactate were not added to the incubation medium.

‘Means with no common superscript are signiﬁcantly different (p<0.05).

73

60 mg bovine serum albumin disisolved in 2.4m] distilled (d)H20

V

1,000 nmoles free C1420 i1issolved in IOpL ethanol

V

Incubate for 5 min at room temperature

V

6.2 mg NAD dissolved in 1 ml dHZO
(NAD:C14:O rirole ratio of 9)

v

2.6 mg sodium lactate dissolved in 1 ml deO
(Na lactatezNAD'mole ratio of 2.5)

V

0.94 mg MgClz dissiilved in 0.1ml dHZO

V

Adjust p'H to 7.4

V

Add 3 mL hen liver microsomes suspended in the homogenizing butter
(2.66 mg microsomal protelin/ml homogenizing buffer)

V

Mix the digest at 500rpm for 30 min at 37°C in a 50 ml glass beaker (dia.1.5”)
(using a Lightnin Lab Master)

Figure 7— Assay for the effect of bovine serum albumin and microsomal supernatant
(MS) on the direct desaturation of free myristic acid by the A9-desaturase system of
hen liver microsomes. When absent, the bovine serum albumin solution was replaced
by 2.4 ml deO. When the microsomal supernatant (MS) was used in the assay, the
microsomes were suspended in the microsomal supernatant (2.66 mg microsomal
protein / mL MS) instead of the homogenizing buffer. The MS is the source of
cytosolic proteins and it contains 31.5 mg cytosolic protein (CP) / mL MS.

The ﬁrml volume of the incubation medium was ﬁxed at 7.5 ml to keep the mixing
parameters constant.

74

 

0.20 I I I l

0.15

0.10

0.05

 

 

nmol C14:1/min/mg microsomal protein

.9
o

 

 

 

Solubilizing proteins

Figure 8- Effect of bovine serum albumin and microsomal supernatant on the direct
desaturation of myristic acid by hen liver microsomal A9-desaturase. All incubations
were carried out under the experimental conditions described in Figure 7. In the
absence of bovine serum albumin 2.4 ml distilled water were added to keep the ﬁnal
volume of the assay constant (7.5 ml). The microsomal supernatant is the source of
cytosolic proteins and it contains 31.5 mg cytosolic protein/ml. +BSA indicates the
presence of bovine serum albumin, -BSA indicates the absence of bovine albumin.
+MS indicated the presence of microsomal supernatant, -MS indicates the absence of
microsomal supernatant.

"° Means with no common superscript are signiﬁcantly different (p<0.05).

75

the active site of the A9-desaturase (Figure 9). By contrast, when myristoyl-CoA was
the substrate, BSA increased the A9-desaturase speciﬁc activity by protecting the
substrate ﬁ'om the competing enzyme present in the microsomal supernatant, but not
ﬁ'om the A9-desaturase.

Direct desaturation of myristic acid differs from that of myristoyl-CoA in that
it requires not only a mixing speed greater than 250 rpm but also the microsomal
ﬁaction containing F ABP. Collins and Hargis (1989) reported the presence of FABP
in the cytosolic ﬁaction of chicken liver. Moreover, Scapin et al. (1990) determined

the crystal structure of chicken liver basic F ABP by X-ray crystallography.

3. 4. I. 7 Effect of microsomal supernatant

The omission of the microsomal supernatant containing FABP resulted in loss
of desaturation activity regardless of whether BSA was present or not (Figure 8). The
microsomal supernatant was examined for protein that may be involved in the
regulation of the direct desaturation of myristic acid using heat inactivation and
trypsin digestion. The effect of heat inactivation and trypsin digestion on the capacity
of microsomal supernatant to increase the desaturation activity revealed the
involvement of such proteins. The microsomal supernatant is unstable to heat and
undergoes major denaturation when boiled for 30min. The activity of the microsomal
supernatant proteins was lost after 30 min of boiling when evaluated as described in
Figure 10. At the same time, boiling produced precipitation of the cytosolic proteins.
The desaturase speciﬁc activity was completely lost as a result of heat precipitation of

the cytosolic proteins (Figure 11).

76

 

 

FABP BSA

Ca.)

A9-desatuurase

 

 

 

 

 

 

>‘TJ

 

 

 

Figure 9- Qualitative scheme for the movement of C14:0-protein complexes in the
incubation medium.

* FA: Fatty Acid (in this example the fatty acid is myristic acid).

* BSA: bovine serum albumin

* FABP: fatty acid binding protein

77

3 mL microsomal supernatani containing 31.5 mg CP / mL

V

1,000 nmoles free C1420 ilissolved in lOpL ethanol

V

Incubate for 5 min at room temperature

v

6.2 mg NAD dissolved in lml deO
(NAD:C14:0 mole ratio of 9)

V

2.6 mg sodium lactate dissolved in lml dH20
(Na lactatezNAD|mole ratio of 2.5)

V

0.94 mg MgClz dissiﬂved in0.1 ml dHZO

V

Adjust pIH to 7.4

V

Add 2 mL hen liver microsomes suspended in the homogenizing buffer
(4 mg microsomal proteinI/mL homogenizing buﬁer)

V

Mix at 500 rpm for 30 min at 37°C in a 50ml glass beaker (dia. 1.5”)
using a Lightnin Lab Master Mixer

Figure 10— Heat treatment and trypsin digestion of the microsomal supernatant used in
the direct desaturation of free Cl4:0 by the A 9—desaturase system of hen hepatic
microsomes. The microsomal supernatant was trypsin-digested in one set of
experiments and boiled for 30 min in another set. The ﬁnal volume of the digest was
adjusted to 7.5 ml with 0.39 ml dl-IzO. Bovine serum albumin was omitted from the
incubation medium, since its effect was detrimental to the desaturase activity.

78

 

.0
N
o

.0
...L
‘1”

0.05-

b b
l J

Control Heat Trvosin

 

nmol C14:1/min/mg microsomal protein
C
5.5

 

 

.0
0

Figure 11- Effect of heat treatment and trypsin digestion of the microsomal
supernatant on the direct desattuation of he myristic acid by the A9—desaturase
system of hen hepatic microsomes. All incubations were carried out using the
experimental conditions described in Figure 10.

"° Means with no common superscript are signiﬁcantly different (p<0.05).

79

The eﬂ‘ect of trypsin digestion on the microsomal supernatant proteins is
shown in Figure 11. It was found that proteolysis of the microsomal supernatant
containing FABP and other cytosolic proteins resulted in a complete loss of the
desaturase speciﬁc activity compared to the control (Figure 11). The control
experiment was performed with untreated microsomal supernatant. In contrast to free
myristic acid, heat treatment and trypsin digestion of the microsomal supernatant
increased the speciﬁc activity of the A9-desaturase when myritoyl-CoA was the
substrate. These results suggest the presence of a protein factor in the microsomal
supernatant that activates the direct desaturation of myristic acid under the
experimental conditions descrrhed in Figure 10. The protein factor is rather speciﬁc
as it cannot be replaced by BSA. The question to be considered now is whether this
protein factor binds the free myristic acid. If so, then the substrate for the direct
desaturation by the hen microsomal A9-desaturase would be in the form of a myristic

acid:binding protein complex.

3. 4. I .8 Eﬂect of a-bromohexadecanoate

It has been established that a-bromohexadecanoate competes with free fatty
acids for the same binding site of fatty acid-binding proteins (Ockner and Manning,
1976). The eﬂea of a-bromohexadecanoate on the direct desaturation of myristic
acid by hen hepatic microsomal A9-desaturase was examined using different mole
ratios of a-bromohexadecanoate: myristic acid (Figures 12 and 13). The linear
decrease in speciﬁc activity observed with free myristic acid as a substrate in the

presence of varying concentrations of or-bromohexadecanoate probably reﬂects the

80

3 ml of hen liver mitochondrial supernatant
(containing 2.66 mg microsomal protelin and 31.5 mg cytosolic protein / mL)

V

1,000 mnoles of free C14:0 dissolved in lOpL ethanol
1

Add a-bromohexadecanoate
(1000, 2000, and 3000 nmo|les dissolved 10 pL ethanol)

V

6.2 mg NAD dissolved in lml deO
(NAD:C14:0 rirole ratio of 9)

V

2.6 mg sodium lactate dissolved in lml dHZO
(Na lactatezNAD'mole ratio of 2.5)

V

0.94 mg Mgcr2 dissclilved in 0.1m1 dH20

V

Adjust pIH to 7.4

V

Mix at 500 rpm for 30 min at 37°C in a 50m] glass beaker (dia. 1.5”)
using a Lightnin Lab Master Mixer

Figure 12- Assay for the effect of a-bromohexadecanoate on the direct desaturation
of {ice myristic acid by the Ag-desaturase system of hen liver rnicrosome. The ﬁnal
volume of the incubation was adjusted to 7.5 ml with 2.38 ml dHZO. Bovine serum
albumin was not added due to its detrimental effect on the desaturation yield.

81

 

x c14:0 CoA Y= 0.0120x2 - 0.0544x + 0.42 (R22 0.99)
- 014:0 Y= - 0.05x + 0.33 (R’= 1)

 

 

 

 

0-5 l I l j l I

 

.0
N

 

 

 

 

nmol C14:1/min/mg microsomal protein
0
0)

0 1 L L l l l l

'0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5
a—bromohexadecanoate:substrate mole ratio

Figure 13- Effect of a-bromohexadecanoatezsubstrate mole ratios on the direct
desaturation of myristic acid by the A9-desaturase system of hen hepatic microsomes.
The desaturation activity was measured under conditions described in Figure 12.
When myristoyl-CoA was the substrate, 1000 nmol myristoyl-CoA were dissolved in
200pl distilled water.

“I Means with no common superscript are signiﬁcantly different (p<0.05).

82

reduction of myristic acid binding to FABP rather than the inhibition of the
desaturation reaction per se. This study revealed that a—bromohexadecanoate
concentrations did not affect the desaturation of myritoyl-CoA under the experimental
conditions reported in Figure 12.

These results suggest that the speciﬁc protein factor in the microsomal
supernatant is a F ABP because it binds free fatty acid and the binding was completely
inhibited by a-bromohexadecanoate. These results also suggest that the substrate for
the direct desaturation by the hen microsomal A9-desaturase is in the form of a
myristic acid:FABP complex. The decrease in the desaturase speciﬁc activity was
directly related with the decrease in concentration of the myristic acid:FABP
complex. Previous studies reported that hepatic fatty acid binding protein (hFABP)
may speciﬁcally direct the utilization of fatty acids toward either esteriﬁcation (Wu-
Rideout et 1976; Iritani et al., 1980) or oxidative paths(Appelkvist et al., 1980).
Bass (1985) suggested that hFABP effects the transfer of fatty acids to the active site
of the enzyme. Wu-Rideout et al. (1976), Ockner and Manning (1976) found that
albumin was an ineﬂ‘ective replacer of hFABP in stimulating the activity of some
microsomal enzymes utilizing fatty acids. This suggested the role of FABP in
providing fatty acids to their enzymes may be more than a substrate solubilizer.
Information in Figures 8 and 9 reveals that FABP cannot be replaced by BSA due to
the tight binding of BSA-C14:0 complex (multiple binding site) (Parks et al., 1983;
Hamilton et al., 1984; Cistola et al., 1987 a) rmking the C14:0 inaccessible the active
site of the A9-desaturase. The FABP-fatty acid complex has one binding site (Glatz et

al., 1983), thus allowing easier access of C14:0 to the enzyme active site (Figure 9).

83

Glatz et al. (1983), using a radiochernical procedure, reported that FABP

possesses only one binding site per protein molecule for long-chain fatty acids.

Cistola et al.(l987a) using 13C NMR spectroscopy reported that oleic acid binds
hFABP through the carboxyl group. Moreover, these investigators suggested that the
anionic carboxyl group of oleic acid bound to the hFABP experiences only one type -
of binding environment (the aqueous milieu adjacent to the protein surface) free of
speciﬁc ion-pair electrostatic attractions with cationic residues on hFABP. Brecher
et al. (1984) found that the incubation of hFABP with labelled oleate reached an
apparent equihhrium at a fatty acid to protein mole ratio of about 1. Moreover, they
reported that the methyl esteriﬁcation of the carboxyl group of long chain fatty acids
inhibit binding of the fatty acids to FABP. These ﬁndings lead to the conclusion that
the FABP may regulate the accessibility of ﬁatty acids in the desaturation reaction to
the active site of the desaturase rather than just protecting the membrane-bound

desaturase from the cytotoxic effect of ﬂee fatty acids.

3.4. 1.9 Effect of mixing rate

The effect of mixing rate on the desaturation of the myristic acid:FABP
complex by hen hepatic microsomal A9-desaturase was examined using mixing
speeds from 0 to 1800 rpm. Under the conditions outlined in the assay procedure
(Figure 14), no desaturation was detected in the absence of mixing despite the
presence of the microsomal supernatant containing FABP. This is in agreement with

previously reported research (Masoro, 1968; Raju and Reiser, 1972; Cook and

Spence, 1973; Pugh and Kates, 1 973; Brenner, 1974; Catala et al., 1975; Cook, 1978;
Voet, 1990) where no evidence was found for the direct desaturation of ﬁ'ee fatty
acids without being activated to the CoA derivatives. Production of myristoleic acid
was observed at mixing speeds greater than 250mm. The optimal mixing rate for
maximum product formation was 500 rpm (Figure 15). The decline in myristoleic
acid production at mixing speeds greater than 500 rpm could be due to the rupturing
of the microsomal membrane.

To investigate the presence of endogenous myristoyl-CoA, an assay was
performed as described in Figure 14, but with no substrate added. Any formation of
product under these conditions would be due to an endogenous source of myristoyl-
CoA. Myristoleic acid formation was not observed under these conditions (Figure
16). The possibility that the myristic acid was activated to its CoA derivative prior to
the desaturation reaction in the presence of a thiokinase, CoA, and ATP in the
microsomal supernatant is unlikely in view of the following observations: (1)
desaturation was not observed at mixing rates lower than 250 rpm despite the
presence of the microsomal supernatant, knowing that the activation reaction does not
require mixing of the incubation medium; and (2) in the absence of BSA, the
microsomal supernatant itself was detrimental to the desaturase activity when
myristoyl-CoA was the substrate (2.4.1.8).

