a. c}... Hn‘ ‘
n 3.. I . u
‘ ‘ 5.....2;
. . .fi‘wr can:
‘ . . ‘ «r .9.
.i i...
1...... 3.

3
x iwcnhrz.

L... . that",

. .e
1.37:!

i 1...: It?
«5. {It uninvfiwi
: .

‘v 2“;
, HF”... . ,s v
.1 n
s) I
. 1393133.. f."
3.: . . . .V
401
.n

 

 

To: .v 5..

If] n 4‘. ll
12”.}. .\
y w’. ‘4. y.

 

1 003

This is to certify that the

dissertation entitled

MECHANISMS INVOLVED IN THE REPRESSION OF HUMAN INSULIN
GENE PROMOTER ACTIVITY MEDIATED BY CHRONIC
HYPERGLYCEMIA

presented by

Maria Fernanda Pino

has been accepted towards fulfillment
of the requirements for

Ph.D. degree in Cell and Molecular
Biology Program

MM
/ 9/27/az

L. Karl Olson i

Major professor

 

 

 

Date September ”#2002

 

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

 

 

LIBRARY
Michigan State
University

 

 

PLACE IN RETURN BOX to remove thi
To AVOID FINES return on
MAY BE RECALLED with earlier

5 checkout from your record.
or before date due.
due date if requested.

 

DATE DUE

DATE DUE

DATE DUE

 

 

R 2 22005

 

 

 

 

 

 

 

 

 

 

 

 

 

6/01 c:/ClFIC/DareDue.p65-p.15

~—-__._——-—~——_.____ _.

MECHANISMS INVOLVED IN THE REPRESSION OF HUMAN INSULIN
GENE PROMOTER ACTIVITY MEDIATED BY CHRONIC

HYPERGLYCEMIA

By

Maria Fernanda Pino

A DISSERTATION

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

DOCTOR OF PHILOSOPHY
Cell and Molecular Biology Program

2002

ABSTRACT
MECHANISMS INVOLVED IN THE REPRESSION OF HUMAN INSULIN
GENE PROMOTER ACTIVITY MEDIATED BY CHRONIC
HYPERGLYCEMIA
By

Maria Fernanda Pino

Type II diabetes is characterized by insulin resistance and failure of pancreatic [3-
cells to secrete sufficient amounts of insulin to overcome hyperglycemia. Once diabetes
is established, chronic hyperglycemia has been postulated to cause adverse alterations in

B-cell function, thus exacerbating the disease state. Some of the characteristics of

hyperglycemia-induced B-cell damage include both suppression of insulin gene
expression and glucose-induced insulin secretion. The later is associated with decreased
insulin gene promoter activity. Mechanisms accounting for chronic hyperglycemia-
induced suppression of insulin promoter activity still remain unclear. Studies in other
cell systems suggest that stress associated with hyperglycemia such as oxidation could be
involved in tissue damage. Therefore, experiments were performed to understand the
role of c-Jun N-terminal kinase (JNK), a stress-activated kinase, in chronic
hyperglycemia-induced insulin promoter repression.

These studies demonstrated that JNK activity is significantly increased, in parallel
with increased AP-l transcription factor activity, in INS-1 cells cultured in 16.7 mM
glucose compared to 4.0 mM glucose. Over-expression of JNK repressed insulin

promoter activity in INS-1 cells cultured in 4.0 mM glucose. Over-expression of MLK3,

an upstream activator of JNK, increased JNK activity, AP-l transcription factor activity,
and repressed insulin promoter activity in cells cultured in 4.0 mM glucose. These data
show that activation of JNK either by high levels of glucose or upstream activators leads
to functional activation of downstream JNK target proteins and causes reduced insulin
promoter activity. These findings suggest that glucose-induced insulin promoter
repression might be mediated by the JNK Si gnaling pathway.

de-l, an important regulatory transcription factor of insulin gene expression,
partially mediates the repression of glucose-induced insulin promoter activity. Because
over-expression of de-l in insulinoma cells cultured in high levels of glucose is
insufficient to prevent insulin promoter repression, we further characterized other
potential promoter targets. Truncation analysis and functional studies demonstrated that
high levels of glucose mediate promoter repression between —327 and —261 nucleotides.
Mobility shift assays showed that three glucose-sensitive complexes bind to the AS/Core,
palindrome, and E3 elements, respectively within the insulin promoter. Site-specific
mutations of all three elements in the insulin promoter gene partially prevented glucose-
induced insulin promoter repression.

Collectively these studies suggest that elevated extracellular glucose mediates
insulin promoter repression possibly through the JNK signaling pathway and through

novel mechanisms that target the distal promoter region.

iv

T 0 my mom, sister, and family

ACKNOWELEDGEMENTS

I would like to thank my mentor Dr. Karl Olson for his encouragement, his
advice, and his support throughout my graduate career. I am also very grateful to my
committee members Dr. Ron Patterson, Dr. Laura McCabe, Dr. Richard Miksicek, and
Dr. Susan Conrad for their discussions and suggestions with respect to my project. I am
especially thankful to Dr. Richard Miksicek and Dr. Susan Conrad for giving me advice
and being helpful when needed. I also want to thank Dr. Kathy Gallo and Barbara Bock
for their collaboration with one of the projects.

I would like to thank the physiology and CMB staff, especially Gregory Romig
and Rachel Rumsey. I want to express my gratitude to the former and present students of
various departments for helpful discussions and encouragement to keep pursuing my
goals. I am very thankful to the lab members Katrina Linning, Hellen Cirrito, and Diana

Ye. I also want to thank all of my friends and Erik Runkle for their great support.

TABLE OF CONTENTS

Page
LIST OF TABLES ................................................................................. viii
LIST OF FIGURES ................................................................................. ix
CHAPTER I. Introduction ......................................................................... 1
CHAPTER H. Literature review ................................................................... 3
1. Endocrine functions of the pancreas .................................................. 3
2. Role of insulin in daily metabolic control ............................................ 4
2.1 Carbohydrate metabolism ..................................................... 4
2.2 Fat metabolism .................................................................. 5
2.3 Protein metabolism ............................................................ 7
3. Insulin biosynthesis ..................................................................... 7
3.1 Insulin synthesis ................................................................ 7
3.2 Insulin secretion ................................................................ 9
4. Regulation of insulin gene transcription ............................................ 11
5. Diabetes mellitus ....................................................................... 18
5.1 Type I diabetes mellitus ...................................................... 19
5.2 Type II diabetes mellitus ..................................................... 19
5.2.1 B-cell failure during the progression of diabetes. . . . . . . . ....20
6. Pathology of B-cells in Type II diabetes ............................................ 23
6.1 Glucotoxicity .................................................................. 24
6.1.1 Studies in Type II diabetes patients ........................... 24
6.1.2 In vivo studies of animal models of Type II diabetes
and in vitro studies of isolated islets and insulinoma
cell lines .......................................................... 25
6.2 Lipotoxicity .................................................................... 31
7. Mechanisms of hyperglycemia-induced damage .................................. 35
7.1 Oxidative stress ................................................................ 35
7.1.1 Oxidative stress in diabetes .................................... 37
7.1.1.1 Oxidative stress in diabetic individuals and
whole animals ........................................... 37
7.1.1.2 Oxidative stress in isolated and insulinoma cell
lines ...................................................... 39
7.1.2 Mechanisms of oxidative stress-induced B-cell damage...4l
7.2 Protein kinase C activation .................................................. 44
8. Protein kinase regulation of B-cell function ........................................ 45
8.1 Growth ......................................................................... 45
8.2 Insulin gene transcription .................................................... 46

vi

8.3 Cytokines involved in B-cell cytotoxicity ................................. 48

9. INS-1 cell model ....................................................................... 49
CHAPTER III. Materials and Methods ......................................................... 57
1. Material ................................................................................. 57
2. INS-1 cell culture ...................................................................... 58
3. Plasmid DNA constructs .............................................................. 58
4. PCR reactions ........................................................................... 60
5. Ligation reactions ...................................................................... 61
6. Transfections ........................................................................... 61
7. CAT assays ............................................................................. 62
8. Luciferase assays ...................................................................... 63
9. INK kinase assay ...................................................................... 63
10. Nuclear extracts ........................................................................ 64
11. Phosphorylated JNK analysis ........................................................ 65
12. Electrophoretic mobility-shift assays ................................................ 65

CHAPTER IV. Increased c-Jun N-tenninal kinase (JNK) activity in INS-1 cells
exposed to elevated glucose-concentrations may mediate the

repression of insulin promoter activity ....................................... 70
1. Abstract ................................................................................. 70
2. Introduction ............................................................................. 72
3. Results ................................................................................... 75
3.1 Exposure of INS-1 cells to elevated glucose concentrations
increases JNK activity ........................................................... 75
3.2 Exposure of INS-1 cells to elevated glucose concentrations
increases AP-l transcription factor activity ................................... 76

3.3 Expression of JNK in INS-1 cells represses insulin promoter activity. . ..79
3.4 Expression of MLK3 in IN S-l cells increases both basal and glucose-

stimulated JNK activity, and AP-l transcription factor activity. . . . . . . . ....82
3.5 Expression of MLK3 in INS-1 cells represses insulin promoter

activity .............................................................................. 84
3.6 Expression of HP] in INS-1 cells does not prevent glucose-induced
insulin promoter repression ..................................................... 87
3.7 Over-expression of HP] and inhibition of p38 MAP kinase did not
prevent glucose-induced repression of insulin promoter .................... 91
3.8 INK-induced repression of insulin promoter activity does not map
directly to the Al, A3, or C1 elements ........................................ 91
4. Discussion .............................................................................. 95

CHAPTER V. Chronic exposure of INS-1 cells to high glucose concentrations
decreases insulin gene reporter activity through regulatory
elements .......................................................................... 102

vii

1. Abstract ................................................................................ 102
2. Introduction ........................................................................... 104
3. Results ................................................................................. 106

3.1 Glucose regulates insulin promoter activity in a biphasic manner in

INS-1 cells ........................................................................ 106
3.2 The Al, A3, and Cl (RIP3bl) regulatory elements have a minor role

in mediating glucose-induced repression of insulin promoter activity

in INS-1 cells ..................................................................... 107
3.3 Upstream regulatory elements play a major role in glucose-mediated
repression of insulin promoter activity ....................................... 110

3.4 The Z, X, and Y minienhancers contain promoter sequences involved
in glucose-induced biphasic regulation of insulin promoter activity. . ....1 15
3.5 High levels of glucose repress DNA binding activity to the A5/Core

and E3 regulatory elements ..................................................... 122

3.6 Mutation of AS/Core, the palindrome, and E3 elements partially
diminished glucose-induced insulin promoter repression .................. 133
4. Discussion ............................................................................. 135
CHAPTER VI. Conclusions and future studies ............................................... 143
CHAPTER VII. List of references ............................................................... 152

viii

LIST OF TABLES

Table Page

1. Sequence of oligonucleotide primers used to generated mutations in —230 insulin
promoter reporter vector ....................................................................... 67

2. Sequence of oligonucleotide primers used to generate truncated insulin promoter
vectors ............................................................................................ 68

3. Oligonucleotide sequences used for mobility—shift assays and used to generate
mutations in the AS/Core, palindrome, and E3 elements ................................. 69

4. Mutations of all three AS/Core, palindrome, and E3 elements partially diminished
glucose-induced repression of the IN S(-327)CAT ....................................... 139

ix

LIST OF FIGURES

Figures Page
1. Proinsulin and insulin molecules ............................................................... 8
2. Schematic representation of glucose—induced insulin secretion .......................... 10
3. Human insulin promoter ....................................................................... 13
4. Incubation of INS-1 cells in high glucose concentration increases JNK activity ...... 77
5. INS-1 cells cultured in high levels of glucose have increased AP-l transcription

factor activity .................................................................................... 78
6. Over-expression of JNK represses insulin promoter activity ............................. 8O
7. Over-expression of MLK3 in INS-1 cells activates JNK activity ........................ 83
8. Over-expression of MLK3 increases AP-l transcription factor activity ................ 85
9. Over—expression of MLK3 represses insulin promoter activity .......................... 87
10. Expression of JIPl protein does not reverse glucose-induced repression of

insulin promoter activity in INS-1 cells ...................................................... 90
11. Effect of J IPl over-expression and p38 MAP kinase inhibitior on glucose-

induced AP-l transcription factor activity .................................................. 92
12. Over-expression of JIP-l and inhibition of p38 MAP kinase does not prevent

glucose-induced repression of insulin promoter activity .................................. 94
13. Repression of insulin promoter activity does not map solely to the A1, A3, and

C1 elements ...................................................................................... 95
14. Expression of JNK does not repress the multimerized E1 element activity ............ 98
15. Glucose regulates insulin promoter activity in a biphasic manner in INS-1 cells. . . l 10
16. Schematic representation of human insulin promoter and truncated insulin

promoter vectors .............................................................................. 1 l 1
17. The —230 bp insulin promoter has a smaller repression than the —327 bp insulin

promoter in INS-l cells cultured in 16.7 mM glucose .................................... 113

X

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

 

Mutations of both A1 and A3 or C1 prevent glucose-induced repression of the

INS(-230)CAT promoter activity ........................................................... 114
Truncation analysis of —327 bp insulin promoter activity in IN S-l cells cultured

in 4.0 and 16.7 mM glucose .................................................................. 115
Schematic representation of distal insulin promoter sequences ........................ 119

The Z and X minienhancers are repressed when INS-l cells are cultured in
16.7 mM glucose .............................................................................. 121

The Y, la ,and lb minienhancers are repressed when INS-l cells are cultured in
16.7 mM glucose .............................................................................. 122

Glucose-sensitive complex binds to the palindrome sequence and the E3 element..125

High levels of glucose reduces binding activity of the Zdl and E3 probes. . . . . . . . ....126
Glucose-sensitive complex binds to the la element region .............................. 129
High levels of glucose reduce binding activity of the Zal complex .................... 130
Glucose-sensitive complex binds to the Xa element region ............................. 132
Glucose-sensitive complex binds to the A5/Core sequence (Xb probe) ............... 133

Neither de1 nor ka6.1 form part of the complex that binds to the
AS/Core element .............................................................................. 135

High levels of glucose reduce binding activity to the A1 element ..................... 136
Glucose does not decrease binding activity of the complex that binds of the E4

element .......................................................................................... 137

xi

I. INTRODUCTION

Type H diabetes is characterized by the combination of insulin resistance and B-
cell dysfunction. B-Cell failure can be ascribed, at least in part, to the adverse effects of
chronic hyperglycemia on pancreatic B-cell fimction. One of the characteristics of B-cell
failure is the loss of glucose-induced insulin secretion that is associated with loss of
insulin gene expression (1). Reduced binding activity of two important transcription
factors, de-l and Cl activator, is correlated with the loss of insulin gene expression and
the loss of promoter activity (2—4). Studies performed to understand the role of de-l in
glucose-induced repression of the insulin promoter suggest that de-l is not the only
transcription factor affected by hyperglycemia (Olson, LK, unpublished data). The C1
activator has recently been cloned (5), and its role in repression of the insulin promoter
has yet to be investigated. These results led us to hypothesize that high glucose
concentrations mediate insulin promoter repression through other mechanisms besides
affecting binding activity of de-l and the Cl activator.

Previous studies have suggested that glucose-mediated repression of insulin gene
transcription is associated with increased oxidative stress (6-8). Glucose generates
oxidative stress through several pathways including: polyol pathway flux, non-enzymatic
glycation, increased advanced glycation end-product (AGE) formation, hexosamine
pathway flux, and activation of protein kinase C (PKC) (9). Furthermore, B-cells
cultured in high levels of glucose and treated with antioxidants partially prevented
decreased insulin gene promoter activity and de-l binding (7). Even though reactive

oxygen species (ROS) might mediate glucose-induced insulin promoter repression, the

exact mechanism remains unclear. Since ROS can activate stress-activated signaling
pathways, this led us to hypothesize that stress-activated kinases, such as c-Jun N—
terminal kinase (JNK) might mediate glucose-induced repression of insulin promoter
activity.

de-l, an important transcription factor that regulates insulin gene expression,
partially mediates glucose-induced insulin promoter repression. Because over-expression
of de-l is insufficient to prevent insulin promoter repression in cells cultured in high
levels of glucose, we further characterized other potential promoter targets. Functional
analysis of a truncated insulin promoter vector demonstrated that additional distal
promoter elements might be involved in glucose-induced insulin promoter repression.
Therefore we investigated the role of the distal promoter region including sequences from
—327 to —-269 in the repression of insulin promoter activity induced by chronic

hyperglycemia.

II. LITERATURE REVIEW

1. Endocrine functions of the pancreas

The pancreas is an organ with exocrine and endocrine functions. It is derived
from endodermal cells of the upper duodenal region of the foregut (10, 11) that leads to
dorsal and ventral protrusions developing into mature pancreas (12, 13). The exocrine
functions, which constitute more than 98% of the pancreas, consists of secretion of
digestive enzymes and bicarbonate solutions by the pancreatic acini and ducts into the
duodenum. The endocrine functions are performed by islets of Langerhans, which in
humans consist of one to two million round clusters (islets) of cells. The islets are
composed of alpha (or), beta (0), delta (6), and pancreatic peptide (PP) cells. The a—cells,
which constitute 20% to 25% of the islet, are located at the periphery. or-Cells secrete the
hormone glucagon, which regulates carbohydrate metabolism by inducing glycogen
breakdown, gluconeogenesis, and synthesis of ketones. Thus, the overall role of
glucagon is to increase glucose and ketone plasma concentrations. B-Cells, which make
up 60% to 70% of the islet, are located centrally in the islet and secrete the hormone
insulin. Insulin, which is often seen as the most important controller of metabolism,
maintains blood glucose levels within a narrow range of 80 to 130 mg/dl (4.4 to 7.2 mM)
by inducing glucose uptake by muscle cells, adipose tissue, and the liver. 8-Cells, which
constitute 10% of the islet, secrete the hormone somatostatin, which acts within the islets
to inhibit both insulin and glucagon secretion. The remaining cells are PP cells, the

source of pancreatic polypeptide, which has an unclear physiological function.

 

The islets are well vascularized which allows them to secrete hormones into the
blood, monitor blood glucose levels, and ensure efficient paracrine regulation among the
hormones. Each 0- and a-cell has a basal (arterial) and an apical (venous) face. Between
the lateral surfaces of neighboring B-cells run canaliculi that span the distance between
the arteriolar and venous ends of the cell. These connections allow the cells to be

exposed laterally to regulatory molecules, such as glucose.

2. Role of insulin in daily metabolic control

The main function of insulin is to regulate blood glucose levels, and it does so by
regulating the metabolism of carbohydrates, lipids, and proteins. After the consumption
of a meal, insulin is secreted. Insulin induces efficient storage of the excess nutrients
while inhibiting the mobilization of endogenous nutrients. The stored nutrients can be
made available during subsequent fasting periods to maintain required fiiel for the body.

The major targets of insulin action are the liver, adipose tissue, and muscle.

2.1 Carbohydrate metabolism

In the liver, insulin stimulates glucose oxidation and storage and at the same time
inhibits the output of glucose. Insulin stimulates oxidation of glucose by inducing
expression of glucokinase and its actions to phosphorylate glucose. Insulin then
increases storage of glucose by inducing the glycogen synthase enzyme complex. At the
same time, insulin inhibits the release of glucose by lowering the rate of glycogenolysis
by inhibiting glycogen phosphorylase activity and by decreasing glycogen glucose-6-

phosphatase levels (14). The hormone also lowers glucose production through

decreasing gluconeogenesis by inhibiting free fatty acids and amino acid mobilization
from fat and muscle to the liver (15). Insulin also decreases the levels of enzymes
involved in gluconeogenesis, such as pyruvate carboxylase, phosphoenolpyruvate
carboxykinase, and fi'uctose-l,6-diphosphatase. Insulin also decreases glucagon
secretion by or-cells, thus decreasing hepatic glucose output (16).

Insulin stimulates glucose uptake in muscle and adipose tissue by increasing the
net rate of translocation of glucose transporters (Glut-4) from intracellular compartments
to the plasma membrane (17, 18). In muscle, depending on the concentrations of insulin,
a certain percentage of glucose is metabolized through glycolysis and oxidation. The
remaining glucose is stored as glycogen. Insulin induces glycolysis, mainly through
regulation of hexokinase and 6-phosphofructokinase. It has been demonstrated that
insulin increases the expression of hexokinase H, but not hexokinase I in rodents and
humans (19). Studies in intact muscle have shown that insulin increases the flux of
glucose-6-phosphate by increasing the activity of 6-phosphofructokinase (20). Insulin
regulates glycogen synthesis by increasing the activity of glycogen synthase, which

generates glycogen from uridine diphosphate glucose (UDP-glucose) (21, 22).

2.2 Fat metabolism

Fat metabolism and mobilization are highly sensitive to insulin. Overall, insulin
enhances storage and blocks mobilization and oxidation of fatty acids. In adipose tissue,
insulin inhibits hormone-sensitive lipase activity, thus inhibiting lipolysis (23, 24).
Insulin also inhibits fat mobilization by retaining free fatty acids (FFA) and inducing

reesterification of FFA to triglycerides (TG) within liver and adipose tissue (23, 25). By

suppressing of lipolysis and release of FFA, insulin inhibits the generation of ketoacids in
the liver and induces the use of ketoacids in the peripheral tissues.

Triglycerides are transported by lipoproteins in the form of very low-density
lipoprotein (V LDL) from the liver. Insulin inhibits hepatic VLDL secretion and
promotes the hydrolysis of circulating VLDL to transfer FFA into adipose tissue (26). In
the liver, insulin induces FFA synthesis from glucose-derived pyruvate, which is
converted into acetyl-CoA and then into malonyl-CoA. Insulin also favors hepatic

synthesis of cholesterol from acetyl-CoA.

2.3 Protein metabolism

Insulin is known to regulate nitrogen balance. The role of insulin in regulating
nitrogen balance is best described in Type I diabetics because a lack of insulin creates
lean tissue atrophy and hyperarninoacidemia. The mechanism by which insulin regulates
nitrogen balance has been only recently elucidated. One mechanism is insulin enhances
protein and amino acid sequestration into target tissues. Insulin also inhibits proteolysis
(27, 28) and patients with diabetes have an increase in urinary nitrogen.

Insulin is also associated with regulation of protein synthesis (29, 30) and amino
acid oxidation (31). Insulin regulates the initiation of skeletal muscle protein synthesis
(32). Amino acid oxidation is control by the availability of amino acids, which is
decreased due to the inhibition of proteolysis.

Insulin secretion is very important in controlling the metabolic responses of daily
life, including feeding and exercising. The above review emphasizes the importance of

insulin in regulating blood glucose levels and mentions that insulin is also involved in

hr

\‘v.

“i

lipid and protein metabolism. All these pathways can not be separated; instead, they feed

back into and regulate one another.

3. Insulin biosynthesis
3.1 Insulin synthesis

Insulin is among the best understood of the polypeptide hormones, having been
the first protein for which the complete amino acid sequence was determined as well as
the first hormone whose gene was molecularly cloned. Pancreatic B-cells maintain a
readily available pool of insulin that can be rapidly secreted in response to a stimulus,
such as a rise in blood glucose concentration. Any increase in insulin release is
compensated for by a corresponding increase in insulin biosynthesis, so that B—cell insulin
stores are constantly maintained. Thus, insulin biosynthesis and processing is a highly
regulated and a dynamic process.

Biologically active insulin consists of two polypeptide chains, the A-chain (21
amino acids) and B-chain (30 amino acids), joined by two interchain disulfide-linked
bridges at A-Cys/B-Cys and another intrachain disulfide bridge between A-Cys/A-Cys
(33). Insulin is initially synthesized as a precursor molecule preproinsulin (34, 35),
which is composed of a 24-amino-acid N-terminal hydrophobic signal peptide, followed
by the insulin B-chain, then the 3l-amino-acid of the connecting peptide (C-peptide), and
then the insulin A-chain (Fig. 1). Preproinsulin mRNA is transported from the nucleus to
the cytoplasm where co-translation with ER takes place, and the newly synthesized
preproinsulin is located at the rough endoplasmic reticulum (RER) (37). In the RER,

signal peptidases, associated with the lumen side of the RER membrane, remove the

 

 

 

S — S A
/"'I ‘“\ #__.
S § Proinsulin
S S “folded”
. l B
Converting
enzymes

C-peptide

    
      
     

 

   

r S“ S\
I
S
I I
S Insulin S

Endoplasmic
reticulum

Golgi

Secretory
granule

Figure l. Proinsulin and insulin molecules. Insulin is first synthesized as a precursor

molecule named preproinsulin, which is then converted into proinsulin in the rough

endoplasmic reticulum (RER). Proinsulin then is converted into mature insulin by

removing the C-peptide in the secretory granule. Pictured was modified from Beme et a1.

(36)

signal peptide and proinsulin forms in the RER lumen. Once proinsulin is correctly
folded with the disulfide bonds, it is delivered to the ciS-Golgi apparatus and continues to
the trans-Golgi apparatus by vesicular transport. The trans-Golgi containing proinsulin
is coated with clatherin, and leads to immature buds that mature to form insulin granules.
In the insulin granules proinsulin becomes active insulin when the C-peptide is cleaved

by endopeptidases PC2 and PC3. The mature granules contain insulin and soluble C-

peptide (Fig. 1).

3.2 Insulin secretion

The major controlling factor for insulin secretion is the plasma glucose
concentration. Insulin secretion is also regulated by other factors, such as nutrients,
hormones, and acetylcholine. Because of the great importance of glucose metabolism to
induce insulin secretion, this mechanism has been studied extensively. The following
description reviews insulin secretion induced by glucose (Fig. 2).

When blood glucose levels rise, B-cells take up and metabolize the glucose, which

induces insulin release. Glucose enters the B-cells through the low affinity-facilitated
glucose transporter Glut2. The high Km (15 to 20 mM) and Vmax of Glut2 allow for
rapid equilibrium of glucose across the B-cell membrane (38). Glucose is then
phosphorylated into glucose-6-phosphate by glucokinase, which has a high Km (5 to 10
mM), ensuring that glucose phosphorylation rates are proportional to blood glucose
levels. Glucose phosphorylation is thought to be the major rate-limiting step in glucose

metabolism within the B-cell, thus playing an essential role in normal glucose-induced

insulin secretion. Glucose-6-phosphate is then metabolized through glycolysis and the

Fl;

  
  

 
  
 

tag I?

ATP *4 '5— ATP

Glucose ?

glucokinase

   
    

Pyruvate

    

PKC —-) Kinases

  
 

®@®@®

   

®@®

Insulin

Figure 2. Schematic representation of glucose-induced insulin secretion. Glucose is
transported into pancreatic B-cells through the facilitated transporter Glut2 and enters the
glycolytic pathway and then the TCA cycle, resulting in an increase of ATP. The high
ratio of ATP/ADP results in closure of the ATP-sensitive K+ channels, membrane
depolarization, and opening of voltage-dependent Ca 2+ channels, which increases Ca 2+
flux. This rise of intracellular Ca 2+, along with a variety of second messengers that lead
to activation of PKC and other kinases, lead to insulin secretion. Picture was modified

from McKinnon et al. (39) and Howell et al. (40).

10

Krebs cycle, raising the ratio of adenosine—tn'phosphate to adenosine-diphosphate (ATP:
ADP). The resulting high ratio of ATPzADP leads to. ATP binding to ATP-sensitive K+
channels, inducing inhibition (closure) of the channel (41, 42). Closure of ATP-sensitive
K+ channels leads to membrane depolarization and opening of voltage-dependent Ca2+
channels (43), leading to an influx of Ca2+. The rise in intracellular Ca2+, along with
second messengers such as increased cAMP and breakdown of membrane lipids, all lead
to mobilization of intracellular stores of Ca2+. Increased intracellular Ca2+ activates Ca2+/
calmodulin kinases, resulting in insulin exocytosis (44, 45). A rise in intracellular Ca“,
however, is not the only regulator of insulin exocytosis; G-proteins, phospholipase C,
protein kinase C, and other protein kinase activities are also involved in insulin release
(40, 46-49). Recently, Aspinwall et al. (50) demonstrated autocrine stimulation of insulin
secretion by insulin in isolated mouse pancreatic B-cells and clonal B-cell cultures.
Insulin binds to the insulin receptors on the surface of B-cells and activates tyrosine
phosphorylation of insulin receptors and insulin receptor substrates, IRS-l, IRS-2, and
PI3-Kinase (51-53). The activation of these proteins has been correlated with the

increase in intracellular Ca2+ and the secretion of insulin (50). Overall, insulin secretion

is very complex, and most likely involves the coordinated responses of several factors.

4. Regulation of insulin gene transcription

The human insulin gene is a small gene located on chromosome 11p15.5 (54), and
it consists of three exons and two introns (55). In contrast to humans, rats and mice have
two non-allelic insulin genes, which are coordinately expressed and regulated. The

restriction of insulin expression to B-cells is at the level of transcription, which is

11

regulated by the insulin gene promoter (5’ flanking sequences, -328 to +1). Mutational
and deletion studies of the rat insulin I and II gene and the human insulin gene have
identified multiple sequences along the insulin promoter that contribute positively and
negatively to its activity: the A, E, C, cAMP-regulatory element (CRE), CRE/CCAAT,
and the Negative Regulatory Element (NRE)/Z element (Fig. 3).

The A elements, or A boxes, consist of A1, A2, A3 and A5 elements, located in
the human promoter at —79 to —84, -123 to —134, -210 to —216, and —313 to -319,
respectively. All the A elements except the A2 element contain a core TAAT sequence,
where several transcription factors bind including the highly specific homeodomain
transcription factor, the pancreatic duodenal homeobox-l (de-l) (56). de-l is
expressed in B and 8 cells of the islets of Langerhans and in dispersed endocrine cells of
the duodenum. It is involved in regulating the expression of a number of key B-cell
genes in addition to insulin, such as Glut2 (57), glucokinase (58), the islet amyloid
polypeptide (59), and somatostatin (60). Gene knock out experiments of de-l
demonstrate that de-l is essential for development of the pancreas (61). Furthermore,
heterozygous mutations of de-l are found in MODY 4, a form of Type II diabetes (62).

The homeodomain protein ka6.1, which is expressed in developing and mature
B-cells, binds to the A3/A4 element of the rat I promoter (63). Mice deficient for ka6.1
display a dramatic reduction in B-cell numbers with a dramatic decrease in insulin
expression, yet other endocrine and exocrine cell types are not affected. ka6.1
functions as a transcriptional repressor and most likely plays an important role in B-cell
differentiation (63). Many other proteins can bind the A elements. For example, high

mobility group (HMG) proteins can bind to the A3/A4 region of the rat I insulin promoter

12

Pax4 PDXI 1 act .

1312/47 ka6. CREB ‘ P531) 512/47
. RE 3 pox] Pur.l|

// A5 NREI 152 A3 CRE CRE A2 C1 El Al or
II
2

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Figure 3. Human insulin promoter. Cis-acting regulatory elements along the promoter
regions of the human insulin gene are boxed. Protein binding to these elements is
indicated above in a highlighted box. The A1-5, El-2, C1-2, and G1 elements are termed

according to the nomenclature in (64). Pictured was modified from Molloul et al. (65).

13

(71). HMG I expression increases transcriptional synergy of del-bHLH

heterodimers in vivo (71 ).

The E-box has a consensus sequence of CANNTG. There are two E elements in
the human insulin gene; the E1 is located at —104 to —1 12 and the E2 is located at —230 to
—238. Although the E1 element is highly conserved among mammalian insulin
promoters, the E2 element is not well conserved. The E elements bind a ubiquitous class
of the basic helix-loop-helix (bHLH) protein family, E12/E47, which heterodimerizes
with the B-cell and the neuronal specific transcription factor Beta2. Beta2 is important
for both insulin gene transcription (66) and pancreatic development (67). Homozygous
knockout mice lacking Beta (i.e., Beta2") have fewer B-cells and develop diabetes (67).
E12/47 and Beta2 control both insulin and glucagon gene expression. The heterodimer
complex, however, has less binding affinity to the glucagon promoter than the insulin
promoter (68). The E12/47/Beta2 dimer synergyzes with de-l, and activate
transcription of an EA minienhancer in non-B-cells engineered to express these proteins
(69). The human E2 element has also been reported to bind upstream stimulatory factor
(U SF) (70).

Some transcription factors have actually been demonstrated to interfere with the
E12/E47 transactivation potential and induce insulin gene repression. The c-Jun
transcription factor, a member of AP-l family, inhibits the insulin gene by inhibiting the
transactivation potential of E47 (72). Some proteins can compete for binding with

E12/E47 and repress insulin gene expression. The Id proteins, a bHLH family that lacks

l4

the DNA-binding domain, can heterodimerize with BETAZ, resulting in a nonfunctional
heterodimeric complex with decreased DNA binding activity (73).

The C1 element is a cytosine-rich sequence that lies between the A2 and the El
element. The C 1 element binds to the B-cell specific Cl activator/RIPE3bl and to the
RIPE3b2 in the rat insulin gene II (74). Recently the C1 activator was cloned and
identified as a mammalian homologue of avian MafA/L-Maf (mMafA) (5).

The C2 element is a G/C rich conserved element located in the human insulin
gene from -253 to —244 (75). The glucagon and somatostatin promoters also contain this
conserved sequence, and all three genes, insulin, glucagon, and somatostatin, are
regulated by the same islet specific factor, which was termed pancreatic “islet cell-
specific enhancer sequence” (PISCES). The C2 rat I element binds the transcription
factor paired-homeodomain PAX6, which is a transactivator (76). PAX6 is found in all
islet cells, as well as in some neuroendocrine cells (76, 77). Mice homozygous for a
mutation in the PAX6 gene die after birth due to cranial and facial defects, and the
fetuses have a decrease in the number of islet cell types (76, 77). The remaining B-cells
have decreased levels of insulin mRNA and insulin content (76). PAX6 binding to the
human insulin promoter has not been demonstrated. Its role in the human insulin gene is
not known. An additional factor that binds to the C2 element was characterized and
named “D0”. Interestingly, the binding of this transcription factor is sensitive to the
redox-state of B-cells (75).

The cyclic AMP response element (CRE) binds the transcription factor CRE
binding protein (CREB), which is regulated by the second messenger cyclic AMP

(cAMP) through the phosphorylation of protein kinase A. Phosphorylated CREB

15

interacts with other transcription factors and induces transcripiton (reviewed in Ref. 75).
There are four putative CREs in the human insulin gene, two in the promoter and two in
the coding region (79). Glucose and other hormones stimulate cAMP accumulation, but
this has only a modest effect on insulin gene transcription (80). There are several
isoforms of CRE modulators that bind to CRE elements and can act as activators or
repressors (81). CRE modulators named CREM have different isoforms by alternative
splicing which can be activators or repressors. CREM activators interact with basal
transcription machinery more efficiently than CREB. In contrast, the CREM repressor
does not bind to the transcription machinery; instead, it competes for binding with CREM
activators (82, 83). Interestingly, the Goto-Kakisaki (GK) rat, a Type 11 diabetic animal
model, contains higher amounts of CREM repressor in islets than control rat islets.
Therefore, CREM may have a role in the repression of insulin mRNA levels in this
diabetic animal model (83).

The CRE element overlaps with a CCAAT motif, which binds the ubiquitous
transcription factor NF-Y (84). This transcription factor leads to basal transcription
activity and inhibits cAMP-induced transcription activity (84). The CCAAT/enhancer-
binding protein B (C/EBPB) binds to a downstream region of the CCAAT element, the
A2 element (85, 86). It was demonstrated that C/EBPB inhibits insulin gene expression
by interfering with the transactivation domain of E47 (86). C/EBPB expression is
upregulated in B-cells chronically exposed to high glucose concentrations, such as in
Zucker diabetes fatty rats and 90 % pancreatectomized rats, suggesting a possible role of

this transcription factor in repressing insulin transcription in Type II diabetes (85).

16

The negative regulatory element (NRE) is located between —260 and -281 in the
human insulin gene. It has been shown to inhibit insulin gene transcription in insulinoma
cells (87, 88). The NRE has been suggested to act as a silencer because, when linked to a
heterologous promoter, it represses transcription (88). The NRE, however, can be
positively regulated when upstream sequences from NRE are linked to NRE in the
heterologous promoter (88).

Recently, it was demonstrated that the NRE is not a silencer in all cell systems. In
fact, removal of this insulin region causes a marked loss of activity in rat fetal islets (64).
The NRE functions as a potent activator when placed upstream of a minimum rat
promoter or a heterologous promoter in both fetal and adult islets. However, in the same
constructs the NRE acted as a repressor in tumor B-cell lines, non-B-cells, and primary-
cultured fibroblasts (89). Because of the positive effects of the NRE element in the islets,
it was renamed the Z element (89). Several complexes bind to the Z element. In
insulinoma cell lines, it has been demonstrated that both Oct 1 (88) and the
glucocorticoid (90) receptor bind to the ME element. In contrast, these transcription
factors were not contained in the complex that bind to the Z element in fetal islets (89).
Further experiments are needed to clarify the role of M in insulin gene transcription.

The G element is characterized by the GAGA sequence. It is located at positions
-40 to --57 and is essential for gene transcription in fetal rat islets (91), and to a lesser
degree in HIT-T15 cells (92). The ubiquitous zinc finger Purl binds to G1 element in the
rat insulin I and II genes, and stimulates promoter activity (93). The role of the G1
element in the human insulin gene does not seem critical Since its mutation does not

affect promoter activity (94).

17

The complexity of the insulin promoter suggests that it is regulated by a
combinatorial mechanism. The most studied cooperative interactions are between the E
and A elements. Experiments performed in the rat insulin I gene promoter minienhancer,
which contains just the E2 element juxtaposed to the A3/A4 sequence elements,
demonstrated that deletion of either the E2 element or the A3/A4 element eliminate
promoter activity, suggesting synergisms between these elements (95). It has been
demonstrated that transcription factors that bind to the A3/A4 elements, such as de1, act
through a protein-protein interaction domain to recruit multiple proteins, including E47,
BETA2, and HMG I, to the E2A3/A4 minienhancer (71). Interactions between de-l
and E47 can also recruit co-activators such as p300 to induce rat I minienhancer promoter
activity (96).

