Citation
Development of insulin resistance in 3T3-L1 adipocytes

Material Information

Title:
Development of insulin resistance in 3T3-L1 adipocytes
Creator:
Thomson, Michael James, 1972-
Publication Date:
Language:
English
Physical Description:
xiii, 156 leaves : ill. ; 29 cm.

Subjects

Subjects / Keywords:
Adipocytes ( jstor )
Antibodies ( jstor )
Cell membranes ( jstor )
Diabetes ( jstor )
Glycogen ( jstor )
Hexosamines ( jstor )
Insulin ( jstor )
Insulin resistance ( jstor )
Proteins ( jstor )
Type 2 diabetes mellitus ( jstor )
Adipocytes ( mesh )
Department of Biochemistry and Molecular Biology thesis Ph.D ( mesh )
Dissertations, Academic -- College of Medicine -- Department of Biochemistry and Molecular Biology -- UF ( mesh )
Glucosamine -- pharmacology ( mesh )
Glucose -- pharmacology ( mesh )
Insulin -- pharmacology ( mesh )
Insulin Resistance ( mesh )
Monosaccharide Transport Proteins ( mesh )
Research ( mesh )
Genre:
bibliography ( marcgt )
non-fiction ( marcgt )

Notes

Thesis:
Thesis (Ph.D.)--University of Florida, 1999.
Bibliography:
Bibliography: leaves 145-155.
General Note:
Typescript.
General Note:
Vita.
Statement of Responsibility:
by Michael James Thomson.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
The University of Florida George A. Smathers Libraries respect the intellectual property rights of others and do not claim any copyright interest in this item. This item may be protected by copyright but is made available here under a claim of fair use (17 U.S.C. §107) for non-profit research and educational purposes. Users of this work have responsibility for determining copyright status prior to reusing, publishing or reproducing this item for purposes other than what is allowed by fair use or other copyright exemptions. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder. The Smathers Libraries would like to learn more about this item and invite individuals or organizations to contact the RDS coordinator (ufdissertations@uflib.ufl.edu) with any additional information they can provide.
Resource Identifier:
030406754 ( ALEPH )
51619713 ( OCLC )

Downloads

This item has the following downloads:


Full Text













DEVELOPMENT OF INSULIN RESISTANCE IN 3T3-L1 ADIPOCYTES


By

MICHAEL JAMES THOMSON

















A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA


1999






























This is dedicated to God, my parents, my dearest Amy, and

Moxie.
















ACKNOWLEDGMENTS


I would sincerely like to thank the members of my

committee Drs. Michael Kilberg, Charles Allen, Thomas

O'Brien, Mark Atkinson, and my advisor Dr. Susan Frost for

their guidance throughout the course of this project. In

addition, I would like to thank the members of the lab and

my friends for their help throughout the time I have spent

here and for making this such an enjoyable period in my

life. My most heartfelt thanks goes to my parents, without

whose love and support I would never have achieved such an

undertaking. Last, but certainly not least, I would like to

thank my dearest Amy, whose love and companionship over the

last several months has not only marked a highlight in my

graduate career, but in my life as well.
















TABLE OF CONTENTS

page

ACKNOWLEDGMENTS ......................................... iii

LIST OF TABLES .......................................... vii

LIST OF FIGURES ........................................ viii

ABSTRACT .................................................. xi

CHAPTERS

1 INTRODUCTION ......................................... 1

Overview of Diabetes ..................................... 1
Insulin Resistance....................................... 2
Role of Adipose in Insulin Resistance ............... 3
Role of the Hexosamine Biosynthetic
Pathway in Insulin Resistance .................. 6
3T3-L1 Adipocytes ........................................ 8
GLUT Transporter Family................................. 9
Characteristics of GLUT Family...................... 9
Translocation of GLUT4............................. 12
Insulin Receptor and Signaling......................... 13
3T3-L1 Adipocytes: A Model for
Insulin Resistance.................................. 17

2 MATERIALS AND METHODS.................................. 18

Materials ............................................ 18
Methods .............................................. 19
Cell Culture ........................................... 19
Chronic Insulin Treatment and
Insulin Washout................................. 20
Glucose Transport Assay............................ 21
Subcellular Fractionation of
3T3-L1 Adipocytes............................... 21
Markwell Assay for Protein Determination........... 22
Gel Electrophoresis................................ 24
Electrotransfer and Western Blotting............... 25
RNA Isolation and Northern Blotting................ 26
Glycogen Analysis ................................... 27









Glycogen Synthesis .................................. 30
Production and Characterization of
GLUT4 Antibody................................... 30
Peptide Purification of GLUT4 Antibody............. 36
Preparation of Total Membranes ................... .. 39
Metabolic Labeling of 3T3-LI Adipocytes............ 40
Immunoprecipitation of GLUT4....................... 41
ATP-Luciferase Assay............................... 43
Leptin Assay........................................ 44

3 DEVELOPMENT OF INSULIN RESISTANCE .................... 49

Introduction ......................................... 49
Results .............................................. 50
Insulin-resistant Glucose
Transport Activity.............................. 50
Role of Glucose and Glucosamine.................... 58
Effects of Various Inhibitors on the
Development of Insulin Resistance............... 61
Effect of Chronic Insulin on GLUT4
Expression and Translocation.................... 69
Effect of Glucose Deprivation on
GLUT4 Expression................................ 77
Effect of Chronic Insulin on
Glut4 mRNA Expression........................... 77
Reversal of Insulin Resistance..................... 80
Conclusions .......................................... 82

4 MECHANISMS OF DECREASED GLUT4 LEVELS IN
INSULIN-RESISTANT CELLS ............................ 88

Introduction ......................................... 88
Results .............................................. 89
Specificity and Efficiency of
GLUT4 Immunoprecipitation........................ 89
Synthesis of GLUT4 in
Insulin-resistant Cells......................... 90
Degradation of GLUT4 in
Insulin-resistant Cells......................... 94
Effects of Cycloheximide on the
Loss of GLUT4 .................................... 99
Effects of Protease Inhibitors on
GLUT4 Expression................................ 99
Conclusions .......................................... 102









5 ROLE OF THE HEXOSAMINE BIOSYNTHETIC
PATHWAY IN THE DEVELOPMENT OF
INSULIN RESISTANCE................................ 107

Introduction ......................................... 107
Results ........ ............................... .. ... ... .. 110
Effects of Fructose Feeding on the
Glucose Transport System....................... .. 110
Effect of Fructose Feeding on
Glycogen Metabolism............................. 117
Conclusions .......................................... 123

6 CONCLUSIONS AND FUTURE DIRECTIONS .................... 129

Conclusions .......................................... 129
Future Directions ..................................... 132

APPENDICES

A GLUCOSE DEPRIVATION AND GLUT TRANSPORTERS............. 137

B LEPTIN AND INSULIN RESISTANCE......................... 141

REFERENCES .............................................. 145

BIOGRAPHICAL SKETCH...................................... 156
















LIST OF TABLES


Table


5-1 Glycogen Synthesis in 3T3-L1 Adipocytes............. 124


page
















LIST OF FIGURES


Figure page

1-1 Overview of the Hexosamine Biosynthetic Pathway.... 7

1-2 Predicted Secondary Structure of GLUT
Transporters ...................................... 10

2-1 Subfractionation of 3T3-Ll Adipocytes............... 23

2-2 Fluctuations in Glycogen Levels Over a Week........ 29

2-3 Specificity of GLUT4 Antiserum for
GLUT4 Peptide ..................................... 34

2-4 Specificity of GLUT4 Antiserum for
GLUT4 Protein..................................... 35

2-5 Elution Profile of Peptide-Purified
anti-GLUT4 Antibody.............................. 38

2-6 ATP Standard Curve................................... 45

2-7 Leptin Standard Curve............................... 48

3-1 Effect of Chronic Insulin on Glucose
Transport in 3T3-L1 Adipocytes .................. 53

3-2 Time Course of Acute Insulin Stimulation............ 54

3-3 Development of Insulin Resistance ................... 57

3-4 Effects of Glucose and Glucosamine on
Insulin Resistance............................... 60

3-5 Effects of Actinomycin and Cycloheximide on
Insulin Resistance............................... 62

3-6 Effects of Nikkomycin Z on Insulin Resistance...... 64

3-7 Effects of Tunicamycin on Insulin Resistance....... 66









3-8 Effects of E-64 on Insulin Resistance............... 67

3-9 Effects of E-64 on Glycogen Synthesis............... 68

3-10 Effects of Acute and Chronic Insulin on
Total GLUT Protein Levels........................ 70

3-11 Subfractionation of Insulin-Resistant
3T3-Ll Adipocytes.............................. 72-73

3-12 Effects of Chronic Insulin on GLUT4
Expression........................................ 75

3-13 Effects of Chronic Insulin on GLUT4
Expression with Larger Samples ................... 76

3-14 Effect of Glucose Deprivation on
GLUT4 Expression................................. 78

3-15 Effect of Chronic Insulin Treatment on
GLUT4 mRNA Levels................................ 79

3-16 Reversal of Insulin Resistance...................... 81

3-17 Reversal of Insulin Resistance in the
Presence of Cycloheximide........................ 83

4-1 Titration of Peptide-Purified GLUT4 Antibody
for Immunoprecipitation.......................... 91

4-2 Specificity of the Peptide-Purified GLUT4
Antibody in Immunoprecipitation.................. 92

4-3 Efficiency of GLUT4 Immunoprecipitation............. 93

4-4 Effect of Chronic Insulin on the
Synthesis of GLUT4 ............................... 96

4-5 Effect of Chronic Insulin on the
Degradation of GLUT4............................. 98

4-6 Degradation of Total Protein in
Cells Treated with Chronic Insulin............... 100

4-7 Effect of Cycloheximide on the Loss of GLUT4....... 101

4-8 Effects of Protease Inhibitors on
GLUT4 Expression................................ 104









5-1 Effects of Fructose on the Development of
Insulin Resistance............................... 112

5-2 Effect of Fructose on GLUT4 Transporter Levels..... 114

.5-3 Translocation of GLUT4 in Cells Treated with
Fructose .......................................... 116

5-4 Effects of Fructose on ATP Levels ................... 119

5-5 Effect of Fructose and Chronic Insulin on
Glycogen Levels.................................. 120

5-6 Effects of Fructose and Chronic Insulin on
Glycogen Synthesis............................... 122

A-l Effect of Glucose Deprivation and Chronic
Insulin on the Aberrant Glycosylation of
GLUT1 and GLUT4 .................................. 138

A-2 Effect of Glucose Deprivation and Chronic
Insulin on Glycogen Levels...................... 140

B-i Leptin Levels ....................................... 144
















Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy

DEVELOPMENT OF INSULIN RESISTANCE IN 3T3-Ll ADIPOCYTES

By

Michael James Thomson

August 1999


Chairperson: Susan C. Frost
Major Department: Biochemistry and Molecular Biology

Insulin resistance is a manifestation of both diabetes

mellitus and obesity. However, the mechanism is still not

clearly identified. My goal was to determine if an in vitro

system, 3T3-Ll adipocytes, could serve as a model for

insulin resistance. This study describes a procedure that,

indeed, allows the evaluation of the development of insulin

resistance in 3T3-Ll adipocytes. Under these conditions, it

has been shown that the concentration of insulin required

for 50% desensitization of glucose transport activity is 100

pM and maximal desensitization could be achieved with 1 nM.

This demonstrates for the first time that 3T3-Ll adipocytes

develop insulin resistance in response to physiologically

relevant concentrations of insulin. Glucose (or

glucosamine), in addition to insulin, was required to









establish desensitization. The expression of GLUT4 protein

decreased by 50% with exposure to 10 nM insulin. The dose-

dependent loss of GLUT4 was similar to the dose-dependence

for insulin-resistant transport activity. Translocation in

the presence of acute insulin was apparent, but the extent

of recruitment directly reflected the decrease in GLUT4

protein. GLUT4 mRNA also declined, but the K50 was

approximately 5 nM. Together, these data suggest that the

loss of GLUT4 protein underlies the cause of

desensitization. This decrease in GLUT4 protein was found

to be a combination of both decreased synthesis and

accelerated degradation.

Glucose, through its metabolism via the hexosamine

pathway, has been implicated in the regulation of insulin-

sensitive glucose transport. Surprisingly, fructose, an

alternate substrate for this pathway, does not appear to

support the development of insulin-resistant transport. To

study this apparent anomaly, we examined the levels and

subcellular distribution of GLUT4 in 3T3-L1 adipocytes in

response to chronic and acute insulin in the presence of

fructose or glucose. The results indicate that cells

exposed to either glucose or fructose, with chronic insulin,

down-regulate the expression of GLUT4. However, cells

exposed to fructose and chronic insulin and subsequently

stimulated with insulin, reveal levels of GLUT4 in the









plasma membrane which are similar to controls. While the

mechanism underlying this apparent "enhancement" of GLUT4

translocation is unknown, it explains the observation that

the insulin-stimulated glucose transport activity in cells

exposed to fructose does not reflect the insulin-resistant

state.
















CHAPTER 1
INTRODUCTION


Overview of Diabetes


Diabetes is one of the leading causes of death and

disability in the United States. An estimated 16 million

people in the United States have diabetes mellitus and about

half of these people do not know that they have the

disorder. Of these 16 million, 127,000 are under the age of

19. Each year, an additional 650,000 people are diagnosed

with diabetes. In 1992, diabetes contributed to the deaths

of more than 169,000 people and cost $92 billion in direct

and indirect costs. In total, diabetes accounted for 1 in 7

health care dollars spent (1).

Diabetes results when the pancreas produces little or

no insulin or the body's cells do not respond to insulin.

As a result, glucose accumulates in the blood, leading to

kidney overload and glucose excretion into the urine. There

are several types of diabetes: type I, type II, and

gestational diabetes. Type I, or insulin-dependent diabetes

mellitus (IDDM), is considered an autoimmune disease during

which the immune system attacks and destroys the insulin-

producing beta cells of the pancreas. Over time, this leads









to an inability of the pancreas to produce insulin. Type

II, or non-insulin-dependent diabetes mellitus (NIDDM), is

the most prevalent and least understood form of diabetes.

About 90-95% of the people with diabetes have type II. This

form usually develops in adults over 40 and is most common

in adults over 55. Type II diabetes is also associated with

obesity in that 80% of the people with this form of diabetes

are overweight. With this form, the pancreas generally

produces insulin, but the body does not respond to insulin

effectively, a condition known as insulin resistance.

Gestational diabetes develops during pregnancy and generally

disappears afterwards. However, women who develop diabetes

during pregnancy have a greater risk of developing type II

diabetes later in life.


Insulin Resistance


Insulin resistance is one of the determining

characteristics of type II diabetes mellitus (NIDDM) and

obesity and is believed to be the underlying pathology (1).

NIDDM is characterized by both hyperglycemia and

hyperinsulinemia, whereas obesity manifests only

hyperinsulinemia (2). Complications of insulin resistance

and diabetes include retinopathy (3), nephropathy (4),

neuropathy (5), and artherosclerosis (6). The

hyperlipidemia (7), hypercoagulation of platelets (8), and









artherosclerosis associated with diabetes can lead to

coronary artery disease, which is the leading cause of

mortality among diabetics (6).

Insulin resistance is defined as a decreased biological

response to physiological concentrations of insulin in

insulin-responsive tissues (1). In adipocytes, this is

characterized by a reduction in insulin-stimulated glucose

transport activity (9). As of yet, the exact mechanism

leading to the onset of insulin resistance has yet to be

elucidated. However, several mechanisms have been proposed

including decreased autophosphorylation of the insulin

receptor (2,10-13), down-regulation of the insulin-

responsive glucose transporter GLUT4 (2,14), altered

translocation of GLUT4 to the plasma membrane from its

intracellular storage vesicles (15-17), and the actions of

the intermediates or products of the hexosamine biosynthetic

pathway (18-21).


Role of Adipose in Insulin Resistance


As mentioned, many obese subjects also exhibit insulin

resistance. However, when these patients are placed on

calorie-restrictive diets to decrease their weight, insulin

resistance is reversed (22,23). This suggests that adipose

plays an important role in insulin resistance. Another

condition which shows the importance of adipose in insulin









resistance is the metabolic syndrome X. This syndrome

manifests diabetes, hypertension, and dyslipidemia with

obesity as the underlying cause.

Adipose also secretes several factors which may be

involved in insulin resistance. For instance, evidence

exists that adipose tissue of obese mice have elevated TNF-

a mRNA levels and secretion of TNF-c protein (24-26).

Also, administration of TNF-a to Zucker fa/fa rats causes

reduced autophosphorylation of the insulin receptor as well

as decreased insulin-stimulated phosphorylation of IRS-1 in

both muscle and adipose tissue. Neutralization of TNF-a by

the addition of soluble TNF-a receptors prevents these

effects (10) and increases insulin sensitivity (24). It has

also been shown that TNF-a down-regulates GLUT4 mRNA levels

in adipocyte and myocyte cultures (24,27,28) as well as

reduces insulin-stimulated [3H]2-deoxyglucose transport into

3T3-L1 adipocytes (11). These studies provide evidence that

TNF-a plays a role in inducing insulin resistance in

adipose as well as muscle. One can imagine that increased

expression of TNF-a by adipose feeds back directly on

adipose by an autocrine loop mechanism, while its effects on

muscle may occur through a paracrine mechanism (25).

However, it has been shown in db/db mice that the elevated

levels of plasma TNF-a are well below those required to









support insulin resistance in cultured cell systems (24,25).

Therefore, it is very plausible that other proteins or

factors may be secreted by the adipose tissue which induce

insulin resistance.

Fatty acids have also been shown to be involved in the

induction of insulin resistance. In 1963, Randle et al.

(29) hypothesized that increased serum levels of free fatty

acids could interfere with glucose uptake and metabolism in

muscles, resulting in insulin resistance. This has come to

be known as the Randle hypothesis. Since the presentation

of this hypothesis, it has been shown by several groups,

including Randle's, that elevated serum free fatty acids

interfere with glucose utilization in vitro and in vivo (30-

33). In further support of this hypothesis, Bierman et al.

(34) have shown that type II diabetics have elevated fasting

levels of free fatty acids. In contrast, the absence of

white adipose tissue can also lead to diabetes as shown by

studies using a transgenic mouse model (35). These mice

express a dominant negative protein, called A-ZIP/F, which

prevents the DNA binding of B-ZIP transcription factors of

both the C/EBP and Jun families, which are necessary for

white adipose differentiation and formation. These mice are

diabetic, with fatty livers and elevated serum glucose,

insulin, and surprisingly, free fatty acids and

triglycerides. Interestingly, when fat is transplanted into









these mice, they lose the diabetic phenotype. This might

argue against the participation of adipose in the

development of insulin resistance. However, the fact that

obesity leads to insulin resistance demonstrates the

significance of adipose, suggesting that a small amount of

fat is beneficial, but a large excess is detrimental.


Role of the Hexosamine Biosynthetic Pathway in Insulin
Resistance


Marshall and colleagues (18) proposed that the

hexosamine biosynthetic pathway was involved in the

induction of insulin resistance in adipocytes because of the

in vitro requirement for glucose, glutamine, and insulin

(15,20). Figure 1-1 represents a diagram of this pathway.

In this pathway, glucose is transported into the cell by

either GLUT1, the constitutive glucose transporter, and/or

by GLUT4, the insulin-responsive transporter, where it is

rapidly phosphorylated to glucose 6-phosphate. This is then

isomerized to fructose 6-phosphate and the majority (97%) of

this is shunted into glycolysis and oxidative metabolism.

However, 2-3% of the fructose 6-phosphate enters the

hexosamine biosynthetic pathway through the rate-limiting

enzyme, glutamine:fructose 6-phosphate amidotransferase

(GFAT). GFAT transfers the amide group from glutamine to

fructose 6-phosphate to form glucosamine 6-phosphate.

Through several other reactions, UDP-N-acetylglucosamine

















Glucose-6-P


Glucosamine-6-P


Fructose-6-P 97 Glycolytic/TCA
PDfhliwaw


Glutamine
GFAT 1
4, Glutamate

> Glucosamine-6-P


UDP-N-Acetylglucosamine


glycolipids
glycosylated proteins
GPI-linked proteins


*. Insulin
Resistance


Figure 1-1. Overview of the Hexosamine Biosynthetic
Pathway.


= =,i i *f,









is formed, which is used in the synthesis of glycolipids,

glycoproteins, and GPI-linked proteins. It was also shown

that glucosamine was at least 40 times more potent than

glucose in inducing insulin resistance (18). This

phenomenon can be explained by appreciating the fact that

glucosamine, which is transported into cells by the same

transporter as glucose, bypasses GFAT through direct

phosphorylation by hexokinase (18,21). The role of this

pathway was also confirmed by the use of two inhibitors of

GFAT: O-diazoacetyl-L-serine (azaserine) and 6-diazo-5-

oxonorleucine (DON), which are well established glutamine

analogs that irreversibly inhibit reactions in which

glutamine is an amide donor. In isolated rat adipocytes,

administration of these compounds blocked the development of

insulin-resistant glucose transport (18). Therefore, it was

hypothesized that products and/or intermediates of this

pathway are responsible for inducing insulin resistance.


3T3-Ll Adipocytes


The 3T3-Lls derive from murine embryonic fibroblasts

that, under the appropriate cell culture conditions,

differentiate into adipocytes (36,37). In the fibroblast

state, cells can be propagated and/or frozen under cryogenic

conditions. When these cells differentiate into adipocytes,

they exhibit the morphological and biochemical









characteristics of adipocytes. Differentiation is a

terminal process at which point the cells no longer divide.

Rather, the cells accumulate lipid droplets and express

lipogenic enzymes such as glycerophosphate acyltransferase

and malic enzyme (38), ATP-citrate lyase, acetyl Co-A

carboxylase, and fatty acid synthetase (39). These cells,

once differentiated, also express increased numbers of

insulin-binding sites (40). As adipocytes, these cells

express two isoforms of the glucose transporter family:

GLUT1, the constitutive glucose transporter, and GLUT4, the

insulin-sensitive glucose transporter (41,42). This feature

is characteristic of authentic adipocytes (43). In

addition, GLUT4 is only expressed in these cells following

differentiation (42).


GLUT Transporter Family


Characteristics of GLUT Family


GLUT1 and GLUT4 are members of a larger group of

proteins responsible for the facilitated transport of

glucose into cells (44). The members of this family were

discovered using a variety of molecular biology techniques

and named in the order in which they were identified, GLUT1-

5 (for GLUcose Tranporters). Figure 1-2 shows the predicted

secondary structure of these transporters, based on

hydropathy plots, as they might appear in the plasma




























45 1 :1 z? 0 1 PM




N ^c
IN









Figure 1-2. Predicted Secondary Structure of GLUT
Transporters.









membrane of the cell. These proteins are integral membrane

proteins consisting of twelve membrane spanning domains.

Both the amino and carboxy termini are located on the

intracellular side of the membrane. A large intracellular

loop between the sixth and seventh membrane spanning domains

divides the two halves of the protein. There is also a

single N-linked glycosylation site in the first

extracellular loop. The overall amino acid homology within

this family of proteins is 68% (45) and the amino acid

sequence identity between GLUT1 and GLUT4 is 63% (42).

The members of this family have differential tissue

distribution: GLUT1 is ubiquitously expressed at high levels

in fetal tissues, erythrocytes, brain, kidney, colon, and

adipocytes; GLUT2 is expressed in liver, P-cells, kidney,

and small intestine; GLUT3 is expressed in many tissues

including brain, placenta, and kidney; GLUT4 is expressed in

skeletal muscle, heart, and adipocytes; and GLUT5, a

proposed fructose transporter as well, is expressed in the

small intestine (46). The Km'S of these transporters also

differ. The Km's of GLUT1, 2, 3, and 4 for 3-0-

methylglucose, a none metabolizable form of glucose, are 21,

42, 10, and 2 mM, respectively, based on the oocyte

expression system (5). The high Km of GLUT2 is rationalized

by its localization to tissues involved in glucose

homeostasis (liver) and glucose sensing (P-cells).









Therefore, the flux of glucose into these tissues would be

expected to vary in a linear fashion over the typical range

of blood glucose levels (5). Following a meal, the very low

Km associated with GLUT4 would ensure that it operates close

to its Vmax even at decreasing blood glucose levels to ensure

maximum uptake of glucose from the blood into insulin-

responsive tissues (5).


Translocation of GLUT4


GLUT4, also referred to as the insulin-responsive

glucose transporter, resides in the trans-Golgi network of

vesicles that translocate to the plasma membrane in response

to insulin stimulation (47-49). This translocation

increases the number of glucose transporters on the cell

surface, and along with the lower Km of GLUT4, increases the

flux of glucose into the cell. Several groups have been

credited with cloning GLUT4 (42,50-53), which has ultimately

enabled studies investigating the regulation of its

translocation.

The basis for understanding translocation comes from

Rothman and colleagues (54) who proposed the SNARE hypothesis

for vesicular trafficking. This theory proposes that a

unique vesicle-bound molecule (v-SNARE) specifically

recognizes and binds to a unique receptor molecule (t-SNARE)

on the target membrane. Three different v-SNARES (VAMP-l,









VAMP-2, and cellubrevin) and eight different t-SNARES

(Syntaxin-l, Syntaxin-2, Syntaxin-3, Syntaxin-4, Syntaxin-6,

Syntaxin-7, SNAP-25, and Syndet) have been described thus

far (55). VAMP-2 is associated with GLUT4 containing

vesicles and is the v-SNARE that is responsible for binding

to the t-SNARES Syntaxin-4 and Syndet, the murine homologue

of SNAP-25 (56,57). In addition, the protein Muncl8c has

been shown to regulate the translocation of GLUT4 by

inhibiting the association of VAMP-2 with Syntaxin-4 (58).

Insulin relieves this inhibition by inducing the

dissociation of Muncl8c from Syntaxin-4 allowing VAMP-2 to

bind to Syntaxin-4. However, it is not known yet how

insulin stimulates the association and dissociation of these

molecules. In addition, other molecules may be involved in

GLUT4 translocation as investigations continue in this

exciting area.


Insulin Receptor and Signaling


Insulin action is mediated by a specific cell-surface

receptor. The insulin receptor is a heterotetramer

consisting of two c- and two P-subunits. Each a- and 3-

subunit is connected by a disulfide bond to form an a/P

dimer and each a/P dimer is then attached via a single

disulfide bond between the a-subunits to form the tetramer.

The a-subunit has a molecular mass of 135,000 Da and is









located completely on the outer surface of the plasma

membrane and contains the insulin binding site. The

3-subunit has a molecular mass of 95,000 Da and contains a

single transmembrane domain. The intracellular domain of

this subunit contains a juxtamembrane domain, an ATP binding

domain, and several tyrosine residues that are capable of

being phosphorylated. The intracellular domain also

contains a tyrosine kinase that is activated upon binding of

insulin presumably via propagation of a conformational

change from the a-subunit, which is transmitted through the

transmembrane domain of the 3-subunit. Once the tyrosine

kinase has been activated, it is capable of

autophosphorylating tyrosines in specific regions of the

intracellular domain of the P-subunit including the

juxtamembrane domain, the regulatory domain, and the carboxy

terminal tail (1). This tyrosine kinase also phosphorylates

selected proteins on tyrosine residues within the cell such

as IRS-1. The phosphorylated IRS-1 recognizes and binds to

the src homology-2 (SH2) domains of various signal

transduction proteins, two of which are Grb2 and PI 3-

kinase. The activation of these molecules sets up divergent

signaling pathways within the cell.

Activation of Grb allows it to bind to son-of-sevenless

(SOS) through an SH3 domain. The Grb2/SOS complex then

activates p21ras which has been shown to bind directly to









Raf-l serine/threonine kinase. This leads to the activation

of MAP kinase kinase which phosphorylates and activates MAP

kinase which activates transcription factors and increases

gene expression (59). It should be noted that

pharmacological concentrations of insulin are required for

activation of gene transcription. This brings doubt to the

physiological relevance of this path to insulin action (60).

The unique aspect of insulin action is its short-term

regulation of metabolic events including increased glucose

transport, increased glycogen synthesis, increased lipid

synthesis, and decreased lipolysis (1). Recent evidence has

shown that the PI 3-kinase cascade is the major pathway for

GLUT4 translocation and thus the stimulation of glucose

transport (61,62). PI 3-kinase is activated when the

phosphorylated IRS-1 binds to the SH2 domain of its p85a

regulatory subunit, this allows the pll0 subunit to become

active. Activated PI 3-kinase leads to the activation of

the serine/threonine kinase Akt (also known as protein

kinase B). Although Akt is involved in the specific

stimulation of GLUT4 translocation (62), the exact mechanism

by which this occurs is not known. However, inhibitor

studies has shown the importance of the PI 3-kinase pathway

over the p21ras pathway. For instance, wortmannin, an

inhibitor of PI 3-kinase, blocks the stimulation of glucose

transport by insulin (63). In contrast, inhibition of









endogenous p21ras has no effect on insulin-stimulated GLUT4

translocation (61).

The insulin receptor and the insulin signaling

cascade(s) are important in insulin resistance because it

has been shown that defects in autophosphorylation of the

insulin receptor leads to insulin-resistant cells (2,10-13).

This leads to a decrease in the phosphorylation of IRS-1

(2,10-13) which in turn leads to a reduction in signaling

cascades. This effect, however, may primarily be due to

secondary effects of TNF-x and the pharmacological doses of

insulin used. Genetic defects in the insulin receptor also

result in decreased autophosphorylation of the receptor and

decreased phosphorylation of IRS-1. This is only a minor

cause for the development of insulin resistance, as it does

not account for the induction of insulin resistance in

individuals with normal insulin receptors and those

individuals who develop insulin resistance as a result of

becoming obese. However, there is evidence in 3T3-L1

adipocytes that chronic insulin treatment alone can lead to

impaired insulin receptor signaling and down-regulation of

IRS-1 expression and phosphorylation (64). Again, these

studies were performed using pharmacological concentrations

of insulin.









3T3-Ll Adipocytes: A Model for Insulin Resistance


Many investigators have used the 3T3-L1 adipocyte cell

line to investigate insulin action. The majority of these

studies have used pharmacological doses of insulin to define

mechanisms. Under these conditions, insulin acts as a

growth factor sharing many of the mitogenic signaling paths

elicited by other growth factors (60). The metabolic

effects of insulin, observed at physiological

concentrations, are unique to insulin and cannot be

reproduced by other cellular stimuli (65-67). Thus, it

became important to determine the development of insulin

resistance under physiological conditions, the goal of this

research. I show that 1.) insulin-stimulated transport

activity decreases in response to physiological insulin; 2.)

GLUT4 expression is reduced resulting in fewer transporters

for mediating glucose uptake; 3.) insulin specifically

modulates the turnover of GLUT4; and 4.) translocation of

GLUT4 is not defective. These studies are complemented by

the roles of alternative hexoses on the development of

insulin resistance of both transport and glycogen turnover.

The appendices include studies on the effects of glucose

deprivation and chronic insulin on GLUT1 and leptin levels.
















CHAPTER 2
MATERIALS AND METHODS


Materials


Dulbecco's modified Eagle's medium (DMEM) (Cat. No.

12100-061) and glutamine-, glucose-free DMEM (Cat. No.

23800-022) were obtained from Life Technologies, Inc. Fetal

bovine serum (FBS) (Cat. No. 1020-75) and calf serum (Cat.

No. 1100-90) were obtained from Intergen. Glucose-free FBS

was prepared by dialyzing FBS against phosphate-buffered

saline (PBS), pH 7.4, for 48 hours at 4C using dialysis

tubing with a molecular weight cutoff of 13,000 Da. Bovine

serum albumin (Cat. No. A-7030) was purchased from Sigma.

Insulin was a generous gift of Dr. Ronald Chance from Eli

Lilly Corp. L-Glutamine, D-glucosamine, D-fructose,

dexamethasone, and methylisobutylxanthine were obtained from

Sigma and D-glucose was obtained from Fisher. ProMix-35 S-

Label (1000 Ci/mmol) (Cat. No. SJQ0079), 2-deoxy-D-[2,6 3H]

glucose (45 Ci/mmol) (Cat. No. TRK672), D-[U-4C]-glucose

(310 mCi/mmol) (Cat. No. CFB96), and D-[U-14C]-fructose (321

mCi/mmol) (Cat. No. CFB47) were obtained from Amersham.

SulfoLink Kit (Cat. No. 20405) was purchased from Pierce.

Protein A-Sepharose was obtained from Sigma. Anti-rabbit









IgG conjugated horseradish peroxidase was purchased from

Sigma. Luciferin/luciferase (Cat. No. L9134) was obtained

from Sigma. Mouse leptin RIA Kit (Cat. No. ML-82K) was

obtained from Linco. All other reagents were of the highest

quality available.


Methods


Cell Culture


Cells were grown and differentiated as previously

described (68). Briefly, 3T3-L1 fibroblasts were seeded on

polystyrene tissue culture dishes and fed every other day

with DMEM containing 10% calf serum (CS) for seven days

until confluence was reached. The fibroblasts were then

induced to differentiate into adipocytes by feeding with

DMEM containing 10% fetal bovine serum (FBS), 1 |jg/mL

insulin, 0.5 mM methylisobutylxanthine, and 0.25 mM

dexamethasone (69). Two days later, the cells were refed

with DMEM containing 1 [tg/mL insulin. Following this

treatment, cells were maintained by feeding with DMEM

containing FBS every other day. Cells were used 8-12 days

following differentiation.









Chronic Insulin Treatment and Insulin Washout


Fully differentiated 3T3-L1 adipocytes were incubated

in DMEM containing 10% FBS and specific concentrations of

insulin ranging from 1 pM to 10 nM for 12 h at 37C in a 7%

C02 incubator. Care was taken to reduce the loss of insulin

in solution, particularly at low concentrations. Thus,

insulin dissolved in 0.01 N HCl was added to DMEM containing

10% FBS to give a final concentration of lOnM insulin. The

remaining insulin concentrations were achieved by serial

dilution into DMEM containing 10% FBS. In addition,

solutions were stored in plastic containers to prevent

insulin binding which occurs in glass bottles. Finally,

cells were refed every 2 h as adipocytes rapidly degrade

insulin.

After chronic treatment, an insulin washout procedure

was performed with the goal of rapidly returning the cells

to a basal state. Specifically, adipocytes were removed

from the C02 incubator and placed in a 37C waterbath.

Plates were washed three times with 3 mL of Krebs' Ringer

Phosphate buffer (KRP), pH 7.4, containing 0.1% BSA and 5 mM

glucose, every 20 min over 60 min. The final wash at 60 min

was performed with KRP alone. The cells were then assayed

according to the particular experimental protocol as

described below. After determining the optimal

concentration of insulin to induce insulin resistance in









these cells, a concentration of 10 nM insulin was chosen for

chronic insulin treatment for the remaining experiments (see

Chapter 1). This concentration was chosen as it can be

removed from the cells by the washout procedure, but the

medium does not have to be replaced every 2 h to maintain

effective insulin concentrations.


Glucose Transport Assay


Glucose transport activity was performed as described

previously (68). Briefly, medium was removed from 3T3-LI

adipocytes grown in 35-mm tissue culture dishes by washing

three times with 3 mL of KRP buffer. Cells were then

incubated in 1 mL KRP with or without 1 JM insulin for 10

min. This was then followed by addition of 200 JiM [3H]2-

deoxyglucose (0.2 jiCi). After 10 min, transport was

terminated by washing the cells three times with 3 mL of

ice-cold phosphate-buffered saline (PBS). Cells were then

air-dried and lysed with 0.1% SDS and duplicate aliquots of

300 jiL were taken for counting by liquid scintillation. The

rate of 2-deoxyglucose transport is reported as nmoles/106

cells/min.


Subcellular Fractionation of 3T3-L1 adipocytes


The subcellular fractionation technique allowed the

subcellular localization of GLUT4. Plasma membrane (PM),









low density membrane (LDM), and high density membrane (HDM)

fractions were isolated by a modification (70) of a

technique described by Weber et al. (71). The major

modification of this technique is the use of a steel block

homogenizer and tungsten ballbearing originally designed by

Balch and Rothman (72). The complete procedure is outlined

in Figure 2-1. Briefly, control or insulin-treated cells

were scraped into TES buffer (10 mM Tris-HCl, pH 7.4, 1 mM

EDTA, and 250 mM sucrose) at 18C. The cells were then

passed over a tungsten ball ten times in the steel block

homogenizer (at 18C) with a clearance of 0.0025 inch. A

crude plasma membrane fraction was collected at 17,000 X g

for 15 min at 4C. Purified membranes were collected from

this fraction by sucrose gradient centrifugation (71). LDM

and HDM fractions were collected by differential

centrifugation in 300 JiL TES. Membrane fractions were

stored at -20C. Protein was determined by the method of

Markwell et al. (73).


Markwell Assay for Protein Determination


The Lowry procedure is widely used for the assay of

soluble proteins (74). However, detergents, used to

solubilize membranes for the release of integral membrane

proteins, and common compounds such as sucrose and EDTA used

in buffers for the isolation of membranes can interfere with









Scrape 5, 10-cm plates
at 18C with 4 mL TES each


Homogenize in steel block at 18C
(subsequent steps at 4C)

SS-34
12K, 15min


Pellet

I SS-34
12K, 20min


Pellet


Supernatant


Pellet


Sucrose Cushion
SW-28
23K, 65min

Interface


SS-34
20K, 30min


HDM


SS-34
20K, 30min



Pellet

SS-34
20K, 30min


PM


SS-34
20K, 30min

Supernatant
Supernatant


I Ti70.1
65K,75min
W
Pellet

Ti70.1
65K,60min


LDM


Figure 2-1. Subfractionation of 3T3-L1 Adipocytes.
Overview of subcellular fractionation procedure.









protein determination by the Lowry method (73). Therefore,

the Markwell procedure, which is a modification of the Lowry

procedure, was used for the analysis of membrane proteins in

our specific extraction buffer, which contains both sucrose

and EDTA (73).

A standard curve (0-100 )'g) was generated by diluting a

solution of BSA (1 mg/mL) in water to a final volume of 0.1

mL. Ten jiL of membranes suspended in TES was added to 0.1

mL of water. Ten [tL of TES was also added to the standard

curve. One mL of a solution containing 2.0% Na2CO3, 0.4%

NaOH, 0.16% Na+ K+ tartrate, and 1% SDS was added to the

standard curve and unknown samples, mixed, and allowed to

stand for 10 min. Folin reagent (0.1 mL of 1 N) was added,

mixed gently, and allowed to stand for 45 min. The

absorbance of the solution was measured at 650 nm and the

protein concentration of the samples was determined by

comparing them to the standards. A second order equation

was used to solve for the constants and the quadratic

equation was used to determine the protein concentration of

the unknown samples.


Gel Electrophoresis


Electrophoresis was performed as described by Laemmli

et al. (75). Equal amounts of membrane protein (50 j.g) were

mixed with half the volume of 2X Laemmli sample dilution









buffer (4% SDS, 6 M urea, 10% P-mercaptoethanol, 0.15 mg/mL

bromophenol blue, 40% glycerol and 20 mM Tris-base, pH 6.6).

Proteins were separated on 10% SDS-polyacrylamide gels by

running the gels for 15 h at 40 V.


Electrotransfer and Western Blotting


Following electrophoresis, the proteins were

transferred to nitrocellulose (pore size 0.45 pm) at 200 mA

for 2 h in transfer buffer (150 mM glycine, 20 mM Tris-base,

20% methanol, pH 8.2) using the method described by Clancy

and Czech (76). Nitrocellulose membranes were then immersed

in blocking buffer (20 mM Tris-base, 137 mM NaCl, 0.1%

Tween-20, pH 7.5, and 5% non-fat dry milk) for 1 h at room

temperature. Membranes were then incubated in blocking

buffer containing a 1:1000 dilution of either GLUT4 antisera

or GLUT1 antisera for 1 h at room temperature. Three washes

for 1 min followed by two washes for 5 min were performed

with buffer in the absence of non-fat dry milk. Membranes

were then incubated with a 1:100,000 dilution of horse-

radish peroxidase conjugated goat anti-rabbit IgG in

blocking buffer for 1 h. Four 1-min and three 5-min washes

were performed with buffer in the absence of non-fat dry

milk to remove excess secondary antibody. The protein-

antibody complex was visualized using enhanced

chemiluminescence.









RNA Isolation and Northern Blotting


RNA isolation and northern blotting was performed by

Martin Williams. Total cellular RNA was isolated by the

guanidinium thiocyanate procedure (77). Twenty jig of total

RNA was loaded onto a 1% formaldehyde-agarose gel and the

gel was run for 12-16 h at 40 V with constant buffer

recirculation. The RNA was then electrophoretically

transferred to an uncharged nylon membrane and cross-linked

with UV light for 3.5 min. GLUT4 probe was then generated

from cDNA generously provided by Dr. Maureen Charron (Albert

Einstein College of Medicine). The insert was labeled by

primer extension (78). Briefly, the DNA (0.5 jig) was

denatured by boiling at 100C for 5 min. After cooling for

3 min, the insert was incubated with a 6-base primer,

deoxynucleotides, one of which was labeled ([32PJ]dATP, dCTP,

dTTP, dGTP), and DNA polymerase. Free nucleotides were

removed by gel chromatography over Sephadex G-50. The

eluted labeled DNA was used directly to probe the RNA blot.

Hybridization was performed for 12-16 h in a buffer

containing 1% bovine serum albumin, 1 mM EDTA, 0.5 M sodium

phosphate, pH 7.2, and 1% SDS at 60C. The membranes were

washed under high stringency conditions (3 X 10-min washes

in 1 mM EDTA, 40 mM sodium phosphate, pH 7.2, and 1% SDS at

65C) and exposed to film for various lengths of time.









Relative intensity of each band was quantified by video

densitometry within the linear range of the film using the

Biolmage Visage 110.


Glycogen Analysis


Total glycogen was isolated as described previously

(79) based on a method described by Pfleiderer (80). Cells

grown on 10-cm tissue culture dishes were washed at 4C with

PBS, pH 7.2, and scraped into a 15-mL polypropylene tube in

1 mL PBS. The cells were then sonicated for 10 s on power 2

at 50% duty cycle (Branson Sonifier 450). Two mL of 30%

(w/v) KOH was added and the sample was mixed on a vortex

mixer. The sample was then boiled in a water bath for 15

min. After cooling to room temperature, 3.5 mL of 95% EtOH

were added. The samples were mixed and heated in a boiling

water bath for 3 min and cooled again to room temperature.

The precipitated glycogen was collected by centrifugation at

1,300 X g for 5 min at room temperature. The supernatant

was removed by aspiration and the glycogen pellet washed by

resuspension in 1 mL 95% EtOH. This suspension was

transferred to a 1 mL microfuge tube and the glycogen

collected by centrifugation at 13,300 X g for 5 min at 4C.

The supernatant was removed by aspiration and the

precipitate was stored at -20C.









Glucose was released from glycogen by acid hydrolysis.

Specifically, 0.2 mL 2 N H2S04 was added to the glycogen

pellet and incubated in a boiling water bath for 2 h. The

samples were cooled to room temperature and 0.15 mL 2 N NaOH

and 0.65 mL H20 were added. The concentration of glucose in

the hydrolysate was then determined using a hexokinase-based

glucose kit (Glucose HK Kit) from Sigma. For samples fed

with glucose, 0.01 mL of hydrolysate was diluted with 0.99

mL of H2S04/NaOH/H20 mixture (2:1.5:6.5). Samples deprived

of glucose were diluted by adding 0.1 mL of hydrolysate to

0.9 mL of the H2SO4/NaOH/H20 mixture. Duplicate aliquots of

0.1 mL were added to 1 mL of prewarmed assay reagent and

incubated at 37C for 30 min. The absorbance of the samples

was then measured at 340 nm.

To limit the variability in experiments dealing with

glycogen, glycogen levels were monitored during the week in

which the 3T3-L1 adipocytes were fully differentiated. As

these cells are fed every 48 h during their normal

maintenance feeding schedule and they begin to enter

glycogen deprivation on the tail end of this time, it was

hypothesized that glycogen levels would fluctuate between

feedings. Upon collecting glycogen from the cells each day

of the week, at the same time of day, I found that glycogen

levels did indeed fluctuate. As Figure 2-2 shows, glycogen

levels were highest the day after they were fed and lowest
















S1.5
0
'I

^ 1.04
0 I
0


-- 0.5

0.0


1 2 3 4 5

A Day t

Fed Fed


Figure 2-2. Fluctuations in Glycogen Levels Over a Week.
Following differentiation, cells were maintained on their
feeding schedule and fed with DMEM containing 10% FBS on the
days indicated. Cells were collected at the same time every
day during the week and glycogen isolated by ethanol
precipitation. Following acid hydrolysis, glucose present
in glycogen was determined by a commercially available kit
(Sigma). Data represent the average S.D. of duplicate
samples within a single experiment (n=2).









on the days of a scheduled feeding as they became glucose

deprived. In addition, glycogen levels increased over the

course of the week from day 1 to day 5. This shows the

importance of performing a glycogen experiment the day

following a feeding and also performing the experiment at

the same time during the week (usually day 2 or 3 following

a feeding) to obtain consistent data.


Glycogen Synthesis


Cells adherent to 35-mm dishes were extensively washed

with KRP containing 0.1% BSA and 5 mM glucose as described

under "Insulin Washout" to return the cells to a basal level

of glucose transport. These cells were then labeled with 2

iCi/plate of 14C-U-[D-glucose] in 1.5 mL KRP containing 5 mM

glucose (giving a final specific activity of 284 pCi/mmol)

with or without 1 [tM insulin for 1 h. Glycogen was then

collected as described above and the pellet dissolved in 300

P.L of water. Radioactivity present was determined by liquid

scintillation.


Production and Characterization of GLUT4 Antibody


Hydrophilicity of a peptide is a criterion for

selecting a suitable sequence for antibody production

because hydrophilic peptides are more likely to be soluble

for coupling reactions and also more likely to be exposed on









the surface of the native protein (81). We chose to

synthesize a peptide corresponding to the last 13 amino

acids of the carboxy terminus GLUT4 due to both its

hydrophilicity and divergence from other members of this

transporter family and specifically GLUT1, which is also

expressed in 3T3-Ll adipocytes. The carboxy-terminal tail of

this transporter also makes it a good sequence for antibody

production because the predicted secondary structure shows

that the tail is probably exposed (42). However, a peptide

of only 13 amino acids is too small to elicit an immune

response if injected into a rabbit alone. Therefore, the

peptide is typically covalently linked to a larger protein

known as an immune carrier such as keyhole limpet hemacyanin

(KLH) or bovine serum albumin (BSA) (81). Although KLH and

BSA are the two most common immune carriers, others used are

ovalbumin, mouse serum albumin, or rabbit serum albumin

(81). As the peptide used to generate GLUT4 was injected

into a rabbit, KLH was used as the carrier.

Antiserum against GLUT4 was generated using a peptide

(CSTELEYLGPDEND) corresponding to amino acids 498-510 of the

GLUT4 sequence (underlined). This peptide was generated by

the Protein Chemistry Core facility at the University of

Florida and verified by amino acid analysis. An N-terminal

cysteine was added to allow its conjugation to KLH. This

conjugation was performed by using a thiol-specific cross-









linking reagent, sulfo-m-maleimidobenzoyl-N-

hydroxysuccinimide ester (sulfo-MSB). KLH (60 mg) was

dissolved in 10 mL of 10 mM potassium phosphate, pH 7.3, and

dialyzed at 4C against 10 mM potassium phosphate for 48 h.

The dialysate was concentrated by coating the dialysis

tubing with a liberal amount of Aquacide and wrapping in

foil. Protein aggregates were removed from the concentrated

dialysate by centrifugation in a microfuge for 10 min. The

concentration of protein in the dialysate was determined by

measuring its absorbance at 280 nm using the milligram

extinction coefficient of KLH (1.6). Six mg of KLH from a

12 mg/mL solution in 10 mM potassium phosphate, pH 7.3, was

added to 0.6 mg sulfo-MSB and incubated at room temperature

for 45 min. The KLH-MSB conjugate was separated from free

sulfo-MSB by size exclusion chromatography on a G-100 column

pre-equilibrated with 50 mM potassium phosphate, pH 6.0.

The free sulfo-MSB is retarded in the column and the KLH-MSB

conjugate flows through. Ten, 1 mL fractions were collected

and measured at 280 nm. KLH-MSB containing fractions were

pooled and reacted with 6 mg of the GLUT4 peptide for 3 h at

room temperature with rotation. The pH of the mixture was

adjusted with HCl to produce the maximum amount of

precipitate. The mixture was then allowed to stand upright

overnight at 4C to allow the precipitate, which represents

the KLH-peptide conjugate, to settle.









The KLH-GLUT4 conjugate (200 4g) dissolved in Freund's

Complete Adjuvant was then injected into the popliteal lymph

node of a New Zealand white rabbit (82). The rabbit was

boosted 28 days later by intradermal injection of an equal

amount of KLH-GLUT4 conjugate and Freund's Incomplete

Adjuvant. This procedure was repeated 2 weeks later. Test

bleeds were collected every week by laceration of the medial

ear vein. Serum was isolated following coagulation of the

red blood cells. Additional boosts were performed every 6

weeks. The antiserum was divided into 1 mL aliquots and

stored at -20C.

To demonstrate the specificity of the GLUT4 antiserum

to GLUT4 protein, dot blots were first performed. GLUT1 or

GLUT4 peptide was spotted onto separate nitrocellulose

sheets and the western blot procedure followed in which both

sheets were incubated with 1:1000 dilution of the GLUT4

antiserum. Figure 2-3 shows that the GLUT4 antiserum is

specific for the GLUT4 peptide as it does not cross react

with the GLUT1 peptide. Next, the GLUT4 antiserum was shown

to be specific for the GLUT4 protein on a western blot, as

shown in Figure 2-4. The reactivity of the GLUT4 serum with

GLUT4 protein was completely prevented by competition with

GLUT4 peptide but not GLUT1 peptide, showing that the

interaction was specific to GLUT4 and not due to non-

specific binding.



















1 2.5 5 jIg GLUT1 peptide


1 2.5 5 Vig GLUT4 peptide


Figure 2-3. Specificity of GLUT4 Antiserum for GLUT4
Peptide. Panel A, 1, 2.5, and 5 [tg of GLUT1 peptide was
blotted onto nitrocellulose and allowed to dry. Membranes
were blocked and probed with a 1:1000 dilution of the GLUT4
antiserum as described under "Electrotransfer and Western
Blotting". The protein-antibody complex was visualized by
enhanced chemiluminescence. Panel B, GLUT4 peptide was
blotted onto nitrocellulose and treated as in panel A.













4
o 00
o 0
9q


GLUT4 -


1 2 3 4


Figure 2-4. Specificity of GLUT4 Antiserum for GLUT4
Protein. Equal protein (50 [tg) from the LDM fraction of
control cells was loaded per lane and subjected to SDS-PAGE.
Proteins were transferred to nitrocellulose and western blot
analysis performed under the following conditions: lane 1,
pre-immune serum; lane 2, 1:1000 dilution of GLUT4
antiserum; lane 3, GLUT4 antiserum competed with 10 pig of
GLUT1 peptide; and lane 4, GLUT4 antiserum competed with 10
jg of GLUT4 peptide.









Peptide Purification of GLUT4 Antibody


In order to study the turnover and synthesis of GLUT4,

metabolic labeling with 35S-cysteine and -methionine was

performed followed by immunoprecipitation. In order to

prevent non-specific interactions during

immunoprecipitation, GLUT4-specific IgG was purified from

the anti-GLUT4 rabbit serum using a modification of the

technique described by Dankert et al. (83). Total IgG was

purified from 15 mL of serum by adjusting the pH to 5 with 3

M acetic acid and adding 0.75 mL of capryllic acid while

vigorously stirring. After stirring for 30 min, the

precipitate was collected by centrifugation at 41,000 X g

for 30 min at 4C. The supernatant was removed and its

volume measured. An equal volume of saturated ammonium

sulfate was added to the supernatant and stirred overnight

at 4C. The next day, the solution was divided equally

between two 15-mL corex tubes and the precipitate was

collected by centrifugation at 41,000 X g for 30 min at 4C.

The supernatant was discarded and each pellet was

resuspended in 0.5 mL of PBS containing 0.1% sodium azide.

The ammonium sulfate was removed from the samples by

dialyzing against three changes of PBS containing 0.1%

sodium azide over 48 h using tubing with a molecular weight

cut off of 13,000 Da. The concentration of the total IgG

solution was determined by measuring the absorbance of the









solution at 280 nm using a milligram extinction coefficient

of 1.4.

GLUT4 antibody was purified from the total IgG solution

by peptide chromatography. For this technique, we used a

SulfoLink Kit from Pierce. This kit contains a column

composed of 6% agarose beads cross-linked to iodoacetate.

The acetyl group binds irreversibly to free sulfhydryl

groups. Two mg of the GLUT4 peptide were bound to this

matrix through the terminal cysteine residue following the

manufacturer's instructions. GLUT4-specific antibody was

then purified by placing 2-3 mL of the total IgG solution on

the column and rotating at room temperature for 2 h. The

column was then washed with 16 mL of PBS and the GLUT4

antibody was eluted with 8 mL of 0.1 M glycine, pH 3.0. One

mL fractions were collected in tubes containing 100 )iL 1 M

Tris-base. The absorbance of the fractions was measured at

280 nm and fractions containing antibody were pooled.

Figure 2-5 shows a typical elution profile of peptide-

purified anti-GLUT4 antibody. The concentration of the

resulting antibody solution was determined by its absorbance

at 280 nm using a milligram extinction coefficient of 1.4.

The purified GLUT4 antibody was then stored as 0.1 mL

aliquots at -20C.















0.5

0.4

0 0.3

0.2

0.1

0.0 IMP7
01 "57
o~o ^ VA ^ y/ ^Ty

0 1 22 3 4 5 6 7 8

Fraction




Figure 2-5. Elution Profile of Peptide-Purified anti-GLUT4
Antibody. Total IgG was added to the column and rotated for
2 h. Unbound antibody was washed from the column using PBS.
GLUT4 antibody was eluted using 0.1 M glycine, pH 3.0. One
mL fractions were collected and neutralized with 1 M Tris-
base. Absorbance of the fractions was measured at 280 nm
and the concentration determined using a milligram
extinction coefficient of 1.4.









