By Joe Yasa Bsc (Hons)

IN POST-PNEUMONECTOMY

LUNG REGENERATION

by Joe Yasa BSc (Hons)

This thesis is presented for the degree of Doctor of Philosophy under the School of Veterinary and Life Sciences, Murdoch University, Perth Western Australia. -2019-

DECLARATION

I declare that this thesis is on account of my research conducted during the Doctor of Philosophy program and contains as its main content, work that has not been previously submitted for a degree at any tertiary institution. In instances where work has been performed by other parties, permission has been granted from these parties and acknowledged by referencing.

Joe Yasa

This thesis is dedicated to my late co-supervisor and mentor

Professor Geoffrey John Laurent

(1948-2018)

Whose vision and work inspired the beginning of this project.

ABSTRACT

Insulin Growth Factor-1 (IGF-1), is a key and highly regulated molecule which stimulates somatic growth. The level of serum IGF-1 in humans peaks at adolescence and declines with age. IGF-1 expression is also critical for embryonic lung development and is expressed in the regenerating lung of young animals following pneumonectomy (PNX), the surgical removal of a lung. The murine left-lung PNX model was used to investigate the hypothesis that IGF-1 enhances the regenerative capacity of the lung. The potential interactions of IGF-1 and the transcription factors early growth response protein 1 (EGR-1) and hypoxia-inducible factor-1α

(HIF-1α) in post-PNX lung growth was also investigated.

I demonstrated that following left-lung PNX in young mice (aged 2-3 months) pre-operative total lung volume and tissue volume is restored by day 21 post-surgery. IGF-1 mRNA and protein levels were significantly induced in the remaining lung, with a transient but significant increase in IGF-1+, pIGF-1R+ and pERK-1/2+ lung cells, at day three post-PNX compared to

SHAM treated mice. I then showed that intraperitoneal administration of IGF-1 following PNX significantly increases the rate of lung volume and tissue volume recovery and the level of lung cell proliferation, when assessed at day seven post-surgery. In contrast, blocking IGF-1 activity by pharmacological inhibition of IGF-1R signalling, significantly attenuated post-PNX lung growth.

In young mice receiving continuous subcutaneous infusion of IGF-1 following PNX, the rate of lung volume recovery to pre-operative levels was similar to age-matched PBS-treated PNX mice. However, lung sections assessed at day 23 post-surgery revealed that IGF-1-treated mice lungs had significantly higher numbers of IGF-1+, pIGF-1R+, pERK-1/2+ lung cells, proliferating SpC+ type two alveolar epithelial cells and number of alveoli per unit area, suggesting that IGF-1 treatment was associated with additional regenerative activity.

In old PNX mice (aged 22-24 months), there was a significant increase in aerated lung volume following continuous subcutaneous administration of IGF-1, compared to age-matched PBS- treated controls. However, unlike young mice, there was no evidence of restoration of pre- operative total lung volume or tissue volume by day 21 post-surgery in either treatment groups.

Additionally, lung sections assessed at day 23 post-surgery revealed no differences between

IGF-1 and PBS treated lungs in the numbers of IGF-1+, pIGF-1R+ and pERK-1/2+ lung cells, or proliferating SpC+ alveolar type two epithelial cells or the number of alveoli per unit area.

Finally, I investigated the potential for interactions of IGF-1 and the transcription factors EGR-

1 and HIF-1α in post-PNX lung growth. I demonstrated that IGF-1 treatment can induce EGR-

1 and HIF-1α protein expression in cultured 3T3 fibroblasts. Furthermore, both EGR-1 and

HIF-1α were transiently increased in the parenchyma of the lung at day three post-PNX, coinciding with the peak of IGF-1 expression. Pharmacological inhibition of either EGR-1 or

HIF-1α activity significantly reduced the density of CD31+ cells and CD31 protein levels in the lung. Additionally, pharmacological inhibition of HIF-1α activity in the lung following

PNX significantly reduced lung cell proliferation at day seven post-surgery. Intraperitoneal administration of IGF-1 was able to restore the level of lung cell proliferation in HIF-1α inhibited PNX lungs to the levels of untreated PNX lungs.

These results collectively point to a temporally coordinated involvement IGF-1, EGR-1 and

HIF-1α during post-PNX regenerative lung growth, and that additional IGF-1 treatment can enhance such growth following PNX in young mice, but not in aged mice. Furthermore, these studies have identified a potential age-dependent defect in responsiveness to IGF-1 in the lung, which warrants further investigation.

ACKNOWLEDGEMENTS

I have a lot of people to thank for helping me complete my PhD thesis. First and foremost, I would like to acknowledge my principle supervisor Dr. Andrew Lucas, who has been a father figure and friend to me throughout my candidature. He has been an invaluable source of knowledge, guidance, mentorship and support beyond the scope of my PhD thesis. His enthusiasm, passion and patience, when things were seemingly frustrating, inspired and encouraged me to completion. I thank him so much for teaching me science and for all that he has done for me. I thoroughly enjoyed working with him and I hope this does not mark the end of our relationship.

I am also deeply indebted to my co-supervisor A/Prof. Cecilia Prêle who took me under her wing as an unsuspecting Summer scholarship student, mentored me through my Honours year and was gracious and patient enough to guide me throughout my PhD candidature. Her critical input and wisdom in editing my thesis, manuscript and presentations has helped shape the direction of this project.

I count myself privileged to have had the opportunity to be mentored and supervised by the late Prof. Geoffrey Laurent, a brilliant and world-renowned scientist in the field of regenerative medicine and fibrosis, whose vision and work inspired the genesis of this project. Though we met on rare occasions, his tremendous experience, passion for science and wisdom had a huge impact on shaping the direction of this project and left a lasting impression on me that has helped shape my passion for science.

I also wish to thank my Murdoch co-supervisor Prof. Robert Mead for his help, guidance and encouragement throughout my PhD and Honours years. I thank him for expressing confidence in my abilities and inspiring me to undertake medical research.

III

I would also like to thank Murdoch University for granting me a Murdoch International Post- graduate Scholarship that supported me financially throughout the course of my PhD. I also acknowledge the Centre for Cell Therapy and Regenerative Medicine (UWA), who financially supported this project through the provision of laboratory equipment and consumables.

I owe a tremendous debt of gratitude to the micro-surgery team headed by Prof. Michaela Lucas whose equipment and resources facilitated the mouse pneumonectomy surgeries. In particular,

I would like to thank Ms. Liu Liu and Dr. Wenhua Huang who performed the surgeries.

I also wish to thank all the staff at the Centre for Microscopy, Characterisation and Analysis

(CMCA) in particular Ms. Diana Patalwala for her assistance with CT imaging and analysis.

I am also thankful for the help, support and friendship I received from staff and students at the

Institute for Respiratory Health and Harry Perkins Institute of Medical Research. I am especially grateful to all past and present members of the Tissue Repair Group and Burns

Research Unit, whom I have worked with over the years. Your advice and critical input in my work was instrumental in guiding the overall direction of my project. Thank you for your friendship and support, it certainly made my PhD an enjoyable experience.

Last but certainly not least, I would like to say a huge thank you to all my family and friends for their constant encouragement and support. I count myself blessed to have been born and raised in a large family filled with so much love and support. To my Dad John Yasa, words cannot describe how grateful I am for his amazing generosity and sacrifice. Not having had the opportunity to attain a decent education himself, he made a tremendous and risky sacrifice by personally sponsoring my education abroad. I thank him so very much for believing in me and the power of education which he constantly preached. I shall forever be indebted to him and my entire family for giving me this wonderful opportunity that has enabled me to pursue my scholarly passions. I hope to make them proud.

TABLE OF CONTENTS

Preface Abstract I

Acknowledgements III

Table of Contents V

List of Figures XIII

List of Tables XVII

Abbreviations XVIII

Chapter 1- 1 Introduction

1.1- Introduction 2

1.2- The unmet clinical burden of chronic lung diseases 3

1.3- Lung repair and regeneration strategies 4

1.4- Pneumonectomy, a clinical perspective 6

1.5- Lung regeneration after PNX 7

1.6- The murine PNX model 9

1.7- Factors affecting regenerative lung growth after PNX 11

1.7.1- Mechanical force 11

1.7.2- Hypoxia 14

1.7.3- Age 14

1.7.3.1- Evidence of lung regeneration in adult humans 16

1.7.4- Transcription factors 18

1.7.4.1- Early growth response (EGR)-1 19

1.7.4.2- Hypoxia-inducible factor (HIF)-1α 20

1.8- Cellular mediators of post-PNX lung growth 22

1.8.1- Alveolar epithelial cell progenitors 22

1.8.1.1- Club cells 22

1.8.1.2- Bronchoalveolar stem cells (BASCs) 23

1.8.1.3- Alveolar type two epithelial cells (AECIIs) 24

1.8.1.4- Alveolar type one epithelial cells (AECIs) 24

1.8.1.5- p63+/Krt-5+ alveolar progenitors 25

1.8.2- Mesenchymal stromal cells in lung regeneration 27

1.8.3- Endothelial cells in lung regeneration 28 1.8.4- Macrophages in regenerative lung growth 29

1.9- Growth factors in post-PNX lung growth 32

1.10- The IGF family of proteins 35

1.10.1- IGF-1 and IGF-2 ligands 35

1.10.2- IGF-1 receptor (IGF-1R) and IGF-1 signalling 36

1.10.3- IGF-2R 37

1.10.4- IGF binding proteins (IGFBPs) 39

1.10.4.1- General structure of IGFBPs 39

1.10.4.2- Mechanism of action 39

1.10.5- Sources of IGF-1 41

1.10.5.1- Endocrine regulation of IGF-1 41

1.10.5.2- Autocrine and paracrine regulation of IGF-1 41

1.10.5.3- IGF-1 levels decline with age in humans 43

1.10.6- Generation of multiple isoforms from the Igf-1 gene 44

1.11- IGF-1 in foetal lung development 47

1.12- IGF-1 in post-PNX lung growth 49

1.13- Hypothesis and Aims of thesis 52

Chapter 2- 54 Methods

2.1- Animal ethics statement 55

2.2- Left lung pneumonectomy and sham thoracotomy surgery 55

2.3- Implantation of the ALZET® Mini-osmotic pump in vivo 57

2.4- Micro-Computed Tomography (µCT) X-ray imaging protocol 58

2.5- µCT Image analysis 59

2.5.1- Image Sorting 59

2.5.2- Reconstruction 59

2.5.3- Optimisation of mean lung density measurement 61

2.5.4- Lung volume and density measurement using CTAn 62

2.5.5- Description of the µCT algorithm used to enumerate lung volume and 62 density

2.6- Estimation of total lung volume by the water displacement method 65

2.7- Quantitative Real time PCR (qRT-PCR) 66

2.7.1- RNA Extraction 66

2.7.2- Complementary DNA (cDNA) Synthesis 67

2.7.3- Real Time (RT) PCR 67

2.8- Western blotting 68

2.8.1- Whole cell lysate protein extraction from homogenised lung tissue 68

2.8.2- Whole cell lysate protein extraction from cultured cells 68

2.8.3- Nuclear protein extraction 69

VII

2.8.4- Protein Quantification 69

2.8.5- Western blot analysis 70

2.8.6- Membrane Stripping 71

2.9- Histology 72

2.9.1- Preparation of lung specimens for histological analysis 72

2.9.2- Immunohistochemistry (IHC) staining 72

2.9.3- Haematoxylin and Eosin (H&E) staining 74

2.9.4- Image analysis 74

2.10- Cell maintenance 76

2.10.1- Propagation of cell line 76

2.10.2- Cell counting 76

2.10.3- Freezing and thawing cells 77

2.10.4- Mycoplasma testing 77

2.11- Methylene blue (MB) cell proliferation assay 78

2.12- Applying equibiaxial stretch to cultured cells 79

2.12.1- Construction of the equibiaxial stretch device 79

2.12.2- Extracting RNA and protein from stretched cells 81

2.12.3- Re-coating Bioflex plates with collagen for re-use 81

2.13- Statistics 82

VIII

Chapter 3- 83 IGF-1 signalling enhances regenerative lung growth after left lung pneumonectomy in young adult mice

3.1- Introduction 84

3.2- Results 87

3.2.1- Lung volume and tissue content are restored to near pre-operative 86 levels after PNX in young adult mice

3.2.2- PNX does not result in pathological features such as inflammation or 89 fibrosis

3.2.3- Post-PNX lung growth response is accompanied by increased cell 91 proliferation

3.2.4- IGF-1 expression is up-regulated in response to PNX 97

3.2.5- Immunohistochemical analysis reveals a role for IGF-1R signalling 103 in PNX lungs

3.2.6- IGFBP-3 positive cells are prominent in pneumonectomised mouse 108 lung tissue

3.2.7- IGF-1 supplementation enhances post-PNX lung growth 110

3.2.8- IGF-1R inhibition reduces regenerative lung growth post-PNX 110

3.3- Discussion 114

Chapter 4- 122 The differential effect of IGF-1 on the regenerative response post-PNX in aging

4.1- Introduction 123

4.2- Results 125

4.2.1- Sustained infusion of IGF-1 increased the aerated lung volume of old 125 mice but not young mice

4.2.2- IGF-1 treatment in old PNX mice transiently reduces the density of 128 the lung

4.2.3- Sustained infusion of IGF-1 induces IGF-1 signalling components in 131 young but not old mice

4.2.4- Sustained infusion of IGF-1 increased cell proliferation and lung 135 septation in young mice but not old mice

4.3- Discussion 142

Chapter 5- 147 Investigating EGR-1 and HIF-1α in post-PNX lung growth

5.1- Introduction 148

5.2- Results 151

5.2.1- Stretch induces Egr-1, Igf-1r and Igfbp-3 mRNA expression in 151 cultured lung fibroblasts

5.2.2– IGF-1 signalling and fibroblast proliferation in vitro can be 153 effectively modulated by the IGF1R antagonist, BMS-536294

5.2.3- IGF-1 induces EGR-1 and HIF-1α protein expression in cultured 153 lung fibroblasts

5.2.4- Pneumonectomy induces EGR-1 and HIF-1α expression in the lung 157

5.2.5- Pharmacological inhibition of EGR-1 or HIF-1α did not significantly 159 alter lung weight but increased lung volume

5.2.6- Sustained inhibition of HIF-1α does not alter lung volume but 162 significantly reduces lung cell proliferation at day seven post-PNX

5.2.7- Pharmacological inhibition of EGR-1 or HIF-1α alters distribution 166 of CD31+ cells in the lung parenchyma post-PNX

5.3- Discussion 169

Chapter 6- 175 General Discussion, Conclusions and Future directions

6.1- General Discussion 176

6.2- Conclusions 183

6.3- Implications of findings 184

6.3.1- Clinical relevance of findings 184

6.3.2- Potential challenges of translating research findings to the clinic 186

6.4- Future directions 187

Chapter 7- 191 Appendices

Appendix A- Materials 192

A.1- Buffers and Solutions 192

A.2- Western Blot Gels 197

A.3- Tissue Culture Reagents 197

A.4- Real-time PCR reagents 198

A.4.1- Complementary DNA (cDNA) Mastermix 198

A.4.2- Real-time PCR mastermix 198

A.4.3- Taqman probe sets 199

A.5- Kits 199

A.6- Immunohistochemistry (IHC) and Western blot (WB) 200 Antibodies

A.7- Equipment and Instruments used 204

A.8- Components of the Equibiaxial Stretch Device 206

A.9- Drugs used 206

A.10- Other Reagents 207

Appendix B- µCT Image acquisition, Sorting and Reconstruction settings 209

Appendix C- Image J Macro for analysing single coloured IHC stains 213

Appendix D- Image J Macro for analysing dual coloured IHC stains 214

Appendix E- Image J Macro for generating pseudocoloured fluorescent 217 images

Appendix F- Primary Antibody controls used in IHC Experiments 218

Chapter 8- 219 References

XII

LIST OF FIGURES

Figure 1.1: Alveolar formation after PNX is similar to post-natal lung 8 growth

Figure 1.2: The murine left-lung PNX model 10

Figure 1.3: Effect of mechanical forces on lung growth and remodelling 13

Figure 1.4: Evidence of neoalveolarisation in an adult human 17

Figure 1.5: Structure and mechanism of EGR-1 transcription factor 21

Figure 1.6: Epithelial cell progenitors that contribute to neoalveolarisation 26

Figure 1.7: Mesenchymal stromal cells involved in alveolar regeneration 31

Figure 1.8: The IGF family of proteins and IGF-1 signalling 38

Figure 1.9: General structure of IGFBPs and mechanism of IGF liberation 40

Figure 1.10: Endocrine and paracrine sources of IGF-1 42

Figure 1.11: Structure of the mouse and human Igf-1 gene 46

Figure 1.12: Proposed mechanism for the involvement of IGF-1 signalling in 53 post-PNX lung growth

Figure 2.1: Sequence of images showing left-lung pneumonectomy surgery 56 in mice Figure 2.2: Coronal and transverse views of reconstructed images from 60 PNX mice lungs during inspiration and expiration.

Figure 2.3: Processes involved in estimation of lung volume and density in 63 CTAn

Figure 2.4: Simple apparatus for measuring lung volume by water 65 displacement

Figure 2.5: Construction of the equibiaxial stretch device 80

Figure 3.1: Schematic of experiments done in chapter three 86

Figure 3.2: Restoration of lung volume and tissue content after PNX young 88 mice

Figure 3.3: H&E images highlighting prominent histological features 90 occurring in the right lung after left lung pneumonectomy in mice

XIII

Figure 3.4: Identification of Ki-67/SpC+ cells across the post-PNX time 92 course

Figure 3.5: Increased cell proliferation in post-PNX lung 93

Figure 3.6: Increase in parenchymal α-SMA positive cells in PNX lungs 95

Figure 3.7: CD31 expression was not altered in response to PNX 96

Figure 3.8: Temporal changes in IGF-1 protein expression in post-PNX 99 lung tissue

Figure 3.9: Immunohistochemical identification of cell types expressing 100 IGF-1 in post-PNX lung tissue

Figure 3.10: Igf-1 mRNA is upregulated in post-PNX lung tissue 102

Figure 3.11: Immunohistochemical localisation (A) and quantification (B) of 104 IGF-1R expressing cells in the lung parenchyma post-PNX

Figure 3.12: PNX induces a transient decline in Igf-1r mRNA 105

Figure 3.13: IHC analysis reveals significant increase in pIGF-1R and pERK 107 expression in post-PNX lung tissue

Figure 3.14: Heightened IGFBP-3 protein expression in post-PNX mouse 109 lungs

Figure 3.15: IGF-1 Supplementation increases lung volume and cell 112 proliferation post-PNX

Figure 3.16: Pharmacological inhibition of IGF-1R reduces total lung 113 volume and tissue volume post-PNX

Figure 4.1: Schematic of the IGF-1 supplementation experiments in young 126 and old PNX mice

Figure 4.2: Longitudinal changes in total lung volume and air volume in 127 young and old PNX mice supplemented with IGF-1 or PBS control

Figure 4.3: IGF-1 supplementation transiently reduces mean lung density 130 and has no significant effect on lung tissue volume in old PNX mice

Figure 4.4: Infusion of IGF-1 increases the number of IGF-1 positive cells 132 in pneumonectomised young mice but not old mice

Figure 4.5: IGF-1 supplementation activates IGF-1R in young 133 pneumonectomised mice but not old mice

XIV

Figure 4.6: Activation of downstream pERK signalling in young 134 pneumonectomised mice supplemented with IGF-1

Figure 4.7: The number of IGF-1R expressing cells within the lungs of 136 young and old pneumonectomised mice are similar and are not significantly altered by IGF-1 supplementation

Figure 4.8: IGF-1 supplementation does not significantly alter the number 137 of IGFBP-3 expressing cells in the lungs of young and old mice post-PNX

Figure 4.9: Continuous infusion of IGF-1 enhances cell proliferation in the 138 lungs of young but not old mice post-PNX

Figure 4.10: Sustained administration of IGF-1 does not significantly alter α- 140 SMA and CD31 expressing cell numbers in the lungs of young and old mice post-PNX

Figure 4.11: IGF-1 supplementation enhances lung septation in young but 141 not old pneumonectomised mice

Figure 5.1: Proposed mechanism driving the IGF-1 signalling pathway 150 post-PNX

Figure 5.2: Stretch induces mRNA expression of Egr-1, Igfbp-3 and Igf-1r 152

Figure 5.3: The IGF-1R antagonist BMS-536924 attenuates 154 phosphorylation of IGF-1R and downstream pERK signalling

Figure 5.4: IGF-1R inhibition reduces IGF-1-mediated fibroblast 155 proliferation

Figure 5.5: The kinetics of IGF-1-induced EGR-1 and HIF-1α expression in 156 3T3 fibroblasts

Figure 5.6: PNX induces a transient increase in EGR-1 and HIF-1α 158 expression in mouse lung tissue

Figure 5.7: Pharmacological inhibition of EGR-1 and HIF-1α did not alter 160 lung mass but significantly increased lung volume at day 15 post-PNX

Figure 5.8: Inhibition of EGR-1 increases alveolar diameter in PNX mice 161

Figure 5.9: Inhibition of HIF-1α has no effect on total lung volume and 163 lung tissue volume in the presence or absence of IGF-1 at day seven post PNX

Figure 5.10: Lung cell proliferation at day seven post-PNX is significantly 165 inhibited by HIF-1α inhibition but restored to baseline by IGF-1 co-treatment

Figure 5.11: Inhibition of EGR-1 or HIF-1α reduces the percentage of 167 CD31+ cells in post-PNX lung tissue

Figure 5.12: Reduction of CD31 protein levels in PNX mice treated with 168 acriflavine or mithramycin

XVI

LIST OF TABLES

Table 1.1: Growth factors implicated in post-PNX lung growth 34

Table 7.1- General purpose buffers and solutions made in the laboratory 195

Table 7.2: SDS-PAGE gel composition 196

Table 7.3: Mastermix constituents for synthesising cDNA 197

Table 7.4: Mastermix composition for real time PCR 197

Table 7.5: TaqMan primer probes used for real time PCR 198

Table 7.6: Primary and secondary antibodies used for 202 immunohistochemistry and western blot analysis

Table 7.7: Specifications of the Equibiaxial stretch device 205

Table 7.8: Drugs administered in in vivo experiments 205

XVII

ABBREVIATIONS

AA amino acid

AECI alveolar epithelial type I cell

AECII alveolar epithelial type II cell

AGER advanced glycosylation end-product specific receptor

AKT rac-alpha serine/threonine-protein kinase

ALS acid labile subunit

ANOVA analysis of variance

Aqp5 aquaporin 5

BADJ bronchoalveolar duct junction

BALF bronchoalveolar lavage fluid

BASCs bronchoalveolar stem cells

BMP bone morphogenetic protein

BrdU Bromodeoxyuridine

BSA bovine serum albumin

ºC degrees Celsius

CBP cAMP response element binding protein (CREB)-binding protein

CCR2 c-c chemokine receptor type 2

CCSP club cell secretory protein

CD cluster of differentiation cDNA complementary DNA

CGRP calcitonin gene-related peptide

CLG compensatory lung growth

COPD chronic obstructive pulmonary disease

XVIII

CSF colony stimulating factor

CT cycle threshold

µCT microcomputed tomography

Cyp2f2 cytochrome P450, family 2, subfamily f, polypeptide 2

2D/3D 2 dimensional/ 3 dimensional

DAB Diaminobenzidine dH2O deionised water

DMEM Dulbecco’s modified eagle medium

DNA deoxyribonucleic acid

DNase Deoxyribonuclease dNTPs deoxyribonucleoside triphosphates

ECM extracellular matrix

EDTA ethylenediaminetetraacetic acid

EGF epidermal growth factor

EGFR epidermal growth factor receptor

EGR-1 early growth response protein 1

ELISA enzyme-linked immunosorbent assay

EMT epithelial-to-mesenchymal transition

ERK extracellular signal-regulated kinase

EpCAM epithelial cell adhesion molecule

EPO-R erythropoietin receptor

FCS foetal calf serum

FGF fibroblast growth factor

FGFR fibroblast growth factor receptor

Foxa2 forkhead box protein A2

XIX

Foxj forkhead box j

GH growth hormone

GKLF gut-enriched Kruppel-like factor

Grb-2 growth factor receptor bound protein

GSK-3β glycogen synthase kinase 3 beta

H&E haematoxylin and eosin

HGF hepatocyte growth factor

HIF hypoxia inducible factor

HPSCs hematopoietic stem cells

HopX homeodomain-only protein homeobox

Hrs or hrs Hours

HU Hounsfield units

Id2 inhibitor of DNA binding 2

IGF-1 insulin-like growth factor 1

IGFBP insulin-like growth factor binding protein

IGF-1R insulin-like growth factor 1 receptor

IHC immunohistochemistry

IkBα inhibitor of nuclear kappa B alpha

IL Interleukin

ILD interstitial lung disease i.p. Intraperitoneal

IPF Idiopathic Pulmonary Fibrosis iPSCs induced pluripotent stem cell

IR insulin receptor

IRS insulin receptor substrate

JNK c-jun NH2 terminal kinase kbp kilo base pairs kDa kilo Daltons

KGF keratinocyte growth factor

Krt-5 cytokeratin 5

Lamp-3 lysosome-associated membrane glycoprotein 3

LRG leucine-rich α-2-glycoprotein

LRMSCs lung-resident mesenchymal stromal cells

MAPK mitogen-activated protein kinase

MGF mechano growth factor

MMP matrix metalloproteinase mRNA messenger ribonucleic acid

MRI magnetic resonance imagings

MSCs mesenchymal stem cells mTORC1 mammalian target of rapamycin complex 1

NAB nerve growth factor inhibitor binding protein

NFκ-B nuclear factor kappa B

NGFI nerve growth factor inhibitor

Nkx2.1 nk2 homeobox 1

NSILAs non-suppressible insulin-like factors p63 transformation-related protein 63

PBS phosphate buffered saline

PCEC pulmonary capillary endothelial cells

PDGF platelet-derived growth factor

PDGFR platelet-derived growth factor receptor

XXI

Pdpn Podoplanin

PET positron-emission tomography

PFA Paraformaldehyde pH power of hydrogen

PI3K phosphtidylinositol-3-kinase

PIP2 phosphoinositol diphosphate

PIP3 phosphoinositol triphosphate

PNX Pneumonectomy

PPARγ peroxisome proliferator-activated receptor gamma proSP-C prosurfactant protein C

PVDF polyvinylidene fluoride qRT-PCR quantitative real time polymerase chain reaction

RA retinoic acid

Raf rapidly accelerated fibrosarcoma

Ras rat sarcoma rhIGF-1 recombinant human insulin-like growth factor 1

RLG regenerative lung growth

RNA ribonucleic acid rpm rotations per minute s.c. Subcutaneous

Sca-1 stem cell antigen 1

SDF-1 stromal cell-derived factor 1

SDS sodium dodecyl sulphate

Scgb1a1 secretoglobin family 1A member 1

SEM standard error of mean

XXII

SftpC/SpC surfactant protein C

SHP-2 src homology-2 containing phosphatase

α-SMA alpha smooth muscle actin

Sos son of sevenless

Sox sex determining region Y high mobility group box

Sp1 specificity protein 1

SRF serum response factor

STAT signal transducer and activator of transcription

TAZ transcriptional coactivator with PDZ-binding motif

TBS tris-buffered saline

TF transcription factor

TGFβ1 transforming growth factor beta 1

TTF-1 thyroid transcription factor 1

TNF tumour necrosis factor

VEGF vascular endothelial growth factor

VEGFR vascular endothelial growth factor receptor v/v volume per volume

Wnt wingless/integrated

WT1 Wilm’s tumour suppressor 1 w/v weight per volume

YAP yes-associated protein

ZF zinc finger

XXIII

Chapter 1

Introduction

1 1.1- Introduction

Partial or complete resection of the lung (PNX) is sometimes the best treatment option for destructive lung diseases such as lung fibrosis, chronic obstructive pulmonary disease (COPD) and progressive lung carcinoma, although this is associated with permanent loss of lung function. In rare cases in adult humans [1] and more commonly in children [2], PNX triggers a compensatory growth response with neoalveolarisation leading to improvement of lung function. Following unilateral PNX in mice, the contralateral lung increases in mass, volume, protein and DNA content reaching levels comparable to that of control littermates within two- to-three weeks post-PNX [3]. A successful regenerative growth response to surgery appears largely age dependent, with young animals showing a more robust capacity to regenerate than adult animals [4]. However, the mechanisms that govern this process are largely unknown. It is thought that mechanical force created by PNX activates early changes in gene expression via transcription factors that lead to expression of growth factors that in turn promote proliferation and differentiation of endogenous alveolar and mesenchymal progenitor cells and subsequently growth of new lung tissue.