Different mixing rates were used to facilitate the transfer of the myristic
acid:FABP complex (substrate) to the active site of the desaturase by direct collision

to form productive enzyme-substrate complexes. Kim and Storch (1992) showed that

85

3 ml of hen liver mitochondrial supernatant
(containing 2.66 mg microsomal protelin and 31.5 mg cytosolic protein / mL)

V

1,000 nmoles of ﬁee C141; dissolved in 10pL ethanol

V

6.2 mg NAD dissolved in lml dHZO
(NAD:C14:0 rirole ratio of 9)

V

2.6 mg sodium lactate dissolved in lml deO
(Na lactatezNADlmole ratio of 2.5)

V

0.94 mg MgClz diSSitlved in 0.1m] deO

V

Adjust pIH to 7.4

V

Mix at 500 rpm for 30 min at 37 °C in a 50ml glass beaker (dia. 1.5”)
using a Lightnin Lab Master Mixer

Figure 14- Final assay for the direct desaturation of myristic acid:FABP complexes

by the A9-desaturase system of hen liver microsomes in the absence of CoA and ATP.
The ﬁnal volume of the assay was adjusted to 7.5 ml with 2.39 ml distilled water.

86

 

 

 

 

 

c I I I I I I
E 0-4 . I I I I I I
e C
o.
E

_ —J
o 0.3
(D
2
.9
E
a) 0 2 — _‘
E
.E
E
v- b
,3; 0.1 -
,_
0
"g x

a a I
c 0.0 ‘ ' m 1 ‘

0.0 50 250 500 1000 1500 1800

Mixing rate (rpm)

Figme 15- Effect of mixing rate on the direct desaturation of myristic acid: FABP
complexes by the A9-desaturase system of hen hepatic microsomes. The desaturation
activity was measured under conditions described in Figure 14 using different mixing
speeds. The complete system contained 8 mg microsomal protein, 94.5 mg cytosolic
protein, 40 mM potassium phosphate (pH 7.4), 120 mM sucrose, 133 pM ﬂee C14:0,
1.24 mM NAD, 3.09 mM sodium lactate, 1.31 mM MgC12, pH 7.4, in a total volume
of7.5 mL. The digest was mixed at 37°C for 30 min.

"° Means with no common superscript are signiﬁcantly different (p<0.05).

87

 

 

 

 

 

E
E3 03 I I I T F I I I I
2
0.
"<6
E
8
E 0.2~ -
.9
E
O)
E
.E
g 0.1e -
g
0
'6' .
g 00 I 3 l l I I I
0 133

Substrate concentration (pM)

Figure 16- Effect of substrate concentrations on the direct desaturation of myristic
acid:FABP complexes by the A9-desaturase system of hen liver microsomes. The
desaturation activity was measured at 37°C under conditions described in Figure 14
with different substrate concentrations. The complete system contained 8 mg
microsomal protein, 94.5 mg cytosolic protein, 40 mM potassium phosphate (pH 7.4),
120 mM sucrose, 1.24 mM NAD, 3.09 mM sodium lactate, 1.31 mM MgC12, pH 7.4,
in a total volume of 7.5 mL. The digest was mixed at 500 rpm for 30 min.

"° Means with no common superscript are signiﬁcantly different (p<0.05).

88

the rate of transfer of the ﬁee fatty acid from hFABP to egg phosphotidycholine
membranes is inﬂuenced by both the structure of the ﬁee fatty acid ligand and the
properties of the aqueous phase (pH, ionic strength, temperature) through which the
ﬁ'ee fatty acid must travel. Working with puriﬁed stearyl-CoA desaturase and acyl-
CoA derivatives as substrates, Enoch et al. (1976) reported that free CoA and free
fatty acids did not bind to the A? -desaturase enzyme and did not inhibit the
desaturation reaction. Brudett and Zeikus (1994) reported that "the secondary alcohol
dehydrogenase provides an NADPH-dependent route to ethanol from acetyl CoA and
acetyldehyde ". They did not use mixing.

The data presented here lead to the conclusion that when free myristic acid is
bound to FABP in the form of protein-monomer complexes, its activation to the CoA
derivative is not necessary in order to be desaturated by the A9-desaturase when using
a mixing rate greater than 250 rpm. Myristic acid:FABP complexes probably serve as

substrates for the [lg-desaturase at high mixing rates.

3. 4. 2 Substrate speciﬁcity studies
The variation in molecular structure between fatty acyl CoA and the free fatty
acid gives rise to variation in reactivity. The only difference between the two

molecules is the CoA esteriﬁed to the acyl group.

II
R-C-OH Free fatty acid

89

O H H OH CH3 0

II I I l I II
R - c - s -(cr12)2 - N - C - (c1192 - N -C- C - c - CH2- 0 - P - o -(CH2)-Adenine

II II I I I
o ' OH CH3 0'

Fatty acyl CoA
This observation might be explained by the guiding and reaction side hypothesis:
“The formation of productive enzyme-substrate complexes occurs if S1 is
complementary in shape, size, and chemical nature to E I , and 52 when needed, is

properly guiding S1 to E1 without any steric hindrance

Let us consider S = S1+ 82 and E = E1+ E2(Figure 17)
S : substrate
8]: reaction side, determines the type of products.

82: guiding side, determines the reactivity.

E : enzyme

E1: active site, determines the type of reaction

E2: non-catalytic side, maintains the enzyme in its tertiary structure.

The guiding side (S2) affects reactivity in two general ways:

(1) by the eﬂect of long range forces between SZ---E2 (mainly electrostatic and ion-
dipole interaction) that facilitate the traﬁcking of the reaction side (S 1) toward the

active site (El) and that activate the short range forces (London-van der Waal’s

90

 

E2

52.

 

 

Figure 17- An enzyme-substrate complex illustrating both the geometrical and the
physical complementarity between active site (E1) and reaction side (S 1 ). Where

E1 is the active site of the enzyme, E2 is the no-catalytic side of the enzyme, S1 is the
reaction side of the substrate, and SZ is the guiding side of the substrate.

91

dispersion) between S1---El to induce a precise ﬁt; and (2) by its steric effect which
prohibits the formation of active site-reaction side (El—S 1) complexes , which lead
to the formation of product. The steric effect of 82 is dependent on the conﬁguration

of E2. E2 is not a dead weight, it plays an important in the catalytic activity of B.

We investigated several scenarios to verify the validity of this hypothesis.

First scenario: fatty acyl C oA derivative and Ag-desaturase

S :C14:0CoA (substrate) E : A9-desattrrase
S1: C14:0 (reaction side) El : active site
S2: CoA (guiding side) E2 : non-catalityc side

The transfer of C14:0 ($1) to the catalytic domain of the A9-desaturase (E1)
was accomplished via direct collision interaction through the electrostatic attractions
between the water-soluble CoA moiety (S2) and the hydrophilic part (E2) of the
membrane bound A9-desaturase which exposes the catalytic domain to the aqueous
environment. Enoch and Strittmatter (1978) reported that positively charged arginyl
residues of the stearyl-CoA desaturase may be involved in the binding of the CoA
moiety of the negatively charged substrate, as the substrate (stearyl-CoA) has three
phosphate groups and bears three to four negative charges.

The guiding side CoA (S2) facilitates the trafﬁcking of the hydrophobic
myristic acid (S 1) to traverse the aqueous phase to reach the membrane bound A9-
desaturase. There was no steric effect observed with CoA. Myristoyl CoA did meet
the conditions of the proposed concept. The A9-desaturase speciﬁc activity expressed
as nmol C14:1/min per mg microsomal protein was 0.50. Enoch et al. (1976) showed

that liposome bound stearyl-CoA is the true substrate for the puriﬁed Ag—desaturase

92

in the absence of BSA. Bound stearyl-CoA undergoes lateral diffusion in the bilayer
to reach the active site of the enzyme. However, in the presence of BSA, the
possibility that myristoyl-CoA is reaching the enzyme by lateral diffusion is unlikely
due to the ability of the albumin to remove the fatty acyl-CoA from the liposomes.
The transfer of substrate by direct collision is more likely.

Lichtenstein et al. (1982) reported that the addition of albumin to unilamellar
vesicles at a molar ratio equivalent to that of oleoyl CoA effectively removed oleoyl
CoA ﬁ'om the vesicle and result in oleoyl CoA-albumin complex. Jeﬂ‘coat et al.
(1977) reported that albumin is not necessary for the desaturatioon reaction per se.
The role of albumin is to protect the acyl CoA derivatives from the acyl transferase
present in the cytosol. In agreement with the reported literature, our data show that
BSA is not necessary for the desaturation myristoyl-CoA by hen liver microsomal A9-

desaturase (Figure 9, Chapter 2).

Second scenario: free fatty acids and 219- desaturase

S : Cl4:0 E : 119-desaturase system
S1: C14:0 (reaction side) E1: active site
S2: absent (guiding side) E2: non catalytic side

Myristic acid as a ﬁ'ee monomer did not meet the conditions of the
hypothesis. The desaturation reaction did not occur because the guiding side (S2) was
absent. The reaction is substrate-transfer limited, instead of being limited to the

kinetics of the reaction. However, when the microsomal supernatant containing

93

FABP, was added to the incubation medium coupled with a mixing speed of 500 rpm,
the speciﬁc activity observed was 0.268 nmol/min per mg microsomal protein. In this
system, neither CoA nor ATP was added. Free myristic acid binds to FABP to form a
myritic acid-FABP complex which ultimately dissociates in the aqueous phase.
Brecher et al. (1984) reported that when exposed to protein (BSA or FABP), 50% of
the fatty acids contained within liposomes could become protein-bound, and the
newly formed fatty acid-protein complex dissociates ﬁom the bilayer surface to the
aqueous space. They also reported that the movement of fatty acid from protein to
liposomes is a much slower process.

The possibility that myristic acid reaches the active site of the Ag-desaturase
by lateral diffusion in the bilayer is unlikely due to the effect of binding proteins.
The transfer of myristic acid to the [Ag-desaturase catalytic domain was accomplished
via random collision between the myristic acid—FABP complexes and the membrane
bound A9-desaturase. The random collisions were generated using diﬁerent mixing
rates. A mixing rate of 500mm was necessary to overcome diffusion obstacles due to
the properties of the aqueous phase (pH, ionic strength, temperature), through which
the myristic acid-FABP complexes has to travel. The desaturation reaction did not
occur in the absence of FABP despite the use of different mixing rates. These results
suggest that F ABPs play a key role in the formation of enzyme substrate complexes
probably via a physical effect on the myristate.

The collision between myristic acid—FABP complex and the membrane-bound
Ag-desaturase activates the short-range forces between the myristic acid (S l) and the

active site (E1) of the desaturase, therefore inducing a precise ﬁt. The short-range

94

forces are London Van-der-Waal’s dispersions and weak polar attraction due to the
polarizable , 1r electrons of the carbonyl group. In this case, the FABP is the guiding
side (82) and myristic acid is the reaction side (S1). This observed effect was speciﬁc
for FABP and could not be duplicated by albumin (Figure 8), giving more weight to
the argument that the effect of FABP is speciﬁc in regulating the desaturase activity. .
The reactivity of the reaction is determined by the nature of the guiding side (82).
Under the experimental conditions used, the guiding side CoA was more reactive than

the guiding side F ABP since speciﬁc activity of the desaturase was higher

(Figure 18).

Third scenario: trimyristin and A9-desaturase system

s : trimyristin E : A9-desaturase
Sl: C14:0 (reaction side) E1: active site
S2: glycerol backbone (guiding side) E2: non catalytic side

The desaturation reaction did not occur even in the presence of FABP and the
use of different mixing rates to overcome diﬂirsion obstacles due to the properties of
the aqueous phase (pH, Ionic strength, temperature) through which the substrate must
travel. This lack of desaturation might be due to the steric eﬂ‘ect exerted by the
glycerol backbone (S2) on (E2). Trirnyristin did not meet the conditions of the
proposed hypothesis and thus no desaturation was observed. The same effect was
observed with l-monomyristoyl-rac-glycerol (C14:0), 1,2 dimyristoyl-rac-glycerol

(C14:0), myristic acid methyl ester, and L-a-phosphotidyl choline dimyristoyl

95

(C14:0) (Figure18).

3. 4. 3 Desaturation of milkfat

As the A9-desaturase is inactive on triacylglycerol substrates, the application
of the desaturation process (outlined in Figure 14) on milkfat necessitates a lipase-
catalyzed hydrolysis of milkfat triacylglycerols to liberate the ﬁee fatty acids (FFAs)
prior to the desaturation reaction and ﬁnally re-esterify the FFAs. They would not all
return to their original positions in the triacylglycerol moiety, hence the reaction of

the ﬁnal material to milkfat would be somewhat tenuous.

3.5 CONCLUSION

The current research effort was directed at understanding the regulation of the
A9-desaturase system of hen liver microsomes using myristoyl-CoA and myristic
acid:FABP complexes as substrates. Experimental evidence has been presented to
show that the variation in molecular structure between myristoyl-CoA and myristic
acid:FABP complexes gives rise to variation in reactivity. CoA and FABP are
guiding molecules. There are positive inﬂuences pulling reactants, but negative
repulsion plays a major role in the way reactants are guided. Attraction and repulsion
can act over short-and long-ranges. Short-range requires contacts with membrane
bound Ag-desaturase, and long-range interactions occur by diﬂusible factors that set
up gradients. Unraveling the mechanism of these interactions is complicated by the

fact that guidance molecules can have biﬁrnctional roles. The same molecule can be

96

O

O

O)
—-I
—-I
—-r
—l
—
-l
—

 

.0
or

.0
4x

.0
N

P
.3

 

 

 

b b b b
I l I I

 

nmol C14:1/min/mg microsomal protein
0
OD

.0
o

0 Q, Ix . x. . x. . x. . x.
0" six 9 e
0‘ OKs-Q OKs-“0° ostegsf a 39*“ ,\ 30““)

Substrates

Figure 18- Effect of different substrates on the speciﬁc activity of the hen microsomal
Ail-desaturase system. The desaturation activity was measured at 37°C under
conditions described in Figure 14 using different substrates. The complete system
contained 8 mg microsomal protein, 94.5 mg cytosolic protein, 40 mM potassium
phosphate (pH 7.4), 120 mM sucrose, 1.24 mM NAD, 3.09 mM sodium lactate, 1.31
mM MgC12, pH 7.4, in a total volume of 7.5 mL. The digest was mixed at 500 rpm
for 30 min.