Studies of the human insulin promoter have demonstrated that the E1 element
interacts with the A1 element and A2/C1 elements (64). The A2/C1-E1-A1 region alone,
however, does not explain all of the activity of the human insulin promoter, suggesting

that other transcription factors and DNA elements are required for full promoter activity.

5. Diabetes mellitus
Diabetes mellitus used to be considered a disease of minor significance for global
health, but is now one of the largest threats to human health. Changes in the human
environment and human lifestyle have led to large increases in both obesity and diabetes.
This disease now affects 151 million people worldwide and 14.2 million Americans (97).
There are two main forms of diabetes mellitus, insulin dependent-diabetes (IDDM

or Type I) and non-insulin dependent diabetes mellitus (NIDDM or Type II).

18

5.1 l

med
dent
Em'i
desti
med
lilI‘c
insu
11 d

hes

the
list
PEI

fact

5.1 Type I diabetes mellitus

Type I diabetes, generally observed in children, is primarily due to autoimmune-
mediated destruction of pancreatic B-cell islets, resulting in an absolute insulin
deficiency. The etiologic agents that induce this disease are not well understood.
Environmental factors such as virus infections are most likely to be involved in the
destruction of B-cells. There is also Type I diabetes that is not related to an autoimmune-
mediated destruction and is most likely due to an inherited predisposition that does not
involve the histocompatibility genes. People with Type I diabetes must take exogenous
insulin for survival. This form of diabetes has a very low frequency compared with Type
II diabetes. A comprehensive review of Type I diabetes is beyond the scope of this

thesis.

5.2 Type II diabetes mellitus

Glucose homeostasis is maintained by a balance between glucose production by
the liver and glucose utilization by insulin-dependent tissues (e.g., fat and muscle) and
insulin—independent tissues (e.g., brain and kidney). Glucose utilization is regulated
primarily by insulin and glucagon. Glucose utilization generally depends on three
factors: 1) the ability of the body to secrete insulin both acutely and in a sustained
fashion, 2) the ability of insulin to inhibit hepatic glucose output and to promote glucose
disposal, and 3) and the ability of glucose to enter cells that do not require insulin. Type
11 diabetic patients are generally characterized by two pathological defects. One is
decreased insulin action in peripheral tissues that need to increase the uptake glucose or

to inhibit hepatic glucose output, a phenomenon known as insulin resistance (98). The

19

other is the inability of B-cells to secrete sufficient amounts of insulin necessary to
compensate for insulin resistance (99).

These two primordial defects in the pathogenesis of Type II diabetes are caused
by a combination of genetic and nongenetic factors. The genetic factors are not well
characterized, while the nongenetic factors include increased age, high caloric intake,
obesity, central adiposity, sedentary lifestyle, and low birth weight. Type II diabetics
may control their blood glucose levels with a balanced diet and exercise, pharmacological
therapies, and sometimes with exogenous insulin. This type of diabetes accounts for 90%

of total global diabetes cases.

5.2.1 B—Cell failure during the progression of diabetes
The inability of B-cells to secrete sufficient amounts of insulin necessary to

compensate for insulin resistance is generally termed B-cell failure. Studies performed
on Zucker diabetic fatty (ZDF) male rats and 90% pancreatectomized rats, both Type 11
diabetic animal models, suggest that there are four phases that lead to B-cell deterioration
(reviewed in Ref. 97, 98). These studies associate changes in islet morphology, insulin
secretion, and gene expression with the hyperglycemic state. The description of the
following phases is hypothetical, and results that lead to these hypotheses can be found in
the above-mentioned citations.
B-Cell compen_SLfion and dysfunction in the progression of in_sulin resistance and Type II
Mes:

Following there is a general description of the characteristics of the Zucker diabetic

fatty (ZDF) male rats and the 90% pancreatectomized rats models, and their relative

20

controls, which are used in the description of the four phases of B-cell failure during the
development of diabetes. Zucker fatty diabetic male rats have a genotype of homozygous
fa/fa leptin receptor. They develop insulin resistance and glucose intolerance between 3
and 8 weeks of age (prediabetic at 7 weeks old) and usually become overtly diabetic
between 8 and 10 weeks. Lean heterozygous +/fa or homozygous +/+ normal leptin
receptor litterrnates develop neither insulin resistance nor diabetes, and they are called
Zucker lean controls (ZLC). A B-cell function analysis was performed in ZLC,
prediabetic, and diabetic rats. The serum glucose levels of prediabetic ZDF rats were
higher than in controls, but the difference was not significant. The serum triglyceride
(TG) levels of prediabetic animals were normal, in contrast with the diabetic animals,
which had a high level of TG. The body weight of the prediabetic animal was only 14 %
higher than that of the lean animal at 7 weeks old, as well as the diabetic rat compared to
ZLC 9-12 week old. Both prediabetic and diabetic animals are insulin resistant, with
insulin levels that are 8 to 12-fold higher than those of controls.

In the 90 % pancreatectomized (Px) animal model, 85 to 95 % of the pancreas is
removed. Regeneration of islets lasts no longer than ~ 10 days after surgery, but there is
a stable mature population of islets exposed to hyperglycemia from the first week post-
Px. The animals gain normal weight and are metabolically stable without needing special
diets or hypoglycemic therapies that might affect the B-cells.

Phase I: B-cell adaptation for insulin resistance. To compensate for insulin
resistance, the B-cells of prediabetic ZDF animals and Px animals increase in mass with
increased neogenesis (regeneration of islets from precursor cells) and replication. The B-

cells also become hypertrophic with multiple irregular projections into the surrounding

21

exocrine pancreas. There is significantly increased insulin secretion at each glucose
concentration tested, with a left shift in the dose-response curve relating glucose
concentration and insulin secretion. The B-cells have defects in the normal oscillatory
pattern of insulin secretion, indicating impairment of the normal feedback mechanism
between glucose and insulin secretion.

Phase II: decompensation of B-cells exposed to mild hyperglycemia. The B-cells
lose the ability to secrete insulin in response to acute glucose, but there is preservation of
insulin secretion to other secretagogues such as arginine. There are changes in important
gene expression, such as decreased Glut2, glucokinase, pyruvate carboxylase, genes
involved in ion channels, and transcription factors (de-l, HNFs, ka6.1, and Pax6).
There is increased expression of lactate dehydrogenase (LDH), glucose-6-phosphatase,
and the transcription factor c-Myc. There are some differences in gene expression
between both models, such as an increase in hexokinase in B-cells of Px, in contrast to the
hexokinase repression in B-cells of the ZDF rat.

Phase III: decompensation of B-cells exposed to severe hyperglycemia. B-Cells
show total loss of insulin response to glucose and impairment of insulin response to other
non-glucose secretagogues. There is B-cell degranulation, which is correlated with
decreased insulin mRNA. There are alterations in metabolic genes, such as further
decreased expression of Glut2, glucokinase, and glycerol'phosphate dehydrogenase, as
well as decreased expression of potassium channel Kir6.2 and calcium ATPase. In the B-
cells of Px, there is a marked increased in the expression of nitric oxide synthase (iNOS),
antioxidant genes such as heme oxygenase-1, Mn-superoxide, and glutathione

peroxidase, and antiapoptotic genes such as gene A20. These phenotypic changes can be

22

beneficial or detrimental to B-cells. For example, the induction of these protective genes
may be a compensatory mechanism that promotes survival of B-cells. In contrast to B-
cells of Px, B-cells of ZDF rat have a marked decreased in iNOS expression. This
contradictory result can be explained if the isolated islets from Px were contaminated
with ductal cells, which could led to an increase in iNOS expression. In addition, it has
been demonstrated that NO can impair glucose-induced insulin secretion (102). Even
though all these alterations occur, some B-cells still remain intact and are able to maintain
some insulin secretion and prevent more profound B-cell deterioration.

Phase IV: decompensation with structural damage. Insulin granules also contain
amylin peptide, which is co-released with insulin. The granule content of amylin is only
1 to 2 % those of insulin. Amylin tends to polymerize, and amylin fibrils accumulate
extracellularly within the islets of Type II diabetics. The mechanisms of amylin fibril
formation are not well understood, but they can have destructive effects on B-cell
function. In this phase, glycogen deposits and lipid droplets are also observed, and
investigators have correlated these accumulations of glycogen and lipids to lipid toxicity.
Finally B-cells go through apoptosis, a process that has been difficult to quantify since in
a chronic situation it is short-lived.

As the 90 % pancreatectomized rat and the Zucker fatty rat models demonstrate,
there are many phenotypic changes in B-cells that lead to the loss of glucose-induced

insulin secretion, and eventually to B-cell apoptosis.

23

6. Pathology of B-cells in Type II diabetes

Type II diabetes results from the failure of B-cells to compensate for the increased
insulin demand due to increased insulin resistance. It has been demonstrated that once
diabetes is established, chronic hyperglycemia and hyperlipidemia can induce deleterious
effects on B-cell firnction, referred to as glucose toxicity and lipotoxicity, respectively.
This section will focus on in vitro and in vivo Type II diabetes models where it has been

shown that gluco-lipo toxicity induces B-cell dysfunction.

6.] Glucotoxicity

As early as 1948, Dohan and Luckens (103) suggested that high levels of glucose
induce some changes in B-cell phenotype. They administered large doses of glucose to
normal cats, inducing permanent hyperglycemia and degeneration of islet of Langerhans,
and ketonuria was observed. Dohan and Luckens ( 103) proposed that hyperglycemia
could play a role in the pathogenesis of diabetes. In the last two decades, the hypothesis
that hyperglycemia could be responsible, at least in part, for B-cell dysfunction has
received considerable attention, and studies in different Type 11 animal models have been

performed.

6.1.1 Studies in Type II diabetic patients

Studies in the mid-19705 demonstrated that normalization of blood glucose levels
for short periods of time restored glucose-induced insulin secretion in Type II diabetics
(104). Brunzell et al. (105) demonstrated that the first-phase of glucose-induced insulin

secretion is lost in humans with high plasma glucose concentrations. Vogue et al. (106)

24

showed that decreased glucose-induced insulin secretion is partially restored after 20 hrs
of insulin infusion in hyperglycemic Type II diabetes patients. These studies clearly
point to the important role that hyperglycemia plays in the defects of B-cell function
associated with Type II diabetes and demonstrate that insulin response to glucose is

partially restored upon normalization of plasma glucose levels.

6.1.2 In viva studies of animal models of Type II diabetes and in vitra studies of
isolated islets and insulinoma cells.
In the late 19805, in viva Type H diabetic models suggested that hyperglycemia
leads to decreased glucose-stimulated insulin secretion. Leahy J.L et al. (107, 108)
infused normal rats in viva with various concentrations of glucose for 48 hrs and then
measured insulin response to glucose (2.0 mM and 16.7 mM) in the in vitra isolated
perfused pancreas. During the infusion period, the animal became hyperglycemic.
Glucose-induced insulin release was blunted afier 48hrs of glucose infusion, and insulin
content was decreased. In contrast, in a second protocol, when phlorizin (a reagent used
to decrease plasma glucose levels by inhibiting renal tubular reabsorption of glucose) was
added during a second 48hrs of infusion, glucose-induced insulin release was completely
restored. These results indicate that high levels of glucose decrease glucose-induced
insulin release and possible biosynthesis.
Studies in the 90 % pancreatectomized rat model have provided further evidence
that hyperglycemia induces loss of glucose-induced insulin release (109). The ninety
percent pancreatectomized rats showed moderate hyperglycemia 4 weeks after surgery.

Insulin response to glucose was blunted but arginine-induced insulin response was intact

25

in intraperitoneal and intravenous glucose tolerance tests and intravenous arginine
challenge given 6-7 weeks after surgery. Similarly, when the pancreatic remnant was
perfused in vitra, insulin release was markedly reduced after challenge with high glucose
concentrations and remained the same after challenge with arginine. These data suggest

that chronic stimulation of B-cells can lead to loss of glucose stimulation of insulin

secretion.

Glucose toxicity on B-cell function was demonstrated in additional in viva studies
by transplanting islets grown in high glucose levels under the kidney capsule of syngenic
mice that were made diabetic with streptozotocin (110). The islets grown in media
containing high levels of glucose (22.0 mM glucose) failed to restore blood sugar levels
of diabetic animals. In contrast, diabetic animals became normoglycemic with islets
grown in normal media (5.0 mM glucose).

Investigators have demonstrated that altered glucose-induced insulin secretion in
diabetes is associated with reduced glucose transporter Glut2. Perifused islet studies in
Zucker fat diabetic rats show impairment in the ability of the islets to secrete insulin in
respond to a glucose stimulus (101). Studies in this animal model demonstrated that loss
of glucose-induced insulin secretion is accompanied by a marked reduction of Glut2
protein (111), which is associated with reduced levels of Glut2 mRNA (112). Reduced
Glut2 expression and mRNA were also observed in islets of the 90% pancreatectomized
rat model (113). Whether Glut2 contributes to the secretory impairment of diabetes is
unclear. Some reduction of Glut2 transport capacity has been demonstrated in Zucker
diabetic fatty rats (111, 114), but the changes may not be sufficient for transport to

become rate limiting for glucose metabolism. The rate-limiting step in glucose

26

metabolism is the phosphorylation of glucose by glucokinase, and glucose transport
capacity of B-cells exceeds that of glucose phosphorylation by lO-fold (115). In addition,
De Vos er al. demonstrated that human B—cells have very little Glut2 but abundant Glutl.
Studies have demonstrated that Glut2 interacts with glucokinase to increase glucose
phosphorylation and thus to enhance insulin secretion (116, 117). Nevertheless, loss of
Glut2 expression, at least in rodents, may be involved in B-cell dysfunction in diabetes.

In vitra studies of cultured human pancreatic islets, isolated rat islets, and
insulinoma cell lines have also demonstrated that hyperglycemia mediates B-cell
dysfunction. Eizirik et al. demonstrated that prolonged exposure of human pancreatic
islets to high levels of glucose in vitra impairs B-cell function (118). Isolated islets from
adult cadaveric organ donors were cultured for seven days in media containing 5.6, 11, or
28 mM glucose. Insulin content was decreased in islet cultured in 11 or 28 mM glucose,
compared to 5.6 mM glucose. The isolated human islets were submitted to a 60-min
stimulation with a low (1.7 mM) followed by a high (16.7 mM) concentration of glucose.
The islets cultured in the high concentration of glucose (28 mM) had a reduced insulin
secretion compared to the islets incubated in 5.6 mM glucose. The rates of insulin
biosynthesis, glucose oxidation, and total protein biosynthesis were significantly lowered
in islets cultured in 28 mM glucose, compared to 11 and 5.6 mM glucose. There was no
change in islet DNA content in the three treatments. In addition, Briaud et al. (119)
demonstrated that isolated rat islets exposed to high levels of glucose for 6 weeks showed
decreased insulin mRNA levels. These results suggest multiple mechanisms by which

hyperglycemia may induce B-cell dysfunction.

27

Studies with insulinoma cell lines have also demonstrated some of the
mechanisms involved in hyperglycemia-induced B-cell dysfirnction. The insulinoma
HIT-T15 cell line, a Syrian hamster pancreatic islet transfected with SV-40 large T
antigen, shows similar characteristics of B-cell dysfunction as observed in in viva
experiments when cells are incubated in high levels of glucose (11.1 mM) for prolonged
periods of time (6 months). Insulin content, insulin mRNA, and insulin stimulated
secretion are reduced in HIT-T15 cells incubated serially in high levels of glucose (120).
These phenotypic changes in HIT-T15 cells were reversible when cells were cultured in
low levels of glucose (120). To understand the molecular basis for the reduction in
insulin mRNA, Olson et al. transiently transfected HIT-T15 cells with a chloromphenicol
acetyl transferase (CAT) reporter gene controlled by the 5’regulatory sequences of the
human insulin gene (INS(-327)CAT) and demonstrated that cells serially cultured in 11.1
mM glucose have decreased insulin promoter activity (121). In addition, decreased
insulin promoter activity was associated with a reduction in binding activity of two
important insulin gene transcription factors, de-l and C1 activator (RIPE3b1) (3, 121).
Olson et al. also demonstrated that high levels of glucose reduce de-l mRNA by a post-
transcriptional mechanism (3). These studies suggest that B-cells exposed to chronic
hyperglycemia decrease glucose-induced insulin secretion, in part, because of reduced
insulin mRNA levels, and this correlated with reduced insulin promoter activity.

Subsequent work by Harmon et al. (122) assessed the temporal loss in binding
activity of de-l and C1 activator in HIT-T15 cells cultured in high levels of glucose.
HIT-T15 cells serially cultured in high levels of glucose reduced insulin promoter activity

by passage 80 to 85. The C1 activator binding activity was reduced by passage 81; in

28

contrast, de-l binding activity was reduced by passage 106. These results suggest that
the loss of Cl activator may play a dominant role in glucose toxicity of B-cells. Studies
analyzing whether reconstitution of de-l would restore insulin promoter activity in HIT-
T15 cultured in high levels of glucose have indicated that de-l partially reconstitutes
insulin promoter activity (122). Experiments reconstituting insulin promoter activity
with C1 activator have not been yet performed because it was not cloned until recently.
These studies suggest that high levels of glucose affects insulin gene transcription by
decreasing de-l and Cl activator binding activity.

Similar effects of glucose toxicity on B-cell function were observed when ETC-6
cells, a mouse B-cell transformed with the large T antigen of SV40 driven by the rat
insulin II promoter (123, 124), were incubated in elevated glucose for prolong periods of
time (125). B-TC-6 cells, incubated for up to 41 weeks in 11.1 mM glucose, had
decreased insulin content and insulin mRNA levels compared to cells incubated in 0.8
mM glucose (125). Insulin promoter activity was also decreased and this was associated
with a loss in C1 activator binding activity (125). Interestingly, no change in de-l
binding activity was observed in BTC6 cells cultured serially in high levels of glucose.
These results suggest that chronic exposure of BTC-6 cells to high levels of glucose
concentrations decreases insulin gene transcription, in part, by reducing C1 binding
activity.

Studies in INS-1 cells, another insulinoma cell line, also demonstrated that high
levels of glucose decrease insulin mRNA, and that decreased insulin mRNA is recovered

When cells are cultured in low levels of glucose (126). The investigators also

29

demonstrated that reduced insulin mRNA is associated with reduced de-l and C1

binding activities.

Lu et al. (86) demonstrated that chronically exposing HIT-T15 cells and INS-1
cells, both insulinoma cells lines, to high levels of glucose leads to increase in the
CAAT/enhancer-binding protein B (C/EBPB) gene expression, suggesting that glucose
toxicity in B-cells may be related to increased levels of C/EBPB. Interestingly, over-
expressed C/EBPB represses rat I insulin promoter activity by interacting with E47
transcription factor and inhibiting dimerization and DNA binding of E47 (86). In viva
studies in the Zucker diabetic fatty (ZDF) rat and the 90 % pancreatectomized rat models
demonstrated similar findings in the regulation of C/EBPB expression. Seufert et al. (85)
demonstrated that de-l expression is downregulated while C/EBPB is upregulated in
ZDF rats and in 90 % pancreatectomized rats. These results suggest that hyperglycemia-

induced repression of insulin expression may be mediated by upregulation of C/EBPB.

c-Myc transcription factor has also been shown to be involved in the glucose-
induced repression of insulin gene expression (4). c-Myc expression is induced in
diabetic rats following partial pancreatectomy and in rats made hyperglycemic with
glucose clamps (4). Over-expression of c-Myc represses insulin gene promoter activity
in insulinoma cells and in primary rat islets by inhibiting Neuro D/BETAZ-mediated
transcriptional activation (127). Furthermore, Kaneto et al. demonstrated that glucose-
induced c-Myc expression is activated by PKC 02 in primary rat islets (127). Other
potential roles of induced c-Myc during diabetes have been correlated to cell replication.

c-Myc is involved in cell cycle progression, differentiation and apoptosis (128-130). B-

30

cell hypertrophy is a compensatory mechanism induced by hyperglycemia, as described
in the section of B-cell adaptation to diabetes. Recently, Laybutt et al. (131)
demonstrated that the acetyl-CoA carboxylase gene, a gene involved in fatty acid
oxidation, is upregulated in the 90 % pancreatectomized rat. Acetyl-CoA carboxylase
gene promoter has c-Myc binding sites, and thus increased c-Myc could upregulate
acetyl-CoA carboxylase expression. c-Myc may be involved in a variety of phenotypic

changes in B—cell dysfunction during diabetes.

AS described so far, hyperglycemia causes a variety of changes in B—cells that
trigger B-cell dysfunction. This idea was further explored by Jonas et al. (4) using the in
viva 9O % pancreatectomized animal model. The authors showed that hyperglycemia
decreases gene expression involved in glucose-induced insulin release in parallel with the
reduction of transcription factors necessary for B-cell development and differentiation.
At the same time, hyperglycemia induced the expression of genes that are only minimally
expressed in B-cells, such as lactate dehydrogenase A and hexokinase I. These changes
occurred in parallel with increased B-cell hypertrophy, which is a typical characteristic of
B-cells compensating for insulin resistance. The B-cell changes were specific to
hyperglycemia, because normalizing of blood glucose levels with phlorizin prevented
phenotypic changes. In conclusion, chronic hyperglycemia leads to B-cell dysfunction
and B-cell hypertrophy by a variety of alterations in gene expression that regulate B-cell

development and differentiation.

31

6.2 Lipotoxicity

Increased incidence of type II diabetes is associated with obesity, and in many
patients, weight loss can significantly control the condition. There are high levels of
plasma free fatty acids (FFA) and increased fat oxidation in many obese diabetics.
Increased plasma FFA has been associated with increased insulin resistance and
increased liver gluconeogenesis (132). In obese Type II diabetic animal models there is
excessive FFA within the islets and this has been suggested to contribute to B-cell
dysfunction, known as lipotoxicity. Prolonged exposure of islets to FFA stimulates basal
insulin secretion (133, 134) and inhibits glucose-induced insulin secretion (133, 135).
Elks et al. (135) demonstrated that isolated rat islets chronically perfused with palmitate
suppress glucose-stimulated insulin release. Islets were perfused with 1 mM palmitate
for up to 4 hrs and then perfused with 3 or 17 mM glucose for 20 min. The results
showed that chronic exposure of islets to palmitate suppresses first and second phase
insulin release. Inhibitors of fat oxidation including or-bromostearate and methyl-3-
tetradecylglycidate reversed this suppression. In vitra studies by Gremlich et al. (136)
demonstrated that isolated rat islets chronically exposed to palmitate decreased insulin
gene transcription in the presence of 30 mM glucose by inhibiting de-l mRNA and
protein expression. Palrnitate also inhibited de-l binding activity to insulin and Glut2
gene promoter, as well as Glut2 and glucokinase mRNAs and protein expression. These
experiments suggest that the second phase of glucose-induced insulin secretion is
inhibited by palmitate through negative regulation of insulin and glucose sensing genes.

Thus, free fatty acids can contribute to B-cell dysfunction as observed in diabetes.

32

As previously discussed, both hyperlipidemia and hyperglycemia induce B-cell
dysfunction. The toxic effects of both conditions on B-cell fimction have been
controversial. The following review will provide evidence that suggests that
hyperglycemia is required for hyperlipidemia to damage B-cells. In vitra experiments by
Jacqueminet et al. (137) demonstrated that inhibition of insulin gene expression by
chronic exposure of isolated rat islets to palmitate requires the presence of elevated
glucose concentrations. Isolated rat islets were incubated for a week in the presence of
2.8 or 16.7 mM glucose with or without 0.5 mM palmitate. Insulin mRNA was increased
from 2.0 to 16.7 mM glucose and palmitate inhibited this glucose-induced increase in
insulin mRN A. The authors associated palmitate reduction in insulin mRNA to reduction
in insulin promoter activity through studies using HIT-T15 cells. The conclusion of these
experiments was that long-term exposure to palmitate coupled with the presence of high
levels of glucose, represses insulin mRNA. In viva experiments have demonstrated that a
rise in TG and FFA causes B-cell dysfunction in Zucker diabetic fat rats. These animals
have increased FFA and TC in the prediabetic phase and increased islet triglyceride
content immediately before the hyperglycemia state (138). In addition, diet restriction
before hyperglycemia appears to reduce hyperlipidemia, hypertriglyceridemia, and
accumulation of fat in islets, as well as prevent of hyperglycemia and B-cell dysfunction
(1 38)._ These results indicate that increased TG and FFA in Type 11 diabetic rats induce
B-cell dysfunction independently.

Recently, Harmon et al. (139) investigated the roles of hyperglycemia and

hYIJerlipidemia on B-cell dysfunction in the Zucker diabetic fatty rat (ZDF). The ZDF

rats were treated with bezafibrate, a lipid-lowering drug that does not affect plasma

33

glucose levels, or phlorizin, a drug that reduces plasma glucose without affecting lipids
levels, after they had become hyperglycemic and hyperlipidemic. Treating rats with
bezafibrate lowered plasma TG levels but did not prevent the rise of TG levels in the
islets and did not prevent the decrease in insulin mRNA. In contrast, treating ZDF rats
with phlorizin lowered plasma glucose levels and lowered islet TG content as well as
preserved insulin mRNA levels. These results indicate that high levels of glucose, and
not high levels of lipids, induces B-cell dysfunction such as decreased insulin mRNA and
elevated islet TG content. In addition, Briaud et al. (140) investigated the role of
hyperlipidemia and hyperglycemia on B-cell dysfunction in the Goto-Kakizaki (GK) rat,
a lean Type H diabetic animal model. Feeding GK rats with a high fat diet for 6 weeks
led to an increase in epididyrnal fat weight, plasma TG, and FFA levels. High-fat fed GK
rats showed impaired insulin secretion, decreased insulin mRNA, and low insulin content
in contrast to high-fat fed Wister rats, a non-diabetic. lean rat. Islet TG content and islet
glucose oxidation were not affected by the high fat diet. High-fat fed GK rats treated
with insulin prevented reduction of glucose-induced insulin secretion. These results
indicate that hyperlipidemia-induced B-cell dysfunction requires the presence of
hyperglycemia. Thus, glucotoxicity and lipotoxicity are closely interdependent, in the
sense that lipotoxicity does not exist without hyperglycemia. This is consistent with the
clinical observation that hyperlipidemic individuals are not necessarily diabetic, and their
B—cell function is typically normal. These results are very important for understanding
glucose toxicity on B-cell function in the Zucker diabetic fatty rat model, since high

levels of lipids are always present in these animals.

34

F

One of the mechanisms by which lipotoxicity induces B-cell dysfunction is
through generation of reactive oxygen species and uncoupling respiratory chain reactions.
~ Carlsson et al. (141) demonstrated that isolated pancreatic islets exposed to palmitate
have decreased glucose-induced insulin secretion, decreased ATP levels, decreased islet
cell mitochondria membrane potential, and increased, yet, uncoupled respiration. They
also observed an increase in B-cell mitochondrial volume in islets exposed to palmitate.
Use of a mitochondrial uncoupler reagent (carbonyl p-phenylbydrazone) at
concentrations that decreased mitochondrial membrane potential to a similar level as
palmitate reduced glucose-induced insulin secretion. In addition, islets exposed to
palmitate increased the generation of reactive oxygen species but not of nitric oxide.
These results indicate that islets exposed to fatty acids mediate B-cell dysfunction, in part,
through oxidative stress and uncoupling oxidative phosphorylation. Interestingly,
chronic hyperglycemia also induces B-cell dysfunction, in part, through reactive oxygen
species, which will be explained in the next section. Taken together these findings
suggest that chronic hyperglycemia and hyperlipidemia interdependently may mediate B-

cell dysfunction through oxidative stress.

6. Mechanisms of hyperglycemia-induced damage

7.1 Oxidative stress.

Several experiments have demonstrated that oxidative stress may be a mechanism
through which hyperglycemia causes B-cell dysfunction. Oxidative stress occurs when

there is an imbalance between reactive oxygen species (ROS) and antioxidant levels

35

(142). High levels of glucose increase ROS levels, leading to oxidative stress. ROS are
eliminated by scavenger proteins such as catalases, superoxide dismutases, and
glutathione peroxidases, as well by antioxidants like glutathione and vitamins (143). It
has been demonstrated that hyperglycemia can lead to an increase of both activity and
mRNA levels of antioxidant enzymes such as Cu, Zn-superoxide dismutase, catalase, and
glutathione peroxidase (144). Thus, over-expression of antioxidant enzymes could be a
mechanism to compensate for glucose-induced oxidative stress. Three main mechanisms
are involved in the development of oxidative stress in the presence of hyperglycemia: 1)
protein glycation and formation of advanced glycation end products (AGES), 2) glucose

autoxidation, and 3) the polyol pathway.

A. Protein glycation and advanced glycation end products

Protein glycation results from the formation of covalent binding between the
aldehyde glucose group and the amino group of proteins. This binding is generated
through non-enzymatic glycation reactions, known as Maillard reaction (145). In the
presence of transition metals such as copper and iron, glycated proteins can donate an
electron to oxygen, leading to oxygenated free radicals (146, 147). Glycated proteins can
undergo irreversible modifications leading to advanced glycosylated end products
(AGES); this is a slow process and AGES are formed only on long-lived macromolecules

(142,145)

B. Glucose autoxidation

Glucose can oxidize when catalyzed by trace amounts of transition metals; this

process generates superoxide anions and carbonyl compounds. Superoxide anions can

36

generate hydrogen peroxide, which in the presence of transition metals produces reactive
hydroxyl radicals (149, 150). The rate of glucose oxidation is very slow, but it has been
demonstrated in diabetes that collagen breakdown occurs during autoxidative glycation

(151).

C. Polyol pathway

Glucose is converted to sorbitol by aldose reductase, which is then converted to
fructose by sorbitol dehydrogenase; this pathway is known as the polyol pathway (152).
Aldose reductase activity requires NADPH. Therefore, increased activity of the polyol
pathway can result in depletion of intracellular NADPH, which is also required for
antioxidant enzymes to eliminate R08 (153). In addition, enhancement of the polyol
pathway leads to an increased concentration of intracellular sorbitol due to its slow
diffusion rate. Therefore, hyperglycemia induces the accumulation of sorbitol inside the
cell, which induces osmotic stress, as observed in lens (154). Enhanced polyol pathway
leads to the generation of fructose, which can readily go through non-enzymatic

glycosylation, generating more ROS.

7.1.1 Oxidative stress in diabetes

Many investigators have reported that chronic exposure to hyperglycemia causes
tissue damage through oxidative stress (155-158). Increased glycated tissues such as
kidney, liver, brain, and lung are observed in diabetic conditions (159, 160). Glycated

proteins such as albumin, lens crystalline, and hemoglobin are also observed in diabetic

37

individuals, as well as advanced glycated end products (AGES) in LDL (161-163).
Increased glycated and AGE proteins can further generate ROS, which can exacerbate
diabetic conditions. In addition to other tissue damaged through ROS, pancreatic B-cells
are very susceptible to ROS because they have low expression of antioxidants such as

glutathione and catalases (164).

7.1.1.1 Oxidative stress in diabetic individuals and whole animals

Oxidative markers such as 8-hydroxy-2’deoxyguanosine (8-OHdG) have been
identified in the urine and blood of patients with Type II diabetes (165, 166). 8-OHdG is
a mutation induced by ROS that results in C:G to A:T transversions during DNA
replication. In addition, Ihara et al. (157). demonstrated increased levels of 8-OHdG in
pancreatic B-cells of the Goto-Kakizaki (GK) rat, which develops diabetes spontaneously.
Increased 4-hydroxy-2-neonal (HNE)-modified proteins are also observed in pancreatic
B-cells of GK rats (157). HNE is an or,[3-unsaturated aldehyde formed by lipid
peroxidation, which is induced by ROS. The authors showed further induction of 8-
OHdG and HNE when the GK rats were fed a 30 % sucrose solution for 4 weeks. In
contrast, 8-OHdG and HNE were reduced when GK rats were fed with voglibose, a
postprandial blood glucose suppressor that delays carbohydrate digestion and reduces the

rate of glucose absorption by inhibiting intestinal or-glucosidase.

Since hyperglycemia causes oxidative stress, one can hypothesize that antioxidants

could prevent hyperglycemia’s toxic effects on different organs and B-cells. Effects of

38

antioxidants were analyzed in Type 11 diabetic rat models. Tanaka et al. (7) demonstrated
that Zucker fat rats treated with antioxidants, N-acetyl-L-cysteine (NAC) or
aminoguanidine (AG), prevented a rise of 8-OHdG and HNE in the blood, and partially
prevented hyperglycemia, changes in glucose tolerance, defective insulin secretion, and
insulin content. Irnportantly, the authors also demonstrated that nitric oxide is not
involved in B-cell dysfunction, since an inhibitor of nitric oxide synthase had no

beneficial effects on B-cell dysfimction. This distinction is important because nitric oxide

is involved in cytokine-induced B-cell apoptosis observed in Type I diabetes.

Further studies by Kaneto et al. (6) demonstrated that antioxidants could improve
diabetic conditions. Diabetic C57BL/KsJ-db/db mice are an obese Type 11 animal model
where hyperglycemia is induced because of increased insulin resistance and subsequent
insufficient B-cell compensation. Mice were treated with NAC for 4 to 10 days after they
had become hyperglycemic. Mice treated with NAC showed improved glucose-
stimulated insulin secretion and an improved intraperitoneal glucose tolerance test.
Antioxidant treatment also preserved insulin content and insulin mRNA levels, as well as
improved detection of de-l in nuclei of islet cells. Overall, these results indicate that

hyperglycemia-induced B-cell dysfunction may be mediated by oxidative stress.

7.1.1.2 Oxidative stress in isolated islets and insulinoma cell lines

Under diabetic conditions, oxidative stress is also produced through the polyol
pathway (167, 168). Fructose is a reduced sugar and can be readily glycated to amine
groups, a process that generates ROS and AGES. Kaneto et al. (169) demonstrated that

treating HIT-T15 cells with 50 mM fructose or 25 mM D-ribose, a reduced sugar, induces

39

oxidative stress and apoptosis. In this study, B-cell apoptosis was prevented in cells
incubated with NAC. These results indicate that reducing sugars triggers oxidative stress,

which may explain B-cell deterioration in diabetes.

In support of the hypothesis that high levels of glucose induces B-cell dysfunction
through oxidative stress, Tajiri er al. (170) demonstrated that in isolated rat islets,
glucose-induced repression of insulin secretion is restored by aminoguanidine (AG), an
inhibitor of glycation and formation of AGES. Isolated islets were cultured in 38 mM
glucose with or without AG for 6 weeks, and then had a wash-out period of continued
culture at 11 mM glucose for 24 hrs without AG. Islets were subject to analysis of
insulin secretion at 3.3 or 27 mM glucose. AG treatment enhanced insulin response to 27
mM glucose two-fold compared to cells treated without AG. These results indicate that
inhibition of glycosylation reactions might prevent hyperglycemia-induced B-cell

dysfunction.

Effects of antioxidants on hyperglycemia-induced repression of insulin gene
expression and insulin promoter activity have been analyzed. Matsuoka et al. (158)
demonstrated that insulin promoter activity was decreased in HIT-T15 cells cultured in
40 mM D-ribose for three days, and this decrease was associated with reduced de-l
binding activity. Reduced insulin promoter activity and de-l binding activity were
prevented in HIT-T15 cells incubated in 10 mM NAC or lmM AG. Further experiments
by Tanaka et al. (7) also demonstrated that high levels of glucose reduce insulin gene
expression through ROS. HIT-T15 cells were cultured serially with 11.1 mM glucose

and antioxidants, NAC or AG. Insulin mRNA level, insulin content, insulin promoter

40

activity, de-l binding activity, and glucose-induced insulin secretion were all
moderately increased by NAC or AC. Overall, these studies indicate reduced insulin

gene expression by hyperglycemia is mediated through oxidative stress.

Glucose can also modify and activate transcription factors by 0-linked
glycosylation through the hexosamine pathway. In the hexosamine pathway, 0-linked
glycosylation is catalyzed by O-linked N-acetylglucosamine transferase (OGT), which
attaches the N-acetylglucosamine monosacharides (GlcNAc) to the hydroxyl group of
serine or threonine residues of intracellular proteins. Glucose is converted to N-
acetylglucosamine through several reactions, and one of the enzymes involved is
glutarnine: fructose-6-phosphate aminotransferase (GFAT), which converts fructose-6-
phosphate to N-acetylglucosamine-6-phosphate. Recently, Kaneto et al. (171)
investigated whether B-cell dysfunction, as induced by hyperglycemia, can also be
observed by artificially inducing the hexarnine pathway. Isolated pancreatic B-cells were
infected with adenovirus-mediated over-expression of GFAT or treated with
glucosarnine. Both treatments impaired glucose-stirnulated insulin secretion and reduced
the expression levels of insulin, Glut2, and glucokinase, as well as reduced de-l binding
activity. Importantly, glucosarnine increased hydrogen peroxide levels, and the
phenotypic B—cell changes induced by GFAT were prevented by treatment with N-acetyl-
L-cysteine (NAC) but not with an inhibitor of 0-linked glycation. These results
demonstrated that GFAT induces B-cell dysfunction as observed in B-cells treated with
high levels of glucose and that these changes are prevented by antioxidants. These
results suggest that hyperglycemia-induced B-cell dysfunction may be mediated through

the hexosamine pathway by inducing oxidative stress.