This thesis examined the role of Insulin-like Growth Factor -1 (IGF-1) in regenerative lung growth using the mouse PNX model. IGF-1 is indispensable for lung development, growth and repair [5,

6]. Increased levels of IGF-1 have been detected in serum and lungs of PNX animals [7-9] and a recent transcriptomic gene profile analysis of left-PNX mice identified IGF-1 as a central player [10]. However, the cellular and molecular mechanisms, source(s) and effect of IGF-1 modulation have not been investigated. Since IGF-1 levels decline with age [11], which mirrors the reduced capacity for lung regeneration, an attractive therapeutic opportunity exists in supplementing aged lungs with rhIGF-1 (which is already approved in the clinic [12]) to improve their regenerative potential and help ameliorate the sequalae of chronic lung diseases.

2 1.2- The unmet clinical burden of chronic lung diseases

The interruption of normal lung function by age, injury or disease has profound consequences on an individual’s health and well-being. In Australia, and in the world at large, lung diseases pose a huge burden on the healthcare system and society. A report published by the Woolcock

Institute of Medical Research in 2014 documented over 20,000 deaths attributed to lung diseases in the year 2012, which represented close to 14% of all deaths in Australia in that year

[13]. Nearly 300,000 hospitalisations in the period from 2011 to 2012 were attributed to lung- related diseases. The spectrum of lung diseases varies across various age groups. In infants, pneumonia and bronchiolitis represent the major health threats, whilst many children and young adults suffer from asthma. Lung cancer, smoking related-chronic obstructive pulmonary disease (COPD) and interstitial lung diseases (ILD) are the prevalent respiratory ailments commonly encountered in adults aged above 50 years old. Lung cancer accounts for approximately 40% of deaths from lung disease whilst COPD contributes almost one-third of morbidities and mortalities of the respiratory tract and is predicted to be the third leading cause of death worldwide by 2020 [14].

Despite the significant effort by researchers and clinicians, there are no effective cures for many chronic lung diseases. The drugs currently available are primarily focused on managing symptoms or halting disease progression, but little is being done to reverse the damage induced by destructive lung diseases. Lung resection has been shown to improve the survival and quality of life of lung cancer, COPD and emphysema patients. Most lung resection surgeries are performed on lung cancer patients, particularly non-small cell lung cancer (NSLC, accounting for 80% of lung cancers) where lung resection is the gold standard treatment for early stage localised NSLC [15]. Although improved surgical and imaging techniques have significantly improved the survival of patients undergoing lung resection surgery, multiple

3 comorbidities and poor lung function contribute to high morbidity and mortality post-surgery

[15]. For many patients with end stage lung diseases, the ultimate solution at present is lung transplantation. Australia has one of the World’s best practises in lung transplant donation, utilisation and surgery with more than 200 successful lung transplants performed annually [16].

However, lung transplantation suffers from many limitations including a limited number of donor organs, organ rejection, morbidity and an overall poor prognosis. Any treatment strategy that can partially reverse the damage induced by destructive lung diseases or restore lung function after resection will be of tremendous benefit to patients suffering from degenerative lung ailments.

1.3- Lung repair and regeneration strategies

Lung regeneration presents an alternative therapeutic approach for treating intractable lung diseases and great strides have been made in the field of regenerative medicine in recent years.

Several strategies are being studied in efforts to rebuild parts of the lung or encourage whole organ regeneration. For a more extensive discussion on this subject the reader is referred to excellent reviews by Hogan et.al [17], Garcia et.al [18], Petersen et.al [19] and others [20-23].

Here I attempt to summarise some of the strategies employed in lung tissue and whole organ regeneration.

Exogenously supplied stem cells derived from various sources including the bone marrow, embryonic tissue, amniotic fluid, placenta, Wharton’s jelly and fat tissue have shown promising results in reversing lung damage and facilitating gas exchange in various preclinical injury models [24-28] and patients with chronic lung diseases such as Idiopathic Pulmonary Fibrosis

(IPF) and COPD [14, 29]. In lieu of cell-based therapies, growth factors (e.g. retinoic acid,

VEGF, etc.), cytokines (e.g. IL-2, IL-10, TNF, etc.) or essential matrix elements could be administered to injured or resected lungs to promote endogenous lung repair/regeneration [20].

4 However, many more questions remain to be addressed before we see the routine use of such therapies in the clinic. For example, issues of immune-rejection, tumorigenicity, host-tissue integration and long-term survival of exogenous supplemented stem cells need to be addressed

[21].

The use of tissue engineering to generate new functional lung structures is an area of active research. A few groups have attempted to build a complete lung using decellularised matrices in mice and rats [30-32]. Lungs from decellularised matrices often fail within a few hours to a few days due to a combination of factors including incomplete vascularisation, intravascular coagulation and poor diffusion capacity because of poorly differentiated mesenchyme and airways [30, 31]. In humans, trachea and bronchi have been made using induced pluripotent stem cells (iPSCs; reprogrammed stem cells) cultured on cadaveric donor decellularised matrices [33]. Synthetic scaffolds using materials such as steel, teflon and hydroxyapatite have also been used to construct prosthetic tracheal replacements in patients [34]. Upper airway structures have been much easier to construct compared to more complex alveolar parenchyma components. Despite the limitations and challenges faced by lung tissue engineering approaches, they represent an innovative alternative for regenerating the lung and soon could be the treatment of choice for chronic and congenital degenerative lung diseases.

The choice and success of treatment for regenerative purposes will depend on several factors including the extent of disease, timing of diagnosis, age and underlying genetic factors [18].

Such an approach will require a more complete understanding of the structural biology of the lung, the important stem cell niche components and factors that limit regeneration of the lung in various settings. The PNX model is a useful tool for investigating the mechanisms of postnatal alveolarisation.

5 1.4- Pneumonectomy, a clinical perspective

The lung is divided into lobules and bronchopulmonary segments that receive individual air and blood supply [35]. This means that if a lesion is restricted to a bronchopulmonary segment, a sleeve/wedge lobectomy may be performed with minimal impact on the remaining pulmonary segments. Lobectomy involves removal of an entire lobe, whilst surgical removal of the whole left or right lung is called unilateral pneumonectomy (PNX). Indicators of PNX in humans range from severe cases of lung trauma, non-small cell lung carcinoma, mesothelioma, mycobacteria induced fibrosis, bronchiectasis, IPF and congenital lobar emphysema [36]. In most cases, PNX is only recommended as a last resort because the permanent loss of a lung severely diminishes respiratory capacity, limiting the patient to simple tasks that demand less exertion. Furthermore, PNX surgery may be associated with many risks and complications including post-PNX syndrome (displacement of structures within the thorax that can cause hemodynamic instability, leakage of blood vessels and lymph and airflow occlusion [37]), abnormal heart rhythms, shortness of breath, fever, pneumonia, empyema, bronchopleural fistula (an abnormal opening between the cut bronchus and pleural space), pulmonary edema, and general pain and discomfort [38]. Insertion of a plombage (inert material e.g. Lucite, fat, paraffin wax, inflatable prosthetic, etc. used to fill the chest cavity [39]) may help manage some these symptoms but studies have shown that plombage limits growth of the remaining lung therefore does not confer long term advantage [40, 41]. Any treatment that can partially restore lost tissue could significantly improve the quality of life of PNX patients.

6 1.5- Lung regeneration after PNX

In response to PNX, the lung can adapt in three main ways: i) recruit underutilised alveolar units ii) undergo tissue remodelling changes such as expansion of existing alveoli, dilation of airways and blood vessels and rearrangement of matrix components to maximise the surface area for gas exchange or iii) undergo compensatory lung growth (CLG) by formation of new alveoli (neoalveolarisation) and blood vessels (neovascularisation) if the functional demand exceeds the available units of gas exchange [42].

CLG after PNX has been documented in a number of species including dogs [43], ferrets [44], mice [4], rabbits [45], rats [7] and humans [1]. The ability of animals to undergo CLG post-

PNX is dependent on several factors (discussed in Section 1.7) but as a general rule, smaller and younger animals have a greater regenerative potential than larger and older animals [46].

Neoalveolarisation post-PNX involves formation of secondary crests from existing primary septa similar to developmental alveolarisation (Figure 1.1A). Collagen and elastin fibres are laid at septal tips of invading secondary septa in concert with new double layered capillaries formed by intussusceptive angiogenesis [47]. Later during septal maturation, the double layered capillaries fuse into a single capillary to ensure newly formed septa are thin enough to allow for efficient gas exchange. The result of this process is subdivision of pre-existing alveoli spaces, reducing the mean linear intercept but increasing the number of alveoli and surface area for gas exchange (Figure 1.1B). Neoalveolarisation is predominant in the subpleural edges of the lung where mechanical forces impart the most amount of strain on tissue abutting the subpleura [48]. Respiratory bronchioles have also been shown to participate in regenerative lung growth; in contrast, the pattern of larger airways and blood vessels remains unchanged

[47].

Figure 1.1: Alveolar formation after PNX is similar to post-natal lung growth. (A) During post-PNX lung growth and developmental alveolarisation, new alveoli are formed by subdivision of alveolar spaces by formation of new septa. Collagen and elastin fibres (green dots) are laid at septal tips by myofibroblasts as secondary tips are formed from existing primary septa. This is accompanied by formation of double layered capillaries by intussusceptive microvascular growth. The capillaries later undergo maturation and fuse to form a single capillary. Image adapted with permission from Schittny et.al [49] with minor modifications. (B) Scanning electron micrograph showing formation of new alveoli in the rat lung. ‘a’= alveoli, ‘ad’ =alveolar duct, ‘s’= saccules. Adapted with permission from P.H Burri [50].

8 1.6- The murine PNX model

In this thesis, we used the unilateral left lung PNX mouse model to study the cellular and molecular mechanisms regulating lung regeneration after PNX (Figure 1.2). The left lung of mice consists of a single lobe that contributes approximately 30-35% of total lung capacity

[51]. In young adult mice (2-3 months old), it is well established that after left lung PNX, right lung volume, mass and function is restored to near pre-operative levels within three weeks post-surgery [52]. This growth response is driven by formation of new alveoli as well as remodelling of existing lung tissue [53]. A study by Fehrenbach et.al [54] showed that the contribution of these two processes (i.e. neoalveolarisation and tissue remodelling) in mice is estimated to be equivalent and that most of the neoalveolarisation takes place at the subpleural edges of the lung because most of the mechanical strain is imposed on this region of the lung

[55]. Furthermore, computed tomography (CT) and positron-emission tomography (PET) imaging studies on the post-PNX mouse lung have further revealed that the cardiac lobe contributes to most of the lung growth seen after PNX [56, 57].

The post-PNX growth response seen in young mice after left-PNX, although not as remarkable as the limb regeneration in newts and salamanders or whole-body regeneration in planaria, is a classic form of regeneration in its own right given that tissue content is largely restored in such a short space of time. This makes the murine PNX model a valuable tool for understanding mechanisms of post-natal alveolarisation. Compared to other injury models such as bleomycin, naphthalene or influenza-induced lung injury models, the post-PNX response is accompanied by minimal inflammation and little evidence of structural damage or fibrosis in the residual lung [58]. Therefore, the cellular and molecular events occurring during neoalveolarisation can be studied in the absence of extensive tissue damage. Additionally, the tissue resected is well defined and reproducible and therefore readily quantified [54].

9 The chief drawback of using the murine PNX model lies in the inherent differences between mouse and human lung structure and physiology [59]. Dog and primate PNX models more accurately resemble the time frame and physiological response to PNX in humans [60].

Additionally, PNX dogs, unlike mice, can be trained on treadmills to assess lung function parameters similar to humans [61]. Nonetheless, the tractability and short healing timeframe of the murine PNX model makes mice favourable for biomedical research. Furthermore, the availability of genetically engineered mice strains makes it possible to individually assess the contribution of a particular cell type or gene of interest in the regenerative process [58].

Therefore, despite the differences between human and mouse anatomy and physiology, studying regenerative mechanisms in mice can offer great insights into human lung regeneration.

Potential Regulators: ▪ Mechanical strain ▪ Hypoxia ▪ Transcription factors ▪ Growth factor release ▪ Age ▪ Progenitor cells

Superior lobe

Middle lobe Left

PNX Inferior lobe

Cardiac lobe

Right lung Left lung

Figure 1.2: The murine left-lung PNX model.

The mouse right lung consists of four lobes (superior, middle, inferior and cardiac lobes) whilst the left lung is a single lobe that accounts for ~30-35% of total lung capacity. Left lung PNX creates a robust enough mechanical force to re-initiate new tissue growth. However, the ability of animals to undergo regenerative lung growth is dependent on several other factors including age, hypoxia, presence of growth factors and activation of progenitor cells. In young adult mice (2-3 months old) left-PNX results in restoration of lung volume and tissue content to near pre- operative levels within three weeks post-surgery.

10 1.7- Factors affecting regenerative lung growth after PNX

The capacity for lungs to undergo regeneration after PNX is influenced by a multitude of factors interacting in a spatial and temporal manner. Some of the important factors that have been cited in the literature include mechanical force, age, vascularisation, hypoxia, sex, species, activation of transcription factors, presence of endogenous progenitor cells, growth factors, hormones and cytokines [42, 46, 62]. The following discussion will be limited to six important factors relevant to work performed and presented in this thesis. These factors are: mechanical force, hypoxia, age, transcription factors, presence of progenitor cells and growth factors

(Figure 1.2). For a more extensive discussion on these and many other factors affecting post-

PNX lung growth the reader is referred to reviews by Hsia [42], Paisley et.al [62], and Thane et.al [46].

1.7.1- Mechanical force

Mechanical forces influence many cellular processes including signal transduction, protein turnover, ion flux, proliferation, apoptosis and growth factor production [42]. During post-natal lung development, mechanical interactions between the thorax and the lung play an important role in promoting lung growth and remodelling [37, 48]. As the thorax expands due to somatic maturity, it stretches the lung, creating an outward strain force that is counteracted by the inward elastic recoil of the lung that imposes stress on lung tissue (Figure 1.3). Stress and strain forces activate cell signalling pathways that promote growth of the lung, relieving stress on the lung and ensuring that its size matches that of the thoracic cavity. This feedback loop continues until somatic maturity when the ribcage ceases to grow due to closure of epiphyses

[48].

Following PNX, airflow and ventilatory forces that were previously apportioned to the two lungs are suddenly directed to the residual lung, imposing a significant amount of stress on

11 parenchymal lung components and activating intrinsic cell signalling pathways. To compensate for the loss of respiratory capacity, an increased respiratory force is applied to improve oxygenation from the remaining lung tissue. This typically results in stretching and growth of the remaining lung to fill the vacated thoracic space [42] (Figure 1.3). Distention of blood vessels increases endothelial permeability, enhancing availability of nutrients, growth factors and migration of cells from the circulation that support acinar growth [36]. Increased airflow resistance in the remaining lung generates a traction force that causes airways to dilate and lengthen in an attempt to increase airway surface area, thus facilitating gas exchange [42].

Interestingly, studies performed by Hsia and colleagues using the canine PNX model, have suggested that mechanical stimuli related to air inflation and those resulting from increased perfusion, equally contribute to post-PNX lung growth [63-66]. The researchers further demonstrated that ventilatory mechanical stimuli preferentially promote parenchymal growth over capillary growth, whereas perfusion-related mechanical stimuli promote growth of double capillary profiles by intussusceptive microvascular growth [65, 67]. Thus, mechanical forces play an important role in initiating lung growth after PNX and there appears to be a differential role for ventilatory and perfusion-related mechanical stimuli.

The impact of blunting mechanical forces during regenerative lung growth post-PNX has been demonstrated in experimental animals inserted with an inflatable intrathoracic prosthesis after

PNX [63, 68]. In these animals, regenerative lung growth lags behind that of animals without a prosthesis. When the prosthesis is deflated, lung growth is enhanced indicating that plombage constrains lung growth but does not completely abrogate lung regeneration. Unilateral diaphragmatic paralysis reduces mechanical stretch of the ribcage and blunts lung growth following PNX [69]. Conversely, airway distension with perfluorocarbon accelerates postnatal lung growth [70]. Furthermore, genetic depletion of elastin, a key component of the lung’s

ECM that makes it complaint, reduces mechanical force of the lung parenchyma and negatively

12 impacts lung regeneration post-PNX [40]. Collectively, these studies demonstrate that mechanical force is a crucial determinant of a successful regenerative response following PNX.

Interestingly, in plombage experiments where the prosthesis is inserted at the midline of the thorax to limit lateral lung expansion, the lung is still able to grow by about 20-30% via caudal displacement [63, 68]. This shows that even though mechanical factors are integral in initiating regenerative lung growth, other non-mechanical factors such as growth factors, hormones and hypoxia make a significant contribution to the post-PNX growth response.

Figure 1.3: Effect of mechanical forces on lung growth and remodelling. The outward recoil of the thorax and inward recoil of the lung create mechanical stress that provides signals for lung growth and remodelling, which in turn relieve stress. Adapted from Hsia [48].

13 1.7.2- Hypoxia

Hypoxia has been associated with improved regeneration of the lung following PNX. Sekhon and colleagues compared lung regeneration in rats exposed to either hyperoxia or hypoxia [71].

Rats exposed to hypoxia had greater lung volumes and surface area compared to hyperoxic rats. Consistent with an effect of oxygenation correlating with lung size, animals living at higher altitudes have bigger lung volumes than animals of the same species at lower altitudes, suggesting that hypoxia is a stimulant for postnatal lung growth. Following PNX, hypoxia- inducible transcription factors (HIFs) are upregulated [72-74]. Downstream targets of HIFs, including erythropoietin receptor (EPO-R) and vascular endothelial growth factor (VEGF), have been shown to be important in neovascularisation and neoalveolarisation post-PNX [73,

75]. Furthermore, induction of HIFs by hypoxia contributes to proliferation and differentiation of AECIIs [72]. Therefore, hypoxia supports many facets of lung regeneration including vascularisation and neoalveolarisation, thus is a critical determinant of successful lung regeneration post-PNX.

1.7.3- Age

In both humans and experimental animals, the propensity for lung regeneration diminishes with age. Numerous studies have shown that compensatory lung growth occurs in children and adolescents following lung resection [2, 76-78]. For example, in a longitidunal study published by Laros and Westermann [2], 230 patients ranging in age from 2-40 years at the time of PNX surgery, were followed for over 30 years to assess their lung function post surgery. Excluding

32 patients with an abnormal vital capacity less than 50%, there was an age-dependent restoration of lung function. In children less than five years of age, vital capacity values were nearly equivalent to those predicted for two lungs, suggesting compensation by neoalveolarisation. Patients aged 6-20 years were reported to have undergone greater compensatory lung growth than older patients in an age-dependent manner, although it was

14 also suggested that hypertrophy rather than hyperplasia was the predominant mechanism of compensation. As highlighted earlier (Section 1.7.1), the ability of young children to undergo neoalveolarisation is mainly attributed to the fact that their unfused epiphyses permit thoracic expansion and increased potential for new lung growth [42]. It has also been proposed that developmental pathways may still be active in children that may contribute to postnatal lung regeneration [79].

Mice do not undergo epipheseal closure therefore retain the capacity to expand their lungs throughout adulthood [60]. However, an interesting study by Paxson et.al [4] demonstrated an age-dependent decline in lung regeneration post-PNX in mice. They compared post-PNX lung growth in mice aged 3, 9 and 24 months old and demonstrated that the regenerative capacity of the lung starts to decline at 9 months and is almost absent at 24 months. [2, 77]. Paxson et.al.

[4, 80] further suggested that diminished capacity for lung regeneration in adults may be attributed to loss of fibroblast clonogenicity with a bias towards a myofibroblastic phenotype.

Fibroblasts synthesise lung collagen and elastin, thus age-related loss of fibroblast activity results in loss of ECM elasticity and failure to transmit mechanical signals necessary for lung regeneration [4].

Further insight into the contribution of age to regenerative lung growth has been gained from a study by Jackson and colleagues [81] that compared the post-PNX response of telomerase- deficient mice and wild-type controls. Telomere shortening is a natural consequence of aging and telomerase dysfunction is associated with lung diseases [81]. Telomerase null mice showed diminished lung regeneration and survival following PNX, illustrating that functional telomerase confers a survival advantage to young animals after PNX and its absence contributes to diminished regeneration in aged animals.

15 1.7.3.1- Evidence of lung regeneration in adult humans

Though the adult human lung is thought to have a limited capacity to regenerate, case studies of adult lung regeneration have been published that challenge this dogma. Brody and co- workers reported evidence of post-natal lung growth in six acromegalic men (condition caused by excess growth hormone) involved in the study [82]. The acromegalic men had disproportionately larger lung volumes, higher than predicted vital capacity, lung compliance and anatomic dead space. Lung elastic recoil and transpulmonary pressures were normal with no evidence of air trapping, suggesting that the lung was expanding with new tissue being formed. However, the normal diffusion capacity suggests that the lung alveoli were expanding in size rather than forming new alveoli.

A recent case study by Butler et.al [1] presented a more convincing argument for adult lung regeneration using in vivo imaging data. The case involved a 33-year-old woman who had undergone a right-PNX procedure to treat adenocarcinoma. Over a period of 15 years, CT images showed progressive expansion of the patient’s remaining lung, attaining a volume increase of approximately 77% (Figure 1.4A). The patient’s spirometry values were significantly higher than predicted for her age and lung disability. More interestingly, MRI scans acquired after inhalation of hyperpolarised helium-3 gas suggested indirect evidence of neoalveolarisation based on the finding that the radial axial alveolar dimension was similar to that of normal lungs and uniformly distributed across the lung, although the depth was smaller than normal (Figure 1.4B, C). More recently, Ghammad et.al [83] reported compensatory lung growth after left-PNX in a 16-year old patient, although it must be pointed out that the patient’s relatively young age may have contributed to the observed growth response. These cases, although rare, demonstrate that spontaneous lung regeneration after PNX is possible even in adults, and understanding mechanisms of this growth response could yield novel therapeutic targets for treating lung diseases.

A B

Figure 1.4: Evidence of neoalveolarisation in an adult human. (A) CT scans showing an increase in lung volume over a period of 15 years in a woman who had undergone right lung pneumonectomy. (B) Hyperpolarised Helium-3 MRI map of radial acinar airway dimension (R) relative to its mean of 330 µm. R is approximately equivalent to the mean across most of the lung, indicating evidence of neoseptation/neoalveolarisation. (C) Similar map to (B) showing distribution of alveolar depth (h) relative to the mean of 70 µm. Most alveoli were shallower than the mean. Images adapted with minor modifications from Butler et.al [1].

