MSU

LIBRARIES

RETURNING MATERIALS:

 

 

 

 

Place in book drop to
remove this checkout from
your record. FINES wi11
be charged if book is
returned after the date
stamped be1ow.

 

 

 

 

 

 

 

EFFECTS OF ESSENTIAL FATTY ACIDS ON IMMUNE

FUNCTIONS IN MICE

BY

James William DeWille

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

1982

Cn'tt"1'~“-‘Ho

ABSTRACT

EFFECTS OF ESSENTIAL FATTY ACIDS ON IMMUNE

FUNCTIONS IN MICE

BY
James William DeWille

Recent reports indicate that polyunsaturated fatty
acids (PUFA) impair immunity. This research was aimed
at investigating the effects of essential fatty acid de-
ficiency (EFAD) and dietary PUFA on immune functions. De-
layed type hypersensitivity was reduced 30% in mice fed the
EFAD diet compared with mice fed the control or high PUFA
diet. Numbers of splenic and thymic mononuclear cells and
proportions of T and B cells were unaffected by diet.

In long term cultures (60-108 hours) containing fetal
calf serum (FCS), 3H-thymidine incorporation into lympho-
cytes from mice fed the three diets was similar in response
to mitogens and to one-way mixed lymphocyte reactions (MLR).
In short term cultures (1-4 hours) containing FCS, Con A
induced l4C-choline and 3H-inositol incorporation into

lipids was also unaffected by diet. But when a lipid free

media was used, Con A induced 14C-choline and 3H—inositol

James William DeWille

incorporation into lipids of lymphocytes from mice fed the
EFAD diet was less than 50% of the incorporation into
lymphocyte lipids of the other treatments. After 4-8
hours in culture, treatment differences were no longer
detectable.

In lipid-free cultures accumulation of aminoisobutyric
acid (AIB) by lymphocytes from mice fed the EFAD diet was
reduced by 30% at all time points (30-240 minutes) compared
with the other treatments. AIB uptake Vmax was similar
across the treatments, but the Km for the Na+ dependent
component of AIB uptake was 70-85% higher in Con A ac-
tivated EFAD lymphocytes compared with the other two treat-
ments.

Viability, 3H-uridine incorporation into RNA, and
14C-leucine incorporation into protein were unaffected by
diet, but the protein content of lymphocytes from mice
fed the EFAD diet was nearly double that of the other two
groups.

These results indicate that: (l) EFAD impairs in vivo
immune functions, but high PUFA diets do not, (2) EFAD
does not alter lymphocyte populations, or proliferative
responses to mitogens and MLR, but, (3) phospholipid
metabolism and neutral amino acid transport are impaired
in EFAD lymphocytes. These impairments may play a role in

the EFAD effects on in vivo immunity.

This Dissertation is dedicated to my Dad
who taught me a whole lot of

things you can't learn from books

ii

ACKNOWLEDGMENTS

I would like to acknowledge the help, guidance, and
patience of my thesis advisor(s) Dr. Dale Romsos and Dr.
Pam Fraker. Also thanks to my committee members, Dr.
Maurice Bennink, Dr. Walter Esselmann and a great pinch-
hitter, Dr. Maije Zile. Special thanks to my three "in-
vestors", Mike Dunn, Bryan Krantz, and Mike Sheahan for
putting friendship above money.

Also, special thanks to all the members of my family
for their patience and support throughout these past
years.

Finally, special thanks to my wife, Normanella, for

making it all fun.

iii

TABLE OF CONTENTS

Chapter

LIST OF TABLES. . . . . .
LIST OF FIGURES . . . . .
INTRODUCTION. . . . . . .
OVERVIEW AND OBJECTIVES .
Overview. . . . . .
Objectives. . . . .

LITERATURE REVIEW . . . .

Essential Fatty Acids:

Perspective . . . .

Historical

Essential Fatty Acid Metabolism .

EFA, Prostaglandins and Cells of the

Immune System . . .

EPA and the Lymphocyte Membrane

Lymphocyte Activation: Phospholipid

Metabolism. . . . .

EPA and Lymphocyte Membrane Receptors,

Enzymes, and Transport.

Lipoproteins and Immunity

EPA and Disease . .

EXPERIMENTAL. . . . . . .

Effects of Essential Fatty Acid
and Diets High in Polyunsaturated Fatty Acids,

Deficiency,

on Delayed Type Hypersensitivity in Mice.

Introduction. . . .

iv

Page

vii

ix

13

17

19

20
24
27

31

31
31

Chapter

Materials and Methods . . . . . .
Resu1ts O O O O O O O O O O 0 O 0
Discussion. . . . . . . . . . . .
Essential Fatty Acids and
Lymphocyte Activation I. Effects
on Proliferative Responses and
Phospholipid Metabolism . . . . . . . .
Introduction. . . . . . . . . . .
Materials and Methods . . . . . .
Results 0 O O O O O O O C O O O 0
Discussion. . . . . . . . . . . .
Essential Fatty Acids and Lymphocyte
Activation II. Effects on Neutral
Amino Acid Transport. . . . . . . . . .
Introduction. . . . . . . . . . .
Materials and Methods . . . . . .
Results 0 O O O O O O O O O O O O
DiscuSSion. O O O O O O O O O O 0
Essential Fatty Acid Deficiency and
Lymphocyte Protein Synthesis, RNA
Synthesis, Total Protein, and Viability
Introduction. . . . . . . . . . .
Materials and Methods . . . . . .
Results . . . . . . . . . . . . .

Discussion. . . . . . . . . . . .

Effect of EFAD and Control Mouse
Serum on Mitogen Responses. . . . . . .

Introduction. . . . . . . . . . .

Materials and Methods . . . . . .

Page

33
35
40

44
44
46
52

64

71
71
72
76

86

91
91
91
93
98

102
102

103

Chapter

Results

Discussion.

CONCLUSIONS

REFERENCES.

vi

Page

103
105

107

LI ST OF TABLES

Table Page
1 Composition of Diets. . . . . . . . . . . . 34
2 Body Weight and Delayed Type Hyper-

Sensitivity Response of Mice Fed the Ex-
perimental Diets for 10 Weeks . . . . . . . 37
3 Mononuclear Cell Counts, and Percentage

of T and B Cells in Lymphoid Organs of

Mice. . . . . . . . . . . . . . . . . . . . 53
4 One Way Mixed-Lymphocyte Reactions

(MLR) . . . . . . . . . . . . . . . . . . . 60
5 Km and Vmax Values for Na+ Dependent

AIB uptake into Con A Activated Lympho-

cytes . . . . . . . . . . . . . . . . . . . 85
6 Effect of Time in Lipid Free Cultures on

% Viability of Spleen Cells from Mice

Fed the EFAD,Control, or High PUFA

Diet. . . . . . . . . . . . . . . . . . . . 99
7 3H-uridine Incorporation into RNA of

Spleen Cells from Mice Fed the EFAD,

Control, or High PUFA Diet. . . . . . . . . 100

vii

Table Page

8 Effects of EFAD and Control Serum on
Mitogen Responses of Spleen Cells from

Mice Fed the EFAD or Control Diet . . . . . 104

viii

Figure

LIST OF FIGURES

Page

Unsaturated fatty acid metabolism . . . . . ll
Delayed-type hypersensitivity response

of mice fed an essential fatty acid de-

ficient diet (0% corn oil). Response

expressed as a percentage of control (13%

corn oil) response. Each point repre-

sents the mean of 10 mice. A = signi-

ficantly different from control (p<0.05);

B = significantly different from refed

(p < 0.05). . . . . . . . . . . . . . . . . 37
3H-thymidine incorporation into TCA
precipitated DNA of mitogen activated
spleen cells from mice fed an EFAD, con-
trol or high PUFA diet. Spleen cells
(2.5 x 105) were cultured for 60 hours
in DMEH plus 5% FCS and optimal doses

of each mitogen. 3H-thymidine (0.1

uCi) was present during the final 12

3H-

hours. The basal (unstimulated)
thymidine incorporation was 299:36 (EFAD),

232123 (control) and 269:37 cpm/well

ix

Figure

3

Page

(high PUFA). All results represent the

mean i SEM of triplicate observations

from each of 10 mice/treatment after a

56-70 day feeding period which was

initiated when the mice were 21:3 days

of age. . . . . . . . . . . . . . . . . . . 56
Effects of the addition of BSA:FA con-

jugates on 3H thymidine incorporation into

TCA precipitated DNA of Con A activated

spleen cells from mice fed an EFAD, con-

trol, or high PUFA diet. Spleen cells

(2.5 x 105) from each treatment were in-
cubated for 60 hours in 5% FCS (upper

panel) or serum free (lower panel) cul-

tures with the addition of BSA: 20:4

or BSA: 18:0 at a molar ratio of 2:1.
3H-thymidine (0.1 uCi) was present for
the final 12 hours. Data represent the
mean i SEM of duplicate or triplicate
observations on each of 8-10 mice per
treatment after a 56-70 day feeding period.
Mice were 21i3 days of age at onset of the

feeding period. . . . . . . . . . . . . . . 58

Figure

Page

Incorporation of phospholipid precursors
into total lipids of Con A activated
lymphocytes from mice fed the EFAD,
control or high PUFA diet cultured in
media containing either FCS or delipidated
FCS (dl-FCS). Splenic mononuclear cells
(1 x 106) were incubated with optimal
doses of Con A, or without mitogen (un-

14C-choline

stimulated) plus 0.5 uCi of
or 1.0 uCi of 3H-inositol. Harvested
lymphocytes were extracted with 8:4:1
(chloroform:methanol:10% TCA) and radio-
activity associated with the chloroform
layer was counted. Radioactivity
specifically associated with Con A ac-
tivation (Acpm) was determined by sub-
tracting the counts incorporated into
total lipids from unstimulated cells

from counts incorporated into the total
lipids Con A activated cells after back-
ground was subtracted from both. Data
represent the mean i SEM of duplicate ob-
servations on each of 8-10 mice/treatment

after a 56-70 day feeding period begun when

the mice were 21:3 days of age. . . . . . . 63

xi

Figure

7A

Page

Time course of AIB uptake. Rest—

ing (---) and Con A activated (———)

(2.5 ug/ml) splenic lymphocytes (10 x 106/
ml) were incubated in a total volume of

0.2 ml with 0.10 uCi of 14

C-AIB in RPMI
1640 plus 0.15% dl-FCS. At the time
points indicated, incubations were
stopped by placing the cultures in an

ice bath and adding 1.0 ml of ice cold
PBS-AIB. Cellular accumulation of AIB
was assessed after collecting cells on
glass microfiber filters and washing with
PBS—AIB. Data represent the mean i SEM
of six experiments. . . . . . . . . . . . . 78
Total AIB uptake. Following a 240 minute
preincubation (Materials and Methods)

Con A activated splenic lymphocytes

6

(10 x 10 ml) were washed and resuspended

in 200 ul of IM-l (Materials and Methods)

plus 0.10 uCi of 14

C-AIB and various
amounts of cold AIB. After a 120 second
incubation in a 370 water bath, incuba-
tions were stopped by adding 1.0 m1 of
ice-cold PBS-AIB and placing the cultures

on ice. Cell-associated radioactivity

xii

Figure

7A

7B

Page

was determined after cells were pelleted
by centrifugation through a 20% Percoll
gradient. Data represent the mean

i SEM of five experiments . . . . . . . . . 81
Na+ dependent AIB uptake. To determine
Na+ dependent AIB uptake parallel experi-
ments were performed as described for
total AIB uptake, except Na+ free medium
(IM—2) was used. The difference between
values obtained in IM—l and IM-2 are con-
sidered to represent Na+ dependent AIB
uptake. Data represent the mean i SEM

of five experiments . . . . . . . . . . . . . 81
Initial rate kinetics of Na+ dependent

AIB uptake into Con A activated splenic
lymphocytes. Suspensions of Con A
activated splenic lymphocytes (10 x 106/ml)
were incubated for 120 seconds in various
concentrations of unlabelled AIB plus 0.1
uCi 14C-AIB. Data obtained are plotted

by the Eadie-Hofstee method and represent
mean values from five experiments. Best
fit lines were obtained by linear regres-
sion of the experimental values with the
following r2 values; EFAD=.608, control=

.675, and high PUFA=.855. . . . . . . . . . 84

xiii

Figure

10

Page

14C leucine incorporation into TCA

precipitated protein of splenic mono-
nuclear cells from mice fed the experi-
mental diets. 1 x 106 splenic mono-
nuclear cells were incubated with optimal
mitogenic doses of Con A (solid lines)

or without mitogen (dotted lines). At
the times indicated, cells were removed,
washed, and precipitated with 10% TCA.
Each data point represents the mean

i SEM of duplicate observations from each
of 3-5 animals/treatment after a 70-84
day feeding period beginning when the
mice were 21i3 days of age. . . . . . . . . 95
Protein content of splenic mononuclear
cells from mice fed the experimental
diets. l x 106 splenic mononuclear

cells were incubated for the specified
times with optimal mitogenic doses of
Con A (solid lines), or without Con A,
(dotted lines). Data represent the mean
i SEM of five cultures/dietary treat-

ment. . . . . . . . . . . . . . . . . . . . 97

xiv

INTRODUCTION

Essential fatty acids (EFA) are structural components
of biological membranes and precursors in the synthesis
of prostaglandins and related lipids (1). In recent years
EFA metabolism in immunocompetent cells has been intensely
studied (2-5), leading to major advances in tracing and
identifying EFA metabolites and their biological effects.
In spite of this progress in EFA metabolism, basic informa-
tion about the relationship between dietary EPA and immune
functions is generally lacking. Several experimental and
clinical observations indicate the importance of obtaining
a better understanding of the relationship between dietary
EFA and the immune response. Some of these studies have
been reviewed previously (6), and others will be discussed
in the literature review section.

A few authors have attempted to look directly at the
relationship between dietary EFA intake and immune func-
tions. But one of these authors used diets that differed
in energy content (7), one did not use quantitative assess-
ments of immune responses (8), and others administered fatty
acids by injection, rather than in the diet (9-11). In
spite of these difficulties these authors generally con-

cluded that "PUFA excess" impaired immunity and "PUFA

deficiency" enhanced immunity (10).

My previous studies were not in agreement with these
reports (12). I found that an essential fatty acid de-
ficient (EFAD) diet reduced in vivo plaque forming responses,
while elevated levels of dietary PUFA had no adverse effects
on these responses (12).

The overall purpose of the experiments reported here
was to further clarify the relationship between dietary EPA
and the immune response and to investigate the cellular
defect underlying the EFAD induced impairment of in vivo

immune functions.

OVERVIEW AND OBJECTIVES

Previous experiments showed that the humoral immune
response of mice was sensitive to dietary EFAD (12). But
these findings differed from those reported by others.

The differences between experimental results, however, may
be a reflection of differences in experimental protocol.
For example, as opposed to subcutaneous injection (9—11),
or gavage (10), my approach was to administer the EPA as

a part of an isocaloric diet. Thus, this would seem to be
the more physiological approach.

I further studied the effects of the experimental diets
on another in vivo immune function, delayed type hyper-
sensitivity. These results were in close agreement with
those from the plaque assay experiments (13). Since re-
sults from both in vivo studies were roughly equal, the
remaining experiments were aimed at investigating the
mechanism by which EFAD alters immune functions.

The majority of the in vitro experiments performed
used mitogens to assess lymphocyte activation. The most
obvious reason for this is that large numbers of lympho-
cytes within a population respond to a particular mitogen,
as opposed to the small number that respond to a specific

mitogen. With a large number of responding cells, more

and varied types of experiments can be performed. Although
some differences exist between mitogen activation and antigen
activation, there are many similarities: (1) both activate
lymphocytes by binding to the lymphocyte surface, (2)
neither require internalization for cell activation, (3)
both initiate a series of cellular events that can be moni-
tored during activation, and (4) after stimulation, both
lead to some functional response by the activated cell
(i.e., antibody production, lymphokine production) that can
be measured under certain experimental conditions (14).
Significant differences do exist, however. Mitogens
activate large numbers of T or B cells, whereas, antigens
activate only a few cells. Also, the initiating event in
mitogen activation differs from antigen activation. Mito-
gens activate lymphocytes in a nonspecific manner, trigger-
ing the cell by cross-linking membrane glycoproteins with
certain sugar residues (14). Antigens, on the other hand,
trigger lymphocytes by binding to specific receptors on the
lymphocyte surface (15-16). On B cells, the antigen re-
ceptor is surface 19 (15);on T cells, the antigen receptor
is an Ig-like protein complex (16). In addition, the anti-
gen activated lymphocyte is subject to a number of regulating
influences in vivo that are not duplicated by the mitogen
activation in vitro. Furthermore, since certain lympho-
cyte cell functions do not require blast cell formation (3),

defects in lymphocyte function may not be detected in

standard blast cell formation assays.

Despite these drawbacks mitogens continue to be used
in a variety of studies investigating cell activation (17).
Mitogens are particularly useful in studies where basic in-
formation about the relationship between a treatment and
lymphocyte activation is lacking. On this premise it would
seem that mitogens are valid tools for use in the present

study.

Objectives

 

Several recent reports have raised the possibility that
dietary fat, particularly PUFA, may contribute to the etiol-
ogy of some infectious and chronic diseases by influencing
the immune system. The overall goal of this research, there-
fore, was to investigate the relationship between dietary
PUFA and the immune response. To accomplish this, several
specific objectives were outlined. The first objective was
to determine the effect of various levels of dietary PUFA,
including EFAD, on in vivo immune functions. The second
objective was to investigate the nature of this effect using
cell culture.

As the project developed it became clear that EFAD im-
paired in vivo immune functions, but elevated levels of
dietary PUFA had no effect. As a result, much of the cell
culture work was aimed at investigating the cellular defect
induced by EFAD in an effort to determine the role of EPA

in lymphocyte responses.

LITERATURE REVIEW

Essential Fatty Acids: Historical Perspective

 

Although EFA were first described over 50 years ago
by George and Mildred Burr, the precise metabolic func-
tions of EFA are still poorly understood (1,18). One reason
for this is that it took some time before EFAD was recog-
nized as a condition relevant for study in humans. An
early attempt by W. R. Brown, an associate of the Burrs
to induce EFAD in himself proved unsuccessful, although he
reportedly remained on a fat free diet for six months (18).
Sometime later, Hansen, et al.(l9) were successful in in-
ducing severe clinical symptoms of EFAD in otherwise healthy
infants. But because of the prolonged feeding period re-
quired to induce the symptoms, this only seemed to further
characterize EFAD as an improbable nutrient deficiency.
Following the development of total parenteral nutrition
(TPN) by Dudrick, Vars, and colleagues (20), however, the
first reports of EFAD began to appear in the literature.
First, reports of EFAD in infants receiving TPN appeared
(21), followed by reports of EFAD in adult patients after
prolonged TPN treatment (22).
Further study of the TPN patients revealed that the EFAD

accompanying TPN involved a complex interaction between

the EFAD solution and metabolism. It was found that changes
in metabolism associated with the continuous infusion of the
hypertonic glucose solutions played an important role in the
early onset of the EFAD symptoms (23). The high rates of
glucose infused resulted in a continuous and potent stimulii
for insulin secretion. With plasma insulin levels chronic-
ally elevated, adipose tissue lipolysis was severely reduced.
This effectively isolated TPN patients from their own
sources of EFA and many patients exhibiting clinical and
biochemical evidence of EFAD had normal stores of EFA in
adipose tissue (23). This was confirmed by several studies
and one solution to the problem has been the use of "cyclic
hyperalimentation", in which hyperalimenated patients under-
go periods of reduced, or limited infusion allowing for
some lipolysis and release of EFA (23). Another strategy
has been the use of solutions containing EFA (24). These
solutions, because they are isotonic, can be delivered
through a peripheral vein, thus avoiding placement of
a central catheter. At present, EFA-containing solutions
such as Intralipid are currently being used in clinical
situations requiring prolonged TPN.

