Jennifer Kricker Thesis (PDF 3MB)

Structural investigations into the relationships of the

insulin-like growth factors (IGFs) and IGF binding

proteins (IGFBPs) with vitronectin (VN)

By

Jennifer Ann Kricker

Bachelor of Applied Science (Hons), QUT 1999

Associate Diploma of Clinical Techniques, QUT 1994

School of Life Sciences

Queensland University of Technology

Brisbane, Australia

A thesis submitted for the degree of Doctor of Philosophy

of the Queensland University of Technology

2004

ii

STATEMENT OF ORIGINALITY

The work contained in this thesis has not been previously submitted for a degree or

diploma at any other higher education institution. To the best of my knowledge and

belief, the thesis contains no material previously published or written by another

person except where due references are made.

SIGNED

DATE

iii

ACKNOWLEDGEMENTS

This work was supported by the Queensland Cancer Fund and the Queensland

University of Technology Small Grants Scheme. Travel was funded by the School

of Life Sciences, QUT Grants-In-Aid and the Biotechnology Innovation Scheme,

while my scholarship was funded with a Dean’s Doctoral Scholarship and by Assoc.

Prof. Zee Upton.

I would like to thank my supervisors Assoc. Prof. Zee Upton and Prof. Adrian

Herington for giving me the opportunity to partake in this project. Their endless

support has guided me through the last four years, which have been very trying at

times. Both Zee and Adrian have been extremely busy throughout my PhD yet their

door was almost always open. Zee has been a fantastic supervisor who provided me

with opportunities to push myself harder and further than I normally would. She has

encouraged me and made it possible for me to travel to conferences both nationally

and overseas, as well as introducing me to many great researchers. She has been an

inspiration and I thank her for that and her emotional support. In my times of panic,

Adrian has always smiled and managed to make me smile, and make the problem

seem simple and calm me down! He also has been a great supervisor, and always

had 5 minutes spare to hear my concerns when he was flat out, or just to have a chat

about anything and everything. Thanks also to Dr Terry Walsh for his help,

especially during my trying times of the BIAcore and while Adrian was away.

I would also like to thank the guys at uni… I don’t know if I could have kept my

sanity without Caz. Her friendship and support both in and out the lab has been

enormous. Thanks also to the TBRI group – especially the ‘nadding’ team, mr

oCean, Tony, Brett and Jos for their help and support, escorts to the pub, pancakes

and pirates! Many thanks to Alex “Stats Man” Anderson and Sean McElwain, who

both patiently tried to show me there could be a method to the madness when it

came to numbers! I’m also grateful to those in and around the lab, for their help and

friendship, especially Marcus, Cyril, Dan, Scott and participants of research

meetings. Thanks also to those from the level 5 & 6 labs and the Biochem, Micro &

AEMF staff!

iv

Lastly, but definitely not least, BIG, BIG thanks to my family who have given me

endless support and tried to understand what it is that I do – I can’t thank you

enough! They have always believed in me and listened to me grumble, and just

patiently waited while one storm passed after another. Thanks also to my best

friends, especially Ady and Jimmy… who were also in it for the long-haul and do

what friends do best ☺ “If in doubt, go to the beach!”

To be inspired. That is the thing.

To be possessed; to be bewitched.

To be obsessed. That is the thing.

To be inspired.

From Tiger’s Eye vol.I no.5

“Quickly, bring me a beaker of wine,

so that I may wet my mind and say something clever”

ARISTOPHANES

v

ABSTRACT

Previous studies demonstrated that IGF-II binds directly to vitronectin (VN) while

IGF-I binds poorly. However, binding of VN to integrins has been demonstrated to

be essential for a range of IGF-I-stimulated biological effects including IGF binding

protein-5 (IGFBP-5) production, IGF type-1 receptor autophosphorylation and cell

migration. Thus, this study examined the hypothesis that a link between IGF-I and

VN must occur and may be mediated through IGFBPs. Studies using competitive

binding assays with VN and [125I]-labelled IGFs in the absence and presence of

IGFBPs revealed IGFBP-4, IGFBP-5 and non-glycoyslated IGFBP-3 significantly

enhance binding of IGF-I to VN, while IGFBP-2 and glycosylated IGFBP-3 had a

smaller effect. Furthermore, binding studies with analogues indicate that

glycosylation status of IGFBP-3 and the heparin-binding domains of IGFBP-3 and

IGFBP-5 are important in this interaction. The functional significance of IGFs

binding to VN on cell migration in MCF-7 breast carcinoma cells was examined and

cell migration was found to be enhanced when VN was pre-bound to IGF-I in the

presence of IGFBP-3, -4 and -5. The effect required IGF:IGFBP:VN complex

formation; this was demonstrated by use of a non-IGFBP-binding analogue, des(1-

3)IGF-I. Additionally, higher doses of IGFs in the presence of VN also could

stimulate cell migration. Together, these data indicated the importance of IGFBPs

in modulating IGF-I binding to VN and that this binding has functional

consequences in cells. Future directions for this work include investigations into the

mechanisms underlying formation of the trimeric complex and the associated

signalling pathways involved.

vi

TABLE OF CONTENTS

Statement of Originality ...….….……ii

Acknowledgements ..….….……iii

Abstract ...…….……..v

Table of Contents ...….….……vi

List of Figures ....….….…...xi

List of Tables ...…………xiv

List of Abbreviations ...….……....xv

List of Publications and Presentations ...……..….xvii

CHAPTER 1: LITERATURE REVIEW

1.1 Introduction ...….….…….2

1.2 Insulin-like Growth Factors ...….….…….2

1.2.1 Structure of the IGFs ...….….…….3

1.2.2 Biological Activities of the IGFs ...….….…….4

1.3 IGF Modulation ...….….…….6

1.3.1 Type-1 IGF Receptor ...….….…….7

1.3.2 Type-2 IGF Receptor / CIMPR ...….….…….9

1.4 IGF-Binding Proteins ...….……....12

1.4.1 Structure of the IGFBPs ...….……....12

1.4.2 Functions of the IGFBPs ...….……....12

1.5 Post-translational Modification of the IGFBPs ...….……....15

1.5.1 Glycosylation ...….……....16

1.5.2 Phosphorylation ...….……....17

1.5.3 Proteolysis ...….……....18

1.6 Inhibition and Potentiation of IGF Action by IGFBPs ...….……....20

1.7 Novel IGF Binding Proteins ...….……....21

1.8 Vitronectin ...….……....22

1.8.1 Structure of VN ...….……....22

1.8.2 VN and its Integrin Receptors ...….……....25

1.8.3 Various Functions of VN ...….……....26

1.9 Interactions between the IGF Axis and VN ...….……....28

vii

1.10 Conclusions ...….……....30

1.11 Outline of Project ...….……....32

1.11.1 Hypotheses ...….……....32

1.11.2 Aims ...….……....32

CHAPTER 2: MATERIALS AND METHODS

2.1 Materials ...….……....34

2.2 Proteins and Cell Lines ...….……....34

2.3 Radiolabelling of Proteins ...….……....35

2.4 SDS-PAGE ...….……....36

2.5 Radioligand Blots ...….……....36

2.6 Solid Plate Binding Assay (SPBA) ...….……....37

2.7 Polyethylene Glycol (PEG) Precipitation Assay ...….……....37

2.8 Cell Culture ...….……....38

2.9 Protein Synthesis Assay ...….……....39

2.10 Transwell Migration Assay ...….……....39

2.11 Scanning Electron Microscopy ...….……....40

2.12 Statistical Analysis ...….……....41

CHAPTER 3: EFFECT OF IGFBPs ON IGFs BINDING TO VN

3.1 Introduction ...….……....43

3.2 Experimental Procedures ...….……....44

3.3 Results ...….……....45

3.3.1 Effects of IGFs and IGF-II analogues on [125I]-

IGF-II binding to VN

...….……....45

3.3.2 Effect of IGFBPs on modulating binding of

[125I]-IGF-II to VN ...….……....47

3.3.3 Effect of IGFBPs on modulating binding of

[125I]-IGF-I to VN ...….……....47

3.3.4 Ability of IGF peptides to compete for binding

of [125I]-IGF-I to VN in the presence of

IGFBP-3 or IGFBP-5 ...….……....47

3.4 Discussion ...….……....51

viii

CHAPTER 4: EFFECT OF GLYCOSYLATION AND HEPARIN-

BINDING DOMAINS ON IGF:IGFBP:VN

INTERACTIONS



4.3 Results ...….……....62

4.3.1 Comparison of the effects of glycosylated and

non-glycosylated IGFBP-3 on [125I]-IGF-I

binding to VN

...….……....62

4.3.2 Comparison of the ability of IGFBP-3

preparations with different states of

glycosylation to bind [125I]-IGF-I

...….……....62

4.3.3 Comparison of the effects of glycosylated and

non-glycosylated IGFBP-6 on [125I]-IGF-I

binding to VN

...….……....66

4.3.4 Binding of non-glycosylated IGFBP-6 and

glycosylated IGFBP-6 to [125I]-IGF-II

...….……....68

4.3.5 Importance of IGFBP-3 HBD in mediating

[125I]-IGF-I binding to VN

...….……....71

4.3.6 Comparison of [125I]-IGF-I binding to wild-

type IGFBP-3 and HBD mutants of IGFBP-3

...….……....71

4.3.7 Importance of IGFBP-5 HBD in mediating

[125I]-IGF-I binding to VN

...….……....74

4.3.8 Comparison of [125I]-IGF-I binding to wild-

type IGFBP-3 and HBD mutants of IGFBP-5

...….……....74

4.3.9 Examination of IGFBP binding to an IGF-I

analogue containing a HBD (MBF)

...….……....77

4.4 Discussion ...….……....80

ix

CHAPTER 5: FUNCTIONAL RESPONSES TO IGF:IGFBP:VN

INTERACTIONS



5.3 Results ...….……....88

5.3.1 Effects of IGFs and IGFBPs in the presence of

VN on protein synthesis in CHO-K1 cells

...….……....88

5.3.2 Migration of MCF-7 cells following exposure

to increasing amounts of IGF-I in the

presence of VN alone or in combination with

IGFBP-5

...….……....90

5.3.3 Migration of MCF-7 cells following exposure

to increasing amounts of IGFBP-5 in the

presence of VN alone or in combination with

IGF-I

...….……....92

5.3.4 Binary complexes of VN and IGFBP-5,

IGFBP-5 mutants or IGF-I do not stimulate

MCF-7 cell migration

...….……....92

5.3.5 Migration of MCF-7 cells in response to IGF-I

or des(1-3)IGF-I pre-bound to VN in the

presence of intact IGFBP-5 or HBD mutant

IGFBP-5

...….……....95

5.3.6 Effect of heparin on migration of MCF-7 cells

in response to IGF-I pre-bound to VN in the

presence of IGFBP-5

...….……....96

5.3.7 Effects of IGFs and MBF on MCF-7 cell

migration

...….……....98

5.3.8 MCF-7 cell migration responses to MBF in

the presence of IGFBPs with or without

heparin

...….….….101

5.3.9 Effect of VN on MCF-7 cell migration in

response to IGFBP-4

...….….….101

x

5.3.10 Comparison of MCF-7 cell migration

responses to VN alone or in the presence of

IGF-I or des(1-3)IGF-I with IGFBP-4

...….…......104

5.3.11 Effect of heparin on MCF-7 cell migration in

response to IGFBP-4 trimeric complexes

...….…......104

5.3.12 MCF-7 cell migration responses to IGFBP-3

preparations with or without VN

...….…......107

5.3.13 Comparison of MCF-7 cell migration

responses to VN alone or in the presence of

IGF-I or des(1-3)IGF-I with IGFBP-3 or HBD

mutant IGFBP-3

...….…......107

5.3.14 Effect of heparin on MCF-7 cell migration in

response to IGFBP-3

...….…......110

5.3.15 Scanning electron micrographs of MCF-7

cells in Transwell migration assay in response

to IGF-I:IGFBP-5:VN complex combinations

...….…......112

5.4 Discussion ...….…......120

CHAPTER 6: GENERAL DISCUSSION ...….…......127

CHAPTER 7: APPENDIX

7.1 Buffer Recipes ...….…......136

CHAPTER 8: REFERENCES ...….…......139

xi

LIST OF FIGURES

1.1 Comparison of human IGF sequences ...….…........4

1.2 Schematic representation of IGF and insulin receptors ...….…........6

1.3 IGF-1R and the associated signalling cascades ...….…........8

1.4 Secondary structure of the CIMPR ...….…......10

1.5 Location of key sites within the primary sequences of the

IGFBPS

...….…......13

1.6 Schematic diagram and NMR images identifying key

regions within VN

...….…......24

1.7 Binding domains of vitronectin towards various ligands ...….…......25

1.8 Major biological functions in which vitronectin has been

implicated

...….…......28

1.9 The complexity of the interactions of the IGFs with VN

and their mediation via the IGFBPs

...….…......31

3.1 Effect of IGFs and IGF-II analogues on binding of [125I]-

IGF-II to VN

...….…......46

3.2 Effect of IGFBPs on the binding of [125I]-IGF-II to VN ...….…......48

3.3 Effect of IGFBPs on the binding of [125I]-IGF-I to VN ...….…......49

3.4 Ability of IGF peptides to compete for binding of [125I]-

IGF-I to VN in the presence of IGFBP-3 or IGFBP-5

...….…......50

3.5 Proposed model of solid-plate binding assay for IGF-

II:VN

...….…......53

3.6 Proposed model of solid-plate binding assays for IGF-

I:IGFBP:VN

...….…......53

3.7 Proposed model to explain IGF:IGFBP:VN bell-shaped

binding curves

...….…......54

3.8 Schematic of binding model ...….…......55

4.1 Comparison of glycosylated and non-glycosylated IGFBP-

3 on [125I]-IGF-I binding to VN

...….…......63

4.2 Comparison of glycosylated and non-glycosylated [125I]-

IGFBP-3 binding to VN

...….…......64

xii

4.3 Silver stain of IGFBP-3 from various sources run on a 4%

stacking/12% resolving SDS-PAGE

...….…......65

4.4 Comparison of the ability of IGFBP-3 from various

sources to bind to [125I]-IGF-I

...….…......67

4.5 Comparison of the effects of glycosylated and non-

glycosylated IGFBP-6 on A) [125I]-IGF-I or B) [125I]-IGF-

II binding to VN

...….…......69

4.6 Binding of glycosylated and non-glycosylated IGFBP-6 to

[125I]-IGF-II in the absence of VN

...….…......70

4.7 Effect of the HBD in IGFBP-3 on IGF-I binding to VN ...….…......72

4.8 Binding of [125I]-IGF-I to wild-type IGFBP-3 compared to

HBD mutants of IGFBP-3

...….…......73

4.9 Comparison of wild-type IGFBP-5 with IGFBP-5 HBD

mutants in their ability to mediate [125I]-IGF-I binding to

VN.

...….…......75

4.10 Binding of [125I]-IGF-I to wild-type IGFBP-5 and HBD

mutants of IGFBP-5

...….…......76

4.11 Binding of [125I]-MBF to VN using the PEG precipitation

assay

...….…......78

4.12 Comparison of binding of either [125I]-IGF-I or [125I]-MBF

(an IGF-I analogue containing a HBD) to IGFBP-3 and -5

...….…......79

5.1 Effects of IGFs and IGFBPs stimulating de novo protein

synthesis in CHO-K1 cells measured by [4,5-3H]-leucine

incorporation

...….…......89

5.2 Dose response curve of increasing amounts of IGF-I in the

presence of VN alone or in combination with IGFBP-5

...….…......91

5.3 Dose response curve of increasing concentrations of

IGFBP-5 in the presence of VN alone or in combination

with IGF-I

...….…......93

5.4 Effect of IGFBP-5 and IGF-I on MCF-7 cell migration

with or without VN

...….…......94

xiii

5.5 Migration of MCF-7 cells in response to IGF-I or des(1-

3)IGF-I pre-bound to VN in the presence of intact IGFBP-

5 or HBD mutant IGFBP-5

...….…......97

5.6 Effect of heparin on IGFBP-5 trimeric complex-stimulated

migration

...….…......99

5.7 Effects of IGFs and an IGF analogue, MBF, on migration

of MCF-7 cells

...….…....100

5.8 MCF-7 cell migration responses to MBF in the presence of

IGFBPs with or without heparin

...….…....102

5.9 Effect of VN on MCF-7 cell migration in response to

IGFBP-4 complexes

...….…....103

5.10 Comparison of MCF-7 migration responses to VN alone

or in the presence of IGF-I or des(1-3)IGF-I

...….…....105

5.11 Effect of heparin on migration of MCF-7 cells in response

to IGFBP-4 trimeric complexes

...….…....106

5.12 MCF-7 cell migration in response to IGFBP-3 with or

without VN

...….…....108

5.13 Comparison of MCF-7 cell migration responses to VN

alone or in the presence of IGF-I or des(1-3)IGF-I with

IGFBP-3 or mutant IGFBP-3

...….…....109

5.14 Effect of heparin on MCF-7 cell migration in response to

wild-type IGFBP-3 and HBD mutant IGFBP-3

...….…....111

5.15 Scanning electron micrographs of MCF-7 cell migration in

response to IGF-I:IGFBP-5:VN complex combinations

...….114-119

5.16 Proposed model of IGF-I:IGFBP-4:VN complex, with or

without heparin

...….…....123

6.1 Proposed model of IGF-I binding to VN via IGFBP-3

or -5

...….…....132

6.2 Proposed model of IGF-I binding to VN via IGFBP-5 ...….…....132

6.3 Synthetic peptides ...….…....133

6.4 Proposed model of IGF-II binding to VN ...….…....133

xiv

LIST OF TABLES

1.1 Summary of key IGFBP characteristics ...….…......14

1.2 Post-translational modifications of IGFBPs ...….…......19

4.1 Summary of key characteristics of each IGFBP-3 ...….…......60

5.1 Comparison of migration results: IGF-I:IGFBP-5:VN

with or without heparin

...….…......99

xv

LIST OF ABBREVIATIONS

˚C Degrees Celsius

β-2ME Beta-2 Mercaptoethanol

γ Gamma

µCi Micro Curie

µg / mL Micrograms per millilitre

µL Microlitre

µm Micrometre

µM Micromoles

APS Ammonium persulphate

BP Insulin-like growth factor binding protein (figures only)

BSA Bovine serum albumin

BV Baculovirus

CO2 Carbon dioxide

CPM Counts per minute

DMEM Dulbecco’s modified Eagles medium

DMEM/F-12 Dulbecco’s modified Eagles medium / Ham’s F-12

ECM Extracellular matrix

EDTA Ethylenediamine tetraacetic acid

FBS Foetal bovine serum

g Gram

GAG Glycosaminoglycan

HBB HEPES binding buffer

HBD Heparin-binding domain

HBSS Hanks’ balanced salt solution

Hep Heparin (figures only)

HEPES 4-2(2-hydroxyethyl)-1-piperazineethanesulfonic acid

Hr Hours 125I Iodine-125

I Insulin-like growth factor-I (figures only)

IGF-I Insulin-like growth factor-I

IGF-II Insulin-like growth factor-II

IGFBP Insulin-like growth factor-binding protein

xvi

IGFBP-rP IGFBP-related protein

II Insulin-like growth factor-II (figures only)

MBF Matrix binding factor

mg/mL Milligrams per millilitre

min Minutes

mL Millilitre

MW Molecular weight

ng/µL Nanograms per microlitre

nm Nanometres

NSB Non-specific binding

PAI-1 Plasminogen activator inhibitor-1

PBS Phosphate buffered saline

PVDF Polyvinylidene difluoride

RGD Arginine-Glycine-Aspartate

RT Room temperature

SDS Sodium dodecyl sulphate

SDS-PAGE SDS-Polyacrylamide electrophoresis

sec Seconds

SEM Standard error of the mean

SFM Serum-free medium

SMC Smooth muscle cell

SPBA Solid-plate binding assay

TCA Trichloroacetic acid

TEMED N,N,N',N'-tetramethylethylenediamine

Tris Tris[hydroxymethyl] amino methane

Triton-X100 Iso-octylphenoxypolyethoxyethanol

Tween 20 Polyethylene 20-sorbitan monolaurate

VN Vitronectin

vSMC Vascular SMC

v/v Volume per volume

w/v Weight per volume

xvii

LIST OF PUBLICATIONS AND PRESENTATIONS

1. Kricker J.A., Towne C.L., Firth S.M., Herington A.C., Upton Z. (2003)

Structural and Functional Evidence for the Interaction of Insulin-like Growth

Factors (IGFs) and IGF Binding Proteins with Vitronectin. Endocrinology,

144(7): 2807-15.

Erratum: (2004) Endocrinology, 145(1): 193.

2. Upton Z, Kricker J.A. (2002) International Patent Application

WO0224219A1. Published 28 March 2002.

3. Kricker J.A., Hyde C.E., Herington A.C., Upton Z. (2001) Investigations into

the structural and functional relationships of the insulin-like growth factors

with vitronectin. Australian Society of Medical Research Student

Conference, Wesley Hospital, Brisbane AUSTRALIA (Poster presentation)

4. Kricker J.A., Hyde C.E., Herington A.C., Upton Z. (2001) Structural and

functional investigations into the interactions of insulin-like growth factors

(IGFs) with vitronectin. Matrix Biology Society of Australia and New

Zealand Annual Scientific Meeting, Canberra, AUSTRALIA (Oral

presentation)

5. Kricker J.A., Noble A.M., Herington A.C., Upton Z. (2002) Structural and

functional investigations into the interactions of insulin-like growth factors

(IGFs) with vitronectin. ASMR Student Conference, Wesley Hospital,

Brisbane AUSTRALIA (Poster presentation)

6. Kricker J.A., Noble A.M., Hyde C.E., Firth S.M., Herington A.C., Upton Z.

(2002) Structural and functional evidence for the interaction of insulin-like

growth factors (IGFs) and IGF binding proteins with vitronectin. BioScience

Frontiers Conference, QUT Brisbane, AUSTRALIA (Poster presentation)

7. Kricker J.A., Noble A.M., Firth S.M., Herington A.C., Upton Z. (2002)

Structural and functional evidence for the interaction of insulin-like growth

xviii

factors (IGFs) with IGF-binding proteins and vitronectin. First Joint

Symposium GH-IGF 2002. Boston, MA, USA (Plenary poster presentation)



factors (IGFs) and IGF binding proteins with vitronectin. ASMR Student

Conference, Wesley Hospital, Brisbane AUSTRALIA (Oral presentation)



factors (IGFs) and IGF binding proteins with vitronectin. MBSANZ

Conference, Acheron Valley, Victoria, AUSTRALIA (Oral presentation)

10. Kricker J.A., Clemmons D.R., Herington A.C., Upton Z. (2004) A trimeric

complex containing IGF-I, IGFBPs and vitronectin: examination of the

structural motifs. Second Joint Symposium GH-IGF 2004. Cairns,

Queensland, AUSTRALIA (Poster presentation)

1

CHAPTER 1: LITERATURE REVIEW

2

1.1 INTRODUCTION The insulin-like growth factor (IGF) system contributes to cellular proliferation and

differentiation as well as exerting insulin-like metabolic effects. Although the IGFs

are well studied, appreciation of their full effects and the exact nature of their

binding interactions with a number of proteins remain incomplete. Recent evidence

suggests that various extracellular matrix (ECM) proteins modulate key actions of

the IGFs, which result in a variety of cellular functions. This review focuses on

connections between the IGF system and its interactions with ECM components

such as vitronectin (VN) and its associated receptors. While addressing the structure

and functions of the IGFs, associations of the IGFs with their modulators, such as

the IGFBPs and the ECM protein VN are also considered.

1.2 INSULIN-LIKE GROWTH FACTORS The insulin-like growth factors (IGFs) are small hormone-like polypeptides that act

to regulate various physiological processes such as cell proliferation, differentiation

and other metabolic activities. The IGF family consists of two members, IGF-I and

IGF-II, and as their name indicates, they share structural homology with proinsulin

(for review, see Cohick and Clemmons, 1993a; Leventhal and Feldman, 1997). The

majority of IGF is produced in the liver whereby IGFs can enter the circulation to

execute endocrine functions, while IGFs produced locally in many tissues, including

lung, kidney and bone, provide additional paracrine and autocrine activity.

Circulating serum levels of IGFs account for 75-100% of total IGF found in the

body. Moreover, 75-80% of the circulating IGFs are bound in a 150-200 kDa

ternary complex, less than 1% exist in the free state and 20-25% are associated in a

binary complex with an IGF-binding protein (IGFBP) (Rajaram et al., 1997). IGF-II

is found in higher concentrations than IGF-I in foetal serum, whereas both are

present at high levels in adult mammalian serum with the exception of rat serum

where IGF-II levels decline postnatally (Bennett et al., 1983).

3

1.2.1 Structure of the IGFs The IGF proteins result from transcription of several mRNA species of individual

genes on the short and long chromosome arms. IGF-I is encoded by a single gene

located on 12q22→23 that contains 5 exons, separated by 4 introns. Alternative

mRNA splicing results in the formation of two unique isoforms known as IGF-IA

and IGF-IB of 153 and 195 amino acids respectively. Similar diversity exists with

IGF-II. The gene encoding IGF-II, located on 11p15.5, contains eight exons that are

transcribed into several mRNA species with three predominant species of 5.3, 6.0

and 4.9kb (for review, see Daughaday and Rotwein, 1989). Differential splicing

also is responsible for the substitution of a tetrapeptide of Arg-Leu-Pro-Gly for Ser

29 and a tripeptide Cys-Gly-Asp for Ser 33 (Zumstein et al., 1985; Hampton et al.,

1989).

The IGF polypeptides are organised into five domains, A to E. Domains B, C and A

are similar to those in insulin, however the IGFs also have an extended C domain as

well as domains D and E. The E domain of IGF-II is cleaved post-translationally to

yield the mature protein formed by the remaining domains B, C, A and D, similar to

IGF-I. The two IGFs are similar in structure sharing 62% homology in amino acid

sequence with each other and 40% homology with proinsulin.

IGF-I is a 7649 Da basic peptide while IGF-II is 7469 Da and is slightly acidic

(Daughaday and Rotwein, 1989; Leventhal and Feldman, 1997) (See Figure 1.1).

IGF-I is synthesised as a pre-protein and the mature peptide circulates as a 70 amino

acid peptide (Laron, 2001). IGF-II is also synthesised as a pre-protein, with the 88-

amino acid residue E domain removed via proteolytic cleavage resulting in the 67

amino acid mature IGF-II protein. Isoforms of IGF-II have been identified which

have variable O-linked sugars but are similar in terms of receptor and ligand binding

interactions and biological activity (Valenzano et al., 1997).

4

Figure 1.1 Comparison of human IGF sequences. Comparison of the sequences for human IGF-I and IGF-II allows for recognition of homologous amino acids and the number of basic and acidic residues. Within the sequence of IGF-II lies a short heparin-binding domain consensus (XBBXBX) highlighted in the box. It is also clear from this sequence alignment that the Cys residues are conserved between the IGFs.

1.2.2 Biological Activities of the IGFs IGFs are best known for their potent mitogenic capabilities as well as their abilities

to promote DNA synthesis and cell cycle progression in numerous cell types

(Humbel, 1990), and involvement in metabolic action (Holt et al., 2003). IGF-I, in

particular, is known to synergise with competence factors such as platelet-derived

growth factor (PDGF) and fibroblast growth factor (FGF) enabling them to act as

progression factors in the cell cycle (Cohick et al., 1993). In addition, IGFs have

been demonstrated to promote cell motility by enhancing actin polymerisation at the

leading edge of the cell (Leventhal and Feldman, 1997). Moreover, IGF-I together

with epidermal growth factor (EGF) has been shown to enhance keratinocyte

migration (Krane et al., 1991; Ando and Jensen, 1993). These actions of the IGFs

often involve the integrins, a family of cell surface proteins, as well as several

distinct signalling pathways. For example, activation of the phosphatidylinositol 3

(PI3)-kinase and mitogen activated protein (MAP)-kinase pathways are necessary

for both IGF-II-stimulated trophoblast migration (McKinnon et al., 2001) and IGF-I-

stimulated porcine vascular smooth muscle cell (vSMC) migration (Arai et al.,

1996a; Imai and Clemmons, 1999). As many tumours produce IGFs (Perks and

1 11 21 I GPETLCGAELVDALQFVCGDRGFYFNKPTGY 1 11 21 II AYRPSETLCGGELVDTLQFVCGDRGFYFSR 32 41 51 61 70 I GSSSRRAPQTGIVDECCFRSCDLRRLEMYCAPLKPAKSA 31 41 51 61 67 II PASRVSRRSRGIVEECCFRSCDLALLETYCATPAKSE K R Basic residues (positively charged) E D Acidic residues (negatively charged) C Cysteine residues

5

Holly, 2003; Fottner et al., 2004), and IGFs induce cell motility, it has been

suggested that IGFs may have a critical role in the metastatic nature of some

tumours.

IGF-I is also commonly associated with cell survival and is generally regarded as a

key anti-apoptotic agent in a number of cell lines (Foulstone et al., 2001; Jamali et

al., 2003). More specifically, and surprisingly, in the presence of tumour necrosis

factor-alpha (TNFα) IGFs were shown to enhance induction of apoptosis and to

inhibit skeletal muscle myoblast differentiation (Foulstone et al., 2001). This

contrasts with other data whereby IGFs are considered to be anti-apoptotic (Niesler

et al., 2000; Galvan et al., 2003), hence raising the question as to why TNFα

together with IGFs has this effect, and what other cytokines may act similarly.

There is increasing evidence to suggest that cell survival is initiated by IGF binding

to the type-1 IGF receptor (IGF-1R) and that the PI3-kinase pathway is central

(Buckley et al., 2002; Galvan et al., 2003). Indeed, IGF-I related cell apoptosis may

not be due to IGF-I itself, but rather, IGF-I is prevented from binding to its receptor

to activate cell survival signalling, such as in the case with IGFBP-3 (Schmid et al.,

2001).

In contrast to the study by Foulstone et al. (2001), IGFs have often been shown to

positively affect cell differentiation. For example, IGF-I induces differentiation of

mesenchymal cells into chondrocytes (Oh and Chun, 2003) and of mouse tongue

myoblasts (Yamane et al., 2002). More recently, IGF-I-induced myoblast

differentiation was demonstrated using i28 mouse myogenic cells by Halevy and

Cantley (2004). Furthermore, this was due to a fine balance between activation of

the PI3-kinase and MAP-kinase signalling pathways. IGF-II has also been shown to

enhance myoblast differentiation (Stewart et al., 1996), and to increase glucose

uptake in differentiating cells (Bhaumick and Bala, 1991). The IGFs are also

commonly associated with metabolism and glucose uptake in skeletal muscle in both

normal physiology (Rotwein, 2003) and in diabetic patients (Frystyk, 2004).

6

1.3 IGF MODULATION The IGFs are modulated by a number of regulators including two receptors known to

mediate the biological actions of IGFs: the IGF-1R and the type-2 IGF receptor, also

known as the cation-independent mannose-6 phosphate receptor (CIMPR). The

IGF-1R binds IGF-I (Kd = 0.16 nM) and IGF-II (Kd = 0.7 nM) with similar affinity

and insulin with a lower affinity (Kd = 160 nM) (Sakano et al., 1991; Schumacher et

al., 1991) as determined by cell-receptor binding assays. Comparably, BIAcore

analysis reveals that the IGF-1R binds IGF-I and IGF-II with relatively similar

affinity (Kd = 4.45 nM and 23 nM) (Forbes et al., 2002) and insulin with 10-fold

difference to cell-based assays (Kd = 16 nM). Only IGF-II, however, has high

binding affinity for the CIMPR, with much reduced, if any, binding observed with

IGF-I and insulin (Kiess et al., 1994). IGF-II has also been shown to bind a variant

of the insulin receptor (IR) lacking exon 11 (IR-A), which also has a 2-fold higher

affinity for insulin. The IR isoform containing exon 11 (IR-B) can only bind to

insulin. Another complexity is the ability of hemireceptors receptors to form

hybrids (Soos et al., 1990): IGF-1R is able to dimerize with either isoform of the IR

(see Figure 1.2). Hybrid IGF-1R/IR-A receptors retain high affinity for IGF-I but

have a much reduced affinity for insulin (Soos et al., 1993), while IGF-I only has

high affinity for IGF-1R-IR-B (Pandini et al., 2002).

Adapted from LeRoith and Roberts (2003)

IGF-1R IR Hybrid R CIMPR

IGF-I IGF-IIInsulin Figure 1.2 Schematic representation of IGF and insulin receptors. IGF action is in part regulated by binding to a number of receptors. IGF-I interacts with its receptor, the IGF-1R and with hybrid receptors, both IGF-1R/IR-A and IGF-1R/IR-B. IGF-II can bind to its receptor the CIMPR, the IGF-1R and IR, while insulin can only bind its receptor, IR, and the hybrid receptor IGF-1R/IR-A.

7

1.3.1 Type-1 IGF Receptor (IGF-1R) The molecular organisation of the human IGF-1R is quite similar to that of the

insulin receptor. The gene, located on chromosome 15q25→26, spans more than

100 kilobases of genomic DNA and contains 21 exons (Abbott et al., 1992). The

IGF-1R is expressed as a single precursor molecule composed of 1367 amino acids

and a 30-residue signal peptide. It is subsequently cleaved at the proteolytic site

(Arg-Lys-Arg-Arg) located between residues 717 and 710 to produce the mature

heterotetrameric receptor comprised of two disulphide bond-linked α (706 residues)

and β subunits (626 residues) (Ullrich et al., 1986).

The mature, functional IGF-1R is a member of the tyrosine kinase receptor family

and is structurally related to the insulin receptor (Wood, 1995). The α-subunit,

which lies extracellularly and contains a cysteine rich domain, is important for high-

affinity IGF binding (LeRoith et al., 1995a; Ward et al., 2001). The β-subunit

exhibits the most homology with the insulin receptor and is associated with receptor

function. The essential tyrosine protein kinase region is located in the cytoplasmic β

domain (Blakesley et al., 1998) that also contains a transmembrane region.

Upon binding of IGF-I, the receptor undergoes induced autophosphorylation. This

centres on a cluster of cytoplasmic tyrosine residues, Tyr 1131, 1135 and 1136, and

results from activation of the intrinsic tyrosine kinase activity of the IGF-1R (Kato et

al., 1994). This in turn activates a cascade involving Ras, Grb2-mSos signalling

pathways (Lowenstein et al., 1992; Egan et al., 1993) (see Figure 1.3). Although the

evidence identifying the exact site where IGFs bind to the IGF-1R is inconclusive,

amino acid residues Leu24 and Leu27 in the IGFs are important for IGF-I and IGF-II

binding, respectively (Cascieri et al., 1988; Bayne et al., 1990; Roth et al., 1991).

There is evidence suggesting that the IGF-1R displays heterogeneity due to

structural variation, differential glycosylation and hybrid formations with the insulin

receptor subunits (LeRoith et al., 1995a). Depending on the nature of the receptor

diversity, a range of alterations in receptor function can arise. For example,

formation of hybrid receptors results in reduced insulin binding (Soos et al., 1993)

and reduced metabolic activity (Federici et al., 1997).

8

Adapted from Leventhal and Feldman (1997) and Rhodes (2000)

Figure 1.3 IGF-1R and the associated signalling cascades. Binding of IGF-I or IGF-II stimulates tyrosine kinase activity within the intracellular domains of the receptor, which autophosphorylates sites on the beta subunits. SH2 domains of IRS-1 bind to these tyrosine phosphates enabling phosphorylation of IRS-1. SH2 domains on the p85 regulatory subunit of PI3-K can bind to IRS-1, resulting in activation of the p110 catalytic subunit. PI3-K can bind directly to Rac to promote exchange of GTP to GDP. FAK and paxillin are also phosphorylated from IGF1-R activation. Shc and Grb2 are also phosphorylated leading to signal transduction through Ras, Raf-1, MEK, MAPK and p90, resulting in mitogenesis.

The ubiquitously expressed IGF-1R is implicated in a range of physiological

processes including a role in mediating IGF-I induced cell motility (Leventhal and

Feldman, 1997). The key role of this receptor in maintaining normal growth and

development is also evident with its implication in tumour-related and disease states.

An important aspect contributing to this is its regulation. Regulation of IGF-1R

occurs at the gene and mRNA expression levels as well as at the hormonal levels

and with each level there is a susceptibility to malfunction and hence, receptor-

related disease states (Krane et al., 1991; LeRoith et al., 1995b). Any change in the

YP

YP

IRS-1

SH

2S

H2 YP

p85 p110

SH

2

PI 3-Kinase

FAK rac

GTP GDP

paxillin

Cell adhesion Actin polymerisation

IGF-1R

YPYPYPYP

YPYPYPYP

Shc

YP

Grb2mSOS

Ras

Raf-1

MEK

MAPK

p90RSK Mitogenesis

YPYP

YPYP

IRS-1

SH

2S

H2 YPYP

p85 p110

SH

2

p110

SH

2

PI 3-Kinase

FAK rac

GTP GDP

paxillin

Cell adhesion Actin polymerisation

IGF-1R

YPYPYPYP

YPYPYPYPYPYPYPYP

YPYPYPYP

YPYPYPYPYPYPYPYP

ShcShc

YPYP

Grb2mSOS

RasRas

Raf-1Raf-1

MEKMEK

MAPKMAPK

p90RSKp90RSK Mitogenesis

α subunit

β subunit

9

receptor expression or structure can lead to altered IGF effects (Kim et al., 2001),

for example, alterations to key IGF-1R tyrosine residues involved in

phosphorylation have been shown to disrupt the actin cytoskeleton as well as inhibit

proliferation and anchorage-independent growth (Blakesley et al., 1998).

Conversely, manipulation of IGF-1R expression can be used to control disease states

such as epidermal proliferation in psoriasis (Wraight et al., 2000).

1.3.2 Type-2 IGF Receptor (CIMPR) In addition to binding to the IGF-1R, IGF-II can also bind to the CIMPR. The gene

for this receptor is localised to chromosome 6, region 6q25→27, and encodes a 2450

amino acid transmembrane receptor of 275 to 300 kDa (Laureys et al., 1988). The

CIMPR is a single polypeptide chain that forms a type I transmembrane receptor

comprised of four domains: an amino-terminal signal sequence; a large extracellular

domain; a transmembrane domain, and a short cytoplasmic domain.

The large extracellular domain consists of 15 homologous repeats, each

approximately 147 amino acids in length which exhibit ~20% identity and are

cysteine rich (Kornfeld, 1992). This domain contains the distinct binding sites for

IGF-II and mannose 6-phosphate (M 6-P)-containing ligands (Kiess et al., 1994;

Schmidt et al., 1995; Devi et al., 1998; Nykjaer et al., 1998; Grimme et al., 2000).

Repeats 3 and 9 are responsible for binding proteins containing mannose 6-

phosphate while repeat 11 is responsible for IGF-II binding (Schmidt et al., 1995).

Further evidence has indicated that repeat 13 also is involved in IGF-II binding and

it acts as an IGF-II affinity-enhancing domain (Devi et al., 1998; Linnell et al.,

2001) (see Figure 1.4). Repeat 11 is also responsible for mediating the

internalisation of IGF-II and its delivery to lysosomes for degradation. High affinity

binding of the IGF-II and M 6-P-containing ligands cannot occur simultaneously, at

least efficiently, either due to conformational changes within the receptor or steric

hindrance. The CIMPR also has a short cytoplasmic region that retains no tyrosine

kinase activity and a single transmembrane domain.

10

Figure 1.4 Secondary structure of the CIMPR. The M 6-P-containing ligand’s binding site is comprised of repeats 3 and 9 while the ‘core’ IGF-II site is at repeat 11. Upon internalisation of the ligands, IGF-II is degraded in lysosomes and phosphorylated M 6-P ligands are transported from the Golgi to lysosomes to promote activation of TGF-β.

The CIMPR is a multifunctional receptor found in the Golgi apparatus, the cell

membrane and the endosomes of most cell types (Nykjaer et al., 1998). The CIMPR

primarily functions as a lysosomal enzyme targeting protein with few published

details having been described on the effects of the binding of non-enzyme ligands.

