Stimulation of Glucose Transport: Potential Role of Aktlffrotein Kinase Ba

Elena Bogdanovic

A thesis submitted in conformity with the requirements for the degree of Master of Science,

Graduate Department of Physiology, University of Toronto

O Copyright by Elena Bogdanovic (2000)

National Library 191 of Canada Bibliothbque nationale du Canada

Acquisitions and Acquisitions et Bibliographie Services services bibliographiques

395 Weiiiigton Street 395. rue Wdlirigtm Ottawa ON K l A ôN4 OitawilON K l A W canada canada

The author has granted a non- exclusive licence aüowing the National Library of Canada to reproduce, loan, distribute or seii copies of this thesis in microform, paper or electronic formats.

The author retains ownership of the copyright in this thesis. Neither the thesis nor substantial extracts fiom it may be printed or otherwise reproduced without the author's permission.

L'auteur a accordé une licence non exclusive permettant a la Bibliothèque nationale du Canada de reproduire, prêter, distribuer ou vendre des copies de cette thèse sous la forme de microfiche/film, de reproduction sur papier ou sur format électronique.

L'auteur conserve la propriété du droit d'auteur qui protège cette thèse. Ni la thèse ni des extraits substantiels de celle-ci ne doivent être imprimés ou autrement reproduits sans son autorisation.

Stimulation of Glucose Transport: Potential Role of Aktl/Protein Kinase B a

A thesis by Elena Bogdanovic submitted in conformity with the requirements for the degree of Master of Science, Graduate Department of Physiology, University of

Toronto, January 2000.

ABSTRACT

Aktlffrotein Kinase Bu (AktUPKBa) has been shown to lie downstream of

phosphatidylinositol 3-kinase (PI3K) in the insulin action pathway and has been

implicated in the stimulation of glucose transport and Glut4 translocation by insulin. To

obtain further insight into this role of Akt l/PKBa, two models were used.

Vanadate and pewanadate (pV) are protein tyrosine phosphatase inhibitors that

can stimuiate glucose uptake and Glut4 translocation in L6 myotubes independently of

PI3K. The present study showed that while insutin and pV increase AktlPKBu activity

in a PI3K dependent manner, vanadate did not. These data suggest that Aktl/PKBu is

important for insulin but not for vanadate or pV stirnulated glucose uptake.

The second mode1 exarnined Akt l/PKBcr activation in insuIin resistant rat

adipocytes. Rat adipocytes rendered insulin resistant by chronic exposure to

hyperglycemia and hyperinsulinemia (High GA) showed decreased insulin-stimulated

glucose transport and Glut4 translocation and AktlPKBa activation. This was

associated with normal IRS- 1 tyrosine phosphorylation and IRS- 1 associated p85. Co-

incubation with NAC (N-acetyl-cysteine) prevented the insulin resistance in glucose

uptake, Glut4 translocation and Akt l/PKBa.

Taken together, these findings suggest a role for Akt l/PKBa to mediate insulin-

stimulated glucose transport. However, additional pathways must also exist that cm

promote glucose uptake and Glut4 translocation.

ACKNOWLEDGEMENTS

1 would like to thank my supervisor Dr. George Fantus for giving me the

opportunity to compIete a Master of Science degree in his laboratory under his

supervision, and for his support, both academic and non-academic.

I would like to thank Dr. Amira KJip and individuals in her laboratory, especially

Jean-Francois St. Denis, Toolsie Ramlal, and Romel Somwar for support and technical

assistance which were of immeasurable value.

In addition, I would also like to thank Dr- Jim Woodgett and his laboratory for

technical assistance.

Finally 1 would li ke to acknowledge the individuals in my own laboratory for

training and help.

III

TABLE OF CONTENTS

ABSTRACT ACKNO WLEDGEMENTS TABLE OF CONTENTS LIST OF FIGURES LIST OF ABBREVIATIONS

II III IV

VI1 vnr

............................................................................................................... CHAPTER ONE. 1 INSULIN STIMULATION OF GLUCOSE TRANSPORT....... ...............................W... L

7 1.1 Overview of Glucose Uptake Stimulation by Insulin ............................................. - ............................................................ 1.1.1 The lnsulin Receptor and IRS Proteins 4

.......................................................................... 1.1.2 Phosphatidylinositol 3-Kinase 4 1.1.3 Downstrem Targets of Phoshatidylinositol 3-Kinase ..................................... 5

........................................................................................ 1.1.4 Glucose Transporters 6 1.1.5 Intrinsic Activityof GlucoseTransporters ....................................................... 7 1.1.6 Subcellular Localization ................................................................................... 8

INTRODUCTION TO CHAPTER 2 A N D CHAPTER 3 ............................................. 9

CHAPTER TWO ............................................................................................................. 12 ...... Raie of Aktl/PKBa in Vanadate and Pervanadate Stimulated Glucose Uptake 12

ABSTRACT ................................................................................................................. 13

2.1 Introduction ........................................................................................................... 14

2.2 Materials and Methods ......................................................................................... 16 2.2.1 Materials ........................................................................................................ 16 2.2.2 Ce11 Culture ................................................................................................... 16

............................................................................................ 2.2.3 Dmg Treatments 17 .................................. 2.2.4 Aktl/PKBa Immunoprecipitation and Kinase Assay 18

...................... ..................................................... 2.2.5 Lmmunoblot Analysis ... 18 2.2.6 Statistical Analysis ........................................................................................ 19

............................................................................................................. 2.4 Discussion 21

........................................................................................................ CHAPTER THREE 26 The Antioxidant N-Acetyl-Cysteine Prevents the Impairment in Glucose Transporter Translocation and AktlmKBa Activation in Hyperglycemia and ...................................................................... Glucosamine Induced Insulin Resistance 26

ABSTRACT .................................................................................................................. 27

........................................................................................................... 3.1 Introduction 28 3.1.1 Pathogenesis of Hyperglycemia-Induced Insulin Resistance: The

............................................................... Hexosamine Biosynthesis Pathway 29 3.1.2 Potential Defects in the Insulin Sipaling Pathway ...................................... 32 3.1.3 Mechanisms of Hexosamine lnduced Insulin Resistance ............................. 34 3.1.4 Potential Role of Oxidative Stress in Insulin Resistance .............................. 35

3 -2 Materials and Methods .......................................................................................... 38 3.2.1 Materials ......................................................................................................... 38

.............................................................. 3.2.2 Adipocyte Isolation and Preparation 39 3.2.3 Pnrnary Culture of Adipocytes .................................................................... 39 3.2.4 Subcellular Fractionation ..................... .. ...................................................... 40

................................................................. 3.2.5 Preparation of Whole Cell Lysates 41 3.2.6 P rotein Determination .................................................................................... 41 3.27 Determination of Glut4 Translocation and Glut 1 Levels ............................... 42

............................................................................ 3.2.8 IRS- 1 Immunoprecipi tation 42 3.2.9 Determination of IRS-1 Tyrosine Phosphorylation and Association with p85 ..

..................................................................................................................... 43 ................................................ 3 2.10 Akt l/PKBu Activation and Phosphory lation 44

C) 3.2.1 1 5'-Nucleotidase ............................................................................................. 44 3.2.12 Statistical Analysis ....................................................................................... 45

3.3 Results ................................................................................................................... 46 3.3.1 Effect of NAC on Glut4 Translocation in Cells Treated with Hïgh GA and

.................................................................................................. Glucosamine -46 3.3.2 Effect of NAC on Glutl Expression in Cells Treated with High GA and

................................................................................................... Glucosami ne 49 3.3.3 Effect of High Gfl and NAC on Akt lPKB a Activation and Phosphorylation .

...................................................................................................................... -49 3.3.4 Effect of High GA and NAC on IRS-1 Tyrosine Phosphorylation and

Association with p85 in the LDM .................................................................. 59 3.3.5 Effect of High G/I and NAC on IRS-1 Tyrosine Phosphorylation in Total

.................................................................................................... Ce11 Lysates 62

3.4 Discussion ............................................................................................................. 67

GENERAL DISCUSSION ....~.....................~..w.......w~.~w............~......................m....~....... 71

References .........................o.....................................................o................................ 77

Copyright Release Authorkation.. ...... ... . . . .. .... . . ..

LIST OF FIGURES

CHAPTER ONE

FIGURE 1.1 Stimulation of Glut4 Translocation

CHAPTER TWO

FIGURE S. 1 Effect of Insulin, Vanadate, and pV on Akt l/PKBu Activity FIGURE 2.2 Serine 473 Phosphorylation of Akt 1/PKBa

CHAPTER THREE

FIGURE 3.1 The Hexosamine Biosynthesis Pathway FIGURE 3.2 Glut4 Translocation in Hïgh GA and NAC Treated Adipocytes FIGURE 3.3 Glut4 Translocation in Glc and NAC Treated Adipocytes F I G W 3.4 Effect of NAC Alone on Glut4 Translocation FIGURE 3.5 Effect of High G/I and NAC on Glutl Expression FIGURE 3.6 Effect of Glc and NAC on Glutl Expression FIGURE 3.7 Effect of High G/I and NAC on Aktl/PKBa Activity FIGURE 3.8 Effect of NAC on Aktl/PKBa Activity FIGURE 3.9 Effect of High G/I and NAC on Serine 473 Phosphorylation FIGURE 3.10 Effect of NAC on Serine 473 Phosphorylation of Akt l/PKBu FIGURE 3.11 Effect of High G/I and NAC on Total Levels of Akt l/PKBu FIGURE 3.12 Effect of High G/I and NAC on RS-1 Levels in the LDM FIGURE 3.13 Effect of High Gn and NAC on IRS-1 Tyrosine Phosphorylation

In the W M FIGURE 3.14 Effect of High GA and NAC on IRS-1 Assoçirited p85

In the LDM FIGURE 3.15 Effect of High GO and NAC on iRS-1 Tyrosine Phosphorylation

In Total Ce11 Lysates FIGURE 3.16 Effect of High G/I and NAC on iRS-1 Levels in Total Cell

Lysates

LIST OF ABBREVIATIONS

AMP ANOVA ATP BSA CAMP DAG DMEM DNA ECL EDTA EGTA FBS GFA Glc GLDH Glut GSH HBP HBS HDM HEPES High G/I IRS KRBH LDM MAPK MAPKAP cc-MEM MOPS NAC NAD NADH NIDDiM NIH NMR 8-OHdG PDGF PDK PH Pi PI3 K PKB

adenosine monophosphate analysis of variance adenosine triphosphate bovine serum albumin cyclic adenosi ne monophosphate diac ylgl ycerol Dulbecco's Modified Eagles Medium deoxyribonucleic acid enhanced cherniluminescence eth ylenediaminetetraacetic acid ethylenebis(oxyethylenenitrilo)tetraacetic acid fetal bovine semm Glutamine: fructose-6-phosphate amidotransferase gl ucosamine L-glutamate deh ydrogenase glucose transporter reduced glutathione hexosamine biosynthesis pathway HEPES buffered sali ne high-density microsome N-[2-Hydroxye<hyl]pipera~ine-N'-[2-ethanesulfonic acid] combined high glucose and high insulin concentrations insulin receptor substrate Kreb's Ringer Bicarbonate HEPES buffer low-densi ty microsome mitogen activated protein kinase mitogen activated protein kinase activated protein alpha Minimum Essential Medium (3-[N-Morpholino]propanesulfonic acid] N-acetyl-cysteine nicotinamide adenine dinucleotide reduced NAD non-insulin dependent diabetes me1 litus National Institute of HeaIth nuclear magnetic resonance 8-hydroxydeoxyguanosine platelet denved growth factor phosphatidylinositol dependent protein kinase plec kstrin homology inorganic phosphate phosphatidylinositol 3-kinase protein kinase B

PKC PM Ptdins-3-P Ptdhs-3 ,4-Pz PtdIns-3,4,5-P3 PtdIns-3.5-Pz PTP PV RNA SDS-PAGE SE SH2

protein kinase C plasma membrane phosphatidylinositol 3-phosphate phosphatidylinosi tol 3.4-diphosphate phosphatidylinositol3,4,5-triphosphate phosphatidylinositol3.5-diphosphate protein tyrosine phosphatase pervanadate nbonucleic acid sodium dodecyl sulphate-polyacrylamide gel electrophoresis standard error src hornotogy 2

CHAPTER ONE

INSULIN STIMULATION OF GLUCOSE TRANSPORT

CHAPTER ONE

INSULIN STIMULATION OF GLUCOSE TRANSPORT

Insulin, a peptide hormone, was first discovered by Frederick Banting and Charles

Best in 1921 (1). Insulin is an anabolic hormone and the principal regulator of energy

metabolism (2). One action of insulin is to promote glucose entry into muscle and

adipose tissue (2). Insulin increases glucose entry into these tissues by promoting the

translocation of glucose transporters from intracellufar low-density microsornes (LDM)

to the plasma membrane (PM) (3). Once recruited, the glucose transporters insert into the

PM and are likely activated (4) allowing the uptake of glucose into the cell. This action

of insulin is particularly important for glucose homeostasis in normal physiology. This is

evident in diseases such as Type 2 diabetes in which the decreased ribility of insulin to

promote glucose uptake in target tissues is a major defect and may contribute to the

development of the disease (2,4).

How the activation of the insulin receptor promotes glucose transporter

translocation and glucose uptake is not entirely known but several signahg proteins have

been i mpl icated.

1.1 Overview of Glucose Uptake Stimulation by Insulin

Full stimulation of glucose transport by insulin requires translocation of glucose

transporters (Gluts) from internai membranes to the PM (3), associated with an increase

in intnnsic activity of the transporters (4). Insulin stimulates both processes likely by

parallel pathways (4), although more is known about the pathway that leads to glucose

transporter translocation (Fig. 1.1).

IRS PROTEINS

INSULIN RECEPTOR

1 PIJ-KINASE 1

GLUT4

PKB/Akt ATYPICAL PKCs

GLUT4 TRANSLOCATION

FIG. 1.1 STIMULATION OF GLUT4 TRANSLOCATION BY INSULIN

3

1- 1- 1 The I~zsrtlNt Recepfor and IRS Proreins

The insulin receptor contains two a and two P subunits ( 5 ) . The u subunîts

are located entirely on the extracellular side of the plasma membrane and contain the

insulin binding sites (5). The B subunits are transmembrane proteins that contain tyrosine

kinase activity in the cytoplasmic portion (5). lnsulin binding to the cjr subunit leads to

the activation of the receptor tyrosine kinase. Once active. the insulin receptor kinase

phosphorylates numerous substrates on tyrosine residues which then function as docking

proteins (5). There are at least four insulin receptor substrates (IRSs) identified, IRS-1

(6), IRS-2 (7), IRS-3 (8), and IRS-4 (9). Once phosphorylated on tyrosine residues, the

IRS proteins recruit and activate Src homology 2 domain (SH2) containing proteins. For

the stimulation of glucose transport, the p85 subunit of Class IA phoshatidylinositol 3-

kinases appears to be the central protein recmited to the tyrosine phosphorylated IRS

proteins (l0,ll).

1.1.2 Plzospharidylit~ositol3- Kitzase

Phosphatidylinositol 3-kinases (PI3K) are a family of lipid kinases that catalyze

the phosphorylation of phosphoinositides on the 3-position of the inositol ring (12). Four

different lipid products of the P13K family have been identified; PtdIns-3-P, PtdIns-3,4-

PI, PtdIns-3,s-PI, and PtdIns-3,4,5-P; (12). The Class IA P13K have been demonstrated to

play a role in the stimulation of glucose uptake by insulin, most likely by regulating the

trafficking of glucose transporter containing vesicles (13). Class IA PI3K are composed

of two subunits: the p85 regulatory subunit, and the pl 10 catalytic subunit (12). The p85

subunit contains two SH2 domains which bind to specific tyrosine phosphorylated motifs

on the R S proteins (10) leading to the activation of the pl IO catalytic subunit (14). in

vitro, p l 10 can catalyze the formation of the above mentioned lipids but i)t vivo only

PtdIns-3.4-Pz and PtdIns-3,4,5-P3 levels are increased by insulin and other extracellular

signals (12). Ptdns-3-P is constitutively present inside the ce11 (12).

1.1.3 Downsrrearn Targets of Pl1osllaridylinosiral3-Kinase

There are 2 important serinehhreonine kinases that have been shown to act

downstream of PI3K and may play a rote in insulin-stimulated glucose transporter

translocation. The first kinase is AktPPKB (15). There are 3 isoforms discovered to date:

Akt 1PKBa (16- 18), Akt2/PKBB (19), and Akt3PKBy (20). AI 1 isoforms possess an N-

terminal pleckstrin homology (PH) domain, followed by a catalytic domain and a C-

terminal taii (21). Akt3/PKBy is truncated at the C terminal compared to the other 2

isoforms (20). Full activation of Akt/PKB requires phosphorylation on threonine

(Thr308 on Akt l/PKBcc, Thr309 on AktZPKBP, Thr305 on Akt3RKBy) and senne

residues (Ser473 on AktlPKBa, Ser474 on Akt2PKBQ) (22). Phosphorylation of

these residues has been shown to be dependent on PUK (21). The enzyme which

phosphorylates the threonine residue has been purified from rabbit skeletal muscle and

shown to be dependerit on PtdIns-3,4-Pz and PtdIns-3.4,5,-P3 for activation. It was

therefore termed phosphoinositide-dependent protein kinase 1 (PDK 1 ) (23). Li ke

Akt/PKB, PDKl also contains a PH domain (34). The kinase which phosphorylates the

senne residue, termed PDK2, has not yet been identified but it will likely be regulated in

the same manner as PDKl and also contain a PH domain (23). The mechanism of

Akt/PKB activation via PI3K is not entirely understood, but a mode1 has been proposed

for the activation of Akt/PKB irl vivo. In this mode!, Ptdns-3,4-Pz and PtdIns-3,4,5-P3

recruit Akt/PKB to the plasma membrane (24). This is accomplished by the presence of

the PH domain in Akt/PKB (24). Once at the PM, PtdIns-3,4-Pz and Ptdns-3,4,5-P3 may

promote conformational changes in Akt/PKB which would allow PDKl and PD= to

phosphoryiate and activate the kinase (12).