One problem associated with studies of EFAD has been
defining its symptoms. Clinically, external symptoms
such as dermatitis, weight loss, reduced growth, and in-
creased water loss leading to increased water consumption
are commonly described consequences of EFAD (1). Other

abnormalities that have been described include erythrocyte

fragility (l), abnormal mitochondrial function (25), altera-
tions in membrane enzymes (26-28), and increased synthesis
of surface glyc0proteins (29). For the most part these
defects appear to be associated with the structural function
of EFA in biological membranes, but this is not firmly
established.

Another observation common to EFAD animals is fatty
liver (1). It has been proposed that fatty liver in EFAD
may be due to abnormalities in the formation of very low
density (VLDL) and/or in the secretion of VLDL (30,31).
However, a recent study by Huang and Williams (32) found
that hepatic triglyceride secretion was actually higher in
EFAD rats than in control rats. Thus, the origin of fatty
liver in EFAD remains controversial. But these studies do
point out that EFA function in normal lipid transport and
metabolism.

Another aspect of EFA metabolism receiving increased
attention in recent years is the oxidative transformation
of arachidonic acid (AA) to prostaglandins and related
lipids (3,4,33). The biochemistry and physiological sig-
nificance of these compounds in regulating immune func-
tions are discussed later in the literature review.

Major advances in EFA biochemistry have occurred
largely because of the development of sophisticated chromato-
graphic methods, particularly gas liquid chromatography

(GLC). The development of GLC in the early 1950s enabled

investigators to obtain more detailed structural information
about a wide variety of compounds, particularly fatty acids.
While understanding the chemistry of EFA is important,
much remains to be learned about the physiological roles of
EFA. In the remainder of this literature review, further
aspects of EFA metabolism and its physiological and immuno-

logical effects will be discussed.

Essential Fatty Acid Metabolism

 

Over 50 years ago Burr and Burr described the essential
fatty acids as a group of three polyunsaturated fatty acids:
linoleic acid,18:2w6; linolenic acid 18:3w3, and arachi-
donic acid,20:4w6 (1,18). Since that time, however, it has
become apparent that linolenic acid is not a dietary essen-
tial. In a recent investigation of linolenic acid require-
ments, rats were fed a linolenic acid free diet through
2-3 generations (34). Normal growth, organ weights, and
reproduction, but lower phospholipid lineolenic acid content
were found in the linolenate free group (34). It is known,
however, that linolenic acid incorporates in membrane phos-
pholipids of the central nervous tissue and retinal photo—
receptors (35), but as yet a unique function for linolenic
acid, or cellular defect induced by its absence, is lack-
ing.

Unlike linolenic acid (n=3) numerous studies over the

years have demonstrated the essentiality of the n=6, or

10

w6 PUFA (1,18). Mammals cannot synthesize 18:2w6, however,
if 18:2w6 is provided in the diet, most mammals can syn-
thesize 20:4w6 from this precursor by a series of elonga-
tion and desaturation reactions (Figure 1). Animals that
are exceptions to this are cats, lions, and turbot (18).
They must obtain their EFA as 20:4w6 because they lack the
necessary deltaG-desaturase enzyme (18). In most mammals,
therefore, tissue 20:4w6 content results from dietary in-
take of 20:4w6, or conversion from 18:2w6 (1,18).
Intracellular pools of free 20:4w6 (AA), however, are
very low. Therefore, cleavage of AA from membrane phos-
pholipids is the initiating step in the synthesis of PG
and related lipids (2,6,36). This membrane cleavage ap-
pears to be the rate-limiting step in further AA metabolism
and it is catalyzed by phospholipase A2 in most tissues
(2,6,36). In platelets, however, this reaction is catalyzed
by either phospholipase A2 (37), or phospholipase C and di-
glyceride lipase (38). The newly released AA is then
rapidly metabolized by the first reaction in prostaglandin
synthesis, PG endoperoxide synthetase (33,36). This en-
zyme complex is composed of a cyclooxygenase which catalyzes

the cyclization of AA to form a C hydroperoxy, C9, C11,

15
endoperoxide called PGGZ. The second component of the PG
endoperoxide synthetase complex is a peroxidase which

catalyzes the reduction of the C15 hydroperoxy group to a

hydroxyl, forming PGH2 (33,36). At this point PG synthesis

11

18:2w6 18:2w6

<? <>
18:2w9 18:3w6

<9 <>
20:2w9 20:3w6

o O
20:3w9 20:4w6

<>
20:4w6
Q Q o o

leukotriene Q 5
CL
b
14 HPETE
O
12 HETE
D

'B

b.

TXA2

O
TXB2

Figure l.

Unsaturated fatty acid

18:3w3
18:4w3
20:4w3

20:5w3

metabolism.

12

may proceed to form PGEZ, PGFZa' or in the case of vascu-
lar endothelial cells PGIZ, or prostacyclin (Figure l).
Platelets metabolize PGH2 through the enzyme throm-
boxane synthetase to form a highly unstable proplatelet
aggregating factor called thromboxane A2 (TXAZ). TXA2 is

rapidly degraded to its inactive form TXB (33,36).

2x
Recent studies have shown that ingestion of a diet
high in 20:5w3, such as the high fish diet of Greenland
eskimos, has profound effects on tissue PUFA levels and
EFA metabolism (35). Serum lipids from these eskimos con-
tain elevated levels of phospholipid 20:5w3, and rela-
tively low levels of 20:4w6. They have a low incidence of
cardiovascular heart disease, and markedly elevated clott-
ing times (35). This is believed to result from the sub-
stitution of 20:5w3 for 20:4w6 in platelet phospholipids.
As a result of this substitution, TXA3 is produced in stimu-
lated platelets instead of TXA2 (Figure 1). In contrast
to the potent proaggregating effects of TXAZ, TXA3 impairs
platelet aggregation. Overall, incorporation of 20:5w3
into platelet phospholipids reduces platelet aggregation by
at least 3 ways: (1) 20:5w3 competes with remaining 20:4w6
for cyclooxygenase, thus, increased 20:5w3 decreases 20:4w6

metabolism to TXA (2) increased 20:5w3 increases the pro-

2'
duction of antiplatelet aggregating TXA3, and (3) 20:5w3
is metabolized in endothelial cells to PGI3, which appears

to be a potent antiaggregating PG, possibly even more potent

13

than the normally produced PGIZ.

Recent progress in the biochemistry of EFA metabolism
has led to the elucidation of an entire new pathway of EFA
metabolism. This pathway, which is initiated by the enzyme
lipoxygenase, metabolizes AA to monohydroperoxy and mono-
hydroxy eicosatetraenoic acids (HPETE and HETE). Some of
these recently described compounds are potent chemotactic
agents (2-4), and they appear to be generated early in the
activation of immunocompetent cells (2-4), possibly func-
tioning as intracellular messengers (2-4). One of these,
lZ-HPETE, enhances PMN hexose transport (39,40).

Other compounds formed through the lipoxygenase path-
way are the dihydroxy metabolites of AA (2,3,41,42). Since
most of the evidence for the existence of these dihydroxy
AA compounds comes from studies with leucocytes they have
been named Leukotrienes (3). They appear to influence a
variety of functions in immunocompetent cells and one,
Leukotriene C, appears to have the biological function of
slow-reacting substance of anaphalaxis (SRSA) (2,4,41,42),

confirming an earlier report that SRSA was derived from AA

(43) .

EFA, Prostaglandins and Cells of the Immune System

 

A large body of evidence indicates that macrophages

of several species actively metabolize AA to a variety of

l4

metabolites, including most PG and TX (2-4). The data for
lymphocytes, however, is less clear (3). One problem in
all studies of purified cell populations is obtaining that
p0pulation without contamination by other cell types (2).
Despite this difficulty, the bulk of the available evidence
indicates that macrophages are the primary source of PG
among mononuclear cells (2-4). Even though macrophages
synthesize greater quantities of PG and TX than lympho-
cytes, membrane AA content is approximately equal between
these two heterogenous cell populations (3). The sig-
nificance of this is unclear, but since lymphocytes release
significant amounts of AA following activation, portions of
this released AA may be taken up by macrophages and metabol-
ized to TX or PG (3). This could provide a basis for
lymphocyte regulation by macrOphage PG synthesized from AA
released from activated lymphocytes. But this type of self
regulation is entirely speculative.

Regulation of lymphocyte function by exogenous PG has
been extensively reported (2-5,44). Many of these reports
indicate that PG inhibit immune reactivity in vitro (3,4,-
45). Much of this evidence, however, was obtained in cul-
tures in which exogenous PG were added, often to concentra-

5 4

tions of 10- -10- M (3,45). If PG are physiologic effec—

tors of lymphocyte functions, then effects should be seen

7

at doses of PG in the range of 10-9-10- M (3,46). But even

within these physiological dose ranges the effects of PG

15

on lymphocyte responses are inconsistent, with results rang—
ing from 75% inhibition of lymphocyte function to no effect
at all (3,47,48). Since PG bind to serum proteins it has
been suggested that the variation in reported results may
reflect differences in experimental methods (3).

Besides adding PG to cultures,numerous investigators
have added inhibitors of PG metabolism to block PG synthesis
and test cell functions in the absence of PG (3,44). Indo-
methacin, which blocks cyclooxygenase, has been shown to
significantly increase T cell blast formation (3,44). But
indomethacin may affect other aspects of lymphocyte activa-
tion, making the results difficult to interpret on the
basis of PG inhibition alone (3,49). Also, one recent report
has shown that indomethacin augments mitogenesis only when
suboptimal doses of the mitogen are used. When Optimal
doses are employed, indomethacin has no effect (3,50).
Another aspect of this problem lies with the heterogeneity
of T cells. It is now clear that T cells are a functionally
diverse pOpulation of cells and that the various T cell
subpopulations may differ in their sensitivity to PG (3).

The effect of PG on T cell subpopulations is particu-
larly important in evaluating the evidence relating PG
effects on antibody production. Several studies have shown

that PGE impairs in vitro antibody synthesis (3,4,51,52).

2
There is now experimental evidence showing that if PGEl

is present in the first 24 hours of the in vitro culture,

16

antibody production is increased at 5 days (3,53). Another
study showed that increased PG production in vivo immed-
iately following antigen injection correlated with higher
antibody production at 5 days (3,51). One possible eXplana-
tion for these results is that early in the antibody response
PG may either enhance T cell helper function (TH) or reduce

T suppressor (TS) function, thus enhancing antibody pro-
duction.

Of central importance in discussing the effects of PG
on T and B cell function is the role of the macrophage in
PG production (2-4). As already mentioned, macrophages
constitute the major PG producing immunocompetent cell (2-4).
There is also evidence showing that macrophage PG production
can influence the functional activity of T and B cells as
well as macrophages themselves (2-4). Because of the require-
ment for macrOphage participation in a number of lymphocyte
responses it is possible that macrophages modulate T and B
cell responses through PG synthesis.

In closing, sufficient data exist indicating that
prostaglandins and other oxygenated metabolites of AA in-
fluence macrophage and lymphocyte function. With more in-
formation about the heterogeneity of macrophage and lympho-
cyte suprpulations, future studies should be able to dis-
cern the major prostaglandin producing and prostaglandin
sensitive cell subpopulations and relate this information

to the effects of prostaglandins on immune functions in

17

vivo. At this point it seems fairly clear that rather than
initiate monocyte responses, PG and related lipids regulate
or modulate these responses. This implies then, that EFA
metabolites regulate immune functions. It is not possible,
however, to tie together EFA intake, PG production, and
regulation of the immune response. Part of the problem is
the relationship between EFA intake and PG production is
not well understood (54). And, as already discussed, the
regulation of macrophage and lymphocyte subpopulations by

PG and related lipids is just beginning to be understood.

EFA and the Lymphocyte Membrane

 

The lymphocyte plasma membrane is approximately 50%
lipid by weight (17,55). About 70% of this lipid is
phospholipid, with phosphotidlylcholine and phophotidyl—
ethanolamine comprising almost 70% of total phospholipids
(55). Studies by Ferber and Resch (55), indicate that
the fatty acid composition of the resting lymphocyte
membrane has a lower AA content than liver cells, indi-
cating that the resting lymphocyte has a more rigid mem-
brane structure. Following lymphocyte stimulation, how—
ever, the membrane fatty acid composition changes dram-
atically (55). Within the first hour of stimulation,
enhanced incorporation of AA into membrane phospholipids
occurs (55). This rapid and specific transfer of AA

to the number two position of membrane phospholipids is

18

catalyzed by a plasma membrane enzyme, acyl Coa:lysoleci-
thin aclytransferase (55). This enzyme exhibits a high
affinity for AA, and its increased activity following
lymphocyte activation changes the fatty acid composition
of the activated lymphocyte to a more unsaturated, more
fluid, state (55).

It has been postulated that a more fluid membrane
may create a favorable environment for patching and capping
of cross linked membrane receptors, thus enhancing lympho-
cyte activation (55). But currently the role of fluidity
in receptor movement is unclear (17). Also, patching and
capping are not obligatory events in lymphocyte activa-
tion, indicating that increased fluidity may not be es-
sential to cell activation (17). But these studies do
point out that regulation of membrane EFA in lymphocytes
may differ from that of other cell types. This suggests
that lymphocyte membrane EFA may perform different functional
duties within the membrane than EFA in other cells. The
effects of EFAD on lymphocyte membrane functions has not
been studied. But in light of evidence indicating the
importance of EFA in the lumphocyte membrane, this in-
formation could be important in understanding much of

how EFA function in the lymphocyte membrane.

19

Lymphocyte Activation: Phospholipid Metabolism

 

Membrane lipids comprise the structural element of
biological membranes. Since 70% of the lymphocyte membrane
lipid component is phospholipid (55), changes in phospho-
lipid metabolism are clearly important in membrane function.
It is particularly significant, then, that one of the
earliest events in lymphocyte activation is enhanced mem-
brane phospholipid turnover (17,55-57). Although all re-
ports are not in total agreement, enhanced turnover of
phosphatidylcholine (17,55,56), and phosphatidylinositol
(57) have been most often reported. Although the precise
function of enhanced phospholipid metabolism is not known,
it appears to be correlated with DNA synthesis in several
experimental systems (17,55-57). In studies of 14C-choline
uptake into total lipids following Con A activation, it was
shown that incorporation of the label into lipids: (1)
occurred early following exposure to Con A (within 3 hours),
(2) did not occur following exposure to a nonmitogenic
lectin, (3) was proportional to mitogenic doses of the
lectin, and (4) could be inhibited by a competitive in-
hibitor of Con A, (a-methyl-mannoside) (56).

As previously mentioned, membrane phospholipid fatty
acid metabolism is also stimulated in activated lympho-
cytes (17,55). Results from several experiments indicate
that phospholipid fatty acid turnover significantly affects

membrane physical properties, and may affect the function

20

of membrane bound enzymes and receptors (58—59).

A direct link between enhanced phosphotidylinositofl\\
turnover and a membrane associated enzyme has recently beeh
reported (59). In this report a Ca2+ dependent protei
kinase (termed protein kinase C) was described. This enzyme
was shown to be active when associated with membranes angd"
dependent on diacylglycerol, unsaturated in the number 2
position (59). Although the physiologic substrates of:\\
protein kinase C are yet to be determined, phosphorylatiod
of a number of cytoplasmic proteins accompanies lymphocyte
activation (59). This may be a first step in unravelling
the physiological significance of early phospholipid turn-
over in lymphocyte activation. Even without a complete
definition of the role of protein kinase C in lymphocyte
activation, the association cf phospholipid turnover and
enzyme activity suggests the importance of phospholipid

metabolism during the lymphocyte activation process.

EFA and Lymphocyte Membrane Receptors, Enzymes, and Transport

 

While lipids are largely responsible for membrane struc-
ture, recent evidence suggests that specific membrane fatty
acids may exist in "domains" (60), and protein lipid inter-
action within these "domains" may determine membrane protein
function (60,61). This is particularly significant in

lymphocytes since the initiating event in lymphocyte

21

activation is the binding of the external ligand to the
lymphocyte membrane receptor (17). Several reports indicate
that membrane receptors can be affected by alterations in
membrane lipids (29,62-64). For example, it has been shown
that EFA are required in the maintenance and induction of
prolactin receptors (62). This implies that changes in
membrane fatty acid composition resulting from EFAD directly
affects the expression of a specific receptor. EFAD has

125I-Con A binding to purified

also been shown to effect
liver plasma membranes, but the effects are opposite to
those observed with prolactin receptors. In this case
125I-Con A binding was increased in the plasma membrane
preparations from animals fed the EFAD diet compared with
preparations from animals fed a high fat diet (29). Total
number of binding sites was increased, while binding affinity
was unaffected by the diet. The increased binding ap-

peared to be due to an increase in membrane surface glyco-
protein content in the EFAD cells (29). This suggests

that an increase in glycoprotein synthesis accompanies EFAD,
but the physiologic significance of this is unknown.

EFAD induces changes in the number of Na+, K+-ATPase
units (28) and enzyme activity (27). Since increased
activity of Na+, K+-ATPase occurs early in lymphocyte ac-
tivation, any external factor which affects its activity

may affect lymphocyte activation. The bulk of available

evidence indicates that mitogens bound to lymphocyte

22

membranes activate Na+-K+-ATPases, either by a direct
interaction between the mitogen receptor complex and the
enzyme, or indirectly by some general membrane perturba-
tion induced by mitogen binding (17). Since oubain blocks
all aspects of lymphocyte activation, regardless of stimu-
lant used, it appears that Na+, K+-ATPase activity is vital
to lymphocyte activation (17). At present it is unknown if
the changes in Na+,K+-ATPase induced by EFAD affect the
function of the enzyme in lymphocyte activation. It seems
plausible though, that EFAD could alter lymphocyte Na+,
K+—ATPase activity and this could contribute to reduced
lymphocyte responses observed in EFAD animals.

Another series of membrane associated events occurring
early in lymphocyte activation is enhanced metabolite
transport (17). Enhancement of glucose transport during
lymphocyte activation has been extensively reported (17).
This increased transport appears to be due to increased
transport Vmax values, but not Km, in mitogen activated
lymphocytes (17). The extent to which mitogens increase
glucose facilitated diffusion has been difficult to assess,
however, because in most studies only a portion of the cells
in the culture are activated by the mitogen. A large num-
ber of cells remain in the resting, or unstimulated state.
Thus, total transport values reflect the sum of transport
into these different populations of cells within an indi-

vidual culture.