17 1.7.4- Transcription factors It has been postulated that immediate and early activation of transcription factors (TFs) provides a vital link between PNX-induced mechanical force/hypoxic tension and changes in gene expression that subsequently lead to lung growth [36]. Stretch/stress-associated cell signalling pathways such the ERK-MAPK pathway, p38 and c-jun NH2 terminal kinases

(JNK), that play crucial roles in cell proliferation, differentiation and apoptosis, act upstream of TFs to regulate gene expression [46, 84, 85]. The role of TFs in post-PNX lung growth was first reported by Gilbert and Rannels [86] who showed increased expression of c-fos and junB,

30 min after left-PNX in rats. Later, Landesberg et.al [87] conducted a cDNA microarray analysis on the right lungs of PNX mice at 2, 6 and 24 hrs after left-PNX. They identified six

TFs that were upregulated in response to PNX: early growth response gene-1 (Egr-1), Nurr77, tristetraprolin, the primary inhibitor of nuclear factor-kB (IkB-α), gut-enriched Kruppel-like factor (GKLF), and leucine-rich α-2-glycoprotein (LRG)-21. They found that among the six

TFs, EGR-1 was the most highly induced TF (4.7-fold increase relative to controls) with maximum expression occurring at 2 hrs post-PNX. More recently, other TFs that have been implicated in post-PNX lung growth in mice including; Yes-associated protein (YAP)[88],

HIF-1α [72-74], thyroid transcription factor 1 (TTF-1) [89] and nuclear factor kappa B (NFκ-

B) [90]. In this thesis, we investigated the role of EGR-1 and HIF-1α TFs in post-PNX lung growth and how they relate to the IGF-1 signalling pathway.

18 1.7.4.1- Early growth response (EGR)-1

EGR-1, also known as NGFI-A, ZNF-225, ZIF-268 or KROX-24 [87], is a an 80-82 kDA protein with 533 amino acids that was discovered nearly 30 years ago when searching for genes activated by nerve growth factor in the rat PC12 adrenal tumour cell line [85]. EGR-1 is a zinc finger (ZF) transcription factor that belongs to a family of related proteins designated EGR-1-

4 that are defined by the presence of three ZF domains each consisting of two cysteine and histidine motifs that recognise GC-rich promoters with the consensus sequence

GCG(G/T)GGGCG [91] (Figure 1.5A). EGR-1 is expressed at low levels in many tissues and is primarily located in the nucleus but occasionally in the cytoplasm in cancerous cells [92]. It is rapidly and strongly induced by injurious stimuli, growth factors, cytokines, mechanical force as well as hypoxia [91, 93]. The mechanisms by which EGR-1 induces gene expression in response to these stimuli is not well understood. A proposed model based on studies examining the effect of mechanical injury in endothelial cells and venous smooth muscle cells is shown in Figure 1.5B [94-97]. Stretch, growth factors and hypoxia activate stress-activated protein kinases, including p38, MAPK, ERK and JNK, lead to phosphorylation of ternary complex factors and their association with serum response factors and the EGR-1 basal transcriptional apparatus, resulting in transcription of EGR-1 at serum response elements [85].

At the basal transcriptional machinery of target genes, EGR-1 competitively displaces the constitutively activate TF, specificity protein 1 (Sp1) or other TFs that recognise GC consensus sites similar to EGR-1 such as Wilm’s tumour suppressor (WT-1) [85]. EGR-1, unlike Sp1, requires help of transcriptional co-activators CREB-binding protein (CBP) and p300 to augment gene expression. It may also require help from additional TFs or alternatively interact with other transcriptional co-activators in the absence of p300/CBP [85]. The activity of EGR-

1 is negatively regulated by NGFI-A binding proteins 1 and 2 (NAB-1 and NAB-2) (Figure

19 1.5A). Whereas NAB-1 is constitutively and ubiquitously expressed, NAB-2 expression is restricted to specific tissues and only induced by EGR-1 activation [85].

1.7.4.2- Hypoxia-inducible factor (HIF)-1α

HIF-1 is a basic helix-loop-helix TF that comprises two subunits, HIF-1α and HIF-1β [98].

HIF-1β is constitutively expressed whereas HIF-1α expression is induced by hypoxia. Under normoxic conditions, HIF-1α is constantly synthesised but hydroxylated and targeted for degradation by E3 Ubiquitin ligases [98]. Under hypoxic conditions, prolyl- and asparaginyl hydroxylases which require oxygen and iron for activity, are inhibited, thus allowing for accumulation of HIF-1α, its dimerisation with HIF-1β, nuclear translocation and transcriptional activation of genes with hypoxia response elements [98]. HIF-1α is not only regulated by changes in oxygen tension but can also be activated in response to injury, stress, anaemia and mechanical stretch [74]. Mechanical force can regulate HIF-1α activity through oxygen- independent mechanisms. Mechanical strain distorts ion channels that produce reactive oxygen species, creating an oxygen tension gradient that promotes HIF-1α activity in the absence of overt organ hypoxia [99]. However, it is important to note that while hypoxia regulates HIF-

1α via a post-translational mechanism that involves HIF-1α stabilisation, mechanical force can induce HIF-1α at both a transcriptional level (mRNA expression) and post-translational level

[100]. Therefore, mechanical tension created by PNX surgery has the potential to induce both

EGR-1 and HIF-1α expression and activity.

20 A EGR-1

EGR-2 EGR-3 EGR-4 Legend = Repression site = Zinc Fingers = NGF1-A binding protein

Figure 1.5: Structure and mechanism of EGR-1 transcription factor.

(A) General structure of EGR-1 family of proteins. EGR proteins are defined by the presence of three zinc finger (ZF) binding domains that target the GC-rich region of target genes. EGR expression is regulated by NGFI-A binding protein (NAB) which binds to a repressor site present in all EGRs except EGR-4. Image adapted with minor modifications from [101]. (B) In response to physical forces, growth factors and hypoxia, stress activated kinases including the JNK and ERK pathways are activated. This leads to phosphorylation of ternary complex factors (TCF), which recruits serum response factor (SRF) to induce Egr-1 transcription in concert with its basal (B) transcriptional machinery. EGR-1 displaces the transcription factor (TF) Sp1 at GC-rich target genes then recruits CBP and p300 to augment transcription of target genes. It may also require other TFs. EGR-1 transcribes several target genes most notably growth factors which in turn lead to cell proliferation, migration, growth and survival. Image adapted with major modifications from Silverman and Collins [85].

21 1.8- Cellular mediators of post-PNX lung growth

Repair and regeneration of the lung following injury is facilitated by residual differentiated cells and resident progenitor cells with stem-cell like properties. Cells of the epithelial, mesenchymal lineages and cells recruited from circulation, as well as their progenitors, have all been shown to play a vital role in reconstituting the architecture of the lung after PNX.

1.8.1- Alveolar epithelial cell progenitors

Using injury models induced by agents such as bleomycin, sulphur dioxide, naphthalene and influenza virus infection, researchers have shown that the airway is populated by multiple progenitor cells including basal cells, club cells and neuroendocrine cells that facilitate airway repair and homeostasis (reviewed in [17, 102-104]). Since post-PNX lung growth is a process that primarily involves neoalveolarisation, only progenitor cells that contribute to alveolarisation will be reviewed.

1.8.1.1- Club cells

Club cells (previously known as Clara cells) are non-ciliated secretory cells that populate most of the tracheal and bronchial airway (Figure 1.6). They are identified by the expression of club cell secretory protein (CCSP, also known as Scgb1a1). Club cells are known to mediate repair of the upper and lower respiratory tract after injury by differentiating to ciliated cells and goblet cells, however their contribution to alveolarisation is highly debated [105-108]. Using a

Scgb1a1-CreER knock-in mouse line to track the lineage of club cells, Rawlins et.al [108] argued that club cells do not participate in alveolar homeostasis, development or repair following hyperoxia, SO2 or naphthalene-induced injury. However, a subsequent study by the same group [109] and two separate studies by Zheng and colleagues [110, 111] used a similar lineage tracing technique to show that club cells contribute to AECI and AECII repair

22 following bleomycin and influenza-induced injury, supporting a role for club cells in alveolar repair. Similar club cell lineage tracing experiments are yet to be reported in the PNX model, however Hoffman et.al [40] showed that club cell proliferation is a feature of post-PNX lung growth and that blunting the mechanical response generated by PNX inhibits club cell proliferation, raising the possibility that club cells contribute to neoalveolarisation post-PNX.

1.8.1.2- Bronchoalveolar stem cells (BASCs)

The bronchoalveolar duct junction (BADJ) is made up of poorly defined cuboidal epithelial cells that open into the alveolar lining (Figure 1.6). A minority of BADJ cells (<1%) called

BASCs, are thought to possess stem cell like properties [17]. BASCs are identified by co- expression of CCSP and surfactant protein C (SftpC) and can potentially give rise to both bronchial airway cells and alveolar cells [112]. Nolen-Walston et.al [113] used the murine PNX model to show that BASCs contribute to reconstitution of AECIIs and AECI differentiation during re-alveolarisation post-PNX. They reported that BASC proliferation peaks at 14 days after PNX and can contribute up to 25% of epithelial regrowth post-PNX. However, the presence of BASCs and their contribution to bronchiolar or alveolar repair/regeneration still remains a subject of contention [108] due to absence of markers that can uniquely identify this cell population. It is argued that BASCs may just be mislabeled club cells or AECIIs due to tamoxifen overdose during the CreER lineage tracing system [17]. Identification of novel markers and methods of isolating BASCs will help in the characterisation of BASCs and their function.

23 1.8.1.3- Alveolar type two epithelial cells (AECIIs)

AECIIs form part of the structural unit of the alveolar lining (Figure 1.6). They are cuboidal cells that are responsible for production of surfactant which helps lower the surface tension of the fluid lining alveoli thus preventing lung collapse [17]. The ability of mature SftpC+ AECIIs to undergo continued renewal overtime and give rise to AECI following injury was recognised more than 40 years ago [114, 115]. This finding has been confirmed in more recent studies employing a SftpC-CreERT2 knock-in allele that tags AECIIs and their progenitors [116]. It is now understood that in the absence of injury, AECIIs undergo limited expansion and differentiation. However, injury induced by agents such as bleomycin, hyperoxia and diphtheria toxin increases the rate of clonal expansion of AECIIs and their differentiation to

AECI, whereas stress due to cellular damage or aging attenuates the stem cell capabilities of

AECIIs [17].

Numerous studies have shown that proliferation of AECIIs and their differentiation to AECIs is a prominent feature of post-PNX lung growth that helps in functional recovery [113, 116-

118]. Additionally, Chapman et.al. [119] identified a unique population of AECIIs within the alveolar epithelium, characterised by SftpC-/α6β4+ expression, that can give rise to SpC+

AECIIs, thus act as an additional progenitor for AECIIs and subsequently AECIs (Figure 1.6).

Therefore, AECIs may be derived from multiple sources during lung regeneration. However, it remains to be determined whether AECIIs in the alveolar lining are the principle source of

AECIs following injury and whether there is variation in the regenerative potential among

AECIIs.

1.8.1.4- Alveolar type one epithelial cells (AECIs)

Mature AECI cells have a thin and flat shape and lie near capillaries for efficient gas exchange

(Figure 1.6). AECIs are characterised by expression of aquaporin-5 (Aqp5) and podoplanin

(Pdpn) [120]. Traditionally, AECIs are regarded as terminally differentiated cells with limited

24 proliferative potential. However, discovery of novel AECI markers homeodomain-only protein homeobox (Hopx) and advanced glycosylation end-product specific receptor (AGER), combined with the CreER lineage tracing approach, has enabled researchers to assess the ability of AECI to proliferate and de-differentiate to AECIIs [120-123]. Jain and colleagues

[120] showed that Hopx+ AECIs proliferate and de-differentiate to AECIIs and contribute to alveolarisation following left-PNX in mice. Interestingly, a recent study by Wang et.al [123] used RNAseq and a Igfbp2-CreER mouse line to show that Hopx+/Igfbp2+ AECIs are terminally differentiated whereas Hopx+/Igfbp2- AECIs are abundant during post-natal alveolarisation and can give rise to AECIIs after PNX (Figure 1.6). These findings illustrate an unusual plasticity of AECIs and forms an important basis for future studies into the potential roles of AECIs, especially in pathological conditions such as emphysema where there is a massive depletion of AECIs and AECIIs.

1.8.1.5- p63+/Krt-5+ alveolar progenitors

Current models of lung regeneration/repair favour the hypothesis that airway progenitors make little or no contribution to alveolar regeneration. However, recent findings from influenza injured mice have challenged this dogma. Although transformation-related protein 63 (p63) and cytokeratin-5 (Krt-5) are basal cell markers, which in mice are only located in the trachea and proximal bronchi, Kumar and co-workers [124] reported evidence of a p63+ distal airway epithelial progenitor that gives rise to Krt-5+ pods that migrate to denuded areas of the alveolar parenchyma to contribute to alveolarisation following H1N1 influenza-virus infection (Figure

1.6). The origin of p63/Krt-5+ cells remains a subject of contention [125-127] and it appears that these progenitors may only be activated after diffuse alveolar damage, as there no reports of Krt-5+ pod formation in other injury models including the PNX model.

25 Neuroendocrine cells (Cgrp+) Variant Club cells (Scgb1a1+, Cyp2f2-) Ciliated cells p63+ distal airway progenitor

p63/Krt-5+ pods Club cells (Scgb1a1/CCSP+, Cyp2f2+)

BASCs (Scgb1a1+, SftpC+)

AECII (SftpC+, Abca3+, Lamp3+) Secondary septum

Primary septum Alveolar Macrophage

AECII atypical progenitor (SftpC-, α6β4+)

Capillary endothelial cells (CD31+) AECI (Aqp5, Pdpn, AGER) AECI progenitor (Hopx+, Igfbp2-)

Figure 1.6: Epithelial cell progenitors that contribute to neoalveolarisation.

This figure depicts the relative location of epithelial progenitor cells that have been shown to participate in neoalveolarisation using various injury models including bleomycin, naphthalene, influenza and PNX mouse models. Markers used to characterise each cell type are indicated in parenthesis. Lineage tracing studies have highlighted a novel role for airway progenitor cells in alveolar repair and regeneration. CCSP+ club cells can self-renew and give rise to ciliated cells to propagate airway repair. Through mechanisms that are not well established, club cells can also give rise to AECIIs and AECIs. Club cells are also characterised by expression of Cyp2f2, a variant of cytochrome P450. Club cells with low expression of Cyp2f2 (variant club cells located close to neuroendocrine bodies) are resistant to naphthalene-induced injury and help regenerate denuded airways and alveolar regions. More recently, p63+ distal airway progenitors have been identified using the influenza-injury mouse model. Activation of p63+ progenitors following H1N1 influenza virus infection gives rise to p63+/Krt-5+ pods that can migrate from the distal airway epithelium to denuded alveolar regions to participate in neoalveolarisation. BASCs at the bronchoalveolar duct junction can differentiate to airway club cells and AECIIs. SftpC+ AECIIs surfactant producing cells are regarded as the classic progenitor for AECIs. However, other cells in the alveolar region contribute to AECI differentiation. A variant of AECIIs characterised by SftpC-/α6β4+ expression, constitutes an additional source of AECIIs and AECIs. AECIs, which cover 90% of the alveolar epithelium, are regarded as terminally differentiated and characterised by expression of Aqp5, Pdpn, Hopx and AGER. Plasticity of AECIs has recently been demonstrated using lineage tracing techniques tagging Hopx and Igfbp-2 markers [123]. It has been demonstrated that Hopx+/Igfbp2- AECIs can dedifferentiate to AECIIs.

26 1.8.2- Mesenchymal stromal cells in lung regeneration

The mesenchyme plays an important role in supporting alveolarisation during lung development and post-natal lung regeneration. Matrix components synthesised by lung resident mesenchymal cells provide a scaffold for alveolar cell growth. Mesenchymal cells also produce essential ligands that stimulate growth of alveoli.

Several studies have demonstrated the involvement and necessity of mesenchymal stromal cells in post-PNX lung growth. Paxson et.al [4] reported that proliferation of lung resident mesenchymal stromal cells (LRMSCs) precedes that of AECIIs and endothelial cells, peaking at day three post PNX, at least two days before non-LRMSCs. They also compared mRNA expression profiles of LRMSCs between young mice (3 months old) and older mice (9 and 17 months old) and found that aged mice depicted a genetic signature consistent with reduced matrix clonogenicity and elasticity, which correlates with reduced capacity to regenerate lung post-PNX in old mice.

Platelet derived growth factor (PDGF)-A signalling through PDGF receptor (PDGFR)α is required for secondary alveolar septation during postnatal lung development. PDGFRα expressing interstitial lung fibroblasts home to the distal tips of alveolar septa, express high levels of α-SMA and produce elastin required for alveolar septa formation [128] (Figure 1.7).

A subset of PDGFRα expressing cells give rise to α-SMA myofibroblasts which support alveolar septa formation post-PNX [129]. Chen and colleagues [129] proposed that lowly expressing PDGFRα cells are progenitors of myofibroblasts whereas high PDGFRα expression reduces α-SMA expression. This was confirmed using a dominant negative FGFR line which exhibited high expression of PDGFRα, dysregulated α-SMA expression and reduced re- alveolarisation, illustrating that FGF signalling regulates PDGFR signalling during post-natal alveolarisation. In another study, suppression of FGFR signalling using a dominant negative

27 receptor, similarly upregulated PDGFRα, reduced α-SMA and impaired retinoic acid-induced re-alveolarisation [130]. Administration of rosiglitazone, a PPARγ agonist, which has been shown to protect against bleomycin induced fibrosis [131-133], increases PDGFRα expression, attenuating α-SMA expression [129]. Therefore, it appears that although PDGFRα+ mesenchymal precursors are important for α-SMA myofibroblast differentiation and alveolar septa formation, over-expression of PDGFRα impairs myofibroblast differentiation.

1.8.3- Endothelial cells in lung regeneration

The close association of capillaries and alveoli suggests that endothelial cell proliferation is indispensable for functional recovery of the lung after PNX and that communication between the epithelium and vasculature is likely to be of mutual benefit to both cellular compartments

(Figure 1.7). Previous studies on the murine PNX model have shown that endothelial cell proliferation and neovascularisation is an integral feature of post-PNX lung growth [47, 134,

135]. Endothelial cell proliferation peaks at day seven post-PNX and is particularly evident in areas of active alveolarisation. Angiogenesis related genes, such as endothelial nitric oxide synthase (eNOS) [79], VEGF [136], HIF-1α [74] and others [137] are upregulated in response to PNX. Additionally, regenerative lung growth after PNX is impaired in the eNOS deficient mouse model [138].

More recently, Ding and colleagues [139] showed that after PNX, endothelial cells upregulate

VEGFR2 and FGFR1 receptor expression which promotes secretion of matrix metalloproteinase (MMP)-14. MMP14 in turn cleaves EGF ligands which promote epithelial cell proliferation (including AECIIs) and re-alveolarisation. Rafii et.al [140] reported similar findings in addition to demonstrating that SDF-1 (also known as CXCL12) derived from platelets augments endothelial cell MMP-14 expression and subsequent neoalveolarisation in

28 response to PNX (Figure 1.7). Therefore, a potential mechanism by which platelets and endothelial cells participate in alveolarisation is by secretion of vital paracrine growth factors that promote alveolarisation. Taken together, these studies show that vascularisation is critical for successful lung regeneration after PNX.

1.8.4- Macrophages in regenerative lung growth

Macrophages play a vital role in regeneration of several organ types including the liver, kidney and skeletal muscle [141-143]. Anti-inflammatory M2 type macrophages in particular are associated with repair and regeneration [141-143]. Macrophages are regarded as the major source of extrahepatic IGF-1 and production of colony stimulating factor (CSF)-1 contributes to macrophage differentiation, IGF-1 expression, post-natal growth and organ development

[144]. Of note, Tonkin et.al [141] showed that conditional knockout of IGF-1 in myeloid cells reduced local muscle IGF-1 expression and significantly compromised macrophage M2 macrophage polarisation, resolution of inflammation and consequently impaired tissue regeneration. Using the murine PNX model, Chamoto and colleagues [145] detected a significant increase of alveolar macrophages (AM) in bronchoalveolar lavage fluid (BALF) 14 days after PNX. Parabiosis experiments suggested that less than 1% of AMs are derived from peripheral blood, indicating that increase in AMs resulted from local proliferation rather than blood derived myeloid cells. However, a later study by the same group [146] showed that

CD11b/CD18-/- null mice show an impaired regenerative response following PNX, indicating that migrating CD11b leucocytes are required for optimal lung regeneration. However, CD11b expression is not restricted to migrating leucocytes; it is also expressed by interstitial and airspace leucocytes. A more recent study by Lechner et.al. [147] more specifically showed that

CCR2 expressing monocytes accumulate in regenerating lungs and are required for AECII proliferation and post-PNX lung growth (Figure 1.7). They further demonstrated the

29 requirement of two cytokines (IL-4 and IL-13) in macrophage M2 polarisation to support

AECII proliferation. Thus, it appears that both locally derived and blood-derived macrophages make a vital contribution to post-PNX lung growth.

To summarise, regeneration of the lung following PNX is a complex and dynamic process that involves cellular participants of many different types, reflecting the complex architecture of the lung. Regeneration of the alveolar epithelium from progenitor cells is regarded as the gold standard measure of lung regeneration. However, other cell types such as endothelial cells, interstitial fibroblasts and macrophages are clearly important to ensure successful regeneration of the lung, thus necessitating communication between these different cellular compartments.

Alveolar Macrophage

Primary septum with Secondary septum with Collagen and elastin fibres at septal tips Collagen and elastin fibres at septal tips Fibroblasts Myofibroblasts (α-SMA+)

Capillary endothelial cells (CD31)

Platelets

Interstitial macrophage AECII progenitor (SftpC-, α6β4+) Monocytes (CCR2+) Lipofibroblasts (PPARγ+)

Figure 1.7: Mesenchymal stromal cells involved in alveolar regeneration. Messenchymal cells play a supportive but integral role in neoalveolarisation. PPAR-γ+ lipofibroblasts often located at the base of AECIIs support stem-cell function of AECIIs. α- SMA myofibroblasts lay down ECM elements at septal tips during secondary septa formation, a process that increases alveolar surface area. FGFR2 and PDFRα signalling plays an important role in lipofibroblasts and myofibroblasts differentiation from fibroblast progenitors. Capillary endothelial cells mediate gas exchange and equally make a vital contribution to neoalveolarisation. Capillary endothelial cells invade secondary septa during neoseptation by intussusceptive microvascular growth. Platelet-derived SDF-1 activates endothelial cells to promote release of ligands that support alveolarisation. Macrophages, in particular M2 type macrophages, secrete vital ligands such IGF-1 that promotes AECII proliferation. After PNX, alveolar and interstitial macrophages are mostly derived from local cells, but a recent study has highlighted a critical role for CCR2+ blood derived monocytes.

31 1.9- Growth factors in post-PNX lung growth

Growth factors (GFs) play an important role in post-PNX lung growth by converting mechanical and hormonal signals to proliferative signals. Examples of GFs that have been reported to be involved in post-PNX lung growth include epidermal growth factor (EGF) [148], keratinocyte growth factor (KGF) [149], retinoic acid (RA) [150], VEGF [151], hepatocyte growth factor (HGF) [152], platelet-derived growth factor (PDGF) [153], and IGF-1[7]. Table

1.1 summarises the role of specific GFs in post-PNX lung growth.

Many GFs involved in early lung development have been implicated in post-PNX lung regeneration and repair. There is a lot of functional overlap among GFs. However, some GFs have unique attributes that cannot be substituted by other ligands. IGF-1 is proposed as one such unique GF. Unlike many GFs which are limited to mitogenic effects on a particular cell type, IGF-1 has pleiotropic effects and participates in both proliferation and differentiation cellular activities [154]. IGF-1 is also critical for somatic growth and targets nearly every cell type in the body [5]. For this and other reasons that will be discussed, IGF-1 was selected as the primary focus of this study in our bid to understand mechanisms of adult lung regeneration post-PNX. Moreover, the role of IGF-1 in post-PNX lung growth has not been thoroughly investigated.

32 Growth Effect on post pneumonectomy lung growth Factor Mouse model Other models - Exogenous EGF increased lung weight, lung volume and - Increased EGFR, AECII proliferation two weeks after lobectomy EGF EGFR expression post-PNX in rats [148]. in pigs [155].

- Upregulation of Hgf mRNA and protein in the right lung and liver plus increased serum levels of HGF and increased Hgfr mRNA expression in the right lung after left-PNX in HGF mice [152]. - HGF administration increased cell proliferation whereas anti-HGF administration attenuated cell proliferation [152].

- Increased expression of lung Igf-1 mRNA identified by - Increased IGF-1 in rat lavage at 2 and 6 days post left-PNX [7]. microarray and qRT-PCR analysis in young adult mice - No significant difference in lung Igf-1 mRNA and serum IGF-1 following left PNX [10]. levels between PNX rats and sham controls [156]. IGF-1 - Increased in lamb lungs 21 days after PNX [9]. - No significant increases in serum IGF-1 identified in PNX rabbits relative to shams [157].

- Delivery of KGF (i.p.) increased lung weight, volume, cell proliferation and neoalveolarisation in a rat PNX model [149].

KGF - KGF gene transfection following trilobectomy in rats increased KGF/KGFR expression, cell proliferation and neoalveolarisation [158].

- PDGFRα kinase activity regulates - Pdgf-b mRNA and PDGF-BB protein were moderately elevated myofibroblast/lipofibroblast differentiation of interstitial but not significantly different from controls 21 days post-PNX PDGF lung fibroblast after left-PNX in mice [159]. in rats however adsorption of PDGF-BB with a soluble PDGFRβ decoy attenuated PNX-induced CP [153].