Preparation of Total Membranes


3T3-L1 adipocytes grown in 10-cm tissue culture dishes

were washed three times with 4 mL of KRP at 37C. Cells

were then scraped into 4 mL of TES buffer (10 mM Tris-HCl,

pH 7.4, 1 mM EDTA, 250 mM sucrose, and 20 [.g/mL PMSF) on

ice. The cells were homogenized using 20 strokes of a

teflon pestle in a 10 mL Potter-Elvejhem flask. The

homogenate was then centrifuged at 1,300 X g for 5 min to

remove nuclei and unbroken cells. The supernatant was

placed into polycarbonate tubes and membranes were collected

by centrifugation at 212,000 X g for 1 h in a Ti70.1 rotor

at 4C. The supernatant was discarded. For

immunoprecipitation, the pellet was resuspended in 1 mL of

extraction buffer (PBS containing 2% C12E9, 0.1% SDS,

1 mM EDTA, and 20 tg/mL PMSF) using 10 strokes of a teflon

pestle in a 2 mL Potter-Elvejhem homogenizing flask.

Insoluble material was then removed by centrifugation at

13,300 X g in a microfuge for 5 min. The supernatant was

transferred to a separate microfuge tube and protein

concentration was determined as described above. For

western blotting, the pellet was resuspended in 300 |JL of

TES and protein concentration was determined as described

above.









Metabolic Labeling of 3T3-L1 Adipocytes


To measure the synthesis and turnover of proteins,

radiolabeling of proteins with [35S]cysteine/methionine was

performed to "mark" a population of proteins within a

specific period of time. The accumulation of radiolabel

into proteins was determined to measure synthesis. To

measure turnover, cells were pulsed with radiolabel followed

by an extended chase with nonradiolabeled medium. The

disappearance of the labeled protein was then monitored. To

examine synthesis of GLUT4, cells were incubated in complete

DMEM with 10% FBS in the presence or absence of 10 nM

insulin for 11 h. The medium was then changed to cysteine-

/methionine-free DMEM, also in the presence or absence of 10

nM insulin, for 1 h. This medium was then aspirated and 2

mL of the cysteine-/methionine-free DMEM containing 400 iCi

of [35S]cysteine/methionine 10 nM insulin was added to

each plate. The plates were incubated at 37C for 10, 20,

30, 45, or 60 min at which time total membranes were

collected and extracted as described above.

Immunoprecipitation was performed on this solution.

Degradation of GLUT4 was examined by first incubating

cells in cysteine-/methionine-free DMEM for 1 h. Following

aspiration of this medium, 2 mL of the same medium

containing 400 piCi of [35S]cysteine/methionine were added to









each plate and incubated at 37C for 3 h. This medium was

aspirated and replaced with complete DMEM containing 10% FBS

with or without 10 nM insulin for the chase. This medium

was refreshed every 12 h and total membranes were collected

at 0, 6, 12, 24, 48, and 72 h. Total membranes were

collected and immunoprecipitation performed.


Immunoprecipitation of GLUT4


Immunoprecipitation provides a means of studying the

turnover of a particular protein. This technique depends on

the existence of an antibody specific for the protein of

interest. Addition of the antibody to a mixture of proteins

results in specific antibody-protein complex formation.

This complex can then be collected by the addition of

Sepharose beads conjugated with protein A, a bacterial

protein that binds to the heavy chain of IgG. The beads are

precipitated by a short centrifugation and the bound

proteins are analyzed. Although in these studies this

technique was employed to study the dynamics of GLUT4,

protein-protein interactions can be determined by analyzing

proteins which co-immunoprecipitate with the protein of

interest.

To examine the synthesis of GLUT4, extracted membrane

protein of specific concentration (1.5 mg) was brought to 1

mL with additional extraction buffer. To investigate the









degradation of GLUT4, equal volumes (750 pL) of each sample,

which represents equal cell number, were brought to 1 mL

instead of equal protein. This was done as the total cell

protein content increases over extended periods of chronic

insulin treatment. Non-specific interactions with protein

A-Sepharose were reduced by preincubation of extracted

material with 50 pL of a 50% slurry of protein A-Sepharose

at 4C for 1 h. The sepharose beads were removed by brief

centrifugation. Five Vg of peptide-purified GLUT4 antibody

was added to the supernatants. The samples were rotated

overnight at 4C. The next morning, 25 pL of a 50% slurry

of protein A-Sepharose were added to each sample and

rotation was continued for an additional 2 h at 4C. The

sepharose bead complexes were collected by brief

centrifugation. The supernatant was discarded. The beads

were then washed three times with 1 mL of extraction buffer

for 10 min followed by four, 10-min washes with 1 mL of

extraction buffer containing 1 M NaCl. Samples were then

washed with 1 mL TES buffer. The final pellet was

resuspended in 25 pL of TES buffer. Sample dilution buffer

(30 pL) was added to the beads and incubated at 37C for 30

min. Proteins in the entire sample were resolved on a 10%

polyacrylamide gel by electrophoresis as described above.

The gel was then fixed in 100 mL of 10% (w/v)

trichloroacetic acid and 60% (v/v) methanol for 30 min. The









solvent was removed by soaking the gel in water for 30 min.

The gel was then soaked in 100 mL of 1 M sodium salicylate

for 1 h. The gel was dried under vacuum onto 3MM Whatman

paper and juxtaposed to x-ray film at -80C. The P

particles emitted by the 35S bombard the ring structure of

the sodium salicylate causing it to fluoresce and this

emission exposes the x-ray film.


ATP-Luciferase Assay


ATP levels were measured in the 3T3-L1 adipocytes by

using an ATP-luciferin/luciferase assay with the assistance

of Dr. James Gardner. The cleavage of luciferin by

luciferase requires ATP and is the basis of the assay. When

luciferin is cleaved, light is emitted and the intensity of

this light correlates with the amount of ATP in the sample.

Cells (grown in 35-mm dishes) were first rinsed three

times with 3 mL of KRP at 37C. One mL of 8% perchloric

acid at 4C was added to each plate and the cells scraped

directly into 1.5-mL eppendorf tubes. The samples were

neutralized by the addition of 3 M KOH containing 0.5 M

triethanolamine base dropwise until a pH of 6.0 was

obtained. The samples were kept on ice during this

procedure and the volume of the samples before and after the

addition of KOH was measured to calculate the volume of KOH

added. The precipitate formed was removed by centrifugation









in a microfuge for 5 min at 4C. The supernatant was

diluted 1:250.

Ten iL of the 1:250 diluted sample or ATP standard was

added to a plastic 12 x 75 mm tube. To this, 500 pL of

assay buffer (60 mM Tris-Acetate, pH 7.75, 10 mM MgCI2, 1 mM

KC1l, 1.5 mM EDTA, 2.5 mM 3-mercaptoethanol, 0.4 mg/mL

luciferin/luciferase stock) were added. The V-

mercaptoethanol was added after the initial solution was

brought to pH 7.75. The luciferin/luciferase stock was

prepared by adding 50 mg of luciferin/luciferase to 2.5 mL

of 50% glycerol. This was divided into 300 pL aliquots and

stored in liquid nitrogen and added to the assay buffer just

prior to use. Following addition of the assay buffer to the

sample, the tube was flicked for 10 sec then placed in the

luminometer and the fluorescence measured for 20 sec. ATP

in the samples was calculated by comparison against the

standard curve seen in Figure 2-6.


Leptin Assay


Cells (5, 10-cm plates) were fed with serum-free DMEM

for 12 h. The medium was then collected, pooled from the

five plates (a total of 40 mL), and concentrated using the

"Centriplus 3" centrifugal concentrating tubes from Amicon,

which have a molecular weight cut-off of 3,000 Da. The













16000 -

S12000 -

8000 -

S 4000 -

0 I
0 1 2 3

pmol ATP



Figure 2-6. ATP Standard Curve. Ten [tl of specific
concentrations of ATP were added to 12 x 75 mm plastic
tubes. The luciferin/luciferase assay buffer was added, the
tube was flicked for 10 sec., and the fluorescence measured
in a luminometer for 20 sec.









medium was separated equally into four tubes and

concentrated by centrifugation at 3,000 x g (4,900 rpm) for

4.8 h at 4C in an SS-34 rotor. The retained solution was

collected from the four concentrators and pooled. The

pooled solution was then placed into a single new tube and

concentrated further by centrifugation at equal force and

time. The retained solution was collected again and brought

to 4 mL using PBS.

The leptin radioimmunoassay (RIA) was performed as per

the manufacturer's instructions. Briefly, 100 [iL of assay

buffer (0.05 M PBS pH 7.4 containing 0.025 M EDTA, 0.1%

sodium azide, 0.05% Triton X-100 and 1% RIA grade BSA) were

added to 12 x 75 mm borosilicate glass tubes. Leptin

standards (100 4L) and unknown samples (100 ptL) were added

to the appropriate tubes. Next, mouse leptin antibody (100

|LL) was added to the tubes. The tubes were mixed on a

vortex incubated overnight at 4C. The next morning, 125s-

mouse leptin (100 pL) was added to all the tubes, mixed, and

incubated overnight at 4C. The following morning, 1 mL of

cold precipitating reagent was added to all the tubes and

mixed. The tubes were incubated for 20 min at 4C and the

precipitate collected by centrifugation for 15 min at 3,000

x g (5,000 rpm) in an SS-34 rotor at 4C. The supernatant

was decanted immediately and the tubes drained for 1 min.

Excess fluid was blotted from the lip of the tubes and the






47


pellet was counted in a gamma counter. Leptin levels

present in the samples were calculated by comparing against

the standard curve seen in Figure 2-7.














100







1



1 0I , , ,I I II
1 10


Leptin (ng/mL)

Figure 2-7. Leptin Standard Curve. Mouse leptin antibody
was added to specific concentrations of leptin provided in
the kit and incubated overnight at 40C. 125I-mouse leptin
was added and incubated at 4C overnight. One mL of cold
precipitating reagent was added and incubated for 20 min at
40C. The precipitate was collected by centrifugation, the
supernatant was decanted and the radioactivity present in
the pellet determined by a gamma counter.
















CHAPTER 3
DEVELOPMENT OF INSULIN RESISTANCE


Introduction


Insulin resistance has been studied in animal models,

freshly isolated cells, and immortalized cells. Using a

cell line has several advantages. One of these advantages

is the ease of harvesting cultured cells. Another is the

consistency and reproducibility of data, the lack of which

plagues in vivo studies.

Although 3T3-Ll adipocytes have been the focus of many

insulin action studies, in each case, pharmacological doses

of insulin were used. In contrast to other classes of

hormones, insulin does not exhibit biphasic effects, thus

the rationale for using pharmacological doses. However,

caution must be taken as insulin at high concentrations in

cell culture activates the mitogenic paths held in common

with other growth factors, the relevance of which is

questioned (60). That said, with long term exposure to

pharmacological doses of insulin, GLUT4 expression (both

mRNA and protein) is reduced (14). This reduction is caused

by both down-regulation of transcription and enhanced

turnover of mRNA. However, the concentration of insulin









required to affect a 50% change in expression of message was

reported as 23 nM. This level of insulin is at least 2

orders of magnitude higher than the concentration of

circulating insulin in humans (2). No comparable dose-

response studies have examined the development of insulin-

resistant glucose transport activity or GLUT4 expression in

these cells. Therefore, I have developed a procedure that

has allowed the measurement of glucose transport after

chronic exposure (12 h) to physiological concentrations of

insulin.


Results


Insulin-resistant Glucose Transport Activity


Chronic insulin exposure elevates glucose transport

activity in 3T3-LI adipocytes. This can be prevented by

protein synthesis inhibitors like cycloheximide (Risch and

Frost, unpublished data). More recent studies have shown

that pharmacological doses of insulin induces GLUT1

transcription (14) leading to elevated GLUT1 protein and

thus transport activity. In the face of elevated GLUT1,

insulin-sensitive glucose transport activity can not be

observed. While this may imply insulin resistance, elevated

basal transport activity complicates interpretation. My

first challenge was to develop a protocol to reestablish

basal transport activity after chronic exposure to insulin.









I relied initially on studies by Garvey et al. (15). These

investigators showed that incubation of cultured rat

adipocytes with insulin for 12 h was sufficient to complete

the desensitization process, resulting in a 50% reduction in

insulin-sensitive glucose transport activity. Therefore,

cells were first treated with either 10 nM or 1 [iM insulin

for 12 h. The cells were then washed in KRP buffer

containing 5 mM glucose and 0.1% defatted BSA at 20-min

intervals over a 140-min time course. At specific times

during this washout period, cells were rinsed in KRP alone

and glucose transport activity determined in a 2-min pulse.

Figure 3-1 A shows the comparison between control cells

(washed in an identical manner) and those treated

chronically with either 1 tM or 10 nM insulin. Cells

treated chronically with 1 ItM insulin showed significantly

elevated transport at the start of washout (time 0) but

never achieved basal levels despite the extensive time of

washing. However, those cells treated with 10 nM insulin

returned to basal levels within 60 min of washing. When

these latter cells were subsequently restimulated with an

acute insulin challenge following the 60-min wash, the rate

of glucose transport was reduced by 50% compared to control

cells (Figures 3-1 B and 3-2). In addition to the decreased

rate, the cells were less sensitive to insulin in that the


















Figure 3-1. Effect of Chronic Insulin on Glucose Transport
in 3T3-L1 Adipocytes. Panel A, fully differentiated 3T3-L1
adipocytes were incubated in DMEM containing 10% FBS in the
absence of added insulin (0) or with 1 |LM (A) or 10 nM (U)
insulin for 12 h. Plates were washed as described in
Chapter 2 under "Insulin Washout Procedure". Glucose
transport activity was measured in KRP (in the absence of
bovine serum albumin and glucose) by the addition of 200 jiM
[3H]2-deoxyglucose (0.2 jiCi). After 2 min, transport was
terminated by the addition of ice-cold phosphate-buffered
saline. Cells were lysed with a 0.1% solution of SDS and
duplicate aliquots of 300 jiL were taken for quantitating
radioactivity. Panel B, cells were incubated as above in
the absence (control) or presence (chronic) of 10 nM
insulin. Removal of insulin was accomplished within 60 min
(see panel A), then 1 (JM insulin was added back, or not, for
10 min, and glucose transport activity was determined.
Panel C, cells were incubated for 12 h in the absence
(control) or presence (chronic) of 10 nM insulin. The
cells were washed for 60 min to remove insulin. In order
to maintain the low levels of insulin in solution, glucose
transport activity was measured in KRP containing 0.1% BSA.
Therefore, following the final wash at 60 min, 950 [iL of KRP
containing 0.1% BSA were added to each plate. Various
concentrations of insulin, dissolved in 0.1%, were then
added for 10 min for acute stimulation. Glucose transport
activity was measured by the addition of [3H]2-deoxyglucose.
Each panel represents the average S.E. of two independent
experiments (n=4).














0.5


rD.S
tO


Q -
- 0


^^o
g 2


1.5

1.0


0.5

0.0
Acute Insulin


0 50 100 150
Time of Washout (min)


- +


-


Control Chronic


100 -c/ 1
80 control
60 -
40 -chronic



0 0.01 1 100
Insulin Concentration (nM)


B


f// /

v7n //. / /_













-^ 2.0

ot
+-

^ ~1.5- r


--X0 1.0


o0.5


00 I I.
0 5 10 15

Time (min)





Figure 3-2. Time Course of Acute Insulin Stimulation.
Cells were incubated in DMEM containing 10% FBS 10 nM
insulin for 12 h. Plates were then washed as described in
Chapter 2 under "Insulin Washout Procedure". Cells were
then stimulated for specific times with 1 [tM insulin and
glucose transport activity determined. Data represent the
average S.D. of duplicate samples within a single
experiment (n=2). control; N chronic insulin.









dose-response curve to acute insulin challenge was shifted

to the right by an order of magnitude (Figure 3-1 C). This

is the first demonstration of "true" insulin-resistant

glucose transport activity in 3T3-L1 adipocytes because of

the ability to reestablish the basal state.

To determine if physiological insulin could establish

the insulin-resistant state, cells were exposed to specific

concentrations of insulin for 12 h. Cells were refed every

2 h to maintain the extracellular insulin levels,

particularly important at the lower concentrations of

insulin as these cells possess the ability to rapidly

degrade insulin. As shown in Figure 3-3 A, the

concentration of insulin that elicits a 50% reduction in

insulin-sensitive glucose transport was approximately 100

pM. This is extremely interesting because the fasting level

of insulin in non-diabetic humans is about 40 pM while that

in obese individuals is about 70 pM and individuals with

non-insulin-dependent diabetes have insulin levels around

200 pM (84). Insulin as low as 1 nM was sufficient to

completely desensitize the transport system. At 10 nM,

maximal desensitization was achieved whether the cells were

refed every 2 h or not. Thus, we chose a concentration of

10 nM to examine the time required for the development of

insulin resistance. As shown in Figure 3-3 B, the

phenomenon of desensitization was completely established

























Figure 3-3. Development of Insulin Resistance. Panel A,
cells were incubated for 12 h with specific concentrations
of insulin as indicated. During this incubation, the
medium was replaced every 2 h. The cells were then washed
for 60 min and glucose transport activity determined
following acute (10 min) stimulation with 1 iM insulin. The
"fractional difference" was determined by subtracting the
glucose uptake rate at 10 nM insulin from the glucose
uptake rate at each point divided by the difference in
uptake rates between 0 and 10 nM insulin. Panel B, cells
were incubated with 10 nM insulin for specific times.
Medium was replaced every 2 h. At appropriate times, the
cells were washed and acutely stimulated with insulin, and
glucose transport activity was measured. The fractional
difference in activity was determined as in panel A. Data
represent the average S.E. of three independent
experiments (n=6).







A
1.0
o = 0.8
0.6-
0.4 -
S0.2 -

u^ 0.0-

0- 0J

Insulin Concentration (nM)

B

z' 1.0
I 0.8-
3 0.6-
0.4-
0.2-
n nn


0 2 4 6 8 10 12


Time (h)









within 8 h of the initial exposure to insulin. This result

is similar to that described in isolated adipocytes (21)


Role of Glucose and Glucosamine


Marshall and his colleagues (15,20,21) have shown in a

series of elegant experiments the requirement of glucose and

glutamine, as well as insulin, for the expression of insulin

resistance in isolated rat adipocytes implicating the N-

acetylglucosamine biosynthetic pathway in this phenomenon.

To test if the same is true in 3T3-L1 adipocytes, I

performed similar experiments. One complication in my

experiments that was not encountered in isolated adipocytes

is the time-dependent activation of glucose transport

activity in the absence of glucose (85-88). This difference

between rat adipocytes and 3T3-Ll adipocytes may result from

higher glycogen stores in the former (89,90) compared with

3T3-Ll adipocytes (79), which provides a metabolic buffer

from external glucose deprivation. We therefore minimized

the time that the cells were exposed to glucose-free medium

but suffered in that only 75% of maximal desensitization was

achieved in these experiments. Importantly, though, the

basal rates of transport were not affected such that true

resistance could be evaluated. Figure 3-4 shows that in the

absence of glucose, insulin was unable to induce

desensitization. Either glucose, in the presence of

























Figure 3-4. Effects of Glucose and Glucosamine on Insulin
Resistance. Panel A, cells were incubated for 6 h in DMEM
containing specific concentrations of glucose in the
absence (control) or presence (chronic) of 10 nM insulin.
Cells were washed and glucose transport activity was
determined in the presence of 1 gM insulin. Panel B, cells
were incubated for 6 h in glucose-free and glutamine-free
DMEM containing specific concentrations of glucosamine in
the absence (control) or presence (chronic) of 10 nM
insulin. Washes were performed on the cells as described
earlier, and glucose transport activity following acute
stimulation with 1 iM insulin was determined. Data
represent the average S.E. of two independent experiments
(n=4). Basal glucose transport activity in control and
glucose-deprived cells was 0.167 0.02 and 0.167 0.01
nmol/106 cells/min, respectively.





A
S1.4

|1.0
I ~0.8
0.6
Q 0.4 -- Control
N, 0.2 -- Chronically treated
m^ 0.0 -------
S0 5 10 15 20

Glucose Concentration (mM)
B
. 1.4


|1.0
S 0.8
o 0.6
0.4- Control
<^ 0.2 -U- Chronically treated
m 0.0 -------
S0.0 0.5 1.0 1.5 2.0

Glucosamine Concentration (mM)









glutamine (Figure 3-4 A), or glucosamine, in the absence of

both glucose and glutamine (Figure 3-4 B), provided

appropriate substrate for the development of insulin

resistance.


Effects of Various Inhibitors on the Development of Insulin
Resistance


Inhibitors of DNA transcription, protein synthesis,

glycosylation and proteases were used to characterize the

development of insulin resistance in these cells. Cells

incubated for 12 h with 4 j.M actinomycin D, an inhibitor of

DNA transcription, exhibited an increase in basal glucose

transport activity and a reduction in insulin-stimulated

glucose transport in both control and chronically-treated

cells (Figure 3-5). The underlying mechanism for this

elevation is unknown. Given elevated basal activity, the

effect of actinomycin D on insulin-sensitive transport is

difficult to determine, although changes in the presence and

absence of acute insulin were fairly small. In contrast,

studies in isolated rat adipocytes (91) showed that

incubation with up to 200 nM actinomycin D for 18 h actually

decreased basal levels. However, in these studies,

actinomycin D prevented the development of insulin

resistance. It is believed that actinomycin D prevented

insulin resistance in these studies by inhibiting GFAT

activity, as insulin resistance induced by incubation with











1.2 T
S1.0


Q- 0.8
006

0.6


0.2-
o.o ^^-^ ^ ^--2^^

acute insulin + + + + + +
actinomycin + + +
cycloheximide + + + +
I I
control chronic






Figure 3-5. Effects of Actinomycin and Cycloheximide on
Insulin Resistance. 3T3-L1 adipocytes were incubated in the
presence or absence of 4 gM actinomycin D or 20 IJM
cycloheximide also in the presence (chronic) or absence
(control) of 10 nM insulin for 12 h. Insulin was removed
from the cells by washing as described previously. Basal
and insulin-stimulated rates of glucose transport were
measured by incubating the cells in KRP with 1 jiM insulin,
or not, for 10 min. The data represent the average S.E.
of two independent experiments (n=4).









glucosamine, which bypasses GFAT, could not be prevented by

actinomycin D (91). As Figure 3-5 shows, in 3T3-L1

adipocytes, 20 (JM cycloheximide had no effect on basal

transport, but inhibited insulin-stimulated glucose

transport in both control and chronically-treated cells.

Therefore, cycloheximide did not appear to prevent insulin

resistance. Studies in isolated rat adipocytes also showed

that cycloheximide was unable to prevent insulin resistance

(92).

To further test the involvement of the hexosamine

biosynthetic pathway in the development of insulin

resistance, the effects of nikkomycin Z on insulin-resistant

glucose transport activity were tested. Nikkomycin Z is a

peptide-nucleoside antibiotic that competitively inhibits

chitin synthase in fungi and insects due to its structural

similarity to UDP-N-acetylglucosamine (93). I hypothesized

that this compound might prevent the onset of insulin

resistance by competing for enzymes which use UDP-N-

acetylglucosamine in mammalian cells. As Figure 3-6 shows,

neither concentration of nikkomycin Z tested had an effect

on glucose transport. Although I have no evidence that

nikkomycin is taken up by these cells, the concentrations

used are in far excess of 0.1 tM, the Ki of nikkomycin for


chitin synthase (94).













S0.8



0.6


d) 0.4 ,,

hx J- -/ ^ / T







acute insulin + +
20M Nikkomycin +
200 iM Nikkomycin + + +
I I
control chronic






Figure 3-6. Effects of Nikkomycin Z on Insulin Resistance.
Cells were incubated in the presence or absence of specific
concentrations of nikkomycin Z also in the presence
(chronic) or absence (control) of 10 nM insulin for 12 h.
Plates were washed to remove insulin and basal and insulin-
stimulated rates of glucose transport were measured as
previously described. Data represent the average + S.D. of
duplicate samples within a single experiment (n=2).









Because the hexosamine biosynthetic pathway provides

substrate for glycoprotein synthesis, I tested the effects

of tunicamycin on the development of insulin resistance. As

shown in Figure 3-7, 2.5 J.g/mL tunicamycin, which inhibits

N-linked glycosylation, decreased the insulin-stimulated

rate of glucose transport in both control and chronically-

treated cells. This is similar to the effect of

cycloheximide and again makes it difficult to determine if

tunicamycin was able to prevent the decrease in glucose

transport in insulin-resistant cells.

It has been hypothesized by Knutson et al. (95) that a

proteolytic fragment of the P-subunit of the insulin

receptor is involved in the development of insulin

resistance. To test this, I used a thiol protease inhibitor

(E-64) to prevent cleavage of the 3-subunit (95). As is

shown in Figure 3-8, 100 |JM E-64 had no effect on glucose

transport activity in control or chronically-treated cells.

As the Knutson group showed the largest effect of E-64 on

glycogen synthesis, I also investigated this insulin-

sensitive process. As Figure 3-9 shows, 100 tM E-64 had no

effect on rates of glycogen synthesis in control or insulin-

resistant cells.













1.0 -

< E 0.8 -
0 r.f /

S 0.6 "0.




//
gr 0.4 E I I
//" I I11 I -




tunicamycin a i t p

Control Chronic







Figure 3-7. Effects of Tunicamycin on Insulin Resistance.
Cells were incubated in the presence or absence of 2.5 Ig/mL
tunicamycin also in the presence (chronic) or absence
(control) of 10 nM insulin for 12 h. Cells were washed to
remove insulin and basal and insulin-stimulated rates of
glucose transport were measured as previously described.
Data represent the average + S.D. of duplicate samples
within a single experiment (n=2).











1.2


1.0
0.8


c(

I
o _
o '-


acute insulin


E-64


- +


+ +


control


c +
chronic


Figure 3-8. Effects of E-64 on Insulin Resistance. Cells
were incubated in the presence or absence of 100 iM E-64
also in the presence (chronic) or absence (control) of 10 nM
insulin for 12 h. Cells were then washed to remove insulin
and basal and insulin-stimulated rates of glucose transport
were determined as previously described. Data represent the
average S.E. of two independent experiments (n=4).


0.6
0.4
0.2
0.0


Jigj

















-
c,)0

c,)0^


40


30


20


10


0


acute insulin
chronic insulin
E-64


+ + + +


+ +


+ +
+ + + +


Figure 3-9. Effects of E-64 on Glycogen Synthesis. Cells
were incubated in the presence or absence of 100 [M E-64
10 nM insulin for 12 h. Plates were then washed to remove
insulin and labeled with 14C-U-[D-glucose] (2 tCi/plate) in
KRP containing 5 mM glucose in the presence or absence of 1
gM insulin for 1 h. Glycogen was then collected as
described and radioactivity counted by liquid scintillation.
Data represent a single experiment.









Effect of Chronic Insulin on GLUT4 Expression and
Translocation


A total membrane fraction revealed that insulin-

resistant cells (i.e. cells exposed to 10 nM insulin for 12

h) expressed 2.4-fold less GLUT4 than control cells while

GLUT1 increased by 2.2-fold (Figure 3-10). Another

important point to make as shown by Figure 3-10 is that

acute insulin stimulation did not alter the total levels of

GLUT1 or GLUT4. Only chronic insulin treatment affected the

total amounts of these proteins. To examine the subcellular

distribution of these changes, I used a subcellular

fractionation technique recently developed in this lab (70)

to isolate three membrane fractions: plasma membrane (PM),

low density membranes (LDM), and high density membranes

(HDM). The LDM consists primarily of the small endosomal

storage vesicles containing GLUT4 and the HDM consists

primarily of endoplasmic reticulum and Golgi. Figure 3-11

shows the distribution of GLUT4 and GLUT1 among these three

membrane fractions in control and insulin-resistant cells.

Each set went through the washout procedure prior to

membrane fractionation. Control cells, which were

stimulated acutely with 1 jiM insulin, showed redistribution

of both GLUT4 and GLUT1; GLUT4 increased by about 6-fold in

the PM (Figure 3-11 A and B) while GLUT1 increased by about

2-fold (Figure 3-11 C and D). These data are similar to


















Acute Insulin
Chronic Insulin


GLUT1 -



GLUT4 -


+ +


Figure 3-10. Effects of Acute and Chronic Insulin on Total
GLUT Protein Levels. Equal protein of the PM, LDM, and HDM
fractions of Figure 3-11 were combined and subjected to SDS-
PAGE and transferred to nitrocellulose. Immunoblot
detection of GLUT1 and GLUT4 was carried out using C-
terminal specific antibodies. The protein-antibody complex
was visualized by enhanced chemiluminescence.

























Figure 3-11. Subfractionation of Insulin-Resistant 3T3-L1
Adipocytes. Cells were treated for 12 h in the absence
(control) or presence (chronic) of 10 nM insulin and
subsequently washed. Following acute stimulation with 1 (JM
insulin, PM, LDM, and HDM were collected as described in
Chapter 2. SDS-polyacrylamide gels of equal protein (70 [tg)
transferred to nitrocellulose allowed immunoblot detection
of GLUT1 and GLUT4 using C-terminal specific antibodies.
The protein-antibody complex was visualized by enhanced
chemiluminescence. Bands were quantitated by video
densitometry. Panel A, immunoblot of membrane fractions
probed with anti-GLUT4 antibody; panel B, densitometry of
GLUT4 immunoblot; panel C, immunoblot of membrane fractions
probed with anti-GLUTi antibody; panel D, densitometry of
GLUT1 immunoblot. I] control; M control + acute insulin; S
chronic insulin treatment; M chronic insulin treatment +
acute insulin. Data represent a single experiment. A
duplicate experiment gave similar results.













A PM LDM HDM
Chronic Insulin + + + + + +
Acute Insulin + + + + + +
GLUT4 i--.n



B GLUT4 Levels
50

S40 -

30 p





PM LDM HDM
S20 | X rN
H /, ^V./N
10 o- XNl k
rOl N\ I l
PM LDM HDM


















PM LDM HDM


Chronic Insulin
Acute Insulin

GLUT1 -*


+ +
+ +


GLUT1 Levels


100


80

r 60

* S 40

20

0


1 ^^ ^,
,'R-^


PM LDMHD


+ +


+ +


+ +


+ +









those that analyzed translocation using the cell surface

photolabel, ATB[2-3H]BMPA [2-N-4-(l-azi-2,2,2-

trifluoroethyl)benzoyl-1,3-bis(D-mannose-4-yloxy)-2-

propylamine] (16). Cells chronically exposed to insulin

(followed by washout) showed levels of GLUT4 in the plasma

membrane equivalent to that of controls (Figure 3-11 A and

B). While acute insulin challenge stimulated translocation,

the amount of GLUT4 was reduced by about 50% compared with

controls. This reflects the down-regulation of GLUT4 and

correlates well with the loss of insulin-stimulated glucose

transport activity. Cells chronically exposed to 10 nM

insulin showed a 2-fold increase in the level of GLUT1 in

the PM after washout compared with controls (Figure 3-11 C

and D), despite the equivalent rates of glucose transport.

Acute insulin challenge stimulated translocation but to a

much more limited degree than in control cells.

As the LDM fraction reflects the loss of GLUT4, we used

this fraction to examine the dose-dependent loss in cells

chronically treated with specific concentrations of insulin.

As shown in Figure 3-12 A, the level of GLUT4 decreased over

time in response to increasing insulin. Figure 3-13 shows a

similar experiment with a larger sample (5, 10-cm plates per

condition compared to 1, 10-cm plate). This larger sample

allows for the somewhat more accurate quantitation of the

LDM fraction. The numbers listed below the blot in Figure















Insulin
[nM]


GLUT4 -4


0 W


0


00
I-


0 0.1 1 10 100 1000
Insulin [nM]


Figure 3-12. Effects of Chronic Insulin on GLUT4
Expression. Panel A, cells were incubated with specific
concentrations of insulin for 12 h (fed with fresh medium
every 2 h). Cells were then washed for 60 min and
subfractionated to isolate the LDM fraction. Proteins were
separated by SDS-polyacrylamide gel electrophoresis and
subsequently transferred to nitrocellulose membrane (70 p[g
of protein were loaded per lane). The membrane was then
probed for GLUT4 and visualized by enhanced
chemiluminescence. Panel B, GLUT4 bands were quantitated by
densitometry. Fractional difference in GLUT4 expression was
calculated from five independent experiments performed as in
panel A. Data represent the average S.E.



















Insulin [nM] 0 (j K- N


GLUT4 --


100 93


67 37


Figure 3-13. Effects of Chronic Insulin on GLUT4 Expression
with Larger Samples. Cells were incubated as described in
Figure 3-12 with the exception that 5, 10-cm plates were
used per condition compared to 1, 10-cm plate in Figure 3-
12. The LDM fraction was collected and treated as described
in Figure 3-12. The numbers listed below the figure are the
percentages of GLUT4 present in each sample compared to the
0 nM insulin control based on densitometric analysis.









3-13 are the percentages of GLUT4 in each sample compared to

the 0 nM insulin control as determined by densitometric

analysis. The dose dependence of this down-regulation

(Figure 3-12 B and Figure 3-13) yielded a K50 of about

600pM, slightly higher than the K50 of insulin-resistant

glucose transport, but a technically more difficult

parameter to measure.


Effect of Glucose Deprivation on GLUT4 Expression


Based on the observation that glucose deprivation

prevented the loss in insulin sensitivity (see Figure 3-4

A), we examined the expression of GLUT4 in the LDM fraction

of cells exposed to glucose-free medium. Figure 3-14 shows

that glucose deprivation blocked the loss of GLUT4 in

chronically treated cells. Thus, this shows for the first

time that glucose is important in regulating the expression

of GLUT4 in response to chronic insulin.


Effect of Chronic Insulin on GLUT4 mRNA Expression


These studies were conducted by Martin Williams to

evaluate the underlying mechanism of the reduction in GLUT4

protein. The level of GLUT4 mRNA was measured after

exposure to specific concentrations of insulin. As shown in

Figure 3-15 A, the level of GLUT4 mRNA decreases with

increasing insulin concentration. However, the





















Glucose + +
Chronic Insulin +


GLUT4 --


Figure 3-14. Effect of Glucose Deprivation on GLUT4
Expression. Cells were maintained in medium for 12 h in the
absence or presence of 10 nM insulin and/or 25 mM glucose.
The LDM fraction was isolated and GLUT4 analyzed by
immunoblot analysis. Data are representative of three
independent experiments.














Insulin
[nM]


GLUT 4 -



Actin -.


W ...qw.e


B

a.)
0


S0.0 i 1. .i ... 1 iil lw
0.1 1 10 100 1000
Insulin [nM]


Figure 3-15. Effects of Chronic Insulin Treatment on GLUT4
mRNA Levels. Panel A, cells were incubated for 12 h in the
presence of specific concentrations of insulin (refed every
2 h). Cells were then washed three times with 8 mL KRP at
which time RNA was extracted using the phenol:chloroform
extraction method. Twenty jig of total RNA were loaded onto
a 1% formaldehyde-agarose gel and subsequently transferred
to a nylon membrane. The membranes were probed with a 32P-
labeled cDNA for GLUT4. Panel B, densitometric analysis
represented as fractional difference in sensitivity. Data
shown represent a single experiment replicated three times.









concentration of insulin required to elicit a 50% loss in

GLUT4 mRNA was about 5 nM (Figure 3-15 B), which is 15 times

greater than that required for the equivalent loss of

insulin-sensitive glucose transport activity and 10-times

that for GLUT4 expression.


Reversal of Insulin Resistance


To determine if the effects of chronic insulin on 3T3-

Ll adipocytes could be reversed, insulin-resistant cells

were incubated with complete medium following the washout

procedure. Both basal and insulin-stimulated rates of

glucose transport were measured at specific times. Eight

hours after insulin removal, glucose transport activity of

previously insulin-resistant cells was equal to that of

controls (Figure 3-16). An interesting point is that the

reversal appears to occur between 4-8 h, as no improvement

in insulin-sensitive transport occurs before that time.

Also, the rate of basal transport between control cells and

resistant cells is identical over this time while only the

insulin-stimulated rates are changing. This shows that the

increase in insulin-stimulated transport seen in resistant

cells over this time is not due to increasing basal rates,

but improved insulin sensitivity.

To determine if this reversal in insulin resistance

required protein synthesis, cycloheximide was included in













a.)
M 2.0-


SI 1-5
- 0
S 1.0


Q 0.5
cIS~

rr 0.0 I
0 2 4 6 8

Time (h)





Figure 3-16. Reversal of Insulin Resistance. Cells were
incubated in the presence (chronic) or absence (control) of
10 nM insulin for 12 h. Plates were washed to remove
insulin and refed with fresh medium lacking insulin. At
specific times, basal and insulin-stimulated rates of
glucose transport were determined. Data represent the
average S.E. of two independent experiments (n=4).
control; U control + acute insulin; A chronic; V chronic +
acute insulin.









this experiment. As is shown in Figure 3-17, cycloheximide

completely blocked the reversal of insulin-resistant glucose

transport activity. It should be noted that cycloheximide

decreased the insulin-stimulated rate of glucose transport

in control cells (closed squares versus inverted triangles).

However, the cycloheximide-induced block in reversal is

considerably larger (open squares versus inverted

triangles). This suggests that the reversal of insulin-

resistant glucose transport is dependent upon protein

synthesis. It is not clear, however, if this reversal is

due to the synthesis and recovery of GLUT4, as those studies

have not yielded reliable results.


Conclusions


In the above studies, I have tested the hypothesis that

3T3-L1 adipocytes can serve as a model for studying the

development of insulin resistance under conditions that

might be realized in a physiological setting. Support for

this hypothesis has been gained from the following

observations. Chronic exposure to physiological levels of

insulin decreased the ability of an insulin challenge to

stimulate glucose transport. Interestingly, postprandial

concentrations of insulin in normal, obese, and diabetic

humans (84) plot along the inflection in the dose-response

curve between no change in insulin responsiveness and that









S2.0

+1.5



0-.
0^


>1.


~ ~0.5

0.0
0 2 4 6 8


Time (h)

-Control
U Control + acute ins
A Control + cyclo
Control + cyclo + acute ins
Chronic
Chronic + acute ins
Chronic + cyclo
-v-- Chronic + cyclo + acute ins





Figure 3-17. Reversal of Insulin Resistance in the Presence
of Cycloheximide. Cells were incubated and washed as
described in the previous figure. Plates were then refed
with fresh medium 20 jM cycloheximide. At specific times,
basal and insulin-stimulated rates of glucose transport were
determined. Data represent the average S.D. of duplicate
samples within a single experiment (n=2).









of maximal resistance. Thus, I have shown for the first

time that 3T3-L1 adipocytes develop insulin resistance in

response to physiologically relevant concentrations of

insulin. I have extended previous work by demonstrating

that insulin challenge of resistant cells stimulates

translocation, although the extent of recruitment is

suppressed relative to controls due to the reduction in the

total expression of GLUT4. Together, these data suggest

that the loss of GLUT4 protein underlies the inability of

3T3-L1 adipocytes to respond to insulin after chronic

exposure. This mimics the clinical manifestation of human

obesity and non-insulin-dependent diabetes where loss of

GLUT4 protein has been observed in adipose tissue (96),

although not in muscle (97).

It should be pointed out that transporter expression

differs in adipose tissue relative to 3T3-Ll adipocytes. In

isolated rat adipocytes, GLUT4 represents 97% of the GLUT

transporter pool (43). In 3T3-Ll adipocytes, GLUT4

represents only 33% of the pool (98), indicating the

substantially higher expression of GLUT1 relative to GLUT4

in this cell line. In control 3T3-L1 adipocytes, the PM

fraction contains about 25% of the GLUT1 pool. Chronic

insulin treatment increases the total pool of GLUT1, which

in turn doubles the GLUT1 content of the PM fraction.

Despite this 2-fold increase in GLUT1 in the PM of resistant









cells, no difference is observed in basal transport activity

(after washout) compared with controls. Resistant cells

treated acutely with insulin show little additional change

in GLUT1 in the PM. This argues that GLUT1 plays but a

small role in insulin-resistant glucose transport. In

contrast, only 3% of the GLUT4 pool resides in the PM of

either control or resistant cells (again, after washout).

When insulin is added acutely, GLUT4 content in the PM

reveals significant translocation; resistant cells show 50%

less GLUT4 following translocation compared to controls

reflecting the difference in the total pool. Again, this

suggests that GLUT4 expression determines insulin

resistance, even with the elevated levels of GLUT1 in the

3T3-L1 adipocyte cell line.

There are some differences between these data and those

reported previously on 3T3-L1 adipocytes. Flores-Riveros et

al. (14) reported that the concentration of insulin required

to induce a 50% change in GLUT4 mRNA was 23 nM in contrast

to the value I calculated, which was about 5 nM, close to

the Kd for the insulin receptor (99). This difference can

be explained by my refeeding protocol during chronic insulin

treatment to maintain the level of extracellular insulin,

particularly important at low hormone concentration, in the

face of extensive degradation of insulin by these cells. In

addition, cells in this study were exposed to insulin for









only 12 h compared to the 24-h exposure in the Flores-

Riveros study (14), which further lessens the impact of

insulin degradation. This temporal difference (24 versus 12

h of insulin treatment) also accounts for the smaller

magnitude of the increase in GLUT1 and decrease in GLUT4 in

our study relative to previous studies (14,76).

Importantly, the down-regulation of GLUT4 mRNA occurs at

insulin concentrations that are not likely to persist in the

physiological state. These insulin concentrations also do

not correlate with those required for the development of

insulin-resistant glucose transport or decrease in GLUT4

protein. Thus, GLUT4 expression appears to be regulated

transcriptionally, but this regulation may not be relevant

to insulin resistance.

Other studies in 3T3-Ll adipocytes have shown varying

results in transporter protein expression. Tordjman et al.

(100) and Koska et al. (16) showed that chronic insulin

treatment did not affect total GLUT4 protein expression

while studies of Flores-Riveros et al. (14) and Clancy and

Czech (76) showed a marked decrease. These latter data along

with my studies are consistent with the accelerated turnover

of GLUT4 in the presence of chronic insulin as measured by

Sargeant and Paquet (101), which will be discussed in the

next chapter. Ricort et al. (64,102) showed a small

decrease in the expression of GLUT4 but also very little









translocation to the PM with acute insulin stimulation.

These authors interpreted their data to mean that GLUT4

translocation was blocked, which clearly differs from our

studies. Kozka et al. (16) interpreted their cell surface

ligand binding experiments similarly, even though they

demonstrated a 50% reduction in cell surface GLUT4, which

would agree with my studies. I can only speculate as to the

cause for the different results. In both of these latter

studies, the loss in the GLUT4 pool was determined by

analyzing homogenate protein, revealing only modest changes

in expression. As the translocatable GLUT4 resides in the

LDM fraction, it may be that the loss was substantially

underestimated. Neither study separated the LDM fraction

from the HDM fraction; thus, this possibility can not be

evaluated. Finally, it is important to point out that my

experiments are the first to show that basal transport

activity can be achieved after chronic insulin treatment,

which allowed me to evaluate true insulin resistance. Data

collected under these conditions are consistent with the

hypothesis that the onset of insulin resistance (i.e.

depressed insulin-sensitive glucose transport) is a

reflection of the reduced GLUT4 pool, not a defect in

translocation. In addition, this decrease in insulin-

sensitive glucose transport can be reversed within 8 h and

is protein synthesis dependent.




Full Text
19
IgG conjugated horseradish peroxidase was purchased from
Sigma. Luciferin/luciferase (Cat. No. L9134) was obtained
from Sigma. Mouse leptin RIA Kit (Cat. No. ML-82K) was
obtained from Lineo. All other reagents were of the highest
quality available.
Methods
Cell Culture
Cells were grown and differentiated as previously
described (68). Briefly, 3T3-L1 fibroblasts were seeded on
polystyrene tissue culture dishes and fed every other day
with DMEM containing 10% calf serum (CS) for seven days
until confluence was reached. The fibroblasts were then
induced to differentiate into adipocytes by feeding with
DMEM containing 10% fetal bovine serum (FBS), 1 (0.g/mL
insulin, 0.5 mM methylisobutylxanthine, and 0.25 mM
dexamethasone (69). Two days later, the cells were refed
with DMEM containing 1 |j,g/mL insulin. Following this
treatment, cells were maintained by feeding with DMEM
containing FBS every other day. Cells were used 8-12 days
following differentiation.


142
(147), and 3T3-L1 adipocytes (148). In addition, leptin
expression is up-regulated by insulin in 3T3 cells
(149,150). Further, leptin may impair insulin action in
isolated rat adipocytes (151). Recent data has shown that
leptin administration decreases glucose uptake, GLUT4 mRNA
and GLUT4 protein levels in white adipose tissue (152).
However, an opposite response was observed in two other
insulin-sensitive tissues, namely, muscle and brown adipose
tissue.
Therefore, I measured the concentration of leptin
secreted into the medium by the 3T3-L1 adipocytes following
chronic insulin treatment using a radioimmunoassay kit
supplied by Lineo. However, low concentrations of leptin in
the medium or serum has complicated its measurement. Leptin
levels measured by ELISA have yielded leptin concentrations
in the plasma of obese individuals of 20 ng/mL while those
in normal individuals ranged from 1-7 ng/mL (153). Levels
measured by other groups using the same RIA kit as used in
the present studies yielded varying results. Values
measured by this method gave leptin levels ranging from 20
pg/ml (154) to 7 ng/ml (144). Following concentration of
the medium overlaying my 3T3-L1 adipocytes, values of
approximately 2.5 ng/ml were obtained (Figure B-l).
However, neither chronic insulin nor incubation with
fructose caused any significant change in the levels of


155
146. Slieker, L.J., Sloop, K.W., Surface, P.L., Kriauciunas,
A., LaQuier, F., Manetta, J., Bue-Valleskey, J.
and Stephens, T.W. (1996) J. Biol. Chem. 271,
5301-5304.
147. Zhang, B. et al. (1996) J. Biol. Chem. 271, 9455-9459.
148. Kallen, C.B. and Lazar, M.A. (1996) Proc. Natl. Acad.
Sci. U.S.A. 93, 5793-5796.
149. MacDougald, O.A., Hwang, C., Fan, H. and Lane, M.D.
(1995) Proc. Natl. Acad. Sci. U.S.A. 92, 9034-
9037 .
150. Leroy, P., Dessolin, S., Villageois, P., Moon, B.C.,
Friedman, J.M., Ailhaud, G. and Dani, C. (1996) J.
Biol. Chem. 271, 2365-2368.
151. Muller, G., Ertl, J., Gerl, M. and Preibisch, G. (1997)
J. Biol. Chem. 272, 10585-10593.
152. Wang, J., Chinookoswong, N., Scully, S., Qi, M. and
Shi, Z. (1999) Endocrinology. 140, 2117-2124.
153. Hebebrand, J., Heyden, J., Devos, R., Kopp, W.,
Herpertz, S., Remschmidt, H. and Herzog, W. (1995)
Lancet. 346, 1624-1625.
154. Barthel, A., Kohn, A.D., Luo, Y. and Roth, R.A. (1997)
Endocrinology. 138, 3559-3562.


65
Because the hexosamine biosynthetic pathway provides
substrate for glycoprotein synthesis, I tested the effects
of tunicamycin on the development of insulin resistance. As
shown in Figure 3-7, 2.5 |j.g/mL tunicamycin, which inhibits
N-linked glycosylation, decreased the insulin-stimulated
rate of glucose transport in both control and chronically-
treated cells. This is similar to the effect of
cycloheximide and again makes it difficult to determine if
tunicamycin was able to prevent the decrease in glucose
transport in insulin-resistant cells.
It has been hypothesized by Knutson et al. (95) that a
proteolytic fragment of the (3-subunit of the insulin
receptor is involved in the development of insulin
resistance. To test this, I used a thiol protease inhibitor
(E-64) to prevent cleavage of the (3-subunit (95) As is
shown in Figure 3-8, 100 (J.M E-64 had no effect on glucose
transport activity in control or chronically-treated cells.
As the Knutson group showed the largest effect of E-64 on
glycogen synthesis, I also investigated this insulin-
sensitive process. As Figure 3-9 shows, 100 (J.M E-64 had no
effect on rates of glycogen synthesis in control or insulin-
resistant cells.


117
plasma membrane as the controls when both groups were
stimulated acutely with insulin. Consistent with this was
the additional drop in GLUT4 in the LDM fraction relative to
that found in the LDM of control cells (Figure 5-3 A).
With the concern raised by Hresko et al. (120)
regarding glucosamine-induced ATP depletion, I measured the
concentration of ATP in cells chronically exposed to glucose
or fructose in the presence of insulin. As shown in Figure
5-4 A, the concentration of ATP did not change revealing an
intracellular concentration of about 7.6 mM 0.35. To
ascertain that ATP levels could indeed be affected, cells
were treated with 1 mM iodoacetate for 30 min which inhibits
glyceraldehyde phosphate dehydrogenase and thus glycolysis.
Under these conditions, ATP concentration dropped
significantly to 1.2 mM (Figure 5-4 B). These studies
conclude that fructose treatment had no deleterious effect
on the energy pool.
Effect of Fructose Feeding on Glycogen Metabolism
While there is evidence that fructose is metabolized in
3T3-L1 adipocytes (85), there are differences in the ability
of these cells to utilize glucose and fructose as
illustrated in Figure 5-5. In the following experiments,
total glycogen was measured after 12 h of exposure to either
glucose or fructose in the absence or presence of chronic


Figure 3-4. Effects of Glucose and Glucosamine on Insulin
Resistance. Panel A, cells were incubated for 6 h in DMEM
containing specific concentrations of glucose in the
absence (control) or presence (chronic) of 10 nM insulin.
Cells were washed and glucose transport activity was
determined in the presence of 1 p,M insulin. Panel B, cells
were incubated for 6 h in glucose-free and glutamine-free
DMEM containing specific concentrations of glucosamine in
the absence (control) or presence (chronic) of 10 nM
insulin. Washes were performed on the cells as described
earlier, and glucose transport activity following acute
stimulation with 1 fiM insulin was determined. Data
represent the average S.E. of two independent experiments
(n=4). Basal glucose transport activity in control and
glucose-deprived cells was 0.167 0.02 and 0.167 0.01
nmol/106 cells/min, respectively.


9
characteristics of adipocytes. Differentiation is a
terminal process at which point the cells no longer divide.
Rather, the cells accumulate lipid droplets and express
lipogenic enzymes such as glycerophosphate acyltransferase
and malic enzyme (38), ATP-citrate lyase, acetyl Co-A
carboxylase, and fatty acid synthetase (39). These cells,
once differentiated, also express increased numbers of
insulin-binding sites (40). As adipocytes, these cells
express two isoforms of the glucose transporter family:
GLUT1, the constitutive glucose transporter, and GLUT4, the
insulin-sensitive glucose transporter (41,42). This feature
is characteristic of authentic adipocytes (43). In
addition, GLT4 is only expressed in these cells following
differentiation (42).
GLUT Transporter Family
Characteristics of GLUT Family
GLUTl and GLUT4 are members of a larger group of
proteins responsible for the facilitated transport of
glucose into cells (44). The members of this family were
discovered using a variety of molecular biology techniques
and named in the order in which they were identified, GLUT1-
5 (for GLUcose Tranporters). Figure 1-2 shows the predicted
secondary structure of these transporters, based on
hydropathy plots, as they might appear in the plasma


31
the surface of the native protein (81). We chose to
synthesize a peptide corresponding to the last 13 amino
acids of the carboxy terminus GLUT4 due to both its
hydrophilicity and divergence from other members of this
transporter family and specifically GLUTl, which is also
expressed in 3T3-L1 adipocytes. The carboxy-terminal tail of
this transporter also makes it a good sequence for antibody
production because the predicted secondary structure shows
that the tail is probably exposed (42). However, a peptide
of only 13 amino acids is too small to elicit an immune
response if injected into a rabbit alone. Therefore, the
peptide is typically covalently linked to a larger protein
known as an immune carrier such as keyhole limpet hemacyanin
(KLH) or bovine serum albumin (BSA) (81). Although KLH and
BSA are the two most common immune carriers, others used are
ovalbumin, mouse serum albumin, or rabbit serum albumin
(81). As the peptide used to generate GLUT4 was injected
into a rabbit, KLH was used as the carrier.
Antiserum against GLUT4 was generated using a peptide
(CSTELEYLGPDEND) corresponding to amino acids 498-510 of the
GLUT4 sequence (underlined). This peptide was generated by
the Protein Chemistry Core facility at the University of
Florida and verified by amino acid analysis. An N-terminal
cysteine was added to allow its conjugation to KLH. This
conjugation was performed by using a thiol-specific cross-


Total Membrane
Protein Degradation
100
Figure 4-6. Degradation of Total Protein in Cells Treated
with Chronic Insulin. Total radioactivity and total protein
were determined from membrane samples collected in Figure 4-
5. Total radioactivity per pig of protein was then
calculated. control; chronic insulin.


Figure 4-5. Effect of Chronic Insulin on the Degradation
of GLUT4. Panel A, cells were incubated in DMEM lacking
cysteine and methionine and FBS for 1 h. DMEM,
supplemented with [35S] cysteine/methionine (400 (j,Ci) was
added and the cells incubated for 3 h. Cells were then
incubated in complete DMEM containing 10% FBS 10 nM
insulin for specific times. The medium was refreshed every
12 h and at the indicated times, total membranes were
collected and GLUT4 immunoprecipitated from equal volume
rather that equal protein. Immunoprecipitated GLUT4 was
treated as described in previous figures. Panel B,
densitometric analysis. The data represent the average of
three independent experiments. control; chronic
insulin.