23

Whereas the role of increased membrane fluidity in
receptor movement may be less than originally thought, it
now appears that increased fluidity correlates with in-
creased glucose transport (65-67). This has been demon-
strated in intact cells (67) and purified membrane vesicles
(65,66). With experimental evidence supporting a role for
membrane fluidity in glucose transport, it seems possible
that this transport may be affected by EFAD.

Along with enhanced glucose transport, enhanced amino
acid transport also occurs early in lymphocyte transforma-
tion (68-71). One of the most thoroughly studied of the
multiple amino acid transport systems is the so called
"A“ system for neutral amino acid transport (70,71). This
system has been characterized using the labelled non-
metabolizable amino acid analogue a-aminoisobutyric acid
(AIB). It has been clearly established that enhanced AIB
transport occurs early in lymphocyte activation and the
total amount of AIB taken up by the lymphocyte is the sum
of that transported by a saturable Na+ dependent component,
and that incorporated by Na+ independent diffusion (67-71).
In studies using Erhlich's ascites tumor cells grown in
animals fed a diet high in saturated fat (the diet was
EFAD (72)) it was shown that AIB transport is sensitive to
dietary fat intake (73). The Km value for AIB uptake into
tumor cells from mice fed the EFAD diet was almost double
that of tumor cells from mice fed EFA sufficient diets

(73). This shows that EFAD alters AIB transport into tumor

24

cells and indicates that this could occur in other cell

types, including lymphocytes.

Lipoproteins and Immunity

 

Several recent reports have proposed that lipoproteins
inhibit immune functions (74-81). Among the different lipo-
protein classes, a subpopulation of LDL, termed LDL-In,
was reported to contain the highest immunoinhibitory effect.
In these studies human LDL-In was found to suppress mitogen
responses and MLR of mouse spleen cells as well as in vivo
plaque forming responses (74-76). Further experiments
revealed that the biological activity of the LDL-In resided
in the protein component, as the immunoregulatory functions
remained after extraction with solvents (78,79). This in-
dicated that the interaction of the LDL-In particle with
the lymphocyte surface and the resulting impaired responses
were independent of the lipid portion of the lipoprotein
complex. One difficulty with these reports is that the
LDL-In is a trivial name and it is not completely clear
what the composition of LDL-In is. If the immunoregulatory
prOperties of LDL-In reside solely in its protein moiety,
then it should be fairly easy to assess this experimentally
as the protein component of LDL is 98% apoprotein B (apo B)
(82). This has recently been looked at in a study assess-

ing the immunoregulatory properties of apOprotein B and

25

apoprotein E (apo E) (77). It was found that both apo B
and apo E inhibited PHA stimulation of human peripheral
lymphocytes. On a molar basis, apo E was approximately
four times more effective in inhibiting PHA mitogenesis
than apo B (77). It was suggested that apo E and apo B
both bind to a common receptor on the lymphocyte surface
("immunoregulatory receptor") and this binding initiates the
suppressive effect. Internalization of the lipoproteins,
although it occurs, is not required for the immunosuppres-
sion as LDL linked to Sepharose also was immunosuppres-
sive (77).

Another recent study has looked at the immunosuppres-
sive effects of mouse lipoproteins (81). In general, the
results agree with the previous reports, except these
authors found that HDL was the predominant immunosup-
pressive lipoprotein, as opposed to LDL or an LDL subfrac-
tion.

In contrast to these reports, there is also evidence
indicating that LDL enhance in vitro immune responses (83).
Also, one report has found that oxidized LDL inhibits
mitogenesis, but non-oxidized LDL has no inhibitory ef-
fect (84).

Another approach employed has been to use serum from
animals fed different levels of dietary fat. Serum from
rats fed a high PUFA diet was found to inhibit mitogen

responses (85). Further experiments showed that the

26

lipoprotein fraction in general was responsible for the
inhibition. This study also showed that spleen cells from
rats fed a low fat diet (about 5% of calories from linoleic
acid) responded better to mitogens than spleen cells from
mice fed a high PUFA diet (about 40% of calories from strip-
ped corn oil): and both groups of cells responded best to
mitogens when cultures were performed in media supplemented
with serum from rats fed the low fat diet (85). This sug-
gests that the high PUFA diet may reduce immune functions,
possibly by affecting lipoproteins. No evidence for shifts
in the lipoprotein profiles between the two dietary treat-
ments was presented, however. On the other hand, in short
term studies, triglyceride rich serum from healthy indi-
viduals after consuming a high fat meal had no inhibitory
effect on lymphocyte mitogenesis (86).

At present, the physiological significance of immuno-
regulation by lipoproteins is unclear. The bulk of the
available evidence comes from studies in which exogenous
lipoproteins have been added to in vitro cultures, with
little discussion of how lipoproteins could regulate immune
functions in vivo. Lipoproteins are not synthesized by
cells of the immune system, nor is their synthesis in-
duced by immunization. Thus, it is difficult to under-
stand how lipoproteins could regulate immunity.

But dietary studies have shown that serum from animals

fed moderate to high levels of PUFA impair mitogen responses

27

(85). This is puzzling as animals fed high PUFA diets are
not immunosuppressed (12,13). In agreement with this,
serum from patients with type IV or type V hyperlipoprotein-
emia inhibits mitogenesis, but there is no evidence that
these individuals are immunodeficient (87). In the ab-
sence of any in vivo immunodeficiency it is difficult to
understand the physiological significance of in vitro im-
munoregulation by serum from animals fed high PUFA diets.
On the other hand numerous in vitro studies have shown
that serum or lipoproteins from animals fed moderate to
high levels of PUFA inhibit in vitro responses (74-81,
85). In particular, studies with lipoproteins have shown
immunoinhibitory effects at physiological concentrations,
indicating that under normal conditions lipoproteins may
affect lymphocyte function (88). Studies correlating
dietary PUFA intake, lipoprotein profiles, and immunity

are needed to put this complicated problem in order.

EFA and Disease

 

Epidemiological evidence indicates that societies
that consume low fiber diets relatively high in fat, par-
ticularly of animal origin, (Western type diet) have a higher
incidence of colon and breast cancer than societies con-
suming a high fiber diet lower in animal products and fat
(89-91). On the other hand, lower incidences of stomach

and liver cancer are seen in societies consuming the

28

Western type diet (90). Immigrant studies show that mi-
grating peoples assuming the customs and eating habits of
their new home also assume the types of cancer characteris-
tic of that area of the world (90,91). All this points to
a relationship between dietary habits and certain cancers.
But the picture is complicated by more than eating habits.
Other factors such as smoking, environmental pollution,
alcohol consumption, caffeine consumption, and stress have
been implicated (90,91). Continued efforts in both epid-
emiological and experimental investigations are needed to
define the relationship between dietary patterns and sus-
ceptibility to certain cancers.

Experimental studies, however, do indicate a relation-
ship between dietary fat and certain types of tumors. Over
the past 15 years a number of reports have appeared demon-
strating a relationship between dietary PUFA intake and
increased appearance and growth of chemically induced and
transplanted mammary tumors in rodents (92-95). Also, in
the past several years a relationship between dietary fat
and the induction of chemically induced colon cancer has
been described (96). It has been proposed that dietary fat
acts as a tumor promoter, and that this promotion may involve
the immune system (94), presumably by affecting tumor
sureillance.

Although dietary PUFA are correlated with the develop-
ment of chemically induced and transplanted mammary tumors,

it appears that tumor growth is approximately equal

29

regardless of the PUFA level fed, as long as the EFA re-
quirements are met (93,95). With chemically induced colon
cancer, a high fat diet similar to that of high risk popu-
lations correlates with an increased incidence of chemi-
cally induced tumors (89,96). While these studies indi-
cate a link between dietary fat intake and certain types of
tumors, there is no evidence that the immune system is
involved.

Current progress in tumor immunology indicates that
the concept of a thymus-dependent tumor surveillance mechan-
ism is undergoing revision (97). A better understanding of
anti-tumor surveillance is necessary to determine if the
immune system is implicated in the relationship between
dietary fat and cancer.

In an earlier review several diseases were associated
with dietary EFA intake (6). In one of these diseases,
multiple sclerosis (MS), some additional work has been
reported in the past few years. The basis of the rela-
tionship between dietary EFA intake, immunity, and MS lies
with the observation that MS patients respond to immuno—
suppressive agents (98,99), and they have low serum EFA
levels (100, 101). These observations, together with those
from animal models of MS, indicated that EFA may reduce the
symptoms of MS (102-104). Results from clinical studies,
however, showed that patients receiving EFA supplements

(3-8 g/day linoleic acid) faired no better than their

30

non-supplemented free-living fellow patients (105-107).
Other reports previously discussed (6), have also been
inconclusive. Thus, the link between EFA and MS remains
to be proven.

Other topics previously reviewed include EFA, immunity,
and transplantation, and cardiovascular disease. Few
reports further exploring links between dietary EFA, the
immune response, and these two conditions have appeared in
the literature in recent years.

Also discussed were reports looking at the direct
relationship between EFAD and resistance to infection (6).
Despite the possibility that there may be a relationship
between the EFAD of TPN patients and their increased sus-
ceptibility to infection, the most recent report of a direct
relationship between EFAD and increased susceptibility to

infection was in 1948 (108).

EXPERIMENTAL
Effects of Essential Fatty Acid Deficiency, and
Diets High in Polyunsaturated Fatty Acids, on

Delayed Type Hypersensitivity in Mice

Introduction

 

Several recent reports have proposed that PUFA regulate,
or modulate, the immune response by specifically impairing
T cell responses (9-ll,109,110). For example, mice receiv-
ing daily injections of PUFA demonstrated impaired allo-
graft rejection (9-11) and cytoxicity (9-11); while addi-
tion of PUFA dissolved in ethanol to lymphocyte cultures
reduced in vitro T cell mitogenesis (111). In only a few
cases, however, has the relationship between dietary PUFA
and in vivo T cell responses been evaluated. But these
results are difficult to interpret due to improper diet
formulation (7), and lack of quantitative presentation of
immune reSponse data (8). Effects of dietary saturated
fatty acids and PUFA on in vitro T cell transformation have
also been reported, however the results are contradictory
(110,112).

I showed earlier that in vivo antibody-mediated im-

munity was not impaired in mice fed a high PUFA diet (50%

31

32

corn oil (C0)). Reduced antibody responses were observed
only in mice fed an essential fatty acid deficient (EFAD)
diet 0% (CO) (12). The assay used in these experiments
(Jerne plaque assay) measured the number of B cell progeny
secreting antibody in response to immunization with sheep
red bloom cells (SRBC). Since it is well known that this
B cell response requires the active participation of T
helper (TH) cells (113), my data indicated that at least

one aspect of in vivo T cell function (T cell function)

H
was not impaired in mice fed a high PUFA diet.

Before using in vitro assays to assess mechanisms
whereby dietary PUFA may affect T cell function, I wished
to further determine if the dietary manipulation affected T
cell responses in vivo. Taking this approach, I assessed
the T cell dependent delayed-type hypersensitivity (DTH)
response as described by Vadas, et al (114).

The DTH response is an inflammatory reaction resulting
from a complex interaction between sensitized T cells (TD),
which bear the Ly 1 surface alloantigen, B cells, and macro-
phages (114-116). Although both TD cells and macrophages
are involved in this response, macrophage participation
depends upon the specific activation of sensitized TD
cells (114-116). In the experiments reported here, I

tested the DTH response of mice fed the EFAD (0% CO),

control (13% CO), or high PUFA (50% CO) diets.

33

Materials and Methods

 

Animals and Diets

 

Male A/J mice 21 days of age were housed, five per
cage, in plastic solid bottom cages in a temperature
(24 1°), light (12 hours light per day), and humidity con-
trolled room. Diets and water were provided ad libitum.
Three diets were formulated on an equal energy basis,
providing 0, 13 or 50% of dietary energy from corn oil,
and 24% of dietary energy from casein. Complete composition

of the diets is in Table 1.

Experiment 1. To determine effects of the EFAD and

 

high PUFA diets on DTH, mice were fed the diets for 70 days.
To measure DTH, a modification of the method of Vadas et al.

(114) was used. Body weights were also assessed.

Experiment 2. Two aspects of the relationship between

 

EFAD and TD cell function were examined: 1) the length of
time required to observe losses in the DTH response after
introduction of the EFAD diet, and 2) the length of time

required to reverse these losses by refeeding the control
diet. Mice were fed the EFAD diet (0% CO) or the control

diet (13% CO) for 77 days. A third group of mice was fed
the EFAD diet for 63 days and then switched to the control
diet for 7 or 14 days. In vivo DTH was assessed at 7 day

intervals from 17 to 77 days.

34

Table 1. Composition of Diets.

 

 

 

 

Diet
Ingredient EFAD Control High PUFA
Basal mix, g1 35 35 35
Glucose, g 65 54 20
Corn oil, 9 0 5 20
% Energy from
corn oil 0 13 50

 

 

lBasal mix contained (in g per 35 g basal): casein, 20.0;

methionine, 0.3; vitamin mix, 0.4; choline chloride, 0.2;
mineral mix, 4.0; and cellulose, 10.1.

35

Determination of DTH. DTH was measured byaamodification

 

of the method of Vadas, et al (114). Mice were sensitized
to dinitrofluorobenzene (DNFB) by spreading 50 ul of a 2%
solution of DNFB onto shaved areas (2 cm diameter) of their
backs and abdomens. Five days later 10 ul of a 1% DNFB
solution was applied to the right ear, and a nonspecific
agent (turpentine) was applied to the left (control ear.
Ten hours after the second administration of DNFB, mice

were injected with 2 uCi of 125

I-iodo-deoxyuridine (5
Ci/mmol, Amersham). Twenty—six hours after the second ad-
ministration of DNFB mice were killed and both ears were

removed and counted in a gamma counter. Results were cal-

culated as follows:

cpm (test ear) - cpm (background)
cpm (control ear) - cpm (background)

 

Stimulation index =

Data analysis. All data were treated statistically

 

by either the student's t-test, or one way analysis of
variance with treatment differences assessed by Tukey's
test (117).

Results

Experiment 1. As I had observed previously (12), body

 

weights of mice fed the EFAD diet for 70 days were reduced

compared to weights of mice in the control and high PUFA

36

groups (Table 2). Food intake was not measured, but in
an earlier study (12) mice fed the high PUFA and control
diets consumed slightly, but not significantly, more than
mice fed the EFAD diet. DTH was impaired in mice fed the
EFAD diet (Table 2), but the response of mice fed the
high PUFA diet did not differ from the controls. These
findings are in close agreement with my previous report on
the humoral response (12), and demonstrate that when rela-
tively high levels of PUFA are fed, no inhibitory effects
on in vivo immune functions are seen. On the other hand,
when mice consume an EFAD diet, in vivo immunity is im-

paired.

Experiment 2. To examine the temporal relationship

 

between consumption of an EFAD diet and in vivo T cell
function I monitored the DTH response of mice over a 77

day period (Figure 2). After 42 days, a significant re-
duction in DTH was observed in mice fed the EFAD diet.

This reduction remained fairly constant at about 75% of the
control response for the remainder of the experiment (77
days).

To determine if the inhibitory effects of the EFAD
diet on DTH were reversible, two groups of mice fed the
EFAD diet for 63 days were switched to the control diet
(13% CO). After only 7 days of refeeding, the DTH res-

ponse was completely restored (Figure ), and even exceeded

37

Table 2. Body Weight and Delayed Type Hypersensitivity
Response of Mice Fed the Experimental Diets for

 

 

 

 

10 Weeks.1
% Dietary Energy From Corn Oil
EFAD Control High PUFA
Body weight2 (g) 23.0:o.7a 27511.2b 29.8:1.4b
313 1.73:0.13a 2.51:0.20b 2.33:0.13b

 

 

1Means for 10 mice. Numbers with different superscripts are
significantly different (p 0.05).

2Mean initial body weights were 9.6 grams.

3 cpm 125I-iododeoxyuridine incorporated into stimulated

SI = ear-cpm (background

 

-_

cpm leI-iododeoxyuridine incorporated into control

ear-cpm (background)

See Materials and Methods for a description of the method.

Figure 2.

38

Delayed-type hypersensitivity response of
mice fed an essential fatty acid deficient
diet (0% corn oil). Response expressed as a
percentage of control (13% corn 011) response.
Each point represents the mean of 10 mice.

A = significantly different from control

(p < 0.05); B = significantly different from
refed (p < 0.05).

”awqakm 1.... -L-7-v-L—L—— . _ ,

 

39

(% of Controls)

Delayed -Type Hypersensitivity ReSponse

 

 

 

 

 

Soillllilllilhk
«m1
moi
Til. mw>c
. ..... . :12. 06:26. So»
Nmi
nu . . 1 _ _ _ d . . m

7— NH mm mm AM A@ mm mw #6 Nu

dam A 6331

wwocnm N.

40

the control response by about 15%. The DTH response after
two weeks of refeeding was approximately equal to the con-
trol response. At both time points the DTH response of mice
remaining on the EFAD diet was significantly reduced com-
pared to the response of the control or the refed groups.
Thus, the deleterious effects of the EFAD diet on DTH are
reversible, with recovery occurring rapidly (within 7

days) after refeeding a diet containing EFAs.

Discussion

 

This report extends previous observations (12) related
to the influence of EFAD and high PUFA diets on in vivo
immune responses. In both studies, consumption of an EFAD
diet (0% CO) depressed in vivo responses, whereas consump-
tion of a high PUFA diet (50% CO) for an extended period
of time (70 days) had no adverse effects on in vivo res-
ponses. These findings, particularly those related to PUFA
and in vivo T cell responses, do not agree with the con-
clusions of others (9-11). Several reasons for these
differences are possible.

Others reported a consistent pattern of immunosup-
pression when mice fed stock diets were injected daily with
10 ul of PUFA (linoleic or arachidonic acid) (9-11). Based
on a daily intake of 12 kcal/day (12), and assuming 90%

absorption of PUFA (118), mice fed my control diet (13%

41

of energy fromCX)(60% linoleic acid)) consumed about 94 mg
linoleic acid/day and mice fed the high PUFA diet received
about 360 mg linoleic acid/day. If the daily intake of
stock diet fed to mice injected with PUFA contained ap-
proximately 100 mg PUFA, then injection of 10 mg PUFA in-
creased intake only 10%. Since mice fed my high PUFA diet
increased PUFA intake by nearly 300% without a reduction in
in vivo T cell function, it is difficult to attribute the
immunosuppression observed after injection of PUFA to
PUFA alone. Although two reports suggest that dietary PUFA
suppress in vivo T cell responses, interpretation of these
studies is difficult because in one case the experimental
diets were grossly different in their nutrient density (7),
and in the other, quantitative assessments of T cell res-
ponses were not used (8).

Conflicting reports have appeared on effects of high
PUFA diets on T cell mitogenesis in vitro (110,112). Erick-
son, et al (110), reported that a high PUFA diet (approxi-
mately 44% of energy from safflower oil) impaired mitogenic
responses of spleen cell from young mice to Con A. How-
ever, Ossman, et al (112), presented data showing that Con
A responses of spleen cells from adult mice increase with
increasing amounts of corn oil in the diet. Another in
vitro approach has been to add PUFA, or PUFA bound to
albumin, directly to cell cultures. Addition of PUFA has

been shown to decrease (119), increase (120), or have no

42

effect on mitogen responses (121). Thus, these data have
not helped clarify affects of high PUFA intake on T cell
function.