The second serinekhreonine kinase that has been implicated in the stimulation of

glucose transporter translocation by insulin is a group of enzymes known as the atypical

PKCs, PKCE and PKCh (25). Insulin activation of PKCC and PKCh has been shown to

be dependent on increased PtdIns-3,4,5-P3, and PDK- 1 dependent phosphorylation of

activation loop sites in PKCC and PKCA (26). Adipocytes transiently transfected with

kinase inactive mutants of PKCC and PKCA displayed a 40-60% inhibition of insulin-

stimulated epitope-tagged Glut translocation (25) .

The glucose transporter (Glut) family belonts to a large superfamily of proteins

involved in transport of hexoses and other carbon compounds (27). The gIucose

transporters are transmembrane proteins that transport glucose down its concentration

gradient. There are six Glut isofoms: Glut 1, Glut2, Glut3, Glut4, Glut5, and Glut7,

which are specifically expressed in various tissues and are ail products of different genes

(27).

Glut4 is the insulin responsi ve gIucose transporter and accounts for the increases

in glucose uptake observed with insulin stimulation. It is highly expressed in skeletal

muscle, heart and adipocytes (28). Inside the cell, the majority of Glut4 is contstined in

the high- and low- density microsomes (29). In the basal state, Glut4 containing vesicles

continuously cycle between the low-density microsomes and the plasma membrane (30).

Upon stimulation with insulin, the exocytosis of Glut4 vesicles is greatly increased, with

a smaller decrease in Glut4 vesicle endocytosis (31). The net result is an increase in the

plasma membrane content of Glut4. Glutl is expressed in al1 tissues (27). In muscle

ceils and adipocytes, Glutl is constitutively present in the plasma membrane (27) ruid

provides the ce11 with glucose under basa1 conditions (27).

I. 1.5 Iirtrinsic Activity of GIztcose Tramponers

Several lines of evidence suggest that insulin stimulates another pathway(s) which

leads to increased intrinsic activity of the transporters. First, Katagiri et al. (32) have

shown a discrepancy between PI3K activation and glucose transport activity. In this

study, a dominant negative p85a regulatory subunit (Ap85u) of PUK was overexpressed

in 3T3-LI adipocytes. Overexpression of Ap85a markedly inhibited insulin stimulated

PUK activity and glucose transporter translocation, but the inhibition of glucose uptake

was less rernarkable. In another study (33), the addition of exogenous PtdIns-3,4,5-P;

lipids to 3T3-L1 adipocytes largely restored hexose uptake blocked by the PI3K inhibitor

wortmannin but treatment of basal cells with PtdIns-3,4,5-P3 alone did not stimulate

hexose uptake. Recently, Sweeney et al. (4) have shown that insulin activates the p38

mitogen activated protein (MAP) kinase in 3T3-LI adipocytes and L6 myotubes.

Preincubation of 3T3-L1 adipocytes and L6 rnyotubes with the p38 MAP kinase

inhibitor SB203580 inhibited insulin stirnulated glucose uptake but did not affect glucose

transpofler translocation. SB203580 had no effect on CRS-1 tyrosine phosphorylation or

its association with p85. AktlPKBa, Akt2/PKB$, or Akt3fPKBy. Together these

findings suggest the presence of another siugnaling pathway(s), separate from the one

described above, that involves p38 MAP kinase and results in the increased intnnsic

activi ty of glucose transponers (4).

1.1.6 Sr~bcellrilur Localizcction

For insulin to stimulate Glut4 translocation, it appears that the proteins involved

in insulin signal transduction must be activated in the correct subcellular fraction (13).

For example, platelet-derived growth factor (PDGF) increases PDK activity in 3T3-L1

adipocytes but does not activate glucose transport (34). This observation is explained by

the failure of PDGF to activate PI3K in the LDM (13). Kelly et al. (35) have shown that

insulin predominately activates P13K in the LDM fraction, and to a lesser extent in the

PM. IRS-1 and IRS-2, located mainly in the LDM in rat adipocytes, have been

implicated in the stimulation of glucose transport by insulin (36). whereas ES-3 located

in the PM was not essential for glucose transport in these cells (37). In addition, the

majonty (75%) of anti-phosphotyrosine immunoprecipitable PUK activity was found in

the LDM in response to insulin, with on1 y a minor portion found in the PM (35).

Cytosolic PI3K activity was not increased in response to insulin, nor was it

imrnunoprecipitable with anti-phosphotyrosine antibodies (35). Together these

observations suggest that for insulin to stimulate glucose transport, the subcellular

distribution, and not just the presence of the proteins is important (22) .

INTRODUCTION TO CHAPTER 2 AND CHAFTER 3

The discovery in 1995 that Akt/PKB can be activated by insulin via P13K (15) has

led to the implication of this enzyme in a variety of insulin stimulated processes (22).

One such process is the stimulation of Glut4 translocation by insulin. There are several

lines of evidence which support a role for Akt/PKB in the translocation of Glut4. First,

Akt/PKB becomes constitutively active upon the addition of a src myristoylation signal

which targets the kinase to the membrane (38). Kohn er al. (38) have shown that

expression of such a constitutively active Akt/PKB where the PH domain was replaced

by the src mynstoylation signal stimulated glucose uptake and Glut4 transiocation to the

same extent as insulin in basal 3T3-Ll adipocytes. Similar results were found using

another constitutive1 y active mutant Akt/PKB, Gag-PKB (39,40,4 1). Gag-PKB is created

by fusion of the Gag protein with the PKB N-terminus. This construct corresponds to the

constitutively active retroviral protein AKT8 (15). Gag-PKB is also targeted to the

membranes (39). Transfection of Gag-PKB into rat adipocytes (39) and L6 muscle cells

(41) increased the ceIl surface content of Glut4 in the absence of insulin. In addition,

transfection of Gag-PKB into L6 muscle ceils and 3T3-L1 adipocytes increased basal

glucose uptake (40). The second line of evidence to support a role for Akt/PKB in the

stimulation of Glut4 translocation by insulin came from using the dominant-negative

inhibitor AAA-PKB (41), a mutant AktlPKBa which has alanine residues substituted at

the two phosphorylation sites (Thr308 and Ser473) and at the ATP binding site (Lys 179).

This mutant is kinase inactive and phosphorylation deficient. Wang et al. (41) have

shown that AAA-PKB inhibited the activation of CO-transfected wild type and

endogenous Akt/PKB by insulin in L6 muscle cells. AAA-PKB markedly inhibited the

increase in Glut4 at the cell surface in response to insulin. In these same cells, CO-

transfection and overall expression of wild type PKB restored Glut4 translocation.

Additionally, ~ k t 3 P K B p has been shown to CO-localize with Glut4 containing vesicles

in the basal state and upon insulin stimulation the level of Akt?/PKBB in the Glut4

vesicles increased, suggesting translocation (29). Kupriyanova et al. (42) also showed

that .i\kt2ffKBp was recmited to the Glut4 containing vesicles in response to insulin and

phosphorylated several vesicle proteins including Glut4. Taken together these studies

implicate AktPKB in the signa! transduction pathway leading to Glut4 translocation in

response to insulin.

The hypothesis tested in this thesis are (1) that Aktl/PKBcc play a critical role in

insulin-stimulated glucose transport and Glut4 translocation, and (2) that this enzyme also

plays a role in vanadate and pervanadate stimulated glucose transport. The studies

described in Chapter 2 and Chapter 3 use 2 approaches and two different modeis to test

these hypotheses and determine the importance of Akt ~ / P K B u in the stimulation of Glutcl

translocation.

The specific aim of Chapter 2 was to determine the potential role of Aktl/PKBa

in vanadate and pervanadate stimulated Glut4 translocation in skeletal muscle. Vanadate

and pervanadate (pV) are protein tyrosine phosphatase (PTP) inhibitors which can mimic

the action of insulin to stimulate Glut4 translocation and glucose uptake in these cells

(43)-

In Chapter 3, the activity of Akt lPKBa was determined in an established rnodel

of insulin resistance in which insulin-stirnulated glucose transport is defective. The

specific aims of Chapter 3 were to:

1) determine if the defect in insulin-stimulated Glut4 translocation found in rat

adipocytes rendered insulin resistant by treatment with high gIucose and

insulin, was associated with a concomitant defect in insulin stimulation of

Akt l/PKBa acti vi ty,

2) detemine whether N-acetyl-cysteine (NAC), an antioxidünt and glutathione

(GSH) precursor (a), which had previously k e n found to prevent glucose

transport defects in this rnodef, could also prevent any defect in Akt l/PKBu

activi ty,

3) investigate the upstream activators of AktUPKBa, namely iRS-1 and IRS-1

associated p85 in cells treated with high glucose/insulin and NAC, to localize

the site of the defect in the insulin signaling pathway,

The results from these studies would provide further insight into the role of Akt l/PKBa

in the stimulation of Glut4 translocation.

CHAPTER TWO

Role of AktlmKBu in Vanadate and Pervanadate Stimulated Glucose Uptake

ABSTRACT

Vanadate and pervanadate (pV) are protein tyrosine phosphatase inhibitors and

insulin mimetic agents. Treatment of L6 myotubes with vanadate and pV increased

glucose transport and glucose transporter (Glut) translocation to the s m e extent as

insulin but independently of PI3K. The serinekhreonine kinase Aktl /PKBu which lies

downstream of PI3K, has been implicated in the stimulation of Glut4 translocation by

insulin. Although Aktl/PKBu is activated by insulin in a PI3K sensitive manner, other

agents can activate the kinase independently of PI3K. Therefore, the present study

exarnined whether Akt l/PKBa was activated by vanadate and pV in L6 myotubes.

Insulin and p V increased the activity of Aktl /PKBa by 2.3M.4 and 17k2.1 fo1d above

control respectively (p<0.05). We previously found that while pretreatment with the

PI3K inhibitor wortmannin inhibited insulin-stimulated glucose transport, it did not

inhibit that stimulated by vanadate o r pV. However, in the present study wonmannin

pretreatrnent prevented the activation of A k t l P K B a by insulin (1.08M.3) and pV

(1.8k0.6). On the other hand, vanadate treatrnent for 30 minutes (1.15M.18) or 60

minutes (O.9L-û. 1) did not significantly activate Akt l f fKBa. These resuIts were

confirmed by immunoblot analysis of Ser373 phosphorylated Akt LPKBcc. Vanadate

and pV stirnulate gIucose transport activity in L6 myotubes by a rnechanism independent

of PI3K and AktUPKBa. The mechanisrn of action of these agents may involve a

separate signaling pathway which converges with that of insulin.

CHAPTER TWO

Role of Aktl/PKBa in Vanadate and Pewanadate Stimulated Glucose Uptake

2.1 Introduction

Vanadate (vol3') is an oxidized forrn of vanadium (45) and stmcturally resernbles

phosphate (46). Vanadate is a protein tyrosine phosphatase (PTP) inhi bitor (47) and has

been shown to mimic the action of insulin on glucose uptake and metabolkm in insulin

target tissues in vitro (48). In vivo, vanadate has been used as a therapeutic agent in

several rodent models of Type 1 and Type II diabetes. and improved glucose control in

human subjects with Type 1 and Type II diabetes (49). Upon mixing vanadate with

hydrogen peroxide (Hz02), a peroxovanadium compound is formed termed pervanadate

(pV) (50). pV is a potent PTP inhibitor, and like vanadate, has been shown to mimic the

metabolic actions of insulin both in vitro and in vivo (5 1,52). Tsiani et al. (53) have

shown that vanadate and pV stimulate glucose transport in cultured L6 myotubes in a

time and dose dependent mrinner.

The mechanism of action of vanadate and pV remains uncertain. The effects of

vanadate or pV on glucose transport in L6 rnyotubes are not additive with insulin

suggesting that these agents activate the same sigaling pathway or different pathways

that converge (53). Treatment of cells with pV stimulates the insulin receptor tyrosine

kinase and increases the overall cellular tyrosine phosphorylation of proteins, including

IRS-1 (54)- most likely due to powerful inhibition of PTPs (51). Vanadate and pV have

been shown to activate PI3K (43). Previous work in our laboratory has s h o w that

vanadate and pV stimulate glucose uptake in L6 rnyotubes to the same extent as insulin

but independently of PUK (43). In this study (43), Li3 myotubes were pretreated with or

without the PI3K inhibitor wortmannin prior to stimulation by insulin, vanadate, or pV.

Wortmannin is a fungal metabolite that binds irreversibly to the pl IO catalytic subunit of

PUK (55) thereby abolishing the ability of the enzyme to produce 3-phophoinositide lipid

products. In the absence of wortmannin, al1 3 agents activated PI3K and stimulated

slucose transporter translocation and glucose uptake. Wortmannin pretreatment inhibited

subsequent PUK activation by al1 three agents but onIy the insulin-stimulated glucose

uptake was inhibited. Vanadate and pV mediated glucose uptake was unaffected by

wortmannin pretreatment. Therefore PUK activation is important for insulin, but not

vanadate or pV mediated glucose transport.

In addition to PI3K, Akt/PKB can aIso be activated by stimuli which do not

activate PI3K. Okadaic acid, a serinekhreonine phosphatase inhibitor, increases

A kt/PKB acti vi ty and glucose transport in 3T3-L l adipoc ytes and isolated rat ridipoc ytes,

but does not activate PI3K (39,56). Cellular stress induzed by heat shock and

hyperosmolarity (57), agents which raise intracellular CAMP levels (58) , and

isoproterenoi (59), have al1 shown to increase Akt/PKB activity in the presence of the

PU K inhi bitor wortmannin. These findings suggest an alternate path way(s) leading to

activation of the kinase.

Since Akt/PKB has been implicated in the stimulation of Glut4 translocation by

i nsulin, and various agents can activate the kinase independently of PI3K, the purpose of

the following study was to detemine whether the insulin mimetic agents vanadate and

pV activated Akr/PKB independently of PI3K in L6 myotubes. The rationale for this

study was based on previous findings in Our laboratory which showed that vanadate and

pV stimulated glucose uptake and glucose transporter translocation in L6 myotubes by a

mechanism independent of PI3K (43).

2.2 Materials and Methods

2.2.1 Materials

a-Minimum essential medium (cc-MEM), fetal bovine serum (FBS), and anti biotics were

obtained from Gibco (Burlington, ON). Sodium orthovanadate and wortmmnin were

purchased from Sigma (St. Louis, MO). Human insulin was a gift from Eli Lilly

(Indianapolis, IN). Antibodies against the PH domain of Akt l f fKBu, and full Iength

PKB/Akt were obtained from Upstate Biotechnology (Lake Placid, NY) and Dr. J.

Woodgett (Ontario Cancer Institute, Toronto, ON) respectively. The antibody

recognizing Senne 173 phosphorylated Akt l/PKBa was purchased from New England

Biolabs (Boston, MA). 3 ' ~ - ~ ~ ~ ( 3 0 0 0 ~ i / m r n o l ) was obtained from Dupont-New

England Nuclear (Boston, MA). The Akt l/PKE%a [mmunoprecipitation and Kinase

Assay Kit was purchased from Upstate Biotechnology. Anti-rabbit secondary antibodies

conjugated to horseradish peroxidase, and the enhanced chemiluminescence (ECL)

reagent were purchased from Arnersham (Oakville, ON). Ni trocel lulose membranes

(0.45pm) were obtained from Schleicher & SchueIl (Keene, NH).

2 - 2 2 Cell Cult~rre

LG myoblasts were grown in 175cm2 flasks in a-MEM containing 10% (vollvol) FBS and

1 Ir, (vol/vol) antibiotic-antimycotic (10,000 U/ml penicillin G, 10,000 p g m l

streptomycin, 25 p@ml amphotericin B). Upon reaching confluence the cells were

trypsinized (0.25% trypsin) and transferred to 15cm petri dishes where they were grown

in 2% (vol/vol) FBS and 1% (vol/voI) antibiotic-antimycotic. The cells were harvested

once they had differentiated into myotubes.

22.3 Dnrg Trearmerzts

A 25mM stock solution of vanadate was prepared in a 50mM HEPES buffer. pV was

prepared by rnixing LOOW of vanadate from the stock with lOOpM H?Oz. Prior to the

drug treatments, the cells were washed twice with HEPES-buffered saline (HBS)

(140mM NaCI, 5mM KCI, 20rnM HEPES, 2.5mM MgS04, ImM CaCI?, pH 7.4) and

preincubated in the presence or absence of lûûnM wortmannin for 10 minutes. This

concentration of wortmannin has been shown to effectively inhibit PUK activity and the

production of PtdIns-39-Pz and PtdIns-3-4,s-P3 (43). Followi ng the preincubation, the

cells were stimulated with either lOOnM insulin for 5 minutes, 5mM vanadate for 30 or

60 minutes, or 10Op~M pV for 15 minutes. All incubations were done at 37°C in HBS

containing 5m1M glucose. The controt sample was incubated for 25 minutes. Fol lowing

the treatments, the cells were washed two times with ice-cold HBS, scraped, and lysed

with Buffer A (5OmM Tris-HC1 pH 7.5, IrnM EDTA, IrnM EGTA, 0.5mM Na3VOa,

O. 1% 2-mercaptoethanol, 1 % Tri ton X- 100, 50mM NaF, 5mM Na-pyrophosphate, lOmM

sodium P-gl ycerophosphate. O. 1 mM phen ylrneth ylsul fon yl fluoride, 1 pgml of aprotini n,

pepstatin, leupeptin, and l p M microcystin). The lysates were centrifuged 10,000 x g for

10 minutes. The supernatant was extracted, snap frozen in liquid nitropen, and stored at -

80°C until further analysis.

2.2.3 Akrl/PKBar I~nnzrtnoprecipitution and Kitlase Assay

Akt/PKB immunoprecipitation and kinase activity was detennined frorn ce11 1 ysates.

Briefly, Aktl/PKBa was immunoprecipitated from ce11 lysates by incubating 200pg of

cell lysate with 4pg of antibody against the PH domain of AktlPKBcx bound to Protein

G agarose for 90 minutes. Following the incubation, the samples were centrifuged for 3

minutes at 10,000 X g. The immunocornplexes were then washed 3X with 500pl of

Buffer A (Section 2.2.3) containing 0.5M NaCl, 2X with 50011 of Buffer B (50rnM Tns-

HCI, pH 7.5.0.03% (w/v) Brij-35, O. lrnM EGTA and 0.1% 2-mercaptoethanol), and 1X

with lOOpl of Assay Dilution Buffer (20mM MOPS, pH 7.2,25mM B-glycerol

phosphate, 5rnM EGTA, 1mM sodium orthovanadate, 1mM dithiothreitol). To determine

kinase activity, the immunoprecipitates were incubated in a shüking water-bath for 10

minutes at 30°C with [$'PI ATP (lOpCi), 20mM MOPS, pH 7.2,25mM P-glycerol

phosphate, SmM EGTA, 5mM sodium orthovanadate, 5mM dithiothreitol, CAMP-

dependent protein kinase inhibitor peptide (lOpM), and the substrate peptide

(RPRAATF) (lOOpM), which is a relatively specific substrate for Akt i/PKBa because it

is not phosphorylated to a significant extent by MAPKAP-Ki or p70 S6 kinase (60).