86
only 12 h compared to the 24-h exposure in the Flores-
Riveros study (14), which further lessens the impact of
insulin degradation. This temporal difference (24 versus 12
h of insulin treatment) also accounts for the smaller
magnitude of the increase in GLUT1 and decrease in GLUT4 in
our study relative to previous studies (14,76).
Importantly, the down-regulation of GLUT4 mRNA occurs at
insulin concentrations that are not likely to persist in the
physiological state. These insulin concentrations also do
not correlate with those required for the development of
insulin-resistant glucose transport or decrease in GLUT4
protein. Thus, GLUT4 expression appears to be regulated
transcriptionally, but this regulation may not be relevant
to insulin resistance.
Other studies in 3T3-L1 adipocytes have shown varying
results in transporter protein expression. Tordjman et al.
(100) and Koska et al. (16) showed that chronic insulin
treatment did not affect total GLUT4 protein expression
while studies of Flores-Riveros et al. (14) and Clancy and
Czech (76) showed a marked decrease. These latter data along
with my studies are consistent with the accelerated turnover
of GLUT4 in the presence of chronic insulin as measured by
Sargeant and Paquet (101), which will be discussed in the
next chapter. Ricort et al. (64,102) showed a small
decrease in the expression of GLUT4 but also very little


CHAPTER 1
INTRODUCTION
Overview of Diabetes
Diabetes is one of the leading causes of death and
disability in the United States. An estimated 16 million
people in the United States have diabetes mellitus and about
half of these people do not know that they have the
disorder. Of these 16 million, 127,000 are under the age of
19. Each year, an additional 650,000 people are diagnosed
with diabetes. In 1992, diabetes contributed to the deaths
of more than 169,000 people and cost $92 billion in direct
and indirect costs. In total, diabetes accounted for 1 in 7
health care dollars spent (1).
Diabetes results when the pancreas produces little or
no insulin or the body's cells do not respond to insulin.
As a result, glucose accumulates in the blood, leading to
kidney overload and glucose excretion into the urine. There
are several types of diabetes: type I, type II, and
gestational diabetes. Type I, or insulin-dependent diabetes
mellitus (IDDM), is considered an autoimmune disease during
which the immune system attacks and destroys the insulin-
producing beta cells of the pancreas. Over time, this leads
1


21
these cells, a concentration of 10 nM insulin was chosen for
chronic insulin treatment for the remaining experiments (see
Chapter 1). This concentration was chosen as it can be
removed from the cells by the washout procedure, but the
medium does not have to be replaced every 2 h to maintain
effective insulin concentrations.
Glucose Transport Assay
Glucose transport activity was performed as described
previously (68). Briefly, medium was removed from 3T3-L1
adipocytes grown in 35-mm tissue culture dishes by washing
three times with 3 mL of KRP buffer. Cells were then
incubated in 1 mL KRP with or without 1 (iM insulin for 10
min. This was then followed by addition of 200 (J.M [3H] 2
deoxyglucose (0.2 pCi). After 10 min, transport was
terminated by washing the cells three times with 3 mL of
ice-cold phosphate-buffered saline (PBS). Cells were then
air-dried and lysed with 0.1% SDS and duplicate aliquots of
300 (0.L were taken for counting by liquid scintillation. The
rate of 2-deoxyglucose transport is reported as nmoles/106
cells/min.
Subcellular Fractionation of 3T3-L1 adipocytes
The subcellular fractionation technique allowed the
subcellular localization of GLUT4. Plasma membrane (PM),


148
42. Kaestner, K.H., Christy, R.J., McLenithan, J.C.,
Braiterman, L.T., Cornelius, P., Pekala, P.H. and
Lane, M.D. (1989) Proc. Natl. Acad. Sci. U.S.A.
86, 3150-3154.
43. Oka, Y., Asano, T., Shibasaki, Y., Kasuga, M.,
Kanazawa, Y. and Takaku, F. (1988) J. Biol. Chem.
263, 13432-13439.
44. Olson, A. and Pessin, J.E. (1996) Annu. Rev. Nutr. 16,
235-256.
45. Carruthers, A. (1990) Physiol. Rev. 70, 1135-1176.
46. Gould, G.W. and Bell, G.I. (1990) TIBS. 15, 18-23.
47. Blok, J., Gibbs, E.M., Lienhard, G.E., Slot, J.W. and
Geuze, H.J. (1988) J. Cell Biol. 106, 69-76.
48. Cushman, S.W. and Wardzala, L.J. (1980) J. Biol. Chem.
255, 4758-4762.
49. Suzuki, K. and Kono, T. (1980) Proc. Natl. Acad. Sci.
U.S.A. 77, 2542-2545.
50. Birnbaum, M.J. (1989) Cell. 57, 305-315.
51. Charron, M.J., Brosius, F.C., Alper, S.L. and Lodish,
H.F. (1989) Proc. Natl. Acad. Sci. U.S.A. 86,
2535-2539.
52. Fukumato, J., Kayano, L., Buse, J.B., Edwards, Y.,
Pilch, P.F., Bell, G.I. and Seino, S. (1989) J.
Biol. Chem. 264, 7776-7779.
53. James, D.E., Strube, M. and Mueckler, M. (1989) Nature.
338, 83-87.
54. Sollner, T., Whiteheart, S.W., Brunner, M., Erdjument-
Bromage, H., Geromanos, S., Tempst, P. and
Rothman, J.E. (1993) Nature. 362, 318-324.
55. Bock, J.B. and Scheller, R.H. (1997) Nature. 387, 133-
134 .
56. Martin, L.B., Shewan, A., Millar, C.A., Gould, G.W. and
James, D.E. (1998) J. Biol. Chem. 273, 1444-1452.


Glucose Transport
(fractional difference)
o o o o o
O M 4^ CD 00 O
Insulin Concentration (nM)
>
Glucose Transport
(fractional difference)
o o o o o
o hJ 4^ CD oo o


85
cells, no difference is observed in basal transport activity
(after washout) compared with controls. Resistant cells
treated acutely with insulin show little additional change
in GLUT1 in the PM. This argues that GLUT1 plays but a
small role in insulin-resistant glucose transport. In
contrast, only 3% of the GLUT4 pool resides in the PM of
either control or resistant cells (again, after washout).
When insulin is added acutely, GLUT4 content in the PM
reveals significant translocation; resistant cells show 50%
less GLUT4 following translocation compared to controls
reflecting the difference in the total pool. Again, this
suggests that GLUT4 expression determines insulin
resistance, even with the elevated levels of GLUT1 in the
3T3-L1 adipocyte cell line.
There are some differences between these data and those
reported previously on 3T3-L1 adipocytes. Flores-Riveros et
al. (14) reported that the concentration of insulin required
to induce a 50% change in GLUT4 mRNA was 23 nM in contrast
to the value I calculated, which was about 5 nM, close to
the Kd for the insulin receptor (99). This difference can
be explained by my refeeding protocol during chronic insulin
treatment to maintain the level of extracellular insulin,
particularly important at low hormone concentration, in the
face of extensive degradation of insulin by these cells. In
addition, cells in this study were exposed to insulin for


75
A
Insulin
[nM]
GLUT 4
B
G
'53
o
CL,
^t

D
J
O
Insulin [nM]
Figure 3-12. Effects of Chronic Insulin on GLUT4
Expression. Panel A, cells were incubated with specific
concentrations of insulin for 12 h (fed with fresh medium
every 2 h). Cells were then washed for 60 min and
subfractionated to isolate the LDM fraction. Proteins were
separated by SDS-polyacrylamide gel electrophoresis and
subsequently transferred to nitrocellulose membrane (70 [ig
of protein were loaded per lane). The membrane was then
probed for GLUT4 and visualized by enhanced
chemiluminescence. Panel B, GLUT4 bands were quantitated by
densitometry. Fractional difference in GLUT4 expression was
calculated from five independent experiments performed as in
panel A. Data represent the average + S.E.


114
A
PM
LDM
HDM
Glucose
+ +
+ +
+ +
Fructose
+ +
+ +
+ +
Chronic Insulin
+ +
+ +
+ +
GLUT4 -

fi;
GLUT1
B GLUT4
Figure 5-2. Effect of Fructose on GLUT Transporter Levels.
Panel A, cells were incubated as described in Figure 5-1 for
12 h. Plates were then washed as described and the PM, LDM,
and HDM fractions collected. Proteins (50 pg) were then
separated by SDS-polyacrylamide electrophoresis, transferred
to nitrocellulose and probed with antibodies to either GLUTl
or GLUT4. The antibody-protein complex was then visualized
by enhanced chemiluminescence. Panel B, Bands were
quantitated by video densitometry. These data are
representative of three independent experiments. || glucose; ggg
glucose + chronic insulin; ES3 fructose; E3! fructose +
chronic insulin.


55
dose-response curve to acute insulin challenge was shifted
to the right by an order of magnitude (Figure 3-1 C). This
is the first demonstration of "true" insulin-resistant
glucose transport activity in 3T3-L1 adipocytes because of
the ability to reestablish the basal state.
To determine if physiological insulin could establish
the insulin-resistant state, cells were exposed to specific
concentrations of insulin for 12 h. Cells were refed every
2 h to maintain the extracellular insulin levels,
particularly important at the lower concentrations of
insulin as these cells possess the ability to rapidly
degrade insulin. As shown in Figure 3-3 A, the
concentration of insulin that elicits a 50% reduction in
insulin-sensitive glucose transport was approximately 100
pM. This is extremely interesting because the fasting level
of insulin in non-diabetic humans is about 40 pM while that
in obese individuals is about 70 pM and individuals with
non-insulin-dependent diabetes have insulin levels around
200 pM (84). Insulin as low as 1 nM was sufficient to
completely desensitize the transport system. At 10 nM,
maximal desensitization was achieved whether the cells were
refed every 2 h or not. Thus, we chose a concentration of
10 nM to examine the time required for the development of
insulin resistance. As shown in Figure 3-3 B, the
phenomenon of desensitization was completely established


CHAPTER 2
MATERIALS AND METHODS
Materials
Dulbecco's modified Eagle's medium (DMEM) (Cat. No.
12100-061) and glutamine-, glucose-free DMEM (Cat. No.
23800-022) were obtained from Life Technologies, Inc. Fetal
bovine serum (FBS) (Cat. No. 1020-75) and calf serum (Cat.
No. 1100-90) were obtained from Intergen. Glucose-free FBS
was prepared by dialyzing FBS against phosphate-buffered
saline (PBS), pH 7.4, for 48 hours at 4C using dialysis
tubing with a molecular weight cutoff of 13,000 Da. Bovine
serum albumin (Cat. No. A-7030) was purchased from Sigma.
Insulin was a generous gift of Dr. Ronald Chance from Eli
Lilly Corp. L-Glutamine, D-glucosamine, D-fructose,
dexamethasone, and methylisobutylxanthine were obtained from
Sigma and D-glucose was obtained from Fisher. ProMix-35S-
Label (1000 Ci/mmol) (Cat. No. SJQ0079), 2-deoxy-D-[2,6 3H]
glucose (45 Ci/mmol) (Cat. No. TRK672), D-[U-14C]-glucose
(310 mCi/mmol) (Cat. No. CFB96) and D-[U-14C]-fructose (321
mCi/mmol) (Cat. No. CFB47) were obtained from Amersham.
SulfoLink Kit (Cat. No. 20405) was purchased from Pierce.
Protein A-Sepharose was obtained from Sigma. Anti-rabbit
18


11
membrane of the cell. These proteins are integral membrane
proteins consisting of twelve membrane spanning domains.
Both the amino and carboxy termini are located on the
intracellular side of the membrane. A large intracellular
loop between the sixth and seventh membrane spanning domains
divides the two halves of the protein. There is also a
single W-linked glycosylation site in the first
extracellular loop. The overall amino acid homology within
this family of proteins is 68% (45) and the amino acid
sequence identity between GLUT1 and GLUT4 is 63% (42).
The members of this family have differential tissue
distribution: GLUT1 is ubiquitously expressed at high levels
in fetal tissues, erythrocytes, brain, kidney, colon, and
adipocytes; GLUT2 is expressed in liver, (3-cells, kidney,
and small intestine; GLUT3 is expressed in many tissues
including brain, placenta, and kidney; GLUT4 is expressed in
skeletal muscle, heart, and adipocytes; and GLUT5, a
proposed fructose transporter as well, is expressed in the
small intestine (46) The Km's of these transporters also
differ. The Km's of GLUT1, 2, 3, and 4 for 3-0-
methylglucose, a none metabolizable form of glucose, are 21,
42, 10, and 2 mM, respectively, based on the oocyte
expression system (5). The high Km of GLUT2 is rationalized
by its localization to tissues involved in glucose
homeostasis (liver) and glucose sensing (|3-cells) .


17
3T3-L1 Adipocytes: A Model for Insulin Resistance
Many investigators have used the 3T3-L1 adipocyte cell
line to investigate insulin action. The majority of these
studies have used pharmacological doses of insulin to define
mechanisms. Under these conditions, insulin acts as a
growth factor sharing many of the mitogenic signaling paths
elicited by other growth factors (60). The metabolic
effects of insulin, observed at physiological
concentrations, are unique to insulin and cannot be
reproduced by other cellular stimuli (65-67). Thus, it
became important to determine the development of insulin
resistance under physiological conditions, the goal of this
research. I show that 1.) insulin-stimulated transport
activity decreases in response to physiological insulin; 2.)
GLUT4 expression is reduced resulting in fewer transporters
for mediating glucose uptake; 3.) insulin specifically
modulates the turnover of GLUT4; and 4.) translocation of
GLUT4 is not defective. These studies are complemented by
the roles of alternative hexoses on the development of
insulin resistance of both transport and glycogen turnover.
The appendices include studies on the effects of glucose
deprivation and chronic insulin on GLUT1 and leptin levels.


76
Insulin [nM]
\ O C\
^ Qy V -V3
GLUT4
100 93 67 37
Figure 3-13. Effects of Chronic Insulin on GLUT4 Expression
with Larger Samples. Cells were incubated as described in
Figure 3-12 with the exception that 5, 10-cm plates were
used per condition compared to 1, 10-cm plate in Figure 3-
12. The LDM fraction was collected and treated as described
in Figure 3-12. The numbers listed below the figure are the
percentages of GLUT4 present in each sample compared to the
0 nM insulin control based on densitometric analysis.


63
glucosamine, which bypasses GFAT, could not be prevented by
actinomycin D (91). As Figure 3-5 shows, in 3T3-L1
adipocytes, 20 (j,M cycloheximide had no effect on basal
transport, but inhibited insulin-stimulated glucose
transport in both control and chronically-treated cells.
Therefore, cycloheximide did not appear to prevent insulin
resistance. Studies in isolated rat adipocytes also showed
that cycloheximide was unable to prevent insulin resistance
(92) .
To further test the involvement of the hexosamine
biosynthetic pathway in the development of insulin
resistance, the effects of nikkomycin Z on insulin-resistant
glucose transport activity were tested. Nikkomycin Z is a
peptide-nucleoside antibiotic that competitively inhibits
chitin synthase in fungi and insects due to its structural
similarity to UDP-N-acetylglucosamine (93). I hypothesized
that this compound might prevent the onset of insulin
resistance by competing for enzymes which use UDP-N-
acetylglucosamine in mammalian cells. As Figure 3-6 shows,
neither concentration of nikkomycin Z tested had an effect
on glucose transport. Although I have no evidence that
nikkomycin is taken up by these cells, the concentrations
used are in far excess of 0.1 [J.M, the Ki of nikkomycin for
chitin synthase (94).


25
buffer (4% SDS, 6 M urea, 10% p-mercaptoethanol, 0.15 mg/mL
bromophenol blue, 40% glycerol and 20 mM Tris-base, pH 6.6).
Proteins were separated on 10% SDS-polyacrylamide gels by
running the gels for 15 h at 40 V.
Electrotransfer and Western Blotting
Following electrophoresis, the proteins were
transferred to nitrocellulose (pore size 0.45 |j,m) at 200 mA
for 2 h in transfer buffer (150 mM glycine, 20 mM Tris-base,
20% methanol, pH 8.2) using the method described by Clancy
and Czech (76). Nitrocellulose membranes were then immersed
in blocking buffer (20 mM Tris-base, 137 mM NaCl, 0.1%
Tween-20, pH 7.5, and 5% non-fat dry milk) for 1 h at room
temperature. Membranes were then incubated in blocking
buffer containing a 1:1000 dilution of either GLUT4 antisera
or GLUT1 antisera for 1 h at room temperature. Three washes
for 1 min followed by two washes for 5 min were performed
with buffer in the absence of non-fat dry milk. Membranes
were then incubated with a 1:100,000 dilution of horse
radish peroxidase conjugated goat anti-rabbit IgG in
blocking buffer for 1 h. Four 1-min and three 5-min washes
were performed with buffer in the absence of non-fat dry
milk to remove excess secondary antibody. The protein-
antibody complex was visualized using enhanced
chemiluminescence.


53
Insulin Concentration (nM)


Figure 4-8. Effects of Protease Inhibitors on GLUT4
Expression. Panel A, cells were incubated 10 nM insulin
and 100 pM leupeptin for 12 h. Insulin was removed by
washing the cells and membrane fractions collected.
Proteins were subjected to SDS-PAGE and transferred to
nitrocellulose for western blot analysis of GLUTl and
GLUT4. Panel B, cells were treated or not with 10 nM
insulin for 12 h in the presence or absence of 50 p.M
chloroquine, 100 |J,M leupeptin, 10 (J.M lactacystin, or 10 (J.M
MG132. Cells were washed and membrane fractions collected.
Shown is the densitometric analysis of the GLUT4 bands in
the LDM fractions expressed relative to the control.
control; ggg control + inhibitor; chronic insulin; g-g
chronic + inhibitor.


Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy
DEVELOPMENT OF INSULIN RESISTANCE IN 3T3-L1 ADIPOCYTES
By
Michael James Thomson
August 1999
Chairperson: Susan C. Frost
Major Department: Biochemistry and Molecular Biology
Insulin resistance is a manifestation of both diabetes
mellitus and obesity. However, the mechanism is still not
clearly identified. My goal was to determine if an in vitro
system, 3T3-L1 adipocytes, could serve as a model for
insulin resistance. This study describes a procedure that,
indeed, allows the evaluation of the development of insulin
resistance in 3T3-L1 adipocytes. Under these conditions, it
has been shown that the concentration of insulin required
for 50% desensitization of glucose transport activity is 100
pM and maximal desensitization could be achieved with 1 nM.
This demonstrates for the first time that 3T3-L1 adipocytes
develop insulin resistance in response to physiologically
relevant concentrations of insulin. Glucose (or
glucosamine), in addition to insulin, was required to
xi


68
chronic insulin
E-64
+ + + +
+ + + +
Figure 3-9. Effects of E-64 on Glycogen Synthesis. Cells
were incubated in the presence or absence of 100 p.M E-64
10 nM insulin for 12 h. Plates were then washed to remove
insulin and labeled with 14C-U-[D-glucose] (2 |j,Ci/plate) in
KRP containing 5 mM glucose in the presence or absence of 1
(j.M insulin for 1 h. Glycogen was then collected as
described and radioactivity counted by liquid scintillation
Data represent a single experiment.


33
The KLH-GLUT4 conjugate (200 (j,g) dissolved in Freund's
Complete Adjuvant was then injected into the popliteal lymph
node of a New Zealand white rabbit (82). The rabbit was
boosted 28 days later by intradermal injection of an equal
amount of KLH-GLUT4 conjugate and Freund's Incomplete
Adjuvant. This procedure was repeated 2 weeks later. Test
bleeds were collected every week by laceration of the medial
ear vein. Serum was isolated following coagulation of the
red blood cells. Additional boosts were performed every 6
weeks. The antiserum was divided into 1 mL aliquots and
stored at -20C.
To demonstrate the specificity of the GLUT4 antiserum
to GLUT4 protein, dot blots were first performed. GLUTl or
GLUT4 peptide was spotted onto separate nitrocellulose
sheets and the western blot procedure followed in which both
sheets were incubated with 1:1000 dilution of the GLUT4
antiserum. Figure 2-3 shows that the GLUT4 antiserum is
specific for the GLUT4 peptide as it does not cross react
with the GLUTl peptide. Next, the GLUT4 antiserum was shown
to be specific for the GLUT4 protein on a western blot, as
shown in Figure 2-4. The reactivity of the GLUT4 serum with
GLUT4 protein was completely prevented by competition with
GLUT4 peptide but not GLUTl peptide, showing that the
interaction was specific to GLUT4 and not due to non
specific binding.


LIST OF FIGURES
Figure page
1-1 Overview of the Hexosamine Biosynthetic Pathway.... 7
1-2 Predicted Secondary Structure of GLUT
Transporters 10
2-1 Subfractionation of 3T3-L1 Adipocytes 23
2-2 Fluctuations in Glycogen Levels Over a Week 29
2-3 Specificity of GLUT4 Antiserum for
GLUT4 Peptide 34
2-4 Specificity of GLUT4 Antiserum for
GLUT4 Protein 35
2-5 Elution Profile of Peptide-Purified
anti-GLUT4 Antibody 38
2-6 ATP Standard Curve 45
2-7 Leptin Standard Curve 48
3-1 Effect of Chronic Insulin on Glucose
Transport in 3T3-L1 Adipocytes 53
3-2 Time Course of Acute Insulin Stimulation 54
3-3 Development of Insulin Resistance 57
3-4 Effects of Glucose and Glucosamine on
Insulin Resistance 60
3-5 Effects of Actinomycin and Cycloheximide on
Insulin Resistance 62
3-6 Effects of Nikkomycin Z on Insulin Resistance 64
3-7 Effects of Tunicamycin on Insulin Resistance 66
viii


136
and therefore has enough energy supply. As a result, the
elevated hexosaraines signal the cell to down-regulate the
transport of glucose into the cell.
Once these questions are answered, one can find ways to
block the degradation of GLUT4 and prevent the onset of
insulin resistance. If, in fact, the degradation is due to
the action of a single protease, such as the insulin-
responsive aminopeptidase, compounds could potentially be
designed to inhibit this protease specifically. With
today's new technology of combinatorial chemistry and high
throughput assays, which can design and test thousands of
compounds at a time, this task should be less difficult than
in years past. As long as such a compound was not cytotoxic
or have any other deleterious effects on the cell, the
compound could be tested in clinical trials and marketed for
individuals with diabetes. However, the basic understanding
of the mechanisms and regulatory pathways of insulin
resistance must be realized first. The knowledge gained
from this venture may spark other ideas on how best to treat
this disease. For instance, once the translocation of GLUT4
is understood completely, stimulation of its translocation
by compounds other than insulin may provide viable options
for treating diabetes. Only the coming years and research
by talented scientists will yield the data necessary to
overcome this debilitating disease.


128
its calories from fructose, which leads to an increase in
plasma triglycerides (129) The liver in fact uses the
excess fructose to make triglycerides. This insulin
resistance could be completely ameliorated by the
administration of benfluorex, an agent that reduces hepatic
triglyceride output (129). This suggests that elevated
plasma triglyceride (or free fatty acids derived from
lipoprotein lipase activity on VLDL from the liver) is
responsible for the impaired insulin action rather than
elevated fructose per se. In contrast to animal studies,
those conducted in humans have shown beneficial effects from
dietary fructose. In fact, fructose administration
decreases postprandial serum glucose and insulin levels in
both normal and diabetic subjects (131-135). As well,
Koivisto et al. (131) found that dietary fructose increases
insulin sensitivity by 34% in subjects with type II
diabetes. In these human studies, the amount of fructose in
the diets represented only 13% of caloric content compared
to the 35-66% in animal diets (136). This lower level of
fructose does not lead to hypertriglyceridemia. Perhaps the
primary benefit of fructose in humans is to elicit a low
blood glucose and insulin secretory response (137) which
maintains insulin sensitivity and glycemic control.


39
Preparation of Total Membranes
3T3-L1 adipocytes grown in 10-cm tissue culture dishes
were washed three times with 4 mL of KRP at 37C. Cells
were then scraped into 4 mL of TES buffer (10 mM Tris-HCl,
pH 7.4, 1 mM EDTA, 250 mM sucrose, and 20 |j,g/mL PMSF) on
ice. The cells were homogenized using 20 strokes of a
teflon pestle in a 10 mL Potter-Elvejhem flask. The
homogenate was then centrifuged at 1,300 X g for 5 min to
remove nuclei and unbroken cells. The supernatant was
placed into polycarbonate tubes and membranes were collected
by centrifugation at 212,000 X g for 1 h in a T70.1 rotor
at 4C. The supernatant was discarded. For
immunoprecipitation, the pellet was resuspended in 1 mL of
extraction buffer (PBS containing 2% C12E9, 0.1% SDS,
1 mM EDTA, and 20 |o,g/mL PMSF) using 10 strokes of a teflon
pestle in a 2 mL Potter-Elvejhem homogenizing flask.
Insoluble material was then removed by centrifugation at
13,300 X g in a microfuge for 5 min. The supernatant was
transferred to a separate microfuge tube and protein
concentration was determined as described above. For
western blotting, the pellet was resuspended in 300 (J.L of
TES and protein concentration was determined as described
above.


62
acute insulin
actinomycin
cycloheximide
+ + +
+ +
+ +
control
+ + +
+ +
+ +
chronic
Figure 3-5. Effects of Actinomycin and Cycloheximide on
Insulin Resistance. 3T3-L1 adipocytes were incubated in the
presence or absence of 4 jj,M actinomycin D or 20 ¡J.M
cycloheximide also in the presence (chronic) or absence
(control) of 10 nM insulin for 12 h. Insulin was removed
from the cells by washing as described previously. Basal
and insulin-stimulated rates of glucose transport were
measured by incubating the cells in KRP with 1 (J.M insulin,
or not, for 10 min. The data represent the average S.E.
of two independent experiments (n=4).


99
shows that the degradation of GLUT4 was accelerated in
insulin-resistant cells. In fact, the half-life of GLUT4
under these conditions decreases from 29 h to 16 h in the
presence of chronic insulin. Figure 4-6 shows that this
increased degradation of GLUT4 was specific as the
degradation of total protein was not different under these
conditions.
Effects of Cycloheximide on the Loss of GLUT4
To determine if the loss of GLUT4 had a protein
synthesis component, the effects of cycloheximide on the
loss of GLUT4 were examined. Figure 4-7 shows that 20 p,M
cycloheximide completely blocked the loss in GLUT4 seen with
chronic insulin treatment. This shows that the loss in
GLUT4 was dependent on new protein synthesis. Most likely,
cycloheximide blocks the synthesis of a protease specific
for degrading GLUT4, or it blocks the synthesis of an
activator of a preexisting protease which is responsible for
the degradation of GLUT4.
Effects of Protease Inhibitors on GLUT4 Expression
The above result led me to consider whether GLUT4
degradation was sensitive to various protease inhibitors
which would aid in defining the route of degradation.
Therefore, two lysosomal degradative inhibitors, chloroquine


79
Insulin
[nM] O Q*' o \v V* ^ ^
GLUT 4
,7 ^ 1? $> S> $> nO
Actin
B
§
P
o
Figure 3-15. Effects of Chronic Insulin Treatment on GLUT4
mRNA Levels. Panel A, cells were incubated for 12 h in the
presence of specific concentrations of insulin (refed every
2 h). Cells were then washed three times with 8 mL KRP at
which time RNA was extracted using the phenol:chloroform
extraction method. Twenty (ig of total RNA were loaded onto
a 1% formaldehyde-agarose gel and subsequently transferred
to a nylon membrane. The membranes were probed with a 32P-
labeled cDNA for GLUT4. Panel B, densitometric analysis
represented as fractional difference in sensitivity. Data
shown represent a single experiment replicated three times.


8
is formed, which is used in the synthesis of glycolipids,
glycoproteins, and GPI-linked proteins. It was also shown
that glucosamine was at least 40 times more potent than
glucose in inducing insulin resistance (18). This
phenomenon can be explained by appreciating the fact that
glucosamine, which is transported into cells by the same
transporter as glucose, bypasses GFAT through direct
phosphorylation by hexokinase (18,21). The role of this
pathway was also confirmed by the use of two inhibitors of
GFAT: O-diazoacetyl-L-serine (azaserine) and 6-diazo-5-
oxonorleucine (DON), which are well established glutamine
analogs that irreversibly inhibit reactions in which
glutamine is an amide donor. In isolated rat adipocytes,
administration of these compounds blocked the development of
insulin-resistant glucose transport (18). Therefore, it was
hypothesized that products and/or intermediates of this
pathway are responsible for inducing insulin resistance.
3T3-L1 Adipocytes
The 3T3-Lls derive from murine embryonic fibroblasts
that, under the appropriate cell culture conditions,
differentiate into adipocytes (36,37). In the fibroblast
state, cells can be propagated and/or frozen under cryogenic
conditions. When these cells differentiate into adipocytes,
they exhibit the morphological and biochemical


REFERENCES
1. Sacks, D.B. and McDonald, J.M. (1996) Am. J. Clin.
Path. 105, 149-156.
2. Olefsky, J.M., Garvey, W.T., Henry, R.R., Brillon, D.,
Matthaei, S. and Freidenberg, G.R. (1988) Am. J.
Med. 85, 86-105.
3. Kahner, E.A., Porta, M. and Hyer, S.L. (1994) in:
Chronic Complications of Diabetes, pp. 52-62
(Pickup, J.C.a.W., G., Ed.) Blackwell Scientific
Publications, Oxford.
4. Walker, J.D. and Viberti, G.C. (1994) in: Chronic
complications of Diabetes, pp. 146-161 (Pickup,
J.C.a.W., G., Ed.) Blackwell Scientific
Publications, Oxford.
5. Thomas, P.K. (1994) in: Chronic Complications of
Diabetes, pp. 101-111 (Pickup, J.C.a.W., G., Ed.)
Blackwell Scientific Publications, Oxford.
6. Stout, R.W. (1992) in: Diabetes and Artherosclerosis,
pp. 53-88 (Stout, R.W., Ed.) Kluwer Academic
Publishers, The Netherlands.
7. Trimble, E.R. and McDowell, I.F.W. (1992) in: Diabetes
and Artherosclerosis, pp. 111-140 (Stout, R.W.,
Ed.) Kluwer Academic Publishers, The Netherlands.
8. Mayne, E.E. (1992) in: Diabetes and Artherosclerosis,
pp. 219-236 (Stout, R.W., Ed.) Kluwer Academic
Publishers, The Netherlands.
9. Garvey, W.T., Huecksteadt, T.P., Matthaei, S. and
Olefsky, J.M. (1988) J. Clin. Invest. 81, 1528-
1536.
10. Hotamisligil, G.S., Budavari, A., Murray, D. and
Spiegelman, B.M. (1994) J. Clin. Invest. 94, 1543-
1549.
145


101
PM LDM
Insulin + + + +
Cydohieximide + + + +
GLUT1
GS.UT4
* no wash out
Figure 4-7. Effect of Cycloheximide on the Loss of GLUT4.
Cells were incubated in DMEM containing 10 % FBS and 10 nM
insulin and 20 )j,M cycloheximide for 12 h. Insulin was not
removed by washing, membrane fractions were immediately
collected. Proteins were subjected to SDS-PAGE and
transferred to nitrocellulose for western blot analysis of
GLUT1 and GLUT4. The data are representative of three
independent experiments.


establish desensitization. The expression of GLUT4 protein
decreased by 50% with exposure to 10 nM insulin. The dose-
dependent loss of GLT4 was similar to the dose-dependence
for insulin-resistant transport activity. Translocation in
the presence of acute insulin was apparent, but the extent
of recruitment directly reflected the decrease in GLUT4
protein. GLUT4 mRNA also declined, but the K5o was
approximately 5 nM. Together, these data suggest that the
loss of GLUT4 protein underlies the cause of
desensitization. This decrease in GLUT4 protein was found
to be a combination of both decreased synthesis and
accelerated degradation.
Glucose, through its metabolism via the hexosamine
pathway, has been implicated in the regulation of insulin-
sensitive glucose transport. Surprisingly, fructose, an
alternate substrate for this pathway, does not appear to
support the development of insulin-resistant transport. To
study this apparent anomaly, we examined the levels and
subcellular distribution of GLUT4 in 3T3-L1 adipocytes in
response to chronic and acute insulin in the presence of
fructose or glucose. The results indicate that cells
exposed to either glucose or fructose, with chronic insulin,
down-regulate the expression of GLUT4. However, cells
exposed to fructose and chronic insulin and subsequently
stimulated with insulin, reveal levels of GLUT4 in the


DEVELOPMENT OF INSULIN RESISTANCE IN 3T3-L1 ADIPOCYTES
By
MICHAEL JAMES THOMSON
A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
1999


147
26. Hofmann, C., Lorenz, K., Braithwaite, S.S., Coica,
J.R., Palazuk, B.J., Hotamisligil, G.S. and
Spiegelman, B.M. (1994) Endocrinology. 134, 264-
270 .
27. Cornelius, P., Lee, M.D., Marlowe, M. and Pekala, P.H.
(1989) Biochem. Res. Common. 165, 429-436.
28. Szalkowski, D., White-Carrington, S., Berger, J. and
Zhang, B. (1995) Endocrinology. 136, 1474-1481.
29. Randle, P.J., Garland, P.B., Hales, C.N. and Newsholme,
E.A. (1963) Lancet. 1, 785-789.
30. Schlach, D.S. and Kipnis, D.M. (1965) J. Clin. Invest.
44, 2010-2020.
31. Randle, P.J., Newsholme, E.A. and Garland, P.B. (1964)
Biochem. J. 93, 652-655.
32. Ferrannini, E., Barrett, E.J., Bevilacqua, S. and
DeFronzo, R.A. (1983) J. Clin. Invest. 72, 1737-
1747 .
33. Arslanian, S.A. and Kalhan, S.C. (1994) Diabetes. 43,
908-914.
34. Biermann, E.L., Dole, V.P. and Roberts, T.N. (1959)
Diabetes. 6, 475-479.
35. Moitra, J. et al. (1998) Genes Dev. 12, 3168-3181.
36. Green, H. and Meuth, M. (1974) Cell. 3, 127-133.
37. Green, H. and Kehinde, 0. (1976) Cell. 7, 105-113.
38. Kuri-Harcuch, W. and Green, H. (1977) J. Biol. Chem.
252, 2158-2160.
39. Mackall, J.C., Student, D.K., Polakis, S.E. and Lane,
M.D. (1976) J. Biol. Chem. 251, 6462-6464.
40. Reed, B.C., Kaufmann, S.H., Mackall, J.C., Student,
A.K. and Lane, M.D. (1977) Proc. Natl. Acad. Sci.
U.S.A. 74, 4876-4880.
41. de Herreros, A.G. and Birnbaum, M.J. (1989) J. Biol.
Chem. 264, 19994-19999.


140
Figure A-2. Effect of Glucose Deprivation and Chronic
Insulin on Glycogen Levels. Cells were incubated in either
DMEM containing 10% FBS or glucose-free DMEM containing 10%
dialyzed FBS and in the presence or absence of 10 nM insulin
for up to 12 h. Cells were collected at specific times and
disrupted by sonication. Glycogen was isolated by ethanol
precipitation and then hydrolyzed using 2 N H2SO4 and the
concentration of glucose determined by a commercially
available kit (Sigma). Data represent the average S.D. of
duplicate plates within a single experiment. glucose;
glucose + chronic insulin; A glucose deprived; glucose
deprived + chronic isulin.


43
solvent was removed by soaking the gel in water for 30 min.
The gel was then soaked in 100 mL of 1 M sodium salicylate
for 1 h. The gel was dried under vacuum onto 3MM Whatman
paper and juxtaposed to x-ray film at -80C. The P
particles emitted by the 35S bombard the ring structure of
the sodium salicylate causing it to fluoresce and this
emission exposes the x-ray film.
ATP-Luciferase Assay
ATP levels were measured in the 3T3-L1 adipocytes by
using an ATP-luciferin/luciferase assay with the assistance
of Dr. James Gardner. The cleavage of luciferin by
luciferase requires ATP and is the basis of the assay. When
luciferin is cleaved, light is emitted and the intensity of
this light correlates with the amount of ATP in the sample.
Cells (grown in 35-mm dishes) were first rinsed three
times with 3 mL of KRP at 37C. One mL of 8% perchloric
acid at 4C was added to each plate and the cells scraped
directly into 1.5-mL eppendorf tubes. The samples were
neutralized by the addition of 3 M KOH containing 0.5 M
triethanolamine base dropwise until a pH of 6.0 was
obtained. The samples were kept on ice during this
procedure and the volume of the samples before and after the
addition of KOH was measured to calculate the volume of KOH
added. The precipitate formed was removed by centrifugation


12
Therefore, the flux of glucose into these tissues would be
expected to vary in a linear fashion over the typical range
of blood glucose levels (5). Following a meal, the very low
Km associated with GLUT4 would ensure that it operates close
to its Vmax even at decreasing blood glucose levels to ensure
maximum uptake of glucose from the blood into insulin-
responsive tissues (5).
Translocation of GLUT4
GLUT4, also referred to as the insulin-responsive
glucose transporter, resides in the trans-Golgi network of
vesicles that translocate to the plasma membrane in response
to insulin stimulation (47-49). This translocation
increases the number of glucose transporters on the cell
surface, and along with the lower Km of GLUT4, increases the
flux of glucose into the cell. Several groups have been
credited with cloning GLUT4 (42,50-53), which has ultimately
enabled studies investigating the regulation of its
translocation.
The basis for understanding translocation comes from
Rothman and collegues (54) who proposed the SNARE hypothesis
for vesicular trafficking. This theory proposes that a
unique vesicle-bound molecule (v-SNARE) specifically
recognizes and binds to a unique receptor molecule (t-SNARE)
on the target membrane. Three different v-SNARES (VAMP-1,


78
Glucose + +
Chronic Insulin +
+
GLUT4
Figure 3-14. Effect of Glucose Deprivation on GLUT4
Expression. Cells were maintained in medium for 12 h in the
absence or presence of 10 nM insulin and/or 25 mM glucose.
The LDM fraction was isolated and GLUT4 analyzed by
immunoblot analysis. Data are representative of three
independent experiments.


72
PM
Chronic Insulin + +
Acute Insulin + +
LDM HDM
+ + + +
+ + + +
GLUT4 -
B GLUT4 Levels


89
possibilities will be investigated in more detail in the
studies described in this chapter.
In addition to investigating the synthesis and
degradation of GLUT4, the degradative pathway responsible
for GLUT4 degradation was also examined. Two major
degradative pathways exist in mammalian cells, the lysosomal
pathway and the ATP-dependent proteasome pathway (103,104).
Several inhibitors exist to distinguish between these
pathways. Chloroquine and leupeptin inhibit the lysosomal
pathway. Chloroquine, an acidotropic amine, exerts its
effects by accumulating in and deacidifying lysosomes and
late endosomes (105) Another acidotropic amine, which
works in much the same way as chloroquine is ammonium
chloride. Leupeptin is a competitive inhibitor of the
lysosomal serine proteases cathepsins B, H, and L (106,107).
Lactacystin, a Streptomyces metabolite isolated by Omura et
al. (108), specifically inhibits the proteasome (109-111).
I also used MG132 (carbobenzoxyl-leucinyl-leucinyl-leucinal)
which, like lactacystin, inhibits protein degradation by the
proteasome (112,113).
Results
Specificity and Efficiency of GLUT4 Immunoprecipitation
Before the synthesis and degradation of GLUT4 could be
investigated, the immunoprecipitation procedure for


DEVELOPMENT OF INSULIN RESISTANCE IN 3T3-L1 ADIPOCYTES
By
MICHAEL JAMES THOMSON
A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
1999

This is dedicated to God, my parents, my dearest Amy, and
Moxie.

ACKNOWLEDGMENTS
I would sincerely like to thank the members of my
committee Drs. Michael Kilberg, Charles Allen, Thomas
O'Brien, Mark Atkinson, and my advisor Dr. Susan Frost for
their guidance throughout the course of this project. In
addition, I would like to thank the members of the lab and
my friends for their help throughout the time I have spent
here and for making this such an enjoyable period in my
life. My most heartfelt thanks goes to my parents, without
whose love and support I would never have achieved such an
undertaking. Last, but certainly not least, I would like to
thank my dearest Amy, whose love and companionship over the
last several months has not only marked a highlight in my
graduate career, but in my life as well.

TABLE OF CONTENTS
page
ACKNOWLEDGMENTS iii
LIST OF TABLES vii
LIST OF FIGURES viii
ABSTRACT xi
CHAPTERS
1 INTRODUCTION 1
Overview of Diabetes 1
Insulin Resistance 2
Role of Adipose in Insulin Resistance 3
Role of the Hexosamine Biosynthetic
Pathway in Insulin Resistance 6
3T3-L1 Adipocytes 8
GLUT Transporter Family 9
Characteristics of GLUT Family 9
Translocation of GLUT4 12
Insulin Receptor and Signaling 13
3T3-L1 Adipocytes: A Model for
Insulin Resistance 17
2 MATERIALS AND METHODS 18
Materials 18
Methods 19
Cell Culture 19
Chronic Insulin Treatment and
Insulin Washout 20
Glucose Transport Assay 21
Subcellular Fractionation of
3T3-L1 Adipocytes 21
Markwell Assay for Protein Determination 22
Gel Electrophoresis 24
Electrotransfer and Western Blotting 25
RNA Isolation and Northern Blotting 26
Glycogen Analysis 27
iv

Glycogen Synthesis 30
Production and Characterization of
GLUT4 Antibody 30
Peptide Purification of GLUT4 Antibody 36
Preparation of Total Membranes 39
Metabolic Labeling of 3T3-L1 Adipocytes 40
Immunoprecipitation of GLUT4 41
ATP-Luciferase Assay 43
Leptin Assay 44
3 DEVELOPMENT OF INSULIN RESISTANCE 4 9
Introduction 49
Results 50
Insulin-resistant Glucose
Transport Activity 50
Role of Glucose and Glucosamine 58
Effects of Various Inhibitors on the
Development of Insulin Resistance 61
Effect of Chronic Insulin on GLUT4
Expression and Translocation 69
Effect of Glucose Deprivation on
GLUT4 Expression 77
Effect of Chronic Insulin on
Glut4 mRNA Expression 77
Reversal of Insulin Resistance 80
Conclusions 82
4 MECHANISMS OF DECREASED GLUT4 LEVELS IN
INSULIN-RESISTANT CELLS 88
Introduction 88
Results 89
Specificity and Efficiency of
GLUT4 Immunoprecipitation 89
Synthesis of GLUT4 in
Insulin-resistant Cells 90
Degradation of GLUT4 in
Insulin-resistant Cells 94
Effects of Cycloheximide on the
Loss of GLUT4 99
Effects of Protease Inhibitors on
GLUT4 Expression 99
Conclusions 102
v

5 ROLE OF THE HEXOSAMINE BIOSYNTHETIC
PATHWAY IN THE DEVELOPMENT OF
INSULIN RESISTANCE 107
Introduction 107
Results 110
Effects of Fructose Feeding on the
Glucose Transport System 110
Effect of Fructose Feeding on
Glycogen Metabolism 117
Conclusions 123
6 CONCLUSIONS AND FUTURE DIRECTIONS 129
Conclusions 129
Future Directions 132
APPENDICES
A GLUCOSE DEPRIVATION AND GLUT TRANSPORTERS 137
B LEPTIN AND INSULIN RESISTANCE 141
REFERENCES 145
BIOGRAPHICAL SKETCH 156
vi

LIST OF TABLES
Table page
5-1 Glycogen Synthesis in 3T3-L1 Adipocytes 124

LIST OF FIGURES
Figure page
1-1 Overview of the Hexosamine Biosynthetic Pathway.... 7
1-2 Predicted Secondary Structure of GLUT
Transporters 10
2-1 Subfractionation of 3T3-L1 Adipocytes 23
2-2 Fluctuations in Glycogen Levels Over a Week 29
2-3 Specificity of GLUT4 Antiserum for
GLUT4 Peptide 34
2-4 Specificity of GLUT4 Antiserum for
GLUT4 Protein 35
2-5 Elution Profile of Peptide-Purified
anti-GLUT4 Antibody 38
2-6 ATP Standard Curve 45
2-7 Leptin Standard Curve 48
3-1 Effect of Chronic Insulin on Glucose
Transport in 3T3-L1 Adipocytes 53
3-2 Time Course of Acute Insulin Stimulation 54
3-3 Development of Insulin Resistance 57
3-4 Effects of Glucose and Glucosamine on
Insulin Resistance 60
3-5 Effects of Actinomycin and Cycloheximide on
Insulin Resistance 62
3-6 Effects of Nikkomycin Z on Insulin Resistance 64
3-7 Effects of Tunicamycin on Insulin Resistance 66
viii

3-8 Effects of E-64 on Insulin Resistance 67
3-9 Effects of E-64 on Glycogen Synthesis 68
3-10 Effects of Acute and Chronic Insulin on
Total GLUT Protein Levels 70
3-11 Subfractionation of Insulin-Resistant
3T3-L1 Adipocytes 72-73
3-12 Effects of Chronic Insulin on GLUT4
Expression 75
3-13 Effects of Chronic Insulin on GLUT4
Expression with Larger Samples 76
3-14 Effect of Glucose Deprivation on
GLUT4 Expression 78
3-15 Effect of Chronic Insulin Treatment on
GLUT4 mRNA Levels 79
3-16 Reversal of Insulin Resistance 81
3-17 Reversal of Insulin Resistance in the
Presence of Cycloheximide 83
4-1 Titration of Peptide-Purified GLUT4 Antibody
for Immunoprecipitation 91
4-2 Specificity of the Peptide-Purified GLUT4
Antibody in Immunoprecipitation 92
4-3 Efficiency of GLUT4 Immunoprecipitation 93
4-4 Effect of Chronic Insulin on the
Synthesis of GLUT4 96
4-5 Effect of Chronic Insulin on the
Degradation of GLUT4 98
4-6 Degradation of Total Protein in
Cells Treated with Chronic Insulin 100
4-7 Effect of Cycloheximide on the Loss of GLUT4 101
4-8 Effects of Protease Inhibitors on
GLUT4 Expression 104
ix

5-1 Effects of Fructose on the Development of
Insulin Resistance 112
5-2 Effect of Fructose on GLUT4 Transporter Levels 114
.5-3 Translocation of GLUT4 in Cells Treated with
Fructose 116
5-4 Effects of Fructose on ATP Levels 119
5-5 Effect of Fructose and Chronic Insulin on
Glycogen Levels 120
5-6 Effects of Fructose and Chronic Insulin on
Glycogen Synthesis 122
A-l Effect of Glucose Deprivation and Chronic
Insulin on the Aberrant Glycosylation of
GLUTl and GLUT4 138
A-2 Effect of Glucose Deprivation and Chronic
Insulin on Glycogen Levels 140
B-l Leptin Levels 144
x

Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy
DEVELOPMENT OF INSULIN RESISTANCE IN 3T3-L1 ADIPOCYTES
By
Michael James Thomson
August 1999
Chairperson: Susan C. Frost
Major Department: Biochemistry and Molecular Biology
Insulin resistance is a manifestation of both diabetes
mellitus and obesity. However, the mechanism is still not
clearly identified. My goal was to determine if an in vitro
system, 3T3-L1 adipocytes, could serve as a model for
insulin resistance. This study describes a procedure that,
indeed, allows the evaluation of the development of insulin
resistance in 3T3-L1 adipocytes. Under these conditions, it
has been shown that the concentration of insulin required
for 50% desensitization of glucose transport activity is 100
pM and maximal desensitization could be achieved with 1 nM.
This demonstrates for the first time that 3T3-L1 adipocytes
develop insulin resistance in response to physiologically
relevant concentrations of insulin. Glucose (or
glucosamine), in addition to insulin, was required to
xi

establish desensitization. The expression of GLUT4 protein
decreased by 50% with exposure to 10 nM insulin. The dose-
dependent loss of GLT4 was similar to the dose-dependence
for insulin-resistant transport activity. Translocation in
the presence of acute insulin was apparent, but the extent
of recruitment directly reflected the decrease in GLUT4
protein. GLUT4 mRNA also declined, but the K5o was
approximately 5 nM. Together, these data suggest that the
loss of GLUT4 protein underlies the cause of
desensitization. This decrease in GLUT4 protein was found
to be a combination of both decreased synthesis and
accelerated degradation.
Glucose, through its metabolism via the hexosamine
pathway, has been implicated in the regulation of insulin-
sensitive glucose transport. Surprisingly, fructose, an
alternate substrate for this pathway, does not appear to
support the development of insulin-resistant transport. To
study this apparent anomaly, we examined the levels and
subcellular distribution of GLUT4 in 3T3-L1 adipocytes in
response to chronic and acute insulin in the presence of
fructose or glucose. The results indicate that cells
exposed to either glucose or fructose, with chronic insulin,
down-regulate the expression of GLUT4. However, cells
exposed to fructose and chronic insulin and subsequently
stimulated with insulin, reveal levels of GLUT4 in the

plasma membrane which are similar to controls. While the
mechanism underlying this apparent "enhancement" of GLUT4
translocation is unknown, it explains the observation that
the insulin-stimulated glucose transport activity in cells
exposed to fructose does not reflect the insulin-resistant
state.
xiii

CHAPTER 1
INTRODUCTION
Overview of Diabetes
Diabetes is one of the leading causes of death and
disability in the United States. An estimated 16 million
people in the United States have diabetes mellitus and about
half of these people do not know that they have the
disorder. Of these 16 million, 127,000 are under the age of
19. Each year, an additional 650,000 people are diagnosed
with diabetes. In 1992, diabetes contributed to the deaths
of more than 169,000 people and cost $92 billion in direct
and indirect costs. In total, diabetes accounted for 1 in 7
health care dollars spent (1).
Diabetes results when the pancreas produces little or
no insulin or the body's cells do not respond to insulin.
As a result, glucose accumulates in the blood, leading to
kidney overload and glucose excretion into the urine. There
are several types of diabetes: type I, type II, and
gestational diabetes. Type I, or insulin-dependent diabetes
mellitus (IDDM), is considered an autoimmune disease during
which the immune system attacks and destroys the insulin-
producing beta cells of the pancreas. Over time, this leads
1

2
to an inability of the pancreas to produce insulin. Type
II, or non-insulin-dependent diabetes mellitus (NIDDM), is
the most prevalent and least understood form of diabetes.
About 90-95% of the people with diabetes have type II. This
form usually develops in adults over 40 and is most common
in adults over 55. Type II diabetes is also associated with
obesity in that 80% of the people with this form of diabetes
are overweight. With this form, the pancreas generally
produces insulin, but the body does not respond to insulin
effectively, a condition known as insulin resistance.
Gestational diabetes develops during pregnancy and generally
disappears afterwards. However, women who develop diabetes
during pregnancy have a greater risk of developing type II
diabetes later in life.
Insulin Resistance
Insulin resistance is one of the determining
characteristics of type II diabetes mellitus (NIDDM) and
obesity and is believed to be the underlying pathology (1) .
NIDDM is characterized by both hyperglycemia and
hyperinsulinemia, whereas obesity manifests only
hyperinsulinemia (2). Complications of insulin resistance
and diabetes include retinopathy (3), nephropathy (4),
neuropathy (5), and artherosclerosis (6). The
hyperlipidemia (7), hypercoagulation of platelets (8), and

3
artherosclerosis associated with diabetes can lead to
coronary artery disease, which is the leading cause of
mortality among diabetics (6).
Insulin resistance is defined as a decreased biological
response to physiological concentrations of insulin in
insulin-responsive tissues (1). In adipocytes, this is
characterized by a reduction in insulin-stimulated glucose
transport activity (9). As of yet, the exact mechanism
leading to the onset of insulin resistance has yet to be
elucidated. However, several mechanisms have been proposed
including decreased autophosphorylation of the insulin
receptor (2,10-13), down-regulation of the insulin-
responsive glucose transporter GLUT4 (2,14), altered
translocation of GLUT4 to the plasma membrane from its
intracellular storage vesicles (15-17), and the actions of
the intermediates or products of the hexosamine biosynthetic
pathway (18-21) .
Role of Adipose in Insulin Resistance
As mentioned, many obese subjects also exhibit insulin
resistance. However, when these patients are placed on
calorie-restrictive diets to decrease their weight, insulin
resistance is reversed (22,23). This suggests that adipose
plays an important role in insulin resistance. Another
condition which shows the importance of adipose in insulin

4
resistance is the metabolic syndrome X. This syndrome
manifests diabetes, hypertension, and dyslipidemia with
obesity as the underlying cause.
Adipose also secretes several factors which may be
involved in insulin resistance. For instance, evidence
exists that adipose tissue of obese mice have elevated TNF-
a mRNA levels and secretion of TNF-a protein (24-26).
Also, administration of TNF-a to Zucker fa/fa rats causes
reduced autophosphorylation of the insulin receptor as well
as decreased insulin-stimulated phosphorylation of IRS-1 in
both muscle and adipose tissue. Neutralization of TNF-a by
the addition of soluble TNF-a receptors prevents these
effects (10) and increases insulin sensitivity (24). It has
also been shown that TNF-a down-regulates GLUT4 mRNA levels
in adipocyte and myocyte cultures (24,27,28) as well as
reduces insulin-stimulated [3H]2-deoxyglucose transport into
3T3-L1 adipocytes (11). These studies provide evidence that
TNF-a plays a role in inducing insulin resistance in
adipose as well as muscle. One can imagine that increased
expression of TNF-a by adipose feeds back directly on
adipose by an autocrine loop mechanism, while its effects on
muscle may occur through a paracrine mechanism (25).
However, it has been shown in db/db mice that the elevated
levels of plasma TNF-a are well below those required to

5
support insulin resistance in cultured cell systems (24,25).
Therefore, it is very plausible that other proteins or
factors may be secreted by the adipose tissue which induce
insulin resistance.
Fatty acids have also been shown to be involved in the
induction of insulin resistance. In 1963, Randle et al.
(29) hypothesized that increased serum levels of free fatty
acids could interfere with glucose uptake and metabolism in
muscles, resulting in insulin resistance. This has come to
be known as the Randle hypothesis. Since the presentation
of this hypothesis, it has been shown by several groups,
including Randle's, that elevated serum free fatty acids
interfere with glucose utilization in vitro and in vivo (30-
33). In further support of this hypothesis, Bierman et al.
(34) have shown that type II diabetics have elevated fasting
levels of free fatty acids. In contrast, the absence of
white adipose tissue can also lead to diabetes as shown by
studies using a transgenic mouse model (35). These mice
express a dominant negative protein, called A-ZIP/F, which
prevents the DNA binding of B-ZIP transcription factors of
both the C/EBP and Jun families, which are necessary for
white adipose differentiation and formation. These mice are
diabetic, with fatty livers and elevated serum glucose,
insulin, and surprisingly, free fatty acids and
triglycerides. Interestingly, when fat is transplanted into

6
these mice, they lose the diabetic phenotype. This might
argue against the participation of adipose in the
development of insulin resistance. However, the fact that
obesity leads to insulin resistance demonstrates the
significance of adipose, suggesting that a small amount of
fat is beneficial, but a large excess is detrimental.
Role of the Hexosamine Biosynthetic Pathway in Insulin
Resistance
Marshall and colleagues (18) proposed that the
hexosamine biosynthetic pathway was involved in the
induction of insulin resistance in adipocytes because of the
in vitro requirement for glucose, glutamine, and insulin
(15,20). Figure 1-1 represents a diagram of this pathway.
In this pathway, glucose is transported into the cell by
either GLUT1, the constitutive glucose transporter, and/or
by GLUT4, the insulin-responsive transporter, where it is
rapidly phosphorylated to glucose 6-phosphate. This is then
isomerized to fructose 6-phosphate and the majority (97%) of
this is shunted into glycolysis and oxidative metabolism.
However, 2-3% of the fructose 6-phosphate enters the
hexosamine biosynthetic pathway through the rate-limiting
enzyme, glutamine:fructose 6-phosphate amidotransferase
(GFAT). GFAT transfers the amide group from glutamine to
fructose 6-phosphate to form glucosamine 6-phosphate.
Through several other reactions, UDP-N-acetylglucosamine

7
Glucosamine-6-P
Glucose-6-P
i
Fructose-6-P
97%
GFAT
C"
Glutamate
* Glucosamine-6-P
UDP-N-Acetylglucosamine
*
glycolipids
glycosylated proteins
GPI-linked proteins
Figure 1-1.
Pathway.
Glycolytic/TCA
Pathway
Insulin
Resistance
Overview of the Hexosamine Biosynthetic

8
is formed, which is used in the synthesis of glycolipids,
glycoproteins, and GPI-linked proteins. It was also shown
that glucosamine was at least 40 times more potent than
glucose in inducing insulin resistance (18). This
phenomenon can be explained by appreciating the fact that
glucosamine, which is transported into cells by the same
transporter as glucose, bypasses GFAT through direct
phosphorylation by hexokinase (18,21). The role of this
pathway was also confirmed by the use of two inhibitors of
GFAT: O-diazoacetyl-L-serine (azaserine) and 6-diazo-5-
oxonorleucine (DON), which are well established glutamine
analogs that irreversibly inhibit reactions in which
glutamine is an amide donor. In isolated rat adipocytes,
administration of these compounds blocked the development of
insulin-resistant glucose transport (18). Therefore, it was
hypothesized that products and/or intermediates of this
pathway are responsible for inducing insulin resistance.
3T3-L1 Adipocytes
The 3T3-Lls derive from murine embryonic fibroblasts
that, under the appropriate cell culture conditions,
differentiate into adipocytes (36,37). In the fibroblast
state, cells can be propagated and/or frozen under cryogenic
conditions. When these cells differentiate into adipocytes,
they exhibit the morphological and biochemical

9
characteristics of adipocytes. Differentiation is a
terminal process at which point the cells no longer divide.
Rather, the cells accumulate lipid droplets and express
lipogenic enzymes such as glycerophosphate acyltransferase
and malic enzyme (38), ATP-citrate lyase, acetyl Co-A
carboxylase, and fatty acid synthetase (39). These cells,
once differentiated, also express increased numbers of
insulin-binding sites (40). As adipocytes, these cells
express two isoforms of the glucose transporter family:
GLUT1, the constitutive glucose transporter, and GLUT4, the
insulin-sensitive glucose transporter (41,42). This feature
is characteristic of authentic adipocytes (43). In
addition, GLT4 is only expressed in these cells following
differentiation (42).
GLUT Transporter Family
Characteristics of GLUT Family
GLUTl and GLUT4 are members of a larger group of
proteins responsible for the facilitated transport of
glucose into cells (44). The members of this family were
discovered using a variety of molecular biology techniques
and named in the order in which they were identified, GLUT1-
5 (for GLUcose Tranporters). Figure 1-2 shows the predicted
secondary structure of these transporters, based on
hydropathy plots, as they might appear in the plasma

10
Figure 1-2. Predicted Secondary Structure of GLUT
Transporters.

11
membrane of the cell. These proteins are integral membrane
proteins consisting of twelve membrane spanning domains.
Both the amino and carboxy termini are located on the
intracellular side of the membrane. A large intracellular
loop between the sixth and seventh membrane spanning domains
divides the two halves of the protein. There is also a
single W-linked glycosylation site in the first
extracellular loop. The overall amino acid homology within
this family of proteins is 68% (45) and the amino acid
sequence identity between GLUT1 and GLUT4 is 63% (42).
The members of this family have differential tissue
distribution: GLUT1 is ubiquitously expressed at high levels
in fetal tissues, erythrocytes, brain, kidney, colon, and
adipocytes; GLUT2 is expressed in liver, (3-cells, kidney,
and small intestine; GLUT3 is expressed in many tissues
including brain, placenta, and kidney; GLUT4 is expressed in
skeletal muscle, heart, and adipocytes; and GLUT5, a
proposed fructose transporter as well, is expressed in the
small intestine (46) The Km's of these transporters also
differ. The Km's of GLUT1, 2, 3, and 4 for 3-0-
methylglucose, a none metabolizable form of glucose, are 21,
42, 10, and 2 mM, respectively, based on the oocyte
expression system (5). The high Km of GLUT2 is rationalized
by its localization to tissues involved in glucose
homeostasis (liver) and glucose sensing (|3-cells) .