It has also been prOposed that PUFA "deficiency" is
immunOpotentiating (10) because graft rejection was faster
in mice fed an autoclaved (supposedly to specifically destroy
PUFA double bonds) diet. I never found the EFAD diet (which
could be termed "PUFA deficient") to be immunopotentiating
in any in vivo study. On the contrary, in vivo immunity
has been consistently depressed in mice fed the EFAD diet.
Furthermore, since autoclaving would be expected to destroy
several water soluble vitamins, in addition to PUFA double
bonds, it is difficult to see how this diet could enhance
the immune response.

I found that both DTH and antibody-mediated responses
were reduced within 21-28 days after feeding the EFAD diet.
This reduced capacity to respond in vivo plateaued at ap-
proximately 60-75% of the control response, where it re-
mained for the duration of both experiments (70-77 days).
Even though I have not studied the effect of EFAD past 77
days, it appears that deSpite an extended period of EFA
deficiency, the immune system maintains some capacity to
reSpond in vivo. Also, after 63 days of consuming the
EFAD diet, the DTH response of mice switched to the control
diet was rapidly (7 days) restored to control levels. This

is consistent with my previous findings (12), and indicates

43

that EFAD-induced alterations in immune functions are
rapidly reversed by feeding a diet contaiing EFAs.

Results presented here, and in my previous report
(12), indicate that feeding mice high levels of PUFA has
no adverse effects on in vivo immune functions. On the
other hand, in both reports, in vivo immune functions
were reduced in mice fed the EFAD diet. Both in vivo anti-
body-mediated immunity and DTH responses result from Com-
plex interactions among several cell types, including T and
B lymphocytes and macrophages (113,114). Because of the
complexity of these responses, it is not possible from
available data to identify the precise cell population
affected by the EFAD diet. Further studies are needed to
determine the cell type and mechanism(s) responsible for
the effects of EFAD on immune functions.

Finally, my results do not agree with previous reports
proposing that PUFA decrease immunity (10,110) and "PUFA
deficiency" increase immunity (10,110). However, dif-
ferences in experimental approach probably account for
the disparity in results. Therefore, PUFA may impair
lymphocyte responses under certain experimental conditions
such as subcutaneous injection (9-ll), or direct addition
to cell cultures (111). But when PUFA are consumed in the

diet, no adverse effects on immune functions are seen.

44

Essential Fatty Acid Deficiency and Lymphocyte Activation

I . Effects on Proliferative Responses and Phospholipid Metabolism

Introduction

 

As structural components of biological membranes es-
sential fatty acids (EFA) influence membrane fluidity and
membrane bound enzymes and receptors (1,55). In addition,
EFA are precursors for cellular synthesis of prostaglandins
and related lipids (1,2,33). In immunocompetent cells,
incorporation of EFA into membrane phospholipids and the
production of a wide range of EFA metabolites has been
amply documented (2-5). Despite this understanding of
the biochemistry of EFA metabolism in immunocompetent cells,
information about the relationship between dietary EFA and
in vivo immune functions is generally lacking.

Several studies have proposed that polyunsaturated fatty
acids (PUFA), particularly EFA, inhibit in vivo immune
functions (9-11). Much of the evidence supporting this
proposal however, comes from studies in which graft rejec-
tion was prolonged in mice injected subcutaneously with
purified PUFA (linoleic acid,18:2; or arachidonic acid
20:4) (10,11). In other studies graft rejection was also
delayed in mice receiving purified PUFA by stOmach tube
(10). Based on a general estimation of the daily energy
intake of adult mice consuming a stock diet and the experi-

mental protocols used, the injected PUFA represented an

45

increase in daily PUFA intake of about 10%, and the

gavage an increase in PUFA intake of about 50%. Since these
protocols differ significantly from normal feeding, the
question arose as to whether or not similar effects would
occur if mice were fed elevated levels of PUFA, or EFA,

in a controlled dietary study.

I investigated this question and showed that in vivo
plaque forming and delayed type hypersensitivity reSponses
of mice fed a high PUFA diet (50% of energy from corn oil)
was equal to responses of mice fed an isocaloric control
diet (13% of energy from corn oil). In comparing this
dietary protocol with the injection and gavage experiments
itijsnoteworthy that the daily PUFA intake of mice fed the
high PUFA diet exceeded the intake of control fed mice by
nearly 300% without adversely affecting in vivo immune func-
tions. These observations make it difficult to attribute the
immunoinhibition reported in the PUFA injection studies to
PUFA alone.

In addition to proposing that PUFA inhibit immune
functions it has also been reported that "PUFA deficiency"
enhances immunity (10,11,110). Since animals do not have a
requirement for PUFA in general, but do require a subset
of PUFA, called EFA, it is difficult to test this proposal.
However, I have looked at the effects of an EFAD diet on in
vivo plaque forming and delayed type hypersensitivity res-

ponses in mice. Rather than being enhanced, the responses

46

of mice fed the EFAD diet was markedly depressed compared
with the control and high PUFA groups (12,13).

The purpose of the present investigation was to further
examine the relationship between dietary EFA and immune
functions using cell culture. Although in vitro studies
provide controlled conditions for manipulating experimental
variables, culture conditions can influence cell responses
in vitro (122-124). Thus, one of the difficulties in
interpreting in vitro results is that as the time in cul-
ture lengthens, cell responses begin to reflect the cul-
ture environment more than in vivo dietary effects. Recog-
nizing this, I assessed a series of events in lymphocyte
activation ranging from those occurring within the first
few hours of stimulation to those occurring after a few days
in culture. In addition to looking at different time points
during lymphocyte activation in vitro, I altered the in
vitro environment to study the responses of cells from

EFAD mice in the presence and absence of EFA.

Materials and Methods

 

Animals and Diets. Male A/J mice 21:3 days of age were

 

housed in plastic cages, five/cage in a temperature (24:10),
light (12 hours per day), and humidity controlled room.
Diets and water were provided ad libitum. Composition of

the diets is presented in Table l.

47

T and B Cell Counts. After a 70 day feeding period
spleen and thymus cell suspensions were prepared in DMEH,
(M. A. Bioproducts, Walkersville, MD) and mononuclear cells
isolated over Lympholyte M (Accurate Chemical & Scientific
Corporation, Westbury, NY). For B cell counts, 100 ul
of mononuclear cells (107/ml) were incubated with 50 ul
of Immunobeads (BioRad, Richmond, CA) coated with rabbit
antimouse IgG. After 60 minutes at room temperature, the
mixture was centrifuged at 150xg and gently resuspended.
Ten ul of 0.15% toluidine blue was added and rosettes (B
cells) were read as positive if 3 or more beads were at-
tached to a cell. For T cell counts, mononuclear cells
(107/ml) plus rabbit antimouse T cell antiserum (Accurate
Chemical and Scientific Corp., Westbury, NY) were incubated
at 40 for 60 minutes. The mixture was then centrifuged,
resuspended in medium plus rabbit complement (Accurate
Chemical & Scientific Corporation, Westbury, NY), and in-
cubated at 370 for an additional 60 minutes. Fifty ul of
1% trypan blue was added and dead cells (T cells) were
counted. Results are expressed at % B or T cells. T & B
cell percentages were calculated on the basis of total

mononuclear cells.

Mitogen Proliferation Experiments. After a 56-70 day feed-
ing period, mice were killed and spleen cell suspensions

were prepared under sterile conditions in RPMI 1640 or DMEH

48

with 25 mM HEPES supplemented with 1 mM glutamine, and 0.2
ml of penicillin streptomycin (M. A. Bioproducts, Walkers-

ville, MD) and 5 x 10-5

M 2-mercaptoethanol (Sigma, St.
Louis, MO). FCS (Grand Island Biological, Madison, WI),
bovine serum albumin (BSA) (Sigma, St. Louis, MO) or d1
FCS protein prepared by the method of Rothblatt et al.
(125), were added to the culture media in selected ex-

6 nucleated

periments. Spleen cell suspensions (2.5 x 10
cells/ml) were cultured in a total volume of 0.2 ml in
round-bottom microculture plates (Linbro, Hamden, Cn) in

a humidified Mishel-Dutton atmosphere (10% CO 7% O

2' 2'

83% N2). Both unstimulated (no mitogen) and stimulated
(containing mitogen) culture wells were prepared in trip-
licate. Viability was determined by trypan blue exclusion
at the beginning and end of each experiment. Optimal mito-
genic doses of Con A, PHA, PWM, and LPS (all from Sigma,
St. Louis, M0) were used in all experiments. At 48 hours

3H-thymidine and

all cultures were pulsed with 0.1 uCi of
harvested 12 hours later on glass microfiber filters
(Whatman, Clifton, NJ) using a sampling manifold (Milli-
pore, Bedford, MA). Radioactivity associated with TCA-
precipitated DNA was counted in a liquid scintillation

counter. Results are expressed as cpm 3H-thymidine in DNA/

culture well.

49

Mixed Lymphogyte Reaction (MLR). MLR was performed by

 

a modification of the method of Meo (126). Briefly, spleen
cell suspensions from A/J mice (reSponder cells) fed the
three experimental diets were prepared at a concentration
of 2.5 x 106 nucleated cells/ml. Stimulator cells were
spleen cells from male C57/bl mice. Stimulator cells

(7.5 x 106/ml) were treated with mitomycin C (50 ug/ml)
(Sigma, St. Louis, M0) for 30 minutes at 37°, washed 3
times with cold culture medium, counted, and resuspended,
at a final concentration of 5 x 106/ml. One hundred micro-
liters of both responder (2.5 x 105) (A/J) spleen cells and
stimulator (5 x 105) (C57/bl) spleen cells were then added
together in microtiter plates and cultured for 96 hours in
a humidified Mishell-Dutton atmosphere. Control responses

were assessed by incubating 2.5 x 105

A/J spleen cells
without the addition of the stimulator cells. To check
the effect of the addition of Mitomycin C treated cells on
3H thymidine uptake an additional control was run in which
Mitomycin C treated A/J spleen cells (5 x 105) were in-
cubated with 2.5 x 105 A/J responder cells. Since 3H
thymidine uptake into these cultures did not differ from
uptake into cultures containing unstimulated A/J spleen
cells alone, these data were omitted. At 96 hours each
culture well was pulsed with 0.1 uCi of 3H-thymidine and
12 hours later cells were harvested, DNA precipitated,

radioactivity counted and results expressed as previously

described for the mitogen experiments.

50

Incorporation of Phospholipid Precursors into Total
Lipids. Spleens were removed from mice fed the experimental
diets and single cell suspensions were prepared in RPMI
1640 plus 0.15% dl-FCS protein. Spleen cell suspensions
were then layered over Lympholyte-M and splenic mononuclear
cells recovered from the interface between media and the
Lympholyte-M after centrifugation. Mononuclear cells were
then washed twice with cold culture medium, counted, and
viability determined by trypan blue exclusion. Viability
was similar for all treatment groups and exceeded 95%.

Mononuclear cells (10 x 106/ml) were cultured in
triplicate in a humidified Mishell-Dutton atmosphere at a
final volume of 0.2 ml in round-bottom microculture plates.

14C-choline

Optimal mitogenic doses of Con A and 0.5 uCi of
or 1.0 uCi of 3H-inositol (Amersham, Arlington Heights,

IL) were added at the initiation of the cultures. Un-
stimulated cultures received label only. After the
specified time in culture lymphocytes were removed and
washed with cold culture medium. Cell pellets were then
extracted with chloroformzmethanol:10% trichloroacetic acid
(8:4:1) and the chloroform layer was then dried under air
and counted in a liquid scintillation counter. In pre-
liminary experiments lipids were separated by thin layer

14

chromatography. Approximately 90% of the C-choline and

3H-inositol recovered in total lipids was associated with

phospholipids. This indicated that the incorporation of

51

the labelled phospholipid precursors into total lipids pre-
dominantly reflected uptake into phOSpholipids and in sub-
sequent experiments separation of lipid classes was omitted.
Results in these experiments were expressed as: (Acpm)
Acpm = (cpm incorporated into lipids of Con A activated
wells-background)

- (cpm incorporated into lipids of unstimulated wells-

background)

Addition of Fatty Acids (FA) to Bovine Serum Albumin
(BSAL. A modification of the method of Spector and Hoak
(127) was used. Briefly, 20 mg of 20:4 or 18:0 (99% pure,
Sigma, St. Louis) was dissolved in 8 ml hexane and added
to beakers containing 600 mg of Celite (Jons Manville,
Madison, WI). The mixture was dried under nitrogen and
vortexed. Twelve ml of Spector's balanced salt solution
(127) containing 200 mg of BSA (essentially fatty acid-free,
Sigma, St. Louis, MO) was then added to the Celite:FA
particles and the solution was mixed for 30 minutes at
room temperature, spun at 20 at 15,000 x g for 10 minutes,
and the supernatant recovered. Protein was analyzed by

the method of Lowry, (128) and FA by the method of Mosinger

(129).

Preparation of dl-FCS. Dl-FCS was prepared by the

method of Rothblatt et a1, (16). Twenty ml of FCS was

52

slowly added to 200 ml of ice-cold acetone-ethanol (1:1)
while gently swirling. The mixture was allowed to stand
overnight at 0°. The delipidated protein was collected on
Whatman #1 filter paper by suction over a Buchner funnel.
The protein was washed with 50 m1 of ethyl ether. After
15 minutes the dried powder was scraped from the filter

paper and stored dessicated until use.

Results

Experiment 1. In this experiment I looked at whether

 

or not EFAD directly alters either the total number of mono-
nuclear cells or the percentage of T and B cells in two
lymphoid tissues, the thymus and the spleen. I found no
differences across the three dietary treatments in either
the total number of mononuclear cells or the percentage of
T and B cells in either tissue (Table 3). This indicates
that consumption of neither the EFAD, nor the high PUFA
diet results in loss of mononuclear cells or lymphocyte
p0pulations. The immunoinhibitory effects of the EFAD

diet, therefore, appear to be directed at events occurring

during the lymphocyte activation process.

Experiment 2. Spleen cells from mice fed the experi-

 

mental diets were stimulated with optimal doses of Con A,
PHA, PWM, and LPS, in cultures supplemented with 5% FCS.
Although not statistically different, mitogenic responses

of spleen cells from mice fed the EFAD diet were

53

 

 

 

 

 

 

Table 3. Mononuclear Cell Counts, and Percentage of
T and B Cells in Lymphoid Organs of Mice.
Spleen
Total # of
Diet Mononuclear Cells % B Cells % T Cells
EFAD 3.65x107 30.5t1.7 32.5:0.6
Control 3.47x107 32.512.8 34.5:1.4
High PUFA 3.50x107 33.4:3.2 34.0:o.9
Thygus
EFAD 1.03x107 3.3 96
Control l.04x107 2.5 95
High porn 0.93x107 3.2 95
1

Mean 1 SEM of 4 mice per group.

were fed the diets for 70 days.

2

Mice 2113 days of age

T and B cells as percentage of total mononuclear cells.

54

consistently higher than the other two dietary treatment
groups (Figure 3).

One explanation for the slightly higher mitogenic
responses of spleen cells from mice fed the EFAD diet was
that they may have "repaired" during the culture period
with exposure to the small amount of EFA contained in FCS.
To further test the importance of in vitro "repair" I
supplemented FCS containing cultures with BSA:20:4. For
comparison, BSA:18:0, a SFA, was added to parallel cul-
tures. Optimal stimulation of mitogen proliferation was
observed in both EFAD and control spleen cells when cultures
were supplemented with BSA:FA conjugates at a molar ratio
of 1:2 (BSA:FA) and at a final concentration of 0.1 umoles/
ml: 0.2 umoles/ml (data not shown). Addition of BSA con-

3H-

jugated fatty acids (20:4 or 18:0) did not increase
thymidine incorporation into spleen cells from EFAD or
control mice above levels observed in cultures containing
FCS alone (Figure 4). Next, spleen cells were cultured in

3H-thymi-

lipid free media. Even under these conditions
dine incorporation into spleen cells from EFAD mice was
comparable to values in spleen cells from control mice
(Figure 3). Addition of BSA:20:4 to the lipid free cul-
tures improved Con A responses of both EFAD and control
spleen cells, but not to levels obtained in FCS containing

cultures. BSA:18:0 had no effect on Con A responses.

In general, viability was similar across the treatments

Figure 3.

55

3H-thymidine incorporation into TCA precipitated
DNA of mitogen activated spleen cells from
mice fed an EFAD, control or high PUFA diet.
Spleen cells (2.5 x 105) were cultured for

60 hours in DMEH plus 5% FCS and optimal doses
of each mitogen. 3H-thymidine (0.1 uCi) was
present during the final 12 hours. The basal
(unstimulated) 3H-thymidine incorporation was
299136 (EFAD), 232123 (control) and 269137
cpm/well (high PUFA). All results represent
the mean 1 SEM of triplicate observations from
each of 10 mice/treatment after a 56-70 day
feeding period which was initiated when the
mice were 2113 days of age.

56

cpm 3H-thymidine in DNA x “(C's/well

 

 

 

002 >

 

fl...)

mwocnm w.

I 1216

D 60230.. ..
E -15: new»

 

 

 

 

Figure 4.

57

Effects of the addition of BSA:FA conjugates on
3H thymidine incorporation into TCA precipitated
DNA of Con A activated spleen cells from mice
fed an EFAD, control, or high PUFA diet. Spleen
cells (2.5 x 105) from each treatment were in-
cubated for 60 hours in 5% FCS (upper panel) or
serum free (lower panel) cultures with the addi-
tion of BSA: 20:4 or BSA: 18:0 at a molar ratio
of 2:1. 3H-thymidine (0.1 uCi) was present for
the final 12 hours. Data represent the mean

1 SEM of duplicate or triplicate observations on
each of 8-10 mice per treatment after a 56-70
day feeding period. Mice were 2113 days of age
at onset of the feeding period.

58

 

 

 

PCONTROL

 

—
—

 

--EFAD

 

 

 

 

 

 

 

 

 

 

 

5% FCS

 

 

 

 

 

 

 

- p p b p
0 O O O 0
4 2 3 2 1-

60"

=o3\n..O—. x <20 c. 052522....» Eco

BSA: 20:4 BSA: 18:0

0.15% BSA

Figure 4.

59

and reduced in cultures that did not contain FCS. Vi-
ability was reduced from approximately 75% in FCS cultures
to about 40% in BSA containing cultures. This reduced
viability in BSA containing cultures probably contributed

to the reduced responses in these cultures.

Exppriment 3. Results from Experiment 2 indicated
that in vitro mitogenic responses from mice fed the experi-
mental diets were similar; whereas earlier reports had
shown that in vivo immune functions were reduced in mice
fed the EFAD diet compared with mice fed the control or
high PUFA diets. In this experiment I assessed the pro-
liferative response of spleen cells from mice fed the ex-
perimental diets to allogeneic spleen cells in a one-way
mixed lymphocyte reaction (MLR). Mitomycin C treated
spleen cells from C57/bl mice served as the allogeneic
stimulator cells. The results (Table 4) showed no dif-

ferences across the three treatments for the one—way MLR.