"° Means with no common superscript are signiﬁcantly different (p<0.05).

C14:0 : Free myristic acid

C14:0CoA : Myristoyl coenzyme A

T.myrst : Trirnyristin

D.myrst : 1,2 dimyristoyI-rac-glycerol (C14:0)

M.myrst : l-monomyritoyl-rac-glycerol (Cl4:0)
C14:0ME : Myristic acid methyl ester

P.myrst. : L-a- phosphotidyl choline dimyristoy1(Cl4:0)

97

attractive or repulsive, short- and long-range.

The frustrations of designing enzyme inhibitors are nothing new. This study
shed some light on the effect of the substrate guiding side (S2) on the reactivity of
enzymatic reactions. Every enzyme has an inhibitor and may be with a better
understanding of the role of the substrate guiding side in enzymatic reactions

researchers will be able to ﬁnd a more efﬁcient way to design enzyme inhibitors.

98

CHAPTER 4

REDUCTION OF THE HYPERCHOLESTEROLEMIC EFFECT

OF ANHYDROUS MILKF AT

4.1 ABSTRACT

Alternatives to trans-rich fats are currently being sought that will allow a
return to animal fat without eliciting deleterious health effects. Removing cholesterol
from anhydrous milkfat to minimize its cholesterol—raising effect, has produced a viable
alternative to native milkfat and trans-rich margarine spreads. A process to modify
anhydrous milkfat to decrease its hypercholesterolemic effect was developed using
B-cyclodextrin, KOH, CaClz, and mixing. The process was optimized by response
surface methodology based on cholesterol and ﬁee fatty acid (FFA) reduction.
Reductions in cholesterol and FFAs with this process were 95.8 and 92.4%, respectively.
Gerbil feeding studies showed that modiﬁed milkfat (i.e., cholesterol-reduced milkfat)

reduced serum total cholesterol by 28% and triacylglycerols by 47%.

99

4.2 INTRODUCTION

Multiple lines of evidence indicate that saturated fatty acids and cholesterol raise
plasma cholesterol levels. Both animal (Hegsted and Gallagher, 1967) and human studies
(Keys et al., 195 7) have demonstrated elevation in serum cholesterol levels following
increased saturated ﬁt intake. These observations, coupled with epidemiological
evidence of increased risk of coronary heart disease (Cl-ID) in populations consuming
excessive amounts of saturated ﬁts and cholesterol, explain the emphasis on the
reduction of saturated fat and cholesterol by most health agencies.

Milkﬁt plays an important role in the structure, mouthfeel, ﬂavor and stability of
dairy products and a host of foods in which milkﬁt is added as a functional ingredient.
However, milkﬁt has a high proportion of hypercholesterolemic saturated fatty acids and
cholesterol which has resulted in decreased consumption of dairy products (USDA,
1997).

The ﬁt and oil industry is continuously looking at new techniques to alter the
composition and ﬁmctionality of ﬁt molecules. Several methods for improving the
nutritional and physical properties of milkfat have been developed over the past few
years including changing the fatty acid composition by altering the diet of cows (Banks et
al., 1990). Feeding protected soybean oil to cows has resulted in a 3.6 fold increase in
linoleic acid (Banks et al., 1990) compared to the values obtained on feeding a basal diet.
Milkfat ﬁactionation using physical processes such as short-path distillation (SPD), melt
crystallization, and supercritical ﬂuid extraction has been applied in conjunction with
cholesterol extraction. Arul et al. (1988) fractionated milkﬁt by SPD into four ﬁactions

at temperatures of 245 and 265 °C and pressures of 220 and 100 pm Hg. However, the

100

use of high temperature might decompose or polymerize the triacylglycerols even when
the milkﬁt is distilled under vacuum. Makhlouf et al. (1987) ﬁactionated milkfat into 7
fractions by melt crystallization at temperatures ranging from 26 to 9 °C. The rmjor
drawback of this technique is the signiﬁcant overlapping of molecular weights of
triacylglycerols. Bhaskar et al. (1993) fractionated milkﬁt using super critical ﬂuid ,
extraction into 5 different fractions (S1 - S5) with differing cholesterol levels at 3.5 -
24.1 MPa and 40 - 60 °C.

Dietary cholesterol is hypercholesterolemic (National Academy of Sciences,
1989b) and prone to oxidation. Increased attention has been focused on speciﬁc effects
associated with certain cholesterol oxidation products and the need to eliminate or
prevent their formation in the human diet. Peng et al. (1979) reported that oxides of
cholesterol are toxic and cause degeneration of aortic smooth muscle cells in tissue
culture, and may lead to the development of atherosclerosis (Peng et al., 1991).
Epidemiological studies further suggest that dietary cholesterol increases the risk for
coronary heart disease beyond its serum cholesterol-raising effect (Stamler and Shekelle,
1988).

Several investigators have used cyclodextrin (CD) to reduce the cholesterol levels
in animal products. CD are oligosaccharides, consisting of seven molecules of glucose
units arranged in a donut shaped ring (Szejtli, 1982a,b). Courregelongue et al. (1989)
used B—CD to remove 41 % of the sterols from anhydrous milkﬁt. Bayol et al. (1989)
reported the removal of 80 % A4-cholesten-3-one from anhydrous milkﬁt, while

Roderbourg et al. (1993) demonstrated that at least 37% of the cholesterol could be

101

removed from anhydrous milkﬁt in one single step. Graille et al. (1991) , , reported the
simultaneous removal of 50% cholesterol and 52 % F FAs ﬁom cream. Okenfull et al.
(1991) invented a new process called SIDOAK to reduce cholesterol in dairy products.
The process consisted of adding CD to milk and mixing below 10°C. The maximum
cholesterol reduction obtained was 80-90 %. Yen and Tsai (1995), using B-CD, indicated
the removal of 95 % cholesterol simultaneously with 50 % FFA ﬁ'om lard.

The objectives of the present study were:

1) to develop a process to effectively remove cholesterol and F F As ﬁ'om milkfat;

2) to test the effect of cholesterol-reduced milkﬁt on blood lipids
4.3 MATERIALS AND METHODS

4. 3. 1 Materials

Commercial grade unsalted butter was obtained ﬁ'om a local food store and
converted into anhydrous milkfat (AMF) by melting at 55°C, centriﬁrging at 5000 x g for
10 min at room temperature, and ﬁltering the top ﬁt layer through Whatman No. 1 ﬁlter
paper. The AMF was stored at -20°C for ﬁxture treatments. Potassium hydroxide and
calcium chloride were purchased from VWR Scientiﬁc Products (Batavia, IL), while

B—CD was obtained ﬁ'om Cerestar (Hammond, IN).

4. 3. 2 Preliminary process
After several exploratory experiments, a preliminary process was
developed. AMF (10g) containing 0.395% cholesterol and 0.293% FFA was mixed in a

SOmL beaker with 10mL of a reﬁning solution (0.0583 % (w/v) KOH and 0.763 % (w/v)

102

CaClz) at 1000 rpm using a Lightnin Lab Master (SI) mixer (model L1U03) and an
impeller (type A100) (Matoon and Lee Equipment, Farmington, MI) until the

temperature reached 50°C using a controlled temperature water bath. B—CD was added to
the mixture to provide a B—Cchholesterol mole ratio of 5.58. Mixing was continued for
another 10 min. The resulting soap and B—CD :cholesterol complexes were immediately
centrifuged at 8700 x g for 10 min at room temperature using a Sorvall RC 2-B centrifuge
(Sorvall Inc., Norwalk). The modiﬁed AMF, i.e., reduced cholesterol and FFAs, was
recovered in the upper supernatant phase. Percentage cholesterol and F FA reductions

were 54 and 92 %, respectively.

4. 3. 3 Experimental design

When many factors and interactions affect desired responses, response surﬁce
methodology (RSM) is an effective tool for optimizing the process (Hunter, 1959). Thus,
the optimization of cholesterol and FFA reduction was accomplished using response

surface methodology.

Two responses were measured: percentage cholesterol reduction (Yr) was
deﬁned as the ratio of the total amount of cholesterol in the sample to the total amount of
cholesterol in the control multiplied by 100. Percentage FFA reduction (Y2) was deﬁned
as the ratio of the total amount of FFAs in the sample to the total amount of FFAs in the
control multiplied by 100. Based on preliminary experiments, the ﬁve independent
variables to be evaluated were KOH:F FA mole ratio, CaCIzzFFA mole ratio, mixing
time, B—CchhoIesterol mole ratio, and mixing rate (Table 1). The other important

independent variables were held ﬁxed: temperature and centrifugation speed of 8700 x g

103

Table 1- Variables and their levels for central composite design used in optimizing the
process for reducing cholesterol and free fatty acids in anhydrous milkﬁt.

 

 

 

Coded-variable levels
Variable Symbol -2 -l 0 + 1 +2
KOHzFFAs (nele ratio) X1 0 0.5 1 1.5 2
CaC12:FFAs (rmle ratio) X2 0 2.5 5 7.5 10
Mix'ng tine (nin) )0 5 7.5 10 12.5 15
CDichobsterol (rmle ratio) X4 4.72 5.15 5.58 6.01 6.4
Mixing rate (rpm) X5 200 600 1000 1400 1800

 

for 10 min at room temperature.

The experimental design adopted was a S-factor, 5-level central composite
(Box and Hunter, 1957). The coded values of the independent variables were -2 (lowest
level), - 1.0 (medium level), 1.0, and 2 (highest level). For each independent variable
studied, the central value (0) was chosen according to the preliminary explanatory
preliminary experiments. The complete design shown in Table 2 consisted of 32

experimental points which included 6 replications of the center (0, 0, 0, 0, 0).

4. 3. 4 Extraction and determination of total cholesterol

Total cholesterol was extracted and quantiﬁed according to the AOAC Ofﬁcial
Method 994.10 (1995). Total cholesterol was quantiﬁed in underivatized form using a
Hewlett Packard 5890A gas chromatograph (GC) (Avondale, PA) equipped with a ﬂame
ionization detector and a fused silica capillary column (15 m x0.25mm i.d.) (J & W

Scientiﬁc, Folsom, CA) coated with a 0.1 pm ﬁlm of DB-l (100% methyl polysiloxane).

4. 3.5 Quantitation of free fatty acids (FF/1s)

FFAs were quantiﬁed according to the AOAC Oﬂicial Method 969.17
(1995)

4. 3. 6 Determination of the hypercholesterolemic effect of anhydrous milkfat
Feeding studies on gerbils to determine the hypercholesterolemic eﬂ‘ect of
anhydrous milkfat were done according to the protocol described by Hajri et al. (1998).

Adult male Mongolian gerbils (Meriones unguiculatus) weighing 75:5 g were housed in

l05

Table 2- Central composite design arrangement and responses used in optimizing the
process for reducing cholesterol and free ﬁtty acids in anhydrous milkfat.

 

 

 

 

 

Variable levels' Responsesb
Run X1 X2 X3 X4 X5 Y1 Y2
l -1 -1 -1 -1 1 61.78 69.53
2 -1 -1 -l 1 -1 28.58 58.07
3 -1 -1 1 -1 -1 46.46 68.06
4 -1 -1 1 1 1 77.3 75.3
5 -1 1 -1 -1 -1 38.53 69.16
6 -1 1 -1 I l 63.66 79.12
7 -l 1 1 -1 l 68.22 77.6
8 -1 1 1 1 -1 40.5 70.79
9 1 -1 -1 -l -1 20.41 92.42
10 1 -1 -1 1 1 65.41 92.28
11 l -1 l -1 1 65.98 92.38
12 1 -1 1 1 -1 42.84 92.21
13 1 1 -1 -1 1 54.28 96.16
14 I 1 -l l -l 38.21 90.36
15 1 1 1 -1 -1 32.28 92.35
16 1 1 1 1 1 74.23 92.48
17 -2 0 0 0 0 59.98 31.35
18 2 0 0 0 0 53.52 92.33
19 0 -2 0 0 0 38.03 91.92
20 0 2 0 0 0 53.32 92.38
21 0 0 -2 0 0 37.49 92.24
22 0 0 2 0 0 63.64 92.63
23 0 0 0 -2 0 46.85 92.64
24 0 0 0 2 0 55.7 92.45
25 0 0 0 0 -2 1 1.91 81.6
26 0 0 0 0 2 73.7 92.61
27 0 0 0 0 0 52.27 92.45
28 0 0 0 0 0 56.27 92.22
29 0 0 0 0 0 53.02 92.53
30 0 0 0 0 0 54.48 92.1
31 0 0 0 0 0 54.94 92.45
32 0 0 0 0 0 54.02 91.93
' Coded variables

b
Y1 is cholesterol reduction (%). Y2 is FFA reduction (%)

individual cages in an air-conditioned room with a 12-h reversed light-dark cycle (lights
on at 18:00h). The gerbils were assigned randomly to 6 groups (n=6 per group) and were
fed two types of ﬁts for 4 weeks: native anhydrous milkﬁt (control) and modiﬁed
milkﬁt. Fat provided 40% of the total dietary energy. The basal composition of these
diets (g/ 100g of total dry weight) was as follows: 22.2 casein, 23 cornstarch, 13.3
glucose, 15.0 cellulose, 20 fat, 5.0 mineral mix (Ausman-Hayes), 1.2 vitamin mix
(Hayes-Cathcart), and 0.3 choline chloride. At the end of the feeding study, each gerbil
was exsanguinated under anesthesia, the livers weighed and inspected for cholesterol, and
plasma collected for total cholesterol, LDL-C, HDL-C, and plasma triacylglycerols. The
feeding studies were done at Brandeis University under the direct supervision of Dr. K.C.