12
Therefore, the flux of glucose into these tissues would be
expected to vary in a linear fashion over the typical range
of blood glucose levels (5). Following a meal, the very low
Km associated with GLUT4 would ensure that it operates close
to its Vmax even at decreasing blood glucose levels to ensure
maximum uptake of glucose from the blood into insulin-
responsive tissues (5).
Translocation of GLUT4
GLUT4, also referred to as the insulin-responsive
glucose transporter, resides in the trans-Golgi network of
vesicles that translocate to the plasma membrane in response
to insulin stimulation (47-49). This translocation
increases the number of glucose transporters on the cell
surface, and along with the lower Km of GLUT4, increases the
flux of glucose into the cell. Several groups have been
credited with cloning GLUT4 (42,50-53), which has ultimately
enabled studies investigating the regulation of its
translocation.
The basis for understanding translocation comes from
Rothman and collegues (54) who proposed the SNARE hypothesis
for vesicular trafficking. This theory proposes that a
unique vesicle-bound molecule (v-SNARE) specifically
recognizes and binds to a unique receptor molecule (t-SNARE)
on the target membrane. Three different v-SNARES (VAMP-1,

13
VAMP-2, and cellubrevin) and eight different t-SNARES
(Syntaxin-1, Syntaxin-2, Syntaxin-3, Syntaxin-4, Syntaxin-6,
Syntaxin-7, SNAP-25, and Syndet) have been described thus
far (55). VAMP-2 is associated with GLUT4 containing
vesicles and is the v-SNARE that is responsible for binding
to the t-SNARES Syntaxin-4 and Syndet, the murine homologue
of SNAP-25 (56,57). In addition, the protein Muncl8c has
been shown to regulate the translocation of GLUT4 by
inhibiting the association of VAMP-2 with Syntaxin-4 (58) .
Insulin relieves this inhibition by inducing the
dissociation of Muncl8c from Syntaxin-4 allowing VAMP-2 to
bind to Syntaxin-4. However, it is not known yet how
insulin stimulates the association and dissociation of these
molecules. In addition, other molecules may be involved in
GLUT4 translocation as investigations continue in this
exciting area.
Insulin Receptor and Signaling
Insulin action is mediated by a specific cell-surface
receptor. The insulin receptor is a heterotetramer
consisting of two a- and two (3-subunits. Each a- and 13-
subunit is connected by a disulfide bond to form an a/p
dimer and each a/p dimer is then attached via a single
disulfide bond between the a-subunits to form the tetramer.
The asubunit has a molecular mass of 135,000 Da and is

14
located completely on the outer surface of the plasma
membrane and contains the insulin binding site. The
p-subunit has a molecular mass of 95,000 Da and contains a
single transmembrane domain. The intracellular domain of
this subunit contains a juxtamembrane domain, an ATP binding
domain, and several tyrosine residues that are capable of
being phosphorylated. The intracellular domain also
contains a tyrosine kinase that is activated upon binding of
insulin presumably via propagation of a conformational
change from the a-subunit, which is transmitted through the
transmembrane domain of the p-subunit. Once the tyrosine
kinase has been activated, it is capable of
autophosphorylating tyrosines in specific regions of the
intracellular domain of the P-subunit including the
juxtamembrane domain, the regulatory domain, and the carboxy
terminal tail (1). This tyrosine kinase also phosphorylates
selected proteins on tyrosine residues within the cell such
as IRS-1. The phosphorylated IRS-1 recognizes and binds to
the src homology-2 (SH2) domains of various signal
transduction proteins, two of which are Grb2 and PI 3-
kinase. The activation of these molecules sets up divergent
signaling pathways within the cell.
Activation of Grb allows it to bind to son-of-sevenless
(SOS) through an SH3 domain. The Grb2/SOS complex then
activates p21ras which has been shown to bind directly to

15
Raf-1 serine/threonine kinase. This leads to the activation
of MAP kinase kinase which phosphorylates and activates MAP
kinase which activates transcription factors and increases
gene expression (59). It should be noted that
pharmacological concentrations of insulin are required for
activation of gene transcription. This brings doubt to the
physiological relevance of this path to insulin action (60).
The unique aspect of insulin action is its short-term
regulation of metabolic events including increased glucose
transport, increased glycogen synthesis, increased lipid
synthesis, and decreased lipolysis (1). Recent evidence has
shown that the PI 3-kinase cascade is the major pathway for
GLUT4 translocation and thus the stimulation of glucose
transport (61,62). PI 3-kinase is activated when the
phosphorylated IRS-1 binds to the SH2 domain of its p85a
regulatory subunit, this allows the pllO subunit to become
active. Activated PI 3-kinase leads to the activation of
the serine/threonine kinase Akt (also known as protein
kinase B). Although Akt is involved in the specific
stimulation of GLUT4 translocation (62), the exact mechanism
by which this occurs is not known. However, inhibitor
studies has shown the importance of the PI 3-kinase pathway
over the p21rss pathway. For instance, wortmannin, an
inhibitor of PI 3-kinase, blocks the stimulation of glucose
transport by insulin (63). In contrast, inhibition of

16
endogenous p21ras has no effect on insulin-stimulated GLUT4
translocation (61).
The insulin receptor and the insulin signaling
cascade(s) are important in insulin resistance because it
has been shown that defects in autophosphorylation of the
insulin receptor leads to insulin-resistant cells (2,10-13).
This leads to a decrease in the phosphorylation of IRS-1
(2,10-13) which in turn leads to a reduction in signaling
cascades. This effect, however, may primarily be due to
secondary effects of TNF-a and the pharmacological doses of
insulin used. Genetic defects in the insulin receptor also
result in decreased autophosphorylation of the receptor and
decreased phosphorylation of IRS-1. This is only a minor
cause for the development of insulin resistance, as it does
not account for the induction of insulin resistance in
individuals with normal insulin receptors and those
individuals who develop insulin resistance as a result of
becoming obese. However, there is evidence in 3T3-L1
adipocytes that chronic insulin treatment alone can lead to
impaired insulin receptor signaling and down-regulation of
IRS-1 expression and phosphorylation (64). Again, these
studies were performed using pharmacological concentrations
of insulin.

17
3T3-L1 Adipocytes: A Model for Insulin Resistance
Many investigators have used the 3T3-L1 adipocyte cell
line to investigate insulin action. The majority of these
studies have used pharmacological doses of insulin to define
mechanisms. Under these conditions, insulin acts as a
growth factor sharing many of the mitogenic signaling paths
elicited by other growth factors (60). The metabolic
effects of insulin, observed at physiological
concentrations, are unique to insulin and cannot be
reproduced by other cellular stimuli (65-67). Thus, it
became important to determine the development of insulin
resistance under physiological conditions, the goal of this
research. I show that 1.) insulin-stimulated transport
activity decreases in response to physiological insulin; 2.)
GLUT4 expression is reduced resulting in fewer transporters
for mediating glucose uptake; 3.) insulin specifically
modulates the turnover of GLUT4; and 4.) translocation of
GLUT4 is not defective. These studies are complemented by
the roles of alternative hexoses on the development of
insulin resistance of both transport and glycogen turnover.
The appendices include studies on the effects of glucose
deprivation and chronic insulin on GLUT1 and leptin levels.

CHAPTER 2
MATERIALS AND METHODS
Materials
Dulbecco's modified Eagle's medium (DMEM) (Cat. No.
12100-061) and glutamine-, glucose-free DMEM (Cat. No.
23800-022) were obtained from Life Technologies, Inc. Fetal
bovine serum (FBS) (Cat. No. 1020-75) and calf serum (Cat.
No. 1100-90) were obtained from Intergen. Glucose-free FBS
was prepared by dialyzing FBS against phosphate-buffered
saline (PBS), pH 7.4, for 48 hours at 4C using dialysis
tubing with a molecular weight cutoff of 13,000 Da. Bovine
serum albumin (Cat. No. A-7030) was purchased from Sigma.
Insulin was a generous gift of Dr. Ronald Chance from Eli
Lilly Corp. L-Glutamine, D-glucosamine, D-fructose,
dexamethasone, and methylisobutylxanthine were obtained from
Sigma and D-glucose was obtained from Fisher. ProMix-35S-
Label (1000 Ci/mmol) (Cat. No. SJQ0079), 2-deoxy-D-[2,6 3H]
glucose (45 Ci/mmol) (Cat. No. TRK672), D-[U-14C]-glucose
(310 mCi/mmol) (Cat. No. CFB96) and D-[U-14C]-fructose (321
mCi/mmol) (Cat. No. CFB47) were obtained from Amersham.
SulfoLink Kit (Cat. No. 20405) was purchased from Pierce.
Protein A-Sepharose was obtained from Sigma. Anti-rabbit
18

19
IgG conjugated horseradish peroxidase was purchased from
Sigma. Luciferin/luciferase (Cat. No. L9134) was obtained
from Sigma. Mouse leptin RIA Kit (Cat. No. ML-82K) was
obtained from Lineo. All other reagents were of the highest
quality available.
Methods
Cell Culture
Cells were grown and differentiated as previously
described (68). Briefly, 3T3-L1 fibroblasts were seeded on
polystyrene tissue culture dishes and fed every other day
with DMEM containing 10% calf serum (CS) for seven days
until confluence was reached. The fibroblasts were then
induced to differentiate into adipocytes by feeding with
DMEM containing 10% fetal bovine serum (FBS), 1 (0.g/mL
insulin, 0.5 mM methylisobutylxanthine, and 0.25 mM
dexamethasone (69). Two days later, the cells were refed
with DMEM containing 1 |j,g/mL insulin. Following this
treatment, cells were maintained by feeding with DMEM
containing FBS every other day. Cells were used 8-12 days
following differentiation.

20
Chronic Insulin Treatment and Insulin Washout
Fully differentiated 3T3-L1 adipocytes were incubated
in DMEM containing 10% FBS and specific concentrations of
insulin ranging from 1 pM to 10 nM for 12 h at 37C in a 7%
C02 incubator. Care was taken to reduce the loss of insulin
in solution, particularly at low concentrations. Thus,
insulin dissolved in 0.01 N HC1 was added to DMEM containing
10% FBS to give a final concentration of lOnM insulin. The
remaining insulin concentrations were achieved by serial
dilution into DMEM containing 10% FBS. In addition,
solutions were stored in plastic containers to prevent
insulin binding which occurs in glass bottles. Finally,
cells were refed every 2 h as adipocytes rapidly degrade
insulin.
After chronic treatment, an insulin washout procedure
was performed with the goal of rapidly returning the cells
to a basal state. Specifically, adipocytes were removed
from the C02 incubator and placed in a 37C waterbath.
Plates were washed three times with 3 mL of Krebs' Ringer
Phosphate buffer (KRP), pH 7.4, containing 0.1% BSA and 5 mM
glucose, every 20 min over 60 min. The final wash at 60 min
was performed with KRP alone. The cells were then assayed
according to the particular experimental protocol as
described below. After determining the optimal
concentration of insulin to induce insulin resistance in

21
these cells, a concentration of 10 nM insulin was chosen for
chronic insulin treatment for the remaining experiments (see
Chapter 1). This concentration was chosen as it can be
removed from the cells by the washout procedure, but the
medium does not have to be replaced every 2 h to maintain
effective insulin concentrations.
Glucose Transport Assay
Glucose transport activity was performed as described
previously (68). Briefly, medium was removed from 3T3-L1
adipocytes grown in 35-mm tissue culture dishes by washing
three times with 3 mL of KRP buffer. Cells were then
incubated in 1 mL KRP with or without 1 (iM insulin for 10
min. This was then followed by addition of 200 (J.M [3H] 2
deoxyglucose (0.2 pCi). After 10 min, transport was
terminated by washing the cells three times with 3 mL of
ice-cold phosphate-buffered saline (PBS). Cells were then
air-dried and lysed with 0.1% SDS and duplicate aliquots of
300 (0.L were taken for counting by liquid scintillation. The
rate of 2-deoxyglucose transport is reported as nmoles/106
cells/min.
Subcellular Fractionation of 3T3-L1 adipocytes
The subcellular fractionation technique allowed the
subcellular localization of GLUT4. Plasma membrane (PM),

22
low density membrane (LDM), and high density membrane (HDM)
fractions were isolated by a modification (70) of a
technique described by Weber et al. (71). The major
modification of this technique is the use of a steel block
homogenizer and tungsten ballbearing originally designed by
Balch and Rothman (72). The complete procedure is outlined
in Figure 2-1. Briefly, control or insulin-treated cells
were scraped into TES buffer (10 mM Tris-HCl, pH 7.4, 1 mM
EDTA, and 250 mM sucrose) at 18C. The cells were then
passed over a tungsten ball ten times in the steel block
homogenizer (at 18C) with a clearance of 0.0025 inch. A
crude plasma membrane fraction was collected at 17,000 X g
for 15 min at 4C. Purified membranes were collected from
this fraction by sucrose gradient centrifugation (71). LDM
and HDM fractions were collected by differential
centrifugation in 300 ¡j,L TES. Membrane fractions were
stored at -20C. Protein was determined by the method of
Markwell et al. (73).
Markwell Assay for Protein Determination
The Lowry procedure is widely used for the assay of
soluble proteins (74). However, detergents, used to
solubilize membranes for the release of integral membrane
proteins, and common compounds such as sucrose and EDTA used
in buffers for the isolation of membranes can interfere with

23
r-
Pellet
Scrape 5, 10-cm plates
at 18C with 4 mL TES each

Homogenize in steel block at 18C
(subsequent steps at 4C)
SS-34
12K, 15min
Supernatant
SS-34
12K, 20min
SS-34
20K, 30min
f
1
Pellet
Pellet
Supernatant
Sucrose Cushion
SS-34
T70.1
SW-28
20K, 30min
65K,7 5min
23K, 65min
>
i '
t
Interface HDM Pellet
SS-34
20K, 30min
Ti7 0.1
65K,60min
v
Pellet
LDM
SS-34
20K, 30min
PM
Figure 2-1. Subfractionation of 3T3-L1 Adipocytes.
Overview of subcellular fractionation procedure.

24
protein determination by the Lowry method (73). Therefore,
the Markwell procedure, which is a modification of the Lowry
procedure, was used for the analysis of membrane proteins in
our specific extraction buffer, which contains both sucrose
and EDTA (73).
A standard curve (0-100 pig) was generated by diluting a
solution of BSA (1 mg/mL) in water to a final volume of 0.1
mL. Ten piL of membranes suspended in TES was added to 0.1
mL of water. Ten piL of TES was also added to the standard
curve. One mL of a solution containing 2.0% Na2C3, 0.4%
NaOH, 0.16% Na+ K+ tartrate, and 1% SDS was added to the
standard curve and unknown samples, mixed, and allowed to
stand for 10 min. Folin reagent (0.1 mL of 1 N) was added,
mixed gently, and allowed to stand for 45 min. The
absorbance of the solution was measured at 650 nm and the
protein concentration of the samples was determined by
comparing them to the standards. A second order equation
was used to solve for the constants and the quadratic
equation was used to determine the protein concentration of
the unknown samples.
Gel Electrophoresis
Electrophoresis was performed as described by Laemmli
et al. (75) Equal amounts of membrane protein (50 jag) were
mixed with half the volume of 2X Laemmli sample dilution

25
buffer (4% SDS, 6 M urea, 10% p-mercaptoethanol, 0.15 mg/mL
bromophenol blue, 40% glycerol and 20 mM Tris-base, pH 6.6).
Proteins were separated on 10% SDS-polyacrylamide gels by
running the gels for 15 h at 40 V.
Electrotransfer and Western Blotting
Following electrophoresis, the proteins were
transferred to nitrocellulose (pore size 0.45 |j,m) at 200 mA
for 2 h in transfer buffer (150 mM glycine, 20 mM Tris-base,
20% methanol, pH 8.2) using the method described by Clancy
and Czech (76). Nitrocellulose membranes were then immersed
in blocking buffer (20 mM Tris-base, 137 mM NaCl, 0.1%
Tween-20, pH 7.5, and 5% non-fat dry milk) for 1 h at room
temperature. Membranes were then incubated in blocking
buffer containing a 1:1000 dilution of either GLUT4 antisera
or GLUT1 antisera for 1 h at room temperature. Three washes
for 1 min followed by two washes for 5 min were performed
with buffer in the absence of non-fat dry milk. Membranes
were then incubated with a 1:100,000 dilution of horse
radish peroxidase conjugated goat anti-rabbit IgG in
blocking buffer for 1 h. Four 1-min and three 5-min washes
were performed with buffer in the absence of non-fat dry
milk to remove excess secondary antibody. The protein-
antibody complex was visualized using enhanced
chemiluminescence.

26
RNA Isolation and Northern Blotting
RNA isolation and northern blotting was performed by
Martin Williams. Total cellular RNA was isolated by the
guanidinium thiocyanate procedure (77) Twenty fig of total
RNA was loaded onto a 1% formaldehyde-agarose gel and the
gel was run for 12-16 h at 40 V with constant buffer
recirculation. The RNA was then electrophoretically
transferred to an uncharged nylon membrane and cross-linked
with UV light for 3.5 min. GLUT4 probe was then generated
from cDNA generously provided by Dr. Maureen Charron (Albert
Einstein College of Medicine). The insert was labeled by
primer extension (78). Briefly, the DNA (0.5 pig) was
denatured by boiling at 100C for 5 min. After cooling for
3 min, the insert was incubated with a 6-base primer,
deoxynucleotides, one of which was labeled ([32P]dATP, dCTP,
dTTP, dGTP), and DNA polymerase. Free nucleotides were
removed by gel chromatography over Sephadex G-50. The
eluted labeled DNA was used directly to probe the RNA blot.
Hybridization was performed for 12-16 h in a buffer
containing 1% bovine serum albumin, 1 mM EDTA, 0.5 M sodium
phosphate, pH 7.2, and 1% SDS at 60C. The membranes were
washed under high stringency conditions (3 X 10-min washes
in 1 mM EDTA, 40 mM sodium phosphate, pH 7.2, and 1% SDS at
65C) and exposed to film for various lengths of time.

27
Relative intensity of each band was quantified by video
densitometry within the linear range of the film using the
Bioimage Visage 110.
Glycogen Analysis
Total glycogen was isolated as described previously
(79) based on a method described by Pfleiderer (80). Cells
grown on 10-cm tissue culture dishes were washed at 4C with
PBS, pH 7.2, and scraped into a 15-mL polypropylene tube in
1 mL PBS. The cells were then sonicated for 10 s on power 2
at 50% duty cycle (Branson Sonifier 450). Two mL of 30%
(w/v) KOH was added and the sample was mixed on a vortex
mixer. The sample was then boiled in a water bath for 15
min. After cooling to room temperature, 3.5 mL of 95% EtOH
were added. The samples were mixed and heated in a boiling
water bath for 3 min and cooled again to room temperature.
The precipitated glycogen was collected by centrifugation at
1,300 X g for 5 min at room temperature. The supernatant
was removed by aspiration and the glycogen pellet washed by
resuspension in 1 mL 95% EtOH. This suspension was
transferred to a 1 mL microfuge tube and the glycogen
collected by centrifugation at 13,300 X g for 5 min at 4C.
The supernatant was removed by aspiration and the
precipitate was stored at -20C.

28
Glucose was released from glycogen by acid hydrolysis.
Specifically, 0.2 mL 2 N H2SO4 was added to the glycogen
pellet and incubated in a boiling water bath for 2 h. The
samples were cooled to room temperature and 0.15 mL 2 N NaOH
and 0.65 mL H20 were added. The concentration of glucose in
the hydrolysate was then determined using a hexokinase-based
glucose kit (Glucose HK Kit) from Sigma. For samples fed
with glucose, 0.01 mL of hydrolysate was diluted with 0.99
mL of H2S04/Na0H/H20 mixture (2:1.5:6.5). Samples deprived
of glucose were diluted by adding 0.1 mL of hydrolysate to
0.9 mL of the H2S04/Na0H/H20 mixture. Duplicate aliquots of
0.1 mL were added to 1 mL of prewarmed assay reagent and
incubated at 37C for 30 min. The absorbance of the samples
was then measured at 340 nm.
To limit the variability in experiments dealing with
glycogen, glycogen levels were monitored during the week in
which the 3T3-L1 adipocytes were fully differentiated. As
these cells are fed every 48 h during their normal
maintenance feeding schedule and they begin to enter
glycogen deprivation on the tail end of this time, it was
hypothesized that glycogen levels would fluctuate between
feedings. Upon collecting glycogen from the cells each day
of the week, at the same time of day, I found that glycogen
levels did indeed fluctuate. As Figure 2-2 shows, glycogen
levels were highest the day after they were fed and lowest

29
Figure 2-2. Fluctuations in Glycogen Levels Over a Week.
Following differentiation, cells were maintained on their
feeding schedule and fed with DMEM containing 10% FBS on the
days indicated. Cells were collected at the same time every
day during the week and glycogen isolated by ethanol
precipitation. Following acid hydrolysis, glucose present
in glycogen was determined by a commercially available kit
(Sigma). Data represent the average S.D. of duplicate
samples within a single experiment (n=2).

30
on the days of a scheduled feeding as they became glucose
deprived. In addition, glycogen levels increased over the
course of the week from day 1 to day 5. This shows the
importance of performing a glycogen experiment the day
following a feeding and also performing the experiment at
the same time during the week (usually day 2 or 3 following
a feeding) to obtain consistent data.
Glycogen Synthesis
Cells adherent to 35-mm dishes were extensively washed
with KRP containing 0.1% BSA and 5 mM glucose as described
under "Insulin Washout" to return the cells to a basal level
of glucose transport. These cells were then labeled with 2
[iCi/plate of 14C-U-[D-glucose] in 1.5 mL KRP containing 5 mM
glucose (giving a final specific activity of 284 (j,Ci/mmol)
with or without 1 |j,M insulin for 1 h. Glycogen was then
collected as described above and the pellet dissolved in 300
[j,L of water. Radioactivity present was determined by liquid
scintillation.
Production and Characterization of GLUT4 Antibody
Hydrophilicity of a peptide is a criterion for
selecting a suitable sequence for antibody production
because hydrophilic peptides are more likely to be soluble
for coupling reactions and also more likely to be exposed on

31
the surface of the native protein (81). We chose to
synthesize a peptide corresponding to the last 13 amino
acids of the carboxy terminus GLUT4 due to both its
hydrophilicity and divergence from other members of this
transporter family and specifically GLUTl, which is also
expressed in 3T3-L1 adipocytes. The carboxy-terminal tail of
this transporter also makes it a good sequence for antibody
production because the predicted secondary structure shows
that the tail is probably exposed (42). However, a peptide
of only 13 amino acids is too small to elicit an immune
response if injected into a rabbit alone. Therefore, the
peptide is typically covalently linked to a larger protein
known as an immune carrier such as keyhole limpet hemacyanin
(KLH) or bovine serum albumin (BSA) (81). Although KLH and
BSA are the two most common immune carriers, others used are
ovalbumin, mouse serum albumin, or rabbit serum albumin
(81). As the peptide used to generate GLUT4 was injected
into a rabbit, KLH was used as the carrier.
Antiserum against GLUT4 was generated using a peptide
(CSTELEYLGPDEND) corresponding to amino acids 498-510 of the
GLUT4 sequence (underlined). This peptide was generated by
the Protein Chemistry Core facility at the University of
Florida and verified by amino acid analysis. An N-terminal
cysteine was added to allow its conjugation to KLH. This
conjugation was performed by using a thiol-specific cross-

32
linking reagent, sulfo-m-maleimidobenzoyl-N-
hydroxysuccinimide ester (sulfo-MSB). KLH (60 mg) was
dissolved in 10 mL of 10 mM potassium phosphate, pH 7.3, and
dialyzed at 4C against 10 mM potassium phosphate for 48 h.
The dialysate was concentrated by coating the dialysis
tubing with a liberal amount of Aquacide and wrapping in
foil. Protein aggregates were removed from the concentrated
dialysate by centrifugation in a microfuge for 10 min. The
concentration of protein in the dialysate was determined by
measuring its absorbance at 280 nm using the milligram
extinction coefficient of KLH (1.6). Six mg of KLH from a
12 mg/mL solution in 10 mM potassium phosphate, pH 7.3, was
added to 0.6 mg sulfo-MSB and incubated at room temperature
for 45 min. The KLH-MSB conjugate was separated from free
sulfo-MSB by size exclusion chromatography on a G-100 column
pre-equilibrated with 50 mM potassium phosphate, pH 6.0.
The free sulfo-MSB is retarded in the column and the KLH-MSB
conjugate flows through. Ten, 1 mL fractions were collected
and measured at 280 nm. KLH-MSB containing fractions were
pooled and reacted with 6 mg of the GLUT4 peptide for 3 h at
room temperature with rotation. The pH of the mixture was
adjusted with HC1 to produce the maximum amount of
precipitate. The mixture was then allowed to stand upright
overnight at 4C to allow the precipitate, which represents
the KLH-peptide conjugate, to settle.

33
The KLH-GLUT4 conjugate (200 (j,g) dissolved in Freund's
Complete Adjuvant was then injected into the popliteal lymph
node of a New Zealand white rabbit (82). The rabbit was
boosted 28 days later by intradermal injection of an equal
amount of KLH-GLUT4 conjugate and Freund's Incomplete
Adjuvant. This procedure was repeated 2 weeks later. Test
bleeds were collected every week by laceration of the medial
ear vein. Serum was isolated following coagulation of the
red blood cells. Additional boosts were performed every 6
weeks. The antiserum was divided into 1 mL aliquots and
stored at -20C.
To demonstrate the specificity of the GLUT4 antiserum
to GLUT4 protein, dot blots were first performed. GLUTl or
GLUT4 peptide was spotted onto separate nitrocellulose
sheets and the western blot procedure followed in which both
sheets were incubated with 1:1000 dilution of the GLUT4
antiserum. Figure 2-3 shows that the GLUT4 antiserum is
specific for the GLUT4 peptide as it does not cross react
with the GLUTl peptide. Next, the GLUT4 antiserum was shown
to be specific for the GLUT4 protein on a western blot, as
shown in Figure 2-4. The reactivity of the GLUT4 serum with
GLUT4 protein was completely prevented by competition with
GLUT4 peptide but not GLUTl peptide, showing that the
interaction was specific to GLUT4 and not due to non
specific binding.

34
A
1 2.5 5 \ig GLUT1 peptide
B
1 2.5 5 ¡ig GLUT4 peptide
Figure 2-3. Specificity of GLUT4 Antiserum for GLUT4
Peptide. Panel A, 1, 2.5, and 5 |_ig of GLUTl peptide was
blotted onto nitrocellulose and allowed to dry. Membranes
were blocked and probed with a 1:1000 dilution of the GLUT4
antiserum as described under "Electrotransfer and Western
Blotting". The protein-antibody complex was visualized by
enhanced chemiluminescence. Panel B, GLUT4 peptide was
blotted onto nitrocellulose and treated as in panel A.

35
GLUT4
. o o
& & d
S' S'
A A
£ £
& &
'O "O 'O
x v x
.<$? aJ" aJ* aJ"
^ ^y ^y ^y
V
Figure 2-4. Specificity of GLT4 Antiserum for GLUT4
Protein. Equal protein (50 |J,g) from the LDM fraction of
control cells was loaded per lane and subjected to SDS-PAGE.
Proteins were transferred to nitrocellulose and western blot
analysis performed under the following conditions: lane 1,
pre-immune serum; lane 2, 1:1000 dilution of GLUT4
antiserum; lane 3, GLUT4 antiserum competed with 10 (ig of
GLUT1 peptide; and lane 4, GLUT4 antiserum competed with 10
(_ig of GLUT4 peptide.

36
Peptide Purification of GLUT4 Antibody
In order to study the turnover and synthesis of GLUT4,
metabolic labeling with 35S-cysteine and -methionine was
performed followed by immunoprecipitation. In order to
prevent non-specific interactions during
immunoprecipitation, GLUT4-specific IgG was purified from
the anti-GLUT4 rabbit serum using a modification of the
technique described by Dankert et al. (83). Total IgG was
purified from 15 mL of serum by adjusting the pH to 5 with 3
M acetic acid and adding 0.75 mL of capryllic acid while
vigorously stirring. After stirring for 30 min, the
precipitate was collected by centrifugation at 41,000 X g
for 30 min at 4C. The supernatant was removed and its
volume measured. An equal volume of saturated ammonium
sulfate was added to the supernatant and stirred overnight
at 4C. The next day, the solution was divided equally
between two 15-mL corex tubes and the precipitate was
collected by centrifugation at 41,000 X g for 30 min at 4C.
The supernatant was discarded and each pellet was
resuspended in 0.5 mL of PBS containing 0.1% sodium azide.
The ammonium sulfate was removed from the samples by
dialyzing against three changes of PBS containing 0.1%
sodium azide over 48 h using tubing with a molecular weight
cut off of 13,000 Da. The concentration of the total IgG
solution was determined by measuring the absorbance of the

37
solution at 280 nm using a milligram extinction coefficient
of 1.4.
GLUT4 antibody was purified from the total IgG solution
by peptide chromatography. For this technique, we used a
SulfoLink Kit from Pierce. This kit contains a column
composed of 6% agarose beads cross-linked to iodoacetate.
The acetyl group binds irreversibly to free sulfhydryl
groups. Two mg of the GLUT4 peptide were bound to this
matrix through the terminal cysteine residue following the
manufacturer's instructions. GLUT4-specific antibody was
then purified by placing 2-3 mL of the total IgG solution on
the column and rotating at room temperature for 2 h. The
column was then washed with 16 mL of PBS and the GLUT4
antibody was eluted with 8 mL of 0.1 M glycine, pH 3.0. One
mL fractions were collected in tubes containing 100 (J.L 1 M
Tris-base. The absorbance of the fractions was measured at
280 nm and fractions containing antibody were pooled.
Figure 2-5 shows a typical elution profile of peptide-
purified anti-GLUT4 antibody. The concentration of the
resulting antibody solution was determined by its absorbance
at 280 nm using a milligram extinction coefficient of 1.4.
The purified GLUT4 antibody was then stored as 0.1 mL
aliquots at -20C.

38
Fraction
Figure 2-5. Elution Profile of Peptide-Purified anti-GLUT4
Antibody. Total IgG was added to the column and rotated for
2 h. Unbound antibody was washed from the column using PBS.
GLUT4 antibody was eluted using 0.1 M glycine, pH 3.0. One
mL fractions were collected and neutralized with 1 M Tris-
base. Absorbance of the fractions was measured at 280 nm
and the concentration determined using a milligram
extinction coefficient of 1.4.

39
Preparation of Total Membranes
3T3-L1 adipocytes grown in 10-cm tissue culture dishes
were washed three times with 4 mL of KRP at 37C. Cells
were then scraped into 4 mL of TES buffer (10 mM Tris-HCl,
pH 7.4, 1 mM EDTA, 250 mM sucrose, and 20 |j,g/mL PMSF) on
ice. The cells were homogenized using 20 strokes of a
teflon pestle in a 10 mL Potter-Elvejhem flask. The
homogenate was then centrifuged at 1,300 X g for 5 min to
remove nuclei and unbroken cells. The supernatant was
placed into polycarbonate tubes and membranes were collected
by centrifugation at 212,000 X g for 1 h in a T70.1 rotor
at 4C. The supernatant was discarded. For
immunoprecipitation, the pellet was resuspended in 1 mL of
extraction buffer (PBS containing 2% C12E9, 0.1% SDS,
1 mM EDTA, and 20 |o,g/mL PMSF) using 10 strokes of a teflon
pestle in a 2 mL Potter-Elvejhem homogenizing flask.
Insoluble material was then removed by centrifugation at
13,300 X g in a microfuge for 5 min. The supernatant was
transferred to a separate microfuge tube and protein
concentration was determined as described above. For
western blotting, the pellet was resuspended in 300 (J.L of
TES and protein concentration was determined as described
above.

40
Metabolic Labeling of 3T3-L1 Adipocytes
To measure the synthesis and turnover of proteins,
radiolabeling of proteins with [35S]cysteine/methionine was
performed to "mark" a population of proteins within a
specific period of time. The accumulation of radiolabel
into proteins was determined to measure synthesis. To
measure turnover, cells were pulsed with radiolabel followed
by an extended chase with nonradiolabeled medium. The
disappearance of the labeled protein was then monitored. To
examine synthesis of GLUT4, cells were incubated in complete
DMEM with 10% FBS in the presence or absence of 10 nM
insulin for 11 h. The medium was then changed to cysteine-
/methionine-free DMEM, also in the presence or absence of 10
nM insulin, for 1 h. This medium was then aspirated and 2
mL of the cysteine-/methionine-free DMEM containing 400 [iCi
of [35S]cysteine/methionine 10 nM insulin was added to
each plate. The plates were incubated at 37C for 10, 20,
30, 45, or 60 min at which time total membranes were
collected and extracted as described above.
Immunoprecipitation was performed on this solution.
Degradation of GLUT4 was examined by first incubating
cells in cysteine-/methionine-free DMEM for 1 h. Following
aspiration of this medium, 2 mL of the same medium
containing 400 [iCi of [35S] cysteine/methionine were added to

41
each plate and incubated at 37C for 3 h. This medium was
aspirated and replaced with complete DMEM containing 10% FBS
with or without 10 nM insulin for the chase. This medium
was refreshed every 12 h and total membranes were collected
at 0, 6, 12, 24, 48, and 72 h. Total membranes were
collected and immunoprecipitation performed.
Immunoprecipitation of GLUT4
Immunoprecipitation provides a means of studying the
turnover of a particular protein. This technique depends on
the existence of an antibody specific for the protein of
interest. Addition of the antibody to a mixture of proteins
results in specific antibody-protein complex formation.
This complex can then be collected by the addition of
Sepharose beads conjugated with protein A, a bacterial
protein that binds to the heavy chain of IgG. The beads are
precipitated by a short centrifugation and the bound
proteins are analyzed. Although in these studies this
technique was employed to study the dynamics of GLUT4,
protein-protein interactions can be determined by analyzing
proteins which co-immunoprecipitate with the protein of
interest.
To examine the synthesis of GLUT4, extracted membrane
protein of specific concentration (1.5 mg) was brought to 1
mL with additional extraction buffer. To investigate the

42
degradation of GLUT4, equal volumes (750 fo.L) of each sample,
which represents equal cell number, were brought to 1 mL
instead of equal protein. This was done as the total cell
protein content increases over extended periods of chronic
insulin treatment. Non-specific interactions with protein
A-Sepharose were reduced by preincubation of extracted
material with 50 fiL of a 50% slurry of protein A-Sepharose
at 4C for 1 h. The sepharose beads were removed by brief
centrifugation. Five |j,g of peptide-purified GLUT4 antibody
was added to the supernatants. The samples were rotated
overnight at 4C. The next morning, 25 |iL of a 50% slurry
of protein A-Sepharose were added to each sample and
rotation was continued for an additional 2 h at 4C. The
sepharose bead complexes were collected by brief
centrifugation. The supernatant was discarded. The beads
were then washed three times with 1 mL of extraction buffer
for 10 min followed by four, 10-min washes with 1 mL of
extraction buffer containing 1 M NaCl. Samples were then
washed with 1 mL TES buffer. The final pellet was
resuspended in 25 (^L of TES buffer. Sample dilution buffer
(30 |J,L) was added to the beads and incubated at 37C for 30
min. Proteins in the entire sample were resolved on a 10%
polyacrylamide gel by electrophoresis as described above.
The gel was then fixed in 100 mL of 10% (w/v)
trichloroacetic acid and 60% (v/v) methanol for 30 min. The

43
solvent was removed by soaking the gel in water for 30 min.
The gel was then soaked in 100 mL of 1 M sodium salicylate
for 1 h. The gel was dried under vacuum onto 3MM Whatman
paper and juxtaposed to x-ray film at -80C. The P
particles emitted by the 35S bombard the ring structure of
the sodium salicylate causing it to fluoresce and this
emission exposes the x-ray film.
ATP-Luciferase Assay
ATP levels were measured in the 3T3-L1 adipocytes by
using an ATP-luciferin/luciferase assay with the assistance
of Dr. James Gardner. The cleavage of luciferin by
luciferase requires ATP and is the basis of the assay. When
luciferin is cleaved, light is emitted and the intensity of
this light correlates with the amount of ATP in the sample.
Cells (grown in 35-mm dishes) were first rinsed three
times with 3 mL of KRP at 37C. One mL of 8% perchloric
acid at 4C was added to each plate and the cells scraped
directly into 1.5-mL eppendorf tubes. The samples were
neutralized by the addition of 3 M KOH containing 0.5 M
triethanolamine base dropwise until a pH of 6.0 was
obtained. The samples were kept on ice during this
procedure and the volume of the samples before and after the
addition of KOH was measured to calculate the volume of KOH
added. The precipitate formed was removed by centrifugation

44
in a microfuge for 5 min at 4C. The supernatant was
diluted 1:250.
Ten |uL of the 1:250 diluted sample or ATP standard was
added to a plastic 12 x 75 mm tube. To this, 500 p,L of
assay buffer (60 mM Tris-Acetate, pH 7.75, 10 mM MgCl2, 1 mM
KC1, 1.5 mM EDTA, 2.5 mM p-mercaptoethanol, 0.4 mg/mL
luciferin/luciferase stock) were added. The p-
mercaptoethanol was added after the initial solution was
brought to pH 7.75. The luciferin/luciferase stock was
prepared by adding 50 mg of luciferin/luciferase to 2.5 mL
of 50% glycerol. This was divided into 300 (J,L aliquots and
stored in liquid nitrogen and added to the assay buffer just
prior to use. Following addition of the assay buffer to the
sample, the tube was flicked for 10 sec then placed in the
luminometer and the fluorescence measured for 20 sec. ATP
in the samples was calculated by comparison against the
standard curve seen in Figure 2-6.
Leptin Assay
Cells (5, 10-cm plates) were fed with serum-free DMEM
for 12 h. The medium was then collected, pooled from the
five plates (a total of 40 mL), and concentrated using the
"Centriplus 3" centrifugal concentrating tubes from Amicon,
which have a molecular weight cut-off of 3,000 Da. The

45
0 12 3
pmol ATP
Figure 2-6. ATP Standard Curve. Ten |4l of specific
concentrations of ATP were added to 12 x 75 mm plastic
tubes. The luciferin/luciferase assay buffer was added, the
tube was flicked for 10 sec., and the fluorescence measured
in a luminometer for 20 sec.

46
medium was separated equally into four tubes and
concentrated by centrifugation at 3,000 x g (4,900 rpm) for
4.8 h at 4C in an SS-34 rotor. The retained solution was
collected from the four concentrators and pooled. The
pooled solution was then placed into a single new tube and
concentrated further by centrifugation at equal force and
time. The retained solution was collected again and brought
to 4 mL using PBS.
The leptin radioimmunoassay (RIA) was performed as per
the manufacturer's instructions. Briefly, 100 piL of assay
buffer (0.05 M PBS pH 7.4 containing 0.025 M EDTA, 0.1%
sodium azide, 0.05% Triton X-100 and 1% RIA grade BSA) were
added to 12 x 75 mm borosilicate glass tubes. Leptin
standards (100 j_iL) and unknown samples (100 |j,L) were added
to the appropriate tubes. Next, mouse leptin antibody (100
pL) was added to the tubes. The tubes were mixed on a
vortex incubated overnight at 4C. The next morning, 125I-
mouse leptin (100 jo.L) was added to all the tubes, mixed, and
incubated overnight at 4C. The following morning, 1 mL of
cold precipitating reagent was added to all the tubes and
mixed. The tubes were incubated for 20 min at 4C and the
precipitate collected by centrifugation for 15 min at 3,000
x g (5,000 rpm) in an SS-34 rotor at 4C. The supernatant
was decanted immediately and the tubes drained for 1 min.
Excess fluid was blotted from the lip of the tubes and the

47
pellet was counted in a gamma counter. Leptin levels
present in the samples were calculated by comparing against
the standard curve seen in Figure 2-7.

48
Figure 2-7. Leptin Standard Curve. Mouse leptin antibody
was added to specific concentrations of leptin provided in
the kit and incubated overnight at 4C. 125I-mouse leptin
was added and incubated at 4C overnight. One mL of cold
precipitating reagent was added and incubated for 20 min at
4C. The precipitate was collected by centrifugation, the
supernatant was decanted and the radioactivity present in
the pellet determined by a gamma counter.

CHAPTER 3
DEVELOPMENT OF INSULIN RESISTANCE
Introduction
Insulin resistance has been studied in animal models,
freshly isolated cells, and immortalized cells. Using a
cell line has several advantages. One of these advantages
is the ease of harvesting cultured cells. Another is the
consistency and reproducibility of data, the lack of which
plagues in vivo studies.
Although 3T3-L1 adipocytes have been the focus of many
insulin action studies, in each case, pharmacological doses
of insulin were used. In contrast to other classes of
hormones, insulin does not exhibit biphasic effects, thus
the rationale for using pharmacological doses. However,
caution must be taken as insulin at high concentrations in
cell culture activates the mitogenic paths held in common
with other growth factors, the relevance of which is
questioned (60). That said, with long term exposure to
pharmacological doses of insulin, GLUT4 expression (both
mRNA and protein) is reduced (14). This reduction is caused
by both down-regulation of transcription and enhanced
turnover of mRNA. However, the concentration of insulin
49

50
required to affect a 50% change in expression of message was
reported as 23 nM. This level of insulin is at least 2
orders of magnitude higher than the concentration of
circulating insulin in humans (2). No comparable dose-
response studies have examined the development of insulin-
resistant glucose transport activity or GLUT4 expression in
these cells. Therefore, I have developed a procedure that
has allowed the measurement of glucose transport after
chronic exposure (12 h) to physiological concentrations of
insulin.
Results
Insulin-resistant Glucose Transport Activity
Chronic insulin exposure elevates glucose transport
activity in 3T3-L1 adipocytes. This can be prevented by
protein synthesis inhibitors like cycloheximide (Risch and
Frost, unpublished data). More recent studies have shown
that pharmacological doses of insulin induces GLUT1
transcription (14) leading to elevated GLUT1 protein and
thus transport activity. In the face of elevated GLUT1,
insulin-sensitive glucose transport activity can not be
observed. While this may imply insulin resistance, elevated
basal transport activity complicates interpretation. My
first challenge was to develop a protocol to reestablish
basal transport activity after chronic exposure to insulin.

51
I relied initially on studies by Garvey et al. (15). These
investigators showed that incubation of cultured rat
adipocytes with insulin for 12 h was sufficient to complete
the desensitization process, resulting in a 50% reduction in
insulin-sensitive glucose transport activity. Therefore,
cells were first treated with either 10 nM or 1 |xM insulin
for 12 h. The cells were then washed in KRP buffer
containing 5 mM glucose and 0.1% defatted BSA at 20-min
intervals over a 140-min time course. At specific times
during this washout period, cells were rinsed in KRP alone
and glucose transport activity determined in a 2-min pulse.
Figure 3-1 A shows the comparison between control cells
(washed in an identical manner) and those treated
chronically with either 1 [xM or 10 nM insulin. Cells
treated chronically with 1 p,M insulin showed significantly
elevated transport at the start of washout (time 0) but
never achieved basal levels despite the extensive time of
washing. However, those cells treated with 10 nM insulin
returned to basal levels within 60 min of washing. When
these latter cells were subsequently restimulated with an
acute insulin challenge following the 60-min wash, the rate
of glucose transport was reduced by 50% compared to control
cells (Figures 3-1 B and 3-2). In addition to the decreased
rate, the cells were less sensitive to insulin in that the

Figure 3-1. Effect of Chronic Insulin on Glucose Transport
in 3T3-L1 Adipocytes. Panel A, fully differentiated 3T3-L1
adipocytes were incubated in DMEM containing 10% FBS in the
absence of added insulin (#) or with 1 |nM (A) or 10 nM (H)
insulin for 12 h. Plates were washed as described in
Chapter 2 under "Insulin Washout Procedure". Glucose
transport activity was measured in KRP (in the absence of
bovine serum albumin and glucose) by the addition of 200 |aM
[3H] 2-deoxyglucose (0.2 |J.Ci) After 2 min, transport was
terminated by the addition of ice-cold phosphate-buffered
saline. Cells were lysed with a 0.1% solution of SDS and
duplicate aliquots of 300 |i,L were taken for quantitating
radioactivity. Panel B, cells were incubated as above in
the absence (control) or presence (chronic) of 10 nM
insulin. Removal of insulin was accomplished within 60 min
(see panel A), then 1 |aM insulin was added back, or not, for
10 min, and glucose transport activity was determined.
Panel C, cells were incubated for 12 h in the absence
(control) or presence (chronic) of 10 nM insulin. The
cells were washed for 60 min to remove insulin. In order
to maintain the low levels of insulin in solution, glucose
transport activity was measured in KRP containing 0.1% BSA.
Therefore, following the final wash at 60 min, 950 ¡0.L of KRP
containing 0.1% BSA were added to each plate. Various
concentrations of insulin, dissolved in 0.1%, were then
added for 10 min for acute stimulation. Glucose transport
activity was measured by the addition of [3H]2-deoxyglucose.
Each panel represents the average S.E. of two independent
experiments (n=4).

53
Insulin Concentration (nM)

54
Figure 3-2. Time Course of Acute Insulin Stimulation.
Cells were incubated in DMEM containing 10% FBS 10 nM
insulin for 12 h. Plates were then washed as described in
Chapter 2 under "Insulin Washout Procedure". Cells were
then stimulated for specific times with 1 |j,M insulin and
glucose transport activity determined. Data represent the
average S.D. of duplicate samples within a single
experiment (n=2). control; chronic insulin.

55
dose-response curve to acute insulin challenge was shifted
to the right by an order of magnitude (Figure 3-1 C). This
is the first demonstration of "true" insulin-resistant
glucose transport activity in 3T3-L1 adipocytes because of
the ability to reestablish the basal state.
To determine if physiological insulin could establish
the insulin-resistant state, cells were exposed to specific
concentrations of insulin for 12 h. Cells were refed every
2 h to maintain the extracellular insulin levels,
particularly important at the lower concentrations of
insulin as these cells possess the ability to rapidly
degrade insulin. As shown in Figure 3-3 A, the
concentration of insulin that elicits a 50% reduction in
insulin-sensitive glucose transport was approximately 100
pM. This is extremely interesting because the fasting level
of insulin in non-diabetic humans is about 40 pM while that
in obese individuals is about 70 pM and individuals with
non-insulin-dependent diabetes have insulin levels around
200 pM (84). Insulin as low as 1 nM was sufficient to
completely desensitize the transport system. At 10 nM,
maximal desensitization was achieved whether the cells were
refed every 2 h or not. Thus, we chose a concentration of
10 nM to examine the time required for the development of
insulin resistance. As shown in Figure 3-3 B, the
phenomenon of desensitization was completely established

Figure 3-3. Development of Insulin Resistance. Panel A,
cells were incubated for 12 h with specific concentrations
of insulin as indicated. During this incubation, the
medium was replaced every 2 h. The cells were then washed
for 60 min and glucose transport activity determined
following acute (10 min) stimulation with 1 JJ.M insulin. The
"fractional difference" was determined by subtracting the
glucose uptake rate at 10 nM insulin from the glucose
uptake rate at each point divided by the difference in
uptake rates between 0 and 10 nM insulin. Panel B, cells
were incubated with 10 nM insulin for specific times.
Medium was replaced every 2 h. At appropriate times, the
cells were washed and acutely stimulated with insulin, and
glucose transport activity was measured. The fractional
difference in activity was determined as in panel A. Data
represent the average S.E. of three independent
experiments (n=6).

Glucose Transport
(fractional difference)
o o o o o
O M 4^ CD 00 O
Insulin Concentration (nM)
>
Glucose Transport
(fractional difference)
o o o o o
o hJ 4^ CD oo o

58
within 8 h of the initial exposure to insulin. This result
is similar to that described in isolated adipocytes (21).
Role of Glucose and Glucosamine
Marshall and his colleagues (15,20,21) have shown in a
series of elegant experiments the requirement of glucose and
glutamine, as well as insulin, for the expression of insulin
resistance in isolated rat adipocytes implicating the N~
acetylglucosamine biosynthetic pathway in this phenomenon.
To test if the same is true in 3T3-L1- adipocytes, I
performed similar experiments. One complication in my
experiments that was not encountered in isolated adipocytes
is the time-dependent activation of glucose transport
activity in the absence of glucose (85-88). This difference
between rat adipocytes and 3T3-L1 adipocytes may result from
higher glycogen stores in the former (89,90) compared with
3T3-L1 adipocytes (79), which provides a metabolic buffer
from external glucose deprivation. We therefore minimized
the time that the cells were exposed to glucose-free medium
but suffered in that only 75% of maximal desensitization was
achieved in these experiments. Importantly, though, the
basal rates of transport were not affected such that true
resistance could be evaluated. Figure 3-4 shows that in the
absence of glucose, insulin was unable to induce
desensitization. Either glucose, in the presence of

Figure 3-4. Effects of Glucose and Glucosamine on Insulin
Resistance. Panel A, cells were incubated for 6 h in DMEM
containing specific concentrations of glucose in the
absence (control) or presence (chronic) of 10 nM insulin.
Cells were washed and glucose transport activity was
determined in the presence of 1 p,M insulin. Panel B, cells
were incubated for 6 h in glucose-free and glutamine-free
DMEM containing specific concentrations of glucosamine in
the absence (control) or presence (chronic) of 10 nM
insulin. Washes were performed on the cells as described
earlier, and glucose transport activity following acute
stimulation with 1 fiM insulin was determined. Data
represent the average S.E. of two independent experiments
(n=4). Basal glucose transport activity in control and
glucose-deprived cells was 0.167 0.02 and 0.167 0.01
nmol/106 cells/min, respectively.