33 - RA administration increased lung weight, volume, cell proliferation and EGFR expression at 21 days post-PNX in rats [160]. - RA administration after left-PNX in adult male dogs had no RA significant effect on compensatory lung growth (CLG) [161]. - RA administration after right-PNX in adult male dogs increased capillary endothelial cell growth [162, 163].

- Daily intranasal delivery of VEGF increased endothelial cell proliferation, lung volume and alveolar count 4 days after left-PNX in mice [164]. - With VEGF systemic administration post left-PNX in mice, CLG was completed in 4 days as opposed to 10 in controls [151]. VEGF - VEGF levels were increased in the lung and plasma post- PNX; VEGF transgenic mice have enhanced CLG; VEGFR knockout mice show reduced CLG after left-PNX; VEGFR+ proangiogenic progenitors recruited from the bone marrow [165].

Table 1.1: Growth factors implicated in post-PNX lung growth.

EGF=epidermal growth factor, HGF= hepatocyte growth factor, IGF-1= insulin-like growth factor-1, KGF= keratinocyte growth factor, PDGF= platelet-derived growth factor, RA= retinoic acid, VEGF= vascular endothelial growth factor.

34 1.10- The IGF family of proteins

Insulin-like Growth Factors (IGFs) are a family of evolutionarily conserved proteins with high homology to insulin and are believed to have evolved from insulin [166]. IGFs are involved in a diverse range of cellular processes including proliferation, survival, migration and differentiation, throughout development and adulthood [154]. The existence of IGFs was first suggested by Salmon and Daughaday in 1957 whilst experimenting on the effects of growth hormone (GH) in hypophysectomised rats [167]. Since their discovery, IGFs have been known by many different names including sulphation factors, somatomedins (because they mediate effects of GH) and non-suppressible insulin-like factors (NSILAs). Changes in nomenclature reflected the controversy and uncertainty surrounding characterisation of IGFs at the time. It was not until 1987 that a consensus was reached to finally change the nomenclature from somatomedins or NSILAs to IGF-1 and IGF-2 [168].

The IGF family of proteins comprises the two IGF ligands IGF-1 and IGF-2, their receptors

IGF-1R and IGF-2R and at least six IGF binding proteins designated IGFBP-1-6 that regulate the bioavailability of IGFs in serum and interstitial spaces (Figure 1.8).

1.10.1- IGF-1 and IGF-2 ligands

IGF-1 and IGF-2 share about 65% sequence similarity to each other and 50% sequence homology to insulin [169]. Mature IGF-1 is 70 amino acids (AA) long and slightly basic whereas IGF-2 is 67AA long and slightly acidic [170]. Given the structural similarity between

IGFs and insulin, the ligands can cross activate their respective canonical receptors (albeit with lower affinity) and utilise similar intracellular pathways, although they elicit quite disparate effects, with insulin mainly involved in metabolism while IGFs are primarily involved in somatic growth (Figure 1.8).

35 IGF-1 and IGF-2 both contribute to embryonic development but it is thought that IGF-2 plays a more important role during prenatal development but has a less significant role in post-natal maturity [171, 172]. The gene encoding IGF-2 is maternally imprinted and aberrant imprinting is implicated in the somatic overgrowth syndrome Beckwith-Wiedemann, obesity, sporadic

Wilm’s tumour and many other cancers [173]. However, most research on IGFs has primarily focused on IGF-1 and less is known on the role of IGF-2 in post-natal development.

1.10.2- IGF-1 receptor (IGF-1R) and IGF-1 signalling

IGF-1R is a member of a family of transmembrane tyrosine kinases which includes the insulin receptor (IR) and orphan insulin receptor-related receptor (IRR) [174]. IGF-1R is ubiquitously expressed in foetal and postnatal tissues and is essential for foetal growth and postnatal development as demonstrated by the lethal phenotype of IGF-1R knockout mice [175-177].

IGF-1R is a heterotetrameric receptor consisting of two ligand binding α-subunits and two membrane spanning β-subunits attached by disulphide bonds akin to the structure of the insulin receptor [178] (Figure 1.8). All four subunits are derived from a single mRNA consisting of

21 exons encoded at chromosome 15q25-26 [179]. IGF-1R pre-peptide is synthesised as a single-chain polypeptide (about 1367AA) that undergoes proteolytic cleavage, glycosylation and dimerisation to yield an α2β2 receptor [180].

IGF-1R binds IGF-1 with a much higher affinity compared to IGF-2 and insulin [179]. IGF-1 binds to cysteine rich domains in the α-subunit of IGF-1R triggering a conformational change in the β-subunits that leads to phosphorylation of three tyrosine residues (Tyr 1131, 1135 and

1136) in the intracellular portion of the IGF-1R β subunit, called the activation loop (ALP)

[181] (Figure 1.8). Phosphorylation of the two β subunits occurs by a trans- autophosphorylation mechanism where one half of the receptor tyrosine kinase phosphorylates the contralateral half and vice versa. The activated IGF-1R can then bind to multiple adaptor

36 proteins that lead to activation of an array of intracellular signalling pathways. The best characterised adaptor proteins are the Src Homology (SH)-2 domain-containing proteins

Insulin Receptor Substrate (IRS)-1, IRS-2 and Shc [180] (Figure 1.8).

IRS-1/2 have multiple tyrosine phosphorylation residues (over 20) that interact with SH-2 domain containing proteins including Grb-2 and p85, the regulatory subunit of phosphoinositol-3-kinase (PI3K) [179]. PI3K activates the threonine/serine protein kinase Akt

(also known as protein kinase B) by phosphorylating phosphoinositol diphosphate (PIP2) to phosphoinositol triphosphate (PIP3) [180]. Akt in turn interacts with multiple cellular pathways that are related to protein synthesis, apoptosis, cell proliferation and glucose metabolism [182]

(Figure 1.8). Grb-2 when activated (phosphorylated) by Shc or IRS-1/2, activates the membrane bound Sos, a GDP to GTP exchange protein which interacts with p21 ras. Ras activates the MAPK-ERK pathway which is integral for cell proliferation [180] (Figure 1.8).

1.10.3- IGF-2R

IGF-2R is a single chain membrane spanning glycosylated protein that belongs to the mannose-

6-phosphate (M-6-P) family of receptors [171]. It is about 2,264aa long and is encoded by chromosome 6q25-q27 in humans [180]. IGF-2R binds to IGF-2 with high affinity at M-6-P repeat domains but can also bind other M-6-P bearing proteins such as thyroglobulin and latent

TGFβ1 [180]. However, unlike IGF-1R, IGF-2R does not transmit intracellular signals; instead it is involved in transportation of lysosomal enzymes from their sites of synthesis to the endosome via endocytosis [171] (Figure 1.8). Mice lacking IGF-2R have accelerated somatic growth indicating that IGF-2R is involved in clearance of IGF-2 [183].

Figure 1.8: The IGF family of proteins and IGF-1 signalling. IGF-1 and IGF-2 principally signal through IGF-1R although they can bind with lower affinity to insulin receptors (IR) and IR/IGF-1R hybrids. IGF-1R similar to IR is a tyrosine kinase constituting two extrinsic alpha subunits and two membrane spanning beta subunits. IGF-2R does not activate cell signalling pathways but instead targets bound ligands for lysosomal degradation. Following ligand binding, IGF-1R tyrosine kinases are activated by an auto-transphosphorylation mechanism within the activation loop (ALP) comprising three main tyrosine (Tyr) residues (Tyr 1131, 1135 and 1136). This activates Src homology containing proteins including Src homology-2 (Shc) and Insulin receptor substrate (IRS-1/2). Shc activates the Ras-Raf-MERK-1/2-ERK-1/2 pathway, leading to nuclear translocation of transcription factors that lead to cell proliferation, differentiation and migration. IRS- 1/2 on the other hand activates the PI3K-PIP3-Akt kinases which are involved in a diverse array of cellular processes including suppression of apoptosis by inhibition of caspases and Bad or promoting Bcl-2 expression, promoting protein synthesis via the mTORC1 pathway or contributing to glucose metabolism via Gsk-3β kinase. The mechanism by which PNX activates IGF-1 signalling is yet to be investigated. We hypothesise that stretch and hypoxic tension created by PNX activates the transcription factors EGR-1 and HIF-1α. Whist it is not novel that these transcription factors are activated by these stimuli, how they relate to the IGF-1 pathway, particularly in the context of PNX, remains to be explored.

38 1.10.4- IGF binding proteins (IGFBPs)

IGFBPs are secreted glycoproteins that regulate the bioavailability of IGFs in serum and interstitial fluids by binding to IGFs with equal or higher affinity than IGF-1R [184]. IGFBPs also minimise potential cross binding of IGFs to the IR [185]. To date, at least six IGFBPs have been described in the literature; these are designated IGFBP 1-6 according to the sequence in which they were discovered [186].

1.10.4.1- General structure of IGFBPs

IGFBPs are structurally similar proteins ranging in size from 24-45 kDA [184]. IGFBP peptides have a conserved N and C terminus with a variable central domain [184]. The N and

C termini possess twelve and six cysteine residues respectively that give IGPBPs a globular structure by formation of disulphide bonds (Figure 1.9). The N-terminus is the binding site for

IGFs whilst the C-terminus binds other proteins such as acid labile subunit (ALS) in the case of IGFBP-3 and -5 [185]. The mid (L) domain contains sites for post-translational modification that can undergo phosphorylation, glycosylation and proteolytic cleavage [184].

1.10.4.2- Mechanism of action

About 99% of IGFs in the circulation are bound to IGFBPs [187]. The majority of IGFs are bound to IGFBP-3 which forms a tripeptide complex with liver-derived ALS to significantly increase the half-life of IGF-1 from minutes to over 12 hrs [185]. IGFs can be released from

IGFBPs by several mechanisms including protease-mediated release, binding of IGFBPs to

ECM, by creation of a concentration equilibrium favouring binding of IGFs to IGF-1R, or by phosphorylation of IGFBPs [184] (Figure 1.9).

The function of IGFBPs is not limited to endocrine regulation of IGFs. IGFBPs have been shown to have ligand-independent functions. IGFBP-5 has been shown to act as a growth factor

39 in IGF-1 null mice and is also capable of inducing smooth muscle migration independent of

IGF-1 [188]. Receptors for IGFBP-3 and IGFBP-5 have been reported that mediate signal transduction of these ligands [189, 190]. IGFBP-3 and -5 also possess a nuclear localisation signal (NLS) that favours nuclear transportation over secretion of these ligands, enabling their participation in signal transduction [191]. The presence of an RGD sequence in the C-terminus of IGFBP-1 enables it to interact with α5β1 integrin to promote cell migration [192]. Therefore, the activity of IGFBPs is not limited to modulation of IGF activities; they can function as independent ligands.

Figure 1.9: General structure of IGFBPs and mechanism of IGF liberation. In serum and interstitial spaces IGF-1 forms a complex with IGFBPs (mainly IGFBP-3). Formation of a tripeptide complex with liver-derived acid labile subunit (ALS) further increases the half -life of IGF-1 in serum. IGFBP general structure comprises a mid-terminal L-domain with phosphorylation (P), glycosylation (G) and splice sites, an N-terminus (N) with 12 cysteine (Cys.) residues and a C-terminus (C) with 6 Cys. residues that bind to give IGFBP a globular structure by formation of disulphide bonds (s). IGF-1 binds to the N-terminus while ALS binds to the C-terminus. IGF-1 can be released from IGFBPs to trigger IGF-1R activation by at least four mechanisms (listed 1-4 in diagram). 1- by phosphorylation of IGF-1; 2- by increasing the concentration of IGF-1 creating a concentration gradient that saturates IGFBPs and favours binding to IGF-1R on cell membranes; 3- adhering to ECM proteins; 4- cleavage of IGFBP-IGF-1 bonds by matrix proteases.

40 1.10.5- Sources of IGF-1 The liver is the principle source of endocrine IGF-1 and contributes about 70% of the total pool of IGF-1 in circulation in response to GH stimulation [193]. IGF-1 is also produced by extrahepatic tissues including the lung, where it acts locally via autocrine and paracrine signalling (Figure 1.10).

1.10.5.1- Endocrine regulation of IGF-1

Liver hepatocytes secrete IGF-1 under the influence of GH which is produced by the anterior pituitary gland in the brain in a pulsatile manner [194, 195]. A negative feedback loop exists for the regulation of pituitary GH and hepatic IGF-1 secretion. Neuroendocrine cells in the hypothalamus, when appropriately stimulated, release growth hormone releasing hormone

(GHRH) that is transported by the hypophyseal portal vascular bed to the anterior pituitary where it stimulates secretion of GH by somatotrophs [194, 195] (Figure 1.10). Once GH is secreted, it is transported to the liver where it binds to GH receptors on hepatocytes to promote secretion of IGF-1 via the transcription factor signal transducer and activator of transcription

(STAT)-5b [196]. Excess IGF-1 and GH promote secretion of somatostatin (SS) a neuroendocrine hormone that inhibits GH secretion at the anterior pituitary level, thereby suppressing IGF-1 production [194, 195].

1.10.5.2- Autocrine and paracrine regulation of IGF-1

In an elegant study by Yakar and colleagues [197], it was demonstrated that liver specific IGF-

1 knockout mice, despite having severely reduced serum IGF-1, grow normally. This showed that locally derived IGF-1 can support somatic growth and complements the endocrine IGF-1 pool to support growth and development. GH is also known to influence extrahepatic secretion of IGF-1 but it is thought that local factors such as the presence of IGFBPs, matrix proteases and injurious stimuli play a more important role in determining local availability of IGF-1 [184]

41 (Figure 1.10). Potential cellular sources of IGF-1 include alveolar macrophages, airway cells, parenchymal cells, smooth muscle cells and fibroblasts [198, 199]. The IGF-1R has been identified on almost all lung cell types indicating the ability of IGF-1 to modulate cellular processes in all lung cells via autocrine and paracrine signalling [200]. However, the profile of

IGF-1/IGF-1R and associated binding proteins has not yet been defined in the PNX model. The main source of IGF-1 in the post-PNX lung is also yet to be defined, although prior studies point to alveolar macrophages [7, 147].

Figure 1.10: Endocrine and paracrine sources of IGF-1. Growth hormone (GH) secretion by the anterior pituitary gland in the brain is triggered by growth hormone releasing hormone (GHRH) from the hypothalamus. Once released, GH binds to GH receptors (GHR) on liver hepatocytes to promote release of IGF-1 into serum via the transcription factor STAT-5b. In serum, IGF-1 forms a tripeptide complex with the acid labile subunit (ALS) and IGFBP-3 (also produced by the liver) to prolong the half-life of IGF-1 as it is transported to extrahepatic tissues including the lung. Extrahepatic tissues can also produce local sources of IGF-1 which acts via autocrine and paracrine signalling. Endocrine-derived IGF-1 contributes about 70% of IGF-1 in serum. When in excess, GH and IGF-1 promote secretion of somastatin (SS) by the hypothalamus, which inhibits GH secretion and consequently lowers IGF-1 production.

42 1.10.5.3- IGF-1 levels decline with age in humans

Availability of IGF-1 in serum is not only regulated by the GH-IGF-1 feedback loop and

IGFBPs, other factors such as age, sex, hormones and nutritional status also play a vital role in determining serum levels of IGF-1 [201-203]. However, age is the key determinant of serum

IGF-1 and GH levels in humans. With advancement in age, the frequency and amplitude of GH secretion by the anterior pituitary decreases, contributing to the decline in serum IGF-1 levels, a phenomenon known as somatopause [204]. Bidlingmaier and co-workers [11] conducted a multicenter study involving over 15,000 subjects to establish reference intervals for serum IGF-

1 levels from birth to senescence. Prenatal IGF-1 concentration was measured from cord blood and no differences were found between sexes, although IGF-1 concentration was correlated to birthweight. Close to birth, IGF-1 levels decline and remain low in the first year of life. After the first year, IGF-1 levels increase, peaking at puberty between the ages of 12-15 years old with serum concentrations ranging from a median of 120-260 ng/mL. Discrepancies are seen between males and females as they age towards puberty. Peri-pubertal girls have significantly higher concentrations of IGF-1 and tend to reach peak IGF-1 levels earlier than boys. After puberty, IGF-1 levels decline in both males and females, however women experience a slightly greater rate of decline during the fifth and sixth decade of life due to decline in sex steroids past menopause [11]. In mice, IGF-1 levels peak at 3-4 weeks of age (the adolescent phase in mice). However, unlike humans, mice do not show significant reductions in serum IGF-1 post puberty; as levels of IGF-1 remain relatively high throughout life [205-208].

43 1.10.6- Generation of multiple isoforms from the Igf-1 gene

The human Igf-1 gene is similar to the murine Igf-1 gene and consists of six exons spanning over 100 Kbp of genomic DNA located on chromosome 12q22-24 [209]. Through a combination of different mechanisms including use of multiple transcription start sites, alternative splicing and use of different polyadenylation signals, the Igf-1 gene gives rise to multiple mRNA transcripts of varying lengths that translate to pre-IGF-1 peptides with a unique amino terminal signal peptide and an extension (E) peptide on the carboxyl terminal [154]

(Figure 1.11). Pre-IGF-1 peptides undergo post-translational modifications, including proteolytic cleavage to remove the N-terminal signal peptide and C-terminal E-peptide. The resulting mature IGF-1 protein is a 70 aa long single chain polypeptide linked by three intracellular disulphide bonds with an estimated size of 7.65 kDa [210].

The Igf-1 gene possesses two promoters; one at the start of exon 1 (P1) and another at the start of exon 2 (P2) [154] (Figure 1.11). Usage of P1 and P2 is mutually exclusive and on this basis of the transcription start site, IGF-1 peptides are categorised into class I (P1 start) and class II

(P2 start site) transcripts. Upstream of exon 1 and exon 2 there are multiple transcription start sites which generate signal peptides of varying length. Four transcription start sites (382, 343,

245 and 40bp) upstream of exon 1 have been identified whereas two transcription start sites

(50 and 20 bp) upstream of exon 2 have been reported [210]. Both P1 and P2 promoters can initiate transcription at multiple sites. Studies have shown that Class II transcripts are highly expressed in the liver and are known to have a higher binding affinity to IGFBPs compared to class I transcripts, indicating that they are more amenable to endocrine function [211-213].

IGF-1 isoforms are further classified on the basis of the C-terminal E-peptide which is generated by alternative splicing of exon 5 and exon 6. Exon 3 and exon 4 encode the mature

IGF-1 protein and are conserved in all transcripts [154] (Figure 1.11). Exon 3 contributes to

25 aa of the mature IGF-1 protein and encodes part of the N-terminal signal peptide whilst

44 exon 4 encodes 45 aa of the mature IGF-1 protein and the first 16 aa of all E-peptides [210]. In both mice and humans, transcripts that have exon 5 excluded so that exon 6 is joined to exon 4 are termed Ea transcripts and translate a 35 aa long Ea peptide (Figure 1.11A, B). In humans,

Eb transcripts exclude exon 6, thus terminate with exon 5 [210] (Figure 1.11A) whereas the

Eb transcript of rodents has exons 5 and 6 included (Figure 1.11B). However, inclusion of exon 6 in the mice Eb transcript introduces a frameshift mutation in the sequence such that a stop codon is present within exon 6 resulting in a shorter 41 aa peptide from 52 base pairs

[210]. In humans, the presence of a cryptic splicing site in exon 5 (Figure 1.11A) gives rise to an additional IGF-1 isoform called an Ec transcript [154]. The Ec transcript is made up of both exons 5 and 6 corresponding to the Eb transcript in rodents. The Ec peptide or Eb peptide in rodents is sometimes called mechano growth factor (MGF) because it was initially identified in muscle tissue where it is known to be upregulated in response to exercise and injury [154].

Figure 1.11: Structure of the mouse the Figureof Structure and 1.11: proteolyticcleavage to removeN the IG The transcript. Eb mouse the to similar is that isoform Ec additional an generates 5 exon presence the humansincluded.In6 exontruncated a have transcripts mice, Eb In5. exon to 4joined exonhavetranscripts Eb lengthsgenerated byalternative spli to common are 4 and transcri 3 one Exons transcripts. class called are (P1) 1 exon to prior promoters have that Transcripts The human and mouse

Igf

gene are similar in many respects. They both consist of six exons wi

cingthese of exons. Transcripts thathave exon removed 5 that exon so 4 is joined exon to 6 aretermed Ea trans - signal terminal peptide and C

IGF mature the encode and transcripts all human IGF human

gene. 1 -

E terminal

peptide

pts whereas transcripts that start at exon 2 (P2) are called class two class called are (P2) 2 exon at start that transcripts whereas pts

- 1 protein. Exons 5 and 6 encode C encode 6 and 5 Exons protein. 1

yielding70aa a mature IGF

- 1 pre 1

th two transcription start sites that are mutually exclusive. - undergoes transcripts these from generated peptide

protein. 1

terminal E terminal

of a crypt a of

peptides of various of peptides

ic splicing site insplicing site ic cripts.

46 1.11- IGF-1 in foetal lung development

Several studies have suggested that signalling pathways involved in lung development are also involved in neoalveolarisation post-PNX. Microarray analysis of PNX lungs revealed a significant proportion of overlapping pathways that were similarly activated during the latter stages of lung development and PNX alveolarisation in mice [214]. Park et.al. [215] reported up-regulation of lung developmental markers Sox2, Sox17 and Foxa2, Foxj1 and β-catenin in the airway epithelium of PNX mice. Additionally, Takahashi et.al. [89] more recently reported that knockdown of TTF-1, an early lung developmental marker which was up-regulated after

PNX, delayed the early phase of post-PNX regeneration. Therefore, an understanding of the process of lung development is required to better understand mechanisms governing post-PNX lung regeneration.

Targeted deletion of the IGF genes has shown their relevance in lung development, homeostasis and repair. Mice with a homozygous deletion of the IGF-1 gene (Igf-1-/-) are born

60% the size of wild-type (WT) littermates and have defects in major organ systems including the lung. Most of the pups die shortly after birth apparently due to respiratory failure accompanied by dystrophic respiratory muscles and atelectatic lungs [216, 217]. Igf-1R-/- mice are born 45% the size of WT mice, display delayed saccular development, have collapsed lungs and are unable to breath thus die shortly after birth [177, 218]. On the other hand, a recent study by Galvis and co-workers [219] showed that loss of the histone methyltransferase Ezh2 in the lung epithelium leads to surplus expression of IGF-1 which contributes to defective basal cell differentiation, lung deformation and perinatal lethality. Taken together, these studies suggest that a proper balance of IGF-1 expression is essential for correct lung development.

Whole lung transcriptomic microarray analysis of Igf-1-/- and Igf-1R-/- mice revealed alterations in expression of genes involved in vascularisation, cell growth and cell adhesion, supporting a

47 role for IGF-1 in these important cellular processes which are necessary for lung development

[6, 218]. Interestingly, immune related genes were found to be upregulated pointing to a role for IGF-1 in regulation of inflammation. In Igf-1-/- mice, none of the genes in the IGF signalling axis except Igfbp2 (which was slightly but significantly reduced) were altered, suggesting that

IGF-1 deficiency during embryonic development does not evoke a compensatory mechanism

[6]. These findings were further corroborated by immunohistochemical analysis of the lungs at

E18.5 which revealed a thick mesenchyme, thin smooth muscles and dilated blood vessels showing that absence of IGF-1 affects a broad range of cell types in the lung. The rate of proliferation and apoptosis was significantly increased compared to WT controls. Increase in cell proliferation was proposed to be a compensatory mechanism for delayed lung development during the pseudoglandular stage when proliferation is normally active [6]. Therefore, delay in cellular differentiation and arrest at the canalicular stage of lung development contributes to lung dysmorphogenesis in Igf-1-/- and Igf-1R-/- mutant mice.

In humans, haplo-sufficient mutations in IGF-related genes have been reported and are associated with intrauterine and post-natal growth retardation [220-224]. One of the patients with a deletion mutation of the Igf-1R gene was reported to have lung hypoplasia among many other developmental abnormalities [225]. Additionally, IGF-1 expression is dysregulated in infants with congenital hypoplastic lung conditions such as bronchopulmonary dysplasia and congenital diaphragmatic hernia [226, 227]. Conversely, acromegalic and pituitary gigantic patients who have high serum levels of GH and IGF-1 have been reported to have disproportionately large lungs [82, 228].

Collectively, studies from transgenic IGF mice and cases of human IGF mutations have highlighted a critical role for IGF-1 in lung growth and development. Given the importance of

IGF-1 in lung development, it is reasonable to hypothesise that IGF-1 has an equally important role in adult lung regeneration.

48 1.12- IGF-1 in post-PNX lung growth

Research into the contribution of IGF-1 in post-PNX lung growth extends as far back as the

1980s when Smith and colleagues [229] first reported that serum samples from rabbits at 9-21 days post-PNX induced proliferation of AECIIs in culture but had no effect on lung or skin fibroblast proliferation. Further biochemical analysis of the serum samples revealed the presence of fractions commensurate to the size of somatomedins (IGFs) which they proposed were likely to be responsible for the mitogenic activity of AECIIs. However, in a similar study published by the same group four years later [157], direct measurement of serum immunoreactive somatomedin C (IGF-1) revealed no difference between PNX and controls suggesting that IGF-1 was unlikely to be involved in post-PNX lung growth.