Hayes.

4. 3. 7 Statistical analysis

Statistical analyses of the cholesterol and the FFA data were performed using software
packages to ﬁt the second order polynomial equation to the experimental data (SAS

Institute, 1990). The model proposed for each response (Y) was:

2
Y = 13 + 213x. + 23.x. + 213..x.x.
o 1 1 11 1 13 1 3

where Bo, Bi, B“, Bij are constants and regression coeﬂicients of the model, and Xi the

independent variables in coded values. The criteria for eliminating a variable ﬁom the

ﬁll] regression were based on R2 values and the signiﬁcance of the F-test.

107

4.4 RESULTS & DISCUSSION
4. 4. 1 Process development and optimization

Response surface methodology was used to optimize the preliminary process for
cholesterol and FF A reduction in anhydrous milkfat. Analysis of variance (AN OVA) for
the regression was performed to test the goodness of ﬁt. Results of the analysis of
variance of each of the responses with their corresponding coefﬁcient of multiple
determination (R2) and coefﬁcient of variation (CV) are shown in Table 3. From the
analysis of the F values and R2, only four independent variables (KOHzFFA mole ratio,
mixing time, CD:cholesterol mole ratio, and mixing rate) out of ﬁve appeared to
inﬂuence cholesterol reduction (Y1). In ﬁct, despite the exclusion of CaClz ﬁom the
model, the R2 remained high (095) and the F-test signiﬁcant to <0.001. The estimated
regression coemcients for Y. and the results of the signiﬁcance test on the coeﬁcient for
the four selected factors are indicated in Table 4. Also, ﬁom the analysis of the F values
and R2, only three independent variables (KOH:F FA mole ratio, CaClzzFFA mole ratio,
and mixing rate) out of ﬁve appeared to inﬂuence FFA reduction (Y2). The exclusion of
mixing time and CD did not shoot the model, since the R2 remained high (0.96) and the
F—test was signiﬁcant to <0.001. The estimated regression coeﬁcients for Y2 and the
results of the signiﬁcance test on the coeﬁcient for the three selected factors are
indicated in Table 5. None of the models exhibited lack of ﬁt. Both models were
considered adequate and usable with satisﬁctory R2 values and a signiﬁcant F-test at
<0.001. The predicted Y1, Y2, and the residual values using the model are reported in

Tables 6 and 7.

108

Table 3- Parameter estimates and analysis of variance of the second order polynomial

regression for ﬁve factors and two responses (cholesterol (Y1) and F A (Y2) reduction)

 

 

 

Y1 Y2

Source Coeﬂicient prob>l Tl Coeﬂicient prob>| Tl
Intercept - 18.66 0.91 29.09 0.87
X1 -139.81 <0.01 160.62 <0.01
X2 12.67 0.07 -3.75 0.06
X3 10.3 0.14 -4.98 0.5
X4 9.97 0.85 - 13.76 0.82
X5 -0.01 0.68 0.06 0.16
X1 *X1 3 .8 0.26 -30.68 <0.01
X2*X2 -0.29 0.04 -0.01 0.92
X3*X3 -0.09 0.48 <-0.01 0.98
X4*X4 -2.25 0.63 0.14 0.97
X5*X5 <-0.01 0.01 <-0.01 0.16
X1*X2 1.38 0.14 -2.26 0.04
X1*X3 0.85 0.35 -1.96 0.07
X1‘X4 21.15 <0.01 -7.71 0.2
X1 ‘X5 <-0.01 0.55 <-0.01 0.69
X2*X3 -0.55 <0.01 0.04 0.82
X2*X4 -0.94 0.37 1.37 0.25
X2*X5 <-0.01 0.94 <-0.01 0.38
X3"‘X4 -1.1 0.3 1.71 0.16
X3*X5 <0.01 0.19 <-0.01 0.1
X4*X5 0.01 0.05 <-0.01 0.7
R2 0.97 0.95
F-ratio 20.07 1 1.49
prob F <0.05 <0.05
CV (%) 8.77 5.76

 

s 5 5
Y = B0+ZBiXi+ ZBiiXiXi +ZBiniXi

PI F1 F1

109

Table 4- Parameter estimates and analysis of variance of the second order polynomial

regression for four ﬁctors and one response (cholesterol reduction (Y1))

 

 

Source Coefﬁcient prob> I T I
Intercept 51.91 <0.01
x1 -3.99 0.05
x3 10.76 <0.01
X4 4.97 0.01
X5 30.44 <0.01
x1*x1 5.23 0.14
x3*x3 -0.94 0.78
X4‘X4 -0.06 0.98
XS‘XS -8.7 0.02
x1*x3 -0.72 0.87
X1*X4 12.86 0.01
x1*xs 2.31 0.62
X3*X4 0.26 0.95
x3*xs 1.06 0.82
X4*X5 _ 4.36 0.35
R2 = 0.95 F-ratio = 22.79 prob F <0.05 cv (%) = 9.1

 

4 4 4
Y = B°+2BiXi+ ZBnXiXi +213inin

F1 F1 F1

2
Y1 = 51.91 - 3.99 Xl + 10.76 X3+ 4.97 X4 + 30.44X5 - 8.7 X5+ 12.86 XIX4

llO

Table 5- Parameter estimates and analysis of variance of the second order polynomial

regression for 3 factors and one response (FFA reduction (Y2))

 

 

Source Coeﬂicient prob>l TI
Intercept 92.11 <0.01
X1 24.58 <0.01
X2 2.39 0.07
X5 5.28 <0.01
X1 *X1 -31.5 <0.01
X2‘X2 - 1.19 0.6
X5‘X5 -624 0.01
X1*X2 -5.91 0.07
X1‘X5 -7.37 0.02
X2*X5 0.99 0.75
R2 = 0.96 ; F-ratio = 65.34 prob F <0.05 CV (%) = 3.70

 

3 3 3
Y = Bo+2BiXi+ ZBiixnri +2133an

181 F1 F1

2
Y2 = 92.11+ 24.58 x] + 5.28 x5 - 31.5 x, - 6.24 xi - 7.37 x1 x,

111

Table 6- Cholesterol reduction in anhydrous milkfat as predicted by the second order

 

 

polynomial regression for 4 ﬁctors
Run X1 X3 X4 X5 Y1 Y1 Residuals
(actual) (estimated) (act - est)
1 -1 -1 -1 1 61.78 61.31 0.46
2 -1 -l 1 -1 28.58 30.82 -2.24
3 -l l -1 -1 46.46 45.15 1.3
4 -l 1 1 1 77.3 73.71 3.58
5 -1 -1 -1 -1 38.53 34.68 3.84
6 -1 -1 1 1 63.66 61.92 1.73
7 -1 1 - 1 1 68.22 72.84 -4.62
8 -1 1 l -1 40.5 41.55 -l.05
9 1 -1 -l -1 20.41 23.61 -3.2
10 1 -1 1 1 65.41 66.33 -0.92
l 1 1 1 -1 1 65.98 63.36 2.61
12 1 1 1 - l 42.84 42.92 -0.08
13 1 -1 -1 1 54.28 52.55 1.72
14 1 -1 l -1 38.21 32.92 5.28
15 1 1 -1 -1 32.28 33.35 -l.07
l6 1 1 1 1 74.23 77.4 -3.17
17 -2 0 0 0 59.98 60.96 -0.98
18 2 0 O 0 53.52 53.57 -0.05
19 0 0 - 0 0 38.03 52.03 -14
20 0 0 0 0 53.32 52.03 1.28
21 0 -2 0 0 37.49 40.31 -2.82
22 0 2 0 O 63.64 61.85 1.78
23 0 0 -2 0 46.85 46.87 -0.02
24 0 0 2 0 55.7 56.81 -1.11
25 0 0 0 -2 11.91 12.77 -0.86
26 0 O 0 2 73.7 73.87 -0.17
27 0 0 0 0 52.27 52.03 0.23
28 0 0 0 0 56.27 52.03 4.23
29 0 0 0 0 53.02 52.03 0.98
30 0 0 0 0 54.48 52.03 2.44
31 0 0 0 0 54.94 52.03 2.9
32 0 0 0 0 54.02 52.03 1.98

 

112

Table 7- Reduction of ﬁ'ee ﬁtty acids as predicted by the second order polynomial

regression for 3 ﬁctors

 

 

Run X1 X2 X5 Y2 Y2 Residuals
(actual) (estimated) (act - est)
1 -1 -1 -1 69.53 71.65 -2.12
2 -1 -l -1 58.07 63.17 -5.1
3 -1 -1 -1 68.06 63.17 4.88
4 -l -1 1 75.3 71.65 3.64
5 -1 1 -1 69.16 68.02 1.13
6 -1 1 1 79.12 77.5 1.61
7 -1 1 1 77.6 77.5 0.09
8 -l 1 -1 70.79 68.02 2.76
9 1 -1 -1 92.42 94.4 -1.98
10 1 -1 1 92.28 95.5 -3.22
11 l -l 1 92.38 95.5 -3.12
12 1 -1 -1 92.21 94.4 -2.19
13 1 l 1 96.16 95.43 0.72
14 1 1 -1 90.36 93.34 -2.98
15 l 1 -l 92.35 93.34 -0.99
16 1 1 1 92.48 95.43 -2.95
17 -2 0 0 31.35 36.03 -4.6
18 2 0 0 92.33 85.19 7.13
19 0 -2 0 91.92 88.53 3.38
20 0 2 0 92.38 93.31 -0.93
21 0 0 0 92.24 92.11 0.12
22 0 0 0 92.63 92.1 1 0.51
23 0 0 0 92.64 92.11 0.52
24 0 0 0 92.45 92.1 1 0.33
25 0 0 -2 81.6 80.58 1.01
26 0 0 2 92.61 91.16 1.44
27 0 0 0 92.45 92.1 1 0.33
28 0 0 0 92.22 92.1 1 0.1
29 0 0 0 92.53 92.1 1 0.41
30 0 0 0 92.1 92.11 -0.01
31 0 0 0 92.45 92.1 1 0.33
32 0 0 0 91.93 92.11 -0.18

 

113

4. 4.2 Optimization based on Y1 (Percentage cholesterol reduction)

The model Y1 is useﬁrl for indicating the direction-to-clmnge variables to
maximize cholesterol reduction. Judging from the signiﬁcant regression coeﬂicients
(Table 4), the most important factors inﬂuencing cholesterol reduction are KOH:FFA
mole ratio, mixing time, CD:cholesterol mole ratio, and mixing rate. They showed '

signiﬁcant ﬁrst order, quadratic, and two-ﬁctor interaction terms.

4. 4.2.1 KOH:FFA mole ratio

KOH was used to neutralize the FF As and remove them from the system. The range
of KOH:FFA mole ratios examined is shown in Table 2. The negative slope
B1 = -3.99 (Table 4) indicated that cholesterol reduction decreased with increased KOH
concentrations at any mixing time or rate (Figures 1 and 2). The negative eﬂ‘ect of KOH
on cholesterol reduction might be due to its ability to convert the FFAs present in the
system to soap molecules. The soap molecules aggregate to form micelles because their
concentration in the system (10.39 mM) surpasses the critical nricelle concentration
(cmc) which is reported to be 1.5 mM for sodium oleate (Small, 1995). The negatively
charged carboxylate groups lie scattered over the surface of the micelle and form
hydrogen bonds with the hydroxyl groups on the external rim of the B-cyclodextrin

molecules, thus competing with cholesterol for the B-CD molecules.

The use of higher B-cyclodextrin:FFA mole ratios to counteract the negative effect of
KOH (Figure 3), is not economically feasible. The shape of the response surface

(Figure 3) and the slope of the two factor interaction terms (B IB 4 = 12.86) are

characteristic of the strong interaction between these two variables.

114

 

 

 

 

 

 

 

 

 

 

 

 

 

\ I\
/I ‘ \\ \
\\\\‘\'\‘\\\ \
‘3 / / \‘QQQ\\\\\\\\\\\\‘
«2‘ 70‘ / /\ax\§\\§§\\§\§\§\\§\\ \\
'7 \\ \\\\\\\\ \
1 ...//// \\\
"' / / ‘ W \
a 50‘ / / .“t‘\\\\\\\\\\\\\\\‘
c / / \\\\\\\\\\\\\\\\\\
8' 40‘ / / ..\\\\\\\\\\\\\\\\\
s / / “W
’3 30‘ / - ‘§.\\\ - 3
°\ --
" / O h 2
2°‘ . .

 

10‘ “5.36 a"

Figure 1- Response surface methodology for percent reduction of cholesterol (Y1) in
anhydrous milkﬁt at a mixing rate of 1000 rpm and a B-cyclodextrin:cholesterol mole

ratio of 5.58.

115

1001

’I’ I
”I I
I

.I

80‘
60‘

\\\\A

40‘
20‘

\\\\A

 

 

(%) "099mm 10191891093

\\\

-20‘
-40‘
-60‘

 

\\\

 

 

 

 

 

 

Figure 2- Response surface methodology for percent reduction of cholesterol (Y 1) in
anhydrous milkfat using a mixing time of 10 min and a B-cyclodextrin:cholesterol mole

ratio of 5.58.

116

 

 

A///////

 

m/////M/

m m m m 4 2 m
Ore—833. Become: Acev

-40‘

Figure 3- Response surface methodology for percent cholesterol reduction (Y1) in

anhydrous milkfat using a mixing time of 10 min and a mixing rate of 1000 rpm.

117

Calcium salts are carboxylates acceptors. They react with the soluble potassium soaps to
form insoluble calcium soaps, thus counteracting the detrimental effect of the soap
micelles on cholesterol reduction as shown in runs 18 and 19 (Table 2).