To date, the recognised list of CIMPR functions include: - facilitating transport of

mannose 6-phosphate ligands from the Golgi network to lysosomal compartments;

facilitating endocytosis of extracellular mannose-6 phosphate-tagged lysosomal

enzymes; facilitating proteolytic activation of TGF-β, a negative growth factor

regulator of epithelial cells; and targeting IGF-II for lysosomal degradation, thereby

preventing IGF-1R activation (as reviewed by DaCosta et al., 2000).

In contrast to the IGF-1R, which mediates mitogenic and metabolic IGF actions, the

only clearly defined action of the CIMPR following binding of IGF-II at the cell

MM

M

IGF-IIM 6-P

3

11

9

protein

N

C

TGF-βP

P

P IGF-II

CIMPR

Activation of latent TGF-β

MMMM

MM

IGF-IIM 6-P

3

11

9

protein

N

C

TGF-βPP

PP

PPP IGF-IIIGF-IIIGF-II

CIMPR

Activation of latent TGF-β

11

surface, is the internalisation and delivery of IGF-II to lysosomes where it is

degraded (Grimme et al., 2000). The CIMPR, therefore, appears to be critical in

managing IGF-II levels. Loss of the receptor leads to abnormally high levels of

IGF-II and excessive stimulation of the type-1 IGF receptor. This has led to

speculation as to whether this receptor is a tumour suppressor candidate (Devi et al.,

1998; Oates et al., 1998; DaCosta et al., 2000). Moreover, down regulation and

mutations of the receptor are commonly reported in tumours (Blakesley et al., 1996;

Schnarr et al., 2000; Bostedt et al., 2001).

In addition to IGF-II and M 6-P binding to the CIMPR, several other proteins have

also been shown to bind to the receptor. For example, Nykjær et al. (1998) reported

the urokinase plasminogen activator receptor (uPAR) binds the CIMPR. In addition,

the CIMPR binds transforming growth factor beta (TGF-β) precursor (Kovacina et

al., 1989), thyroglobulin, proliferin (Lee and Nathans, 1988), retinoic acid and LIF

(leukaemia inhibitory factor) (Blanchard et al., 1998). Moreover, retinoic acid has

been shown to stimulate CIMPR-mediated internalisation of IGF-II and to increase

lysosomal enzyme sorting (Kang et al., 1997). It is interesting to note that the

CIMPR is shown to be specific to cell types and tissues of mainly embryonic or

tumour origin (Kiess et al., 1994). This could, therefore, suggest a possible role for

IGF-II in growth and development. However, whether the receptor has any

physiological role arising from binding IGF-II, other than removal of IGF-II from

the cell surface, remains to be clearly determined.

12

1.4 IGF-BINDING PROTEINS There are six members of the IGF-binding protein family, IGFBP-1 through to

IGFBP-6. These proteins have been shown to bind the IGFs with high affinity and

specificity.

1.4.1 Structure of the IGFBPs The IGFBPs range in size from 216 to 289 amino acids in length and they share a

high degree of homology (see Figure 1.5). Within the amino- and carboxy-terminals

of the IGFBPs there is approximately 40-60% sequence similarity (Rajaram et al.,

1997). These conserved regions harbour binding sites for the IGFs and a number of

other proteins such as components of the uPA system and cell-surface associated

proteins. Included in these regions is the heparin-binding domain, which resides in

the carboxy terminal of IGFBP-2, -3, -5 and –6. Also within the terminal domains

lie 18 cysteine residues (2 extra in IGFBP-4 and 2 less in IGFBP-6) that form

bridges that are thought to maintain the overall structure of the IGFBPs. The

variable central domain gives each IGFBP its uniqueness and is also responsible for

maintaining the proteins’ structure required for effective binding interactions.

1.4.2 Function of the IGFBPs The majority of circulating IGFs exist as part of a 150 kDa - 200 kDa complex

consisting of the IGF, an IGFBP (IGFBP-3 or –5) and an acid-labile subunit (ALS)

(~85kDa). It is reported that the IGFBPs have a number of diverse roles including

increasing the half-lives of the IGFs, storage of IGFs, transporting circulating serum

IGFs, modulating binding of IGFs to their receptors and localising IGFs to tissues,

as well as IGF-independent activities. Summarised in Table 1.1 are the main

characteristics of the IGFBPs including their relative preferences for IGF-I and –II.

From Binoux et al. (1999)

Figure 1.5 Location of key sites within the primary sequences of the IGFBPs

13

halla

This figure is not available online. Please consult the hardcopy thesis available from the QUT Library

14

Table 1.1 Summary of IGFBP key characteristics

(RGD = Arg-Gly-Asp; HBD = Heparin-Binding Domain; NLS = nuclear localisation sequence)

IGFBPs are variably expressed in serum, different tissues and fluids where they

confer specificity on the IGFs (Cohick and Clemmons, 1993b). Tissues expressing

IGFBPs include vascular endothelial cells (Bar et al., 1987), fibroblasts (Conover et

al., 1994), smooth muscle cells (McCusker et al., 1991; Cohick et al., 1993),

hepatocytes (Arany et al., 1994) and keratinocytes (Murashita et al., 1995). There

are many reports detailing the association of IGFBPs with cell surface proteins. In

particular, IGFBP-1 and –2 contain the binding sequence Arg-Gly-Asp (RGD) that

is recognised by integrins, a family of heterodimeric transmembrane glycoproteins,

which are primarily involved in maintenance of cell-cell or cell-ECM contact

(Lawler et al., 1988; Rickard et al., 1991; Ramachandrula et al., 1992). Independent

of its RGD sequence, IGFBP-2 as well as IGFBP-3, which lacks this sequence, have

been found to associate with cells, with binding mediated via various proteins and

proteoglycans present (Yamanaka et al., 1999; Mishra and Murphy, 2003).

Although the identity of many of these proteins remains unknown, proteoglycans on

the cell surface have been demonstrated to mediate IGFBP-2 binding in the rat brain

(Russo et al., 1997), while IGFBP-3 association with the surface of breast cancer

IGFBP IGF

affinity

Sequence /

Characteristic NLS

Ternary

Complex

with ALS

Potentiate

/ Inhibit

IGF

Action

Molecular

Weight

kDa

1 I = II RGD - - - / + 30

2 I < II RGD, HBD - - - 31 – 36

3 I = II HBD + + - / + 29, 37 – 48

4 I = II 2 extra Cys - - - 24 & 28

5 I < II HBD + + + 29 - 34

6 I < II HBD, 2 less

Cys - - - 23, 28 - 34

15

cells (Yamanaka et al., 1999; Mishra and Murphy, 2003) has implicated TGF-β (Oh

et al., 1995; Yamanaka et al., 1999; Mishra and Murphy, 2003).

IGFBPs are also known to mediate IGF-independent actions and this is believed to

relate to cell-surface sequences such as the RGD motif in IGFBP-1 and IGFBP-2

(Jones et al., 1993b; Yamanaka et al., 1999; Schutt et al., 2004). Employing non-

IGF-binding mutants or the combination of IGF-I knockout mice with exogenously

added IGFBP-4 to inhibit any remaining IGF-II activity, inreasing studies on cell

survival and apoptosis describe IGF-independent action of IGFBP-3 and -5, which

lack the RGD sequence. In such studies, IGFBP-3, has been widely demonstrated to

induce apoptosis in the absence of IGFs in three cancer cell lines (Hollowood et al.,

2002; Hong et al., 2002) and this involves activation of caspases (Kim et al., 2004).

IGFBP-5, in contrast, displays the ability to inhibit apoptosis during myogenesis

(Cobb et al., 2004) and to inhibit ceramide-induced apoptosis in breast cancer cells

(Perks et al., 2000; Perks et al., 2002). IGFBP-5 has also been shown to stimulate

bone formation independently of IGFs (Mohan et al., 1995; Andress, 2001).

IGFBP-3 is of particular importance in the IGF system as the majority of IGFs found

in the circulation are bound to it in the form of the 150 kDa – 200 kDa complex

containing ALS referred to earlier. Modulation of IGF activity is maintained via

competition for IGF binding by IGFBP-3 and the IGF-1R. Thus, it has been

hypothesised that IGFBP-3 inhibits IGF activity by making it unavailable for

interaction with its biological receptor. More recently, IGFBP-5 has also been

shown to bind to the ALS and IGF complex suggesting it too is a key to regulating

the action of IGFs (Twigg et al., 1998).

1.5 POST-TRANSLATIONAL MODIFICATION OF THE

IGFBPS Regulation of the IGF activity by IGFBPs has been demonstrated to be both

inhibitory and potentiating depending on the IGFBP involved, the status of

phosphorylation, glycosylation and binding protein proteolysis (see reviews by

16

Rajaram et al., 1997; Conover, 1999) and whether IGFBP is added prior to, or

together with, exogenous IGF.

1.5.1 Glycosylation Glycosylation sites are shown to be present in four IGFBPs, this feature also being

reflected in the varied relative molecular weights observed for the proteins (Table

1.1). In particular, IGFBP-3, -4, -5 and –6 have been shown to exist in glycosylated

forms in vivo (as reviewed by Conover, 1999). For example, IGFBP-3 is found as a

29 kDa protein as well as ~40 – 50 kDa forms in serum and in other body fluids

(Campbell et al., 1998) with the differences being contributed to N-linked sugars as

proven by studies using N-glycosidases.

In various studies examining binding of IGFBPs with IGFs, the effect of

glycosylation of the IGFBPs has received little attention. This is presumably due to

non-glycosylated IGFBPs being considered likely to be physiologically irrelevant

and to reports suggesting glycosylation does not play a role in IGFBP-3-mediated

IGF effects. Indeed, it has been reported that the carbohydrate moieties on IGFBP-3

do not influence binding and also are not required for cell association (Conover,

1991; Firth et al., 1998). Nevertheless, it is commonly accepted that glycosylated

IGFBPs exhibit greater resistance to proteolysis over their non-glycosylated counter-

parts (Neumann et al., 1998).

Interestingly, more recent data has been published outlining dramatic effects of

IGFBP glycosylation on binding affinity to the IGFs, at least in the case of IGFBP-6.

Thus, Marinaro et al. (2000) reported that non-glycosylated IGFBP-6 has a greater

than three-fold increased affinity for IGF-II than glycosylated IGFBP-6, and this in

turn decreases as much as 10-fold when IGFBP-6 binds to glycosaminoglycans

(GAGs). This contrasts with the initial study that identified IGFBP-6 as being O-

glycosylated in which it was reported that the affinity for IGF-II did not change

between glycosylated and non-glycosylated IGFBP-6 (Bach et al., 1992). This may,

however, be a reflection of the source of IGFBP-6, as the study by Bach et al. (1992)

enzymatically deglycosylated their IGFBP-6 that was purified from cerebrospinal

17

fluid, while the study by Marinaro et al. (2000) examined recombinant IGFBP-6

produced by E.coli and CHO (Chinese hamster ovary) cells. This source of IGFBP-

6 may also explain another study that challenged the concept that IGFBP-6

preferentially bound IGF-II over IGF-I, 20-100-fold that led to the suggestion that

IGFBP-6 is an inhibitor of IGF-II. Employing biosensor analysis to determine

affinity constants, Wong et al. (1999) observed similar low KAs of IGFBP-6 for both

IGF-I and IGF-II. Again the discrepancy in the affinity data may be attributable to

the source of the proteins.

1.5.2 Phosphorylation The IGFBPs all contain potential sites for phosphorylation by serine or threonine

kinases yet only IGFBP-1, -3 and –5 have known phospho-isoforms (Jones et al.,

1993a; Hoeck and Mukku, 1994; Coverley and Baxter, 1997). Phosphorylation is a

mechanism that helps regulate protein function and activity and is widely observed

and understood in signalling pathways. IGFBP phosphorylation, however, remains

less well understood, although the conformational change in the protein properties

associated with the introduced negative charge is appreciated (Conover, 1999;

Nicholas et al., 2002). While earlier reports indicated that phosphorylation of

IGFBP-3 did not affect its affinity for the IGF-I (Hoeck and Mukku, 1994; Coverley

and Baxter, 1997), a recent study demonstrated that IGF-I binding was enhanced

when IGFBP-3 was phosphorylated by proteins present in or near the cell membrane

(Mishra and Murphy, 2003).

IGFBPs display resistance to proteases by several mechanisms including

phosphorylation (Gibson et al., 1999b). Indeed, the effects of IGFBP-1 and -3 on

IGF actions are modulated by phosphorylation (Pattison et al., 1999; Sakai et al.,

2001). Thus, Coverley et al. (2000) demonstrated that phospho-IGFBP-3 was

resistant to plasmin and cysteine protease degradation in MCF-7 breast cancer cells.

In addition to inhibiting proteolysis, phosphorylation also hinders the binding of

IGFBPs to cells and to the ALS, but not to the IGFs themselves (Coverley and

Baxter, 1997). In fact, phosphorylation has been shown to increase the affinity of

IGFBP-1 for IGF-I approximately six-fold (Jones et al., 1991). This alteration of

18

IGFBP function appears to regulate IGF activity and prevent IGF-mediated effects

(Siddals et al., 2002). This has also been shown with non-phosphorylated forms of

IGFBP-1 which potentiate IGF-I-stimulated DNA synthesis (Elgin et al., 1987) and

IGF-I-stimulated amino acid uptake (Yu et al., 1998).

1.5.3 Proteolysis Proteolysis of the IGFBPs, which occurs in all biological fluids, decreases their

affinity for the IGFs and thus is another mechanism through which IGF activity can

be regulated (Bunn and Fowlkes, 2003; Sadowski et al., 2003). IGF receptors and

other IGF binding proteins, whose affinity for the IGF is higher than the fragmented

IGFBP, compete for IGF binding and this in turn results in the IGF action being

regulated (see review by Firth and Baxter, 2002). Proteolysis has also been

suggested to be an important mechanism in the regulation of IGF-independent

functions of IGFBPs (Maile et al., 1999; Singh et al., 2004). An increasing number

of studies demonstrate that IGFBP fragments are involved in normal cell

physiology. For example, Bernard et al. (2002) demonstrated that the amino-

terminal domain of IGFBP-3 is responsible for inducing apoptosis in the MCF-7

breast cancer cell line. Such fragmentation, whether the cleaved product is

functional or not, is observed in various physiological states such as pregnancy

(Hossenlopp et al., 1990; Davenport et al., 1992; Kubler et al., 1998) and chronic

illnesses (Davies et al., 1991; Remacle-Bonnet et al., 1997) where protease levels

are raised.

Protease levels in normal serum are quite low compared to other physiological fluids

such as interstitial, synovial and peritoneal fluids (as reviewed by Maile and Holly,

1999). The most common physiological state associated with raised protease levels

is that of pregnancy. Pregnancy-associated proteases, including distinct enzymes of

the serine protease family, are known to modify IGFBP-1, -2, -4, and in particular

IGFBP-3 (Davenport et al., 1992; Miell et al., 1997; Zheng et al., 1998; Gibson et

al., 1999b). Proteolysis of the IGFBPs is likely to occur so that IGFs are released

from the binary complex to allow increased mitogenic IGF activity to facilitate

foetal growth.

19

There are a number of proteases known to be responsible for non-specifically

generating IGFBP fragments. These include members of the urokinase plasminogen

activator (uPA) system such as plasmin (Lalou et al., 1995; Campbell and Andress,

1997b) and thrombin (Zheng et al., 1998), prostate specific antigen (PSA)

(Koistinen et al., 2002), matrix metalloproteinases (Fowlkes et al., 1995; Kubler et

al., 1998; Sadowski et al., 2003) and cathepsins (Claussen et al., 1997). While some

of these proteases can act on more than one IGFBP, other proteases have been

shown to be specific for a particular IGFBP (Chernausek et al., 1995). In particular,

pregnancy-associated plasma protein-A (PAPP-A) cleaves IGFBP-2, -4 and -5, with

IGFBP-2 and -4 requiring the presence of IGFs (Bayes-Genis et al., 2001; Byun et

al., 2001; Laursen et al., 2001; Monget et al., 2003). Common to each of these

IGFBP-cleaving proteases, however, is that the cleavage sites in the IGFBPs appear

to be restricted to the variable central domain (Chernausek et al., 1995; Claussen et

al., 1997; Zheng et al., 1998).

Table 1.2 Post-translational modifications of IGFBPs

Glycosylation Phosphorylation Proteolysis Cell / ECM *

IGFBP-1 - + + +

IGFBP-2 - - + +

IGFBP-3 N + + +

IGFBP-4 N - + -

IGFBP-5 O + + +

IGFBP-6 O - + -

*Cell or extracellular matrix association. Adapted from Conover (1999)

20

1.6 INHIBITION AND POTENTIATION OF IGF ACTION BY

IGFBPS

As mentioned above, post-translational modification of IGFBPs can influence IGF

action, either inhibiting or potentiating IGF action. In particular, glycosylation,

phosphorylation or proteolysis affects IGF affinity for various proteins and therefore

regulates potential IGF-mediated effects. Moreover, IGFBP affinities for the IGFs,

differential regulation of the IGFBPs by the IGFs, and association of the IGFBPs

with the cell-surface also contribute to IGF action.

Proteolysis of IGFBPs, whereby fragmentation of the binding protein releases the

bound IGF enabling IGF action, is often implicated in modulating IGF action (Jacot

and Clemmons, 1998). Several recent studies have revealed that in porcine vSMCs,

IGFBP-4 and IGFBP-5 have opposing roles in regulating IGF-I due to differential

regulation by the growth factor (Duan and Clemmons, 1998; Hsieh et al., 2003).

Hence, Duan and Clemmons (1998) suggested the decrease in the activity of an IGF-

I-regulated IGFBP-4 protease resulted in raised IGFBP-4 levels causing inhibition of

IGF-I-stimulated DNA synthesis. In addition, IGFBP-5 levels were decreased

resulting in potentiation of IGF-I effects. Another study in uterine myometrial cells

(Huynh, 2000) also exemplifies the dynamic regulation of IGF-I and IGFBP in

modulating IGF activity. In this study IGF-I enhanced IGFBP-4 proteolytic activity

decreasing IGFBP-4 levels, while IGFBP-3 levels were maintained via cell-

association, allowing initiation of IGF-I mitogenic activity (Huynh, 2000).

Cell surface association of IGFBP-3 and –5 has also been shown to both inhibit and

potentiate IGF actions. Cell bound IGFBP-3 is associated with enhanced IGF action

in endothelial cells (Booth et al., 1995) and fibroblasts (Conover, 1992). This

prompted speculation that IGFBP-3 might help deliver IGF to its receptor

(McCusker et al., 1990; Conover, 1991). However, a study by Forsten et al. (2001)

indicated that cell-associated IGFBP-3 bound with IGF-I inhibits IGF-I-induced

bovine mammary epithelial cell proliferation suggesting that IGFBP-3 does not

transfer IGF-I to its receptor. Instead, it has been proposed that membrane-bound

IGFBP-3 and -5 have lowered affinities for the IGFs thereby releasing IGFs to bind

to the IGF-1R (Mishra and Murphy, 2003).

21

Varying affinities of the IGFBPs for the IGFs also act as a mechanism which can

potentiate or inhibit IGF action. For example, IGFBP-6, which appears to have a 20

– 100-fold higher binding affinity for IGF-II over IGF-I, acts as a relatively specific

inhibitor of IGF-II (Bach et al., 1992). In addition, O-glycosylation of IGFBP-6

protects it from proteolysis and from associating with cells, which also contributes to

its inhibition of IGF-II actions (Marinaro et al., 2000). Similarly, IGFBP-1 is

reported to inhibit IGF actions in this way by interrupting the interaction between

IGF-I and its receptor (Yee et al., 1994). In contrast, non-phosphorylated forms of

IGFBP-1 potentiate IGF actions in certain cells, such as adipocytes and decidua due

to a reduced affinity for the IGFs (Koistinen et al., 1993; Coverley and Baxter, 1997;

Siddals et al., 2002).

1.7 NOVEL IGF BINDING PROTEINS In addition to the IGFBPs, IGFBP-related proteins have also been described, leading

to the proposal of an IGFBP superfamily as reviewed by Hwa et al. (1999). There

are a number of recognised members in the IGFBP-related protein family including:

connective tissue growth factor (CTGF) (Kim et al., 1997), L56 and the proteins

encoded by the genes for mac25, cyr61 and nov-oncogene. Similar to the IGFBPs

there is a cysteine-rich domain at the N-terminus contributing to 20% overall

identity (Collet and Candy, 1998). Nevertheless it appears that these proteins have

much lower binding affinities for the IGFs than the IGFBPs (Kim et al., 1997; Collet

and Candy, 1998; Rosenfeld et al., 1999). In addition to the provisional members of

the IGFBP-related protein family, another eight proteins displaying similar

characteristics have been identified including WISP-1, -2 and –3, tumour adhesion

factor (TAF) and prostacyclin stimulating factor (PSF) (Hwa et al., 1999).

The exact function of the IGFBP-related proteins is unclear. However, it has been

proposed that they may also have IGF-independent actions on growth regulation.

Wilson et al. (2002) indicated that IGFBP-related protein 1, or Mac25, was an

inhibitor of breast cancer cell proliferation. Functions commonly associated with

both CYR61 and CTGF are their angiogenic and adhesive capabilities (Chen et al.,

2000; Chen et al., 2001; Lin et al., 2003). Furthermore, Chen et al. (2001)

22

suggested they might also play a role in matrix remodelling due to their activation of

angiogenesis and wound healing. While this would suggest a beneficial role, it has

also been reported that up-regulation of CTGF can lead to expansion of the ECM, or

fibrosis, which has dire consequences in diabetic patients (Twigg et al., 2001).

Functions of the IGFBP-related proteins are believed to be mediated via integrin

binding (Chen et al., 2000; Grzeszkiewicz et al., 2001; Grzeszkiewicz et al., 2002).

For example, Grzeszkiewicz and co-workers demonstrated CYR61 mediates

fibroblast migration via αvβ5 (Grzeszkiewicz et al., 2001) and vSMC adhesion

through α6β1 (Grzeszkiewicz et al., 2002). Although IGFBP-related proteins are

structurally similar to a degree, their different functional roles and lack of high

affinity IGF binding suggest that an IGFBP superfamily is not an ideal classification.

1.8 VITRONECTIN (VN) Upton et al. (1999) reported the identification of another protein that bound IGFs,

namely VN. VN was found to bind IGF-II with similar affinity to that observed

with IGF-II binding to the IGF-1R (Upton et al., 1999). This is of particular interest

as it is structurally dissimilar to the previously documented IGFBPs and IGFBP-

related proteins. In addition, VN only binds IGF-II and not IGF-I or insulin.

Moreover, VN binds des(1-6)-IGF-II, an IGF-II analogue with reduced affinity for

‘classic’ IGFBPs, further indicating that the interaction of IGF-II with VN is

different to that of IGF-II with the IGFBPs.

1.8.1 Structure of VN

VN is a 75 kDa multifunctional glycoprotein with functions in the plasma and

extracellular matrix. There have been several names for VN in the past including

epibolin (Brown et al., 1991), S protein (Stanley, 1986; Jenne et al., 1989) and

nectinepsin (Blancher et al., 1996), each of which identifies VN as an adhesive

glycoprotein. Indeed the amino-terminal segment of VN was originally thought to

be mitogenic (Barnes et al., 1984) and is now known as the somatomedin B domain.

VN is primarily synthesised in the liver and is secreted into the plasma where it is

found in a monomeric form. A multimeric form is found in the extravascular matrix

23

and in platelets (Seiffert and Schleef, 1996; Francois et al., 1999; Gibson and

Peterson, 2001; Podor et al., 2001). The multimer is a clipped form composed of

two chains of 65 kDa and 10 kDa held together by a disulphide bridge. The term

multimeric VN has been used interchangeably with denatured VN, as has the term

monomeric with native or plasma VN. The complete amino acid sequence of VN

has been deduced from human liver cDNA where the open reading frame (ORF)

encodes a 19 amino acid signal peptide plus 459 amino acids that include three

glycosylation sites (Schvartz et al., 1999).

VN is known to interact with numerous proteins, hence enabling VN to regulate

diverse physiological processes including complement-mediated cell lysis, cell

adhesion, coagulation and fibrinolysis. Binding to VN occurs in three main domains

(Figure 1.6). Firstly, the amino terminal residues 1-47 comprise the somatomedin B

domain. Residues 47-130 include binding sites for thrombin-antithrombin (TAT)

complexes and collagen while the central domain is largely comprised of a region

sharing homology with hemopexin (Gibson et al., 1999a) which extends from amino

acid residues 131-323. A similar hemopexin-like domain is also found in matrix

metalloproteinases (Massova et al., 1998) including collagenase. The third major

domain is the carboxy terminus, comprised of amino acid residues 332-459.

Binding of various ligands to VN can alter the structural conformation, which in turn

affects the ability of VN to bind alternative proteins. An example is heparin,

whereby upon its binding, cryptic sites are revealed exposing other sites within VN

to which other ligands can bind (Morris et al., 1994; Seiffert and Smith, 1997) (see

Figure 1.6).

24

Figure 1.6 Schematic diagram and NMR images identifying key regions within VN. The positive amino acids within the HBD (heparin-binding domain) bind the negative amino acids within the acidic region, enabling VN to fold over on itself to give a closed conformation. Alternatively, upon binding certain ligands, VN opens up its conformation exposing both the acidic region and HBD. RGD (Arg-Gly-Asp integrin recognition sequence). Adapted from Xu et al. (2001) Many studies detail the numerous roles of VN, including its involvement with the

urokinase-type plasminogen activator (uPA) system (Ciambrone and McKeown-

Longo, 1990; Lawrence et al., 1994; Wei et al., 1994; Kanse et al., 1996; Gechtman

et al., 1997; Stahl and Mueller, 1997; Chavakis et al., 1998; Podor et al., 2000).

Furthermore, throughout the length of VN lie a number of binding sites linking VN

with components of the uPA system (such as plasminogen, PAI-1, uPAR) indicating

a critical role for VN in cell migration and invasion (Figure 1.7). The amino

terminal of VN, residues 1-44, contains the binding sites for PAI-1 and uPAR

(Seiffert and Loskutoff, 1991; Deng et al., 1996). Directly adjacent is the Arg-Gly-

Asp (RGD) integrin recognition sequence. The RGD sequence facilitates binding of

VN with the integrins αvβ3, αvβ5, αIIbβ3 and αvβ1 (Klemke et al., 1994; Yebra et

al., 1996; Horton, 1997; Huang et al., 1998; Schvartz et al., 1999), these interactions

having a significant role in cell migration as demonstrated in keratinocytes by

Huang et al. (1998).

Heparin-binding Somatomedin B 4-bladed β propeller

N C1-53 354-456 129-323

RGD ACIDIC REGION HBD

+ + + - - -

25

Figure 1.7 Binding domains of VN towards various ligands. Regions of VN which bind various ligands (PAI-1 = plasminogen activator inhibitor-1; uPAR = urokinase plasminogen activator receptor; TAT = thrombin anti-thrombin complex). Adapted from Schvartz et al. (1999)

Downstream from the RGD sequence at residues 53-64, is an expanse of acidic

amino acids where the binding site for thrombin-antithrombin (TAT) complexes

(Gechtman et al., 1997) and collagen lie (Izumi et al., 1988). Another site for

collagen binding is located adjacent to the heparin-binding site that is situated at the

C-terminus (Morris et al., 1994; Schvartz et al., 1999). Also at the carboxy end is a

putative binding site for PAI-1 (Kost et al., 1992; Gechtman et al., 1997) as well as

sites for plasminogen and GAGs. VN also harbours consensus sequences for

phosphorylation by various proteins at the carboxy terminal end (Seger et al., 2001;

Schvartz et al., 2002). The central hemopexin-like/carbohydrate domain is less

characterised and displays considerable variation between species (Nakashima et al.,

1992).

1.8.2 VN and its Integrin Receptors VN has been implicated in a number of processes due to its ability to bind a diverse

range of molecules, including those promoting cellular migration. Although the

mechanisms are not known, VN’s use of integrin receptors seems to be involved

(Leavesley et al., 1993; Jones et al., 1996; Huang et al., 1998). The integrins, a

family of cell-surface proteins, have a heterodimeric structure composed of non-

covalently associated α and β subunits. Combinations of the heterodimer complex

define their ligand specificity and hence, their functional interactions (Leavesley et

NH 2 COOH

1 45 47 132 332 361 348 370 459

PAI-1, uPAR Integrins

TAT, Collagen

Plasminogen Heparin

PAI-1

6 hemopexin repeats

RG

D

26

al., 1992; Horton, 1997). Their key functions include cell-to-cell and cell-to-ECM

adhesion, along with acting as receptors for transmitting signals to the cell interior

upon ligand binding (Jones et al., 1995; Jones and Walker, 1999). Common to VN,

IGFBP-1 also contains the RGD integrin-recognition sequence, which can mediate

binding to the cell surface resulting in rapid spreading and migration of cells as

shown in a number of cell lines including CHO and vSMCs (Jones et al., 1993b;

Gockerman et al., 1995). Of interest, VN has been reported to protect glioma cells

from drug-induced apoptosis and therefore, enhance cell survival, via a mechanism

involving VN-associated integrins (Uhm et al., 1999). Similarly, VN is reported to

reduce microvascular endothelial cell apoptosis via the αvβ3 or αvβ5 integrins (Isik

et al., 1998). In particular, it seems that the αv component is critical to VN-

associated apoptosis yet the β3 subunit appears to be critical for IGF-I-stimulated

cell migration (Maile et al., 2001).

Occupancy of the integrin αvβ3 by VN has been shown to enhance the ability of

IGF-I to activate the IGF-1R (Jones et al., 1996; Clemmons et al., 1999). Moreover,

blocking of the integrin with antagonists inhibits IGF-mediated effects such as

vSMC migration and DNA synthesis (Zheng and Clemmons, 1998; Clemmons et

al., 1999). Hence, it appears that significant cross talk between the VN and IGF

receptor occurs. Recent reports by Maile and Clemmons (2002) indicate that IGF-I

binding to its receptor stimulates redistribution of the integrin-associated protein

from the cell cytosol to the VN receptor, αvβ3, which in turn increases the affinity

of VN for αvβ3. A combination of IGF and VN binding to their respective

receptors in turn leads to a tyrosine phosphorylation cascade resulting in mediation

of IGF-I-stimulated responses (Maile and Clemmons, 2002; Maile et al., 2002).

Despite these recent findings, the exact mechanisms through which IGF and VN

interact to mediate IGF-stimulated effects still remain unclear.

1.8.3 Various Functions of VN VN is involved in a number of cellular processes including its promotion of cell

adhesion and spreading of anchorage-dependent cells, examples of which have been

demonstrated in vSMCs (Dufourcq et al., 2002; Stepanova et al., 2002) and

27

melanoma cells (Stahl and Mueller, 1997). Taliana and co-workers (2000)

substantiated that VN has a principal role in tissue repair and wound healing using

corneal myo-fibroblasts, as did Jang et al. (2000) using mouse VN knockout models.

They found dermal wound healing was delayed in mice lacking VN and changes

were evident in the fibrinolytic balance that is also associated with uPA and PAI-1.

In addition, studies in knockout mice have demonstrated that VN and PAI-1 act

together to help develop thrombotic activity in response to vascular injury (Eitzman

et al., 2000).

To examine the various functions of VN, a number of studies have used VN

receptor-blocking antibodies to prevent the VN-induced cellular responses. One

such example is reported by Dufourcq et al. (2002), whereby they demonstrated that

VN is up-regulated after arterial injury and is involved in mediating the formation of

neointima. Another study employing the same strategy documented that VN acts to

reduce microvascular endothelial cell apoptosis (Isik et al., 1998).

VN is also known to interact with a number of proteins including GAGs such as

heparin (Kost et al., 1992; Francois et al., 1999; Gibson et al., 1999a; Hocking et al.,

1999). The HBD enables VN to interact with GAGs causing allosteric effects to the

VN molecule making the complex more stable (Francois et al., 1999). Indeed, these

changes within VN are reported to expose additional binding sites on VN, which in

turn enable binding of other proteins (Morris et al., 1994; Seiffert, 1997). While the

GAGs function to anchor VN to the ECM, binding to additional sites on VN may be

linked to its role in cell adhesion, spreading and migration (see Figure 1.8). More

recent studies have documented the ability of VN to interfere with the functions of a

number of ECM proteins such as TGF-β (Schoppet et al., 2002).

28

Figure 1.8 Major biological functions in which VN has been implicated.

Adapted from Schvartz et al. (1999)

As seen in Figure 1.7, PAI-1 and uPAR share binding sites on VN (Deng et al.,

1996; Schvartz et al., 1999). The PAI-1 interaction with VN stabilises the inhibitor

in its active conformation and is implicated in competition with uPAR for binding to

VN. It is proposed that PAI-1, independent of its protease inhibitor role, governs

uPAR-mediated cell adhesion and detachment (Deng et al., 1996; Kanse et al.,

1996; Waltz et al., 1997) as uPAR and PAI-1 bind to the same site on VN. The

integrins also contribute to the dynamic balance between cell adhesion and

detachment governed by competition between PAI-1 and uPAR (Kjoller et al.,

1997).

1.9 INTERACTIONS BETWEEN THE IGF AXIS AND VN

VN and IGFs are often implicated in the same physiological processes such as

wound healing where cell adhesion, migration and establishment are required. As

revealed earlier, binding of both IGF-I and VN to their respective receptors results in

signal transduction cascades. Integrin signaling occurs as a result of VN binding to

its receptor, αvβ3, and the subsequent re-assembling and clustering of the integrins

to form focal adhesions (Ria et al., 2002; Chandhoke et al., 2004). It has been

reported that formation of VN-associated uPA:PAI-1 complexes can also be

incorporated into the focal adhesions which act to localise uPA to its receptor

(Ciambrone and McKeown-Longo, 1992). Indeed, this mechanism may also explain

VITRONECTIN

Extracellular Anchoring

(GAG, collagen)

Haemostasis (Thrombin, Factor Xa)

Immune Defense (Complement)

Cell Adhesion, Spreading & Migration

(Integrins, uPAR)

Cell Proliferation (Integrins, Growth factors)

Fibrinolysis (PAI-1, uPAR)

29

the synchronised signaling of VN and IGFs that results in IGF-mediated responses,

as IGF alone (Kabir-Salmani et al., 2003), or associated with an IGFBP, could

complex with VN and thus be directed to the IGF-1R clustered within or adjacent to

a focal adhesion.

The number of links reported between the uPA system and the IGF system are

increasing in number. In addition to IGF-I up-regulating the production of uPA

(Dunn et al., 2000), associations between PAI-1 and IGFBPs have been made. For

example, PAI-1 also binds IGFBP-5 (Nam et al., 1997), and in turn, IGFBP-5 can be

degraded by plasmin (Campbell and Andress, 1997b). Moreover it has been

reported that plasminogen binds the HBD of IGFBP-3, which consequently releases

IGFs to act on local tissues (Campbell et al., 1998). In addition, this same group of

investigators reported that IGFBP-3 binds fibrin and fibrinogen, directly and

indirectly while it is bound to plasminogen (Campbell et al., 1999). Of interest,

recent studies by Nam and co-workers have demonstrated that IGFBP-5 binds to

VN, and furthermore, this interaction can modulate IGF-I action (Nam et al., 2002).

More recently, IGFBP-2 has been shown to specifically interact with the VN

receptor, αvβ3, and this interaction on negatively regulating breast tumour cell

growth (Pereira et al., 2004). Clearly there is an association between the IGF system

and the uPA system and increasing studies suggest that VN may be the link.

30

1.10 CONCLUSIONS This review summarises the various components of the IGF system and its

involvement with the uPA system, integrins and ECM proteins. In particular, it

highlights the complexity of these interactions with regard to the IGFs and VN and

their modulation of their activity via the IGFBPs (see Figure 1.9). Clearly there are

a number of knowledge gaps that need to be addressed to fully appreciate the

mechanisms involved. Through identification of the major binding motifs for the

interactions between IGFs and/or IGFBPs and VN, we can translate these findings

into in vitro models to further understand the biological relevance of these

interactions. Clearly any contribution to understanding these complex mechanisms

may have clinical significance in diseases/tissues related to where these proteins

exist and function. For example, IGFs and VN have been identified at sites of injury

and could be used to improve wound healing.

31

d e-

P

I GF-

1R

YP

YP

YP

YP

YP

YP

YP

YP

SHP

S-1

P

SHP

-2d e-

P

Cel

l ad h

e sio

n

Mit o

g ene

s is

IRS

-1

SH2

SH2

YPI R

S-

1 SH2

SH2

p 110

SH2

PI 3

-Kin

ase

IGF-

I∗

∗∗

I AP

SHP

-2

βα

αvβ3

βα

Shc F A

K

Ras

I RS

-1 S

H2

SH2

p 110

SH2PI 3

-K

AKT

P

I AP∗

V it ro

nect

in∗

Vitr o

nect

in

Actin

po l

y me r

isa t

ion

IGF-

IIO

RIG

FBP

I GF-

I∗

IGF-

II

d e-

Pd e

-PP

I GF-

1R

YP

YP

YP

YP

YP

YP

YP

YP

YP

YP

YP

YP

YP

YP

YP

YP

YP

YP

YP

YP

YP

YP

YP

YP

YP

YP

YP

YP

YP

YP

YP

YP

SHP

S-1

P

SHP

-2d e-

P

SHP

S-1

P

SHP

S-1

PP

SHP

-2S

HP-2d e

-P

d e-

Pd e

-PP

Cel

l ad h

e sio

n

Mit o

g ene

s is

IRS

-1

SH2

SH2

YP

IRS

-1

SH2

SH2

YPYPI R

S-

1 SH2

SH2

p 110

SH2

PI 3

-Kin

ase

I RS

-1 S

H2

SH2

p 110

SH2

PI 3

-Kin

ase

I RS

-1 S

H2

SH2

p 110

SH2

PI 3

-Kin

ase

IGF-

I∗IG

F-I

IGF-

I∗

∗∗

I AP

I AP

SHP

-2S

HP-2

βα

βα

αvβ3

βα

βα

Shc

Shc F A

KF A

K

Ras

Ras

I RS

-1 S

H2

SH2

p 110

SH2PI 3

-K

AKT

I RS

-1 S

H2

SH2

p 110

SH2PI 3

-K I RS

-1 S

H2

SH2

p 110

SH2PI 3

-K

AKT

AKT

PP

I AP

I AP∗

V it ro

nect

in∗

Vitr o

nect

in

Actin

po l

y me r

isa t

ion

IGF-

IIO

RIG

FBP

IGFB

PI G

F-I

∗I GF-

II G

F-I

∗IG

F-II

Figu

re 1

.9 T

his

diag

ram

sum

mar

ises

the

com

plex

ity o

f the

inte

ract

ions

of t

he IG

Fs w

ith V

N a

nd th

eir m

edia

tion

via

the

IGFB

Ps.

Bin

ding

of

ligan

ds to

the

IGF-

1R a

nd α

vβ3

inte

grin

initi

ates

a si

gnal

tran

sduc

tion

invo

lvin

g ph

osph

oryl

atio

n (∗

) of a

num

ber o

f pro

tein

s tha

t res

ults

in IG

F-m

edia

ted

resp

onse

s. In

the

pres

ence

of I

GF-

I, au

toph

osph

oryl

atio

n of

IGF-

1R b

y SH

PS-1

lead

s to

the

rem

oval

of a

sup

pres

sor f

rom

αvβ

3 as

w

ell a

s th

e re

dist

ribut

ion

of I

nteg

rin A

ssoc

iatio

n Pr

otei

n (I

AP)

at α

vβ3.

Th

is o

ccur

s vi

a a

PI3

kina

se p

athw

ay a

nd c

ause

s an

incr

ease

in th

e af

finity

of V

N fo

r its

rece

ptor

. Th

e su

bseq

uent

pho

spho

ryla

tion

casc

ade

trigg

ers

SHP-

2 to

de-

phos

phor

ylat

e IG

F-1R

and

FA

K.

Blo

ckin

g th

e PI

3-k

path

way

inhi

bits

IAP

to c

lust

er a

t αvβ

3 an

d as

resu

lt in

hibi

ts IG

F-I s

timul

ated

mig

ratio

n.

32

1.11 OUTLINE OF PROJECT

1.10.1 Hypotheses The underlying hypotheses explored in my PhD studies were:

A) A link between IGFs and VN exists and may be mediated through IGFBPs; and

B) The heparin-binding regions are important for facilitating this interaction.

C) That these interactions are physiologically relevant.