41

7.1.2 Mechanisms of oxidative stress-induced B-cell damage

Several investigators have demonstrated that hyperglycemia induces B-cell
dysfunction through reactive oxygen species (ROS), although the exact mechanism of
action for ROS-induced damage in B-cells is not known. There are at least four
mechanisms by which ROS can damage B-cells. First, ROS can modify DNA by forming
8-0HdG (157), which is a pro-mutagenic lesion. Second, ROS can modify membrane
lipids, inducing lipid peroxidation and further modifications to form aldehydes such as
HNE (157). Third, ROS can modify mitochondrial membrane potential, which
eventually interrupts the coupling of glucose metabolism to insulin secretion, as
demonstrated by Maechler et al. (172). Maechler et al. showed that pancreatic B-cell
dysfunction is induced in isolated rat islets and INS-1 cells (an insulinoma cell line)
treated with 200 11M H202 for 10 min. Insulin secretion was inhibited when isolated rat
islets were pretreated with H202. In contrast, H202 did not affect the secretory response
induced by plasma membrane depolarization. Similar results were obtained in INS-1
cells treated with H202. In addition, H202 treatment induced depolarization of INS-1
mitochondrial membrane but inhibited the hyperpolarization induced by glucose. H202
also inhibited glucose-induced increase of mitochondrial Ca2+. Consequently, H202
inhibited the increase of cytosolic ATP. These experiments demonstrate that H202
targets mitochondria and leads to inhibition of glucose-induced insulin secretion. Fourth,
Kaneto et a1. (8) demonstrated that ROS induces B-cell dysfunction by activating c-Jun
N-temlinal kinase (JNK). Insulin mRNA levels and DNA binding activity of de-l were
reduced in isolated rat islets treated with 50 [EM H202 for 48 hrs. Furthermore, H202

activated p38 kinase and protein kinase C. H202 -induced insulin mRNA repression was

42

prevented in rat islets infected with an adenovirus-mediate expression of a dominant-
negative form of JNK. In contrast, inhibition of p38 kinase or protein kinase C did not
prevent repression of insulin mRNA by H202. Islets treated with N-acetyl-L-cysteine
(NAC) prevented H202 induced repression of insulin mRNA. These results indicate that
H202 induces B-cell dysfunction by activating the INK pathway, suggesting that high

levels of glucose might mediate B-cell dysfunction by activating the JNK pathway.

The toxic effects of chronic hyperglycemia on B-cell function are caused, in part,
by oxidative stress. A compensatory mechanism to this oxidative stress is to enhance
expression of antioxidant and antiapoptotic genes. Using the 90 % pancreatectomized rat
model, Laybutt et al. (173) showed that the expression of antioxidant genes such as
glutathione peroxidase, and antiapoptotic genes such as A20 increased within the first
week of surgery. The changes in gene expression were reversed when blood glucose
levels were reduced by treatment with phlorizin. In conclusion, chronic exposure of B-
cells to hyperglycemia increases antioxidant and antiapoptotic gene expression and this is
likely a mechanism to compensate for induced oxidative stress and to protect against

apoptosis.

7.2 Protein Kinase C activation

Protein kinase C (PKC) is an important mediator of signal transduction in response to
several cellular signals (174), such as in glucose-induced insulin secretion (175, 176). In

the diabetic state, PKC is induced in a variety of tissues including aorta, retina, heart,

43

renal, brain, and peripheral nerve (177). Some PKC isoforms activities are increased by
hyperglycemia and associated with abnormalities observed in diabetes, especially in
cardiovascular, retinal, and renal tissues (178),(179, 180). Recently, Kaneto er al. (127)
demonstrated that PKC activity is increased in isolated pancreatic B-cell cultured in high
glucose concentrations. The authors showed that PKC B is involved in glucose-induced c-
Myc expression, which has been shown to repress insulin gene expression and alter the

expression of other B-cell genes.

7. Protein kinase regulation of B-cell function

8.1 Growth

A small percentage of normal adult pancreatic B-cells go through mitogenesis
(181, 182). This mitogenesis is about 0.5 % of the population of B-cells in a pancreatic
islet (181). Glucose and other secretagogous can induce proliferation (182).
Interestingly, B-cell growth is also observed during B-cell compensation for insulin
resistance in the prediabetic phase (100). It has been postulated that transient and mild
hyperglycemia induces B-cell growth (100). However, as the hyperglycemia and
hyperlipidemia states increase, 0-cell growth is not sufficient to compensate for insulin
resistance (183). Therefore, understanding the molecular mechanisms of B-cell grth
and maintaining this state have been of great interest to prevent loss of B-cell mass during

the development of Type II diabetes.

 

It has been demonstrated that physiological glucose concentrations induce B-cell
proliferation, which is mediated by protein kinase A (PKA), Caz+/calrnadulin, and PKC
(45, 184). This glucose-induced mitogenesis seems to be a result of a glucose metabolite,

since pyruvate and mitochondrial fuels also promote B-cell proliferation (185).

Recently, Hiigl et al. (186) demonstrated that insulin-like growth factor I (IGF-I)
induces B-cell growth in a glucose-dependent manner. This study showed that INS-1
cells, an insulinoma cell line, incubated in 15 mM glucose and 10 mM IGF-l increase
cell proliferation more than 50 —fold. IGF-l-induced B-cell proliferation was mediated
by the insulin receptor substrate (IRS) family, PI3K, and the 70-kDa S6 kinase (p70 56k).
The activation of IRS proteins by IGF-I or insulin receptors leads to activation of PIBK
and subsequently to activation of p70 56". IGF-l mitogenesis activity does not require the
activation of the Janus kinase-2/Signal transducer and activator of transcription-5
(JAK2/STAT5) signaling pathway, nor activation of INK, or p38 MAP kinases (Cousin,

SP; Hiigl, SR & Rhodes, C, unpublished observations).

Growth hormone (GH) is another potent inducer of glucose-dependent B-cell
proliferation via JAK2/STAT5 with no cross talk with IRS proteins (187). It appears that

PI3K also is required for GH to provide glucose-dependent B-cell proliferation (187).

Interestingly, prolonged exposure of B-cells to high levels of glucose or FFA
induces apoptosis, causing a reduction in B-cell mass (188, 189). The signaling pathway
involved in FFA reduction of IGF-1/ glucose-dependent proliferation is by activation of
protein kinase B (PKB) and chronic activation of atypical PKC zeta (185). Interestingly,

FFA induces insulin resistance through PKB and certain isoforms of PKC (190, 191).

45

Thus, it seems that the same signaling pathways mediate FFA inhibition of B-cell growth
and induction of insulin resistance, two fundamental pathologic characteristics of Type II

diabetes.

Recently, Lingohr et al. (192) demonstrated that prolonged (>24 hrs) activation of
extracellular-regulated kinase-l/2 (Erk-1/2) and PI3K signaling pathways are important
to induce glucose and IGF-l-induced B-cell proliferation. They also demonstrated that
transforming growth factor-a (TGF) and epidermal growth factor (EGF) transiently
activated Erkl/2 (< 20 min) and this activation was not correlated with B-cell
proliferation. Importantly, it can be postulated that hyperglycemia and hyperlipidemia
can induce B-cell proliferation-dependent on prolonged activation of specific signaling
pathways. The prolonged activation of certain pathways not only induces B-cell
proliferation but also might induce other signaling pathways and eventually lead to
inhibition of glucose-induced insulin secretion. Thus, these kinases might also be
involved in reduced insulin secretion and decreased insulin gene transcription during the
hyperglycemia and hyperlipidemia states. Future studies need to be performed to
elucidate the roles of IRS proteins, PKC, and PKB in glucose-induced repression of

insulin gene expression.

8.2 Insulin gene transcription

Insulin is essential to maintain blood glucose levels within a narrow range. In

turn, glucose regulates insulin secretion. This process requires regulation not only of

46

insulin exocytosis but also of insulin translation and transcription (193, 194). It is
commonly believe that in the short-term (< 20 min), glucose regulates insulin at
postranslational and postranscriptional levels, and in the long-term (> 2 hrs), glucose
regulates insulin gene at transcriptional levels (195). It has been recently demonstrated
that glucose also has an immediate regulatory fimction at the transcriptional level (196).
In addition, glucose metabolism is required for regulation of insulin transcription and
insulin promoter activity (52). Recently, studies have demonstrated that insulin increases
insulin mRNA (197, 198) and insulin promoter activity (199, 200). Insulin also induces
an increase in binding activity of de-l (200). Insulin regulates insulin gene transcription
via B-cell insulin receptors and its downstream targets, the IRS-2/PI-3 kinase/p70 36k and
calmodulin kinase signaling pathways (52). p38 kinase has also been shown to regulate
de-l phosphorylation and its translocation from the cytoplasm to the nucleus (201).
However, these results have been challenged by studies in which over-expression of an
upstream p38 kinase and use of a p38 kinase inhibitor did not support the involvement of
this kinase in the regulation of insulin mRNA (202). In contrast, over-expression of PI3K
and use of a PI3K inhibitor suggested that this kinase mediates de-l activation and

translocation (202).

One can postulate that prolonged exposure of B-cells to high levels of glucose can
alter the activation of many of kinases and eventually lead to decreased insulin gene
expression. In the first stages of B-cell dysfunction, increased insulin secretion is
observed, a phenomenon termed “hyperinsulinemia”. The prolonged exposure of B-cells

to high levels of insulin might also alter insulin gene regulation.

47

Glucose-induced insulin secretion is associated with a small stimulation of mitogen-
activated protein (MAP) kinase (203). In addition, glucose causes transcriptional
activation of AP-l family members such as c-fos, jun B, and c-jun (204). The
accumulation of these genes is dependent on cAMP, which by itself has only a small
effect on the stimulation of c-fos, jun B, c-jun genes, and glucagon-like peptide-1 (GLP-
l). The accumulation of c-jun and c-fos occurs at physiological glucose concentrations
(3 to 11 mM) and requires one to two hrs. The authors conclude that the accumulation of
AP-l family members induced by the synergy of glucose and GLP-1 may be involved in
facilitating induction of insulin gene, and may play an important role in the adaptive
process of B-cells to hyperglycemia. It has also been documented that over-expression of
c-Jun represses cAMP-induced activation of human insulin promoter activity (79). If e-
Jun represses insulin promoter activity then glucose-induced accumulation of c-fos and
jun B might prevent insulin gene repression by sequestering c-jun through dimerization

and consequently facilitate glucose/cAMP induction of insulin gene promoter (204).

8.3 Cytokines involved in B-cell cytotoxicity

It is well document that certain cytokines are involved in the destruction of
pancreatic B-cells in Type I diabetes (205, 206). Culturing islets with IL-IB induces
inhibition of glucose-induced insulin secretion, impairs islet oxidation of glucose, and
increases apoptosis (207-210). The cytotoxic effect of IL-10 on B-cells is mediated, in
part, by expression of nitric oxide synthase (NOS) and overexpression of nitric oxide

(N 0) (208). Recently it was demonstrated that IL-10 increases INK activity, and

48

phosphorylation of c-Jun and ATF-2 in RINmSF B-cells, an insulinoma cell line (211,
212). INS-1 cells treated with IL-10 for two days leads to apoptosis (213). Bonny et al.
(213) demonstrated that IL-lB-induced apoptosis is mediated by increased INK
expression and decreased INK-interacting protein-1 (JIP-l) content. A series of
experiments either by over-expressing JlP-l protein or inhibiting JIP-l protein with an
inducible IIP-l antisense RNA led to the conclusion that JIP-l protein is an anti-
apoptotic agent that can protect cells fiom the cytotoxic effects of IL-0 (213). It has also
been demonstrated that p38 MAP kinase and ERKl/2 pathways are involved in IL-IB-
induced N0 in isolated pancreatic islets and RINmSF cells (214). Overall, these
experiments demonstrated that cytokines induce B-cell apoptosis by activating p38 MAP

kinase and INK pathways, which increases the expression of iNOS and production of

N0.

9. IN S-l cell model

IN S-l cells are a rat insulinoma cell line that was derived from cells isolated from
an X-ray-induced rat transpantable insulinoma (215). INS-1’s cell growth is dependent
on Seeding density, and the optimal conditions are 3.5 X 104 cells/ cmz. Cell growth
depends on 2-mercaptoethanol (2-ME, 50 11M). Removal of 2-ME causes a 15-fold drop
in tOtal cellular glutathione levels (216). Thus, 2-ME maintains high levels of total
glutathione, which probably prevents cell damage from oxidative stress since it has been
reportfid that B-cells contain low levels of antioxidants (164). Trypsinization and

rePlatin g are required for continuous propagation.

49

Morphological studies of INS-1 cells show that the cells are very similar to
normal B-cells. INS-l cells contain several granules that exhibit a central dense core and
a peripheral clear halo, although the degree of granularity varies between cells (215).
Immunoflourescence staining demonstrated that most of the cells contain cytosolic
insulin but the staining varies between cells. No glucagon, somatostatin, or pancreatic
polypeptide was detected in INS-l cells, indicating that only pancreatic B-cells are
contained in this cloned (215). INS-l cells contain high levels of insulin (~ 8 — 10 pg per
million cells), which can be maintained for at least two and a half years of continuous
culture (215). Although, INS-1 cells have high amounts of insulin compared to other
insulinomas, it has ~ 4 times less than normal B-cells (40 ug/ 106 cells). INS-l cells
synthesize both insulin I and II peptides. However, glucose does not stimulate the
incorporation of amino acids into newly synthesized hormone, a characteristic that is

observed in normal B-cells. This could indicate that there is abnormal packaging,

storage, or secretion of insulin in INS-1 cells compared to normal B-cells.

INS-1 cells secrete insulin in response to glucose (215). Insulin secretion studies
were analyzed by static incubation of cells clusters in suspension or attached to plastic
CUlture dishes, and by perifusion (215). In static analysis, cells were preincubated in a
KmbS-Ringer-bicarbonate-Hepes-buffer (KRBH) for 30-60 min. Cells were then washed
and incubated for 30 or 60 min in KRBH buffer containing 2.8, 11.2, or 16.7 mM
glucose_ In the perifusion studies, INS-1 cells were detached fiom culture flasks by
t1VI’SiIlization and then spun for 3 hrs in a culture flask containing regular media. Cells
were then transferred into perifusion chambers and perifused with KRBH containing

same Concentrations of glucose as in the static incubations for 45 min. INS-1 cells have a

50

high insulin response at 11.2 mM of glucose compared to no glucose or 2.8 mM glucose,
and a slight but not significant decrease of insulin secretion at 16.7 mM glucose
compared to 11.2 mM glucose. Thus, INS-l cells secrete insulin in response to glucose
during an acute challenge. However, this secretion is much reduced compared to normal
B-cells (217). INS-1 cells are also sensitive to other secretagogous such as arginine,
leucine, and KC], and inhibitory substances including somatostatin, epinephrine, and

diazoxide (an ATP-sensitive K+ channel opener).

INS-l cells also have the capacity to induce membrane depolarization, which is a
prerequisite for glucose-induced insulin secretion (215). Glucose (1 mM) depolarizes
INS-l cell’s membrane and higher glucose concentrations led to an extended
depolarization. The depolarization was correlated with increased cytosolic
Ca2+concentrations when challenge to glucose (215). These results suggest that INS-1
cells are able to metabolize glucose and this leads to membrane depolarization most
likely by closure of ATP-sensitive K+ channel and subsequently an increase in cytosolic

+ . . .
Ca2 and Insulin secretion.

One important player in sensing high levels of blood glucose is the glucose
transporters in B-cells. B-Cells express the high Km (~ 17 mM) Glut2 and to a lesser
degree of Glutl expression (218). INS-1 cells exhibit high levels of the Glut2 transporter
(219). Studies understanding Glut2 regulation by glucose have been performed in INS-1
cells (219). These studies demonstrated that INS-1 cells incubated in 20 mM glucose for
24 hrs increased Glut2 mRNA levels compared to 2 mM glucose. These results are in

agreement with studies performed in HIT-T15 cells incubated in 22.2 versus 11.1 mM

51

glucose for 24 hrs (220), and in rat islets cultured in 11.1 versus 5.5 mM glucose for 24

hrs (221). These results suggest that Glut2 expression and regulation in INS-l cells may

be similar to normal B-cells.

Glucose-induced insulin secretion is markedly enhanced by glucagon-like
peptide-l (GLP-l) (reviewed in Ref. 219, 220). GLP-1 is secreted by the

enteroendocrine L-cells in response to nutrient intake and acts on B-cells as a potent

insulin secretagogue (222, 223). GLP-l also up-regulates other genes involved in B-cell
function such as insulin gene expression and promoter activity (224), and de-l gene
expression (225). Some of the characteristics of GLP-1 in enhancing glucose-induced
insulin secretion have been identified in INS-l cells (226). INS-1 cells treated with
stimulatory GLP-l concentrations induce insulin promoter activity (226). These results

also imply that INS-1 cells express GLP-l receptors as observed normal B-cells.

One of the key components linking glucose metabolism to insulin secretion is the
ATP-sensitive potassium channel (KATp). Upon glucose metabolism, KAT? channels are
closed in response to an increase in ATP/ADP ratio, resulting in membrane
depolarization, which opens voltage-dependent Ca2+ channels leading to a increase in
intracellular Ca2+ levels and subsequent insulin secretion (227). KAT]: channels are
composed of a sulfonylurea receptor-l (SURl), a domain for ATP binding, and a
potassium channel (Kir6.2), which forms the pore of the channel (227). Studies
understanding the regulation of KATP channels have been performed in INS-1 cells (228).
These studies indicate that INS-1 cells expressed the SURl and Kir6.2. In addition

Moritz et al. (228) demonstrated that INS-l cells incubated in high concentrations of

52

glucose for more than 24 hrs lead to downregulation of SUR1 and Kir6.l mRNA levels.
Studies in Zucker diabetic fatty rats (229), in 90 % pancreatectomized rats (4), and in
isolated rat islets exposed to high levels of glucose for 24 hrs (230) also showed
decreased expression of Kir6.2 mRNA levels. These studies Show that IN S-l cells have

some similar characteristics observed in Type 11 diabetic models.

INS-1 cells have also been used as a model to study regulation of essential genes
in the glycolytic pathway by hyperglycemia (231). INS-l cells express
PhOSphofructokinase-l (PFKI), which phosphorylates fructose-6-phosphate to fructose-
1,6-bisphosphate, glyceraldehyde-3-phosphate dehydrogenase (GAPDH), which provides
Cytosolic NADH, and L-pyruvate kinase (L—PK), which catalyzes the formation of ATP.
In agreement with previous in viva and in vitra studies (99, 232), INS-1 cells cultured in
25 HIM glucose for three days led to an increase in insulin secretion at low (2-5 mM)

gIUCOSe and did not respond to higher glucose concentrations (231). The prolonged
incubation of cells led to an increase in gene expression of PFKl, GAPDH, and L-PK
cor“pared to cells cultured in 5 mM glucose (231). In contrast, glucokinase (GK) and 6-
p hosphofructo—Z-kinase transcripts remain unchanged. IN S-l cells incubated for prolong
p eriOds in high levels of glucose also accumulate glycogen, like islet tissues (233) and the
ZDF animal model (7). The authors conclude that the increase of these enzymes and
accumulation of glycogen are involved in the compensatory mechanisms of B-cells
eXDOSed chronically to hyperglycemia. These compensatory mechanisms are a high rate
of inSulin secretion at low glucose concentrations, hypertrophy, and hyperplasia, which

haVe been observed in Type 11 diabetic models (100, 101).

53

Important for our studies, INS-1 cells exhibit other characteristics observed in
Type II diabetes such as decreased insulin gene expression. Studies in INS-l cells have
demonstrated that insulin mRNA and insulin promoter activities are reduced in cells
exposed to high levels of glucose for 24 to 48 hrs (126). Reduced insulin promoter
activity is correlated to reduced de1 binding activity in IN S-l cells (126). These results
have also been shown in viva in 90 % pancreatectomized rat (113) and Zucker diabetic
rat models (101). When INS-1 cells are returned from high glucose levels to low glucose

levels or when Zucker rats blood glucose levels are normalized with phlorizin, insulin

mRN A levels is no longer repressed (113, 126).

INS-1 cells have also been used as a model for hyperlipidemia-induced B-cell

dYSfunction studies. Hyperlipidemia is associated with B-cell dysfirnction because Type
H diabetic patients frequently exhibit high triglycerides and free fatty acid blood levels
(234) and display reduced glucose-induced insulin secretion (235). Studies in INS-1 cells
demonstrated that prolonged exposure of cells to fatty acids markedly represses glucose-

induced insulin secretion (236)-

Pancreatic B-cells are known to be very susceptible to oxidative stress because of
their low expression levels of antioxidant enzymes like superoxide dismutase (SOD),
glutathione peroxidase (GPx), and catalase in comparison to other tissues (161, 234).
RINITISF, a cell line derived from the same islet tumor as INS-l cells (238, 239), express
10‘” levels of the above-mentioned antioxidant enzymes (164). These results suggest that
INS‘ 1 cells might also express low levels of antioxidant enzymes, making the cells

vulnerable to reactive oxygen species. INS-l cells express peroxiredoxins (240), another

54

group of antioxidant enzymes that catalyze reduction of both hydrogen peroxide and
alkyl peroxides to water or corresponding alcohol by using tlrioredoxin as a hydrogen
donor (241, 242). In addition, peroxiredoxins can also protect against nitrogen radicals
(243). Studies in INS-1 cells demonstrated that oxidative stress induces peroxiredoxin
levels (240). INS-cells treated with cytokines including IL-lB, IFN-y, and TNF-or for 48
hrs led to an up-regulation of peroxiredoxins, which has also been observed in isolated
pancreatic islets (240). These results indicate that the induction of peroxiredoxin
expression might be a compensatory response to oxidative and nitrosative stress. Other
studies have used INS-1 cells as a model to investigate whether over-expression of
antioxidant enzymes protects the cells against the cytotoxic effects of nitric oxide donors
(244). Over-expression of Cu/Zn superoxide dismutase protected INS-1 cells against

nitric oxide cytotoxicity.

Overall, INS-1 cells behave in a similar manner as pancreatic B-cells. They
metabolize glucose and secrete insulin through the same mechanism as normal B-cells.
hnportantly for our studies, INS-1 cells secrete insulin at physiological glucose
conCentrations and have similar phenotypic characteristics as Type H diabetic models.
Thus’ INS-1 cells appear to be a good model to investigate mechanisms involved in the

epressron of glucose-Induced Insulin secretion as observed 1n Type II diabetes.

INS-1 cells, however, do not completely mimic all the responses observed in [3-
Cells exposed to hyperglycemia. First, glucose toxicity on B-cells is thought to occur
afiel. a prolong exposure of cells to high glucose concentrations that leads to a progressive

1
033 of B-cell differentiation. Isolated pancreatic islets incubated in high (> 11 mM)

55

glucose levels for 7 days Show reduced glucose-induced insulin secretion, islet content of
insulin, and insulin mRNA levels (245). In contrast, INS-1 cells cultured for 24 to 48 hrs
shows reduced insulin mRNA levels, which are associated with reduced insulin promoter
activity (126). The discrepancy in the behaviors of both isolated islets and INS-lcells is
not well understood. One can speculate that isolated islets contain in addition to B-cells,
or-cells and 5-cells, which can positively affect B-cell function. In addition, INS-l cells
require 2-ME for growth and replication, in contrast, isolated islets are typically cultured
without 2-ME. This could indicate that INS-1 cells might have even lower levels of
antioxidant enzymes than isolated islets. Thus, INS-1 cells may be even more vulnerable
to oxidative stress than isolated islets, thus allowing for shorter periods of time to observe

Phenotypic changes induced by oxidative stress.

56

III. MATERIALS AND METHODS

1. Materials

Cell culture media and Lipofectamine were purchased from Invitrogen
(Gaitherburg, MD). Luciferase (LUC) assay kits were from Promega (Madison, WI).
Anti-JNK] (C17) antibodies, anti-phosphorylated-INK (G-7) antibodies, glutathione S-
transferase (GST)-c-Jun fusion protein and Protein-A agarose were from Santa Cruz
Biotechnology Inc. (Santa Cruz, CA). Nitrocellulose membranes were from BioRad
(Hercules, CA). Chemiluminescence detection kits were from Pierce (Rockford, IL). [7-
32P] ATP and ['4C]chloramphenicol were from NEN Life Science Products (Boston,
MA). Ligase enzyme was from Invitrogen Corp (Carlsbad, CA). Taq DNA polymerase
was from Invitrogen Corporation (Carlsbad, CA). Vent-DNA polymerase was from New
England Biolabs-Inc (Beverly, MA). DNA purification kit, Geneclean-Turbo, was
purchased from BiolOl (Vista, CA). Quick spin columns B-50 sephadex was purchased
from Roche, as well as the inhibitors aprotinin and leupeptin (Indianapolis, DI). The
substrate for CAT assays, butyryl CoA, was from Sigma (St. Louis, MO). pRK, MLK3,
MLK3K144A, and anti-MLK3 antibody were provided by Dr. Kathleen Gallo, Michigan
State University (MSU). The 2XTRE-LUC reporter gene was provided by Dr. Maduker,
MSU. JNKlal and JIP-l expression vectors were provided by Dr. Roger Davis,
University of Massachusetts Medical School and Howard Hughes Medical Institute,
Worcester, MA. pMMTVCAT vector was provided by Dr. Richard Mikcisek, MSU.

PK(-l97)CAT vector was provided by Dr. Howard C. Towle, University of Minnesota,

57

MN. The Z, X, Y, Za, and Zb minienhancer reporter vectors were provided by Dr.

Michael German, University of San Francisco, CA.

2. INS-1 cell culture

INS-1 cells (215) (kindly provided by Dr. C. Wollheim) were routinely cultured in 5%

CO2-95% air at 37 °C in RPMI-1640 media containing 11.1 mM glucose and supplemented with

10% fetal-bovine serum, 1 mM pyruvate, 10 mM Hepes, 50 11M 2-mercaptoethanol, 100 units

penicillin/ ml and 100 11g streptomycin/ ml. Cells were passed weekly by trypsin-EDTA

detachment. All experiments were performed on INS—1 cells between passages 70 and 85 .

3. Plasmid DNA constructs

The plasmid INSCAT contains the chloromplenicol acetyltransferase (CAT) gene
under transcriptional regulation by the human insulin gene sequences —327 to +30 as
previously described (121). The —230 INSCAT (INS(-230)CAT) plasmid contains the
CAT gene under transcriptional regulation by the human insulin gene sequences -230 to
+30. The mClINS(-230)CAT vector, which contains mutations in the Cl element, the
mA1A3 INS(—230)CAT, which contains mutations in both A1 and A3 elements, the
mAlA3C1 INS(-230)CAT, which contains all three mutated elements, were generated by
PCR amplification using INS(-230)CAT, mAlINS(—230)CAT, and mA3AllNS(-
230)CAT as a template, respectively. Oligonucleotide primers used to generate these
specific mutant elements are listed in Table 1. mAlINS(-230)CAT was previously made
by Dr. Olson where the A1 element was mutated from CCCTAATGGG to

CCGCGCGCGG. The mutations in the A3 element were TAAT to TQT and the

58

mutations in the C1 element were GC_C_T_ CA to GCflfiCA. Truncated insulin promoter
vectors were also generated by PCR amplification. The template used in the PCR
reactions was INS(-327)CAT and the respective primers are shown in Table 2. INS(-
250)CAT was previously made by Dr. Olson. The mAS/Core/Pal/E3 INSCAT vector
was generated by first making mPalINSCAT vector using INS(-327)CAT as a template.
Then, mPalINSCAT was used as a template to generate mPal/E3INSCAT, which was
finally used as a template to produce mA5/Core/pal/E3 INSCAT vector. Primers used to
generate the mA5/Core/pal/E3 INSCAT vector are shown in Table 3. The plasmid
2XTRE-LUC contains the luciferase gene under transcriptional regulation of the prolactin
promoter from —36 to +37 and two copies of the alpha collagenase AP-l promoter
enhancer (246). pCDN3-JNK10L1 and pCMVS-flag-IIPI expression plasmid have been
previously described by Derijard et al. (247) and Dickens et al. (248), respectively. pRK-
MLK3 and pRK-MLK3K144A expression plasmids have been previously described by
Gallo et al. (249). MLK3 was also cloned into pCR3.1 vector because transfections
containing the pRK vector significantly reduced basal INSCAT activity. To clone MLK3
cDNA into pCR3.l vector, MLK3 was amplified by PCR from the pRK-MLK3 vector,
and then ligated into pBluescript 11 KS vector. MLK3 cDNA was then cut out of the
pBluescript 11 KS vector and ligated into the pCR3.l vector. The multimer
(E1)3RSV103CAT plasmid was made by ligating three hybridized E1 element sequences,
which contain BamHl restriction site at both 5’and 3’ends (5’ GAT CCG GGG TCG
GCA GAT GGC TGG GGG CG 3’) upstream to the minimal promoter (only the TATA
box) of Rous sarcoma virus driving CAT expression. All vectors were verified by

sequencing on an ABI Prism 3700 DNA Analyzer.

59

4. PCR reactions

In general, all mutated and truncated insulin promoter vectors were generated by
PCR using primers containing Xba I or Xho I restriction sites. The insert was then
ligated into poCAT vector that contained these sites in the multilinker region.

In general, PCR reactions contained 2 ng/ul template, 20 pmol of appropriate
oligonucleotide primers, 10 mM dATP, 10 mM dGTP, 10 mM dTTP, 10 mM dCTP, and

buffer and polymerase according to manufacturer’s instructions (Vent polymerase was

 

used for mutated INS(-230)CAT vectors, and T aq polymerase was used for truncated
INSCAT vectors and mA5/Core/Pal/E3 INSCAT vector). The thermal cycle profile
employed a 1 min of denaturation at 94 0C followed by 20 to 25 amplification cycles (1
min of denaturation at 94 °C, 30 sec annealing at 58 OC, and 30 sec extension at 68 °C)
and a extension step for 4 min at 68 °C. The mA1C1A3 INSCAT and mA5/Core/Pal/E3
INSCAT vectors were generated by PCR amplification using a “PCR-bridge reaction”.
The “PCR-bridge reaction” consisted of two PCR products that both contained the same
mutated element, and both were used as templates to generate a PCR product with the
mutated element. One PCR product contained sequences from 5’ end, which had a XbaI
restriction site, to the mutated element. The other PCR product contained sequences
from the mutated element to the 3’ end, which had a XhoI restriction site. The PCR
“bridge reaction” contained 5 111 to 10 pl of each PCR template, 20 pmol of the
appropriate primer that contained the Xba I or Xho I restriction site, and the additional
cOrnponents of the PCR reaction as described above. The mutated insulin promoter

cDNA inserts were then digested with respective restriction enzymes. Following

60

digestion and verification of sizes by 1% agarose gel, the cDNA inserts were ligated into
linearized poCAT vector that had Xba I and X1101 sites.

To generate pCR3.l-MLK3, MLK3 was amplified by PCR from 20 ng pRK-
MLK3 vector by using 20 pmol primer that contained the replication start sequence and a
BamHI restriction site, and another 20 pmol primer that contained the replication stop
sequence and a EcoRI restriction site (see Table l for sequences). The PCR reaction also
contained 1 pl of DMSO to disrupt disulfide bonds, 10 mM of each dNTP, 5 pl buffer,
and 1 pl of pfu DNA polymerase. The thermal cycle profile employed 1 nrin of
denaturation at 94 °C followed by 25 amplification cycles (1 min of denaturation at 94
°C, 1 min of annealing at 69 °C, and 6 min of extension at 72 °C) and an extension step of
10 rrrin at 72 °C. The MLK3 cDNA insert was digested with EcoRI and BamHI

restriction sites and ligated into linearized pBluescript 11 KS vector.

5. Ligation reactions and purification of DNAs

The general reaction had a volume of 15 pl containing 0.1 pmol of digested
vector, 0.3 pmol, 0.6 pmol, or 0.9 pmol of cDNA insert, and buffer and ligase according
to manufacturer’s instructions. Reaction was then incubated overnight at 16 °C. cDNA
Vectors were amplified by competent E. cali DH 5 a. Vectors were then purified by using

Qiagen kit and vectors were further purified by cesium chloride method.

6- Transfections
For all reporter gene studies, INS-1 cells were subcultured for two days before

transfection at a density of 1.5 x 106 cells per well (diameter 3.5 cm) in RPMI-1640

61

media supplemented as described above. Cells were transfected for 5 hrs according to
manufacture protocols using a ratio of 1 pg plasmid to 2 pl Lipofectamine and were then
incubated in RPMI-l640 media containing 4.0 or 16.7 mM glucose (as indicated in figure
legends). In experiments designed to test the effect of JNKlorl or MLK3 on promoter
activity, 1 pg of reporter plasmid was used and increasing concentrations of expression
plasmid (as indicated in figure legends). All cells were transfected with equal amounts
of DNA by use of control expression plasmids containing no inserts. Cells were
harvested either 24 or 48 hrs after transfection (as indicated in figure legends) and CAT

or luciferase activity was assayed.

7. CAT assays.

After treatment, cells were washed twice with 1X PBS. CAT assays were
performed as follows. In general, 1 ml of TEN (40 mM Tris, 150 mM and 1 mM EDTA)
was added to cells for 5 min on ice. Cells were then scraped and transfer to Eppendorf
tube, then spun at 5000 rpm for 30 sec. TEN was aspirated and 100 pl of 250 mM Tris
pH 7.5 was added. Cells were then lysed by three cycles of freezing and thawing.
Broken cells were spun down at 14,000 rpm for 5 min at 4 °C. Lysates were aliquoted
into new Eppendorf tubes and stored at -20 °C.

In general, CAT assays were performed by combining the appropriate amount of

extracts and 250 mM Tris to a 50 pl total volume. Then 100 pl of CAT reaction ( 250
mM Tris, 2 pl /rxn of 1"C-Chloromphenicol (54 mCi/mmol, 0.05 mCi/ml), and 3 pl of
O- 01 M Butyl CoA) was added. The reaction was incubated usually for 2 hrs at 37 °C.

Amount of extracts and incubation times were determined in such a way that the CAT

62

activity was in the linear range of the reaction. Extraction of labeled butyl CoA was
performed as follows: first, 300 pl xylene was added, then vortexed for 30 sec, and spun
for 3 min at 14000 rpm. Second, xylene (upper layer) was removed and added to an
Eppendorf tube that contained 300 pl of 250 mM Tris, then vortexed for 30 sec, and Spun
for 3 min at same speed as before. Then 150 pl of the xylene (upper layer) was added to
a scintillation counter vial containing 5 ml of scintillation cocktail. CAT units were

normalized to protein concentrations. All experiments were carried out in duplicate.

8. Luciferase assays

After treatment, cells were washed twice with 1X PBS. Luciferase activity was
measured according to the protocol from Promega. In general, 200 pl of luciferase lysis
buffer was added per well and incubated for 15 min at room temperature. Cells were
then scraped and frozen at —20 °C. Cells were then spun down and supernatant was
transferred to a new Eppendorf tube. 20 pl of supernatant was assayed with 100 pl
luciferase reaction buffer. A Turner TD 20E luminometer (Turner Designs) was used to

measure luciferase activity. All experiments were carried out in duplicate.

9. JN K kinase assay
INS-1 cells were subcultured for 2 days at a density of 5 x 106 cells per plate
(diameter 6.0 cm) in RPMI-1640 media supplemented as described above. In
experiments designed to test the effect of MLK3 on INK activity, cells were transfected
With 3 pg expression plasmid and 6 pl Lipofectamine. Cells were then incubated for 24

hrs in RPMI-1640 media containing either 4.0 mM or 16.7 mM glucose and the

63

supplements as described above. Cells were then lysed for 5 min on ice in 1 ml lysis
buffer (50 mM HEPES pH 7.5, 150 mM NaCl, 1.5 mM MgCl2, 2 mM EGTA, 1% Triton
X-100, 10% glycerol, 10 mM NaF, 1 mM Na4PPi , 100 pM B-glycerophosphate, 1 mM
Na3VO4, 2 mM PMSF, and 0.15 U/ml aprotinin). Lysates were centrifuged at 4 °C for 20
min at 4,000 rpm in an Eppendorf centrifuge. Lysates (200 pg protein) were incubated at
4° C for 90 min with 1 pg of anti-INK-l antibody bound to 20 pl of Protein -A-agarose.
JNK was then precipitated by centrifugation and washed three times with HNTG buffer
(20 mM HEPES pH 7.5, 150 mM NaCl, 0.1% Triton-X-lOO, 10% glycerol) containing 1
M LiCl, three times with HNTG buffer, and twice with kinase reaction buffer (50 mM
Tris-HCl pH 7.5, 100 mM NaCl, 1 mM MnCl2, 0.1 mM Na3V04 ). Irnmunoprecipitates
were resuspended in 20 p1 of kinase reaction buffer containing 5 pCi per reaction [y-32P]-
ATP (3000 Ci/mmol), 50 pM ATP and 1 pg GST-c-Jun protein and then incubated at
room temperature for 30 min. JNK enzymatic reactions were terminated by the addition
of an equal volume of 2 x SDS sample buffer (100 mM Tris pH 6.8, 4% SDS, 20%
glycerol, 0.2% bromophenol blue, 100 mM DTT, 1% B-mercatoethanol, 50 mM EDTA).
Proteins were then separated by 10% SDS-PAGE, transferred to nitrocellulose
membranes, and quantitated on a Phospholmager (Molecular Dynamics). All
nitrocellulose membranes were then immunoblotted with a 1:1000 dilution of anti-INK

antibodies to ensure that equal levels of INK were immunoprecipitated and transferred.