The CaC12:FFA mole ratio used within the experimental domain had no signiﬁcant effect

on the model. The range of CaC12:FFA mole ratios tested is shown in Table 2.

4. 4.2.2 Mixing time
Cholesterol reduction (Y1) was strongly dependent on mixing time. The

positive slope, B 3 = 10.76, indicated that that cholesterol reduction increased with

increased mixing time at any B-cyclodextrinzcholesterol mole ratio or any mixing rate

(Figures 4 and 5). The range of mixing times examined is summarized in Table 2.

4. 4. 2. 3 ﬂ-CD.'cholesterol mole ratio

Theoretically, 1 mole of B-CD will complex 1 mole of compound
(Szejtli, 1982a). Some larger molecular weight compounds may complex with more than
one cyclodextrin molecule (Szjetli, 1982a). The range of B- CD:cholesterol mole ratios
tested is shown in Table 2. Cholesterol reduction was strongly dependent on the B-
CD:cholesterol mole ratio. The positive slope B4 = 4.97, indicated that cholesterol

reduction increased with increased B-cyclodextrinzcholesterol mole ratio (Figure 6).

4. 4.2.4 Mixing rate
Mixing is necessary to increase the contact between the B-cyclodextrin molecules

and cholesterol molecules. The greater the contact the greater the rate of complex

118

 

 

 

Owe—owned— noaeoaoa Anew

Figure 4- Response surﬁce methodology for percent cholesterol reduction (Y1) in

anhydrous milkﬁt using a KOH:FFA mole ratio of l and a mixing rate of 1000 rpm.

119

(%) "011091191 101919010‘13

 

 

 

Figure 5' Response surface methodology for percent cholesterol reduction (Y1) in
anhydrous milkfat using a KOH:FFA mole ratio of 1 and a 3 -, ‘ ‘ 4 ' L ' .

mole ratio of 5.58.

I20

     
   

 
             

v’ .
J

     

   

2‘53" 4&5‘

l
‘4'.
4',

.

9
94:0“
’56,?

N“
R“
m“
5M 'I
82"»?
4:9 ‘5
20'

D
O

 
   
  
  
      
     

 

   

80 .géz:§$§%§§
a a pipe.; g.
3 ‘ ‘35:: <_ M
: 20 ““‘53§5 0 “,3
-- .2. 0-8.31.) . 9°
’33 .40 ﬁ....‘\§§g§};.. 3 $7
& if“ 6‘

-60 .. ‘88 ‘93

~80 ’ 56'

_ .6;
., ' of
grate (1pm) [$0
Q

Figure 6- Response surﬁce methodology for percent cholesterol reduction (Y1) in
anhydrous milkﬁt using a KOH:FFA mole ratio of 1 and a mixing time of 10 min.

|2I

formation between cholesterol and B—cyclodextrin molecules (Awad et aL, 1996).
Moreover, mixing helps stabilize the oil-in-water emulsion. As soon as the temperature
reached 50 °C, B-cyclodextrin was added and mixed for 10 min. The cholesterol
reduction was strongly dependent on the mixing rate. The positive slope B5 = 30.44
strongly indicated that cholesterol reduction increased with increased mixing rate at any,
0- cyclodextrinzcholesterol mole ratio . The shape of the response surface (Figure 6) was
characteristic of the strong interaction between B-cyclodextrinzcholesterol mole ratio and

the mixing rate. The range of mixing rates tested is shown in Table 2.

4.4.3 Optimization based on Y2(percent FFA reduction).

The model (Y2) is useful for indicating the direction-to-change variables to
maximize FFA reduction. Judging from the signiﬁcant regression coeﬁcients (Table 5),
the most important ﬁctors inﬂuencing FFA reduction are KOH:FFA mole ratio,
CaClzzFF As mole ratio, and mixing rate. They showed signiﬁcant ﬁrst order, quadratic,

and two factor interaction terms.

4. 4. 3.1 KOH:FFA mole ratio

The diluted alkali metal hydroxide solution of known concentration (KOH:FFA

mole ratio of 1) reacts with FFAs as follows:

 

RCOOH + KOH , RCOOK + H20

The amount of KOH used is a function of the FF A concentration in anhydrous milkﬁt. A

KOH:FFA mole ratio of 1 is enough to neutralize all the FFAs present in the system

122

under the experimental conditions described in the preliminary process. An excess of
alkali solution is detrimental to the yield due to loss of neutral ﬁt by saponiﬁcation
(Braae, 1976) and to the ﬂavor because of hydrolysis of lactones to their corresponding
hydroxy acids. Reduction in FFA content was strongly dependent on the KOH:FFA mole
ratio. The positive slope, B 1 = 24.58 (Table 5) indicated that F FA reduction increased
with increased KOH:FFA mole ratio at all mixing rates (Figure 7). The range of

KOH:FFA mole ratios examined is shown in Table 2.

4. 4. 3.2 Calcium Chloride:FFA mole ratio
Calcium chloride was used as a carboxylate acceptor. Its addition to the reﬁning
solution is critical due its ability to:

(1) react with the carboxylates to form insoluble salts similar to limestone. Their
separation from the system by centriﬁrgation is therefore very efﬁcient;

(2) react with the soluble potassium or sodium soaps to form insoluble soaps and
minimize the losses of neutral ﬁt by emulsiﬁcation, thus improving the yield;

(3) destroy the soap micelles and minimize the ﬂoculation of the B-cyclodextrin on the
electronegatively charged micelle surface, thus improving cholesterol reduction
(Table 2);

(4) synergistically increase the reduction of F FA in conjunction with KOH;

(5) saturate the system with cations and thus improve FF A reduction by prohibiting the

occurrence of the following reaction:

 

 

RCOO' + H20 « > RCOOH + OH'
The excess of cations should be provided by the calcium salts and not by the

123

   

   
  
   
  
    
 

 
 

  
    
   
 

 

    
     
 

 

   

  

 

.50 jj
100] ﬁWﬁ‘i‘g
33 5° / / Eﬂisﬁtsﬁéﬁgtﬂ-
3* o/ / rapaggg 18:91.
2. i l7 ' is ﬁt??? . -
5. '5°'// = ~§§§:.-
e\ ' , a {55‘
"’ -1oo ./ ﬁg? if. ..
‘ g ‘35, .~ 3
.150 5!: .~ 2 'O
-200' *4?
C

Figure 7- Response surface methodology for FFA reduction (Y2) in anhydrous milkfat.

124

caustic solution due to the detrimental effect of excess hydroxyl groups on yield and

ﬂavor.
The range of CaClzzFFA mole ratios examined is shown in Table 2. The effect of
calcium chloride on F FA reduction was confounded by the B-cyclodextrin present in the

system, since B—cyclodextrin also reduce FF A concentrations.

4. 4. 3.3 Mixing rate

Mixing rate is an important ﬁctor to assure contact between the solutes of the
reﬁning solution and FFAs. The mixing rate used within the experimental range had a
slight effect on FFA reduction (Figure 7). The positive slope B5 = 5.28 (Table 5)
indicated that FFA reduction increased with increased mixing rate. The range of mixing

rates tested is shown in Table 2.

4.4.4 Process evaluation

The model for Y1 showed that the greater the B-cyclodextrin:cholesterol mole
ratio, mixing time, and mixing rate, the greater the percentage of cholesterol reduction.
The joint eﬂect of these three independent variables used in runs 4 and 26 (Table2)
reduced cholesterol in rnilkﬁt by 77.3 and 73.7%, respectively. Further optimization of
these treatments may improve overall cholesterol reduction. The amount of B-
cyclodextrin used in run 4 (Table 2) was greater than the amount used in run 26
(Table2).

Taking into account the cost-effectiveness of the process, further optimization

was performed on run 26 (Table2) with regard to mixing time, as increasing the amount

125

of B-cyclodextrin is not economically feasible. In addition, use of higher mixing rates
may increase the viscosity of the mdium which could cause problems during
centrifugation. It was found that the reduction of cholesterol could be ﬁlrther improved
by increasing the mixing time. In fact, the theoretical mathematical model for cholesterol
reduction showed that the cholesterol could be completely removed with 35 min of ‘

mrxmg. The observed cholesterol reduction with 35 min of mixing was 95.8% (Figure 8).

4.4.5 Effect of anhydrous milkfat on blood lipid levels

The eﬂ‘ects of diet on circulating lipids in blood were among the earliest
observations of factors inﬂuencing lipid metabolism and the atherogenic process. A
voluminous research literature has accumulated over the past 75 years on the relationship
between diet, lipoproteins, atherogenesis, and coronary heart disease (CHD). The major
conclusions of this body of research are that (1) the higher the total blood cholesterol
level, the greater the severity of atherosclerosis and the greater the risk for CHD, (2)
dietary saturated ﬁt and cholesterol raise total blood cholesterol and low density
lipoprotein (LDL) cholesterol levels, and (3) polyunsaturated ﬁt lowers total blood
cholesterol and LDL cholesterol levels (Howell et al., 1997).

Although the synthesis of cholesterol by the liver is under feedback regulation
ﬁ'om dietary cholesterol, a reduction of dietary cholesterol intake will cause a net plasma
cholesterol decrease despite increased endogenous synthesis (Hegsted, 1986). The ability
with which dietary cholesterol suppresses endogenous cholesterol synthesis varies greatly
among individuals (Quintao and Speretto, 1987). Dietary cholesterol appears to suppress

LDL receptor function in humans (Appelbaum-Bowden et al. 1984).

126

10 (g) anhydrous milkﬁt containing 0.234% FFAs

i

10 ml distilled water containing 0.0475% KOH and 0.6% CaClz

l

Heat to 50°C in a water bath while mixing at 1400 rpm for 2.5 min
(using 50 mL beaker and Lightnin Lab Master (SI))

I

At 50°C, add B—cyclodextrin
(cyclodextrinzcholesterol mole ratio of 5.58)

I

Mix at 1400 rpm, 32.5 min at 50°C

Centrifuge at 8700 X g, 10min, room temperature

1

Fat is recovered in the supernatant phase
(95.8 % cholesterol reduction, and 92.4 % FFA reduction)

Figure 8- Optimized process to remove cholesterol and FFA in anhydrous milkﬁt. The
number of moles of KOH and CaClz added is function of the number of moles of FFA
originally present in AMF. The mole ratio of KOH:FFA used is 1:1. The mole ratio of
CaC12:FFA used is 5:1.

127

Most animal species are more sensitive than humans to dietary cholesterol and develop
high plasma cholesterol concentrations and aortic lesions that resemble human atheroma
(McGill, 1979a; 1979b).

The results of the present study conﬁrm previous observations of Hegsted (1986)
that dietary cholesterol does affect blood lipids. In fact, as can be seen from Figure 9,1
when gerbils were fed a modiﬁed milkﬁt, total plasma cholesterol and triacylglycerols
were reduced by 28 and 47%, respectively.

There are side effects to any intervention. The purpose of dietary therapy is to
reduce LDL cholesterol levels, but dietary therapy could have other effects that are
undesirable. For example, a cholesterol-lowering diet, while lowering LDL cholesterol
levels, will also lower HDL cholesterol levels (Denke, 1993). The decrease in HDL
cholesterol observed in this study (Figure 9) is mainly due to the ﬁct that HDL represents

the major transport system for cholesterol in gerbils (Hajri et al., 1998).

4.5 CONCLUSION

The process proposed in this study has a number of attributes which could result
in considerable practical advantages over the prior-art processes used for reducing
cholesterol and FFAs in milkﬁt. The process is eﬁcient since the cholesterol was almost
completely (>98%) removed. Free ﬁtty acid reduction was greater than 92.40 %, which
was more than 45 % greater than the prior-art processes. The process is simple and
eﬁcient in removing cholesterol. The process can be done in a closed system to protect

anhydrous milkﬁt ﬁom exposure to air to minimize oxidative deterioration. This process

does not require expensive specialized equipment It requires only heating and mixing

128

 

200 I

Blood lipids (mg/dL)
8 a:

D O

I

U!
0

 

 

I I

Native Milkfat
a Modiﬁed Milkfat
e

     

/’/, ﬂ;
/

\\\-NCTN‘ -

\ .\\ ~\\\
bx :\ \\\
\ ‘ t
\\‘x\ \ ‘5, \
.

   

\

     

\

  

\\\

 

 

\ \\ \r \
\\\ \ ‘

IIIII IIIIIIIIII IIIIIIIII

.\ \
\\
‘\$\\ in (\\\\

\
\ \\C-‘V \E\\~\ \
\‘.

\

\ \\\st

\
\

‘ \\‘\:§‘\

 

 

 

’6»

Figure 9- Blood lipid response of gerbils to modiﬁed milkfat

"' Means with no common superscript are signiﬁcantly different (p<0.05).

TC: total cholesterol

HDL: high-density lipoprotein
LDL: low-density lipoprotein
TG: triglycerides

129

followed by centrifugation. Modiﬁed milkﬁt has all the beneﬁts of native milk ﬁt with
one exception, it does not increase blood cholesterol and therefore decreases the risk of
heart disease.

Modiﬁed milkﬁt is great for cooking, baking, toppings and spreading. Also, it could be
used for whole milk reconstitution and consequently the whole spectrum of

dairy products including ice cream, yogurt, and cheese.

130

SUMMARY AND CONCLUSIONS

It is in the area of consumer needs that we encounter some of the most diﬂicult
problems in fostering the rational development of food technologies. One key area of
development is the modiﬁcation of animal ﬁts, particularly dairy products, to satisfy
the changing dietary habits of consumers. Concerns about cholesterol and saturated
ﬁt originate ﬁom the ﬁct that high serum cholesterol, especially LDL, is one of the
risk ﬁctors associated with atherosclerosis. This study addressed alternatives to
hypercholesterolemic anhydrous milkfat by: (1) developing an enzymatic process to
convert the hypercholesterolemic myristic acid to myristoleic acid using the A9-
desaturase system of hen hepatic micro somes, (2) and reducing the cholesterol
content of animal ﬁts by chemical means, followed by their blending them with

vegetable oils.