1.10.2 Aims Thus, the aims of my PhD studies were:

1. To establish if IGFBPs can mediate interactions of IGF-I with VN.

2. To identify the effect of each IGFBP on the interaction of IGFs with VN.

3. To examine the effects of structural motifs, such as glycosylation and HBDs, on

the interaction of IGFs ± IGFBPs with VN.

4. To examine the functional effects of the IGF:IGFBP:VN interaction using a

cellular model.

33

CHAPTER 2: MATERIALS AND METHODS

34

2.1 MATERIALS General reagents were all highest laboratory grade and sourced from various

companies. Hydrochloric acid, formaldehyde and iso-amyl acetate were purchased

from APS Group (Seven Hills, NSW, Australia). Acetic acid, di-sodium hydrogen

orthophosphate, ethanol, glucose, glycerol, methanol, polyethylene glycol,

potassium chloride, sodium chloride, trichloroacetic acid and triton X-100 were

from BDH Laboratory Supplies (Poole, England). Other general reagents included:

magnesium sulphate and sodium hydroxide (Chem Supply, Gillman, SA, Australia),

and β-2 mercaptoethanol, bromophenol blue, BSA fraction V RIA grade,

chloramine-T, crystal violet, γ-globulin, glycerol, glycine, HBSS, heparin sodium

salt (~13 kDa), HEPES, Sigmacote, sodium azide, sodium metabisulphite, suberic

acid, Tris and trypan blue (Sigma-Aldrich Pty Ltd, Castle Hill, NSW, Australia).

Reagents used in electrophoretic experiments were from BIORAD Laboratories Pty

Ltd (Regents Park, NSW, Australia) and included acrylamide, ammonium

persulphate, bromophenol blue, Coomassie brilliant blue G-250 and TEMED.

Experiments involving tissue culture used plasticware from Nagle Nunc (Roskilde,

Denmark), Ready Safe scintillation fluid from Beckman (Gladesville, NSW,

Australia), while DMEM and DMEM/F-12 medium, and relevant supplements were

purchased from GIBCO Invitrogen (Mt Waverley, Vic, Australia). COONs basal

medium and FBS were from ThermoTrace (Noble Park, Vic, Australia).

2.2 PROTEINS AND CELL LINES IGF-I, IGF-II, des(1-3)IGF-I, des(1-6)IGF-II, [Leu27]IGF-II, IGFBP-1, -2, -4 and -6

were purchased from GroPep Ltd (Adelaide, SA, Australia). An IGF-I analogue

with a HBD engineered at the carboxy-terminal end (Patent: WO9954359A1)

known as Matrix Binding Factor (MBF) was provided by Prof. Leanna Read (TGR

BioSciences, Adelaide, SA, Australia). IGFBP-5, glycosylated IGFBP-3, HBD

mutant IGFBP-3 and mutant non-glycosylated IGFBP-3 were produced by

Adenoviral-mediated expression in 911 human embryonic retina cells and were

provided by Dr Sue Firth (Firth et al., 1999). E.coli produced His-tagged non-

glycosylated IGFBP-6 was kindly donated by Dr Leon Bach (Department of

Medicine, University of Melbourne, Vic, Australia) (Neumann et al., 1998).

35

IGFBP-5 HBD mutants were a kind gift from Prof. David Clemmons (Department

of Medicine, University of North Carolina, NC, USA) (Arai et al., 1996b; Parker et

al., 1996). Baculovirus-expressed wild-type IGFBP-3 and HBD mutant IGFBP-3

were a kind gift from Dr Susan Durham and Prof. David Powell (Baylor College of

Medicine, Houston, TX, USA). Human VN was purchased from Promega

Corporation (Madison, WI, USA). Human breast carcinoma cells (MCF-7; ATCC#

HTB-22) were obtained from the American Type Culture Collection (Manassas,

VA, USA), as were the Chinese Hamster Ovary cells (CHO-K1; ATCC# CCL-61).

2.3 RADIOLABELLING OF PROTEINS IGF-I, IGF-II and MBF were iodinated according to the chloramine-T method as

described by GroPep Ltd for IGFs (Protocol #3001) while IGFBP-3 (glycosylated

and non-glycosylated) was iodinated as per a protocol provided by Dr Janet Martin

(personal communication, Kolling Institute of Medical Research, Sydney, NSW,

Australia). The chloramine-T reactions were performed for 1 minute for the IGFs

(10 µg) and 15 seconds for IGFBP-3 (5 µg). Labelled IGFs were purified using size

exclusion on Sephadex G-50 (Amersham Pharmacia, Buckinghamshire, England),

with 50 mM sodium phosphate, 150 mM NaCl, 0.25% w/v BSA, pH 6.5 as the

elution buffer, while labelled IGFBP-3 was purified using heparin affinity

chromatography (Amersham Pharmacia) and 50 mM sodium phosphate, 0.1% w/v

BSA, pH 6.5 as the equilibration buffer. The IGFBP-3 was eluted using elution

buffers 1-3, consisting of the equilibration buffer containing 1) 0.4 M NaCl, 2) 0.75

M NaCl and 3) 1.0 M NaCl, all at pH 6.5 respectively. Confirmation that the [125I]-

IGFBP-3 was the correct molecular size and was not fragmented during the labelling

procedure was obtained by non-reducing SDS-polyacrylamide gel electrophoresis.

Ten thousand cpm of [125I]-IGFBP-3 fractions from peaks in the iodination elution

profile were run on a 4% stacking/10% separation Tris-glycine gel, dried and then

exposed to autoradiographic film (Agfa, Mortsel, Belgium) for 1-7 days. TCA

precipitation was used as a guide in selecting peak fractions for both IGFs and

IGFBPs: those that TCA precipitated above 90% were pooled into aliquots and

stored at -20°C. The average specific activity of labelled IGF was in the range of

50-80 µCi/g and 75 – 90 µCi/g for IGFBP-3.

36

2.4 SODIUM DODECYL SULPHATE-POLYACRYLAMIDE

ELECTROPHORESIS (SDS-PAGE)

Samples were run on 4% stacking/12% resolving or pre-cast 4 – 20% gradient SDS-

PAGE gels (Gradipore, Frenchs Forest, NSW, Australia) under reducing or non-

reducing conditions (Laemmli, 1970). Samples were mixed 1:4 with 5x sample

loading dye (0.05% bromophenol blue, 0.225M Tris-HCl, 5% SDS, 50% glycerol) ±

20% β-2-mercaptoethanol and heated at 95°C for 3 min. Gels were run in Tris-

glycine buffer at 80 V for 15 min, then 120 V for 70 – 90 min. Gels were stained

with 0.2% Coomassie Blue (5:4:1 – water: methanol: ethanol) or with SilverSNAP

Stain from Pierce (Rockford, IL, USA). Molecular weights of the proteins were

estimated using pre-stained molecular weight markers (BIORAD Laboratories Pty

Ltd).

2.5 RADIOLIGAND BLOTS Proteins on SDS-PAGE gels were transferred to Immobilon-PSQ PVDF transfer

membrane (Millipore, Billerica, MA, USA) using either the wet transfer or semi-dry

transfer method. For wet transfer, proteins on gels were transferred using carbonate

transfer buffer (10mM NaHCO3, 3mM Na2CO3, 20% methanol) (Dunn, 1986) while

the semi-dry transfer used buffers containing 15 mM Tris, 120 mM glycine, 20%

methanol (cathode) and 0.3 M Tris (anode) (Hossenlopp et al., 1986). The transfer

membrane was then washed for 0.5 hr in Triton wash buffer (0.01 M Tris, 0.14 M

NaCl, 1% Triton X-100, pH 7.4) on a shaker at RT. The wash buffer was then

replaced with 50 mL blocking buffer (0.01 M Tris, 0.14 M NaCl, 1% Tween 20, 1%

BSA, pH 7.4). After 4 hr, the blocking buffer was replaced with another 50 mL

containing 1 - 2 million cpm [125I]-IGF and left agitating on the membrane overnight

at RT. The blocking buffer was removed the following day and the membrane was

washed with wash buffer (0.01 M Tris, 0.14 M NaCl, 1% Tween 20, pH 7.4) for 4

hr, with the buffer changed every hour. The dried membrane was then exposed to x-

ray film for 2 – 7 days and developed using an Agfa automated film developer CP-

1000 (Mortsel, Belgium).

37

2.6 SOLID-PLATE BINDING ASSAY (SPBA) The SPBA was used to examine binding of ternary complexes, namely

IGF:VN:IGFBP, where VN was pre-bound to wells and the IGFs and IGFBPs were

added in solution and complexes remaining in the well were measured. The binding

times and doses were optimised and developed by others within the Upton

laboratory. The binding assays were performed in removable 4 HBX Immulon wells

(Dynex Technologies, Chantilly, VA, USA) coated with or without 300 ng of VN

(Promega Corporation) in 100 μL DMEM at 37°C, 5% CO2 for a mininum of 2 hr,

up to 4 hr. Wells were rinsed twice with HEPES Binding Buffer (HBB: 0.1 M

HEPES, 0.12 M NaCl, 5 mM KCl, 1.2 mM MgSO4, 8 mM glucose containing 0.5%

w/v BSA, pH 7.6) to prevent non-specific binding. [125I]-labelled protein (IGF-I,

IGF-II, glycosylated IGFBP-3 or non-glycosylated IGFBP-3) (10000 cpm) in HBB

+ 0.5% BSA in the absence or presence of increasing concentrations of unlabelled

IGFs (0.2 – 300 ng), IGF analogues (0.1 – 100 ng) and/or IGFBPs (0.05 – 100 ng)

were incubated overnight at 4°C in a final volume of 100 μL (Ballard et al., 1986;

Upton et al., 1999). Unbound radiolabelled protein was then removed by aspiration

and the wells were washed three times with HBSS. Radioactivity remaining bound

in each well was then determined using a γ-counter (Packard Cobra Auto-Gamma

counter, Global Medical Instrumentation Inc., Albertville, Minnesota, USA). Each

sample was measured in triplicate and the experiment repeated at least three times.

Student’s paired t-test was used to compare amounts of labelled protein in test wells

to the amount in the control wells (absence of VN and presence of tracer).

Differences were considered significant if the p value was less than 0.05.

2.7 PEG (Polyethylene Glycol) PRECIPITATION ASSAY The PEG precipitation assay was used to assess the formation of dimeric complexes

of proteins while in solution. In 5 mL tubes, 20,000 cpm of [125I]-IGF-I, [125I]-IGF-

II or [125I]-MBF were allowed to bind to IGFBP-3, -4, -5, -6 or VN at various

concentrations ranging from 0 – 100 ng in buffer (0.1 M HEPES, 44 mM NaH2PO4,

0.1% Triton X-100, 0.1% BSA, pH 6.5) up to a final volume of 250 µL and allowed

to incubate overnight at 4ºC. The next day, 250 µL 0.25% γ-globulin was added to

radiolabelled IGF/MBF:IGFBP solution, followed by 500 µL 25% PEG. After 15

38

min at 4ºC, the tubes were spun at 2,200 rpm for 15 min, then the supernatant

decanted. The remaining pellet was then washed with 1 mL of 6.25% PEG and

centrifuged at 4,200 rpm for 10 min. Again, the supernatant was decanted and the

remaining pellet was counted in the γ-counter. Each sample was tested in duplicate

and the experiment repeated at least twice. Results were expressed as % Bound over

Total (%B/T), where B0 was radiolabelled IGF/MBF in the absence of IGFBP.

2.8 CELL CULTURE Stocks of Chinese Hamster Ovary (CHO-K1) and human breast carcinoma cells

(MCF-7) were stored in liquid nitrogen. When required, an ampoule of cells was

thawed rapidly at 37ºC and then transferred to a T-25 culture flask in the appropriate

medium (DMEM/F-12 for MCF-7 cells and COONs Basal Medium for CHO-K1

cells) supplemented with 10% FBS, penicillin (50 units/mL), streptomycin (0.1

μg/mL) and gentamycin (1 μg/mL), as well as 4 mM L-glutamine for the CHO-K1

cells. The cells were then incubated at 37°C in a humidified environment with 5%

CO2.

Apart from the initial resuscitation of cells from liquid nitrogen, cells were always

grown in T-80 culture flasks. Cells were passaged at approximately 70 – 80%

confluency with a medium change every 2 – 3 days. Prior to an experiment, the

medium was changed 2 days before commencement. Routine passaging of cells

involved washing the cells for 3 min with 3 mL sterile PBS containing 0.5 mM

EDTA that had been pre-warmed to 37ºC, then adding approximately 1 mL of pre-

warmed 0.05% trypsin/0.53 mM EDTA to the flask of cells and agitating. After 1

minute, an appropriate volume of medium containing 10% FBS was added to the

flask. The MCF-7 cells were then split either 1:4 or 1:6 and the CHO-K1 cells were

split either 1:4 or 1:8 to new flasks and the volumes were adjusted to 10 – 12 mL

with medium. The cells were then returned to the incubator.

Storage of cell aliquots involved the same steps as passaging except the cells were

added to a 15 mL tube and resuspended in 10 mL of medium. This tube was then

centrifuged at 1000 rpm for 3 min. The medium was aspirated and replaced with

39

medium containing 15% FBS and 10% DMSO. One mL was transferred to a

cryovial and gradually frozen at approximately 1ºC/hr for 24 hr in an isopropanol-

containing cryovessel at -80ºC. The vials were then stored in liquid nitrogen.

2.9 PROTEIN SYNTHESIS ASSAY

CHO-K1 cells were grown to 70 – 80% confluence at 37°C in a humidified

environment with 5% CO2 in COONs Basal Medium supplemented with 10% FBS,

penicillin (50 units/mL), streptomycin (0.1 μg/mL), gentamycin (1 μg/mL) and L-

glutamine (4 mM). Protein synthesis was measured in CHO-K1 cells by the

incorporation of [4,5-3H]-leucine (Amersham Pharmacia) into de novo synthesized

protein. Briefly, 24-well plates were coated with or without 300 ng of VN in 300 μL

COONs medium at 37°C, 5% CO2 for 5 hr. Wells were washed three times with

HBB + 0.5% BSA to prevent non-specific binding, then either 30 or 300 ng of IGF-

II alone, or IGF-I in the presence or absence of 300 ng IGFBP-3 or IGFBP-5, were

added to duplicate wells in a final volume of 400 μL. IGFs and IGFBPs were

allowed to bind for 2 hr at 37°C prior to addition of the cells. CHO-K1 cells that

had been serum-starved for 4 hr were trypsinised and labelled with [4,5-3H]-leucine

were seeded into wells (50,000 cells / well) and incubated at 37°C, 5% CO2 for 2

days. After incubation, the cells were lysed by washing twice with HBSS, then two

15 minute washes with 5% TCA and a final wash with water as described in Francis

et al. (1986). Lysed cells were then treated with 0.5 M NaOH in 0.1% triton

solution (caustic triton) for 2 hr. One hundred μL subsamples of the caustic triton

solution were added to poly-Q scintillation vials containing 5 mL of Ready Safe

scintillant and counted in a β-scintillation counter (Beckman scintillation counter

LS500TA, Gladesville, NSW, Australia).

2.10 TRANSWELL CELL MIGRATION ASSAY MCF-7 cells were grown in DMEM-F12 medium supplemented with 10% FBS,

penicillin (50 units/mL), streptomycin (0.1 μg/mL) and gentamycin (1 μg/mL).

Cells were grown to 70 – 80% confluence at 37°C in a humidified environment with

40

5% CO2. Cell migration assays using Transwells (Corning COSTAR, New York,

NY, USA) were performed using cells from passages 24 to 34.

The lower chambers of 12 µm pore polycarbonate tissue culture treated Transwells

(12-well plate) were pre-coated with 1 µg VN in serum-free DMEM-F12 and

incubated at 37° C for 2 hr. Medium containing unbound VN was then removed and

the lower chambers washed twice with HBB containing 0.5% BSA. IGF-I, des(1-

3)IGF-I, IGF-II or an IGF-I analogue in DMEM-F12 + 0.05% BSA was added to the

lower chamber in the absence or presence of IGFBPs and/or heparin and allowed to

bind to the pre-coated VN overnight at 4°C. The medium containing unbound

growth factors was removed and the lower chambers washed twice with DMEM-

F12 + 0.05% BSA. MCF-7 cells that had been serum-starved for 4 hr were

trypsinised and seeded on to the microporous membrane in the upper chamber of the

Transwell inserts (200,000 cells/well) and incubated at 37° C in 5% CO2 for 5 hr.

Cells that had attached to the upper surface were removed with a cotton swab while

cells that had migrated to the lower surface of the porous membrane were then fixed

in 37% formaldehyde and stained with 0.01% crystal violet in PBS (pH 7.3). The

number of cells that had migrated to the lower side of the membrane was quantitated

by extracting the crystal violet stain in 10% acetic acid and determining the optical

density of these extracts at 595 nm (Leavesley et al., 1993). Treatments were

expressed as a percentage of the response observed with VN alone. Data were

pooled from duplicate samples and the entire experiment replicated twice, unless

otherwise stated.

2.11 SCANNING ELECTRON MICROSCOPY Samples used for scanning electron microscopy imaging included the Transwell

inserts without physical removal of cells from the upper surface that had not

migrated. In this situation the cells remaining on the inserts were fixed overnight in

3% glutaraldehyde (0.1 M sodium cacodylate pH 7.4, 0.4 M sucrose in cacodylate

buffer, 1.0 M calcium chloride in cacodylate buffer, 25% EM grade glutaraldehyde)

to allow improved cross-linking of the sample molecules and stabilisation of

structures.

41

The fixed samples were washed twice for 10 min each in cacodylate buffer, fixed in

1% osmium tetroxide for 30 min followed by two 5 minute washes in distilled water

as described by the QUT Analytical Electron Microscope Facility protocol (QUT,

Brisbane, Australia). The samples were then dehydrated with the following 10

minute washes, which were repeated twice: 50%, 70%, 90% and 100% ethanol then

100% amyl acetate. Once dehydrated, the samples were critically point dried and

then mounted on stubs with orientations exposing both the upper and lower

membrane surfaces. Samples were finally coated with gold using a sputter coater

and stored in a desiccator until viewed under the scanning electron microscope (FEI

Quanta200 Environmental Scanning Electron Microscope, FEI Company, Hillsboro,

Oregon, USA).

2.12 STATISTICAL ANALYSIS Values in the figures are means ± standard error of the mean (SEM). Differences

from the controls were analysed by Student’s paired t-test and differences among

groups were analysed using 1-way ANOVA and post-hoc Tukey’s analysis of

multiple means. Results expressed as percentage bound over total (%B/T) were

analysed using the Mann-Whitney U-test. Significance was accepted at p < 0.05.

42

CHAPTER 3: EFFECT OF IGF-BINDING PROTEINS ON

IGF BINDING TO VN

43

3.1 INTRODUCTION

The mitogenic effects of insulin-like growth factors (IGFs) are modulated by

members of the IGF-binding protein (IGFBP) family. These proteins have been

demonstrated to both inhibit and potentiate IGF action (Jones and Clemmons, 1995).

In addition to the six IGFBPs, another group of proteins termed IGFBP-related

proteins (IGFBP-rPs), have also been shown to bind the IGFs, although with a much

lower affinity. Upton et al. (1999) have reported the identification of another

endogenous protein complex consisting of VN and IGF-II. This is particularly

interesting as VN is structurally unrelated to both the IGFBPs and IGFBP-rPs.

VN, a multifunctional protein found in plasma and extracellular matrix, is a

component of the urokinase system. A number of proteins bind to VN, including

glycosaminoglycans (Kost et al., 1992; Francois et al., 1999), which bind via a

heparin-binding domain in VN, and integrins, which bind via an RGD sequence {D,

1993 #61;Boettiger, 2001 #610}. It is through the binding of various proteins to

these motifs, as well as other domains within VN, that diverse physiological

processes such as extracellular anchoring, cell spreading and migration are mediated

(de Boer et al., 1992; Wilkins-Port and McKeown-Longo, 1996; Deng et al., 2001).

IGF-II has been shown to bind directly to VN, whereas only minimal binding of

IGF-I to VN occurs (Upton et al., 1999). Nevertheless, VN appears to be critical for

a number of IGF-I-related effects including cellular DNA synthesis, type-1 IGF

receptor autophosphorylation and cell migration (Jones et al., 1995; Zheng and

Clemmons, 1998; Maile et al., 2001). More specifically, Clemmons and co-workers

have shown that VN binding to the integrin αvβ3 is critical for IGF-I stimulated

smooth muscle cell migration (Clemmons et al., 1999). In addition, inhibition of

IGFBP-5 binding to porcine smooth muscle cell (SMC) extracellular matrix also

reduces cellular responses to IGF-I (Rees and Clemmons, 1998). Furthermore, the

potentiating effects of IGFBPs on IGF action appear to require interaction with, as

yet unidentified, cell-surface associated proteins which may include VN (Forsten et

al., 2001). For example, IGFBP-5 has been demonstrated to facilitate binding of

IGF-I to bone cells independently of the IGF receptors (Mohan et al., 1995) and

44

IGFBP-3 has also been shown to potentiate IGF action markedly following binding

to the cell surface (Conover, 1992).

Given the importance of IGFBPs and VN for regulation of IGF action, it was

hypothesised that IGFBPs may mediate in direct binding of IGFs to VN, and in

particular, that IGFBPs may be necessary for functional interaction of IGF-I with

VN. This chapter provides substantial evidence to support this hypothesis based

upon studies examining binding of labelled-IGF-I and -IGF-II to VN in the absence

and presence of IGFBPs and enhanced cell migration in the presence of these

complexes.

The experiments in this chapter were designed to verify and further characterise that:

1) IGF-II can bind directly to VN, 2) demonstrate that IGFBPs affect IGF-I and –II

binding to VN; and 3) show that the involvement of IGFBPs in mediating IGF-I

binding to VN is specific.

3.2 EXPERIMENTAL PROCEDURES Brief descriptions of proteins and procedures specific to this chapter are provided;

Chapter 2 contains full details.

Materials

IGF-I, IGF-II, des(1-3)IGF-I, des(1-6)IGF-II, [Leu27]IGF-II, IGFBP-1, -2, -4 and -6

were purchased from GroPep Ltd while IGFBP-5 and glycosylated IGFBP-3 were

produced as described previously by Firth et al. (1999). Human VN was purchased

from Promega Corporation. General plasticware used in experiments containing

IGFBPs and VN was siliconised with Sigmacote (Sigma) and left to air-dry

overnight to prevent adhesion and loss of proteins onto the plastic.

Radiolabelling of Proteins

IGF-I and IGF-II were iodinated using the chloramine-T method (Section 2.3). The

reactions were performed for 1 minute for the IGFs (10 µg), then the labelled IGFs

were purified using size exclusion chromatography.

45

Solid-Plate Binding Assay

As described in Section 2.6, IGF:VN:IGFBP binding assays were performed in

removable 4 HBX Immulon wells coated with or without 300 ng of VN for 2 - 4 hr.

[125I]-labelled IGF-I and IGF-II (10,000 cpm) in the absence or presence of

increasing concentrations of unlabelled IGFs, IGF analogues and/or IGFBPs were

incubated overnight at 4°C. Remaining radioactivity bound in each well was then

determined using a γ-counter. Each sample was measured in triplicate and the

experiment repeated at least three times. Student’s paired t-test was used to compare

amounts of labelled protein in test wells to the amount in the control wells (absence

of VN and presence of tracer). Differences were significant if the p value was less

than 0.05.

3.3 RESULTS 3.3.1 Effects of IGFs and IGF-II analogues on binding of [125I]-IGF-II to VN

To demonstrate that the interaction of IGF-II with VN is specific, [125I]-IGF-II

binding assays were conducted in the presence of IGF-II analogues with varying

affinities for IGFBPs and/or IGF receptors. Des(1-6)IGF-II (which has a low

affinity for IGFBPs) (Francis et al., 1993) and [Leu27]IGF-II (which has low affinity

for the type-1 IGF receptor and IGFBP-3) (Roth et al., 1991) were equipotent with

native IGF-II in their ability to displace [125I]-IGF-II bound to VN (Figure 3.1).

Half-maximal competitive effects were observed at approximately 3 ng. IGF-I, on

the other hand, was much less effective at displacing [125I]-IGF-II, achieving

approximately a 20% reduction at 0.2 ng with no further reduction at higher doses

up to 100 ng. This indicates that IGF-II can bind to VN, while IGF-I binds VN

poorly as it competes ineffectively for [125I]-IGF-II binding to VN.

46

Figure 3.1 Effect of IGFs and IGF-II analogues on binding of [125I]-IGF-II to

VN. Ten thousand cpm of IGF-II tracer were added to wells containing pre-bound

VN (as described in Section 2.6) in the absence or presence of increasing amounts of

IGF-II, des(1-6)IGF-II, [Leu27]IGF-II and IGF-I. Approximately 4,500 cpm were

bound in the absence of added unlabelled IGF (or analogue) and has been designated

as 100%. Each data point is the mean ± SEM of triplicate wells from 3 experiments,

which have been corrected for non-specific binding (400 cpm).

0 0.1 1 10 100

120

100

80

60

40

20

0

IGF (ng)

% o

f Con

trol

IGF-II des (1-6) IGF-II Leu27 IGF-II IGF-I

47

3.3.2 Effect of IGFBPs on modulating binding of [125I]-IGF-II to VN

The ability of IGFBPs to modulate IGF-II binding to VN was then investigated

(Figure 3.2). All six IGFBPs were examined (Panels A to F) and each IGFBP was

found to compete with radiolabelled IGF-II for binding to VN. However, IGFBP-5

was only effective at the highest dose tested (100 ng). On the other hand, IGFBP-1,

-2, -3, -4 and –6 competed effectively, even at the lowest doses tested (0.05 ng and

0.2 ng), although IGFBP-2 and –4 had much less dramatic effects on binding of

[125I]-IGF-II to VN compared with IGFBP-1, -3 and -6.

3.3.3 Effect of IGFBPs on modulating binding of [125I]-IGF-I to VN

The effect of IGFBPs on modulating binding of [125I]-IGF-I to VN was also

determined using the solid plate-binding assay (Figure 3.3). IGF-I binding was very

low (380 cpm) compared to that observed with IGF-II (4,500 cpm) in the presence

of VN alone. Addition of IGFBP-1 (panel A) had a significant inhibitory effect on

the very small amount of [125I]-IGF-I binding directly to VN. IGFBP-6 (panel F)

was also inhibitory, but less so. In stark contrast, IGFBP-2, -3, -4 and –5 (panels B,

C, D and E) share similar binding patterns, whereby they enhance binding of

radiolabelled IGF-I to VN by 3-fold, 2-fold, 3.5-fold and 8-fold respectively, at their

particular optimal concentrations. Maximum binding of labelled IGF-I to VN was

observed with 0.5 ng IGFBP-3 while maximum binding of labelled IGF-I was found

at 5 ng for IGFBP-2, -4 and –5.

3.3.4 Ability of IGF peptides to compete for binding of [125I]-IGF-I to VN in

the presence of IGFBP-3 or IGFBP-5

In order to demonstrate that the enhanced binding of [125I]-IGF-I to VN in the

presence of IGFBPs was specific and involved formation of an IGF-I:IGFBP:VN

complex, competitive solid-plate binding studies were undertaken in the presence of

IGFBP-3 or IGFBP-5 with unlabelled IGF-I or the IGF-I analogue, des(1-3)IGF-I,

which has very low affinity for these IGFBPs (Francis et al., 1993). Des(1-3)IGF-I

was much less effective than IGF-I at competing for binding of labelled IGF-I to VN

in the presence of IGFBP-3 (Figure 3.4A). Half-maximal competition

48

Figure 3.2 Effect of IGFBPs on the binding of [125I]-IGF-II to VN. Panels A to

F show radiolabelled IGF-II binding to VN in the absence and presence of IGFBPs.

Ten thousand cpm of IGF-II tracer were added to wells containing pre-bound VN

with increasing amounts of IGFBPs. Data is expressed as percentage of control

([125I]-IGF-II and VN alone) where 100% is approximately 4,500 cpm. Each data

point is the mean ± SEM of triplicate wells from 3 experiments which have been

corrected for non-specific binding (400 cpm). Significant differences from the ‘VN

only’ value are indicated by * p < 0.05 and ** p < 0.01.

120 100

80 60 40 20

0

% o

f Con

trol

A: IGFBP-1

****

**

**

****

**

B: IGFBP-2

****

****

****

*

120 100

80 60 40 20

0

% o

f Con

trol

C: IGFBP-3

****

** ** **

**

D: IGFBP-4

**** **

**

F: IGFBP-6

****

** ** **

** **

0 0.05 0.2 0.5 2 5 20 100 IGFBP (ng)

120 100

80 60 40 20

0

% o

f Con

trol

E: IGFBP-5

**

**

0 0.05 0.2 0.5 2 5 20 100 IGFBP (ng)

49

Figure 3.3 Effect of IGFBPs on the binding of [125I]-IGF-I to VN. Panels A to F

show radiolabelled IGF-I binding to VN in the absence and presence of IGFBPs.

Ten thousand cpm of IGF-I tracer were added to wells containing pre-bound VN

with increasing amounts of IGFBPs. Data is expressed as percentage of control

([125I]-IGF-I and VN alone) where 100% is approximately 380 cpm. Each data point

is the mean ± SEM of triplicate wells from 3 experiments which have been corrected

for non-specific binding (220 cpm). In the absence of VN, [125I]-IGF-I binding to

IGFBP-5 was less than that of the non-specific binding. Significant differences from

the ‘VN only’ value are indicated by * p < 0.05 and ** p < 0.01.

% o

f Con

trol

600 500 400 300 200 100

0

C: IGFBP-3 D: IGFBP-4

**

****

**

600 500 400 300 200 100

0

% o

f Con

trol

A: IGFBP-1

********** **

B: IGFBP-2

******

*

**

0 0.05 0.2 0.5 2 5 20 100 IGFBP (ng)

F: IGFBP-6

******% o

f Con

trol

0 0.05 0.2 0.5 2 5 20 100 IGFBP (ng)

600 500 400 300 200 100

0

808

E: IGFBP-5

**

**

**

****

50

Figure 3.4 Ability of IGF peptides to compete for binding of [125I]-IGF-I to VN

in the presence of IGFBP-3 or IGFBP-5. Panel A shows binding of IGF-I tracer to

VN in the presence of 0.5 ng IGFBP-3 (the optimal dose, see Figure 3.3) with

increasing amounts of either IGF-I or des(1-3)IGF-I while Panel B is in the presence

of 5.0 ng IGFBP-5 (the optimal dose, see Figure 3.3). Data are represented as

percentage of control (IGFBP in the presence of IGF-I tracer and VN) whereby

additions of IGF-I or its analogue reduce the additive effects of the complex. In the

absence of VN, binding of [125I]-IGF-I to IGFBP-3 or –5 was less than the value for

non-specific binding (220cpm). Values shown are the mean ± SEM of triplicate wells

from 3 experiments. Significant differences from the control are indicated by * p <

0.05 and ** p < 0.01.

% o

f Con

trol

0 0.1 1 10 100

A: IGFBP-3 (0.5 ng) IGF-I des(1-3)IGF-I

**

**

**

****

****

**

**

** ** ** ****

IGF (ng) 0 0.1 1 10 100

% o

f Con

trol

B: IGFBP-5 (5.0 ng)

** *** ** **

*

**

**

** **** **

**

IGF (ng)

120

100

80

60

40

20

0

120

100

80

60

40

20

0

IGF-I des(1-3)IGF-I

51

for [125I]-IGF-I binding to VN occurred at much lower levels for IGF-I (much less

than 0.1 ng) than for des(1-3)IGF-I (0.3 ng). Nevertheless, at the highest dose tested,

both peptides were able to negate any enhancing effects of IGFBP-3, in terms of

facilitating binding of IGF-I to VN. In the presence of IGFBP-5 (Figure 3.4B), des(1-

3)IGF-I was ineffective in reducing binding of [125I]-IGF-I to VN. Half-maximal

displacement for IGF-I occurred at approximately 1 ng while des(1-3)IGF-I reduced

binding by only 40% at the highest dose. These data strongly suggest that the binding

of [125I]-IGF-I to VN requires the formation of IGF-I:IGFBP:VN complexes.

3.4 DISCUSSION The studies reported here extend previous observations in which VN was identified as

a novel high-affinity IGF-II binding protein (Upton et al., 1999) that may be

responsible for mediation of many effects of IGF-II in the extracellular environment.

The same earlier studies (Upton et al., 1999) revealed that IGF-I did not bind directly

to VN. This was somewhat surprising given the increasing evidence suggesting a key

role for VN in mediating a number of core cellular effects of IGF-I such as cellular

DNA synthesis, type-1 IGF receptor autophosphorylation and cell migration (Jones et

al., 1995; Zheng and Clemmons, 1998; Maile et al., 2001; Grulich-Henn et al., 2002).

To explain this, we proposed that IGFBPs may be specifically required to mediate

binding of IGF-I to VN.

This study provides evidence to support this hypothesis by demonstrating that IGF-I

can only interact with VN via the intermediate involvement of IGFBPs. This

investigation has shown for the first time that: i) direct binding of IGF-II to VN does

not require IGFBPs but is competitively inhibited by IGFBPs; ii) IGF-I binding to VN

is significantly enhanced by all IGFBPs except for IGFBP-1 and –6; and iii) the role

of IGFBPs is specific since des(1-3)IGF-I is a poor competitor for binding of labelled

IGF-I to VN in the presence of IGFBPs.

IGF-II binding to VN is independent of IGFBPs, this being demonstrated by two

means. First, by the equivalent competitive inhibition of [125I]-IGF-II binding by

wild-type IGF-II and by two analogues, des(1-6)IGF-II and [Leu27]IGF-II, which have

reduced affinity for IGFBPs and the type-1 IGF receptor, respectively (Roth et al.,

52

1991; Francis et al., 1993). Second, all six IGFBPs were shown to inhibit IGF-II

binding to VN, at least at the higher levels tested. The most effective IGFBPs were

IGFBP-1, -3 and –6. Competition for binding of IGF-II to VN may occur either by

IGFBPs directly competing with IGF-II for the IGF-II binding site on VN or by the

IGFBPs binding and sequestering IGF-II in solution. The studies reported in this

chapter cannot distinguish between these possibilities for IGFBP-3. For IGFBP-1 and

–6, however, since these IGFBPs also inhibited IGF-I binding to VN, it is likely that

these IGFBPs bind IGF-I and/or IGF-II in solution and hence primarily sequester the

IGFs away from VN (see Figure 3.5). It is likely that the very efficient inhibition of

IGF-II:VN binding by IGFBP-6 reflects its high affinity for IGF-II (Bach et al.,

1993). It was interesting to note that IGFBP-2 and -5, which also have a preference

for IGF-II over IGF-I (Bach et al., 1993), had a much lesser effect on IGF-II binding

to VN. However, this may reflect the fact that IGFBP-6 has a much greater affinity

for IGF-II, whereas IGFBP-2 and -5 have a moderate preference for IGF-II over IGF-

I. Previous biochemical data (Upton et al., 1999) have demonstrated by 2-D gel

electrophoresis that the purified VN used for the IGF-II binding studies was devoid of

any traces of contaminating IGFBPs. This is further substantiated by the inability of

IGF-I to bind to VN.

The finding that IGF-I binding to VN is markedly enhanced in the presence of

IGFBPs is of particular interest. All IGFBPs, except for IGFBP-1 and –6, enhanced

binding of IGF-I to VN to varying degrees (Figure 3.6). The minor inhibitory effect

of IGFBP-1 and -6 suggests that the minor amount of binding of IGF-I directly to VN

was blocked by these IGFBPs, presumably via sequestration of IGF-I in solution

(equivalent to Figure 3.5D). Alternatively, it and/or may be also be an indication of a

low affinity for IGF-I, as is the case with IGFBP-6 (Bach et al., 1993). These data,

taken together with the presence of an RGD integrin-binding motif in IGFBP-1 (Jones

et al., 1993b) and the finding that IGFBP-1 can bind directly to integrins to effect cell

migration and proliferation (Gockerman et al., 1995; Gleeson et al., 2001;

Chakraborty et al., 2002), indicate that these IGFBP-1-stimulated cellular responses

are unlikely to also involve VN. Interestingly, in Figure 3.3 panels C, D and E that

describe the binding of IGF-I to VN in the presence of IGFBPs, there is a noticeable

bell-shape to the curves. In this type of assay where there are increasing amounts or

53

Figure 3.5 Proposed model of solid-plate binding assay for IGF-II:VN. IGFBPs

compete with [125I]-IGF-II for binding to VN, which may occur via one of two

mechanisms: IGFBPs can bind directly to VN (A) as can IGF-II (B). When both

IGFBPs and IGF-II are present in solution, they may compete for binding at the

same or adjacent sites on VN (C) or the IGFBPs sequester IGF-II away from VN

(D).

Figure 3.6 Proposed model of solid-plate binding assays for IGF-I:IGFBP-2 →

IGFBP-5:VN. IGFBPs mediate [125I]-IGF-I binding to VN: IGFBPs alone can bind

to VN (A) while IGF-I binds poorly (B). Together, IGF-I can bind to VN via select

IGFBPs (IGFBP-2, -3, -4 and -5) (C). IGFBP-1 and -6, which may not bind VN,

inhibit [125I]-IGF-I binding to VN, which represents a situation similar to 3.5D

where the IGFBPs sequester the IGFs and together do not bind VN.

C D

A B

II IGFBP

VN -- ++

VN -- ++

IGFBP II IGFBP II

VN -- ++

VN -- ++

CA B

I IGFBP

VN -- ++

VN -- ++

VN -- ++

I

IGFBP

HBD region polyanionic region

54

concentrations of a component of the complex, it is typical to observe a sigmoidal-

shape curve indicating saturation of the complex. It is not unusual to observe bell-

shape curves in cell-based binding reactions where it can be explained in terms of

transition between dimers and monomers, receptor dimerisation via sequential binding

as well as down-regulation of receptors (De Meyts et al., 1994; Uchida et al., 1999;

Stone et al., 2001). In this instance, however, there may be three likely scenarios.

Firstly, it is likely the complex will reach equilibrium when allowed to bind

overnight, and can be summarised by the model in Figure 3.7. In this case, it is

possible the IGFBPs are in excess of either the [125I]-IGF-I or the VN and thereby the

number of trimeric complexes that can be formed is limited (Figure 3.8).

Figure 3.7 Proposed model to explain IGF:IGFBP:VN bell-shaped binding curves. Formation of IGF-I:IGFBP:VN trimeric complexes can form via different pathways. Pre-formed binary complexes may bind a third protein or binding of complex components may be sequential.

The above model also describes a number of scenarios for the formation of IGF-

I:IGFBP:VN complexes. A number of binary complexes may form in the

intermediate steps leading to ternary complex formation. While the lack of binding

between IGF-I and VN eliminates IV complexes, other binary complexes that may

form include IGF-I:IGFBP (IB) and IGFBP:VN (BV). This model describes the

formation of trimeric complexes as sequential, that is, BV can bind to IGF-I

resulting in IBV, or preformed IB can bind to V. Although data to support the

formation of the active complex by both pathways have not been shown here, both

scenarios result in the formation of an active product, IBV. As mentioned above,

I

V

B

I B

V B

V V B I

B B V B B

I

B I = IGF-I B = IGFBP V = VN

55

bell-shaped curves have been reported in a number of binding studies where

sequential binding can contribute to this shaped curve. However, the possible

formation of inactive IGFBP dimers (BB) (Koedam et al., 1997), alone or with VN,

may be the significant factor to describe the bell-shape curve seen in this study. The

increase in the amount of IGFBP which may lead to BB or BBV limits IGF-I

binding to IGFBP, which results in a decrease in IGF-I bound indirectly to VN at the

higher doses of IGFBP (Figure 3.8). Nevertheless, none of these explanations can

be substantiated by the studies reported here, but merely provide a possible model

for the nature of the curves observed.

Figure 3.8 Schematic of binding model. IGFBP can bind to VN or IGF-I; IGFBP bound to VN can then bind IGF-I; when IGFBP amount is increased, IGFBP is in excess leaving IGFBP to either bind VN, form dimers or complex with VN and IGF-I or IGF-I alone.