Experiment 4. The lack of agreement between the in

 

vitro results from Experiment 2 and 3 and previous in vivo
studies (12,13) suggested the further possibility that the
culture conditions may have influenced the in vitro results.
Therefore, two variables were changed. First, in an effort
to reduce time in culture I switched from 3H—thymidine

incorporation (which requires several days for expression)

60

Table 4. One Way Mixed-Lymphocyte Reactions (MLR)1
3H-Thymidine Uptake cpm/well.

 

 

Responders &

 

Diet Responders Stimulators
EFAD 12541324 441711141
Control 10441293 41481901
High PUFA 11421354 37391937

 

 

1Spleen cells (2.5 x 105) from A/J mice fed the three ex-
perimental diets were incubated with 5.0 x 105 mitomycin-
C treated spleen cells from C57/Bl mice. Cells were in-
cubated for 108 hours, with 0.1 uCi of 3H-thymidine presen
for the final 12 hours. Data represent the mean 1 SEM of
duplicate observations from each of 10 mice/treatment.

61

to phospholipid metabolism (which occurs within a few hours
of lymphocyte activation) (55-57). Second, to study lym-
phocyte activation under EFA-free conditions I used a
serum-free, lipid-free culture supplement, dl-FCS.

In preliminary experiments I found that 3

H-thymidine
uptake in response to Con A was 40% higher in dl-FCS than

in BSA-containing cultures, and background stimulation in
dl-FCS was reduced 75% compared with BSA cultures. Since
Optimal 3H-thymidine uptake in response to Con A and minimal
background values were obtained using 0.15% dl-FCS, this
level was used in all experiments discussed.

First, I assessed Con A induced 14

C-choline uptake into
cellular lipids after 240 and 480 minutes in culture (Fig-
ure 5). After 240 minutes in FCS containing cultures, no
differences in l4C-choline uptake were seen across the
treatments. However, after only 240 minutes in lipid free

culture, Con A induced l4

C-choline uptake into total lipids
of spleen cells from mice fed the EFAD diet was signi-
ficantly reduced (approximately 60%) compared with uptake
into cells from mice fed the control or high PUFA diet.
If cultures were extended to 480 minutes, no treatment dif-
ferences were seen in either the lipid free of FCS con-
taining cultures.

Similarly, Con A induced 3H-inositol uptake into cell

lipids was assessed (Figure 5). In agreement with the

results for choline no differences were seen across the

Figure 5.

62

Incorporation of phospholipid precursors into
total lipids of Con A activated lymphocytes
from mice fed the EFAD, control or high PUFA
diet cultured in media containing either FCS

or delipidated FCS (dl-FCS). Splenic mono-
nuclear cells (1 x 106) were incubated with
Optimal doses of Con A, or without mitogen
(unstimulated) plus 0.5 uCi of 14C-choline or
1.0 uCi of 3H-inositol. Harvested lymphocytes
were extracted with 8:4:1 (chloroform:methanol:
10% TCA) and radioactivity associated with the
chloroform layer was counted. Radioactivity
specifically associated with Con A activation
(Acpm) was determined by subtracting the counts
incorporated into total lipids from unstimulated
cells from counts incorporated into the total
lipids Con A activated cells after background
was subtracted from both. Data represent the
mean 1 SEM of duplicate observations on each of
8-10 mice/treatment after a 56-70 day feeding
period begun when the mice were 2113 days of
age. (*) = significantly different (pl 0.05)
than the control or high PUFA groups.

C-chollno In total llplds/woll

14

A cpm

63

‘ -_EFAD '24'—'o MINUTESI

[:J-CONTBOL
--mcH PUFA

 

 

480 MINUTES

1900

1,000

    

 

 

FCS dI-FCS

‘ "——_]oo unw'rss

 

‘ 240 MINUTES

 

 

 

FCS dI-FCS

Figure 5.

0|
0

N
O
O

100

A cpm 3H-Inotuol In_ total lipids/well

64

treatments in FCS containing cultures at 60 minutes. How-
ever, after only 60 minutes in the lipid-free culture,
Con A induced 3H-inositol uptake was significantly re-
duced in lymphocytes from mice fed the EFAD diet compared

with the other groups.

Discussion

 

I previously showed that in vivo plaque forming res-
ponses and delayed type hypersensitivity were impaired in
mice fed an EFAD diet compared with mice fed a control or
high PUFA diet (12,13). In this section, I investigated
the effects of these dietary treatments on lymphocyte activa-
tion in vitro.

EFAD could impair in vivo immune responses by alter-
ing total mononuclear cells, or the proportions of lympho-
cyte populations. In this experiment I determined total
mononuclear cell numbers and percentages of T and B cells
in two lymphoid tissues (thymus and spleen) of unimmunized
mice fed the three experimental diets. Values were similar
across the three treatments for all measurements (Table
3). The absence of any dietary effects on mononuclear
cell numbers, or T and B cell proportions, suggests that
EFAD affects in vivo immunity by altering the functional
capacity of existing cells, rather than by eliminating

specific cell populations. However, the assays used to

65

determine T and B cell counts do not rule out the pos-
sibility that EFAD could alter T and B cell subpopulations,
and in this manner influence in vivo immunity.

The presence or absence of certain T and B cell sub-
populations can be estimated, to some degree, by their
response to specific mitogens. T helper cells (TH), for
example, bear the Ly 1 surface antigen and constitute the
major responding T cell subpopulation to Con A and in the
MLR (130,131). T cells bearing Ly 1, 2, 3, are PHA res-
ponsive (130). Also, LPS appears to preferentially stimu-
late immature B cells (132). Since proliferative responses
to all these mitogens were similar across the three dietary
treatments under standard culture conditions, this sug-
gested that the diets did not alter lymphocyte subpopula-
tions.

Previous studies had shown that when EFAD mice were
switched to a diet containing EFA, in vivo plaque forming
and delayed type hypersensitivity responses returned to
normal, and even overshot control responses (12,13).

Since the proliferative responses of spleen cells from
mice fed the EFAD diet were equal to, or slightly higher
than the responses from the other two treatments this sug-
gested the possibility that the EFAD spleen cells may have
"repaired" in vitro in the presence of the small amount
of EFA in FCS. But further supplementation of FCS cul-

tures with EFA (20:4) had no effect on Con A responses

66

of either EFAD or control spleen cells. This suggested
that the small amount of EFA provided by FCS was sufficient
for lymphocyte proliferation regardless of the dietary
treatment. Others have reported conflicting results; sup-
plementation of serum containing cultures with additional
fatty acids either free or bound to BSA has been shown to
increase (120), decrease (119), or have no effect (121)

on lymphocyte proliferation. Variations in serum, or

ratio of albumin to fatty acid, may be contributing factors
in the lack of agreement among reports (121).

Although overall proliferative responses of both treat-
ments were reduced by more than 50% in lipid free cultures
compared with FCS cultures, the response Of EFAD spleen cells
remained slightly higher than the controls. Thus, even
in the absence of any exogenous source of EFA, the EFAD
spleen cells retained the capacity to reSpond to Con A as
well as, or slightly better than, control spleen cells.

Con A responses of both treatments improved by about 20%
with the addition of 20:4 to lipid free cultures. This
is in general agreement with previous reports indicating
that adding PUFA, particularly EFA, improves cell growth
and proliferation compared with cultures containing only
BSA alone, or BSA plus a SFA (133). But again, there was
no evidence of treatment differences following the addi-
tion of 20:4 to the cultures.

These experiments demonstrate that spleen cell viability

67

and proliferative responses to mitogens and MLR were not
impaired by EFAD. Several reasons could account for this.
Lymphocyte proliferation may not require EFA, or, when EFA
are present in the culture, the amount required for pro-
liferation may not be altered by dietary EFA status.

Another possibility is that during EFAD lymphocyte
responses are suppressed, either by the presence of a sup-
pressive factor, or by the absence of some activating
factor derived from EFA, and this suppressive effect is
released when cells are cultured and assayed in vitro. It
is also possible that the manner in which EFAD impairs in
vivo immune functions may not be detected by in vitro pro-
liferation assays. Another consideration is the extended
culture time required for assessment of lymphocyte pro-
liferation. Since it is well established that cultured
cells incorporate exogenous lipids over time in culture,
it seems probable that proliferation experiments requiring
several days in culture provide little insight into the
functional prOperties of EFAD lymphocytes as they exist in
vivo.

To minimize affects associated with long term cul-
tures, I investigated the influence of the dietary treat-
ments on lymphocyte phospholipid metabolism, an important
membrane associated event occurring within the first few
hours of lymphocyte activation (55-57). When these short

term cultures contained FCS, no treatment differences

68

were detected. However, in lipid free cultures, there was
a transient reduction in Con A induced uptake of 14C-
choline and 3H-inositol into EFAD spleen cells compared
with the other two treatments. The loss of dietary treat-
ment effects over time in culture strongly supports the
notion that long term culture experiments reflect in vitro
conditions rather than in vivo dietary effects.

This appears to be true for EFAD, but may not be the
case for all nutrient deficiencies. Decreased Con A pro-
liferative responses have been shown in experimental studies
Of zinc deficiency (134,135), vitamin E deficiency (136),
and protein calorie malnutrition (137). In both zinc de-
ficiency and protein calorie malnutrition however, de-
creased Con A responses have been associated with shifts
in the lymphocyte populations induced by the deficiency
(135,137).

Since EFAD does not appear to shift lymphocyte pOpu-
lations or subpopulations my results suggest that impaired
early phospholipid metabolism in spleen cells from mice
fed the EFAD diet results from a membrane-associated defect
induced by EFAD. This defect may not be detected in pro-
liferation experiments because of extensive lipid rearrange-
ments occurring over several days in culture. On the
other hand, phospholipid metabolism in Con A activated
EFAD cells in FCS cultures was similar to the control and
high PUFA groups. A likely explanation for this is

that'some "repair" of the EFAD cells occurred in the

69

presence of EFA in FCS.

Although it seems highly probable that the lack of
EFA in the lipid free cultures contributed to the impaired
early phospholipid metabolism in Con A activated EFAD spleen
cells, other explanations are possible. For example, the
delipidation procedure may have extracted or denatured
growth factors in FCS. It is possible that EFAD spleen
cells may be more sensitive to the loss of growth factors,
or possibily the production of toxic compounds by the delipi-
dation procedure. But since phospholipid metabolism in the
Con A activated EFAD spleen cells was similar to the other
two treatments at a later time point, these alternate ex-
planations for the early impairment seem unlikely. The
loss of treatment differences with time in culture is
consistent with the prOposal that long term in vitro
responses reflect culture conditions to a far greater extent
than in vivo dietary effects, at least in the case of
EFAD.

Recent studies indicate that enhanced phospholipid
metabolism is a precommitment event in the cell cycle which
is proportional to Con A concentrations, even very high
levels (100-200 ug/ml) (56,57). Unlike phOSpholipid
metabolism, DNA synthesis is optimally stimulated within
a very narrow, relatively low dose range of Con A (2.5-10
ug/ml). To facilitate comparisons between early pre-

commitment events and DNA synthesis, all phospholipid

70

cultures discussed here contained doses of Con A optimal
for DNA synthesis. Within this dose range early phospho-
lipid metabolism was significantly delayed in EFAD spleen
cells, but DNA synthesis was not. Whether or not a similar
early impairment follows antigen activation in vivo is not
known. But previous studies following the development of
plaque forming cells (pfc) after in vivo immunization

with SRBC indicated a significant delay in the sequence of
appearance of IgM and IgG pfc, as well as peak pfc response,
in mice fed the EFAD diet (12). The delayed appearance of
pfc would be consistent with an EFAD-induced impairment
early in lymphocyte activation. Also, since EFAD alters
membrane structural lipids (1), it seems likely that in
addition to phospholipid metabolism, other membrane-as-
sociated events may be affected by EFAD. Presumably, the
combined effects of these EFAD induced defects impairs
lymphocyte function.

Finally, others have proposed that PUFA impair the
immune response. In the present report, the responses of
lymphocytes from mice fed the high PUFA diet equalled the
response of the control group in all assays measured. This
supports my previous findings which showed that in vivo im—
mune functions were not impaired in mice fed an isocaloric,

high PUFA diet (12,13).

71

Essential Fatty Acids and Lymphocyte Activation

II. Effects on Neutral Amino Acid Transport

Introduction

 

Because of the importance of EFA in the lipid component
of biological membranes, EFAD could impair lymphocyte
responses by altering the lymphocyte membrane structure
(1). Alterations in structure could affect lymphocyte
responses by influencing a number of membrane-associated
events that occur early in lymphocyte activation. One of
these events, phospholipid metabolism, has already proven
to be sensitive to EFAD, as shown in the previous section.
Another early event in lymphocyte activation is the in-
creased influx of metabolities (17), including amino acids
(17,68-71). In this section I examined neutral amino acid
transport using a-aminoisobutyric acid (AIB). This trans-
port system increases in activity early in lymphocyte ac-
tivation (68-71) and correlates with DNA synthesis (69).
AIB transport in ascites tumor cells is also sensitive to
dietary fat intake (73).

I assessed the capacity of resting (unstimulated) and
Con A activated lymphocytes from mice fed the experimental

14C-AIB. I found that Con A activated

diets to take up
EFAD lymphocytes were defective in their capacity to
transport AIB at low substrate concentrations (0.01 mM).

At 0.01 mM AIB, reduced AIB uptake by lymphocytes from

72

mice fed the EFAD diet elevated Km, but not Vmax values for
initial Na+-dependent transport. Taken together with
results from the phospholipid experiments, this indicates
that lymphocytes from mice fed an EFAD diet have defects in
several membrane-associated events prominent in the first
few hours of lymphocyte activation. These defects may
contribute to the overall impairment in the immune re-

sponse which accompanies EFAD (12,13).

Materials and Methods

 

Animals and Diets. Male A/J mice, 2113 days of age,

 

were housed in plastic cages, in a temperature (24110),
light (12 hours per day), and humidity controlled room.
Mice were fed ad libitum one of the three experimental

diets described in Table 1 for 70-84 days.

Preparation and Preliminary Incubation of Splenic

 

Mononuclear Cell Suspensions. Red cells, dead cells, and

 

granulocytes were removed from spleen cell suspensions by
centrifugation over Lympholyte M. Viability of the mono-
nuclear cell suspension collected from the culture medium/
gradient interface always exceeded 95% and did not differ
across the three dietary treatments.

Before assessing AIB uptake, mononuclear cells (107/

ml) were preincubated with or without Optimal mitogenic

73

doses of Con A for 240 minutes at 370 in a humidified atmos-

phere of 7% CO 10% 02, and 83% N2. All cell preparations

2!
and preliminary incubations were conducted aseptically in
RPMI 1640 plus 25 mM HEPES, 2 mM glutamine, and 0.15%
delipidated fetal calf serum protein (dl-FCS). Dl-FCS

was prepared by the method of Rothblatt et al, (125).

14

Time Course Of C-AIB Accumulation. After the 240

 

minute preincubation, cells were washed twice with cold
culture medium, and resuspended in 0.2 ml of Earle's
Balanced Salt Solution containing 25 mM HEPES and 0.1 uCi

Of 14

C-AIB (Amersham, Arlington Heights, IL). At specified
times, uptake was stopped by the addition of 1.0 ml of

cold phosphate buffered saline containing 5 mM AIB (PBS-
AIB). Cells were then collected on glass microfiber filters
using a sampling manifold, and washed twice with 10 ml of
cold PBS-AIB. After the filters were dried, scintillation
cocktail was added and accumulated radioactivity determined
by counting in a liquid scintillation counter. To correct
for extracellular trapped radioactivity, 0.1 uCi of 14C-
AIB was added to cell suspensions kept at 40 and counts
associated with these tubes subtracted from experimental

values. Results are expressed as total accumulation of AIB

in picomoles/lO6 mononuclear cells.

74

AIB Initial Rate Measurements. Total cellular AIB
uptake includes AIB entering the cell by a saturable Na+
dependent transport system and by simple diffusion (70,138).
The contribution of each of these components to total AIB
uptake can be estimated by performing parallel experiments
in isosmolar Na+ containing and Na+ free media. The active
Na+ dependent component of AIB uptake is calculated by
subtracting AIB uptake in Na+ free media from AIB uptake
in Na+ containing media.

To assess total AIB uptake incubation media (IM-l)
was prepared as previously described by Finklestein and
Adelberg (139). IM-l contained 150 mM NaCl, 1.0 mM

CaC1 ~2H o, 5.0 mM K ~3H o, 1.2 mM MgSO -7H 0, 0.3

2 2 2 4 2 4 2
mM H3PO4, and 5.6 mM D-glucose. The pH was 7.4. Follow-

HPO

ing the 240 minute pre-incubation mononuclear cell suspen-
sions were washed twice with cold IM—l and resuspended in
0.2 ml of IM-l. After a five minute incubation in a 370
shaking water bath, various concentrations of unlabelled AIB

f 14C-AIB were added to each tube. After 120

plus 0.1 uCi 0
seconds 1.0 ml of ice cold IM-l plus 5 mM AIB was added to
each tube as the tubes were placed in ice. Spleen cells were
then centrifuged in a microfuge (Eppendorf, Brinkman In-
struments, Westbury, NY) for 30 seconds over a 20% Percoll
gradient (Pharmacia, Piscataway, NJ). Cell pellets were

then washed twice with ice cold IM-l containing 5 mM AIB,

resuspended in saline, and transferred to scintillation vials

75

to which 15 ml of ACS scintillation cocktail (Amersham,
Arlington Heights, IL) was added. Radioactivity was de-
termined by counting in a liquid scintillation counter.
Results are expressed as picomoles AIB uptake/106 cells/
minute.

To determine the Na+ independent or diffusion com-
ponent of AIB uptake IM-2 was formulated in an identical
manner to IM-l, except 150 mM choline chloride replaced
NaCl. Experiments were performed in an identical fashion
as described for total AIB uptake, except in all instances
IM-l was replaced with IM-2. The active Na+ dependent com-
ponent of AIB transport was then calculated by subtracting
values obtained in incubations using IM-2 from those using
IM-l. Results are expressed as AIB uptake in picomoles/106
cells/minute. By using the Percoll gradient and washing the
pellets as described, negligible counts were detected in the

pellets as extracellular or trapped counts.

Data Analysis. Data were analyzed statistically by one
way analysis of variance with treatment differences de-
termined by Tukey's test (117). Results relating substrate
concentrations to initial velocity were determined by least
squares analysis with graphic transformation shown by

plotting v vs v/[AIB].

76

Results

Time Course of AIB Uptake. Enhanced neutral amino acid

 

transport occurs early in the process of lymphocyte activa-
tion (68-71) and is sensitive to dietary fat saturation
(73), at least in ascites tumor cells (73). Figure 6 shows
the time course of the AIB (0.01 mM) accumulation into splen-
ic lymphocytes from mice fed the EFAD, control or high PUFA
diets. AIB accumulation into unstimulated lymphocytes was
characterized by a rapid initial uptake in the first 120
minutes, followed by little additional uptake between 120
and 240 minutes. Although AIB accumulation into unstimu-
lated lymphocytes from EFAD mice was slightly less than for
lymphocytes from the control or high PUFA groups, no sta-
tistical differences were evident.

AIB accumulation into Con A stimulated lymphocytes,
however, was significantly reduced in the EFAD group com-
pared with the control or high PUFA groups (Figure 6).