The ﬁrst part of the study aimed at identifying the physiological role of the
hepatic A9-desaturase and the way in which this enzyme is inﬂuenced by solubilzing
proteins such as BSA and F ABPs, pH, temperature, enzyme concentration, and
different mixing rates of the incubation system. Although most of the available
evidence implicates the CoA ester of the ﬁtty acid as the substrate for desaturation,
direct desaturation of the ﬁtty acid without prior activation to the CoA derivative was
possible when the incubation medium was mixed at mixing speeds greater than 250

rpm in the presence of FABPs. Experimental evidence has been presented to show

131

that the variation in molecular structure between myristoyl-CoA and myristic
acid:FABP complexes gives rise to variation in desaturation activity. CoA and FABP
are guiding molecules. There are positive inﬂuences pulling reactants, but negative
repulsion plays a major role in the way reactants are guided. Attraction and repulsion
forces can act over short-and long-ranges. Short-range requires contacts with
membrane bound Ail-desaturase, and long-range interactions occur by diﬂ'usible
ﬁctors that set up gradients. Unraveling the mechanism of these interactions is
complicated by the ﬁct that guidance molecules can have biﬁmctional roles. The

same molecule can be attractive or repulsive, short- and long-range.

The second part of this study aimed at developing a process to reduce
cholesterol and free ﬁtty acids in anhydrous milkﬁt. The process to reduce
cholesterol in anhydrous milkﬁt is based on the speciﬁc afﬁnity of B-cyclodextrin for
ﬁ'ee cholesterol and cholesterol esters. Other processes have been proposed in the
US. and overseas, including diet alteration, solvent extraction, and adsorption with B-
cyclodextrin. A major disadvantage of the solvent approach is the deterioration of the
ﬁne ﬂavor components of milkﬁt. The process demonstrated in this study has a
number of attributes that resulted in considerable practical advantages over other
cholesterol-reduction technologies. The process is eﬂicient since the cholesterol was
almost completely (>98%) removed. Free ﬁtty acid reduction was greater than 92.40
%, which was more than 45 % greater than other prior-art processes. The process can
be done in a closed system to protect anhydrous milkﬁt ﬁ'om exposure to air to

minimize oxidative deterioration. This process does not require expensive and

132

specialized capital equipment It requires only heating and mixing followed by
centriﬁrgation. Modiﬁed milkﬁt has all the beneﬁts of native milkﬁt and it does not
increase blood cholesterol. Instead, it actually lowers it and therefore decreases the
risk of heart disease. It is great for cooking, baking, topping and spreading. Also,
modiﬁed milk ﬁt could be used for whole milk reconstitution and consequently the

whole spectrum of dairy products such as ice cream, yogurt, and cheese.

I33

FUTURE RESEARCH

Some suggestions for future research are listed below:

1- Bridging the gaps in our understanding of unsaturated ﬁtty acid biosynthesis, namely
the detailed chemistry of desaturation reactions and the biochemical control mechanism

that inﬂuence the activity of the enzymes involved.

2- The mechanism by which alterations in dietary ﬁtty acids aﬁect athero genesis and the
risk of coronary heart disease, including degree of saturation, chain length, and ﬁtty acid

series.

3- Clariﬁcation of the role of dietary cholesterol and cholesterol oxide products in
atherogenesis, including variability in response, effects on cholesterol metabolism in both
ﬁting and postprandial states, and interactions of postprandial lipoprotein remnants with

cells of the arterial wall.

134

003.13CU'D ”(n-ﬂﬁo—ﬂm

announced) ﬂC-uo'OD—Q”

Appendix I

 

§

 

 
 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

, a
1 i
ear .
. 74
so? .
I a? '
4e .
j z a . I
. I
213- .
I 3f I
01 : LLL. . 4‘: . , r - 4: a“ T - ;-+‘r.—- - r5 - - - , T i
so 100 150 200 250 see 350 400 450
n/z
Mass spectrum of authentic standard cis-9-C14:1
100‘ 5
i
so- .
d 74 -
601 94 .
4&1l 208 ..
1 a
1 1 4 1-6 ’
20 I.
I 1 240 ,
j 1 9 l
a . I v v a L- IV a..-' 1:.4‘ AVL‘J '!A‘a W L all
so roe 150 200 250 see 350 1160

Mass spectrum of sample cis-9-Cl4zl produced by the desaturation of C14:0 by hen

liver desaturase system.

135

REFERENCES

Introduction

Ascherio, A., Hennekens, H.C., and Willett, W.C. 1994. Trans-ﬁtty acid intake and
risk of myocardial inﬁrction. Circulation 89(1):94-101.

Expert Panel, 1994. The export panel report on detection, evaluation, and
treatment of high blood cholesterol in adults (ATP 11) (1994). National
Cholesterol Education Program, National Heart, Lung,and Blood Institute,
US. Department of Health and Human Services, Bethesda, MD.

National Cholesterol Education Program (NCEP), 1993. Second report of the expert
panel on detection, evaluation and treatment of high blood cholesterol in adults.
(Adult Treatment Panel 11). NIH Publication no.93-3095 .

Chapter I: Literature Review

Apgar, J.L., Shivel, CA, and Tarka jr., SM. 1987. Digestibility of cocoa butter and
corn oil and their inﬂuence on fatty acid distribution in rats. J. Nutr. 117:
660-665.

A1111, 1., Bourdeau, A., Makhlouf, J ., Tardif, R., and Sahasrabudhe, MR. 1987.
Fractionation of anhydrous milkfat by supercritical carbon dioxide. J. Food
Sci. 52:1231-1236.

Am], J., Bourdeau, A., Makhlouf, J ., Tardif, R, and Bellavia, T. 1988. Fractionation
of anhydrous milkfat by short-path-distillation. .1. Am. Oil Chem. Soc. 65:
1642-1646.

Ascherio, A., Hennekens, HO, and Willett, W.C. 1994. Trans-fatty acid intake and
risk of myocardial infarction. Circulation 89(1):94-101.

Awad, AC. and Smith, D.M. 1996. Method for reduction of cholesterol in egg
materials. United States patent, USP 5,484,624.

Banks, W., Clapperton, J.L., And Girdler, AK. 1990. Effect of dietary unsaturated

fatty acids in various forms on the do novo synthesis of fatty acids in the
bovine mammary gland. J. Dairy Res. 57:179-185.

136

Bhaskar, A.R., Rizvi, SSH, and Sherbon, J.W. 1993. Anhydrous rnilkﬁt
ﬁactionation with continuous countercurrent supercritical carbon dioxide. J.
Food Sci. 58:748-752.

Bass, NM. 1985. Function and regulation of hepatic and intestinal ﬁtty acid binding
proteins. Chem. Phys. Lipids. 38:95-114.

Bayol, A., Franlinet, J., and Gonzalez, B. 1989. Procédé d’élimination de composés
steroidiques contenus dans une substance d’origine biologique. European
Patent, EP 326,469.

Berk, K.N. 1998. Introductory Statistics with SYSTAT. Prentice Hall, New Jersey.

Bloomer, S., Adlercreutz, P., and Mattiasson, B. 1990. Triglyceride interesteriﬁcation

by lipases: Cocoa butter equivalent from a ﬁaction of palm oil. J. Am. Chem.
Soc. 67:519-524.

Bonanome, A. and Grundy, SM. 1988. Effect of dietary stearic acid on plasma
cholesterol and lipoprotein levels. N. Engl. J. Med. 318:1244-1248.

Bonanome, A. and Grundy, SM. 1989. Intestinal absorption of stearic acid after
consumption of high ﬁt meals in humans. J. Nutr. 119:1556-1560.

Brenner, RR. 1971. The desaturation step in the animal biosythesis of
polyunsaturated ﬁtty acids. Lipids. 6:567-575.

Brenner, RR. 1974. The oxidative desaturation of unsaturated ﬁtty acids in animals.
Mol. Cel. Biochem. 3:41-52.

Brophy, RI. and Vance, DE. 1976. The synthesis and hydrolysis of long chain ﬁtty
acyl-coenzyme A thioesters by soluble and microsomal ﬁactions ﬁ'om the
brain of the developing rat. Biochem. J. 160:247-251.

Catala, A., Nervi, A.M., and Brenner, RR. 1975. Separation of a protein ﬁctor
necessary for the oxidative desaturation of ﬁtty acids in the rat. J. Biol.
Chem. 250:7481-7484.

Chang, M.K., Abraham, 0., and John, V.T. 1990. Production of cocoa butter-like ﬁt
ﬁom interesteriﬁcation of vegetable oils. J. Am. Chem. Soc. 67:832-834.

Clement, I.P, Scirneca, J. A., and Thompson, H.J. 1994. Conjugated linoleic acid: A
powerful anticarcinogen from animal ﬁt sources. Cancer Supplement 74(3):
1050-1054.

Courregelongue, J. and Mamand, J. P. 1989. Process for eliminating cholesterol
contained in a ﬁtty substance of animal origin and the ﬁtty substance with

137

reduced cholesterol obtained. United States Patent, USP 4,880,573.

Denke, M.A. and Grundy, SM. 1991. Effects of ﬁts high in stearic acid on lipid and
lipoprotein concentrations in man. Am. J. Clin. Nutr. 54:1036—1040.

Denke, M.A. and Grundy, SM. 1992. Comparison of effects of lauric and palmitic
acid on plasma lipids and lipoproteins. Am. J. Clin. Nutr. 56:895-898.

Elovson, J. 1965. Immediate ﬁte of albumin-bound [1"4C] stearic acid following its ‘
intraportal injection into carbohydrate refed rats. Biochim. Biophys. Acta.
106:480-494.

Enock, H.G., Catala, A., and Strittmatter, P. 1976. Mechanism of rat liver microsomal
stearyl-CoA desaturase. J. Biol. Chem. 251 :5095-5103.

Faulkner, A. and Pollock, HT. 1989. Changes in the concentration of metabolites in
milk ﬁom cows fed on diets supplemented with soybean oil or ﬁtty acids. J.
Dairy Res. 56:179-183.

Friedrich, J .P. and Pryde, EH. 1984. Supercritical C02 extraction of lipid-bearing
nﬁterials and characterization of the products. J. Am. Oil. Chem. Soc. 61 :223-
228.

Graille, J., Pioch, D., Serpelloni, M., and Mentink, L. 1991. Procédé de ﬁbrication de
produits Iaitiers a teneur réduite en stérols, et produits ainsi obtenus. European
Patent , EP 406,101.

Grumrner, RR 1991. Effect of feed on the composition of milkﬁt. J. Dairy Sci. 74:
3244-3257. '

Grundy, SM. and Vega, G.L. 1988a. Hypertriglyceridemia causes and relations to
coronary heart disease. Semin Thromb. Hemost. 14:149-156.

Grundy, SM. and Vega, G.L. 1988b. Plasma cholesterol responsiveness to saturated
ﬁtty acids. Am. J. Clin. Nutr. 47:822-824.

Gurr, MI. 1994. Nutritional signiﬁcance of lipids. C11. 9 in Advanced Dairy
Chemistry, Lipids. P.F. Fox (Ed.), p. 349-432. Chapman &Hall, London.

Hayes, KC. 1995. Saturated ﬁt and blood lipids: new slant on old story. Can. J.
Cardiol. 11:39G-46G.

Hegsted, D.M. and Gallagher, A 1967. Dietary ﬁt and cholesterol and serum
cholesterol in gerbil. J. Lipid Res. 8:210-214.

Holloway, P.W. 1971. A requirement for the three protein components in microsorrral

138

stearyl coenzyme A desaturation. Biochem. 10(9): 1556-1560.

Iacono, J.M. and Dougherty, RM. 1991. Lack of effect of linoleic acid on the
high density lipoprotein cholesterol ﬁ'action of plasma lipoproteins. Am. J.
Clin. Nutr. 53: 660-664.

Jeﬂ‘coat, R., Brawn, RR, and James, A.T. 1976. The effect of soluble proteins on the
activity of microsomal stearoyl CoA and linoleoyl CoA desaturase.
Biochimica. Biphysica. Acta. 431 :33-44.

Jeffcoat, R., Brown, P.R., Safford, R. 1977. Properties of rat liver microsomal staroyl
coenzyme A desaturase. Biochem. J. 161:431-437.

Jeffcoat, R. 1979. The biosynthesis of unsaturated ﬁtty acids and its control in
mammalian liver. Essays Biochem. 15:1-37.

Jones, P.D., Holloway, P.W., Peluﬂ‘o, R.O., and Wakil, SJ. 1969. A requirement for
lipids by the microsomal stearoyl-coenzyme A desaturase. J. Biol. Chem.
244:744-754.

Kennedy, J. 1991. Structured lipid for the future. J. Food T echnol. 11:76-78.

Keys, A., Anderson, J.T. and Grande, F. 1957. Prediction of serum-cholesterol
responses of man to changes in ﬁts in the diet. Lancet (Nov. 16):955-966.

Keys, A., Anderson, J.T. and Grande, F. 1965. Serum cholesterol response to changes
in the diet. IV. Particular saturated ﬁaty acids in the diet. Metabolism 14:776-
787.

Khosla, P. and Hayes, KC. 1993. Dietary palmitic acid raises LDL cholesterol
relative to oleic acid only at a high intake of cholesterol. Biochim. Biophys
Acta 1210:13-22.

Kromhout, D. Bosscheiter, E. D., and Delezzene, C. 1985. The inverse relation
between ﬁsh oil consumption and 20 year mortality from coronary heart
disease. N. Engl. J. Med. 312: 1205- 1209.

Larry, L.S., James, OH, and John, GP. 1990. Medium chain ﬁtty acids: evidence
for incorporation into chylomicron triglycerides inhuman. Am. J. Clin. Nutr.
52:834-836.

Leaf, A. and Webber, RC. 1988. Cardiovascular effects of n-3 ﬁtty acids. N. Engl. J.
Med. 3 18: 549-556.