60
%
Ph
'P
co
O
O
"bb,
X
8 08 I-
cn
] 2-Deo
(nmol/l
o o
k>
Control
B Chronically treated
K
i i i i
0
10 15 20
Glucose Concentration (mM)
B
CD
CD
GO
O
o
J2
"bb
X
o
CD
9
CN
K
GO
1.4
1.2
1.0
a
2 06
o 0.4
cn
0.0
0.0 0.5 1.0 1.5 2.0
Glucosamine Concentration (mM)

61
glutamine (Figure 3-4 A), or glucosamine, in the absence of
both glucose and glutamine (Figure 3-4 B) provided
appropriate substrate for the development of insulin
resistance.
Effects of Various Inhibitors on the Development of Insulin
Resistance
Inhibitors of DNA transcription, protein synthesis,
glycosylation and proteases were used to characterize the
development of insulin resistance in these cells. Cells
incubated for 12 h with 4 p,M actinomycin D, an inhibitor of
DNA transcription, exhibited an increase in basal glucose
transport activity and a reduction in insulin-stimulated
glucose transport in both control and chronically-treated
cells (Figure 3-5). The underlying mechanism for this
elevation is unknown. Given elevated basal activity, the
effect of actinomycin D on insulin-sensitive transport is
difficult to determine, although changes in the presence and
absence of acute insulin were fairly small. In contrast,
studies in isolated rat adipocytes (91) showed that
incubation with up to 200 nM actinomycin D for 18 h actually
decreased basal levels. However, in these studies,
actinomycin D prevented the development of insulin
resistance. It is believed that actinomycin D prevented
insulin resistance in these studies by inhibiting GFAT
activity, as insulin resistance induced by incubation with

62
acute insulin
actinomycin
cycloheximide
+ + +
+ +
+ +
control
+ + +
+ +
+ +
chronic
Figure 3-5. Effects of Actinomycin and Cycloheximide on
Insulin Resistance. 3T3-L1 adipocytes were incubated in the
presence or absence of 4 jj,M actinomycin D or 20 ¡J.M
cycloheximide also in the presence (chronic) or absence
(control) of 10 nM insulin for 12 h. Insulin was removed
from the cells by washing as described previously. Basal
and insulin-stimulated rates of glucose transport were
measured by incubating the cells in KRP with 1 (J.M insulin,
or not, for 10 min. The data represent the average S.E.
of two independent experiments (n=4).

63
glucosamine, which bypasses GFAT, could not be prevented by
actinomycin D (91). As Figure 3-5 shows, in 3T3-L1
adipocytes, 20 (j,M cycloheximide had no effect on basal
transport, but inhibited insulin-stimulated glucose
transport in both control and chronically-treated cells.
Therefore, cycloheximide did not appear to prevent insulin
resistance. Studies in isolated rat adipocytes also showed
that cycloheximide was unable to prevent insulin resistance
(92) .
To further test the involvement of the hexosamine
biosynthetic pathway in the development of insulin
resistance, the effects of nikkomycin Z on insulin-resistant
glucose transport activity were tested. Nikkomycin Z is a
peptide-nucleoside antibiotic that competitively inhibits
chitin synthase in fungi and insects due to its structural
similarity to UDP-N-acetylglucosamine (93). I hypothesized
that this compound might prevent the onset of insulin
resistance by competing for enzymes which use UDP-N-
acetylglucosamine in mammalian cells. As Figure 3-6 shows,
neither concentration of nikkomycin Z tested had an effect
on glucose transport. Although I have no evidence that
nikkomycin is taken up by these cells, the concentrations
used are in far excess of 0.1 [J.M, the Ki of nikkomycin for
chitin synthase (94).

64
3
P-: O'
& .s
0)
on
O
o
2
02
o
g 2
(D -th
9 I
W W
cn
acute insulin
20|iM Nikkomycin
200 |iM Nikkomycin
+ + +
+ +
+ +
i
control
+ + +
+ +
+ +
chronic
Figure 3-6. Effects of Nikkomycin Z on Insulin Resistance.
Cells were incubated in the presence or absence of specific
concentrations of nikkomycin Z also in the presence
(chronic) or absence (control) of 10 nM insulin for 12 h.
Plates were washed to remove insulin and basal and insulin-
stimulated rates of glucose transport were measured as
previously described. Data represent the average S.D. of
duplicate samples within a single experiment (n=2).

65
Because the hexosamine biosynthetic pathway provides
substrate for glycoprotein synthesis, I tested the effects
of tunicamycin on the development of insulin resistance. As
shown in Figure 3-7, 2.5 |j.g/mL tunicamycin, which inhibits
N-linked glycosylation, decreased the insulin-stimulated
rate of glucose transport in both control and chronically-
treated cells. This is similar to the effect of
cycloheximide and again makes it difficult to determine if
tunicamycin was able to prevent the decrease in glucose
transport in insulin-resistant cells.
It has been hypothesized by Knutson et al. (95) that a
proteolytic fragment of the (3-subunit of the insulin
receptor is involved in the development of insulin
resistance. To test this, I used a thiol protease inhibitor
(E-64) to prevent cleavage of the (3-subunit (95) As is
shown in Figure 3-8, 100 (J.M E-64 had no effect on glucose
transport activity in control or chronically-treated cells.
As the Knutson group showed the largest effect of E-64 on
glycogen synthesis, I also investigated this insulin-
sensitive process. As Figure 3-9 shows, 100 (J.M E-64 had no
effect on rates of glycogen synthesis in control or insulin-
resistant cells.

66
acute insulin
tunicamycin
+
+
+
+ +
J L
+
+ +
l
Control
Chronic
Figure 3-7. Effects of Tunicamycin on Insulin Resistance.
Cells were incubated in the presence or absence of 2.5 p,g/mL
tunicamycin also in the presence (chronic) or absence
(control) of 10 nM insulin for 12 h. Cells were washed to
remove insulin and basal and insulin-stimulated rates of
glucose transport were measured as previously described.
Data represent the average + S.D. of duplicate samples
within a single experiment (n=2).

67
acute insulin
E-64
+ + + +
+ + + +
control chronic
Figure 3-8. Effects of E-64 on Insulin Resistance. Cells
were incubated in the presence or absence of 100 pM E-64
also in the presence (chronic) or absence (control) of 10 nM
insulin for 12 h. Cells were then washed to remove insulin
and basal and insulin-stimulated rates of glucose transport
were determined as previously described. Data represent the
average S.E. of two independent experiments (n=4).

68
chronic insulin
E-64
+ + + +
+ + + +
Figure 3-9. Effects of E-64 on Glycogen Synthesis. Cells
were incubated in the presence or absence of 100 p.M E-64
10 nM insulin for 12 h. Plates were then washed to remove
insulin and labeled with 14C-U-[D-glucose] (2 |j,Ci/plate) in
KRP containing 5 mM glucose in the presence or absence of 1
(j.M insulin for 1 h. Glycogen was then collected as
described and radioactivity counted by liquid scintillation
Data represent a single experiment.

69
Effect of Chronic Insulin on GLUT4 Expression and
Translocation
A total membrane fraction revealed that insulin-
resistant cells (i.e. cells exposed to 10 nM insulin for 12
h) expressed 2.4-fold less GLUT4 than control cells while
GLUT1 increased by 2.2-fold (Figure 3-10). Another
important point to make as shown by Figure 3-10 is that
acute insulin stimulation did not alter the total levels of
GLUT1 or GLUT4. Only chronic insulin treatment affected the
total amounts of these proteins. To examine the subcellular
distribution of these changes, I used a subcellular
fractionation technique recently developed in this lab (70)
to isolate three membrane fractions: plasma membrane (PM),
low density membranes (LDM), and high density membranes
(HDM). The LDM consists primarily of the small endosomal
storage vesicles containing GLUT4 and the HDM consists
primarily of endoplasmic reticulum and Golgi. Figure 3-11
shows the distribution of GLUT4 and GLUT1 among these three
membrane fractions in control and insulin-resistant cells.
Each set went through the washout procedure prior to
membrane fractionation. Control cells, which were
stimulated acutely with 1 (J.M insulin, showed redistribution
of both GLUT4 and GLUTl; GLUT4 increased by about 6-fold in
the PM (Figure 3-11 A and B) while GLUTl increased by about
2-fold (Figure 3-11 C and D). These data are similar to

70
Acute Insulin
Chronic Insulin + +
GLUT1 +
GLUT4
Figure 3-10. Effects of Acute and Chronic Insulin on Total
GLUT Protein Levels. Equal protein of the PM, LDM, and HDM
fractions of Figure 3-11 were combined and subjected to SDS-
PAGE and transferred to nitrocellulose. Immunoblot
detection of GLUT1 and GLUT4 was carried out using C-
terminal specific antibodies. The protein-antibody complex
was visualized by enhanced chemiluminescence.

Figure 3-11. Subfractionation of Insulin-Resistant 3T3-L1
Adipocytes. Cells were treated for 12 h in the absence
(control) or presence (chronic) of 10 nM insulin and
subsequently washed. Following acute stimulation with 1 |iM
insulin, PM, LDM, and HDM were collected as described in
Chapter 2. SDS-polyacrylamide gels of equal protein (70 p,g)
transferred to nitrocellulose allowed immunoblot detection
of GLUT1 and GLUT4 using C-terminal specific antibodies.
The protein-antibody complex was visualized by enhanced
chemiluminescence. Bands were quantitated by video
densitometry. Panel A, immunoblot of membrane fractions
probed with anti-GLUT4 antibody; panel B, densitometry of
GLUT4 immunoblot; panel C, immunoblot of membrane fractions
probed with anti-GLUTl antibody; panel D, densitometry of
GLUT1 immunoblot. control; E23 control + acute insulin; ESI
chronic insulin treatment; chronic insulin treatment +
acute insulin. Data represent a single experiment. A
duplicate experiment gave similar results.

72
PM
Chronic Insulin + +
Acute Insulin + +
LDM HDM
+ + + +
+ + + +
GLUT4 -
B GLUT4 Levels

Chronic Insulin
Acute Insulin
D GLUT1 Levels
05

74
those that analyzed translocation using the cell surface
photolabel, ATB [2-3H]BMPA [2-N-4-(l-azi-2,2,2-
trifluoroethyl)benzoyl-1,3-bis(D-mannose-4-yloxy)-2-
propylamine] (16). Cells chronically exposed to insulin
(followed by washout) showed levels of GLUT4 in the plasma
membrane equivalent to that of controls (Figure 3-11 A and
B). While acute insulin challenge stimulated translocation,
the amount of GLUT4 was reduced by about 50% compared with
controls. This reflects the down-regulation of GLUT4 and
correlates well with the loss of insulin-stimulated glucose
transport activity. Cells chronically exposed to 10 nM
insulin showed a 2-fold increase in the level of GLUT1 in
the PM after washout compared with controls (Figure 3-11 C
and D), despite the equivalent rates of glucose transport.
Acute insulin challenge stimulated translocation but to a
much more limited degree than in control cells.
As the LDM fraction reflects the loss of GLUT4, we used
this fraction to examine the dose-dependent loss in cells
chronically treated with specific concentrations of insulin.
As shown in Figure 3-12 A, the level of GLUT4 decreased over
time in response to increasing insulin. Figure 3-13 shows a
similar experiment with a larger sample (5, 10-cm plates per
condition compared to 1, 10-cm plate). This larger sample
allows for the somewhat more accurate quantitation of the
LDM fraction.
The numbers listed below the blot in Figure

75
A
Insulin
[nM]
GLUT 4
B
G
'53
o
CL,
^t

D
J
O
Insulin [nM]
Figure 3-12. Effects of Chronic Insulin on GLUT4
Expression. Panel A, cells were incubated with specific
concentrations of insulin for 12 h (fed with fresh medium
every 2 h). Cells were then washed for 60 min and
subfractionated to isolate the LDM fraction. Proteins were
separated by SDS-polyacrylamide gel electrophoresis and
subsequently transferred to nitrocellulose membrane (70 [ig
of protein were loaded per lane). The membrane was then
probed for GLUT4 and visualized by enhanced
chemiluminescence. Panel B, GLUT4 bands were quantitated by
densitometry. Fractional difference in GLUT4 expression was
calculated from five independent experiments performed as in
panel A. Data represent the average + S.E.

76
Insulin [nM]
\ O C\
^ Qy V -V3
GLUT4
100 93 67 37
Figure 3-13. Effects of Chronic Insulin on GLUT4 Expression
with Larger Samples. Cells were incubated as described in
Figure 3-12 with the exception that 5, 10-cm plates were
used per condition compared to 1, 10-cm plate in Figure 3-
12. The LDM fraction was collected and treated as described
in Figure 3-12. The numbers listed below the figure are the
percentages of GLUT4 present in each sample compared to the
0 nM insulin control based on densitometric analysis.

77
3-13 are the percentages of GLUT4 in each sample compared to
the 0 nM insulin control as determined by densitometric
analysis. The dose dependence of this down-regulation
(Figure 3-12 B and Figure 3-13) yielded a K50 of about
600pM, slightly higher than the K50 of insulin-resistant
glucose transport, but a technically more difficult
parameter to measure.
Effect of Glucose Deprivation on GLUT4 Expression
Based on the observation that glucose deprivation
prevented the loss in insulin sensitivity (see Figure 3-4
A), we examined the expression of GLUT4 in the LDM fraction
of cells exposed to glucose-free medium. Figure 3-14 shows
that glucose deprivation blocked the loss of GLUT4 in
chronically treated cells. Thus, this shows for the first
time that glucose is important in regulating the expression
of GLUT4 in response to chronic insulin.
Effect of Chronic Insulin on GLUT4 mRNA Expression
These studies were conducted by Martin Williams to
evaluate the underlying mechanism of the reduction in GLUT4
protein. The level of GLUT4 mRNA was measured after
exposure to specific concentrations of insulin. As shown in
Figure 3-15 A, the level of GLUT4 mRNA decreases with
increasing insulin concentration. However, the

78
Glucose + +
Chronic Insulin +
+
GLUT4
Figure 3-14. Effect of Glucose Deprivation on GLUT4
Expression. Cells were maintained in medium for 12 h in the
absence or presence of 10 nM insulin and/or 25 mM glucose.
The LDM fraction was isolated and GLUT4 analyzed by
immunoblot analysis. Data are representative of three
independent experiments.

79
Insulin
[nM] O Q*' o \v V* ^ ^
GLUT 4
,7 ^ 1? $> S> $> nO
Actin
B
§
P
o
Figure 3-15. Effects of Chronic Insulin Treatment on GLUT4
mRNA Levels. Panel A, cells were incubated for 12 h in the
presence of specific concentrations of insulin (refed every
2 h). Cells were then washed three times with 8 mL KRP at
which time RNA was extracted using the phenol:chloroform
extraction method. Twenty (ig of total RNA were loaded onto
a 1% formaldehyde-agarose gel and subsequently transferred
to a nylon membrane. The membranes were probed with a 32P-
labeled cDNA for GLUT4. Panel B, densitometric analysis
represented as fractional difference in sensitivity. Data
shown represent a single experiment replicated three times.

80
concentration of insulin required to elicit a 50% loss in
GLUT4 mRNA was about 5 nM (Figure 3-15 B), which is 15 times
greater than that required for the equivalent loss of
insulin-sensitive glucose transport activity and 10-times
that for GLUT4 expression.
Reversal of Insulin Resistance
To determine if the effects of chronic insulin on 3T3-
L1 adipocytes could be reversed, insulin-resistant cells
were incubated with complete medium following the washout
procedure. Both basal and insulin-stimulated rates of
glucose transport were measured at specific times. Eight
hours after insulin removal, glucose transport activity of
previously insulin-resistant cells was equal to that of
controls (Figure 3-16). An interesting point is that the
reversal appears to occur between 4-8 h, as no improvement
in insulin-sensitive transport occurs before that time.
Also, the rate of basal transport between control cells and
resistant cells is identical over this time while only the
insulin-stimulated rates are changing. This shows that the
increase in insulin-stimulated transport seen in resistant
cells over this time is not due to increasing basal rates,
but improved insulin sensitivity.
To determine if this reversal in insulin resistance
required protein synthesis, cycloheximide was included in

81
Time (h)
Figure 3-16. Reversal of Insulin Resistance. Cells were
incubated in the presence (chronic) or absence (control) of
10 nM insulin for 12 h. Plates were washed to remove
insulin and refed with fresh medium lacking insulin. At
specific times, basal and insulin-stimulated rates of
glucose transport were determined. Data represent the
average S.E. of two independent experiments (n=4).
control; control + acute insulin; A chronic; chronic +
acute insulin.

82
this experiment. As is shown in Figure 3-17, cycloheximide
completely blocked the reversal of insulin-resistant glucose
transport activity. It should be noted that cycloheximide
decreased the insulin-stimulated rate of glucose transport
in control cells (closed squares versus inverted triangles).
However, the cycloheximide-induced block in reversal is
considerably larger (open squares versus inverted
triangles). This
suggests that the reversal of insulin-
resistant glucose
transport is dependent upon protein
synthesis. It is
not clear, however, if this reversal is
due to the synthesis and recovery of GLUT4, as those studies
have not yielded reliable results.
Conclusions
In the above
studies, I have tested the hypothesis that
3T3-L1 adipocytes
can serve as a model for studying the
development of insulin resistance under conditions that
might be realized in a physiological setting. Support for
this hypothesis has been gained from the following
observations. Chronic exposure to physiological levels of
insulin decreased the ability of an insulin challenge to
stimulate glucose transport. Interestingly, postprandial
concentrations of insulin in normal, obese, and diabetic
humans (84) plot along the inflection in the dose-response
curve between no change in insulin responsiveness and that

83
Time (h)
Control
Control + acute ins
Control + cyclo
Control + cyclo + acute ins
--
Chronic
e-
Chronic + acute ins
-a-
Chronic + cyclo
Chronic + cyclo + acute ins
Figure 3-17. Reversal of Insulin Resistance in the Presence
of Cycloheximide. Cells were incubated and washed as
described in the previous figure. Plates were then refed
with fresh medium 20 |iM cycloheximide. At specific times,
basal and insulin-stimulated rates of glucose transport were
determined. Data represent the average S.D. of duplicate
samples within a single experiment (n=2).

84
of maximal resistance. Thus, I have shown for the first
time that 3T3-L1 adipocytes develop insulin resistance in
response to physiologically relevant concentrations of
insulin. I have extended previous work by demonstrating
that insulin challenge of resistant cells stimulates
translocation, although the extent of recruitment is
suppressed relative to controls due to the reduction in the
total expression of GLUT4. Together, these data suggest
that the loss of GLUT4 protein underlies the inability of
3T3-L1 adipocytes to respond to insulin after chronic
exposure. This mimics the clinical manifestation of human
obesity and non-insulin-dependent diabetes where loss of
GLUT4 protein has been observed in adipose tissue (96),
although not in muscle (97).
It should be pointed out that transporter expression
differs in adipose tissue relative to 3T3-L1 adipocytes. In
isolated rat adipocytes, GLUT4 represents 97% of the GLUT
transporter pool (43). In 3T3-L1 adipocytes, GLUT4
represents only 33% of the pool (98), indicating the
substantially higher expression of GLUTl relative to GLUT4
in this cell line. In control 3T3-L1 adipocytes, the PM
fraction contains about 25% of the GLUTl pool. Chronic
insulin treatment increases the total pool of GLUTl, which
in turn doubles the GLUTl content of the PM fraction.
Despite this 2-fold increase in GLUTl in the PM of resistant

85
cells, no difference is observed in basal transport activity
(after washout) compared with controls. Resistant cells
treated acutely with insulin show little additional change
in GLUT1 in the PM. This argues that GLUT1 plays but a
small role in insulin-resistant glucose transport. In
contrast, only 3% of the GLUT4 pool resides in the PM of
either control or resistant cells (again, after washout).
When insulin is added acutely, GLUT4 content in the PM
reveals significant translocation; resistant cells show 50%
less GLUT4 following translocation compared to controls
reflecting the difference in the total pool. Again, this
suggests that GLUT4 expression determines insulin
resistance, even with the elevated levels of GLUT1 in the
3T3-L1 adipocyte cell line.
There are some differences between these data and those
reported previously on 3T3-L1 adipocytes. Flores-Riveros et
al. (14) reported that the concentration of insulin required
to induce a 50% change in GLUT4 mRNA was 23 nM in contrast
to the value I calculated, which was about 5 nM, close to
the Kd for the insulin receptor (99). This difference can
be explained by my refeeding protocol during chronic insulin
treatment to maintain the level of extracellular insulin,
particularly important at low hormone concentration, in the
face of extensive degradation of insulin by these cells. In
addition, cells in this study were exposed to insulin for

86
only 12 h compared to the 24-h exposure in the Flores-
Riveros study (14), which further lessens the impact of
insulin degradation. This temporal difference (24 versus 12
h of insulin treatment) also accounts for the smaller
magnitude of the increase in GLUT1 and decrease in GLUT4 in
our study relative to previous studies (14,76).
Importantly, the down-regulation of GLUT4 mRNA occurs at
insulin concentrations that are not likely to persist in the
physiological state. These insulin concentrations also do
not correlate with those required for the development of
insulin-resistant glucose transport or decrease in GLUT4
protein. Thus, GLUT4 expression appears to be regulated
transcriptionally, but this regulation may not be relevant
to insulin resistance.
Other studies in 3T3-L1 adipocytes have shown varying
results in transporter protein expression. Tordjman et al.
(100) and Koska et al. (16) showed that chronic insulin
treatment did not affect total GLUT4 protein expression
while studies of Flores-Riveros et al. (14) and Clancy and
Czech (76) showed a marked decrease. These latter data along
with my studies are consistent with the accelerated turnover
of GLUT4 in the presence of chronic insulin as measured by
Sargeant and Paquet (101), which will be discussed in the
next chapter. Ricort et al. (64,102) showed a small
decrease in the expression of GLUT4 but also very little

87
translocation to the PM with acute insulin stimulation.
These authors interpreted their data to mean that GLUT4
translocation was blocked, which clearly differs from our
studies. Kozka et al. (16) interpreted their cell surface
ligand binding experiments similarly, even though they
demonstrated a 50% reduction in cell surface GLUT4, which
would agree with my studies. I can only speculate as to the
cause for the different results. In both of these latter
studies, the loss in the GLUT4 pool was determined by
analyzing homogenate protein, revealing only modest changes
in expression. As the translocatable GLUT4 resides in the
LDM fraction, it may be that the loss was substantially
underestimated. Neither study separated the LDM fraction
from the HDM fraction; thus, this possibility can not be
evaluated. Finally, it is important to point out that my
experiments are the first to show that basal transport
activity can be achieved after chronic insulin treatment,
which allowed me to evaluate true insulin resistance. Data
collected under these conditions are consistent with the
hypothesis that the onset of insulin resistance (i.e.
depressed insulin-sensitive glucose transport) is a
reflection of the reduced GLUT4 pool, not a defect in
translocation. In addition, this decrease in insulin-
sensitive glucose transport can be reversed within 8 h and
is protein synthesis dependent.

CHAPTER 4
MECHANISMS OF DECREASED GLUT4 LEVELS
IN INSULIN-RESISTANT CELLS
Introduction
As described in Chapter 3, I showed that a decrease in
GLUT4 protein was responsible for the decrease in insulin-
sensitive glucose transport activity. Reduced levels of
GLUT4 could be explained by one of several mechanisms. One
possibility is that GLUT4 mRNA levels are decreased leading
directly to reduced GLUT4 protein. While I have shown this
to be true, the K5o for the decrease in GLUT4 mRNA is 15
times greater than the concentration of insulin required for
the loss in insulin-sensitive glucose transport activity and
10 times greater than that for GLUT4 expression. A second
possibility is that the synthesis of GLUT4 protein is
decreased under chronic insulin conditions, despite
equivalent mRNA template. In this case, reduced
translational efficiency would lead to a diminished GLUT4
pool. A third possibility is that the degradation of GLUT4
is accelerated under these conditions again decreasing the
cellular levels of GLUT4 protein. These latter two
88

89
possibilities will be investigated in more detail in the
studies described in this chapter.
In addition to investigating the synthesis and
degradation of GLUT4, the degradative pathway responsible
for GLUT4 degradation was also examined. Two major
degradative pathways exist in mammalian cells, the lysosomal
pathway and the ATP-dependent proteasome pathway (103,104).
Several inhibitors exist to distinguish between these
pathways. Chloroquine and leupeptin inhibit the lysosomal
pathway. Chloroquine, an acidotropic amine, exerts its
effects by accumulating in and deacidifying lysosomes and
late endosomes (105) Another acidotropic amine, which
works in much the same way as chloroquine is ammonium
chloride. Leupeptin is a competitive inhibitor of the
lysosomal serine proteases cathepsins B, H, and L (106,107).
Lactacystin, a Streptomyces metabolite isolated by Omura et
al. (108), specifically inhibits the proteasome (109-111).
I also used MG132 (carbobenzoxyl-leucinyl-leucinyl-leucinal)
which, like lactacystin, inhibits protein degradation by the
proteasome (112,113).
Results
Specificity and Efficiency of GLUT4 Immunoprecipitation
Before the synthesis and degradation of GLUT4 could be
investigated, the immunoprecipitation procedure for

90
isolating GLUT4 had to be optimized. The first step was to
determine the appropriate amount of peptide-purified GLUT4
antibody required for the complete precipitation of GLUT4.
Figure 4-1 shows that 5 pig of antibody is the optimal amount
for maximal GLUT4 recovery.
It was also necessary to prove that the protein that
was being precipitated by the antibody was indeed GLUT4.
Figure 4-2 addresses this question by showing that the
protein precipitated could be competed by an equal molar
amount of GLUT4 peptide (compare lanes 3 and 4). This
proves that the peptide-purified GLUT4 antibody is
precipitating GLUT4 and that this interaction is specific.
Finally, the last step in optimizing the
immunoprecipitation procedure was to show that a single
"round" of immunoprecipitation could precipitate all of the
GLT4 in the sample. Figure 4-3 shows that a successive
round of immunoprecipitation did not yield significant
amounts of GLUT4 (see lanes 3 and 4). This proves that one
round of immunoprecipitation is sufficient to isolate 98% of
the GLUT4 in the sample.
Synthesis of GLUT4 in Insulin-resistant Cells
The synthesis of GLUT4 in control and insulin-resistant
cells was examined to determine if a reduced synthesis in
resistant cells could explain the reduced levels of GLUT4

91
|ig GLUT4 Ab 0 1 2.5 5 10
GLUT4
Figure 4-1. Titration of Peptide-Purified GLUT4 Antibody
for Immunoprecipitation. Cells were incubated in DMEM
lacking cysteine and methionine and FBS for 1 h. Fresh
DMEM, supplemented with [35S] cysteine/methionine (400 |J.Ci)
was added and the cells incubated for 3 h. A total membrane
fraction was then isolated and GLUT4 was immunoprecipitated
from this fraction using specific amounts of peptide-
purified GLUT4 antibody. Proteins released from the immune
complex were subjected to SDS-PAGE and the radiolabeled
proteins were visualized by autoradiography. The graph
depicts the densitometric analysis of the autoradiogram.
The data represent a single experiment.

92
GLUT4
12 3 4
Figure 4-2. Specificity of the Peptide-Purified GLUT4
Antibody in Immunoprecipitation. Cells were incubated and
radiolabeled as described in Figure 4-1. Total membranes
were collected and GLUT4 was immunoprecipitated under the
following conditions: lane 1, no GLUT4 antibody added;
lane 2, 5 pig of total IgG from pre-immune serum added; lane
3, 5 |ig of peptide-purified GLUT4 antibody competed with
molar equivalent of GLUT4 peptide added; and lane 4, 5 pig
of peptide-purified GLUT4 antibody added. These data
represent a single experiment.

93
GLUT4
12 3 4
Figure 4-3. Efficiency of GLUT4 Immunoprecipitation. Total
membranes were collected from duplicate plates (A and B) of
cells. Immunoprecipitation was performed using 5 pig of
peptide-purified GLUT4 antibody (1st round). The
supernatant was saved from this immunoprecipitation and
immunoprecipitation was performed again (2nd round) on this
supernatant. Proteins were resolved by SDS-PAGE and
transferred to nitrocellulose for western blot analysis of
GLUT4. Lanes 1 and 2, immunoprecipitated GLUT4 from the
initial procedure. Lanes 3 and 4, GLUT4 immunoprecipitated
from the supernatant of the first immunoprecipitation. The
graph shows the densitometric analysis of the immunoblot.
These data represent a single experiment.

94
present in these cells. Following the onset of insulin
resistance, cells were radiolabeled with
[35S]cysteine/methionine for specific times and a total
membrane fraction collected. GLUT4 was then
immunoprecipitated from this membrane fraction and samples
subjected to SDS-PAGE and autoradiography. As Figure 4-4
shows, at each time point, less GLUT4 was synthesized in the
insulin-resistant cells compared to the controls. However,
the synthesis in these cells is not reduced by 50% as seen
by the densitometric analysis. Therefore, this suggests
that the reduction in the total pool of GLUT4 in insulin-
resistant cells is only in part due to reduced synthesis of
GLUT4.
Degradation of GLUT4 in Insulin-resistant Cells
The degradation of GLUT4 under conditions of chronic
insulin was then examined to determine if accelerated
degradation of GLUT4 in insulin-resistant cells could
contribute to the reduced pool in these cells. In order to
achieve equal incorporation of [35S]cysteine/methionine into
both control and resistant cells, cells were radiolabeled
prior to incubation with insulin. Cells were then chased
with complete medium in the presence or absence of insulin
for up to 72 h. Total membranes were collected at specific
times and GLUT4 immunoprecipitated as above. Figure 4-5

Figure 4-4. Effect of Chronic Insulin on the Synthesis of
GLUT4. Panel A, cells were incubated in DMEM containing
10% FBS in the presence (chronic) or absence (control) of
10 nM insulin for 11 h. Cells were then incubated in DMEM
lacking methionine and cysteine and FBS for 1 h, also in
the presence or absence of 10 nM insulin. Fresh DMEM,
supplemented with [35S] cysteine/methionine (400 (j,Ci) and
10 nM insulin, was then added and the cells incubated for
specific times. A total membrane fraction was then
isolated and GLUT4 immunoprecipitated from this fraction.
Immunoprecipitated proteins were then subjected to SDS-PAGE
and radiolabeled proteins visualized by autoradiography.
pGLUT4 represents precursor GLUT4 and mGLUT4, mature GLUT4,
based on its glycosylation state. Panel B, densitometric
analysis was performed. The values represent total GLUT4,
both pGLUT4 and mGLUT4. The data represent the average
S.D. of three independent experiments. 9 control;
chronic insulin.

96
A
Time (min) 10
20
30
45
60
Chronic Insulin
r
i r
i r
i r
mGLUT4
pGLUT4
B
Time (min)

Figure 4-5. Effect of Chronic Insulin on the Degradation
of GLUT4. Panel A, cells were incubated in DMEM lacking
cysteine and methionine and FBS for 1 h. DMEM,
supplemented with [35S] cysteine/methionine (400 (j,Ci) was
added and the cells incubated for 3 h. Cells were then
incubated in complete DMEM containing 10% FBS 10 nM
insulin for specific times. The medium was refreshed every
12 h and at the indicated times, total membranes were
collected and GLUT4 immunoprecipitated from equal volume
rather that equal protein. Immunoprecipitated GLUT4 was
treated as described in previous figures. Panel B,
densitometric analysis. The data represent the average of
three independent experiments. control; chronic
insulin.

98
A
B

99
shows that the degradation of GLUT4 was accelerated in
insulin-resistant cells. In fact, the half-life of GLUT4
under these conditions decreases from 29 h to 16 h in the
presence of chronic insulin. Figure 4-6 shows that this
increased degradation of GLUT4 was specific as the
degradation of total protein was not different under these
conditions.
Effects of Cycloheximide on the Loss of GLUT4
To determine if the loss of GLUT4 had a protein
synthesis component, the effects of cycloheximide on the
loss of GLUT4 were examined. Figure 4-7 shows that 20 p,M
cycloheximide completely blocked the loss in GLUT4 seen with
chronic insulin treatment. This shows that the loss in
GLUT4 was dependent on new protein synthesis. Most likely,
cycloheximide blocks the synthesis of a protease specific
for degrading GLUT4, or it blocks the synthesis of an
activator of a preexisting protease which is responsible for
the degradation of GLUT4.
Effects of Protease Inhibitors on GLUT4 Expression
The above result led me to consider whether GLUT4
degradation was sensitive to various protease inhibitors
which would aid in defining the route of degradation.
Therefore, two lysosomal degradative inhibitors, chloroquine

Total Membrane
Protein Degradation
100
Figure 4-6. Degradation of Total Protein in Cells Treated
with Chronic Insulin. Total radioactivity and total protein
were determined from membrane samples collected in Figure 4-
5. Total radioactivity per pig of protein was then
calculated. control; chronic insulin.

101
PM LDM
Insulin + + + +
Cydohieximide + + + +
GLUT1
GS.UT4
* no wash out
Figure 4-7. Effect of Cycloheximide on the Loss of GLUT4.
Cells were incubated in DMEM containing 10 % FBS and 10 nM
insulin and 20 )j,M cycloheximide for 12 h. Insulin was not
removed by washing, membrane fractions were immediately
collected. Proteins were subjected to SDS-PAGE and
transferred to nitrocellulose for western blot analysis of
GLUT1 and GLUT4. The data are representative of three
independent experiments.

102
and leupeptin, and the two proteasome inhibitors,
lactacystin and MG132, were studied. As Figure 4-8 shows,
none of these inhibitors, when incubated with the cells in
the presence of chronic insulin, seemed to prevent the loss
of GLUT4 seen in the LDM fraction. Figure 4-8 A shows the
immunoblot of an experiment testing the effects of
leupeptin. This is representative of the immunoblots
obtained from the other inhibitors tested. The results of
these experiments are displayed in Figure 4-8 B which shows
the densitometric data from the LDM fractions of these
experiments normalized to the control (minus inhibitor and
chronic insulin).
Conclusions
I have shown with these studies that the
immunoprecipitation protocol is specific for GLUT4 and
optimized for complete precipitation from a given extract
with a single round of immunoprecipitation. Using this
technique, I showed that the synthesis of GLUT4 following
chronic insulin treatment was reduced and the degradation
was accelerated in the presence of chronic insulin. This
suggests that the reduction in the total pool of GLUT4 seen
in insulin-resistant cells is a combination of both reduced
synthesis and accelerated degradation.

Figure 4-8. Effects of Protease Inhibitors on GLUT4
Expression. Panel A, cells were incubated 10 nM insulin
and 100 pM leupeptin for 12 h. Insulin was removed by
washing the cells and membrane fractions collected.
Proteins were subjected to SDS-PAGE and transferred to
nitrocellulose for western blot analysis of GLUTl and
GLUT4. Panel B, cells were treated or not with 10 nM
insulin for 12 h in the presence or absence of 50 p.M
chloroquine, 100 |J,M leupeptin, 10 (J.M lactacystin, or 10 (J.M
MG132. Cells were washed and membrane fractions collected.
Shown is the densitometric analysis of the GLUT4 bands in
the LDM fractions expressed relative to the control.
control; ggg control + inhibitor; chronic insulin; g-g
chronic + inhibitor.

104
A PM LDM
Insulin
Leupeptin
GLUT1 *
+ + + +
+ + + +
GLUT4
3

105
Preliminary results with cycloheximide suggested that
the synthesis or activation of a protease was responsible
for the loss of GLUT4 in the LDM fraction of insulin-
resistant cells. This hypothesis was pursued by attempting
to identify a pathway responsible for GLUT4 degradation.
However, experiments using inhibitors specific for the
lysosomal and proteasomal degradative pathways revealed no
clues as to the route of degradation. In fact, these data
were negative despite the metabolic labeling data which
clearly shows an accelerated degradation of GLUT4 in the
presence of insulin. There are several possible
explanations for these negative results. For example,
leupeptin inhibits cathepsin L activity by only 50% (107)
and inhibits overall protein degradation in the lysosomes by
only 45% (114). Ammonium chloride also inhibits protein
degradation by only 45% with no additive effects obtained
with chloroquine (114). However, an additive effect is seen
with the addition of leupeptin, albeit small (<20%) (114).
Therefore, these inhibitors may exert a small effect in this
system, but one which cannot be measured. Alternatively, it
is possible that the inhibition of one protein degradative
pathway is compensated for by another, even though it may
not be the preferred pathway for degrading a particular
protein. Finally, there is evidence that an aminopeptidase
colocalizes with GLUT4 containing vesicles in vivo (115-

106
117). Activation of this aminopeptidase or a yet unknown
protease which colocalizes with GLUT4 may be responsible for
its accelerated degradation.

CHAPTER 5
ROLE OF THE
HEXOSAMINE BIOSYNTHETIC PATHWAY
IN THE DEVELOPMENT OF INSULIN RESISTANCE
Introduction
Marshall and his colleagues showed that glucose and
glutamine were required for the insulin-dependent
development of insulin-resistant glucose transport activity
in isolated rat adipocytes (15,20,21) As shown in Chapter
3, I have confirmed this in the 3T3-L1 adipocyte cell line
(118). Because glucose (via its metabolism to fructose 6-
phosphate) and glutamine are co-substrates of glutamine,
fructose 6-phosphate amidotransferase (GFAT), this
implicates the intermediates or products of the hexosamine
biosynthetic pathway in the development of insulin
resistance. The hexosamine biosynthetic pathway is
responsible for the synthesis of N-acetylglucosamine, one of
the sugars used in the post-translational modification of
proteins.
Several lines of evidence support the Marshall
hypothesis. First, glucosamine appears to desensitize
adipocytes to the actions of insulin at lower concentrations
than does glucose in both isolated adipocytes (18)
107

108
and 3T3-L1 adipocytes (118). Recall that glucosamine 6-
phosphate, the product of GFAT, can also be produced by
hexokinase-mediated phosphorylation of glucosamine,
functionally bypassing GFAT. In addition, overexpression of
GFAT in fibroblasts inhibits insulin-sensitive glycogen
synthase activity (119) Further, glucose deprivation in
the presence of chronic insulin prevents both the loss of
GLUT4 (118) and the development of insulin-resistant glucose
transport (15,118).
There is also evidence to suggest that intermediates or
products of the hexosamine biosynthetic pathway are not
involved in the development of insulin resistance. Hresko
et al. (120) reported that glucosamine in the presence of
insulin depletes intracellular ATP in 3T3-L1 adipocytes,
suggesting that glucosamine 6-P accumulates preventing
phosphate recycling. This leads to an inability of the
cells to respond to insulin, which normally initiates an
ATP-dependent kinase cascade. ATP depletion would also
prevent fusion of intracellular GLUT4 vesicles with the
plasma membrane, as demonstrated by investigators in the
mid-80's (121,122). Thus, glucosamine affects metabolism
independently from its flux through the hexosamine pathway.
A second study by Garvey et al. (15) showed that rat
adipocytes incubated with fructose in the presence of
insulin did not exhibit insulin-resistant glucose transport

109
activity. While one could argue that adipocytes metabolize
fructose at a slower rate than glucose, early studies showed
that fructose uptake and utilization in adipose tissue was
nearly equivalent to that of glucose (123) Further,
Kitzman et al. (85) has shown that fructose prevents
aberrant protein glycosylation in response to glucose
deprivation which demonstrates that fructose is metabolized
specifically by the hexosamine pathway. Thus, three
possibilities arise. The specific presence of fructose
might prevent the down-regulation of GLUT4. Alternatively,
fructose may support the insulin-dependent, down-regulation
of GLUT4, but change the relative distribution of GLUT4
between the plasma membrane and the endosomal fraction
(LDM). Finally, GLUT4 might be down-regulated but exhibit
"activation" at the cell surface. The data in this chapter
will show that fructose, in the presence of chronic insulin,
does in fact induce down-regulation of GLUT4. However,
GLUT4 translocation is enhanced in response to an acute
insulin challenge leading to equivalent amounts of GLUT4 on
the plasma membrane as compared to the controls. This
prevents the manifestation of insulin-resistant transport
activity despite the insulin-dependent, down-regulation of
GLUT4.
To compare these results to another insulin-sensitive
system, I examined the effect of the chronic treatment

110
paradigms on glycogen synthesis. Chronic exposure of cells
to fructose, alone, accelerated the rate of glycogen
synthesis significantly over cells exposed to glucose,
alone. This difference appears to be correlated, in part,
to the size of the glycogen pool. Subsequent exposure to
insulin stimulated synthesis in both sets of cells. With
the additional exposure to chronic insulin, insulin
resistance developed in the presence of either glucose or
fructose although to different degrees. The level of
insulin-stimulated glycogen synthesis in cells exposed to
glucose and chronic insulin fell 90% compared to the
controls, while the level of synthesis in cells exposed to
fructose and insulin fell to only 50% compared to controls.
This suggests that the development of insulin resistance in
the glycogen synthetic path, while observed in both glucose-
and fructose-treated cells, appears to be greater in cells
exposed to glucose and chronic insulin.
Results
Effects of Fructose Feeding on the Glucose Transport System
To examine the potential role of fructose in the
development of insulin-dependent, insulin resistance, I have
first examined its effect in transport activity assays
(Figure 5-1). In these experiments, 3T3-L1 adipocytes were
incubated in DMEM containing 25 mM glucose or 25 mM fructose

Figure 5-1. Effects of Fructose on the Development of
Insulin Resistance. Cells were incubated in DMEM
containing 10% FBS or in glucose-free DMEM containing 10%
dialyzed FBS and 25 mM fructose in the presence or absence
of 10 nM insulin for 6 h. The cells were then washed as
described in Chapter 2. Basal and insulin-simulated rates
of glucose transport were measured by incubating the cells
in KRP with 1 |o,M insulin, or not, for 10 min followed by the
addition of 200 |o.M [3H] 2-deoxyglucose (0.2 jnCi) After 10
min, transport was terminated by the addition of ice-cold
phosphate-buffered saline. Cells were lysed with a 0.1%
solution of SDS and duplicate aliquots of 300 (J.L were taken
and [3H]2-deoxyglucose uptake determined by liquid
scintillation counting. The data represent the average
S.E. of three independent experiments (n=3).

[ H] 2-Deoxyglucose Uptake
112
acute insulin + + + +
chronic insulin + + + +
glucose + + + +
fructose
+ + + +

113
in the presence or absence of 10 nM insulin for 6 h. While
it takes at least 8 h for these cells to exhibit full
insulin-resistant glucose transport, I chose 6 h in these
experiments to avoid the accelerated transport rates seen in
the absence of glucose (85). Following the washout
procedure to remove insulin, glucose transport activity was
measured in the presence or absence of acute insulin
stimulation. Glucose in the presence of chronic insulin
induced a 50% reduction in subsequent insulin-stimulated
glucose transport activity compared to glucose controls, as
expected (see Chapter 3). In agreement with Garvey et al.
(15), cells exposed to chronic fructose and insulin did not
show reduced acute insulin-stimulated glucose transport
activity.
To examine the expression of GLUT4, cells were treated
as in Figure 5-1 but for 12 h. Following washout, cells
were then subfractionated to obtain three membrane
fractions: plasma membrane (PM), low density membranes
(LDM), and high density membranes (HDM). Note that the LDM
fraction consists of the small intracellular vesicles
containing the translocatable pool of GLUT4 and the HDM
consists of membrane vesicles from Golgi and endoplasmic
reticulum (70). Figure 5-2 A shows the distribution of both
GLUT4 and GLUT1. In cells incubated in glucose and chronic
insulin, GLUT1 levels were increased, especially in the PM.

114
A
PM
LDM
HDM
Glucose
+ +
+ +
+ +
Fructose
+ +
+ +
+ +
Chronic Insulin
+ +
+ +
+ +
GLUT4 -

fi;
GLUT1
B GLUT4
Figure 5-2. Effect of Fructose on GLUT Transporter Levels.
Panel A, cells were incubated as described in Figure 5-1 for
12 h. Plates were then washed as described and the PM, LDM,
and HDM fractions collected. Proteins (50 pg) were then
separated by SDS-polyacrylamide electrophoresis, transferred
to nitrocellulose and probed with antibodies to either GLUTl
or GLUT4. The antibody-protein complex was then visualized
by enhanced chemiluminescence. Panel B, Bands were
quantitated by video densitometry. These data are
representative of three independent experiments. || glucose; ggg
glucose + chronic insulin; ES3 fructose; E3! fructose +
chronic insulin.

115
This is noteworthy in that transport activity in control and
chronically-treated cells after insulin washout was nearly
identical (see Figure 3-2 B and Figure 5-1). Similar
results were seen in cells treated with fructose and chronic
insulin. GLUT4 levels were decreased in the LDM fraction of
cells exposed to glucose and chronic insulin, as expected.
Surprisingly, GLUT4 was reduced to the same level in cells
treated with either glucose or fructose in the presence of
chronic insulin (Figure 5-2 A and B). Thus, despite normal
insulin-stimulated glucose transport in cells chronically
treated with insulin and fructose, GLUT4 levels dropped in
the endosomal fraction.
The apparent contradiction between GLUT4 expression and
insulin-sensitive transport activity led us to examine
translocation of GLUT4 from the LDM fraction to the PM.
Figure 5-3 shows the comparison in translocation between
cells treated with fructose and chronic insulin and control
cells, which were only treated with acute insulin. As with
glucose, cells exposed to fructose in the presence of
chronic insulin again showed reduced GLUT4 in the LDM
fraction. Remember, in the presence of acute insulin, the
PM isolated from cells incubated with glucose and chronic
insulin contained about half the GLUT4 as in controls (see
Figure 3-10). However, cells incubated with fructose and
chronic insulin contained the same amount of GLUT4 in the

116
A
PM
LDM
HDM
Glucose
Fructose
Chronic Insulin
Acute Insulin
+ +
+
+
+
+ +
+ +
+ +
+ +
+ +
+ + +
+ + +
+ + +
GLUT4 *
HNt
B GLUT4
Figure 5-3. Translocation of GLUT4 in Cells Treated with
Fructose. Panel A, cells were incubated under conditions as
described in Figure 5-2. Following the wash, cells were
acutely stimulated, or not, with 1 (J.M insulin for 10 min.
Membrane fractions were isolated and treated as described
above. Panel B, bands were quantitated by video
densitometry. These data are representative of three
independent experiments. II glucose; glucose + acute
insulin; ES9 fructose + chronic insulin; ^3 fructose +
chronic insulin + acute insulin.

117
plasma membrane as the controls when both groups were
stimulated acutely with insulin. Consistent with this was
the additional drop in GLUT4 in the LDM fraction relative to
that found in the LDM of control cells (Figure 5-3 A).
With the concern raised by Hresko et al. (120)
regarding glucosamine-induced ATP depletion, I measured the
concentration of ATP in cells chronically exposed to glucose
or fructose in the presence of insulin. As shown in Figure
5-4 A, the concentration of ATP did not change revealing an
intracellular concentration of about 7.6 mM 0.35. To
ascertain that ATP levels could indeed be affected, cells
were treated with 1 mM iodoacetate for 30 min which inhibits
glyceraldehyde phosphate dehydrogenase and thus glycolysis.
Under these conditions, ATP concentration dropped
significantly to 1.2 mM (Figure 5-4 B). These studies
conclude that fructose treatment had no deleterious effect
on the energy pool.
Effect of Fructose Feeding on Glycogen Metabolism
While there is evidence that fructose is metabolized in
3T3-L1 adipocytes (85), there are differences in the ability
of these cells to utilize glucose and fructose as
illustrated in Figure 5-5. In the following experiments,
total glycogen was measured after 12 h of exposure to either
glucose or fructose in the absence or presence of chronic

Figure 5-4. Effects of Fructose on ATP Levels. Panel A,
cells were incubated with DMEM containing 10% FBS or
glucose-free DMEM containing 10% dialyzed FBS and 25 mM
fructose in the presence or absence of 10 nM insulin for 12
h. Cells were then collected and ATP concentrations
determined by a luciferin/luciferase assay as described.
Data represent the average S.D. of three independent
experiments (n=6) control; E23 chronic insulin; ¡S33
fructose; E-S3 fructose + chronic insulin. Panel B,
iodoacetate (1 mM) was added to control cells and incubated
for 30 min. ATP was then determined as described
previously. Data represent a single experiment.

03
ATP Concentration (mM)
O N> G) 00
ATP Concentration
(mM)
o hO CT> 00

120
Figure 5-5. Effect of Fructose and Chronic Insulin on
Glycogen Levels. Cells were incubated with DMEM containing
10% FBS or glucose-free DMEM containing 10% dialyzed FBS and
25 mM fructose in the presence or absence of 10 nM insulin
for 12 h. Cells were collected and disrupted by sonication.
Glycogen was isolated by ethanol precipitation then
hydrolyzed using 2 N H2S04 and the concentration of glucose
determined by a commercially available kit (Sigma). Data
represent the average S.D. of five experiments.
glucose; E3 glucose + chronic insulin; ESS fructose; SS3
fructose + chronic insulin.

121
insulin. Glycogen content of cells incubated with 25 mM
glucose averaged 1.2 |omol of glucose/106 cells. In
contrast, glycogen levels in cells incubated with 25 mM
fructose fell substantially, confirming earlier work by
McMahon and Frost (79). Chronic exposure to glucose and
insulin increased the amount of glycogen by about 2-fold.
Cells chronically exposed to fructose and insulin still
exhibited reduced glycogen although not to the same extent
as in the presence of fructose, alone. When the apparent
rates of glycogen synthesis were investigated under these
conditions, I found that in cells pre-exposed to fructose,
glucose incorporation into glycogen was significantly higher
than in cells pre-exposed to glucose (Figure 5-6). In cells
exposed to glucose, alone, acute insulin treatment
stimulated the rate of glucose incorporation by 50-fold over
controls. Acute insulin exposure of fructose-treated cells
stimulated glucose incorporation by only 9-fold, and yet the
apparent rate was 4 times that in control. After chronic
insulin treatment, subsequent insulin-stimulated glucose
incorporation in glucose-fed cells dropped by 90%. Chronic
insulin exposure to fructose-fed cells reduced the
incorporation by 40%. As the intracellular pool of fructose
6-phosphate, glucose 6-phosphate, and glucose 1-phosphate
were not measured, these rates must not be overinterpreted.
However, the cells were incubated in KRP containing 5 mM

122
GO
ii
GO
43
GO
13
o
£ 2
m
o
o
O
co
O
o
3
bb
o
a
acute insulin + +
chronic insulin - + +
Figure 5-6. Effects of Fructose and Chronic Insulin on
Glycogen Synthesis. Cells were incubated in DMEM containing
10% FBS or glucose-free DMEM containing 10% dialyzed FBS and
25 mM fructose in the presence or absence of 10 nM insulin
for 12 h. Following "washout" to remove insulin, cells were
labeled with 14C-U-[D-glucose] (2 |j,Ci/plate) in KRP
containing 5 mM glucose and in the presence or absence of 1
(J.M insulin for 1 h. Glycogen was then collected as
described and radioactivity counted by liquid scintillation.
These data represent the average + S.D. of three independent
experiments (n=3). glucose; ¡^j fructose.

123
glucose for 1 h prior to the glycogen synthesis assay which
should serve to normalize the intracellular concentrations
of these metabolites. Indeed, I have measured the actual
mass of glycogen synthesized during the time in which cells
are exposed to [14C]-glucose (Table 5-1). Despite the
significant changes in the rates of glycogen synthesis in
response to acute insulin, the mass of glycogen changes very
little during this time. In fact, the only statistically
significant change in mass occurred in cells exposed to
fructose and acute insulin when compared to cells exposed to
fructose alone.
Conclusions
I have further investigated the role that the
hexosamine biosynthetic pathway plays in the induction of
insulin resistance in adipose cells. I have shown that 3T3-
L1 adipocytes, exposed to fructose and insulin for a chronic
period of time, down-regulate their expression of GLUT4 with
a specific loss from the endosomal fraction in a manner
similar to that observed in cells exposed to glucose and
insulin. However, when cells chronically treated with
fructose and insulin were acutely stimulated with insulin,
the amount of GLUT4 in the PM is equal to that seen in
control cells. This is in contrast to cells chronically
exposed to glucose and insulin, which exhibit a 50%

124
Glycogen Synthesis in 3T3-L1 Adipocytes
Apparent Rate of
[14C]-Glucose Incorp.
into Glycogen
(nmol/106 cells/h)
Apparent Rate
Change in Response
to Acute Insulin
Glycogen Mass
(p.mol/10s cells)
Control
basal
+ acute insulin
0.57 0.14
29.61 5.53 *
51.9x
0.85 0.30
1.04 0.29
Chronic Insulin
basal
+ acute insulin
0.90 0.44
3.94 0.93|
4.4x
1.88 0.56
2.00 0.53
Fructose
basal
+ acute insulin
11.14 4.05
100.96 27.18*
9.1x
0.33 0.07
0.60 0.22 t
Fructose and
Chronic Insulin
basal
+ acute insulin
23.89 8.33
64.37 18.63$
2.7x
0.72 0.19
0.84 0.22
Table 5-1. Glycogen Synthesis in 3T3-L1 Adipocytes. The
values presented in the "Apparent Rate of [i4C]-Glucose
Incorp. into Glycogen" column are the rates S.D. of
glycogen synthesis found in Figure 5-6. The "Glycogen Mass"
was determined by treating 10-cm plates of 3T3-L1 adipocytes
as in Figure 5-6 without the addition of 14C-U-[D-glucose] .
Total glycogen was then isolated from the cells and treated
as in Figure 5-5. The data from this column represent the
average S.D. of five independent experiments.
Significance of differences: basal versus + acute insulin,
*P<0.001, fP<0.01, JP<0.05 .