Following the 10 minute incubation, the samples were centrifuged and the supernatants

were spotted on P8 1 phosphocellulose paper. Radioactivity was determined by

scintillation counting.

2.2.5 Ir~znrrrrzoblor Analysis

Cells were serum starved for 3 hours prior to stimulation. Cell lysates (60pg) were

subjected to SDS-PAGE (10% polyacrylamide) followed by electro transfer of proteins to

nitrocellulose. Following the transfer, the membranes were incubated for 1 hour at 22°C

in Tris buffered saline (TBS) (25mM Tris, pH 7.5, 154m.M NaCl) containing O. L%

Tween 20 and 5% non-fat dry milk. The membrane was then incubated overnight with a

polyclonal phospho-specific anti4473 antibody to detect active Akt l/PKBa, or with an

antibody against the full length of the kinase to quantify total levels. Following the

overnight incubation the membrane was treated for 1 hour wi th an anti-rabbit secondary

antibody conjugated to horseradish peroxidase. Proteins were then visualized by

enhanced cherniluminescence and quantified by NIH Image.

3.26 Srarisrical Analysis

Results are expressed as MEANS k SE. The significance between groups was

determined using analysis of variance (ANOVA, ~ 4 . 0 5 ) .

2.3 Results

Stimulation with IOOnM insulin for 5 minutes resulted in a 2.30M.37 fold activation

of Akt l/PKBct above basal (Fig. 2.1). Treatment of cells with 5mM vanadate for 30

minutes resulted in a slight but insignificant activation of AktlPKBa, 1.15s. 17 above

control, while stimulation with 5mM vanadate for 60 minutes did not activate

AktUPKBa, 0.90d. 10. pV was a potent stimulator of Akt l/PKBu, 17e.11 fold

activation was observed after incubating the cells with 100pM pV for 15 minutes.

15 - O- wortmannin :

I + wortmannin

Con Ins Van30 Van60

FIG. 2.1 Effect of insulin, vanadate, and pV on Akt l/PKBa activity. L6 myotubes were inc ubated in the presence or absence of 1 OOnM wortmannin for 1 0 minutes. The cells were then stirnulated with either lOOaM insulin for 5 minutes, 5mM vanadate for 30 or 60 minutes, or 100pM pV for 15 minutes. Cells were lysed and Aktl/PKBa activity was determined using the irnrnunoprecipitation b a s e assay described in Section 2.2.4. Insulin and pV increased Akt l/PKBa activity which was inhibited by wortmannin pretreatment. Vanadate had no significant effect. Results are MEANkSE of three to five expenments (Op < 0.05 compared to untreated Control).

Pretreatment with lOOnM wortmannin for 10 minutes blocked Akt l /PKBa activation by

insulin (1 .O8M.X), vanadate (30 minutes, O.86-CI.l l), and pV (L.8Odl.57). Although

not statistically significant, pretreatment of control cells with wortmannin reduced

Akt l/PH3u activity in the basal state (O.88d. 19). Similar results were obtained by

immunoblotting with an anti phospho-Ser473 anti body which recognized the

phosphorylated form of Aktl/PKBa (Fig.2.2).

2.4 Discussion

Insulin stimulates Akt 1 P K B a activity in a time and dose dependent manner (6 1).

The stimulation protocol used in this study (100nM insulin for 5 minutes) would produce

maximal activation of the kinase by the hormone. Pretreatment of cells with wortmannin

inhibited subsequent activation of the kinase by insulin. This finding is well docurnented

and supports Aktl/PKBu as a downstream target of PI3K. Kohn er al. (38) have shown

that inhibition of insulin stimulated Aktl/PKBu activity by using several concentrations

of wortmannin closely paralied the inhibition of PI3K in anti-phosphotyrosine

immunoprecipitates, indicating that Akt l/PKBa activation by insulin is closel y

associated with the activation of P13K.

pV was a potent activator of Akt l/PKBa and induces phosphorylation of

Senne373. This observation is consistent with other reports (62,63). pV has been s h o w

to potently activate and phosphorylate Aktl/PKBa by a wortmannin sensitive mechanism

Wortmannin - O - - + + + + Treatment C 1 V p V C I v PV

FIG. 2.2 L6 myotubes were pretreated in the presence or absence of 100nM wortrnannin followed by the addition of either 1OOn.M insulin for 5 minutes, 5mM vanadate for 30 minutes, or 100pM pV for 15 minutes. Cells were then lysed and proteins (60pg) were separated by SDS-PAGE, transferred to nitrocellulose, and probed with (A) an anti-Ser473 antibody to examine phosphorylated Akt l/PKBa, or (B) an anti-Akt 1 /PKBa antibody to determine total levels. Insulin and pV increased the Ser473 phosphoryiation of Akt l/PKBa, which was prevented by wortmannin. Vanadate treatment had a minimal eEect. Total levels of AktlPKBa were similar in al1 conditions. Relative intensities of stimulation corrected for total Aktl/PKBa detennined by densitometry in arbitrary units; insuiin, 0.64; vanadate, 0.065; pV, 1.32.

in both 3T3 fibroblasts (62) and isolated rat adipocytes (63). In isolated rat adipocytes

(63) pV stimulated Aktl/PKBa activity more potently than vanadate. Taken together, the

results of the present study and previous studies indicate that pV, and to a much lesser

extent vanadate, induce phoshorylation and activation of Akt l/PKBa by ri PI3K

mechanism.

The difference in activation of AktlffKBu by vanadate and pV may be due to the

different mechanism of PTP inhibition by each agent. pV is a much more potent PTP

inhibitor than vanadate (5 l,52). Vanadate, a completely reversi ble inhibitor, inhi bits

PTPs by competing with phosphate for binding at the active site (64). In contrast. pV

inhibits PTPs irreversibly (64) by oxidizing the cysteine residue Iocated at the catalytic

site (65). The irreversible inhibition by pV may account for the fact that pV is much

more potent in mimicking the actions of insulin than vanadate. Fantus et al. (52) have

shown that pV is 10'-l0' times more potent than vanadate in stimulating lipo, uenesis,

protein synthesis, and inhibiting epinephrine-stirnulated lipolysis in isolated rat

adipocytes. In isolated rat skeletal muscle, pV treatment stimulated glucose utilization

significantly more than vanadate alone (66). In addition, due to the different mechanisms

of inhi bition, vanadate and pV may have different FTP specificities. Fantus et al. (52)

have shown that vanadate, but not pV, inhibited tyrosine dephosphorylation of the insulin

receptor P-subunit by alkaline phosphatase. In the same experiment, pV powerful ly

inhibited tyrosine dephosphorylation of the insulin receptor by a crude preparation of

phosphotyrosine phosphatase, while vanadate had minimal effects. In the case of the

present study, pV may be inhibiting a key phosphatase(s) important in regulating the

activation of Akt l/PKBa.

Pretreatment of L6 myotubes with wortmannin inhibited insulin stimulated Glut

translocation and glucose uptake (43). Pretreatment with wortrnannin at the

concentration and duration used in this study effectively inhibited PI3K but did not block

vanadate and pV induced glucose transport or Glut4 translocation (43). suggesting that

Akt i/PKBa is also not essential for these agents to increase glucose transport.

Tt is interesting to note that other stimuli can also increase cellular glucose uptake

by a pathway(s) that does not involve P13K or Akt l/PKBa. Examples are muscle

contraction (67,68), hypoxia (67) and osmotic shock (69). Muscle contraction and

hypoxia have been shown to stimulate Glut4 translocation and glucose uptake as potently

as insulin (67). Lee et al. (67) have shown that wortmannin pretreatrnent blocked insulin

but not contraction or hypoxia stimulated glucose uptake in rat skeletal muscle.

Furthermore, Brozinick et al. (68) have shown that muscle contraction does not activate

Akt l/PKBa suggesti ng that contraction induced increases in glucose transport are not

mediated by Akt/PKB. Chen et al. (69) have shown that osrnotic shock increases glucose

transport and Glut translocation in 3T3-LL adipocytes by a wortmannin insensitive

tyrosine kinase pathway that did not involve Akt/PKB.

In summary, pV, but not vanadate, potently activates AktlPKBa in L6 myotubes

by a wortmannin sensitive mechanism, In contrast to insulin, both vanadate and pV

stimulate glucose transport and Glut translocation by a mechanism independent of PUK

and Aktl/PKBa in L6 myotubes. Since glucose transport is not further stimulated by

these agents above that achieved by maximum concentration of insulin, this mechanism

rippears to involve a separate signaling pathway that converges with the insulin action

pathway (53). This is supponed by the finding that cytochalasin D, an actin cytoskeleton

dismpting agent, inhibited glucose transport stimulation by al1 three agents (53).

CHAPTER THREE

The Antioxidant N-Acetyl-Cysteine Prevents the Impairment in Glucose Transporter Translocation and Aktl/PKBa Activation in Hyperglycemia and

Glucosamine Induced Insulin Resistance

ABSTRACT

To examine the potential role of oxidative stress in high glucose/insulin and

glucosamine induced insulin resistance, nt adipocytes were incubated for 18 h in media

containing 20mM glucose/lOOnM insulin (High Gn) or 5mM glucosamine (Glc). Each

incubation was done in the presence or absence of 5mM N-acetyl-cysteine (NAC). High

GfI and Glc impaired the translocation of Glutl from low-density microsornes (LDM) to

the plasma membrane (PM). which could be prevented by NAC. High GR. but not Glc

treatment, decreued Glut 1 expression in the PM, which was also prevented by NAC.

Funher examination of the insulin signaling cascade in High G/I treated cells revealed

that [RS-1 tyrosine phosphorylation and association with the p85 subunit of P13K in

response to insulin was not impaired, whether examined in the LDM or in total ce11

lysates. In contrast, Aktl/PKBu activation and Ser173 phosphorylation in response to

insulin were signi ficantl y impaired in resistant cells, which was prevented upon co-

incubation with NAC.

In conclusion, NAC. an antioxidant and glutathione (GSH) precursor was able to

prevent both h yperglycemia and glucosamine induced defects in insu1 in action. The

results from this study suggest a role for oxidative stress in the induction of insulin

resistance in this model. NAC may serve as a potential design for novel therapeutic

agents.

CHAPTER THREE

The Antioxidant N-Acetyl-Cysteine Prevents the Impairment in Glucose Transporter Translocation and Aktl/PKBct Activation in Hyperglycemia and

Glucosamine Induced Insulin Resistance

3.1 Introduction

Diabetes mellitus is a metabolic disease affecting 5% of the population

worldwide (2,701- Type 2 diabetes or non-insulin dependent diabetes mellitus (NIDDM)

accounts for over 90% of diabetic patients (2,70) and is charactenzed by (1) insulin

resistance in peripheral target tissues, namely, muscle and fat. (2) decreased pancreatic P

ce11 function, and (3) increased hepatic glucose production (7 1). Chronic complications

of diabetes include kidney failure, lower Iirnb amputations, blindness, and increased risk

of cardiovascular disease and stroke (2,70)-

Genetic and environmental factors contribute to the development of Type 2

diabetes (70). One of the earliest defects observed. and likely a key factor in the

progression of the disease, is insulin resistance (70). Insulin resistance c m be genetic or

acquired or both, and is chardcterized by impaired stimulation of glucose uptake by

insulin in target tissues (71). The exact cause or mechanism of insulin resistance is not

entirely known. Genetic analysis of insulin resistance examined the effects of mutations

in the insulin receptor gene and genes encoding the proteins in the insulin action pathway

(2). Mutations in the insulin receptor gene result in severe insulin resistance, but this

defect accounts for only ri very small percentage of Type II diabetic patients (2) . Type II

diabetes has not been shown to be associated with mutations in genes encoding proteins

in the insulin action pathway (2). At the present time, the genetic cause of insulin

resistance is not known and multiple inhented traits may be important (2). Numerous

factors have been shown to be able to induce insulin resistance in target tissues such as

hormones, cytokines, and metabolites (72). One of them is hyperglycemia (73). in Type

II diabetic patients, hyperglycemia is present, and may exacerbate the already present

insulin resistance (7 i).

3 1. i Pathogenesis of Hyperglycenlia- Induced Insulin Resisfurtce: The Hexosamine

Biosyn thesis Parlz ivay

In t h e 1 ate 1980s, Garvey et al . (73) showed using primary cultured rat adipocytes

that insulin and glucose CO-regulate the glucose transport system. In these experiments,

isolated adipocytes were incubated in media for 24 h in the absence or presence of

various glucose and insulin concentrations. At the end of the incubation, basal and

maximal insuli n stimulated gtucose transport was assayed. Glucose alone had no effect

on basal or maximal insulin stimulated glucose transport. Incubation of adipocytes with

increasing concentrations of insulin in the absence of glucose slightly decreased insulin

sti mulated glucose transport but on1 y at the highest insulin concentration tested, wi th no

decrease in basal uptake. In contrast, CO-incubation with glucose and insulin impaired

subsequent basal and insulin stimulated glucose uptake in a dose-dependent manner. At

the highest glucose and insulin concentration tested (20mM glucose, 50ng/ml insulin),

basal and insulin stimulated glucose transport were decreased by 89% and 63%

respectively. In a subsequent study by Traxinger et al. (74), it was discovered that insulin

resistance induced by glucose and insulin required the presence of the amino acii: L-

glutamine. Together these findings led to the hypothesis that the hexosamine

biosynthesis pathway (HBP) was imponant in the induction of insulin resistance by high

glucose (75). The HBP is thought to act as a 'glucose sensor' coupled to a negative

feedback system which would sense high glucose availability and subsequently prevent

excess glucose from entering the ce11 (75).

When adipocytes are exposed to high concentrations of glucose and insulin,

(insulin facilitates glucose entry into cells), the incoming glucose saturates the glucose

utilizing pathways such as glycogen s ynthesis and glycol ysis. The rernaining glucose

(usually 2-3% of total incoming glucose) fluxes through the HBP. Glucose enters the ce11

by the glucose transporters and is phosphorylated to glucose-6-phosphate which can enter

the glycogen synthesis pathway or be further processed to fructose-6-phosphate.

Fructose-6-phosphate then enters the glycolytic pathway or the HBP. The first step of the

HBP is the conversion of fructose-6-phosphate in the presence of L-glutamine to

glucosamine-6-phosphate by the rate-Iimiting enzyme Glutamine: fructose-6-phosphate

amidotransferase (GFA). Glucosamine-6-phosphate then undergoes a series of

biochemical steps to form UDP-N-acetyl-hexosamines as the end products. The main

product of this pathway, UDP-N-acetyl-glucosamine, is used in the synthesis of

gl ycoproteins, proteogl ycans, and gl ycolipids (Fig. 3.1). Several lines of evidence

support the role of the HBP in high glucose induced insulin resistance. First, GFA

inhi bi tors, azaserine and 6-diazo-5-oxonorleucine, which inhibit the conversion of

fructose-6-phosphate to glucosamine-6-phosphate, also prevented insulin resistance

induced by high glucose (75). This inhibition of GFA by azasenne closely paralleled the

prevention of glucose induced insulin resistance (75). Second, overexpression of GFA in

GLUCOSAMINE GLUCOSE

N-ACETYL-GLUCOSAMINE

GLYCOPROTEINS PEPTIDOGLYCANS GLYCOLLPKDS

I I

? ' + IMPAIRED GLUCOSE TRANSPORT

FIG. 3.1 THE HEXOSAMINE BIOSYNTHESIS PATHWAY

s keletal muscle and adipoc ytes in transgenic mice resulted in decreased insulin mediated

glucose disposa1 in vivo in the absence of hyperplycemia (76). Third, adipocytes treated

wi th glucosamine become insulin resistant. Glucosamine enters adipocytes via the

glucose transporters. Once inside the ceIl, glucosamine is phosphorylated to

glucosamine-6-phosphate and enters the HBP distal to GFA, thereby bypassing the

enzyme (75). Glucosamine is 40x more potent than glucose in inducing insulin resistance

(75). Lastly, hyperglycemia and glucosamine induced insulin resistance are non-additive

suggesting that both agents operate through the same pathway (77).

In addition to primary cultured rat adipocytes, chronic hyperglycemia and

glucosamine treatment have been shown to induce insulin resistance in 3T3-Ll

adipocytes (78) and in skeletal muscle preparations irt vitro (79,80). In vivo. studies

using the euglycemic hyperinsulinemic clamp technique have shown that prolongeci

h ypergl ycemia (8 1) or glucosamine infusions (8 1 $2) markedl y impair insulin mediated

olucose uptake at the whole body level and in skeletal muscle and adipose tissue (82). b

3.12. Poterztial Defects in the Insrilin Sigrzaiir~g P a r / w q

Insulin resistance induced by chronic hyperglycemia and glucosamine treatment

in isolated rat adipocytes is associated with impaired translocation of Glut4 to the plasma

membrane without a change in total transporter nurnber (73). Impaired Glut4

translocation has also been demonstrated in skeletal muscle (83), and in GFA

overexpressing transgenic mice (76). This type of defect is also seen in Type II diabetic

patients. Using NMR spectroscopy, C h e et al. (84) have shown that glucose transport

stimulation by insulin is defective in these patients, which was not due to decreased

levels of Glut4 (85) . These findings suggest that impaired translocation of Glut4 to the

plasma membrane is due to defective insulin signaling (83).

Studies examining the more proximal steps in the insulin signaling cascade in

hyperglycernia and glucosarnine induced resistance have led to controversial results.