The specificity of the requirement for IGFBPs to facilitate IGF-I binding to VN was

also demonstrated. Through competitive inhibition studies, it was shown that while

unlabelled IGF-I was effective in reducing the enhancing effects of both IGFBP-3

and –5 on binding of [125I]-IGF-I to VN, des(1-3)IGF-I, which has a much reduced

affinity for IGFBPs, especially IGFBP-5 (Francis et al., 1993), was a great deal less

effective. These data show the likelihood for IGFBP (-3 or –5) to mediate the

binding of IGF-I to VN.

The findings presented here, along with recent data of others (Grulich-Henn et al.,

2002; Maile et al., 2002; Nam et al., 2002), provide important new insights into the

mechanism by which IGF-I mediates its effects via VN and VN-binding integrins.

Although it has not been specifically addressed in this study, these findings also

offer an explanation as to how IGF-II and IGF-I can exert different functions as

I B

V

IB

V

I

BV

I

B

V

B

B

BB

IB

56

IGF-II appears to bind directly to VN, whereas IGF-I binds indirectly via select

IGFBPs. Thus, despite their structural similarity, the IGFs have clearly evolved

different regulatory mechanisms to provide the capacity for different cellular

functional roles.

It is here proposed that four of the IGFBPs, namely IGFBP-2, -3, -4 and -5, enhance

IGF-I binding to VN by forming a heterotrimeric complex comprised of IGF-I:

IGFBP:VN, and that this complex may be responsible for a number of cellular

responses. Previous studies by Clemmons and co-workers indicated there is a

functional and specific connection between IGF-I and VN, as blocking of the VN

receptor, αvβ3, inhibited IGF-I mediated cellular responses (Rees and Clemmons,

1998; Clemmons et al., 1999). Grulich-Henn et al. (2002) have also recently

demonstrated that transport of IGF-I across endothelial cell monolayers required

IGF-I interacting with VN. These investigations also suggested that VN was not

likely to be a primary binding site for IGF-I and that IGFBPs could be implicated.

The results from the study reported here, in which IGF-I is linked to VN via

IGFBPs, can potentially explain the observation that VN is critical in a number of

IGF-I-stimulated cellular responses such as those reported by Clemmons et al.

(1999) and Grulich-Henn et al. (2002), despite there being only minimal direct

binding of IGF-I to VN (Upton et al., 1999). Together, these findings give insights

as to how IGF-I can mediate diverse effects such as cell migration and cellular DNA

synthesis and, moreover, suggest that VN may have a critical role in linking effects

requiring both activation of integrins and the type-1 IGF receptor as demonstrated

by Maile et al. (2002). Thus, the IGF:IGFBP:VN complex appears to be important

in normal growth and development and further functional and structural

investigation of this complex may provide mechanisms for maintaining these

physiologies in altered diseased states.

57

CHAPTER 4: EFFECT OF GLYCOSYLATION AND

HEPARIN-BINDING ON VN:IGF:IGFBP

INTERACTIONS

58

4.1 INTRODUCTION An increasing number of studies are demonstrating a link between the IGFs and the

ECM protein, VN. In addition to the previous chapter demonstrating binding of

IGFs to VN in the presence of IGFBPs as is the case with IGF-I, two studies have

provided functional evidence to support the theory that IGFs and VN together are

involved in cellular proliferation in endothelial and muscle cells. In particular,

Grulich-Henn et al. (2002) demonstrated that in human vascular endothelial cells

IGF-I bound the ECM, via involvement of VN, and this interaction resulted in

increased cell survival. In addition, it has been demonstrated that blocking the VN

receptor, the αvβ3 integrin, results in attenuation of IGF-I-mediated effects such as

DNA synthesis in smooth muscle cells (Clemmons et al., 1999).

The data presented in this section progresses the studies reported in Chapter 3

describing the interaction between IGFs, IGFBPs and VN. In particular, these

studies focus on the potential role that post-translational modifications may have in

the IGF:IGFBP:VN interaction, and also, identifies protein domains that may be

involved in complex formation. As outlined in the literature review (Chapter 1),

there are relatively few studies on the subject of glycosylation of IGFBPs as it is

generally accepted that glycosylation has little relevance in the physiological setting.

However, many commercial sources of IGFBPs are produced in expression systems

that vary in their ability to glycosylate proteins. For example, glycosylation is

absent from E.coli-produced protein (Bagnall et al., 2003), while the extent of

glycosylation is variable and simple in baculovirus/insect systems. In this chapter,

IGFBPs produced in mammalian systems are compared with those produced in both

E.coli and baculovirus in their ability to bind IGFs and to form a complex with IGFs

and VN.

Additionally, a structural feature that is regarded as important in many ECM protein

interactions is the heparin-binding domain (HBD). HBDs are commonly

characterised by heparin-binding consensus sequences which contain a large

constituent of basic residues: lysine (Lys - K) and arginine (Arg - R). The presence

of either a short sequence: XBBXBX or a larger sequence: XBBBXXBX (where B is

a basic amino acid and X is any non-basic amino acid) can be used to putatively

identify a heparin-binding region (Cardin and Weintraub, 1989). These regions or

59

domains are commonly important for many extracellular protein interactions (Booth

et al., 1995; Campbell and Andress, 1997a). For example, IGF-I has been shown to

interact with HBD-containing proteins such as epidermal growth factor (EGF). The

synergism of EGF together with IGF-I, appears to be important in re-

epithelialisation during wound repair (Marikovsky et al., 1996). In addition, both

VN and some of the IGFBPs (2, 3, 5 and 6) contain heparin-binding consensus

sequences (see Figure 1.4), which may also have key roles in the regulation of

cellular responses.

The data presented here addresses the role of both glycosylation and heparin-binding

domains in the interactions between IGF:IGFBP:VN complexes. Therefore the

specific aims of the studies reported in this chapter were to:

1. Examine the role of glycosylation of IGFBP-3 and -6 on trimeric complex

formation with either IGF-I:VN or IGF-II:VN

2. Characterise the binding of IGFBP glycosylation mutants to IGFs

3. Determine if the HBD of IGFBP-3 and -5 contributes to trimeric complex

formation with IGF-I:VN

4. Characterise the binding of the HBD mutants to IGF-I

4.2 EXPERIMENTAL PROCEDURES Brief descriptions of proteins and procedures specific to this chapter are provided;

Chapter 2 contains full details.

Materials

IGF-I, IGF-II, IGFBP-1, -2, -4 and glycosylated IGFBP-6 were purchased from

GroPep Ltd. IGFBP-5 and glycosylated IGFBP-3 were purchased from the Kolling

Institute of Medical Research (University of Sydney, NSW, Australia), while

glycosylated HBD mutant IGFBP-3 and mutant non-glycosylated IGFBP-3 were

produced as described previously by Firth et al. (1999). Baculovirus-expressed

wild-type IGFBP-3 and HBD mutant IGFBP-3 were a kind gift from Dr Susan

Durham and Prof David Powell (Baylor College of Medicine, Houston, TX, USA)

(see Table 4.1 for details of mutations). Non-glycosylated IGFBP-6 was kindly

60

Table 4.1 Summary of key characteristics of each IGFBP-3.

Binding

Protein# Expression Mutation Glycosylation

Approximate

MW from

Figure 4.3

gly BP-3 Adenovirus in

mammalian cells None Full 42

BV.BP3 Baculovirus in

insect cells None Variable or Partial 41

N109D

BP3 Escherichia coli N109D None 35

non-gly

BP3

Adenovirus in

mammalian cells

N89→A / N109→A /

N172→A None 35

HBDm

BP3

Adenovirus in

mammalian cells

K228GRKR →

MDGEA Full 41

BV.HBDm

BP3

Baculovirus in

insect cells

BP1: K183NGFYHSR

QCETSMDGEA200 →

BP3: K215KGFYKKK

QCRPSKGR KR232

Variable or Partial 40

Binding Protein#: References for the various IGFBP-3 preparations include: gly

BP3 (Firth et al., 1999), BV.BP3 and BV.HBDm BP3 (Durham et al., 1999), N109D

BP3 (Upstate data sheet: Cat# 01-204), non-gly BP3 (Firth and Baxter, 1999),

HBDm BP3 (Firth et al., 1999).

61

donated by Dr Leon Bach (Austin and Repatriation Medical Centre, University of

Melbourne, Vic, Australia). IGFBP-5 heparin-binding mutants were a kind gift

from Prof. David Clemmons (Department of Medicine, University of North

Carolina, NC, USA). MBF (IGF-I analogue with a HBD from basic fibroblast

growth factor tagged at the carboxy-terminus) was provided by Prof. Leanna Read

(TGR BioSciences, Adelaide, SA, Australia). Human VN was purchased from

Promega Corporation. All other reagents are as described in Section 2.1. General

plasticware used in experiments containing IGFBPs and VN was siliconised with

Sigmacote and left to air-dry overnight and low-adhesion plasticware was purchased

from Quantum Scientific Pty Ltd.

Radiolabelling of Proteins

IGF-I, IGF-II, MBF and IGFBP-3 (glycosylated and non-glycosylated) were

iodinated according to the method described in Section 2.3. The chloramine-T

reactions were performed for 1 minute for the IGFs (10 μg) and IGF analogue, MBF

(10 μg), and 15 seconds for IGFBP-3 (5 µg) which was then purified using heparin

affinity chromatography.

Solid-Plate Binding Assay

Please refer to Section 2.6 for a full description of the method. Briefly, IGF:

VN:IGFBP binding assays were performed in removable 4 HBX Immulon wells

coated with or without VN in DMEM for 2 - 4 hr. Ten thousand cpm of [125I]-

labelled protein (IGF-I, IGF-II, glycosylated IGFBP-3 or non-glycosylated IGFBP-

3) in HBB + 0.5% BSA in the absence or presence of increasing concentrations of

unlabelled IGFs, IGF analogues and/or IGFBPs ± heparin were incubated overnight

at 4°C. Unbound radiolabelled protein was then removed and remaining

radioactivity bound in each well was then determined using a γ-counter.

PEG Precipitation Assay

Full details are described in Section 2.7. Briefly, 20,000 cpm of either [125I]-IGF-I,

[125I]-IGF-II or [125I]-MBF were added to increasing amounts of IGFBPs or VN in 5

mL tubes and allowed to bind overnight at 4˚C. Binary complexes that form were

precipitated out of solution and radioactivities remaining in the pellets were counted

using a γ-counter.

62

4.3 RESULTS 4.3.1 Comparison of the effects of glycosylated and non-glycosylated IGFBP-3

on [125I]-IGF-I binding to VN

To examine whether glycosylation of IGFBP-3 was important for the enhancement

of IGF-I binding to VN observed in earlier assays, binding of labelled IGF-I to VN

in the presence of glycosylated IGFBP-3 or non-glycosylated IGFBP-3 was

compared (Figure 4.1). Both forms of IGFBP-3 were produced using Adenovirus

expression in mammalian 911 human embryonic retina cells, but the non-

glycosylated form had amino acid substitutions of alanine for asparagine at the

glycosylation sited located at positions 89, 109 and 172. Non-glycosylated IGFBP-3

was approximately 15-times more effective in enhancing binding of labelled IGF-I

to VN than glycosylated IGFBP-3 at 0.5 ng. However, experiments with

radiolabelled glycosylated and non-glycosylated IGFBP-3 indicated that there was

no significant difference in binding of non-glycosylated IGFBP-3 (695 ± 110 cpm)

directly to VN compared to that observed with glycosylated IGFBP-3 (595 ± 111

cpm) (Figure 4.2). Thus, the enhanced binding of labelled IGF-I to VN found with

non-glycosylated IGFBP-3 appears to be solely related to the ability of glycosylated

or non-glycosylated IGFBP-3 to bind to IGF-I.

4.3.2 Comparison of the ability of IGFBP-3 preparations with different states

of glycosylation to bind [125I]-IGF-I

While comparison of variously glycosylated forms of IGFBP-3 in the physiological

setting has been found by others to not be relevant to IGFBP-3 function (Conover,

1991), the previous results (Figure 4.1) led to the direct examination of whether

glycosylation affected binding of IGF-I to IGFBP-3. Consequently, binding of

[125I]-IGF-I to IGFBP-3 preparations from 3 different sources was compared. The

specific details of the various IGFBP-3 used in this chapter are outlined in Table 4.1

and the size shift due to glycosylation as determined by SDS-PAGE is shown in

Figure 4.3.

63

Figure 4.1 Comparison of glycosylated and non-glycosylated IGFBP-3 on [125I]-

IGF-I binding to VN. IGF-I tracer was added with either glycosylated IGFBP-3 or

non-glycosylated mutant IGFBP-3 to VN pre-bound to wells. Data are expressed as

a percentage of control, which is the binding observed in the presence of IGF-I

tracer and VN alone (no IGFBP-3). Values shown are the mean ± SEM of triplicate

wells from 3 experiments. Significant differences between glycosylated and the

non-glycosylated mutant for the same amount of IGFBP-3 are indicated by ** p <

0.01.

3000

2500

2000

1500

1000

500

0 0 0.05 0.2 0.5 2 5 20 100

IGFBP-3 (ng)

% O

F C

ON

TRO

L **

****

**

**

**

non-gly BP3 gly BP3

64

Figure 4.2 Comparison of glycosylated and non-glycosylated [125I]-IGFBP-3

binding to VN. VN was pre-bound to the wells and 10,000 cpm of either

glycosylated [125I]-IGFBP-3 or mutant non-glycosylated [125I]-IGFBP-3 were added.

Values shown are the corrected mean ± SEM of six replicate treatments from 3

experiments. Non-specific binding was approximately 210 cpm.

900

750

600

450

300

150

0 glycosylated non-glycosylated

RADIOLABELLED IGFBP-3

CO

RR

ECTE

D C

PM

65

Figure 4.3 Silver stain of IGFBP-3 from various sources run on a 4%

stacking/12% resolving SDS-PAGE. Under reducing conditions, the various

IGFBP-3 preparations were run on an SDS-PAGE. The molecular weights of each

major band were approximated through comparison with the MW markers using a

log plot of marker MWs against the migration distance. The variations in size

(Table 4.1) reflect the extent of glycosylation.

→→

→

→

→

114 88

50.7

35.5

28.8

MW (kDa) BV.

HB

Dm

BP3

HB

Dm

BP3

non-

gly

BP3

gly

BP3

N10

9D B

P3

BV.

BP3

MW

Mar

ker

66

In order to characterise the binding ability of the various preparations of IGFBP-3 to

[125I]-IGF-I, binary complexes were precipitated using PEG and the radioactivity in

the pellet was counted. IGFBP-3 binding to IGF-I, expressed as % B/T (percentage

bound of total counts added), steadily increased, as expected, in response to

increasing amounts of IGFBP-3 (Figure 4.4). At Bo (the absence of IGFBP), % B/T

was 4.9% and at the peak response (achieved at 10 ng of IGFBP-3) glycosylated

IGFBP-3 (gly BP3), BV-expressed IGFBP-3 (BV.BP3), E.coli-produced non-

glycosylated IGFBP-3 (N109D BP3) and mutant non-glycosylated IGFBP-3 (non-

gly BP3) gave % B/T values of 36.6%, 49.5%, 42% and 52.2% respectively.

Significant differences were only found when binding of [125I]-IGF-I to non-

glycosylated IGFBP-3 and glycosylated IGFBP-3 were compared. At 0.5 ng and 1

ng of IGFBP-3, there was a 2-fold difference in binding of [125I]-IGF-I when

compared to the glycosylated form of IGFBP-3, whereas at the higher amounts (10

and 25 ng) the difference was approximately 1.3-fold. Although no other significant

differences were detected, a clear trend was noted whereby glycosylated IGFBP-3

had somewhat reduced [125I]-IGF-I binding compared to IGFBP-3 forms that had

lower or zero levels of glycosylation. Again, this trend may help to explain the

noteworthy differences seen in Figure 4.1.

4.3.3 Comparison of the effects of glycosylated and non-glycosylated IGFBP-6

on [125I]-IGF binding to VN

Another IGFBP that is post-translationally glycosylated is IGFBP-6, although it is

O-glycosylated in contrast to IGFBP-3, which is N-glycosylated (Bach et al., 1992).

After observing the dramatic effects of glycosylation on IGF-I binding to VN in the

presence of IGFBP-3 (Figure 4.1), it was examined whether glycosylation of

IGFBP-6 had any effect on IGF-I and IGF-II binding to VN (Figure 4.5). Because

of the essential lack of effect of IGFBP-6 on IGF-I binding to VN (Figure 3.3), it

was predicted that no change would be observed when non-glycosylated and

glycosylated IGFBP-6 were compared in their ability to alter IGF-I binding to VN.

Indeed, this was the case (Figure 4.5A), with both forms of IGFBP-6 causing a

decrease in the very low intrinsic binding of IGF-I to VN. Non-glycosylated

IGFBP-6, however, was less potent at inhibiting IGF-II binding to VN than

67

Figure 4.4 Comparison of the ability of IGFBP-3 from various sources to bind

[125I]-IGF-I. Twenty thousand cpm of [125I]-IGF-I was added to increasing amounts

of IGFBP-3, allowed to bind overnight at 4˚C and then binary complexes were

precipitated using PEG. Results are expressed as a percentage of bound cpm over

total cpm added/tube (% B/T) and values are the mean of duplicate treatments from

3 experiments. Using the Mann-Whitney test, significant differences were found in

binding of labelled IGF-I to non-gly IGFBP-3 versus wild-type IGFBP-3 (gly BP3),

and are indicated by * p < 0.05.

% B

/ T

60

50

40

30

20

10

0 0 0.1 1 10 100

IGFBP-3 (ng)

N109D BP3gly BP3

BV BP3 non-gly BP3 X

*

*

*

*

68

glycosylated IGFBP-6 (Figure 4.5B). Only at the higher doses of non-glycosylated

IGFBP-6 tested (100 ng and 300 ng) were there notable reductions in IGF-II

binding. The strong competition of glycosylated IGFBP-6 for IGF-II binding to VN

is presumably due to sequestration of IGF-II by IGFBP-6 as a result of its high

affinity for IGF-II. The data presented here is from two experiments with each dose

tested in triplicate; however the 300 ng dose was only tested in one experiment.

Although this data was not repeated, there was only a 2.3-fold difference between

glycosylated and non-glycosylated IGFBP-6 for IGF-II binding, which was found to

be non-significant.

4.3.4 Binding of non-glycosylated IGFBP-6 and glycosylated IGFBP-6 to

[125I]-IGF-II

To determine if the lower potency of non-glycosylated IGFBP-6 in inhibiting direct

IGF-II binding to VN was due to a lower affinity of non-glycosylated IGFBP-6 for

IGF-II direct binding studies were carried out. Binding of radiolabelled IGF-II to

glycosylated and non-glycosylated IGFBP-6 in the absence of VN was compared

(Figure 4.6). Increasing the amounts of IGFBP-6 gave rise to significant differences

between the two forms of IGFBP-6 in their ability to bind [125I]-IGF-II at 0.5 ng, 1

ng, 2.5 ng, 5 ng and 10 ng (p < 0.05). While % B/T for glycosylated IGFBP-6

steadily increased from 17.8% to 27.5% at these data points, binding of [125I]-IGF-II

to non-glycosylated IGFBP-6 increased to only lower values of 13.4% to 18.3%

over the same range of amounts tested. These data denote that non-glycosylated

IGFBP-6 has a significantly lower affinity for IGF-II than glycosylated IGFBP-6,

and that this is the likely explanation for the much reduced effect of non-

glycosylated IGFBP-6 on IGF-II:VN complex formation (Figure 4.5B).

69

Figure 4.5 Comparison of the effects of glycosylated and non-

glycosylated IGFBP-6 on A) [125I]-IGF-I and B) [125I]-IGF-II binding to

VN. IGF tracer was added with either glycosylated IGFBP-6 or non-

glycosylated IGFBP-6 to wells pre-bound with VN. Data are expressed as

bound IGF (corrected cpm) where NSB: [125I]-IGF-I = 125 cpm, [125I]-IGF-II

= 340 cpm. Note differences in scales between panel A) with [125I]-IGF-I

and B) [125I]-IGF-II. Values shown are the mean ± SEM of triplicate wells

from 2 experiments (except for 300 ng which is from 1 experiment).

Tukey’s test was used to detect significant differences between glycosylated

and non-glycosylated IGFBP-6 and are indicated by ** p < 0.01.

200

175

150

125

100

75

50

25

0 0 0.01 0.1 1 10 100 1000

A: [125I]-IGF-I binding to VN

IGFBP-6 (ng)

BO

UN

D IG

F (C

OR

REC

TED

CPM

)

4000

3500

3000

2500

2000

1500

1000

500

0

B: [125I]-IGF-II binding to VN

IGFBP-6 (ng)

******

**

**

0 0.01 0.1 1 10 100 1000

non-gly BP6 gly BP6

non-gly BP6 gly BP6

BO

UN

D IG

F (C

OR

REC

TED

CPM

)

70

Figure 4.6 Binding of glycosylated and non-glycosylated IGFBP-6 to [125I]-IGF-

II in the absence of VN. Twenty thousand cpm of IGF-II tracer was added to either

glycosylated IGFBP-6 or non-glycosylated IGFBP-6 and the complexes were

precipitated using PEG. Data are expressed as % B/T of [125I]-IGF-II added. Values

shown are the mean ± SEM of duplicate samples from 2 experiments. Mann-

Whitney U-test was used to demonstrate significant differences between

glycosylated and non-glycosylated IGFBP-6 and are indicated by * p < 0.05.

IGFBP-6 (ng)

% B

/ T

30

25

20

15

10

5

0

non-gly BP6 gly BP6

0 0.1 1 10

* ** * *

*

71

4.3.5 Importance of the HBD in IGFBP-3 in mediating [125I]-IGF-I binding to

VN

Heparin-binding domains are commonly required for many extracellular protein

interactions (Booth et al., 1995; Incardona et al., 1996; Campbell et al., 1998; Gui

and Murphy, 2001) and interestingly, IGFBP-3 contains such a domain. In order to

determine the role of this domain in IGFBP-3 enhancement of IGF-I binding to VN,

binding studies using glycosylated IGFBP-3 and an HBD mutant IGFBP-3 were

undertaken. The HBD mutant (K228GRKR → MDGEA) has previously been

demonstrated to bind IGF-I with similar affinity to that of the wild-type IGFBP-3

(Firth et al., 1998). As demonstrated earlier (Figure 3.3 Panel C), the presence of

glycosylated IGFBP-3 increases the binding of labelled IGF-I to VN by

approximately 2-fold. However, when the HBD domain of IGFBP-3 is mutated,

IGFBP-3-mediated binding of [125I]-IGF-I to VN is completely negated (Figure 4.7).

Indeed, basal [125I]-IGF-I binding is inhibited, presumably due to the sequestration

in solution of the labelled IGF-I by the added mutant IGFBP-3, since this mutant

IGFBP-3 has been shown to not associate with the cell surface (Firth et al., 1998),

and presumably VN. Although there is no direct evidence for this latter statement, it

seems clear that the HBD in IGFBP-3 does bind to VN since it cannot enhance IGF-

I binding to VN despite being able to bind IGF-I well. These data again indicate the

need for a complex involving IGFBPs to facilitate IGF-I binding to VN.

4.3.6 Comparison of [125I]-IGF-I binding to wild-type IGFBP-3 and HBD

mutants of IGFBP-3

In order to eliminate differences in IGF-I binding for the particular IGFBP-3

preparations (wild-type and HBD mutants) used in the previous figure (Figure 4.7),

the ability of the two different preparations of HBD mutant to bind to [125I]-IGF-I

was compared with wild-type IGFBP-3 (Figure 4.8). One of the mutants was

produced in adenovirus in mammalian cells and the other was produced in

baculovirus in insect cells (see Table 4.1). At 5 ng, 10 ng and 25 ng, both mutants

had significantly reduced binding to radiolabelled IGF-I, despite an earlier report

that affinity for the fully glycosylated IGFBP-3 mutant (HBDm BP3) is similar to

intact IGFBP-3 (Firth et al., 1998).

72

Figure 4.7 Effect of the HBD in IGFBP-3 on IGF-I binding to VN. IGF-I tracer

was added with either glycosylated IGFBP-3 or HBD mutant IGFBP-3 to wells pre-

bound with VN. Data are expressed as percentage of control, which is IGF-I tracer

and VN alone (approximately 220 cpm). Values shown are the mean ± SEM of

triplicate wells from 3 experiments. Significant differences between the wild-type

IGFBP-3 and the HBD mutant IGFBP-3, both of which are produced in mammalian

cells, for the same amount of protein are indicated by * p < 0.05 and ** p < 0.01.

250

200

150

100

50

0 0 0.05 0.2 0.5 2 5 20 100

IGFBP-3 (ng)

% O

F C

ON

TRO

L

********

*

HBDm BP3 gly BP3

73

Figure 4.8 Binding of [125I]-IGF-I to wild-type IGFBP-3 compared to HBD

mutants of IGFBP-3. Asterisks above the curve indicate significant differences

between wild-type IGFBP-3 (gly BP3) and HBD mutant IGFBP-3 (HBDm BP3),

both of which were produced in mammalian cells. Asterisks below the curve

indicate significant differences of wild-type IGFBP-3 to baculovirus HBD mutant

IGFBP-3 (BV.HBDm BP3) (* p < 0.05 and ** p < 0.01 determined by Mann-

Whitney U-test). Values shown are the mean ± SEM of duplicate samples from 3

experiments.

60

50

40

30

20

10

0

IGFBP-3 (ng)

% B

/ T

HBDm BP3 gly BP3

BV.HBDm BP3

0 0.1 1 10 100

**

**

*

** *

74

4.3.7 Importance of IGFBP-5 HBD in IGFBP-5 in mediating [125I]-IGF-I

binding to VN

Similar to IGFBP-3, IGFBP-5 also contains an HBD that may be involved in

mediating [125I]-IGF-I binding to VN. This was tested using the solid-plate binding

assay to measure IGF-I binding to VN in the presence of two IGFBP-5 HBD

mutants and wild-type IGFBP-5 (Figure 4.9). The mutations present at the HBD

were as follows: IGFBP-5 HBD mutant 1 (BP5m1): R201A/K202N/K206N/K208N;

and IGFBP-5 HBD mutant 2 (BP5m2): K202A/K206A/R207A. These and similar

mutants have also been examined elsewhere in their ability to bind to heparin and

the ECM (Arai et al., 1996b; Nam et al., 2002; Hsieh et al., 2003). At each dose

tested between 0.5 ng and 20 ng, wild-type IGFBP-5 increased [125I]-IGF-I binding

to VN to a greater degree than both of the HBD mutants. However, for IGFBP-5

HBDm1, only 2 ng, 5 ng and 20 ng were significantly different to wild-type IGFBP-

5 (p < 0.01). Differences in binding for IGFBP-5 HBD mutant BP5m2 to IGF-I

reached significance at 100 ng in addition to the doses seen for BP5m1 (p < 0.05).

The fold difference at the optimum dose for wild-type IGFBP-5 (5 ng) was 2.3 and 3

for BP5m1 and BP5m2 respectively. Both mutants resulted in maximal binding at a

slightly lower amount (2 ng) where there was only a 1.7-fold difference for each,

compared to the wild-type IGFBP-5.

4.3.8 Comparison of binding of [125I]-IGF-I to wild-type IGFBP-5 and HBD

mutants of IGFBP-5

The differences between wild-type IGFBP-5 and the IGFBP-5 HBD mutants in their

ability to bind to [125I]-IGF-I were also examined (Figure 4.10). Both IGFBP-5

mutants have been reported to bind IGF-I with near-normal affinity (Arai et al.,

1996b) and this was supported when re-investigated here using PEG precipitation.

Although wild-type IGFBP-5 was consistently a slightly better binder, as reflected in

the % B/T, significant differences were only found at 0.5 ng and 1 ng for BP5m1

and 0.1 ng for BP5m2, suggesting the observations in Figure 4.9 reflect a change in

affinity of the IGFBP-5 HBD mutants for VN, rather than differences in affinity for

binding [125I]-IGF-I.

75

Figure 4.9 Comparison of wild-type IGFBP-5 with IGFBP-5 HBD mutants in

their ability to mediate [125I]-IGF-I binding to VN. In the figure the HBD

mutants are designated as BP5m1 (R201A/K202N/K206N/K208N) and BP5m2

(K202A/K206A/R207A). Data are expressed as % of Control ([125I]-IGF-I + VN

only) and are the mean ± SEM of triplicate wells from 3 experiments. Significant

differences between wild-type IGFBP-5 (BP5) and the IGFBP-5 HBD mutants for

the same amount of protein are indicated by * p < 0.05 and ** p < 0.01. There were

no significant differences detected between the two HBD mutants.

0 0.2 0.5 2 5 20 100

1000 900 800 700 600 500 400 300 200 100

0

% O

F C

ON

TRO

L BP5m1 BP5

BP5m2

IGFBP-5 (ng)

**

** **

**

****

*

*

76

Figure 4.10 Binding of [125I]-IGF-I to wild-type IGFBP-5 and HBD mutants

of IGFBP-5. Using the PEG precipitation assay, 20,000 cpm of [125I]-IGF-I were

added to increasing amounts of IGFBP-5, allowed to bind and then binary

complexes were precipitated with PEG. Values shown are the mean ± SEM of

duplicate samples from 3 experiments. Mann-Whitney U-test was used to

demonstrate significant differences between wild-type IGFBP-5 and the two HBD

mutants, indicated by * p < 0.05 and ** p < 0.01 (m1 above curve, m2 below curve).

70

60

50

40

30

20

10

0 0 0.1 1 10

IGFBP-5 (ng)

BP5m1 BP5

BP5m2

**

*

*

% B

/ T

77

4.3.9 Examination of IGFBP binding to an IGF-I analogue containing a HBD

(MBF)

To further explore the role of HBD, MBF, an IGF-I analogue with an HBD from

basic fibroblast growth factor tagged at the carboxy terminus was examined in its

ability to bind VN. As seen in Figure 4.11, there was no binding to VN when

binding was examined using the PEG precipitation assay. This was surprising as

one of the minor differences between IGF-I and IGF-II is the presence of an HBD-

like site of basic amino acids in IGF-II, and as a consequence, was predicted to be

the region responsible for binding VN. The addition of an HBD onto IGF-I was

predicted to enable IGF-I to bind to VN. However, this was not the case and in fact,

there was no binding of [125I]-MBF to VN.

In addition, binding of [125I]-MBF to IGFBP-3, and -5 was compared with [125I]-

IGF-I using the PEG precipitation assay (Figure 4.12). Both IGFBP-3 and IGFBP-5

had a higher affinity for [125I]-IGF-I than [125I]-MBF at each dose tested. Significant

differences (** p < 0.01) were detected at 1, 2.5, 5 and 10 ng for IGFBP-3 while

significant differences (* p < 0.05) with IGFBP-5 were detected at each dose tested

except for 0.1 ng.

78

Figure 4.11 Binding of [125I]-MBF to VN using the PEG precipitation assay.

Twenty thousand cpm of [125I]-MBF were added to increasing amounts of VN,

allowed to bind and then binary complexes were precipitated with PEG. Values

shown are the mean ± SEM of duplicate samples from 2 experiments.

14

12

10

8

6

4

2

0 0 0.1 1 10 100

Vitronectin (ng)

% B

/ T

79

Figure 4.12 Comparison of binding of either [125I]-IGF-I or [125I]-MBF (an

IGF-I analogue containing an HBD) to IGFBP-3 and -5. Values shown are the

mean ± SEM of duplicate samples from at least 2 experiments. Note results are

expressed as % of Control (B/T) in order to compare [125I]-IGF-I with [125I]-MBF, in

contrast to earlier figures of PEG data where the y-axis is %B/T. Mann-Whitney U-

test was used to demonstrate significant differences between wild-type IGF-I and the

IGF-I analogue, MBF, indicated by * p < 0.05 and ** p < 0.01.

0 0.1 0.25 0.5 1 2.5 5 10

900 800 700 600 500 400 300 200 100

0

A [125I]-IGF-I [125I]-MBF

0 0.1 0.25 0.5 1 2.5 5 10

1400

1200

1000

800

600

400

200

0

B

IGFBP-5 (ng)

% O

F C

ON

TRO

L (B

/T)

% O

F C

ON

TRO

L (B

/T) [125I]-IGF-I

[125I]-MBF

** **** **

IGFBP-3 (ng)

* ** * **

80

4.4 DISCUSSION

The previous chapter identified the direct interaction of IGF-II and VN, and the

indirect interaction of IGF-I and VN via particular IGFBPs, namely IGFBP-2, -3, -4

and 5. In this chapter, the interaction between IGFs, IGFBPs and VN was further

explored and in particular, concentrated on the effect of IGFBP post-translational

modifications such as glycosylation on the IGF:VN interaction and the role of the

heparin-binding domain in IGFBPs.

Firstly, data presented here have shown that the ability of IGF-I to bind to VN is

markedly influenced by the glycosylation state of IGFBP-3. In most IGF/IGFBP

studies, the effect of the glycosylation state of IGFBPs has received little attention.

Indeed, glycosylation is reported to play little role in IGFBP-3-mediated IGF effects

(Conover, 1991; Sommer et al., 1993; Firth and Baxter, 1999). In contrast, we

demonstrate here that non-glycosylated IGFBP-3 markedly enhances binding of

labelled IGF-I to VN compared to glycosylated IGFBP-3. However, this was not

due to a significantly greater ability of non-glycosylated IGFBP-3 to bind to VN,

demonstrating that other factors apart from glycosylation are involved. Others have

previously demonstrated that de-glycosylated IGFBP-3 has a higher affinity for the

cell surface (Firth and Baxter, 1999; Firth et al., 1999), but not for IGF-I (Sommer et

al., 1993; Firth and Baxter, 1995). However, in this study examining IGF-I binding

revealed that there were small differences in the ability of IGFBPs with different

glycosylation status to bind labelled IGF-I. Moreover, these differences were shown

to be significant when the mammalian cell produced glycosylated and non-

glycosylated IGFBP-3 mutant were compared. In view of the present data the

difference may also translate to preferential binding of de-glycosylated IGFBP-3 to

bind VN associated with the cell surface. While non-glycosylated IGFBP-3 would

not appear to be especially relevant in the physiological context, the observations in

this study suggest that use of non-glycosylated IGFBP-3 in a trimeric protein

complex with IGF-I and VN may well prove to be a potent way to facilitate delivery

of IGF-I to the cell surface - a phenomenon which potentially could be used in

therapeutic and industrial applications to manipulate cell processes.

81

Another interesting finding highlighted in this chapter that also contrasts with the

literature is the significant difference in the ability of non-glycosylated IGFBP-6 to

compete with IGF-II for binding to VN compared to glycosylated IGFBP-6. This is

worthy to note as Bach et al. (1992) showed that enzymatic deglycosylation of

IGFBP-6 had no effect on its affinity for IGF-II. Subsequently, there have been

several discrepant reports: Marinaro et al. (2000) indicated that recombinant non-

glycosylated IGFBP-6 had a three-fold greater affinity for various

glycosaminoglycans over glycosylated IGFBP-6, while Wong et al. (1999) reported

that both the IGFs have similar low affinity constants for glycosylated/non-

glycosylated IGFBP-6. However, these differing reports on whether IGFBP-6

glycosylation does indeed affect its action may reflect the source and production of

IGFBP-6 used in the experimental procedures. Each study used non-glycosylated

IGFBP-6 from a different source, including enzymatic deglycosylated IGFBP-6

(Bach et al., 1992), recombinant E.coli expressed (Marinaro et al., 2000) and yeast

expressed (Wong et al., 1999) IGFBP-6, whereas the data presented in this chapter

compared IGFBP-6 produced in mammalian cells (glycosylated) and E.coli (non-

glycosylated). The results presented in Figure 4.5B indicate that non-glycosylated

IGFBP-6 has a reduced affinity for IGF-II and therefore is less effective at

sequestering IGF-II away from VN as presumed and discussed in the previous

chapter. Moreover, the reduction in affinity of non-glycosylated IGFBP-6 for IGF-II

may enable non-glycosylated IGFBP-6 to compete directly with IGF-II for binding

to VN rather than predominantly being sequestered by IGF-II in solution. In

summary, these data suggest that it is possible that deglycosylation of IGFBP-6

alters its affinity for IGF-II and the source of IGFBP-6 also plays a role.

Results from this study identify structural characteristics that appear to be important

in the IGF:IGFBP:VN interaction. It is hypothesised that both glycosylation and the

HBD play an important role in IGFBP mediation of IGF-I binding to VN. Building

on data presented in the previous chapter, the studies presented here show that

IGFBP-3 enhancement of IGF-I binding to VN requires an intact IGFBP-3 HBD.

IGFBP-3-mediated binding of IGF-I to VN was abolished when the IGFBP-3 HBD

region was mutated even though the affinity of IGF-I for this IGFBP-3 variant has

been reported to be similar to that of the wild-type IGFBP-3 (Firth et al., 1998). In

contrast, the data presented here demonstrate that IGFBP-3 HBD mutants produced,

82

either in mammalian cells or by baculovirus, have a decreased affinity for IGF-I

compared to wild-type IGFBP-3. This result, taken together with the negated IGF-I

binding to VN in the presence of HBD mutant IGFBP-3, indicates the importance of

this region in the IGF:IGFBP:VN interaction, at least with IGFBP-3. The heparin-

binding site in proteins such as IGFBP-3 and VN have previously been implicated in

cell association processes (de Boer et al., 1992; Rosenblatt et al., 1997; Hocking et

al., 1999; Devi et al., 2000; Mazerbourg et al., 2000; Forsten et al., 2001).

Moreover, the finding that the HBD is important for binding of IGFBP-3 to VN

mirrors other findings that the HBD of IGFBP-3 is required for binding to

fibronectin, a protein with similar functions to VN (Gui and Murphy, 2001).

Likewise, IGFBP-5 has been shown to bind via its HBD to VN with high affinity

(Nam et al., 2002) and the functional effects of IGFBP-5 required trimeric complex

formation with IGF-I. Furthermore, this complex was found to effect IGF-I-

mediated functional responses through the αvβ3 integrin (Nam et al., 2002).

Interestingly, these IGF-I-stimulated responses were decreased in the presence of

heparin, again highlighting the potential involvement of IGFBP basic amino acid

residues that form the characteristic HBD, in binding to VN. The findings in

Sections 4.3.7 and 4.3.8 again corroborate these earlier studies. Thus, mutation of

the HBD of IGFBP-5 significantly reduced IGF-I binding to VN, seemingly from a

decreased affinity of IGFBP-5 for VN as the mutants had a similar affinity for IGF-I

as found with the wild-type IGFBP-5. Interestingly, the binding curve for the

IGFBP-5 HBD mutants (Figure 4.9) is also biphasic as noted in Chapter 3. Again

suggesting a more complex binding model for IGF-I:IGFBP:VN. Indeed, the study

by Nam et al. (2002), in which labelled IGFBP-5 was used to examine VN:IGFBP-5

complex formation, independently validates the findings of the current study which

demonstrate IGF:IGFBP:VN complex formation using labelled IGF-I.

To further investigate the role of HBDs in the IGF:IGFBP:VN interaction, the IGF-I

analogue known as MBF, comprised of an HBD linked to IGF-I, was examined in

its ability to bind to VN. As IGF-II can bind VN while IGF-I does so poorly, it was

hypothesised that this may be due to the presence of basic amino acids within the C-

domain of IGF-II, a motif that is not present in IGF-I (see Figure 1.1). As such, it

was thought that addition of basic residues to IGF-I, for example MBF, may increase

83

IGF-I binding to VN. However there was no binding of radiolabelled MBF to VN.

To examine if the iodination process affected the function of MBF, binding of

labelled MBF to IGFBP-3 and -5 was assessed. Both bound MBF, albeit with lower

affinity than wild-type IGF-I. The IGF-I binding site on IGFBP-3 and -5 involves

both the carboxy and amino regions, and within the carboxy site also lies the major

HBD site. However, through mutagenesis studies, mutations of the IGF-I carboxy-

binding site have been shown to have no effect on heparin binding (Song et al.,

2000; Shand et al., 2003). Following the unexpected MBF binding to VN results, it

is a possible conclusion that the proximal presence of another HBD that is tagged to

IGF-I in addition to VNs HBD, may cause steric hindrance due to the opposing

positive charges characteristic of an HBD sequence. Alternatively, as binding of

VN to MBF was measured in solution as opposed to surface binding, it is possible

VN may not be fully ‘dentaured’ and in a conformation recognisable to MBF.