125
reduction in the level of GLUT4 in the PM upon acute insulin
stimulation. The observation that fructose does not support
the development of insulin-resistant glucose transport
activity is thus explained by this "enhanced" translocation
of GLUT4 to the PM in the presence of fructose.
As many others have shown, insulin stimulates glycogen
biosynthesis. While this process has been studied
mechanistically in adipose, the importance of this glycogen
pool has been dismissed, until recently, because of the
quantitatively small glycogen pools compared to that in
muscle or liver. However, Rigden et al. (124) and Frayn et
al. (90) have provided evidence that as much as 50% of the
glucose extracted by human adipose tissue, during in vivo
perfusion, is stored as glycogen. This suggests that
glycogen may play a potentially important role in adipose
metabolism. In fact, McMahon and Frost (79) have shown that
glycogen provides the carbohydrate for A7-linked protein
glycosylation during short term glucose deprivation of 3T3-
L1 adipocytes. Precursors for oligosaccharides are, in
part, generated through the hexosamine biosynthetic pathway,
the same pathway implicated in the development of insulin
resistance.
In the studies described here, I did observe a decrease
in the rate of glycogen synthesis following incubation of
cells with fructose and chronic insulin.
In control cells,

126
the rate of glycogen synthesis accelerated 50-fold with
acute insulin stimulation (Figure 5-6) and with chronic
exposure to 10 nM insulin, the glycogen mass increased by at
least 2-fold (Figure 5-5). After chronic exposure to
insulin in the presence of glucose, the acute stimulation of
glycogen synthesis by insulin decreased by 90%. This is
even more substantial than the loss in insulin-sensitive
glucose transport, indicating additional control steps
beyond glucose availability. Others have shown that
increased flux through the hexosamine pathway decreases
insulin-sensitive glycogen synthase activity (119) Thus,
inhibited glycogen synthase likely contributes to overall
desensitization of glycogen biosynthesis.
Fructose leads to depletion of the glycogen pool in
3T3-L1 adipocytes, as has been shown previously (79),
suggesting that fructose is a poor substrate for glycogen
biosynthesis in these cells. When cells previously exposed
to fructose were subsequently supplied with glucose, both
the basal and insulin-stimulated rates of glycogen synthesis
were significantly higher that cells previously exposed to
glucose. As all cells were "washed" for 1 h in buffer
containing 5 mM glucose prior to the glycogen synthesis
assay, it is unlikely that these differences are related to
alterations in the specific activity of endogenous substrate
pools. Thus, it appears that the rate of glycogen

127
biosynthesis is inversely related to the size of the
glycogen pool. This phenomenon has been observed in rat
skeletal muscle in which the rate of glycogen synthesis is
highest in those muscle subtypes which manifest the greatest
extent of glycogen depletion during food deprivation (125).
From a clinical perspective, are there lessons to be
learned from these studies? While my studies in 3T3-L1
adipocytes support the ability of fructose to prevent
insulin resistance, it must be pointed out that the
available pool of GLUT4 is substantially reduced in cells
exposed to chronic fructose and insulin despite the normal
sensitivity. The long term consequences of this are
unknown. One could argue that the high concentration used
in these studies is not likely achievable in vivo because of
the efficiency with which the liver absorbs fructose from
the portal system (126). Thus, even with high dietary
fructose, a maximal level of only 2.2 mM has been observed
in the human systemic circulation which is approximately 10
times less than the concentration used in this study.
However, preliminary experiments in 3T3-L1 adipocytes using
2 mM fructose with chronic insulin resulted in 30% down-
regulation of GLT4 in the LDM fraction (data not shown).
Studies in animal models have actually shown that
fructose administration leads to insulin resistance (127-
130). However, the diets in these studies derive 35-66% of

128
its calories from fructose, which leads to an increase in
plasma triglycerides (129) The liver in fact uses the
excess fructose to make triglycerides. This insulin
resistance could be completely ameliorated by the
administration of benfluorex, an agent that reduces hepatic
triglyceride output (129). This suggests that elevated
plasma triglyceride (or free fatty acids derived from
lipoprotein lipase activity on VLDL from the liver) is
responsible for the impaired insulin action rather than
elevated fructose per se. In contrast to animal studies,
those conducted in humans have shown beneficial effects from
dietary fructose. In fact, fructose administration
decreases postprandial serum glucose and insulin levels in
both normal and diabetic subjects (131-135). As well,
Koivisto et al. (131) found that dietary fructose increases
insulin sensitivity by 34% in subjects with type II
diabetes. In these human studies, the amount of fructose in
the diets represented only 13% of caloric content compared
to the 35-66% in animal diets (136). This lower level of
fructose does not lead to hypertriglyceridemia. Perhaps the
primary benefit of fructose in humans is to elicit a low
blood glucose and insulin secretory response (137) which
maintains insulin sensitivity and glycemic control.

CHAPTER 6
CONCLUSIONS AND
FUTURE DIRECTIONS
Conclusions
Insulin resistance is a hallmark of diabetes, a disease
which affects hundreds of thousands of people in the United
States and a vastly growing population around the world.
Unfortunately, type II diabetes is the most common and least
understood form of this disease. The role that adipose
plays in this disease has been underappreciated until
recently. The majority of individuals with type II diabetes
have this disease because they are obese. While this is a
strong statement, studies show that losing weight reverses
the effects of this disease.
The studies described herein examined the development
of insulin resistance in a stable adipose cell line. These
studies have shown that, indeed, the 3T3-L1 adipocytes
developed insulin resistance within physiological
concentrations of insulin. Essential to this discovery was
the experimental design that returned cells to a "basal"
state with an extensive insulin washout procedure. This
procedure removed insulin from the cells and allowed the
129

130
return to control levels of glucose transport activity.
Cells could then be restimulated acutely with insulin. Only
in this could "true" insulin resistance be measured. Not
only did this procedure allow us to examine glucose
transport activity in these cells, but the translocation of
GLUT4 as well. The insulin washout procedure allowed the
retrieval of GLUT4 from the plasma membrane back into the
endosomal (LDM) fraction. This resulted in very low levels
of GLUT4 in the plasma membrane, as were seen in the basal
state. With acute stimulation with insulin, the
translocation of GLUT4 could thus be monitored. Another
technique essential for measuring translocation of GLUT4 was
the subcellular fractionation procedure, which allowed the
separation of plasma membrane (PM), low density microsomes
(LDM), and high density microsomes (HDM).
The results of these studies showed that insulin
resistance in these cells resulted in a 50% decrease in
insulin-stimulated glucose transport activity. This effect
was both dose- and time-dependent. The development of
insulin resistance required glucose, glutamine, and insulin,
although glucosamine could substitute for glucose and
glutamine as shown also in isolated rat adipocytes
(15,18,20).
Although GLUT4 mRNA was reduced in response to insulin
levels, the K50 was 15 times greater than the equivalent

131
decrease in glucose transport activity. Therefore, the
level of GLUT4, the insulin-responsive glucose transporter,
was examined in resistant cells to determine if changes in
this protein could explain the reduced glucose transport
activity. In fact, GLUT4 levels were reduced 2.4-fold in
insulin-resistant cells. Translocation experiments revealed
that GLUT4 was reduced by 50% in the PM of resistant cells,
reflecting the reduction in the LDM fraction. This proved
that the loss of GLUT4 was responible for the reduced
insulin-stimulated glucose transport activity seen in these
cells. The mechanism responsible for the reduced GLUT4
levels was investigated and found to be a combination of
reduced synthesis and increased degradation of GLUT4
protein.
While the hexosamine biosynthetic pathway has been
implicated in the development of insulin resistance, the
anomalous effect of fructose cast some doubt. As fructose
is the substrate of GFAT, the rate-limiting enzyme in the
hexosamine biosynthetic pathway, incubation with this and
insulin should lead to decreased glucose transport activity,
if the hexosamine bisynthetic pathway is important in
inducing insulin resistance. However, no decrease in
insulin-stimulated glucose transport was observed following
incubation with insulin and fructose. This confirmed
previous data from isolated rat adipocytes (15).

132
Therefore, the levels and translocation of GLUT4 were
examined following incubation with fructose and insulin to
explain this apparent anomaly. GLUT4 levels were indeed
reduced by 50% in the LDM fraction of these cells as they
were with glucose. However, upon acute stimulation of these
cells with insulin, GLUT4 translocation was enhanced. This
resulted in levels of GLUT4 in the PM of fructose and
chronic insulin treated cells equal to the levels present in
the PM of control cells. This enhanced translocation
explained why glucose transport activity was not reduced in
cells treated with fructose and chronic insulin even in the
presence of reduced levels of GLUT4. These results may, in
part, explain the beneficial effects of fructose in the
diabetic diet.
Future Directions
These studies show that the down-regulation of GLUT4
protein underlies the manifestation of insulin resistance in
these cells. Therefore, I believe it is important to
explore further the mechanisms leading to the decreased
synthesis and increased degradation of GLUT4 in insulin-
resistant cells. As the decreased GLUT4 mRNA levels do not
seem to be involved in inducing insulin resistance in these
studies, perhaps the reduced synthesis is due to
translational control. The investigation of the mechanism

133
responsible for the degradation of GLUT4 is likely to prove
difficult. Of the several protease inhibitors used, none
yielded any positive results. However, the concentrations
of lactacystin and MG132 could be increased even though
concentrations used in this study have been sufficient in
other cells (111,113). Another compound used to study the
degradation of proteins in the lysosomes is ammonium
chloride. Although the effects of this compound were not
investigated in this study, it might prove beneficial to
test it for any effects.
If increasing the concentrations of lactacystin and
MG132 or testing of ammonium chloride have no effects on the
degradation of GLUT4, then this would point to the
possibility of a specific protease responsible for degrading
GLUT4 under conditions of chronic insulin. One such
candidate could be the insulin-responsive aminopeptidase
(IRAP) which co-localizes with GLUT4 containing vesicles.
IRAP also known as gpl60, for glycoprotein of 160 kDa, and
vpl65, for vesicle protein of 165 kDa, was identified by two
independent groups (116,117) and has been shown to
translocate to the plasma membrane in response to insulin
(116). This protein was shown to have aminopeptidase
activity (115) and to be a member of the family of zinc-
dependent membrane aminopeptidases (117). It contains a
large extracellular catalytic domain, single transmembrane

134
domain, and a unique extended cytoplasmic domain (117).
This cytoplasmic domain contains two dileucine motifs, which
are similar to the motifs found in the carboxy terminus of
GLUT4, and may be responsible for its similar intracellular
trafficking to GLUT4 (117). As the major protein component
of GLUT4-containing vesicles, IRAP could be responsible for
the degradation of GLT4. Although in previous studies, the
aminopeptidase activity of this protein did not change with
acute insulin (115), there may be other regulatory steps.
For instance, longer exposure to insulin may increase the
aminopeptidase activity by causing other factors, not seen
with acute insulin exposures, to affect the activity of this
protein. These factors may explain the data shown in Figure
4-7 in which incubation with cycloheximide during chronic
insulin treatment blocked the loss of GLUT4 in the LDM
fraction. Chronic insulin treatment could stimulate the
synthesis of a protein which increases the activity of this
aminopeptidase causing the degradation of GLUT4. In the
presence of cycloheximide, the synthesis of this protein is
blocked, preventing the activation of the aminopeptidase and
consequently the degradation of GLUT4.
Another area which I believe should be investigated
further is the reversal of insulin resistance. I have shown
that glucose transport activity can be restored in insulin-
resistant cells after 8 h following the removal of insulin

135
from the cells. This reversal is also protein synthesis
dependent in that cycloheximide blocks the return of normal
insulin-stimulated glucose transport. An important
observation is that this reversal takes place between 4-8 h
following the removal of insulin, suggesting again that the
synthesis of a protein is involved. Therefore, I think it
is important to examine the GLUT4 protein levels during this
time to ascertain if the recovery of GLUT4 protein is
responsible for this reversal.
In addition, other metabolites involved in the
hexosamine biosynthetic pathway could be investigated to
identify a specific culprit involved in inducing insulin
resistance by this pathway. This pathway seems to be
important in insulin resistance, but no products or
intermediates of this pathway as yet have been identified as
the causitive agent in inducing insulin resistance.
However, this pathway has been hypothesized to be a glucose
sensor (18,138,139) and therefore may regulate glucose
transport through negative feedback inhibition. As only 2-
3% of the fructose-6-P enters the hexosamine biosynthetic
pool, increased levels of hexosamines, such as UDP-N-
acetylglucosamine, may act as a sensor of energy in the cell
and decrease glucose uptake. In other words, high levels of
hexosamines would signal that the cell has a saturating
amount of substrate fluxing through the glycolytic pathway

136
and therefore has enough energy supply. As a result, the
elevated hexosaraines signal the cell to down-regulate the
transport of glucose into the cell.
Once these questions are answered, one can find ways to
block the degradation of GLUT4 and prevent the onset of
insulin resistance. If, in fact, the degradation is due to
the action of a single protease, such as the insulin-
responsive aminopeptidase, compounds could potentially be
designed to inhibit this protease specifically. With
today's new technology of combinatorial chemistry and high
throughput assays, which can design and test thousands of
compounds at a time, this task should be less difficult than
in years past. As long as such a compound was not cytotoxic
or have any other deleterious effects on the cell, the
compound could be tested in clinical trials and marketed for
individuals with diabetes. However, the basic understanding
of the mechanisms and regulatory pathways of insulin
resistance must be realized first. The knowledge gained
from this venture may spark other ideas on how best to treat
this disease. For instance, once the translocation of GLUT4
is understood completely, stimulation of its translocation
by compounds other than insulin may provide viable options
for treating diabetes. Only the coming years and research
by talented scientists will yield the data necessary to
overcome this debilitating disease.

APPENDIX A
GLUCOSE DEPRIVATION AND GLUT TRANSPORTERS
Studies examining the requirement of glucose for the
development of insulin resistance led to studies
investigating the expression of the GLUTS in response to
glucose deprivation. Previously, our lab has shown that
glucose deprivation, in the absence of chronic insulin
treatment, results in the appearance of the aberrantly
glycosylated form of GLUT1 after 18 h (85). This aberrantly
glycosylated form was named p37 because it migrates as a 37
kDa protein compared to the normal glycosylated form of
GLUT1 which migrates as a 46 kDa protein. In the presence
of chronic insulin, p37 exhibits significant expression
after only 12 h (Figure A-l). Under these conditions, the
distribution between p46 and p37 across the three membrane
fractions was 86% and 14% respectively. Interestingly, no
lower molecular weight form of GLUT4 was seen in the
presence or absence of chronic insulin with glucose
deprivation.
Although the half-life of GLUT1 under glucose
deprivation increases from 14 h to greater than 50 h (140)
the rate of synthesis does not change resulting in increased
137

138
PM
Glucose + +
Chronic Insulin + +
LDM
HDM
+
+ +
+ +
+
Figure A-l. Effect of Glucose Deprivation and Chronic
Insulin on the Aberrant Glycosylation of GLUT1 and GLUT4.
Cells were incubated in the presence or absence of 10 nM
insulin and DMEM containing 10% FBS or glucose-free DMEM
containing 10% dialyzed FBS for 12 h. Cells were washed to
remove insulin and membrane fractions collected. Proteins
were subjected to SDS-PAGE and transferred to nitrocellulose
for western blot analysis of GLUTl and GLUT4.

139
GLUTl (p46 and p37) levels after 24 h (85,140) Again,
chronic insulin under glucose-deprived conditions
accelerated these changes leading to a 2.2-fold increase in
the total GLUTl pool after 12 h (Figure A-l). Because
chronic insulin, alone, increases GLUTl synthesis in the
presence of glucose (Figure A-l), the elevation in GLUTl in
cells exposed to both chronic insulin and glucose
deprivation may result from both decreased degradation and
accelerated synthesis.
An important point to make is that the glycogen pool,
which provides glucose for the glycosylation of proteins
under glucose deprivation, is depleted after 12 h of glucose
deprivation (Figure A-2). In addition, regardless of the
presence of insulin, the glycogen pool is depleted under
conditions of glucose deprivation at the same rate. This
suggests that glucose, derived from glycogen, is diverted
from the hexosamine pathway in the presence of insulin.
Based on these observations, the combination of the
increased synthesis of GLUTl and depleted glycogen pool,
under chronic insulin conditions, likely leads to the
accelerated appearance of the aberrantly glycosylated form
of GLUTl, p37.

140
Figure A-2. Effect of Glucose Deprivation and Chronic
Insulin on Glycogen Levels. Cells were incubated in either
DMEM containing 10% FBS or glucose-free DMEM containing 10%
dialyzed FBS and in the presence or absence of 10 nM insulin
for up to 12 h. Cells were collected at specific times and
disrupted by sonication. Glycogen was isolated by ethanol
precipitation and then hydrolyzed using 2 N H2SO4 and the
concentration of glucose determined by a commercially
available kit (Sigma). Data represent the average S.D. of
duplicate plates within a single experiment. glucose;
glucose + chronic insulin; A glucose deprived; glucose
deprived + chronic isulin.

APPENDIX B
LEPTIN AND INSULIN RESISTANCE
Leptin, a 16 kDa protein which is the product of the OB
gene, was discovered in 1994 (141,142) This protein is
produced and secreted by white adipose tissue and 3T3-L1
adipocytes (143,144). Mice which have a homozygous nonsense
mutation in this gene, termed ob/ob mice, are unable to
produce leptin and are extremely obese (141,142) Another
well characterized mouse mutation is the db/db mouse. These
mice have a mutation in the hypothalamic receptor for leptin
and therefore, do not respond to this protein. As a result,
these mice are also obese and have elevated levels of leptin
(145.146). Leptin is believed to act as a regulator of
adiposity by acting as a satiety factor and regulator of
energy expenditure when it binds its receptor in the
hypothalamus (141,142).
Some evidence suggests that leptin may be involved in
insulin resistance. Both ob/ob and db/db mice exhibit
characteristics of insulin resistance and diabetes
(141.142.145.146). As well, leptin expression is down-
regulated with administration of an antidiabetic
thiazolidinedione in Zucker diabetic fatty rats, db/db mice
141

142
(147), and 3T3-L1 adipocytes (148). In addition, leptin
expression is up-regulated by insulin in 3T3 cells
(149,150). Further, leptin may impair insulin action in
isolated rat adipocytes (151). Recent data has shown that
leptin administration decreases glucose uptake, GLUT4 mRNA
and GLUT4 protein levels in white adipose tissue (152).
However, an opposite response was observed in two other
insulin-sensitive tissues, namely, muscle and brown adipose
tissue.
Therefore, I measured the concentration of leptin
secreted into the medium by the 3T3-L1 adipocytes following
chronic insulin treatment using a radioimmunoassay kit
supplied by Lineo. However, low concentrations of leptin in
the medium or serum has complicated its measurement. Leptin
levels measured by ELISA have yielded leptin concentrations
in the plasma of obese individuals of 20 ng/mL while those
in normal individuals ranged from 1-7 ng/mL (153). Levels
measured by other groups using the same RIA kit as used in
the present studies yielded varying results. Values
measured by this method gave leptin levels ranging from 20
pg/ml (154) to 7 ng/ml (144). Following concentration of
the medium overlaying my 3T3-L1 adipocytes, values of
approximately 2.5 ng/ml were obtained (Figure B-l).
However, neither chronic insulin nor incubation with
fructose caused any significant change in the levels of

143
leptin secreted from these cells. I can only hypothesize
that differences in experimental procedure have led to such
a wide range of reported concentrations of leptin secreted
by 3T3-L1 adipocytes.

144
Figure B-l. Leptin Levels. Cells were incubated in
glucose- and serum-free medium containing 25 mM glucose or
fructose and 10 nM insulin for 12 h. Medium was collected
from the plates and concentrated using concentrator spin
columns (Amicon). Leptin present in the medium was then
determined by an 125I-leptin radioimmunoassay kit from Lineo.
Data represent a single experiment. CD glucose; 1231 glucose +
chronic insulin; ¡S9 fructose; ES3 fructose + chronic insulin.

REFERENCES
1. Sacks, D.B. and McDonald, J.M. (1996) Am. J. Clin.
Path. 105, 149-156.
2. Olefsky, J.M., Garvey, W.T., Henry, R.R., Brillon, D.,
Matthaei, S. and Freidenberg, G.R. (1988) Am. J.
Med. 85, 86-105.
3. Kahner, E.A., Porta, M. and Hyer, S.L. (1994) in:
Chronic Complications of Diabetes, pp. 52-62
(Pickup, J.C.a.W., G., Ed.) Blackwell Scientific
Publications, Oxford.
4. Walker, J.D. and Viberti, G.C. (1994) in: Chronic
complications of Diabetes, pp. 146-161 (Pickup,
J.C.a.W., G., Ed.) Blackwell Scientific
Publications, Oxford.
5. Thomas, P.K. (1994) in: Chronic Complications of
Diabetes, pp. 101-111 (Pickup, J.C.a.W., G., Ed.)
Blackwell Scientific Publications, Oxford.
6. Stout, R.W. (1992) in: Diabetes and Artherosclerosis,
pp. 53-88 (Stout, R.W., Ed.) Kluwer Academic
Publishers, The Netherlands.
7. Trimble, E.R. and McDowell, I.F.W. (1992) in: Diabetes
and Artherosclerosis, pp. 111-140 (Stout, R.W.,
Ed.) Kluwer Academic Publishers, The Netherlands.
8. Mayne, E.E. (1992) in: Diabetes and Artherosclerosis,
pp. 219-236 (Stout, R.W., Ed.) Kluwer Academic
Publishers, The Netherlands.
9. Garvey, W.T., Huecksteadt, T.P., Matthaei, S. and
Olefsky, J.M. (1988) J. Clin. Invest. 81, 1528-
1536.
10. Hotamisligil, G.S., Budavari, A., Murray, D. and
Spiegelman, B.M. (1994) J. Clin. Invest. 94, 1543-
1549.
145

146
11. Hotamisligil, G.S., Murray, D.L., Choy, L.N. and
Spiegelman, B.M. (1994) Proc. Natl. Acad. Sci.
U.S.A. 91, 4854-4858.
12. Hotamisligil, G.S., Peraldi, P., Budavari, A., Ellis,
R., White, M.F. and Spiegelman, B.M. (1996)
Science. 271, 665-668.
13. Maddux, B.A. et al. (1995) Nature. 373, 448-451.
14. Flores-Riveros, J.R., McLenithan, J.C., Ezaki, 0. and
Lane, M.D. (1993) Proc. Natl. Acad. Sci. U.S.A.
90, 512-516.
15. Garvey, W.T., Olefsky, J.M., Matthaei, S. and Marshall
S. (1987) J. Biol. Chem. 262, 189-197.
16. Kozka, I.J., Clark, A.E. and Holman, G.D. (1991) J.
Biol. Chem. 266, 11726-11731.
17. Mller, G., Dearey, E.A. and Pnter, J. (1993) Biochem
J. 289, 509-521.
18. Marshall, S., Bacote, V. and Traxinger, R.R. (1991) J.
Biol. Chem. 266, 4706-4712.
19. Marshall, S., Garvey, W.T. and Traxinger, R.R. (1991)
FASEB. 5, 3031-3036.
20. Traxinger, R.R. and Marshall, S. (1989) J. Biol. Chem.
264, 20910-20916.
21. Traxinger, R.R. and Marshall, S. (1991) J. Biol. Chem.
266, 10148-10154.
22. Su, H.-Y., Sheu, W.H.-H., Chin, H.-M.L., Jeng, C.-Y.,
Chen, Y.-D.I. and Reaven, G.M. (1995) Am. J.
Hyper. 8, 1067-1071.
23. Colman, E., Katzel, L.I., Rogus, E., Coon, P., Muller,
D. and Goldberg, A.P. (1995) Metabolism. 44, 1502
1508 .
24. Hotamisligil, G.S., Shargill, N.S. and Spiegelman, B.M
(1993) Science. 259, 87-91.
25. Hotamisligil, G.S. and Spiegelman, B.M. (1994)
Diabetes. 43, 1271-1278.

147
26. Hofmann, C., Lorenz, K., Braithwaite, S.S., Coica,
J.R., Palazuk, B.J., Hotamisligil, G.S. and
Spiegelman, B.M. (1994) Endocrinology. 134, 264-
270 .
27. Cornelius, P., Lee, M.D., Marlowe, M. and Pekala, P.H.
(1989) Biochem. Res. Common. 165, 429-436.
28. Szalkowski, D., White-Carrington, S., Berger, J. and
Zhang, B. (1995) Endocrinology. 136, 1474-1481.
29. Randle, P.J., Garland, P.B., Hales, C.N. and Newsholme,
E.A. (1963) Lancet. 1, 785-789.
30. Schlach, D.S. and Kipnis, D.M. (1965) J. Clin. Invest.
44, 2010-2020.
31. Randle, P.J., Newsholme, E.A. and Garland, P.B. (1964)
Biochem. J. 93, 652-655.
32. Ferrannini, E., Barrett, E.J., Bevilacqua, S. and
DeFronzo, R.A. (1983) J. Clin. Invest. 72, 1737-
1747 .
33. Arslanian, S.A. and Kalhan, S.C. (1994) Diabetes. 43,
908-914.
34. Biermann, E.L., Dole, V.P. and Roberts, T.N. (1959)
Diabetes. 6, 475-479.
35. Moitra, J. et al. (1998) Genes Dev. 12, 3168-3181.
36. Green, H. and Meuth, M. (1974) Cell. 3, 127-133.
37. Green, H. and Kehinde, 0. (1976) Cell. 7, 105-113.
38. Kuri-Harcuch, W. and Green, H. (1977) J. Biol. Chem.
252, 2158-2160.
39. Mackall, J.C., Student, D.K., Polakis, S.E. and Lane,
M.D. (1976) J. Biol. Chem. 251, 6462-6464.
40. Reed, B.C., Kaufmann, S.H., Mackall, J.C., Student,
A.K. and Lane, M.D. (1977) Proc. Natl. Acad. Sci.
U.S.A. 74, 4876-4880.
41. de Herreros, A.G. and Birnbaum, M.J. (1989) J. Biol.
Chem. 264, 19994-19999.

148
42. Kaestner, K.H., Christy, R.J., McLenithan, J.C.,
Braiterman, L.T., Cornelius, P., Pekala, P.H. and
Lane, M.D. (1989) Proc. Natl. Acad. Sci. U.S.A.
86, 3150-3154.
43. Oka, Y., Asano, T., Shibasaki, Y., Kasuga, M.,
Kanazawa, Y. and Takaku, F. (1988) J. Biol. Chem.
263, 13432-13439.
44. Olson, A. and Pessin, J.E. (1996) Annu. Rev. Nutr. 16,
235-256.
45. Carruthers, A. (1990) Physiol. Rev. 70, 1135-1176.
46. Gould, G.W. and Bell, G.I. (1990) TIBS. 15, 18-23.
47. Blok, J., Gibbs, E.M., Lienhard, G.E., Slot, J.W. and
Geuze, H.J. (1988) J. Cell Biol. 106, 69-76.
48. Cushman, S.W. and Wardzala, L.J. (1980) J. Biol. Chem.
255, 4758-4762.
49. Suzuki, K. and Kono, T. (1980) Proc. Natl. Acad. Sci.
U.S.A. 77, 2542-2545.
50. Birnbaum, M.J. (1989) Cell. 57, 305-315.
51. Charron, M.J., Brosius, F.C., Alper, S.L. and Lodish,
H.F. (1989) Proc. Natl. Acad. Sci. U.S.A. 86,
2535-2539.
52. Fukumato, J., Kayano, L., Buse, J.B., Edwards, Y.,
Pilch, P.F., Bell, G.I. and Seino, S. (1989) J.
Biol. Chem. 264, 7776-7779.
53. James, D.E., Strube, M. and Mueckler, M. (1989) Nature.
338, 83-87.
54. Sollner, T., Whiteheart, S.W., Brunner, M., Erdjument-
Bromage, H., Geromanos, S., Tempst, P. and
Rothman, J.E. (1993) Nature. 362, 318-324.
55. Bock, J.B. and Scheller, R.H. (1997) Nature. 387, 133-
134 .
56. Martin, L.B., Shewan, A., Millar, C.A., Gould, G.W. and
James, D.E. (1998) J. Biol. Chem. 273, 1444-1452.

149
57. Rea, S., Martin, L.B., McIntosh, S., Macaulay, S.L.,
Ramsdale, T., Baldini, G. and James, D.E. (1998)
J. Biol. Chem. 273, 18784-18792.
58. Thurmond, D.C., Ceresa, B.P., Okada, S., Elmendorf,
J.S., Coker, K. and Pessin, J.E. (1998) J. Biol.
Chem. 273, 33876-33883.
59. White, M.F. and Kahn, C.R. (1994) J. Biol. Chem. 269,
1-4 .
60. Saltiel, A.R. (1996) Am. J. Physiol. 270, E375-E385.
61. Haruta, T., Morris, A.J., Rose, D.W., Nelson, J.G.,
Mueckler, M. and Olefsky, J.M. (1995) J. Biol.
Chem. 270, 27991-27994.
62. Kohn, A.D., Summers, S.A., Birnbaum, M.J. and Roth,
R.A. (1996) J. Biol. Chem. 271, 31372-31378.
63. Okada, T., Kowano, Y., Sakakibara, T., Hazeki, O. and
Ui, M. (1994) J. Biol. Chem. 269, 3568-3573.
64. Ricort, J.-M., Tanti, J.-F., Van Obberghen, E. and Le
Marchand-Brustel, Y. (1995) Diabetologia. 38,
1148-1156.
65. Robinson, L.J., Razzack, Z.F., Lawrence, J.C., Jr. and
James, D.E. (1993) J. Biol. Chem. 268, 26422-
26427.
66. Lin, T., Kong, X., Saltiel, A.R., Blackshear, P.J. and
Lawrence, J.C., Jr. (1997) J. Biol. Chem. 270,
18531-18538.
67. Wiese, R.J., Mastick, C.C., Lazar, D.F. and Saltiel,
A.R. (1995) J. Biol. Chem. 270, 3442-3446.
68. Frost, S.C. and Lane, M.D. (1985) J. Biol. Chem. 260,
2646-2652.
69. Rubin, C.S., Hirsch, A., Fung, C. and Rosen, O.M.
(1978) J. Biol. Chem. 253, 7570-7578.
70. Fisher, M.D. and Frost, S.C. (1996) J. Biol. Chem. 271,
11806-11809.

150
71. Weber, T.M., Joost, H.-G., Simpson, I.A. and Cushman,
S.W. (1988) in: Methods for Assessment of Glucose
Trasport Activity and the Number of Glucose
Transporters in Isolated Rat Adipose Cells and
Membrane Fractions, pp. 171-187 (Kahn, R.C. and
Harrison, L.C., Eds.) Alan R. Liss, Inc., New
York.
72. Balch, W.E. and Rothman, J.E. (1985) Arch. Biochem.
Biophys. 240, 413-425.
73. Markwell, M.A.K., Haas, S.M., Bieber, L.L. and Tolbert,
N.E. (1978) Anal. Biochem. 87, 206-210.
74. Lowry, O.H., Rosenbrough, N.J., Farr, A.L. and Randall,
R.J. (1951) J. Biol. Chem. 193, 265-275.
75. Laemmli, U.K. (1970) Nature. 227, 680-685.
76. Clancy, B.M. and Czech, M.P. (1990) J. Biol. Chem. 265,
12434-12443.
77. Church, G.M. and Gilbert, W. (1984) Proc. Natl. Acad.
Sci. U.S.A. 81, 1991-1995.
78. Feinberg, A.P. and Vogelstein, B. (1983) Anal. Biochem.
132, 6-13.
79. McMahon, R.J. and Frost, S.C. (1996) Am. J. Phys. 270,
E640-E645.
80. Pfleiderer, G. (1963) in: Methods of Enzymatic
Analysis, pp. 91-103 (Bergmeyer, H.U., Ed.)
Academic, New York.
81. Harlow, E. and Lane, D. (1988) pp. 55-77 Cold Springs
Harbor Laboratory, New York.
82. Sigel, M.B., Sinha, Y.N. and VanderLaan, W.P. (1983)
in: Methods in Enzymology, Vol. 93, pp. 3-12
Academic Press, Inc., New York.
83. Dankert, J.R., Shiver, J.W. and Esser, A.F. (1985)
Biochemistry. 24, 2754.
84. Garvey, W.T., Maianu, L., Hancock, J.A., Golichowski,
A.M. and Baron, A. (1992) Diabetes. 41, 465-475.

151
85. Kitzman, H.H.J., McMahon, R.J., Williams, M.G. and
Frost, S.C. (1993) J. Biol. Chem. 268, 1320-1325.
86. van Putten, J.P.M. and Krans, H.M.J. (1985) J. Biol.
Chem. 260, 7996-8001.
87. Reed, B.C., Shade, D., Alperovich, F. and Vang, M.
(1990) Arch. Biochem. Biophys. 279, 261-274.
88. Tordjman, K.M., Leingang, K.A. and Mueckler, M. (1990)
Biochemistry. J. 271, 201-207.
89. Froesch, E.R., Burgi, H., Bally, P. and Labhart, A.
(1965) Mol. Pharmacol. 1, 280-296.
90. Frayn, K.N., Coppack, S.W. and Humphreys, S.M. (1989)
Biochem. Soc. Trans. 17, 1091.
91. Marshall, S., Bacote, V. and Traxinger, R.R. (1991) J.
Biol. Chem. 266, 10155-10161.
92. Traxinger, R.R. and Marshall, S. (1989) J. Biol. Chem.
264, 8156-8163.
93. Decker, H., Zahner, H., Heitsch, H., Konig, W.A. and
Fiedler, H.-P. (1991) J. Gen. Micro. 137, 1805-
1813 .
94. Causier, B.E., Milling, R.J., Foster, S.G. and Adams,
D.J. (1994) Microbiology. 140, 2199-2205.
95. Knutson, V.P., Donnelly, P.V., Baiba, Y. and Lopez-
Reyes, M. (1995) J. Biol. Chem. 270, 24972-24981.
96. Garvey, W.T. (1992) Diab. Care. 15, 396-417.
97. Pedersen, O., Bak, J.F., Andersen, P.H., Lund, S.,
Moller, D.E., Flier, J.S. and Kahn, B.B. (1990)
Diabetes. 39, 865-870.
98. Calderhead, D.M., Kitagawa, K., Tanner, L.I., Holman,
G.D. and Lienhard, G.E. (1990) J. Biol. Chem. 265,
13800-13808.
99. Reed, B.C. and Lane, M.D. (1980) Adv. Enzyme Regul. 18,
97-117.

152
100. Tordjman, K.M., Leingang, K.A., James, D.E. and
Mueckler, M.M. (1989) Proc. Natl. Acad. Sci.
U.S.A. 86, 7761-7765.
101. Sargeant, R.J. and Pquet, M.R. (1993) Biochem. J. 290,
913-919.
102. Ricort, J.-M., Tanti, J.-F., Cormont, M., Van
Obberghen, E. and Le Marchand-Brustel, Y. (1994)
FEBS. 347, 42-44.
103. Dice, J.F. (1987) FASEB. 1, 349-357.
104. Hochstrasser, M. (1995) Curr. Opin. Cell Biol. 7, 215-
223 .
105. Poole, B. and Ohkuma, S. (1981) J. Cell Biol. 90, 665-
669.
106. Kirschke, H., Danger, J., Wiederander, B., Ansorge, S.,
Bohley, P. and Broghammer, U. (1976) Acta Biol.
Med. Germ. 35, 285-299.
107. Kirschke, H., Langner, J., Wiederanders, B., Ansorge,
S. and Bohley, P. (1977) Eur. J. Biochem. 74, 293-
301.
108. Omura, S., Keiichi, M., Tomoko, F., Kazuchito, K.,
Toshio, F., Shigeo, F. and Akira, N. (1991) J.
Antibio. 44, 113-118.
109. Fenteany, G., Standaert, R.F., Lane, W.S., Choi, S.,
Corey, E.J. and Schreiber, S.L. (1995) Science.
268, 726-731.
110. Craiu, A., Gaczynska, M., Akopian, T., Gramm, C.F.,
Fentean, G., Goldberg, A.L. and Rock, K.L. (1997)
J. Biol. Chem. 272, 13437-13445.
111. Dick, L.R. et al. (1997) J. Biol. Chem. 272, 182-188.
112. Rock, K.L., Gramm, C., Rothstein, L., Clark, K., Stein,
R., Dick, L., Hwang, D. and Goldberg, A.L. (1994)
Cell. 78, 761-771.
113. Lee, D. and Goldberg, A.L. (1996) J. Biol. Chem. 211,
27280-27284.

153
114. Seglen, P.O., Grinde, B. and Solheim, A.E. (1979) Eur.
J. Biochem. 95, 215-225.
115. Kandror, K.V., Yu, L. and Pilch, P.F. (1994) J. Biol.
Chem. 269, 30777-30780.
116. Kandror, K.V. and Pilch, P.F. (1994) Proc. Natl. Acad.
Sci. U.S.A. 91, 8017-8021.
117. Keller, S.R., Scott, H.M., Mastick, C.C., Aebersold, R.
and Lienhard, G.E. (1995) J. Biol. Chem. 270,
23612-23618.
118. Thomson, M.J., Williams, M.G. and Frost, S.C. (1997) J.
Biol. Chem. 272, 7759-7764.
119. Crook, E.D., Zhou, J., Daniels, M., Neidigh, J.L. and
McClain, D.A. (1995) Diabetes. 44, 314-320.
120. Hresko, R.C., Heimberg, H., Chi, M. and Mueckler, M.
(1998) J. of Biol. Chem. 273, 20658-20668.
121. Sato, N., Irie, M., Kajinuma, H. and Suzuki, K. (1990)
Endocrinology. 127, 1072-1077.
122. Toyoda, N., Robinson, F.W., Smith, M.M., Flanagan, J.E.
and Kono, T. (1986) J. Biol. Chem. 261, 2117-2122.
123. Froesch, E.R.a.G., J.L. (1962) J. Biol. Chem. 237,
3317-3324.
124. Rigden, D.J., Jellyman, A.E., Frayn, K.N. and Coppack,
S.W. (1990) Eur. J. Clin. Nutr. 44, 689-692.
125. Holness, M.J., Shuster-Bruce, M.J.L. and Sugden, M.C.
(1988) Biochem. J. 254, 855-859.
126. Mayes, P.M. (1993) Ami. J. Clin. Nutr. 58 (suppl) 754S-
65S .
127. Tobey, T.A., Mondon, C.E., Zavaroni, I. and Reaven,
G.M. (1982) Metabolism. 31, 608-612.
128. Vrna, A. and Kazdov, L. (1970) Life Sci. 9, 257-265.
129. Storlien, L.H., Oakes, N.D., Pan, D.A., Kusunoki, M.
and Jenkins, A.B. (1993) Diabetes. 42, 457-462.

154
130. Luo, J., Rizkalla, S.W., Lerer-Metzger, M., Boillot,
J., Ardeleanu, A., Bruzzo, F., Chevalier, A. and
SIama, G. (1995) J. Nutr. 125, 164-171.
131. Koivisto, V.A. and Yki-Jarvinen, H. (1993) J. Inter.
Med. 233, 145-153.
132. Crapo, P.A., Kolterman, O.G. and Henry, R.R. (1986)
Diab. Care. 9, 111-119.
133. Bantle, J.P., Laine, D.C., Castle, G.W., Thomas, J.W.,
Hoogwerf, B.J. and Goetz, F.C. (1983) N. Eng. J.
Med. 309, 7-12.
134. Crapo, P.A. and Kolterman, O.G. (1984) Am. J. Clin.
Nutr. 39, 525-534.
135. Akgn, S. and Ertel, N. (1980) Diab. Care. 3, 582-585.
136. Henry, R.R. and Crapo, P.A. (1991) Ann. Rev. Nutr. 11,
21-39.
137. Crapo, P.A., Kolerman, O.G. and Olefsky, J.M. (1980)
Diab. Care. 3, 575-582.
138. Cooksey, R.C., Hebert, L.F., Zhu, J., Wofford, P.,
Garvey, W.T. and McClain, D.A. (1999)
Endocrinology. 140, 1151-1157.
139. Hawkins, M., Angelov, I., Liu, R., Barzilai, N. and
Rossetti, L. (1997) J. Biol. Chem. 272, 4889-4895
140. McMahon, R.J. and Frost, S.C. (1995) J. Biol. Chem.
270, 12094-12099.
141. Zhang, Y., Proenca, R., Maffei, M., Barone, M.,
Leopold, L. and Friedman, J.M. (1994) Nature. 372
425-432.
142 .
Halaas, J.L.
et al.
(1995)
Science. 269, 543-546.
143 .
Masuzaki, H.
et al.
(1995)
Diabetes. 44, 855-858.
144 .
Yoshida, T.,
Monkawa, T.,
Hayashi, M. and Saruta, T
(1997) Biochem. Biophys. Res. Comm. 232, 822-826.
145. Tartaglia, L.A. et al. (1995) Cell. 83, 1263-1271.

155
146. Slieker, L.J., Sloop, K.W., Surface, P.L., Kriauciunas,
A., LaQuier, F., Manetta, J., Bue-Valleskey, J.
and Stephens, T.W. (1996) J. Biol. Chem. 271,
5301-5304.
147. Zhang, B. et al. (1996) J. Biol. Chem. 271, 9455-9459.
148. Kallen, C.B. and Lazar, M.A. (1996) Proc. Natl. Acad.
Sci. U.S.A. 93, 5793-5796.
149. MacDougald, O.A., Hwang, C., Fan, H. and Lane, M.D.
(1995) Proc. Natl. Acad. Sci. U.S.A. 92, 9034-
9037 .
150. Leroy, P., Dessolin, S., Villageois, P., Moon, B.C.,
Friedman, J.M., Ailhaud, G. and Dani, C. (1996) J.
Biol. Chem. 271, 2365-2368.
151. Muller, G., Ertl, J., Gerl, M. and Preibisch, G. (1997)
J. Biol. Chem. 272, 10585-10593.
152. Wang, J., Chinookoswong, N., Scully, S., Qi, M. and
Shi, Z. (1999) Endocrinology. 140, 2117-2124.
153. Hebebrand, J., Heyden, J., Devos, R., Kopp, W.,
Herpertz, S., Remschmidt, H. and Herzog, W. (1995)
Lancet. 346, 1624-1625.
154. Barthel, A., Kohn, A.D., Luo, Y. and Roth, R.A. (1997)
Endocrinology. 138, 3559-3562.

BIOGRAPHICAL SKETCH
Michael James Thomson was born in Kenmore, New York, in
March 1972, to James and Donna Thomson. He is one of four
children, a brother and sister of which, including Michael,
make up a set of triplets, and an older brother. He was
raised in Niagara Falls, New York, until he moved with his
family to Lake City, Florida, in the summer of 1985. He
attended Lake City Community College and graduated with an
associate of arts degree in 1992. He continued his
education at the University of Florida, where he graduated
with a bachelor of science degree in Zoology in 1994 and a
doctor of philosophy in biochemistry and molecular biology
in 1999. He will now attend the School of Veterinary
Medicine at the University of Florida.
156

I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality, as
a dissertation for the degree of Doctor of Philosophy.
yC-
Susan C. Frost, Chair
Associate Professor of
Biochemistry and
Molecular Biology
I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality, as
a dissertation for the degree of Doctor of Philosc
MiCTbael S. Kilb^rg
Professor of Biochemistry
and Molecular Bioloc
I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality,
a dissertation for the degree of Doctor of Philosophy.
hi. i/M.
Allen
as
Charles M,
Professor of Biochemistry
and Molecular Biology
I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality, as
a dissertation for the degree of Doctor of Philosophy.
Thomas W. O'Brien
Professor of Biochemistry
and Molecular Biology
I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality, as
a dissertation for the degree of Docto^Q^-pk^hqsophy.
Mark A.'hAtkinson
Professor of Pathology

This dissertation was submitted to the Graduate Faculty
of the College of Medicine and to the Graduate School and
was accepted as partial fulfillment of the requirements for
the degree of Doctor of Philosophy.
August 1999
Dean, College of Medicine
Dean, Gradute School



51
I relied initially on studies by Garvey et al. (15). These
investigators showed that incubation of cultured rat
adipocytes with insulin for 12 h was sufficient to complete
the desensitization process, resulting in a 50% reduction in
insulin-sensitive glucose transport activity. Therefore,
cells were first treated with either 10 nM or 1 |xM insulin
for 12 h. The cells were then washed in KRP buffer
containing 5 mM glucose and 0.1% defatted BSA at 20-min
intervals over a 140-min time course. At specific times
during this washout period, cells were rinsed in KRP alone
and glucose transport activity determined in a 2-min pulse.
Figure 3-1 A shows the comparison between control cells
(washed in an identical manner) and those treated
chronically with either 1 [xM or 10 nM insulin. Cells
treated chronically with 1 p,M insulin showed significantly
elevated transport at the start of washout (time 0) but
never achieved basal levels despite the extensive time of
washing. However, those cells treated with 10 nM insulin
returned to basal levels within 60 min of washing. When
these latter cells were subsequently restimulated with an
acute insulin challenge following the 60-min wash, the rate
of glucose transport was reduced by 50% compared to control
cells (Figures 3-1 B and 3-2). In addition to the decreased
rate, the cells were less sensitive to insulin in that the


I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality, as
a dissertation for the degree of Doctor of Philosophy.
yC-
Susan C. Frost, Chair
Associate Professor of
Biochemistry and
Molecular Biology
I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality, as
a dissertation for the degree of Doctor of Philosc
MiCTbael S. Kilb^rg
Professor of Biochemistry
and Molecular Bioloc
I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality,
a dissertation for the degree of Doctor of Philosophy.
hi. i/M.
Allen
as
Charles M,
Professor of Biochemistry
and Molecular Biology
I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality, as
a dissertation for the degree of Doctor of Philosophy.
Thomas W. O'Brien
Professor of Biochemistry
and Molecular Biology
I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality, as
a dissertation for the degree of Docto^Q^-pk^hqsophy.
Mark A.'hAtkinson
Professor of Pathology


28
Glucose was released from glycogen by acid hydrolysis.
Specifically, 0.2 mL 2 N H2SO4 was added to the glycogen
pellet and incubated in a boiling water bath for 2 h. The
samples were cooled to room temperature and 0.15 mL 2 N NaOH
and 0.65 mL H20 were added. The concentration of glucose in
the hydrolysate was then determined using a hexokinase-based
glucose kit (Glucose HK Kit) from Sigma. For samples fed
with glucose, 0.01 mL of hydrolysate was diluted with 0.99
mL of H2S04/Na0H/H20 mixture (2:1.5:6.5). Samples deprived
of glucose were diluted by adding 0.1 mL of hydrolysate to
0.9 mL of the H2S04/Na0H/H20 mixture. Duplicate aliquots of
0.1 mL were added to 1 mL of prewarmed assay reagent and
incubated at 37C for 30 min. The absorbance of the samples
was then measured at 340 nm.
To limit the variability in experiments dealing with
glycogen, glycogen levels were monitored during the week in
which the 3T3-L1 adipocytes were fully differentiated. As
these cells are fed every 48 h during their normal
maintenance feeding schedule and they begin to enter
glycogen deprivation on the tail end of this time, it was
hypothesized that glycogen levels would fluctuate between
feedings. Upon collecting glycogen from the cells each day
of the week, at the same time of day, I found that glycogen
levels did indeed fluctuate. As Figure 2-2 shows, glycogen
levels were highest the day after they were fed and lowest


152
100. Tordjman, K.M., Leingang, K.A., James, D.E. and
Mueckler, M.M. (1989) Proc. Natl. Acad. Sci.
U.S.A. 86, 7761-7765.
101. Sargeant, R.J. and Pquet, M.R. (1993) Biochem. J. 290,
913-919.
102. Ricort, J.-M., Tanti, J.-F., Cormont, M., Van
Obberghen, E. and Le Marchand-Brustel, Y. (1994)
FEBS. 347, 42-44.
103. Dice, J.F. (1987) FASEB. 1, 349-357.
104. Hochstrasser, M. (1995) Curr. Opin. Cell Biol. 7, 215-
223 .
105. Poole, B. and Ohkuma, S. (1981) J. Cell Biol. 90, 665-
669.
106. Kirschke, H., Danger, J., Wiederander, B., Ansorge, S.,
Bohley, P. and Broghammer, U. (1976) Acta Biol.
Med. Germ. 35, 285-299.
107. Kirschke, H., Langner, J., Wiederanders, B., Ansorge,
S. and Bohley, P. (1977) Eur. J. Biochem. 74, 293-
301.
108. Omura, S., Keiichi, M., Tomoko, F., Kazuchito, K.,
Toshio, F., Shigeo, F. and Akira, N. (1991) J.
Antibio. 44, 113-118.
109. Fenteany, G., Standaert, R.F., Lane, W.S., Choi, S.,
Corey, E.J. and Schreiber, S.L. (1995) Science.
268, 726-731.
110. Craiu, A., Gaczynska, M., Akopian, T., Gramm, C.F.,
Fentean, G., Goldberg, A.L. and Rock, K.L. (1997)
J. Biol. Chem. 272, 13437-13445.
111. Dick, L.R. et al. (1997) J. Biol. Chem. 272, 182-188.
112. Rock, K.L., Gramm, C., Rothstein, L., Clark, K., Stein,
R., Dick, L., Hwang, D. and Goldberg, A.L. (1994)
Cell. 78, 761-771.
113. Lee, D. and Goldberg, A.L. (1996) J. Biol. Chem. 211,
27280-27284.


16
endogenous p21ras has no effect on insulin-stimulated GLUT4
translocation (61).
The insulin receptor and the insulin signaling
cascade(s) are important in insulin resistance because it
has been shown that defects in autophosphorylation of the
insulin receptor leads to insulin-resistant cells (2,10-13).
This leads to a decrease in the phosphorylation of IRS-1
(2,10-13) which in turn leads to a reduction in signaling
cascades. This effect, however, may primarily be due to
secondary effects of TNF-a and the pharmacological doses of
insulin used. Genetic defects in the insulin receptor also
result in decreased autophosphorylation of the receptor and
decreased phosphorylation of IRS-1. This is only a minor
cause for the development of insulin resistance, as it does
not account for the induction of insulin resistance in
individuals with normal insulin receptors and those
individuals who develop insulin resistance as a result of
becoming obese. However, there is evidence in 3T3-L1
adipocytes that chronic insulin treatment alone can lead to
impaired insulin receptor signaling and down-regulation of
IRS-1 expression and phosphorylation (64). Again, these
studies were performed using pharmacological concentrations
of insulin.


Figure 4-4. Effect of Chronic Insulin on the Synthesis of
GLUT4. Panel A, cells were incubated in DMEM containing
10% FBS in the presence (chronic) or absence (control) of
10 nM insulin for 11 h. Cells were then incubated in DMEM
lacking methionine and cysteine and FBS for 1 h, also in
the presence or absence of 10 nM insulin. Fresh DMEM,
supplemented with [35S] cysteine/methionine (400 (j,Ci) and
10 nM insulin, was then added and the cells incubated for
specific times. A total membrane fraction was then
isolated and GLUT4 immunoprecipitated from this fraction.
Immunoprecipitated proteins were then subjected to SDS-PAGE
and radiolabeled proteins visualized by autoradiography.
pGLUT4 represents precursor GLUT4 and mGLUT4, mature GLUT4,
based on its glycosylation state. Panel B, densitometric
analysis was performed. The values represent total GLUT4,
both pGLUT4 and mGLUT4. The data represent the average
S.D. of three independent experiments. 9 control;
chronic insulin.


83
Time (h)
Control
Control + acute ins
Control + cyclo
Control + cyclo + acute ins
--
Chronic
e-
Chronic + acute ins
-a-
Chronic + cyclo
Chronic + cyclo + acute ins
Figure 3-17. Reversal of Insulin Resistance in the Presence
of Cycloheximide. Cells were incubated and washed as
described in the previous figure. Plates were then refed
with fresh medium 20 |iM cycloheximide. At specific times,
basal and insulin-stimulated rates of glucose transport were
determined. Data represent the average S.D. of duplicate
samples within a single experiment (n=2).


139
GLUTl (p46 and p37) levels after 24 h (85,140) Again,
chronic insulin under glucose-deprived conditions
accelerated these changes leading to a 2.2-fold increase in
the total GLUTl pool after 12 h (Figure A-l). Because
chronic insulin, alone, increases GLUTl synthesis in the
presence of glucose (Figure A-l), the elevation in GLUTl in
cells exposed to both chronic insulin and glucose
deprivation may result from both decreased degradation and
accelerated synthesis.
An important point to make is that the glycogen pool,
which provides glucose for the glycosylation of proteins
under glucose deprivation, is depleted after 12 h of glucose
deprivation (Figure A-2). In addition, regardless of the
presence of insulin, the glycogen pool is depleted under
conditions of glucose deprivation at the same rate. This
suggests that glucose, derived from glycogen, is diverted
from the hexosamine pathway in the presence of insulin.
Based on these observations, the combination of the
increased synthesis of GLUTl and depleted glycogen pool,
under chronic insulin conditions, likely leads to the
accelerated appearance of the aberrantly glycosylated form
of GLUTl, p37.


54
Figure 3-2. Time Course of Acute Insulin Stimulation.
Cells were incubated in DMEM containing 10% FBS 10 nM
insulin for 12 h. Plates were then washed as described in
Chapter 2 under "Insulin Washout Procedure". Cells were
then stimulated for specific times with 1 |j,M insulin and
glucose transport activity determined. Data represent the
average S.D. of duplicate samples within a single
experiment (n=2). control; chronic insulin.