1 0. Nuclear extracts
Nuclear extracts were made from INS-1 cells according to the method of

Schreiber et al. (250). In general, cells were washed with 10 ml cold PBS. Cells were

64

then scraped and pelleted at 1500 g for 5 min in a 15 ml tube. Pellets were resuspended
in 1 ml Tris-buffered saline, transferred to Eppendorf tube and pelleted for 15 sec at 4 °C.
Pellets were resuspended in 800 pl buffer A (10 mM Hepes pH 7.9, 10 mM KCl, 0.1 mM
EDTA, 0.1 mM EGTA, 1 mM PMSF, 0.2 pg/ pl leupeptin, 0.2 pg/ pl aproptinin) by gently
pipeting up and down. Cells were allowed to swell on ice for 15 min. Then, 50 pl of 10
% NP40 was added and samples were vortexed for 10 sec. Lysates were spun for 30 sec
at 14000 rpm at 4 °C. The supematants were removed and the nuclear pellets were
resuspended in 100 pl buffer B (20 mM Hepes pH 7.9, 400 mM NaCl, 1 mM EDTA, 1
mM EGTA, 1 mM PMSF, 1 mM DTT, 0.2 pg/pl leupeptin, 0.2 pg/pl aproptinin).
Nuclear pellets were rocked vigorously for 15 min on ice. Nuclear debris was spun down

for 5 min at 4 °C, and finally nuclear extracts were aliquoted and stored at -70 0C.

1 1 - Ph osphorylated JNK analysis
Proteins (30 pg) were resolved on 10% SDS-PAGE, transferred to nitrocellulose

membranes, and phosphorylated-INK was detected using a 1:1000 dilution of an anti-

phosphorylated-INK antibody. Protein levels were visualized by chemiluminescence.

l 2- Electrophoretic mobility-shift assays

Nuclear extracts were prepared as described above. Oligonucleotides were

hybl‘idized as follow: 1.2 pM Oligonucleotides, 5.0 pl React 2, and 20 pl water were
heated for 10 min at 68 — 72 °C, and then cooled slowly at room temperature.
0 -

llgonucleotides were either 32p-5’-end-labeled or 32p-3’-end-labeled-incorporation. The

r eact' 32 . - - - 32
Ion to generate p-S -end-labeled Oligonucleotides contained 5 pl [y- p]ATP (3000

65

Ci/mmol), 1 pl T4 polynucleotide kinase, and 0.1 pl hybridized oligonucleotide in a 20 pl

total volume. The reaction to generate 32p-3’-end-incorporation-labeled oligonucleotides

contained 2.0 pl [or-32p]dCTP (3000 Ci/mmol), 3 pl 10 mM dNTP (except dCTP), 0.5 pl
hybridized oligonucleotide, and 1.0 pl Klenow Fragment of E. cali polymerase I in a 50
pl reaction volume. Reactions were incubated for 30 to 45 min at room temperature for
Klenow Fragment or at 37 °C for T4 kinase. For the reaction with Klenow Fragment, 1
p1 dCTP was added to complete the filling reaction and incubated for 10 min at room

temperature. The reactions were then incubated at 68 0C for 10 min and cooled slowly to

room temperature. Unincorporated nucleotides were removed by using a Quick spin

column according to manufacture protocol. 1 pl of probe was counted by using a

Scintillation counter. Labeled oligonucleotides were mixed with nuclear extracts in a
reaction that contained 10 pg nuclear extract protein, labeled oligonucleotides (30,000
0pm), 3.5 pl of 10 X binding buffer (50 % glycerol, 10 mM EDTA, 500 mM NaCl, 100
HIM Tris pH 7.5, 10 mM DTT), 2 pg/pl polydIdC in a 25 pl total volume. The reaction
Was incubated for 30 min at room temperature. For super-shift assays, 1 pl or 2 pl
atltibody was added to labeled oligonucleotide before addition of proteins and binding
buffer, and was incubated for 20 min on ice. After incubation, the binding reactions were
SeIDarated by a 5 % nondenaturing polyacrylamide gels in 1X TGE buffer (40 mM Tris,
3 84 mM glycine, 2 mM EDTA, pH 8.3). Gel was dried and nuclear protein binding was

Vi Snalized and quantitated by Phospholmaging (Molecular Dynamics).

66

 

Name 5’end Sequence

 

-230INS/ 5 ’Xbal -230 GCGTCTAGACCCCTGGTTAAGACT

-2301NS/ 3’ X1101 +30 GCGCTCGAGCTCTTCTGATGCAGC
CTGTC

mA3/Xba 1/ forward -230 GCGTCTAGACCCCTGGTTAAGACT
CTC_CTGACCCGCTGG

mCl/Pst 1/ forward -126 CCGGAAATTGCAGCIQCAGCCCCC
AGCCATCTG

rnCl/Pst I/ reversed ~96 CAGATGGCTGGGGGQGCAGCTGC
AATTTCCGG

 

Table 1. Sequences of oligonucleotide primers. This table shows the primers used to
generate the mA3A1(-230)1NSCAT, mCl(-230)INSCAT, and mA3C1A1(-230)INSCAT
Vectors. —2301NS/ 5’ Xba I is the wild type 5’ primer containing a Xba I restriction site.
~2301NS/ 3’ Xho I is the wild type 3’ primer containing a Xho I restriction site.
mAB/Xba I is the forward primer that contains mutations in the A3 element. mCl/Pst I is

the primer that contains mutations in the C1 element and a Pst I restriction site.

67

 

Name S’end Sequence

INS(-292)CAT -292 GCCTCTAGAGGCTTTGCTCTCCTGGAGACA
INS(-270)CAT -279 GCCTCTAGATGGAGACATTTGCCCCCAGCT
1NS(-261)CAT -26l GCCTCTAGAGCTGTGAGCAGGGACAGGTCT
INS(-250)CAT -250 GCCTCTAGAGGACAGGTCTGGCCACCGGGC

Table 2. Sequences of oligonucleotide primers used to generate truncated insulin
promoter vectors. Each primer contains six extra nucleotides upstream of the 5’ end

with a Xba I restriction site.

 

68

 

iName 5 ’end Sequence

 

Za -292 GGCTTTGCTCTCCTGGAGACATTTGCCCCCA
Zam-271 -292 GGCTTTGCTCTCCTGGAGACAQTTGCCCCCA
Zam-273 -292 GGCTTTGCTCTCCTGGAGAQATTTGCCCCCA
Zam-282/ 3 -292 GGCTTTGCT_A_G_CCTGGAGACATTTGCCCCCA
Zd -289 TTTGCTCTCCTGGAGACATTTGCCCCCAGCTG
TGAGC
Palindrome -289 TTTGCTCTCCTGGAGACATTT
mPalindrome -2 89 TTTGCTAQCCTGGAGACATTT
E3 -278 GGAGACATTTGCCCCCAGCTGTGAGC
mE3 -278 GGAGAGQTTGCCCCCAGLGITGAGC
A5/Core/E4 -327 TCTCCTGGTCTAATGTGGAAAGTGGCCCAGG
TGAGGGCT
A5/ Core -323 CTGGTCTAATGTGGAAAGTG
mAS/Core -323 CTGGTCTAC_GITGGAAAGTG
mAS/Core/Xba I -323 GCCTCTAGACTGGTCTAQQITGGAAAGTG
E4 -306 GTGGCCCAGGTGAGGGCT

 

 

Table 3. Oligonucleotide sequences used for mobility-shift assays and used to
generate mutations in the A5/Core, palindrome, and E3 elements. mAS/Core/Xba I,
I'l’lparlindrome, and mE3 oligonucleotides contain mutations in the respective elements,
Which are underlined. These oligonucleotide primers were used to generate
mAS/Core/Pal/E3 INS(-327)CAT vector. All the oligonucleotides shown except

I"IlAIS/Core/Xba I were used for mobility-shift assays.

69

IV. INCREASED C-JU N N-TERMINAL KINASE (JN K) ACTIVITY IN INS-1
CELLS EXPOSED TO ELEVATED GLUCOSE CONCENTRATIONS MAY

MEDIATE THE REPRESSION OF INSULIN PROMOTER ACTIVITY

1 - Abstract
Chronic exposure of pancreatic islets and B-cell lines to elevated glucose

concentrations causes D-cell dysfunction including reduced insulin gene transcription.

Studies using antioxidants indicate that elevated glucose levels reduce insulin gene

transcription by mechanisms involving reactive oxygen species (ROS), thus suggesting a

I‘OIe for stress-activated signaling pathways. This study was designed to investigate the

possible involvement of c-Jun N-tenninal kinase (JNK), a stress-activated kinase, in
mediating the reduction of insulin gene transcription observed in B-cells incubated in
elevated glucose concentrations. We have reported that exposure of INS-l cells to 16.7
mM glucose for 24 to 48 hrs leads to a 70 % reduction in insulin promoter activity. This
Stlldy demonstrated that incubation of INS-1 cells in 16.7 mM glucose for 24 hrs
Sigl'lificantly increased INK activity compared to cells incubated in 4.0 mM glucose.
Treatment of cells with 16.7 mM glucose for 48 hrs also increased AP-l reporter gene
e)‘KIDTession, indicating that glucose-induced activation of INK functionally activates
dQWnstream signaling components such as AP-l transcription factors. Transient
trEtnsfection of cells with increasing concentrations of a JNK] expression plasmid
recluced insulin promoter activity in cells cultured in 4.0 mM glucose and this repression
was more pronounced in cells cultured in 16.7 mM glucose. Over-expression of MLK3,

an Upstream activator of JNK, led to an increase in JNK activity in cells incubated in both

70

4.0 and 16.7 mM glucose. Over-expression of MLK3 also led to an increase in AP-l
reporter gene expression and a reduction in insulin promoter activity in cells cultured in
both 4.0 and 16.7 mM glucose. These data show that activation of JNK, either by high
levels of glucose or upstream activators, leads to functional activation of downstream
JNK target proteins and causes reduced insulin promoter activity. Truncation and
mutational analysis indicated that JNK might reduce insulin promoter activity through a
generalized target such as a co-activator. In conclusion, increased JNK activity in INS-1

cells exposed to high glucose levels may mediate glucose-induced repression of insulin

promoter activity.

71

2. Introduction

Type II diabetes is characterized by insulin resistance and failure of pancreatic [3-
cells to secrete sufficient levels of insulin necessary to overcome hyperglycemia. Once
diabetes is established, chronic hyperglycemia has been postulated to fiirther damage B-
cells, thus exacerbating the disease state. Adverse effects of hyperglycemia on B-cells
include suppression of insulin gene expression and glucose-induced insulin secretion (2-
4, 100, 101, 109, 113, 121, 126). Reduction in insulin gene expression has been
correlated with decreased insulin gene promoter activity and reduced binding activity of
two important transcription factors, de-l and C1 activator (2-4, 121). Studies in INS-1
Cells have demonstrated that insulin promoter activity is markedly repressed in cells
Cultured in 16.7 mM glucose for 48 hrs, and this coincides with a reduction in de-l and

C 1 activator binding (126).
Although some mechanisms involved in glucose-induced repression of insulin
gene expression have been described, e.g. regulation of key transcription factors, the
causative pathway remains unknown. Three possible mechanisms have been implicated
in glucose-induced tissue damage in a diabetic state. First, glucose has been shown to
1T10dify proteins through non-enzymatic glycosylation, which generates reactive oxygen
sI>ecies (ROS) and advanced glycosylated end products (AGES) (251). Autoxidative
g1)v'cation has also been observed in diabetes, which further increases oxidative stress
State (156). Glycated proteins are present in normal tissues but some tissues are more
Setlsitive to this phenomenon. For example, increased glycosylation in tissues such as
kidney, liver, brain, and lung are observed in diabetic conditions (159, 160). Second,

hYperglycemia can increase protein kinase C activity, which has been associated with

72

diabetic vascular complications in the retina (252), aorta (253), and renal glomeruli (254).
Third, glucose can be converted to sorbitol by aldose reductase through the polyol
pathway (152). Aldose reductase uses NADPH as a co-factor and an increased activity of
this enzyme can deplete intracellular NADPH levels, which is detrimental to cell function
(153). For example, decreased NADPH levels inhibit the ability of reduced glutathione
to eliminate reactive oxygen species (153). In addition, because sorbitol is not readily
exported and has a low rate of conversion to fructose, sorbitol accumulates within the cell
creating osmotic stress, and leading to swelling and rupture of the lens fiber cells (25 5).
Many investigators have studied the hypothesis that reactive oxygen species are
involved in the toxic effects of glucose on B—cell function (7, 157, 169, 170). B-Cells are
very vulnerable to ROS since they have low levels of glutathione, superoxide dismutase,
and catalase (164). Some of the mechanisms by which ROS might induce B-cell
dysfunction include DNA modifications such as 8-hydroxy-2’deoxyguanosine (8-OHdG)
(256), and membrane lipid peroxidation, which forms aldehydes such as 4-hydroxy2-
neonal (HNE) (257). 8-OHdG and HNE are oxidative markers and 8-OHdG has been
identified in the urine and blood of patients with Type II diabetes (165, 166). Ihara et al.
(157) demonstrated that B-cells of Goto-Kakizaki (GK), a Type II diabetic rat model,
have high levels of 8-OHdG. In addition, afier GK rats were fed with a diet high in
sucrose, increased oxidative markers were observed, suggesting that chronic
hyperglycemia might be involved in the oxidative stress-related dysfunction of GK rats
B-cells. ROS have also been shown to inhibit glucose-induced increases in mitochondria
and plasma membrane potential, which are required for metabolic signaling for insulin

secretion (172, 258). Changes in B-cell function induced by hyperglycemia, reduced

73

sugars, or H202 has been shown to be partially or fully prevented by antioxidants such as
N-Acetyl-L-cyteine (NAC) or aminoguanidine (AG) (7, 169, 170). Overall, these
experimental results address the important role of oxidative stress in mediating adverse
changes in B-cell function observed in the diabetic state.

Stress-activated protein kinases (SAPK) are a group of serine/threonine specific
kinases that are activated by dual phosphorylation on threonine and tyrosine residues in
response to a variety of extracellular stress factors such as UV, ROS, osmolarity, heat
shock, and treatment of cells with cytokines. Two main groups of SAPK, c-Jun N-
terminal kinase (JNK) (247, 259, 260) and p38 MAP kinase (261, 262) regulate cell
proliferation, differentiation, and apoptosis. There are 12 isoforms of JNK encoded by
three genes (JNK-1, 2 and 3) (263). Targets of JNKs are mostly transcription factors,
including c-Jun, activating transcription factor-2 (ATF-2), and ETS-containing factors
such as Elkl. JNK has natural regulators named IBl/JNK-interacting proteins (JIP),
which are scaffolding proteins that interact with upstream components of JNK signaling
pathway. There are three isoforms of HP (JIP1-3). HP] and JIP2 are closely related
proteins mainly expressed in pancreatic B-cells and in neurons (reviewed in 260).
Although JIPl is now understood to mediate JNK activation (264), it was originally
identified as a cytosolic anchor protein that binds specifically to JNK (248). Over-
expression of JIPl in insulin-producing cells prevented JNK-mediated activation of
transcription factors such as c-Jun, ATF2, and Elkl (248, 265). Interestingly, missense
mutations in the JIPI gene have been correlated with Type II diabetes (266). Studies by

Bonny et al. (213) demonstrated that HP] is involved in B-cell apoptosis. These data

suggest that the JNK signaling pathway maybe involved in B-cell pathology. J IP3 protein

74

is
me
res

3C1

3.]

3.1

is structurally unrelated to J [Pl and J IP2 and is ubiquitously expressed. The other SAPK,
p38 MAP kinase, has four different isoforms and shares common targets with JNK such
as ATF-2, as well as specific targets such as heat shock protein 25 (Hsp 25) (263).

The formation of ROS under hyperglycemic conditions suggests a possible role
for SAPK in mediating B-cell dysfunction. In agreement with this hypothesis, Kaneto et
al. (8) recently demonstrated that ROS suppresses insulin gene expression and secretion
in isolated rat islets. The repression of insulin gene expression was mediated through
JNK. Kaneto et al. (8) also demonstrated that adenovirus-mediated JNK expression in
isolated rat islets represses insulin mRNA. In contrast, adenovirus-mediated dominant-
negative JNK expression prevents ROS-induced repression of insulin mRNA. These
results suggest that high levels of glucose might mediate B-cell dysfunction through JNK
activity.

Herein, we demonstrated that high levels of glucose increased JNK activity and

this may mediate glucose-induced repression of insulin promoter activity in B-cells.

3. Results
3.1 Exposure of INS-1 cells to elevated glucose concentrations increases JNK activity
Recent studies report that antioxidants can partially prevent B-cell dysfunction
associated with chronic exposure to high glucose concentrations (6, 7, 158, 170). These
studies raise the possibility that stress-activated protein kinases such as JNK may mediate
B-cell dysfunction caused by high glucose concentrations. To determine whether
BXposure of INS-1 cells to elevated glucose affects INK activity, INS-l cells were

incubated in 4.0 mM or 16.7 mM glucose for 24 hrs and JNK activity was then assayed.

75

Incubation of INS-1 cells in 16.7 mM glucose led to a 4.5 i 0.4-fold (n=4) increase in

JNK activity compared to cells incubated in 4.0 mM glucose (Fig. 4, Panel A). Western
blot analysis for JNK demonstrated that glucose did not increase JNK activity by
elevating INK protein levels. As observed with JNK activity, incubation of INS-1 cells
in 16.7 mM glucose led to increased phosphorylation of JNK protein levels in nuclear
extracts compared to cells grown in 4.0 mM glucose (Fig. 4, Panel C). These results
indicate that incubation of INS-1 cells in an elevated glucose concentration caused a

significant increase in JNK activity.

3.2 Exposure of INS-l cells to elevated glucose concentrations increase AP-l
transcription factor activity
In many cells types, activation of JNK is associated with activation of AP-l
transcription factors. To test whether the increased JNK activity observed in IN S-l cells
in response to elevated glucose leads to functional changes in AP-l activity, INS-1 cells
were transfected with a luciferase reporter gene regulated by two consensus TREs (AP-1
consensus binding sites) inserted upstream of the rat prolactin minimal promoter
(2XTRE-LUC) (246). Incubation of INS-1 cells in 16.7 mM glucose for 48 hrs led to a
3.6 i 0.6 fold (n=3) increase in 2XTRE-LUC expression compared to cells treated with
4.0 mM glucose (Fig. 5). In contrast, incubation of INS-1 cells in 12.7 mM mannitol plus
4.0 mM glucose did not increase the expression of 2XTRE-LUC (data not shown). These
data suggest that the mechanism whereby 16.7 mM glucose increased 2XTRE-LUC
CXpression was not due to osmotic stress. Moreover, these data indicate that glucose-

induced activation of JNK firnctionally activates downstream signaling

76

 

ea

 

A
Glucose (mM) 4.0 16.7

 

 

C

6' *
5
Kinase activity ' * P'C'JUN S 4
Western blot " i A 'JNK 8 3
22
£1
0

Western blot aim “ P'JNK 16.
. mM glucose

 

 

 

 

Figure 4. Incubation of INS-l cells in high glucose concentration increases JNK
activity. INS-1 cells were incubated in 4.0 or 16.7 mM glucose for 24 hrs, then JNK
activity and phosphorylation were measured. Panel A (upper top) shows a representative
JN K kinase assay performed as described in the Materials and Methods. Panel A (lower
part) shows JNK protein levels in the cytosolic extracts used for the JN K assay depicted
above. Panel B, shows the mean t SE of JNK activity in four independent experiments
(n=4). Asterisk indicates that JNK activity was significantly different between 4.0 mM
and 16.7 mM glucose (p < 0.001 ). Panel C, shows a Western blot of phosphorylated JNK

in nuclear extracts derived from INS-1 cells.

77

 

 

 

 

 

 

 

 

 

 

5
a:
4 4 I
E
e 3 .
U
5
1:
.E
w:
:2
1 q —
0 t .
4.0 16.7
Glucose (mM)

Figure 5. INS-1 cells cultured in high levels of glucose have increased AP-l
transcription factor activity. INS-1 cells were transfected with 2XTRE-LUC
expression plasmid and incubated for 48 hrs in 4.0 or 16.7 mM glucose. Luciferase

activity was measured according to the Materials and Methods. The data shown are the
mean i SE of three independent experiments done in duplicates (n=3). Asterisk indicates
that AP-l activity was significantly different between 4.0 and 16.7 mM glucose (p <

0.03).

78

components such as AP-l transcription factors.

3.3 Expression of JN K in INS-1 cells represses insulin promoter activity

Incubation of INS-1 cells in high glucose concentrations causes a pronounced
decreased in insulin mRNA and insulin promoter activity (126). The mechanism by
which high glucose represses insulin promoter activity has not been fully determined and
may involve glucose-induced activation of JNK. To test whether JNK activation is
sufficient to repress insulin promoter activity, INS-1 cells were co-transfected with a
human insulin promoter-regulated CAT reporter gene (INSCAT) and increasing
concentrations of JNK] expression plasmid (0, 0.1, 0.25, or 0.5 ug). As previously
reported, incubation of INS-1 cells in 16.7 mM glucose caused a 74.4 % reduction in
INSCAT expression compared to cells incubated in 4.0 mM glucose (Fig. 6A).
Expression of JNK] for 48 hrs reduced insulin promoter activity up to 78.4 % (n=6) in
cells cultured in 4.0 mM glucose (Fig. 6A). The ability of JNK to reduce INSCAT
expression was also observed when cells were incubated in 16.7 mM glucose
concentrations (Fig. 6A), suggesting that elevated glucose concentrations can activate the
over-expressed JNK leading to further suppression of the insulin promoter. To analyze
whether JNK-induced repression of insulin promoter activity was specific, INS-l cells
were transiently co-transfected with different reporter vectors controlled by promoters
from the Rous Sarcoma Virus (RSVlOl), Mouse Mammary Tumor Virus (MMTV) or
liver pyruvate kinase (PK(-197)) and JNK (0.1 ug) expression vector. Expression of JNK

did not significantly repress RSV101CAT, MMTVCAT or PK(-l97)CAT expression

79

Figure 6. Over-expression of JN K represses insulin promoter activity. A, cells were
co-transfected with INSCAT and pCMV (control vector) or increasing concentrations of
JNK expression plasmid (as shown in the figure). Afier transfection, cells were
incubated for 8 hrs in 4.0 mM glucose to allow for JNK expression. Cells were then
incubated in 4.0 or 16.7 mM glucose for 48 hrs. Cells were then harvested and CAT
activity was measured. Values are the mean i SE of three individual experiments done in
duplicates (n=6). Asterisk indicates that over-expressed JNK values are significantly
different compared to control for the given glucose concentration (p < 0.05). B, INS-1
cells were co-transfected with RSVlOlCAT vector, pMMTVCAT vector or PK(-
197)CAT vector and pCMV or JNKl (0.1 ug) expression plasmid. Cells were then
cultured for 48 hrs in 4.0 mM glucose. Values represent mean t SE of two independent

experiments done in duplicates (n=4).

80

cpmlug protein/hr

cpm/ug protein/hr

180

 

160 .
14o «
120 4
100 J
80 ~
60 .
401

204

 

 

2500

- 0.50 ug pCMV
12:] 0.10 ug JNK]
- 0.25 11ng
I: 0.50 ug JNK]

 

 

 

 

 

Glucose (mM)

 

2250 e

2000*

1750 ‘

1500 4

\
\\

150 ‘
100 ‘
50W

 

 

 

 

 

 

RSVIOICAT

—pCMV
CZIJNKI

\\
iv

 

 

 

 

 

pMMTV PK(-l97)CAT
Vector

81

(Fig. 6B) and indicates that repression of the insulin promoter by JNK is not through a
global, non-specific mechanism. Overall, these results support the involvement of JNK
in the repression of insulin promoter activity in INS-1 cells expose to high glucose

concentrations.

3.4 Expression of MLK3 in INS-l cells increases both basal and glucose-stimulated
JNK activity, and AP-l transcription factor activity.

Activation of JNK by high glucose may be associated with decreased insulin promoter
activity. If this is true, then direct upstream activators of JNK should also have the
capacity to suppress insulin promoter activity. To test this hypothesis, INS-1 cells were
transfected with an upstream activator of JNK termed MLK3. MLK3 is a
serine/threonine kinase that belongs to the mixed—lineage protein kinase family (249).
MLK3 has been shown to activate both MKK4 (267) and MKK7 (268), which can
directly phosphorylate and activate JNK (269). Expression of MLK3 for 24 hrs led to a
2.1 i 0.5 fold increase in INK activity in INS-1 cells incubated in 4.0 mM glucose (Fig.
7). MLK3 further increased JNK activity up to 5.0 i 1.0 fold in cells cultured in 16.7
mM glucose compared to cells cultured in 4.0 mM glucose. In contrast, a kinase inactive
form of MLK3 termed MLK3K144A (249) was not able to activate INK. To assess
whether MLK3-mediated activation of JNK leads to functional changes in downstream
targets such as AP-l transcription factor activity, INS-l cells were co-transfected with
2XTRE-LUC and MLK3 or MLK3K144A for 24 hrs. MLK3 led to a 2.2 i 0.2 fold

(n=3) increase of 2XTRE-LUC expression in cells incubated in 4.0 mM glucose (Fig. 8).

82

F igurc 7. Over-expression of MLK3 in INS-l cells activates JNK activity. INS-1
cells were transfected with pRK (control vector), MLK3 or MLK3K144A for 24 hrs in
4.0 or 16.7 mM glucose. Cells were harvested and JNK kinase activity was assayed as
described in the Materials and Methods. Panel, A shows a representative JNK kinase
assay. Panel B, shows the mean :5 SE of JNK activity from five independent experiments
(n=5). The asterisk indicates that values were significantly increased between 4.0 and
16.7 mM glucose when compared to control for the given glucose concentration (p <
0.04). Data are normalized to the level of JNK activity measured in cells treated with 4.0
mM glucose. One experiment was performed by B. Bock, Department of Physiology,

Michigan State University.

83

CONTROL MLK3 MLK-3K144A

 

 

 

GLUCOSE (mM) 4.0 16.7 4.0 16.7 4.0 16.7

If; e’.‘ -.

..b'v"
I

    

p-C-IUN amt, first M

 

 

7

64—pRK *
EZZIMLKB I

54 —MLK3K144A

4.

 

 

Relative JN K activity

 

 

 

4.0 16.7

Glucose (mM)

84

 

l4

 

—pRK *
12‘ CZZIMLKB T
10 —MLK3K144A '
E
*5 8~
=
“D
..=.
:9 6*
n2

 

 

 

 

 

4.0 16.7
Glucose (mM)

Figure 8. Over-expression of MLK3 increases AP-l transcription factor activity.
INS-1 cells were co-transfected with 2XTRE-LUC (1.0 pg) and pRK (control vector,
0.5pg), MLK3 or MLK3K144A expression plasmid. Cells were then incubated in 4.0 or
16.7 mM glucose for 24 hrs. Cells were harvested and luciferase assay was performed.
Data are normalized to the level of control reporter gene expression measured in cells
treated with 4.0 mM glucose. Values are the mean i SE from three independent
experiments done in duplicates (n=3). Asterisk indicates that values were significantly
different between 4.0 and 16.7 mM glucose when compared to control for the given

glucose concentration (p < 0.002). # indicates that values were significantly different

compared to pRK values of the given glucose concentration (p < 0.05).

85

Addition of 16.7 mM glucose increased MLK3-induced 2XTRE-LUC expression 4.0 i
0.4 fold (n=3) over that observed in cells treated with 4.0 mM glucose. In contrast
MLK3K144A only marginally increased 2XTRE-LUC expression in cells incubated in
either 4.0 or 16.7 mM glucose. These data demonstrate that MLK3 activates INK which,
in turn, leads to functional changes in signal pathways dependent upon activation of AP-l
transcription factor activity. Thus, expression of MLK3 can be used as a tool to activate
JNK and investigate the potential role of JNK in the repression of insulin promoter

activity.

3.5 Expression of MLK3 in INS-1 cells represses insulin promoter activity

If JNK is involved in repression of insulin promoter activity when INS-1 cells are
chronically incubated in high glucose, then activation of JNK by MLK3 should also
repress insulin promoter activity. To test this hypothesis, IN S-l cells were co-transfected
with INSCAT and MLK3 or MLK3K144A expression plasmids and incubated for 48 hrs
in 4.0 or 16.7 mM glucose. As described above, incubation of INS-l cells in 16.7 mM
glucose led to a marked decrease in IN SCAT expression (F ig. 9A). Transfection of INS-
] cells with increasing concentrations of MLK3 expression plasmid reduced insulin
promoter activity up to 64.4 i 3.4 % (n=3) in cells cultured in 4.0 mM glucose (Fig. 9A).
As observed with JNK expression, MLK3-mediated repression of insulin promoter
activity was greater when cells were incubated in 16.7 mM glucose (Fig. 9A). In
contrast, expression of kinase inactive MLK3, MLK3K144A, increased INSCAT

expression in cells incubated in either 4.0 or 16.7 mM glucose (Fig. 9B). These data

86

Figure 9. Over-expression of MLK3 represses insulin promoter activity. Panel A,
INS-cells were co-transfected with INSCAT vector and pCR3.1 (control) and increasing
concentrations of MLK3 expression plasmid. Panel B, cells were co-transfected with
INSCAT vector and pRK (control), MLK3 or MLK3K144A. Data shown are the mean 1-
SE from three independent experiments done in duplicates (n=3). Asterisk indicates that
the values are significant different compared to control for the given glucose

concentration (p < 0.02).

87

cpm/ ug protein/hr

cpm/ ug protein/hr

 

 

 

 

 

 

 

 

30
A — 0.50 ug pCR3.1
25 - [2:23 0.10 ug MLK3
— 0.25 ug MLK3
20 q :1 0.50 ug MLK3
15 ‘
*
10 ‘
5 -1
0 -
120 *
B - 0.50 ug pRK
100 _ :3 0.50 ug MLK3

— 0.50 ug MLK3K144A

 

 

 

 

Glucose (mM)

88

support the hypothesis that increased INK activity induced by either chronic glucose

exposure or expression of MLK3 can suppress insulin promoter activity.

3.6 Expression of HP] in INS-l cells does not prevent glucose-induced insulin
promoter repression

JIPl protein is a specific JNK scaffolding protein that was first identified as a
cytoplasmic inhibitor of JNK (248). Over-expression of JIPl causes retention of JNK in
the cytoplasm and prevents its signaling activity (248). To further investigate whether
JNK mediates glucose-induced insulin promoter repression, we tested whether over-
expression of JIPl can prevent repression of insulin promoter activity. INS-l cells were
co-transfected with INSCAT and increasing concentrations of a plasmid encoding JIPl
protein. After transfection, cells were incubated in 4.0 mM glucose for ~ 8 hrs to allow
expression of JIPl under low glucose conditions. Transfection of cells with HP]
expression plasmid (0.25 pg to 1.0 pg) did not affect INSCAT expression in cells
cultured in 4.0 mM glucose for 48 hrs (Fig. 10). In contrast, cells transfected with 1.0 pg
of HP] expression plasmid had reduced expression of INSCAT when cultured in 16.7
mM glucose (Fig. 10). In conclusion, over-expression of HP] protein in INS-1 cells
cultured in 16.7 mM glucose was insufficient to prevent glucose-induced repression of
insulin promoter activity. The inability of HP] to prevent repression of insulin promoter
activity by glucose could be explained by three possibilities. First, JIPl protein may not
be expressed to levels necessary for complete inhibition of JNK. Second, JIPl protein is
actually an activator of JNK, thus over-expressed JIPl protein leads to further repression

of insulin promoter activity in cells cultured in 16.7 mM glucose where JNK pathway is

89

140

 

— 1.00 ug pCMV
12° ‘ :2: 0.25 ug JIPl
- 0.50 ug JIPl
E 100 d [:3 1.00ugJIP1
.E
9 ..
E 80
O-
go 60 q
E
9
40 ~
20 a
O ..

 

 

 

 

 

 

 

 

16.7
Glucose (mM)

Figure 10. Expression of HP] protein does not reverse glucose-induced repression
of insulin promoter activity in INS-l cells. Cells were co-transfected with INSCAT
vector and pCMV (control) or JIPl expression plasmid. Afier transfection, cells were
incubated for 8 hrs in 4.0 mM glucose and then for 48 hrs in 4.0 or 16.7 mM glucose.
Data shown are the mean :t SE from two independent experiments done in duplicates

(n=2). Asterisk indicates that the value is significant different compared .to control for

the given glucose concentration (p < 0.003).

90

activated. Third, other signaling pathways in addition to JNK are involved in insulin
promoter activity repression.

To test whether over-expression of JIPl protein effectively inhibits JNK activity
in INS-l cells, we analyzed whether glucose-induced activation of APl transcription
factor activity is inhibited by J 1P1. To accomplish this, INS-l cells were co-transfected
with 2XTRE-LUC and JIPl expression plasmids. Over-expression of HP] protein only
partially prevented increased 2XTRE-LUC expression in cells cultured in 16.7 mM
glucose (Fig. 11). The inability of HP] protein to fully prevent glucose-increased
2XTRE-LUC expression could be explained by two possibilities. First, glucose activates
2XTRE-LUC expression through additional signaling pathways. Second, insufficient
levels of J IPl protein are being expressed for complete inhibition of JNK.

p38 MAP kinase is another stress-activated protein kinase that is able to
phosphorylate and activate AP-l family members such as ATF2. ATF2 can
heterodimerized with AP-l transcription factors and bind to TRE elements (270). Thus,
p38 MAP kinase may be involved in glucose-induced increase of 2XTRE-LUC
expression. To test this hypothesis, INS-1 cells were transiently transfected with
2XTRE-LUC and then treated for 24 hrs with or without a p38 MAP kinase inhibitor,
SB203580 (10 pM) (271). p38 MAP kinase inhibitor did not affect 2XTRE-LUC
expression in cells incubated in 4.0 or 16.7 mM glucose (Fig. 11), indicating that p38

MAP kinase is not likely to be involved in glucose activation of AP-l in INS-1 cells.

91

 

3.0

_ control
2.5 ‘ 88203580
— HP]

2.0 -

1.5 -

Fold induction

1.0 ..

0.5 T

 

 

 

 

0.0 -

 

16.7

Glucose (mM)

Figure 11. Effect of HP] over-expression and p38 MAP kinase inhibitor on glucose-
induced APl transcription factor activity. INS-l cells were co-transfected with
2XTRE-LUC and 2 ug pCMV (control) or 2 pg JIPl expression plasmid. After
transfection, cells were treated with DMSO (control and over-expressed JIP] cells) or
SB203580 (10 M) in 4.0 or 16.7 mM glucose for 24 hrs. Cells were harvested and
luciferase activity was measured. Values represent mean i SE from three independent
experiments done in duplicates (n=3). The asterisk indicates that expression of JIPl
significantly reduces APl activity compared to control in 16.7 mM glucose (p < 0.01). #

indicates that the values are significantly increased compared to control in 4.0 mM

glucose (p < 0.05).

92

3.7 Over-expression of JIPl and inhibition of p38 MAP kinase did not prevent
glucose-induced repression of the insulin promoter.

p38 MAP kinase might mediate insulin promoter repression since it has been
shown to regulate insulin gene expression (272). Thus, a combination of p38 MAP
kinase and JNK activity may mediate glucose-induced repression of the insulin promoter.
To test this hypothesis, INS-1 cells were co-transfected with INSCAT and /or 2.0 ug JIPl
expression plasmid, then treated with or without SB203580 (10 M). Expression of HP]
protein or treatment with p38 MAP kinase inhibitor significantly reduced INSCAT
expression when cells were cultured in both 4.0 and 16.7 mM glucose (Fig. 12).
Expression of JIPl in combination with the p38 MAP kinase inhibitor did not prevent

INSCAT repression in cells cultured 16.7 mM glucose (Fig. 12).

3.8 JNK-induced repression of insulin promoter activity does not map directly to the
A1, A3, or C1 elements.

Next, insulin promoter element(s) involve in JNK-induced insulin promoter
repression were investigated. To overcome the complexity of interactions within the
intact promoter (INSCAT that contains from —327 bp to + 30 bp), a truncated version of
the insulin promoter linked to CAT, termed lNS(-230)CAT (containing —230 bp to +30
bp) was tested. INS-l cells were co-transfected with INS(-230)CAT and 0.1 ug JNKl
expression plasmid. Over-expression of JNKl repressed INS(-230)CAT when cells were
cultured in 4.0'mM glucose (Fig. 13). These data indicate that promoter elements

contained within the —230 promoter are sufficient for JNK-induced repression.

93

 

140.0
- control
120.0 . 12:1 SBZO3580
- JIPl
b 100.0 ~ [:1 SBZO3580+JIP1
.=
E
8 80.0 .
E
a.
E” 60.0 -
E
3
40.0 .
20.0 1
0.0 -

 

 

 

 

   

Glucose (mM)

Figure 12. Over-expression of JIP-l and inhibition of p38 MAP kinase does not
prevent glucose-induced repression of insulin promoter activity. INS-1 cells were co-
transfected with INSCAT vector and 2 ug pCMV (control vector) or 2 ug JIPl expression
plasmid. Cells were incubated for 8 hrs in 4.0 mM glucose to allow for JIPl protein
expression. Cells were then incubated with DMSO or 10 M SBZO3580 in 4.0 or 16.7
mM glucose for 48 hrs. Values represent the mean i SD of one experiment performed in
duplicate (n=2). Asterisk indicates that the values are significantly different compared to

control for a the given glucose concentration (p < 0.02).

94

 

 

Wt INS(-230)CAT £31 fl a CAT
mCl [j

mAlmClmA3 8 a E 4 EAE ]

 

 

 

 

 

 

 

 

     

 

 

 

 

 

 

50
— pCMV
I::1 JNK]
4O .
I-
E
.5 30 -4
‘é
a.
:9
E 20 - *
D.
U
* a:
10 «
0 .. ' I
INS(-230)CAT mA3Al mCl mA3C1A1
Vector

Figure 13. Repression of insulin promoter activity does not map solely to the A1,
A3, and C1 elements. INS-1 cells were co-transfected with INS(-230)CAT, mA3A1(-
230)INSCAT, mC1(-230)INSCAT, or mAlClA3(-230)INSCAT and 0.1 ug pCMV
(control) or 0.1 ug JNK expression vector. Cells were then incubated for 48 hrs in 4.0
mM glucose. Values represent means i SE of three independent experiments done in
duplicates (n=6). Asterisk indicates that values were significantly different compared to

control (p < 0.04).