Since viability was similar across the three treatments
(approximately 80% of the original cell number), the reduced
uptake into the EFAD cells was not due to decreased cell

numbers in the EFAD cultures.

AIB Uptake Kinetics. Next I investigated initial AIB

 

uptake rates in Con A activated lymphocytes from mice
fed the three diets. After 240 minutes in lipid—free
medium containing Optimal mitogenic doses of Con A, total,

Na+-dependent, and Na+-independent AIB uptake, as well as

Figure 6.

77

Time course Of AIB uptake. Resting (---)

and Con A activated (———) (2.5 ug/ml) splenic
lymphocytes (10 x 106/ml) were incubated in a
total volume of 0.2 ml with 0.10 uCi of 14c-
AIB in RPMI 1640 plus 0.15% dl-FCS. At the time
points indicated, incubations were stopped by
placing the cultures in an ice bath and adding
1.0 ml of ice cold PBS—AIB. Cellular accumula-
tion of AIB was assessed after collecting cells
on glass microfiber filters and washing with
PBS-AIB. Data represent the mean i SEM of six
experiments.

78

 

AIB uptake: plcomolosHQ’ cells

r

 

o -EFAD

 

 

*
C

.......... ‘

* ...... i :123131231W'I [nu-.1 ............... E

cocoon-oc:::..::::..:" I r v

M\\\\':'{-.':':: .....

1 J J # ;— l

30 00 1 20 2‘0

Figure 6.

 

79

initial transport rates of Na+-dependent uptake were as-
sessed. A 120 second incubation period was used to measure
initial uptake rates.

Over the doses of AIB used (0.01-4 mM), I observed a
characteristic biphasic curve for AIB uptake (Figure 7A).
Total uptake into EFAD lymphocytes was significantly reduced
compared with the other two treatments at the .01 mM AIB
concentration (Figure 7A insert). At this level of AIB,
which was identical to the AIB concentration used in the
time course experiments, the capacity of activated EFAD
lymphocytes to transport AIB was impaired. However, no
differences occurred among the treatments for AIB uptake
at AIB concentrations from 0.035-4 mM.

As these experiments were run in Na+ buffer (Materials
and Methods) parallel experiments were also run in Na+-
free buffer (Materials and Methods) to determine the Na+-
independent, or diffusion component of total AIB uptake.
When this component is subtracted from total AIB accumula-
tion, values for Na+ dependent AIB transport are obtained
(Figure 7B). As previously shown with total AIB uptake,
Na+ dependent AIB transport was significantly lower in
the EFAD cells at AIB concentration of 0.01 mM. (Figure 7B
(insert)). No differences were seen across the three
treatments in Na+ dependent AIB transport at AIB concentra-
tions ranging from 0.035-4.0 mM.

The kinetics of AIB transport were assessed in lymphocytes

Figure 7A.

7B.

80

Total AIB uptake. Following a 240 minute pre-
incubation (Materials and Methods) Con A acti-
vated splenic lymphocytes (10 x 106 ml) were
washed and resuspended in 200 ul of IM-l (Materi-
als and Methods) plus 0.10 uCi of l4C-AIB and
various amounts of cold AIB. After a 120 second
incubation in a 370 water bath, incubations were
stopped by adding 1.0 ml of ice-cold PBS-AIB
and placing the cultures on ice. Cell-associated
radioactivity was determined after cells were
pelleted by centrifugation through a 20% Percoll
gradient. Data represent the mean 1 SEM of five
experiments.

Na+ dependent AIB uptake. To determine Na+
dependent AIB uptake parallel experiments were
performed as described for total AIB uptake,
except Na free medium (IM-2) was used. The
difference between values Obtained in IM-l and
IM-2 are considered to represent Na+ dependent
AIB uptake. Data represent the mean 1 SEM of
five experiments.

AIB Uptake: picomoles/10° collo/mlnuto

81

 

 

 

 

     

 

 

 

2.0.
100 v- "
‘00 P
r I
80 r- ) .s": ‘
.-"*
P ' J a:
0.01
00 '- '
40 '- “EFAD "‘
A-CONTROL
i' D-NIGH PUFA "
20 - "'
J l J l
0 1 0.5 1 0 2 O 4.0
1.00"

 

 

_L 1 J

0.1 0.5 1.0 2.0
(AIB, mM)

 

Figure 7.

 

 

82

from mice fed the three experimental diets. Results are
presented in Figure 8, with initial velocity (v) of Na+
dependent transport plotted against v/[AIB]. Lines derived
from least squares analysis of experimental values for

each treatment were plotted, and the Km and Vmax values
estimated graphically. From this analysis the Km value for
the EFAD lymphocytes was almost twice as high as the values
for the control or high PUFA groups (2.88 mM vs 1.54 and
1.66 mM), and the Vmax value for the EFAD group was ap-
proximately 50% higher than the other two treatments (92
pmol vs. 63 and 61 pmol 106 cells/min).

To facilitate statistical analysis, Km and Vmax values
obtained from each experiment (n = 5) were derived and mean
values plus standard errors calculated. The results (Table
5) are essentially the same as those obtained graphically
(Figure 8). The Km value for AIB transport in the EFAD
lymphocytes were significantly greater than the Km value
for the control or high PUFA groups. However, the vmax
value for the EFAD group, while remaining almost 50% higher
than the Vmax values for the other two groups, did not

differ statistically from these two treatments (p < .05).

83

Figure 8. Initial rate kinetics of Na+ dependent AIB up-
take into Con A activated splenic lymphocytes.
Suspensions of Con A activated splenic lympho-
cytes (10 x 106/ml) were incubated for 120 seconds
in various concentrationscflfunlabelled AIB plus 0.1
uCi l4C-AIB. Data obtained are plotted by the
Eadie-Hofstee method and represent mean values
from five experiments. Best fit lines were ob-
tained by linear regression of the experimental
values with the following r2 values; EFAD =
.608, control = .675, and high PUFA = .855.

84

 

 

 

 

 

0'

fl

:3

.E

E . -EFAD

a _ A-CONTROL
= o-HIGH PUFA
o .
' 0 80 r-

O

2 - .

E

. A

a: so

a.

“a __ _ /EFAD

:

Q 40 - /

< HIGH PUFA CONTROL

*3 P

5

u b

c 20 SD

3 .

c b

‘c ' . .
* A 1

a 10 20 so 40
2

5‘ v/(AIB,mM)

Figure 8.

85

Table 5. Km and Vmax Values for Na+ Dependent AIB Uptake
Into Con A Activated Lymphocytes.l

 

 

 

Vmax

Diet Km (mMol) (pmol/106 cells/min)
EFAD 2.3610.36a 83:14a
Control 1.54:0.17b 64110a
High PUFA 1.43:0..13b 55:9a

 

 

1Data represent mean 1 SEM of 5 experiments. Numbers in the
same column with different subscripts are significantly dif-
ferent (p < 0.05).

86

Discussion

 

The purpose of the experiments in this section was to
further study the effects of dietary EFA and EFAD on early
events in lymphocyte activation. I found that when extra-
cellular AIB concentrations were low (0.01 mM), AIB uptake
into Con A activated EFAD lymphocytes was impaired. This
was true for AIB uptake over an extended time course
(0-240 minutes), for total AIB uptake under conditions
approaching initial rates (120 seconds), and for the Na+
dependent component of initial AIB uptake. AIB uptake
(0.01 mM) into Con A-activated EFAD lymphocytes was sig-
nificantly reduced compared with uptake into Con A-activated
lymphocytes from the control or high PUFA groups: but,
again, only when AIB concentrations were low (0.01 mM).

At higher AIB concentrations (0.035-4.0 mM) AIB uptake was
similar across the three treatments. As a result, the Km
value for Na+ dependent AIB transport into Con A-activated
EFAD lymphocytes was almost twice the Km values for the
control and high PUFA groups, but Vmax values did not
differ. This finding is in close agreement with the re-
sults obtained by Kaduce, et a1, (73) in their study of AIB
transport into Ehrlich's ascites tumor cells grown in mice
fed diets differing in fat saturation. They found sig-
nificantly elevated Km, but not vmax, values for Na+
dependent AIB transport into tumors grown in mice fed a

diet high in hydrogenated coconut oil (EFAD).

87

The importance of the high affinity (low Km) component
of neutral amino acid transport in mitogen activated lympho-
cytes has recently been demonstrated by Borghetti, et al.
(71). They showed that two independent saturable components
systems ACS and A) contribute to the entry of neutral amino
acids into mitogen activated lymphocytes, and enhanced
activity of the high affinity component (system ACS) is
important for subsequent steps in lymphocyte activation.
Given the importance of the high affinity component of
neutral amino acid transport in lymphocyte activation, it
seems plausible that its disruption in mitogen activated
EFAD lymphocytes may alter their response capacity. This
could occur in several ways.

The most direct way that EFAD could influence membrane
transport would be through alterations in membrane struc-
tural lipids (26,27). EFAD reduces membrane arachidonic
acid (AA) levels, but has no effect on membrane phospho-
lipid content (27,72,73). Reducing membrane AA levels could
affect membrane fluidity (55).

Several years ago Ferber and Resch (55) showed that
after lymphocyte activation by mitogens or antigens, there
was a rapid and preferential incorporation of unsaturated
fatty acids (predominantly AA) into membrane phospholipids.
This increased the ratio of unsaturated fatty acids to
saturated fatty acids (USFA:SFA) in the activated lympho-

cyte membrane, thereby increasing membrane fluidity (55).

88

Enhanced membrane fluidity may be important for lymphocyte
activation by directly affecting a number of cell processes,
including membrane transport. Recently, Pilch et a1. (66)
showed that unsaturated fatty acids stimulated glucose up-
take when added directly to suspensions of adipocyte plasma
membranes. Also, in a detailed study of membrane fluidity
and glucose transport kinetics, Yuli (67) showed that the
maximal rate of glucose transport was closely correlated
with increased membrane fluidity up to a peak value, after
which further increases in fluidity decreased transport.
But membrane fatty acid rearrangements accompanying EFAD
may actually increase the ratio of USFA:SFA in membrane
phOSpholipids (27), suggesting that EFAD may alter lympho-
cyte function in a manner unrelated to membrane fluidity.
Besides affecting membrane structural properties such
as fluidity, membrane lipids may also affect membrane protein
function. It is becoming increasingly clear that membrane
fatty acids are organized into discrete domains within the
lipid bilayer (60). Within these domains, specific fatty
acids are complexed to membrane proteins (61). This close
association between membrane lipids and proteins suggests
that alterations in membrane fatty acid composition may
affect membrane protein function. This is supported by
studies of EFAD and membrane-bound enzymes (26-28) and
receptors (62,63). In lymphocytes, EFAD could alter antigen

receptor binding, or the binding of membrane transport

89

proteins to metabolites.

In addition, by lowering membrane AA content, EFAD
may reduce the synthesis of prostaglandins and related AA
metabolites (54). This could influence lymphocyte activa-
tion by directly affecting a number of cell processes,
including metabolite transport. In two recent reports
Bass, et al (39,40) showed that AA metabolism through the
lipoxygenase pathway to form lZ-HPETE enhances hexose trans-
port into human polymorphonuclear leucocytes. This suggests
a direct role for AA metabolites in transport processes. In
a related report, when an inhibitor of AA metabolism, ETYA
(5,8,1l,l4-eicosatetraynoic acid) was added to mast cell
cultures, phospholipid turnover and mediator release were
blocked (58). This suggests that AA metabolism, phospho-
lipid turnover, and a cellular effector response are linked.
It further demonstrates the importance of AA metabolism in
cell function. In EFAD, decreased availability of AA could
reduce lymphocyte responses in a similar manner.

Although EFAD impairs AIB uptake into Con A-activated
lymphocytes, AIB uptake into unstimulated or resting EFAD
lymphocytes was similar across the treatments. This sug-
gests a unique role for EFA in activated cells. Since en-
hanced activity of the high affinity component of neutral
amino acid transport accompanies mitogen activation (71),
it's possible that EFA may be needed for the full expression

of this augmented transport component.

90

In addition to studying the effects of EFAD on lympho-
cyte functions, I also looked at the effects of a high PUFA
diet on these responses. As in previous reports (12,13)
and earlier parts of this thesis, I found that lymphocytes
from mice fed the high PUFA diet responded similar to those

from control fed mice.

91

Essential Fatty Acid Deficiency and Lymphocyte Protein

Synthesis, RNA Synthesis, Total Protein, and Viability

Introduction

 

The activation of lymphocytes by mitogens leads to en-
hanced turnover of membrane components (17), increased
metabolite fluxes (17,68-71), and increased synthesis of
cellular macromolecules such as protein, RNA, and DNA (17).
In previous sections I found that within the first few
hours of lymphocyte activation certain aspects of metabolite
transport and membrane phospholipid metabolism were reduced
in EFAD spleen cells compared with the control or high PUFA
groups,provided experiments were performed in lipid free
media. After several days in culture, no differences were
observed across the treatments in 3H-thymidine incorporation
in response to T and B cell mitogens and MLR. The next
series of experiments assessed events occurring near the
midpoint of lymphocyte activation : protein and RNA syn-
thesis. In addition, total protein and viability were also
assessed. All of these experiments were performed in lipid

free cultures.

Materials and Methods

 

14C-Leucine Incorporation into TCA Precipitated Protein.

 

Spleen cells were isolated from mice as described in earlier

92

experiments and mononuclear cells separated from red cells,
dead cells, and granulocytes by centrifugation over Lympho-

lyte M. At the onset of the cultures Con A activated cells

received an optimal dose of Con A and 0.15 uCi of 14C-

leucine, resting or unstimulated cells received the 14C-

leucine alone. All cultures were carried out at a final

6

cell concentration of 5 x 10 cells/ml in a final volume of

0.2 ml. Culture medium contained RPMI plus 25 mM HEPES, 1 mM
glutamine, 0.2 m1 penicillin streptomycin/100 ml culture

medium, 5 x 10"5

M 2-mercaptoethanol, and .15% dl-FCS. At
the specified times cells were removed from incubation

and washed twice with 10 volumes of ice cold culture medium.
Cell pellets were then solubilized with l N NaOH and placed
in a 560 shaking water bath for 10 minutes. Two ml of ice
cold 10% TCA was then added to the tubes and they were
placed in the cold (40) for 24 hours. Precipitated protein
was collected on filter papers using a Millipore sampling
manifold, washed twice with cold 10% TCA, dried, and counted.

14

Results are expressed as cpm C-leucine in protein/well.

3H-Uridine Incorporation into RNA. Cultures of splenic

 

mononuclear cells were prepared as previously described at
a final cell density of 2.5 x 106 cells/ml. After 24

hours in culture medium identical to that used in the pro-
tein synthesis experiments, 0.1 uCi of 3H-uridine was added

and the culture was continued for an additional 18 hours.

93

At this point cells were collected on filter paper using

the sampling manifold, washed twice with 10 ml of PBS, 5 ml
of 10% TCA, and 5 ml of methanol. Filter papers were then
dried and counted. Results are expressed as cpm 3H-uridine

incorporated into RNA/well.

Viability. Viability was determined by trypan blue

 

exclusion throughout all experiments. In most instances

10 ul Of 1% trypan blue in PBS was added to cell suspensions
in a 1-2 ml volume. After 1-2 minutes a drop of the cell
suspension was placed on a hemocytometer and two fields

were counted. Cells that stained blue were counted

(dead cells) as well as those excluding the dye (live

cells). The final calculation was: (mean of two fields
counted) x dilution factor x 104 (volume of the hemocytometer
field) = number of cells/m1. Total numbers of live and

dead cells were counted and the results were expressed as

% viable cells.

Results

Although not statistically different, 3H leucine in-
corporation into TCA precipitated protein of resting and
Con A activated spleen cells from EFAD mice exceeded uptake
into cells from mice fed control and high PUFA diets at
all time points measured (Figure 9). Cellular protein

was higher (Figure 10) in the EFAD cells, but viability

Figure 9.

94

14C leucine incorporation into TCA precipitated
protein of splenic mononuclear cells from mice
fed the experimental diets. l x 106 splenic
mononuclear cells were incubated with optimal
mitogenic doses of Con A (solid lines) or without
mitogen (dotted lines). At the times indicated,
cells were removed, washed, and precipitated with
10% TCA. Each data point represents the mean

1 SEM of duplicate observations from each of 3-5
animals/treatment after a 70-84 day feeding period
beginning when the mice were 2113 days of age.

50

25

cpm MC-Ieucine x lo’glculture well

 

95

'. EFAD
A CONTROL

 

HIGH PUFA

’.

o
o
0".
o
a..'
...............
------

'
0’.
o
.9
o
.9
o
.0 ,0
a ,a
'.
v
..
o
‘C
v
..

.O
u‘ l
.. .OOOC.....I.............“..

TIME (hours)

Figure 9.

Figure 10.

96

Protein content of splenic mononuclear cells
from mice fed the experimental diets. l x 106
splenic mononuclear cells were incubated for the
specified times with optimal mitogenic doses

of Con A (solid lines), or without Con A,
(dotted lines). Data represent the mean 1 SEM
of five cultures/dietary treatment.

97

0 EFAD
A CONTROL
I HIGH PUFA >1:

(“g/well)

 

total protein

 

 

T If . 221
TIME (hours)

Figure 10.

98

did not differ across the treatments (Table 6). This points
to a markedly higher protein content in the EFAD cells com-
pared with the other two treatments. RNA synthesis was
estimated by measuring the incorporation of 3H-uridine into
TCA precipitated RNA. As shown in Table 7, incorporation

of 3H-uridine into RNA did not differ across the treatments.

Discussion

 

During lymphocyte activation by mitogens a number of
cellular events occur which have been correlated with blast
cell formation (17,69). In this section, protein and RNA
synthesis were assessed. Although not statistically dif-
ferent, 3H-leucine incorporation into cellular protein was
higher in both unstimulated (resting) and Con A activated
EFAD spleen cells. This was unexpected because neutral
amino acid (NAA) transport was impaired in EFAD cells.
Presumably if decreased NAA transport affects lymphocyte
function, it would be detected through assessment of protein
synthesis. Although this seems logical it is important to
consider that these two experiments measured different as-
pects of protein metabolism. The NAA transport experiment,
which used AIB, indicated that the Km but not the Vmax for
NAA uptake was defective. On the other hand, in the present

l4C-leucine into TCA preci-

experiment, incorporation of
pitated protein was measured. The conditions of the AIB

study included measurements of uptake when extracellular

99

Table 6. Effect of Time in Lipid Free Cultures on %
Viability of Spleen Cells From Mice Fed the
EFAD Control, or High PUFA Diet.1

 

 

 

Time (Hours) EFAD Control High PUFA

4 88 85 83

14 15 15

8 85 76 71

14 15 15

16 70 67 64

15 15 14

24 73 65 64

13 14 13

 

 

1Values represent the mean 1 SEM of 5 observations per
treatment.

100

Table 7. 3H-Uridine Incorporation Into RNA of Spleen Cells
From Mice Fed the EFAD, Control, or High PUFA

 

 

 

Diet.1
Diet Unstimulated Con A
EFAD 865 3291
163 1214
Control 790 3800
155 1202
High PUFA 579 3767
136 1200

 

 

1Data represent cpm 3H-uridine in TCA-precipitated RNA/

culture well. Values are means 1 SEM of duplicate ob-
servations from 8 mice/treatment.