It was not until 1992 that IGF-1 was first directly implicated in post-PNX lung growth in a seminal study published by McAnulty and co-workers [7]. They showed that following left-

PNX in rats, BALF collected at day two and day six but not day 14 post-PNX significantly increased fibroblast proliferation in culture compared to sham and unoperated controls. They then demonstrated the presence of molecular fractions in BALF that had similar weight and biochemical properties to IGF-1 and IGF-1-IGFBP-3 complexes. Indeed, total assayed IGF-1 in BALF from PNX rats at day two and day six was significantly more than in sham and unoperated controls. More interestingly, proliferation of fibroblasts in culture was significantly attenuated in the presence of anti-IGF-1 antibodies but not anti-PDGF antibodies, suggesting that IGF-1 was responsible for increased mitogenic activity of fibroblasts in culture, and its release in the lung following PNX contributed to regenerative lung growth. Furthermore, an independent study by Khadempour and colleagues [8] published in the same year also reported increased levels of IGF-1 in serum of 14 day old pregnant rats two-to-five days post-PNX.

Additionally, Nobuhara and colleagues [9] reported increased expression of IGF-1 in lungs of

49 pneumonectomised lambs 21 days after PNX accompanied by neoalveolarisation, pointing to the involvement of IGF-1 in regenerative lung growth after PNX.

Contrary to the findings by McAnulty et.al. [7] and others [8, 9], a later study by Price et.al.

[156] using the rat left-PNX model reported no significant differences in IGF-1/IGFBPs mRNA expression in the lung (quantified by northern blot densitometry analysis) between

PNX and control animals. Serum levels of IGF-1 and IGFBPs were also not significantly altered compared to sham and unoperated controls. Thus, they concluded, based on these observations, that changes in the IGF-1 signalling axis do not represent a major event of the post-PNX process. The discrepancies in the early findings on the role of IGF-1 in post-PNX lung growth could be attributed to the differences in methodologies, strains of animals used or their hormonal status [156]. It is also worth noting that many of the studies discussed thus far measured IGF-1 protein levels using radioimmunoassay techniques that involve separation of

IGFs from IGFBPs prior to quantification. This can greatly interfere with IGF activity and concentration, thus contributing to variations in results [156].

In a more recent study by Paxson and colleagues [10], transcriptomic microarray analysis of pneumonectomised right lung tissue from 10-12-week-old mice, compared to sham operated controls, revealed upregulation of IGF-1 pathway-related genes with concomitant activation of mesenchymal associated genes and downregulation of pro-inflammatory genes. The researchers further validated these findings by qRT-PCR and showed a 1.4-fold increase in

IGF-1 mRNA in day one PNX mice compared to sham controls at the same timepoint, suggesting that increased expression of IGF-1 is a prominent event of post-PNX regenerative lung growth in mice.

To summarise, data in the literature thus far regarding the role of IGF-1 in post-PNX lung growth strongly indicates that PNX in a number of species results in induction of IGF-1

50 expression leading to increased levels of IGF-1 in BALF and serum [7-9]. However, a few studies have published contrasting results [156, 157], lending this matter to further clarification.

The non-biased transcriptomic microarray analysis by Paxson et.al. [10] showing involvement of the IGF-1 signalling complex in PNX mice, has consolidated the argument for the significance of IGF-1 in post-PNX lung growth. However, many more questions still need to be addressed. No study has examined the effect of genetic/pharmacological modulation of IGF-

1 in regenerative lung growth post-PNX. Such studies will inform us of the reliance of regenerative lung growth on the presence of IGF-1. The cellular and molecular mechanisms by which PNX induces IGF-1 expression, its source and how it contributes to post-PNX lung growth still remains to be investigated. Furthermore, the potential for IGF-1 to modulate lung growth in regeneration-deficient aged animals has not yet been explored. Addressing these questions will help in understanding the mechanisms behind IGF-1 signalling post-PNX and its significance in lung regeneration, and hopefully reveal novel targets that may be used to enhance lung regeneration in chronic lung diseases.

51 1.13- Hypothesis and Aims of thesis

This thesis tests the hypothesis that IGF-1 enhances the regenerative capacity of the lung following left-lung PNX. Furthermore, we test the hypothesis that activation of the transcription factors EGR-1 and HIF-1α in response to PNX, contributes to IGF-1-driven post-

PNX lung growth (Figure 1.12).

More specifically, this thesis aims to: i) Characterise changes in IGF-1 expression in post-PNX lung growth in young adult mice. ii) Determine whether administration of IGF-1 or inhibition of IGF-1R signalling alters regenerative lung growth post-PNX. iii) Determine whether sustained exogenous infusion of IGF-1 can restore regenerative lung growth in aged mice. iv) Determine the effect of EGR-1 and HIF-1α inhibition on post-PNX lung growth in the presence or absence of IGF-1 supplementation.

Figure 1.12: Proposed mechanism for the involvement of IGF-1 signalling in post-PNX lung growth. PNX creates mechanical stretch in the lung that leads to activation of immediate and early transcription factors EGR-1 and HIF-1α. Activation of EGR-1 and EGR-1 induces IGF-1 and IGFBP-3 expression. The IGF-1 ligand through autocrine and paracrine signalling activates the IGF-1R by phosphorylation, triggering activation of downstream signalling pathways primarily the ERK and Akt pathways, resulting in increased cell proliferation (measured by Ki-67 expression) and eventually growth of lung involving formation of new blood vessels (measured by CD31 expression), proliferation of alveolar epithelial progenitors (AECIIs) and AECI differentiation as well as formation of new matrix elements by α-SMA+cells.

Chapter 2

Methods

54 2.1- Animal ethics statement

Animal procedures conducted in this project were approved by the University of Western

Australia (UWA) Animal Ethics Committee (Approval numbers: RA/3/100/1365 (2015-2017);

RA/3/100/1557 (2018)). Additional approval was sought from the Murdoch University Animal

Ethics Committee (Approval number N2795/15; RAMP number RAMP0456_05_16).

2.2- Left lung pneumonectomy and sham thoracotomy surgery

Animals used in this study were C57BL/6 female wild type mice (Mus musculus). Young adult mice ranged in age from 2-3 months and old mice were aged 22-24 months. Mice were purchased from the Animal Resources Centre (Perth, Australia) and housed at the UWA M- block Animal Care Facility where all surgical procedures were performed.

Mice were weighed and then anaesthetised in an induction chamber with 5% isoflurane/1L of oxygen (ISOTEC3, Advanced Anaesthetic Specialists, Sydney, Australia). Orotracheal intubation was performed with the aid of the Leica M651 dissection microscope, (Leica

Microsystems Pty Ltd., Macquarie Park, Australia), using a 22G cannula connected to a

MiniVent respirator set at 200µL volume/250 strokes/minute (Type 845; Hugo Sachs

Elektronik, March, Germany). A mixture of 2% isoflurane/ 1L of oxygen was supplied to maintain anaesthesia. The left chest wall between the fifth and sixth ribs was shaved and sterilised using a chlorhexidine/alcohol antiseptic. Using sterile scissors, an incision (1cm deep) was made in the skin at the fifth intercostal space (Figure 2.1A). The ribs were retracted gently using blunt ended forceps exposing the left lung (Figure 2.1B). The left lung was gently exteriorised making it easier to locate the hilum which was then ligated with a pre-knotted 6/0 silk (Figure 2.1C-E). The portion distal to the ligature including the left lung was excised leaving 1-2mm of tissue (Figure 2.1F, G).

( (D (B cav thethoracic expose to intercostal sixth and fifth between made is chest theon left (A) 2. Figure

C) C)

F) F)

Anaesthe

) Left lung is excised and the tho and isexcised Left lung ) Base of the lung is located and the knot is tied tied is the knot and oflocated the lung Base is

Lefti lung

: Sequence of images showing left images showing of Sequence :

t ised mice laying prone with orotracheal tube connected to venitilator to maintain a steady breathing rate under anaesthesia. anaesthesia. under rate breathing steady a maintain to venitilator to connected tube orotracheal with prone laying mice ised

pre into a forceps withended blunt externalised scarefully

racic cavity and surgi and cavity racic

- lung pneumonectomy surgery surgery pneumonectomy lung

cal site are sutured. are site cal

made 6/0 silk knot placed on the surgical site ready to pass the lung through. lung the pass ready to site theon surgical placed knot silk made 6/0

in mice. in

ity.

deep) (1cm Inscision

56 After ensuring that there was no air leakage, the chest cavity and surgical site were sutured

(Figure 2.1H). Sham operated mice underwent the same incision and wound closure procedure except that the lung was not removed. The analgesic buprenorphine (50µL at 0.1mg/kg) was subcutaneously administered immediately after surgery to minimise post-surgical stress and discomfort. Animals were housed in individual cages on a heating pad and closely monitored at 15 min intervals in the first hour and at 30 min intervals up to three hours after surgery.

Animals were then transferred to a temperature controlled HEPA-filtered cabinet where they were housed in individual cages for at least a week. Animals were then monitored twice daily for the first two days and daily thereafter until the conclusion of the experiment. Monitoring tasks included: checking animal weight, activity, gait, posture, grimace, respiration, and the surgical sight as well as administration of buprenorphine in the first two days after surgery.

Over the course of this study, mortality rate during and 24hrs post-surgery was less than 5%, and less 2% at 48hrs post-surgery.

2.3- Implantation of the ALZET® Mini-osmotic pump in vivo

For longitudinal µCT experiments described in Chapter four, IGF-1 was subcutaneously administered using the ALZET® Mini-osmotic pump (Model 2004, ALZET®, DURECT ™

Corporation, Cupertino CA). The pump was implanted subcutaneously immediately after PNX surgery via a small incision made below the scapula with the flow moderator pointing away from the incision. The stainless-steel tubing of the flow moderator was replaced with a polyetheretherketone microtubing of the same dimensions to avoid poor image resolution during CT imaging due to the presence of a metallic material. Recombinant human IGF-1

(rhIGF-1) (PeproTech, Rocky Hill, NJ) was prepared at a concentration of 0.5 mg/mL resulting in an estimated serum concentration of 500 ng/mL/day delivered for 21 days. IGF-1 containing solution or phosphate buffered saline was filled in the ALZET pump 24 hrs before surgery according to the manufacturer’s instructions and stored at 37°C.

57 2.4- Micro-Computed Tomography (µCT) X-ray imaging protocol

Lung volume in live mice was measured using the SkyScan small animal µCT scanner

(SkyScan 1176, software version 1.1, Bruker micro-CT, Kontich, Belgium). Prior to each CT scan, mice were weighed to account for differences in body size. The animal was anaesthetised with 5% isoflurane and 1L of oxygen in an induction chamber and then transferred to the imaging cradle with its snout placed in a nose cone that supplied 1-2% isoflurane in 1L of oxygen to maintain anaesthesia during imaging. A styrofoam cushion approximately 10mm2 was lightly tethered to the spine with translucent tape at the level of the diaphragm to track breathing movements captured with the help of a video camera (Princeton Instruments, Trenton

NJ) placed in front of the animal. The breathing movements captured by the camera produced a wave pattern which was displayed in a physiological monitoring window on the computer screen. This wave form was based on the change in light intensity within a box drawn at the edge of the styrofoam as the animal inhaled and exhaled. To accurately track the breathing rate, the threshold and amplitude of the signal as well as the position and size of the box delineated on the edge of the styrofoam were appropriately adjusted to obtain a consistent wave pattern with maximal peaks and troughs. Once an optimal wave pattern had been achieved, the percentage of isoflurane was adjusted accordingly (~ at 2%) to maintain a steady breathing rate ranging from 0.8-1 breaths per second during image acquisition. As soon as the breathing rate had stabilised, the instrument chamber was closed and a whole-body scan (scout scan) was conducted to locate the position of the animal on the bed. The camera, X-ray source and detector were then centred around the thorax. A flatfields correction command was performed using the software to minimise background noise; this command was only executed once in each imaging session. The scan parameters used are listed in Appendix B. This imaging protocol results in a dose of less than 1Gy per scan and has previously been shown not to result in radiotoxicity of the lungs [230, 231].

58 2.5- µCT Image analysis

2.5.1- Image Sorting

The first step in analysing lung µCT images from spontaneously breathing animals involves grouping images according to the stage of the respiratory cycle at which the image was captured. This ensures that lung volumes between animals are compared at similar stages of the respiratory cycle. This form of image partitioning is termed extrinsic retrospective respiratory gating and is distinct from a prospective respiratory gating approach which relies on the use of a mechanical ventilator to acquire images at specific phases of the breath cycle

[232]. Lung projection images were sorted into four bins (bin 0 to bin 3) corresponding to the end of expiration and start of inspiration respectively, using the Dataviewer software (version

1.5.2.4, SkyScan).

2.5.2- Reconstruction

Raw projection images acquired after µCT imaging need to be processed to reduce background noise and blurring before 3D images can be reconstructed and analysed from tomography slices. Reconstruction was performed using the NRecon software (version 1.6.10.4, SkyScan) which uses the filtered Feldkamp cone-beam algorithm to create a stack of cross-sectional slices from tomography images [233]. Reconstruction settings used are listed in Appendix B.

An example of a coronal and transverse section of a reconstructed µCT image of the lung at expiration and inspiration is depicted in Figure 2.2. During inspiration, the edges of the lung appear ‘fuzzy’ due to shadow effects caused by motions of the diaphragm and anaesthesia- induced gasping [232] (Figure 2.2). This effect is largely attenuated during the expiration phase giving a clear boundary between the lungs and external tissues and thus yielding a more accurate estimate of lung volume. For this reason, only lung volume measures at the end of expiration were analysed.

A- Inspiration B- Expiration

Coronal view Coronal

Transverse view Transverse

Figure 2.2: Coronal and transverse views of reconstructed images from PNX mice lungs during inspiration and expiration.

This figure is representative of an in vivo μCT slice taken at T8 vertebra highlighting the lung (dark pixels) in the thoracic cavity at the end of the inspiration (A) and expiration (B) in a pneumonectomised mouse seven days after surgery. Grey pixels represent solid structures within and outside the lungs whereas white pixels represent bone. Fat has a similar density to that of lungs because of its long carbon chains and low water content, therefore appears black outside the thorax (red arrow) and must be excluded during μCT analysis to yield a more accurate estimation of aerated lung volume [234]. During expiration, the edges of the lung in a μCT scan are more clearly defined (red hatched line) whereas edges of the lung at inspiration appear fuzzy (orange hatched line). Fuzzy lung edges can result from anaesthesia-induced gasping or shadowing caused by motions of the diaphragm [232]. This effect is largely attenuated in the expiratory phase of the breathing cycle giving a clear boundary between the lungs and the diaphragm. For this reason, end of expiration volume, also known as functional residual capacity, is more useful for evaluating lung volume. The implanted osmotic pump (green arrows) can be seen protruding on the dorsal-lateral aspect of the body.

60 2.5.3- Optimisation of mean lung density measurement

To estimate the mean lung density, reconstructed datasets were calibrated in Hounsfield Units

(HU) using the CTAn software (version 1.11.0, SkyScan). HU is a measure of lung density in

CT which is based on a scale where the density of water is designated zero and air minus 1000

[235]. As the lung is comprised of both tissue (which has a density close to zero HU) and air, the density of the lung is less than zero. However, the density of the lung fluctuates quite significantly at different phases of the respiratory cycle [235], therefore it is imperative to analyse images at synchronised phases of the breathing cycle.

HU calibration was performed according to the Bruker-microCT method for applying HU calibration in Skyscan CTAn [236]. In brief, a water-filled phantom tube was scanned and reconstructed using the same settings used for live in vivo imaging. A circular ROI was then drawn within the water portion of the phantom over 10 layers and a density histogram of the dataset was plotted. The mean grey index of water was equated to its theoretical HU value

(zero) whilst zero grey index (that is zero attenuation) corresponded to -1000 HU. This operation normalised the HU scale so that the minimum grey index (zero) equated to minus

1000 whilst maximum grey index value (255) corresponded to a HU number of 2437. The number ‘2437’ was entered in the density calibration window to match the 255 grey index value in every dataset analysed to calibrate lung images for density analysis.

61 2.5.4- Lung volume and density measurement using CTAn

Following reconstruction and density calibration, images were exported to the CTAn software

(version 1.11.0, SkyScan) to estimate total lung volume, aerated lung volume and mean lung density. To analyse reconstructed datasets in CTAn, first the top and bottom slices corresponding to areas occupied by lung pixels on the raw coronal CT image were defined. An elliptical region of interest (ROI) was drawn around thorax at every slice selected then images were HU calibrated as described in Section 2.5.3. Calibrated ROIs were saved and then re- opened in a new CTAn window where the algorithm that was generated to calculate lung volume and density (described in Section 2.5.5) was applied to the dataset.

2.5.5- Description of the µ-CT algorithm used to enumerate lung volume and density

The µCT analysis algorithm used in this study was adapted from the Bruker µCT method [237] with modifications to allow simultaneous analysis of total lung volume, aerated lung volume and mean lung density from HU calibrated datasets.

To analyse lung volume, images were first thresholded from 60-255 greyscale indices to segment out the body from the rest of the image (Figure 2.3A). A despeckle manoeuvre was run in 2D to remove pores inside the body as detected by image borders. Another despeckle manoeuvre that swept all but the largest object in the image was run in 2D to remove residual speckles and the imaging bed. This created an ROI encompassing the thorax (Figure 2.3B).

Using a bitwise operation, the ROI was superimposed on the original raw dataset without altering any lung pixels thus allowing for analysis of the lung within the ROI (Figure 2.3C-

D). Images were then thresholded from 0-60 grey indices to select lung pixels within the ROI.

outline outline is run estimate to mean lung density HU. in Lastly file a of the structu 3D ves blood major excluding but contents tissue including lung the of volume total ( analysisplug pixels)exe is the size the of manoeuvre on (based ( ( 2D 3D. and The outl ( ( ( Figure 2.

G E D C B

) Fat and other components of similar density outsidensitysimilarcomponents ofother andFat ) Air ) pixels are highlighted black.

) ) The imagesthen are threshol ) Black pixels within the region of interest (ROI) are identified. Res Raw ) images are thresholded to identifyair and tissueco

) To estimate total lung volume (TLV) a morphological operation is run to close black pores within the lung smaller than 10 r 10 than smaller lung the within pores black close to run is operationmorphological a (TLV) volume lung total estimate To )

: : Processesinvolved estim in

in.

ineof the lung is retained (although not in shown

ded ded againto identify air pixels within the lung

ation of lung volumedensity and CTAn in

cuted resulting in a clear outline of the lung ( lung the of outline clear a cuted resultingin

de the thorax are still present within the ROI at this stage. To remove these pixels, another despeckleanotherpixels, removethese stage.To this ROI at withinpresentthe still are thoraxthe de

mponentsboth within outside and the lung.

idual speckles outside the ROI are removed by running a despeckle manouver in both

) and ) is then superimposed the on original sections depicted as in D.

re,which must exported be toCTVol for viewing, is created (

sels. A histogram analysis of lung density distribution within the TLV TLV the within distribution density lung of analysis histogram A sels.

) from which the air volume is calculatedbyrunningvolumefrom air is the which)

adiu

s. This estimates the estimates This s.

a 3D a

63 To limit the ROI to borders of the lung, a bitwise operation that combined the image generated by 0-60 thresholding and the thorax ROI was applied so that the resulting image displayed pixels common between these two images. A despeckle manoeuvre which swept all but the largest object (the lung) was applied in 3D to remove fat and other pixels of similar density to the lung (Figure 2.3F). Another despeckle manoeuvre was applied in 3D to remove voxels less than 33 voxels within the lung before finally estimating the aerated volume of the lung (Figure

2.3F). To estimate the total lung volume, a morphology-based operation that closed all edges of the lung but removed dark pixels greater than 10 (major blood vessels) in 2D was applied to the dataset to calculate the area covered by the lung at every projection (Figure 2.3G). After running this morphological operation, a 3D analysis plug-in was run to calculate the total lung volume (TLV). To estimate the mean lung density from TLV, a bitwise operation based on binary arithmetic was applied to copy the TLV ROI onto the dataset. Images were reloaded and then re-imposed on the TLV ROI before finally running a histogram plugin which enabled calculation of lung density, based on HU calibrated values described in Section 2.5.3. All the above steps were combined into one task list algorithm that was uniformly applied to all datasets to simultaneously estimate the aerated lung volume, TLV and mean lung density from tomography images. In addition, running this task list generated a ‘cmv’ file that could be exported to the CTvol software (version 2.3.2.0, SkyScan) to generate a 3D model of the lung

(Figure 2.3H). 3D models were analysed to visually assess how well the lung volume measures output by CTAn reflected the structure of the lung and also ensuring that none of the lung lobes were omitted in the analysis or that no extra non-lung voxels were included as part of the lung.

Where the volume of the lung could not be automatically derived, the total lung volume was estimated by manually outlining the lung at every projection, and then calculating the aerated lung volume, TLV and mean lung density as described above.

64 2.6- Estimation of total lung volume by the water displacement method

An alternative method used to estimate lung volume in this study was the water displacement method [4, 238]. Following euthanasia, the mouse lung was inflated with 4% PFA as outlined in Section 2.9.1. After overnight fixation at 4°C, fat and connective tissue were carefully dissected away prior to determining lung volume. The apparatus used to measure lung volume by this method is depicted in Figure 2.4. It consisted of a copper coiled wire attached to a spatula to suspend the coil as it was immersed in a 50mL water-filled beaker. To measure lung volume, the beaker was filled with at least 40mL of water then the whole apparatus was placed on a balance and tared. The fixative-filled lung was gently rolled on a paper towel to remove excess liquid and then carefully immersed in water under the copper coil thus keeping it wholly immersed in water. The reading on the balance in mg was noted and directly converted to µL since the density of water is approximately 1g/mL at RT. It was assumed that the larger the lung volume, the greater the amount of fixative it could uptake and thus the greater amount of water it could displace. Measurements were repeated at least three times and the mean calculated.

Figure 2.4: Simple apparatus for measuring lung volume by water displacement.

A copper coiled wire (thin arrow) attached to a spatula (bold arrow) placed on top of a beaker keeps the lung immersed in water (triangle arrow) to estimate the amount of water it displaces (measured in mg) which was read as the volume of the lung in µL.

65 2.7- Quantitative Real time PCR (qRT-PCR)

2.7.1- RNA Extraction

To extract RNA from mouse lung tissue, the animal was culled by cervical dislocation, the lung carefully dissected, snap frozen in liquid nitrogen and stored at -80°C until required. Frozen lung tissue was and then ground up in the presence of liquid nitrogen using a mortar and pestle.

BL/TG RNA lysis buffer supplied in the Relia Prep™ RNA Cell MiniPrep System RNA extraction kit (Promega, Alexandria, Australia) was prepared according to the manufacturer’s instructions and 800µL of the buffer was added to 10mg of each sample, in a 1.5mL microfuge tube. The tissue was homogenised by passing through a 22G needle approximately 10 times.

The sample was centrifuged at 13,200rpm for five minutes and the supernatant transferred to a sterile 1.5mL microfuge tube. Absolute isopropanol was added to the sample (35% v/v), mixed by vortexing for five seconds and then transferred to the spin column for RNA extraction according to instructions provided in the Relia Prep™ RNA Cell MiniPrep System RNA extraction kit (Promega, Alexandria, Australia). An on-column DNAse treatment step (Qiagen,

Chadstone, Australia) was performed to eliminate excess genomic DNA. The concentration and purity of RNA was assessed by measuring UV absorbance (A260/A280) using the

NanoDrop® 2000 U.V-Vis spectrophotometer (Thermo Scientific, Wilmington, DE).

To extract RNA from lung fibroblasts, following a PBS wash, 200µL of BL/TG buffer was added to 2x105 cells in a 12 well plate and pipetted multiple times on the cell monolayer to detach and homogenise cells. The cell suspension was transferred to a 1.5mL microfuge tube to extract RNA as outlined above according to instructions provided in the Relia Prep™ RNA

Cell MiniPrep System RNA extraction kit.

66 2.7.2- Complementary DNA (cDNA) Synthesis

One microgram of RNA was reverse transcribed into cDNA using the Omniscript RT Kit

(Qiagen, Chadstone, Australia) in a 20µL reaction comprising 7μL of cDNA mastermix components listed in Appendix A (Table 7.3) and 13μL of 1μg RNA solution diluted in

RNAase free water. A no-reverse transcriptase control and no-RNA control were included to check for genomic contamination and mastermix integrity respectively. After adding the appropriate volumes of RNA and mastermix to each sample in 0.2mL PCR®-STRIP tubes

(Axygen), the tubes were tightly capped, vortexed briefly and given a quick spin before being transferred to the Eppendorf thermocycler (Quanta Biosciences, Gaithersburg MD) where the samples were incubated at 37°C for one hr to synthesise cDNA.

2.7.3- Real Time (RT) PCR

Quantitative real time-PCR was performed using the StepOne Plus™ Real-Time PCR system

(Software version 2.2.2, Applied Biosystems, Carlsbad, CA) in a 10µL real-time PCR mixture consisting of 2.5µL of cDNA (diluted 1 in 4), pre-designed Taqman probes (Applied

Biosystems Life Technologies, Australia) directed against the gene of interest (FAM-labelled,

0.5µL) and the house keeping gene, mouse phosphoglycerate kinase (PGK)-1 (VIC-labelled,

0.5µL), 1X Taqman gene expression mastermix (5µL) and 1.5µL of RNAse free water. The genes of interest probed for are listed in Appendix A (Table 7.5). All samples, including a no- cDNA control, were run in duplicates in MicroAmp® Fast Optical 96-well Reaction Plates

(Applied Biosystems by Life Technologies) covered with a PCR Compatible MicroAmpTM

Optical Adhesive film (Applied Biosystems). After sealing the plate with adhesive film, it was centrifuged at 1200 rpm for two minutes and then placed in the Applied Biosystems Step-One-

Plus Real-Time PCR machine for the RT-PCR reaction with cycling conditions as follows: one cycle at 50°C for two mins, one cycle at 95°C for 10 mins, followed by 40 cycles of

67 denaturation at 95°C for 15 secs, annealing and elongation at 60°C for one min. The CT values were measured and relative gene expression calculated using the ΔCt method where the CT value of the gene of interest (GOI) was subtracted from house-keeping gene (HKG) CT value and expressed as 2ΔCT indicative of the ratio of GOI:HKG [239] which was plotted in GraphPad

Prism 5.