95

The —230 bp promoter region contains four important regulatory elements, the A1 and A3
elements where de-l binds, the E1 element where 1312/1347 and Beta2 bind, and the Cl
element where C1 activator binds (273). To further investigate whether the A1, A3 and
Cl elements are involved in JNK-induced repression, the Al, A3, and/ or Cl elements
were mutated. mAlA3-INSCAT contains mutated A1 and A3 elements, mCl-INSCAT
contains mutated C1 element, and mA1C1A3-1N SCAT contains all three A1, A3, and Cl
mutated elements (mutations are described in the Materials and Methods). INS-1 cells
were co-transfected with INS(-230)CAT, mAlA3-INSCAT, mCl-INSCAT or
mAlClA3-INSCAT, and 0.1 ug JNK expression plasmid. The basal activity of all the
mutated promoter vectors was decreased compared to wild type (Fig. 13). Surprisingly,
expressed JNK significantly repressed the activities of all the mutated insulin promoters
(Fig 13). These results can be interpreted in two possible ways. First, the A1, A3, and
C1 elements are not involved in JNK-induced repression of insulin promoter activity,
suggesting that other elements such as E1 or CRE within the —230 bp promoter may
mediate the repression. The E1 element can be a possible target of JNK, since this
element is markedly regulated by acute glucose (91) and c-Jun inhibits the transactivation
potential of E47, which binds to E1 element (72). The second possibility is that JNK can
affect a general target such as a transcriptional co-activator.

Previous experiments have demonstrated that the E1 element is a very important
regulatory element of insulin promoter activity, and deletion of this element causes a
marked reduction in promoter activity (87, 91). To analyze whether JNK represses the
insulin promoter through the El element, a multimer of three E1 elements linked to the

RSV-TATA box (RSV103CAT) reporter vector was constructed. INS-1 cells were co-

96

transfected with the multimerized El ((E1)3RSV103CAT) vector or RSV103CAT
(control), and 0.1 ug JNK expression plasmid, and incubated in 4.0 or 16.7 mM glucose.
In contrast to INSCAT, (E1)3RSV103CAT expression was not repressed in cells cultured
in 16.7 mM glucose (Fig. 14). These results are in agreement with previous results that
show that glucose induces binding of factors to the E1 element (126). Expression of INK
protein repressed RSV103CAT (control) expression in cells cultured in 16.7 mM glucose.
In contrast, expression of INK protein did not significantly inhibit (E1)3RSV103CAT
expression in cells cultured in 4.0 and 16.7 mM glucose (Fig. 14). These data indicate
that over-expression of JNK protein does not repress the activity of multimerized E1
element, suggesting that El element does not mediate INK-induced repression of the

insulin promoter.

4. Discussion

The role of INK, a stress-activated protein kinase, in mediating repression of
insulin promoter activity in INS-1 cells exposed to high concentrations of glucose was
studied. INS-l cells cultured in 16.7 mM glucose markedly increased JNK activity in
parallel with increased AP-l transcription factor activity. Over-expression of JNK]
protein repressed insulin promoter activity in INS-1 cells cultured in 4.0 mM glucose.
These results suggest that the JNK signaling pathway might mediate hyperglycemic toxic
effects on the insulin promoter. Our results agree with previous reports that demonstrated
that over-expression of c-Jun, a downstream target of JNK, can repress insulin promoter

activity in HIT cells (79). Over-expression of c-Jun was shown to inhibit insulin gene

97

 

 

25.0
— RSV103CAT + pCMV
RSV103CAT + 1in
20.0 ~ _ (El)3RSVCAT+pCMV
h [:1 (E1)3RSVCAT + 1le
“S
.5 15.0 -
E
a.
g .l
E 10.0
D-
O
5.0 ~
0.0 -

 

 

 

 

 

 

 

 

16.7

Glucose (mM)

Figure 14. Expression of JNK does not repress the multimerized E1 element
activity. INS-1 cells were co-transfected with (El)3RSV103CAT and pCMV (0.1 ug) or
JNK] expression vector. Then cells were incubated for 48 hrs in 4.0 or 16.7 mM
glucose. Values represent means i SE of three independent experiments done in

duplicates (n=3). Asterisk indicates that the value was significant different compared to

control in 4.0 mM glucose (p < 0.0007).

98

transcription by reducing the transactivation potential of the El2/E47 transcription factors
which bind to the E-box (72). Recently, Kaneto et al. (8) demonstrated the involvement
of JNK in suppressing insulin gene expression in isolated rat islets treated with H202, a
strong activator of oxidative stress. The authors showed that H202 treated isolated islets
had increased JNK phosphorylation and repressed insulin mRNA levels. In addition,
over-expression of JNK lowered insulin mRNA levels in the absence of H202 and over-
expression of a dominant negative form of JNK helped to maintain insulin mRNA levels
from the toxic effects of H202. Many investigators have demonstrated that high levels of
glucose increase ROS in B-cells (7, 157, 158). Furthermore, treatment of B-cells with
antioxidants provides some protection against B-cell dysfunction such as partial recovery
of de-l activity and insulin mRNA expression (7). Overall these reports highly suggest
that hyperglycemia increases reactive oxygen species which, in turn, activate the JNK
signaling pathway and this may mediate repression of insulin promoter activity.

Because of lack of specific JNK inhibitors, over-expression of HP] protein was
tested as a JNK inhibitor to investigate the involvement of JNK in glucose-mediated
repression of insulin promoter. JIPl protein is a scaffolding protein that was first
identified as a cytosolic anchor protein that binds specifically to JNK (248). In some cell
systems over-expression of J IPl protein retains JNK in the cytoplasm thereby inhibiting
JNK regulation of gene expression (248). Over-expression of JIPl in INS-1 cells failed
to prevent the repression of insulin promoter activity by high levels of glucose. In fact,
high concentrations of JIPl led to further repression of insulin promoter activity. The
inability of HP] protein to prevent the repression insulin promoter activity can be

interpreted in three possible ways. First, the amounts of plasmid used to express JIP-l

99

protein (0.25 to 2.0 ug) may have been insufficient to retain JNK in the cytoplasm. In-
deed, these concentrations of JIPl may have actually enhanced INK activity through its
scaffolding activity and led to a further repression of insulin promoter activity. Second,
over-expressed J 1P1 protein may have had non-specific effects on the INS—1 cell and may
have inhibited essential kinases or transcription factors required for insulin gene
expression. Third and more likely is that JIPl protein was unable to reverse glucose-
induced repression of insulin promoter activity because other mechanisms may be
involved in this repression.

Several investigators have demonstrated that high levels of glucose repress insulin
expression by a variety of mechanisms. These various mechanisms include c-Myc,
CCAAT/Enhancer-binding protein B (C/EBPB), de-l, and C1 activator (3, 85, 86, 121,
126, 274). c-Myc expression has been shown to increase in islets from diabetic 90 %
pancreatectomized rats (4) and from rats made hyperglycemic with glucose clamps (275).
It was demonstrated that over-expression of c-Myc suppresses insulin gene expression by
inhibiting BETAZ-mediated transcriptional activation (274). Investigators have also
reported that supraphysiological glucose concentrations increase the expression of the
transcription factor C/EBPB in different pancreatic B-cell lines (86) as well as in Zucker
diabetic fatty rats and 90 % pancreatectomized rats (85). In these diabetic models,
repression of insulin mRNA was correlated with increased C/EBPB mRNA (85). In
addition, over-expression of C/EBPB repressed insulin promoter activity by inhibiting
binding activity of E47 transcription factor (86). de-l and C1 activator are also
associated with the repression of insulin gene expression (2, 123). High levels of glucose

reduce DNA binding activities of de-l and Cl activator to insulin promoter activity (2,

100

3, 121, 126). Overall, these results strongly suggest that repression of insulin gene
expression involves various mechanisms, which might include JNK signaling pathway.
Interestingly, Noguchi et al. (276) demonstrated that c-Myc is specifically
phosphorylated by JNK in numerous cells types treated with UV radiation. One can
postulate that hyperglycemia increases JNK activity that leads to phosphorylation and
activation of c-Myc, which can repress insulin gene expression.

p38 MAP kinase is another stress-activated protein and that is involved in acute
glucose regulation of de-l (272), suggesting a possible role of p38 MAP kinase in
glucose-induced repression of insulin promoter activity. Inhibition of p38 MAPK,
however, did not prevent glucose-induced insulin promoter repression. In contrast,
inhibition of p38 MAPK lowered insulin promoter activity when cells were cultured in
either 4.0 or 16.7 mM glucose. These results can be interpreted in three different ways.
First, p38 MAP kinase may be required for insulin promoter activity, thus inhibition of
p38 MAP kinase would reduce insulin promoter activity irrespective of glucose
concentration. Such as interpretation is supported by previous reports that demonstrated
exposure of B-cells to acute glucose induces de-l phosphorylation by p38 MAP kinase,
resulting in activation and translocation of de-l to the nucleus where it stimulates
insulin gene expression (272). Second, the concentration of p38 kinase inhibitor
(88203580) used may have had non-specific effects and may be inhibiting essential
kinases that regulate insulin gene expression. Third, additional mechanisms besides JNK
and p38 kinase may be involved in the repression of insulin promoter activity.

To investigate which elements are involved in JNK-induced repression of insulin

promoter activity, a truncated insulin promoter containing sequences fiom --230 to + 30

101

(termed INS(-230)CAT) was analyzed. Over-expression of JNK repressed INS(-
230)CAT activity in cells cultured in low glucose concentrations. Unexpectedly, over-
expression of JNK also repressed the —230 insulin promoter activity containing mutations
in the A1, A3, and Cl elements. These results can be explained if JNK affects an
upstream target such as a co-activator that is necessary for de-l and C1 activator to bind
DNA and transactivate insulin gene.

The lack of identifiable JNK target(s) on the insulin promoter suggests that JNK
protein affects an upstream target(s) that regulates insulin gene transcription such as co-
activators and co-repressors. Recently, Pessah, M et al. (277) demonstrated that JNK
represses transcriptional activity by inhibiting the interactions among transcription factors
and co-activators by inducing transcription factors to interact with co-repressors.
Furthermore, Qiu et al. (96) demonstrated that the synergistic interactions between de-l ,
Beta2, and E47 are mediated through the co-activator p300. They showed that the
adenovirus ElA protein interacts with p300 and disrupts the interaction between p300
and de-l, thus inhibiting insulin gene transcription (96). A possible mechanism by
which JNK inhibits insulin promoter activity could be by inhibiting the ability of co-
activators to interact with several transcription factors such as de-l, Beta2, and E47.
Over-expression of increasing concentrations of p300 and/ or cAMP-response element-
binding protein (CBP), both well known co-activators, did not prevent JNK inhibitory
effects on INS(-230)CAT activity (data not shown). These results, however, do not rule
out the possibility that JNK can affect other co-activators.

It is noteworthy that lNS(-230)CAT is repressed only 37.3 % (n=3) in IN S-l cells

cultured in high levels of glucose (data shown in next chapter), while the INS(-327)CAT

102

is repressed 70 to 80 % (126). These results further suggest that high levels of glucose
repress insulin promoter activity through several potential sites involving various
mechanisms. In addition, characterization of the 5’ region of the insulin promoter
suggests that the repression of insulin promoter activity involves other mechanisms that
affect the distal insulin promoter as will be discussed in the next chapter.

In conclusion, glucose-induced B-cell dysfunction likely involves the generation
of reactive oxygen species that can activate stress-activated signaling pathways including
JNK and this ultimately leads to detrimental changes in pancreatic B-cell gene

expression.

103

V. CHRONIC EXPOSURE OF INS-1 CELLS TO HIGH GLUCOSE
CONCENTRATIONS DECREASES INSULIN GENE REPORTER ACTIVITY

THROUGH NOVEL REGULATORY ELEMENTS

1. Abstract

Chronic hyperglycemia induces B-cell dysfunction that is associated, in part, with
reduced insulin gene promoter activity. We have reported that insulin promoter activity
is markedly repressed in INS-1 cells cultured in 16.7 mM glucose compared to 4.0 mM
glucose. We now report that insulin promoter activity is gradually increased when INS-1
cells are cultured for 48 hrs in 4 and 6 mM glucose compared to 2.0 mM glucose. In
contrast, glucose concentrations above 8 mM cause a concentration-dependent reduction
in promoter activity. In INS-1 cells, repression of insulin promoter activity by elevated
glucose levels has been associated with reduced binding activity of de-l and the C1
activator (RIPE3bl). Studies designed to assess the role of de-l and C1 activator in
glucose-induced insulin promoter repression led to the discovery that the majority of the
repression mapped to the sequences upstream of the de-l and C1 activator binding
elements. Truncation analysis indicated that a strong repression site is located between —
327 and —261 nucleotides. This region includes the X minienhancer, which contains the
AS/Core and E4 elements, and the Z minienhancer, which includes a palindrome
sequence and the E3 element. As observed with insulin promoter activity, the activities
of the X and Z minienhancers had a biphasic response to glucose. Thus, the X and Z
minienhancer activity increased in cells cultured in media containing 6.0 mM glucose

compared to 2.0 mM glucose, and decreased in cells cultured in glucose concentrations

104

above 8.0 mM. DNA binding activities to the AS/Core element, palindrome/E3a
element, and palindrome/E3a-E3b elements indicated three different-sized complexes,
which were markedly reduced in cells cultured in 16.7 mM glucose. Super-shift mobility
assays demonstrated that de-l, ka6. 1, and Pax6 do not form part of the three different-
sized complexes, and these complexes were not effectively competed by the rat I E2
element. These data demonstrate that high levels of glucose can functionally repress the
X and Z minienhancers, and this repression is correlated with decreased binding activity
of their binding complexes. Glucose-induced insulin promoter repression was partially
prevented when all three AS/Core, palindrome, and E3 elements were mutated,
suggesting that these elements mediate some of the glucose-induced insulin promoter
repression. In conclusion, these data indicate the existence of additional transcriptional
mechanisms by which elevated glucose concentrations can repress insulin promoter

activity.

105

2. Introduction

Mammalian glucose homeostasis is tightly controlled by insulin production and
release. Glucose, on the other hand, is one of the major physiological regulators of
insulin gene expression and secretion (278). Acute changes in glucose concentrations
increase insulin synthesis by controlling insulin transcription (80), insulin mRNA
stability (193), and insulin gene translation (279). Glucose responsiveness of insulin
gene transcription is associated with increased insulin promoter activity (95). Several
glucose-responsive elements in the — 400 insulin promoter have been identified (91).
These elements belong to three classes of sequences, A elements, E elements, and the C1
element (Fig. 16). The A elements, Al-AS, except for the A2, are characterized by AT
rich sequences with a consensus sequence of TAAT (280), and bind several transcription
factors including the pancreatic duodenal homeobox-l (de-l) transcription factor (281).
de-l is essential for pancreas development and B-cell differentiation (282), and is a
major transactivator of the human insulin gene (91, 283-285). The A3 element is highly
regulated by glucose and its deletion has a detrimental effect on insulin transcription
(281). The E elements have a consensus sequence of CANNT G. This sequence is
recognized by a heterodimer of two helix-loop-helix factors, including the ubiquitously
expressed proteins E12/E47 (286) and the B-cell specific factor, Beta2/Neuro D (66).
There are two main E elements, E1 and E2, and both are regulated by glucose. Mutations
within these elements decrease insulin promoter activity (92). Transcriptional activity of
an isolated E1 element is very low, but is potentiated by the addition of adjacent
regulatory elements (95, 287, 288). Therefore, E elements are required for insulin gene

transcription, and their activity is dependent on other elements such as the A1 and the Cl

106

element. In addition to the A and E elements, the C1 element is also regulated by glucose
(289). The C1 element, with a C-rich sequence, binds the Cl activator (RIPE3b1) (74,
94, 290) that has only recently been cloned (5). Finally, an additional glucose-responsive
element in the distal human insulin promoter, the Z minienhancer (-292/-243), was
recently described using primary culture islets (89). The Z minienhancer contains the
previously described negative regulatory element (NRE) (87, 88) and binds the ZaI
complex that has not been well characterized.

The NRE is located between positions —279 and —261 on the human insulin
promoter and deletion of sequences from —270 to —258 leads to a 25-fold increase in
promoter activity in HIT-T15 cells (87). In this cell line, the NRE activity is modulated
positively by sequences located from -279 to — 341 (88). The NRE resembles a silencer
because its negative activity is partially independent of location and orientation, and can
suppress a variety of promoters (88). The role of NRE has not been fully described but
may be involved in restricting expression of the insulin gene to B-cells.

Chronic hyperglycemia, as observed in Type II diabetes mellitus, causes fi-cell
dysfunction including decrease glucose-stimulated insulin secretion and suppression of
insulin gene expression. In vivo and in vitro studies have demonstrated that
hyperglycemia suppresses insulin gene expression by decreasing insulin promoter
activity. Decreased promoter activity is correlated with reduced binding activity of de-l
and the C l activator (RIPE-3bl) (2, 85, 121, 126, 291).

Our laboratory has shown that chronic exposure of INS-1 cells to high levels of
glucose decreases insulin promoter activity and this is associated with reduced binding

activity of de-l and the C1 activator (126). Decreased de-l binding activity is

107

associated with a post-translational decrease in de-l mRNA levels (126). The decrease
in de-l and C1 activator binding activity is readily reversible by incubation of INS-1
cells in low glucose concentrations (126). Nevertheless, over-expression of de-l is not
sufficient to prevent chronic hyperglycemia from inhibiting insulin promoter activity
(L.K. Olson, unpublished data). Furthermore, other investigators have suggested that
hyperglycemia can increase c‘Myc and C/EBPB levels, thus suggesting a more
complicated mechanism (86, 274). Thus, the mechanism by which insulin gene
expression is decreased in B-cells chronically exposed to supraphysiological glucose
concentration remains to be completely elucidated. Studies described within this chapter
were performed to investigate other possible promoter elements in the insulin gene that

are involved in glucose-induced repression of insulin promoter activity in INS-l cells.

3. Results

3.1 Glucose regulates insulin promoter activity in a biphasic manner in INS-1 cells.
Our laboratory has reported that exposure of INS-1 cells to 16.7 mM glucose for
24 to 48 hrs leads to marked suppression of insulin promoter activity (126). We have
also shown that insulin mRNA levels start to diminish when cells are incubated in
glucose concentrations at or above 8.0 mM glucose (126). We analyzed insulin promoter
activity at different concentrations of glucose and tested whether promoter activity has
the same pattem as insulin mRNA levels. To assess promoter activity, INS-1 cells were
transiently transfected with an insulin promoter CAT reporter gene (INS(-327)CAT), in
which chloramphenicol acetyl transferase (CAT) gene expression is regulated by

sequences from —327 to +30 of the human insulin promoter (3). Cells were then treated

108

with 2.0, 4.0, 6.0, 8.0, 14.0 or 16.7 mM glucose for 48 hrs. Insulin promoter activity
increased 2.2 i 0.3 or 3.6 i 0.6-fold in cells incubated in 4.0 or 6.0 mM glucose,
respectively, compared to 2.0 mM glucose (Fig. 15). In contrast, incubation of cells in
concentrations greater than 8.0 mM glucose led to a concentration-dependent reduction in
insulin promoter activity. Thus, incubation of INS-l cells in 16.7 mM glucose led to an
82.9 i 1.4% reduction in insulin promoter activity, compared to cells incubated in 4.0

mM glucose.

3.2 The A1, A3, and C1 (RIPE3b1) regulatory elements have a minor role in

mediating glucose-induced repression of insulin promoter activity in INS-1 cells.

The decrease in insulin mRNA and promoter activity observed in INS-l cells
incubated in high levels of glucose is correlated with a reduction in DNA binding activity
of de-l and the C1 activator (126). To determine the relative roles of de-l and C1
activator binding elements in insulin promoter repression in cells treated with high levels
of glucose, a simplified insulin promoter vector termed IN S(-230)CAT was constructed.
INS(-230)CAT is a CAT reporter gene regulated by insulin promoter sequences from —-
230 to + 30 (Fig. 16). This reporter gene contains the A1 (-79 to ~84) and A3 (-210 to —-
216) elements where de-l binds, and the C1 element (—116 to -124) where the Cl

activator (RIPE3bl) binds. Unexpectedly, INS-l cells transfected with INS(-230)CAT

and incubated in 16.7 mM glucose demonstrated only a small (~ 34 %) suppression of

109

500

 

400 ~

300 a

200-

cpm/ug protein/hr

100 -

 

 

 

 

rnM glucose

Figure 15. Glucose regulates insulin promoter activity in a biphasic manner in INS-
] cells. INS-l cells were transiently transfected with INS(-327)CAT vector, and then
were incubated in different concentrations of glucose for 48 hrs. Cells were then
harvested and assayed for CAT activity. Values are the mean i SE of four individual
experiments done in duplicates (n=4). # indicates that the values are significantly
increased compared to 2.0 mM glucose (p < 0.01). Asterisk indicates that the values are

significantly reduced compared to 4.0 mM glucose (p < 0.005).

110

doc: .3 32.823“

2m 2985 88:58 5:55 .xon some 96% @8865 En 3:256 82: 9 wE—EB 86on .338: 2a $5520 58an

boa—swam 6.8.3.» .5382; 5.3.: 68.8.5.5 an: .83an 5.3.... 525:. ue 53358.59. czanaom .3 0.53..—

 

 

 

 

 

 

 

 

F<U

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

_cn~-
J cmn-
_ SN-
_ aha-
_ «an-
3
5n.
<<H<h r N< ll ”.55 1 m< an.— N 9893‘ 1|
. 8 .o. o 8 6
736m «Somkvm c3958 5 36¢ NSomRvm
_ i. _ _ _ _ r
I. an. and... cmu- EN. EN. Na- an-

111

promoter activity compared to cells incubated in 4.0 mM glucose (Fig. 17B). Consistent
with our previous report (126), INS(-327)CAT expression was markedly suppressed (~
80 %) in cells incubated in 16.7 mM glucose compared to 4.0 mM glucose (Fig. 17A).
To analyze the role of de-l and the C1 activator in the repression of INS(-230)CAT, the
Al, A3 and/ or C1 elements were mutated. As expected, mutation of both Al and A3
elements led to a large (71.2 i 3.3 %) decrease in basal INS(-230)CAT expression, and
importantly, the repression of glucose-induced insulin promoter activity was prevented
(Fig. 18). In a similar manner, mutation of the C1 element (mCl) led to a 74.5 i 3.9 %
decrease in basal 1NS(-230)CAT expression and also prevented glucose-induced insulin
promoter repression (Fig. 18). In conclusion, these data indicate that the Al, A3, and C1
elements are involved in glucose-induced repression of lNS(-230)CAT, however, most of

the repression of the insulin promoter lies upstream of —230.

3.3 Upstream regulatory elements play a major role in glucose-mediated repression

of insulin promoter activity.

To investigate the upstream elements involved in glucose-induced repression of
insulin promoter, activity, a series of truncated insulin reporter genes were
constructed(Fig. 16). Consistent with previously reported results (126), the INS(-
327)CAT expression was repressed 79.1 i 1.6 % when INS-1 cells were incubated in
16.7 mM glucose compared to 4.0 mM glucose (Fig. 19). Deletion of the promoter

region from — 327 to —292, INS(-292)CAT, which removes the A5 element (—292 to —

112

 

350

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

A INS(-327)CAT B INS(-230)CAT
300 -» I ._ «
250
E
2 200 r i
E
S
=3 150 r
E
3 It at:
100 «- .. y
80 % T 33 %
50 l" T T
0 I v 1 T
4.0 16.7 4.0 16.7

Glucose (mM)

Figure 17. The —230 insulin promoter has a smaller repression than the —327 insulin
promoter in INS-l cells cultured in 16.7 mM glucose. A. INS-l cells were transiently
transfected with INS(-327)CAT vector (A) or INS(-230)CAT (B) and then cultured in 4.0
or 16.7 mM glucose for 48 hrs. Cells were then harvested and assayed for CAT activity.
Values are the means i SE of five independent experiments done in duplicates (n=5).
Asterisk indicates that values were significantly different between 4.0 and 16.7 mM

glucose (p < 0.02).

113

100

 

- 4.0 mM glucose

30 4 1:22] 16.7 mM glucose

 

60-

4o-

cpm/ug protein/hr

20-

 

 

 

 

 

 

 

 

lNS(-230)CAT mAlmA3 mCl mAlmClmA3

VCCIOI'

Figure 18. Mutations of both A1 and A3 or Cl prevent glucose-induced repression
of the INS(-230)CAT promoter activity. INS-l cells were transiently transfected with
different vectors, shown in the figure, and were then cultured in 4.0 or 16.7 mM glucose
for 48 hrs. Cells were then harvested and CAT activity was assayed. mAlmA3
represents mutated Al and A3 elements. mCl represents mutated C1 element and

mAlmClmA3 represents the mutations of the three elements. INS(-230)CAT and mCl
values are the mean i SE of 5 independent experiments done in duplicates (n=10).
mAlmA3 and mAlmClmA3 values are the mean of i SE of 4 independent experiments

done in duplicates (n=8). Asterisk indicates that values were significantly different

between 4.0 and 16.7 mM glucose (p< 0.05).

114

Figure 19. Truncation analysis of —327 insulin promoter activity in INS-1 cells
cultured in 4.0 and 16.7 mM glucose. INS-l cells were transiently transfected with
different truncated insulin promoter vectors shown in Figure 16 and in the above
schematic representation of insulin gene promoter (not to scale). The cells were then
cultured in 4.0 or 16.7 mM glucose for 48 hrs. Values represent 5 experiments done in
duplicates (n=10) with the means i SE. Asterisk indicates that values were significantly

different between 4.0 and 16.7 mM glucose (p< 0.05).

115

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

-327 -292 -279 -261 -250 -230 -80
l I I l I j I
“'ASICoreI" E4 — Pal 123 E2 CRE ‘A2/C1 E1 n Al
2
-327 INSCAT i i [:1 4.0 mM lucose
I . 16 7 glucose
'2”- ‘NSCAT _ . l '
-279 INSCAT F
i
461““ _:H
-250 INSCAT
*
-230 INSCAT *
0 100 200 300 400 500
cpm/ug protein/hr

116

 

314), the enhancer core element (Core) (-317 to —309), and an E-box like element named
the E4 element (-300 to -294), caused an approximately two-fold increase in CAT
expression in both 4.0 and 16.7 mM glucose compared to INS(-327)CAT (Fig. 19).
INS(-292)CAT expression was repressed 59.1 i 2.5 % in cells incubated in 16.7 mM
glucose compared to 4.0 mM glucose. Deletion of the promoter region from —327 to —
279, lNS(-279)CAT, which removes the A5/Core, E4 and destroys a palindrome
sequence (-284 to -273) contained in the Z minienhancer (89) and negative regulatory
element (NRE) (88), led to only a 27.9 i 5.4 % repression of CAT expression in cells
cultured in 16.7 mM glucose compared to 4.0 mM glucose. Removal of promoter
sequences from —327 to —261, INS(-261)CAT, which deletes the A5/Core, E4 element,
the palindrome, and the E3 element (273 to —257), did not further decrease the ability of
16.7 mM glucose to suppress insulin promoter activity. Further removal of promoter
sequences up to -—230 did not lead to any additional decrease of the high glucose-induced
suppression of insulin promoter activity. These data suggest that regulatory elements
within —327 and —279 promoter sequences are involved in glucose-induced repression of

insulin promoter activity.

3.4 The Z, X, and Y minienhancers contain promoter sequences involved in glucose-

induced biphasic regulation of insulin promoter activity.

Sander et al. (89) recently characterized glucose regulation of distal human
insulin promoter sequences from —341 to —243 using multimerized (5 copies) regions of

promoter (named minienhancers) linked to the —-85 bp minimal promoter from the rat

117

insulin I gene promoter. They divided this region into X (-342 to —293), Y (-3 l 7 to —268)
and Z (-292 to —243) minienhancers (Fig. 20). The X and Y minienhancers both contain
the A5/Core element, while the Y minienhancer also contains a palindrome and the 5’
region of the E3 element. The Z minienhancer contains a palindrome sequence and the
E3 element. The E3 element contains two E-like sequences that we have termed E3a and
E3b, separated by four base pairs (Fig. 20). The Z minienhancer was further divided into
the Za (-292 to —263) and Zb (-274 to -—246) minienhancers. The Za minienhancer (-292
to —261) contains a palindrome sequence and the E3a (-273 to —268), while the Zb
minienhancer (-275 to —245) contains both the E33 and E3b (-263 to —257) and not the
palindrome (Fig. 20). Sander et a1. (89) demonstrated that the Z minienhancer had a very
strong transcriptional activity and its activity was further increase in primary islets
cultured in 16.0 mM glucose compared to 2.0 mM glucose. They concluded that the Z
minienhancer is a glucose-responsive element only in primary cultured islets. In contrast,
the X and Y minienhancers had very weak transcriptional activity when transfected in

fetal islets and did not respond to glucose.

In order to investigate possible upstream regulatory elements involved in the
glucose-induced repression of insulin promoter activity, INS-1 cells were transiently
transfected with the multimerized (5 copies) Z, X, or Y minienhancers and were then
incubated in 2.0, 4.0, 6.0, 8.0, or 16.7 mM glucose for 48 hrs. The -—85 to +1 rat insulin

promoter (—85 rlNSCAT/pFOXCATZ) was found to be only slightly glucose-responsive

in INS-l cells, with a 1.8 i 0.2 or 2.7 i 0.1-fold increase in CAT expression in cells

118

.388 «£93239: Sm mom:

moDOhQ mm :0}, ma .muGDEQ—D MDHOEOHQ MO numofihofig $32.? 0.8 @2509» notomoh UDNCDEmSSE am UOm: mwfioEO—D Hoocwfivmfimz

6382 2a $5820 88an 5.35 $35.53 .8359:— ===m£ .55. no 5585830.. 3382.5 .3 9:.»3

nmmémmN

Scamp—:22? >

111.1 11 1 111.1-.1. i , 114

 

 

 

 

 

1@ L a __ _1 am- .a _ g... a E
omm- SN- 2h- Nam- Rm-
— W p p b
- d d u — 4 q
omm- ova- com- own- com- own. 03..

119

incubated in 6.0 or 16.7 mM glucose, respectively, compared to 2.0 mM glucose (Fig.
21 ). We believe this small response is not insulin promoter specific because we have also
found a small response with the Rous sarcoma virus minimal promoter (Fig. 14, chapter
I). In contrast, the Z minienhancer had a large activity compared to the —85 rINSCAT.
The Z minienhancer had a 3.1 i 0.2-fold increase when cells were incubated in 6.0 mM
glucose compared to incubation in 2.0 mM glucose (Fig. 21). As observed with INS(-
327)CAT, glucose concentrations greater than 8.0 mM glucose led to a decrease in Z
minienhancer activity compared to cells incubated in 4.0 or 8.0 mM glucose. Incubation
of INS-l cells in 16.7 mM glucose led to a 38.2 i 5.1 °/o reduction in Z minienhancer
activity compared to cells incubated in 4.0 mM glucose (Fig. 21). In order to determine
whether the palindrome and/ or E3a-E3b elements were involved in the glucose-induced
repression of the Z minienhancer, we analyzed the ability of glucose to repress the la
minienhancer, which contains the palindrome and the E3a element, and the Zb
minienhancer, which contains only the E3a and E3b elements (Fig. 20). As observed
with the Z minienhancer, both the Za and lb minienhancers activity had a biphasic
response to glucose with a ~2.7 and ~4.1-fold increase, respectively, when cells were
incubated in 6.0 mM glucose compared to 2.0 mM glucose (Fig. 22). When cells were
incubated in 16.7 mM glucose, Za and Zb minienhancers activity were also repressed by
~ 80.0 % and ~ 73.0 %, respectively, compared to 4.0 mM glucose. These data suggest
that the E3a element, which is contained in both the Za and lb minienhancers, may be
involved in the repression of insulin promoter activity by high levels of glucose.
Alternatively, independent elements within the Za and lb minienhancers may be

involved in glucose-induced insulin promoter repression.

120

.Cod v5 882w 28 5.3 98 oi :83on “couobmv
bamocmcwfi Bo? 83? on“ 35 8:8me xmtowmax .Cod v 3 882m 28 ON 8 389:8 BENCE DES—.«Ewa 803
mos—S, 05 :2: 82865 n .Avnav $28296 5 28¢ mEoEtono “covenants: 95 .3 mm H :38 05 Pa 839, EC. .8833
28 od 5 3:53 £3 E .869» 223830 NH<UXOHE 05 8 9528 Ba 82? of. 6ch 05 E 550% 3 882m
mo mcocmbcoocoo E E: wv do.“ 33:20 :2: 203 £60 .NH<UXOmn 5 28m .50 2: wE>EV 58805 H “E mwl wEEmEoo
c309» 3550 m S 8505 2005::358 2: .3 838 823:: 96 $35ch mEEmmE 53, 368mg: 32565.5 0.5?

$8 TmZH .382» 2:. 5.3 5 69.3.3 «.8 £3 TmZ— 5:3 e883?— oha 28.—«€85:— X was N 2E. AN 95E”—

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

AEEV 8.820
3. SW 3 3. cm 62 3 ca 3. o.~ 3. 3w 3 3. o.~

t . t t . O O r . 1 1 r O
. r . N com W
H . — m
v v m.
2:. H 3
u o H n f N W
coo c u e
. w a n
Z n w”
. 2 gm H .M

a “usages? x a 505,225,: N a $82: mffifixofi
N. 82 v

 

 

 

121

Amod v 5 882m 28 52 was 06 5253 “:20wa hangout—ma 203 mos—9» 05
35 832?: €133. Amod v5 882m 28 ON 2 uoSQEoo Eofibfi figment—ma 803 was?» 2: 35 82865 t .AvHE
882.96 E 25v 3585?... “newcoaovfi So.“ go mm H :35 2: 8m 82? och 683% 28 od E @8330 £8 5 £20?»
:23898 05 9 038—2 Ba mos—g 2F Esme 2: E 850% mm 588% .3 303826280 Hecate E ME 3 go.“ 3553
:05 803 £30 .8505 Egan—€258 2: .«o 8&8 83:3 96 mfifificoo mEEmaE 53> 6200355 325853 203 £30

~1WZH 6255—“ SE h.©~ fl. fichflu—flu Qua m=ao —Imz~ Eon—>9 fiQmehQOh Qhfl whouflwflflomflms AN mafia . «N .> 05H. .NN Ohflwmnm

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

A25 8820
5.2 o6 3. QN we. od 3. o.~ E: Qo ed o.~

> > » p O 1 p p p O t t . O
y — c a. .J
H .. — N p
. Z w...
v N A A
H . a
. N . . m

u
1 1
xx 7 M w e
, a. m.
m.
. w v c w ,M
u Coca—EEEE 2N w Scamp—:BEE «N u BosmacoEmE >
o m o.

 

 

 

 

 

 

 

 

 

122

Deletion of insulin promoter sequences from —327 to —292, INS(-292)CAT,
demonstrated that high levels of glucose mediated some promoter repression through this
region. This region contains the AS/Core and the E4 elements. To determine the effect
of glucose on this general region, INS-1 cells were transiently transfected with the X
minienhancer reporter gene and incubated in 2.0, 4.0, 6.0, 8.0, or 16.7 mM glucose for 48
hrs. The X minienhancer activity was markedly lower than the Z minienhancer activity
(Fig. 21), as previously observed in fetal rat islets (89). Interestingly, the X minienhancer
also had a biphasic response to glucose, with a 5.3 i 0.7-fold increase when cells were
incubated in 6.0 mM glucose compared to 2.0 mM glucose (Fig. 21). The X
minienhancer activity was also markedly reduced when cells were incubated in 8.0 or
16.7 mM glucose (Fig. 21). The X minienhancer activity was decreased by 74.4 i 0.2 %
in cells incubated in 16.7 mM glucose compared to 4.0 mM glucose. These data suggest
that high glucose concentrations repress the insulin promoter activity, in part, through the

promoter region from -327 to —292.

We also analyzed the ability of glucose to repress the Y minienhancer activity,
which contains the AS/Core, the E4, the palindrome, and the E3a element (Fig. 20). INS-
] cells were transiently transfected with Y minienhancer reporter vector and cells were
then incubated in media containing 2.0, 4.0, 6.0, or 16.7 mM glucose for 48 hrs. As
expected, the Y minienhancer also had a biphasic glucose response with a 7.1 i 2.1-fold
increase in cells incubated in 6.0 mM glucose compare to 4.0 mM glucose (Fig. 22). The
Y minienhancer was also repressed by ~ 82.0 % in cells incubated in 16.7 mM glucose

compared to 4.0 mM glucose. Overall, these data suggest that the AS/Core, the E4

123

element, the palindrome, and the E3 element all have possible roles in mediating glucose-

induced repression of insulin promoter activity.

3.5 High levels of glucose repress DNA binding activity to the AS/Core and E3

regulatory elements.