Examination of VN ligands in the SPBA studies involved ‘pre-binding’ VN to the

well, where upon interaction with the polystyrene surface, VN denatures

(Underwood et al., 1993) exposing both its HBD and polyanionic region. Although

none of these explanations can be supported, the former rationalisation is unlikely as

the HBD added onto IGF-I does not contain Tyr residues and therefore would not

directly be affected by the iodination process.

Of particular interest, HBDs reside in the carboxy-terminal domain of IGFBP-2, -3, -

5 and –6 (see Figure 1.5). These regions are thought to double as cell-association

binding sites (Booth et al., 1995; Parker et al., 1996; Fowlkes et al., 1997; Firth et

al., 1998). Rees and Clemmons (1998) were able to demonstrate reduced cell-

associated IGFBP-5 using a synthetic peptide whose sequence was identical to that

of the basic residues in the C-terminal of IGFBP-5. In addition, substitution of

amino acids within this region has been shown to change the affinity of IGFBP-5

and -3 for heparin, although it was also noted that this altered the affinity for IGF-I

(Arai et al., 1996b). Putative heparin-binding motifs have also been reported within

the variable central region of IGFBP-3 and -5. Interestingly, other functions have

been associated with the heparin-binding region of these IGFBPs and in particular

include this region being a target site for various proteases (Arai et al., 1994a;

Campbell and Andress, 1997b; Campbell et al., 1998; Booth et al., 2002).

84

In summary, the results presented here suggest that the HBD of IGFBP-3 and -5 are

important in the IGF:IGFBP:VN interaction. More specifically, the region plays a

critical role in mediating IGF-I binding to VN. In contrast, the presence of an HBD

on or linked to the IGF molecule does not result in direct binding of IGF to VN,

suggesting that either IGF-II binds VN differently than do the IGFBPs, or that the

protein used in this study is unable to bind to VN due to interference from the new

protein conformation. In addition, it appears that glycosylation may also play a role

in the interaction, with less glycosylated IGFBP-3 binding IGF-I with greater

affinity than fully glycosylated IGFBP-3.

85

CHAPTER 5: FUNCTIONAL RESPONSES TO TRIMERIC

COMPLEXES

86

5.1 INTRODUCTION Cell migration is critical for normal development and wound healing. It is a

complex process that requires numerous interactions between the cells including

formation of cell junctions and focal adhesions, the presence of integrins and

cadherins, as well as cytoskeleton rearrangement (Bauer et al., 1992; Guvakova et

al., 2002). Any alteration from normal cell physiology can result in abnormal cell

states such as cancer, for example. If such disease states arise, increases in cell

migration can lead to metastasis.

A number of growth factors, including the IGFs, have been shown to stimulate cell

migration via a number of receptors (growth factor receptors and integrins) (Tysnes

et al., 1997; Clemmons et al., 1999; Mira et al., 1999; Podar et al., 2002).

Currently, much remains unknown regarding how activation of the IGF-1R can lead

to diverse cellular responses via the downstream signalling pathways such as the

PI3-kinase, MAP and protein kinase C signal transduction pathways (Kuemmerle

and Bushman, 1998; Meyer et al., 2001; Pozios et al., 2001; Hermanto et al., 2002).

In order for the IGFs to activate their receptor, the IGF-1R, the IGFBPs that exist in

molar excess in tissues and in the circulation must release them. To do this,

affinities of the IGFBPs for the IGFs are altered, most commonly via post-

translational proteolysis (see review by Clemmons, 1998). Although other post-

translational events, for example, glycosylation and phosphorylation, may modify

the affinity of IGFBPs for IGFs, few studies to date have shown significant changes

in the affinity of IGFBPs for IGFs as a result of these changes (Sakai et al., 2001;

Siddals et al., 2002; Schedlich et al., 2003).

More recent studies have suggested IGF-independent effects of the IGFBPs in

cellular responses such as DNA synthesis and cellular migration. These reports have

described both interactive and independent effects of IGF-I and IGFBP-5 on

migration in a number of cell lines (Abrass et al., 1997; Hsieh et al., 2003).

Important steps in the cascade of tissue remodelling and cell migration are the

interaction of IGFs with various ECM components, such as VN, fibronectin, laminin

and collagen. These processes involve adhesion of ECM components to their cell

surface receptors, namely VN interacting with its integrin receptors. Earlier studies

87

have shown that subsequent activation and co-ordination of IGF receptors with

integrin receptors is critical for cell migration in tumour models (Brooks et al., 1997;

Mira et al., 1999), as well as in normal cells (Jones et al., 1996; Kabir-Salmani et

al., 2003). Although it has recently been reported that VN is not up-regulated in

breast cancer, its distrubtion varies from that in normal tissue (Aaboe et al., 2003).

Yet the VN-integrin receptors have shown increased expression (Silvestri et al.,

2002). In particular, cell-surface expression of the αv subunits appears to correlate

with varying metastatic potential, and hence cell migration. Thus, both the VN

receptors are implicated: integrin αvβ3 is often present in the most aggressive breast

cancer cell lines (MDA-MB-231 and MDA-MB-435) while the αvβ5 is expressed in

the less metastatic MCF-7 cell line (Wong et al., 1998). Of pertinence to the studies

outlined here, the breast carcinoma cell line, MCF-7, has been shown to be a good

model to study migration and the effects of the IGF axis and the related urokinase

system (Doerr and Jones, 1996; Carriero et al., 1999). In addition to being

oestrogen receptor (ER) and αvβ5 positive, MCF-7 cells express IGFs and IGFBP-2,

-3, -4 and -5 (Yee et al., 1991; Adamo et al., 1992; Sheikh et al., 1992; McGuire et

al., 1994). Expression of IGFBP-3 and -5, however, has only been detected at the

mRNA level and protein level following exposure of the cells to IGF-I or retinoic

acid (Adamo et al., 1992; Sheikh et al., 1992). This cell line has been demonstrated

to be migration responsive to both IGF-I and VN, via involvement of both the IGF-

1R and the integrin αvβ5 (Carriero et al., 1997; Bartucci et al., 2001; Guvakova et

al., 2002). Hence MCF-7 cells were selected as the primary model of choice to

examine effects of IGFs, IGFBPs and VN in the studies reported here. In some early

experiments CHO-K1 cells were used to measure effects on cell proliferation, but as

these cells are primarily used for industrial cell culture, and are generally grown in

suspension and produce little ECM, it was decided that they were not an ideal

biological model to study IGF:IGFBP:VN complexes. Specifically, the studies

described in this chapter were directed at exploring the mechanism of IGFBP

modulation of IGF-I-stimulated migration in the presence of VN. Using the MCF-7

cell line, combinations of the proteins were examined for their effects on cell

migration. In particular, the IGF, IGFBP and VN complexes described in chapters 3

and 4 were examined.

88

5.2 EXPERIMENTAL PROCEDURES Descriptions and suppliers of proteins used in this chapter can be found in Section

2.2 and Table 4.1. The culturing of cells, protein synthesis and Transwell migration

assays used in this chapter are described in full in Sections 2.8, 2.9 and 2.10.

Briefly, the protein synthesis assays involved seeding 50,000 cells that had been

serum starved for 4 hr into 24-well culture dishes in 1 mL SFM containing [4,5-3H]-

leucine and incubating these cells at 37ºC/5% CO2 for 40 hr. After washing and

trichloroacetic-acid precipitation of protein in the cell monolayer, incorporation of

[3H]-leucine into de novo synthesized protein was determined by subsampling

solubilized protein precipitate for β-scintillation counting.

For the migration assays, 200,000 cells that had been serum starved by incubation in

SFM for 4 hr were seeded into the upper chamber of a 12.0 µm pore Transwell.

After 5 hr incubation, at which time cell proliferation is minimal, cells that had

migrated through the porous membrane were fixed and stained with crystal violet.

The number of cells attached was estimated by extracting the crystal violet stain in

10% acetic acid and measuring the absorbance of these extracts via

spectrophotometry.

5.3 RESULTS To examine whether the structural findings described in chapters 3 and 4 resulted in

altered functional responses, protein synthesis assays with CHO-K1 cells and

Transwell migration assays with MCF-7 cells were employed. Significant

differences for all migration experiments were analysed using the Student’s paired t-

test.

5.3.1 Effects of IGFs and IGFBPs in the presence of VN on protein synthesis

in CHO-K1 cells

In order to determine if the enhanced binding of IGF-I to VN in the presence of

IGFBP-3 and –5 had functional consequences, the effects of the complexes on

stimulating protein synthesis were examined in the CHO-K1 cell line (Figure 5.1).

89

Figure 5.1 Effects of IGFs and IGFBPs on stimulating de novo protein

synthesis in CHO-K1 cells measured by [4,5-3H]-leucine incorporation. Fifty

thousand cells were seeded into wells pre-coated with or without VN (300 ng) and

IGFs (30 or 300 ng), IGFBP-3 (300 ng) or IGFBP-5 (300 ng), added alone or in

combination as indicated. Data are expressed as percentage stimulation of protein

synthesis compared with that obtained with control wells (VN alone). Values shown

are the mean ± SEM of duplicate wells from 3 experiments. ** p < 0.01 indicate

significant differences to the control (VN alone) while ª p < 0.01 indicates

differences between IGFBP treatments without IGF-I to those with IGF-I, and # p <

0.01 indicates significant differences between 30 ng and 300 ng treatments.

0 30 300 0 30 300 30 300

300

250

200

150

100

50

0

% O

F C

ON

TRO

L (+

VN

)

IGFBP-5 (300 ng) IGFBP-3 (300 ng)

IGF-II IGF-I IGF-I

TREATMENT (ng)

- VN+ VN

/ +VN _

**

**

**ª **ª

**ª **ª**#

** ** ** ****

****

** **

90

In the absence of VN, there was minimal protein synthesis regardless of treatment,

with responses less than 40% of the ‘plus VN’ control. However, both IGF-I:IGFBP

and IGF-II enhanced protein synthesis in the presence of VN, with IGF-II

stimulating similar responses to that found with the addition of IGFBP-3 and –5 to

IGF-I and VN. IGFBP-5 and IGFBP-3 alone stimulated protein synthesis of 101.6 ±

7.2% and 125.3 ± 4.4%, respectively, compared to the control. For both IGFBP-3

and -5, the addition of IGF-I dose-dependently increased protein synthesis. At 30 ng

IGF-I, responses were 221.9 ± 7.9% and increased to 250.2 ± 15.4% in the presence

of IGFBP-5 and VN. Similar values were observed for IGFBP-3 and VN with 219.9

± 14.5% and 234.5 ± 27.7% for 30 ng and 300 ng respectively. Again, IGF-II gave

similar results at 30 ng (158 ± 10.4%) and 300 ng (236.7 ± 11%). The lack of a

significant increase in protein synthesis between 30 ng and 300 ng suggests that the

amount of IGF-I added has almost reached saturation. In contrast, with IGF-II

significant difference (p < 0.01) was detected.

For the subsequent studies reported in this chapter the biological responses to

IGF:IGFBP:VN complexes were examined in MCF-7 cells. The CHO-K1 assays

are clearly highly dependent on VN to mediate cell attachment and as such, it was

felt that these cells were not the most ideal model for examining the functional

effects of the complexes. The MCF-7 cells on the other hand, could attach in the

absence of serum, this presumably being a reflection of integrin expression and

ECM production by the cells. Therefore, it was felt the MCF-7 cells would provide

a more convincing analysis of cell function in the presence of a heterotrimeric

complex. Thus MCF-7 cells were used for the subsequent experiments.

5.3.2 Migration of MCF-7 cells following exposure to increasing amounts of

IGF-I in the presence of VN alone or in combination with IGFBP-5

Using the MCF-7 cell line, a preliminary experiment examined whether the trimeric

complex comprised of IGF-I:IGFBP-5:VN could stimulate cell migration. One

thousand ng of VN and 1000 ng IGFBP-5 in the presence of an increasing amount of

IGF-I was assessed for its ability to stimulate migration (Figure 5.2). The amount of

91

Figure 5.2 Dose response curve of increasing amounts of IGF-I in the presence

of VN alone or in combination with IGFBP-5. Two hundred thousand MCF-7

cells were seeded into Transwells in which the lower wells had been coated with VN

and increasing amounts of IGF-I ± 1000 ng IGFBP-5 and allowed to migrate

through the porous membrane for 5 hr as described in Section 2.10. Cells traversing

the membrane were stained and this was measured as an absorbance. The data are

expressed as a percentage of cells that migrated on VN alone (100%). Values shown

are the mean ± SEM of duplicate wells from 2 experiments. * p < 0.05 compared to

the control (VN alone).

Figure provided by Chris Towne as used in Kricker et al. (2003)

% A

BO

VE C

ON

TRO

L

40

60

80

20

0

100

-20 IGFBP-5

alone IGF-I (ng)

** *VN + I

VN + I + BP5

1 3 10 30 100 +VN

92

VN used per well was determined by proportionally increasing the amount used in

24-well plate experiments according to the increase in surface area. IGFBP-5 alone

was inhibitory compared to the VN alone control, as were 1 and 3 ng treatments of

IGF-I alone. At the higher amounts of IGF-I tested (10, 30 and 100 ng), there was a

dose-dependent response in IGF-I-stimulated cell migration. These IGF-I

treatments, however, did not stimulate migration as greatly as the trimeric complex

of IGF-I:IGFBP-5:VN at any amount of IGF-I added. In fact, as the amount of IGF-

I increased, there is a slight decrease in migration. The only treatments significantly

different to the control (VN alone) were the trimeric complex with 1, 3 and 10 ng of

IGF-I.

5.3.3 Migration of MCF-7 cells following exposure to increasing amounts of

IGFBP-5 in the presence of VN alone or in combination with IGF-I

In order to determine the optimum amount of IGFBP-5 to use in future assays,

increasing amounts of IGFBP-5 were added to 10 ng IGF-I and 1000 ng VN (Figure

5.3). Increasing the added amount of IGFBP-5 alone to VN did not significantly

increase migration above the control. However, as seen in Figure 5.2, IGFBP-5 in

the presence of 10 ng IGF-I resulted in considerable changes between the treatments

and the control. In particular, the addition of 100 ng and 300 ng IGFBP-5 led to

significant responses and stimulated migration of 174.9 ± 10.7% (p < 0.01) and

153.2 ± 21.5% (p < 0.05) respectively, in contrast to 1000 ng (146.5 ± 36.3%) which

did not result in responses that were significantly difference to the control. Overall,

there was a minor downward trend correlating with the increase in IGFBP. This

result, taken together with the previous data, resulted in the use of 1000 ng VN, 10

ng IGF-I and 100 ng IGFBP in future assays.

5.3.4 Binary complexes of VN and IGFBP-5, IGFBP-5 mutants or IGF-I do

not stimulate MCF-7 cell migration

In order to further explore the earlier results demonstrating that both IGFBP-5 and

IGF-I are needed for maximal MCF-7 cell migration, the effect of VN on the growth

factor and binding protein alone was determined (Figure 5.4). All results presented

in this chapter are expressed as a ‘percentage of’ or ‘percentage above’ VN as in the

93

Figure 5.3 Dose response curve of increasing concentrations of IGFBP-5 in the

presence of VN alone or in combination with IGF-I. The lower chambers of the

Transwells were coated with increasing amounts of IGFBP-5 in combination with

VN (1 μg) ± IGF-I (10 ng). Data are expressed as percentage of migration observed

in the control, VN alone, which is defined as 100%. Values shown are the mean ±

SEM of duplicate wells from 2 experiments. * p < 0.05, ** p < 0.01 compared to

the control.

200

175

150

125

100

75

50

25

0 0 100 300 1000

IGFBP-5 (ng) -VN

VN VN + I

% O

F C

ON

TRO

L (+

VN

)

**

**

*

94

Figure 5.4 Effect of IGFBP-5 and IGF-I on MCF-7 cell migration with or

without VN. A representative experiment demonstrating the lack of migration

observed in the absence of VN. Cells were exposed to 100 ng IGFBP-5, IGFBP-5

m1 (R201A/K202N/K206N/K208N) or IGFBP-5 m2 (K202A/K206A/R207A) or 10

ng IGF-I (I), with or without VN (1 µg). Values shown are the mean ± SEM of

duplicate wells of one experiment for ‘minus VN’ treatments, while ‘plus VN’

treatments are of duplicate wells from 3 experiments. ** p < 0.01 of treatments with

VN compared to the those without VN.

10 ng

+ VN - VN

0 I BP5 m1 m2

** ****

**

150

125

100

75

50

25

0

**

% O

F C

ON

TRO

L (+

VN

)

100 ng

95

absence of VN, few cells migrate, presumably due to the lack of cell attachment

sites in the absence of VN. Due to the consistent finding that migration is minimal

in the ‘minus VN’ results, the sample number for all ‘minus VN’ data is from only 2

replicate wells. Figure 5.4 also confirms that none of wild-type IGFBP-5, IGFBP-5

HBD mutants or IGF-I alone have much effect on cell migration at the amounts

used, even when VN is present. The cell migration response observed in the

absence of any treatment was 40 ± 4.9%, which showed little variation when cells

were exposed to IGF-I or IGFBP-5 preparations in the absence of VN. These

responses ranged from 30% to 42% of the control value of VN alone (100 ± 2.0%).

When VN was present, IGFBP-5 preparations were still ineffective at stimulating

cell migration above the control with wild-type IGFBP-5 and the HBD mutants of

IGFBP-5 (m1 and m2) (desribed in detail in the next section) resulting in responses

of 96.0 ± 18.9%, 85.9 ± 10.8% and 93.1 ± 14.9%, respectively. IGF-I alone with

VN resulted in migration (118.6 ± 4.8%, p < 0.05) slightly above control values.

Together, this data suggests that presence of VN is critical for cell migration to

occur regardless of the treatment.

5.3.5 Migration of MCF-7 cells in response to IGF-I or des(1-3)IGF-I

pre-bound to VN in the presence of intact IGFBP-5 or HBD mutant

IGFBP-5

The HBD term refers to motifs in proteins that bind heparin. However, amino acids

in the HBD of IGFBP-5 have also been shown to be involved in IGF-I binding

(Parker et al., 1998; Bramani et al., 1999). Although both IGFBP-5 HBD mutants

used in these studies were previously characterised by Arai et al. (1996b) and shown

to have an ability to bind IGF-I similar to wild-type IGFBP-5, the results in Section

4.3.8 indicate that binding of [125I]-IGF-I to both HBD mutants was reduced.

Taking into account this conflicting data in IGF-I affinity, experiments addressing

the following questions were pursued: 1) did the increase in cell migration following

addition of IGF-I and IGFBP-5 involve the specific formation of a ternary IGF-

I:IGFBP-5:VN complex? ; 2) does the affinity of IGFBP-5 for IGF-I effect

migration responses? ; and 3) is the presence of the HBD critical for cell migration?

96

Migration responses of MCF-7 cells to native IGF-I were compared with des(1-

3)IGF-I, an analogue that has reduced affinity for IGFBPs while retaining its ability

to activate the IGF-I receptor (Francis et al., 1993). In the absence of IGFBP-5,

assays with either 10 ng of native IGF-I or des(1-3)IGF-I with VN resulted in

migration of 118.6 ± 4.8% and 111.2 ± 6.5% of the control (VN), respectively

(Figure 5.5). However, migration of cells in the Transwells when native IGF-I was

tested in the presence of IGFBP-5 (166.8 ± 17.3 %, p < 0.01) were significantly

different to the control (VN alone: 100 ± 2.0%). Responses observed with des(1-

3)IGF-I in the presence of IGFBP-5 remained unchanged (122.1 ± 9.9 %).

Previous studies have also shown that the HBD region is important for several

extracellular protein interactions (Booth et al., 1995; Campbell and Andress, 1997a)

hence, the role of the HBD in IGFBP-5 in formation of the trimeric complex and

subsequent effect on cell migration was investigated. The mutants, designated as

IGFBP-5 HBDm1 (R201A/K202N/K206N/K208N) and IGFBP-5 HBDm2 (K202A/

K206A/R207A) have been examined in several studies (Arai et al., 1996b; Parker et

al., 1996; Parker et al., 1998; Hsieh et al., 2003). The studies reported here indicate

that the HBD mutation in IGFBP-5 m1 had no effect on the ability of IGFBP-5 +

IGF-I + VN to stimulate cell migration (181.8 ± 16.6%, p < 0.01), this result being

similar to complexes containing wild-type IGFBP-5. However, the second mutant,

IGFBP-5 m2, had reduced ability to stimulate migration when compared with IGF-I

+ VN and was not significantly different to the ‘plus VN’ control (129.4 ± 14.5%).

With both mutants, the substitution of des(1-3)IGF-I for IGF-I in the complexes

resulted in a much reduced ability to stimulate migration, again indicating that the

binding of the IGF to IGFBP is critical for complex-stimulated migration.

5.3.6 Effect of heparin on migration of MCF-7 cells in response to IGF-I pre-

bound to VN in the presence of IGFBP-5

Continuing examination of the role of the HBD, 1 µg heparin (Sigma: heparin salt of

~13 kDa) was added to either VN ± IGFBP or VN:IGFBP-5:IGF-I. Firstly, the

effect of heparin together with VN was found to have no effect on MCF-7 cell

97

Figure 5.5 Migration of MCF-7 cells in response to IGF-I or des(1-3)IGF-I pre-

bound to VN in the presence of intact IGFBP-5 or HBD mutant IGFBP-5. In

the presence of VN (1 μg), the ability of IGFBP-5/IGFBP-5 mutants (100 ng) and

IGF-I (10 ng) or des(1-3)IGF-I (10 ng), alone or in combination, to stimulate MCF-7

cell migration was assessed. IGFBP-5 mutants are designated as m1 (R201A/

K202N/K206N/K208N) and m2 (K202A/K206A/R207A). Values are the means ±

SEM of duplicate samples from 2 separate experiments. * p < 0.05 and ** p < 0.01

indicate significant differences from the control (VN alone) while ª p < 0.05

indicates differences of treatment containing des(1-3)IGF-I compared to IGF-I

treatment, and # p < 0.01 indicates significant differences with the addition of

IGFBP-5 to either IGF-I or des(1-3)IGF-I.

250

200

150

100

50

0 -VN +VN I / des I BP5 m1 m2 alone

% O

F C

ON

TRO

L (+

VN

) VN + I VN + des I

**

* ª

**#

100 ng

**#

* ª**

98

migration when examined alone or when added to any form of IGFBP-5 (Figure

5.6). The effect of IGF-I:IGFBP-5:VN complexes on cell migration in the presence

of heparin were then examined for their ability to stimulate migration (Figure 5.6,

response values summarised in Table 5.1). Again, the addition of heparin had no

effect on any complex with any of the three IGFBP-5 proteins examined. Although

the addition of heparin slightly increased the cell migration response obtained with

wild-type IGFBP-5 and BP5m2, these differences were not statistically significant.

5.3.7 Effects of IGFs and an IGF analogue, MBF, on MCF-7 cell migration

The effects of IGF-I and IGF-II, as well as the IGF analogue, Matrix Binding Factor

(MBF, Patent WO9954359A1: an IGF-I analogue with an HBD from basic

fibroblast growth factor tagged at the carboxy terminus), at two amounts (10 ng or

100 ng) in the presence of VN (1 μg), were compared in their ability to stimulate

MCF-7 cells to migrate (Figure 5.7). As demonstrated with the ‘minus VN’ control

(18.4 ± 4.5%), there is minimal MCF-7 cell migration when VN is absent compared

to the control of ‘plus VN’ (100 ± 3.0%). In the absence of VN, regardless of the

treatment, similar results were observed to the ‘minus VN’ control (data not shown,

see Figures 5.4, 5.9 and 5.12). Each IGF was able to stimulate migration above the

control, with maximal migration observed at the highest amount tested (100 ng).

Treatments of IGF-I, IGF-II and MBF at both 10 and 100 ng were significantly

different to the ‘plus VN’ control (p < 0.01). The increase in amount of growth

factor from 10 ng to 100 ng resulted in a 1.7-fold, 1.4-fold and 1.3-fold increase in

MCF-7 cell migration for IGF-I (p < 0.01), IGF-II (p < 0.05) and MBF (p < 0.05),

respectively. No differences were detected between the 3 growth factors when

compared to each other at either 10 or 100 ng. The stimulatory effect of IGF-I and

MBF, in the absence of IGFBPs, was somewhat surprising given the demonstrated

inability of both IGF-I and MBF to bind effectively to VN in the structural studies

(Figures 3.3 and 4.11).

99

Figure 5.6 Effect of heparin on IGFBP-5 trimeric complex-stimulated

migration. MCF-7 cell migration was measured in the presence of VN (1 μg),

IGFBP-5/IGFBP-5 mutants (100 ng), IGF-I (10 ng) and/or heparin (1 µg) alone or in

combination. Values are the means ± SEM of duplicate samples from 2 separate

experiments. * p < 0.05 and ** p < 0.01 indicate treatments that were significantly

different from the control (VN alone). No significant differences in migration were

observed when heparin was added to the IGFBP-5 treatments.

Table 5.1 Comparison of migration results: IGF-I:IGFBP-5:VN with or

without heparin. Results are expressed as a percentage of the control (VN alone).

IGFBP-5 Trimeric complex Trimeric complex +

heparin

wt BP5 166.8 ± 17.3 200.7 ± 19

BP5 m1 181.8 ± 16.6 176 ± 38.1

BP5 m2 129.4 ± 14.5 159.9 ± 6.2

0 BP5 m1 m2

250

200

150

100

50

0

% O

F C

ON

TRO

L (+

VN

) VN + I VN + Hep VN + I + Hep

** *

****

**

100 ng

*

100

Figure 5.7 Effects of IGFs and MBF on MCF-7 cell migration. Treatments of

IGF-I and IGF-II and an IGF analogue (MBF) at two amounts (10 ng or 100 ng) in

the presence of VN (1μg) were compared for their ability to stimulate MCF-7 cell

migration. Values shown are the mean ± SEM of duplicate wells from 3

experiments. Significant differences to the control (VN alone) are indicated by * p <

0.05 and ** p < 0.01, while ª p < 0.05 indicates differences of treatment using 100

ng to same treatment containing 10 ng. No significant differences were detected

between the 3 growth factors (IGF-I, IGF-II or MBF) at either 10 ng or 100 ng.

- VN + VN I II MBF

% O

F C

ON

TRO

L (+

VN

) VN + GF 10 ng VN + GF 100 ng

**

**

**ª

****

**ª **ª250

200

150

100

50

0

101

5.3.8 MCF-7 cell migration responses to MBF in the presence of IGFBPs with

or without heparin

In the presence of VN (1 μg) and MBF (10 and 100 ng), the ability of IGFBP-5,

IGFBP-5 mutant, IGFBP-3 and an IGFBP-3 HBD mutant (100 ng), alone or in

combination with heparin (1 μg), to stimulate MCF-7 cell migration was assessed

(Figure 5.8). All treatments containing either 10 ng or 100 ng MBF resulted in

migration that was significantly different to the control (VN alone) (see Figure 5.8

for p values) whether IGFBPs were present or not. Although cell migration

responses with treatments containing 100 ng MBF were consistently higher than

those with 10 ng MBF, differences were only detected in 3 treatments: MBF +

IGFBP-5 + VN (161.8 ± 11.8% vs 207.6 ± 8.9%, p < 0.05), MBF + IGFBP-3

HBDm + VN (151.4 ± 12.6% vs 197.5 ± 14.5%, p < 0.05) and MBF + IGFBP-3 +

VN (139.0 ± 10.9% vs 209.8 ± 11.9%, p < 0.01). Surprisingly, the addition of

heparin to any of the treatments had no effect on reducing the stimulated MCF-7 cell

migration, other than the treatment of MBF alone.

5.3.9 Effect of VN on MCF-7 cell migration in response to IGFBP-4

Earlier structural studies in Figure 3.3 indicated that IGFBP-4 could enhance IGF-I

binding to VN. To determine whether this had a functional effect, complexes

comprised of various combinations of IGF-I, IGFBP-4 and VN were pre-coated to

the lower chambers of the Transwell plates. As seen in Figure 5.9, in the absence of

VN, minimal cell migration is observed regardless of the treatment when compared

to the ‘plus VN’ control (100 ± 4.3 %). The addition of IGF-I or IGFBP-4 to VN

resulted in migration similar to that of the ‘plus VN’ control, 118.6 ± 4.8 % (p <

0.05) and 104.9 ± 13.2 % respectively. However, migration with the combination of

all three proteins, VN:IGF-I:IGFBP-4 gave significantly higher responses (p < 0.01,

244.3 ± 35%) compared to the control value.

102

Figure 5.8 MCF-7 cell migration responses to MBF in the presence of IGFBPs

with or without heparin. In the presence of VN (1 μg) and MBF (10 and 100 ng),

the ability of IGFBP-5 and the IGFBP-5 m2 mutant (K202A/K206A/R207A)

(Figure A), and IGFBP-3 and IGFBP-3 HBD mutant (100 ng) (Figure B), alone or in

combination with heparin (1 μg), were assessed for their ability to stimulate MCF-7

cell migration. Values shown are the mean ± SEM of duplicate wells from 2

experiments. Except for MBF 10 ng + heparin, all treatments were significantly

different (p < 0.01; MBF 10 ng + BP5m2: p < 0.05) to the control. No significant

differences were observed when treatments of VN + MBF + IGFBP were compared

to those same treatments when heparin was added, nor were there any differences

detected between treatments when 10 ng was compared to 100 ng, except for VN +

MBF + BP5 (* p < 0.05) and VN + MBF + BP3 (** p < 0.01).

250

200

150

100

50

0 MBF Hep BP5 BP5+ BP5m2 BP5m2+ alone Hep Hep

% O

F C

ON

TRO

L (+

VN

)

250

200

150

100

50

0 MBF alone BP3 BP3+ Hep BP3 HBDm

*

**

A: IGFBP-5

A: IGFBP-3

% O

F C

ON

TRO

L (+

VN

)

VN + MBF 10 VN + MBF 100

VN + MBF 10 + Hep VN + MBF 100 + Hep

103

Figure 5.9 Effect of VN on MCF-7 cell migration in response to IGFBP-4

complexes. In the presence of VN (1 µg), the ability of IGF-I (10 ng), IGFBP-4

(100ng), alone or in combination, to stimulate MCF-7 cell migration was assessed.

The number of cells traversing the membrane in the presence of the treatments was

then expressed as a percentage of those that migrated on VN alone. Values are

means of duplicate samples ± SEM (n = 2 for ‘minus VN’, n = 6 for ‘plus VN’).

* p < 0.05, ** p < 0.01 compared to the control (VN alone).

300

250

200

150

100

50

0 I BP4 BP4 + I

+ VN - VN

% O

F C

ON

TRO

L (+

VN

)

**

**** **

+ VN VN _

*

104

5.3.10 Comparison of MCF-7 migration responses to VN alone or in the

presence of IGF-I or des(1-3)IGF-I with IGFBP-4

To test whether the increase in cell migration following addition of IGF-I and

IGFBP-4 to VN involved the formation of a ternary IGF-I:IGFBP-4:VN complex,

responses were compared between native IGF-I and the des(1-3)IGF-I analogue that

has reduced affinity for IGFBPs while retaining its ability to activate the IGF-I

receptor (Francis et al., 1993). In the absence of IGFBP-4, assays with 10 ng of

native IGF-I or des(1-3)IGF-I in the presence of VN resulted in migration on VN of

118.6 ± 4.8 % and 111.2 ± 6.5 % respectively (Figure 5.10). However, only the

migration of cells in the Transwells with native IGF-I + VN treatment was

significantly increased (p < 0.01) by the addition of IGFBP-4 + VN (see above in

Section 5.3.9). Responses observed with the des(1-3)IGF-I in the presence of

IGFBP-4 remained unchanged (111.8 ± 17.7 % compared to VN).

5.3.11 Effect of heparin on MCF-7 migration in response to IGFBP-4 trimeric

complexes

It has been suggested that glycosaminoglycans may shift the affinities of the IGFBPs

for the IGFs (Arai et al., 1994b). More specifically, this has been demonstrated with

IGFBP-3 and -5, but not IGFBP-1 or -4 as they lack the conserved HBD found in the

other two binding proteins. To test the hypothesis that heparin should not interfere

with the VN:IGF-I:IGFBP-4 complex, heparin (1 µg) was added to various

combinations of the above complex (Figure 5.11). Interestingly, when heparin was

added to IGFBP-4 in the presence of VN, it was able to significantly increase the

migration response compared to the control (VN alone) in the absence of IGF-I.

This was not due to the presence of the individual complex components as the

heparin:VN treatment response (84.0 ± 4.9%) was similar to control levels (100 ±

4.3%), as was IGFBP-4:VN (104.9 ± 13.2%). Indeed, addition of heparin to the

treatment IGF-I:IGFBP-4:VN reduced migration to some extent (244.3 ± 35.0% vs

166.7 ± 13.1 %), however, no significant difference was detected when the two

treatments were compared (p = 0.15).

105

Figure 5.10 Comparison of MCF-7 migration responses to VN alone or in the

presence of IGF-I or des(1-3)IGF-I. In the presence of VN (1 μg), the ability of

IGFBP-4 (100 ng), IGF-I (10 ng) and des(1-3)IGF-I (10 ng), alone or in

combination, to stimulate MCF-7 cell migration was assessed. Values shown are the

mean ± SEM of duplicate wells from 2 experiments. ** p < 0.01 indicates

significant difference compared to the control (VN alone), while # p < 0.05 indicates

significant differences of treatment with IGFBP-4 to the same treatment without

IGFBP-4. ª p < 0.05 indicates a significant difference between the treatment

containing des(1-3)IGF-I compared to IGF-I treatment.

+/ - 10 ng IGF-I / des(1-3)IGF-I

- VN 0 BP4

300

250

200

150

100

50

0

VN + I VN

VN + des I

**

**#

ª

% O

F C

ON

TRO

L (+

VN

)

106

Figure 5.11 Effect of heparin on migration of MCF-7 cells in response to

IGFBP-4 trimeric complexes. The presence of VN (1 μg), IGFBP-4 (100 ng),

IGF-I (10 ng) and/or heparin (1 µg), alone or in combination, to stimulate MCF-7

cell migration was assessed. Values shown are the mean ± SEM of duplicate wells

from 2 experiments. ** p < 0.01 indicate significant differences to the control (VN

alone) while differences between treatments without IGFBP-4 compared to the same

treatments with IGFBP-4 are indicated by # p < 0.05. No differences in migration

were observed when heparin was added to the IGF-I:IGFBP-4:VN treatment.

300

250

200

150

100

50

0

% O

F C

ON

TRO

L (+

VN

)

0 IGFBP-4

VN + IVN + Hep VN + I + Hep

**#

**# **

107

5.3.12 MCF-7 cell migration responses to IGFBP-3 preparations with or

without VN

As observed with IGFBP-4, minimal cell migration was observed in the absence of

VN. In the presence of VN, IGFBP-3 from three sources including glycosylated

IGFBP-3 and IGFBP-3 HBD mutants were compared to the control and the

treatment VN + IGF-I (Figure 5.12). Full descriptions of the IGFBP-3 preparations

are listed in Table 4.1 Briefly, the proteins used are mammalian-produced

glycosylated IGFBP-3 (gly BP3), baculovirus-expressed IGFBP-3 (BV.BP3), E.coli-

produced non-glycosylated IGFBP-3 (N109D), mammalian-produced mutant non-

glycosylated IGFBP-3 (non-gly), mammalian-produced HBD mutant IGFBP-3

(HBDm) and baculovirus-expressed HBD mutant IGFBP-3 (BV.HBDm). Each

IGFBP-3 treatment exposed to VN resulted in migration responses that were similar

to the control of VN alone (100 ± 3.8%).

5.3.13 Comparison of MCF-7 migration responses to VN alone or in the

presence of IGF-I or des(1-3)IGF-I with IGFBP-3 or HBD mutant

IGFBP-3

In order to determine if the enhanced binding of IGF-I to VN in the presence of

IGFBP-3 has functional consequences, the effects of the complexes on stimulating

cells to migrate were examined using the MCF-7 cell line. In addition, to determine

if the findings from Chapter 4, specifically, if mutation of HBD in IGFBP-3 and if

differential glycosylation of IGFBP-3 have a physiological effect, cell migration

experiments were undertaken comparing wild-type IGFBP-3 with the various forms

of IGFBP-3. The presence of 10 ng IGF-I exposed to 100 ng IGFBP-3 and 1 μg VN

resulted in significant increases (p < 0.01) above control (VN alone) for each

IGFBP-3 preparation tested (Figure 5.13). In Figure 5.13A, trimeric complexes

comprised of either glycosylated IGFBP-3 or glycosylated IGFBP-3 HBD mutant

were compared. Neither IGF-I or des(1-3)IGF-I were significantly different from

the ‘plus VN’ control. The addition of the IGFBP-3 preparations to IGF-I and VN

resulted in responses of 252.8 ± 39.4% and 360.2 ± 49.6% of the control, for wild-

type IGFBP-3 and the HBD mutant respectively. When des(1-3)IGF-I was

substituted for native IGF-I, enhanced migration was abolished and the results were

similar to the VN alone control.

108

Figure 5.12 MCF-7 cell migration in response to IGFBP-3 with or without VN.

MCF-7 cell migration was measured in response to treatments of either VN (1 µg),

IGF-I (10 ng), IGFBP-3 or IGFBP-3 mutants (100 ng). The number of cells

traversing the membrane in the presence of these treatments was then expressed as a

percentage of those that migrated on VN alone. Values are means of duplicate

samples ± SEM from 1 experiment for ‘minus VN’ treatments and 2 experiments for

‘plus VN’ treatments. * p < 0.05, ** p < 0.01 indicate responses that are

significantly different to the control (VN alone).

150

125

100

75

50

25

0

% O

F C

ON

TRO

L (+

VN

) + VN - VN

*

****

*

** **

0 I gly BP3 BV.BP3 N109D non-gly HBDm BV.HBDm

***

100 ng 10 ng

109

Figure 5.13 Comparison of MCF-7 cell migration responses to VN alone or in the presence of IGF-I or des(1-3)IGF-I with IGFBP-3 or mutant IGFBP-3. A) In the presence of VN (1μg), the ability of IGFBP-3 or IGFBP-3 HBD mutant (100 ng), IGF-I (10 ng) and des(1-3)IGF-I (10 ng), alone or in combination, to stimulate MCF-7 cell migration was assessed. Values shown are the mean ± SEM of duplicate wells from 2 experiments. * p < 0.05 and ** p < 0.01 indicate significant differences to the control (VN alone) while ª p < 0.05 indicates significant differences of treatments containing des(1-3)IGF-I compared to those containing IGF-I. There were no differences observed between the wild-type IGFBP-3 and the HBD mutant. B) Migration was used to functionally compare IGFBP-3 glycosylation mutants from different expression systems to wild-type IGFBP-3 expressed in mammalian cells (BP3) in the presence of VN and IGF-I. Migration that was significantly enhanced above the control (VN alone) is denoted by * p < 0.05 and ** p < 0.01. Significant differences of other IGFBP-3 preparations to wild-type ‘gly BP3’ are indicated as ‘#’ (p < 0.05).

0-VN 0+VN I gly BP3 HBDm

450 400 350 300 250 200 150 100

50 0

% O

F C

ON

TRO

L (+

VN

) VN + I VN + des I

100 ng 10 ng

**

**

ªª

**

% O

F C

ON

TRO

L (+

VN

)

IGF-I gly BP3 N109D non-gly BV.BP3 BV.HBDm

600

500

400

300

200

100

0

100 ng 10 ng

**

**#

A

B VN + I

*

**

**# **#

110

In Figure 5.13B, other preparations of IGFBP-3 (see Table 4.1 for details) were

compared to the wild-type glycosylated IGFBP-3. Each IGFBP-3 included as part

of a trimeric complex resulted in migration greater than 250% with the N109D

mutant (E.coli non-glycosylated) giving the greatest migration (519.8 ± 14.1%).

Nevertheless, there is a noticeable trend across the IGFBP-3 treatments: the less

glycosylated the binding protein, the greater the enhancement of cell migration. An

exception to this is mammalian cell–produced mutant non-glycosylated IGFBP-3

(non-gly). Interestingly, the response observed for this non-glycosylated mutant was

significantly different (p < 0.01) to the other non-glycosylated protein, N109D

which was expressed in E.coli.