2.8- Western blotting

2.8.1- Whole cell lysate protein extraction from homogenised lung tissue

To extract proteins from lung tissue for Western blot analysis, approximately 10mg of ground frozen lung tissue was weighed out in a sterile 1.5mL microfuge tube and placed on dry ice.

Pierce® RIPA lysis buffer (100µL per sample, ThermoFisher Scientific, Rockford, IL) supplemented with protease inhibitor cocktail tablets (cOmplete Tablets Mini EDTA-free,

Roche, Indianapolis, IN) and 100mM phenylmethylsulfonyl fluoride (PMSF) was added to the tissue lysate and placed on ice for 45 mins, vortexing for 10 seconds every 15 mins. The tissue lysate was then homogenised by passing through a 22G needle approximately 10 times on ice.

The sample was centrifuged at 13,200rpm for 10 mins at 4°C. The resulting supernatant was transferred to a fresh 1.5mL microfuge tube, placed on dry ice for five minutes and then stored at -20°C until required.

2.8.2- Whole cell lysate protein extraction from cultured cells

To extract proteins from lung fibroblasts, culture medium was aspirated, and cells were washed twice with ice-cold PBS. Protein lysis buffer (50-100µL) supplemented with phosphatase and protease inhibitors was added to the cells which were lysed by scraping the plate. Protein lysates were transferred to 1.5mL tubes and incubated on ice for 30 mins, vortexing for 10

68 seconds every 10 mins. The protein lysate was centrifuged at 13,200 rpm for 10 min at 4°C and the resulting supernatant transferred to a new 1.5mL microfuge tube, kept on ice or stored at -20°C prior to quantification.

2.8.3- Nuclear protein extraction

To probe for EGR-1 and HIF-1α transcription factors by Western blot, whole cell lysates were enriched for nuclear proteins using hypertonic and hypotonic buffers (recipes in Appendix A,

Table 7.1). First, cell lysates were incubated in 80-100µL of hypotonic buffer for 30 mins on ice, vortexing every 10 mins. Cell lysates were then homogenised by passing through a 22G needle ~10 times and then centrifuged at 13,200rpm for 10 mins at 4°C. The supernatant

(cytoplasmic protein fraction) was transferred to a separate tube for storage. The cell pellet was resuspended in 30-50µL of hypertonic buffer and mixed on a rotating rack for 30 mins at 4°C.

The sample was then centrifuged at 13,200rpm for 10 mins at 4°C. The resulting supernatant which was enriched for nuclear proteins was transferred to a separate tube and stored on ice ready for quantification.

2.8.4- Protein Quantification

Protein concentration was determined using the PierceTM Coomassie plusTM (Bradford) Assay

(Fisher Scientific, Scoresby, Australia). A BSA standard curve was generated by serially diluting 20μL of stock 2mg/mL Bovine Serum Albumin (BSA) (Thermo Scientific, Scoresby,

Australia) in dH2O to create BSA standards with concentrations ranging from 0.016mg/mL to

1mg/mL. Protein samples to be quantified were diluted 1 in 20 in dH2O then 10μL of each sample and each of the BSA standards were added in duplicate to a 96-well ELISA plate. Using a multichannel pipette, 200μL of PierceTM Coomassie plusTM reagent (BIO-RAD Hercules,

CA) previously equilibrated to room temperature, was added to the samples. The plate was gently tapped to mix contents of the wells and then incubated at room temperature for 15

69 minutes protected from light. The absorbance of each sample was then measured at 595nm

2 using the Wallac Victor V™ Plate Reader (Perkin Elmer Life Sciences, Waltham, MA). Using the absorbance values of BSA standards with known concentrations, a standard curve was generated in Microsoft Excel and the absorbance values of the protein samples were used to extrapolate the protein concentration of each sample from the standard curve.

2.8.5- Western blot analysis

For Western blot analysis, approximately 30µg of protein was mixed with DTT reducing agent

(Bolt™ Sample reducing agent (10X), Novex by Life Technologies, Carlsbad, CA) and sample loading buffer (Bolt™ Sample buffer (4X), Novex by Life Technologies, Carlsbad, CA) in a

20µL volume and then heated at 95°C for five minutes to denature the proteins. Samples, including 5µL of Precision Plus Protein™ dual coloured standard ladder (BIO-RAD, Hercules,

CA) were loaded into wells of the Bolt™ 4-12% Bis-Tris Plus gel and run at 200V for approximately 22 minutes in the Bolt™ Mini Gel tank (ThermoFisher Scientific, Rockford, IL) with the NuPAGE MES SDS Running buffer (Novex by Life Technologies, Carlsbad, CA).

Proteins were then transferred for seven minutes onto a nitrocellulose membrane using the iBlot® transfer system (ThermoFisher Scientific, Rockford, IL).

Alternatively, proteins were resolved using 10-12% acrylamide/bisacrylamide running gels with 5% stacking gels, cast in the laboratory (gel and buffer recipes in Appendix A, Tables

7.1 and 7.2). Gels were run in 1x SDS running buffer initially at 40mA until proteins reached the bottom of the stacking gel then at 100V for 1.5hrs. After resolving proteins, the gel was equilibrated in 1x transfer buffer for 20 minutes at 4°C. A transfer cassette was then assembled in 1x transfer buffer as follows: black side of cassette lay down followed by one pre-soaked sponge, three wet filter papers, the running gel, transfer membrane (nitrocellulose or PVDF, which needs to be pre-incubated in 100% methanol for at least five minutes prior to use), three

70 more wet filter papers and finally another sponge. The cassette was closed and fitted into the transfer tank containing 1x transfer buffer at 4°C and an ice block. The black side of the cassette faced the negative electrode (black part of the tank). Transfer was performed for 1.5 hours at room temperature.

After transfer, the membrane was blocked for one hour at room temperature in 5% non-fat skim milk diluted in tris-buffered saline/0.05% tween-20 (TBST) followed by primary antibody incubation overnight at 4°C. Primary antibodies directed against phosphorylated antigens were blocked and diluted in 5% BSA/TBST. Primary antibodies used for Western blots and their dilutions are listed in Appendix A (Table 7.6). After overnight incubation in primary antibody, the membrane was washed four times for five minutes in TBST followed by the addition of an appropriate species specific HRP-conjugated secondary antibody diluted in 5% milk/TBST for one hour at room temperature. The membrane was then washed again four times for five minutes in TBST prior to detection using the Immobilon™ Western chemiluminescent HRP

Substrate (Millipore Corporation, Billerica, MA) according to the manufacturer’s instructions.

The signal was developed on CL-Xposure™ film (ThermoFisher Scientific, Rockford, IL) using the CP1000 X-ray Processor (AGFA Healthcare, Mortsel, Belgium). Semi-quantitative densitometry analysis was performed by comparing the signal intensity of the protein of interest and that of the loading control (α-tubulin) using NIH Image J software [240].

2.8.6- Membrane Stripping

To re-probe a membrane for another protein of interest, the membrane was incubated in two changes of glycine stripping buffer (Appendix A, Table 7.1) for 10 mins at room temperature and washed three times for fives in TBST. The membrane was then blocked for 1hr in 5% milk/TBST prior to addition of primary antibody and detection as outlined in Section 2.8.5.

71 2.9- Histology

2.9.1- Preparation of lung specimens for histological analysis

Mice were anaesthetised with 5% isoflurane/oxygen and then euthanised by cervical dislocation. A median sternotomy followed by a tracheostomy with a 21G needle was performed to inflate the lungs with 4% paraformaldehyde (PFA) at 25cmH2O pressure for five minutes. A knot was tied mid-way along the trachea and the inflated lung was carefully dissected and immersed in 4% PFA for storage at 4°C overnight. Lungs were then dehydrated in 15% sucrose/PBS overnight at 4°C, processed using the Leica automatic tissue processor

(model TP 1020, Leica Biosystems, Nussloch, Germany) and then embedded in paraffin blocks using the Leica tissue embedder (model EG1150C, Leica Biosystems, Nussloch, Germany).

Tissue sections (5µm thick) were cut from paraffin blocks using a Leica microtome (model

RM2245, Leica Biosystems, Nussioch, Germany) and attached onto adhesive glass slides

(HURST Scientific PTY Ltd, Canning Vale, WA). Prior to use, slides were baked at 60°C overnight to enhance adhesion of the tissue to the slide.

2.9.2- Immunohistochemistry (IHC) staining

To stain lung tissue by IHC, tissue sections were dewaxed in two washes of xylene and one wash of xylene/ethanol (1:1) for five minutes each. Slides were then rehydrated in decreasing grades of ethanol (2x100%, 90%, 70% and 30%) and finally deionised water for five minutes in each wash, prior to antigen retrieval. Antigen retrieval was performed as follows: a coplin jar was filled with 10mM citrate buffer (pH 6) or 1mM EDTA (pH 8) antigen retrieval buffer and warmed in boiling water using a Cuisinart electric pressure cooker (model CPC-600A, East

Windsor, NJ). Slides were then added to the warm antigen retrieval buffer and boiled at high pressure for 20 minutes. Ki-67, SpC and pIGF-1R antigens were retrieved in EDTA buffer

72 whereas all other antigens probed were retrieved in citrate buffer. Following antigen retrieval, the slides were cooled by placing the coplin jars in excess running water and then allowed to equilibrate to room temperature. Slides were washed twice for five minutes in TBS before a hydrophobic ring was drawn around the tissue using a wax pen (Dako Cytomation, Australia) to concentrate solutions around the tissue on the slide. Endogenous peroxidase activity was blocked by incubation in 3% H2O2 for five minutes followed by two five-minute TBS washes.

Prior to overnight primary incubation at 4°C, 10% turbo serum/TBS was applied to slides for

30 minutes to block non-specific antibody binding. The primary antibodies used are listed in

Appendix A (Table 7.6). After overnight primary antibody incubation, slides were washed three times for five minutes in TBS/0.05% tween-20 (TBST), the appropriate species specific biotinylated secondary antibody (diluted 1 in 200 TBS/0.1% triton-X-100) was applied to slides for 45 minutes at room temperature. Slides were washed in TBST three times for five minutes each, incubated with Streptavidin-HRP (DakoCytomation, Australia) diluted 1 in 200

TBS/0.1% triton-X-100 for 30 minutes and washed again three times for five minutes, before the positive signal was developed with 3,3’-Diaminobenzidine solution (DAB; Sigma-Aldrich,

Melbourne Australia). For Ki67/SpC dual staining, the antigen to be detected with DAB was developed first then slides were washed three times for five minutes in TBST before the secondary antibody for the second antigen was added and its signal developed with the Liquid

Fast-Red Substrate kit (Abcam, Melbourne, Australia) or the Stay Green/AP Plus kit (Abcam,

Melbourne, Australia) according to the manufacturer’s instructions. Slides were then washed four times for five minutes, counterstained with Gill’s hematoxylin (Grade 2, Polysciences Inc.,

Warrington, PA) dehydrated in increasing concentrations of ethanol (30%, 70%, 90% and 2x

100%) followed by two xylene washes for five minutes each and finally coverslipped with

DePex polymer mounting media (Merck, USA) ready for analysis.

73 2.9.3- Haematoxylin and Eosin (H&E) staining

Tissue sections were deparaffinised and rehydrated as outlined in Section 2.9.2 and then stained in Gill’s haematoxylin (Grade 2, Polysciences Inc., Warrington, PA) for one minute.

Slides were washed in running tap water for two minutes followed by 30 second washes in

Scott’s tap water, deionised water and 95% ethanol respectively. Slides were then immersed in eosin Y solution for two minutes and then rinsed in excess deionised water for one minute.

Finally, slides were dehydrated, mounted and coverslipped as outlined in Section 2.9.2.

2.9.4- Image analysis

IHC images were captured at x200 or x400 magnification using the Nikon Eclipse Ni-E motorised brightfield/fluorescent microscope (Nikon Instruments Inc., Japan) fitted with a

Nikon Ds-fi2 camera and operated with Nis Elements AR software (v 4.20.00, 64 bit, Nikon

Instruments Inc., Melville, NY). For each animal, 20 non-overlapping images (5/lobe) were randomly sampled twice from lung sections at least 50µm apart. A mask was used to orient the objective to a random field. Major blood vessels, airways and alveolar ducts were avoided. For

α-SMA analysis, images were captured at x600 magnification. Positive staining was evaluated using a custom written Image J macro (Appendix C) that deconvolved images to automatically differentiate and count DAB stained cells from unstained cells. Positive cells were expressed as a percentage of the number of nuclei or as the number of positive cells per field of view, where manual counts were done. DAB and Stay-red dual stained sections were analysed using a separate custom written Image J macro (Appendix D, E). Briefly, the macro first ran a colour deconvolution algorithm that identified and differentiated pixels based on whether they were predominately blue, red or brown. DAB positive cells were then pseudo-coloured green, Stay- red cells were coloured red and nuclei were coloured blue. Images were then merged and

74 thresholded to identify yellow pixels which represented areas of overlap between DAB (green) and Stay-red (red) pixels, that is dual labelled cells. All images were evaluated in a blind fashion to prevent bias in the analysis. Additionally, to evaluate the quality of the staining, isotype controls and no primary antibody stained tissue sections were included in every experiment (Appendix F).

Alternatively, slides were analysed by first scanning areas occupied by tissue on the slide at x200 or x400 magnification with the aid of the Aperio AT2 slide scanner (Leica Biosystems,

Vista, CA). This resulted in creation of a digital map of the entire tissue section that was then analysed with the Aperio ImageScope software (version 11.2.0.780, Lecia Biosystems). Within the program various pre-programed and user-created algorithms such positive pixel count and colour deconvolution algorithms can be used to identify positively stained cells, measure and annotate stained areas. To analyse IHC and H&E stains, boundaries of the lung were defined using the positive pen tool provided in the software. Using a negative pen tool, large airways and blood vessels were negatively selected. Using either the colour deconvolution or positive pixel count algorithm, positively stained cells were identified within the delineated region and expressed as a percentage of the total region of interest. Alternatively, a camera tool provided in the software was used to capture images at 200 or x400 magnification and the digital files exported to Image J for further analysis.

75 2.10- Cell maintenance

2.10.1- Propagation of cell line

The NIH 3T3 embryonic fibroblast cell line (ATCC® CRL-1658™) was used for all in vitro experiments. Cells were grown in T75cm2 tissue culture flasks (Falcon, USA) in 10mL of

Dulbecco’s Modified Eagles Medium (DMEM; Invitrogen Life Technologies, USA) supplemented with 10% (v/v) foetal calf serum (FCS; Invitrogen Life Technologies, USA),

2mM L-glutamine (Invitrogen Life Technologies, USA), 100 units/mL penicillin and

100µg/mL streptomycin (Invitrogen Life Technologies, USA). Cells were maintained in a tissue culture incubator at 37°C and 5% CO2 (v/v) of air volume. After reaching 90% confluency, cells were subcultured by first washing with PBS twice followed by addition of

0.05% Trypsin-EDTA-PBS solution (2mL) and incubation at 37°C/5% CO2 for five mins. An equivalent amount of growth medium was added to neutralise trypsin. Cells were then transferred to a 15mL falcon tube and centrifuged at 1200rpm for five mins. The supernatant was discarded, and the cell pellet was resuspended in 1mL of growth medium by repeated pipetting. A quarter of the cell suspension was transferred to a new T75cm2 flask with 10mL of fresh growth medium.

2.10.2- Cell counting

Cell counts were performed using a Neubauer Improved Bright-line Haemocytometer

(Hirschmann® Eberstadt, Germany). To estimate the concentration of viable cells in a cell suspension, 10μL of the cell suspension was diluted in an equivalent volume of 0.4% trypan blue and observed under the Nikon Eclipse Ti-S inverted microscope (Nikon, Tokyo Japan).

The cell concentration (in cells/mL) was obtained by multiplying the average number of cells counted in four quadrants on the chamber by 10,000 and the dilution factor.

76 2.10.3- Freezing and thawing cells

Cells were frozen at a density of 1x106 cells/mL in medium containing 50% FCS, 40% DMEM and 10% DMSO. One millilitre of the cell suspension was transferred to each Nalgene®

Cryoware Cryogenic vial (Thermo Scientific, Rockford, IL) and placed in a Nalgene cryo1oC freezing container (Thermo Scientific) containing isopropyl alcohol to avoid rapid freezing of cells. Cells were kept at -80°C for 24hrs prior to long term storage in liquid nitrogen.

To thaw cells, the cryogenic vial was warmed in a water bath at 37°C. One millilitre of growth medium was added dropwise to the vial and then transferred to a 15mL falcon tube where a further 3mL of growth medium was gently pipetted under sterile conditions. The cell suspension was spun at 1200rpm for five minutes after which medium was aspirated and the cells resuspended in 1mL of growth medium. The cell suspension was transferred to a T25cm2 flask with 4mL of growth medium and gently swirled to ensure a homogenous distribution of cells. The cells were then maintained in the 5% CO2/37°C incubator. After reaching confluency, cells were expanded into a T75cm2 flask and maintained as outlined in Section

2.10.1.

2.10.4- Mycoplasma testing

Contamination of cell cultures with mycoplasma can distort cell physiology, compromising the quality of experimental results [241]. Therefore, cells were regularly tested for the presence of mycoplasma species by PCR. To prepare cells for mycoplasma testing, 1x106 cells were aliquoted into 1.5mL microfuge tubes, pelleted by centrifugation at 1200rpm for five minutes and then washed three times in 0.9% NaCl solution. The cell suspension was centrifuged again, and the resulting cell pellet was sent to the mycoplasma testing facility at the Harry Perkins

Institute of Medical Research.

77 2.11- Methylene blue (MB) cell proliferation assay

Cells were seeded at 3x103cells/100µL/well in a 96-well plate, assayed in six replicates per condition. The outside wells of the plate were filled with 100µL of phosphate buffered saline

(PBS) to prevent cells from drying. A separate T0 plate was set up to account for cell numbers at the start of the experiment. Cells were incubated at 37°C/5% CO2 overnight to allow them to adhere to the plate. The following day, the medium was aspirated from the wells, cells were washed with PBS and then replaced with 0.4% FCS/DMEM serum starve medium. A control column (n=6) containing 10% FCS/DMEM was included to measure cell doubling under standard conditions. On the third day, cells were stimulated with appropriate growth factor/drug combination in a volume of 5µL. After 48 hrs, the medium was aspirated from the wells, cells were then washed with PBS once prior to fixation with 4% PFA (100µL/well) for

10 minutes on ice. Following fixation, cells were washed three times in PBS and then stored in

PBS at 4°C until ready for analysis. Prior to spectrophotometric analysis, the plate was equilibrated to room temperature, PBS was aspirated and replaced by 100μl of 2% w/v methylene blue-borate buffer solution incubated for 30 mins. Excess dye was washed from the wells by washing in four changes of 0.01M borate buffer, blotting the plate on paper towels between each wash. The dye was then eluted with 1:1 v/v, ethanol:0.1M HCl solution and the plate read at 650nm using the Wallac Victor2V™ spectrophotometer (Perkin Elmer Life

Sciences, Waltham, MA).

78 2.12- Applying equibiaxial stretch to cultured cells

To investigate the effect of stretch on changes in gene expression of molecules related to the

IGF-1 signalling pathway, a stretch device was constructed, based on a design published by

Rana and co-workers [242]. The apparatus used is a simple and cost-effective means of applying homogenous static stretch to cells cultured on a stretchable elastic membrane and similar designs have been used to investigate stretch of lung cells in vitro [88, 243].

2.12.1- Construction of the equibiaxial stretch device

Components of the stretch device included: two acrylic plates, polyvinyl chloride (PVC) discs

(1-6), four screws, four butterfly nuts and a Bioflex 6-well plate pre-coated with Collagen I

(Flexcell® Int. Corp., Burlington, NC) to culture cells on. Specifications of each of the device components is listed in Appendix A (Table 2.7). All components of the device were sprayed with 70% ethanol prior to use and assembled on a flat surface.

To construct the device, four holes (one in each corner) were drilled into the two acrylic plates that form the top and base of the apparatus (Figure 2.5A). Screws were then inserted in all four holes of the bottom acrylic plate and secured by hexagonal nuts (Figure 2.5A). PVC discs were then placed on the bottom acrylic plate positioned centrally underneath wells that were intended to be stretched on the Bioflex 6-well plate (Figure 2.5B). The Bioflex plate was then carefully placed on top of the discs, ensuring that the discs remained at the centre of the well (Figure

2.5B). The upper acrylic plate was then placed on top of the Bioflex plate (Figure 2.5C). The butterfly nuts were then inserted in the screws till they just met the upper plate (Figure 2.5D).

Before applying any rotation, the starting position of the butterfly nuts was marked on the shaft of the screws and the position of the wings was noted on the plate (Figure 2.5D-F). To apply strain to the flexible membrane, the nut was rotated for 360º three times to achieve ~10-12% strain according to measurements determined microscopically by Rana and colleagues [242].

This number of rotations in our experience often corresponded to the point at which the Bioflex

F G

Figure 2.5: Construction of the equibiaxial stretch device.

(A) Four holes are drilled at each corner of the plexiglass through which screws are inserted and secured by hexagonal nuts. PVC discs are placed on the plexiglass plate at positions directly below the Bioflex plate wells. (B) Next, the Bioflex plate is placed on top of the discs followed by another plexiglass plate (C). Butterfly nuts are inserted and rotated for 360° three times (D) to push the upper plexiglass and Bioflex plate against the PVC disc thus stretching the silicone membrane and attached cells (E). Images A-E were adapted from Rana and co- workers’ publication [242], the authors that originally reported this design. (F) is an original image of the fully assembled stretch device and (G) shows its individual components.

80 plate was pushed just far enough to touch the bottom acrylic plate. Rana and colleagues previously demonstrated that the number of screw rotations was directly proportional to the percentage of strain and that a maximum of 21% strain could be achieved with five rotations

[242]. Only three rotations (~10% strain) were applied to cells in this study to minimise cell detachment.

2.12.2- Extracting RNA and protein from stretched cells

Prior to addition of cells in wells of the Bioflex plate, wells were pre-warmed in growth medium (10% FCS/DMEM) at 37º C for at least five minutes. The medium was then aspirated, and the cell suspension added. For RNA extraction, cells were seeded at a density of 1.5 x105 cells/mL and a density of 2.5x105 was used for protein extraction. Cells were grown for 24 hrs to allow them to attach to the silicone membrane. The next day growth medium was replaced by 0.4% FCS/DMEM starve medium following one PBS wash. After overnight serum starvation, treatments were added and finally equibiaxial strain was applied as outlined in

Section 2.12.1. Cells were stretched for 1-24 hrs before harvesting RNA and protein as outlined in Sections 2.7 and 2.8 respectively.

2.12.3- Re-coating Bioflex plates with collagen for re-use

Bioflex plates were sterilised and re-coated with collagen and re-used up to three times following guidelines published by Feng et.al [244] and Wang et.al [245]. Plates were first washed for five minutes in 0.5% w/v Alconox® detergent (Alconox Inc. NY) then washed in excess sterile deionised water. Next, plates were immersed in 70% ethanol for five minutes and then dried under U.V light for 30 mins. Collagen IV (Sigma Aldrich, MO) coating solution diluted to a working concentration of 5µg/mL in 0.05N hydrochloric acid, was added to each well (2mL/well). The lid was replaced, and the plate wrapped in cling wrap for storage at 4°C

81 overnight on a flat surface to ensure homogenous distribution of the coating reagent. The next day, the collagen solution was aspirated, the wells washed three times with PBS and then dried under U.V. light for at least 30 mins. Finally, growth medium was added to each well and the plate stored in the CO2 incubator at 37°C to acclimatise the well surface for cell attachment.

2.13- Statistics

Statistical analysis was performed using the GraphPad Prism 5 software (GraphPad Software

Inc., La Jolla CA). Numerical data in figures is graphed as mean ± standard error of mean

(SEM). Means of two groups were compared using the unpaired Student’s t test. Means of more than two groups were assessed by One-Way Analysis of Variance (ANOVA) when comparing one variable or Two-Way ANOVA when assessing two variables, with

Bonferroni’s multiple comparison post-hoc test [246]. A p- value less than 0.05 was considered significant.

Chapter 3

IGF-1 signalling enhances regenerative lung growth after left lung pneumonectomy in young adult mice

83 3.1- Introduction

Unilateral pneumonectomy (PNX), surgical removal of the whole left or right lung, is a model of post-natal lung growth that has been used to investigate mechanisms of lung regeneration and is used clinically for the treatment of lung cancers, fibrosis, bronchiectasis and other chronic degenerative lung diseases [46, 247]. In young adult mice (aged 2-3 months), left lung

PNX induces a rapid growth response that results in restoration of lung volume and function

[3, 52]. Although the mechanisms that regulate this process are complex and not well understood, it is accepted that increased mechanical force created by lung resection, initiates a number of growth-promoting intracellular signalling cascades that lead to accumulation of growth factors and hormones that play a vital role in post-PNX lung growth by translating physical cues to pro-proliferative signals [42]. Indeed, a number of growth factors including

IGF-1, KGF, VEGF and others have been implicated in post-PNX lung growth [7, 149, 151,

153, 155, 158] (Chapter one, Section 1.9).