INS-1 cells incubated in 16.7 mM glucose significantly repressed the Z, X, and Y
minienhancer activities. To analyze whether glucose-induced repression of Z, X, and Y
minienhancer promoter activities correlate with a reduction in DNA binding activity, we
tested nuclear extracts of INS-l cells incubated in 4.0 or 16.7 mM glucose for the ability
to bind labeled. double-stranded oligonucleotides within the Z and X minienhancers.
Because the palindrome (-284 to —279) and the E3a-E3b elements (-273 to —258) are
potential regulatory elements within the Z minienhancer, we analyzed DNA/nuclear
extract binding activity with a probe termed Zd (~289 to —-252) (Fig. 20). As shown in
Figure 23, four specific DNA/protein complexes (Zdl-4) were observed in nuclear
extracts of INS-1 cells incubated in either 4.0 or 16.7 mM glucose for 48 hrs. The fastest
migrating complex (Zdl) consistently showed reduced binding activity in cells cultured
in 16.7 mM glucose (Fig. 23, compare lanes 2 and 3, and Fig. 24A). Binding activities of
all four complexes were specific to the Zd probe because they were completely competed
with IOO-fold molar excess of the unlabeled Zd probe (Fig. 23, compare lane 2 and 3 with
4 and 5). The palindrome sequence alone was not responsible for Zdl-4 complexes
because competition with the excess unlabeled palindrome probe (Zpal) did not affect

binding (Fig. 23A, lanes 6 and 7). Competition analysis showed that the slowest

124

A B
Competition: Zd Zpal E3 28 Far-box E3 mE3

Glucose(mM): 4 I67 4 16.7 4 16.7 4 16.7 4 16.7 4 [6.7 4 l6.7 4 16.7 4 l6.7
t5'”' . . -. 7.1” 0.». . . ' ‘

.--~. , p-A1~ a)‘-‘ .1 . - ‘a.

mi

 

Zd4

    

Zd3 h i: i ’
Zdz .9. . n .2: . ., . .. a.
Zdl L4 um ind ha

 

 

Figure 23. Glucose-sensitive complex binds to the palindrome sequence and the E3
element. Equal amounts of nuclear extracts from INS-1 cells incubated in media
containing 4.0 or 16.7 mM glucose were analyzed for binding to a 32P-labeled Zd probe
(A) or E3 probe (B) (see Table 3). (A) Competition with 100-fold molar excess of
unlabeled Zd probe (wt) (lanes 4 and 5), palindrome probe (lanes 6 and 7), E3 probe
(lanes 8 and 9), Za probe (lanes 10 and 11) and Far-box probe (lanes 12 and 13). (B)
Competition with 100-fold molar excess of unlabeled E3 probe (wt) (lanes 4 and 5) and
mE3 probe (lanes 6 and 7). Shown is a representative experiment of 3 three independent

experiments.

125

 

>
u—k A
'o N
l

.0
on
1
‘4»

relative binding activity
0
'0)

.0
b
l

.0
N
1

 

 

.0
o

 

 

ea
3

A
O
1

p
(D
1

_{

relative binding activity
_0 o
h» 0)

.0
N
1

 

 

0.0

 

4.0 16.7
glucose (mM)

Figure 24. High levels of glucose reduces binding activity of the Zdl and E3 probes.
Fig. 24A and 24B show the binding activity of the slowest migrating complex that bind to
the Zd and E3 probes, respectively. Values represent mean :t SE of three independent
experiments (n=3). Asterisk indicates that 16.7 mM glucose significantly decreases

binding activity compared to 4.0 mM glucose (p < 0.03).

126

migrating complex (Zd4) was effectively competed for by the la probe, which contains
the palindrome and the E3a elements (Fig. 23A, lanes 10 and 11). These data suggest
that either the E3a element alone or the E3a element and palindrome sequence are
required for Zd4 binding. Competition analysis also showed that the E3 probe, which
contains the E3a and E3b elements effectively competed for Zdl-4 binding complexes
(Fig. 23A, lanes 8 and 9). To test whether the E12/E47 and Beta2 transcription factors
are contained in the Zdl-4 complexes, the Zd probe was competed for binding with an E-
like element fiom the rat I promoter, the rat FAR-box. The Zdl-4 complexes were not
competed for by the rat I FAR-box probe (Fig. 23A, lanes 12 and 13), indicating that the
E12/E47 and Beta2 transcription factors do not bind to these sequences. Interestingly,
protein binding activity to a labeled E3 element (E3a and E3b elements), termed 253.1531,
probe, showed similar pattern of DNA/protein complexes to those observed for the'Zd
probe (Fig. 23B, lanes 2 and 3 compared to Fig. 23A, lanes 2 and 3). High levels of
glucose reduced binding activity of the fastest migrating complex (Fig. 23B, lanes 2 and
3 and Fig. 24B). The DNA/protein binding activities were specific to the ZE3aE3b probe
because the 100-fold molar excess of unlabeled ZE3aE3b competed efficiently with all
complexes (Fig 233, lanes 4 and 5). When the Zg3ag3b probe was competed with a probe
containing mutations on E3a and E3b elements, the Zdl-4 complexes were not competed
(Fig. 23B, lanes 6 and 7). These data strongly suggest that the DNA binding complexes
of the Zd probe require the E3a and E3b elements, and there is a trend that high levels of

glucose decrease Zdl binding activity.

High concentrations of glucose also reduced the Za minienhancer activity

(Fig. 22B). Since the la probe lacks the E3b element, it was important to determine

127

whether DNA binding activity to the Za probe was regulated by glucose. As shown in
Fig. 25A, three complexes bind to the Za probe (Zal-3). The Za2 complex binding
activity was markedly decreased in nuclear extracts from INS-l cells incubated in 16.7
mM glucose compared to 4.0 mM glucose (Fig 25A, compare lanes 2 and 3 and Fig. 26).
These complexes were specific to Za probe since addition of excess unlabeled Za probe
prevented formation of Zal-3 complexes (Fig. 25A, lanes 4 and 5). Mutation of —271 (T
to G), which changed the third base-pair of the E3a element, did not affect binding of the
three Za complexes (Fig. 25B, lanes 4 and 5). In contrast, a mutation at -273 (C to G),
which changed the first base-pair of the E33 element, led to loss of both Zal and Za2
complexes (Fig 25B, lanes 6 and 7). More over, a mutation at both —283 (C to A) and -
282 (T to G), both contained in the palindrome, led to a complete loss in binding activity
for all three Za complexes (Fig. 25B, lanes 8 and 9). Overall, these data suggest that the
Za2 complex, which is a glucose-sensitive complex, requires both the palindrome and the
E3a element for binding. In addition, these data demonstrated that separation of the
palindrome sequence and the E3a element from the E3b element led to different
DNA/protein complexes than the complex bound to the three elements together. These
data suggest that there could be some synergic interactions among the palindrome, E3a,
and E3b elements for binding transcription factors. Similar observations have been made

by Sander et al. (89).

Experiments performed in this study showed that high levels of glucose
repressed the X minienhancer activity (Fig. 21). Therefore DNA/binding activity to

elements within the X minienhancer was examined. The Xa probe (-323 to -—288), which

128

Competition: Za

Glucose(mM): 4 16.7 4 16.7
A

233 —>‘

Za2 ‘> 3:2; . ,
£16444
Zal _>‘ | u w 4 he

Probe:

 

Za3 —’

4 16.7 4 16.7 4 16.7 4 16.7

Za2—*"

Zal —’

 

 

Za m-27l m-273 m-282/3

Figure 25. Glucose-sensitive complex binds to the la element region. Equal amounts

of nuclear extracts from INS-l cells incubated in media containing 4.0 or 16.7 mM

glucose were analyzed for binding to a 32P-labeled Za probe (A), and to a 32P-labeled m-

271, m-273 or m-282/3 (B). See Table 3 for sequences of the probes. Panel A, lanes 4

and 5 contained 100 X-fold unlabeled Za probe. Shown is a representative experiment of

4 independent experiments.

129

 

1.2

 

 

 

 

1.0-l *
.E‘
.2 0.8-
‘6
(U
D
.2 . *
E 0.6- ’~4 « ,
Q .
g _.T,
E o.4~   -.
'2

0.24

0.0 u .

 

 

 

 

 

4.0 16.7

glucose (mM)

Figure 26. High levels of glucose reduce binding activity of the Zal complex. High
levels of glucose reduce binding activity of the slowest migrating complex of the la
probe. Data shown are the mean i SE of four independent experiments (n=4). Asterisk
indicates that 16.7 mM glucose significantly decreases binding activity compared to 4.0

mM glucose (p < 0.0002).

130

contains the A5/Core (—317 to —309) and the E4 element (-300 to —294) (Fig. 20), was
labeled and analyzed for DNA/protein binding activity with nuclear extracts from INS-1
cells incubated in 4.0 or 16.7 mM glucose for 48 hrs. As shown in Figure 27, multiple
DNA/protein complexes bound to the Xa probe. Importantly, a slow migrating complex,
Xal, was markedly decreased in nuclear extracts derived from cells cultured in 16.7 mM
glucose (Fig. 27, compared lanes 2 and 3). Addition of 100-fold molar excess of
unlabeled Xa probe competed with Xal binding, demonstrating that the binding of this

complex was specific to the Xa probe.

To determine whether the glucose-sensitive complex, Xal, requires the
A5/Core and/ or E4 element contained in the Xa probe, nuclear factor binding activity to
a probe that contained only the AS/Core sequence, termed the Xb probe (-323 to —303),
was examined (Fig. 20). As shown in Figure 28, only a single slow-migrating complex
formed with the Xb probe. Importantly, this complex was markedly lower in nuclear
extracts from cells incubated in 16.7 mM glucose (Fig. 28, lanes 2 and 3). This complex
was readily competed by excess unlabeled Xb probe (Fig. 28, lanes 4 and 5), indicating
the specificity of this complex. A mutant AS/Core probe, where the third base-pair of A5
and two overlapping sequences of A5/Core were mutated (ATG to CGT), did not
compete for binding (Fig. 28, lanes 6 and 7), suggesting that this glucose-sensitive
complex binds directly to A5 and Core elements. Since the A5 element is similar in
sequence to a de-l binding site (TAAT), and de-l binding activity is regulated by
glucose, it was necessary to examine whether de-l was part of this complex.

Competition analysis with 100-fold molar excess of unlabeled A1 and A3 probes, which

131

A B
Competition: Xa
Glucose: .-.4....i.617 4 16.7

Xal —>Q

 

“Iii.

  

1.2
1.0 -
0.8 -
0.6 -

 

 

0.4 -
0.2 <

relative binding activity

 

 

0.0

 

4.0 V 416.7'
glucose (mM)

 

Figure 27. Glucose-sensitive complex binds to the Xa element region. A. Equal
amounts of nuclear extracts from INS-1 cells incubated in media containing 4.0 or 16.7
mM glucose were analyzed for binding to a 32P-labeled Xa probe. Lanes 4 and 5 contain
100-fold molar excess unlabeled Xa probe. Shown is a representative experiment of 4
independent experiments. B. Summary of the Xal binding activity of 4 independent
experiments with values of the mean :t SE (n=4). Asterisk indicates that 16.7 mM

glucose significantly decreases binding activity compared to 4.0 mM glucose (p < 0.004).

132

A B
Competition: Wt DIAS/Core

Glucose (mM): 4. 16.7 4 16.7 4 16.7

-‘ a
i I , .r I ' .4
Z ‘I' 7' -I "If-J ‘ . iitfi
_ .1 -4 , e -';"1
‘ 4 ' 'i - , rat:
.~ .g *g- o

 

 

 

0...;
(DON

I

9.0
50)

relative binding activ

.0
N

 

 

40 16.7
glucose (mM)

 

.9
o

 

Figure 28. Glucose-sensitive complex binds to the A5/Core sequence (Xb probe). A.
Equal amounts of nuclear extracts from INS-1 cells incubated in media containing 4.0 or
16.7 mM glucose were analyzed for binding to a 32P-labeled Xb probe (AS/Core). Lanes
4 and 5 contain 100 -fold unlabeled Xb probe. Lanes 6 and 7 contain 100 -fold molar
excess unlabeled mutated Xb probe (mAS/CORE). B. Summary of binding activity of
three independent experiments with values of the mean i SE (n=3). Asterisk indicates
that 16.7 mM glucose significantly decreases binding activity compared to 4.0 mM

glucose (p < 0.002).

133

have de-l binding sites, did not diminish binding of the glucose-sensitive complex
(Fig. 29, lanes 6, 7 and 8, 9, respectively). Furthermore, addition of a de-l antibody did
not supershift or disrupt the glucose-sensitive complex (Fig. 29, lanes 10 and 11). It was
also necessary to determine whether ka6.1 or Pax6 formed part of the glucose-sensitive
complex, since both transcription factors bind to A-like elements (76, 292). Addition of a
ka6.1 or Fax-6 antibody did not supershifi or disrupt binding of the glucose-sensitive
complex (Fig. 29, lanes 12 and 13, data not shown for Pax-6). A positive control that
demonstrates that high levels of glucose reduce binding activity of de-l to the A1
element, and that an antibody that recognizes de-l supershifi de-l are shown in Figure
30. High levels of glucose reduce A1 binding activity (Fig. 30 comparing lanes 2 and 3).
de-l antibody recognizes de-l within this complex (Fig. 30 comparing lanes 10 and

11).

Because the Xa element contains the A5/Core and the E4 element, it was
necessary to determine whether there was any glucose-regulated binding activity to the
E4 element (-300 to —294). Nuclear extracts from INS-1 cells incubated in 4.0 or 16.7
mM glucose were analyzed for the ability to bind to the E4 probe (-306 to -288) (Fig.
20). A single-complex bound the E4 probe and its binding activity was not decreased in
nuclear extracts from cells incubated in 16.7 mM glucose (Fig. 31, lanes 2 and 3). This
complex was specific for the E4 probe because, a 100-fold molar excess of unlabeled E4

probe effectively competed for the complex (Fig. 31, lanes 4 and 5). '

134

Competition: Xb Al A3 Ab-del Ab-ka6.l
Glucose(mM): 4 16:7 4 16.7 4 16.7 4 16.7 4 16.7 4 16.7

P..‘.1.

 

 

9 10' "1i 1'2’ 13‘;

Figure 29. Neither de1 nor ka6.l form part of the complex that binds to the
A5/Core element. Equal amounts of nuclear extracts from INS-1 cells incubated in
media containing 4.0 or 16.7 mM glucose were analyzed for binding to a 32P-labeled Xb
probe (AS/Core). Competition with IOO-fold molar excess of unlabeled Xb probe (wt)
(lanes 4 and 5), A1 probe (lanes 6 and 7), A2 probe (lanes 8 and 9), and the addition of
1p] anti-de-l antibody (lanes 10 and 11) or the 1 pl anti-ka6.l antibody (lanes 12 and
13). Shown is a representative experiment of three independent experiments where as the

super-shift experiments were only performed once.

135

Competition: A1 mlAl m2A1 Ab-del
Glucose(mM): 4 16.7 4 16.7 4 16.7 4 16.7 4 16.7

 

 

Figure 30. High levels of glucose reduce binding activity to the A1 element. Equal
amounts of nuclear extracts from INS-l cells cultured in 4.0 or 16.7 mM glucose were
analyzed for binding to a 32P-labeled A1 probe. Lanes 4 and 5 contain lOO-fold molar
unlabeled A1 probe. Lanes 6-7, and 8-9 contain 100-fold molar excess unlabeled
mutated A1 probe, mlAl and m2A1, respectively. Shown is a representative experiment

of 2 independent experiments.

136

A B

Competition: E4 E2
Glucose(mM): 4 16.74 16.74 16.7

 

 

_’~‘ ' 1.8

1.6< I NS
1.4<

1.2<

 

0.8 «
0.6-
0.4 «

relative binding activity

 

 

 

0.0 . .
m 2,, m, ~, g, 4.0 16.7
glucose(mM)

’1 ‘2 3 4 5 6 7‘
Figure 31. Glucose does not decrease binding activity of the complex that binds to
the E4 element. A. Equal amounts of nuclear extracts from INS-1 cells incubated in
media containing 4.0 or 16.7 mM glucose were analyzed for binding to a 32P-labeled E4
probe. Competition analysis with 100-fold molar excess of unlabeled E4 probe (wt)
(lanes 4 and 5) or FAR-box (rat E2 element) (lanes 6 and 7). B. Summary of binding
activity of three independent experiments with values of the mean i SE (p < 0.17). NS,

not significantly different.

137

An unlabeled Far-box probe also competed for this complex (Fig. 31, lanes 6 and 7),
indicating that the El2/E47/Beta2 heterodimer possibly binds to this site. Overall, these
data suggest that there are three insulin promoter elements, A5/Core, palindrome, and E3

element, through which high levels of glucose might mediate insulin promoter repression.

3.6 Mutation of the A5/Core, the palindrome, and the E3 elements partially

diminished glucose-induced insulin promoter repression

To determine whether the A5/Core, the palindrome and/ or the E3 elements are involved
in mediating glucose-induced insulin promoter repression, site-specific mutations of all
these elements were made within the full length of insulin promoter, INS(-327)CAT.
The INS(-327)CAT vector containing mutations 'in the palindrome, in both the
palindrome and the A5/Core, or in both the palindrome and the E3 element were termed
mPalINS(-327)CAT, ‘ mPal/AS/CoreINS(-327)CAT, and mPal/E31NS(-327)CAT,
respectively. Mutations of all three elements, A5/Core, palindrome, and E3 within the
INS(-327)CAT vector was termed mA5/Core/Pal/E3H‘IS(-327)CAT. All mutated
sequences of these elements are shown in Table 3. As expected, INS(-327)CAT
expression was repressed 86.6 % in INS-1 cells cultured in 16.7 mM glucose compared to
4.0 mM glucose (Table 4). Mutations of the palindrome sequence as well as mutations of
both the palindrome and the AS/Core element did not prevent glucose-induced repression
of the insulin promoter activity. The mPalINS(-327)CAT and the mAS/Core/Pal INS(-

327)CAT expressions were repressed 80.0 % and 74.8 %, respectively. The repression of

138

 

4.0 mM glucose 16.7 mM glucose % repression

 

Name CAT activity CAT activity CAT activity
INS(-327)CAT 127.2 3: 6.2 17.0 i 2.5 86.6 *
mPal 1NS(-327)CAT 210.8 1 9.5 42.8 i 4.2 80.0 *
mAS/COFC/Pal 1NS(-327)CAT 182.2 2‘. 27.3 45.9 :t 9.4 74.8 *
mPal/E3 1NS(-327)CAT 96.2 :t 7.4 34.5 i 2.7 64.2 *
mA5/Core/Pal/E3 INS(-327)CAT 37.7 :1: 6.8 20.1 i 4.0 46.6 *

 

Table 4. Mutations of all three ASICore, palindrome, and E3 elements partially
diminished glucose-induced repression of the INS(-327)CAT. INS-1 cells were
transfected with insulin promoter vectors. The cells were then cultured in 4.0 or 16.7 mM
glucose for 48 hrs. mPal represents mutated palindrome sequence. mA5/Core/Pal
represents mutated AS/Core and palindrome sequences. mPal/E3 represents mutated
palindrome and E3 sequences. mAS/Core/Pa1/E3 represents mutated A5/Core,
palindrome, and E3 sequences. Values are the means 1- SE of three individual
experiments (n=3). Asterisk indicates that values at 16.7 mM glucose were significantly
lower than at 4.0 mM glucose of given vector (p < 0.04). Two experiments were

performed by Diana Ye, Department of Pharmacology, Michigan State University.

139

the insulin promoter activity by high glucose concentrations was slightly prevented when
both the palindrome and E3 elements were mutated. The mPal/E3INS(-327)CAT
expression was repressed by 64.2 % in cells cultured in 16.7 mM glucose compared to
4.0 mM glucose. Importantly, mutation of all three elements, mA5/Core/Pal/E3 INS(-
327)CAT expression was only 46.6 % repressed in cells cultured in 16.7 mM, compared
to 4.0 mM glucose. These results strongly suggest that the palindrome, AS/Core, and E3
elements mediate part of glucose-induced insulin promoter suppression, and that Al, A3,

and C1 elements might mediate the rest of the suppression.

4. Discussion

In vitro and in vivo models of hyperglycemia have shown that insulin gene
expression is suppressed when B-cells are chronically exposed to high glucose
concentrations (110, 118). Reduced expression and/ or binding activity of transcription
factors including de-l and C1 activator (2, 3, 113, 121, 126) and increased expression of
C/EBPB and c-Myc (85, 274) have been suggested to mediate some of the glucose-
induced changes in promoter activity. To gain a more in-depth understanding of the role
of de-l and C1 activator on these events we examined a more simplified insulin
promoter vector that contains sequences from —230 to + 30, termed lNS(-230)CAT. High
levels of glucose led to only a minor repression of INS(-230)CAT expression compared
to the large repression observed on the full length promoter INS(-327)CAT. Mutations of
the de-l and! or C1 activator binding sites abolished the suppression found in the IN S(-

230)CAT expression. These results indicate that de-l and C1 activator are involved,

140

but have a minor role in glucose-induced repression of the INS(-327)CAT expression,
and suggest that additional mechanisms mediate insulin promoter repression. The
participation of additional mechanisms are also illustrated in INS—1 cells transfected with
a de-l expression vector, because de-ldoes not prevent decreased insulin promoter
activity in cells cultured in high levels of glucose (L.K. Olson, unpublished data).
Studies in HIT-T15 cells, a Syrian hamster insulinoma, also demonstrated that over-
expression of de-l alone is not sufficient for full recover of promoter activity (122).
Although loss of C1 activator binding likely plays an important role in the
downregulation of insulin promoter activity, its exact role has not been determined since
it has only recently been cloned. Overall, these studies suggest that there are other
mechanisms by which exposure of B-cells to high levels of glucose represses insulin
promoter activity and that these mechanisms may mediate repression through sequences

upstream of the —230 insulin promoter.

Deletion of human insulin promoter sequences from ~327 to —279 reduced insulin
promoter repression by glucose to similar levels observed by deleting sequences fiom —
327 to —230. Deleting sequences upstream of —279 removes the A5/Core element, the E4
element, and disrupts the palindrome sequence by dividing it in half. It has been shown
that palindrome sequences can bind transcription factors such as glucocorticoid receptor
(GC) and estrogen receptor (293), suggesting that similar transcription factors can bind to
the insulin palindrome sequence. In fact, Goodman et al. (90) demonstrated that two
complexes from nuclear extracts of two insulinoma cells lines, HIT-T15 and RIN-mSF,
bind to the human insulin NRE and that a consensus sequence GC oligonucleotide

efficiently competes for binding of these complexes. Interestingly, the insulin

141

palindrome sequence is adjacent to the E3 element, and Sander et al. (89) have proposed
that transcription factors that bind to these adjacent sites might interact and that their
binding activity might be dependent upon one another. Therefore, disruption of the
palindrome may disrupt binding to the E3 element. If this is true, deletion of sequences
upstream of —279, which disrupts the palindrome, could be predicted to have similar
effects as deleting sequences upstream —261 that removes the E3 element. This
hypothesis is in agreement with our deletion studies that showed that deleting upstream
sequences from —279 had similar glucose-induced promoter repression as the reporter
gene with upstream sequences of —261 deleted. Overall, these results demonstrate that
high levels of glucose suppress insulin promoter through sequences from —327 to —261.
The A5/Core, E4, palindrome, and E3 elements are the most likely sites to mediate

glucose-induced repression of the insulin promoter.

The Z minienhancer, which contains the palindrome sequence and the E3
element, was markedly repressed in INS-1 cells cultured in 16.7 mM glucose, compared
to 4.0 mM glucose. In contrast to our results, Sander et al. (89) demonstrated that the Z
minienhancer was a potent transcriptional enhancer in primary fetal islet cells cultured in
16.0 mM glucose compared to 2.0 ran glucose. The inconsistency with Sander et al.
(89) results of the Z minienhancer in fetal islets and our results in INS-l cells might be
explained by the biphasic glucose regulation of this minienhancer. Sander et al. (89) only
cultured fetal islet cells in 2.0 or 16.0 mM glucose, potentially missing the large glucose
induction of the Z minienhancer activity that occurred in INS-l cells cultured in 2.0 to
6.0 mM glucose. Therefore, the Z minienhancer could have shown repression in fetal

islet cells incubated in 16.7 mM glucose compared to 4.0 or 6.0 mM glucose as was

142

observed in INS-1 cells. In fact, if we just compare the Z minienhancer activity in INS-1
cells incubated in 2 mM vs. 16.7 mM glucose we could have concluded that the Z
element was induced by glucose as observed in fetal islets. Our studies showed that most
of the suppression of the Z minienhancer activity occurred at glucose concentrations
between 6.0 to 16.7 mM glucose and these concentrations were not examined in fetal islet

cells.

Sander et al. (89) reported that the Z minienhancer activity is repressed compared
to controls (pFOXCAT) in two B-cell lines transformed by Simian virus large T-antigen.
The authors suggested that the Z minienhancer activity is repressed in these transformed
cell lines because they have a high proliferation rate, in contrast to differentiated fetal
islets, which do not divide rapidly in culture. In contrast, we found that INS-l cells,
which are transformed by gamma irradiation, express the Z minienhancer at high levels
compared to the pFOXCAT control. Our results suggest that rapid proliferation does not
block the Z minienhancer activity, but that T-antigen transformation may block Z
minienhancer activity. Nevertheless, INS-1 cells do show enhanced proliferation when
cultured at high levels of glucose concentrations, but this is unlikely to account for
changes in promoter activity because G1 arrested INS-1 cells still had decreased insulin
mRNA expression when cultured in elevated glucose (126). Consistently with this idea,
Clark et al. (88) reported binding of a possible repressor, Oct-1, to the NRE that is within
the Z minienhancer in HIT-T15 cells. Further experiments need to be performed to

elucidate differences between Z minienhancer activity in different tumor B-cell lines.

143

Mobility-shift assays performed with nuclear extracts from INS-1 cells
demonstrated a binding complex to the palindrome and E33 element, termed the Za2
complex, and its binding activity is decreased from 4.0 to 16.7 mM glucose. In contrast,
nuclear extracts from fetal islet cells revealed a binding complex to the same region that
its binding activity is increased from 2.0 to 16.7 mM glucose (89). The difference
between the binding activities of these complexes could be explained by the biphasic
glucose regulation of the Z minienhancer activity. In INS-l cells, Z minienhancer
activity markedly increases from 2.0 to 6.0 mM glucose, and slightly increases from 2.0
to 16.7 mM glucose, suggesting that binding activity to the palindrome and E3a element
most likely increases as observed in nuclear extracts from fetal islet. Additional
experiments need to be performed to elucidate Z element binding activity of nuclear

extracts from INS-l cells and fetal islets.

Mobility-shift assays performed with nuclear extracts from INS-1 cells also
demonstrated that there are two different glucose-sensitive complexes that bind to the Z
element: one complex binds to the palindrome and E3a-E3b elements, and a different
complex binds to the palindrome and only the E3a element. These results indicate that
removing the E3b element from the palindrome-E3a-E3b sequences shows different
complexes, suggesting that the palindrome and the E3 element work in a synergistic
manner, as suggested by Sander et al. (89). As discuss above, these results also agree
with our truncation analysis that showed that deleting sequences upstream from -—279,
which disrupts the palindrome, had the same glucose-induced repression as deletion of
sequences upstream from —269. Overall, these results highly suggest that transcription

factors that bind to the palindrome, E3a, and E3b elements interact as has been shown for

144

other transcription factors that regulate insulin promoter activity. For example, E47/E12,
which binds to the E1 element, interacts with de-l that binds to Al element (64), and

C1 activator that binds to the C1 element (289).

Activity of the X minienhancer, which contains the AS/Core and the E4 element,
also showed a biphasic glucose regulation. Mobility-shift assays demonstrated a glucose-
sensitive complex that binds to the A5/Core element, we termed Xal. Interestingly, this
complex appears to be distinct from other known transcription factors that bind to A-like
elements within insulin promoter. These results suggest a novel transcription factor that
binds to the A5 element and whose binding activity is diminished by high levels of
glucose. In contrast to the A5/Core, palindrome, and E3 elements, binding activity to the
E4 element showed a glucose-stimulated binding complex. Interestingly, this E4 element
behaves like the E2 element in that glucose enhances its binding activity (88, 123). In
addition, the E2 element (Far-box element) fiom the rat I insulin promoter competed for
the complex bound to E4 element. E12/E47 and Beta2 transcription factors bind to E
elements, but super-shift mobility assays demonstrated that Beta2 did not bind to E4
element (data not shown). These results indicate that transcription factors that recognize

E2 rat element may bind and regulate the E4 element.

DNA binding activity analyses of sequences from —327 to —242 of insulin
promoter demonstrated three major glucose-sensitive elements, the A5/Core, the
palindrome, and the E3. Site-specific mutations of all these three elements in the —327
insulin promoter markedly diminished the ability of glucose to repress the insulin
promoter activity. These results demonstrate that, in fact, glucose mediates insulin

promoter repression, in part, through the A5/Core, palindrome, and E3 elements. These

145

results indicate that additional mechanism besides changes in de-l, C1 activator, c-Myc,

and C/EBPB mediate glucose repression of insulin promoter.

Indeed, recently experiments have demonstrated that hyperglycemia is causing a
more global change in gene expression and phenotype. For example, it has been
demonstrated that hyperglycemia impairs glucose-induced insulin secretion by altering
gene expression involved in glucose metabolism and insulin secretion. In Zucker
diabetic fatty rats and 90 % pancreatectonrized rats, genes involved in glucose
metabolism including glucose transporter Glut2, glucokinase, mitocondrial glycerol-3-
phosphate dehydrogenase, and pyruvate carboxylase decreased progressively as
hyperglycemia increased (100, 101). The same phenomenon was observed in genes
involved in insulin secretion such as potassium channel Kir6.2 and voltage dependent
Ca2+ channel (4, 101, 113). In 90 % pancreatectomized rats there is also an increase in
glucose-6-phosphatase, hexokinase I, and lactate dehydrogenase A (LDH) gene
expression, which normally are expressed at low levels in islets (4). The implications of
the changes of these genes lead to reduced glucose-induce insulin secretion. The changes
in other genes are not clearly understood, but one can postulate that increased LDH
expression, which produces lactate from pyruvate by using NADH, depletes NADH

levels, which are required for mitochondrial ATP production.

Expression of important genes involved in B-cell differentiation is also altered by
chronic hyperglycemia. In the 90 % pancreatectomy rat diabetic model, gene expression
of hepatic nuclear factor (HNF) —3B, 4a, 10., paired-box homeodomain-6 (Pax-6), NK-

homeodomain factor ka6.1, Beta2, and de-l were gradually decreased with increasing

146

 

levels of hyperglycemia (4). One can postulate that these global alterations of gene
expression may contribute not only to decrease glucose-stimulated insulin secretion but

also to the degeneration of B-cells that eventually leads to B-cell apoptosis.

Chronic hyperglycemia induces B-cell dysfunction by altering a wide-ranging of
gene expression involved in glucose metabolism, glucose-induced insulin secretion, and
B-cell differentiation. Because of this broad effect of hyperglycemia in B-cell function, it
is not surprising that hyperglycemia mediates insulin promoter repression through several
promoter elements. The mechanisms by which high levels of glucose induce such

diverse changes in B-cell function are poorly understood.

147

V]. CONCLUSIONS AND FUTURE STUDIES

Pancreatic B-cells control blood glucose levels in a narrow range by responding to
even small changes of blood glucose, which is the major secretagogue, and secrete the
right amount of insulin into circulation. Failure of B-cells to secrete insulin leads to the
development of diabetes. Type II diabetes is characterized by a combination of insulin
resistance and altered glucose-induced insulin secretion (294). As a compensatory
mechanism for insulin resistance, B-cell hypertrophy and changes in expression of key
glucose metabolism enzymes are observed that lead to an increase in insulin secretion.
Hypertrophy is found in 90 % pancreatectomized rats (4), in Zucker diabetic fatty rats
(188), and after 96 hr of glucose infusion (295). Hypertrophy and hyperinsulinimea have
been associated with increased c-Myc and hexokinase I expression, respectively (4).

Failure of the B-cell to compensate for insulin resistance because of inability to
fiirther increase B-cell mass leads to increased blood glucose levels and this has been
associated with B-cell dysfunction. An important characteristic of B-cell dysfunction is
decreased glucose-induced insulin secretion, which further exacerbates the diabetic
condition. Decreased glucose stimulation of insulin secretion has been extensively
studied in the last decade and is correlated with alterations in genes involved in glucose
metabolism and insulin secretion. Recent studies in the 90 % pancreatectomized rat
model demonstrated downregulation of a variety of genes involved in glucose
metabolism such as insulin, Glut2, gluocokinase, mitochondrial glycerol phosphate

dehydrogenase, and pyruvate carboxylase (4). Genes involved in insulin release are also

148

downregulated such as the potassium channel Kir6.2, the voltage-dependent calcium
channel a 1D, and calcium ATPase channel (SERCAB) (4).

The pathogenesis of glucose-induced insulin secretion during diabetes involves a
variety of genes that lead to downregulation of insulin gene expression. Decreased
binding activity and/ or expression of de-l and C1 activator (3, 126), (113) are
associated with downregulation of insulin gene expression. This thesis demonstrated that
additional mechanisms are involved in glucose-induced repression of insulin promoter
activity. Functional analysis and mobility-shift assays showed that the A5/Core,
palindrome, and E3 elements are also involved in the insulin promoter repression. Some
of the mechanisms involved in the repression of insulin promoter activity are correlated
with increased c-Myc and C/EBPB expression (86, 274). In addition, we demonstrated
that hyperglycemia might repress insulin promoter activity through the JNK pathway. In
line with our results, Kaneto et al. (8) showed that inhibition of JNK by over-expressing a
dominant negative form of JNK in H202 treated isolated rat islets prevented repression of
insulin mRNA and recovered de-l binding activity. Since high levels of glucose
increase ROS, it is likely that elevated ROS levels by hyperglycemia activates JNK and
subsequently represses insulin promoter activity.

Another potential mechanism by which high levels of glucose might repress
insulin promoter activity is through chronic exposure of B-cell to insulin. As mentioned
above, during the development of diabetes high levels of insulin are secreted to
compensate for insulin resistance. Insulin is known to induce insulin gene expression
(197) and one can postulate that chronic exposure of B-cells to insulin might repress

insulin gene expression. In our experiments, INS-1 cells are cultured in 11.1 mM glucose

149

prior to treatment with 16.7 mM glucose for 48 hrs, which induces insulin secretion.
Therefore, in addition to chronic exposure of cells to high levels of glucose, cells are also
chronically exposed to insulin, which might contribute to the reduction of insulin
promoter activity.

Repression of glucose-induced insulin secretion by hyperglycemia is also
accompanied by changes of important genes involved in B-cell differentiation. In the 90
% pancreatectomized rat model a decrease in de-l, ka6.1, Pax6, Beta2, HNF 1a,
HNF4a, and HNFBB gene expression is observed (4). Recently, it was demonstrated that
expression of genes that are normally suppressed in the B-cell are increased in correlation
with increasing glucose levels (131). Increased expression of lactate dehydrogenase-l
(LDHA), glucose-6-phosphatase, fiuctose-1,6-bisphosphatase, peroxisome proliferator-
activated receptor (PPAR 7), and uncoupling protein 2 (UCP-2) were observed in the 90
% pancreatectomized rat model (131). The role of the alteration of all these gene’s
expressions is not understood yet and it is not known if changes in mRNA levels are
correlated with changes in protein activities. The important point is that there is a global
change in gene expression that can be involved in B-cell adaptation to hyperglycemia or
B-cell decompensation due to hyperglycemia.

Hyperlipidemia has been also associated with detrimental effects on B-cell
function (296). Chronic exposure of B-cell to free fatty acids (FA) increases basal insulin
release but decreases glucose-induced insulin secretion (reviewed in Ref. 294).
Hyperlipidemia also alters glucose-induced insulin secretion in the presence of high
levels of glucose, which has been correlated with decreased insulin gene expression and

furthermore with decreased de-l expression (133, 134, 295). Recently, in vivo studies

150

demonstrated that hyperglycemia is required for hyperlipidemia to alter insulin gene
expression (139). This was demonstrated in the Zucker diabetic fatty rat, which contains
high content of triglycerides (TG) in both blood and islets. When the rats were treated
with a reagent, phlorizin, that lowers blood glucose levels and islet TG content without
affecting blood TG content, clearly shows that hyperglycemia is associated with
increased TG islet content and decreased insulin mRNA levels (139). The role of
hyperlipidemia in the repression of glucose-induced insulin secretion is not clear since
the 90 % pancreatectomized rat model does not have high levels of TG and there is
repression of insulin secretion. Nevertheless, the possibility can not be ruled out that
chronic exposure of B-cells to TG contributes to decrease glucose-induced insulin

secretion when the diabetic state is more severe.

The biochemical mechanisms by which hyperglycemia alters B-cell function are
not well understood. One potential hypothesis is that there is a generalized mechanism,
which activates other pathways that lead to the global changes in gene expression. A
likely mechanism is reactive oxygen species since hyperglycemia can generate reactive
oxygen species by several mechanisms including glucose autoxidation, glycation,
advanced glycation end-products, and induction of the polyol pathway (reviewed in Ref.
8). In a non-hyperglycemic state, superoxide molecules are produced by mitochondria
during the process of oxidative phosphorylation. During hyperglycemia there is an
increased in superoxide levels, which can inhibit glyceraldehyde-3-phosphate
dehydrogenase (GAPDH) (299) that subsequently leads to accumulation of glycolytic
metabolites such as glyceraldehydeo3-phosphate (GAP) and fi'uctose-6-phosphate (F 6?).

GAPDH can lead to de novo synthesis of diacylglycerol (DG), which leads to activation

151

of protein kinase C that is known to be activated in diabetic conditions (177).
Accumulation of F6P leads to the activation of the hexosamine pathway, which increases
modification of proteins by O-linked N-acetylglucosamine (300). The AGE pathway can
also be increased by GAP (299). One can hypothesize that inactivation of superoxide
production by the michondria could inhibit activation of PKC, hexosamine, polyol, and
AGE pathways, that subsequently meliorate detrimental effects in diabetes. In fact,
Nishikawa et al. (301) demonstrated that inhibition of electron transport in the
mitochondria prevents activation of PKC, formation of AGE, and sorbitol accumulation
in vascular endothial cells exposed to 30 mM glucose. These results indicate that there is
a generalize mechanism, production of ROS, that leads to the activation of several
pathways involved in the pathogenesis of diabetes.