Lim, S. Lim, G. B. and Rizvi, S. S. H. 1991. Continuous SC-COz processing of
milkﬁt. In Proceeding of 2"” International Symposium on Supercritical Fluids,

139

M.A. McHugh (Ed), p. 292-296, Boston, MA, May 20-22. Dept. of Chemical
Engineering, Johns Hopkins Univ., Baltimore, MD.

Maclennan, K, Jensen, O.M., Mosbach, J., and Vuori, H. 1978. Diet, thransit time,
stool weight, and colon cancer in two Scandinavian populations. Am. J. Clin.
Nutr. 31:S 239-242.

Mahley, RW. 1985. Atherogenic lipoproteins and coronary artery disease. Concepts
derived from recent advances in cellular and molecular biology. Circulation
72: 943-948.

Makhlouf, J., Arul, J., Bourdeau, A., Verret, P., and Saharabudhe, MR. 1987.
Fractionement de la matiere grasse laitiére par cristallisation simple et son
utilisation dans la ﬁbrication de beurres mous. Can. Inst. Food Sci. Technol. J.
20: 236-247.

Masoro, E.J., 1968. Biosynthesis of fatty acids. Ch.4 In Physiological Chemistry of
Lipid in Mammals, E. J. Masoro (Ed.), p.58-73.W.b.Saunders Company,
Philadelphia.

Mattson, RH. and Grundy, SM. 1985. Comparison of eﬁ‘ects of dietary saturated,
monounsaturated, and polyunsaturated ﬁtty acids on plasma lipids and
lipoproteins in man. J. Lipid Res. 26:194-202.

Mensink, RP and Katan, MB. 1989. Effect of diet enriched with monounsaturated or
polyunsaturated ﬁtty acids on levels of low-density lipoprotein cholesterol in
healthy women and men. N. Engl. J. Med. 321: 436-441.

Mensink, RP and Katan, MB. 1992. Effects of dietary ﬁtty acids on serum lipids
and lipoproteins: a meta-analysis of 27 trials. Arterio. Thromb. 12:911-919.

Nawar, W.W. 1985. Lipids. Ch.4. In Food Chemistry. O.R. Fennema (Ed.), p. 139-
244. Marcel Dekker, New York.

Oakenfull, D.G. and Sidhu, G. S. 1991. Cholesterol reduction. International Patent
(Patent Corporation Treaty) WO 91/11114.

Oshino, N., Imai, Y., and Sato, R. 1971. Function of cytochrome b5 in ﬁtty acid
desaturation. J. Biochem. 69:155-167.

Peng, S.K., Tham, P., Taylor, CB, and Mikkelson, B. 1979. Cytotoxicity of
oxidation derivatives of cholesterol on cultured aortic smooth muscle cells and
their eﬂ‘ect on cholesterol biosynthesis. Am. J. Clin. Nutr. 32:1033-1042.

Peng, S.K., Hu, B., and Mortin, R. 1991. Angiotoxicity and atherogenicity of
cholesterol oxides. J. Clin. Lab. Anal. 5:144-152.

140

Prarnod, K. and Hayes, KC. 1992. Comparison between the effect of dietary
saturated (16:0) monounsaturated (18:1), and polyunsaturated (18:2) ﬁtty
acids on plasma lipoproteins metabolism in Cebus and Rheus monkeys fed
cholesterol ﬁ'ee diet. Am. J. Clin. Nutr. 55:51-62.

Pronczuk, A., Khosla, P., and Hayes, KC. 1994. Dietary myristic, palmitic and
linoleic acids modulate cholesterolemia in gerbils. FASEB J. 8:1191-1200.

Raju, P.K. and Reiser, R. 1972. Stimulation of stearoyl CoA desaturase activity by
sn-glycerol-3-phosphate in mouse liver microsomes. Biochem. Biophys. Acta.
280: 267-274.

Rasmussen, J.T., Bdrchers, T., and Knudsen, J. 1990. Comparison of the binding
afﬁnities of acyl-CoA binding protein and ﬁtty-acid-binding protein for long-
chain acyl-CoA esters. Biochem. J. 265:849-855.

Ratnayake, W.M. and Pelletier, G. 1992. Positional and geometrical isomers of
linoleic acid in partially hydrogenated oils. J. am. oil Chem. Soc. 69(2): 95-105.

Roderbourg, H., Dalemans, H., and Bouhon, R. 1993. Process for reducing the
content of cholesterol and free ﬁtty acids in an animal ﬁt. United States
Patent, USP 5,232,725.

Rogers, M.J. and Strittmatter, P. 1974. Evidence for random distribution and
translational movement of cytochrome b5 in endoplasmic reticulum. J. Biol.
Chem. 249:895-900. '

Spartz, L. and Strittmatter, P. 1973. A form of reduced nicotinamide adenine
dinucleotide cytochrome b5 reductase containing both the catalytic site and an
additional hydrophobic membrane-binding segment. J. Biol. Chem. 248:793-
799.

Sreenivasan, B.1978. Interesteriﬁcation of ﬁts. J. Am. Oil Chem. Soc. 55:796-805.

Stamler J, Shekelle R. 1988. Dietary cholesterol and human coronary heart disease:
the epidemiological evidence. Arch Pathol. Lab. Med. 112:1032-1040.

Strittmatter, P., Spartz, L., Corcoran, D., Rogers, M.J., Setlow, B., and Reline, R.
1974. Puriﬁcation and properties of rat liver microsomal stearoyl coenzyme A
desaturase. Proc. Natl. Acad Sci. USA. 71:4565-4569.

Strittmtter, P. and Rogers, M.J. 1975. Apparent dependence of interactions between
cytochrome b5 and cytochrome b5 reductase upon translational diffusion in
dirnyristoyl lecithin liposomes. Proc. Natl. Acad Sci. USA. 72:2658-2661.

Sundrarn, K, Hayes, KC, and Siru, OH. 1994. Dietary palmitic acid results in a

141

lower serum cholestrol than a lauric-myristic acid combination in
norrnolipidemic humans. Am. J. Clin. Nutr. 59:841-846.

Sundram, K., Hayes, KC, and Siru, OH. 1995. Both dietary 18:2 and 16:0 may be
required to improve the serum LDL/HDL cholesterol ratio in
normocholesterolemic men. J. Nutr. Biochem. 6:179-187.

Szejtli, J. 19823. Chemistry and preparation of cyclodextrins. In Cyclodextrins and
Their Inclusion Complexes, p. 17-43. Akademiai Kiado. Budapest.

Szejtli, J. 1982b. Types, formation and structures of inclusion complexes. In
Cyclodextrins and Their Inclusion Complexes, p. 93-108. Akademiai Kiado.

Budapest.
USDA 1991. Nutrient content of foods, Handbook 8-1, Washington, DC.

USDA 1997: Food Consumption, Prices, and Expenditures. Statistical Bulletin No.
965.

Walstra, P. and Jenness, R. 1984. Dairy Chemistry and Physics, Wiley, New York.

Walstra, P., Vandiet, T., and Kloek, W. 1994. Crystalliztion and rheological
properties of milk ﬁt. Ch.5. in Advanced Dairy Chemistry, Lipids. P.F.
Fox (Ed.), p.179-212. Chapman &Hall, London.

Willett, W.C. 1994. Diet and health: What should we eat? Science 264:532-537.

Yen, G.C. and Tsai, L]. 1995. Cholesterol removal from a lard-water mixture with
B—cyclodextrin. J. Food Sci. 60(3): 561-564.

Zock, PL. and Katan, MB. 1992. Hydrogenation alternatives. Effects of trans ﬁtty
acid and stearic acid versus linoleic acid on serum lipids and lipoproteins in
humans. J. Lipid Res. 33: 399-408.

Zock, P., de Vries, J., and Katan, M. 1994. Impact of myristic acid versus palrnitic

acid on serum lipid and lipoprotein levels in healthy women and men.
Arterioscler. Thromb. 14:567-575.

Chapter 11:

Bass, NM. 1985. Function and regulation of hepatic and intestinal ﬁtty acid binding
proteins. Chem. Phys. Lipids 38:95-114.

Berk, K.N. 1998. Introductory Statistics with Systat. Prentice Hall, New Jersey.

142

Brenner, RR. 1971. The desaturation step in the animal biosythesis of
polyunsaturated ﬁtty acids. Lipids. 6:567-575.

Brenner, RR. 1974. The oxidative desaturation of unsaturated ﬁtty acids in animals.
Mol. Cel. Biochem. 3:41-52.

Brett, D., Morris, L.J., and James, A.T. 1971. Speciﬁcity of the ﬁtty acid desaturases.
The conversion of saturated to monoenoic acids. Arch. Biochem. Biophys.
143:535-547.

Brophy, P.J. and Vance, DE. 1976. The synthesis and hydrolysis of long chain ﬁtty
acyl-coenzyme A thioesters by soluble and microsomal ﬁactions from the
brain of the developing rat. Biochem. J. 160:247-251.

Cook, H.W. and Spence, M.W. 1973. Formation of monoenoic ﬁtty acids by
desaturation in rat brain homogenate. J. Biol. Chem. 248:1786-1793.

Enock, H.G., Catala, A., and Strittmatter, P. 1976. Mechanism of rat liver microsomal
stearyl-CoA desaturase. J. Biol. Chem. 251:5095-5103.

Holloway, P.W. 1971. A requirement for the three protein components in nricrosomal
stearyl coenzyme A desaturation. Biochem. 10(9): 1556-1560.

Jeﬂ‘coat, R., Brawn, RR, and James, A.T. 1976. The effect of soluble proteins on the
activity of microsomal stearoyl CoA and linoleoyl CoA desaturase.
Biochim. Biophys. Acta. 431:33-44.

Jeﬂ‘coat, R., Brown, P.R., Saﬁord, R. 1977. Properties of rat liver microsomal staroyl
coenzyme A desaturase. Biochem. J. 161:431-437.

Jeﬂcoat, R., Dunton, AP, and James, A.T. 1978. Evidence for the different
responses of A9-, A°- and As-ﬁtty acyl-CoA desaturases to cytoplasmic
proteins. Biochim. Biophys. Acta. 528:28-35.

Lowry, O.H., Rosebrough, N.J., Farr, A.L., and Randall, R.J. 1951. Protein
measurement with the folin phenol reagent. J. Biol. Chem. 193:265-275.

Masoro, E.J., 1968. Biosynthesis of ﬁtty acids. Ch.4 In Physiological Chemistry of
Lipid in Mammals, E. J. Masoro(Ed.), p.58-73.W.b.Saunders Company,
Philadelphia.

Morrison, W.R. and Smith, L.M. 1964. Preparation of ﬁtty acid methyl esters and

dimthacetals from lipids with boron triﬂuoride-methanol. J. Lipid Res.
5:600-608.

143

Oshino, N., Imai, Y., and Sato, R. 1966. Electron-transfer mechanism associated with
ﬁtty acid desaturation catalysed by liver microsomes. Biochim. Biophys. Acta.
128:13-28.

Oshino, N., Imai, Y., and Sato, R. 1971. Function of cytochrome b5 in ﬁtty acid
desaturation. J. Biochem. 69:155-167.

Pand, S.V. and Mead, J .F. 1968. Long-chain ﬁtty acid activation in subcellular
preparations from rat liver. J. Biol. Chem. 243:352-361.

Paulsrud, J.R., Steewart, S.E., Graﬂ‘, G., and Holman, RT. 1970. Desaturation of
saturated ﬁtty acids by rat liver microsomes. Lipids 5(7):611-616.

Raju, RR. and Reiser, R. 1972. Stimulation of stearoyl CoA desaturase activity by
sn-glycerol-3-phosphate in mouse liver microsomes. Biochim. Biophys. Acta.
280: 267-274.

Salem, L. 1962. The role of long-range forces in the cohesion of lipoproteins. Can.
J. Biochem. Physiol. 40:1287-1298.

Spartz, L. and Strittmatter, P. 1973. A form of reduced nicotinamide adenine
dinucleotide cytochrome b, reductase containing both the catalytic site and an
additional hydrophobic membrane-binding segment. J. Biol. Chem. 248:793-
799.

SPSS Inc.1997. SYATAT, version 7.0, SPSS Inc., Chicago IL.

Strittmatter, P., Spartz, L., Corcoran, D., Rogers, M.J., Setlow, B., and Reline, R.
1974. Puriﬁcation and properties of rat liver microsomal stearoyl coenzyme A
desaturase. Proc. Natl. Acad Sci. USA. 71:4565-4569.

Voet, D. and Voet, J.G. 1990. Lipid metabolism. Ch. 23 in Biochemistry, p.618-665.

John Wiley and Sons Inc., NY.

Chapter 111: Direct desaturation of free myristic acid by hen liver microsomal
A9-desaturase without prior activation to myristoyl-CoA derivative.

AppleKvist, EL. and Dallner, G. 1980. Possible involvement of ﬁtty acid binding

protein in peroxisomal B-oxidation of ﬁtty acids. Biochim. Biophys. Acta.
617:156-160.

Bass, NM. 1985. Function and regulation of hepatic and intestinal ﬁtty acid binding
proteins. Chem. Phys. Lipids 38:95-114.

144

Berk, K.N. 1998. Introductory Statistics with Systat. Prentice Hall, New Jersey

Brecher, P., Saouaf, R, Sugarman, J.M., Eisenberg, D., and Larosay, K. 1984. Fatty
acid transfer between multilamellar liposomes and ﬁtty acid binding proteins.
J. Biol. Chem. 259:13395-13401.

Brenner, RR. 1971. The desaturation step in the animal biosythesis of
polyunsaturated ﬁtty acids. Lipids. 6:567-575.

Brett, D., Morris, L.J., and James, A.T. 1971. Speciﬁcity of the ﬁtty acid desaturases.
The conversion of saturated to monoenoic acids. Arch. Biochem. Biophys.
143:535-547.