The role of IGF-1 in regenerative lung growth has not yet been thoroughly addressed despite its vital role in lung development and repair [6, 248]. Data reported to date on the role of IGF-

1 in the PNX model remains contradictory, with a range of studies reporting that IGF-1 levels are elevated in serum and bronchoalveolar lavage of pneumonectomised animals (reviewed in

Chapter one, Section 1.12) [7, 9, 157], whilst others report no effect [156]. More recently,

Paxson and colleagues detected increased expression of IGF-1 using transcriptomic microarray analysis, supporting the argument that IGF-1 plays a role in regulating post-PNX lung growth

[10]. To date, no study has reported the effect of modulating IGF-1 levels or blocking its signalling in post-PNX lung growth.

84 Therefore, the aims of this chapter were:

i) To characterise the mouse PNX model, examining changes in lung volume and

cellularity post-PNX.

ii) To characterise IGF-1 signalling in response to PNX.

iii) To determine whether administration of rhIGF-1 enhances lung growth post-PNX

in mice.

iv) To examine the effect of pharmacological inhibition of IGF-1R on post-PNX lung

growth.

A schematic of the experiments done to address aims in this chapter is presented in Figure 3.1.

Figure 3.1: Schematic of experiments done in chapter three. (A) Characterisation of the post-PNX response. Following PNX or Sham surgery, mice were sacrificed at indicated timepoints and their lungs embedded in paraffin for histology and Immunohistochemical (IHC) analysis of IGF-1 pathway related molecules (IGF-1, IGF-1R, IGFBP-3, pIGF-1R, pERK), α-SMA, CD31 or H&E to assess lung morphology. In a different cohort of mice, lungs were weighed, homogenised and then RNA and protein were prepared as described in the methods section, to analyse IGF-1 pathway transcripts (tIgf-1, Igf-1Eb, Igf- 1r, Igfbp-3) and protein (IGF-1, IGF-1R, pIGF-1R, IGFBP-3) respectively. A different cohort of mice was used to analyse longitudinal changes in lung volume by µ-CT. (B) Supplementation or Inhibition of IGF-1 signalling. Groups of pneumonectomised (PNX) mice received daily treatment of IGF-1R antagonist (BMS-536294) or rhIGF-1 (administered i.p), beginning a day prior to PNX surgery. In the IGF-1R antagonist treated cohort and their respective controls, lung volume was assessed longitudinally until day 14. Mice were then sacrificed, and their wet weights measured. In the cohort of mice treated with rhIGF-1 and their corresponding controls, lung volume was only assessed at day seven by µ-CT before lungs were embedded in paraffin for IHC analysis of Ki-67 proliferating cells.

86 3.2- Results 3.2.1- Lung volume and tissue content are restored to near pre-operative levels after PNX in young adult mice

This study aimed to characterise changes that occur in the remaining right lung following left- lung PNX in young adult mice (aged 2-3 months). To track growth of the lung overtime in individual mice, state-of-the art µCT imaging was employed. This approach enabled visualisation of the topography of the lung in vivo as well as quantification of lung volume and tissue content in live mice. The in vivo 3D model recreated from left thoracotomised (sham) operated lungs (shown in blue in Figure 3.2A) revealed an intact multilobed right lung and a unilobed left lung similar to an unoperated control lung. The left lung of post-PNX mice (red) was noticeably absent (see arrow in Figure 3.2A). Following excision of the left lung, the right lung progressively expanded, approaching the total lung volume of two lungs in sham operated controls by day 14 (Figure 3.2B). Owing to its anatomic location, it is predominantly the cardiac lobe (encircled in Figure 3.2A) that grows into the thoracic space created by removal of the left lung [57]. Longitudinal analysis of lung tissue volume measured in vivo by µCT revealed a gradual increase in tissue volume in post-PNX lungs, which started off approximately 40% lower than that of sham operated lungs at day two (33.07 ±2.91mm3 for

PNX vs 57.30 ±2.36mm3 for shams, p<0.001, n=3/group) and was equivalent to the tissue volume of sham operated lungs by day 14 (Figure 3.2C). The right lung of post-PNX mice not only increased in size but also in mass. By day seven, the mass of post-PNX lung was nearly twice the mass of the right lung of corresponding sham operated controls (178.25 ±28.75mm3 for PNX vs 109 ±21mm3 for shams, p<0.01, n=4/group, Figure 3.2D).

A B 600 SHAM

PNX

) 3 R.lung (SHAM) * ** 400

$ $$ 200

Total lung volume (mm volume lung Total 0 2 4 7 14 Days after surgery

C 80

) 3 ** 60 *** **

20 SHAM

PNX Tissue volume (mm volume Tissue 0 2 4 7 14 Days after surgery D

250 **

Control ** )

g Sham

200

m (

PNX

h g

i 150

n 100

h g

i 50 R

0 0 1 3 7 15 Days after surgery

Figure 3.2: Restoration of lung volume and tissue content after PNX young mice.

(A) Reconstructed 3D µCT models of PNX (red) and sham operated (blue) lungs on days 2, 4, 7 and 14 after surgery. The left lung is absent in PNX lungs (black arrow). Much of the growth in PNX lungs occurs in the cardiac lobe (circled). (B) Quantitative analysis of the total lung volume measured by µCT, comparing the right lung of PNX mice (red line, n=3) and both lungs of sham operated mice (black solid line, n=3) as well as the right lung only of sham operated mice (black dotted line) over a period of 14 days. **p<0.01 (both lungs of sham vs PNX), $p<0.05, $$p<0.05 (PNX vs right lung of sham). (C) Quantitative analysis of in vivo tissue volume by µCT in left-PNX and sham lungs on days 2, 4, 7 and 14 post-surgery. **p<0.01, ***p<0.001 (PNX vs sham at indicated timepoints). (D) Changes in the ratio of the right lung weight following left-lung PNX (n=4-8) compared to sham operated (n=4-5) and unoperated controls (n=3) on day 1, 3, 7 and 15 after surgery. **p<0.01 (PNX vs Sham at a specific timepoint).

88 3.2.2- PNX does not result in pathological features such as inflammation or fibrosis

To further understand changes occurring in the mouse lung following PNX surgery, tissue sections from post-PNX and sham operated control lungs were stained with haematoxylin and eosin (H&E), to observe changes in lung tissue architecture and morphology. At a gross histological level, the morphology of PNX lungs appears similar to sham operated controls comprising of airways that progressively branch giving rise to patent airspaces lined by alveolar epithelial cells (Figure 3.3). However, on closer observation, histological features that distinguish post-PNX lungs from sham operated controls were identified. In the first three days after PNX surgery, areas of expanded airspaces were observed in the alveolar regions (Figure

3.3A-B). Between day three and day seven post-PNX, the lung histology changed from emphysematous to a more consolidated phenotype, with an increase in cell density. This increase in cell density was due in part to increased cell proliferation as well as a mild inflammatory response (Figure 3.3B-C). By day 15, these changes were largely resolved such that the architecture of post-PNX lungs largely resembled that of sham operated and unoperated control lungs (Figure 3.3D-G). However, there were patches of consolidated areas in the subpleural edges still evident as late as day 21 post-PNX (Figure 3.3E). These regions are presumably still undergoing growth and remodelling as the lung expands in size.

day 15 the lung has resumed its its resumed has lung the 15 day =100µm. bar timepoint.Scale per n=4 of representative Images timepoints. respective imagesthe of power subpleural regions of the lung (encircled area in PNX D2 inflam mild a by mediated part in density cell parenchymal the in increase an is there PNX, after seven and three day Between parenchyma. Patchy areas of expanded airspaces (d cell club by lined airways Patent unperturbed an much like appears level histological gross a on pneumonectomy lung left after lung right morphologythe of The 3.3: Figure

H&E

t in occurring features histological prominent highlighting images

s (red dotted outlines) can be seen giving rise to alveolar ducts (red arrows) which eventually terminate in alveoli (airspac alveoli in terminate eventually which arrows) (red ducts alveolar to rise giving seen be can outlines) dotted (red s normal architecture, however random patches of consolidated areas, presumably areas of remodelling or regeneration can be see be can regeneration or remodelling of areas presumably areas, consolidated of patches random however architecture, normal

otted otted blue circles in PNXD D1 image) are evident in the early phas

1 image). Black arrows point towards eosinophilic areas of platele

he right lung after left lung pneumonectomy in mice in pneumonectomy left lung after lung right he

es es of the healing response, perhaps induced by stretch.

ts in major blood vessels. Insert images are higher

matory re matory lung at mos lungat

sponse and cell proliferation. By By proliferation. cell and sponse

process. healing the of stages t

es) in the lung lung the in es)

n in the the in n

90 3.2.3- Post-PNX lung growth is accompanied by increased cell proliferation

The ability of cells to proliferate in response to PNX is a crucial determinant of a successful regenerative response. Therefore, the dynamics of cell proliferation was examined in PNX lungs and compared to lungs of sham operated and unoperated controls on day 1, 3, 7, 15 and

21 post-surgery. Lung tissue sections were simultaneously stained for Ki-67 (a marker for proliferating cells) and surfactant protein (SpC), a marker for type-two pneumocytes which are the main progenitors of type one alveolar cells, the cell type that makes up the majority of the alveolar surface epithelium [116]. There was little evidence of cell proliferation in sham operated and unoperated control lungs throughout the 21-day time course, with the percentage of Ki-67+ cells not exceeding 1.2 ±0.32% of the total cell population (Figures 3.4 and 3.5A).

PNX surgery increased cell proliferation beginning on day three post-PNX and peaking on day seven (1.2 ±0.5% for sham controls vs 5.3 ±1.5% for PNX, p<0.05, n=4/group, Figure 3.5A).

By day 15, cell proliferation had significantly declined although it appeared to remain slightly elevated in PNX lungs compared to sham operated controls (Figure 3.5A). The kinetics of Ki-

67/SpC+ dual labelled cells was similar to that of Ki-67+ cells, peaking on day seven post-

PNX (0.9 ±0.4% for sham controls vs 7.5 ±1.4% for PNX, p<0.01, n=4/group) and returning to baseline by day 15 (Figure 3.5B). In addition to SpC+ alveolar type-two cells, many different cell types were positive for Ki-67 including alveolar type-one cells, alveolar macrophages, stromal cells and endothelial cells (identified by their likely regional location and morphology in the lung parenchyma rather than dual staining, Figure 3.4).

Figure3 solid arrow solid PNX in D3) and endothelial cells (hatchedblack arrows PNX in D7). alveolar t (incellsD7), alveolar one type PNX arrowtypescell proliferatemanyin (thin cells including PNX,club airwaydifferent t type highlight (brown) sham thoracotomy (n=3 show figure This

ype two cells (black solid arrows and inset in PNX D3), alveolar macrophages (blue solid arrow in PNX D3), stromal cells (yel

.4 : :

Ki of Identification s representative s Ki

4 4 per group) and wo pneumocytes. Tissue sections were counterstained with haematoxylin (blue) to highlight nuclei. Following nuclei. highlight to (blue) haematoxylin with counterstained were sections Tissue pneumocytes. wo

67/SpC+ cells across 67/SpC+ the post 67/SpC

noperated

stained lung tissue taken from mice(8 takenfrom lungtissue stained

(Unp.)

control lungs (n=4). Ki

- PNXtime course.

Scalebar -

67+ 67+ cells (red) highlight proliferating cells. SpC+ cell 12 weeks old) on days 1, 3, 7, 15 and 21 post 21 and 15 7, 3, 1, days on weeksold) 12

30 = μ

set in PNXD1), in set

orPNX low

A 10 PNX SHAM 8 * Unoperated Control

Ki-67+ cells in the lung (%) the lung in cells Ki-67+ 0 0 1 3 7 15 21

Days after surgery

B 10 **

%Ki-67/SpC+ (SpC total) 0 0 1 3 7 15 21

Days after surgery

Figure 3.5: Increased cell proliferation in post-PNX lung. (A) Immunohistochemical quantification of Ki-67+ cells and (B) quantification of Ki-67/SpC dual labelled cells derived from mice lung sections harvested on days 1, 3, 7, 15 and 21 post- PNX or sham thoracotomy (n≥3 per group) and unoperated controls (n=4). In sham operated and unoperated control lungs, there was little evidence of cell proliferation. In post-PNX lung tissue, total cell proliferation and proliferation of alveolar type two progenitors increased with time and was maximal on day seven, returning to baseline levels by day 15 (B, C). Statistical analysis was performed by Two-way ANOVA followed by Bonferroni's multiple comparison test. * p< 0.05, ** p< 0.01.

93 The lung’s extracellular matrix (ECM), is primarily composed of collagen and elastin fibres, and provides a three-dimensional scaffold, imparting biochemical and mechanical cues on resident cells that govern cell shape, signalling and function [249]. Given the significance of a supporting matrix in neoalveolarisation after PNX [40], the percentage of α-SMA+ myofibroblasts, the principle cell type responsible for ECM deposition [250], was evaluated.

α-SMA myofibroblasts were regionally defined as cells positive for α-SMA in the lung parenchyma excluding α-SMA expressed by parabronchial and perivascular smooth muscle cells [251] (Figure 3.6A, B). In response to PNX, α-SMA+ cells in the lung parenchyma were transiently increased on day seven post-PNX (1.2 ±0.2% for shams vs 3.6 ±0.6% for PNX, p<0.05, n=3/4 respectively) with no significant difference observed between post-PNX and sham operated controls at all other sampled timepoints (Figure 3.6B).

Equally indispensable for neoalveolarisation is the growth of a network of blood vessels that closely intertwines with the alveolar surface for efficient gas exchange. Lung sections were therefore immunostained for CD31, a marker for endothelial cells, to highlight blood vessels in the lung parenchyma. CD31 was expressed throughout the lung parenchyma in both post-

PNX and sham operated control lung tissue (Figure 3.7A). Western blot analysis revealed no significant difference in the level of CD31 protein expressed in post-PNX lungs compared to sham operated control lungs (Figure 3.7B-C).

E 5 Unp. Control

SHAM * 4 PNX

-SMA+ -SMA+ (%) 3  2

Parenchymal 0 0 1 3 7 15 21 Days after surgery Figure 3.6: Increase in parenchymal α-SMA positive cells in PNX lungs.

(A-D) Lungs sections from post-PNX (n=3-4/group), sham operated (n=3-4/timepoint) and unoperated control mice (n=4) on days 1, 3, 7, 15 and 21 were stained for α-SMA. α-SMA expression was prominent in parabronchial smooth muscle cells (arrow in A) and vascular smooth muscle cells (arrow in B). In the lung parenchyma, α-SMA staining was rare but could be detected at the tip of alveolar rings (arrow in C) and base of alveolar septa (arrow in D). (E) Quantitation of α-SMA expression levels in the lung parenchyma demonstrated a significant increase in a-SMA expression on day seven post-PNX. Scale bar in A, B = 50μm, in C and D scale bar = 10μm, *p<0.05.

A PNX D7 SHAM D7

C 4 Unp. Control SHAM 3 PNX

-tubulin ratio -tubulin  1

CD31/ 0 Day 0 Day 1 Day 3 Day 7 Day 15

Figure 3.7: CD31 expression was not altered in response to PNX.

(A) Representative CD31 staining pattern of pneumonectomised and sham operated mice lung tissue sections on day seven post-surgery (n=3-4/timepoint). CD31 was expressed on endothelial cells of major blood vessels (red hatched line) and alveolar capillaries (black arrows). (B-C) CD31 protein expression was analysed by western blot and quantified by densitometry. CD31 expression levels were not significantly different post- PNX compared to sham-operated and unoperated controls (n=3/group). Scale bar=50µm.

96 3.2.4- IGF-1 expression is up-regulated in response to PNX

Having gained an understanding of the morphological and cellular changes induced by PNX, the role of IGF-1 signalling in regenerative lung growth post-PNX was examined. Using immunohistochemistry (IHC), the spatial-temporal dynamics of IGF-1 expression in PNX and sham operated control lung tissue was examined over a 21-day time course. One day after surgery, IGF-1+ cells could be detected in both post-PNX and sham operated lungs although post-PNX lungs exhibited relatively higher IGF-1 expression levels (Figure 3.8A). After day one post-surgery, IGF-1+ cells were rarely detected in sham operated control lungs (Figure

3.8A). In contrast, IGF-1 expression in post-PNX lungs continued to increase, peaking on day three post-PNX (9.4 ±0.2% for PNX vs 0.35 ±0.1% for sham operated controls, p<0.001, n=3/group) and then returning to baseline levels by day seven post-PNX (Figure 3.8A, B). The pattern of IGF-1 expression on lung tissue sections was not uniform with a preponderance towards the subpleural edges and the distal airway epithelium (Figure 3.8A). Based on the morphology and relative location of cell types in the lung, it was observed that in the alveolar regions, IGF-1 was predominantly expressed by alveolar macrophages and alveolar type two cells (Figure 3.8). Alveolar type-one cells, interstitial mesenchymal cells and vascular endothelial cells rarely expressed IGF-1 (Figure 3.9). However, smooth muscle cells adjacent to alveolar and endothelial cells were positive for IGF-1, which may suggest that these cells act as a local source of IGF-1 for adjacent cells (Figure 3.9).

A SHAM PNX

IGF

1 stained 1

Figure legend on next page

15 Unp. Control SHAM Operated 12 PNX *** 9

(%)IGF-1+ 3

0 0 1 3 7 15 21

Days after surgery

Figure 3.8: Temporal changes in IGF-1 protein expression in post-PNX lung tissue. (A) Representative images showing temporal changes in IGF-1 expression in post-PNX lung tissue and sham operated control lungs on days 1, 3, 7, 15 and 21 post-surgery. Inset images highlight specific cell associated staining. Scale bar=50µm. (B) Immunohistochemical quantification of IGF-1 expression shows that IGF-1 expression is upregulated early on day 1 and day 3 post-PNX and then returns to baseline by day 15. Images are representative of n=3-4/treatment group. Statistical comparison between groups was performed using Two-way ANOVA, ***p<0.001 above sham and unoperated controls.

day expressIGF end vascular ( subpleura the lining cells Mesenchymal alveolar regions, c (A Figure3.

three post three

shows IGF shows

: Immunohistochemical : cell expressing types of identification IGF

othelial cell othelial

1. 1. However adjacent paravascular cellssmooth (black solid arrows in

ells that pre

1 expression in airway in as well club expression 1 as cells underlying parabronchial cells muscle smooth (arrow in

. Scale bar Scale .

s ( s

shown in in shown

dominantly express IGF

for all images for

, ,

and

in in

=20µm

) also )

, ,

just above hatched line) and alveolar type I epithelial cells (encircled in in (encircled cells epithelial I type alveolar and line) hatched above just

(

shows

shown only shown in

1 1 include alveolar macrophages (

prominent expression of IGF of expression prominent

. .

1 in in 1

) ) do expressIGF

post

) ) and alveolar type II epithelial cells (

1 (double edged arrow in in arrow edged (double 1

PNXtissue lung

1. 1. All images were sampled

). In contrast, In ). B

). ( ).

C) C)

) rarely )

In the

, ,

I ).

100 To further examine the role of IGF-1 in PNX-induced regenerative lung growth, IGF-1 mRNA and protein levels were quantified by qRT-PCR and Western blot respectively. Differential splicing of the Igf-1 precursor transcript yields two main isoforms in mice, Igf-1Ea and Igf-

1Eb, with the latter shown to have a higher affinity for binding lung ECM [252] (reviewed in

Chapter One, Section 1.9.6). To this end, total Igf-1 and Igf-1Eb transcripts were measured.

Total Igf-1 and Igf-1Eb mRNA levels increased on day three post-PNX and peaked on day seven post-PNX, at which point total Igf-1 mRNA expression in the post-PNX lung was 1.5- fold more than that of sham operated controls (p<0.05, n=4/group, Figure 3.10A) while Igf-

1Eb mRNA levels were four times more than that of corresponding sham operated controls

(p<0.001, n=4/group, Figure 3.10B). By day 15, Igf-1 and Igf-1Eb mRNA expression in post-

PNX lung tissue had declined to levels on day three post-PNX (Figure 3.10A, B). Throughout the 15-day time course, Igf-1 and Igf-1Eb mRNA expression levels in sham operated control lungs were not significantly changed. Densitometry analysis of IGF-1 Western blots revealed a gradual increase in IGF-1 protein levels in post-PNX lungs until day 15 (Figure 3.10C, D).

However, no statistically significant difference between post-PNX and sham operated control lungs was observed at the timepoints studied although there appeared to be a trend towards higher IGF-1 protein levels in post-PNX lung tissue (Figure 3.10C, D).

101 A

Figure 3.10: Igf-1 mRNA is upregulated in post-PNX lung tissue.

RNA and protein were extracted from lung tissue homogenates of pneumonectomised (PNX), sham operated C57BL/6 female wildtype mice (aged 8-12 weeks) on days 1, 3, 7 and 15 post-surgery (n=3-4/group/timepoint). Total RNA was reverse transcribed and real- time PCR performed using Taqman probes specific for total Igf-1 mRNA (A) and Igf-1Eb mRNA (B) normalised to the house-keeping gene Pgk-1. (C, D) IGF-1 protein was quantified by Western blot analysis and normalised to α-tubulin loading control. All values represented as mean ±SEM, *p<0.05, **p<0.01, ***p<0.001.

102 3.2.5- Immunohistochemical analysis reveals a role for IGF-1R signalling in PNX lungs

IGF-1 mediates its effects through its cognate receptor IGF-1R, which when phosphorylated activates two signalling pathways, the pERK-MAPK and PI3K-Akt pathways. The pERK is primarily associated with cell proliferation whilst the PI3K-Akt pathway is associated with cell survival and protein synthesis (Chapter one, Section 1.9.2). Given that IGF-1 expression was detected in post-PNX lung tissue by IHC, it is reasonable to speculate that IGF-1R expression might similarly be elevated or activated in response to PNX. Thus, IGF-1R, pIGF-1R and pERK-1/2 expression in the lung was examined over the 21-day time course following PNX or sham thoracotomy surgery.

IGF-1R was ubiquitously expressed in both post-PNX and sham operated lung tissue, with highest expression noted in airway epithelial and subpleural mesenchymal cells (Figure

3.11A). In unoperated control lung tissue, the percentage of IGF-1R+ cells, quantified by IHC, was 3±0.5% (Figure 3.11B). On day one post-surgery, both PNX and sham surgery similarly induced IGF-1R expression in the lung but not significantly more than the levels in unoperated controls (11.53 ±9% for sham operated controls; 18.98 ±2.8% for PNX, p>0.05 compared to unoperated controls, n=3/group, Figure 3.11B). However, from day three to day 21 post- surgery, PNX operated lungs exhibited significantly higher percentage of IGF-1R+ cells than sham operated control lungs, with maximum expression of IGF-1R on day three post-PNX

(34.4 ±14% for PNX vs 1.2 ±0.3% for shams, p<0.001, n=3/group, Figure 3.11B). qRT-PCR analysis revealed a significant reduction in Igf-1r mRNA expression levels in post-PNX lungs on day three and day seven post-surgery (Figure 3.12A). In contrast Western blot analysis of total IGF-1R protein levels showed no significant changes in IGF-1R expression (Figure

3.12B, C).

103 A

and graph and All values All represented mean as *p<0.05, ±SEM, **p< post between cells expressing in arrow thin mesenchymal cells ( ( Figure 3.1

A I rsos to response In )

: :

Immunohistochemicallocalisation (A)

h Scale bar in in bar Scale

( cells endothelial vascular and )

PNX

c, g c,

), ),

alveolarmacrophages (

ugr, IGF surgery,

and

PNX, sham operated and unoperated control lungs (n=3/gr lungs control unoperated and operated sham PNX,

e = e

50µm. For all other images scale bar =20µm. ( =20µm. bar scale images other all For 50µm.

R xrsin s peuae i vros ug el ye icuig iwy lb cells club airway including types cell lung various in upregulated is expression 1R

). Under quiescent conditions, IGF conditions, quiescent Under ).

),alveolar typeII epithelial cells(black solid arrow in

0.01,***p<0.001.

andquantification of (B) IGF

) Graph comparing the temporal changes in percentage of IGF of percentage in changes temporal the comparing Graph )

1Rexpressing cells in the parenchyma lung post

expression is normally low or absen or low normally is expression

oup/timepoint) quantified from immunohistochemical images.immunohistochemical from quantified oup/timepoint)

), alveolar ), type I epithelial cells(multi B

IGF-1R+ve cells (%)

Dayssurgery after

t as exemplified in image in exemplified as t

***

(a, b b (a,

and

PNX.

, s ),

Unp. Control

PNX

SHAM

ubpleural

headed

R R e e

104

B IGF-1R 95 kDa

α-tubulin 50 kDa

PNX D7

PNXD3

PNX D1

Sham Sham D7

Sham Sham D1 Sham Sham D3

PNX D15

Sham Sham D15 Unp. Unp. Cont. C

Figure 3.12: PNX induces a transient decline in Igf-1r mRNA.

(A) Comparison of Igf-1r mRNA measured by qRT-PCR and (B, C) IGF-1R protein levels measured by Western blot in the lungs of PNX, sham operated controls and unoperated controls. Igf-1r mRNA levels were lower in PNX lungs compared to Sham operated controls on day three and day seven post-surgery (A). However, no significant difference was observed in IGF-1R protein expression levels detected by western blot (B, C). All values represented as mean ±SEM, *p<0.05, n=3/group.

105 Consistent with the IGF-1R IHC data, enumeration of pIGF-1R and pERK-1/2+ cells

(indicative of IGF-1R activation) by IHC revealed a trend that matched the profile of IGF-

1/IGF-1R expression with peak expression occurring on day three post-PNX (Figure 3.13A-

C). Relative to sham operated control lungs at the same time point, PNX induced a 15-fold increase in the percentage of pIGF-1R+ cells (3.6 ±1.74% for shams vs 60.84 ±3.82% for PNX, p<0.001, n=3/group, Figure 3.13B) and approximately a 70% increase in pERK+ cells (11.74

± 2.1% for shams vs 39.13 ±13.89% for PNX, p<0.05, n=3/group, Figure 3.13C) on day three post-surgery. pIGF-1R and pERK were prominently expressed by club cells, alveolar type two cells and alveolar macrophages, identified by their morphology and location in the lung

(Figure 3.13A). We attempted to further validate these findings using Western blot analysis and found that similar to the IGF-1R data, no significant increase in pIGF-1R expression was noted by this method (Figure 3.13D, E).