5.3.14 Effect of heparin on MCF-7 migration in response to IGFBP-3 trimeric

complexes

In contrast to IGFBP-4, IGFBP-3 contains an HBD, thus as for IGFBP-5, it is

hypothesised that heparin may disrupt the enhanced migratory effect stimulated by

the trimeric complexes. To test this hypothesis, IGFBP-3 (100 ng) in combination

with IGF-I (10 ng) and VN (1 µg), was compared to the combination containing the

IGFBP-3 HBD mutant, with or without heparin (1 µg) (Figure 5.14). As shown in

the previous figures (5.13A and 5.13B), both the IGFBP-3 and IGFBP-3 HBD

mutant in complex with VN and IGF-I stimulate enhanced MCF-7 cell migration

when compared to the control (VN alone). As shown in Figure 5.14, the presence of

heparin with either IGFBP-3 preparation plus VN resulted in migration responses

similar to those found with the control (BP3, 106.0 ± 13.4%; HBDm BP3, 115.3 ±

16.9%). When heparin was added to IGFBP-3 trimeric complexes (IGF-I:IGFBP-

3:VN), migration remained significantly increased when compared to the control

(p < 0.01). However, when these responses were compared to those IGFBP-3

treatments without heparin, only the IGFBP-3 HBD mutant resulted in significantly

different responses (p < 0.05). Despite a reduction in migration obtained for IGFBP-

3 in the presence of heparin (252.8 ± 39.4% vs 180.2 ± 20.8%), migration with wild-

type IGFBP-3 was not significantly reduced in contrast to the IGFBP-3 HBD mutant

(360.2 ± 49.6% vs 187.9 ± 20.0%).

111

Figure 5.14 Effect of heparin on MCF-7 cell migration in response to wild-type

IGFBP-3 and HBD mutant IGFBP-3. Migration was measured in the presence of

VN (1 μg), IGFBP-3 (100 ng), IGF-I (10 ng) and/or heparin (1 µg) alone or in

combination. Values shown are the mean ± SEM of duplicate wells from 3

experiments. ** p < 0.01 indicates significant differences to the control (VN alone)

while the addition of IGFBP-3 was significantly different to treatments with IGFBP-

3 as indicated by #. The addition of heparin only resulted in a significant difference

with the HBD mutant when compared to the same treatment without heparin, as

denoted by ‘a’.

0 gly BP3 HBDm

450 400 350 300 250 200 150 100

50 0

VN + I VN + Hep VN + I+ Hep

** **ª#

% O

F C

ON

TRO

L (+

VN

)

** #

**#

112

5.3.15 Scanning electron micrographs of MCF-7 cells in Transwell migration

assay in response to IGF-I:IGFBP-5:VN complex combinations

In order to assess the relevance of cell migration through the 12 µm pore inserts, as

well as morphological differences in cells exposed to different treatments, the

migrated cells were assessed using scanning electron microscopy. MCF-7 cells on

microporous inserts were fixed, gold-coated and viewed using an FEI Quanta200

Environmental scanning electron microscope. Treatments examined included no

treatment (no VN), VN alone, IGF-I – VN, IGF-I + VN, IGFBP-5 – VN, IGFBP-5 +

VN and IGF-I + IGFBP-5 + VN. Each sample was photographed at 400x, 800x and

1600x magnification (Figure 5.15).

As found in the Transwell migration assay results measured by absorbance

spectrophotometry reported in earlier sections of this chapter, there was no

migration of cells from the upper surface of the insert to the bottom surface when

VN was absent. At the low magnification (400x) cells were randomly distributed

and were found to aggregate mainly in clusters on the upper surface (Figure 5.15,

panels A1,A2 → G1,G2). Images of the top surface reveal rounded cells and in the

treatments containing VN, cells can be seen entering pores on the top surface (see

Panels B.1 and B.3). On the reverse side, the bottom of the insert, cells that have

attached appear more as single-cells rather than clusters – these observations are

clearer at higher magnifications (Panels A3, A4, A5, A6 → G3, G4, G5, G6).

Several interesting observations were noted relating to the microporous membrane

itself. Although it can be seen at 400x (Panels A1,A2 → G1,G2), it is clearer in the

images at 800x magnification Panels A3,A4 → G3,G4) that the surface topography

between the top and bottom surfaces are different. The top surface appears quite

smooth whereas the bottom surface is quite rough. In addition, the pores, which are

randomly positioned and at times overlap (for example, Panel E.6), have defined

edges on the top surface and on the bottom surface are not and are crater-like in

shape. Presumably this reflects how the pores are created and that most likely, pores

are created from the upper surface and when ‘punched’ through to the bottom

surface, some of the membrane from the ‘punched’ hole takes with it some of the

remaining intact membrane.

113

At 1600x magnification (Panels A5,A6 → G5,G6), cell morphology is more

identifiable. On the top surface and in the absence of VN, the cells are clustered

together and have very fine spindly projections. When VN is present, ECM

production appears to be increased as seen by the presence of sheet- and thread-like

debris surrounding the cells. Images of cells treated with IGF-I, IGFBP-5 and VN

alone appear to have similar amounts, and cells can clearly be seen to enter pores.

The treatment with the trimeric complex IGF-I:IGFBP-5:VN results in maximal

ECM production compared to all other treatments. On the bottom surface, the cells

once again are not in clusters and appear to be behaving separately. There is an

obvious production of ECM and many spindle-like processed emerging from the

cells. Cells are mainly fibroblastic in shape and appear quite flattened, with some

cells almost indiscernible from the membrane (see Panel G.6). In addition, the

visual estimate of the number of cells that have migrated to the bottom surface

appears to correlate well with the data from the Transwell migration assays. That is,

cell numbers present with exposure to the trimeric complex are greater than cells

exposed to either dimeric complexes or single protein treatments.

114

Top Surface Bottom Surface

400x

A.1 No VN A.2 No VN

B.1 VN B.2 VN

C.1 No VN + IGF-I C.2 No VN + IGF-I

D.1 VN + IGF-I D.2 VN + IGF-I

200 µm 200 µm

200 µm 200 µm

200 µm 200 µm

200 µm 200 µm

115

Top Surface Bottom Surface 400x

E.1 No VN + IGFBP-5 E.2 No VN + IGFBP-5

F.1 VN + IGFBP-5 F.2 VN + IGFBP-5

G.1 VN + IGF-I + IGFBP-5 G.2 VN + IGF-I + IGFBP-5

200 µm 200 µm

200 µm 200 µm

200 µm 200 µm

116


A.3 No VN A.4 No VN

B.3 VN B.4 VN



100 µm 100 µm

100 µm 100 µm

100 µm 100 µm

100 µm 100 µm

117




G.3 VN + IGF-I + IGFBP-5 G.4 VN + IGF-I + IGFBP-5

100 µm 100 µm

100 µm 100 µm

100 µm 100 µm

118


A.5 No VN A.6 No VN

B.5 VN B.6 VN



50 µm 50 µm

50 µm 50 µm

50 µm 50 µm

50 µm 50 µm

119




G.5 VN + IGF-I + IGFBP-5 G.6 VN + IGF-I + IGFBP-5 Figure 5.15 Scanning electron micrographs of MCF-7 cell migration in

response to IGF-I:IGFBP-5:VN complex combinations. In the Transwell assay,

200,000 cells were added to the upper surface of the membrane insert. Wells were

treated with the samples as stated and images were taken on both the top and the

bottom of the microporous membrane insert. Pore sizes are given at 12 µm by the

manufacturer.

50 µm 50 µm

50 µm 50 µm

50 µm 50 µm

120

5.4 DISCUSSION

The studies in this chapter have shown that the formation of IGF:IGFBP:VN

complexes do indeed have a functional role in cell physiology. It has been revealed

that trimeric complexes containing IGFBP-3 and IGFBP-5 stimulate protein

synthesis in CHO-K1 cells and migration in MCF-7 cells. In addition, IGFBP-4,

IGF-I and –II, and an IGF analogue can also stimulate cell migration and these

responses are due to the combination of proteins rather than individual proteins by

themselves. Furthermore, the studies presented in this chapter have highlighted a

number of structural regions within the IGFBPs that demonstrate functional

relevance.

Initially, it was established that trimeric complexes containing either IGFBP-3 or

IGFBP-5 could stimulate protein synthesis as examined in the CHO-K1 cell line.

However, further assays were not pursued with this cell line as CHO-K1 cells have

been shown to be highly dependent on VN to mediate cell attachment (Franco et al.,

2001) and as such did not turn out to be a useful model for examining the

complexes. The MCF-7 cells on the other hand, could attach in the absence of

serum, this presumably being a reflection of integrin expression and ECM

production by the cells (Wong et al., 1998; Morini et al., 2000; Zhang et al., 2004).

Therefore, it was felt the MCF-7 results provided a more appropriate model to

examine whether the presence of a heterotrimeric complex was essential for

functional responses.

Through use of the weakly metastatic breast cancer cell line, MCF-7, cell migration

responses were demonstrated with the various combinations of the IGF:IGFBP:VN

complex. While optimising the assay conditions to determine the amount of protein

to use, preliminary results demonstrated the ability of IGFBP-5 to markedly enhance

cell migration when added in combination with IGF-I and VN. These early assays

also demonstrated that in the presence of VN high levels of IGF-I could stimulate

increases in cell migration, whereas only low levels of IGF-I were required to

maximize migration in the presence of IGFBP-5. It was also identified that VN was

critical for any cell migration to occur, as in the absence of VN, levels of migration

were minimal. Similarly, through the use of des(1-3)IGF-I, which has a

121

considerably lower affinity for IGFBPs than IGF-I (Francis et al., 1993), it was

demonstrated that IGF-I needs to interact with IGFBPs and this is essential for

maximal cell migration.

Further examination of IGFBPs, specifically IGFBP-3, -4 and -5, in complex with

VN and IGF-I undertaken using the cell migration assay revealed all were able to

stimulate migration. Binary complexes composed of either low dose IGF-I:VN or

IGFBP:VN, were unable to effectively enhance cell migration. The comparison of

IGF-I with des(1-3)IGF-I clearly demonstrates that formation of the IGF-

I:IGFBP:VN ternary complex is required for eliciting enhanced responses in

migration. This is consistent with the structural studies as reported in Chapter 3 as

well as recent data from other laboratories. For example, these findings are

supported by Nam et al. (2002) who recently demonstrated (via immuno-

precipitation) direct binding of IGFBP-5 to VN and furthermore, demonstrated

functional relevance of this interaction. Their use of different biochemical and

functional methods further validates this study. Similarly, Xu et al. (2004) have

demonstrated direct protein-protein interaction between IGFBP-5 and fibronectin, a

glycoprotein with similar functions to VN, as well as the formation of a ternary

complex with IGF-I. Interestingly, they have shown IGF-I did not bind fibronectin

but, IGFBP-5 bound both IGF-I and fibronectin simultaneously yet independently.

HBD mutants of IGFBP-3 and IGFBP-5 were used to examine the physiological role

of the HBD in complex formation. The amino acid substitutions in these HBD

mutants overlap with a region that contributes to IGF binding (Parker et al., 1998;

Bramani et al., 1999; Song et al., 2000), yet prior characterisation of the IGFBP-5

and IGFBP-3 HBD mutants demonstrated that these proteins bind IGF-I to a similar

extent as their wild-type IGFBPs (Arai et al., 1996b; Firth et al., 1998). While

IGFBP-5 HBD mutants were shown to bind IGF-I in a similar manner as found with

wild-type IGFBP-5 (Figure 4.10), the studies reported in this thesis with IGFBP-3

HBD mutants indicate that binding of [125I]-IGF-I to HBD mutants was reduced

(Figure 4.8). Despite these conflicting data, the mutants were able to enhance MCF-

7 cell migration to a similar extent, or better than the wild-type IGFBPs when added

to IGF-I and VN. An earlier study by Nam et al. (2002) compared VN affinity for

wild-type IGFBP-5 with that for the identical IGFBP-5 HBD mutants. Using

122

competition assays, they found wild-type IGFBP-5 could inhibit IGFBP-5:VN

binding almost completely (~100%), while the equivalent mutant HBDm1 could

only disrupt binding 28%, and the equivalent mutant HBDm2, 76%. Furthermore,

Nam et al. (2002) also examined the effect of these IGFBP-5 proteins in the

presence of IGF-I and VN to stimulate cell migration across a wound-line using

smooth muscle cells (vSMC). These investigators found that protein combinations

containing wild-type IGFBP-5 enhanced vSMC migration 2-fold when compared to

cells plated on VN alone, while HBDm1-like protein increased migration 1.3-fold.

In the studies presented in this chapter, the opposite was shown; migration with the

IGFBP-5 HBDm1 was comparable to wild-type IGFBP-5 which in turn was

significantly increased compared to the responses found with the control. With

IGFBP-5 HBDm2 on the other hand, it had a similar response to that obtained with

the control (VN alone), and migration was significantly decreased when compared

to wild-type IGFBP-5. It was also interesting to note that the addition of heparin to

the IGF-I:IGFBP-5 HBDm2:VN complex resulted in an increased cell migration

response that was significantly different to the control (p < 0.01) and to IGF-I:VN

alone ((p < 0.05). It appears here that heparin was able to ‘rescue’ the diminished

cell migration response caused by the mutation within HBDm2. Although it is not

clear how this occurs, it supports the notion that residues within the HBD of IGFBP-

5 are involved in the IGF-I:IGFBP-5:VN interaction. The discrepancy between the

study presented here and that of Nam et al. (2002) most likely reflects the assay used

to measure migration. Migration in the wound-scrape assay was measured over 48

hr and the pre-treatment of the vSMC differed appreciably from the Transwell assay

used here, highlighting the importance of assay conditions when comparing data.

The migration results presented here using IGFBP-3 and the mammalian cell-

produced IGFBP-3 HBD mutant mirror the findings found when using IGFBP-5 and

the IGFBP-5 HBDm1. Again, MCF-7 migration was statistically increased

compared to the control, with both IGFBP-3 proteins. In brief, it would appear that

the role of the HBD, which structurally appeared to affect IGF-I:IGFBP-3/-5:VN

complex formation as measured by competitive assays in Sections 4.3.5 and 4.3.7,

has no consequence in these cell assays as the HBD mutants were able to enhance

cell migration greater than the control. To further support this, the effect of heparin

on trimeric complex-stimulated cell migration was assessed. The addition of

123

heparin to complexes containing either IGFBP-3 or -5 had no effect on cell

migration. This was surprising from the viewpoint that heparin has been reported to

disrupt IGFBP complexes with various ECM proteins and to modulate IGFBP-

mediated effects where binding has been localised to the HBD (Campbell et al.,

1998; Gui and Murphy, 2001; Hsieh et al., 2003). It is possible, however, that the

lack of effect observed with heparin is due to heparin binding to the HBD within

VN. The amount of complex formed may occur through IGFBP-3 and -5 binding to

the polyanionic region of VN, leaving the HBD of VN exposed and available for

heparin to bind (see Figure 3.6). Therefore, the complex remains unaffected by the

presence of heparin. In the case of IGFBP-4, the opposite may occur. As IGFBP-4

lacks an HBD, complex formation may occur through the positively-charged amino

acids present at the N-terminus of IGFBP-4 binding to the HBD region on VN. The

reduction in migration that is observed when heparin is added to IGF-I:IGFBP-4:VN

is possibly due to disruption of the trimeric complex at the HBD of VN whereby

heparin competes directly with IGFBP-4 for binding to VN (Figure 5.16).

Alternatively, glycosaminoglycans present in the ECM are also known to modulate

cellular activity and it has been suggested that binding of GAGs to IGFBP-3 and -5

lowers their affinity for IGF-I, allowing IGF-I to bind to its receptor (Arai et al.,

1994b). This too, may be true here, and may apply when IGF-I is bound to IGFBPs

and VN, although there is no evidence to support it. Specific to the MCF-7 cell

model, production of GAGs have been shown to be important in the mechanism of

breast cell growth (Lambrecht et al., 1998).

Figure 5.16 Proposed model of IGF-I:IGFBP-4:VN complex, with or without heparin. IGFBP-4 mediates [125I]-IGF-I binding to VN through an interaction between the N-terminus of IGFBP-4 with the HBD of VN: IGFBP-4 alone can bind to VN (A) while IGF-I binds poorly (B). Together, IGF-I can bind to VN via IGFBP-4, however, heparin may compete with IGFBP-4:IGF-I for binding to VN (C).

CA B

I

VN -- ++

VN -- ++

VN -- ++

I

IGFBP-4 --

IGFBP-4 --

heparin

124

To further define the role of the HBD in enhanced cell migration, IGFs were

compared to an IGF-I analogue with an HBD added at its N-terminus (MBF).

Several lines of evidence support IGF-I-induced cell migration in a number of cell

lines (Ando and Jensen, 1993; Gockerman et al., 1995; Mira et al., 1999;

Puglianiello et al., 2000) as has also been shown here. Both IGF-I and IGF-II dose-

dependently increased cell migration as did the IGF-I analogue, MBF. Stimulation

of MCF-7 cells to migrate in response to IGF-I and MBF in the presence of VN is

surprising given that earlier studies have shown that neither bind VN well. Also, the

presence of the HBD did not seem to affect the ability of IGF-I to stimulate cell

migration. This may reflect that the minimal amount of IGF-I and MBF that does

bind to VN, may be sufficient for cells to migrate to a greater degree than those cells

exposed to VN alone.

When IGF-I was replaced with des(1-3)IGF-I, cell migration responses were

substantially less, indicating that IGF-I that can interact with IGFBPs and that the

IGFBPs are critical for maximal cell migration. Interestingly, the reduced affinity of

the IGFBP-3/-5 HBD mutants for IGF-I, does not appear to affect migration

responses, indicating the complex can still form. This too, may suggest that less

IGFBP and/or IGF-I is sufficient for maximal cell migration. Another concern that

arises in the functional assays is the effect of endogenous protein production, which

may independently contribute to basal cell migration and therefore compensate for

the less functional proteins, such as the IGFBP mutants. However, this is unlikely to

be a contributing factor due to the short 5 hr incubation in the Transwell assay used

here. On the other hand, the decreased affinity that the mutant IGFBPs have for

IGF-I may more readily release IGF-I and/or VN to their respective receptors, which

may explain the increased migration responses seen with IGFBP-5 HBDm1 and

IGFBP-3 HBDm.

Several studies have described the cross-talk between the IGF-1R and integrins

which result in IGF-I-mediated effects (Jones et al., 1995; DePasquale, 1998; Maile

and Clemmons, 2002). Co-activation of the IGF-1R and VN receptor, αvβ5 (an

integrin present in MCF-7 cells) (DePasquale, 1998), has been reported previously

by Doerr and Jones (1996). Moreover, levels of αvβ5 have been shown to be

regulated by IGF-I concentration, which in turn has been shown to correlate with

125

cell migration responses (Mira et al., 1999; Bartucci et al., 2001). In addition

Bartucci et al. (2001) demonstrated that MCF-7 cell migration was also influenced

by the number of IGF-1R. Similarly, the synergistic responses observed in this

chapter are likely to be an indirect consequence of co-activation of the IGF-1R and

the VN receptor, αvβ5 integrin, an integrin present of MCF-7 cells (Wong et al.,

1998). In solution IGFBP-3 and -5 bind IGF-I, and therefore IGF-I is not released to

activate the IGF-1R, which also renders the IGFBPs unavailable to elicit IGF-

independent effects such as those described by Hsieh et al. and others (Hsieh et al.,

2003; Kim et al., 2004; Ricort and Binoux, 2004). It is possible that in the presence

of VN, IGFBPs in complex with IGF-I can also bind VN, thereby decreasing the

affinity of the IGFBPs for IGF-I and the complex acts as a delivery system for the

proteins and their respective receptors.

The effect of glycosylation on trimeric complex-enhanced MCF-7 cell migration

was also investigated. The presence and extent of carbohydrate moieties on IGFBPs

has been reported to be physiologically irrelevant with several studies asserting that

glycosylation does not affect the affinity of IGFBPs for IGF-I (Conover, 1991;

Sommer et al., 1993; Firth and Baxter, 1999). Nevertheless, Firth and Baxter (1995)

suggested that non-glycosylated IGFBP-3 may have a higher rate of turnover due to

increased susceptibility to proteolysis. This chapter reports that

non-glycosylated IGFBP-3 can significantly enhance MCF-7 cell migration to a

greater extent than fully glycosylated IGFBP-3. Furthermore, there was a noticeable

correlation with the extent of glycosylation and migration responses: thus, the less

glycosylated the protein, the greater migration. The exception to this was the non-

glycosylated IGFBP-3 mutant in which the glycosylation sites had been mutated

rather than the protein being expressed in a system which does not glycosylate

proteins (Firth and Baxter, 1995).

The studies presented in this chapter have provided insight into the effects of

IGFBP modulation of IGF-I-stimulated migration in the presence of VN, as well as

the effects of IGF interaction with VN in the absence of IGFBPs. Although

functional studies of individual complex components are well documented, little is

known of the signalling mechanisms involved when these components are presented

together as a complex, or of the regions that are important for these enhanced

126

cellular responses. Inhibitory and regulatory actions of IGFBPs on IGF action are

well characterised, in particular, IGFBP-2 and -4 are known to inhibit IGF-I effects

(Rees et al., 1998; Gustafsson et al., 1999; Hsieh et al., 2003; Pereira et al., 2004).

IGFBP-2, which contains the integrin recognition sequence RGD, has been shown to

negatively regulate IGF-I-stimulated cellular responses (Pereira et al., 2004).

Conversely, IGFBP-4 displays resistance to proteolysis thereby regulating IGF-I

activation of its receptor (Rees et al., 1998). In contrast, these studies demonstrate

IGFBP-4 in complex with IGF-I and VN is a potent stimulator of cell migration,

similar to IGFBP-3 and IGFBP-5. Studies by Hsieh et al. (2003) validate these

findings with IGFBP-3 and -5, also using vSMCs in the Transwell migration assay.

However, they also demonstrated IGF-independent effects of IGFBP-5 on cell

migration. Although the study reported in this thesis did not specifically address

IGF-independent migration, it would be interesting to further investigate Hsieh’s

findings, and relate their study to the one presented here by examining how

migration is enhanced. In addition, investigations into the delivery of proteins to

their receptors and the subsequent signalling cascades initiated are clearly worthy of

further investigation.

In summary, these results, demonstrating a functional role for IGF:IGFBP:VN

complexes in cell migration, highlight the complexity of protein interactions on cell

behaviour. This is particularly relevant to models where cell migration occurs,

either in normal physiology such as wound healing, or as demonstrated here, in

breast cancer. Both of these physiological processes require the action of IGFs and

VN to stimulate cell migration to serve different purposes: either closure of wounds,

or in breast cancer metastasis. Increasing the knowledge base of the IGF:IGFBP:VN

protein interactions and their effects on cell migration may provide critical

information leading to the development of targets for therapeutics.

127

CHAPTER 6: GENERAL DISCUSSION

128

The aim of this project was to investigate and expand on current knowledge that

IGFs could interact with the extracellular matrix protein VN. This thesis focused on

interactions between IGF-I and –II, IGFBPs and VN, both structurally and

functionally. A number of novel findings are reported here, as well as studies

corroborating the initial investigations revealing the association of IGF-II with VN,

and the poor ability of IGF-I to bind VN (Upton et al., 1999). Evidence has been

provided to support the initial hypotheses proposed in Chapter 1 as described below.

The first hypothesis suggested there was a link between the IGFs and VN which, at

least in the case of IGF-I, may be mediated via the IGFBPs. Indeed, this was

supported when the effect of the 6 IGFBPs was determined and it was revealed that

in the presence of IGFBP-2, -3, -4 and -5, IGF-I could interact with VN (Chapter 3).

In contrast, all six IGFBPs inhibited the interaction between IGF-II and VN, albeit to

varying degrees. It was interesting to note that of the IGFBPs that had the greatest

potency in competing with IGF-II for binding to VN (IGFBP-1, -3 and -6), IGFBP-1

and -6 were the least effective in mediating IGF-I binding to VN. Similarly, IGFBP-

5, which only competed with IGF-II binding to VN at the highest amounts tested,

increased IGF-I binding the greatest, while to a lesser extent both IGFBP-2 and -4

competed with IGF-II binding and increased IGF-I binding to VN. Whether these

relationships indicate a characteristic pertinent to each IGFBP that is involved in

VN/IGF binding, is not known. However, the critical intermediate involvement of

the IGFBPs in mediating IGF-I binding to VN was shown to be specific for IGFBP-

3 and -5 through the use of IGF analogues that bind the IGFBPs poorly. These IGF-

I analogues were unable to bind VN.

The second hypothesis was that the heparin-binding regions are important in the

IGF:VN interaction. This hypothesis was later expanded to include glycosylation as

an important factor. The structural relationship of the IGFs with VN and the

involvement of these characteristics of the IGFBPs were scrutinised (Chapter 4).

Results from Chapter 4 demonstrated that the presence of carbohydrate moieties on

the IGFBPs plays a role in binding of IGFs to VN. The use of an IGFBP-3

glycosylation mutant revealed that non-glycosylated IGFBP-3 mediated binding of

IGF-I to VN more effectively than glycosylated IGFBP-3. Surprisingly, this was not

due to a preference of either glycosylated or non-glycosylated IGFBP-3 for VN or

129

IGF-I, which both supports and contradicts the literature that has described

glycosylation effects of IGFBP-3. Reports to date claim that less glycosylated

IGFBP-3 has a higher affinity for the cell surface (which may reflect the affinity of

IGFBP-3 for VN) than fully glycosylated IGFBP-3 (Firth and Baxter, 1999), yet

there is no distinction between affinity of non-glycosylated IGFBP-3 and

glycosylated IGFBP-3 for IGF-I binding (Sommer et al., 1993). The comparison of

different IGFBP-3 glycosylation preparations described here indicated a slight trend

for non-glycosylated IGFBP-3 to exhibit enhanced binding of IGF-I, though the

differences were not statistically significant. Nevertheless, in this study, these minor

differences in binding between glycosylated and non-glycosylated IGFBP-3 may

have compounded when trimeric complex formation was assessed.

In addition, glycosylation was also shown to affect the IGF-II:VN interaction. Here,

non-glycosylated IGFBP-6 was shown to have a lower affinity for IGF-II than

glycosylated IGFBP-6, and hence, was a less potent competitor for IGF-II binding to

VN binding. Results from Chapter 4 clarify existing discrepancies in the current

literature regarding IGFBP-6 glycosylation and IGF-II affinity, as well as identifying

that the source of IGFBP-6 is equally important in terms of its IGF-II affinity (Bach

et al., 1992; Wong et al., 1999; Marinaro et al., 2000).

Analysis of the VN and IGFBP sequences led to a proposal that IGFBP-3 and -5

interact with VN via their basic HBD where they bind to the polyanionic region of

VN. In addition it was further hypothesised that IGFBP-4, which lacks a HBD, may

instead bind via to the HBD within VN, as IGFBP-4 has a complementary region of

acidic residues in its N-terminal domain. Findings reported in Chapter 4 highlighted

the role of the HBD in IGFBP-3 and IGFBP-5 on IGF-I binding to VN. While

mutation of the IGFBP-3 HBD negated IGF-I binding to VN, HBD mutations in

IGFBP-5 only reduced IGF-I binding to VN. PEG precipitation, used to re-

characterise binding of these mutants to IGF-I, revealed binding of IGF-I was

reduced with the IGFBP-3 HBD mutant, but near normal binding of IGF-I was

found with the IGFBP-5 mutants. In addition, it appears that varying IGF-I affinities

for IGFBPs may in turn reflect the potency of IGF binding to VN via the IGFBPs.

Overall, these studies suggest that the IGFBP HBD does indeed play a role in the

IGF:VN association.

130

Having identified that IGFBPs could mediate IGF-I binding to VN and that

glycosylation and IGFBP HBDs were important in the interaction, it was then

explored whether these findings had a functional significance. In Chapter 5,

migration studies with the MCF-7 breast carcinoma cell line were undertaken to

assess this. A number of findings were made in this chapter including: all three

components, that is IGF-I, IGFBP and VN are required for maximal cell migration;

IGFBP-3, -4 or -5 each could stimulate enhanced cell migration when present with

VN and low doses of IGF-I; and IGFs alone at high doses could also enhance cell

migration when present with VN.

When the effects of glycosylation and the HBDs were examined on trimeric

complex-stimulated cell migration, it was found that the use of less glycosylated

IGFBP-3 in the trimeric complex formation resulted in higher migration responses

than fully glycosylated IGFBP-3. This complements the increased binding of IGF-I

to non-glycosylated IGFBPs and VN. The exception to this was the result obtained

when using the mutant non-glycosylated IGFBP-3. However, this may be due to

this preparation being non-glycosylated through mutation of specific Asn residues

rather than as a result of the protein expression system used and hence subsequent

post-translational modification. The IGFBP-3 preparations that follow the trend

were expressed as native IGFBP-3 yet the expression system they were produced in

was the limiting factor in glycosylating the proteins. In contrast, the non-

glycosylated IGFBP-3 preparation that does not follow the trend was produced in

mammalian cells, yet the amino acids that are normally glycosylated were subject to

site-directed mutagenesis preventing post-translational modification to take place.

Chapter 5 findings revealed contradictory results with regards to the function of the

IGFBP HBD. Migration was significantly enhanced compared to the control when

either wild-type IGFBP-3 or HBD mutant IGFBP-3 was added to IGF-I and VN.

Equally, IGFBP-5 HBDm1 had comparable migration responses to wild-type

IGFBP-5, yet IGFBP-5 HBDm2 had a reduced ability to stimulate migration. These

results were quite surprising given the results from Chapter 4, and as such, it was

expected that all the HBD mutants would behave functionally like IGFBP-5 HBD

m2. There does not seem to be any plausible explanation for the differences

observed with the HBD mutant migration responses. The migration results

131

described in Chapter 5 with the IGFBP-5 m2 trimeric complex are similar to those

found by Hsieh et al. (2003), where they too found reduced migration when they

compared their heparin-binding mutant IGFBP-5 with wild-type IGFBP-5. In

contrast, comparison of cell migration responses found with the mutant similar to

IGFBP-5 m1 do not corroborate those found in Chapter 5 (Nam et al., 2002). As

mentioned in the discussion of Chapter 5, this discrepancy with the study by Nam et

al. (2002) is likely to be due to the methodology rather than the IGFBP-5 mutants

themselves.

The studies with heparin in Chapter 5 were designed to further investigate the

hypothesis that the IGFBP HBD is functionally important. Although heparin had no

effect on enhanced migration with IGFBP-3 or -5 trimeric complexes, these results

provided insufficient evidence to reject or accept the hypothesis. In fact, these

findings support the notion that IGFBP-3 and -5 bind to VN via their HBD and the

lack of effect observed with heparin is due to heparin binding the HBD of VN,

rather than competing with the IGFBP-3/-5 binding site (Figure 6.1). In the case of

IGFBP-4, heparin was able to reduce migration stimulated by trimeric complexes.

As IGFBP-4 does not contain an HBD sequence, it was conjectured that it may bind

to VN via the HBD within VN, and therefore, heparin may disrupt binding by

directly competing for the same site (Figure 6.2). This indeed was found to be true,

although the reduction in migration caused by the presence of heparin was found to

be not significantly different to cell responses to IGFBP-4 trimeric complexes in the

absence of heparin.

The studies reported in this thesis have demonstrated a number of findings which

could be further investigated. Future directions for this project include: the use of

synthetic peptides representing key structural regions (see Figure 6.3), such as the

HBDs of IGFBPs and/or VN and the polyanionic region of VN, to assess whether

the peptides can interfere with complex binding and subsequent functional studies;

and evidence to specifically demonstrate direct protein-protein interaction through

immunoligand blots and immuno-precipitation studies. In addition, the BIAcore

instrument could be used to determine binding affinities of IGFs and IGFBPs for

VN. Protein mutations could also provide insight into the IGF:VN interaction –

132

Figure 6.1 Proposed model of IGF-I binding to VN via IGFBP-3 or -5. IGFBP-3 and -5 mediate [125I]-IGF-I binding to VN: IGFBPs alone can bind to VN (A) while IGF-I binds poorly (B). Together, IGF-I can bind to VN via IGFBP-3 and -5, while heparin can bind to VN’s exposed HBD (C).

Figure 6.2 Proposed model of IGF-I binding to VN via IGFBP-4. IGFBP-4 mediates [125I]-IGF-I binding to VN through an interaction between the N-terminus of IGFBP-4 with the HBD of VN: IGFBP-4 alone can bind to VN (A) while IGF-I binds poorly (B). Together, IGF-I can bind to VN via IGFBP-4, however, heparin may compete with IGFBP-4:IGF-I for binding to VN (C).

namely the use of IGF analogues. For example, the lack of ability of IGF-I to bind

to VN in contrast to IGF-II that can bind VN, may be due the presence of a heparin-

binding-like consensus sequence at IGF-II’s C-domain (Figure 6.4). Thus, through

use of IGF-I and IGF-II C-domain swap mutants, the role of the C-domain in

trimeric and dimeric complex formation will help isolate and identify which amino

acid residues are responsible for IGF-II binding to VN (Figure 6.3). IGF chimeras

where the C and D domains were swapped between IGF-I and IGF-II, were

employed in a similar study by Denley et al. (2004), whereby they determined the

region in IGF-II responsible for binding to the insulin receptor. In addition, the

VN -- ++

VN -- ++

VN -- ++

IGFBP-4 --

IGFBP-4 --

heparin I I

CA B

CA B

I IGFBP

VN -- ++

VN -- ++

VN -- ++

++ IGFBP

++

I heparin HBD region polyanionic

region

133

Figure 6.3 Synthetic peptides. Schematic diagrams of VN, IGFBP-5 and -4, and and IGF-I and IGF-II highlighting potential sequences for peptide synthesis. The peptides representing the structural regions can be used in competition studies to help identify the significance of these motifs on IGF:IGFBP:VN complex formation.

Figure 6.4 Proposed model of IGF-II binding to VN. IGFBPs compete with [125I]-IGF-II for binding to VN, which may occur via one of two mechanisms: IGFBPs can bind directly to VN (A) as can IGF-II via its heparin-binding motif (B). When both IGFBPs and IGF-II are present in solution, they may compete for binding at the same or adjacent sites on VN (C) or the IGFBPs sequester IGF-II away from VN (D).

HBD KRKQCK N

N ELAEIE

N DDGEE AKKQRF HBD

Acidic region

Acidic region

VN

BP5

BP4

N RVSRRSR Putative HBD

IGF-II

N RAPQTG Corresponding

sequence

IGF-I

C

C

C

C

C

C D

A B

II IGFBP

VN -- ++

VN -- ++

IGFBP II IGFBP II

VN -- ++

VN -- ++

HBD region polyanionic

region

++

++++

134

generation of synthetic peptides representing the HBD and polyanionic region

within VN may also provide insights into the regions involved in complex

formation.

Furthermore, regulation of cell migration involves proteases, for example, the matrix

metalloproteinases (MMPs). Thus, it would be of interest to examine the expression

of MMPs in unstimulated MCF-7 cells in comparison to types and levels present in

cells exposed to trimeric complexes and the various component combinations. Mira

et al. (1999) have shown MMP-9 to be expressed in MCF-7 cells exposed to VN and

IGF-I. These investigators also demonstrated that αvβ5 integrins were up-regulated

when cells were exposed to IGF-I, indicating that analysis of integrin and MMP

expression could be valuable in order to understand the complexity and the effects of

growth factors on cell proliferation and migration. These future studies would help

to shed new light on physiological conditions such as, growth, cancer metastasis,

wound healing and other situations were IGFs, IGFBPs and VN have all been

implicated to have a role.

In summary, the novel findings in this thesis provide valuable insight into the

interaction of IGFs with VN via IGFBPs. Through understanding this interaction,

comprehension of the diversity, and the relationships between the IGF axis and the

vitronectin in normal physiology, may be appreciated. As the IGF axis and VN play

critical roles in cell growth, migration and regulation, the new relationships reported

here may prove to be useful in the development of improved therapeutics.

135

CHAPTER 7:

APPENDIX

136

BUFFER RECIPES

7.1 IODINATION BUFFERS Chromatography Buffer pH 6.5

50 mM NaH2PO4·2H2O

0.15 M NaCl

0.25 % BSA

Equilibration Buffers 1 – 4 pH 6.5

1. 50 mM NaH2PO4·2H2O

0.1 % BSA

2. As 1) + 0.4 M NaCl

3. As 1) + 0.75 M NaCl

4. As 1) + 1.0 M NaCl

7.2 SPBA AND FUNCTIONAL ASSAYS HEPES Binding Buffer (HBB) pH 7.6

0.1 M HEPES

0.12 M NaCl

5 mM KCl

1.2 mM MgSO4

8 mM Glucose

0.5 % BSA

7.3 PEG PRECIPITATION ASSAY Buffer pH 6.5

0.1 M HEPES

0.44 mM NaH2PO4

0.1 % Triton X-100

0.1 % BSA

137

7.4 SDS-PAGE AND WESTERN BLOT 4x Resolving Gel Buffer pH 8.8

1.5 M Tris-HCl

0.4 % SDS

4x Stacking Gel Buffer pH 6.8

0.5 M Tris-HCl

0.4 % SDS

5x Running Buffer pH 8.3

25 mM Tris

192 mM Glycine

0.1% SDS

5x Sample Buffer pH 6.8

0.05% Bromophenol blue

0.225M Tris-HCl

5% SDS

50% Glycerol

Gels

Resolving Gel % (for 2 gels) 8 10 12 15 20 40% acrylamide 2.0 mL 2.5 3.0 3.75 5.0 Resolving Gel Buffer 2.5 mL 2.5 2.5 2.5 2.5 dd water 5.4 mL 4.9 4.4 3.65 2.4 10% APS 50 μL 50 50 50 50 TEMED 5 μL 5 5 5 5 Stacking Gel # gels (4% gel) 2 4 6 10 14 40% acrylamide 0.8 mL 1.6 2.4 4.0 5.6 Stacking Gel Buffer 2.5 mL 5 7.5 12.5 17.5 dd water 6.6 mL 13.2 19.8 33 46.2 10% APS 100 μL 200 300 400 700 TEMED 10 μL 20 30 40 70

138

Wet Transfer Buffer pH 9.9

10 mM NaHCO3

3 mM Na2CO3

20 % Methanol

Semi-dry Transfer Buffers pH 8.3

1. 15 mM Tris

120 mM Glycine

20% Methanol

2. 0.3 M Tris

7.5 RADIOLIGAND BLOT 10x TBS Buffer pH 7.4 0.1 M Tris

1.4 M NaCl

Triton Wash Buffer pH 7.4 10 % TBS

1 % Triton X-100

Wash Buffer pH 7.4

10 % TBS

0.1 % Tween 20

Blocking Buffer pH 7.4

Wash buffer

1.0 % BSA

7.6 OTHER Caustic Triton

0.5 M NaOH

0.1 % Triton X-100

139

CHAPTER 8: REFERENCES

140

Aaboe, M., Offersen, B. V., Christensen, A., and Andreasen, P. A. (2003). Vitronectin in human breast carcinomas. Biochim Biophys Acta 1638, 72-82.

Abbott, A. M., Bueno, R., Pedrini, M. T., Murray, J. M., and Smith, R. J. (1992). Insulin-like growth factor I receptor gene structure. J Biol Chem 267, 10759-10763.

Abrass, C. K., Berfield, A. K., and Andress, D. L. (1997). Heparin binding domain of insulin-like growth factor binding protein-5 stimulates mesangial cell migration. Am J Physiol 273, F899-906.

Adamo, M. L., Shao, Z. M., Lanau, F., Chen, J. C., Clemmons, D. R., Roberts, C. T., Jr., LeRoith, D., and Fontana, J. A. (1992). Insulin-like growth factor-I (IGF-I) and retinoic acid modulation of IGF-binding proteins (IGFBPs): IGFBP-2, -3, and -4 gene expression and protein secretion in a breast cancer cell line. Endocrinology 131, 1858-1866.

Ando, Y., and Jensen, P. J. (1993). Epidermal growth factor and insulin-like growth factor I enhance keratinocyte migration. J Invest Dermatol 100, 633-639.

Andress, D. L. (2001). IGF-binding protein-5 stimulates osteoblast activity and bone accretion in ovariectomized mice. Am J Physiol Endocrinol Metab 281, E283-288.

Arai, T., Arai, A., Busby, W. H., Jr., and Clemmons, D. R. (1994a). Glycosaminoglycans inhibit degradation of insulin-like growth factor-binding protein-5. Endocrinology 135, 2358-2363.