123
glucose for 1 h prior to the glycogen synthesis assay which
should serve to normalize the intracellular concentrations
of these metabolites. Indeed, I have measured the actual
mass of glycogen synthesized during the time in which cells
are exposed to [14C]-glucose (Table 5-1). Despite the
significant changes in the rates of glycogen synthesis in
response to acute insulin, the mass of glycogen changes very
little during this time. In fact, the only statistically
significant change in mass occurred in cells exposed to
fructose and acute insulin when compared to cells exposed to
fructose alone.
Conclusions
I have further investigated the role that the
hexosamine biosynthetic pathway plays in the induction of
insulin resistance in adipose cells. I have shown that 3T3-
L1 adipocytes, exposed to fructose and insulin for a chronic
period of time, down-regulate their expression of GLUT4 with
a specific loss from the endosomal fraction in a manner
similar to that observed in cells exposed to glucose and
insulin. However, when cells chronically treated with
fructose and insulin were acutely stimulated with insulin,
the amount of GLUT4 in the PM is equal to that seen in
control cells. This is in contrast to cells chronically
exposed to glucose and insulin, which exhibit a 50%


03
ATP Concentration (mM)
O N> G) 00
ATP Concentration
(mM)
o hO CT> 00


15
Raf-1 serine/threonine kinase. This leads to the activation
of MAP kinase kinase which phosphorylates and activates MAP
kinase which activates transcription factors and increases
gene expression (59). It should be noted that
pharmacological concentrations of insulin are required for
activation of gene transcription. This brings doubt to the
physiological relevance of this path to insulin action (60).
The unique aspect of insulin action is its short-term
regulation of metabolic events including increased glucose
transport, increased glycogen synthesis, increased lipid
synthesis, and decreased lipolysis (1). Recent evidence has
shown that the PI 3-kinase cascade is the major pathway for
GLUT4 translocation and thus the stimulation of glucose
transport (61,62). PI 3-kinase is activated when the
phosphorylated IRS-1 binds to the SH2 domain of its p85a
regulatory subunit, this allows the pllO subunit to become
active. Activated PI 3-kinase leads to the activation of
the serine/threonine kinase Akt (also known as protein
kinase B). Although Akt is involved in the specific
stimulation of GLUT4 translocation (62), the exact mechanism
by which this occurs is not known. However, inhibitor
studies has shown the importance of the PI 3-kinase pathway
over the p21rss pathway. For instance, wortmannin, an
inhibitor of PI 3-kinase, blocks the stimulation of glucose
transport by insulin (63). In contrast, inhibition of


10
Figure 1-2. Predicted Secondary Structure of GLUT
Transporters.


91
|ig GLUT4 Ab 0 1 2.5 5 10
GLUT4
Figure 4-1. Titration of Peptide-Purified GLUT4 Antibody
for Immunoprecipitation. Cells were incubated in DMEM
lacking cysteine and methionine and FBS for 1 h. Fresh
DMEM, supplemented with [35S] cysteine/methionine (400 |J.Ci)
was added and the cells incubated for 3 h. A total membrane
fraction was then isolated and GLUT4 was immunoprecipitated
from this fraction using specific amounts of peptide-
purified GLUT4 antibody. Proteins released from the immune
complex were subjected to SDS-PAGE and the radiolabeled
proteins were visualized by autoradiography. The graph
depicts the densitometric analysis of the autoradiogram.
The data represent a single experiment.


130
return to control levels of glucose transport activity.
Cells could then be restimulated acutely with insulin. Only
in this could "true" insulin resistance be measured. Not
only did this procedure allow us to examine glucose
transport activity in these cells, but the translocation of
GLUT4 as well. The insulin washout procedure allowed the
retrieval of GLUT4 from the plasma membrane back into the
endosomal (LDM) fraction. This resulted in very low levels
of GLUT4 in the plasma membrane, as were seen in the basal
state. With acute stimulation with insulin, the
translocation of GLUT4 could thus be monitored. Another
technique essential for measuring translocation of GLUT4 was
the subcellular fractionation procedure, which allowed the
separation of plasma membrane (PM), low density microsomes
(LDM), and high density microsomes (HDM).
The results of these studies showed that insulin
resistance in these cells resulted in a 50% decrease in
insulin-stimulated glucose transport activity. This effect
was both dose- and time-dependent. The development of
insulin resistance required glucose, glutamine, and insulin,
although glucosamine could substitute for glucose and
glutamine as shown also in isolated rat adipocytes
(15,18,20).
Although GLUT4 mRNA was reduced in response to insulin
levels, the K50 was 15 times greater than the equivalent


58
within 8 h of the initial exposure to insulin. This result
is similar to that described in isolated adipocytes (21).
Role of Glucose and Glucosamine
Marshall and his colleagues (15,20,21) have shown in a
series of elegant experiments the requirement of glucose and
glutamine, as well as insulin, for the expression of insulin
resistance in isolated rat adipocytes implicating the N~
acetylglucosamine biosynthetic pathway in this phenomenon.
To test if the same is true in 3T3-L1- adipocytes, I
performed similar experiments. One complication in my
experiments that was not encountered in isolated adipocytes
is the time-dependent activation of glucose transport
activity in the absence of glucose (85-88). This difference
between rat adipocytes and 3T3-L1 adipocytes may result from
higher glycogen stores in the former (89,90) compared with
3T3-L1 adipocytes (79), which provides a metabolic buffer
from external glucose deprivation. We therefore minimized
the time that the cells were exposed to glucose-free medium
but suffered in that only 75% of maximal desensitization was
achieved in these experiments. Importantly, though, the
basal rates of transport were not affected such that true
resistance could be evaluated. Figure 3-4 shows that in the
absence of glucose, insulin was unable to induce
desensitization. Either glucose, in the presence of


6
these mice, they lose the diabetic phenotype. This might
argue against the participation of adipose in the
development of insulin resistance. However, the fact that
obesity leads to insulin resistance demonstrates the
significance of adipose, suggesting that a small amount of
fat is beneficial, but a large excess is detrimental.
Role of the Hexosamine Biosynthetic Pathway in Insulin
Resistance
Marshall and colleagues (18) proposed that the
hexosamine biosynthetic pathway was involved in the
induction of insulin resistance in adipocytes because of the
in vitro requirement for glucose, glutamine, and insulin
(15,20). Figure 1-1 represents a diagram of this pathway.
In this pathway, glucose is transported into the cell by
either GLUT1, the constitutive glucose transporter, and/or
by GLUT4, the insulin-responsive transporter, where it is
rapidly phosphorylated to glucose 6-phosphate. This is then
isomerized to fructose 6-phosphate and the majority (97%) of
this is shunted into glycolysis and oxidative metabolism.
However, 2-3% of the fructose 6-phosphate enters the
hexosamine biosynthetic pathway through the rate-limiting
enzyme, glutamine:fructose 6-phosphate amidotransferase
(GFAT). GFAT transfers the amide group from glutamine to
fructose 6-phosphate to form glucosamine 6-phosphate.
Through several other reactions, UDP-N-acetylglucosamine


116
A
PM
LDM
HDM
Glucose
Fructose
Chronic Insulin
Acute Insulin
+ +
+
+
+
+ +
+ +
+ +
+ +
+ +
+ + +
+ + +
+ + +
GLUT4 *
HNt
B GLUT4
Figure 5-3. Translocation of GLUT4 in Cells Treated with
Fructose. Panel A, cells were incubated under conditions as
described in Figure 5-2. Following the wash, cells were
acutely stimulated, or not, with 1 (J.M insulin for 10 min.
Membrane fractions were isolated and treated as described
above. Panel B, bands were quantitated by video
densitometry. These data are representative of three
independent experiments. II glucose; glucose + acute
insulin; ES9 fructose + chronic insulin; ^3 fructose +
chronic insulin + acute insulin.


48
Figure 2-7. Leptin Standard Curve. Mouse leptin antibody
was added to specific concentrations of leptin provided in
the kit and incubated overnight at 4C. 125I-mouse leptin
was added and incubated at 4C overnight. One mL of cold
precipitating reagent was added and incubated for 20 min at
4C. The precipitate was collected by centrifugation, the
supernatant was decanted and the radioactivity present in
the pellet determined by a gamma counter.


66
acute insulin
tunicamycin
+
+
+
+ +
J L
+
+ +
l
Control
Chronic
Figure 3-7. Effects of Tunicamycin on Insulin Resistance.
Cells were incubated in the presence or absence of 2.5 p,g/mL
tunicamycin also in the presence (chronic) or absence
(control) of 10 nM insulin for 12 h. Cells were washed to
remove insulin and basal and insulin-stimulated rates of
glucose transport were measured as previously described.
Data represent the average + S.D. of duplicate samples
within a single experiment (n=2).


126
the rate of glycogen synthesis accelerated 50-fold with
acute insulin stimulation (Figure 5-6) and with chronic
exposure to 10 nM insulin, the glycogen mass increased by at
least 2-fold (Figure 5-5). After chronic exposure to
insulin in the presence of glucose, the acute stimulation of
glycogen synthesis by insulin decreased by 90%. This is
even more substantial than the loss in insulin-sensitive
glucose transport, indicating additional control steps
beyond glucose availability. Others have shown that
increased flux through the hexosamine pathway decreases
insulin-sensitive glycogen synthase activity (119) Thus,
inhibited glycogen synthase likely contributes to overall
desensitization of glycogen biosynthesis.
Fructose leads to depletion of the glycogen pool in
3T3-L1 adipocytes, as has been shown previously (79),
suggesting that fructose is a poor substrate for glycogen
biosynthesis in these cells. When cells previously exposed
to fructose were subsequently supplied with glucose, both
the basal and insulin-stimulated rates of glycogen synthesis
were significantly higher that cells previously exposed to
glucose. As all cells were "washed" for 1 h in buffer
containing 5 mM glucose prior to the glycogen synthesis
assay, it is unlikely that these differences are related to
alterations in the specific activity of endogenous substrate
pools. Thus, it appears that the rate of glycogen


This is dedicated to God, my parents, my dearest Amy, and
Moxie.


81
Time (h)
Figure 3-16. Reversal of Insulin Resistance. Cells were
incubated in the presence (chronic) or absence (control) of
10 nM insulin for 12 h. Plates were washed to remove
insulin and refed with fresh medium lacking insulin. At
specific times, basal and insulin-stimulated rates of
glucose transport were determined. Data represent the
average S.E. of two independent experiments (n=4).
control; control + acute insulin; A chronic; chronic +
acute insulin.


150
71. Weber, T.M., Joost, H.-G., Simpson, I.A. and Cushman,
S.W. (1988) in: Methods for Assessment of Glucose
Trasport Activity and the Number of Glucose
Transporters in Isolated Rat Adipose Cells and
Membrane Fractions, pp. 171-187 (Kahn, R.C. and
Harrison, L.C., Eds.) Alan R. Liss, Inc., New
York.
72. Balch, W.E. and Rothman, J.E. (1985) Arch. Biochem.
Biophys. 240, 413-425.
73. Markwell, M.A.K., Haas, S.M., Bieber, L.L. and Tolbert,
N.E. (1978) Anal. Biochem. 87, 206-210.
74. Lowry, O.H., Rosenbrough, N.J., Farr, A.L. and Randall,
R.J. (1951) J. Biol. Chem. 193, 265-275.
75. Laemmli, U.K. (1970) Nature. 227, 680-685.
76. Clancy, B.M. and Czech, M.P. (1990) J. Biol. Chem. 265,
12434-12443.
77. Church, G.M. and Gilbert, W. (1984) Proc. Natl. Acad.
Sci. U.S.A. 81, 1991-1995.
78. Feinberg, A.P. and Vogelstein, B. (1983) Anal. Biochem.
132, 6-13.
79. McMahon, R.J. and Frost, S.C. (1996) Am. J. Phys. 270,
E640-E645.
80. Pfleiderer, G. (1963) in: Methods of Enzymatic
Analysis, pp. 91-103 (Bergmeyer, H.U., Ed.)
Academic, New York.
81. Harlow, E. and Lane, D. (1988) pp. 55-77 Cold Springs
Harbor Laboratory, New York.
82. Sigel, M.B., Sinha, Y.N. and VanderLaan, W.P. (1983)
in: Methods in Enzymology, Vol. 93, pp. 3-12
Academic Press, Inc., New York.
83. Dankert, J.R., Shiver, J.W. and Esser, A.F. (1985)
Biochemistry. 24, 2754.
84. Garvey, W.T., Maianu, L., Hancock, J.A., Golichowski,
A.M. and Baron, A. (1992) Diabetes. 41, 465-475.


Figure 3-3. Development of Insulin Resistance. Panel A,
cells were incubated for 12 h with specific concentrations
of insulin as indicated. During this incubation, the
medium was replaced every 2 h. The cells were then washed
for 60 min and glucose transport activity determined
following acute (10 min) stimulation with 1 JJ.M insulin. The
"fractional difference" was determined by subtracting the
glucose uptake rate at 10 nM insulin from the glucose
uptake rate at each point divided by the difference in
uptake rates between 0 and 10 nM insulin. Panel B, cells
were incubated with 10 nM insulin for specific times.
Medium was replaced every 2 h. At appropriate times, the
cells were washed and acutely stimulated with insulin, and
glucose transport activity was measured. The fractional
difference in activity was determined as in panel A. Data
represent the average S.E. of three independent
experiments (n=6).


4
resistance is the metabolic syndrome X. This syndrome
manifests diabetes, hypertension, and dyslipidemia with
obesity as the underlying cause.
Adipose also secretes several factors which may be
involved in insulin resistance. For instance, evidence
exists that adipose tissue of obese mice have elevated TNF-
a mRNA levels and secretion of TNF-a protein (24-26).
Also, administration of TNF-a to Zucker fa/fa rats causes
reduced autophosphorylation of the insulin receptor as well
as decreased insulin-stimulated phosphorylation of IRS-1 in
both muscle and adipose tissue. Neutralization of TNF-a by
the addition of soluble TNF-a receptors prevents these
effects (10) and increases insulin sensitivity (24). It has
also been shown that TNF-a down-regulates GLUT4 mRNA levels
in adipocyte and myocyte cultures (24,27,28) as well as
reduces insulin-stimulated [3H]2-deoxyglucose transport into
3T3-L1 adipocytes (11). These studies provide evidence that
TNF-a plays a role in inducing insulin resistance in
adipose as well as muscle. One can imagine that increased
expression of TNF-a by adipose feeds back directly on
adipose by an autocrine loop mechanism, while its effects on
muscle may occur through a paracrine mechanism (25).
However, it has been shown in db/db mice that the elevated
levels of plasma TNF-a are well below those required to


125
reduction in the level of GLUT4 in the PM upon acute insulin
stimulation. The observation that fructose does not support
the development of insulin-resistant glucose transport
activity is thus explained by this "enhanced" translocation
of GLUT4 to the PM in the presence of fructose.
As many others have shown, insulin stimulates glycogen
biosynthesis. While this process has been studied
mechanistically in adipose, the importance of this glycogen
pool has been dismissed, until recently, because of the
quantitatively small glycogen pools compared to that in
muscle or liver. However, Rigden et al. (124) and Frayn et
al. (90) have provided evidence that as much as 50% of the
glucose extracted by human adipose tissue, during in vivo
perfusion, is stored as glycogen. This suggests that
glycogen may play a potentially important role in adipose
metabolism. In fact, McMahon and Frost (79) have shown that
glycogen provides the carbohydrate for A7-linked protein
glycosylation during short term glucose deprivation of 3T3-
L1 adipocytes. Precursors for oligosaccharides are, in
part, generated through the hexosamine biosynthetic pathway,
the same pathway implicated in the development of insulin
resistance.
In the studies described here, I did observe a decrease
in the rate of glycogen synthesis following incubation of
cells with fructose and chronic insulin.
In control cells,


134
domain, and a unique extended cytoplasmic domain (117).
This cytoplasmic domain contains two dileucine motifs, which
are similar to the motifs found in the carboxy terminus of
GLUT4, and may be responsible for its similar intracellular
trafficking to GLUT4 (117). As the major protein component
of GLUT4-containing vesicles, IRAP could be responsible for
the degradation of GLT4. Although in previous studies, the
aminopeptidase activity of this protein did not change with
acute insulin (115), there may be other regulatory steps.
For instance, longer exposure to insulin may increase the
aminopeptidase activity by causing other factors, not seen
with acute insulin exposures, to affect the activity of this
protein. These factors may explain the data shown in Figure
4-7 in which incubation with cycloheximide during chronic
insulin treatment blocked the loss of GLUT4 in the LDM
fraction. Chronic insulin treatment could stimulate the
synthesis of a protein which increases the activity of this
aminopeptidase causing the degradation of GLUT4. In the
presence of cycloheximide, the synthesis of this protein is
blocked, preventing the activation of the aminopeptidase and
consequently the degradation of GLUT4.
Another area which I believe should be investigated
further is the reversal of insulin resistance. I have shown
that glucose transport activity can be restored in insulin-
resistant cells after 8 h following the removal of insulin


87
translocation to the PM with acute insulin stimulation.
These authors interpreted their data to mean that GLUT4
translocation was blocked, which clearly differs from our
studies. Kozka et al. (16) interpreted their cell surface
ligand binding experiments similarly, even though they
demonstrated a 50% reduction in cell surface GLUT4, which
would agree with my studies. I can only speculate as to the
cause for the different results. In both of these latter
studies, the loss in the GLUT4 pool was determined by
analyzing homogenate protein, revealing only modest changes
in expression. As the translocatable GLUT4 resides in the
LDM fraction, it may be that the loss was substantially
underestimated. Neither study separated the LDM fraction
from the HDM fraction; thus, this possibility can not be
evaluated. Finally, it is important to point out that my
experiments are the first to show that basal transport
activity can be achieved after chronic insulin treatment,
which allowed me to evaluate true insulin resistance. Data
collected under these conditions are consistent with the
hypothesis that the onset of insulin resistance (i.e.
depressed insulin-sensitive glucose transport) is a
reflection of the reduced GLUT4 pool, not a defect in
translocation. In addition, this decrease in insulin-
sensitive glucose transport can be reversed within 8 h and
is protein synthesis dependent.


LIST OF TABLES
Table page
5-1 Glycogen Synthesis in 3T3-L1 Adipocytes 124


124
Glycogen Synthesis in 3T3-L1 Adipocytes
Apparent Rate of
[14C]-Glucose Incorp.
into Glycogen
(nmol/106 cells/h)
Apparent Rate
Change in Response
to Acute Insulin
Glycogen Mass
(p.mol/10s cells)
Control
basal
+ acute insulin
0.57 0.14
29.61 5.53 *
51.9x
0.85 0.30
1.04 0.29
Chronic Insulin
basal
+ acute insulin
0.90 0.44
3.94 0.93|
4.4x
1.88 0.56
2.00 0.53
Fructose
basal
+ acute insulin
11.14 4.05
100.96 27.18*
9.1x
0.33 0.07
0.60 0.22 t
Fructose and
Chronic Insulin
basal
+ acute insulin
23.89 8.33
64.37 18.63$
2.7x
0.72 0.19
0.84 0.22
Table 5-1. Glycogen Synthesis in 3T3-L1 Adipocytes. The
values presented in the "Apparent Rate of [i4C]-Glucose
Incorp. into Glycogen" column are the rates S.D. of
glycogen synthesis found in Figure 5-6. The "Glycogen Mass"
was determined by treating 10-cm plates of 3T3-L1 adipocytes
as in Figure 5-6 without the addition of 14C-U-[D-glucose] .
Total glycogen was then isolated from the cells and treated
as in Figure 5-5. The data from this column represent the
average S.D. of five independent experiments.
Significance of differences: basal versus + acute insulin,
*P<0.001, fP<0.01, JP<0.05 .


42
degradation of GLUT4, equal volumes (750 fo.L) of each sample,
which represents equal cell number, were brought to 1 mL
instead of equal protein. This was done as the total cell
protein content increases over extended periods of chronic
insulin treatment. Non-specific interactions with protein
A-Sepharose were reduced by preincubation of extracted
material with 50 fiL of a 50% slurry of protein A-Sepharose
at 4C for 1 h. The sepharose beads were removed by brief
centrifugation. Five |j,g of peptide-purified GLUT4 antibody
was added to the supernatants. The samples were rotated
overnight at 4C. The next morning, 25 |iL of a 50% slurry
of protein A-Sepharose were added to each sample and
rotation was continued for an additional 2 h at 4C. The
sepharose bead complexes were collected by brief
centrifugation. The supernatant was discarded. The beads
were then washed three times with 1 mL of extraction buffer
for 10 min followed by four, 10-min washes with 1 mL of
extraction buffer containing 1 M NaCl. Samples were then
washed with 1 mL TES buffer. The final pellet was
resuspended in 25 (^L of TES buffer. Sample dilution buffer
(30 |J,L) was added to the beads and incubated at 37C for 30
min. Proteins in the entire sample were resolved on a 10%
polyacrylamide gel by electrophoresis as described above.
The gel was then fixed in 100 mL of 10% (w/v)
trichloroacetic acid and 60% (v/v) methanol for 30 min. The


69
Effect of Chronic Insulin on GLUT4 Expression and
Translocation
A total membrane fraction revealed that insulin-
resistant cells (i.e. cells exposed to 10 nM insulin for 12
h) expressed 2.4-fold less GLUT4 than control cells while
GLUT1 increased by 2.2-fold (Figure 3-10). Another
important point to make as shown by Figure 3-10 is that
acute insulin stimulation did not alter the total levels of
GLUT1 or GLUT4. Only chronic insulin treatment affected the
total amounts of these proteins. To examine the subcellular
distribution of these changes, I used a subcellular
fractionation technique recently developed in this lab (70)
to isolate three membrane fractions: plasma membrane (PM),
low density membranes (LDM), and high density membranes
(HDM). The LDM consists primarily of the small endosomal
storage vesicles containing GLUT4 and the HDM consists
primarily of endoplasmic reticulum and Golgi. Figure 3-11
shows the distribution of GLUT4 and GLUT1 among these three
membrane fractions in control and insulin-resistant cells.
Each set went through the washout procedure prior to
membrane fractionation. Control cells, which were
stimulated acutely with 1 (J.M insulin, showed redistribution
of both GLUT4 and GLUTl; GLUT4 increased by about 6-fold in
the PM (Figure 3-11 A and B) while GLUTl increased by about
2-fold (Figure 3-11 C and D). These data are similar to


93
GLUT4
12 3 4
Figure 4-3. Efficiency of GLUT4 Immunoprecipitation. Total
membranes were collected from duplicate plates (A and B) of
cells. Immunoprecipitation was performed using 5 pig of
peptide-purified GLUT4 antibody (1st round). The
supernatant was saved from this immunoprecipitation and
immunoprecipitation was performed again (2nd round) on this
supernatant. Proteins were resolved by SDS-PAGE and
transferred to nitrocellulose for western blot analysis of
GLUT4. Lanes 1 and 2, immunoprecipitated GLUT4 from the
initial procedure. Lanes 3 and 4, GLUT4 immunoprecipitated
from the supernatant of the first immunoprecipitation. The
graph shows the densitometric analysis of the immunoblot.
These data represent a single experiment.


122
GO
ii
GO
43
GO
13
o
£ 2
m
o
o
O
co
O
o
3
bb
o
a
acute insulin + +
chronic insulin - + +
Figure 5-6. Effects of Fructose and Chronic Insulin on
Glycogen Synthesis. Cells were incubated in DMEM containing
10% FBS or glucose-free DMEM containing 10% dialyzed FBS and
25 mM fructose in the presence or absence of 10 nM insulin
for 12 h. Following "washout" to remove insulin, cells were
labeled with 14C-U-[D-glucose] (2 |j,Ci/plate) in KRP
containing 5 mM glucose and in the presence or absence of 1
(J.M insulin for 1 h. Glycogen was then collected as
described and radioactivity counted by liquid scintillation.
These data represent the average + S.D. of three independent
experiments (n=3). glucose; ¡^j fructose.


121
insulin. Glycogen content of cells incubated with 25 mM
glucose averaged 1.2 |omol of glucose/106 cells. In
contrast, glycogen levels in cells incubated with 25 mM
fructose fell substantially, confirming earlier work by
McMahon and Frost (79). Chronic exposure to glucose and
insulin increased the amount of glycogen by about 2-fold.
Cells chronically exposed to fructose and insulin still
exhibited reduced glycogen although not to the same extent
as in the presence of fructose, alone. When the apparent
rates of glycogen synthesis were investigated under these
conditions, I found that in cells pre-exposed to fructose,
glucose incorporation into glycogen was significantly higher
than in cells pre-exposed to glucose (Figure 5-6). In cells
exposed to glucose, alone, acute insulin treatment
stimulated the rate of glucose incorporation by 50-fold over
controls. Acute insulin exposure of fructose-treated cells
stimulated glucose incorporation by only 9-fold, and yet the
apparent rate was 4 times that in control. After chronic
insulin treatment, subsequent insulin-stimulated glucose
incorporation in glucose-fed cells dropped by 90%. Chronic
insulin exposure to fructose-fed cells reduced the
incorporation by 40%. As the intracellular pool of fructose
6-phosphate, glucose 6-phosphate, and glucose 1-phosphate
were not measured, these rates must not be overinterpreted.
However, the cells were incubated in KRP containing 5 mM


23
r-
Pellet
Scrape 5, 10-cm plates
at 18C with 4 mL TES each

Homogenize in steel block at 18C
(subsequent steps at 4C)
SS-34
12K, 15min
Supernatant
SS-34
12K, 20min
SS-34
20K, 30min
f
1
Pellet
Pellet
Supernatant
Sucrose Cushion
SS-34
T70.1
SW-28
20K, 30min
65K,7 5min
23K, 65min
>
i '
t
Interface HDM Pellet
SS-34
20K, 30min
Ti7 0.1
65K,60min
v
Pellet
LDM
SS-34
20K, 30min
PM
Figure 2-1. Subfractionation of 3T3-L1 Adipocytes.
Overview of subcellular fractionation procedure.


113
in the presence or absence of 10 nM insulin for 6 h. While
it takes at least 8 h for these cells to exhibit full
insulin-resistant glucose transport, I chose 6 h in these
experiments to avoid the accelerated transport rates seen in
the absence of glucose (85). Following the washout
procedure to remove insulin, glucose transport activity was
measured in the presence or absence of acute insulin
stimulation. Glucose in the presence of chronic insulin
induced a 50% reduction in subsequent insulin-stimulated
glucose transport activity compared to glucose controls, as
expected (see Chapter 3). In agreement with Garvey et al.
(15), cells exposed to chronic fructose and insulin did not
show reduced acute insulin-stimulated glucose transport
activity.
To examine the expression of GLUT4, cells were treated
as in Figure 5-1 but for 12 h. Following washout, cells
were then subfractionated to obtain three membrane
fractions: plasma membrane (PM), low density membranes
(LDM), and high density membranes (HDM). Note that the LDM
fraction consists of the small intracellular vesicles
containing the translocatable pool of GLUT4 and the HDM
consists of membrane vesicles from Golgi and endoplasmic
reticulum (70). Figure 5-2 A shows the distribution of both
GLUT4 and GLUT1. In cells incubated in glucose and chronic
insulin, GLUT1 levels were increased, especially in the PM.


102
and leupeptin, and the two proteasome inhibitors,
lactacystin and MG132, were studied. As Figure 4-8 shows,
none of these inhibitors, when incubated with the cells in
the presence of chronic insulin, seemed to prevent the loss
of GLUT4 seen in the LDM fraction. Figure 4-8 A shows the
immunoblot of an experiment testing the effects of
leupeptin. This is representative of the immunoblots
obtained from the other inhibitors tested. The results of
these experiments are displayed in Figure 4-8 B which shows
the densitometric data from the LDM fractions of these
experiments normalized to the control (minus inhibitor and
chronic insulin).
Conclusions
I have shown with these studies that the
immunoprecipitation protocol is specific for GLUT4 and
optimized for complete precipitation from a given extract
with a single round of immunoprecipitation. Using this
technique, I showed that the synthesis of GLUT4 following
chronic insulin treatment was reduced and the degradation
was accelerated in the presence of chronic insulin. This
suggests that the reduction in the total pool of GLUT4 seen
in insulin-resistant cells is a combination of both reduced
synthesis and accelerated degradation.


131
decrease in glucose transport activity. Therefore, the
level of GLUT4, the insulin-responsive glucose transporter,
was examined in resistant cells to determine if changes in
this protein could explain the reduced glucose transport
activity. In fact, GLUT4 levels were reduced 2.4-fold in
insulin-resistant cells. Translocation experiments revealed
that GLUT4 was reduced by 50% in the PM of resistant cells,
reflecting the reduction in the LDM fraction. This proved
that the loss of GLUT4 was responible for the reduced
insulin-stimulated glucose transport activity seen in these
cells. The mechanism responsible for the reduced GLUT4
levels was investigated and found to be a combination of
reduced synthesis and increased degradation of GLUT4
protein.
While the hexosamine biosynthetic pathway has been
implicated in the development of insulin resistance, the
anomalous effect of fructose cast some doubt. As fructose
is the substrate of GFAT, the rate-limiting enzyme in the
hexosamine biosynthetic pathway, incubation with this and
insulin should lead to decreased glucose transport activity,
if the hexosamine bisynthetic pathway is important in
inducing insulin resistance. However, no decrease in
insulin-stimulated glucose transport was observed following
incubation with insulin and fructose. This confirmed
previous data from isolated rat adipocytes (15).


This dissertation was submitted to the Graduate Faculty
of the College of Medicine and to the Graduate School and
was accepted as partial fulfillment of the requirements for
the degree of Doctor of Philosophy.
August 1999
Dean, College of Medicine
Dean, Gradute School


14
located completely on the outer surface of the plasma
membrane and contains the insulin binding site. The
p-subunit has a molecular mass of 95,000 Da and contains a
single transmembrane domain. The intracellular domain of
this subunit contains a juxtamembrane domain, an ATP binding
domain, and several tyrosine residues that are capable of
being phosphorylated. The intracellular domain also
contains a tyrosine kinase that is activated upon binding of
insulin presumably via propagation of a conformational
change from the a-subunit, which is transmitted through the
transmembrane domain of the p-subunit. Once the tyrosine
kinase has been activated, it is capable of
autophosphorylating tyrosines in specific regions of the
intracellular domain of the P-subunit including the
juxtamembrane domain, the regulatory domain, and the carboxy
terminal tail (1). This tyrosine kinase also phosphorylates
selected proteins on tyrosine residues within the cell such
as IRS-1. The phosphorylated IRS-1 recognizes and binds to
the src homology-2 (SH2) domains of various signal
transduction proteins, two of which are Grb2 and PI 3-
kinase. The activation of these molecules sets up divergent
signaling pathways within the cell.
Activation of Grb allows it to bind to son-of-sevenless
(SOS) through an SH3 domain. The Grb2/SOS complex then
activates p21ras which has been shown to bind directly to


64
3
P-: O'
& .s
0)
on
O
o
2
02
o
g 2
(D -th
9 I
W W
cn
acute insulin
20|iM Nikkomycin
200 |iM Nikkomycin
+ + +
+ +
+ +
i
control
+ + +
+ +
+ +
chronic
Figure 3-6. Effects of Nikkomycin Z on Insulin Resistance.
Cells were incubated in the presence or absence of specific
concentrations of nikkomycin Z also in the presence
(chronic) or absence (control) of 10 nM insulin for 12 h.
Plates were washed to remove insulin and basal and insulin-
stimulated rates of glucose transport were measured as
previously described. Data represent the average S.D. of
duplicate samples within a single experiment (n=2).


90
isolating GLUT4 had to be optimized. The first step was to
determine the appropriate amount of peptide-purified GLUT4
antibody required for the complete precipitation of GLUT4.
Figure 4-1 shows that 5 pig of antibody is the optimal amount
for maximal GLUT4 recovery.
It was also necessary to prove that the protein that
was being precipitated by the antibody was indeed GLUT4.
Figure 4-2 addresses this question by showing that the
protein precipitated could be competed by an equal molar
amount of GLUT4 peptide (compare lanes 3 and 4). This
proves that the peptide-purified GLUT4 antibody is
precipitating GLUT4 and that this interaction is specific.
Finally, the last step in optimizing the
immunoprecipitation procedure was to show that a single
"round" of immunoprecipitation could precipitate all of the
GLT4 in the sample. Figure 4-3 shows that a successive
round of immunoprecipitation did not yield significant
amounts of GLUT4 (see lanes 3 and 4). This proves that one
round of immunoprecipitation is sufficient to isolate 98% of
the GLUT4 in the sample.
Synthesis of GLUT4 in Insulin-resistant Cells
The synthesis of GLUT4 in control and insulin-resistant
cells was examined to determine if a reduced synthesis in
resistant cells could explain the reduced levels of GLUT4


143
leptin secreted from these cells. I can only hypothesize
that differences in experimental procedure have led to such
a wide range of reported concentrations of leptin secreted
by 3T3-L1 adipocytes.


44
in a microfuge for 5 min at 4C. The supernatant was
diluted 1:250.
Ten |uL of the 1:250 diluted sample or ATP standard was
added to a plastic 12 x 75 mm tube. To this, 500 p,L of
assay buffer (60 mM Tris-Acetate, pH 7.75, 10 mM MgCl2, 1 mM
KC1, 1.5 mM EDTA, 2.5 mM p-mercaptoethanol, 0.4 mg/mL
luciferin/luciferase stock) were added. The p-
mercaptoethanol was added after the initial solution was
brought to pH 7.75. The luciferin/luciferase stock was
prepared by adding 50 mg of luciferin/luciferase to 2.5 mL
of 50% glycerol. This was divided into 300 (J,L aliquots and
stored in liquid nitrogen and added to the assay buffer just
prior to use. Following addition of the assay buffer to the
sample, the tube was flicked for 10 sec then placed in the
luminometer and the fluorescence measured for 20 sec. ATP
in the samples was calculated by comparison against the
standard curve seen in Figure 2-6.
Leptin Assay
Cells (5, 10-cm plates) were fed with serum-free DMEM
for 12 h. The medium was then collected, pooled from the
five plates (a total of 40 mL), and concentrated using the
"Centriplus 3" centrifugal concentrating tubes from Amicon,
which have a molecular weight cut-off of 3,000 Da. The


70
Acute Insulin
Chronic Insulin + +
GLUT1 +
GLUT4
Figure 3-10. Effects of Acute and Chronic Insulin on Total
GLUT Protein Levels. Equal protein of the PM, LDM, and HDM
fractions of Figure 3-11 were combined and subjected to SDS-
PAGE and transferred to nitrocellulose. Immunoblot
detection of GLUT1 and GLUT4 was carried out using C-
terminal specific antibodies. The protein-antibody complex
was visualized by enhanced chemiluminescence.


82
this experiment. As is shown in Figure 3-17, cycloheximide
completely blocked the reversal of insulin-resistant glucose
transport activity. It should be noted that cycloheximide
decreased the insulin-stimulated rate of glucose transport
in control cells (closed squares versus inverted triangles).
However, the cycloheximide-induced block in reversal is
considerably larger (open squares versus inverted
triangles). This
suggests that the reversal of insulin-
resistant glucose
transport is dependent upon protein
synthesis. It is
not clear, however, if this reversal is
due to the synthesis and recovery of GLUT4, as those studies
have not yielded reliable results.
Conclusions
In the above
studies, I have tested the hypothesis that
3T3-L1 adipocytes
can serve as a model for studying the
development of insulin resistance under conditions that
might be realized in a physiological setting. Support for
this hypothesis has been gained from the following
observations. Chronic exposure to physiological levels of
insulin decreased the ability of an insulin challenge to
stimulate glucose transport. Interestingly, postprandial
concentrations of insulin in normal, obese, and diabetic
humans (84) plot along the inflection in the dose-response
curve between no change in insulin responsiveness and that


110
paradigms on glycogen synthesis. Chronic exposure of cells
to fructose, alone, accelerated the rate of glycogen
synthesis significantly over cells exposed to glucose,
alone. This difference appears to be correlated, in part,
to the size of the glycogen pool. Subsequent exposure to
insulin stimulated synthesis in both sets of cells. With
the additional exposure to chronic insulin, insulin
resistance developed in the presence of either glucose or
fructose although to different degrees. The level of
insulin-stimulated glycogen synthesis in cells exposed to
glucose and chronic insulin fell 90% compared to the
controls, while the level of synthesis in cells exposed to
fructose and insulin fell to only 50% compared to controls.
This suggests that the development of insulin resistance in
the glycogen synthetic path, while observed in both glucose-
and fructose-treated cells, appears to be greater in cells
exposed to glucose and chronic insulin.
Results
Effects of Fructose Feeding on the Glucose Transport System
To examine the potential role of fructose in the
development of insulin-dependent, insulin resistance, I have
first examined its effect in transport activity assays
(Figure 5-1). In these experiments, 3T3-L1 adipocytes were
incubated in DMEM containing 25 mM glucose or 25 mM fructose


37
solution at 280 nm using a milligram extinction coefficient
of 1.4.
GLUT4 antibody was purified from the total IgG solution
by peptide chromatography. For this technique, we used a
SulfoLink Kit from Pierce. This kit contains a column
composed of 6% agarose beads cross-linked to iodoacetate.
The acetyl group binds irreversibly to free sulfhydryl
groups. Two mg of the GLUT4 peptide were bound to this
matrix through the terminal cysteine residue following the
manufacturer's instructions. GLUT4-specific antibody was
then purified by placing 2-3 mL of the total IgG solution on
the column and rotating at room temperature for 2 h. The
column was then washed with 16 mL of PBS and the GLUT4
antibody was eluted with 8 mL of 0.1 M glycine, pH 3.0. One
mL fractions were collected in tubes containing 100 (J.L 1 M
Tris-base. The absorbance of the fractions was measured at
280 nm and fractions containing antibody were pooled.
Figure 2-5 shows a typical elution profile of peptide-
purified anti-GLUT4 antibody. The concentration of the
resulting antibody solution was determined by its absorbance
at 280 nm using a milligram extinction coefficient of 1.4.
The purified GLUT4 antibody was then stored as 0.1 mL
aliquots at -20C.


BIOGRAPHICAL SKETCH
Michael James Thomson was born in Kenmore, New York, in
March 1972, to James and Donna Thomson. He is one of four
children, a brother and sister of which, including Michael,
make up a set of triplets, and an older brother. He was
raised in Niagara Falls, New York, until he moved with his
family to Lake City, Florida, in the summer of 1985. He
attended Lake City Community College and graduated with an
associate of arts degree in 1992. He continued his
education at the University of Florida, where he graduated
with a bachelor of science degree in Zoology in 1994 and a
doctor of philosophy in biochemistry and molecular biology
in 1999. He will now attend the School of Veterinary
Medicine at the University of Florida.
156


[ H] 2-Deoxyglucose Uptake
112
acute insulin + + + +
chronic insulin + + + +
glucose + + + +
fructose
+ + + +


50
required to affect a 50% change in expression of message was
reported as 23 nM. This level of insulin is at least 2
orders of magnitude higher than the concentration of
circulating insulin in humans (2). No comparable dose-
response studies have examined the development of insulin-
resistant glucose transport activity or GLUT4 expression in
these cells. Therefore, I have developed a procedure that
has allowed the measurement of glucose transport after
chronic exposure (12 h) to physiological concentrations of
insulin.
Results
Insulin-resistant Glucose Transport Activity
Chronic insulin exposure elevates glucose transport
activity in 3T3-L1 adipocytes. This can be prevented by
protein synthesis inhibitors like cycloheximide (Risch and
Frost, unpublished data). More recent studies have shown
that pharmacological doses of insulin induces GLUT1
transcription (14) leading to elevated GLUT1 protein and
thus transport activity. In the face of elevated GLUT1,
insulin-sensitive glucose transport activity can not be
observed. While this may imply insulin resistance, elevated
basal transport activity complicates interpretation. My
first challenge was to develop a protocol to reestablish
basal transport activity after chronic exposure to insulin.


149
57. Rea, S., Martin, L.B., McIntosh, S., Macaulay, S.L.,
Ramsdale, T., Baldini, G. and James, D.E. (1998)
J. Biol. Chem. 273, 18784-18792.
58. Thurmond, D.C., Ceresa, B.P., Okada, S., Elmendorf,
J.S., Coker, K. and Pessin, J.E. (1998) J. Biol.
Chem. 273, 33876-33883.
59. White, M.F. and Kahn, C.R. (1994) J. Biol. Chem. 269,
1-4 .
60. Saltiel, A.R. (1996) Am. J. Physiol. 270, E375-E385.
61. Haruta, T., Morris, A.J., Rose, D.W., Nelson, J.G.,
Mueckler, M. and Olefsky, J.M. (1995) J. Biol.
Chem. 270, 27991-27994.
62. Kohn, A.D., Summers, S.A., Birnbaum, M.J. and Roth,
R.A. (1996) J. Biol. Chem. 271, 31372-31378.
63. Okada, T., Kowano, Y., Sakakibara, T., Hazeki, O. and
Ui, M. (1994) J. Biol. Chem. 269, 3568-3573.
64. Ricort, J.-M., Tanti, J.-F., Van Obberghen, E. and Le
Marchand-Brustel, Y. (1995) Diabetologia. 38,
1148-1156.
65. Robinson, L.J., Razzack, Z.F., Lawrence, J.C., Jr. and
James, D.E. (1993) J. Biol. Chem. 268, 26422-
26427.
66. Lin, T., Kong, X., Saltiel, A.R., Blackshear, P.J. and
Lawrence, J.C., Jr. (1997) J. Biol. Chem. 270,
18531-18538.
67. Wiese, R.J., Mastick, C.C., Lazar, D.F. and Saltiel,
A.R. (1995) J. Biol. Chem. 270, 3442-3446.
68. Frost, S.C. and Lane, M.D. (1985) J. Biol. Chem. 260,
2646-2652.
69. Rubin, C.S., Hirsch, A., Fung, C. and Rosen, O.M.
(1978) J. Biol. Chem. 253, 7570-7578.
70. Fisher, M.D. and Frost, S.C. (1996) J. Biol. Chem. 271,
11806-11809.


38
Fraction
Figure 2-5. Elution Profile of Peptide-Purified anti-GLUT4
Antibody. Total IgG was added to the column and rotated for
2 h. Unbound antibody was washed from the column using PBS.
GLUT4 antibody was eluted using 0.1 M glycine, pH 3.0. One
mL fractions were collected and neutralized with 1 M Tris-
base. Absorbance of the fractions was measured at 280 nm
and the concentration determined using a milligram
extinction coefficient of 1.4.


127
biosynthesis is inversely related to the size of the
glycogen pool. This phenomenon has been observed in rat
skeletal muscle in which the rate of glycogen synthesis is
highest in those muscle subtypes which manifest the greatest
extent of glycogen depletion during food deprivation (125).
From a clinical perspective, are there lessons to be
learned from these studies? While my studies in 3T3-L1
adipocytes support the ability of fructose to prevent
insulin resistance, it must be pointed out that the
available pool of GLUT4 is substantially reduced in cells
exposed to chronic fructose and insulin despite the normal
sensitivity. The long term consequences of this are
unknown. One could argue that the high concentration used
in these studies is not likely achievable in vivo because of
the efficiency with which the liver absorbs fructose from
the portal system (126). Thus, even with high dietary
fructose, a maximal level of only 2.2 mM has been observed
in the human systemic circulation which is approximately 10
times less than the concentration used in this study.
However, preliminary experiments in 3T3-L1 adipocytes using
2 mM fructose with chronic insulin resulted in 30% down-
regulation of GLT4 in the LDM fraction (data not shown).
Studies in animal models have actually shown that
fructose administration leads to insulin resistance (127-
130). However, the diets in these studies derive 35-66% of


84
of maximal resistance. Thus, I have shown for the first
time that 3T3-L1 adipocytes develop insulin resistance in
response to physiologically relevant concentrations of
insulin. I have extended previous work by demonstrating
that insulin challenge of resistant cells stimulates
translocation, although the extent of recruitment is
suppressed relative to controls due to the reduction in the
total expression of GLUT4. Together, these data suggest
that the loss of GLUT4 protein underlies the inability of
3T3-L1 adipocytes to respond to insulin after chronic
exposure. This mimics the clinical manifestation of human
obesity and non-insulin-dependent diabetes where loss of
GLUT4 protein has been observed in adipose tissue (96),
although not in muscle (97).
It should be pointed out that transporter expression
differs in adipose tissue relative to 3T3-L1 adipocytes. In
isolated rat adipocytes, GLUT4 represents 97% of the GLUT
transporter pool (43). In 3T3-L1 adipocytes, GLUT4
represents only 33% of the pool (98), indicating the
substantially higher expression of GLUTl relative to GLUT4
in this cell line. In control 3T3-L1 adipocytes, the PM
fraction contains about 25% of the GLUTl pool. Chronic
insulin treatment increases the total pool of GLUTl, which
in turn doubles the GLUTl content of the PM fraction.
Despite this 2-fold increase in GLUTl in the PM of resistant


CHAPTER 4
MECHANISMS OF DECREASED GLUT4 LEVELS
IN INSULIN-RESISTANT CELLS
Introduction
As described in Chapter 3, I showed that a decrease in
GLUT4 protein was responsible for the decrease in insulin-
sensitive glucose transport activity. Reduced levels of
GLUT4 could be explained by one of several mechanisms. One
possibility is that GLUT4 mRNA levels are decreased leading
directly to reduced GLUT4 protein. While I have shown this
to be true, the K5o for the decrease in GLUT4 mRNA is 15
times greater than the concentration of insulin required for
the loss in insulin-sensitive glucose transport activity and
10 times greater than that for GLUT4 expression. A second
possibility is that the synthesis of GLUT4 protein is
decreased under chronic insulin conditions, despite
equivalent mRNA template. In this case, reduced
translational efficiency would lead to a diminished GLUT4
pool. A third possibility is that the degradation of GLUT4
is accelerated under these conditions again decreasing the
cellular levels of GLUT4 protein. These latter two
88


120
Figure 5-5. Effect of Fructose and Chronic Insulin on
Glycogen Levels. Cells were incubated with DMEM containing
10% FBS or glucose-free DMEM containing 10% dialyzed FBS and
25 mM fructose in the presence or absence of 10 nM insulin
for 12 h. Cells were collected and disrupted by sonication.
Glycogen was isolated by ethanol precipitation then
hydrolyzed using 2 N H2S04 and the concentration of glucose
determined by a commercially available kit (Sigma). Data
represent the average S.D. of five experiments.
glucose; E3 glucose + chronic insulin; ESS fructose; SS3
fructose + chronic insulin.


45
0 12 3
pmol ATP
Figure 2-6. ATP Standard Curve. Ten |4l of specific
concentrations of ATP were added to 12 x 75 mm plastic
tubes. The luciferin/luciferase assay buffer was added, the
tube was flicked for 10 sec., and the fluorescence measured
in a luminometer for 20 sec.


ACKNOWLEDGMENTS
I would sincerely like to thank the members of my
committee Drs. Michael Kilberg, Charles Allen, Thomas
O'Brien, Mark Atkinson, and my advisor Dr. Susan Frost for
their guidance throughout the course of this project. In
addition, I would like to thank the members of the lab and
my friends for their help throughout the time I have spent
here and for making this such an enjoyable period in my
life. My most heartfelt thanks goes to my parents, without
whose love and support I would never have achieved such an
undertaking. Last, but certainly not least, I would like to
thank my dearest Amy, whose love and companionship over the
last several months has not only marked a highlight in my
graduate career, but in my life as well.


CHAPTER 6
CONCLUSIONS AND
FUTURE DIRECTIONS
Conclusions
Insulin resistance is a hallmark of diabetes, a disease
which affects hundreds of thousands of people in the United
States and a vastly growing population around the world.
Unfortunately, type II diabetes is the most common and least
understood form of this disease. The role that adipose
plays in this disease has been underappreciated until
recently. The majority of individuals with type II diabetes
have this disease because they are obese. While this is a
strong statement, studies show that losing weight reverses
the effects of this disease.
The studies described herein examined the development
of insulin resistance in a stable adipose cell line. These
studies have shown that, indeed, the 3T3-L1 adipocytes
developed insulin resistance within physiological
concentrations of insulin. Essential to this discovery was
the experimental design that returned cells to a "basal"
state with an extensive insulin washout procedure. This
procedure removed insulin from the cells and allowed the
129


24
protein determination by the Lowry method (73). Therefore,
the Markwell procedure, which is a modification of the Lowry
procedure, was used for the analysis of membrane proteins in
our specific extraction buffer, which contains both sucrose
and EDTA (73).
A standard curve (0-100 pig) was generated by diluting a
solution of BSA (1 mg/mL) in water to a final volume of 0.1
mL. Ten piL of membranes suspended in TES was added to 0.1
mL of water. Ten piL of TES was also added to the standard
curve. One mL of a solution containing 2.0% Na2C3, 0.4%
NaOH, 0.16% Na+ K+ tartrate, and 1% SDS was added to the
standard curve and unknown samples, mixed, and allowed to
stand for 10 min. Folin reagent (0.1 mL of 1 N) was added,
mixed gently, and allowed to stand for 45 min. The
absorbance of the solution was measured at 650 nm and the
protein concentration of the samples was determined by
comparing them to the standards. A second order equation
was used to solve for the constants and the quadratic
equation was used to determine the protein concentration of
the unknown samples.
Gel Electrophoresis
Electrophoresis was performed as described by Laemmli
et al. (75) Equal amounts of membrane protein (50 jag) were
mixed with half the volume of 2X Laemmli sample dilution


61
glutamine (Figure 3-4 A), or glucosamine, in the absence of
both glucose and glutamine (Figure 3-4 B) provided
appropriate substrate for the development of insulin
resistance.
Effects of Various Inhibitors on the Development of Insulin
Resistance
Inhibitors of DNA transcription, protein synthesis,
glycosylation and proteases were used to characterize the
development of insulin resistance in these cells. Cells
incubated for 12 h with 4 p,M actinomycin D, an inhibitor of
DNA transcription, exhibited an increase in basal glucose
transport activity and a reduction in insulin-stimulated
glucose transport in both control and chronically-treated
cells (Figure 3-5). The underlying mechanism for this
elevation is unknown. Given elevated basal activity, the
effect of actinomycin D on insulin-sensitive transport is
difficult to determine, although changes in the presence and
absence of acute insulin were fairly small. In contrast,
studies in isolated rat adipocytes (91) showed that
incubation with up to 200 nM actinomycin D for 18 h actually
decreased basal levels. However, in these studies,
actinomycin D prevented the development of insulin
resistance. It is believed that actinomycin D prevented
insulin resistance in these studies by inhibiting GFAT
activity, as insulin resistance induced by incubation with


APPENDIX A
GLUCOSE DEPRIVATION AND GLUT TRANSPORTERS
Studies examining the requirement of glucose for the
development of insulin resistance led to studies
investigating the expression of the GLUTS in response to
glucose deprivation. Previously, our lab has shown that
glucose deprivation, in the absence of chronic insulin
treatment, results in the appearance of the aberrantly
glycosylated form of GLUT1 after 18 h (85). This aberrantly
glycosylated form was named p37 because it migrates as a 37
kDa protein compared to the normal glycosylated form of
GLUT1 which migrates as a 46 kDa protein. In the presence
of chronic insulin, p37 exhibits significant expression
after only 12 h (Figure A-l). Under these conditions, the
distribution between p46 and p37 across the three membrane
fractions was 86% and 14% respectively. Interestingly, no
lower molecular weight form of GLUT4 was seen in the
presence or absence of chronic insulin with glucose
deprivation.
Although the half-life of GLUT1 under glucose
deprivation increases from 14 h to greater than 50 h (140)
the rate of synthesis does not change resulting in increased
137


77
3-13 are the percentages of GLUT4 in each sample compared to
the 0 nM insulin control as determined by densitometric
analysis. The dose dependence of this down-regulation
(Figure 3-12 B and Figure 3-13) yielded a K50 of about
600pM, slightly higher than the K50 of insulin-resistant
glucose transport, but a technically more difficult
parameter to measure.
Effect of Glucose Deprivation on GLUT4 Expression
Based on the observation that glucose deprivation
prevented the loss in insulin sensitivity (see Figure 3-4
A), we examined the expression of GLUT4 in the LDM fraction
of cells exposed to glucose-free medium. Figure 3-14 shows
that glucose deprivation blocked the loss of GLUT4 in
chronically treated cells. Thus, this shows for the first
time that glucose is important in regulating the expression
of GLUT4 in response to chronic insulin.
Effect of Chronic Insulin on GLUT4 mRNA Expression
These studies were conducted by Martin Williams to
evaluate the underlying mechanism of the reduction in GLUT4
protein. The level of GLUT4 mRNA was measured after
exposure to specific concentrations of insulin. As shown in
Figure 3-15 A, the level of GLUT4 mRNA decreases with
increasing insulin concentration. However, the


Figure 5-4. Effects of Fructose on ATP Levels. Panel A,
cells were incubated with DMEM containing 10% FBS or
glucose-free DMEM containing 10% dialyzed FBS and 25 mM
fructose in the presence or absence of 10 nM insulin for 12
h. Cells were then collected and ATP concentrations
determined by a luciferin/luciferase assay as described.
Data represent the average S.D. of three independent
experiments (n=6) control; E23 chronic insulin; ¡S33
fructose; E-S3 fructose + chronic insulin. Panel B,
iodoacetate (1 mM) was added to control cells and incubated
for 30 min. ATP was then determined as described
previously. Data represent a single experiment.


146
11. Hotamisligil, G.S., Murray, D.L., Choy, L.N. and
Spiegelman, B.M. (1994) Proc. Natl. Acad. Sci.
U.S.A. 91, 4854-4858.
12. Hotamisligil, G.S., Peraldi, P., Budavari, A., Ellis,
R., White, M.F. and Spiegelman, B.M. (1996)
Science. 271, 665-668.
13. Maddux, B.A. et al. (1995) Nature. 373, 448-451.
14. Flores-Riveros, J.R., McLenithan, J.C., Ezaki, 0. and
Lane, M.D. (1993) Proc. Natl. Acad. Sci. U.S.A.
90, 512-516.
15. Garvey, W.T., Olefsky, J.M., Matthaei, S. and Marshall
S. (1987) J. Biol. Chem. 262, 189-197.
16. Kozka, I.J., Clark, A.E. and Holman, G.D. (1991) J.
Biol. Chem. 266, 11726-11731.
17. Mller, G., Dearey, E.A. and Pnter, J. (1993) Biochem
J. 289, 509-521.
18. Marshall, S., Bacote, V. and Traxinger, R.R. (1991) J.
Biol. Chem. 266, 4706-4712.
19. Marshall, S., Garvey, W.T. and Traxinger, R.R. (1991)
FASEB. 5, 3031-3036.
20. Traxinger, R.R. and Marshall, S. (1989) J. Biol. Chem.
264, 20910-20916.
21. Traxinger, R.R. and Marshall, S. (1991) J. Biol. Chem.
266, 10148-10154.
22. Su, H.-Y., Sheu, W.H.-H., Chin, H.-M.L., Jeng, C.-Y.,
Chen, Y.-D.I. and Reaven, G.M. (1995) Am. J.
Hyper. 8, 1067-1071.
23. Colman, E., Katzel, L.I., Rogus, E., Coon, P., Muller,
D. and Goldberg, A.P. (1995) Metabolism. 44, 1502
1508 .
24. Hotamisligil, G.S., Shargill, N.S. and Spiegelman, B.M
(1993) Science. 259, 87-91.
25. Hotamisligil, G.S. and Spiegelman, B.M. (1994)
Diabetes. 43, 1271-1278.