Isolated rat adipocytes exposed to chronic hyperglycemia displayed decreased activation

of the receptor tyrosine kinase and IRS-1 tyrosine phosphorylation in response to insulin

in one study (86). Another study using the sarne cells and sirnilm incubation conditions

showed no defect in insulin receptor kinase activity or IRS-1 phosphorylation. Prolonged

exposure of rat skeletal muscle to 25mM glucose in vitro impaired Akt/PKB activation

and phosphorylation by insulin while PI3K activation was unaffected (79). Glucosamine

treatment of 3T3-L1 adipocytes led to impaired insulin receptor tyrosine kinase acti8/ity

and IRS- 1 tyrosine phosphorylation. impaired IRS- 1 associated PI3K activity and

decreased AktlPKB activation by insulin (87). Ir1 vivo, chronic glucosarnine infusion in

rats did not decrease insulin receptor tyrosine phosphorylation in skeletül muscle (88, 89).

IRS-1 tyrosine phosphorylation in response to insulin was reduced in one study (90) but

not in another (88). Glucosamine treatment has been shown to reduce IRS-1 associated

PI3K activity while having no effect on Akt/PKB activation or phosphorylation (88,89).

Similar findings were recently shown in skeletal muscle of obese Type II diabetic

patients (90). Obese Type II diabetic subjects displayed normal Aktl/PKBa and

Akt2/PKBP activation and phosphorylation in response to insulin while IRS-1 and IRS-3

associated PI3K activities were significantly reduced compared to lean controls and

obese non-diabetic subjects. The reason for the discrepancies among the various studies

and treatments is presently not clear but may be due to the length of the treatment times

or the presence or absence of insulin during the treatment.

3.1.3 Meclianisms ofHe,rc7sarnine lndttced Insrilin Resisrance

How the increased flux through the HBP with the accumulation of UDP-N-

acetylglucosamine impairs Glut4 translocation is not known but several potential

mechanisms have been proposed. One mechanisrn is the activation of protein kinase C

(PKC).

Indirect evidence stems from the finding that the decrease in insulin receptor

tyrosine phosphorylation induced by high glucose in rat adipocytes was prevented by co-

incubation with a PKC inhibitor (86). Hyperglycemia has been shown to activate PKC

(86) likely by increasing intracellular diacylgiycerol (DAG) levels (91). DAG can be

produced from the glycolytic intermediates dihydroxyacetone phosphate and glycerol 3-

phosphate (9 1). With increasing glucose concentrations, more glucase is avai lable for

glycolysis thereby increasing cellular DAG levels. Some evidence implicating PKC in

hexosamine-induced resistance comes from a study done by Filippis et al. (92). In this

study, exposure of rat adipocytes to high glucose or glucosamine activated PKC.

Pretreating the cells with the GFA inhibi tor azasenne almost completel y prevented the

activation of PKC by high glucose while having no effect on glucosamine induced PKC

activation. The residual elevated PKC activity remaining in the presence of azasenne

treatment may have been due to DAG production from glycolysis, suggesting that

hexosamines are powerful PKC activators (92). In this same study, treatment of cells

with the PKC inhibitor RO-3 1-8220 prevented insulin resistance induced by these agents.

It has been suggested that activation of PKC may lead to serine/threonine

phosphorylation of the insulin receptor thereby decreasing its tyrosine kinase activity

towards endogenous substrates such as the iRS proteins (93), ancilor that the IRS proteins

are themselves substrates for PKC which inhibit their tyrosine phosphorylation by the

receptor (94).

Increased flux through the HB P increases the intracel lulx concentration of UDP-

N-acetylglucosarnine, which serves as a substratc for O-linked glycosylation of various

proteins (72). Increased O-linked glycosylation rnay affect the activity of transporters

and enzymes involved in insulin signaling (73). Haltiwanger et al. (95) have shown that

increased O-linked glycosylation of the transcription factor Spl resulted in a reciprocal

decrease in its level of phosphoryIation suggesting that O-linked N-acetylglucosarnine

may compete with phosphate on some proteins. Recently, Patti et al. (89) have shown

that glucosamine infusion resulted in O-linked glycosylation of IRS-1 and iRS-2 in rat

skeletal muscle. UDP-N-acetylglucosamine may aIso modify gene transcription.

Increasing intracellular concentrations of UDP-N-acetylglucosarnine in adipocytes and

skcletal muscle by high glucose o r glucosamine treatment resulted in rapid and marked

increases in leptin messenger RNA and protein levels (96). To date, the significance of

enhanced O-linked protein glycosylation in the induction of insutin resistance remains

uncertain.

3.1.4 Porerttial Rule of O.ridurive Srress in Irzsulin Resistance

Oxidative stress has been implicated in the developrnent of insulin resistance and

diabetic complications (97) and results from either an overproduction of reactive oxygen

free radicals andor from a decreased efficiency of radical scavenger systems (97). The

net result is an increase in intracellular oxygen free radicals, which can be formed in cells

by many processes (98). Oxygen radicals are capable of reversibly or irreversibly

damaging nucleic acids, proteins, and lipids, thereby disrupting normal ce11 function (98).

Oxidative stress has been implicated in the deveioprnent of numerous human diseases

(98) including Type II diabetes. Evidence to support a role of oxidative stress in the

deveiopment of insulin resistance and Type LI diabetes cornes from data from Type II

diabetic subjects, and numerous in vitro and Nt vivo studies.

Oxidative darnage to DNA and membrane Lipids has been observed in Type II

diabetic patients. Type II diabetic subjects displayed higher levels of 8-OHdG (8-

hydroxydeoxyguanosine), an indicator of oxidative damage to DNA in the plasma and

urine (99, 100). These subjects also displayed higher IeveIs of Iipid peroxidation products

in the plasma and urine, suggesting damage to membrane lipids (101). In addition,

plasma free radical (O2-) concentrations were negatively correlated with whole body

glucose disposal (r = -0.53, ref. 97) in diabetic subjects. suggestinp that free radicals were

asswiated with insulin resistance.

Hyperglycemia and elevated free fatty acids (FFA), conditions present in Type II

diabetes, are known to cause insulin resistance in target tissues and have also been shown

to cause oxidative stress. Hyperglycemia is a well-known cause of enhanced plasma free

radical concentrations (102). Cells exposed to prolonged hypergl ycemia di splayed

oxidative damage to DNA and increased levels of lipid peroxidation products (100).

Plasma free fatty acids have also k e n shown to correlate with plasma free radicals (103).

In healthy human subjects, increasing the plasma free fatty acid concentration by a 24 h

infusion of intralipid caused an increase in maiondialdehyde concentration (an indicator

of oxidative stress), and a decline in plasma GSWGSSG ratio (reduced form of

glutathione/oxidized form of glutathione). Hyperl i pidemia has been shown to cause

insulin resistance in skeIetaI muscle, likely by interfering with the glucose/fatty acid

cycle (8 1). Interestingly, intralipid infusions in rats led to an increase in skeletal muscle

UDP-N-acetylglucosamine levels suggesting that free fatty acids may induce insulin

resistance by increasing the flux through the HBP, similar to high glucose (8 1).

In vitra, induction of oxidative stress by prolonged exposure of 3T3-L1

adi poc ytes to micromolar concentrations of h ydrogen peroxide (H102) i mpai red Glut4

translocation to the plasma membrane in response to insulin (104) and depleted the

intracellular levels of reduced glutathione (GSH) (105). Treatment of 3T3-L1 adipocytes

with lipoic acid, an antioxidant, pnor to treatment with Hz02 prevented the impainnent in

Glut4 translocation and maintained GSH levels (105).

In light of the findings that hypergiycemia and hyperlipidemia are associated with

oxidative stress (102,103) and increase cellular UDP-N-acetylhexosamines (81), and that

H2O2 irnpaired Glu4 translocation in 3T3-Ll adipocytes (104), our laboratory has

investigated whether insulin resistance induced by high glucose or glucosamine is due to

oxidative stress. To explore this possibility, isolated rat adipocytes were incubated for 18

h in media containing 20mM glucose/lûûnM insulin (High Gn), or 5mM glucosamine

(Glc). The incubations were done in the presence and absence of 5rnM N-acetyl-cysteine

(NAC). NAC is an antioxidant and glutathione (GSH) precursor (44), which would

scavenge any reactive oxygen species produced. High G/I decreased both basal

(pmol/6x10~ cellsf3rnin; control 240+71; high G/I 55Ç11; p4.01) and insulin-stimulated

(control 8 1621 55; high GA 326+27; p<0.0 1) glucose uptake, which were prevented by

NAC (basal 235547, insulin-stirnulated 885+63). In contrast to High GA, Glc treatment

did not significantly reduce basal glucose uptake (control 307k29; Glc 254I54). but the

insuIin stimulated glucose uptake was reduced (control 9492101, Glc 461+75) which was

prevented by NAC (Glc + NAC basal 335k63; insulin-stimulated 1049t98, pcO.01).

The present study examined the effect of NAC on Glut4 translocation and Glut i

expression in adipocytes treated with High GA and Glc. To gain further insight into the

mechanisrn of NAC action in adipocytes treated with high GD, insulin induced IRS-1

tyrosine phosphorylation, IRS-1 association with the p85 subunit of PDK, and

Akt l/PKBa activation were examined. These measurements were performed in order to

determine which signaling event(s) were altered, and the effect of NAC on these steps in

the insulin-signaling cascade.

3.2 Materials and Methods

3.2.1 Materials

Dulbecco's Modi fied Eagles Medium (DMEM) was obtained from Gibco (Ehrlington,

ON. Canada). Collagenase was purchased from Wonhington Biochernical Corporation

(Lakewood, NJ). Bovine serum albumin (BSA), N-acetyl-cysteine (NAC), and the 5'-

nucleotidase assay kit were obtained from Sigma (St. Louis, MO). Human insulin was a

gift from Eli Lilly (Indianapolis, LN). Antibodies against Glut4 and Glut l were a kind

gift from Dr. Samuel Cushman (Nationai Institutes of Hedth). Antibodies against IRS-1,

the p85 subunit of P13K, and the Akt llPKBa substrate Crosstide were purchased from

Upstate Biotechnology (Lake Placid, NY). Anti-phosphotyrosine antibodies (pY99) were

obtained from Santa-Cruz Biotechnology. Protein A Sepharose was purchased from

Pharrnacia Biotechnoiogy.

3 2 . 2 Adipocyte Isolation and Preparatiori

The method used to obtain isolated adipocytes was as described by Rodbell (106) with

minor modifications (73). Sprague-Dawley rats weighing 200-250g were kiIIed by CO2

inhalation followed by cervical dislocation. The epididymal fat pads were removed and

placed in DMEM containing 25mM HEPES, 1 % antibiotic-rintimycotic solution, 0.5%

FBS, and 3% BSA. isolated adipocytes were obtained by adding collagenase (4mgml)

to the medium and shaking for 1 h at 37OC. The cells were then filtered through nylon

and washed 2 times with the same medium descnbed above followed by 2 more washes

with DMEiM containing 25mM HEPES, 1 % antibiotic solution, 0.5% FBS, and 1 % BSA.

3.2.3 Prinrav Crtlrrcre of Adipocytes

Stock solutions of glucosamine and NAC were prepared fresh before each experiment

and equilibrated at pH 7.5. Freshly isolated adipocytes were incubated for 18 h at 37°C

in DMEM containing 25mM HEPES, 1% antibiotic solution, 0.5% FBS, and 1% BSA.

Cells werc incubated either in media alone (5.6rnM glucose), or containing 20mM D-

giucose/lOOnM insulin (High G/I), High G/I plus 5mM NAC, 5mM glucosamine, or

5mM glucosamine plus 5mM NAC. Insulin, 4.2nM. was added to the glucosamine

conditions to facilitate glucosamine entry into the cells. Al1 incubations were done in

stenle polypropylene flasks and maintained in a humidified atmosphere of 95%

air/S%CO? at 37°C. Following the 18 h incubation, the cells were washed twice with

KR3OH (137m.M NaCl, 5mM KCI, 1.2rnM KHrPOa, MgSOa, 1.25mM CaCI?, 30mM

HEPES, 1mM sodium-pymvate, pH 7.0) containing 3% BSA, and then incubated in the

same buffer for 30 minutes at 37°C to remove al1 extracellular and receptor bound

insulin. This step allows the glucose transport system in the insulin treated cells to return

to basal levels (73). Following the 30 minute incubation, the cells were washed twice in

KRBH (1 18mM NaCI, SrnM KCI, 1.2rnM MgS02, 2.5mii CaCI?, 1.2mM KH-PO4, 5rnM

NaHCO;, 30mM HEPES, 1 mM sodium-pynivate, pH 7.5) containing 3% BSA.

3 - 2 4 S~rbcellrrlar Fractionarion

To obtain plasma membranes (PM) and Iow-density microsomes (LDM), the ceils were

fractionated following a procedure modi fied from Simpson et al. (107). Foltowing

incubation with or without insulin at 37"C, the cells were washed once in TES buffer

(25mM Tris pH 7.4, SmM EDTA, 0.25M sucrose, 0.01 mgml aprotinin, O.Olrng/ml

leupeptin, 0. lmm PMSF) previously equilibrated at 1g0C, and homogenized rapidly

using a glass-teflon homogenizer. Al1 subsequent steps were performed at 4°C. Whole

ceIl homogenates were centrifuged at 17,000 x g for 15 minutes. The fat cake was

carefull y removed and discarded. The supernatant (Supernatant 1 ) was removed and

transferred to another tube. The pelIet (Pellet 1) was re-suspendend in TES buffer and

centrifuged again at 17,000 x g for 15 minutes. Supernatant 1 was centrifuged at 48,000

x g for 15 minutes. The pellet, containing the high-density microsomes, was discarded.

The supernatant (Supernatant 2) was re-suspended in TES buffer and centrifuged at

400,000 x g for 60 minutes. The pellet obtained from this step contained the LDM. To

obtain the PM fraction, Pellet 1 (the first pellet obtained) was resuspended in TES buffer

and layered on a 1.12M sucrose cushion ( l .12M sucrose, 2mM EDTA, 25mM Tris, pH

7.4) and centrifuged at 95,000 x g for 6 5 minutes in a swinging bucket rotor (S W60,

Beckman). The PM fraction was collected at the interface between the two liquid layers,

re-suspended in TES buffer, and centrifuged at 48,000 x g for 15 minutes. The pellet,

containine the PM fraction was re-suspended in TES buffer and centrifuged again at the

same speed. PM and LDM fractions were solubilized in Buffer A (Section 2.2.3) and

stored at -80°C until further analysis. Total protein yields (Sstion 3.2.6) of LDM and PM

fractions were similar in al1 conditions. The purity of the PM fraction was determined by

5'-nucleotidase activity as described in Section 3.2.1 1.

3.2.5 Preparatio~z of Wlole Cell Lysares

FolIowing treatment with or without insulin, the cells were washed once at room

temperature with KRBH (no BSA) and rapidly hornogenized in Buffer A (Section 2.2.3)

using a glass-teflon homogenizer. AI1 subsequent steps were performed at 4°C. The

homogenate was centrifuged at 10,000 x g for 10 minutes. The fat cake was carefully

removed and discarded. The supernatrint was transferred to another tube and stored at -

80°C until funher analysis.

3.2.6 Protein Derern~i~zation

Protein levels were determined by the Bio-Rad method.

3.2.7 Determination of Glut4 Trarrslocariorz and Glrttl Levels

Following the 18 h incubation and washing procedures, the cells were incubated in

KRBH containing 3% BSA for 30 minutes at 37°C in the presence or absence of lOOnM

insulin. Following insulin treatment, LDM and PM fractions were obtained. LDM and

PM samples (30pg) were separated by SDS-PAGE using 10% polyacrylamide, and

transferred to nitrocellulose membranes. The membranes wece first incubated at room

temperature for 1 h in TBS (Section 2 .25) containing O. 1% Tween-20.5% non-fat dry

milk to minirnize non-specific binding, and then incubated for 16 h at 4°C in TBS

containing 3% BSA with antibodies recognizing either Glut4 (1: 1000) or Glutl (1:2000).

The membranes were then incubated for 1 h at room temperature in TBS (3% BSA)

containing an anti-rabbit secondary antibody conjugated to horseradish peroxidase.

Proteins were visualized by enhanced cherniluminescence. The intensities of the bands

were qurintified using NiH Image.

3.2.8 IRS- I Imm~moprecipitation

CeiIs were incubated for 5 minutes at 37OC in KRBH containing 3% BSA in the presence

or absence of lOOnM insulin. Following insulin stimulation. the cells were either

fractionated to obtain the LDM or whole cell lysates were prepared. One hundred pl of

washed Protein A sepharose bead slurry (SOp1 packed beads) was incubated with 4pg of

u-IRS-I for 2 h at 4°C in phosphate buffered saline (PBS) (137mM NaCl, 3rnM KCI,

8mM Na2HP04, 15mM KH2P04, pH 7.4). LDM sarnples (5OOpg) or whole ce11 lysates

(lmg) were added to the tubes which were then left shliking ovemight at 4°C. Following

the overnight incubation, the precipitated irnmunocomplex was washed 3 times with

500pl of ice cold PBS. The sepharose beads were re-suspended in 2x Laemmli sarnple

buffer (108) and boiled for 5 minutes. The beads were then centrifuged at 10,000 x g for

1 minute. The supernatant was collected and analyzed.

3.2.9 Determination of IRS-l firosine Phosp/~arylation and Association wit/z p85

The IRS-1 immunoprecipitated from the above step was divided into 2 aliquots. One

aIiquot was for the detemination of tyrosine phosphorylation, and the other for

association with p85. The aliquots were separated by SDS-PAGE by the procedure

outlined in Section 3-27 , To determine JRS-1 tyrosine phosphorylation, the membrane

was first incubated at room temperature for lh in TBS containing O. 1% Tween-20 with

5% BSA. Anti-phosphotyrosine antibodies (1 : 1000) were then added to the medium and

the membrane was left shaking overnight at 4OC. The membrane was then incubated for

1 h at room temperature in TBS (3% BSA) containing an anti-mouse antibody conjugated

to horseradish peroxidase. Total p85 associated with IRS-1 was determined by the

procedure outlined in Section 3-27 with a-p85 (1:lOOO) as the primary antibody. To

ensure that equal amounts of IRS-1 were immunoprecipitated in al1 conditions, the

membranes were treated with a 15% H107 solution for 15 minutes at room temperature

with continuous shaking. The membranes were then probed for total IRS-1 as described

above. Since the amoumt of IRS-1 immunoprecipitated from the LDM was sirnilar in al1

conditions, the densitometry values from the anti-phosphotyrosine imrnunoblots were

used in the calculations. Due to slight variability in the amount of IRS-1

immunoprecipitated from total cell lysates, the anti-phosphotyrosine values were

corrected for total IRS-1 present.