101

AIB concentrations were limited. At the lower extra-
cellular AIB levels (0.01 mM), defective transport was
observed. The protein synthesis experiments were per-
formed under conditions of NAA excess. Thus, even if the
affinity of the transporter was impaired, in NAA excess
this defect may not be detected and no effect on protein
synthesis would be expected.

What does show, however, is that under conditions of
NAA excess, protein synthesis proceeds normally, if not at
an elevated level in EFAD spleen cells. This indicates
that EFAD does not affect cytoplasmic protein synthesis
machinery.

Both resting and Con A activated EFAD spleen cells
contain almost double the protein content of the control
and high PUFA groups. After 24 hours in culture, the sig-
nificance of the higher protein content of EFAD spleen
cells is unclear, particularly as it relates to lymphocyte
activation. It has been demonstrated that EFAD cells have
increased Na+K+-ATPase units and increased Con A binding
sites (28,29), suggesting that some cellular proteins are
present in greater amounts in EFAD cells compared with
normal cells. In contrast to total protein, 3H-uridine
incorporation into RNA and viability were not elevated in
the EFAD spleen cells compared with the control and high
PUFA groups.

The lack of dietary treatment effects on protein

1

102

and RNA synthesis is consistent with previous experiments
requiring extended culture times. Of particular interest
was the finding that the protein content of the EFAD cells
was almost twice that of the other two groups. Although the
significance of this is unknown, it points out a form of

metabolic adaptation that accompanies EFAD.

Effect of EFAD and Control Mouse Serum

on Mitogen Responses

Introduction

 

As previously discussed EFAD reduces in vivo plaque
forming and delayed type hypersensitivity responses, while
a high PUFA diet has no adverse effects on these responses
(12,13). Attempts to study the relationship between EFAD
and immune functions in long term cultures using FCS
supplemented media were inconclusive, possibly because
of "repair" of the EFAD cells by EFA in the FCS. I
then reasoned that one way to prevent in vitro repair was
to culture spleen cells from the EFAD mice in media sup-
plemented with serum from EFAD mice. Likewise, spleen cells
from control mice were cultured in media supplemented with
serum from mice fed the control diet. In addition, cross-
over studies were performed in which EFAD spleen cells were
cultured in control serum and control spleen cells in EFAD
serum. The results of these experiments are presented in

this section.

103

Materials and Methods

 

Mitogen Experiments. Spleen cell preparations, cell

 

culture, 3H-thymidine labelling, and cell harvesting was
performed exactly as described in earlier mitogen experiments
(p.47). Dose response curves for each of the mitogens used
were Obtained, but only the data from the Optimal mitogenic

doses are presented.

Preparation of Mouse Serum. ICR mice were used as

 

sources of serum for the cultures. These mice were ob-
tained at about 4 weeks of age and fed ad libitum either

the EFAD or control diet for 42-56 days before bleeding.
Only fresh serum was used in the experiments. In most cases
mice were bled the afternoon before an experiment and the
blood left to clot overnight in the refrigerator. The
serum was collected the next morning and used that day for

the experiment.

Results

The data (Table 8) clearly show that the mitogen re-
sponses of EFAD and control spleen cells were significantly
better in EFAD serum supplemented cultures compared with
responses in cultures supplemented with control serum.

Con A responses for both treatments were 50% higher in the

EFAD serum while PHA and LPS responses were ZOO-300% higher

104

Table 8. Effects of EFAD and Control Serum on Mitogen
Responses of Spleen Cells From Mice Fed the EFAD
or Control Diet.l

 

 

 

 

EFAD Serum Mice
EFAD Control
Unstimulated 406175 367161
Con A 42,85211164a 39,30211910a
PHA 32,85511657a 30,58111394a
LPS 3.9171372a 3,8371392a

Control Serum

 

Unstimulated 378142 280132
Con A' 28,13011320 24,09311181
PHA 11,9401972 14,50712484
LPS 1,4971218 1,7741266

 

 

1Values are mean 1 SEM of triplicate observations from each
of 8 mice/treatment.

aSignificantly greater than response in control serum.

105
in the EFAD serum cultures compared with the control serum
cultures. Values for the unstimulated cultures did not

differ between the two serum supplemented cultures.

Discussion

 

In light of in vivo results showing that EFAD impairs
plaque forming responses and delayed type hypersensitivity
it was completely unexpected that EFAD serum would enhance
in vitro mitogen responses. But even though these findings
do not agree with previous in vivo findings, they are in
Close agreement with the findings of Kollmorgen et al (85).
They found that mitogen responses were greater in cultures
with serum from animals fed a low fat diet than mitogen
responses in cultures supplemented with serum from mice
fed a high PUFA diet. As previously discussed in the
literature review, hyperlipoproteinemic serum also inhibits
in vitro lymphocyte transformation (87). But neither my
control mice (nor high PUFA mice), nor the hyperlipopro-
teinemic patients are immunodeficient.

Since Con A binds to serum lipoproteins containing Apo
B (140), it is possible that Con A responses may be reduced
in serum with elevated Apo B. But raising the dose level
of Con A in cultures did not overcome the inhibitory effect
and there is no evidence to date indicating that PHA or
LPS are bound up by serum lipOproteins. Thus, it is unclear

how serum, or lipOporteins, from animals fed moderate to

106

elevated levels of PUFA inhibit lymphocyte responses to
mitogens. Presently there are a number of laboratories
looking at the immunoinhibitory effects of lipOproteins. It
seems that studies should be expanded at this point to
include dietary experiments, rather than focusing exclusively
on in vitro experimentation. Progress in defining the
relationship between dietary PUFA intake, lipoprotein
metabolism, and immune functions should provide a basis for
interpreting some of the seemingly inconsistent findings

reported in these experiments.

CONCLUSIONS

The purpose of this research was to investigate the
effects of dietary EFA on various immune functions. The
first experiments discussed here were in close agreement
with my previous studies (12). Both reports support the
conclusion that high PUFA diets do not adversely affect in
vivo immune functions, but EFAD impairs in vivo immune
functions. That these findings are consistent in plaque
forming and delayed type hypersensitivity responses casts
some doubt on the proposal that PUFA impair immune responses
and "PUFA deficiency" enhances immunity. The findings
reported here could be expanded upon by other assays of in
vivo immunity. In particular, it would be interesting to
look at the effects of the diets used here on allograft
rejection, as this system was used to prOpose an immuno-
inhibitory role for PUFA.

The relationship between EFAD and immunity has not been
extensively studied. The data presented here should be a
useful basis for additional studies. From the standpoint
of applied nutrition it seems worthwhile to continue study-
ing the relationship between EFAD and immunity because of
the possibility of this deficiency developing in patients

receiving total parenteral nutrition.

107

108

At the level of basic research further study of EFA
and immunocompetent cells would also seem apprOpriate.
The evidence presented here point to a role for EFA in
lymphocyte activation. Some effects at the level of the
lymphocyte membrane have been demonstrated. Further studies
could develop the mechanism of this effect by looking at
the effects of EFAD on membrane bound enzymes and receptors.
On the other hand, EFAD may affect lymphocyte function by
altering the generation of EFA metabolites. This may af-
fect lymphocytes by impairing some intracellular process
associated with activation. Alternatively, EFAD may reduce
the generation of PG and related lipids by macrophages, and
loss of these regulators may affect lymphocyte function.

The effects of EFA metabolites on lymphocyte function
is under intense investigation. The findings of this
study indicate that EFAD is an additional factor for con-

sideration in this research.

REFERENCES

10.

REFERENCES

Mead, J. F., 1980. Nutrients with special functions:
Essential fatty acids. IN Human Nutrition. A
comprehensive treatise. Vol. 3A. Nutrition and the
adult. Macronutrients. Edited by R. B. Alfin-
Slater and D. Kritchevsky. Plenum Press, New York
and London. p. 213.

Stenson, W. F. and Parker, C. W. 1980. Prostaglan—
dins, macrOphages, and Immunity. J. Immunol. 125:1.

Goldyne, M. E., and Stobo, J. D. 1981. Immunoregula-
tory role of prostaglandins and related lipids.
Crit. Rev. Immunol. p. 189.

Pelus, L. M., and Strausser, H. R. 1977. Prostaglan-
dins and the immune response. Life Sci. 20:903.

Meade, C. J., and Mertin, J. 1978. Fatty acids and
immunity. £2 Adv. in lipid research. Vol. 16.
Edited by Paoletti, R. and Kritchevsky, D. Academic
Press. New York, p. 127.

DeWille, J. W. 1979. Effects of essential fatty acid
deficiency and various levels of dietary polyun-
saturated fatty acids On humoral immunity in mice.
Ms. Thesis.

Santiago-Delpin, E. A. and Szepsenwol, J. 1977.
Prolonged survival of skin and tumor allografts in
mice on high-fat diets: Brief communication. J. Nat.
Cancer Inst. 59:459.

Friend, J. V., Lock, S. O., Gurr, M. I. and Parish,
W. E. 1980. Effect of different dietary lipids on
the immune responses of Hartley strain guinea pigs.
Int. Arch. Allergy appl. Immunol. 62:292.

Meade, C. J. and Mertin, J. 1976. The mechanism of
immunoinhibition by arachidonic and linoleic acid:
effects on the lymphoid and reticuloendothelial
systems. Int. Archs. Allergy appl. Immunol. 51:2.

Mertin, J. and Hunt, R. 1976. Influence of polyun-
saturated fatty acids on survival of skin allografts
and tumor incidence in mice. Proc. Nat. Acad. Sci.
73:928.

109

11.

12.

13.

14.

15.

16.

17.

18.

19.

20.

110

Mertin, J. 1976. Effect of polyunsaturated fatty
acids on skin allograft survival and primary and

secondary cytotoxic response in mice. Transpl.
21:1.

DeWille, J. W., Fraker, P. J. and Romsos, D. R. 1979.
Effects of essential fatty acid deficiency, and
various levels of dietary polyunsaturated fatty acids
on humoral immunity in mice. J. Nutr. 109:1018.

DeWille, J. W., Fraker, P. J., and Romsos, D. R.
1981. Effects of dietary fatty acids on delayed
type hypersensitivity in mice. J. Nutr. 111:2039.

Stobo, J. D. 1977. Use of lectins as probes of
lymphocyte structure and function. 1p, The lympho-
cyte: structure and function. Part II. Edited by
J. J. Marchalonis. Marcel Dekker, Inc. New York,
p. 493.

Stober, S. 1977. Maturation of B lymphocytes:
changes in cell surface markers, physiological
characteristics, and immune function. Ip, Progress
in Immunology III. North-Holland Publishing Co.,
Amsterdam. p. 183.

Deluca, D., Warr, G. W., and Marchalonis, J. J. 1979.
The immunoglobulin-like T-cell receptor. II. Co-
distribution of Fab determinants and antigen on the
surface of antigen-binding lymphocytes of mouse
thymus. J. Immunogenetics 6:359.

Hume, D. A. and Weidemann, M. J. 1980. Mitogenic
lymphocyte transformation. Volume 2. A general
model for the control of mammalian cell prolifera-
tion and differentiation. Elsevier/North Holland
Biomedical Press. Amsterdam.

Rivers, J. P. W. and Frankel, T. L. 1981. Essential
fatty acid deficiency. Brit. Med. Bull. 37:59.

Hansen, A. E., Haggard, M. E., Boelsche, A. N.,
Adam, D. J. D., and Wiese, H. F. 1958. Essential
fatty acids in infant nutrition. III. Clinical man-
ifestations of linoleic acid deficiency. J. Nutr.
66:565.

Dudrick, S. J., Wilmore, D. W., and Vars, H. M. 1968.
Long-term parenteral nutrition with growth, develop-

ment, and positive nitrogen balance. Surgery, 64:
134.

21.

22.

23.

24.

25.

26.

27.

28.

29.

111

Paulsrud, J. R., Pensler, L., Whitten, C. F., Stewart,
S., and Holman, R. T. 1972. Essential fatty acid
deficiency in infants induced by fat-free intra-
venous feeding. Am. J. Clin. Nutr. 25:897.

Fleming, C. R., Smith, L. M., and Hodges, R. E.
1976. Essential fatty acid deficiency in adults
receiving total parenteral nutrition. Am. J. Clin.
Nutr. 29:976.

Blackburn, G. L. 1976. Adipose tissue as a valuable
source of essential fatty acids. 12' Fat emulsions
in parenteral nutrition. Edited by Meng, H. C., and
Wilmore, D. W. American Medical Association. p. 37.

Jeejeeboy, K. N., Marliss, E. B., Anderson, G. H.,
Greenberg, G. R., Kuksis, A., and Breckenridge, C.
1976. Lipid in parenteral nutrition: Studies of
clinical and metabolic features. In, Fat emulsions
in parenteral nutrition. Edited by_Meng, H. C.,
and Wilmore, D. W. American Medical Association.
p. 45.

Gerschenson, L. E., Mead, J. F., Harary, I., and
Haggerty, D. F. 1967. Studies on the effects of es-
sential fatty acids on growth rate, fatty acid com-
position, oxidative phosphorylation and respiratory
control of HeLa cells in culture. Biochim. et
Biophys. acta: 131:42.

Bloj, B., Morero, R. D., Farias, R. N., and Trucco,

R. E. 1973. Membrane lipid fatty acids and regulation
of membrane bound enzymes. Allosteric behaviour

of Erythrocyte Mg2+ ATPase, (Na+,K+)-ATPase and
acetylcholinesterase from rats fed different fat
supplemented diets. Biochim. et Biophys. acta.
311:67.

Solomonson, L. P., Liepkalns, V. A., and Spector,
A. A. 1976. Changes in (Na+,K+)-ATPase activity of
ehrlich ascites tumor cells produced by alteration
of membrane fatty acid composition. Biochem. 15:
892.

Lin, M. H., Romsos, D. R., Akera, T., and Leveille,
G. A. 1979. Increase in Na+,K+-ATPase enzyme units
in liver and kidneys from essential fatty acid de-

ficient rats. Experientia. 35:735.

Henriquez, D. S., Tepperman, H. M., and Tepperman,
J. 1979. Effects of high glucose and high fat diets
on concanavalin A binding to rat liver plasma mem-
branes and on the amount and pattern of their

glyCOprotein carbohydrates. J. Lipid Res. 20:624.

30.

31.

32.

33.

34.

35.

36.

37..

38.

39.

40.

41.

112

Fukazawa, T., Privett, O. S., and Takahashi, Y.
1970. Effect of essential fatty acid deficiency on
release of triglycerides by the perfused rat liver.
J. Lipid Res. 11:522.

Fukazawa, T., Privett, O. S., and Takahashi, Y.
1971. Effect of EFA deficiency on lipid transport
from liver. Lipids. 6:388.

Huang, M. T. and Williams, M. A. 1980. Essential fatty
acid deficiency and plasma triglyceride turnover.
Am. J. Physiol. 238:E499.

Pace-Asciak. 1977. Oxidative transformation of
arachidonic acid. Prostaglandins. 13:811.

Tinoco, J., Endemann, G., Hincenbergs, I., Medwadowski,
B., Miljanich, P., and Williams, M. A. 1980. Ef-

fects of linolenic acid deficiency on the fatty acid
patterns in plasma and liver cholesteryl esters, tri-
glycerides, and phospholipids in female rats. J.

Nutr. 110:1497.

Budowski, P. 1981. Review: Nutritional effects of
w3 polyunsaturated fatty acids. Isr. J. Med. Sci. 17:223.

Lands, W. E. M. 1979. The biosynthesis and metabolism
of prostaglandins. Ann. Rev. Physiol. 41:633.

McKean, M. L., Smith, J. B., and Silver, M. J.

1981. Formation of 1ySOphosphatidy1choline by human
platelets in response to thrombin. J. Biol. Chem.
256:1522.

Broekman, M. J., Ward, J. W., and Marcus, A. J. 1980.
Phospholipid metabolism in stimulated human plate-
lets. Changes in phOphatidylinositol, phosphatidic
acid, and lysophospholipids. J. Clin. Invest. 66:275.

Bass, D. A., O'Flaherty, J. T., Szejda, P., De-
Chatelet, L. R., and McCall, C. E. 1980. Role of
arachidonic acid in stimulation of hexose transport
by human polymorphonuclear 1eykocytes. Proc. Nat.
Acad. Sci. 77:5125.

Bass, D. A., Thomas, M. J., Goetzl, E. J., DeChatelet,
L. R., and McCall, C. E., 1981. Lipoxygenase derived
products of arachidonic acid mediate stimulation of
hexose uptake in human polymorphonuclear leukocytes.
Biochem. BiOphys. Res. Comm. 100:1.

Samuelsson, B. and Hammarstrom, S. 1981. Slow react-
ing substances and leukotrienes. Immunol. Today. 2:3.

42.

43.

44.

45.

46.

47.

48..

49.

50.

51.

52.

113

Ford-Hutchinson, T. 1981. Leukotrienes and other
lipoxygenase products. Immunol. Today. 2:9i.

Jakschik, B. A., Falkenhein, S., and Parker, C. W.
1977. Precursor role of arachidonic acid in release
of slow-reacting substance from rat baSOphilic
leukemia cells. Proc. Nat. Acad. Sci. 74:4577.

Goodwin, J. S. and Webb, D. R. 1980. Regulation of
the immune response by prostaglandins. Clin. Immunol.
and Immunopath. 15:106.

Smith, J. W., Steiner, A. L., Newberry, W. M., and
Parker, C. W. 1971. Cyclic adenosine 3',5' mono-
phosphate in human lymphocytes. Alterations after
phytohemagglutinin stimulation. J. Clin. Invest.
50:432.

Goodwin, J. S., Wilk, A., Lewis, M., Bankhurst, A.
D., and Williams, R. C. 1979. High affinity binding
sites for prostaglandin E on human lymphocytes.
Cell. Immunol. 43:150.

Muscoplat, C. C., Klausner, D. J., Brunner, C. J.,
Sloane, E. D., Johnson, D. W. 1979. Regulation of
mitogen and antigen stimulated lymphocyte blasto-
genesis by prostaglandins. Infection and Immunity.
26:311.

Burchiel, S. W. 1979. PGIZ and PGD2 effects on cyclic
AMP and human T cell mitogenesis. Prostaglandins and
Med. 3:315.

Goodwin, J. S., Bankhurst, A. D., and Messner, R. P.
1977. Suppression of human T cell mitogenesis by
prostaglandins. Existence of a prostaglandin pro-
ducing suppressor cell. J. Exp. Med. 146:1719.

Goodwin, J. S., Messner, R. P., and Peake, G. T.
1978. Prostaglandin suppression of mitogen stimu-
lated lymphocytes in vitro. Changes with mitogen
dose and preincubation. J. Clin. Invest. 62:753.

Webb, D. R. and Nowoiejski, I. 1977. The role of
prostaglandins in the control of the primary 19s
immune response to sRBC. Cell. Immunol. 33:1.

Zimecki, M. and Webb, D. R. 1977. The role of
prostaglandins in the control of the immune response
to an autologous red blood cell antigen (Hb). Cell.
Immunol. and Immunopathol. 3:420.

53.

54.

55.

56.

57.

58.

59..

60.

61.

62.