Arai, T., Busby Walker, Jr., and Clemmons, D. R. (1996a). Binding of insulin-like growth factor (IGF) I or II to IGF-binding protein-2 enables it to bind to heparin and extracellular matrix. Endocrinology 137, 4571-4575.

Arai, T., Clarke, J., Parker, A., Busby, W., Jr., Nam, T., and Clemmons, D. R. (1996b). Substitution of specific amino acids in insulin-like growth factor (IGF) binding protein 5 alters heparin binding and its change in affinity for IGF-I response to heparin. J Biol Chem 271, 6099-6106.

Arai, T., Parker, A., Busby, W., Jr., and Clemmons, D. R. (1994b). Heparin, heparan sulfate, and dermatan sulfate regulate formation of the insulin-like growth factor-I and insulin-like growth factor-binding protein complexes. J Biol Chem 269, 20388-20393.

Arany, E., Afford, S., Strain, A. J., Winwood, P. J., Arthur, M. J., and Hill, D. J. (1994). Differential cellular synthesis of insulin-like growth factor binding protein-1 (IGFBP-1) and IGFBP-3 within human liver. J Clin Endocrinol Metab 79, 1871-1876.

Bach, L. A., Hsieh, S., Sakano, K., Fujiwara, H., Perdue, J. F., and Rechler, M. M. (1993). Binding of mutants of human insulin-like growth factor II to insulin-like growth factor binding proteins 1-6. J Biol Chem 268, 9246-9254.

141

Bach, L. A., Thotakura, N. R., and Rechler, M. M. (1992). Human insulin-like growth factor binding protein-6 is O-glycosylated. Biochem Biophys Res Commun 186, 301-307.

Bagnall, W., Sharpe, P. M., Newham, P., Tart, J., Mott, R. A., Torr, V. R., Forder, R. A., and Needham, M. R. (2003). Expression and purification of biologically active IGF-binding proteins using the LCR/Mel expression system. Protein Expr Purif 27, 1-11.

Ballard, F. J., Read, L. C., Francis, G. L., Bagley, C. J., and Wallace, J. C. (1986). Binding properties and biological potencies of insulin-like growth factors in L6 myoblasts. Biochem J 233, 223-230.

Bar, R. S., Harrison, L. C., Baxter, R. C., Boes, M., Dake, B. L., Booth, B., and Cox, A. (1987). Production of IGF-binding proteins by vascular endothelial cells. Biochem Biophys Res Commun 148, 734-739.

Barnes, D. W., Foley, T. P., Jr., Shaffer, M. C., and Silnutzer, J. E. (1984). Human serum spreading factor: relationship to somatomedin B. J Clin Endocrinol Metab 59, 1019-1021.

Bartucci, M., Morelli, C., Mauro, L., Ando, S., and Surmacz, E. (2001). Differential insulin-like growth factor I receptor signaling and function in estrogen receptor (ER)-positive MCF-7 and ER-negative MDA-MB-231 breast cancer cells. Cancer Res 61, 6747-6754.

Bauer, J. S., Schreiner, C. L., Giancotti, F. G., Ruoslahti, E., and Juliano, R. L. (1992). Motility of fibronectin receptor-deficient cells on fibronectin and vitronectin: collaborative interactions among integrins. J Cell Biol 116, 477-487.

Bayes-Genis, A., Schwartz, R. S., Lewis, D. A., Overgaard, M. T., Christiansen, M., Oxvig, C., Ashai, K., Holmes, D. R., Jr., and Conover, C. A. (2001). Insulin-like growth factor binding protein-4 protease produced by smooth muscle cells increases in the coronary artery after angioplasty. Arterioscler Thromb Vasc Biol 21, 335-341.

Bayne, M. L., Applebaum, J., Chicchi, G. G., Miller, R. E., and Cascieri, M. A. (1990). The roles of tyrosines 24, 31, and 60 in the high affinity binding of insulin-like growth factor-I to the type 1 insulin-like growth factor receptor. J Biol Chem 265, 15648-15652.

Bennett, A., Wilson, D. M., Liu, F., Nagashima, R., Rosenfeld, R. G., and Hintz, R. L. (1983). Levels of insulin-like growth factors I and II in human cord blood. J Clin Endocrinol Metab 57, 609-612.

Bernard, L., Babajko, S., Binoux, M., and Ricort, J. M. (2002). The amino-terminal region of insulin-like growth factor binding protein-3, (1-95)IGFBP-3, induces apoptosis of MCF-7 breast carcinoma cells. Biochem Biophys Res Commun 293, 55-60.

Bhaumick, B., and Bala, R. M. (1991). Differential effects of insulin-like growth factors I and II on growth, differentiation and glucoregulation in differentiating chondrocyte cells in culture. Acta Endocrinol (Copenh) 125, 201-211.

142

Binoux, M., Lalou, C., and Mohseni-Zadeh, S. (1999). Biological Actions of Proteolytic Fragments of the IGF Binding Proteins. In The IGF system : molecular biology, physiology, and clinical applications, R. G. Rosenfeld, and C. T. Roberts, Jr., eds. (Totowa, N.J., Humana Press), pp. 281-314.

Blakesley, V. A., Kalebic, T., Helman, L. J., Stannard, B., Faria, T. N., Roberts, C. T., Jr., and LeRoith, D. (1996). Tumorigenic and mitogenic capacities are reduced in transfected fibroblasts expressing mutant insulin-like growth factor (IGF)-I receptors. The role of tyrosine residues 1250, 1251, and 1316 in the carboxy-terminus of the IGF-I receptor. Endocrinology 137, 410-417.

Blakesley, V. A., Koval, A. P., Stannard, B. S., Scrimgeour, A., and LeRoith, D. (1998). Replacement of tyrosine 1251 in the carboxyl terminus of the insulin-like growth factor-I receptor disrupts the actin cytoskeleton and inhibits proliferation and anchorage-independent growth. J Biol Chem 273, 18411-18422.

Blanchard, F., Raher, S., Duplomb, L., Vusio, P., Pitard, V., Taupin, J. L., Moreau, J. F., Hoflack, B., Minvielle, S., Jacques, Y., and Godard, A. (1998). The mannose 6-phosphate/insulin-like growth factor II receptor is a nanomolar affinity receptor for glycosylated human leukemia inhibitory factor. J Biol Chem 273, 20886-20893.

Blancher, C., Omri, B., Bidou, L., Pessac, B., and Crisanti, P. (1996). Nectinepsin: a new extracellular matrix protein of the pexin family. Characterization of a novel cDNA encoding a protein with an RGD cell binding motif. J Biol Chem 271, 26220-26226.

Boettiger, D., Lynch, L., Blystone, S., and Huber, F. (2001). Distinct ligand-binding modes for integrin alpha(v)beta(3)-mediated adhesion to fibronectin versus vitronectin. J Biol Chem 276, 31684-31690.

Booth, B. A., Boes, M., Andress, D. L., Dake, B. L., Kiefer, M. C., Maack, C., Linhardt, R. J., Bar, K., Caldwell, E. E., Weiler, J., and et al. (1995). IGFBP-3 and IGFBP-5 association with endothelial cells: role of C-terminal heparin binding domain. Growth Regul 5, 1-17.

Booth, B. A., Boes, M., Dake, B. L., Knudtson, K. L., and Bar, R. S. (2002). IGFBP-3 binding to endothelial cells inhibits plasmin and thrombin proteolysis. Am J Physiol Endocrinol Metab 282, E52-58.

Bostedt, K. T., Schmid, C., Ghirlanda-Keller, C., Olie, R., Winterhalter, K. H., and Zapf, J. (2001). Insulin-like Growth Factor (IGF) I Down-Regulates Type 1 IGF Receptor (IGF 1R) and Reduces the IGF I Response in A549 Non-Small-Cell Lung Cancer and Saos-2/B-10 Osteoblastic Osteosarcoma Cells. Exp Cell Res 271, 368-377.

Bramani, S., Song, H., Beattie, J., Tonner, E., Flint, D. J., and Allan, G. J. (1999). Amino acids within the extracellular matrix (ECM) binding region (201-218) of rat insulin-like growth factor binding protein (IGFBP)-5 are important determinants in binding IGF-I. J Mol Endocrinol 23, 117-123.

143

Brooks, P. C., Klemke, R. L., Schon, S., Lewis, J. M., Schwartz, M. A., and Cheresh, D. A. (1997). Insulin-like growth factor receptor cooperates with integrin alpha v beta 5 to promote tumor cell dissemination in vivo. J Clin Invest 99, 1390-1398.

Brown, C., Stenn, K. S., Falk, R. J., Woodley, D. T., and O'Keefe, E. J. (1991). Vitronectin: effects on keratinocyte motility and inhibition of collagen-induced motility. J Invest Dermatol 96, 724-728.

Buckley, D. A., Cheng, A., Kiely, P. A., Tremblay, M. L., and O'Connor, R. (2002). Regulation of Insulin-Like Growth Factor Type I (IGF-I) Receptor Kinase Activity by Protein Tyrosine Phosphatase 1B (PTP-1B) and Enhanced IGF-I-Mediated Suppression of Apoptosis and Motility in PTP-1B-Deficient Fibroblasts. Mol Cell Biol 22, 1998-2010.

Bunn, R. C., and Fowlkes, J. L. (2003). Insulin-like growth factor binding protein proteolysis. Trends Endocrinol Metab 14, 176-181.

Byun, D., Mohan, S., Yoo, M., Sexton, C., Baylink, D. J., and Qin, X. (2001). Pregnancy-associated plasma protein-A accounts for the insulin-like growth factor (IGF)-binding protein-4 (IGFBP-4) proteolytic activity in human pregnancy serum and enhances the mitogenic activity of IGF by degrading IGFBP-4 in vitro. J Clin Endocrinol Metab 86, 847-854.

Campbell, P. G., and Andress, D. L. (1997a). Insulin-like growth factor (IGF)-binding protein-5-(201-218) region regulates hydroxyapatite and IGF-I binding. Am J Physiol 273, E1005-1013.

Campbell, P. G., and Andress, D. L. (1997b). Plasmin degradation of insulin-like growth factor-binding protein-5 (IGFBP-5): regulation by IGFBP-5-(201-218). Am J Physiol 273, E996-1004.

Campbell, P. G., Durham, S. K., Hayes, J. D., Suwanichkul, A., and Powell, D. R. (1999). Insulin-like growth factor-binding protein-3 binds fibrinogen and fibrin. J Biol Chem 274, 30215-30221.

Campbell, P. G., Durham, S. K., Suwanichkul, A., Hayes, J. D., and Powell, D. R. (1998). Plasminogen binds the heparin-binding domain of insulin-like growth factor-binding protein-3. Am J Physiol 275, E321-331.

Cardin, A. D., and Weintraub, H. J. (1989). Molecular modeling of protein-glycosaminoglycan interactions. Arteriosclerosis 9, 21-32.

Carriero, M. V., Del Vecchio, S., Capozzoli, M., Franco, P., Fontana, L., Zannetti, A., Botti, G., D'Aiuto, G., Salvatore, M., and Stoppelli, M. P. (1999). Urokinase receptor interacts with alpha(v)beta5 vitronectin receptor, promoting urokinase-dependent cell migration in breast cancer. Cancer Res 59, 5307-5314.

Carriero, M. V., Del Vecchio, S., Franco, P., Potena, M. I., Chiaradonna, F., Botti, G., Stoppelli, M. P., and Salvatore, M. (1997). Vitronectin binding to urokinase receptor in human breast cancer. Clin Cancer Res 3, 1299-1308.

144

Cascieri, M. A., Chicchi, G. G., Applebaum, J., Hayes, N. S., Green, B. G., and Bayne, M. L. (1988). Mutants of human insulin-like growth factor I with reduced affinity for the type 1 insulin-like growth factor receptor. Biochemistry 27, 3229-3233.

Chakraborty, C., Gleeson, L. M., McKinnon, T., and Lala, P. K. (2002). Regulation of human trophoblast migration and invasiveness. Can J Physiol Pharmacol 80, 116-124.

Chandhoke, S. K., Williams, M., Schaefer, E., Zorn, L., and Blystone, S. D. (2004). Beta 3 integrin phosphorylation is essential for Arp3 organization into leukocyte alpha V beta 3-vitronectin adhesion contacts. J Cell Sci 117, 1431-1441.

Chavakis, T., Kanse, S. M., Yutzy, B., Lijnen, H. R., and Preissner, K. T. (1998). Vitronectin concentrates proteolytic activity on the cell surface and extracellular matrix by trapping soluble urokinase receptor-urokinase complexes. Blood 91, 2305-2312.

Chen, C. C., Chen, N., and Lau, L. F. (2001). The angiogenic factors Cyr61 and connective tissue growth factor induce adhesive signaling in primary human skin fibroblasts. J Biol Chem 276, 10443-10452.

Chen, N., Chen, C. C., and Lau, L. F. (2000). Adhesion of human skin fibroblasts to Cyr61 is mediated through integrin alpha 6beta 1 and cell surface heparan sulfate proteoglycans. J Biol Chem 275, 24953-24961.

Chernausek, S. D., Smith, C. E., Duffin, K. L., Busby, W. H., Wright, G., and Clemmons, D. R. (1995). Proteolytic cleavage of insulin-like growth factor binding protein 4 (IGFBP-4). Localization of cleavage site to non-homologous region of native IGFBP-4. Journal of Biological Chemistry 270, 11377-11382.

Ciambrone, G. J., and McKeown-Longo, P. J. (1990). Plasminogen activator inhibitor type I stabilizes vitronectin-dependent adhesions in HT-1080 cells. J Cell Biol 111, 2183-2195.

Ciambrone, G. J., and McKeown-Longo, P. J. (1992). Vitronectin regulates the synthesis and localization of urokinase-type plasminogen activator in HT-1080 cells. J Biol Chem 267, 13617-13622.

Claussen, M., Kubler, B., Wendland, M., Neifer, K., Schmidt, B., Zapf, J., and Braulke, T. (1997). Proteolysis of insulin-like growth factors (IGF) and IGF binding proteins by cathepsin D. Endocrinology 138, 3797-3803.

Clemmons, D. R. (1998). Role of insulin-like growth factor binding proteins in controlling IGF actions. Mol Cell Endocrinol 140, 19-24.

Clemmons, D. R., Horvitz, G., Engleman, W., Nichols, T., Moralez, A., and Nickols, G. A. (1999). Synthetic alphaVbeta3 antagonists inhibit insulin-like growth factor-I-stimulated smooth muscle cell migration and replication. Endocrinology 140, 4616-4621.

145

Cobb, L. J., Salih, D. A., Gonzalez, I., Tripathi, G., Carter, E. J., Lovett, F., Holding, C., and Pell, J. M. (2004). Partitioning of IGFBP-5 actions in myogenesis: IGF-independent anti-apoptotic function. J Cell Sci 117, 1737-1746.

Cohick, W. S., and Clemmons, D. R. (1993a). The insulin-like growth factors. Annu Rev Physiol 55, 131-153.

Cohick, W. S., and Clemmons, D. R. (1993b). Regulation of IGFBP secretion and modulation of cell growth in MDBK cells. Growth Regul 3, 20-23.

Cohick, W. S., Gockerman, A., and Clemmons, D. R. (1993). Vascular smooth muscle cells synthesize two forms of insulin-like growth factor binding proteins which are regulated differently by the insulin-like growth factors. J Cell Physiol 157, 52-60.

Collet, C., and Candy, J. (1998). How many insulin-like growth factor binding proteins? Mol Cell Endocrinol 139, 1-6.

Conover, C. A. (1991). Glycosylation of insulin-like growth factor binding protein-3 (IGFBP-3) is not required for potentiation of IGF-I action: evidence for processing of cell-bound IGFBP-3. Endocrinology 129, 3259-3268.

Conover, C. A. (1992). Potentiation of insulin-like growth factor (IGF) action by IGF-binding protein-3: studies of underlying mechanism. Endocrinology 130, 3191-3199.

Conover, C. A. (1999). Posttranslational Modification of the IGF Binding Proteins. In The IGF system: molecular biology, physiology, and clinical applications, R. G. Rosenfeld, Roberts, C.T., Jr., ed. (Totowa, N.J., Humana Press), pp. 355-378.

Conover, C. A., Clarkson, J. T., and Bale, L. K. (1994). Insulin-like growth factor-II enhancement of human fibroblast growth via a nonreceptor-mediated mechanism. Endocrinology 135, 76-82.

Coverley, J. A., and Baxter, R. C. (1997). Phosphorylation of insulin-like growth factor binding proteins. Mol Cell Endocrinol 128, 1-5.

Coverley, J. A., Martin, J. L., and Baxter, R. C. (2000). The effect of phosphorylation by casein kinase 2 on the activity of insulin-like growth factor-binding protein-3. Endocrinology 141, 564-570.

DaCosta, S. A., Schumaker, L. M., and Ellis, M. J. (2000). Mannose 6-phosphate/insulin-like growth factor 2 receptor, a bona fide tumor suppressor gene or just a promising candidate? J Mammary Gland Biol Neoplasia 5, 85-94.

Daughaday, W. H., and Rotwein, P. (1989). Insulin-like growth factors I and II. Peptide, messenger ribonucleic acid and gene structures, serum, and tissue concentrations. Endocr Rev 10, 68-91.

146

Davenport, M. L., Pucilowska, J., Clemmons, D. R., Lundblad, R., Spencer, J. A., and Underwood, L. E. (1992). Tissue-specific expression of insulin-like growth factor binding protein-3 protease activity during rat pregnancy. Endocrinology 130, 2505-2512.

Davies, S. C., Wass, J. A., Ross, R. J., Cotterill, A. M., Buchanan, C. R., Coulson, V. J., and Holly, J. M. (1991). The induction of a specific protease for insulin-like growth factor binding protein-3 in the circulation during severe illness. J Endocrinol 130, 469-473.

de Boer, H. C., Preissner, K. T., Bouma, B. N., and de Groot, P. G. (1992). Binding of vitronectin-thrombin-antithrombin III complex to human endothelial cells is mediated by the heparin binding site of vitronectin. J Biol Chem 267, 2264-2268.

De Meyts, P., Wallach, B., Christoffersen, C. T., Urso, B., Gronskov, K., Latus, L. J., Yakushiji, F., Ilondo, M. M., and Shymko, R. M. (1994). The insulin-like growth factor-I receptor. Structure, ligand-binding mechanism and signal transduction. Horm Res 42, 152-169.

Deng, G., Curriden, S. A., Hu, G., Czekay, R. P., and Loskutoff, D. J. (2001). Plasminogen activator inhibitor-1 regulates cell adhesion by binding to the somatomedin B domain of vitronectin. J Cell Physiol 189, 23-33.

Deng, G., Curriden, S. A., Wang, S., Rosenberg, S., and Loskutoff, D. J. (1996). Is plasminogen activator inhibitor-1 the molecular switch that governs urokinase receptor-mediated cell adhesion and release? J Cell Biol 134, 1563-1571.

Denley, A., Bonython, E. R., Booker, G. W., Cosgrove, L. J., Forbes, B. E., Ward, C. W., and Wallace, J. C. (2004). Structural determinants for high-affinity binding of insulin-like growth factor II to insulin receptor (IR)-A, the exon 11 minus isoform of the IR. Mol Endocrinol 18, 2502-2512.

DePasquale, J. A. (1998). Cell matrix adhesions and localization of the vitronectin receptor in MCF-7 human mammary carcinoma cells. Histochem Cell Biol 110, 485-494.

Devi, G. R., Byrd, J. C., Slentz, D. H., and MacDonald, R. G. (1998). An insulin-like growth factor II (IGF-II) affinity-enhancing domain localized within extracytoplasmic repeat 13 of the IGF-II/mannose 6-phosphate receptor. Mol Endocrinol 12, 1661-1672.

Devi, G. R., Yang, D. H., Rosenfeld, R. G., and Oh, Y. (2000). Differential effects of insulin-like growth factor (IGF)-binding protein-3 and its proteolytic fragments on ligand binding, cell surface association, and IGF-I receptor signaling. Endocrinology 141, 4171-4179.

Doerr, M. E., and Jones, J. I. (1996). The roles of integrins and extracellular matrix proteins in the insulin-like growth factor I-stimulated chemotaxis of human breast cancer cells. J Biol Chem 271, 2443-2447.

147

Duan, C., and Clemmons, D. R. (1998). Differential expression and biological effects of insulin-like growth factor-binding protein-4 and -5 in vascular smooth muscle cells. J Biol Chem 273, 16836-16842.

Dufourcq, P., Couffinhal, T., Alzieu, P., Daret, D., Moreau, C., Duplaa, C., and Bonnet, J. (2002). Vitronectin is up-regulated after vascular injury and vitronectin blockade prevents neointima formation. Cardiovasc Res 53, 952-962.

Dunn, S. D. (1986). Effects of the modification of transfer buffer composition and the renaturation of proteins in gels on the recognition of proteins on Western blots by monoclonal antibodies. Anal Biochem 157, 144-153.

Dunn, S. E., Torres, J. V., Nihei, N., and Barrett, J. C. (2000). The insulin-like growth factor-1 elevates urokinase-type plasminogen activator-1 in human breast cancer cells: a new avenue for breast cancer therapy. Mol Carcinog 27, 10-17.

Durham, S. K., Suwanichkul, A., Hayes, J. D., Herington, A. C., Powell, D. R., and Campbell, P. G. (1999). The heparin binding domain of insulin-like growth factor binding protein (IGFBP)-3 increases susceptibility of IGFBP-3 to proteolysis. Horm Metab Res 31, 216-225.

Egan, S. E., Giddings, B. W., Brooks, M. W., Buday, L., Sizeland, A. M., and Weinberg, R. A. (1993). Association of Sos Ras exchange protein with Grb2 is implicated in tyrosine kinase signal transduction and transformation. Nature 363, 45-51.

Eitzman, D. T., Westrick, R. J., Nabel, E. G., and Ginsburg, D. (2000). Plasminogen activator inhibitor-1 and vitronectin promote vascular thrombosis in mice. Blood 95, 577-580.

Elgin, R. G., Busby, W. H., Jr., and Clemmons, D. R. (1987). An insulin-like growth factor (IGF) binding protein enhances the biologic response to IGF-I. Proc Natl Acad Sci U S A 84, 3254-3258.

Federici, M., Porzio, O., Zucaro, L., Giovannone, B., Borboni, P., Marini, M. A., Lauro, D., and Sesti, G. (1997). Increased abundance of insulin/IGF-I hybrid receptors in adipose tissue from NIDDM patients. Mol Cell Endocrinol 135, 41-47.

Felding-Habermann D, and Cheresh D. (1993). Vitronectin and its receptors. Current Opinion in Cell Biology 5, 864-868.

Firth, S. M., and Baxter, R. C. (1995). The role of glycosylation in the action of IGFBP-3. Prog Growth Factor Res 6, 223-229.

Firth, S. M., and Baxter, R. C. (1999). Characterisation of recombinant glycosylation variants of insulin-like growth factor binding protein-3. J Endocrinol 160, 379-387.

Firth, S. M., and Baxter, R. C. (2002). Cellular actions of the insulin-like growth factor binding proteins. Endocr Rev 23, 824-854.

148

Firth, S. M., Ganeshprasad, U., and Baxter, R. C. (1998). Structural determinants of ligand and cell surface binding of insulin-like growth factor-binding protein-3. J Biol Chem 273, 2631-2638.

Firth, S. M., Ganeshprasad, U., Poronnik, P., Cook, D. I., and Baxter, R. C. (1999). Adenoviral-mediated expression of human insulin-like growth factor-binding protein-3. Protein Expr Purif 16, 202-211.

Forbes, B. E., Hartfield, P. J., McNeil, K. A., Surinya, K. H., Milner, S. J., Cosgrove, L. J., and Wallace, J. C. (2002). Characteristics of binding of insulin-like growth factor (IGF)-I and IGF-II analogues to the type 1 IGF receptor determined by BIAcore analysis. Eur J Biochem 269, 961-968.

Forsten, K. E., Akers, R. M., and San Antonio, J. D. (2001). Insulin-like growth factor (IGF) binding protein-3 regulation of IGF-I is altered in an acidic extracellular environment. J Cell Physiol 189, 356-365.

Fottner, C., Hoeflich, A., Wolf, E., and Weber, M. M. (2004). Role of the insulin-like growth factor system in adrenocortical growth control and carcinogenesis. Horm Metab Res 36, 397-405.

Foulstone, E. J., Meadows, K. A., Holly, J. M., and Stewart, C. E. (2001). Insulin-like growth factors (IGF-I and IGF-II) inhibit C2 skeletal myoblast differentiation and enhance TNFalpha-induced apoptosis. J Cell Physiol 189, 207-215.

Fowlkes, J. L., Thrailkill, K. M., George-Nascimento, C., Rosenberg, C. K., and Serra, D. M. (1997). Heparin-binding, highly basic regions within the thyroglobulin type-1 repeat of insulin-like growth factor (IGF)-binding proteins (IGFBPs) -3, -5, and -6 inhibit IGFBP-4 degradation. Endocrinology 138, 2280-2285.

Fowlkes, J. L., Thrailkill, K. M., Serra, D. M., Suzuki, K., and Nagase, H. (1995). Matrix metalloproteinases as insulin-like growth factor binding protein-degrading proteinases. Prog Growth Factor Res 6, 255-263.

Francis, G. L., Aplin, S. E., Milner, S. J., McNeil, K. A., Ballard, F. J., and Wallace, J. C. (1993). Insulin-like growth factor (IGF)-II binding to IGF-binding proteins and IGF receptors is modified by deletion of the N-terminal hexapeptide or substitution of arginine for glutamate-6 in IGF-II. Biochem J 293, 713-719.

Francis, G. L., Read, L. C., Ballard, F. J., Bagley, C. J., Upton, F. M., Gravestock, P. M., and Wallace, J. C. (1986). Purification and partial sequence analysis of insulin-like growth factor-1 from bovine colostrum. Biochem J 233, 207-213.

Franco, C. R., Rocha, H. A., Trindade, E. S., Santos, I. A., Leite, E. L., Veiga, S. S., Nader, H. B., and Dietrich, C. P. (2001). Heparan sulfate and control of cell division: adhesion and proliferation of mutant CHO-745 cells lacking xylosyl transferase. Braz J Med Biol Res 34, 971-975.

Francois, P. P., Preissner, K. T., Herrmann, M., Haugland, R. P., Vaudaux, P., Lew, D. P., and Krause, K. H. (1999). Vitronectin interaction with glycosaminoglycans. Kinetics, structural determinants, and role in binding to endothelial cells. J Biol Chem 274, 37611-37619.

149

Frystyk, J. (2004). Free insulin-like growth factors -- measurements and relationships to growth hormone secretion and glucose homeostasis. Growth Horm IGF Res 14, 337-375.

Galvan, V., Logvinova, A., Sperandio, S., Ichijo, H., and Bredesen, D. E. (2003). Type 1 insulin-like growth factor receptor (IGF-IR) signaling inhibits apoptosis signal-regulating kinase 1 (ASK1). J Biol Chem 278, 13325-13332.

Gechtman, Z., Belleli, A., Lechpammer, S., and Shaltiel, S. (1997). The cluster of basic amino acids in vitronectin contributes to its binding of plasminogen activator inhibitor-1: evidence from thrombin-, elastase- and plasmin-cleaved vitronectins and anti-peptide antibodies. Biochem J 325, 339-349.

Gibson, A. D., Lamerdin, J. A., Zhuang, P., Baburaj, K., Serpersu, E. H., and Peterson, C. B. (1999a). Orientation of heparin-binding sites in native vitronectin. Analyses of ligand binding to the primary glycosaminoglycan-binding site indicate that putative secondary sites are not functional. J Biol Chem 274, 6432-6442.

Gibson, A. D., and Peterson, C. B. (2001). Full-length and truncated forms of vitronectin provide insight into effects of proteolytic processing on function. Biochim Biophys Acta 1545, 289-304.

Gibson, J. M., Westwood, M., Lauszus, F. F., Klebe, J. G., Flyvbjerg, A., and White, A. (1999b). Phosphorylated insulin-like growth factor binding protein 1 is increased in pregnant diabetic subjects. Diabetes 48, 321-326.

Gleeson, L. M., Chakraborty, C., McKinnon, T., and Lala, P. K. (2001). Insulin-like growth factor-binding protein 1 stimulates human trophoblast migration by signaling through alpha 5 beta 1 integrin via mitogen-activated protein Kinase pathway. J Clin Endocrinol Metab 86, 2484-2493.

Gockerman, A., Prevette, T., Jones, J. I., and Clemmons, D. R. (1995). Insulin-like growth factor (IGF)-binding proteins inhibit the smooth muscle cell migration responses to IGF-I and IGF-II. Endocrinology 136, 4168-4173.

Grimme, S., Honing, S., von Figura, K., and Schmidt, B. (2000). Endocytosis of insulin-like growth factor II by a mini-receptor based on repeat 11 of the mannose 6-phosphate/insulin-like growth factor II receptor. J Biol Chem 275, 33697-33703.

Grulich-Henn, J., Ritter, J., Mesewinkel, S., Heinrich, U., Bettendorf, M., and Preissner, K. T. (2002). Transport of insulin-like growth factor-I across endothelial cell monolayers and its binding to the subendothelial matrix. Exp Clin Endocrinol Diabetes 110, 67-73.

Grzeszkiewicz, T. M., Kirschling, D. J., Chen, N., and Lau, L. F. (2001). CYR61 stimulates human skin fibroblast migration through Integrin alpha vbeta 5 and enhances mitogenesis through integrin alpha vbeta 3, independent of its carboxyl-terminal domain. J Biol Chem 276, 21943-21950.

150

Grzeszkiewicz, T. M., Lindner, V., Chen, N., Lam, S. C., and Lau, L. F. (2002). The angiogenic factor cysteine-rich 61 (CYR61, CCN1) supports vascular smooth muscle cell adhesion and stimulates chemotaxis through integrin alpha(6)beta(1) and cell surface heparan sulfate proteoglycans. Endocrinology 143, 1441-1450.

Gui, Y., and Murphy, L. J. (2001). Insulin-like growth factor (IGF)-binding protein-3 (IGFBP-3) binds to fibronectin (FN): demonstration of IGF-I/IGFBP-3/fn ternary complexes in human plasma. J Clin Endocrinol Metab 86, 2104-2110.

Gustafsson, T., Andersson, P., and Arnqvist, H. J. (1999). Different inhibitory actions of IGFBP-1, -2 and -4 on IGF-I effects in vascular smooth muscle cells. J Endocrinol 161, 245-253.

Guvakova, M. A., Adams, J. C., and Boettiger, D. (2002). Functional role of alpha-actinin, PI 3-kinase and MEK1/2 in insulin-like growth factor I receptor kinase regulated motility of human breast carcinoma cells. J Cell Sci 115, 4149-4165.

Halevy, O., and Cantley, L. C. (2004). Differential regulation of the phosphoinositide 3-kinase and MAP kinase pathways by hepatocyte growth factor vs. insulin-like growth factor-I in myogenic cells. Exp Cell Res 297, 224-234.

Hampton, B., Burgess, W. H., Marshak, D. R., Cullen, K. J., and Perdue, J. F. (1989). Purification and characterization of an insulin-like growth factor II variant from human plasma. J Biol Chem 264, 19155-19160.

Hermanto, U., Zong, C. S., Li, W., and Wang, L. H. (2002). RACK1, an Insulin-Like Growth Factor I (IGF-I) Receptor-Interacting Protein, Modulates IGF-I-Dependent Integrin Signaling and Promotes Cell Spreading and Contact with Extracellular Matrix. Mol Cell Biol 22, 2345-2365.

Hocking, D. C., Sottile, J., Reho, T., Fassler, R., and McKeown-Longo, P. J. (1999). Inhibition of fibronectin matrix assembly by the heparin-binding domain of vitronectin. J Biol Chem 274, 27257-27264.

Hoeck, W. G., and Mukku, V. R. (1994). Identification of the major sites of phosphorylation in IGF binding protein-3. J Cell Biochem 56, 262-273.

Hollowood, A. D., Stewart, C. E., Perks, C. M., Pell, J. M., Lai, T., Alderson, D., and Holly, J. M. (2002). Evidence implicating a mid-region sequence of IGFBP-3 in its specific IGF-independent actions. J Cell Biochem 86, 583-589.

Holt, R. I., Simpson, H. L., and Sonksen, P. H. (2003). The role of the growth hormone-insulin-like growth factor axis in glucose homeostasis. Diabet Med 20, 3-15.

Hong, J., Zhang, G., Dong, F., and Rechler, M. M. (2002). Insulin-like Growth Factor (IGF)-binding Protein-3 Mutants That Do Not Bind IGF-I or IGF-II Stimulate Apoptosis in Human Prostate Cancer Cells. J Biol Chem 277, 10489-10497.

Horton, M. A. (1997). The alpha v beta 3 integrin "vitronectin receptor". Int J Biochem Cell Biol 29, 721-725.

151

Hossenlopp, P., Segovia, B., Lassarre, C., Roghani, M., Bredon, M., and Binoux, M. (1990). Evidence of enzymatic degradation of insulin-like growth factor-binding proteins in the 150K complex during pregnancy. J Clin Endocrinol Metab 71, 797-805.

Hossenlopp, P., Seurin, D., Segovia-Quinson, B., Hardouin, S., and Binoux, M. (1986). Analysis of serum insulin-like growth factor binding proteins using western blotting: use of the method for titration of the binding proteins and competitive binding studies. Anal Biochem 154, 138-143.

Hsieh, T., Gordon, R. E., Clemmons, D. R., Busby, W. H., Jr., and Duan, C. (2003). Regulation of vascular smooth muscle cell responses to insulin-like growth factor (IGF)-I by local IGF-binding proteins. J Biol Chem 278, 42886-42892.

Huang, X., Wu, J., Spong, S., and Sheppard, D. (1998). The integrin alphavbeta6 is critical for keratinocyte migration on both its known ligand, fibronectin, and on vitronectin. J Cell Sci 111, 2189-2195.

Humbel, R. E. (1990). Insulin-like growth factors I and II. Eur J Biochem 190, 445-462.

Huynh, H. (2000). Post-transcriptional and post-translational regulation of insulin-like growth factor binding protein-3 and -4 by insulin-like growth factor-I in uterine myometrial cells. Growth Horm IGF Res 10, 20-27.

Hwa, V., Oh, Y., and Rosenfeld, R. G. (1999). The insulin-like growth factor-binding protein (IGFBP) superfamily. Endocr Rev 20, 761-787.

Imai, Y., and Clemmons, D. R. (1999). Roles of phosphatidylinositol 3-kinase and mitogen-activated protein kinase pathways in stimulation of vascular smooth muscle cell migration and deoxyriboncleic acid synthesis by insulin-like growth factor-I. Endocrinology 140, 4228-4235.

Incardona, F., Lawler, J., Cataldo, D., Panet, A., Legrand, Y., Foidart, J. M., and Legrand, C. (1996). Heparin-binding domain, type 1 and type 2 repeats of thrombospondin mediate its interaction with human breast cancer cells. J Cell Biochem 62, 431-442.

Isik, F. F., Gibran, N. S., Jang, Y. C., Sandell, L., and Schwartz, S. M. (1998). Vitronectin decreases microvascular endothelial cell apoptosis. J Cell Physiol 175, 149-155.

Izumi, M., Shimo-Oka, T., Morishita, N., Ii, I., and Hayashi, M. (1988). Identification of the collagen-binding domain of vitronectin using monoclonal antibodies. Cell Struct Funct 13, 217-225.

Jacot, T. A., and Clemmons, D. R. (1998). Effect of glucose on insulin-like growth factor binding protein-4 proteolysis. Endocrinology 139, 44-50.

Jamali, R., Bao, M., and Arnqvist, H. J. (2003). IGF-I but not insulin inhibits apoptosis at a low concentration in vascular smooth muscle cells. J Endocrinol 179, 267-274.

152

Jang, Y. C., Tsou, R., Gibran, N. S., and Isik, F. F. (2000). Vitronectin deficiency is associated with increased wound fibrinolysis and decreased microvascular angiogenesis in mice. Surgery 127, 696-704.

Jenne, D., Hille, A., Stanley, K. K., and Huttner, W. B. (1989). Sulfation of two tyrosine-residues in human complement S-protein (vitronectin). Eur J Biochem 185, 391-395.

Jones, J. I., Busby, W. H., Jr., Wright, G., and Clemmons, D. R. (1993a). Human IGFBP-1 is phosphorylated on 3 serine residues: effects of site-directed mutagenesis of the major phosphoserine. Growth Regul 3, 37-40.

Jones, J. I., and Clemmons, D. R. (1995). Insulin-like growth factors and their binding proteins: biological actions. Endocr Rev 16, 3-34.

Jones, J. I., D'Ercole, A. J., Camacho-Hubner, C., and Clemmons, D. R. (1991). Phosphorylation of insulin-like growth factor (IGF)-binding protein 1 in cell culture and in vivo: effects on affinity for IGF-I. Proc Natl Acad Sci U S A 88, 7481-7485.

Jones, J. I., Doerr, M. E., and Clemmons, D. R. (1995). Cell migration: interactions among integrins, IGFs and IGFBPs. Prog Growth Factor Res 6, 319-327.

Jones, J. I., Gockerman, A., Busby, W. H., Jr., Wright, G., and Clemmons, D. R. (1993b). Insulin-like growth factor binding protein 1 stimulates cell migration and binds to the alpha 5 beta 1 integrin by means of its Arg-Gly-Asp sequence. Proc Natl Acad Sci U S A 90, 10553-10557.

Jones, J. I., Prevette, T., Gockerman, A., and Clemmons, D. R. (1996). Ligand occupancy of the alpha-V-beta3 integrin is necessary for smooth muscle cells to migrate in response to insulin-like growth factor. Proc Natl Acad Sci U S A 93, 2482-2487.

Jones, J. L., and Walker, R. A. (1999). Integrins: a role as cell signalling molecules. Mol Pathol 52, 208-213.

Kabir-Salmani, M., Shiokawa, S., Akimoto, Y., Sakai, K., Nagamatsu, S., Nakamura, Y., Lotfi, A., Kawakami, H., and Iwashita, M. (2003). alpha(v)beta(3) Integrin Signaling Pathway Is Involved in Insulin-Like Growth Factor I-Stimulated Human Extravillous Trophoblast Cell Migration. Endocrinology 144, 1620-1630.

Kang, J. X., Li, Y., and Leaf, A. (1997). Mannose-6-phosphate/insulin-like growth factor-II receptor is a receptor for retinoic acid. Proc Natl Acad Sci U S A 94, 13671-13676.

Kanse, S. M., Kost, C., Wilhelm, O. G., Andreasen, P. A., and Preissner, K. T. (1996). The urokinase receptor is a major vitronectin-binding protein on endothelial cells. Exp Cell Res 224, 344-353.

Kato, H., Faria, T. N., Stannard, B., Roberts, C. T., Jr., and LeRoith, D. (1994). Essential role of tyrosine residues 1131, 1135, and 1136 of the insulin-like growth factor-I (IGF-I) receptor in IGF-I action. Mol Endocrinol 8, 40-50.

153

Kiess, W., Yang, Y., Kessler, U., and Hoeflich, A. (1994). Insulin-like growth factor II (IGF-II) and the IGF-II/mannose-6-phosphate receptor: the myth continues. Horm Res 41, 66-73.

Kim, H. S., Ingermann, A. R., Tsubaki, J., Twigg, S. M., Walker, G. E., and Oh, Y. (2004). Insulin-like growth factor-binding protein 3 induces caspase-dependent apoptosis through a death receptor-mediated pathway in MCF-7 human breast cancer cells. Cancer Res 64, 2229-2237.

Kim, H. S., Nagalla, S. R., Oh, Y., Wilson, E., Roberts, C. T., Jr., and Rosenfeld, R. G. (1997). Identification of a family of low-affinity insulin-like growth factor binding proteins (IGFBPs): characterization of connective tissue growth factor as a member of the IGFBP superfamily. Proc Natl Acad Sci U S A 94, 12981-12986.

Kim, J. J., Park, B. C., Kido, Y., and Accili, D. (2001). Mitogenic and metabolic effects of type I IGF receptor overexpression in insulin receptor-deficient hepatocytes. Endocrinology 142, 3354-3360.