3.2.10 AX-rl/PKBcr Acrivation and Phospl~orylarion

Cells were incubated in KRBH (1 % BSA) for 10 minutes at 37°C either in the presence

or absence of 1OOnM insulin. Whole ce11 lysates were then prepared as descri bed in

Section 3.2.5. Aktl/PKBa activity was determined by the kinase assay procedure

outlined in Section 2.2.4. The substrate Crosstide (30pM per assay) was used- Serine 473

phopshorylation of Akt l/PKBct was determined by western irnmunoblotting according to

the procedure outlined in 2.2.5.

3.2.1 1 5'-Ntrcleoridase

5'-Nucteotidase (5'-M)) activity was determined by a spectrophotometric assay which

measured the rate of nicotinamide adenine dinucleotide (NAD) formation from the

following series of reactions (abbreviations defined below):

5 ' 4 3

A* - Adenosine + Pi

ADA Adenosine . ) Inosine + Amrnonia

GLDH Ammonia + 2-Oxoglutarate + NADH . - L-Glutamate + NAD

Samples (lûûpI) of whole ceIl lysates, LDM and PM fractions were added to 900~1 of 5 ' -

ND reagent containing 3.2mM adenosine monophosphate (AMP), 0.2mM reduced NAD

(NADH). 3.7mM 2-Oxoglutarate, 1 1 000U/L L-glutamate deh ydrogenase (GLDH),

adenosine deaminase (ADA) 400 Un, fi-glycerophosphate, buffer, and stabilizers. Once

added, the reaction mixture was mixed by inversion and left at room temperature for 5

minutes. After 5 minutes the absorbance (A) was read using a spectrophotometer

(Beckman DU4O) at 340nm versus water as a reference (Initial A). The mixtures were

then placed in a 37°C incubator for 5 minutes after which the absorbance was read again

(Final A). 5'ND activity was calculated as follows:

*A per 5 min (W) = Initial A - Final A

5'-ND = *A per 5 min X 322 X 1.35

Where: Factor 322 = (1 X 1000)/(6.22 X O. 1 X 5)

1 = Total reaction volume (ml) 1000 = Conversion of activity per ml to activity per L 6.22 = Millimolar absorptivity of NADH at 340nm 0.1 = Volume of sample (ml) 5 = Conversion of *A per 5 min to *A per min 1.35 = Temperature correction factor

Enzyme activities were then corrected for total protein in the reaction mixtures and

expressed as U U m g protein. Results are: Whole cell lysate 10, LDiM 6, PM 8 1.

3.2.12 Statistical A~iulysis

Results represent the ME.4N-tSE. To analyze the data, a one-way analysis of variance

(ANOVA) was used. Post hoc comparisons of means were performed using Scheffe's

test and Tukey's test. Statistical signi ficance was defined as p < 0.05.

3.3.1 Ef ic t of NAC on Glirt3 Transfocarion in Cells Treared roith High GA? and Glitcosarnine

Glut4 levels in each fraction are expressed as a percentage of the total GIut4

found in the LDM and PM fractions combined. designated in each experiment as 100%.

In basal cells, almost al1 of the Glut4 was found in the LDM with very little found in the

PM. This was seen in al1 conditions (Fig. 3.2 and Fig. 3.3). In controf cells, insulin

treatment resuIted in the translocation of Gtut4 from the LDM to the PM, increasing the

plasma membrane concentration of Glut4 with a corresponding decrease in the LDM

content (Fig. 3.2 and Fig. 3.3). In cells treated with High GA, Glut4 translocation to the

PM was impaired (InsuIin stimulated PM; Control 74.16.3%, High GA 35.7k5.596,

p~O.05, n=3) with Glut4 being retained in the LDM (Insulin stimulated LDM; Control

35.9I6.3%, High GA 64.3f 5.5%. p<0.05). Co-incubation with NAC prevented the

impairment in Glut4 translocation induced by High GA (Insulin stimulated; PM

69.4_+2.1%, LDM 30.6I2.196). In cells treated with High G/I + NAC, the distribution of

Glut4 between the LDM and PM compartments in response to insulin was simihr to the

control (Fig. 3.2). Similar results were seen in cells treated with glucosamine ( F i g 3.3).

Glucosamine treatment impaired the translocation of Glut4 to the PM in response to

insulin (Insulin stimulated PM; Control 71.3&4.4%, Glc 46.8f6.6%, ~ ~ 0 . 0 5 , n=3) with

Glut4 remaining in the LDM (Insulin stimulated LDM: Control 28.7&4.4%, Glc

53.2I6.6%, p<0.05). Co-incubation with NAC prevented the impairment in Glut4

translocation induced by glucosamine treatment (Insulin stimulated; PM 79.346-0 1 %,

LDM 20.73+6.01%). Incubation with NAC alone had no effect on Glut4 translocation in

LDM PM LDM P M LDM P M LïBM PM LDM PM LDM PM

INSULIN - + + - - + + - + + Control High G/I High G/I +NAC

LDM P M LDM P M LDM P M LDM PM LDM P M LDM P M

INSULIN - + + - - + + - + + Control High GD High G/I + NAC

FIG. 3.2 Glut4 translocation in High GA and NAC treated rat adipocytes. Adipocytes were incubated for 18 h in media containing 20mM glucose/lOOnM insulin in the absence and presence of 5rnM NAC (High GA and High Gn + NAC respectively). Cells were washed, and incubated for 30 min in the presence or absence of 1 OOnM insulin. LDM and PM samples (30pg) were separated by SDS-PAGE, transferred to nitrocelluIose, and immunoblotted using an anti-Glut4 antibody. (A) Representative immunoblot. (B) Data are MEANSE of 3 independent experiments ( > < 0.05, High GA vs others).

LDM P M U>M PM LDM PM

INSULIN - O + Control

+ - - Glc

LDM LDM P M LDM P M

- - + + Glc + NAC

LDM PM LDM P M LDM PM LDM P M LDM PM LDM P M

INSULIN - - + + + + - - + + Control Glc Glc + NAC

HG. 3.3 Glut4 translocation in Glc and NAC treated rat adipocytes. Adipocytes were incubated for 18 h in media containing 5mM glucosamine/4.2nM insulin in the absence and presence of 5rnM NAC (Glc and Glc + NAC respectively). Cells were washed, and incubated for 30 min in the presence or absence of 100nM insulin. LDM and PM samples (30pg) were separated by SDS-PAGE, transferred to nitrocellulose, and inmunoblotted using an anti-GLut4 antibody. (A) Representative immunoblot. (B) Data are MEAN* SE of 3 independent experiments ( *p < 0.05, Glc vs others).

response to insulin (Basal Control; LDM 85.096, PM 15.0%, NAC; LDM 88.0%, PM

12.0%. Insulin Control; LDM 29.0%, PM 7 1.0%, NAC; LDM 28.0%, PM 72.0%) (Fig.

3.4).

3.3.3 Effecr of NAC orr Glrttl Expressiort in CeIls Treared with Higli GA? and Gltr cosam irz e

Glutl was mainiy detected in the PM (Fig. 3.5 and Fig. 3.6), consistent with

previous reports (27). Since Glutl translocation to the PM in response to insulin was not

obsenled, on1 y basal cells were used in the calculations. Ali Glut 1 data is expressed as

fold of control basal which was designated a value of 1.0. In cells treated with High Gn,

Glut 1 levels were sigificantly reduced (Control 1 .O, High G/I 0.2kû. 1, p<O.OS, n=3)

(Fig. 3.5). This finding was consistent with the decrease in basal glucose uptake seen in

cells treated with High GA. Co-incubation with NAC prevented the downregulation in

Glut 1 induced by High GA (High G/T + NAC 1.2t0.1) (Fig. 3.5). In contrast to High G/I,

18 h glucosamine treatment did not significantly reduce Glutl levels and CO-incubation

with NAC had no further effect (Control 1 .O, Glc 1.1S.2, Gic + NAC 1.0k0.1, p = NS,

n=3) (Fig. 3.6). Glucosamine treatment did not significantly affect basal glucose uptake.

Treatment with NAC alone had no effect on Glutl levels (Control 1.0, NAC l.OS)(Fig.

3.4).

3.3.3 Effect of High G/I and NAC on Aktl/PKBaActivution and Phosplioryfation

Treatment of control cells with IOOnM insulin for 10 minutes resulted in a 1422.8

fold increase above basal in Akt lfPKBa activation (Fig. 3.7). In cells treated with High

Ga, Aktl/PKBa activation by insulin was reduced by 44% compared to the insulin

LDM PM LDM PM LDM P M LDM P M

INSULIN - - + + - - + + Control NAC

LDM PM LDM

INSULIN - - + Control

PM LDM PM LDM PM

+ - - + + NAC

FIG. 3.4 Effect of NAC alone on Glut4 translocation and Glutl expression. Adipocytes were incubated for 18 h in media containing SmM NAC. Cells were washed, and incubated for 30 minutes in the presence or absence of lOOnM insulin. LDM and PM samples (30pg) were separated by SDS-PAGE, transferred to nitrocellulose, and immuooblotted using (A) an anti-Glut4 antibody, or (B) an anti-Glut 1 antibody.

LDM PM LDM PM LDM P M LDM PM LDM PM LDM PM

INSULIN - - + + - - + + - - + + Control High G/I High G/I + NAC

Control High G/I High G/I + NAC

FIG. 3.5 Glut content of plasma membranes in High G/I and NAC treated rat adipocytes. Adipocytes were incibated for 18 h in media containhg 20mM glucose/lOOnM insulin in the absence and presence of 5mM NAC (High GA and High GA + NAC respectively). CelIs were washed, and incubated for 30 min in the presence or absence of lOOnM insulin. LDM and PM samples (30pg) were separated by SDS-PAGE, t rans ferred to r,itrocellulose, and immunoblotted using an anti-Glut 1 antibody . (A) Represeatative imrn~moblot. (B) Data are MEANISE of 3 independent experiments ( *p < 0.05, Hi& GA vs others). Results are expressed as the fiaction of total GIutI distributed in the PM in each condition, PM of Control was given the value 1 .O.

LDM P M LDM PM LDM P M LDM PM LDM PM LDM PM

INSULIN - - + + - - + + - t + Control Glc GIc + NAC

Control Glc Glc + NAC FIG. 3.6 Glut l content of plasma membranes in Glc and NAC treated rat adipocytes. Adipocytes were incubated for 18 h in media containing S r n M glucosamuie/4.2nM insulin in the absence and presence of 5mM NAC (Glc and Glc + NAC respectively). Cells were washed, and incubated for 30 min in the presence or absence of l OOnM insulin. LDM and PM samples (30pg) were separated by SDS-PAGE, transferred to nitroceilulose, and immunoblotted using an anti-Glut 1 antibody. (A) Representative irnmunoblor. (B) Data are MEAN* SE of 3 independent experirnents. Results are expressed as the fiaction of total Glut 1 distributed in the PM in each condition. Control PM was assigned the value 1 .O.

i

O L CI

r: 1 4 . U cLir

O 1 2 - = O z E ' O - .- > .- Y

Y 8 -

INSULIN - +

Control

- + High GD

- + Higb G/I + NAC

H G . 3.7 Effect o f High GA and NAC on Akt l/PKBa activity. Adipocytes were incubated for 18 h in media containing 20mM glucose/lOOnM insulin in the absence and presence of 5rnM NAC (High G/I and High G/I + NAC respectively). Following the incubation, cells were washed, and incubated for 10 min in the presence or absence of 1OOn.M insulin. Cells were then lysed, Akt lPKBa immunoprecipitated,and an N> vitro b a s e assay was performed using the substrate Crosstide. Results represent MEANSE of 4 independent experiments ( *p<0.05, High GA vs others).

stimulated control (High GA; 7.8W.95, pc0.05, n=4). Co-incubation with NAC

prevented the decrease in AktlPKBa activation induced by High G/I (High G/I + NAC:

14.7k3.1) (Fig. 3.7). Basal Akt l/PKBu activity was not different mong conditions.

Incubation with NAC alone reduced insulin stimulated AktlIPKBa activation by 34%,

while having no effect on basal AktliPKBa activity (Basal; Control0.09'10.03, NAC

0.05kO.01, p = NS, n 4 . Insulin stimulated; control 1 .O, NAC 0.66f0.05, p < 0.05, n d )

(Fig.3.8).

lmrnunoblot analysis of Ser473 phosphorylated AktlPKBu showed similar results

(Fig. 3.9). In basal cells, Ser473 phosphorylated AktUPKBa was not detected. but upon

stimulation with insulin a prominent band appeared on the irnmunoblot. Therefore, only

insulin treated conditions were used in the calculations. The insulin stimulated control

was designated a value of 1.0. In cells treated with High GA. AktllPKBu

phosphorylation on Ser473 was significantly decreased compared to the control (Control;

1 .O, High GA; 0.3 lkO.15, p<0.05, n=3) (Fig. 3.9). Co-incubation with NAC prevented

the impairment in AktlPKBa phosphorylation seen in cells treated with High G/I @gh

G/I + NAC 1.18M.07) (Fig. 3.9). Incubacion with NAC alone did not have any effect on

the insulin-induced phosphorylation of Ser473 (Control 1 .O, NAC 1.1410.02, p = NS,

n=3)(Fig. 3.10). Total levels of Akt l/PKBa were unchanged by the treatments (Basal;

Control 1.0. High GA 1-04 I 0.23, High G/I + NAC 1.0 r 0.08, p = NS. Insu!in

stimulated; Contro10.97 dl.08, High G/I 0.82 a .05 , High G/I +NAC 0.90 H.08, p =

NS, n=3) (Fig. 3.10 and Fig. 3.11).

INSULIN - + Control

- + NAC

FIG. 3.8 Effect of NAC on Aktl/PKBa activity. Adipocytes were incubated for 18 h in media containing 5mM NAC. Following the incubation, cells were treated with 1OOnM insulin for 10 minutes. Cells were subsequently lysed, Akt l/PKBa irnmunoprecipitated and in vitro kinase activity determined using the substrate Crosstide. Results represent MEANkSE of 4 independent experiments. The insulin treated control was assigned the value 1 .O (Basal; Control0.09f0.03, NAC O.O5+O.O 1, p = NS. Insulin; Control 1 .O, NAC 0.66I0.05, *@.05).

INSULIN - + Control

- + - + High GA High G/I + NAC

Control High G/I High G/I + NAC

FIG. 3.9 E ffect of High G/I and NAC on Ser473 phosphory lation of Akt 1 /PKBa. Adipocytes were incubated for 18 h in media containing 20mM glucose/lOOnM insulin in the absence and presence of SmM NAC (High G/I and High G/I + NAC, respectively). Following the incubation, the cells were washed, stimulated for 10 minutes with 1 OOnM insulin, and lysed. Samples ( 6 0 ~ g ) were separated on SDS-PAGE, transferred to nitrocellulose, and probed for (A) Ser473 phosphorylated Akt LPKBa. (B) Represents the MEANkSE of 3 independent experiments. The insulin treated control was designated the value 1 .O since Ser473 phospho- Akt l/PKBa was only detected with insulin treatment (Control 1 .O; High GA 0.3 1 M. 15; High G/I + NAC 1.17fl.07, *@.OS, Control vs others).

INSULIN - + Control

I + NAC

Control NAC

FIG. 3.10 Effect of NAC alone on Ser473 phosphorylation of Akt l/PKBa. Adipocytes were incubated for 18 h in media containhg 5mM NAC. Following the incubation, the cells were washed, stimulated for 10 minutes with lOOnM insulin, and lysed. Samples (60pg) were separated on SDS-PAGE, transferred to nitrocellulose, and probed for (A) total Akt l/PKBa, or (B) Ser473 phosphorylated AktlPKBa. (C) Represents the MEANHE of 3 independent experiments represented in (B), the insulin treated control was designated the value 1 .O since Ser473 phospho- Akt l/PKBa was only detected with insulin treatment (Control 1 .O; NAC 1.1+0.02, p = NS).

INSULIN - + - + - + Control High G/I High G/I +NAC

O --

INSULIN Control High G/I High G/I + NAC

FIG. 3.11 Effect of High G/I and NAC on total levels of Aktl/PKBa. Adipocytes were incubated for 18 h in media containing 20mM glucose/lOOnM insulin in the absence and presence of 5mM NAC (High GA and High G/I + NAC, respectively). Following the incubation, the cells were washed, stimulated for 10 minutes with 1 OOnM insulin, and lysed. Samples (60pg) were separated on SDS-PAGE, transferred to nitrocellulose, and probed for (A) total AktlPKBa. (B) Represents the MEANSE of 3 independent experiments (Basal; control 1 .O; High GA 1 .O4 + 0.23;High GA + NAC 1 .O + 0.08 p = NS. Insulin stimulated; control0.97 50.08; High GA 0.82 F0.05, Hi& GA +NAC 0.90 k0.08, p = NS).

3.3.4 Eflect of High G/I and NAC on IRS-I Tyrosine Phasphorylatiorr and Associariorz

irpith p85 in the LDM

Several studies have suggested that insulin resistance is associated with

increased serine/threonine phosphorylation of IRS- 1. which may prevent its interaction

with, or render it an inhibitor rather than a substrate of the insulin receptor kinase (94).

This would decrease insulin-stirnulated tyrosine phosphorylation (93, 109). To determine

whether the High GO treatment caused such a defect, iRS-1 was immunoprecipitated

from the LDM and tyrosine phosphorylation rneasured. In rat adipocytes, it has been

dernonstrated previously that the rnajority of IRS-1 is located in the LDM and this

localization was unchanged after insulin stimulation (36). This was confirmed in this

study. Total levels of IRS-1 in the LDM were unchanged by any of the treatments

(Basal; Control 1.0, High G/I 0.75kO.05, High GA + NAC 0.8 lH.07, p = NS. Insulin;

Control 0.97k0.11, High GA 0.78H.02, High GA + NAC 0.99%. 13, p = NS, n=3) (Fig.