Brudett, D. and Zeikus, J.G. 1994. Puriﬁcation of acetaldehyde dehydrogenase and
alcohol dehydrogenases ﬁom Thermoanaerobacter ethanolicus 39B and
characterization of the secondary-alcohol dehydrogenase as a bifunctional
alcohol dehydrogenase-acetyl CoA reductive thioesterase. J. Biochem.
302:163-170.

Catala, A., Nervi, A.M., and Brenner, RR. 1975. Separation of a protein ﬁctor
necessary for the oxidative desaturation of ﬁtty acids in the rat. J. Biol.
Chem. 250:7481-7484.

Cistola, D.P., Small, D.M., and Hamilton, J.A. 1987a. Carbon 13 NMR studies of
saturated ﬁtty acids bound to bovine serum albumin. J. Biol. Chem.
262:10971-10979.

Cook, H.W. and Spence, M.W. 1973. Formation of monoenoic ﬁtty acids by
desaturation in rat‘ brain homogenate. J. Biol. Chem. 248:1786-1793.

Cook, H.W. 1978. In vitro formation of polyunsaturated ﬁtty acids by desaturation in
rat brain: some properties of the enzymes in developing brain and comparison
with liver. J. Neurochem. 30:1327-1334.

Collins, D.M. and Hargis, PS. 1989. Distribution of ﬁtty acid binding proteins in
tissues and plasma of Gallus domesticus. Comp. Biochem. Physiol. 92 B(2):
283-289.

Enock, H.G., Catala, A., and Strittmatter, P. 1976. Mechanism of rat liver microsomal
stearyl-CoA desaturase. J. Biol. Chem. 251:5095-5103.

Enoch, HO. and Strittmatter, P. 1978. Role of Tyrosyl and arginyl residues in rat
liver microsomal stearyl coenzyme A desaturase. Biochem. 57:4927-
4932

Scapin, G., Spadon, P., Mammi, M., Zannotti, G., and Monaco, H. 1990. Crystal

145

structure of chicken liver basic ﬁtty acid-binding protein at 2.7 A resolution.
Mol. Cel. Biochem. 98:95-99.

Glatz, J.F.C. and Veerkamp, J.H. 1983. A radiochemical procedure for the assay of
ﬁtty acid binding by proteins. Anal. Biochem. 132:89-95.

Gurr, M.I., Robinson, MP, and James, AT. 1969. The mohanism of formation of
polyunsaturated ﬁtty acids by photosynthetic tissue. The tight coupling of
oleate desaturation with phospholipid synthesis in Chlorella vulgaris. Eur. J. -
Biochem. 9(1):70-78

Hamilton, J.A., Cistola, D.P., Morrisett, J.D., and S w, J.T. 1984. Interactions of
myristic acid with bovine serum albumin: A 3c NMR study. Proc. Natl.
Acad Sci. USA. 81:3718-3722 1...,

Holloway, P.W. 1971. A requirement for the three protein components in microsomal
stearyl coenzyme A desaturation. Biochem. 10(9): 1556-1560. 1.

 

Iritani, N., Fukuda, E., and Inogushi, K. 1980. A possible role of Z protein in dietary
control of hepatic triacylglycerol synthesis. J. Nutr. Sci. Vitaminol. 26:271-
277.

Kirn, H.K and Storch, J. 1992. Free ﬁtty acid transfer ﬁom rat liver ﬁtty acid-
binding protein to phospholipid vesicles. J. Biol. Chem. 267:77-82.

Lichtenstein, A.H., Small, D.M., and Brecher, P. 1982. Interaction of oleoyl
coenzyme A with phospholipid bilayers. Biochem. 21 :2233-2241.

Lowry, O.H., Rosebrough, N.J., Farr, A.L., and Randall, R.J. 1951. Protein
measurement with the folin phenol reagent. J. Biol. Chem. 193:265-275.

Masoro, E.J., 1968. Biosynthesis of ﬁtty acids. C114 In Physiological Chemistry of
Lipid in Mammals, ﬂ. Masoro(Ed.), p.58-73.W.B.Saunders Company,
Philadelphia.

 

Morrison, W.R. and Smith,L.M. 1964. Preparation of ﬁtty acid methyl esters and
dirnethyacetals ﬁom lipids with boron triﬂuoride-methanol. J. Lipid Res.
5:600-608.

Ockner, R.K. and Manning, J.A. 1976. Fatty acid binding protein. J. Clin. Invest.
58:632-641.
Oshino, N., Imai, Y., and Sato, R. 1966. Electron-transfer mechanism associated with
ﬁtty acid desaturation catalysed by liver microsomes. Biochim. Biophys. Acta.
128:13-28.

Oshino, N., Imai, Y., and Sato, K 1971. Function of cytochrome b5 in fatty acid

146

desaturation. J. Biochem. 69:155-167.

Pand, S.V. and Mead, J.F. 1968. Long-chain ﬁtty acid activation in subcellular
preparations from rat liver. J. Biol. Chem. 243:352-361.

Parks, J.S., Cistola, D.P., Small, D.M., and Hamilton, J.A. 1983. Interactions of the
carboxyl group of oleic acid with bovine serum albumin: ABC NMR study. J.
Biol. Chem. 258:9262-9269.

Pugh, EL. and Kates, M. 1973. Desaturation of phosphatidylcholine and
phosphatidylethanolarnine by a microsomal enzyme system ﬁom Candida
lipotica. Biochim. Biophys. Acta. 316:305-316.

Pugh, EL. and Kates, M. 1977. Direct desaturation of eicosatrienoyl lecithin to
arachidonoyl lecithin by rat liver microsomes. J. Biol. Chem. 252:68-73.

Raju, P.K. and Reiser, R. 1972. Stimulation of stearoyl CoA desaturase activity by
sn-glycerol-3-phosphate in mouse liver microsomes. Biochem. Biophys. Acta.
280: 267-274.

Sackrnann, E. and Trﬁuble, H. 1971. Studies on the crystalline-liquid crystalline phase
transition of liquid model membranes. J. Am. Chem. Soc. 97:4492-4498.

Spartz, L. and Strittmatter, P. 1973. A form of reduced nicotinamide adenine
dinucleotide cytochrome b5 reductase containing both the catalytic site and an
additional hydrophobic membrane-binding segment. J. Biol. Chem. 248:793-
799.

SPSS Inc.1997. SYATAT, version 7.0, SPSS Inc., Chicago IL.

Strittmatter, P., Spartz, L., Corcoran, D., Rogers, M.J., Setlow, B., and Reline, R
1974. Puriﬁcation and properties of rat liver microsomal stearoyl coenzyme A
desaturase. Proc. Natl. Acad Sci. USA. 71:4565-4569.

Voet, D. and Voet, J. G. 1990. Lipid metabolism. Ch. 23, in Biochemistry, p. 618-665.
John Wiley and sons Inc., NY.

Wu-Rideout, M.Y.C., Elson, C., and Shargo, E. 1976. The role of Fatty acid binding
protein on the metabolism of ﬁtty acids ill isolated rat hematocytes. Biochem.
Biophys. Res. Commun. 71:809-816.

Chapter 4: Reduction of the hypercholesterolemic effect of anhydrous milkfat.

AOAC, 1995. Official Methods of Analysis, 16“1 , ed. Association of Ofﬁcial
Analytical Chemists, Washington, DC.

147

Appelbaum—Bowden, D., Haﬂ'ner, S.M., Hartsook,mE., Luk, H.H., Albers, J.J., and
Hazzard, W.R. 1984. Down-regulation of the low-density lipoprotein receptor
by dietary cholesterol. Am. J. Clin Nutr. 39:360-67

Arul, J., Bourdeau, A., Makhlouf, J., Tardif, R., and Bellavia, T. 1988. Fractionation
of anhydrous milkﬁt by short-path-distillation. J. Am. Oil Chem. Soc. 65:
1642-1646.

Awad, AC. and Smith, D.M. 1996. Method for reduction of cholesterol in egg
materials. United States patent, USP 5,484,624.

Banks, W., Clapperton, J.L., And Girdler, AK. 1990. Eﬂect of dietary unsaturated
ﬁtty acids in various forms on the do novo synthesis of ﬁtty acids in the
bovine mammary gland. J. Dairy Res. 57:179-185.

Bayol, A., Franlinet, J ., and Gonzalez, B. 1989. Procédé d’élimination de composés
steroidiques contenus dans une substance d’origine biologique. European
Patent, EP 326,469.

Bhaskar, A.R., Rizvi, SSH, and Sherbon, J.W. 1993. Anhydrous milkﬁt
fractionation with continuous countercurrent supercritical carbon dioxide. J.
Food Sci. 58:748-752.

Box, GE. and Hunter, JS 1957. Multiﬁctor experimental designs for exploring
response surﬁces. Ann Math. Stat. 28:195-141.

Braae, B. 1976. Degumming and reﬁning practices in Europe. J. Am. Oil Chem Soc.
53: 353-357.

Courregelongue, J. and Maﬂiand, J. P. 1989. Process for eliminating cholesterol
contained in a ﬁtty substance of animal origin and the ﬁtty substance with
reduced cholesterol obtained. United States Patent, USP 4,880,573.

Denke, MA 1993. Dietary determinants of high-density lipoprotein-cholesterol levels.
Cardiovasc. Risk Factors. 3(5): 4-8.

Graille, J., Pioch, D., Serpelloni, M., and Mentink, L. 1991. Procédé de ﬁbrication de
produits Iaitiers a teneur réduite en stérols, et produits ainsi obtenus. European
Patent , EP 406,101.

Hajri, T., Khosla, P., Pronczuk, A., Hayes, KC. 1998. Myristic acid-rich ﬁt raises
plasma LDL by stimulating LDL production without affecting functional
clearance in gerbils fed a cholesterol-free diet. J. Nutr. 128:477-484

Hegsted, D.M. and Gallagher, A. 1967. Dietary ﬁt and cholesterol and serum
cholesterol in gerbil. J. Lipid Res. 8:210-214.

148

Hegsted, D.M. 1986. Serum- cholesterol response to dietary cholesterol: A
reevaluation. Am J. Clin. Nutr. 44:299-305.

Howell, H.W., McNamara, D.J., Tosca, M.A., Smith, B.T., and Gaines, J.A. 1997.
Plasma lipid and lipoprotein responses to dietary ﬁt and cholesterol: a meta-
analysis. Am. J. Clin. Nutr. 65:1747-1764.

Hunter, J .S. 1959. Determination of Optimum Operating Conditions by Experimental
Methods. Ind. Qual. Cont. 15(8):6-30. '

Keys, A., Anderson, J.T. and Grande, F. 1957. Prediction of serum-cholesterol
responses of man to changes in ﬁts in the diet. Lancet (Nov. 16):955-966.

Makhlouf, J., Am], J., Bourdeau, A., Verret, P., and Saharabudhe, MIL 1987.
Fractionement de la matiere grasse Iaitiere par cristallisation simple et son
utilisation dans la ﬁbrication de beurres mous. Can. Inst. Food Sci. Technol. J.
20: 236-247.

McGill, H.C.J., Jr. 1 979a. Cholesterol: appraisal of cholesterol as a causation ﬁctor
in atherogenesis. Am. J. Clin. Nutr. 3222632.

McGill, H.C.J. Jr. 1979b. The relationship of dietary cholesterol to serum cholesterol
concentration and to athersclerosis in man. Am. J. Clin. Nutr. 32:2664.

National Academy of sciences 1989b. Committee on diet and health, food and
nutrition board, national research council. Diet and Health: Implications for
Reducing Chronic Disease Risk. National Academy Press, Washington DC,

749pp.

Oakenﬁlll, D.G. and Sidhu, G. S. 1991. Cholesterol reduction. International Patent
(Patent Corporation Treaty) WO 91/11114.

Peng, S.K., Tham, P., Taylor, CB, and Mikkelson, B. 1979. Cytotoxicity of
oxidation derivatives of cholesterol on cultured aortic smooth muscle cells and
their effect on cholesterol biosynthesis. Am. J. Clin. Nutr. 32:1033-1042.

Peng, S.K., Hu, B., and Mortin, R. 1991. Angiotoxicity and atherogenicity of
cholesterol oxides. J. Clin. Lab. Anal. 5:144-152.

Quintao, E.C.R., and Sperotto, G. 1987. The role of dietary cholesterol in the
regulation of human body cholesterol metabolism. Adv. Lipid Res. 22:173-188.

Roderbourg, H., Dalemans, H., and Bouhon, R. 1993. Process for reducing the
content of cholesterol and ﬁee ﬁtty acids in an animal ﬁt. United States
Patent, USP 5,232,725.

SAS Institute, Inc. 1990. SAS/STAT User’s Guide, 4‘“ ed. Version 6. SAS Institute,

149

Inc., Cary, NC.

Small, D.M. 1995. Lipids, structure and biochemistry of . In Molecular Biology and
Biotechnology. R.A. Meyers (Ed.), p. 503-509. VCH Publishers, Inc., New
York.

Spady, D.K., and Dietschy, J .M. 1985. Dietary saturated triacylglycerols suppress
hepatic low density lipoprotein receptor activity in the hamster. Proc. Natl.
Acad Sci. USA 82:4526-4530.

Stamler J, Shekelle R. 1988. Dietary cholesterol and human coronary heart disease:
the epidemiological evidence. Arch. Pathol. Lab. Med 11221032-1040.

Szejtli, J. 1982a. Chemistry and preparation of cyclodextrins. In Cyclodextrins and
Their Inclusion Complexes, p. 17-43. Akademiai Kiado. Budapest.

Szejtli, J. 1982b. Types, formation and structures of inclusion complexes. In
Cyclodextrins and Their Inclusion Complexes, p. 93-108. Akademiai Kiado.
Budapest.

USDA. 1997: Food Consumption, Prices, and Expenditures. Statistical Bulletin No.
965.

Yen, G.C. and Tsai, L.J. 1995. Cholesterol removal from a lard-water mixture with
B-cyclodextrin. J. Food Sci. 60(3): 561-564.

150

  

      
  

U

IIIIIIIIIIII

IIIIIIIIIIIIIII

209