114

Kishimoto, R, and Ishizaka, K. 1976. Regulation of
the antibody response in vitro. J. Immunol. 116:534.

Mathias, M. M. and Dupont, J. 1979. The relationship
of dietary fats to prostaglandin biosynthesis. Lipids.
14:247.

Ferber, E. and Resch, K. 1977. Structure and physiol-
ogic role of lipids in the lymphocyte membrane. 1p,
The lymphocyte: Structure and function. Edited by,

J. J. Marchalonis. Marcel Dekker, Inc. New York.

p. 593.

Chen, S. S. H. 1979. Relationship between phos-
phatidylcholine biosynthesis and cellular commitment
in concanavalin A stimulated lymphocytes. Exp. Cell
Res. 121:283.

Chen, S. S. H. 1979. Enhanced sterol synthesis in
concanavalin A stimulated lymphocytes: correlation
with phospholipid synthesis and DNA synthesis. J.
Cell. Physiol. 100:147.

Marquardt, D. L., Nicolotti, R. A., Kennerly, D. A.,
and Sullivan, T. J. 1981. Lipid metabolism during
mediator release from mast cells: studies of the role
of arachidonic acid metabolism in the control of
phospholipid metabolism, J. Immunol. 127:845.

Ku, T. Y., Kishimoto, A., Takai, Y., Ogawa, Y.,

Kimura, S., and Nishizuka, Y. 1981. A new regulatory
system for protein phosphorylation in human peripheral
lymphocytes. II. Possible relation to phosphatidylino-
sitol turnover induced by mitogens. J. Immunol.
127:1375.

Sackmann, E., Albrecht, 0., Hartmann, W., and Galla,
H. J. 1977. Domain formation in lipid bilayers and
biological membranes. lg, Structural and kinetic
approach to plasma membrane function. Edited by C.
Nicolau, and A. Paraf. Springer Verlag. Berlin. p.
28.

Omary, M. B. and Trowbridge, I. S. 1981. Covalent
binding of fatty acid to the transferring receptor
in cultured human cells. J. Biol. Chem. 256:4715.

Knazek, R. A. and Liu, S. C. 1979. Dietary essen-
tial fatty acids are required for maintenance and
induction of prolactin receptors. Proc. Soc. Exp.
Biol. Med. 162:346.

63.

64.

65.

66.

67.

68.

69..

70.

71.

72.

73.

115

Shoyab, M. and Todaro, G. J. 1981. Perturbation of
membrane phosPholipids alters the interaction between
epidermal growth factor and its membrane receptor.
Arch. Biochem. and Biophys. 206:222.

Marshall, J. D., Heiniger, H. J., and Waymouth, C.
1979. High affinity concanavalin A binding to sterol
depleted L cells. J. Cell. Physiol. 100L539.

Melchior, D. L. and Czech, M. P. 1979. Sensitivity
Of the adipocyte D-glucose transport system to
membrane fluidity in reconstituted vesicles. J.
Biol. Chem. 254:8744.

Pilch, P. F., Thompson, P. A., and Czech, M. P.
1980. Coordinate modulation of D-glucose transport
activity and bilayer fluidity in plasma membranes
derived from control and insulin treated adipocytes.
Proc. Nat. Acad. Sci. 77:915.

Yuli, I. Wilbrandt, W., and Shinitzky, M. 1981.
Glucose transport through cell membranes of modified
lipid fluidity. Biochem. 20:4250.

Mendelsohn, J., Skinner, S. A., and Kornfeld, S.
1971. The rapid induction by phytohemagglutinin of
increased o-aminoisobutyric acid uptake by lympho-
cytes. J. Clin Invest. 50:818.

Van Den Berg, K. J., and Betel, I. 1974. Correlation
Of early changes in amino acid transport and DNA
synthesis in stimulated lymphocytes. Cell. Immunol.
10:319.

Guidotti, G. G., Borghett, A. F., Gazzola, G. C.
1978. The regulation of amino acid transport in
animal cells. Biochim. et Biophys. acta. 515:329.

Borghetti, A. F., Tramacere, M., Ghiringhelli, P.,
Severini, A., and Kay, J. E. 1981. Amino acid trans-
port in pig lymphocytes. Enhanced activity of trans—
port system ASC following mitogenic stimulation.
Biochem. et Biophys. acta. 646:218.

Awad, A. B. and Spector, A. A. 1976. Modification of
the fatty acid composition of ehrlich ascites tumor
plasma membranes. Biochim. et BiOphys., Acta. 426:
723.

Kaduce, T. L. Awad, A. B., Fontenelle, L. J., and
Spector, A. A. 1977. Effect of fatty acid saturation
on a-aminoisobutyric acid transport in ehrlich as-
cites cells. J. Biol Chem. 252:6624.

74.

75.

76.

77.

78.

79.

80.

81.

82.

83.

84.

116

Curtiss, L. K. and Edington, T. S. 1976. Regulatory
serum lipoproteins: regulation of lymphocyte stimu-
lation by a species of low density lipoprotein. J.
Immunol. 116:1452.

Curtiss, L. K., DeHeer, D. H., and Edgington, T. S.
1977. In vivo suppression of the primary immune
response by a species of low density serum lipo-
protein. J. Immunol. 118:648.

Curtiss, L. K., and Edgington, T. S. 1977. Effect
of LDL-In, a normal immunoregulatory human serum low
density lipoprotein, on the interaction of macro-
phages with lymphocytes proliferating in response to
mitogen and allogeneic stimulation. J. Immunol.
118:1966.

Hui, D. Y., Harmony, J. A. K., Innerarity, T. L.,
and Mahley, R. W. 1980. Immunoregulatory plasma
lipOproteins. Role of apoprotein E and apoprotein
B. J. Biol Chem 255:117755.

Curtiss, L. K. and Edgington, T. S. 1981. Immuno-
regulatory plasma low density lipoprotein: the bio-
logical activity and receptor binding specificity is
independent of neutral lipid. J. Immunol. 126:1008.

Curtiss, L. K. and Edgington, T. S. 1981. The bio-
logical activity of the immunoregulatory lipoprotein,

, LDL-In, is independent of its free fatty acid con-

tent. J. Immunol. 126:1382.

Edgington, T. S. and Curtiss, L. K. 1981. Plasma
lipOproteins with bioregulatory prOperties including
the capacity to regulate lymphocyte function and the
immune response. Cancer Res. 41:3786.

Hsu, K. H. L., Ghanta, V. K., and Hiramoto, R. N.
1981. Immunosuppressive effect of mouse serum lipo-
proteins. I. In vitro studies. J. Immunol. 126:1909.

Schaefer, E. J., Eisenberg, S., and Levy, R. L.
1978. LipOprotein apoprotein metabolism. J. Lipid
Res. 19:667.

Shine, T. E., Little, J. R., Shore, V., Medoff, G.,
and Abegg, A. 1980. Effect of exogenous lipids
and lipoproteins on the primary immune response in
vitro. Eur. J. Immunol. 10:714.

Schuh, J., Novogrodsky, A., and Haschmeyer, R. H.
1978. Inhibition of lymphocyte mitogenesis by
autoxidized low density lipoprotein. Biochem. Bio-
phys. Res. Comm. 84:763.

85.

86.

87.

88.

89.

90.

91.

92.

93.

94.

95.

96.

117

Kollmorgen, G. M., Sansing, W. A., Lehman, A. A.,
Fischer, G., Longley, R. E., Alexander, S. S., King,
M. M., and McCay,-P. B. 1979. Inhibition of lympho-
cyte function in rats fed high fat diets. Cancer
Res. 39:3458.

SOppi, E., Viikari, J., Eskola, J., and Ruuskanen,

O. 1979. Meal induced increase in serum triglycerides
has no effect on lymphocyte response to mitogens.

Eur. J. Clin. Invest. 9:281.

Waddell, C. C., Taunton, 0. D., and Twomey, J. J.
1976. Inhibition of lymphoproliferation by hyper-
lipoproteinemic plasma. J. Clin. Invest. 58:950.

Morse, J. H., Witte, L. D., and Goodmen, D. S. 1977.
Inhibition of lymphocyte proliferation stimulated by
lectins and allogeneic cells by normal plasma lipo-

proteins. J. Exp. Med. 146:1791.

Weissburger, J. H., Reddy, B. S., and wynder, E. L.
1977. Colon cancer: its epidemiology and experimental
production. Cancer. 40:2414.

Shils, M. E. 1979. Diet and nutrition as modifying
factors in tumor develOpment. Med. Clin N. Amer.
63:1027.

Gori, G. B. 1979. Dietary and nutritional implica-

_ tions in the multifactorial etiology of certain

prevalent human cancers. Cancer. 43:2151.

Gammal, E. B., Carroll, K. K., Plunkett, E. R.
1967. Effects of dietary fat on mammary carcino-
genesis by 7,12, dimethylbenz(a)anthracene. Cancer
Res. 27:1737.

Rao, A. and Abraham, S. 1976. Enhanced growth rate
of transplanted mammary adenocarcinoma induced in
C3H mice by dietary linoleate. J. Natl. Cancer Inst.
56:431.

Hopkins, G. J. and West, C. E. 1976. Possible roles
of dietary fats in carcinogenesis. Life Sci. 19:1103.

Hopkins, G. J. and Carroll, K. K. 1979. Relationship
between amount and type of dietary fat in promotion
of mammary carcinogenesis induced by 7,12, dimethyl-
benz(a)anthracene. J. Natl. Cancer Inst. 62:1009.

Reddy, B. S., Watanabe, K., and Weisburger, J. H.
1977. Effect of high fat diet on colon carbinogenesis

97.

98.

99.

100.

101.

102.

103.

104.

105.

106.

118

in F344 rats treated with 1,2-dimethylhydrazine,
methylazoxymethanol acetate, or methylnitrosourea.
Cancer Res. 37:4156.

Stutman, O. 1981. NK cells, anti-tumor surveillance
and interleukins. Immunology Today. 2:205.

Owen, R. L., Dau, P. C., Johnson, K. P., and Spitler,
L. E. 1980. Immunologic mechanisms in multiple
sclerosis. Exacerbation by type A hepatitis and skin
test antigens. J. Amer. Med. Assoc. 24:2301.

Merin, J., Knight, S., Rudge, P., Thompson, E. J.,
and Healy, M. J. R. 1980. Double blind, controlled
trial of immunosuppression in treatment of multiple
sclerosis. Lancet. 949.

Belin, J., Pettet, N., Smith, A. D., Thompson, R.

H. S., and Zilkha, K. J. 1971. Linoleate metabolism
in multiple sclerosis. J. Neurol. Neurosurg. Psy-
chiat. 34:25.

Homa, S. T., Belin, J., Smith,A.D., Monro, J. A.,

and Zilkha, K. J. 1980. Levels of linoleate and
arachidonate in red blood cells of healthy individuals
and patients with multiple sclerosis. J. Neurol.
Neurosurg. Psychiat. 43:106.

Mertin, J. and Stackpoole. 1978. Suppression by es-

. sential fatty acids of experimental allergic ence-

phalomyelitis is abolished by indomethacin. Pros-
tagland. and Med. 1:283.

Mertin, J. and Stackpoole, A. 1979. The spleen is
required for the suppression of experimental allergic
encephalomyelitis by prostaglandin precursors. Clin.
exp. Immunol. 36:449.

Hughes, D., Keith, A. B., Mertin, J., and Caspary,
E. A. 1980. Linoleic acid therapy in severe experi-
mental allergic encephalomyelitis in the guinea pig:
suppression by continuous treatment. Clin. exp.
Immunol. 41:523.

Millar, J. H. D., Zilkha, K. J., Langmen, M. J. S.
Wright, H. P. Smith, A. D., Belin, J., and Thompson,
R. H. S. 1973. Double blind trial of linoleate sup-
plementation of the diet in multiple sclerosis. Brit.
Med. J. 1:765.

Bates, D., Fawcett, P. R. W., Shaw, D. A., and
Weightmen, D. 1977. Trial of polyunsaturated fatty
acids in non relapsing multiple sclerosis. Brit. Med.
J. 2:932.

107.

108.

109.

110.

111.

112.

113.

114.

115.

116.

119

Bates, D., Fawcett, P. R. W., Shaw, D. A., and Weight-
man, D. 1978. Polyunsaturated fatty acids in treat-
ment of acute remitting multiple sclerosis. Brit.
Med. J. 2:1390.

Hansen, A. E., Beck, 0., and Weiss, H. F. 1948.
Susceptibility to infection manifested by dogs on a
low fat diet. Fed. Proc. 7:289 (abs).

Beisel, W. R., Edelman, R. Nauss, K., and Suskind,

R. M. 1981. Single nutrient effects on immunologic
functions. Report of a workshop sponsored by the

Department of Food and Nutrition and its nutrition

advisory group of the American Medical Association.
J. Amer. Med. Assoc. 245:53.

Erickson, K. L., McNeill, C. J., Gershwin, M. E.,

and Ossman, J. B. 1980. Influence of dietary fat

concentration and saturation on immune ontogeny in
mice. J. Nutr. 110:1555.

Mertin, J. and Hughes, D. 1975. Specific inhibitory
action of polyunsaturated fatty acids on lymphocyte

transformation induced by PHA and PPD. Int. Archs.

Allergy appl. Immunol. 48:203.

Ossman, J. B., Erickson, K. L. and Canolty, N. L.
1980. Effects of saturation and concentration of
dietary fats on lymphocyte transformation in mice.

1 Nutr. Rep. Int. 22:279.

Katz, D. H. and Benacerraf B. 1972. The regulatory
influence of activated T cells on B cell responses
to antigen. Adv. in Immunol. 15:1.

Vadas, M. A., Miller, J. F. A. P., Gamble, J. and
Whitelaw, A. 1975. A radioisotope method to measure
delayed-type hypersensitivity in the mouse. I.
Studies in sensitized and normal mice. Int. Archs.
Allergy Appl. Immunol. 49:670.

Vadas, M. A., Miller, J. F. A. P., McKenzie, I. F.
C., Chism, S. E., Shen, F. W., Boyse, E. A., Gamble,
J. R. and Whitelaw, A. M. 1976. Ly and Ia antigen
phenotypes of T cells involved in delayed-type hyper-
sensitivity and suppression. J. Exp. Med. 144:10.

Askenase, P. W. 1980. Effector cells in late and
delayed hypersensitivity reactions that are dependent
on antibodies or T cells. 12? Immunology 80, Pro-
gress in Immunology IV. (Fougereau, M. and Dausset.
J., ed.), Academic Press, London. p. 829.

117.

118.

119.

120.

121.

122.

123.

124.

125.

126.

120

Gill, J. L. 1978. Design and analysis of experiments
in the animal and medical sciences. The Iowa State
University Press, Ames, Ia. p. 44.

Clarke, S. D., Romsos, D. R. and Leveille, G. A.
1977. Differential effects of dietary methyl esters
of long-chain saturated and polyunsaturated fatty
acids on rat liver and edipose tissue lipogenesis.
J. Nutr. 107:1170.

Weyman, C., Morgan, S. J., Belin, J. and Smith, A.
D. 1977. Phytohemagglutinin stimulation of human
lymphocytes. Effect of fatty acids on uridine up-
take and phosphoglyccride fatty acid profile. Bio-
chim. Biophys. Acta. 496:155.

Kelly, J. P. and Parker, C. W. 1979. Effects of
arachidonic acid and other unsaturated fatty acids

on mitogenesis in human lymphocytes. J. Immunol.
122:1556.

Tonkin, C. H. and Brostoff, J. 1978. Do fatty acids
exert a Specific effect on human lymphocyte trans-
formation in vitro? Int. Arch. Allergy Appl. Im-
munol. 57:171.

Tombaccini, D., Fallani, A., Mugnai, G., and Ruggieri,
S. 1981. Lipid composition of Balb/c3T3, SV3T3,

and Concanavalin A selected revertant cells grown

in media containing lipid depleted serum. J. Lipid

' Res. 22:590.

Hatten, M. E., Horwitz, A. F., and Burger, M. M.
1977. The influence of membrane lipids on the pro-
liferation of transformed and untransformed cell
lines. Exp. Cell Res. 007:31.

Tombaccini, D., Ruggieri, S., and Mugnai, G. 1980.
Concanavalin A mediated agglitinibility of Balb/

c3T3 cells grown in media supplemented with different
phosphatidylcholines. Biochem. Biophys. Res. Comm.
96:1109.

Rothblatt, G. H., Arbogast, L. Y., Oulette, L., and
Howard, B. V. 1976. Preparation of delipidized serum

protein for use in cell culture systems. In vitro.
12:554.

Meo, T. 1979. The MLR test in the mouse, IN, Im-
munological Methods, Academic Press, New York, p.
227.

127.

128.

129.

130.

131.

132.

133.

134..

135.

136.

137.

138.

121

Spector, A. A. and Hoak, J. C. 1969. An improved
method for the addition of long-chain free fatty acid
to protein solutions. Anal. Biochem. 32:297.

Lowry, O. H., Rosborough, N. J., Farr, A. L., and
Randall, R. J. 1951. Protein measurement with the
Folin phenol reagent. J. Biol. Chem. 193:265.

Mosinger, F. 1965. Photometric adaptation of Dole's
microdetermination of free fatty acids. J. Lipid
Res. 6:157.

Hirst, J. A., Beverley, P. C. L. Kisielow, P. Hoffman,
M. K., and Oettgen, H. F. 1975. Ly antigens: markers
of T cell function on mouse spleen cells. J. Immunol.
115:1555.

Snell, G. D. 1978. T Cells, T cell recognition struc-
tures, and the major histocompatibility complex. Im-
munol. Rev. 38:3.

Gronowicz, E., Coutinho, A. and Moller, G. 1974.
Differentiation of B cells: Sequential appearance of
responsiveness to polyclonal activators. Scad. J.
Immunol. 3:413.

Nilausen, K. 1978. Role of fatty acids in growth
promoting effect of serum albumin on hamster cells
in vitro. J. Cell. Physiol. 96:1.

Gross, R. L., Osdin, N., Fong, L., and Newberne, P.
M. 1979. I. Decreased immunological function in zinc
deprived rats as measured by mitogen response of
spleen, thymus, and peripheral blood. Amer. J.

Clin. Nutr. 32:1260.

Despasquale-Jardieu. P. and Fraker, P. J. (submitted,
J. Immunol.) Assessment of in vitro and adoptive
transfer responses of splenocytes from zinc deficient
adult mice.

Corwin, L. M. and Shloss, J. 1980. Influence of
vitamin E on the mitogenic response of murine lymphoid
cells. J. Nutr. 110:916.

Chandra, R. K. 1980. Cell mediated immunity in
nutrition imbalance. Fed. Proc. 39:3088.

Helmreich, E. and Kipnis, D. M. 1962. Amino acid
transport in lymph node cells. J. Biol. Chem. 237:
2582.

139.

140.

122

Finkelstein, M. C. and Adelberg, E. A. 1977. Neutral
amino acid transport in an established mouse lympho-
cytic cell line. J. Biol. Chem. 252:7101.

McConathy, W. J. and Alaupouic, P. 1974. Studies of
the interaction of Concanavalin A with major density
classes of human plasma lipoproteins. Evidence for

the specific binding of lipOprotein B in its associated
and free forms. FEBS Lett. 41:174.