3.13). Insulin treatment of control cells led to a 13-fold increase in tyrosine

phosphorylation compared to basal cells (ControI; Basal 0.06kO.03, Insulin treated

control was designated 1 .O) (Fig. 3.13). Treatment cf cells with High G/I for 18 h did not

have an effect on insulin-stimulated tyrosine phosphorylation of IRS-1 compared to the

control (Insulin; High G/I 1 .O3M.35, p = NS, n=3). There was a slight increase in brisa

tyrosine phosphorylation of iRS-1 compared to the basal controI but this was not

statistically significant (Basal; High GA 0.22M.07, p = NS). In cells CO-incubated with

NAC and Hiph GA, tyrosine phosphorylation of IRS-1 in basal and insulin treated cells

was similar to the control (High G/I + NAC; Basal 0.03M.01, Insulin 0.89H.32, p = NS)

(Fig. 3.13).

- + - + - + Control High G/I High GA + NAC

0 ---

INSULIN

Control High Gn High GD + NAC

FIG. 3.12 Effect of High GA and NAC on IRS-1 levels in the LDM. Adipocytes were incubated for 18 h in media containing 20mM glucose/100nM insulin in the absence and presence of 5mM NAC (High GA and High G/I + NAC, respectively). Cells were subsequently washed and stimulated with 1 OOnM insulin for 5 minutes. LDM samples (8Opg) were resolved by SDS-PAGE, transferred to nitrocellulose and blotted with an anti-IRS- 1 antibody. (A) Representative immunoblot. (B) Represents the MEANSE of 3 independent experiments. Basal control was assigned the value 1 .O (Basal; Control 1 .O, High GA 0.75k0.05, High GA + NAC 0.8 1k0.07, p = NS. Insulin; Control 0.97k0.11, Hi& G/X 0.78M.02, High GA + NAC 0.99k0.13, p = NS).

INSULIN 9 + - + 9 + Control Higb GA High G/I + NAC

INSULIN - + + + Control High G/I High GD + NAC

HG. 3.13 Effect of High GA and NAC on IRS-1 tyrosine phosphorylation in the LDM. Adipocytes were incubated for 18 h in media containing 20mM glucose/ 100nM insulin in the absence and presence of 5rnM NAC (High GA and High GA + NAC, respectively). Followùig the incubation, cells were washed and treated with lOOnM insulin for 5 minutes. IRS- 1 was imrnunoprecipitated fiom LDM samples (SOOpg), resolved by SDS-PAGE, transferred to nitrocellulose and probed with anti-phosphotyrosine antibodies (pY99). (A) Representative immunoblot. (B) Total IRS-1 imrnunoprecipitated in (A). (C) Results represent MEANkSE of 3 independent experiments representedin (A). The insulin treated control was assigned the value 1 .O (Basal; Control0.06+0.03, High GA 0.22k0.08, High Gd+ NAC 0.03M.01 p = NS. Insulin; Control 1 .O, High G/I 1.03+0.35, High GA + NAC 0.89-tO.32, p = NS).

61

The p85 subunit of PI3K binds to tyrosine phosphorylated IRS-1. Therefore, the amount

of p85 associated with RS-1 was determined. RS-1 was immunoprecipitated from the

LDM, resolved by SDS-PAGE and immunoblotted with p85 antibodies. In control cells,

insulin increased the amount of p85 bound to ES-1 approxirnately 5- fold above basal

(Control; Basal O. 17H.06, Insulin 1.0) (Fig. 3.14). Ceils treated with High G/I displayed

similar ievels of p85 associated with IRS-1 in both basai and insulin treated cells (High

GA: Basal O - Z k O . 19, Insulin i.26kû. 19, p = NS, n=3). In ce1 1s CO-incubated with NAC

and High GA, the amount of p85 associated with RS-1 in basal and insulin treated cells

was similar to the control and High GA treated cells (High G/i + NAC; Basal 0.30kû. 19.

Insulin 0.8 1 s . 16, p = NS) (Fig. 3.14).

3.3.5 Effect of High G/I and NAC on I R S - I Tvrosine Phosphory farion in Total Ce11

Lysares

IRS- 1 tyrosine phosphoryiation was measured in total ce11 lysates. since recent

studies have suggested that disruption of IRS-1 function may Vary among cellular

compartments ( 1 15). In control cells, insulin increased tyrosine phosphorylation of IRS-

1 (Control; Basal 0.04&0.02, Insulin 1 .O) (Fig. 3.15). Cells treated with High GA,

displayed higher basal leveis of tyrosine phosphorylation but lower levels after insulin

treatmenr compared to the control. These differences did not reach statistical significance

(High G/I; Basal 0.30N. 14, Insulin 0.65f0. Il, p = NS, n=3). Ceils CO-incubated with

High GA and NAC displayed a tyrosine phosphorylation pattern similar to the control, in

both basal and insulin treated cells (High G/i + NAC; Basal 0.15M.08; Insulin

1.07k0.43, p = NS). Due to variability, tyrosine phosphorylation of IRS-1 in response to

INSULIN - + + + Control High GA High GD + NAC

cf- INSULIN - + - + - +

Control High G/I High G/I + NAC

FIG. 3.14 Effect of High GA and NAC on IRS-1 associated p85 in the LDM. Adipocytes were incubated for 18 h in media containhg 20mM glucose/lOOnM insulin in the absence and presence of 5mM NAC (High GA and High G A + NAC, respectively). Following the incubation, cells were washed and treated with 100nM insulin for 5 minutes. IRS-1 was immunoprecipitated h m LDM samples (500pg), resolved by SDS-PAGE, transferred to nitrocellulose and probed with an anti-p85 antibody. (A) Representative imrnunoblot. (B) Results represent MEAMSE of 3 independent experiments. The insulin treated control was assigned a value of 1 .O (Basal; Control 0.17t0.06, High GA 0.25k0.19, High G/I + NAC 0.3020.19,p = NS. Insulin; Control 1.0, High GD 1.2620.19, f igh GII +NAC 0.81k0.16, p = NS).

63

Control High G/I High G/I + NAC B

INSULIN - + - + - + Control High G/I High GII + NAC

FIG. 3.15 Effect of High GA and NAC on iRS-1 tyrosine phosphorylation in total ce11 lysates. Adipocytes were incubated for 18 h in media containing 20m.M glucose/100nM insulin in the absence and presence of 5mM NAC (High GA and High G/I + NAC, respectively). Followïng the incubation, cells were washed and treated with l0On.M insulin for 5 minutes. XRS-1 was immuno- precipitated fkom ceIl lysates (lmg), resolved by SDS-PAGE, transferred to nitrocellulose and probed with anti-phosphotyrosine antibodies (pY99). (A) Representative immunoblot. (B) Results represent M E W S E for 3 independent experirnents. The insulin treated control was assigned the value 1 .O (Basal; Control O.O4f 0.02, High G/I 0.30kO. 14, High G/I + NAC O. 15f 0.08, p = NS. Insulin; Control 1 .O, High GA 0.65kO. 1 1, High GA + NAC 1.07H.43, p = NS).

insulin was not significantly different among conditions (Fig. 3.15). 18 h treatment with

NAC alone had no effect on insulin-stimulated IRS-1 tyrosine phosphorylation (Basal;

Controt 0.16, NAC 0.1 1 , insulin; Control 1 .O. NAC 0.88). Total levels of IRS-1 in ce11

l ysates were unchanged by the treatments (Control 1.0, High G/I 1 . O 8 S . Z ? High G/E +

NAC 1.01k0.15, p = NS. n = 3, NAC alone 0.78) (Fig. 3.16).

- Control High Gfl High G/l + NAC NAC

Control High Gfï High GA + NAC NAC

FIG. 3.16 Effect of High GA and NAC on IRS-1 levels in total ce11 lysates. Adipocytes were incubated for 18 h in media containing 20mM glucose/lOOnM insulin in the absence and presence of 5mM NAC (High GA and High GA + NAC, respectively), or with 5mM NAC alone. Ce11 lysates (80pg) were resolved by SDS-PAGE, transferred to nitrocellulose and blotted with an anti-IRS- 1 antibody. (A) Representative immunoblot. (B) Represents the MEANkSE of 3 independent experiments. The control was assigned the vaIue 1 .O (Control 1 .O, High G/I 1.0820.22, High GA + NAC 1 .Ol+O. 15, p = NS, NAC; 0.78).

3.4 Discussion

Previous work in our laboratory has shown that isolated rat adipocytes chronically

treated with high glucose/insulin, or glucosamine, dispiayed marked decreases in insulin-

stimulated glucose uptake which could be prevented by NAC. The present study has

showed that insuiin resistance induced by high plucose/insulin or glucosmine is

associated wi th impaired translocation of Glut4 to the plasma membrane, as reported

previously (73). Co-incubation with NAC prevented the impairment in GIut4

translocation, suggesting that NAC prevents insulin resistance. The fact that NAC

prevented both high glucose/insulin and glucosamine induced insulin resistance implies

that the site of action of NAC is at a common point in the two pathways. In addition, the

ability of NAC to prevent insulin resistance induced by high glucose/insulin or

glucosamine suggests a potential roIe of oxidative stress in the induction of insulin

resistance in rat adipocytes.

In contrast to Glut4 translocation, high gIucose/insulin and glucosamine had

different effects on Glutl expression. High glucose/insulin decreased Glutl levels in the

plasma membrane. High glucose and insulin have shown to have independent effects on

Glut 1 expression. In ceIl cultures prolonged treatment with high glucose decreases Glut 1

expression (27). El-Kebbi et al. (1 10) have shown that treatment of BC3H-1 cells, a

cultured skeletal muscle ceIl line, for 24 h in 25mM glucose resulted in an 80% decrease

in Glutl content. AIthough chronic insulin treatment has been shown to increase Glutl

expression in L6 muscle cells and 3T3-LI adipocytes (1 1 1,112). this effect was not seen

in isolated rat adipocytes (73). Therefore, it appears that high glucose has a dominant

effect on the levels of Glutl in rat adipocytes incubated with high glucose/insulin.

G m e y er al. (73) have shown that high glucose and insulin together decreased Glutl

levels, but chronic insulin treatment alone had no effect suggesting that increased

metabolism of glucose causes a reduction in Glutl protein levels. In contrast to high

gIucose/insulin, glucosamine treatment, either in the presence or absence of NAC, had no

effect on Glutl protein levels. The difference between the effects of high glucose/insulin

and glucosamine may be explained by the different point of entry into the hexosarnine

biosynthesis pathway by these agents. For high glucose to enter the hexosamine

pathway, it first has to saturate glycolysis and glycogen synthesis before being converted

to glucosamine-6-phosphate by GFA (75). Intermediates from any of these pathways

could potentially be involved in the regulation of Glut 1 levels (1 13). Glucosamine

bypasses these steps and enters the hexosarnine pathway distal to GFA (75). NAC was

able to prevent the decrease in Glut 1 induced by high glucose/insulin. How NAC

prevents the decrease in Glutl is not clear. NAC may be acting upstream of GFA thereby

blocking the generation of the glucose signal. In contrast to the present study, Rudich et

ul. (104) have shown that prolonged oxidative stress, induced by incubating cells with

micromolar concentrations of H2O2 increases Glutl levels in 3T3-L1 adipocytes.

IRS-1 tyrosine phosphorylation in response to insulin was not altered with high

glucose/insuiin treatment whether measured in the LDM or in total ce1 1 l ysates,

sugesting that, in this modei, there was no significant serinehhreonine phosphorylation

of IRS-I preventing tyrosine phosphorylation. In addition, the association of p85 with

RS-1 in response to insulin was not altered in the LDM fraction. Due to technical

difficulties, IRS-1 associated p85 in total cell lysates was not measured. Since almost al1

of the IRS-1 is located in the LDM in rat adipocytes (36), the results obtained from the

experiments examining RS- t in the LDM fraction should accurately reflect the status of

these proteins under these conditions. Lima et al. (1 L4), have s h o w that glucose/insulin

induced desensitization of glucose transport i n rat adipocytes occurred without effect on

IRS- 1 tyrosine phosphorylation. High glucose was also without effect on IRS-1 tyrosine

phosphorylation and PI3K activity in rat skeletal muscle (79)- In conuast, treatrnent of

3T3-L1 adipocytes with H202, inhibited insulin stirnulated IRS-1 tyrosine

phosphorylation and IRS-1 associated P13K activity in the LDM (1 15).

Although not statistically significant, basal levels of IRS-1 tyrosine

phosphorylation and association with p85 in high glucose/insuIin, and high

gIucose/insulin + NAC, treated cells tended to be higher compared to control cells. One

possibility for this effect coulci be that following the 18 h incubation there was some

residud effect of insulin remaining after the washing steps (1 14).

In the high glucose/insulin mode1 of insulin resistance, total levels of IRS-1 were

unchanged. Sorne insulin resistance States have shown to disptay decreased levels of

IRS-1 expression, for example, chronic insulin treatrnent of 3T3-Ll adipocytes ied to a

significant reduction in iRS-1 protein levels ( 1 17).

In addition to IRS-1, rat adipocytes also contain IRS-2, which is also located in

the LDM fraction (36). IRS-2 binds to p85 and can mediate the action of insulin to

stimulate Glut4 translocation (1 18). Therefore, it is possible that in rat adipocytes treated

with high glucose/insulin, IRS-2 tyrosine phosphorylation in response to insulin rnay be

impaired. Anai et al. (36) have shown that in rats rendered insulin resistant by high fat

feeding, expression of RS-2, and to a Iesser extent RS-1, was reduced in adipose cells.

. AktllPKBa activation by insulin was impaired in cells treated with high

giucose/insulin which was prevented by NAC. The decrease in kinase activity with high

glucose/insulin treatment was accompanied by decreased phosphorylation of Ser473

suggesting an upsueam signaling defect. PI3K activity and lipid products were not

measured in the present study, therefore the possibility rernains that there might be a

defect at this sep. There may also be a defect in the upstream kinase that phosphorylates

Ser473. Although not measured in this study, phosphorylation of Th308 may also be

reduced with high glucose/insulin treatment which wouid implicate defective PDK-1

activity. High glucose/insulin may activate protein phosphatase 2A (PPZA), a

serinehhreonine phosphatase which could dephosphorylate and inactivate AktlffKBu

(62). Co-incubation with NAC prevented the defect in Aktl/PKBcz activation and Ser473

phosphorylation suggesting that one point at which NAC is acting is upstream of

Akt l/PKBu and downstream of IRS- 1. A discordance between ES- 1fPI3K and

Akt l/PKBa activation has previously been shown in ceramide induced insulin resistance.

Summers et al. (1 19), have shown that exposing 3T3-LI adipocytes to the short-chain

ceramide analog, Cz-ceramide, impaired insulin stimulated glucose uptake, Glutl

translocation, and Akt l/PKBcr and ~kt2/PKBp activities. These defects occurred despite

normal insulin stimulated IRS- 1 tyrosine phosphorylation, association with p85, and

PI3K activity.

High glucose has previously k e n reported to impair Akt l/PKBa activation in

skelctal muscle (79). Hyperglycemia could impair Akt lPKBa by a PKC dependent

rnechanism (79). Bsuthel er al. (1 16) have shown that pnor activation of PKC by PMA

inhibited the subsequent ability of insutin to stimulate A k t l P K B a and Akt3PKBy

activities in 3T3-L1 fibroblasts and adipocytes. Hyperglycemia has been shown to

acti vate PKC (86). Interestingly, Hz02, has also been shown to activate PKC (120). If

oxidative stress is operative in the present high gIucose/insulin model, this rnechanism of

Akt l/PKBcx inactivation remains possible.

lmpaired AktIPKB activation has also been shown to occur in 3T3-L1 adipocytes

treated with HzOz Lipoic acid, an antioxidant like NAC, was able to prevent the

impairment in Aktl/PKBa activation induced by H301. These findings suggest that

Akt l/PKBa activation may be sensitive to the oxidation state of the cell.

Activation of A k t l P K B a has been associated with the translocation of Gfut4 to

the plasma membrane in response to insulin. Insulin resistant cells (high glucose/insulin

treated) displayed approximately 50% lower Akt 1 P K B u activation than the control o r

NAC treated cells. How much of a reduction in activity is necessary to impair Glut4

translocation is not yet known. Wang et al. (4 1) compared various dominant-negative

mutants of AktlPKBa on Glut4 translocation and showed that a 2- fold increase in

Akt l /PKBa activity was enough to fully translocate Glut4 to the plasma membrane. In

the present study, the resistant cells displayed a 7-foid increase in Akt l / P K B a activity

above basal. Whether the decrease in Akt l/PKBcc activity can account for the impairment

in Glut4 translocation is not certain.

In addition to Aktl/PKBa, rat adipocytes also express Akt2/PKw (121). Most

studies in the past have focused mainly on Aktl/PKBa. Recently, numerous studies have

implicated Akt2/PKBP in the stimulation of Glut4 translocation by insulin in rat

adipocytes (29,42). Preliminary data from our laboratory suggest that AktYPKBB

activation by insulin is not impriired with high glucose/insulin treatment.

Treatment with NAC alone led to a reduction in Aktl/PKBa activation by insulin

wi th no change in Ser473 phosphorylation. Why NAC reduced Akt l/PKBa activation is

presently not ciear. There are two possibilities that may explain the discordance between

Aktl/PKBu activation and Ser473 phosphorylation. First, full activation of AktlPKBa

requires phosphorylation on both Ser473 and Thr308 (21). Phosphorylation on either

residue alone results in only partial activation of the kinase (21). The possibility exists

that phosphorylation on Thr308 may be defective. Second, the kinase assay may be a

more sensitive measure that Ser473 imrnunoblotting.

There are several possibilities that may explain the different signaling defects

observed between the present study, from those observed in 3T3-L1 adipocytes exposed

to H?O1 (lO4,lO5,ll5). First, the cells used were different. Although both types represent

adipocytes, the two celi types may respond differently to various treatments. For

example, chronic incubation of 3T3-L1 adipocytes with insulin was able to markedly

decrease insulin-stimulated glucose uptake (78), white having a minimal effect in isolated

rat adipoc ytes (73). In addition, chronic incubation of 3T3-L 1 adipocytes wi th glucose

and insulin decreased Glut4 levels (78), something not seen in isolated adipocytes (73).

Second, the method of generating reactive oxygen species was different between the two

models of insulin resistance. In the HIOz model, reactive oxygen species were generated

in the media surrounding the cells, whereas the high glucose/insulin model suggests that

reactive oxygen species are generated inside the ce11 from metabolism. Reactive oxygen

species generated outside the ce11 may activate different stress pathways. It is possible

that cells may respond differently to various forrns of oxidative stress.