The biochemical mechanisms by which hyperglycemia alters glucose-induced
insulin secretion involves, at least in part, generation of chronic oxidative stress (7, 157,
158, 170). Interestingly, it has been proposed that hyperlipidemia induces B-cell
dysfunction by inducing oxidative stress (141). One can postulate that hyperglycemia
and hyperlipidemia intercommunicate to induce B-cell dysfunction mediated by
generation of oxidative stress. Recently, it was shown that in the 90 %
pancreatectomized rat model there is an increase of antioxidant gene expression such as
heme oxygenase-1 and glutathione peroxide (173). These results indicate that pancreatic
islets induce a protective mechanism against the production of ROS during

hyperglycemia.

152

Future studies

We demonstrated that INK activity increased in INS-1 cells cultured in high
levels of glucose and that over-expression of JNK repressed insulin promoter activity in
cells cultured in low levels of glucose. These results led to the correlation that glucose-
induced insulin promoter repression is mediated by activation of JNK. However, it was
not possible to directly demonstrate that JNK mediated the insulin promoter repression.
Recently, Kaneto et al. (8) demonstrated that isolated islets treated with H202 repress
insulin expression through activation of JNK. Prevention of H202 —induced activation of
INK and repression of insulin gene expression was demonstrated by infecting isolated rat
islets with an adenovirus expressing a dominant negative JNK (DN-JNK). The DN-JNK
is a kinase-inactive JNK where the ATP-binding site is mutated. DN-JNK is
phosphorylated, but it is not able to phosphorylate downstream targets. Therefore, it may
be feasible to prevent glucose-induced activation of JNK in INS-1 cells over-expressing
this DN-JNK. I hypothesize that inhibition of endogenous JNK by over-expressing DN-
JNK may partially prevent repression of insulin gene promoter activity in INS-1 cells
cultured in high levels of glucose.

Recent experiments demonstrated that a generalize pathway, production of
superoxide through the mitochondrial electron transport chain, may be involved in the
complications of diabetes (296, 298). Superoxide production induces activation of PKC,
polyol, hexosamine, and AGE pathways. It would be interesting to test whether
inhibition of the production of superoxide by the mitochondrial electron chain pathway
prevents hyperglycemia-induced insulin promoter activity in INS-1 cells. These

experiments could be achieved by over-expressing a manganese superoxide dismutase

153

gene, which is specific for the superoxides produced by mitochondria. I hypothesize that
over-expression of Mn-superoxide dismutase would prevent repression of insulin
promoter activity in INS-1 cells cultured in high levels of glucose.

Studies have demonstrated that reactive oxygen species are involved in glucose-
induced repression of insulin secretion (7, 158). Tanaka et al. (7) demonstrated that
treatment of HIT-T15 cells with NAC and AG, two antioxidants, prevent repression of
insulin promoter and de-l binding activity by hyperglycemia. Previous experiments
performed in our laboratory demonstrated that INS-1 cells cultured with these two
antioxidants did not prevent insulin promoter repression, in contrast promoter activity
was further repressed. In these experiments, INS-1 cells were treated with higher
concentrations of NAC and AG than used by Tanaka et al. (7). The high concentrations
of NAC and AG could have had non-specific effects and therefore hindered the effects of
ROS on insulin promoter activity. Future experiments investigating the role of ROS in
insulin promoter repression should be designed with a dose concentration response of
NAC and AG on promoter activity. I hypothesize that ROS is involved in glucose-
induced repression of insulin promoter activity, thus NAC and AG should partially
prevent some of this insulin promoter repression.

It was very interesting to discover that high levels of glucose regulated the distal
insulin promoter region. Deletion of sequences from —327 to —292 increased basal
insulin promoter activity, suggesting that there is some basal repression within this small
region. This area contains the A5/Core and E4 element. To our knowledge no one has
demonstrated any transcription factor binding activity in this area. Our studies

demonstrated that proteins that bind to the rat I E2 element recognize the human E4

154

element. Beta2 and E12/47 transcription factors heterodimerize and bind to the E2
elements. Mobility-shifi assays demonstrated that Beta2 does not form part of the
complex that binds to the E4 element. I hypothesize that E12/E47 transcription factors
might bind to the E4 element and possibly other helix-loop-helix transcription factors as
well. Electrophoretic mobility shift assays with an antibody specific for de-l or Pax6
demonstrated that the complex that binds to the A5/Core element does not contain de-l
and Pax6, and the human A1 element does not compete for binding with this complex.
Overall, future experiments can be design to identify transcription factors that bind to the
AS/Core and E4 element. It will be interesting to investigate whether the new
transcription factors that bind to AS/Core element also bind to the other A-elements, and
whether these transcription factors are regulated by glucose.

Hyperglycemia represses insulin promoter activity through several elements
including the palindrome. It has been demonstrated that estrogen and glucocorticoid
receptors can bind to palindromic sequences (293). In fact, glucocorticoid receptors can
bind to the human palindrome sequence (90). Interestingly, glucocorticoids are known to
induce insulin resistance that is usually accompanied with B-cell dysfiinction (302, 303).
HIT-T15 cells and isolated B-cells treated with dexamethasone, a synthetic
glucocorticoid, induces a decrease in insulin secretion and mRNA levels (304).
Treatment of HIT-T15 cells with dexamethasone for 48 to 72 hrs leads repression of a
reporter vector driven by the human palindrome and E3 element, known as the NRE
element (90). It is also interesting that B-cells are the only islet cells that contain
glucocorticoid receptors (305). I hypothesize that the repression of insulin promoter

activity observed during hyperglycemia might be mediated, in part, by glucocorticoids.

155

To test whether glucocorticoids mediate repression of insulin gene expression, diabetic
animal models such as the Zucker diabetic fatty rat can be treated with a glucocorticoid
antagonist like RU-486 from 6 through 12 weeks of age when the animal becomes
diabetic. Then insulin secretion and insulin mRNA can be measured.

Sander et al. (89) showed that high levels of glucose increase binding activity to
the Za element from extract of fetal and adult pancreatic islets. In contrast, nuclear
extracts from INS-1 cells showed that glucose reduced binding activity to the la element.
The difference between these experiments is the basal glucose concentration. It is
important to clarify these results by comparing binding activity to the Z element from
extracts of fetal islet and INS-1 cultured at the same glucose concentrations. I
hypothesize that Z element-binding activity is repressed when both INS-cells and fetal

islets are incubated in 16.7 mM glucose compared to 4.0 or 6.0 mM glucose.

156

10.

VII. LIST OF REFERENCES

Robertson RP, Harmon JS, Tanaka Y, Sacchi G, Tran PO, Gleason CE,
Poitout V 2000 Glucose Toxicity of the Beta cell. In: LeRoith D, Taylor SI,
Olefsky JM (eds) Diabetes Mellitus: A fundamental and clinical text. Lippincott
Williams & Wilkins, Philadelphia, PA:125-131

Sharma A, Olson LK, Robertson RP, Stein R 1995 The reduction of insulin gene
transcription in HIT-T15 beta cells chronically exposed to high glucose
concentration is associated with the loss of RIPE3b1 and STF-l transcription factor
expression. Mol Endocrinol 9:1127-34. .

Olson LK, Sharma A, Peshavaria M, Wright CV, Towle HC, Rodertson RP,
Stein R 1995 Reduction of insulin gene transcription in HIT-T15 beta cells
chronically exposed to a supraphysiologic glucose concentration is associated with
loss of STF-l transcription factor expression. Proc Natl Acad Sci U S A 92:9127—
31.

Jonas JC, Sharma A, Hasenkamp W, Ilkova H, Patane G, Laybutt R, Bonner-
Weir S, Weir GC 1999 Chronic hyperglycemia triggers loss of pancreatic beta cell
differentiation in an animal model of diabetes. J Biol Chem 274: 141 12-21.

Olbrot M, Rud J, Moss LG, Sharma A 2002 Identification of beta-cell-specific
insulin gene transcription factor RIPE3b1 as mammalian MafA. Proc Natl Acad Sci
U S A 99:6737-42.

Kaneto H, Kajimoto Y, Miyagawa J, Matsuoka T, Fujitani Y, Umayahara Y,
Hanafusa T, Matsuzawa Y, Yamasaki Y, Hori M 1999 Beneficial effects of

antioxidants in diabetes: possible protection of pancreatic beta-cells against glucose
toxicity. Diabetes 48:2398-406.

Tanaka Y, Gleason CE, Tran PO, Harmon JS, Robertson RP 1999 Prevention
of glucose toxicity in HIT-T15 cells and Zucker diabetic fatty rats by antioxidants.
Proc Natl Acad Sci U S A 96:10857-62.

Kaneto H, Xu G, Fujii N, Kim S, Bonner-Weir S, Weir GC 2002 Involvement of
c-Jun N-terminal kinase in oxidative stress-mediated suppression of insulin gene
expression. J Biol Chem 14:14

Brownlee M 2001 Biochemistry and molecular cell biology of diabetic
complications. Nature 41 4:813-20.

Pictet RL, Rall LB, Phelps P, Rutter WJ 1976 The neural crest and the origin of
the insulin-producing and other gastrointestinal hormone-producing cells. Science
191:191-2.

157

11.

12.

l3.

14.

15.

16.

l7.

l8.

19.

20.

21.

22.

23.

24.

Fontaine J, Le Douarin NM 1977 Analysis of endoderm formation in the avian
blastoderrn by the use of quail-chick chimaeras. The problem of the neurectodermal
origin of the cells of the APUD series. J Embryol Exp Morphol 412209-22.

Edlund H 1998 Transcribing pancreas. Diabetes 47: 1 817-23.

Edlund H 1999 Pancreas: how to get there from the gut? Curr Opin Cell Biol
11:663-8.

Sindelar DK, Balcom JH, Chu CA, Neal DW, Cherrington AD 1996 A
comparison of the effects of selective increases in peripheral or portal insulin on
hepatic glucose production in the conscious dog. Diabetes 45: 1594-604.

Prager R, Wallace P, Olefsky JM 1987 Direct and indirect effects of insulin to
inhibit hepatic glucose output in obese subjects. Diabetes 36:607-11.

Myers SR, Diamond MP, Adkins-Marshall BA, Williams PE, Stinsen R,
Cherrington AD 1991 Effects of small changes in glucagon on glucose production
during a euglycemic, hyperinsulinemic clamp. Metabolism 40:66-71.

Kahn BB 1996 Lilly lecture 1995. Glucose transport: pivotal step in insulin action.
Diabetes 45: 1644-54.

Klip A, Paquet MR 1990 Glucose transport and glucose transporters in muscle and
their metabolic regulation. Diabetes Care 132228-43.

Katzen HM, Soderman DD, Wiley CE 1970 Multiple forms of hexokinase.
Activities associated with subcellular particulate and soluble fractions of normal
and streptozotocin diabetic rat tissues. J Biol Chem 245:4081-96.

Rossetti L, Giaccari A 1990 Relative contribution of glycogen synthesis and
glycolysis to insulin- mediated glucose uptake. A dose-response euglycemic clamp
study in normal and diabetic rats. J Clin Invest 85: 1785-92.

Mandarino LJ 1989 Regulation of skeletal muscle pyruvate dehydrogenase and
glycogen synthase in man. Diabetes Metab Rev 5:475-86.

Kruszynska YT, Home PD, Alberti KG 1986 In vivo regulation of liver and
skeletal muscle glycogen synthase activity by glucose and insulin. Diabetes 35:662-
7.

Campbell PJ, Carlson MG, Hill JO, Nurjhan N 1992 Regulation of free fatty
acid metabolism by insulin in humans: role of lipolysis and reesterification. Am J
Physiol 263:E1063-9.

Jensen MD, Caruso M, Heiling V, Miles JM 1989 Insulin regulation of lipolysis
in nondiabetic and IDDM subjects. Diabetes 38:1595-601.

158

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

36.

37.

Bonadonna RC, Groop LC, Zych K, Shank M, DeFronzo RA 1990 Dose-
dependent effect of insulin on plasma free fatty acid turnover and oxidation in
humans. Am J Physi01259zE736-50.

Hagstrom-Toft E, Bolinder J, Eriksson S, Amer P 1995 Role of
phosphodiesterase III in the antilipolytic effect of insulin in vivo. Diabetes 44:1170-
5.

Jefferson LS, Rannels DE, Monger BL, Morgan HE 1974 Insulin in the
regulation of protein turnover in heart and skeletal muscle. Fed Proc 33:1098-1104.

Lundholm K, Edstrom S, Ekman L, Karlberg I, Walker P, Schersten T 1981
Protein degradation in human skeletal muscle tissue: the effect of insulin, leucine,
amino acids and ions. Clin Sci (Lond) 60:319-26.

Pain VM, Garlick PJ 1974 Effect of streptozotocin diabetes and insulin treatment
on the rate of protein synthesis in tissues of the rat in vivo. J Biol Chem 249:4510—
4.

Jefferson LS 1980 Lilly Lecture 1979: role of insulin in the regulation of protein
synthesis. Diabetes 29:487-96.

Flakoll PJ, Kulaylat M, Frexes-Steed M, Hourani H, Brown LL, Hill JO,
Abumrad NN 1989 Amino acids augment insulin's suppression of whole body
proteolysis. Am J Physi01257zE839-47.

Kimball SR, Jefferson LS, Fadden P, Haystead TA, Lawrence JC, Jr. 1996
Insulin and diabetes cause reciprocal changes in the association of eIF- 4E and
PHAS-I in rat skeletal muscle. Am J Physiol 270:C705-9.

Steiner DF, Bell GI, Tager HS 1990 Chemistry and biosynthesis of pancreatic

protein hormones. In: L.G D (ed) Endocrinology. WB Saunders, Harcourt Brace
J ovanovich, Philadelphia: 1263

Chan SJ, Keim P, Steiner DF 1976 Cell-free synthesis of rat preproinsulins:
characterization and partial amino acid sequence determination. Proc Natl Acad Sci
U S A 73:1964-8.

Bell GI, Pictet RL, Rutter WJ, Cordell B, Tischer E, Goodman HM 1980
Sequence of the human insulin gene. Nature 284226-32.

Berne RM, Levy MN 1993 Hormones of the pancreas. In: Farrel R (ed) Physiology
Thrid ed. Mosby Year Book, St. Louis, MO

Rhodes CJ 2000 Processing of the insulin molecule. In: LeRoith D, Taylor SI,
Olefsky JM (eds) Diabetes Mellitus: a fundamental and clinical text Second ed.
Lippincott Williams & Wilkins, Philadelphiaz20-38

159

38.

39.

40.

41.

42.

43.

45.

46.

47.

48.

49.

50.

51.

Thorens B, Sarkar HK, Kaback HR, Lodish HF 1988 Cloning and functional
expression in bacteria of a novel glucose transporter present in liver, intestine,
kidney, and beta-pancreatic islet cells. Cell 552281-90.

McKinnon CM, Docherty K 2001 Pancreatic duodenal homeobox-l, PDX-l, a
major regulator of beta cell identity and function. Diabetologia 44: 1203-14.

Howell SL, Jones PM, Persaud SJ 1994 Regulation of insulin secretion: the role
of second messengers. Diabetologia 37 Suppl 2:830—5.

Dunne MJ, Petersen OH 1986 Intracellular ADP activates K+ channels that are
inhibited by ATP in an insulin-secreting cell line. FEBS Lett 208:59-62.

Hopkins WF, Fatherazi S, Peter-Riesch B, Corkey BE, Cook DL 1992 Two
sites for adenine-nucleotide regulation of ATP-sensitive potassium channels in
mouse pancreatic beta-cells and HIT cells. J Membr Biol 129:287-95.

Ashcroft FM, Harrison DE, Ashcroft SJ 1984 Glucose induces closure of single
potassium channels in isolated rat pancreatic beta-cells. Nature 312:446-8.

Ashcroft SJ, Hughes SJ 1990 Protein phosphorylation in the regulation of insulin
secretion and biosynthesis. Biochem Soc Trans 18:116—8.

Prentki M, Matschinsky FM 1987 Ca2+, cAMP, and phospholipid-derived
messengers in coupling mechanisms of insulin secretion. Physiol Rev 67:1185-248.

Wenham RM, Landt M, Easom RA 1994 Glucose activates the multifunctional
Ca2+/calmodulin-dependent protein kinase II in isolated rat pancreatic islets. J Biol
Chem 269:4947-52.

Ashcroft SJ 1994 Protein phosphorylation and beta-cell function. Diabetologia 37
Suppl 22821-9.

Sorenson RL, Brelje TC, Roth C 1994 Effect of tyrosine kinase inhibitors on
islets of Langerhans: evidence for tyrosine kinases in the regulation of insulin
secretion. Endocrinology 134: 1975-8.

Ratcliff H, Jones PM 1993 Effects of okadaic acid on insulin secretion from rat
islets of Langerhans. Biochim Biophys Acta 1175:188-91.

Aspinwall CA, Qian WJ, Roper MG, Kulkarni RN, Kahn CR, Kennedy RT
2000 Roles of insulin receptor substrate-1, phosphatidylinositol 3-kinase, and

release of intracellular Ca2+ stores in insulin-stirnulated insulin secretion in beta -
cells. J Biol Chem 275:22331-8.

Rothenberg PL, Willison LD, Simon J, Wolf BA 1995 Glucose-induced insulin

receptor tyrosine phosphorylation in insulin- secreting beta-cells. Diabetes 44:802-
9.

160

52.

53.

54.

55.

56.

57.

58.

59.

60.

61.

62.

63.

64.

Leibiger IB, Leibiger B, Moede T, Berggren PO 1998 Exocytosis of insulin
promotes insulin gene transcription via the insulin receptor/PI-3 kinase/p70 s6
kinase and CaM kinase pathways. Mol Cell 1:933-8.

Xu GG, Gao ZY, Borge PD, Jr., Wolf BA 1999 Insulin receptor substrate 1-
induced inhibition of endoplasmic reticulum Ca2+ uptake in beta-cells. Autocrine
regulation of intracellular ca2+ homeostasis and insulin secretion. J Biol Chem
274:18067-74.

Owerbach D, Bell GI, Rutter WJ, Shows TB 1980 The insulin gene is located on
chromosome 11 in humans. Nature 286:82-4.

German M 1994 Insulin gene structure and regulation, in Molecular biology of
diabetes. Human Press, Totowa, NJ

Rudnick A, Ling TY, Odagiri H, Rutter WJ, German MS 1994 Pancreatic beta
cells express a diverse set of homeobox genes. Proc Natl Acad Sci U S A 91:12203-
7.

Waeber G, Thompson N, Nicod P, Bonny C 1996 Transcriptional activation of
the GLUT2 gene by the IPF-l/STF-l/IDX-l homeobox factor. Mol Endocrinol
10:1327-34.

Watada H, Kajimoto Y, Umayahara Y, Matsuoka T, Kaneto H, Fujitani Y,
Kamada T, Kawamori R, Yamasaki Y 1996 The human glucokinase gene beta-

cell-type promoter: an essential role of insulin promoter factor l/PDX-l in its
activation in HIT-T15 cells. Diabetes 45:1478-88.

Carty MD, Lillquist JS, Peshavaria M, Stein R, Soeller WC 1997 Identification
of cis- and trans-active factors regulating human islet amyloid polypeptide gene
expression in pancreatic beta-cells. J Biol Chem 272:11986—93.

Miller CP, McGehee RE, Jr., Habener JF 1994 IDX-l: a new homeodomain
transcription factor expressed in rat pancreatic islets and duodenum that
transactivates the somatostatin gene. Embo J 13:1145-56.

Jonsson J, Carlsson L, Edlund T, Edlund H 1994 Insulin-promoter-factor 1 is
required for pancreas development in mice. Nature 371:606-9.

Stoffers DA, Ferrer J, Clarke WL, Habener JF 1997 Early-onset type-II diabetes
mellitus (MODY4) linked to IPFl. Nat Genet 17: 138-9.

Mirmira RG, Watada H, German MS 2000 Beta-cell differentiation factor
ka6.1 contains distinct DNA binding interference and transcriptional repression
domains. J Biol Chem 275: 14743-5 l.

Odagiri H, Wang J, German MS 1996 Function of the human insulin promoter in
primary cultured islet cells. J Biol Chem 271:1909-15.

161

65.

66.

67.

68.

69.

70.

71.

72.

73.

74.

75.

76.

Melloul D, Marshak S, Cerasi E 2002 Regulation of insulin gene transcription.
Diabetologia 451309-26.

Naya FJ, Stellrecht CM, Tsai MJ 1995 Tissue-specific regulation of the insulin
gene by a novel basic helix- loop-helix transcription factor. Genes Dev 9:1009-19.

Naya FJ, Huang HP, Qiu Y, Mutoh H, DeMayo FJ, Leiter AB, Tsai MJ 1997
Diabetes, defective pancreatic morphogenesis, and abnormal enteroendocrine
differentiation in BETA2/neuroD-deficient mice. Genes Dev 11:2323-34.

Dumonteil E, Laser B, Constant I, Philippe J 1998 Differential regulation of the
glucagon and insulin 1 gene promoters by the basic helix-loop-helix transcription
factors E47 and BETA2. J Biol Chem 273:19945-54.

Peers B, Leonard J, Sharma S, Teitelman G, Montminy MR 1994 Insulin
expression in pancreatic islet cells relies on cooperative interactions between the
helix loop helix factor E47 and the homeobox factor STF-l. Mol Endocrinol
821798-806.

Read ML, Clark AR, Docherty K 1993 The helix-loop-helix transcription factor
USF (upstream stimulating factor) binds to a regulatory sequence of the human
insulin gene enhancer. Biochem J 295:233-7.

Ohneda K, Mirmira RG, Wang J, Johnson JD, German MS 2000 The
homeodomain of PDX-l mediates multiple protein-protein interactions in the

formation of a transcriptional activation complex on the insulin promoter. Mol Cell
Biol 20:900-1 1.

Robinson GL, Henderson E, Massari ME, Murre C, Stein R 1995 c-jun inhibits
insulin control element-mediated transcription by affecting the transactivation
potential of the E2A gene products. Mol Cell Biol 15: 1398-404.

Wice BM, BernaI-Mizrachi E, Permutt MA 2001 Glucose and other insulin
secretagogues induce, rather than inhibit, expression of Id-l and Id-3 in pancreatic
islet beta cells. Diabetologia 44:453-63.

Shieh SY, Tsai MJ 1991 Cell-specific and ubiquitous factors are responsible for
the enhancer activity of the rat insulin 11 gene. J Biol Chem 266:16708-14.

Read ML, Masson MR, Docherty K 1997 A RIPE3b1-like factor binds to a novel
site in the human insulin promoter in a redox-dependent manner. FEBS Lett
418:68-72.

Sander M, Neubuser A, Kalamaras J, Ee HC, Martin GR, German MS 1997

Genetic analysis reveals that PAX6 is required for normal transcription of
pancreatic hormone genes and islet development. Genes Dev 11:1662-73.

162

 

77.

78.

79.

80.

81.

82.

83.

84.

85.

86.

87.

Turque N, Plaza S, Radvanyi F, Carriere C, Saule S 1994 Pax-QNR/Pax-6, a
paired box- and homeobox-containing gene expressed in neurons, is also expressed
in pancreatic endocrine cells. Mol Endocrinol 82929-38.

Daniel PB, Walker WH, Habener JF 1998 Cyclic AMP signaling and gene
regulation. Annu Rev Nutr 18:353-83

Inagaki N, Maekawa T, Sudo T, lshii S, Seino Y, Imura H 1992 c-Jun represses
the human insulin promoter activity that depends on multiple cAMP response
elements. Proc Natl Acad Sci U S A 89:1045-9.

Nielsen DA, Welsh M, Casadaban MJ, Steiner DF 1985 Control of insulin gene
expression in pancreatic beta-cells and in an insulin-producing cell line, RIN-SF
cells. I. Effects of glucose and cyclic AMP on the transcription of insulin mRNA. J
Biol Chem 260: 13585-9.

Fonlkes NS, Borrelli E, Sassone-Corsi P 1991 CREM gene: use of alternative
DNA-binding domains generates multiple antagonists of cAMP-induced
transcription. Cell 64:739-49.

Inada A, Someya Y, Yamada Y, Ihara Y, Kubota A, Ban N, Watanabe R,
Tsuda K, Seino Y 1999 The cyclic AMP response element modulator family

regulates the insulin gene transcription by interacting with transcription factor IID. J
Biol Chem 274:21095-103.

Inada A, Yamada Y, Someya Y, Kubota A, Yasuda K, Ihara Y, Kagimoto S,
Kuroe A, Tsuda K, Seino Y 1998 Transcriptional repressors are increased in
pancreatic islets of type 2 diabetic rats. Biochem Biophys Res Cornrnun 253:712-8.

Eggers A, Siemann G, Blume R, Knepel W 1998 Gene-specific transcriptional
activity of the insulin cAMP-responsive element is conferred by NF-Y in
combination with cAMP response element- binding protein. J Biol Chem
273:18499—508.

Seufert J, Weir GC, Habener JF 1998 Differential expression of the insulin gene
transcriptional repressor CCAAT/enhancer—binding protein beta and transactivator

islet duodenum homeobox-1 in rat pancreatic beta cells during the development of
diabetes mellitus. J Clin Invest 101:2528-39.

Lu M, Seufert J, Habener JF 1997 Pancreatic beta-cell-specific repression of
insulin gene transcription by CCAAT/enhancer-binding protein beta. Inhibitory
interactions with basic helix—loop—helix transcription factor E47. J Biol Chem
272:28349-59.

Boam DS, Clark AR, Docherty K 1990 Positive and negative regulation of the
human insulin gene by multiple trans-acting factors. J Biol Chem 265:8285-96.

163

 

88.

89.

90.

91.

92.

93.

94.

95.

96.

97.

98.

99.

100.

Clark AR, Wilson ME, Leibiger 1, Scott V, Docherty K 1995 A silencer and an
adjacent positive element interact to modulate the activity of the human insulin
promoter. Eur J Biochem 232:627-32.

Sander M, Griffen SC, Huang J, German MS 1998 A novel glucose-responsive
element in the human insulin gene functions uniquely in primary cultured islets.
Proc Natl Acad Sci U S A 95:11572-7.

Goodman PA, Medina-Martinez O, Fernandez-Mejia C 1996 Identification of
the human insulin negative regulatory element as a negative glucocorticoid
response element. Mol Cell Endocrinol 120:139-46.

German MS, Wang J 1994 The insulin gene contains multiple transcriptional
elements that respond to glucose. Mol Cell Biol 14:4067-75.

Karlsson O, Edlund T, Moss JB, Rutter WJ, Walker MD 1987 A mutational
analysis of the insulin gene transcription control region: expression in beta cells is
dependent on two related sequences within the enhancer. Proc Natl Acad Sci U S A
84:8819-23.

Epstein J, Cai J, Glaser T, Jepeal L, Maas R 1994 Identification of a Pax paired
domain recognition sequence and evidence for DNA-dependent conformational
changes. J Biol Chem 269:8355-61.

Crowe DT, Tsai MJ 1989 Mutagenesis of the rat insulin H 5'-flanking region
defines sequences important for expression in HIT cells. Mol Cell Biol 9: 1784-9.

German MS, Moss LG, Wang J, Rutter WJ 1992 The insulin and islet amyloid
polypeptide genes contain similar cell- specific promoter elements that bind
identical beta-cell nuclear complexes. Mol Cell Biol 12: 1777-88.

Qiu Y, Guo M, ll-Iuang S, Stein R 2002 Insulin gene transcription is mediated by
interactions between the p300 coactivator and PDX-l, BETA2, and E47. Mol Cell
Biol 222412-20.

Zimmet P, Alberti KG, Shaw J 2001 Global and societal implications of the
diabetes epidemic. Nature 414:782-7.

Olefsky JM 1981 LIlly lecture 1980. Insulin resistance and insulin action. An in
vitro and in vivo perspective. Diabetes 302148-62.

Leahy JL, Bonner-Weir S, Weir GC 1992 Beta-cell dysfunction induced by
chronic hyperglycemia. Current ideas on mechanism of impaired glucose-induced
insulin secretion. Diabetes Care 15:442-55.

Weir GC, Laybutt DR, Kaneto H, Bonner-Weir S, Sharma A 2001 Beta-cell
adaptation and decompensation during the progression of diabetes. Diabetes 50
Suppl 1:S154-9.

164

 

101.

102.

103.

104.

105.

106.

107.

108.

109.

110.

111.

112.

Tokuyama Y, Sturis J, DePaoli AM, Takeda J, Stoffel M, Tang J, Sun X,
Polonsky KS, Bell GI 1995 Evolution of beta-cell dysfunction in the male Zucker
diabetic fatty rat. Diabetes 44:1447-57.

Ma Z, Landt M, Bohrer A, Ramanadham S, Kipnis DM, Turk J 1997
Interleukin-l reduces the glycolytic utilization of glucose by pancreatic islets and

reduces glucokinase mRNA content and protein synthesis by a nitric oxide-
dependent mechanism. J Biol Chem 272: 1 7827-35.

Dohan F, Luken F 1948 Experimental diabetes produced by the administration of
glucose. Endocrinology 42:244-262

Turner RC, McCarthy ST, Holman RR, Harris E 1976 Beta-cell function
improved by supplementing basal insulin secretion in mild diabetes. Br Med J
1:1252-4.

Brunzell JD, Robertson RP, Lerner RL, Hazzard WR, Ensinck JW, Bierman
EL, Porte D, Jr. 1976 Relationships between fasting plasma glucose levels and

insulin secretion during intravenous glucose tolerance tests. J Clin Endocrinol
Metab 42:222-9.

Vague P, Moulin JP 1982 The defective glucose sensitivity of the B cell in non
insulin dependent diabetes. Improvement after twenty hours of norrnoglycaemia.
Metabolism 31:139-42.

Leahy JL, Weir GC 1988 Evolution of abnormal insulin secretory responses
during 48-h in vivo hyperglycemia. Diabetes 371217-22.

Leahy JL, Cooper HE, Weir GC 1987 Impaired insulin secretion associated with
near norrnoglycemia. Study in normal rats with 96-h in vivo glucose infusions.
Diabetes 361459-64.

Bonner-Weir S, Trent DF, Weir GC 1983 Partial pancreatectomy in the rat and
subsequent defect in glucose— induced insulin release. J Clin Invest 71 : 1544-53.

Collier SA, Mandel TE, Carter WM 1982 Detrimental effect of high medium
glucose concentration on subsequent endocrine function of transplanted organ-
cultured foetal mouse pancreas. Aust J Exp Biol Med Sci 60 Pt 4:43 7-45.

Johnson JH, Ogawa A, Chen L, Orci L, Newgard CB, Alarn T, Unger RH 1990
Underexpression of beta cell high Km glucose transporters in noninsulin- dependent
diabetes. Science 250:546-9.

Orci L, Ravazzola M, Baetens D, Inman L, Amherdt M, Peterson RG,
Newgard CB, Johnson JH, Unger RH 1990 Evidence that down-regulation of
beta-cell glucose transporters in non- insulin-dependent diabetes may be the cause
of diabetic hyperglycemia. Proc Natl Acad Sci U S A 87:9953-7.

165

113.

114.

115.

116.

117.

118.

119.

120.

121.

122.

123.

124.

Zangen DH, Bonner-Weir S, Lee CH, Latimer JB, Miller CP, Habener JF,
Weir GC 1997 Reduced insulin, GLUT2, and IDX-l in beta-cells after partial
pancreatectomy. Diabetes 46:258-64.

Unger RH 1991 Diabetic hyperglycemia: link to impaired glucose transport in
pancreatic beta cells. Science 251 : 1200-5.

De Vos A, Heimberg H, Quartier E, Huypens P, Bonwens L, Pipeleers D,
Schuit F 1995 Human and rat beta cells differ in glucose transporter but not in
glucokinase gene expression. J Clin Invest 96:2489-95.

Lachaal M, Spangler RA, Jung CY 1993 High Km of GLUT-2 glucose
transporter does not explain its role in insulin secretion. Am J Physiol 265:E914-9.

Lachaal M, Jung CY 1993 Interaction of facilitative glucose transporter with
glucokinase and its modulation by ADP and glucose-6-phosphate. J Cell Physiol
156:326—32.

Eizirik DL, Korbutt GS, Hellerstrom C 1992 Prolonged exposure of human
pancreatic islets to high glucose concentrations in vitro impairs the beta-cell
function. J Clin Invest 90:1263-8.

Briaud I, Rouault C, Reach G, Poitout V 1999 Long-term exposure of isolated rat

islets of Langerhans to supraphysiologic glucose concentrations decreases insulin
mRN A levels. Metabolism 48:319-23.

Robertson RP, Zhang HJ, Pyzdrowski KL, Walseth TF 1992 Preservation of
insulin mRNA levels and insulin secretion in HIT cells by avoidance of chronic
exposure to high glucose concentrations. J Clin Invest 90:320-5.

Olson LK, Redmon JB, Towle HC, Robertson RP 1993 Chronic exposure of HIT
cells to high glucose concentrations paradoxically decreases insulin gene
transcription and alters binding of insulin gene regulatory protein. J Clin Invest
92:514-9.

Harmon JS, Tanaka Y, Olson LK, Robertson RP 1998 Reconstitution of
glucotoxic HIT-T15 cells with somatostatin transcription factor-1 partially restores
insulin promoter activity. Diabetes 47:900-4.

Hanahan D 1985 Heritable formation of pancreatic beta-cell tumours in transgenic
mice expressing recombinant insulin/simian virus 40 oncogenes. Nature 315:115-
22.

Efrat S, Linde S, Kofod H, Spector D, Delannoy M, Grant S, Hanahan D,
Baekkeskov S 1988 Beta-cell lines derived fi'om transgenic mice expressing a
hybrid insulin gene-oncogene. Proc Natl Acad Sci U S A 85:9037-41.

166

125.

126.

127.

128.

129.

130.

131.

132.

133.

134.

135.

136.

137.

Poitout V, Olson LK, Robertson RP 1996 Chronic exposure of betaTC-6 cells to
supraphysiologic concentrations of glucose decreases binding of the RIPE3b1
insulin gene transcription activator. J Clin Invest 97:1041-6.

Olson LK, Qian J, Poitout V 1998 Glucose rapidly and reversibly decreases INS-1
cell insulin gene transcription via decrements in STF-l and C1 activator
transcription factor activity. Mol Endocrinol 122207-19.

Kaneto H, Suzuma K, Sharma A, Bonner-Weir S, King GL, Weir GC 2002
Involvement of protein kinase C beta 2 in c-myc induction by high glucose in
pancreatic beta-cells. J Biol Chem 277:3680-5.

Dang CV 1999 c-Myc target genes involved in cell growth, apoptosis, and
metabolism. Mol Cell Biol 1921-11.

Facchini LM, Penn LZ 1998 The molecular role of Myc in growth and
transformation: recent discoveries lead to new insights. F aseb J 122633-51.

Bouchard C, Staller P, Eilers M 1998 Control of cell proliferation by Myc.
Trends Cell Biol 82202-6.

Laybutt DR, Sharma A, Sgroi DC, Gaudet J, Bonner-Weir S, Weir GC 2002
Genetic regulation of metabolic pathways in beta-cells disrupted by hyperglycemia.
J Biol Chem 277:10912-21.

Elks ML 1990 Fat oxidation and diabetes of obesity: the Randle hypothesis
revisited. Med Hypotheses 331257-60.

Sako Y, Grill VE 1990 A 48-hour lipid infusion in the rat time-dependently
inhibits glucose- induced insulin secretion and B cell oxidation through a process
likely coupled to fatty acid oxidation. Endocrinology 127:1580-9.

Beme C 1975 The metabolism of lipids in mouse pancreatic islets. The
biosynthesis of triacylglycerols and phospholipids. Biochem J 152:667-73.

Elks ML 1993 Chronic perifusion of rat islets with pahnitate suppresses glucose-
stimulated insulin release. Endocrinology 133:208-14.

Gremlich S, Bonny C, Waeber G, Thorens B 1997 Fatty acids decrease IDX-l
expression in rat pancreatic islets and reduce GLUT2, glucokinase, insulin, and
somatostatin levels. J Biol Chem 272:30261-9.

Jacqueminet S, Briaud I, Rouault C, Reach G, Poitout V 2000 Inhibition of
insulin gene expression by long-term exposure of pancreatic beta cells to pahnitate

is dependent on the presence of a stimulatory glucose concentration. Metabolism
49:532-6.

167

 

138.

139.

140.

141.

142.

143.

144.

145.

146.

147.

148.

149.

150.

Lee Y, Hirose H, Ohneda M, Johnson JH, McGarry JD, Unger RH 1994 Beta-
cell lipotoxicity in the pathogenesis of non-insulin-dependent diabetes mellitus of

obese rats: impairment in adipocyte-beta-cell relationships. Proc Natl Acad Sci U S
A 91 :10878-82.

Harmon JS, Gleason CE, Tanaka Y, Poitout V, Robertson RP 2001 Antecedent
hyperglycemia, not hyperlipidemia, is associated with increased islet triacylglycerol

content and decreased insulin gene mRNA level in Zucker diabetic fatty rats.
Diabetes 50:2481-6.

Briaud I, Kelpe CL, Johnson LM, Tran PO, Poitout V 2002 Differential effects
of hyperlipidemia on insulin secretion in islets of langerhans from hyperglycemic
versus normoglycemic rats. Diabetes 51 :662-8.

Carlsson C, Borg LA, Welsh N 1999 Sodium pahnitate induces partial
mitochondrial uncoupling and reactive oxygen species in rat pancreatic islets in
vitro. Endocrinology 140:3422-8.

Betteridge DJ 2000 What is oxidative stress? Metabolism 4913-8.

Finkel T, Holbrook NJ 2000 Oxidants, oxidative stress and the biology of ageing.
Nature 408:239-47.

Ceriello A, dello Russo P, Amstad P, Cerutti P 1996 High glucose induces
antioxidant enzymes in human endothelial cells in culture. Evidence linking
hyperglycemia and oxidative stress. Diabetes 45:471-7.

Njoroge GF, Monnier VM 1989 The chemistry of the maillard reaction under
physiological conditions: A review. In The Maillard Reaction in Aging, Diabetes,
and Nutrition. Alan R. Liss, Incz85-107

Gillery P, Monboisse JC, Maquart FX, Borel JP 1988 Glycation of proteins as a
source of superoxide. Diabete Metab 14:25-30.