30
on the days of a scheduled feeding as they became glucose
deprived. In addition, glycogen levels increased over the
course of the week from day 1 to day 5. This shows the
importance of performing a glycogen experiment the day
following a feeding and also performing the experiment at
the same time during the week (usually day 2 or 3 following
a feeding) to obtain consistent data.
Glycogen Synthesis
Cells adherent to 35-mm dishes were extensively washed
with KRP containing 0.1% BSA and 5 mM glucose as described
under "Insulin Washout" to return the cells to a basal level
of glucose transport. These cells were then labeled with 2
[iCi/plate of 14C-U-[D-glucose] in 1.5 mL KRP containing 5 mM
glucose (giving a final specific activity of 284 (j,Ci/mmol)
with or without 1 |j,M insulin for 1 h. Glycogen was then
collected as described above and the pellet dissolved in 300
[j,L of water. Radioactivity present was determined by liquid
scintillation.
Production and Characterization of GLUT4 Antibody
Hydrophilicity of a peptide is a criterion for
selecting a suitable sequence for antibody production
because hydrophilic peptides are more likely to be soluble
for coupling reactions and also more likely to be exposed on


Figure 5-1. Effects of Fructose on the Development of
Insulin Resistance. Cells were incubated in DMEM
containing 10% FBS or in glucose-free DMEM containing 10%
dialyzed FBS and 25 mM fructose in the presence or absence
of 10 nM insulin for 6 h. The cells were then washed as
described in Chapter 2. Basal and insulin-simulated rates
of glucose transport were measured by incubating the cells
in KRP with 1 |o,M insulin, or not, for 10 min followed by the
addition of 200 |o.M [3H] 2-deoxyglucose (0.2 jnCi) After 10
min, transport was terminated by the addition of ice-cold
phosphate-buffered saline. Cells were lysed with a 0.1%
solution of SDS and duplicate aliquots of 300 (J.L were taken
and [3H]2-deoxyglucose uptake determined by liquid
scintillation counting. The data represent the average
S.E. of three independent experiments (n=3).


CHAPTER 5
ROLE OF THE
HEXOSAMINE BIOSYNTHETIC PATHWAY
IN THE DEVELOPMENT OF INSULIN RESISTANCE
Introduction
Marshall and his colleagues showed that glucose and
glutamine were required for the insulin-dependent
development of insulin-resistant glucose transport activity
in isolated rat adipocytes (15,20,21) As shown in Chapter
3, I have confirmed this in the 3T3-L1 adipocyte cell line
(118). Because glucose (via its metabolism to fructose 6-
phosphate) and glutamine are co-substrates of glutamine,
fructose 6-phosphate amidotransferase (GFAT), this
implicates the intermediates or products of the hexosamine
biosynthetic pathway in the development of insulin
resistance. The hexosamine biosynthetic pathway is
responsible for the synthesis of N-acetylglucosamine, one of
the sugars used in the post-translational modification of
proteins.
Several lines of evidence support the Marshall
hypothesis. First, glucosamine appears to desensitize
adipocytes to the actions of insulin at lower concentrations
than does glucose in both isolated adipocytes (18)
107


144
Figure B-l. Leptin Levels. Cells were incubated in
glucose- and serum-free medium containing 25 mM glucose or
fructose and 10 nM insulin for 12 h. Medium was collected
from the plates and concentrated using concentrator spin
columns (Amicon). Leptin present in the medium was then
determined by an 125I-leptin radioimmunoassay kit from Lineo.
Data represent a single experiment. CD glucose; 1231 glucose +
chronic insulin; ¡S9 fructose; ES3 fructose + chronic insulin.


94
present in these cells. Following the onset of insulin
resistance, cells were radiolabeled with
[35S]cysteine/methionine for specific times and a total
membrane fraction collected. GLUT4 was then
immunoprecipitated from this membrane fraction and samples
subjected to SDS-PAGE and autoradiography. As Figure 4-4
shows, at each time point, less GLUT4 was synthesized in the
insulin-resistant cells compared to the controls. However,
the synthesis in these cells is not reduced by 50% as seen
by the densitometric analysis. Therefore, this suggests
that the reduction in the total pool of GLUT4 in insulin-
resistant cells is only in part due to reduced synthesis of
GLUT4.
Degradation of GLUT4 in Insulin-resistant Cells
The degradation of GLUT4 under conditions of chronic
insulin was then examined to determine if accelerated
degradation of GLUT4 in insulin-resistant cells could
contribute to the reduced pool in these cells. In order to
achieve equal incorporation of [35S]cysteine/methionine into
both control and resistant cells, cells were radiolabeled
prior to incubation with insulin. Cells were then chased
with complete medium in the presence or absence of insulin
for up to 72 h. Total membranes were collected at specific
times and GLUT4 immunoprecipitated as above. Figure 4-5


26
RNA Isolation and Northern Blotting
RNA isolation and northern blotting was performed by
Martin Williams. Total cellular RNA was isolated by the
guanidinium thiocyanate procedure (77) Twenty fig of total
RNA was loaded onto a 1% formaldehyde-agarose gel and the
gel was run for 12-16 h at 40 V with constant buffer
recirculation. The RNA was then electrophoretically
transferred to an uncharged nylon membrane and cross-linked
with UV light for 3.5 min. GLUT4 probe was then generated
from cDNA generously provided by Dr. Maureen Charron (Albert
Einstein College of Medicine). The insert was labeled by
primer extension (78). Briefly, the DNA (0.5 pig) was
denatured by boiling at 100C for 5 min. After cooling for
3 min, the insert was incubated with a 6-base primer,
deoxynucleotides, one of which was labeled ([32P]dATP, dCTP,
dTTP, dGTP), and DNA polymerase. Free nucleotides were
removed by gel chromatography over Sephadex G-50. The
eluted labeled DNA was used directly to probe the RNA blot.
Hybridization was performed for 12-16 h in a buffer
containing 1% bovine serum albumin, 1 mM EDTA, 0.5 M sodium
phosphate, pH 7.2, and 1% SDS at 60C. The membranes were
washed under high stringency conditions (3 X 10-min washes
in 1 mM EDTA, 40 mM sodium phosphate, pH 7.2, and 1% SDS at
65C) and exposed to film for various lengths of time.


5 ROLE OF THE HEXOSAMINE BIOSYNTHETIC
PATHWAY IN THE DEVELOPMENT OF
INSULIN RESISTANCE 107
Introduction 107
Results 110
Effects of Fructose Feeding on the
Glucose Transport System 110
Effect of Fructose Feeding on
Glycogen Metabolism 117
Conclusions 123
6 CONCLUSIONS AND FUTURE DIRECTIONS 129
Conclusions 129
Future Directions 132
APPENDICES
A GLUCOSE DEPRIVATION AND GLUT TRANSPORTERS 137
B LEPTIN AND INSULIN RESISTANCE 141
REFERENCES 145
BIOGRAPHICAL SKETCH 156
vi


5
support insulin resistance in cultured cell systems (24,25).
Therefore, it is very plausible that other proteins or
factors may be secreted by the adipose tissue which induce
insulin resistance.
Fatty acids have also been shown to be involved in the
induction of insulin resistance. In 1963, Randle et al.
(29) hypothesized that increased serum levels of free fatty
acids could interfere with glucose uptake and metabolism in
muscles, resulting in insulin resistance. This has come to
be known as the Randle hypothesis. Since the presentation
of this hypothesis, it has been shown by several groups,
including Randle's, that elevated serum free fatty acids
interfere with glucose utilization in vitro and in vivo (30-
33). In further support of this hypothesis, Bierman et al.
(34) have shown that type II diabetics have elevated fasting
levels of free fatty acids. In contrast, the absence of
white adipose tissue can also lead to diabetes as shown by
studies using a transgenic mouse model (35). These mice
express a dominant negative protein, called A-ZIP/F, which
prevents the DNA binding of B-ZIP transcription factors of
both the C/EBP and Jun families, which are necessary for
white adipose differentiation and formation. These mice are
diabetic, with fatty livers and elevated serum glucose,
insulin, and surprisingly, free fatty acids and
triglycerides. Interestingly, when fat is transplanted into


32
linking reagent, sulfo-m-maleimidobenzoyl-N-
hydroxysuccinimide ester (sulfo-MSB). KLH (60 mg) was
dissolved in 10 mL of 10 mM potassium phosphate, pH 7.3, and
dialyzed at 4C against 10 mM potassium phosphate for 48 h.
The dialysate was concentrated by coating the dialysis
tubing with a liberal amount of Aquacide and wrapping in
foil. Protein aggregates were removed from the concentrated
dialysate by centrifugation in a microfuge for 10 min. The
concentration of protein in the dialysate was determined by
measuring its absorbance at 280 nm using the milligram
extinction coefficient of KLH (1.6). Six mg of KLH from a
12 mg/mL solution in 10 mM potassium phosphate, pH 7.3, was
added to 0.6 mg sulfo-MSB and incubated at room temperature
for 45 min. The KLH-MSB conjugate was separated from free
sulfo-MSB by size exclusion chromatography on a G-100 column
pre-equilibrated with 50 mM potassium phosphate, pH 6.0.
The free sulfo-MSB is retarded in the column and the KLH-MSB
conjugate flows through. Ten, 1 mL fractions were collected
and measured at 280 nm. KLH-MSB containing fractions were
pooled and reacted with 6 mg of the GLUT4 peptide for 3 h at
room temperature with rotation. The pH of the mixture was
adjusted with HC1 to produce the maximum amount of
precipitate. The mixture was then allowed to stand upright
overnight at 4C to allow the precipitate, which represents
the KLH-peptide conjugate, to settle.


98
A
B


92
GLUT4
12 3 4
Figure 4-2. Specificity of the Peptide-Purified GLUT4
Antibody in Immunoprecipitation. Cells were incubated and
radiolabeled as described in Figure 4-1. Total membranes
were collected and GLUT4 was immunoprecipitated under the
following conditions: lane 1, no GLUT4 antibody added;
lane 2, 5 pig of total IgG from pre-immune serum added; lane
3, 5 |ig of peptide-purified GLUT4 antibody competed with
molar equivalent of GLUT4 peptide added; and lane 4, 5 pig
of peptide-purified GLUT4 antibody added. These data
represent a single experiment.


40
Metabolic Labeling of 3T3-L1 Adipocytes
To measure the synthesis and turnover of proteins,
radiolabeling of proteins with [35S]cysteine/methionine was
performed to "mark" a population of proteins within a
specific period of time. The accumulation of radiolabel
into proteins was determined to measure synthesis. To
measure turnover, cells were pulsed with radiolabel followed
by an extended chase with nonradiolabeled medium. The
disappearance of the labeled protein was then monitored. To
examine synthesis of GLUT4, cells were incubated in complete
DMEM with 10% FBS in the presence or absence of 10 nM
insulin for 11 h. The medium was then changed to cysteine-
/methionine-free DMEM, also in the presence or absence of 10
nM insulin, for 1 h. This medium was then aspirated and 2
mL of the cysteine-/methionine-free DMEM containing 400 [iCi
of [35S]cysteine/methionine 10 nM insulin was added to
each plate. The plates were incubated at 37C for 10, 20,
30, 45, or 60 min at which time total membranes were
collected and extracted as described above.
Immunoprecipitation was performed on this solution.
Degradation of GLUT4 was examined by first incubating
cells in cysteine-/methionine-free DMEM for 1 h. Following
aspiration of this medium, 2 mL of the same medium
containing 400 [iCi of [35S] cysteine/methionine were added to


60
%
Ph
'P
co
O
O
"bb,
X
8 08 I-
cn
] 2-Deo
(nmol/l
o o
k>
Control
B Chronically treated
K
i i i i
0
10 15 20
Glucose Concentration (mM)
B
CD
CD
GO
O
o
J2
"bb
X
o
CD
9
CN
K
GO
1.4
1.2
1.0
a
2 06
o 0.4
cn
0.0
0.0 0.5 1.0 1.5 2.0
Glucosamine Concentration (mM)


Figure 3-11. Subfractionation of Insulin-Resistant 3T3-L1
Adipocytes. Cells were treated for 12 h in the absence
(control) or presence (chronic) of 10 nM insulin and
subsequently washed. Following acute stimulation with 1 |iM
insulin, PM, LDM, and HDM were collected as described in
Chapter 2. SDS-polyacrylamide gels of equal protein (70 p,g)
transferred to nitrocellulose allowed immunoblot detection
of GLUT1 and GLUT4 using C-terminal specific antibodies.
The protein-antibody complex was visualized by enhanced
chemiluminescence. Bands were quantitated by video
densitometry. Panel A, immunoblot of membrane fractions
probed with anti-GLUT4 antibody; panel B, densitometry of
GLUT4 immunoblot; panel C, immunoblot of membrane fractions
probed with anti-GLUTl antibody; panel D, densitometry of
GLUT1 immunoblot. control; E23 control + acute insulin; ESI
chronic insulin treatment; chronic insulin treatment +
acute insulin. Data represent a single experiment. A
duplicate experiment gave similar results.


Glycogen Synthesis 30
Production and Characterization of
GLUT4 Antibody 30
Peptide Purification of GLUT4 Antibody 36
Preparation of Total Membranes 39
Metabolic Labeling of 3T3-L1 Adipocytes 40
Immunoprecipitation of GLUT4 41
ATP-Luciferase Assay 43
Leptin Assay 44
3 DEVELOPMENT OF INSULIN RESISTANCE 4 9
Introduction 49
Results 50
Insulin-resistant Glucose
Transport Activity 50
Role of Glucose and Glucosamine 58
Effects of Various Inhibitors on the
Development of Insulin Resistance 61
Effect of Chronic Insulin on GLUT4
Expression and Translocation 69
Effect of Glucose Deprivation on
GLUT4 Expression 77
Effect of Chronic Insulin on
Glut4 mRNA Expression 77
Reversal of Insulin Resistance 80
Conclusions 82
4 MECHANISMS OF DECREASED GLUT4 LEVELS IN
INSULIN-RESISTANT CELLS 88
Introduction 88
Results 89
Specificity and Efficiency of
GLUT4 Immunoprecipitation 89
Synthesis of GLUT4 in
Insulin-resistant Cells 90
Degradation of GLUT4 in
Insulin-resistant Cells 94
Effects of Cycloheximide on the
Loss of GLUT4 99
Effects of Protease Inhibitors on
GLUT4 Expression 99
Conclusions 102
v


Chronic Insulin
Acute Insulin
D GLUT1 Levels
05


104
A PM LDM
Insulin
Leupeptin
GLUT1 *
+ + + +
+ + + +
GLUT4
3


27
Relative intensity of each band was quantified by video
densitometry within the linear range of the film using the
Bioimage Visage 110.
Glycogen Analysis
Total glycogen was isolated as described previously
(79) based on a method described by Pfleiderer (80). Cells
grown on 10-cm tissue culture dishes were washed at 4C with
PBS, pH 7.2, and scraped into a 15-mL polypropylene tube in
1 mL PBS. The cells were then sonicated for 10 s on power 2
at 50% duty cycle (Branson Sonifier 450). Two mL of 30%
(w/v) KOH was added and the sample was mixed on a vortex
mixer. The sample was then boiled in a water bath for 15
min. After cooling to room temperature, 3.5 mL of 95% EtOH
were added. The samples were mixed and heated in a boiling
water bath for 3 min and cooled again to room temperature.
The precipitated glycogen was collected by centrifugation at
1,300 X g for 5 min at room temperature. The supernatant
was removed by aspiration and the glycogen pellet washed by
resuspension in 1 mL 95% EtOH. This suspension was
transferred to a 1 mL microfuge tube and the glycogen
collected by centrifugation at 13,300 X g for 5 min at 4C.
The supernatant was removed by aspiration and the
precipitate was stored at -20C.


36
Peptide Purification of GLUT4 Antibody
In order to study the turnover and synthesis of GLUT4,
metabolic labeling with 35S-cysteine and -methionine was
performed followed by immunoprecipitation. In order to
prevent non-specific interactions during
immunoprecipitation, GLUT4-specific IgG was purified from
the anti-GLUT4 rabbit serum using a modification of the
technique described by Dankert et al. (83). Total IgG was
purified from 15 mL of serum by adjusting the pH to 5 with 3
M acetic acid and adding 0.75 mL of capryllic acid while
vigorously stirring. After stirring for 30 min, the
precipitate was collected by centrifugation at 41,000 X g
for 30 min at 4C. The supernatant was removed and its
volume measured. An equal volume of saturated ammonium
sulfate was added to the supernatant and stirred overnight
at 4C. The next day, the solution was divided equally
between two 15-mL corex tubes and the precipitate was
collected by centrifugation at 41,000 X g for 30 min at 4C.
The supernatant was discarded and each pellet was
resuspended in 0.5 mL of PBS containing 0.1% sodium azide.
The ammonium sulfate was removed from the samples by
dialyzing against three changes of PBS containing 0.1%
sodium azide over 48 h using tubing with a molecular weight
cut off of 13,000 Da. The concentration of the total IgG
solution was determined by measuring the absorbance of the


67
acute insulin
E-64
+ + + +
+ + + +
control chronic
Figure 3-8. Effects of E-64 on Insulin Resistance. Cells
were incubated in the presence or absence of 100 pM E-64
also in the presence (chronic) or absence (control) of 10 nM
insulin for 12 h. Cells were then washed to remove insulin
and basal and insulin-stimulated rates of glucose transport
were determined as previously described. Data represent the
average S.E. of two independent experiments (n=4).


138
PM
Glucose + +
Chronic Insulin + +
LDM
HDM
+
+ +
+ +
+
Figure A-l. Effect of Glucose Deprivation and Chronic
Insulin on the Aberrant Glycosylation of GLUT1 and GLUT4.
Cells were incubated in the presence or absence of 10 nM
insulin and DMEM containing 10% FBS or glucose-free DMEM
containing 10% dialyzed FBS for 12 h. Cells were washed to
remove insulin and membrane fractions collected. Proteins
were subjected to SDS-PAGE and transferred to nitrocellulose
for western blot analysis of GLUTl and GLUT4.


22
low density membrane (LDM), and high density membrane (HDM)
fractions were isolated by a modification (70) of a
technique described by Weber et al. (71). The major
modification of this technique is the use of a steel block
homogenizer and tungsten ballbearing originally designed by
Balch and Rothman (72). The complete procedure is outlined
in Figure 2-1. Briefly, control or insulin-treated cells
were scraped into TES buffer (10 mM Tris-HCl, pH 7.4, 1 mM
EDTA, and 250 mM sucrose) at 18C. The cells were then
passed over a tungsten ball ten times in the steel block
homogenizer (at 18C) with a clearance of 0.0025 inch. A
crude plasma membrane fraction was collected at 17,000 X g
for 15 min at 4C. Purified membranes were collected from
this fraction by sucrose gradient centrifugation (71). LDM
and HDM fractions were collected by differential
centrifugation in 300 ¡j,L TES. Membrane fractions were
stored at -20C. Protein was determined by the method of
Markwell et al. (73).
Markwell Assay for Protein Determination
The Lowry procedure is widely used for the assay of
soluble proteins (74). However, detergents, used to
solubilize membranes for the release of integral membrane
proteins, and common compounds such as sucrose and EDTA used
in buffers for the isolation of membranes can interfere with


46
medium was separated equally into four tubes and
concentrated by centrifugation at 3,000 x g (4,900 rpm) for
4.8 h at 4C in an SS-34 rotor. The retained solution was
collected from the four concentrators and pooled. The
pooled solution was then placed into a single new tube and
concentrated further by centrifugation at equal force and
time. The retained solution was collected again and brought
to 4 mL using PBS.
The leptin radioimmunoassay (RIA) was performed as per
the manufacturer's instructions. Briefly, 100 piL of assay
buffer (0.05 M PBS pH 7.4 containing 0.025 M EDTA, 0.1%
sodium azide, 0.05% Triton X-100 and 1% RIA grade BSA) were
added to 12 x 75 mm borosilicate glass tubes. Leptin
standards (100 j_iL) and unknown samples (100 |j,L) were added
to the appropriate tubes. Next, mouse leptin antibody (100
pL) was added to the tubes. The tubes were mixed on a
vortex incubated overnight at 4C. The next morning, 125I-
mouse leptin (100 jo.L) was added to all the tubes, mixed, and
incubated overnight at 4C. The following morning, 1 mL of
cold precipitating reagent was added to all the tubes and
mixed. The tubes were incubated for 20 min at 4C and the
precipitate collected by centrifugation for 15 min at 3,000
x g (5,000 rpm) in an SS-34 rotor at 4C. The supernatant
was decanted immediately and the tubes drained for 1 min.
Excess fluid was blotted from the lip of the tubes and the


5-1 Effects of Fructose on the Development of
Insulin Resistance 112
5-2 Effect of Fructose on GLUT4 Transporter Levels 114
.5-3 Translocation of GLUT4 in Cells Treated with
Fructose 116
5-4 Effects of Fructose on ATP Levels 119
5-5 Effect of Fructose and Chronic Insulin on
Glycogen Levels 120
5-6 Effects of Fructose and Chronic Insulin on
Glycogen Synthesis 122
A-l Effect of Glucose Deprivation and Chronic
Insulin on the Aberrant Glycosylation of
GLUTl and GLUT4 138
A-2 Effect of Glucose Deprivation and Chronic
Insulin on Glycogen Levels 140
B-l Leptin Levels 144
x


3
artherosclerosis associated with diabetes can lead to
coronary artery disease, which is the leading cause of
mortality among diabetics (6).
Insulin resistance is defined as a decreased biological
response to physiological concentrations of insulin in
insulin-responsive tissues (1). In adipocytes, this is
characterized by a reduction in insulin-stimulated glucose
transport activity (9). As of yet, the exact mechanism
leading to the onset of insulin resistance has yet to be
elucidated. However, several mechanisms have been proposed
including decreased autophosphorylation of the insulin
receptor (2,10-13), down-regulation of the insulin-
responsive glucose transporter GLUT4 (2,14), altered
translocation of GLUT4 to the plasma membrane from its
intracellular storage vesicles (15-17), and the actions of
the intermediates or products of the hexosamine biosynthetic
pathway (18-21) .
Role of Adipose in Insulin Resistance
As mentioned, many obese subjects also exhibit insulin
resistance. However, when these patients are placed on
calorie-restrictive diets to decrease their weight, insulin
resistance is reversed (22,23). This suggests that adipose
plays an important role in insulin resistance. Another
condition which shows the importance of adipose in insulin


106
117). Activation of this aminopeptidase or a yet unknown
protease which colocalizes with GLUT4 may be responsible for
its accelerated degradation.


151
85. Kitzman, H.H.J., McMahon, R.J., Williams, M.G. and
Frost, S.C. (1993) J. Biol. Chem. 268, 1320-1325.
86. van Putten, J.P.M. and Krans, H.M.J. (1985) J. Biol.
Chem. 260, 7996-8001.
87. Reed, B.C., Shade, D., Alperovich, F. and Vang, M.
(1990) Arch. Biochem. Biophys. 279, 261-274.
88. Tordjman, K.M., Leingang, K.A. and Mueckler, M. (1990)
Biochemistry. J. 271, 201-207.
89. Froesch, E.R., Burgi, H., Bally, P. and Labhart, A.
(1965) Mol. Pharmacol. 1, 280-296.
90. Frayn, K.N., Coppack, S.W. and Humphreys, S.M. (1989)
Biochem. Soc. Trans. 17, 1091.
91. Marshall, S., Bacote, V. and Traxinger, R.R. (1991) J.
Biol. Chem. 266, 10155-10161.
92. Traxinger, R.R. and Marshall, S. (1989) J. Biol. Chem.
264, 8156-8163.
93. Decker, H., Zahner, H., Heitsch, H., Konig, W.A. and
Fiedler, H.-P. (1991) J. Gen. Micro. 137, 1805-
1813 .
94. Causier, B.E., Milling, R.J., Foster, S.G. and Adams,
D.J. (1994) Microbiology. 140, 2199-2205.
95. Knutson, V.P., Donnelly, P.V., Baiba, Y. and Lopez-
Reyes, M. (1995) J. Biol. Chem. 270, 24972-24981.
96. Garvey, W.T. (1992) Diab. Care. 15, 396-417.
97. Pedersen, O., Bak, J.F., Andersen, P.H., Lund, S.,
Moller, D.E., Flier, J.S. and Kahn, B.B. (1990)
Diabetes. 39, 865-870.
98. Calderhead, D.M., Kitagawa, K., Tanner, L.I., Holman,
G.D. and Lienhard, G.E. (1990) J. Biol. Chem. 265,
13800-13808.
99. Reed, B.C. and Lane, M.D. (1980) Adv. Enzyme Regul. 18,
97-117.


135
from the cells. This reversal is also protein synthesis
dependent in that cycloheximide blocks the return of normal
insulin-stimulated glucose transport. An important
observation is that this reversal takes place between 4-8 h
following the removal of insulin, suggesting again that the
synthesis of a protein is involved. Therefore, I think it
is important to examine the GLUT4 protein levels during this
time to ascertain if the recovery of GLUT4 protein is
responsible for this reversal.
In addition, other metabolites involved in the
hexosamine biosynthetic pathway could be investigated to
identify a specific culprit involved in inducing insulin
resistance by this pathway. This pathway seems to be
important in insulin resistance, but no products or
intermediates of this pathway as yet have been identified as
the causitive agent in inducing insulin resistance.
However, this pathway has been hypothesized to be a glucose
sensor (18,138,139) and therefore may regulate glucose
transport through negative feedback inhibition. As only 2-
3% of the fructose-6-P enters the hexosamine biosynthetic
pool, increased levels of hexosamines, such as UDP-N-
acetylglucosamine, may act as a sensor of energy in the cell
and decrease glucose uptake. In other words, high levels of
hexosamines would signal that the cell has a saturating
amount of substrate fluxing through the glycolytic pathway


74
those that analyzed translocation using the cell surface
photolabel, ATB [2-3H]BMPA [2-N-4-(l-azi-2,2,2-
trifluoroethyl)benzoyl-1,3-bis(D-mannose-4-yloxy)-2-
propylamine] (16). Cells chronically exposed to insulin
(followed by washout) showed levels of GLUT4 in the plasma
membrane equivalent to that of controls (Figure 3-11 A and
B). While acute insulin challenge stimulated translocation,
the amount of GLUT4 was reduced by about 50% compared with
controls. This reflects the down-regulation of GLUT4 and
correlates well with the loss of insulin-stimulated glucose
transport activity. Cells chronically exposed to 10 nM
insulin showed a 2-fold increase in the level of GLUT1 in
the PM after washout compared with controls (Figure 3-11 C
and D), despite the equivalent rates of glucose transport.
Acute insulin challenge stimulated translocation but to a
much more limited degree than in control cells.
As the LDM fraction reflects the loss of GLUT4, we used
this fraction to examine the dose-dependent loss in cells
chronically treated with specific concentrations of insulin.
As shown in Figure 3-12 A, the level of GLUT4 decreased over
time in response to increasing insulin. Figure 3-13 shows a
similar experiment with a larger sample (5, 10-cm plates per
condition compared to 1, 10-cm plate). This larger sample
allows for the somewhat more accurate quantitation of the
LDM fraction.
The numbers listed below the blot in Figure


CHAPTER 3
DEVELOPMENT OF INSULIN RESISTANCE
Introduction
Insulin resistance has been studied in animal models,
freshly isolated cells, and immortalized cells. Using a
cell line has several advantages. One of these advantages
is the ease of harvesting cultured cells. Another is the
consistency and reproducibility of data, the lack of which
plagues in vivo studies.
Although 3T3-L1 adipocytes have been the focus of many
insulin action studies, in each case, pharmacological doses
of insulin were used. In contrast to other classes of
hormones, insulin does not exhibit biphasic effects, thus
the rationale for using pharmacological doses. However,
caution must be taken as insulin at high concentrations in
cell culture activates the mitogenic paths held in common
with other growth factors, the relevance of which is
questioned (60). That said, with long term exposure to
pharmacological doses of insulin, GLUT4 expression (both
mRNA and protein) is reduced (14). This reduction is caused
by both down-regulation of transcription and enhanced
turnover of mRNA. However, the concentration of insulin
49


34
A
1 2.5 5 \ig GLUT1 peptide
B
1 2.5 5 ¡ig GLUT4 peptide
Figure 2-3. Specificity of GLUT4 Antiserum for GLUT4
Peptide. Panel A, 1, 2.5, and 5 |_ig of GLUTl peptide was
blotted onto nitrocellulose and allowed to dry. Membranes
were blocked and probed with a 1:1000 dilution of the GLUT4
antiserum as described under "Electrotransfer and Western
Blotting". The protein-antibody complex was visualized by
enhanced chemiluminescence. Panel B, GLUT4 peptide was
blotted onto nitrocellulose and treated as in panel A.


47
pellet was counted in a gamma counter. Leptin levels
present in the samples were calculated by comparing against
the standard curve seen in Figure 2-7.


115
This is noteworthy in that transport activity in control and
chronically-treated cells after insulin washout was nearly
identical (see Figure 3-2 B and Figure 5-1). Similar
results were seen in cells treated with fructose and chronic
insulin. GLUT4 levels were decreased in the LDM fraction of
cells exposed to glucose and chronic insulin, as expected.
Surprisingly, GLUT4 was reduced to the same level in cells
treated with either glucose or fructose in the presence of
chronic insulin (Figure 5-2 A and B). Thus, despite normal
insulin-stimulated glucose transport in cells chronically
treated with insulin and fructose, GLUT4 levels dropped in
the endosomal fraction.
The apparent contradiction between GLUT4 expression and
insulin-sensitive transport activity led us to examine
translocation of GLUT4 from the LDM fraction to the PM.
Figure 5-3 shows the comparison in translocation between
cells treated with fructose and chronic insulin and control
cells, which were only treated with acute insulin. As with
glucose, cells exposed to fructose in the presence of
chronic insulin again showed reduced GLUT4 in the LDM
fraction. Remember, in the presence of acute insulin, the
PM isolated from cells incubated with glucose and chronic
insulin contained about half the GLUT4 as in controls (see
Figure 3-10). However, cells incubated with fructose and
chronic insulin contained the same amount of GLUT4 in the


105
Preliminary results with cycloheximide suggested that
the synthesis or activation of a protease was responsible
for the loss of GLUT4 in the LDM fraction of insulin-
resistant cells. This hypothesis was pursued by attempting
to identify a pathway responsible for GLUT4 degradation.
However, experiments using inhibitors specific for the
lysosomal and proteasomal degradative pathways revealed no
clues as to the route of degradation. In fact, these data
were negative despite the metabolic labeling data which
clearly shows an accelerated degradation of GLUT4 in the
presence of insulin. There are several possible
explanations for these negative results. For example,
leupeptin inhibits cathepsin L activity by only 50% (107)
and inhibits overall protein degradation in the lysosomes by
only 45% (114). Ammonium chloride also inhibits protein
degradation by only 45% with no additive effects obtained
with chloroquine (114). However, an additive effect is seen
with the addition of leupeptin, albeit small (<20%) (114).
Therefore, these inhibitors may exert a small effect in this
system, but one which cannot be measured. Alternatively, it
is possible that the inhibition of one protein degradative
pathway is compensated for by another, even though it may
not be the preferred pathway for degrading a particular
protein. Finally, there is evidence that an aminopeptidase
colocalizes with GLUT4 containing vesicles in vivo (115-


7
Glucosamine-6-P
Glucose-6-P
i
Fructose-6-P
97%
GFAT
C"
Glutamate
* Glucosamine-6-P
UDP-N-Acetylglucosamine
*
glycolipids
glycosylated proteins
GPI-linked proteins
Figure 1-1.
Pathway.
Glycolytic/TCA
Pathway
Insulin
Resistance
Overview of the Hexosamine Biosynthetic


13
VAMP-2, and cellubrevin) and eight different t-SNARES
(Syntaxin-1, Syntaxin-2, Syntaxin-3, Syntaxin-4, Syntaxin-6,
Syntaxin-7, SNAP-25, and Syndet) have been described thus
far (55). VAMP-2 is associated with GLUT4 containing
vesicles and is the v-SNARE that is responsible for binding
to the t-SNARES Syntaxin-4 and Syndet, the murine homologue
of SNAP-25 (56,57). In addition, the protein Muncl8c has
been shown to regulate the translocation of GLUT4 by
inhibiting the association of VAMP-2 with Syntaxin-4 (58) .
Insulin relieves this inhibition by inducing the
dissociation of Muncl8c from Syntaxin-4 allowing VAMP-2 to
bind to Syntaxin-4. However, it is not known yet how
insulin stimulates the association and dissociation of these
molecules. In addition, other molecules may be involved in
GLUT4 translocation as investigations continue in this
exciting area.
Insulin Receptor and Signaling
Insulin action is mediated by a specific cell-surface
receptor. The insulin receptor is a heterotetramer
consisting of two a- and two (3-subunits. Each a- and 13-
subunit is connected by a disulfide bond to form an a/p
dimer and each a/p dimer is then attached via a single
disulfide bond between the a-subunits to form the tetramer.
The asubunit has a molecular mass of 135,000 Da and is


96
A
Time (min) 10
20
30
45
60
Chronic Insulin
r
i r
i r
i r
mGLUT4
pGLUT4
B
Time (min)


plasma membrane which are similar to controls. While the
mechanism underlying this apparent "enhancement" of GLUT4
translocation is unknown, it explains the observation that
the insulin-stimulated glucose transport activity in cells
exposed to fructose does not reflect the insulin-resistant
state.
xiii


132
Therefore, the levels and translocation of GLUT4 were
examined following incubation with fructose and insulin to
explain this apparent anomaly. GLUT4 levels were indeed
reduced by 50% in the LDM fraction of these cells as they
were with glucose. However, upon acute stimulation of these
cells with insulin, GLUT4 translocation was enhanced. This
resulted in levels of GLUT4 in the PM of fructose and
chronic insulin treated cells equal to the levels present in
the PM of control cells. This enhanced translocation
explained why glucose transport activity was not reduced in
cells treated with fructose and chronic insulin even in the
presence of reduced levels of GLUT4. These results may, in
part, explain the beneficial effects of fructose in the
diabetic diet.
Future Directions
These studies show that the down-regulation of GLUT4
protein underlies the manifestation of insulin resistance in
these cells. Therefore, I believe it is important to
explore further the mechanisms leading to the decreased
synthesis and increased degradation of GLUT4 in insulin-
resistant cells. As the decreased GLUT4 mRNA levels do not
seem to be involved in inducing insulin resistance in these
studies, perhaps the reduced synthesis is due to
translational control. The investigation of the mechanism


109
activity. While one could argue that adipocytes metabolize
fructose at a slower rate than glucose, early studies showed
that fructose uptake and utilization in adipose tissue was
nearly equivalent to that of glucose (123) Further,
Kitzman et al. (85) has shown that fructose prevents
aberrant protein glycosylation in response to glucose
deprivation which demonstrates that fructose is metabolized
specifically by the hexosamine pathway. Thus, three
possibilities arise. The specific presence of fructose
might prevent the down-regulation of GLUT4. Alternatively,
fructose may support the insulin-dependent, down-regulation
of GLUT4, but change the relative distribution of GLUT4
between the plasma membrane and the endosomal fraction
(LDM). Finally, GLUT4 might be down-regulated but exhibit
"activation" at the cell surface. The data in this chapter
will show that fructose, in the presence of chronic insulin,
does in fact induce down-regulation of GLUT4. However,
GLUT4 translocation is enhanced in response to an acute
insulin challenge leading to equivalent amounts of GLUT4 on
the plasma membrane as compared to the controls. This
prevents the manifestation of insulin-resistant transport
activity despite the insulin-dependent, down-regulation of
GLUT4.
To compare these results to another insulin-sensitive
system, I examined the effect of the chronic treatment


35
GLUT4
. o o
& & d
S' S'
A A
£ £
& &
'O "O 'O
x v x
.<$? aJ" aJ* aJ"
^ ^y ^y ^y
V
Figure 2-4. Specificity of GLT4 Antiserum for GLUT4
Protein. Equal protein (50 |J,g) from the LDM fraction of
control cells was loaded per lane and subjected to SDS-PAGE.
Proteins were transferred to nitrocellulose and western blot
analysis performed under the following conditions: lane 1,
pre-immune serum; lane 2, 1:1000 dilution of GLUT4
antiserum; lane 3, GLUT4 antiserum competed with 10 (ig of
GLUT1 peptide; and lane 4, GLUT4 antiserum competed with 10
(_ig of GLUT4 peptide.


29
Figure 2-2. Fluctuations in Glycogen Levels Over a Week.
Following differentiation, cells were maintained on their
feeding schedule and fed with DMEM containing 10% FBS on the
days indicated. Cells were collected at the same time every
day during the week and glycogen isolated by ethanol
precipitation. Following acid hydrolysis, glucose present
in glycogen was determined by a commercially available kit
(Sigma). Data represent the average S.D. of duplicate
samples within a single experiment (n=2).


3-8 Effects of E-64 on Insulin Resistance 67
3-9 Effects of E-64 on Glycogen Synthesis 68
3-10 Effects of Acute and Chronic Insulin on
Total GLUT Protein Levels 70
3-11 Subfractionation of Insulin-Resistant
3T3-L1 Adipocytes 72-73
3-12 Effects of Chronic Insulin on GLUT4
Expression 75
3-13 Effects of Chronic Insulin on GLUT4
Expression with Larger Samples 76
3-14 Effect of Glucose Deprivation on
GLUT4 Expression 78
3-15 Effect of Chronic Insulin Treatment on
GLUT4 mRNA Levels 79
3-16 Reversal of Insulin Resistance 81
3-17 Reversal of Insulin Resistance in the
Presence of Cycloheximide 83
4-1 Titration of Peptide-Purified GLUT4 Antibody
for Immunoprecipitation 91
4-2 Specificity of the Peptide-Purified GLUT4
Antibody in Immunoprecipitation 92
4-3 Efficiency of GLUT4 Immunoprecipitation 93
4-4 Effect of Chronic Insulin on the
Synthesis of GLUT4 96
4-5 Effect of Chronic Insulin on the
Degradation of GLUT4 98
4-6 Degradation of Total Protein in
Cells Treated with Chronic Insulin 100
4-7 Effect of Cycloheximide on the Loss of GLUT4 101
4-8 Effects of Protease Inhibitors on
GLUT4 Expression 104
ix


Figure 3-1. Effect of Chronic Insulin on Glucose Transport
in 3T3-L1 Adipocytes. Panel A, fully differentiated 3T3-L1
adipocytes were incubated in DMEM containing 10% FBS in the
absence of added insulin (#) or with 1 |nM (A) or 10 nM (H)
insulin for 12 h. Plates were washed as described in
Chapter 2 under "Insulin Washout Procedure". Glucose
transport activity was measured in KRP (in the absence of
bovine serum albumin and glucose) by the addition of 200 |aM
[3H] 2-deoxyglucose (0.2 |J.Ci) After 2 min, transport was
terminated by the addition of ice-cold phosphate-buffered
saline. Cells were lysed with a 0.1% solution of SDS and
duplicate aliquots of 300 |i,L were taken for quantitating
radioactivity. Panel B, cells were incubated as above in
the absence (control) or presence (chronic) of 10 nM
insulin. Removal of insulin was accomplished within 60 min
(see panel A), then 1 |aM insulin was added back, or not, for
10 min, and glucose transport activity was determined.
Panel C, cells were incubated for 12 h in the absence
(control) or presence (chronic) of 10 nM insulin. The
cells were washed for 60 min to remove insulin. In order
to maintain the low levels of insulin in solution, glucose
transport activity was measured in KRP containing 0.1% BSA.
Therefore, following the final wash at 60 min, 950 ¡0.L of KRP
containing 0.1% BSA were added to each plate. Various
concentrations of insulin, dissolved in 0.1%, were then
added for 10 min for acute stimulation. Glucose transport
activity was measured by the addition of [3H]2-deoxyglucose.
Each panel represents the average S.E. of two independent
experiments (n=4).


41
each plate and incubated at 37C for 3 h. This medium was
aspirated and replaced with complete DMEM containing 10% FBS
with or without 10 nM insulin for the chase. This medium
was refreshed every 12 h and total membranes were collected
at 0, 6, 12, 24, 48, and 72 h. Total membranes were
collected and immunoprecipitation performed.
Immunoprecipitation of GLUT4
Immunoprecipitation provides a means of studying the
turnover of a particular protein. This technique depends on
the existence of an antibody specific for the protein of
interest. Addition of the antibody to a mixture of proteins
results in specific antibody-protein complex formation.
This complex can then be collected by the addition of
Sepharose beads conjugated with protein A, a bacterial
protein that binds to the heavy chain of IgG. The beads are
precipitated by a short centrifugation and the bound
proteins are analyzed. Although in these studies this
technique was employed to study the dynamics of GLUT4,
protein-protein interactions can be determined by analyzing
proteins which co-immunoprecipitate with the protein of
interest.
To examine the synthesis of GLUT4, extracted membrane
protein of specific concentration (1.5 mg) was brought to 1
mL with additional extraction buffer. To investigate the


APPENDIX B
LEPTIN AND INSULIN RESISTANCE
Leptin, a 16 kDa protein which is the product of the OB
gene, was discovered in 1994 (141,142) This protein is
produced and secreted by white adipose tissue and 3T3-L1
adipocytes (143,144). Mice which have a homozygous nonsense
mutation in this gene, termed ob/ob mice, are unable to
produce leptin and are extremely obese (141,142) Another
well characterized mouse mutation is the db/db mouse. These
mice have a mutation in the hypothalamic receptor for leptin
and therefore, do not respond to this protein. As a result,
these mice are also obese and have elevated levels of leptin
(145.146). Leptin is believed to act as a regulator of
adiposity by acting as a satiety factor and regulator of
energy expenditure when it binds its receptor in the
hypothalamus (141,142).
Some evidence suggests that leptin may be involved in
insulin resistance. Both ob/ob and db/db mice exhibit
characteristics of insulin resistance and diabetes
(141.142.145.146). As well, leptin expression is down-
regulated with administration of an antidiabetic
thiazolidinedione in Zucker diabetic fatty rats, db/db mice
141


80
concentration of insulin required to elicit a 50% loss in
GLUT4 mRNA was about 5 nM (Figure 3-15 B), which is 15 times
greater than that required for the equivalent loss of
insulin-sensitive glucose transport activity and 10-times
that for GLUT4 expression.
Reversal of Insulin Resistance
To determine if the effects of chronic insulin on 3T3-
L1 adipocytes could be reversed, insulin-resistant cells
were incubated with complete medium following the washout
procedure. Both basal and insulin-stimulated rates of
glucose transport were measured at specific times. Eight
hours after insulin removal, glucose transport activity of
previously insulin-resistant cells was equal to that of
controls (Figure 3-16). An interesting point is that the
reversal appears to occur between 4-8 h, as no improvement
in insulin-sensitive transport occurs before that time.
Also, the rate of basal transport between control cells and
resistant cells is identical over this time while only the
insulin-stimulated rates are changing. This shows that the
increase in insulin-stimulated transport seen in resistant
cells over this time is not due to increasing basal rates,
but improved insulin sensitivity.
To determine if this reversal in insulin resistance
required protein synthesis, cycloheximide was included in


TABLE OF CONTENTS
page
ACKNOWLEDGMENTS iii
LIST OF TABLES vii
LIST OF FIGURES viii
ABSTRACT xi
CHAPTERS
1 INTRODUCTION 1
Overview of Diabetes 1
Insulin Resistance 2
Role of Adipose in Insulin Resistance 3
Role of the Hexosamine Biosynthetic
Pathway in Insulin Resistance 6
3T3-L1 Adipocytes 8
GLUT Transporter Family 9
Characteristics of GLUT Family 9
Translocation of GLUT4 12
Insulin Receptor and Signaling 13
3T3-L1 Adipocytes: A Model for
Insulin Resistance 17
2 MATERIALS AND METHODS 18
Materials 18
Methods 19
Cell Culture 19
Chronic Insulin Treatment and
Insulin Washout 20
Glucose Transport Assay 21
Subcellular Fractionation of
3T3-L1 Adipocytes 21
Markwell Assay for Protein Determination 22
Gel Electrophoresis 24
Electrotransfer and Western Blotting 25
RNA Isolation and Northern Blotting 26
Glycogen Analysis 27
iv


108
and 3T3-L1 adipocytes (118). Recall that glucosamine 6-
phosphate, the product of GFAT, can also be produced by
hexokinase-mediated phosphorylation of glucosamine,
functionally bypassing GFAT. In addition, overexpression of
GFAT in fibroblasts inhibits insulin-sensitive glycogen
synthase activity (119) Further, glucose deprivation in
the presence of chronic insulin prevents both the loss of
GLUT4 (118) and the development of insulin-resistant glucose
transport (15,118).
There is also evidence to suggest that intermediates or
products of the hexosamine biosynthetic pathway are not
involved in the development of insulin resistance. Hresko
et al. (120) reported that glucosamine in the presence of
insulin depletes intracellular ATP in 3T3-L1 adipocytes,
suggesting that glucosamine 6-P accumulates preventing
phosphate recycling. This leads to an inability of the
cells to respond to insulin, which normally initiates an
ATP-dependent kinase cascade. ATP depletion would also
prevent fusion of intracellular GLUT4 vesicles with the
plasma membrane, as demonstrated by investigators in the
mid-80's (121,122). Thus, glucosamine affects metabolism
independently from its flux through the hexosamine pathway.
A second study by Garvey et al. (15) showed that rat
adipocytes incubated with fructose in the presence of
insulin did not exhibit insulin-resistant glucose transport


153
114. Seglen, P.O., Grinde, B. and Solheim, A.E. (1979) Eur.
J. Biochem. 95, 215-225.
115. Kandror, K.V., Yu, L. and Pilch, P.F. (1994) J. Biol.
Chem. 269, 30777-30780.
116. Kandror, K.V. and Pilch, P.F. (1994) Proc. Natl. Acad.
Sci. U.S.A. 91, 8017-8021.
117. Keller, S.R., Scott, H.M., Mastick, C.C., Aebersold, R.
and Lienhard, G.E. (1995) J. Biol. Chem. 270,
23612-23618.
118. Thomson, M.J., Williams, M.G. and Frost, S.C. (1997) J.
Biol. Chem. 272, 7759-7764.
119. Crook, E.D., Zhou, J., Daniels, M., Neidigh, J.L. and
McClain, D.A. (1995) Diabetes. 44, 314-320.
120. Hresko, R.C., Heimberg, H., Chi, M. and Mueckler, M.
(1998) J. of Biol. Chem. 273, 20658-20668.
121. Sato, N., Irie, M., Kajinuma, H. and Suzuki, K. (1990)
Endocrinology. 127, 1072-1077.
122. Toyoda, N., Robinson, F.W., Smith, M.M., Flanagan, J.E.
and Kono, T. (1986) J. Biol. Chem. 261, 2117-2122.
123. Froesch, E.R.a.G., J.L. (1962) J. Biol. Chem. 237,
3317-3324.
124. Rigden, D.J., Jellyman, A.E., Frayn, K.N. and Coppack,
S.W. (1990) Eur. J. Clin. Nutr. 44, 689-692.
125. Holness, M.J., Shuster-Bruce, M.J.L. and Sugden, M.C.
(1988) Biochem. J. 254, 855-859.
126. Mayes, P.M. (1993) Ami. J. Clin. Nutr. 58 (suppl) 754S-
65S .
127. Tobey, T.A., Mondon, C.E., Zavaroni, I. and Reaven,
G.M. (1982) Metabolism. 31, 608-612.
128. Vrna, A. and Kazdov, L. (1970) Life Sci. 9, 257-265.
129. Storlien, L.H., Oakes, N.D., Pan, D.A., Kusunoki, M.
and Jenkins, A.B. (1993) Diabetes. 42, 457-462.


20
Chronic Insulin Treatment and Insulin Washout
Fully differentiated 3T3-L1 adipocytes were incubated
in DMEM containing 10% FBS and specific concentrations of
insulin ranging from 1 pM to 10 nM for 12 h at 37C in a 7%
C02 incubator. Care was taken to reduce the loss of insulin
in solution, particularly at low concentrations. Thus,
insulin dissolved in 0.01 N HC1 was added to DMEM containing
10% FBS to give a final concentration of lOnM insulin. The
remaining insulin concentrations were achieved by serial
dilution into DMEM containing 10% FBS. In addition,
solutions were stored in plastic containers to prevent
insulin binding which occurs in glass bottles. Finally,
cells were refed every 2 h as adipocytes rapidly degrade
insulin.
After chronic treatment, an insulin washout procedure
was performed with the goal of rapidly returning the cells
to a basal state. Specifically, adipocytes were removed
from the C02 incubator and placed in a 37C waterbath.
Plates were washed three times with 3 mL of Krebs' Ringer
Phosphate buffer (KRP), pH 7.4, containing 0.1% BSA and 5 mM
glucose, every 20 min over 60 min. The final wash at 60 min
was performed with KRP alone. The cells were then assayed
according to the particular experimental protocol as
described below. After determining the optimal
concentration of insulin to induce insulin resistance in


154
130. Luo, J., Rizkalla, S.W., Lerer-Metzger, M., Boillot,
J., Ardeleanu, A., Bruzzo, F., Chevalier, A. and
SIama, G. (1995) J. Nutr. 125, 164-171.
131. Koivisto, V.A. and Yki-Jarvinen, H. (1993) J. Inter.
Med. 233, 145-153.
132. Crapo, P.A., Kolterman, O.G. and Henry, R.R. (1986)
Diab. Care. 9, 111-119.
133. Bantle, J.P., Laine, D.C., Castle, G.W., Thomas, J.W.,
Hoogwerf, B.J. and Goetz, F.C. (1983) N. Eng. J.
Med. 309, 7-12.
134. Crapo, P.A. and Kolterman, O.G. (1984) Am. J. Clin.
Nutr. 39, 525-534.
135. Akgn, S. and Ertel, N. (1980) Diab. Care. 3, 582-585.
136. Henry, R.R. and Crapo, P.A. (1991) Ann. Rev. Nutr. 11,
21-39.
137. Crapo, P.A., Kolerman, O.G. and Olefsky, J.M. (1980)
Diab. Care. 3, 575-582.
138. Cooksey, R.C., Hebert, L.F., Zhu, J., Wofford, P.,
Garvey, W.T. and McClain, D.A. (1999)
Endocrinology. 140, 1151-1157.
139. Hawkins, M., Angelov, I., Liu, R., Barzilai, N. and
Rossetti, L. (1997) J. Biol. Chem. 272, 4889-4895
140. McMahon, R.J. and Frost, S.C. (1995) J. Biol. Chem.
270, 12094-12099.
141. Zhang, Y., Proenca, R., Maffei, M., Barone, M.,
Leopold, L. and Friedman, J.M. (1994) Nature. 372
425-432.
142 .
Halaas, J.L.
et al.
(1995)
Science. 269, 543-546.
143 .
Masuzaki, H.
et al.
(1995)
Diabetes. 44, 855-858.
144 .
Yoshida, T.,
Monkawa, T.,
Hayashi, M. and Saruta, T
(1997) Biochem. Biophys. Res. Comm. 232, 822-826.
145. Tartaglia, L.A. et al. (1995) Cell. 83, 1263-1271.


2
to an inability of the pancreas to produce insulin. Type
II, or non-insulin-dependent diabetes mellitus (NIDDM), is
the most prevalent and least understood form of diabetes.
About 90-95% of the people with diabetes have type II. This
form usually develops in adults over 40 and is most common
in adults over 55. Type II diabetes is also associated with
obesity in that 80% of the people with this form of diabetes
are overweight. With this form, the pancreas generally
produces insulin, but the body does not respond to insulin
effectively, a condition known as insulin resistance.
Gestational diabetes develops during pregnancy and generally
disappears afterwards. However, women who develop diabetes
during pregnancy have a greater risk of developing type II
diabetes later in life.
Insulin Resistance
Insulin resistance is one of the determining
characteristics of type II diabetes mellitus (NIDDM) and
obesity and is believed to be the underlying pathology (1) .
NIDDM is characterized by both hyperglycemia and
hyperinsulinemia, whereas obesity manifests only
hyperinsulinemia (2). Complications of insulin resistance
and diabetes include retinopathy (3), nephropathy (4),
neuropathy (5), and artherosclerosis (6). The
hyperlipidemia (7), hypercoagulation of platelets (8), and


133
responsible for the degradation of GLUT4 is likely to prove
difficult. Of the several protease inhibitors used, none
yielded any positive results. However, the concentrations
of lactacystin and MG132 could be increased even though
concentrations used in this study have been sufficient in
other cells (111,113). Another compound used to study the
degradation of proteins in the lysosomes is ammonium
chloride. Although the effects of this compound were not
investigated in this study, it might prove beneficial to
test it for any effects.
If increasing the concentrations of lactacystin and
MG132 or testing of ammonium chloride have no effects on the
degradation of GLUT4, then this would point to the
possibility of a specific protease responsible for degrading
GLUT4 under conditions of chronic insulin. One such
candidate could be the insulin-responsive aminopeptidase
(IRAP) which co-localizes with GLUT4 containing vesicles.
IRAP also known as gpl60, for glycoprotein of 160 kDa, and
vpl65, for vesicle protein of 165 kDa, was identified by two
independent groups (116,117) and has been shown to
translocate to the plasma membrane in response to insulin
(116). This protein was shown to have aminopeptidase
activity (115) and to be a member of the family of zinc-
dependent membrane aminopeptidases (117). It contains a
large extracellular catalytic domain, single transmembrane