Overall, whife various modets of insulin resistance have a11 shown impaired

translocation of Glut4 to the PM in response to insulin, there have been no consistent

signahg defects observed. One possibility may be that different factors which induce

insulin resistance generate different signals resulting in differential defects in insulin

signaling. Another possibility may be the duration of the treatments. Some defects rnay

occur early on, while others take longer to manifest.

To ssummarize and conclude, NAC. an antioxidant and glutathione (GSH)

precunor, was able to prevent the impairment in Glut4 translocation induced by high

glucose/insuiin and glucosamine. NAC was also able to prevent the decrease in Glutl

expression and Akt l/PKBu activation and phosphorylation observed in ce1 1s treated wi th

high glucoselinsulin. These findings suggest a role for oxidative stress as a potential

mechanism in the induction of insulin resistance in this model. In addition, these findings

are relevant for Type II diabetic patients where hyperglycemia and hypennsulinemia are

IikeIy present and may worsen penpheral insulin resistance. Finally, NAC may provide

the potential design for novel therüpeutic agents.

GENERAL DISCUSSION

The studies described in this thesis (Chapters 2 and 3) were designed to

investigate the potential role of AktUPKBa in the stimulation of glucose uptake and

Glut4 translocation using 2 different models. In the first model (Chapter 2) using insulin,

vanadate, and pV to stimulate glucose uptake, the data support a role for AktlPKBa in

Glut3 translocation mediated by insulin. Thus wortmannin pretreatment of cultured rat

skeletal muscle cells (L6 cells) inhi bited P13K and subsequent activation of Akt l/PKBu

by insulin i n association with inhibition of insuiin-mediated Glut4 translocation and

glucose uptake (43). In the second model, further support for Akt l/PKBu in insulin

stimulated Glut4 translocation came from the study of Akt l/PKBa in rat adipoocytes

rendered insulin resistant by treatment with high glucose/insulin, In these cells insulin

stimulated Glut4 translocation and glucose uptake were impaired by 50% and 60%

respectively. In these same cells, Aktl/PKBa activation by insuIin was reduced by 44%.

Important1 y, CO-incubation with NAC, which prevented the impairment in Glut4

translocation and glucose uptake, also corrected the defect in Akt l/PKBa activation. The

close association between these parameters in both models strongly supports a role for

Akt l/PKBa in mediating insulin induced Glut4 translocation and subsequent glucose

uptake. Chronic treatment of rat adipocytes with glucosamine also resulted in impaired

Glut4 translocation and glucose uptalie stimulated by insulin. Co-incubation with NAC

prevented these defects. The activation of Akt l/PKBa in this third model has not yet

been studied but will be important to detennine. These data strongly suggest but do not

prove that the Akt lPKBa defect is responsible for the insulin resistance of glucose

transport. To funher examine this question, it would be worthwhile to activate

AktlPKBu in these insulin resistant cells. This could be accomplished by using okadaic

acid (56) or by transfecting a constitutively active forrn of the enzyme (38110). The

effects on GIut4 translocation could then be observed. To date, a specific inhibitor of

Akt l/PKBa has not been found (22). The majority of studies implicating AktlPKBu in

Glut4 translocation have used consti tutivel y active or dominant negati ve forms of the

enzyme. The development of a specific pharmacological inhibitor would also contribute

significantty to Our understanding of the role of this enzyme in the stimulation of Glut4

translocation by insulin.

In these studies the site of the glucose transport defect in the insulin resistant

(High GA treated) adipocytes appears to lie at least in part at the leveI of Aktl/PKBoc.

However, the cause of the defect in this enzyme is not yet clear. Thus, insulin-stimulated

tyrosine phosphorylation of IRS-1 and the association of IRS-1 with the p85 regulatory

subunit of PI3K was found to be normal. In general, this association implies an intact

ability of insulin to stimulate PUK and generate the lipid second messengers PtdIns-3,4-

PI and PtdIns-3,4,5-P3. The steps between PI3K and Akt l/PKBa include activation of

PDKl and PD= which are immediately downstrearn of PI3K and upstream of

Akt IPKBa. Future studies must determine whether these enzymes are normally

activated in response to insulin or are defective in the insuiin resistant adipocytes.

Another possibility to explain such a defect is increased activity of a lipid phosphatase

such as SHIP (SH2-domain containing inositol 3' phosphatase). This could potentially

result in a reduction in the active PI3K Iipid products in the face of normal PI3K activity.

Finally, the data presented here (Chapter 2) using vanadate and pV implicate that

at least one novel pathway independent o f PI3K and Aktl/PKBa exists by which glucose

uptake and Glut4 translocation may be stimulated.

References

Leibal, B.S., and G.A. Wrenshall. (1971). Insrrlirt- Published by the Canadian Diabetic Association, Banting and Best Department of Medical Research, University of Toronto.

Taylor, S.I. (1999). Deconstructing Type II diabetes. Cell97: 9-12.

Cushman, S., and Wardzala, L.J. (1980). Potential mechanism of insulin action on glucose transport in the isolated rat adipose cell. J. Biol. Chern. 255(10):4758- 4762.

Sweeney, G., Somwar, R., Rarnlal, T., Volchuk, A., Ueyama, A., and Klip, A. (1999). An inhibitor of p38 rnitogen-activated protein kinase prevents insufin- sti mulated glucose transport but not glucose transporter translocation in 3T3-L 1 adipocytes and L6 myotubes. J. Biol. Chem. 274(15): 10071-10078.

White, M.F. (1997). The insulin signaling system and the iRS proteins. Diubetologia 40: S2-S 1 7.

Myers, M.G., Jr., Xiao, J.S., and White, M.F. (1993). The IRS-1 signaling system. TZBS 19: 289-293.

Sun, X.J., et al. (1995). Role of IRS-2 in insulin and cytokine signaling. Nature 377: 173-177.

Smith-Hall, J., et al. (1997). The 60 kDa insulin receptor substrate functions like an iRS protein (pp601RS3) in adipose cells. Biochemisrry 36: 8304-83 10.

Fantin, V.R. (1998). Characterization of insulin receptor substrate 4 in hurnan embryonic kidney 293 cells. J. BioI. Chem. 273(17): 10726-10732.

Myers, M.G., Jr. er al. (1992). IRS-1 activates phosphatidylinositol 3'-kinase by associating with src homology 2 domains of p85. froc. Narl. Acad. Sci. USA 89: 10350-10354.

Domin, J., and Waterfield, M.D. (1997). Using structure to define the function of phosphoinositide 3-kinase family rnembers. FEBS Lerters 4 10: 9 1-95.

Rameh, L.E., and Cantley, L.C. (1999). The role of phosphoinositide 3-kinase lipid products in ce11 function. J. Biol. Chem. 274(13): 8347-8350.

Holman, G.D., and Kasuga, M. (1997). From receptor to transporter: insulin signaling to glucose transport. Diabetologia 40: 99 1- 1003.

Giorgetti, S., et al. (1993). The insulin and insulin-like growth factor-1 receptor substrate IRS-1 associates with and activates phosphatidylinositol 3-kinase in vitro. J. Biol. Chern. 268( 10): 7358-7364.

Boudewijn, M., Burgering, Th., and Coffer, P.J. ( 1995). Protein kinase B (c-Akt) in phosphatidylinosi tol-3-OH kinase signal transduction. Nature 376: 599-602.

Coffer, P.J., and Woodgett, J.R. (1991). Molecular cloning and characterization of a novel putative protein-serine kinase related to the CAMP-dependent and protein kinase C families. Ertr. J. Bioc/iern. 201: 47548 1.

Jones, P.F., Jakubowicz, T., Pitossi, F.J., Maurer, F., and Hemmings, B.A. (1991). Molecular cloning and identification of a sennehhreonine protein kinase of the second-messenger subfamily. Proc. Natl. Acad Sci. 88: 4 17 1-4 175-

Bellacosa, A., Testa, J.R., Staal, S.P., and Tsichlis, P.N. (1991). A retroviral oncogene, ah, encoding a serine-threonine kinase containing an Sm-like region. Scierice 254: 274-277.

Jones, P.F., Jakubowicz, T., and Hemmings, B.A. (199 1). Molecular cloning of a second form of rac protein kinase. Ce11 Reg. 2: 100 1- lOO9.

Konishi, H., et al. (1995). Molecular cloning and characterization of a new member of the rac protein kinase family: association of the pleckstrin homology domain of three types of rac protein kinase with protein kinase C subspecies and Py subunits of Ci proteins. Biochem. Biophys. Res. Conr. 2 l6(2): 526-534.

Alessi, D.R., Andjelkovic, M., Caudwell, B., Cron, P., Momce, N., Cohen, P., and Hemmings. B.A. (1996). Mechanism of activation of protein kinase 8 by insulin and IGF- 1. EMBO J. l5(23): 654 1-655 1.

Coffer, P., Jin, J., and Woodgett, J.R. (1998). Protein kinase B (c-Akt): a multifunctional mediator of phosphatidylinositol 3-kinase activation. Bioc/le~rz J. 335: 1-13.

Alessi, D.R., James, S.R., Downes, P.C., Holmes, A.B., Gafney, P., Reese, C.B., and Cohen, P. (1997). Characterization of a 3-phosphoinositide-dependent protein kinase which phosphorylates and activates protein kinase Ba. Cur. Biol. 7:26 1-269.

S tephens, L., Anderson, K., Stokoe. D., Erdjurnent-Bromage Hedi ye, Painter G.F., Holmes, A.B., Gaffney, P., Reese, C.B., McCormick, F., Tempst, P., Coadwell, S., and Hawkins, P.T. (1998). Protein kinase B kinases that rnediate phosphatidylinositol 3,4,5-triphosphate-dependent activation of protein kinase B. Scierzce 279: 7 10-7 13.

Bandyopadhyay, G., Standaen, M.L., Kikkawa, U., Ono, Y., Moscat, J., and Farese, R-V. (1999). Effects of transiently expressed atypical (5, A), conventional (a, p) and novel (6, E ) protein kinase C isoforms on insulin-stimulated translocation of epitope-tagged GLüT4 glucose transporters in rat adipocytes: specific interchangeable effects of protein kinases C-6 and C-A. Biocliem. J. 337: 46 1-470.

Standaert, M.L., et al. (1999). Insulin activates protein kinases C-5 and C-h by an autophosphorylation -dependent mechanism and stimulates their translocation to Glut4 vesicles and other membrane fractions in rat adipocytes. J. Biol. Clieni. 274(36): 25308-253 16.

Mueckler, M. ( 1994). Facili tative glucose transporters. Eur. J. Biochenz. 2 19 :7 13-735.

Fukumoto, H., Kayano, T., Buse, J.B., Edwards, Y., Pilch, P.F., BelI, G.I., and Seino, S. (1989). Cloninp and characterization of the major insulin-responsive glucose transporter expressed in human skeletal muscle and other insulin- responsive tissues. J. Biol. Chem. 264(14): 7776-7779.

Calera, M.R., Martinez C., Liu, H., El Jack, A., Birnbaum, M.J., and Pilch, P.F. (1998). Insulin increases the association of Akt-2 with Glut4-containing vesicles. J. Biol. CIze~n. 273(l3): 730 l-72O4.

Yang, J., and Holman, G.D. (1993). Cornparison of GLUT4 and GLUTl subcellular trafficking in basal and insulin-stimulated 3T3-L1 cells. J. Biol. Cllem. 268(7): 4600-4603.

Satoh, S., Nishimura, H., Clark, A.E., Kozka, I.J., Vannucci, S.J., Simpson, I.A., Quon, M.J., Cushman, S.W., and Holman, G.D. (1993). Use of bismannose photolabel to elucidate insuiin-regulated GLUT4 subcelluIar trafficking kinetics in rat adipose cells. Evidence that exocytosis is a critical site of hormone action. J. Biol. Chem. 268(24): 17820-17829.

Katagiri, H., et al. (1997). Roles of PI 3-kinase and ras on insulin-stimulated glucose transport in 3T3-L1 adipocytes. Am. J. Physiol. 372 (Erldocrinol. Metab. 35): E326-E33 1.

Jiang, T., Sweeney, G., Rudolf, M.T., Klip, A., Traynor-Kaplan, A.. and Tsien, R.Y. (1998). Membrane-permeant esters of phosphatidylinositol 3,4,5- trisphosphate. J. Biol. Chen?. 273(l8): 1 1017-1 1024.

Wiese, R.J., Mastick, C C , Lazar, D.F., and Saltiel, A.R. (1995). Activation of mitogen-activated protein kinase and phosphatidylinositol 3-kinase is not sufficient for the hormonal stimulation of glucose transport, lipogenesis, or gl ycogen s ynthesis in 3T3-L 1 adipoc ytes. J. Biol. Chern. 270: 342-3446.

Kelly, K.L., Ruderman, N.B., and Chen, K.S. (1992). Phosphatidylinositol-3- kinase in isolated rat adipocytes. J. Biol. Chern. 267(5): 3423-3428.

Anai, M. et al. (1998). Different subcelluIar distribution and reguiation of expression of insulin receptor substrate (ES)-3 from those of RS-1 and RS-2. J. Biol. Chem. 273(45): 29686-29692.

Liu, S.C.H., Wang, Q., Lienhard, G.E., and Keller, S.R. (1999). Insulin receptor substrate 3 is not essential for growth or glucose homeostasis. J. Biol. Chenl. 274(25): 18093- 18099.

Kohn, AD., Summers, S.A., Birnbaum, M.J., and Roth, R.A. (1996). Expression of a constitutively active Akt sedthr kinase in 3T3-L1 adipocytes stimulates glucose uptake and glucose transporter 4 translocation. J. B i d . Cherit. 271(49): 3 1372-3 1378.

Tanti, J.F., Grillo, S., Gremeaux, T., Coffer, PJ., Obberghen, E., and Marchand- Brustel, Y. (1 997). Potential role of protein kinase B in glucose transporter 4 translocation in adipocytes. Endocrinology 1 38(5): 2005-20 10.

Ueki, K . er al. (1998). Potential role of protein kinase B in insulin-induced glucose transport, glycogen synthesis, and protein synthesis. J. Biol. Cherri. 273(9): 53 15-5322.

Wang, Q., Somwar, R., Bilan, P.J., Liu, Z., Jin, J., Woodgett. J.R., and Klip, A. ( 1999). Protein kinase B/Akt participates in GLUT4 translocation by insulin in L6 rnyoblats. Mol. Cel. Biol. 19(6): 4008-4018.

Kupriyanova T. A., and Kandror, V. (1999). Akt-2 binds to Glut4containing vesicles and phosphorylates their componeni proteins in response to insulin. J. Biol. Cltem. 274(3): 1458- l4W.

Tsiani, E., Bogdanovic, E., Sorisky, A., Nagy, L., and Fantus, I.G. (1998). Tyrosine phosphatase inhibitors, vanadate and pervanadate, stimulate glucose transport and GLUT translocation in muscle cells by a rnechanism independent of phosphatidylinositol 3-kinase and protein kinase C . Diabetes 47: 1676- 1686.

Aruoma, O.J., Halliwefl, B., Hoey, B.M., and Butler, J. (1989). The antioxidant action of N-acetylcysteine: its reaction with hydrogen peroxide, hydroxyl radical, superoxide, and h ypochlorous acid. Free Rad. Biol. Med. 6(6): 593-597.

Heyliger, C.E., Tahiliani, A.G., McNeill, J.H. (1985). Effect of vanadate on elevated blood glucose and depressed cardiac performance of diabetic rats. Science 227: 1474- 1476.

Shechter, Y. (1990). Insulin-mimetic effects of vanadate. Diaberes 39: 1-5.

Swarup, G., Cohen, S., and Garbers, D.L. (1982). Inhibition of membrane phosphotyrosyl-protein phosphatase activity by vanadate. Biochern. Biophys. Res. Conzniun. 107(3): 1 104- 1 109.

Tolman, E., Barris, E., Burns, M., Pansini, A., and Partridge, R. (1979). Effects of vanadium on glucose metabolism in vitro. Lge Sci. 25: 1 159-1 164.

Goldfine, A., Simonson, D.C., Folli, F., Patti, M.E., and Kahn, C.R. (1995). III vivo and in vitro studies of vanadate in human and rodent diabetes mellitus. Mol. Ce11 Bioclzem. 153: 2 17-23 1.

Kadota, S., Fantus, I.G., Deragon, G., Guyda, H., Hersh, B., and Posner, B. (1987). Peroxide(s) of vanadium: a novel and potent insulin-mimetic agent which activates the insulin receptor kinase. Biochern. Biophys. Res. Cornrnun. 147: 259- 266.

Posner, B.I., Faure, R., Burgess, J.W., Bevan, A.P., Lachance, D., Zhang-Sun, G., Fantus, I.G., Ng, J.B., Hall, D.A., and Lum, B.S. (1994). Peroxovanadium compounds. A new class of potent phosphotyrosine phosphatase inhibitors which are insulin mimetics. J. Biol. Chenr. 269(6): 4596-46W.

Fantus, I.G., Kadota, S., Deragon, G., Foster, B., and Posner, B. 1. (1989). Pervanadate [peroxide(s) of vanadate] mimics insulin action in rat adipocytes via activation of the insulin receptor tyrosine kinase. Bioclieriristp 28: 8864-8871.

Tsiani, E. Abdullah, N., and Fantus, I.G. (1997). Insulin-mimetic agents vanadate and pervanadate stimulate glucose but inhibit amino acid uptake. Am. J. Pkysiul. 272 (Cell Pltysiol. 4 1): C 156-C 162.

Wilden, P.A., and Broadway, D. (1995). Combination of insulinomimetic agents Hz02 and vanadate enhances insulin receptor mediated tyrosine phosphorylation of IRS-1 leading to IRS-1 association with the phosphatidylinositol 3-kinase. J. Cell. Bioclrem. 58(3): 279-91.

Ui, M., Okada, T., Hazeki, K., and Hazeki, 0. (1995). Wortmannin as a unique probe for an intracellular signaling protein, phosphoinositide 3-kinase. TIBS 20: 303-307.

Jullien, D., Tanti, J.F., Heydrick, S.J., Gautier, N., Gremeaux, T., Obberghen, E., and Marchan-Brustel, Y. (1993). Differential effects of okadaic acid on insulin- stimulated glucose and arnino acid uptake and phosphatidylinositol 3-kinase activity. J. Biol. Clzem. 268(20): 15246- 1525 1.