Sakurai T, Tsuchiya S 1988 Superoxide production from nonenzymatically
glycated protein. FEBS Lett 236:406-10.

Brownlee M 2000 Negative consequences of glycation. Metabolism 49:9-13.

Hunt JV, Dean RT, Wolff SP 1988 Hydroxyl radical production and autoxidative
glycosylation. Glucose autoxidation as the cause of protein damage in the
experimental glycation model of diabetes mellitus and ageing. Biochem J 256:205-
12.

Hunt JV, Smith CC, Wolff SP 1990 Autoxidative glycosylation and possible
involvement of peroxides and free radicals in LDL modification by glucose.
Diabetes 39: 1420-4.

168

 

151.

152.

153.

154.

155.

156.

157.

158.

159.

160.

161.

162.

163.

164.

165.

Carubelli R, Schneider JE, Jr., Pye QN, Floyd RA 1995 Cytotoxic effects of
autoxidative glycation. Free Radic Biol Med 18:265-9.

Clements RS, Jr. 1986 The polyol pathway. A historical review. Drugs 32:3-5.

Lee AY, Chung SS 1999 Contributions of polyol pathway to oxidative stress in
diabetic cataract. F aseb J 13:23-30.

Varma SD, Kinoshita JH 1974 The absence of cataracts in mice with congenital
hyperglycemia. Exp Eye Res 19:577-82.

Hunt JV, Wolff SP 1991 Oxidative glycation and free radical production: a causal
mechanism of diabetic complications. Free Radic Res Commun 12-132115-23.

Baynes JW 1991 Role of oxidative stress in development of complications in
diabetes. Diabetes 40:405-12.

Ihara Y, Toyokuni S, Uchida K, Odaka H, Tanaka T, Ikeda H, Hiai H, Seino
Y, Yamada Y 1999 Hyperglycemia causes oxidative stress in pancreatic beta-cells
of GK rats, a model of type 2 diabetes. Diabetes 482927-32.

Matsuoka T, Kajimoto Y, Watada H, Kaneto H, Kishimoto M, Umayahara Y,
Fujitani Y, Kamada T, Kawamori R, Yamasaki Y 1997 Glycation-dependent,

reactive oxygen species-mediated suppression of the insulin gene promoter activity
in HIT cells. J Clin Invest 99:144-50.

Myint T, Hoshi S, Ookawara T, Miyazawa N, Suzuki K, Taniguchi N 1995
Immunological detection of glycated proteins in normal and streptozotocin-induced
diabetic rats using anti hexitol-lysine IgG. Biochim Biophys Acta 1272:73-9.

Bunn HF 1981 Nonenzymatic glycosylation of protein: relevance to diabetes. Am J
Med 70:325-30.

Bunn HF 1981 Modification of hemoglobin and other proteins by nonenzymatic
glycosylation. Prog Clin Biol Res 51:223-39

Brownlee M, Vlassara H, Cerami A 1984 Nonenzymatic glycosylation and the
pathogenesis of diabetic complications. Ann Intern Med 101:527-37.

Vlassara H 1997 Recent progress in advanced glycation end products and diabetic
complications. Diabetes 46 Suppl 2:S19-25.

Tiedge M, Lortz S, Drinkgern J, Lenzen S 1997 Relation between antioxidant
enzyme gene expression and antioxidative defense status of insulin-producing cells.
Diabetes 46:1733-42.

Dandona P, Thusu K, Cook S, Snyder B, Makowski J, Armstrong D, Nicotera
T 1996 Oxidative damage to DNA in diabetes mellitus. Lancet 347:444-5.

169

166.

167.

168.

169.

170.

171.

172.

173.

174.

175.

176.

Leinonen J, Lehtimaki T, Toyokuni S, Okada K, Tanaka T, Hiai H, Ochi H,
Laippala P, Rantalaiho V, Wirta O, Pasternack A, Alho H 1997 New biomarker

evidence of oxidative DNA damage in patients with non- insulin-dependent
diabetes mellitus. FEBS Lett 417:150—2.

Kashiwagi A, Obata T, Suzuki M, Takagi Y, Kida Y, Ogawa T, Tanaka Y,
Asahina T, Ikebuchi M, Saeki Y, et al. 1992 Increase in cardiac muscle fi'uctose
content in streptozotocin-induced diabetic rats. Metabolism 41 :1041-6.

Yorek MA, Wiese TJ, Davidson EP, Dunlap JA, Stefani MR, Conner CE,
Lattimer SA, Kamijo M, Greene DA, Sima AA 1993 Reduced motor nerve
conduction velocity and Na(+)-K(+)-ATPase activity in rats maintained on L-
fucose diet. Reversal by myo-inositol supplementation. Diabetes 42: 1401 -6.

Kaneto H, Fujii J, Myint T, Miyazawa N, Islam KN, Kawasaki Y, Suzuki K,
Nakamura M, Tatsumi H, Yamasaki Y, Taniguchi N 1996 Reducing sugars
trigger oxidative modification and apoptosis in pancreatic beta-cells by provoking
oxidative stress through the glycation reaction. Biochem J 320:855-63.

Tajiri Y, Moller C, Grill V 1997 Long-term effects of aminoguanidine on insulin
release and biosynthesis: evidence that the formation of advanced glycosylation end
products inhibits B cell function. Endocrinology 138:273-80.

Kaneto H, Xu G, Song KH, Suzuma K, Bonner-Weir S, Sharma A, Weir GC
2001 Activation of the hexosamine pathway leads to deterioration of pancreatic
beta-cell function through the induction of oxidative stress. J Biol Chem
276:31099—104.

Maechler P, Jornot L, Wollheim CB 1999 Hydrogen peroxide alters
mitochondrial activation and insulin secretion in pancreatic beta cells. J Biol Chem
274:27905-13.

Laybutt DR, Kaneto H, Hasenkamp W, Grey S, Jonas JC, Sgroi DC, Groff A,
Ferran C, Bonner-Weir S, Sharma A, Weir GC 2002 Increased expression of
antioxidant and antiapoptotic genes in islets that may contribute to beta-cell survival
during chronic hyperglycemia. Diabetes 51 :413-23.

Nishizuka Y 1992 Intracellular signaling by hydrolysis of phospholipids and
activation of protein kinase C. Science 258:607-14.

Ganesan S, Calle R, Zawalich K, Greeuawalt K, Zawalich W, Shulman GI,
Rasmussen H 1992 Irnmunocytochemical localization of alpha-protein kinase C in

rat pancreatic beta-cells during glucose-induced insulin secretion. J Cell Biol
119:313-24.

Ito A, Saito N, Taniguchi H, Chiba T, Kikkawa U, Saitoh Y, Tanaka C 1989

Localization of beta II subspecies of protein kinase C in beta-cells. Diabetes
38:1005-11.

170

177.

178.

179.

180.

181.

182.

183.

184.

185.

186.

187.

188.

Koya D, King GL 1998 Protein kinase C activation and the development of
diabetic complications. Diabetes 47:859-66.

Kunisaki M, Bursell SE, Umeda F, Nawata H, King GL 1994 Normalization of
diacylglycerol-protein kinase C activation by vitamin E in aorta of diabetic rats and
cultured rat smooth muscle cells exposed to elevated glucose levels. Diabetes
43:1372-7.

King GL, Ishii H, Koya D 1997 Diabetic vascular dysfunctions: a model of
excessive activation of protein kinase C. Kidney Int Suppl 60:S77-85.

Park JY, Takahara N, Gabriele A, Chou E, Naruse K, Suzuma K, Yamauchi
T, Ha SW, Meier M, Rhodes CJ, King GL 2000 Induction of endothelin-1
expression by glucose: an effect of protein kinase C activation. Diabetes 49:1239-
48.

Brockenbrough JS, Weir GC, Bonner-Weir S 1988 Discordance of exocrine and
endocrine growth after 90% pancreatectomy in rats. Diabetes 37:232-6.

Swenne I 1992 Pancreatic beta-cell growth and diabetes mellitus. Diabetologia
35:193-201.

Unger RH 1995 Lipotoxicity in the pathogenesis of obesity-dependent NIDDM.
Genetic and clinical implications. Diabetes 44:863-70.

Sjoholm A 1997 Glucose stimulates islet beta-cell mitogenesis through GTP-
binding proteins and by protein kinase C-dependent mechanisms. Diabetes
46:1141-7.

Rhodes CJ 2000 IGF-I and GH post-receptor signaling mechanisms for pancreatic
beta- cell replication. J Mol Endocrinol 24:303-11.

Hugl SR, White MF, Rhodes CJ 1998 Insulin-like growth factor I (IGF-I)-
stimulated pancreatic beta-cell growth is glucose-dependent. Synergistic activation

of insulin receptor substrate-mediated signal transduction pathways by glucose and
IGF-1 in INS-1 cells. J Biol Chem 273:17771-9.

Cousin SP, Hugl SR, Myers MG, Jr., White MF, Reifel-Miller A, Rhodes CJ
1999 Stimulation of pancreatic beta-cell proliferation by growth hormone is
glucose-dependent: signal transduction via janus kinase 2 (JAK2)/signal transducer
and activator of transcription 5 (STATS) with no crosstalk to insulin receptor
substrate-mediated mitogenic signalling. Biochem J 344 Pt 3:649-58.

Pick A, Clark J, Kubstrup C, Levisetti M, Pugh W, Bonner-Weir S, Polonsky
KS 1998 Role of apoptosis in failure of beta-cell mass compensation for insulin

resistance and beta-cell defects in the male Zucker diabetic fatty rat. Diabetes
472358-64.

171

 

189.

190.

191.

192.

193.

194.

195.

196.

197.

198.

199.

200.

Shimabukuro M, Zhou YT, Lee Y, Unger RH 1998 Troglitazone lowers islet fat
and restores beta cell function of Zucker diabetic fatty rats. J Biol Chem 273:3547-
50.

Griffin ME, Marcucci MJ, Cline GW, Bell K, Barucci N, Lee D, Goodyear LJ,
Kraegen EW, White MF, Shulman GI 1999 Free fatty acid-induced insulin
resistance is associated with activation of protein kinase C theta and alterations in
the insulin signaling cascade. Diabetes 48: 1270-4.

Schmitz-Peiffer C, Craig DL, Biden TJ 1999 Ceramide generation is sufficient to
account for the inhibition of the insulin-stimulated PKB pathway in C2C12 skeletal
muscle cells pretreated with palmitate. J Biol Chem 274:24202-10.

Lingohr MK, Dickson LM, McCuaig JF, Hugl SR, Twardzik DR, Rhodes CJ
2002 Activation of IRS-2-mediated signal transduction by IGF-l, but not TGF-
alpha or EGF, augments pancreatic beta-cell proliferation. Diabetes 51:966-76.

Welsh M, Nielsen DA, MacKrell AJ, Steiner DF 1985 Control of insulin gene
expression in pancreatic beta-cells and in an insulin-producing cell line, RIN-SF
cells. 11. Regulation of insulin mRNA stability. J Biol Chem 260:13590-4.

Welsh M, Scherberg N, Gilmore R, Steiner DF 1986 Translational control of
insulin biosynthesis. Evidence for regulation of elongation, initiation and signal-
recognition-particle—mediated translational arrest by glucose. Biochem J 235:459-
67.

Bennett DL, Bailyes EM, Nielsen E, Guest PC, Rutherford NG, Arden SD,
Hutton JC 1992 Identification of the type 2 proinsulin processing endopeptidase as
PC2, a member of the eukaryote subtilisin family. J Biol Chem 267:15229-36.

Leibiger B, Moede T, Schwarz T, Brown GR, Kohler M, Leibiger IB, Berggren
PO 1998 Short-term regulation of insulin gene transcription by glucose. Proc Natl
Acad Sci U S A 95:9307-12.

Leibiger B, Wahlander K, Berggren PO, Leibiger IB 2000 Glucose-stimulated
insulin biosynthesis depends on insulin-stimulated insulin gene transcription. J Biol
Chem 275:30153-6.

Xu GG, Rothenberg PL 1998 Insulin receptor signaling in the beta-cell influences
insulin gene expression and insulin content: evidence for autocrine beta-cell
regulation. Diabetes 47:1243-52.

da Silva Xavier G, Varadi A, Ainscow EK, Rutter GA 2000 Regulation of gene
expression by glucose in pancreatic beta -cells (MIN6) via insulin secretion and
activation of phosphatidylinositol 3'- kinase. J Biol Chem 275:36269-77.

Wu H, MacFarlane WM, Tadayyon M, Arch JR, James RF, Docherty K 1999
Insulin stimulates pancreatic-duodenal homoeobox factor-l (PDXl) DNA- binding

172

 

201.

202.

203.

204.

205.

206.

207.

208.

209.

210.

activity and insulin promoter activity in pancreatic beta cells. Biochem J 344 Pt
31813-8.

Macfarlane WM, Smith SB, James RF, Clifton AD, Doza YN, Cohen P,
Docherty K 1997 The p38/reactivating kinase mitogen-activated protein kinase
cascade mediates the activation of the transcription factor insulin upstream factor 1

and insulin gene transcription by high glucose in pancreatic beta-cells. J Biol Chem
272120936-44.

Rafiq 1, da Silva Xavier G, Hooper S, Rutter GA 2000 Glucose-stimulated
preproinsulin gene expression and nuclear trans- location of pancreatic duodenum

homeobox-l require activation of phosphatidylinositol 3-kinase but not p38
MAPK/SAPKZ. J Biol Chem 275:15977-84.

Frodin M, Sekiue N, Roche E, Filloux C, Prentki M, Wollheim CB, Van
Obberghen E 1995 Glucose, other secretagogues, and nerve growth factor

stimulate mitogen- activated protein kinase in the insulin-secreting beta-cell line,
INS-1. J Biol Chem 270:7882-9.

Susini S, Roche E, Prentki M, Schlegel W 1998 Glucose and glucoincretin
peptides synergize to induce c-fos, c-jun, junB, zif-268, and nur—77 gene expression
in pancreatic beta(INS-1) cells. Faseb J 12:1173-82.

Mandrup-Poulsen T 1996 The role of interleukin-1 in the pathogenesis of IDDM.
Diabetologia 39:1005-29.

Robinvitch A 1993 Role of cytokines in IDDM pathogenesis and islet beta-cell
destruction. Diabetologia Rev. 11215-240

Rabinovitch A, Sumoski W, Rajotte RV, Warnock GL 1990 Cytotoxic effects of
cytokines on human pancreatic islet cells in monolayer culture. J Clin Endocrinol
Metab 71:152-6;

Corbett JA, Wang JL, Hughes JH, Wolf BA, Sweetland MA, Lancaster JR,
Jr., McDaniel ML 1992 Nitric oxide and cyclic GMP formation induced by

interleukin 1 beta in islets of Langerhans. Evidence for an effector role of nitric
oxide in islet dysfunction. Biochem J 2871229-35.

Comens PG, Wolf BA, Unanue ER, Lacy PE, McDaniel ML 1987 Interleukin 1
is potent modulator of insulin secretion from isolated rat islets of Langerhans.
Diabetes 361963-70.

Corbett JA, Sweetland MA, Wang JL, Lancaster JR, Jr., McDaniel ML 1993
Nitric oxide mediates cytokine—induced inhibition of insulin secretion by human
islets of Langerhans. Proc Natl Acad Sci U S A 90: 1731 -.5

173

 

211.

212.

213.

214.

215.

216.

217.

218.

219.

220.

221.

Major CD, Wolf BA 2001 Interleukin-lbeta stimulation of c-Jun NH(2)-terminal
kinase activity in insulin-secreting cells: evidence for cytoplasmic restriction.
Diabetes 5012721-8.

Welsh N 1996 Interleukin-l beta-induced cerarnide and diacylglycerol generation
may lead to activation of the c-Jun NH2-terminal kinase and the transcription factor
ATF2 in the insulin-producing cell line RINmSF. J Biol Chem 271 18307-12.

Bonny C, Oberson A, Steinmann M, Schorderet DF, Nicod P, Waeber G 2000
IBl reduces cytokine-induced apoptosis of insulin-secreting cells. J Biol Chem
275116466-72.

Larsen CM, Wadt KA, Juhl LF, Andersen HU, Karlsen AE, Su MS, Seedorf
K, Shapiro L, Dinarello CA, Mandrup-Poulsen T 1998 Interleukin-lbeta-
induced rat pancreatic islet nitric oxide synthesis requires both the p38 and

extracellular signal-regulated kinase 1/2 mitogen-activated protein kinases. J Biol
Chem 273115294-300.

Asfari M, Janjic D, Meda P, Li G, Halban PA, Wollheim CB 1992
Establishment of 2-mercaptoethanol-dependent differentiated insulin- secreting cell
lines. Endocrinology 1301167-78.

Janjic D, Wollheim CB 1992 Effect of 2-mercaptoethanol on glutathione levels,

cystine uptake and insulin secretion in insulin-secreting cells. Eur J Biochem
2101297-304.

Trautmann ME, Blondel B, Gjinovci A, Wollheim CB 1990 Inverse relationship
between glucose metabolism and glucose-induced insulin secretion in rat
insulinoma cells. Horm Res 34: 75- 82

Heimberg H, De Vos A, Pipeleers D, Thorens B, Schuit F 1995 Differences in
glucose transporter gene expression between rat pancreatic alpha- and beta-cells are

correlated to differences 1n glucose transport but not in glucose utilization. J Biol
Chem 27018971-5.

Waeber G, Thompson N, Haefliger JA, Nicod P 1994 Characterization of the
murine high Km glucose transporter GLUT2 gene and its transcriptional regulation
by glucose in a differentiated insulin-secreting cell line. J Biol Chem 269126912-9.

Inagaki N, Yasuda K, Inoue G, Okamoto Y, Yano H, Someya Y, Ohmoto Y,
Deguchi K, Imagawa K, Imura H, et al. 1992 Glucose as regulator of glucose

transport activity and glucose- transporter mRNA in hamster beta-cell line. Diabetes
41 1592-7.

Yasuda K, Yamada Y, Inagaki N, Yauo H, Okamoto Y, Tsuji K, Fukumoto H,
Imura H, Seino S, Seino Y 1992 Expression of GLUTl and GLUT2 glucose

transporter isoforms in rat islets of Langerhans and their regulation by glucose.
Diabetes 41 :76-81.

174

 

 

222.

223.

224.

225.

226.

227.

228.

229.

230.

231.

232.

233.

234.

Kieffer TJ, Habener JF 1999 The glucagon-like peptides. Endocr Rev 20:876-
913.

Drucker DJ 1998 Glucagon-like peptides. Diabetes 471159-69.

Skoglund G, Hussain MA, Holz CG 2000 Glucagon-like peptide 1 stimulates
insulin gene promoter activity by protein kinase A-independent activation of the rat
insulin I gene cAMP response element. Diabetes 49:1156-64.

Wang X, Cahill CM, Pineyro MA, Zhou J, Doyle ME, Egan JM 1999
Glucagon-like peptide-1 regulates the beta cell transcription factor, PDX-l, in
insulinoma cells. Endocrinology 140:4904-7.

Kemp DM, Habener J F 2001 Insulinotropic hormone glucagon-like peptide 1
(GLP-l) activation of insulin gene promoter inhibited by p38 mitogen-activated
protein kinase. Endocrinology 142:1 179-87.

Bryan J, Aguilar-Bryan L 1997 The ABCs of ATP-sensitive potassium channels:
more pieces of the puzzle. Curr Opin Cell Biol 91553-9.

Moritz W, Leech CA, Ferrer J, Habener JF 2001 Regulated expression of
adenosine triphosphate-sensitive potassium channel subunits in pancreatic beta-
cells. Endocrinology 142:129-38.

Tokuyama Y, Fan Z, Furuta H, Makielski JC, Polonsky KS, Bell GI, Yano H
1996 Rat inwardly rectifying potassium channel Kir6.2: cloning
electrophysiological characterization, and decreased expression in pancreatic islets
of male Zucker diabetic fatty rats. Biochem Biophys Res Commun 2201532-8.

Purrello F, Vetri M, Vinci C, Gatta C, Buscema M, Vigneri R 1990 Chronic
exposure to high glucose and impairment of K(+)-channel function in perifused rat
pancreatic islets. Diabetes 391397-9.

Roche E, Assimacopoulos-Jeannet F, Witters LA, Perruchoud B, Yaney G,
Corkey B, Asfari M, Prentki M 1997 Induction by glucose of genes coding for
glycolytic enzymes in a pancreatic beta-cell line (INS-1). J Biol Chem 27213091-8.

Purrello F, Buscema M, Rabuazzo AM, Caltabiano V, Forte F, Vinci C, Vetri
M, Vigneri R 1993 Glucose modulates glucose transporter affinity, glucokinase
activity, and secretory response in rat pancreatic beta-cells. Diabetes 42:199-205.

Marynissen G, Leclercq-Meyer V, Seuer A, Malaisse WJ 1990 Perturbation of
pancreatic islet function in glucose-infused rats. Metabolism 39:87-95.

Lewis B, Mancini M, Mattock M, Chait A, Fraser TR 1972 Plasma triglyceride
and fatty acid metabolism in diabetes mellitus. Eur J Clin Invest 21445-53.

175

 

235.

236.

237.

238.

239.

240.

241.

242.

243.

244.

245.

246.

Porte D, Jr. 1991 Banting lecture 1990. Beta-cells in type H diabetes mellitus.
Diabetes 401166-80.

Brun T, Assimacopoulos-Jeannet F, Corkey BE, Prentki M 1997 Long-chain
fatty acids inhibit acetyl-CoA carboxylase gene expression in the pancreatic beta-
cell line INS-l. Diabetes 461393-400.

Tiedge M, Lortz S, Munday R, Lenzen S 1999 Protection against the co-operative
toxicity of nitric oxide and oxygen free radicals by overexpression of antioxidant
enzymes in bioengineered insulin-producing RINmSF cells. Diabetologia 42:849-
55.

Gazdar AF, Chick WL, Oie HK, Sims HL, King DL, Weir GC, Lauris V 1980
Continuous, clonal, insulin- and somatostatin-secreting cell lines established flour a
transplantable rat islet cell tumor. Proc Natl Acad Sci U S A 77:3519-23.

Chick WL, Warren S, Chute RN, Like AA, Lauris V, Kitchen KC 1977 A
transplantable insulinoma in the rat. Proc Natl Acad Sci U S A 741628-32.

Bast A, Wolf BA, Oberbaumer I, Walther R 2002 Oxidative and nitrosative
stress induces peroxiredoxins in pancreatic beta cells. Diabetologia 451867-876

Kang SW, Chae HZ, Seo MS, Kim K, Baines IC, Rhee SG 1998 Mammalian
peroxiredoxin isoforms can reduce hydrogen peroxide generated in response to
growth factors and tumor necrosis factor-alpha. J Biol Chem 27316297-302.

Verdoucq L, Vignols F, Jacquot JP, Chartier Y, Meyer Y 1999 In vivo
characterization of a thioredoxin h target protein defines a new peroxiredoxin
family. J Biol Chem 274119714-22.

Chen L, Xie QW, Nathan C 1998 Alkyl hydroperoxide reductase subunit C
(AhpC) protects bacterial and human cells against reactive nitrogen intermediates.
Mol Cell 11795-805.

Moriscot C, Patton F, Kerr-Conte J, Richard MJ, Lemarchand P, Benhamou
PY 2000 Contribution of adenoviral-mediated superoxide dismutase gene transfer
to the reduction in nitric oxide-induced cytotoxicity on human islets and INS-l
insulin-secreting cells. Diabetologia 431625-31.

Svensson C, Hellerstrom C 1991 Long-term effects of a high glucose
concentration in vitro on the oxidative metabolism and insulin production of
isolated rat pancreatic islets. Metabolism 401513-8.

Trejo J, Chambard JC, Karin M, Brown JH 1992 Biphasic increase in c-jun
mRNA is required for induction of AP-l- mediated gene transcription: differential
effects of muscarinic and thrombin receptor activation. Mol Cell Biol 1214742-50.

176

 

 

247.

248.

249.

250.

251.

252.

253.

254.

255.

256.

257.

258.

Derijard B, Hibi M, Wu IH, Barrett T, Su B, Deng T, Karin M, Davis RJ 1994
JNKl: a protein kinase stimulated by UV light and Ha-Ras that binds and
phosphorylates the c-Jun activation domain. Cell 76:1025-37.

Dickens M, Rogers JS, Cavanagh J, Raitano A, Xia Z, Halperu JR, Greenberg
ME, Sawyers CL, Davis RJ 1997 A cytoplasmic inhibitor of the JNK signal
transduction pathway. Science 2771693-6.

Gallo KA, Mark MR, Scadden DT, Wang Z, Gu Q, Godowski PJ 1994
Identification and characterization of SPRK, a novel src-homology 3 domain-
containing proline-rich kinase with serine/threonine kinase activity. J Biol Chem
269:15092-100.

Schreiber E, Matthias P, Muller MM, Schaffner W 1989 Rapid detection of
octarner binding proteins with 'mini-extracts', prepared from a small number of
cells. Nucleic Acids Res 1716419.

Eble AS, Thorpe SR, Baynes JW 1983 Nonenzymatic glucosylation and glucose-
dependent cross-linking of protein. J Biol Chem 25819406-12.

Shiba T, Inoguchi T, Sportsman JR, Heath WF, Bursell S, King GL 1993
Correlation of diacylglycerol level and protein kinase C activity in rat retina to
retinal circulation. Am J Physi012651E783-93.

Inoguchi T, Battan R, Handler E, Sportsman JR, Heath W, King GL 1992
Preferential elevation of protein kinase C isoform beta II and diacylglycerol levels
in the aorta and heart of diabetic rats: differential reversibility to glycemic control
by islet cell transplantation. Proc Natl Acad Sci U S A 89:11059-63.

Craven PA, Davidson CM, DeRubertis FR 1990 Increase in diacylglycerol mass
in isolated glomeruli by glucose from de novo synthesis of glycerolipids. Diabetes
391667-74.

Kinoshita JH, Merola LO, Tung B 1968 Changes in cation permeability in the
galactose--exposed rabbit lens. Exp Eye Res 7180-90.

Kasai H 1997 Analysis of a form of oxidative DNA damage, 8-hydroxy-2'-
deoxyguanosine, as a marker of cellular oxidative stress during carcinogenesis.
Mutat Res 387:147-63.

Uchida K, Szweda LI, Chae HZ, Stadtman ER 1993 Irnmunochemical detection
of 4-hydroxynonenal protein adducts in oxidized hepatocytes. Proc Natl Acad Sci U
S A 90:8742-6.

Krippeit-Drews P, Kramer C, Welker S, Lang F, Ammon HP, Drews G 1999
Interference of H202 with stimulus-secretion coupling in mouse pancreatic beta-
cells. J Physiol 514:471-81.

177

259.

260.

261.

262.

263.

264.

265.

266.

267.

268.

269.

270.

Hibi M, Lin A, Smeal T, Minden A, Karin M 1993 Identification of an
oncoprotein- and UV-responsive protein kinase that binds and potentiates the c-Jun
activation domain. Genes Dev 712135-48.

Kyriakis JM, Banerjee P, Nikolakaki E, Dai T, Rubie EA, Ahmad MF, Avruch
J, Woodgett JR 1994 The stress-activated protein kinase subfamily of c-Jun
kinases. Nature 369:156-60.

Han J, Lee JD, Bibbs L, Ulevitch RJ 1994 A MAP kinase targeted by endotoxin
and hyperosmolarity in mammalian cells. Science 2651808-1 1.

Rouse J, Cohen P, Trigon S, Morange M, Alonso-Llamazares A, Zamanillo D,
Hunt T, Nebreda AR 1994 A novel kinase cascade triggered by stress and heat
shock that stimulates MAPKAP kinase-2 and phosphorylation of the small heat
shock proteins. Cell 7811027-37.

Davis RJ 2000 Signal transduction by the JNK group of MAP kinases. Cell
1031239-52.

Whitmarsh AJ, Cavanagh J, Toumier C, Yasuda J, Davis RJ 1998 A
mammalian scaffold complex that selectively mediates MAP kinase activation.
Science 28111671-4.

Bonny C, Nicod P, Waeber G 1998 181, a J IP-l-related nuclear protein present in
insulin-secreting cells. J Biol Chem 273: 1843-6.

Waeber G, lDelplanque J, Bonny C, Mooser V, Steinmann M, Widmann C,
Maillard A, Miklossy J, Dina C, Hani EH, Vionnet N, Nicod P, Boutiu P,
Froguel P 2000 The gene MAPKSIPI, encoding islet-brain-l, is a candidate for
type 2 diabetes. Nat Genet 241291-5.

Rana A, Gallo K, Godowski P, Hirai S, Ohno S, Zon L, Kyriakis JM, Avruch J
1996 The mixed lineage kinase SPRK phosphorylates and activates the stress-
activated protein kinase activator, SEK-l. J Biol Chem 271119025-8.

Merritt SE, Mata M, N ihalani D, Zhu C, Hu X, Holzman LB 1999 The mixed
lineage kinase DLK utilizes MKK7 and not MKK4 as substrate. J Biol Chem
274110195-202.

Tibbles LA, lug YL, Kiefer F, Chan J, Iscove N, Woodgett JR, Lassam NJ
1996 MLK-3 activates the SAPK/JNK and p38/RK pathways via SEKl and
MKK3/6. Embo J 15:7026-35.

Derijard B, Raingeaud J, Barrett T, Wu IH, Han J, Ulevitch RJ, Davis RJ

1995 Independent human MAP-kinase signal transduction pathways defined by
MEK and MKK isoforms. Science 267:682-5.

178

271.

272.

273.

274.

275.

276.

277.

278.

279.

280.

281.

282.

283.

Saldeen J, Lee JC, Welsh N 2001 Role of p38 mitogen-activated protein kinase
(p38 MAPK) in cytokine- induced rat islet cell apoptosis. Biochem Pharrnacol
61:1561-9.

Elrick LJ, Docherty K 2001 Phosphorylation-dependent nucleocytoplasmic
shuttling of pancreatic duodenal homeobox-1. Diabetes 5012244-52.

Ohneda K, Ee H, German M 2000 Regulation of insulin gene transcription. Semin
Cell Dev Biol 111227-33.

Kaneto H, Sharma A, Suzuma K, Laybutt DR, Xu G, Bonner-Weir S, Weir
GC 2002 Induction of c-Myc expression suppresses insulin gene transcription by
inhibiting NeuroD/BETAZ-mediated transcriptional activation. J Biol Chem
277:12998-3006.

Jonas JC, Laybutt DR, Steil GM, Trivedi N, Pertusa JG, Van de Casteele M,
Weir GC, Henquiu JC 2001 High glucose stimulates early response gene c-Myc
expression in rat pancreatic beta cells. J Biol Chem 276135375-81.

Noguchi K, Kitanaka C, Yamana H, Kokubu A, Mochizuki T, Kuchino Y 1999
Regulation of c-Myc through phosphorylation at Ser-62 and Ser—71 by c- Jun N-
tenninal kinase. J Biol Chem 274:32580-7.

Pessah M, Prunier C, Marais J, Ferraud N, Mazars A, Lallemand F, Gauthier
JM, Atfi A 2001 c-Jun interacts with the corepressor TG-interacting factor (TGIF)
to suppress Smad2 transcriptional activity. Proc Natl Acad Sci U S A 98:6198-203.

Docherty K, Clark AR, Scott V, Knight SW 1991 Metabolic control of insulin
gene expression and biosynthesis. Proc Nutr Soc 50:553-8.

Welsh M, Brunstedt J, Hellerstrom C 1986 Effects of D-glucose, L-leucine, and
2-ketoisocaproate on insulin mRNA levels in mouse pancreatic islets. Diabetes
351228-31.

German M, Ashcroft S, Docherty K, Edlund H, Edlund T, Goodison S, Imura
H, Kennedy G, Madsen O, Melloul D, et al. 1995 The insulin gene promoter. A
simplified nomenclature. Diabetes 44:1002-4.

Melloul D, Ben-Neriah Y, Cerasi E 1993 Glucose modulates the binding of an
islet-specific factor to a conserved sequence within the rat I and the human insulin
promoters. Proc Natl Acad Sci U S A 90:3865-9.

Edlund H 2001 Developmental biology of the pancreas. Diabetes 50 Suppl 11S5-9.

Petersen HV, Serup P, Leonard J, Michelsen BK, Madsen OD 1994
Transcriptional regulation of the human insulin gene is dependent on the
homeodomain protein STFl/IPFl acting through the CT boxes. Proc Natl Acad Sci
USA91110465-9.

179

 

284.

285.

286.

287.

288.

289.

290.

291.

292.

293.

294.

295.

Marshak S, Totary H, Cerasi E, Melloul D 1996 Purification of the beta-cell
glucose-sensitive factor that transactivates the insulin gene differentially in normal
and transformed islet cells. Proc Natl Acad Sci U S A 93:15057-62.

Serup P, Petersen HV, Pederseu EE, Edlund H, Leonard J, Petersen JS,
Larsson Ll, Madsen OD 1995 The homeodomain protein IPF-l/STF-l is
expressed in a subset of islet cells and promotes rat insulin 1 gene expression

dependent on an intact E1 helix-loop-helix factor binding site. Biochem J 3101997-
1003.

Nelson C, Shen LP, Meister A, Fodor E, Rutter WJ 1990 Pan: a transcriptional
regulator that binds chymotrypsin, insulin, and AP-4 enhancer motifs. Genes Dev
411035-43.

German MS, Wang J, Chadwick RB, Rutter WJ 1992 Synergistic activation of
the insulin gene by a LIM-homeo domain protein and a basic helix-loop-helix
protein: building a functional insulin minienhancer complex. Genes Dev 612165-76.

Karlsson 0, Walker MD, Rutter WJ, Edlund T 1989 Individual protein-binding
domains of the insulin gene enhancer positively activate beta-cell-specific
transcription. Mol Cell Biol 91823-7.

Sharma A, Stein R 1994 Glucose-induced transcription of the insulin gene is
mediated by factors required for beta-cell-type-specific expression. Mol Cell Biol
141871-9.

Whelan J, Poon D, Weil PA, Stein R 1989 Pancreatic beta-cell-type-specific
expression of the rat insulin II gene is controlled by positive and negative cellular
transcriptional elements. Mol Cell Biol 913253-9.

Marshak S, Leibowitz G, Bertuzzi F, Socci C, Kaiser N, Gross DJ, Cerasi E,
Melloul D 1999 Impaired beta-cell functions induced by chronic exposure of
cultured human pancreatic islets to high glucose. Diabetes 48:1230-6.

Jorgensen MC, Vestergard Petersen H, Ericson J, Madsen 0D, Serup P 1999
Cloning and DNA-binding properties of the rat pancreatic beta-cell- specific factor
ka6. 1. FEBS Lett 461 1287-94.

Klock G, Strahle U, Schutz G 1987 Oestrogen and glucocorticoid responsive
elements are closely related but distinct. Nature 3291734—6.

DeFronzo RA, Bonadonna RC, Ferrannini E 1992 Pathogenesis of NIDDM. A
balanced overview. Diabetes Care 151318-68.

Bonner-Weir S, Decry D, Leahy JL, Weir GC 1989 Compensatory growth of
pancreatic beta-cells in adult rats after short- term glucose infusion. Diabetes 38:49-
53.

180

296.

297.

298.

299.

300.

301.

302.

303.

304.

305.

Unger RH, Zhou YT, Orci L 2000 Lipotoxicity. In: LeRoith D, Taylor SI, Olefsky
JM (eds) Diabetes Mellitus: a fundamental and clinical text Second ed. Lippincott
Williams and Wilkins, Philadelphia: 132-141

McGarry JD, Dobbins RL 1999 Fatty acids, lipotoxicity and insulin secretion.
Diabetologia 421128-38.

Ritz-Laser B, Meda P, Constant I, Klages N, Charollais A, Morales A, Magnan
C, Ktorza A, Philippe J 1999 Glucose-induced preproinsulin gene expression is
inhibited by the free fatty acid palmitate. Endocrinology 14014005-14.

Du XL, Edelstein D, Rossetti L, Fautus IG, Goldberg H, Ziyadeh F, Wu J,
Brownlee M 2000 Hyperglycemia-induced mitochondrial superoxide
overproduction activates the hexosamine pathway and induces plasminogen

activator inhibitor-1 expression by increasing Spl glycosylation. Proc Natl Acad
Sci U S A 97112222-6.

Kolm-Litty V, Sauer U, Nerlich A, Lehmann R, Schleicher ED 1998 High
glucose-induced transforming growth factor betal production is mediated by the

hexosamine pathway in porcine glomerular mesangial cells. J Clin Invest 1011160-
9.

Nishikawa T, Edelstein D, Du XL, Yamagishi S, Matsumura T, Kaneda Y,
Yorek MA, Beebe D, Oates PJ, Hammes HP, Giardino I, Brownlee M 2000
Normalizing mitochondrial superoxide production blocks three pathways of
hyperglycaemic damage. Nature 4041787-90.

Lenzen S, Bailey CJ 1984 Thyroid hormones, gonadal and adrenocortical steroids
and the function of the islets of Langerhans. Endocr Rev 51411-34.

Orland MJ, Permutt MA 1991 Comparative modulations of insulin secretion,
pancreatic insulin content, and proinsulin mRNA in rats. Effects of 50%
pancreatectomy and dexamethasone administration. Diabetes 40:181-9.

Philippe J, Missotteu M 1990 Dexarnethasone inhibits insulin biosynthesis by
destabilizing insulin messenger ribonucleic acid in hamster insulinoma cells.
Endocrinology 127: 1640-5.

Fischer B, Rausch U, Wollny P, Westphal H, Seitz J, Aumuller G 1990
Irnmunohistochemical localization of the glucocorticoid receptor in pancreatic beta-
cells of the rat. Endocrinology 12612635-41.

181