Kjoller, L., Kanse, S. M., Kirkegaard, T., Rodenburg, K. W., Ronne, E., Goodman, S. L., Preissner, K. T., Ossowski, L., and Andreasen, P. A. (1997). Plasminogen activator inhibitor-1 represses integrin- and vitronectin-mediated cell migration independently of its function as an inhibitor of plasminogen activation. Exp Cell Res 232, 420-429.

Klemke, R. L., Yebra, M., Bayna, E. M., and Cheresh, D. A. (1994). Receptor tyrosine kinase signaling required for integrin alpha v beta 5-directed cell motility but not adhesion on vitronectin. J Cell Biol 127, 859-866.

Koedam, J. A., Hoogerbrugge, C. M., van Buul-Offers, S. C., and Van den Brande, J. L. (1997). Solid-phase assay for insulin-like growth factor (IGF) binding to IGF-binding protein-3: application to the study of the effects of antibodies and heparin. J Endocrinol 153, 87-97.

Koistinen, H., Paju, A., Koistinen, R., Finne, P., Lovgren, J., Wu, P., Seppala, M., and Stenman, U. H. (2002). Prostate-specific antigen and other prostate-derived proteases cleave IGFBP-3, but prostate cancer is not associated with proteolytically cleaved circulating IGFBP-3. Prostate 50, 112-118.

Koistinen, R., Angervo, M., Leinonen, P., Hakala, T., and Seppala, M. (1993). Phosphorylation of insulin-like growth factor-binding protein-1 increases in human amniotic fluid and decidua from early to late pregnancy. Clin Chim Acta 215, 189-199.

Kornfeld, S. (1992). Structure and function of the mannose 6-phosphate/insulinlike growth factor II receptors. Annu Rev Biochem 61, 307-330.

Kost, C., Stuber, W., Ehrlich, H. J., Pannekoek, H., and Preissner, K. T. (1992). Mapping of binding sites for heparin, plasminogen activator inhibitor-1, and plasminogen to vitronectin's heparin-binding region reveals a novel vitronectin-dependent feedback mechanism for the control of plasmin formation. J Biol Chem 267, 12098-12105.

154

Kovacina, K. S., Steele-Perkins, G., Purchio, A. F., Lioubin, M., Miyazono, K., Heldin, C. H., and Roth, R. A. (1989). Interactions of recombinant and platelet transforming growth factor-beta 1 precursor with the insulin-like growth factor II/mannose 6-phosphate receptor. Biochem Biophys Res Commun 160, 393-403.

Krane, J. F., Murphy, D. P., Carter, D. M., and Krueger, J. G. (1991). Synergistic effects of epidermal growth factor (EGF) and insulin-like growth factor I/somatomedin C (IGF-I) on keratinocyte proliferation may be mediated by IGF-I transmodulation of the EGF receptor. J Invest Dermatol 96, 419-424.

Kricker, J. A., Towne, C. L., Firth, S. M., Herington, A. C., and Upton, Z. (2003). Structural and Functional Evidence for the Interaction of Insulin-Like Growth Factors (IGFs) and IGF Binding Proteins with Vitronectin. Endocrinology 144, 2807-2815.

Kubler, B., Cowell, S., Zapf, J., and Braulke, T. (1998). Proteolysis of insulin-like growth factor binding proteins by a novel 50-kilodalton metalloproteinase in human pregnancy serum. Endocrinology 139, 1556-1563.

Kuemmerle, J. F., and Bushman, T. L. (1998). IGF-I stimulates intestinal muscle cell growth by activating distinct PI 3-kinase and MAP kinase pathways. Am J Physiol 275, G151-158.

Laemmli, U. K. (1970). Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227, 680-685.

Lalou, C., Lassarre, C., and Binoux, M. (1995). A proteolytic fragment of insulin-like growth factor (IGF) binding protein-3 that fails to bind IGF is a cell growth inhibitor. Prog Growth Factor Res 6, 311-316.

Lambrecht, V., Le Bourhis, X., Toillon, R. A., Boilly, B., and Hondermarck, H. (1998). Alterations in both heparan sulfate proteoglycans and mitogenic activity of fibroblast growth factor-2 are triggered by inhibitors of proliferation in normal and breast cancer epithelial cells. Exp Cell Res 245, 239-244.

Laron, Z. (2001). Insulin-like growth factor 1 (IGF-1): a growth hormone. Mol Pathol 54, 311-316.

Laureys, G., Barton, D. E., Ullrich, A., and Francke, U. (1988). Chromosomal mapping of the gene for the type II insulin-like growth factor receptor/cation-independent mannose 6-phosphate receptor in man and mouse. Genomics 3, 224-229.

Laursen, L. S., Overgaard, M. T., Soe, R., Boldt, H. B., Sottrup-Jensen, L., Giudice, L. C., Conover, C. A., and Oxvig, C. (2001). Pregnancy-associated plasma protein-A (PAPP-A) cleaves insulin-like growth factor binding protein (IGFBP)-5 independent of IGF: implications for the mechanism of IGFBP-4 proteolysis by PAPP-A. FEBS Lett 504, 36-40.

Lawler, J., Weinstein, R., and Hynes, R. O. (1988). Cell attachment to thrombospondin: the role of ARG-GLY-ASP, calcium, and integrin receptors. J Cell Biol 107, 2351-2361.

155

Lawrence, D. A., Berkenpas, M. B., Palaniappan, S., and Ginsburg, D. (1994). Localization of vitronectin binding domain in plasminogen activator inhibitor-1. J Biol Chem 269, 15223-15228.

Leavesley, D. I., Ferguson, G. D., Wayner, E. A., and Cheresh, D. A. (1992). Requirement of the integrin beta 3 subunit for carcinoma cell spreading or migration on vitronectin and fibrinogen. J Cell Biol 117, 1101-1107.

Leavesley, D. I., Schwartz, M. A., Rosenfeld, M., and Cheresh, D. A. (1993). Integrin beta 1- and beta 3-mediated endothelial cell migration is triggered through distinct signaling mechanisms. J Cell Biol 121, 163-170.

Lee, S. J., and Nathans, D. (1988). Proliferin secreted by cultured cells binds to mannose 6-phosphate receptors. J Biol Chem 263, 3521-3527.

LeRoith, D., and Roberts, C. T., Jr. (2003). The insulin-like growth factor system and cancer. Cancer Lett 195, 127-137.

LeRoith, D., Werner, H., Beitner-Johnson, D., and Roberts, C. T., Jr. (1995a). Molecular and cellular aspects of the insulin-like growth factor I receptor. Endocr Rev 16, 143-163.

LeRoith, D., Werner, H., Neuenschwander, S., Kalebic, T., and Helman, L. J. (1995b). The role of the insulin-like growth factor-I receptor in cancer. Ann N Y Acad Sci 766, 402-408.

Leventhal, P. S., and Feldman, E. L. (1997). Insulin-like growth factors as regulators of cell motility: Signaling mechanisms. Trends in Endocrinology & Metabolism 8, 1-6.

Lin, C. G., Leu, S. J., Chen, N., Tebeau, C. M., Lin, S. X., Yeung, C. Y., and Lau, L. F. (2003). CCN3 (NOV) is a novel angiogenic regulator of the CCN protein family. J Biol Chem 278, 24200-24208.

Linnell, J., Groeger, G., and Hassan, A. B. (2001). Real time kinetics of insulin-like growth factor II (IGF-II) interaction with the IGF-II/mannose 6-phosphate receptor: the effects of domain 13 and pH. J Biol Chem 276, 23986-23991.

Lowenstein, E. J., Daly, R. J., Batzer, A. G., Li, W., Margolis, B., Lammers, R., Ullrich, A., Skolnik, E. Y., Bar-Sagi, D., and Schlessinger, J. (1992). The SH2 and SH3 domain-containing protein GRB2 links receptor tyrosine kinases to ras signaling. Cell 70, 431-442.

Maile, L. A., Badley-Clarke, J., and Clemmons, D. R. (2001). Structural analysis of the role of the beta 3 subunit of the alpha V beta 3 integrin in IGF-I signaling. J Cell Sci 114, 1417-1425.

Maile, L. A., and Clemmons, D. R. (2002). The alphaVbeta3 integrin regulates insulin-like growth factor I (IGF-I) receptor phosphorylation by altering the rate of recruitment of the Src-homology 2-containing phosphotyrosine phosphatase-2 to the activated IGF-I receptor. Endocrinology 143, 4259-4264.

156

Maile, L. A., Gill, Z. P., Perks, C. M., and Holly, J. M. (1999). The role of cell surface attachment and proteolysis in the insulin-like growth factor (IGF)-independent effects of IGF-binding protein-3 on apoptosis in breast epithelial cells. Endocrinology 140, 4040-4045.

Maile, L. A., and Holly, J. M. (1999). Insulin-like growth factor binding protein (IGFBP) proteolysis: occurrence, identification, role and regulation. Growth Horm IGF Res 9, 85-95.

Maile, L. A., Imai, Y., Clarke, J. B., and Clemmons, D. R. (2002). Insulin-like growth factor I increases alpha Vbeta 3 affinity by increasing the amount of integrin-associated protein that is associated with non-raft domains of the cellular membrane. J Biol Chem 277, 1800-1805.

Marikovsky, M., Vogt, P., Eriksson, E., Rubin, J. S., Taylor, W. G., Joachim, S., and Klagsbrun, M. (1996). Wound fluid-derived heparin-binding EGF-like growth factor (HB-EGF) is synergistic with insulin-like growth factor-I for Balb/MK keratinocyte proliferation. J Invest Dermatol 106, 616-621.

Marinaro, J. A., Neumann, G. M., Russo, V. C., Leeding, K. S., and Bach, L. A. (2000). O-glycosylation of insulin-like growth factor (IGF) binding protein-6 maintains high IGF-II binding affinity by decreasing binding to glycosaminoglycans and susceptibility to proteolysis. Eur J Biochem 267, 5378-5386.

Massova, I., Kotra, L. P., Fridman, R., and Mobashery, S. (1998). Matrix metalloproteinases: structures, evolution, and diversification. Faseb J 12, 1075-1095.

Mazerbourg, S., Zapf, J., Bar, R. S., Brigstock, D. R., and Monget, P. (2000). Insulin-like growth factor (IGF)-binding protein-4 proteolytic degradation in bovine, equine, and porcine preovulatory follicles: regulation by IGFs and heparin-binding domain-containing peptides. Biol Reprod 63, 390-400.

McCusker, R. H., Busby, W. H., Dehoff, M. H., Camacho-Hubner, C., and Clemmons, D. R. (1991). Insulin-like growth factor (IGF) binding to cell monolayers is directly modulated by the addition of IGF-binding proteins. Endocrinology 129, 939-949.

McCusker, R. H., Camacho-Hubner, C., Bayne, M. L., Cascieri, M. A., and Clemmons, D. R. (1990). Insulin-like growth factor (IGF) binding to human fibroblast and glioblastoma cells: the modulating effect of cell released IGF binding proteins (IGFBPs). Journal of Cellular Physiology 144, 244-253.

McGuire, S. E., Hilsenbeck, S. G., Figueroa, J. A., Jackson, J. G., and Yee, D. (1994). Detection of insulin-like growth factor binding proteins (IGFBPs) by ligand blotting in breast cancer tissues. Cancer Lett 77, 25-32.

McKinnon, T., Chakraborty, C., Gleeson, L. M., Chidiac, P., and Lala, P. K. (2001). Stimulation of human extravillous trophoblast migration by IGF-II is mediated by IGF type 2 receptor involving inhibitory G protein(s) and phosphorylation of MAPK. J Clin Endocrinol Metab 86, 3665-3674.

157

Meyer, G. E., Shelden, E., Kim, B., and Feldman, E. L. (2001). Insulin-like growth factor I stimulates motility in human neuroblastoma cells. Oncogene 20, 7542-7550.

Miell, J. P., Jauniaux, E., Langford, K. S., Westwood, M., White, A., and Jones, J. S. (1997). Insulin-like growth factor binding protein concentration and post-translational modification in embryological fluid. Mol Hum Reprod 3, 343-349.

Mira, E., Manes, S., Lacalle, R. A., Marquez, G., and Martinez, A. C. (1999). Insulin-like growth factor I-triggered cell migration and invasion are mediated by matrix metalloproteinase-9. Endocrinology 140, 1657-1664.

Mishra, S., and Murphy, L. J. (2003). Phosphorylation of insulin-like growth factor (IGF) binding protein-3 by breast cancer cell membranes enhances IGF-I binding. Endocrinology 144, 4042-4050.

Mohan, S., Nakao, Y., Honda, Y., Landale, E., Leser, U., Dony, C., Lang, K., and Baylink, D. J. (1995). Studies on the mechanisms by which insulin-like growth factor (IGF) binding protein-4 (IGFBP-4) and IGFBP-5 modulate IGF actions in bone cells. J Biol Chem 270, 20424-20431.

Monget, P., Mazerbourg, S., Delpuech, T., Maurel, M. C., Maniere, S., Zapf, J., Lalmanach, G., Oxvig, C., and Overgaard, M. T. (2003). Pregnancy-Associated Plasma Protein-A Is Involved in Insulin-Like Growth Factor Binding Protein-2 (IGFBP-2) Proteolytic Degradation in Bovine and Porcine Preovulatory Follicles: Identification of Cleavage Site and Characterization of IGFBP-2 Degradation. Biol Reprod 68, 77-86.

Morini, M., Mottolese, M., Ferrari, N., Ghiorzo, F., Buglioni, S., Mortarini, R., Noonan, D. M., Natali, P. G., and Albini, A. (2000). The alpha 3 beta 1 integrin is associated with mammary carcinoma cell metastasis, invasion, and gelatinase B (MMP-9) activity. Int J Cancer 87, 336-342.

Morris, C. A., Underwood, P. A., Bean, P. A., Sheehan, M., and Charlesworth, J. A. (1994). Relative topography of biologically active domains of human vitronectin. Evidence from monoclonal antibody epitope and denaturation studies. J Biol Chem 269, 23845-23852.

Murashita, M. M., Russo, V. C., Edmondson, S. R., Wraight, C. J., and Werther, G. A. (1995). Identification of insulin-like growth factor binding proteins from cultured human epidermal keratinocytes. J Cell Physiol 163, 339-345.

Nakashima, N., Miyazaki, K., Ishikawa, M., Yatohgo, T., Ogawa, H., Uchibori, H., Matsumoto, I., Seno, N., and Hayashi, M. (1992). Vitronectin diversity in evolution but uniformity in ligand binding and size of the core polypeptide. Biochim Biophys Acta 1120, 1-10.

Nam, T., Moralez, A., and Clemmons, D. (2002). Vitronectin Binding to IGF Binding Protein-5 (IGFBP-5) Alters IGFBP-5 Modulation of IGF-I Actions. Endocrinology 143, 30-36.

158

Nam, T. J., Busby, W., Jr., and Clemmons, D. R. (1997). Insulin-like growth factor binding protein-5 binds to plasminogen activator inhibitor-I. Endocrinology 138, 2972-2978.

Neumann, G. M., Marinaro, J. A., and Bach, L. A. (1998). Identification of O-glycosylation sites and partial characterization of carbohydrate structure and disulfide linkages of human insulin-like growth factor binding protein 6. Biochemistry 37, 6572-6585.

Nicholas, B., Scougall, R. K., Armstrong, D. G., and Webb, R. (2002). Changes in insulin-like growth factor binding protein (IGFBP) isoforms during bovine follicular development. Reproduction 124, 439-446.

Niesler, C. U., Urso, B., Prins, J. B., and Siddle, K. (2000). IGF-I inhibits apoptosis induced by serum withdrawal, but potentiates TNF-alpha-induced apoptosis, in 3T3-L1 preadipocytes. J Endocrinol 167, 165-174.

Nykjaer, A., Christensen, E. I., Vorum, H., Hager, H., Petersen, C. M., Roigaard, H., Min, H. Y., Vilhardt, F., Moller, L. B., Kornfeld, S., and Gliemann, J. (1998). Mannose 6-phosphate/insulin-like growth factor-II receptor targets the urokinase receptor to lysosomes via a novel binding interaction. J Cell Biol 141, 815-828.

Oates, A. J., Schumaker, L. M., Jenkins, S. B., Pearce, A. A., DaCosta, S. A., Arun, B., and Ellis, M. J. (1998). The mannose 6-phosphate/insulin-like growth factor 2 receptor (M6P/IGF2R), a putative breast tumor suppressor gene. Breast Cancer Res Treat 47, 269-281.

Oh, C. D., and Chun, J. S. (2003). Signaling mechanisms leading to the regulation of differentiation and apoptosis of articular chondrocytes by insulin-like growth factor-1. J Biol Chem 278, 36563-36571.

Oh, Y., Muller, H. L., Ng, L., and Rosenfeld, R. G. (1995). Transforming growth factor-beta-induced cell growth inhibition in human breast cancer cells is mediated through insulin-like growth factor-binding protein-3 action. J Biol Chem 270, 13589-13592.

Pandini, G., Frasca, F., Mineo, R., Sciacca, L., Vigneri, R., and Belfiore, A. (2002). Insulin/lGF-l hybrid receptors have different biological characteristics depending on the insulin receptor isoform involved. J Biol Chem.

Parker, A., Clarke, J. B., Busby, W. H., Jr., and Clemmons, D. R. (1996). Identification of the extracellular matrix binding sites for insulin-like growth factor-binding protein 5. J Biol Chem 271, 13523-13529.

Parker, A., Rees, C., Clarke, J., Busby, W. H., Jr., and Clemmons, D. R. (1998). Binding of insulin-like growth factor (IGF)-binding protein-5 to smooth-muscle cell extracellular matrix is a major determinant of the cellular response to IGF-I. Mol Biol Cell 9, 2383-2392.

Pattison, S. T., Fanayan, S., and Martin, J. L. (1999). Insulin-like growth factor binding protein-3 is secreted as a phosphoprotein by human breast cancer cells. Mol Cell Endocrinol 156, 131-139.

159

Pereira, J. J., Meyer, T., Docherty, S. E., Reid, H. H., Marshall, J., Thompson, E. W., Rossjohn, J., and Price, J. T. (2004). Bimolecular interaction of insulin-like growth factor (IGF) binding protein-2 with alphavbeta3 negatively modulates IGF-I-mediated migration and tumor growth. Cancer Res 64, 977-984.

Perks, C. M., and Holly, J. M. (2003). The insulin-like growth factor (IGF) family and breast cancer. Breast Dis 18, 45-60.

Perks, C. M., McCaig, C., Clarke, J. B., Clemmons, D. R., and Holly, J. M. (2002). Effects of a non-IGF binding mutant of IGFBP-5 on cell death in human breast cancer cells. Biochem Biophys Res Commun 294, 995-1000.

Perks, C. M., McCaig, C., and Holly, J. M. (2000). Differential insulin-like growth factor (IGF)-independent interactions of IGF binding protein-3 and IGF binding protein-5 on apoptosis in human breast cancer cells. Involvement of the mitochondria. J Cell Biochem 80, 248-258.

Podar, K., Tai, Y. T., Lin, B. K., Narsimhan, R. P., Sattler, M., Kijima, T., Salgia, R., Gupta, D., Chauhan, D., and Anderson, K. C. (2002). Vascular endothelial growth factor-induced migration of multiple myeloma cells is associated with beta 1 integrin- and phosphatidylinositol 3-kinase-dependent PKC alpha activation. J Biol Chem 277, 7875-7881.

Podor, T. J., Campbell, S., Chindemi, P., Foulon, D. M., Farrell, D. H., Walton, P. D., Weitz, J. I., and Peterson, C. B. (2001). Incorporation of vitronectin into fibrin clots: Evidence for a binding interaction between vitronectin and aA/a' fibrinogen. J Biol Chem 14, 14.

Podor, T. J., Peterson, C. B., Lawrence, D. A., Stefansson, S., Shaughnessy, S. G., Foulon, D. M., Butcher, M., and Weitz, J. I. (2000). Type 1 plasminogen activator inhibitor binds to fibrin via vitronectin. J Biol Chem 275, 19788-19794.

Pozios, K. C., Ding, J., Degger, B., Upton, Z., and Duan, C. (2001). IGFs stimulate zebrafish cell proliferation by activating MAP kinase and PI3-kinase-signaling pathways. Am J Physiol Regul Integr Comp Physiol 280, R1230-1239.

Puglianiello, A., Germani, D., Rossi, P., and Cianfarani, S. (2000). IGF-I stimulates chemotaxis of human neuroblasts. Involvement of type 1 IGF receptor, IGF binding proteins, phosphatidylinositol-3 kinase pathway and plasmin system. J Endocrinol 165, 123-131.

Rajaram, S., Baylink, D. J., and Mohan, S. (1997). Insulin-like growth factor-binding proteins in serum and other biological fluids: regulation and functions. Endocr Rev 18, 801-831.

Ramachandrula, A., Tiku, K., and Tiku, M. L. (1992). Tripeptide RGD-dependent adhesion of articular chondrocytes to synovial fibroblasts. J Cell Sci 101 ( Pt 4), 859-871.

Rees, C., and Clemmons, D. R. (1998). Inhibition of IGFBP-5 binding to extracellular matrix and IGF-I-stimulated DNA synthesis by a peptide fragment of IGFBP-5. J Cell Biochem 71, 375-381.

160

Rees, C., Clemmons, D. R., Horvitz, G. D., Clarke, J. B., and Busby, W. H. (1998). A protease-resistant form of insulin-like growth factor (IGF) binding protein 4 inhibits IGF-1 actions. Endocrinology 139, 4182-4188.

Remacle-Bonnet, M. M., Garrouste, F. L., and Pommier, G. J. (1997). Surface-bound plasmin induces selective proteolysis of insulin-like-growth-factor (IGF)-binding protein-4 (IGFBP-4) and promotes autocrine IGF-II bio-availability in human colon-carcinoma cells. International Journal of Cancer 72, 835-843.

Rhodes, C. J. (2000). IGF-I and GH post-receptor signaling mechanisms for pancreatic beta-cell replication. J Mol Endocrinol 24, 303-311.

Ria, R., Vacca, A., Ribatti, D., Di Raimondo, F., Merchionne, F., and Dammacco, F. (2002). alpha(v)beta(3) integrin engagement enhances cell invasiveness in human multiple myeloma. Haematologica 87, 836-845.

Rickard, K. A., Shoji, S., Spurzem, J. R., and Rennard, S. I. (1991). Attachment characteristics of bovine bronchial epithelial cells to extracellular matrix components. Am J Respir Cell Mol Biol 4, 440-448.

Ricort, J. M., and Binoux, M. (2004). Insulin-like growth factor binding protein-3 stimulates phosphatidylinositol 3-kinase in MCF-7 breast carcinoma cells. Biochem Biophys Res Commun 314, 1044-1049.

Rosenblatt, S., Bassuk, J. A., Alpers, C. E., Sage, E. H., Timpl, R., and Preissner, K. T. (1997). Differential modulation of cell adhesion by interaction between adhesive and counter-adhesive proteins: characterization of the binding of vitronectin to osteonectin (BM40, SPARC). Biochem J 324, 311-319.

Rosenfeld, R. G., Hwa, V., Wilson, L., Lopez-Bermejo, A., Buckway, C., Burren, C., Choi, W. K., Devi, G., Ingermann, A., Graham, D., et al. (1999). The insulin-like growth factor binding protein superfamily: new perspectives. Pediatrics 104, 1018-1021.

Roth, B. V., Burgisser, D. M., Luthi, C., and Humbel, R. E. (1991). Mutants of human insulin-like growth factor II: expression and characterization of analogs with a substitution of TYR27 and/or a deletion of residues 62-67. Biochem Biophys Res Commun 181, 907-914.

Rotwein, P. (2003). Insulin-like growth factor action and skeletal muscle growth, an in vivo perspective. Growth Horm IGF Res 13, 303-305.

Russo, V. C., Bach, L. A., Fosang, A. J., Baker, N. L., and Werther, G. A. (1997). Insulin-like growth factor binding protein-2 binds to cell surface proteoglycans in the rat brain olfactory bulb. Endocrinology 138, 4858-4867.

Sadowski, T., Dietrich, S., Koschinsky, F., and Sedlacek, R. (2003). Matrix Metalloproteinase 19 Regulates Insulin-like Growth Factor-mediated Proliferation, Migration, and Adhesion in Human Keratinocytes through Proteolysis of Insulin-like Growth Factor Binding Protein-3. Mol Biol Cell 14, 4569-4580.

161

Sakai, K., D'Ercole, A. J., Murphy, L. J., and Clemmons, D. R. (2001). Physiological differences in insulin-like growth factor binding protein-1 (IGFBP-1) phosphorylation in IGFBP-1 transgenic mice. Diabetes 50, 32-38.

Sakano, K., Enjoh, T., Numata, F., Fujiwara, H., Marumoto, Y., Higashihashi, N., Sato, Y., Perdue, J. F., and Fujita-Yamaguchi, Y. (1991). The design, expression, and characterization of human insulin-like growth factor II (IGF-II) mutants specific for either the IGF-II/cation-independent mannose 6-phosphate receptor or IGF-I receptor. J Biol Chem 266, 20626-20635.

Schedlich, L. J., Nilsen, T., John, A. P., Jans, D. A., and Baxter, R. C. (2003). Phosphorylation of insulin-like growth factor binding protein-3 by deoxyribonucleic acid-dependent protein kinase reduces ligand binding and enhances nuclear accumulation. Endocrinology 144, 1984-1993.

Schmid, C., Ghirlanda-Keller, C., and Zapf, J. (2001). Effects of IGF-I and -II, IGF binding protein-3 (IGFBP-3), and transforming growth factor-beta (TGF-beta) on growth and apoptosis of human osteosarcoma Saos-2/B-10 cells: lack of IGF-independent IGFBP-3 effects. Eur J Endocrinol 145, 213-221.

Schmidt, B., Kiecke-Siemsen, C., Waheed, A., Braulke, T., and von Figura, K. (1995). Localization of the insulin-like growth factor II binding site to amino acids 1508-1566 in repeat 11 of the mannose 6-phosphate/insulin-like growth factor II receptor. J Biol Chem 270, 14975-14982.

Schnarr, B., Strunz, K., Ohsam, J., Benner, A., Wacker, J., and Mayer, D. (2000). Down-regulation of insulin-like growth factor-I receptor and insulin receptor substrate-1 expression in advanced human breast cancer. Int J Cancer 89, 506-513.

Schoppet, M., Chavakis, T., Al-Fakhri, N., Kanse, S. M., and Preissner, K. T. (2002). Molecular Interactions and Functional Interference between Vitronectin and Transforming Growth Factor-beta. Lab Invest 82, 37-46.

Schumacher, R., Mosthaf, L., Schlessinger, J., Brandenburg, D., and Ullrich, A. (1991). Insulin and insulin-like growth factor-1 binding specificity is determined by distinct regions of their cognate receptors. J Biol Chem 266, 19288-19295.

Schutt, B. S., Langkamp, M., Rauschnabel, U., Ranke, M. B., and Elmlinger, M. W. (2004). Integrin-mediated action of insulin-like growth factor binding protein-2 in tumor cells. J Mol Endocrinol 32, 859-868.

Schvartz, I., Kreizman, T., Brumfeld, V., Gechtman, Z., Seger, D., and Shaltiel, S. (2002). The PKA Phosphorylation of Vitronectin: Effect on Conformation and Function. Arch Biochem Biophys 397, 246-252.

Schvartz, I., Seger, D., and Shaltiel, S. (1999). Vitronectin. Int J Biochem Cell Biol 31, 539-544.

Seger, D., Seger, R., and Shaltiel, S. (2001). The CK2 Phosphorylation of Vitronectin. Promotion of cell adhesion via the alpha vbeta 3-phosphatidylinositol 3-kinase pathway. J Biol Chem 276, 16998-17006.

162

Seiffert, D. (1997). The glycosaminoglycan binding site governs ligand binding to the somatomedin B domain of vitronectin. J Biol Chem 272, 9971-9978.

Seiffert, D., and Loskutoff, D. J. (1991). Evidence that type 1 plasminogen activator inhibitor binds to the somatomedin B domain of vitronectin. J Biol Chem 266, 2824-2830.

Seiffert, D., and Schleef, R. R. (1996). Two functionally distinct pools of vitronectin (Vn) in the blood circulation: identification of a heparin-binding competent population of Vn within platelet alpha-granules. Blood 88, 552-560.

Seiffert, D., and Smith, J. W. (1997). The cell adhesion domain in plasma vitronectin is cryptic. J Biol Chem 272, 13705-13710.

Shand, J. H., Beattie, J., Song, H., Phillips, K., Kelly, S. M., Flint, D. J., and Allan, G. J. (2003). Specific Amino Acid Substitutions Determine the Differential Contribution of the N- and C-terminal Domains of Insulin-like Growth Factor (IGF)-binding Protein-5 in Binding IGF-I. J Biol Chem 278, 17859-17866.

Sheikh, M. S., Shao, Z. M., Chen, J. C., Clemmons, D. R., Roberts, C. T., Jr., LeRoith, D., and Fontana, J. A. (1992). Insulin-like growth factor binding protein-5 gene expression is differentially regulated at a post-transcriptional level in retinoic acid-sensitive and resistant MCF-7 human breast carcinoma cells. Biochem Biophys Res Commun 188, 1122-1130.

Siddals, K. W., Westwood, M., Gibson, J. M., and White, A. (2002). IGF-binding protein-1 inhibits IGF effects on adipocyte function: implications for insulin-like actions at the adipocyte. J Endocrinol 174, 289-297.

Silvestri, I., Longanesi Cattani, I., Franco, P., Pirozzi, G., Botti, G., Stoppelli, M. P., and Carriero, M. V. (2002). Engaged urokinase receptors enhance tumor breast cell migration and invasion by upregulating alphavbeta5 vitronectin receptor cell surface expression. Int J Cancer 102, 562-571.

Singh, B., Charkowicz, D., and Mascarenhas, D. (2004). Insulin-like growth factor-independent effects mediated by a C-terminal metal-binding domain of insulin-like growth factor binding protein-3. J Biol Chem 279, 477-487.

Sommer, A., Spratt, S. K., Tatsuno, G. P., Tressel, T., Lee, R., and Maack, C. A. (1993). Properties of glycosylated and non-glycosylated human recombinant IGF binding protein-3 (IGFBP-3). Growth Regul 3, 46-49.

Song, H., Beattie, J., Campbell, I. W., and Allan, G. J. (2000). Overlap of IGF- and heparin-binding sites in rat IGF-binding protein-5. Journal of Molecular Endocrinology 24, 43-51.

Soos, M. A., Field, C. E., and Siddle, K. (1993). Purified hybrid insulin/insulin-like growth factor-I receptors bind insulin-like growth factor-I, but not insulin, with high affinity. Biochem J 290, 419-426.

163

Soos, M. A., Whittaker, J., Lammers, R., Ullrich, A., and Siddle, K. (1990). Receptors for insulin and insulin-like growth factor-I can form hybrid dimers. Characterisation of hybrid receptors in transfected cells. Biochem J 270, 383-390.

Stahl, A., and Mueller, B. M. (1997). Melanoma cell migration on vitronectin: regulation by components of the plasminogen activation system. Int J Cancer 71, 116-122.

Stanley, K. K. (1986). Homology with hemopexin suggests a possible scavenging function for S-protein/vitronectin. FEBS Lett 199, 249-253.

Stepanova, V., Jerke, U., Sagach, V., Lindschau, C., Dietz, R., Haller, H., and Dumler, I. (2002). Urokinase-dependent human vascular smooth muscle cell adhesion requires selective vitronectin phosphorylation by ectoprotein kinase CK2. J Biol Chem 277, 10265-10272.

Stewart, C. E., James, P. L., Fant, M. E., and Rotwein, P. (1996). Overexpression of insulin-like growth factor-II induces accelerated myoblast differentiation. J Cell Physiol 169, 23-32.

Stone, J. D., Cochran, J. R., and Stern, L. J. (2001). T-cell activation by soluble MHC oligomers can be described by a two-parameter binding model. Biophys J 81, 2547-2557.

Taliana, L., Evans, M. D., Dimitrijevich, S. D., and Steele, J. G. (2000). Vitronectin or fibronectin is required for corneal fibroblast-seeded collagen gel contraction. Invest Ophthalmol Vis Sci 41, 103-109.

Twigg, S. M., Chen, M. M., Joly, A. H., Chakrapani, S. D., Tsubaki, J., Kim, H. S., Oh, Y., and Rosenfeld, R. G. (2001). Advanced glycosylation end products up-regulate connective tissue growth factor (insulin-like growth factor-binding protein-related protein 2) in human fibroblasts: a potential mechanism for expansion of extracellular matrix in diabetes mellitus. Endocrinology 142, 1760-1769.

Twigg, S. M., Kiefer, M. C., Zapf, J., and Baxter, R. C. (1998). Insulin-like growth factor-binding protein 5 complexes with the acid-labile subunit. Role of the carboxyl-terminal domain. J Biol Chem 273, 28791-28798.

Tysnes, B. B., Haugland, H. K., and Bjerkvig, R. (1997). Epidermal growth factor and laminin receptors contribute to migratory and invasive properties of gliomas. Invasion Metastasis 17, 270-280.

Uchida, H., Banba, S., Wada, M., Matsumoto, K., Ikeda, M., Naito, N., Tanaka, E., and Honjo, M. (1999). Analysis of binding properties between 20 kDa human growth hormone (hGH) and hGH receptor (hGHR): the binding affinity for hGHR extracellular domain and mode of receptor dimerization. J Mol Endocrinol 23, 347-353.

Uhm, J. H., Dooley, N. P., Kyritsis, A. P., Rao, J. S., and Gladson, C. L. (1999). Vitronectin, a glioma-derived extracellular matrix protein, protects tumor cells from apoptotic death. Clin Cancer Res 5, 1587-1594.

164

Ullrich, A., Gray, A., Tam, A. W., Yang-Feng, T., Tsubokawa, M., Collins, C., Henzel, W., Le Bon, T., Kathuria, S., Chen, E., and et al. (1986). Insulin-like growth factor I receptor primary structure: comparison with insulin receptor suggests structural determinants that define functional specificity. Embo J 5, 2503-2512.

Underwood, P. A., Steele, J. G., and Dalton, B. A. (1993). Effects of polystyrene surface chemistry on the biological activity of solid phase fibronectin and vitronectin, analysed with monoclonal antibodies. J Cell Sci 104, 793-803.

Upton, Z., Webb, H., Hale, K., Yandell, C. A., McMurtry, J. P., Francis, G. L., and Ballard, F. J. (1999). Identification of vitronectin as a novel insulin-like growth factor-II binding protein. Endocrinology 140, 2928-2931.

Valenzano, K. J., Heath-Monnig, E., Tollefsen, S. E., Lake, M., and Lobel, P. (1997). Biophysical and biological properties of naturally occurring high molecular weight insulin-like growth factor II variants. J Biol Chem 272, 4804-4813.

Waltz, D. A., Natkin, L. R., Fujita, R. M., Wei, Y., and Chapman, H. A. (1997). Plasmin and plasminogen activator inhibitor type 1 promote cellular motility by regulating the interaction between the urokinase receptor and vitronectin. J Clin Invest 100, 58-67.

Ward, C. W., Garrett, T. P., McKern, N. M., Lou, M., Cosgrove, L. J., Sparrow, L. G., Frenkel, M. J., Hoyne, P. A., Elleman, T. C., Adams, T. E., et al. (2001). The three dimensional structure of the type I insulin-like growth factor receptor. Molecular Pathology 54, 125-132.

Wei, Y., Waltz, D. A., Rao, N., Drummond, R. J., Rosenberg, S., and Chapman, H. A. (1994). Identification of the urokinase receptor as an adhesion receptor for vitronectin. J Biol Chem 269, 32380-32388.

Wilkins-Port, C. E., and McKeown-Longo, P. J. (1996). Heparan sulfate proteoglycans function in the binding and degradation of vitronectin by fibroblast monolayers. Biochem Cell Biol 74, 887-897.

Wilson, H. M., Birnbaum, R. S., Poot, M., Quinn, L. S., and Swisshelm, K. (2002). Insulin-like growth factor binding protein-related protein 1 inhibits proliferation of MCF-7 breast cancer cells via a senescence-like mechanism. Cell Growth Differ 13, 205-213.

Wong, M. S., Fong, C. C., and Yang, M. (1999). Biosensor measurement of the interaction kinetics between insulin-like growth factors and their binding proteins. Biochim Biophys Acta 1432, 293-301.

Wong, N. C., Mueller, B. M., Barbas, C. F., Ruminski, P., Quaranta, V., Lin, E. C., and Smith, J. W. (1998). Alphav integrins mediate adhesion and migration of breast carcinoma cell lines. Clin Exp Metastasis 16, 50-61.

Wood, T. L. (1995). Gene-targeting and transgenic approaches to IGF and IGF binding protein function. Am J Physiol 269, E613-622.

165

Wraight, C. J., White, P. J., McKean, S. C., Fogarty, R. D., Venables, D. J., Liepe, I. J., Edmondson, S. R., and Werther, G. A. (2000). Reversal of epidermal hyperproliferation in psoriasis by insulin-like growth factor I receptor antisense oligonucleotides. Nat Biotechnol 18, 521-526.

Xu, D., Baburaj, K., Peterson, C. B., and Xu, Y. (2001). Model for the three-dimensional structure of vitronectin: predictions for the multi-domain protein from threading and docking. Proteins 44, 312-320.

Xu, Q., Yan, B., Li, S., and Duan, C. (2004). Fibronectin binds insulin-like growth factor-binding protein 5 and abolishes Its ligand-dependent action on cell migration. J Biol Chem 279, 4269-4277.

Yamanaka, Y., Fowlkes, J. L., Wilson, E. M., Rosenfeld, R. G., and Oh, Y. (1999). Characterization of insulin-like growth factor binding protein-3 (IGFBP-3) binding to human breast cancer cells: kinetics of IGFBP-3 binding and identification of receptor binding domain on the IGFBP-3 molecule. Endocrinology 140, 1319-1328.

Yamane, A., Urushiyama, T., and Diekwisch, T. G. (2002). Roles of insulin-like growth factors and their binding proteins in the differentiation of mouse tongue myoblasts. Int J Dev Biol 46, 807-816.

Yebra, M., Parry, G. C., Stromblad, S., Mackman, N., Rosenberg, S., Mueller, B. M., and Cheresh, D. A. (1996). Requirement of receptor-bound urokinase-type plasminogen activator for integrin alphavbeta5-directed cell migration. J Biol Chem 271, 29393-29399.

Yee, D., Favoni, R. E., Lippman, M. E., and Powell, D. R. (1991). Identification of insulin-like growth factor binding proteins in breast cancer cells. Breast Cancer Res Treat 18, 3-10.

Yee, D., Jackson, J. G., Kozelsky, T. W., and Figueroa, J. A. (1994). Insulin-like growth factor binding protein 1 expression inhibits insulin-like growth factor I action in MCF-7 breast cancer cells. Cell Growth Differ 5, 73-77.

Yu, J., Iwashita, M., Kudo, Y., and Takeda, Y. (1998). Phosphorylated insulin-like growth factor (IGF)-binding protein-1 (IGFBP-1) inhibits while non-phosphorylated IGFBP-1 stimulates IGF-I-induced amino acid uptake by cultured trophoblast cells. Growth Horm IGF Res 8, 65-70.

Zhang, X., Kamaraju, S., Hakuno, F., Kabuta, T., Takahashi, S., Sachdev, D., and Yee, D. (2004). Motility response to insulin-like growth factor-I (IGF-I) in MCF-7 cells is associated with IRS-2 activation and integrin expression. Breast Cancer Res Treat 83, 161-170.

Zheng, B., Clarke, J. B., Busby, W. H., Duan, C., and Clemmons, D. R. (1998). Insulin-like growth factor-binding protein-5 is cleaved by physiological concentrations of thrombin. Endocrinology 139, 1708-1714.

Zheng, B., and Clemmons, D. R. (1998). Blocking ligand occupancy of the alphaVbeta3 integrin inhibits insulin-like growth factor I signaling in vascular smooth muscle cells. Proc Natl Acad Sci U S A 95, 11217-11222.

166

Zumstein, P. P., Luthi, C., and Humbel, R. E. (1985). Amino acid sequence of a variant pro-form of insulin-like growth factor II. Proc Natl Acad Sci U S A 82, 3169-3172.

Documents

Jennifer Kricker Thesis (PDF 3MB)