Konishi, H., Matsuzalu, H., Tanaka, M., Yoshitaka, O., Tokunaga, C., Kuroda, S., and Kikkawa, U- (1996). Activation of RAC-protein kinase by heat shock and hyperosmolarity stress through a pathway independent of phosphatidylinositol 3- kinase. Proc. Nufl. Acad. Sci. USA 93: 7639-7643.

Sable, CL. , Filippa, N., Hemmings, B., and Obberghen, E. (1997). CAMP stimulates protein kinase B in a wortmannin-insensitive rnanner. FEBS Letrers 409: 253-257.

Moule, S.K., Welsh, G.I., Edgell, N.J., Foulstone, E.J., Proud, C G . , and Denton, R.M. (1997). Regulation of protein kinase B and glycogen synthase kinase-3 by insulin and P-adrenergic agonists in rat epididymal fat cells. J. Biol. Chem. 272(12): 77 13-7719.

Alessi, D.R., Caudwell, B.F., Andjelkovic, M., Hemmings, B.A., and Cohen, P. (1996). Molecular basis for the substrate specificity of protein kinase 8; cornparison with MAPKAP kinase-1 and p70 S6 kinase. FEBS Letfers 399: 333- 338.

Kohn, A.D., Kovacina, K.S., and Roth, R.A. (1995). Insulin stimulates the kinase activity of RAC-PK, a pleckstnn homology domain containing ser/thr kinase. EMBO J. l4(17): 4288-4295.

Andjelkovic, -M., Jakubowicz, T., Cron, P., Ming, X., Han, J., and Hemmings, B.A. (1996). Activation and phosphorylation of a pleckstrin homology domain containing protein kinase (RAC-PWPKB) promoted by semm and phosphatase inhibitors. Proc. Natl. Acad. Sci. USA 93: 5699-5704.

Wijkander, J., Stenson Holst, L., Rahn, T., Resjo, S., Castan, I., Manganiello, V., Belfrage, P., and Degerman, E. (1997). Regulation of protein kinase B in rat adipocytes by insulin, vanadate, and peroxovanadate. J. Biol. Clrern. 272(34): 21520-21526.

Huyer, G., Liu, S., Kelly, J., Moffat, J., Payette, P., Kennedy, B., Tsaprailia, G., Gresser, M.J., and Ramachandran, C. (1997). Mechanism of inhibition of protein- tyrosine phosphatases by vanadate and pervanadate. J. Biol. Chern. 272(2): 843- 85 1.

Shaver., Ng, J.N., Hall, D.A., Soo Lum, B., and Posner, B. 1. (1993). Irzorg. Clzern. 32: 3 109-3 113.

Foot, E., Bliss, T., Femandez, L.C., Da Costa, C., and Leighton, B. (1992). The effects of orthovanadate, vanadyl and peroxides of vanadate o n glucose metabolisrn in skeletal muscle preparations in vitro. Mol. Cell. Biochern. 109: 157-162.

Lee, A.D., Hansen. P.A., and Holloszy, J.O. (1995). Wortmannin inhibits insulin- stimulated but not contraction - stimulated glucose transport activity in skeletal muscle. FEBS Letters 36 1 : 5 1-54.

Brozinick, J.T., and Bimbaum, M.J. (1998). Insulin, but not contraction, activates Akt~PKB iri isolated rat skeletal muscle. J. Biol. Clzern. 273(24): 14679-14682.

Chen, D., Elrnendorf, J.S., Olson, A.L., Li, X., S helton Earp, H., and Pessin, J.E. (1997). Osmotic shock stimulates GLUT4 translocation in 3T3-L1 adipocytes by a novel tyrosine kinase pathway. J. Biol. Clietn. 272(43): 27401-274 10.

Kahn, B.B. (1998). Type 2 diabetes: when insulin secretion fails to compensate for insul in resistance. Cell92: 593-596.

Olefsky, J.M. and Nolan. J.J. (1995). Insulin resistance and non-insulin- dependent diabetes mellitus: ceIlular and moiecular mechanisms. Am. J. CZNz. Nrctr. 6 1 (suppl): 980s-986s.

O'Doheny, R., Stein, D., and Foley, J. (1997). Insulin resistance. Diabetologio 40: B10-B15.

Garvey, W.T., Olefsky, LM., Matthaei, S., and Marshall, S. (1987). Glucose and insulin CO-regulate the glucose transport system in primary cultured adipocytes. J. BioL Chenz. 262(1): 189-197.

Traxinger, R.R., and Marshall, S. (1989). Role of amino acids in modulating glucose-induced desensitization of the glucose transport system. J. Biol. Cllem. 264(35): 209 10-209 16.

Marshall, S., Bacote, V., and Traxinger, R.R. (1991). Discovery of a metabolic pathway mediüting glucose-induced desensitization of the glucose transport systern. J. Biol. Cherri. 266(8): 4706-47 12.

Cooksey, R.C., Hebert, L-F-, Zhu, J., Wofford, P., Garvey, Y., and McClain, D.A. (1999). Mechanism of hexosamine-induced insulin resistance in transgenic mice overexpressing g1utamine:fnictose-6-phosphate amidotransferase: decreased glucose transporter GLUT4 translocation and reversal by treatment with thiazolidinedione. Erldocrinology 140(3): 1 15 1 - 1 157.

Rossetti, L.. Hawkins, M., Chen, W.. Gini, J., and Brazilai, N. (1995). In vivo glucosamine infusion induces insulin resistance in normoglycemic but not in hyperglycemic conscious rats. J. Clirt. Irivest. 96: 132-140.

Thompson, M.J., Williams, M.G., and Frost, S C . (1996). Development of insulin resistance in 3T3-L1 adipocytes. J. Biol. Chem. 272(12): 7759-7764.

Kurowski, T.G., Lin, Y., Luo, Z., Tsichlis, P.N., Buse, M.G., Heydrick, S.J., and Ruderman, N.B. (1999). Hypergiycemia inhibits insulin activation of Akt/protein kinase B but not phosphatidylinositol 3-kinase in rat skeletal muscle. Diabetzs 48: 658-663.

Furnsinn, C., Sanderson, A.L., Radda, G.K., and Leighton, B. (1995). Effects of glucosarnine on insulin-stimulated glucose metabolism in rat soleus muscle. Int. J. Biochem. CelZ BioZ. 27(8): 805-8 14.

Hawkins, M., Barzilai, N., Liu, R., Hu, M., Chen, W., and Rossetti, L. (1997). Role of the glucosamine pathway in fat-induced insulin resistance. J. Clin. Invest. 99: 2 173-3 183.

Virkarnaki, A., Daniels, M.C., Hamalainen, S., Utriainen, T., Mcclain, D., and Yki-Jarvinen, H. (1997). Activation of the hexosamine biosynthesis pathway by glucosamine in vivo induces insulin resistance in multiple insuiin sensitive tissues. Endocrinology 138(6): 3501-2507.

Baron, A.D., Zhu, JS., Zhu, JH., Weldon, H., Maianu, L,, and Garvey, T.W. (1995). Glucosamine induces insulin resistance in vivo by affecting GLUT4 translocation in skeletal muscle. J. Clin. Invest. 96: 2792-2801.

Cline, G.W., er al. (1999). hpaired glucose transport as a cause of decreased insulin-stimulated muscle glycogen synthesis in type 2 diabetes. N. E~zgl. J. Med. 34 1 : 240-246.

Zierath, J.R., et al. (1996). Insulin action on glucose transport and plasma membrane CLUT4 content in skeletal muscle from patients with NiDDM. Diabetologia 39: 1 180- 1 189.

Muller, H.K., Kellerer, iM., Errnel, B., Muhlhofer, A., Oberrnaier-Kusser, B., Vogt, B., and Haring, H.U. (1991). Prevention of protein kinase C inhibitors of glucose-induced insulin receptor tyrosine kinase resistance in rat fat cells. Diabetes 40: 1440- 1448.

Hresko, R.C., Heimberg, H., Chi, M., and Mueckler, M. (1998). Glucosamine- induced insulin resistance in 3T3-LI adipocytes is caused by depletion of intracellular ATP. J. Biol. Chenl. 273(32): 20658-20668.

Kim, YB., Zhu, JS., Zierath, J., Shen, HQ., Baron, A.D., and Kahn, B.B. (1999). Glucosamine infusion in rats rapidly impairs insulin stimulation of phosphoinositide 3-kinase but does not alter activation of Aktlprotein kinase B in skeletal muscle. Diaberes 48: 3 10-320.

Patti, ME., Virkarnaki, A., Landaker, E.J., Kahn, R.C., and Yki-Jarvinen, H. (1999). Activation of the hexosamine pathway by glucosamine in vivo induces

insulin resistance of early postreceptor insulin signaling events in skeletal muscte. Diubetes 48: 1562- 157 1.

Kim, YB., Nikoulina, S.E., Ciaraldi, T.P., Henry, R.R., and Kahn, B.B. (1999). Normal insulin-dependent activation of Aktlprotein kinase B, with diminished activation of phosphoinositide 3-kinase, in muscle in type 2 diabetes. J. Clin. Invest. 104: 733-74 1 .

Lee, TS., Saltsman, KA. , Ohashi, H., and King, G.L. (1989). Activation of protein kinase C by elevation of glucose concentration: proposa1 for a mechanism in the developrnent of diabetic vascular complications. Proc. Natl. Acad. Sei. 86: 5 141-5 145.

Filippis, A., Clark, S., and Proietto, J. (1997). Increased flux through the hexosamine biosynthesis pathway inhi bits glucose transport acutel y by activation of protein kinase C . Biochenz. J. 324: 981-985.

Pillay, T.S., Xiao, S., and Olefsky, J.M. (1995). Glucose-induced phosphorylation of the insulin receptor. Functional effects and characterization of phosphorylation sites. J. Clin. Invest. 97: 613-620.

Kellerer, M., Mushack, J., Seffer, E., Mishack, H., Ullrich, A., and Haring, H.U. (1998). Protein kinase C isoforms alpha, delta, and theta require insulin receptor substrate-1 to inhibit the tyrosine kinase activity of the insulin receptor in human kidney embyonic cells (HEK 293 cells). Diabetologiu 41(7): 833-838.

Haltiwanger, R.S., Grove, K., and Philipsberg, G.A. (1998). Modulation of O- linked N-Acetylglucosarnine levels on nuclear and cytoplasmic proteins in vivo using the peptide O-GlcNAc-B-N-acetylglucosaminidase inhibitor O-(?- Acetanido-2-deoxy-D-glycopyranosylidene)amino-N-phenylcarbamate. J. Biol. Chetn. 273(6): 361 1-3617.

Wang, J., Liu, R., Hawkins, M., Barzilai, N., and Rossetti, L. (1998). A nutrient- sensing pathway regufates leptin gene expression in muscle and fat. Nutr~re 393: 684-688.

Paolisso, G., D'Amore, A., et al. (1994). Evidence for a relationship between oxidative stress and insulin action in non-insulin dependent (type II) diabetic patients. Metabolisni 43(11): 1426- 1429.

Cross, C.E., et al. (1987). Oxygen radicals and human disease. Ann. Inrer. Med. 107: 526-545.

Dandona, P., Thusu, K., Cook, S., Snyder, B., Makowski, J., Armstrong, D., and Nicotera, T. (1 996). Oxidative damage to DNA in diabetes melli tus. Lancer 347: 444-445.

Ihara, Y., Toyokuni, S., et al. (1999). Hyperglycemia causes oxidative stress in pancreatic B-cells of GK rats, a mode1 of type 2 diaberes. Diaberes 48: 927-932.

Zadeh-Nourooz, J., Rahimi, A., Tajaddini-Saadi, J., Tritschler, H., Rosen, P., Halliwell, B., and Betteridge, D.J. (1997). Relationships between plasma measures of oxidative stress and metabolic control in NIDDM. Diaberologia 40: 647-653.

Hunt, J.V., Dean, R.T., Wolff, S.P. (1988). Hydroxyl radical production and autoxidative glycosylation. Biochem. J . 256: 205-2 12.

Paolisso, G., and Giugliano, D. (1996). Oxidtive stress and insulin action: is there a relationship? Diabe~oiogia 39: 357-363.

Rudich, A., Tirosh, A., Potashnik, R., Hemi, R., Kanety, H., and Bashan, N. (1998). Prolonged oxidative stress impairs insulin-induced GLUT4 translocation in 3T3-LI adipoc ytes. Diaberes 47: 1562- 1569.

Rudich, A., Tirosh, A., Potashnik, R., Khamaisi, M., and Bashan, N. (1999). Lipoic acid protects against oxidative stress induced impairment in insulin stimulation of protein kinase B and glucose transport in 3T3-LI adipocytes. Diubetologia 42: 949-957.

Rodbell, M. (1963). Metabolism of isolated fat cells. J. Biol. Cltem. 239(2): 375- 380.

Simpson, I.A., er al. (1983). Insulin-stimulated translocation of glucose transporters in the isolated rat adi poce cells: characterization of subcellular fractions. Bioclrem. Biopltys. Acta. 763: 393-407.

Laemmli, U.K. (L970). Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227: 680-685.

Paz, K., Hemi, R., LeRoith, D., Karasik, A., Elhanany, E., Kanety, H., and Zick, Y. (1997). A rnolecu1a.r basis for insulin resistance. J. Biol. Cltem. 272(47): 299 1 1-299 18.

El-Kebbi, LM., Roser, S., Pollet, R.J., Cushman, S. W., and Wilson, C.M. (1994). Regulation of the Glutl glucose transporter in cultured myocytes: total number and subcellular distribution as determined by photoaffinity labelling. Biociier7r. J. 301: 35-40.

Tordjman, KM., Leingang, KA. , James, D.E., and Meuckler, M.M. (1989).

Differential regulation of two distinct glucose transporter pecies xpressed in 3T3- L1 adipocytes: effect of chronic insulin and tolbutamide treatment. Proc. Narl. Acad Sei. 86: 776 1-7765.

Walker, P.S., Ramlal, T., Sarabia, V., Koivisto, U., Bilan, P.J., Pessin, J.E., and Klip, A. (1989). Glucose transport activity in L6 muscle cells is regulated by the coordinate control of subcellular glucose transporter distribution, biosynthesis, and mRNA transcription. J. Biol. Chent. 265(3): 15 16- 1523.

Klip, A., Tsakiridis, T., Marette, A., and Ortiz, P.A. (1994). Regulation of expression of glucose transporters by glucose: a review of studies in vivo and in ce11 cultures. FASEB J. 8: 43-53.

Lima, F. B., Thies, R.S., and Garvey, T. (1991). Glucose and insulin regulate insulin sensitivity in primary cultured adipocytes without affecting insulin receptor kinase. EndocTi)zology 128: 24 15-2426.

Tirosh, A., Potashnik, R., Bashan, N., and Rudich, A. (1999). Oxidative stress disnipts insulin-induced cellular redistribution of insulin receptor substrate- 1 and phosphatidylinositol 3-kinase in 3T3-LI adipocytes. J. Biol. Cheni. 274(15): 10595- 10602.

Barthel, A., Nakatani, K., Dandekar, A.A., and Roth, R.A. (1998). Protein kinase C modulates the insulin-stimulated increase in Akt 1 and Akt3. Bioclzerii. Bioplrys. Res. Comm. 243(2): 509-5 1 3.

Ricort, J.M., Tanti, J.F., Van Obberghen, E., and Marchand-Brustel, Y.L. ( 1995). Alterations in insulin signalling pathway induced by prolonged insulin treatment of 3T3-Ll adipocytes. Diabetologia 38: 1148-1 156.

Zhou, L., Chen, H., Lin, C H . , Cong, L., McGibbon, M.A., Sciacchitani, S., Lesniak, MA., Quon, M.J., and Taylor, S.I. (1997). Insulin receptor substrate-3 IRS-2) can mediate the action of insulin to stimulate translocation of GLUT4 to the ceIl surface in rat adipose cells. J. Biol. Ciieni. 272(47): 29829-28833.

Summers, S.A., Luis, A.G., Zhou, H., and Birnbaum, M.J. (1998). Regulation of insulin-stimulated glucose transporter GLUT4 translocation and Akt kinase activi ty by ceramide. Mol. Cell. Biol. l8(9): 5457-5464.

Konishi, H., et al. (1997). Activation of protein kinase C by tyrosine phosphorylation in response to Hz02. Proc. Narl. Acad. Sci. USA 94: 1 1233- 11237.

Walker, K.S., Deak, M., Paterson, A., Hudson, K., Cohen, P., and Alessi, D.R.

(1998). Activation of protein kinase B and y isoforms by insulin in vivo and by 3-phosphoinositide-dependent protein kinase- 1 in vitro: cornparison with protein kinase B a. Biochem. J. 33 1: 299-308.

Permission Granted 01-13-00 Carol yn Hilyard, Permissions Editor

Copyright by the Amencan Diabetes Association. I I mrge Fantus, M.D., F.R.C.F!(C)

Faculty (d Mrdicinr University of Tiartwiir~

Divikinn of Endm-rino/m.v and Metabolism Dia bttes Cïiniral R escarch L'nit

December 1 7.1999 F A X 703-683-2890

IBcrrnissions Editor Amcrican Diabetcs Association Inc. 1 70 1 North Beauregard Street Alexandria. VA 223 1 1

To Whom J t May Conctm,

I am writing to scck permission for my student Elena Bogdanovic to rcpubli orgindly published in Our article i n m i n her M.Sc. thesis. The arc Fig.8 and Fig.9 found in Tsiam. E.. Bogdanovic, E., Sorisky, A.. Nagy. L.. Fanius, 1 .G . -'Tyrosine Phosphatase Inhi bitors, Vanadate and Pcrvmadate. Stimulate Gl u c ~ s c 'I'ransport and GLUT Translocation in Muscle Cells by a Mechanism Independeni oï I~hosphatidyi inositol 3-kinase and Protein Kinase C", Diabetcs 47: 1676- 1680. 1998. - 'l'hank you for your consideration of this rcquest.

Sincerely.

%Qv,/L% 1.G. FZIJI~US. M.D., F.RC.P.(C) IGFIbjB

600 University Avcnuc, Suitc 780, Toronto, Ontario MX; 1x5

89