106

A B 80 *** SHAM 60 PNX 40 20 10

8 field of view of field 6 +ve cells/ pIGF-1R 4 2 0 1 3 7 15 21 C Days after surgery 60 *

40 * *

20 pERK +ve cells (%) pERK +ve cells

0 1 3 7 15 21 Days after surgery

E D

Figure 3.13: IHC analysis reveals significant increase in pIGF-1R and pERK expression in post-PNX lung tissue.

(A) Representative immunohistochemical images of pIGF-1R (a, b) and pERK-1/2 (c, d) stained sections taken from PNX and sham operated (SHAM) lung tissue sections on day three post-surgery. pIGF-1R and pERK-1/2 positive cells were evident both in the alveolar regions (arrows in A) and airway (red dotted line in Aa). (B) Immunohistochemical quantification of pIGF-1R expressing cells showing maximal expression on day three (***p<0.001). (C) IHC quantification of pERK-1/2 stained cells revealed that pERK-1/2

expression was significantly increased in PNX lungs (*p<0.05) on days 1, 3 and 15 post- surgery. However, Western blot analysis of pIGF-1R expression from whole lung tissue lysates (D, E) showed no significant increase in pIGF-1R. All values represented as mean ±SEM, n=3/group/timepoint. Scale bar in A = 20μm.

107 3.2.6- IGFBP-3 positive cells are prominent in mouse lung tissue post-PNX

The bioavailability of IGF-1 is modulated by IGF-1 binding proteins (IGFBPs) that are found complexed to IGF-1 in serum and interstitial spaces. At least six types of IGFBPs have been reported sharing similarities in size and structure [253]. Since IGFBP-3 is by far the most abundant IGFBP, its mRNA and protein expression profile in response to PNX was assessed.

IHC analysis demonstrated that IGFBP-3 expression was significantly up-regulated in post-

PNX lungs, specifically in alveolar and airway cells (Figure 3.14A). The percentage of IGFBP-

3+ cells increased from day one post-PNX, peaked on day three post-PNX (0.19 ±0.09% for shams vs 16.47 ±5.49% for PNX, p<0.001, n=3/group, Figure 3.14B) and remaining significantly elevated until day 15 post-PNX. In contrast, the percentage of IGFBP-3+ cells in sham operated control mouse lung tissue was only transiently increased on day one post- surgery then remained at baseline levels throughout the 21-day time course (Figure 3.14A, B).

There was no significant change in Igfbp-3 mRNA expression except on day seven post-surgery where Igfbp-3 mRNA expression levels in post-PNX lung tissue was significantly less than that of the sham operated controls (Figure 3.14C). Total IGFBP-3 protein levels measured by

Western blot revealed a time-dependent increase in IGFBP-3 expression in both post-PNX and sham operated control lungs until day seven with no significant difference observed between the two groups at all timepoints examined (Figure 3.14D, E).

108

B C SHAM 0.4 30 PNX ** Unp. Control *** 0.3

20 *** Pgk-1 0.2

10 expression mRNA 0.1

** to relative IGFBP-3 +ve cells (%) +ve cells IGFBP-3 0 Igfbp-3 0.0 0 1 3 7 15 0 1 3 7 15 21 Days after surgery Days after surgery 4 D E Unp. Control SHAM 3 PNX

-tubulin ratio 2  1

IGFBP-3/ 0 0 1 3 7 15 Days after surgery

Figure 3.14: Heightened IGFBP-3 protein expression in post-PNX mouse lungs.

(A, B) IHC analysis of IGFBP-3 expression in post-PNX and sham operated (SHAM) control mouse lung tissue. Compared to sham operated and non-operated control lung (Unp. Control), IGFBP-3 expression was increased in post-PNX lungs. IGFBP-3 was detected in alveolar and airway cells up until day 15 post-surgery (n=3/timepoint/group). Real-time quantitative PCR analysis of Igfbp-3 mRNA (C) and IGFBP3 protein expression by Western blot (D, E) revealed no significant difference in IGFBP-3 expression in PNX compared to sham and unoperated control mouse lung tissue. All graphed values are represented as mean ±SEM, *p<0.05, **p<0.01, ***p<0.001. Scale bar in A = 20μm.

109 3.2.7- IGF-1 supplementation increases lung growth by day seven post-PNX

Having examined changes in the IGF-1 signalling pathway in response to PNX, the effect of recombinant human IGF-1 (rhIGF-1) supplementation on regenerative lung growth post-PNX was then determined. Importantly, human IGF-1 is highly homologous with mouse IGF-1 and has similar biological activity, such that murine cells are typically used to evaluate the activity of rhIGF-1 activity [254]. A single daily dose of rhIGF-1 (0.2mg/kg/day) was administered to mice by intraperitoneal (i.p.) injection, beginning a day prior to PNX surgery then for seven consecutive days before finally sacrificing them for IHC analysis (Figure 3.15A). After seven days of treatment, analysis of lung volume by µCT revealed that rhIGF-1 treated PNX mice compared to PBS-treated control PNX mice, underwent a significant increase in total lung volume (358.6 ±1.41mm3 for PBS vs 421.9 ±17.36mm3 for IGF-1, p<0.05, n=3 and 5 respectively, Figure 3.15B) and tissue volume (12.63 ±0.54mm3 for PBS vs 25.36 ±1.42mm3 for IGF-1, p<0.01, Figure 3.15C). Additionally, rhIGF-1 treatment significantly increased the percentage of Ki67+ cells (11.3 ±1.5% in PBS vs 19.5 ±1.9% in IGF-1 treated, p<0.001, n=3 and 5 respectively, Figure 3.15D, E).

3.2.8- IGF-1R inhibition reduces regenerative lung growth post-PNX

The role of IGF-1R signalling in post-PNX lung growth was further tested by pharmacological inhibition of IGF-1R activation using BMS-536294, a small molecule competitive inhibitor of

IGF-1R ATP kinases [255]. A single daily dose of BMS-536294 (20mg/kg/day) or its vehicle control (0.2% DMSO/PBS) was administered via oral gavage for 15 consecutive days beginning a day prior to PNX surgery (Figure 3.16A). Mice were longitudinally imaged by

µCT on days 4, 7 and 14 post-surgery to monitor changes in lung volume. In both BMS-

536294-treated and vehicle control PNX mice, there was a gradual increase in total lung

110 volume (p=0.0001, Figure 3.16B) and tissue volume (p=0.0032, Figure 3.16C) consistent with earlier observations in untreated PNX mice (Figure 3.2). However, BMS-536294-treated mice had a lower mean total lung volume and tissue volume than vehicle-treated controls. This was observed at all time points. Statistical comparison of the growth curves by two-way ANOVA revealed that IGF-1R inhibition significantly reduced total lung volume (p=0.0009, Figure

3.16B) and tissue volume (p=0.0376, Figure 3.16C). The most significant difference in total lung volume was recorded on day seven post-surgery (437.8 ±17.45mm3 for PBS vs 391.7

±7.87 mm3, p<0.05, n=4 and 5 respectively, Figure 3.16B). On day 15, the total lung volume of BMS-536294-treated mice was lower but not significantly different from that of vehicle controls, however the total tissue volume was similar between the two groups (Figure 3.16B,

C). In agreement with the in vivo tissue volume measured on day 14, lung wet weights measured on day 15 were not significantly different between BMS-536294-treated mice and vehicle controls (Figure 3.16D). Overall, the results indicated that IGF-1R inhibition delays lung growth post-PNX in mice.

111

500 C ) 3 * **

) 30 3 400 20

300 10

Total lung volume (mm volume lung Total 200 Tissue(mm volume 0 PBS IGF-1 PBS IGF-1

E *** 25

15 ** 10

(%) the lung in cells Ki67+ 0 PBS D7 IGF-1 D7

Figure 3.15: IGF-1 Supplementation increases lung volume and cell proliferation post-PNX.

(A) rhIGF-1 (0.2mg/kg/day) was delivered via daily i.p. injections beginning a day prior to PNX surgery then for seven days before lungs were harvested and weighed. IGF-1 significantly increased total lung volume (B) and tissue volume (C) as evaluated by µCT as well as the percentage of Ki-67+ proliferating cells (D, E). Lung sections were stained with stay green (Ki-67) and counterstained with hematoxylin. Representative images in D have been pseudo-coloured (Ki67, Green, nuclear staining is shown in Blue). Quantitative IHC analysis (E) was performed on the original unaltered images using the Image J algorithm in Appendix C. Values represented as mean± SEM, n=3 for PBS; n=5 for IGF-1 treated PNX mice. Scale bar in D= 100µm. *p<0.05, **p<0.01, ***p<0.001.

112

B ) 550

3 Saline BMS 500  treatment, p= 0.0009  time, p= 0.0001 * 450

400

350

Total lung volume (mm volume lung Total 300 4 7 14 Days after pneumonectomy

C 35 Saline D 150

BMS ) 3  treatment, p= 0.0376 30  time, p= 0.0032 140

25 130

20 120

Tissue(mm volume Right lung weight (mg) weight lung Right

15 110 4 7 14 PBS D15 BMS D15 Days after pneumonectomy

Figure 3.16: Pharmacological inhibition of IGF-1R reduces total lung volume and tissue volume post-PNX.

(A) The IGF-1R antagonist BMS-536294 was delivered via oral gavage at a daily single dose of 20mg/kg/day for 15 consecutive days beginning a day prior to PNX surgery. Lungs were imaged by µCT on days 4, 7 and 15. IGF-1R inhibition significantly reduced total lung volume (B) and tissue volume (C) measured in vivo by µCT. Lung wet weight measures were similar between the two groups (D). Values represented as mean± SEM. Statistical analysis of lung volume was performed using a Two-Way ANOVA followed by Bonferroni’s post hoc-test. Lung weight data was compared using an Unpaired Students T-test. n=3-4 for PBS; n=5 for BMS. *p<0.05. Δ= difference.

113 3.3- Discussion

This study tested the hypothesis that IGF-1 plays a significant role in driving lung regeneration following PNX. Data presented here showed that PNX-induced regenerative lung growth is associated with increased IGF-1 expression and activation of the IGF-1 signalling pathway.

Furthermore, it was shown that supplemental IGF-1 enhanced post-PNX lung growth and that pharmacological blockade of IGF-1R signalling delayed post-PNX lung growth. These findings suggest that IGF-1 plays an important role in ensuring a successful regenerative response post-PNX.

To examine the role of IGF-1 in post-PNX lung growth, it was necessary to first understand the cellular and morphological changes that occur in response to PNX. Consistent with existing literature, it was shown that PNX in young adult mice (aged 2-3 months) leads to restoration of lung volume, lung weight and tissue content within 14 days after PNX surgery (Figure 3.2)

[3, 4, 52, 256]. For the first time, this study has measured longitudinal changes in total lung volume and tissue volume as well as the dynamic changes in the topography of the lung that occurs in vivo after left lung PNX, using live µCT imaging on freely breathing anesthetised mice. This method has been shown to be safe and has previously been used in various rodent lung disease models including lung fibrosis and emphysema [230-232]. The findings reported here are consistent with other µCT studies of murine lungs and are backed up by stereological analysis of the lung from other studies that show evidence of new tissue growth and neoalveolarisation [3, 52, 56, 257]. The resolution of the instrument used, although sufficient to image and quantify the 3D-topography of the lung at a lobular level, was not sufficient to enable evaluation of the anatomical changes at the acinar level. Nonetheless, the method employed in this study contributes to the development of µCT lung imaging analysis and

114 further enhances our understanding of the morphological adaptions of the lung in response to

PNX surgery.

Unlike other injury models such as bleomycin or influenza-induced injury, the PNX model shows little evidence of inflammation and fibrosis [111, 258]. This permits researchers to examine factors contributing to the regenerative lung growth in a niche less complicated by overt histopathological features. Histological assessment of PNX lungs revealed a relatively normal lung at a gross histological level for the entirety of the period of observation (Figure

3.3). Closer examination the lungs revealed some atypical changes. Firstly, the presence of expanded airspaces in parts of the lung early on after surgery was noticed. This was likely caused by increased mechanical tension arising from increased cardiac output and ventilatory forces experienced by the remaining right lung after excision of the left lung [259]. Secondly, on day three and day seven post-PNX, there was an increase in lung density, a feature that may have been manifested by increased cell proliferation and a mild inflammatory response [52].

By day 15 however, these changes were largely resolved and the post-PNX lung appeared much like an unoperated control lung.

Previous studies have shown that neoalveolarisation occurs after left-PNX in young adult mice

[3, 52]. Although no direct evidence of neoalveolarisation was presented in this study, changes in lung volume, lung wet weight and tissue volume were suggestive of new tissue growth.

Further evidence of new tissue growth was obtained from IHC analysis of proliferating cells

(Figure 3.4 and 3.5). Cell proliferation was evident in all major cellular compartments including airway and alveolar epithelial cells, endothelial cells and mesenchymal cells indicating an interdependency and cross-talk between multiple cell types that is crucial for a successful regenerative response. Of note, SpC+ alveolar type two cells, constituted the majority of proliferating cells. Historically, alveolar type two cells have been regarded as the progenitor of type one alveolar cells (AECIs), the cells that comprise the majority of the

115 alveolar epithelium and are involved in gas exchange with adjacent endothelial cells [116].

However, as discussed in detail in Chapter one (Section 1.8) recent studies have revealed that

AECIs can arise from other epithelial progenitor cells among them club cell secretory protein

(CCSP)+ club cells [260] and CCSP/SpC+ BASC cells [40, 261].

Neoalveolarisation post-PNX is reminiscent of developmental alveolarisation involving formation of secondary septa which extend from existing primary septa [262]. Secondary septa formation is guided by α-SMA+ myofibroblasts which deposit elastin fibres at the septal tips supporting newly formed septa [263]. Therefore, expression of α-SMA positive cells was examined. Parenchymal α-SMA+ cells were transiently amplified at the height of maximum cell proliferation on day seven post-PNX suggesting a role for these cells in neoseptation

(Figure 3.6). Similar findings have been reported by Bennett et.al [250] and Ysasi et.al [264] although they detected maximal α-SMA expression on day three post-PNX.

In concert with newly forming secondary septa, post-PNX lung growth involves formation of new blood vessels by a process known as intussusceptive micro-vascular growth which involves sprouting of double layered capillary tributes from a primary capillary in the primary septa [265]. There was no significant difference in the distribution of CD31+ vascular endothelial cells and CD31 total protein levels between post-PNX lung tissue and sham operated control lungs (Figure 3.7). This result does not preclude a role for endothelial cells in post-PNX lung growth and warrants further investigation. Some studies have utilised more advanced ultrastructural imaging techniques such as scanning electron microscopy and synchrotron radiation X-ray tomographic microscopy to study changes to the vasculature during post-natal alveolarisation [266], whereas other studies resort to counting number of vessels [267]. However, the fact that CD31 expression levels remained unchanged could also mean that vascular density matches the epithelial density throughout the duration of post-PNX recovery thereby ensuring that the entire alveolar epithelium including newly formed alveoli

116 are adequately perfused. Additionally, the significance of the vasculature has been demonstrated in a number of landmark studies [139, 140]. It is known from these studies that endothelial cells are not mere by-standers in the regenerative process but produce important ligands including VEGF and EGF that promote epithelial cell proliferation and neoalveolarisation. Taken together, data presented thus far showed that post-PNX lung growth is an active process that not only involves an increase in the volume of the lung, but formation of new tissue, through an expansion of epithelial and mesenchymal cells.

Having gained an understanding of the changes that occur in the regenerating lung, expression of IGF-1 and its signalling components was investigated by IHC, Western blot and qRT-PCR analysis to determine whether IGF-1 signalling plays a role in post-PNX lung growth. By IHC analysis, it was determined that PNX induced robust expression of IGF-1 in a variety of cell types most notably distal airway club cells, alveolar macrophages and AECII progenitors, cell types that play a crucial role in post-natal alveolarisation [116, 268, 269] (Figures 3.8 and 3.9).

Maximum expression of IGF-1 was detected three days post-PNX suggesting that IGF-1 potentially contributes to the increased cell proliferation that occurs on day three and day seven post-PNX. IGF-1 expression has been shown to be up-regulated during alveolar development and could similarly be required for PNX-induced alveolarisation [6, 248]. Analysis of IGF-1 mRNA and protein by qRT-PCR and Western blot respectively also showed increased IGF-1 expression levels in post-PNX lung tissue albeit with different kinetics from that recorded by

IHC (Figure 3.10). Igf-1 and Igf-1 Eb mRNA began to increase on day three and peaked on day seven before returning to baseline levels on day 15 post-PNX. IGF-1 protein expression quantified by Western blot showed that PNX induced a time-dependent increase in total IGF-

1 protein levels with peak expression occurring on day seven post-PNX. However, sham operated control lungs also exhibited a similar but lower increase in IGF-1 protein expression, as a result there was no statistically significant difference in total IGF-1 protein levels between

117 sham operated control lung tissue and post-PNX lung tissue although IGF-1 protein levels appeared higher in post-PNX lungs (Figure 3.10C, D). It was expected that Igf-1 mRNA would peak earlier than day three in response to PNX-induced stretch, preceding the observed increase in IGF-1+ cells on lung tissue sections. Paxson and colleagues [270] showed a 1.4-fold increase in PNX lungs relative to shams on day one after surgery. However, they did not investigate its expression at later timepoints. The apparent increase in IGF-1 mRNA and protein after day three may suggest that there is a second wave of IGF-1 expression which may be required during the post-proliferative phase of neoalveolarisation that involves cell differentiation, septal reorganisation and maturation [6].

Increased IGF-1 expression was accompanied by a coincidental increase in IGF-1R+ cells as revealed by IHC analysis (Figure 3.11). Multiple cell types were positive for IGF-1R including cell types that expressed IGF-1, suggesting a paracrine and autocrine mode of action for IGF-

1 signalling in response to PNX. In unoperated and sham operated control lungs IGF-1R+ expression was relatively low (~3% of total lung cells) indicating tight regulation of IGF-1R signalling, which if unchecked may lead to uncontrolled cell proliferation and neoplastic transformation [271]. Similar to the pattern of IGF-1 expression seen on PNX lung tissue sections, IGF-1R expression peaked on day three post-PNX and was maintained at significantly higher levels until day 21, supporting the notion that IGF-1R signalling may be required past the proliferative phase (Figure 3.11). It was interesting to note transient downregulation of

IGF-1R expression on day seven post-PNX at both mRNA and protein levels (Figures 3.11 and 3.12). It is unclear what causes this decline. One possible explanation is that following ligand binding and activation, the IGF-1R can be internalised and recycled by the endosomal- lysosomal system [272]. When IGF-1 is in excess, as occurs on day three and seven post-PNX, this phenomenon may be exacerbated resulting in reduced expression of IGF-1R, thus acting as a means of regulating IGF-1R activity.

118 Increased expression of IGF-1 in response to PNX not only led to an increased percentage of

IGF-1R+ cells but also resulted in an increased percentage pIGF-1R+ and pERK+ cells as determined by IHC, suggesting that PNX-induced IGF-1 expression activates IGF-1R signalling in several different cell types and actively contributes to the post-PNX lung growth.

Although IHC image analysis showed a robust increase in pIGF-1R+ cells in PNX lungs, this trend was not be replicated by Western blot analysis (Figure 3.13). Similarly, IGFBP-3, the principle binding partner of IGF-1 [253] was highly expressed in multiple cell types on lung tissue sections but no significant differences were observed between post-PNX and sham operated control lungs by qRT-PCR and Western blot analysis (Figure 3.14). However, this does not exclude a role for other IGFBPs in post-PNX lung growth.

The differences between IHC and qRT-PCR/Western blot data may be attributed to several reasons. Firstly, for IHC quantification, regions in the alveolar parenchyma were sampled excluding major airways and blood vessels whereas Western blot and qRT-PCR semi- quantitative analyses were sampled from whole lung homogenates which may include extra- parenchymal components and blood-derived IGF-1/IGFBP-3. Secondly, these differences may suggest that IGF-1 and/or IGFBPs have a broad distribution pattern that may have reduced our ability to detect subtle or spatial-temporal restricted changes in whole lung extracts. Thirdly, it is also possible that the differences witnessed may reflect the variability in availability of the

IGF-1 epitopes depending on the primary antibody used, the protein isolation protocol or antigen retrieval method employed. Furthermore, it is important to remember that regulation of IGF-1 is very complex and is influenced by multiple physiological factors, including stress and nutritional status [273]. Additionally, expression of IGF-1 is regulated by differential splicing of mRNA isoforms, post-transcriptional and post-translational modifications [274]

(reviewed in Chapter one, Section 1.10.6). As an example, the C-terminal Eb peptide, compared to the Ea peptide, has been shown to have higher binding affinity to ECM proteins

119 [252, 275] and as such has a greater probability of being retained in tissue rather than spilling over in serum. Therefore, depending on the isoform preferentially expressed at a particular time, levels of IGF-1 detected in lung tissue may vary. Nonetheless, the consistent patterns of

IGF-1 signalling components on lung sections and the detected increase in IGF-1 mRNA and protein in whole lung homogenates collectively supported the observation that the IGF-1 signalling plays a prominent role in post-PNX lung growth. Additionally, enhanced expression of IGF-1 and its role in promoting lung growth post lung surgery may not be limited to mice.

A recent study by Fu and others [276] reported up-regulation of IGF-1 and IGF-1R in lung specimens as well increased IGF-1 in serum of non-small cell lung cancer patients who had undergone lung resection. Although this was associated with a poor prognosis, it raises the possibility that the sensitivity of IGF-1 to pulmonary mechanical signals is a conserved feature across species [276].

Lastly, the significance of IGF-1 signalling in post-PNX lung growth was tested in two different ways; first by examining the effect of rhIGF-1 supplementation in PNX mice and second by pharmacological blockade of IGF-1R signalling following PNX. IGF-1 supplementation delivered by daily i.p. injections significantly increased lung volume, tissue volume and cell proliferation after seven days of treatment, showing that IGF-1 treatment can enhance regenerative lung growth in PNX mice (Figure 3.15). Conversely, pharmacological inhibition of IGF-1R signalling significantly delayed lung growth post-PNX suggesting that

IGF-1 signalling contributes to a successful regenerative response (Figure 3.16).

In summary, this chapter has revealed many novel aspects about the significance of IGF-1 in post-PNX lung growth. Features of a successful regenerative response in young adult mice were characterised prior to investigating the significance of IGF-1 signalling in this process. It was demonstrated that PNX induces increased expression of IGF-1 that correlates with activation of the IGF-1 signalling complex in the lung, a change that likely contributes to the

120 subsequent increase in cell proliferation ultimately resulting in restoration of lung volume and tissue mass. Finally, the importance of IGF-1 in post-PNX lung growth was demonstrated by addition of IGF-1 which enhanced lung growth and blocking its signalling resulted in delayed lung growth. Therefore, IGF-1 makes a significant contribution to lung regeneration following

PNX in young adult mice.

121

Chapter 4

The effect of age on IGF-1-induced regeneration post-PNX

122 4.1- Introduction The rise in human life expectancy is resulting in an increase of the ageing population in many developed countries [277]. This has been associated with an increase in the incidence of chronic lung diseases such as COPD, IPF and lung cancer, placing an enormous burden on health care

[278]. The lung, like any other organ, declines in function with age. By the age of 20-25 years, the lungs have reached full maturity, after which lung function begins a progressive decline at a rate of 1% per year [279]. Age-associated changes in the lung include reduced strength of respiratory muscles, enlargement of airspaces, reduced lung elasticity and elastic recoil, decreased mucocilliary clearance and a less robust immune system that makes the aged lung susceptible to infection [278, 280]. At a cellular level, ageing is associated with increased oxidative stress, DNA damage, shorter telomeres, mitochondrial dysfunction, increased apoptosis and increased cell senescence, changes that contribute to the aging lung phenotype

[279, 281]. Finding ways to counter age-related chronic lung diseases will undoubtedly help address some of the future challenges arising from an increasing ageing population.

The left lung PNX model is useful for studying the mechanisms of lung adaption and regeneration, i.e. the repair mechanisms within the lung that are initiated to restore the cellular and structural composition of the lung following injury. As discussed earlier (Chapters one and three), PNX in young adult mice and juvenile patients, results in a robust regenerative response whereas aged animals and adult humans have a reduced capacity to regenerate [4].

The limited capacity for adult lung regeneration was challenged in a recent case report in which

Butler et al. reported evidence of lung growth in an adult patient 15 years post-PNX [1].

Mechanistic studies in this area have typically utilised young adult mouse models. Studying the response of geriatric mice to PNX will help unravel these mechanisms and improve our understanding of regeneration in aging.

123 The reduced capacity for lung regeneration in aged animals mirrors the age-related decline in serum levels of growth hormone and IGF-1 in both mice and humans [282]. Therefore, we hypothesised that following left lung PNX, supplemental IGF-1 can enhance the regenerative capacity of lungs of old mice aged 22-24 months, which correspond to the age range of 60-70 years old in humans [283]. In Chapter three, the effect of IGF-1 on post-PNX lung growth was examined by intraperitoneal administration of the growth factor. One of the potential consequences of a bolus infusion of IGF-1 is that it can result in unwanted side effects such as hypotension and edema [164]. This chapter extended the assessment of the role of IGF-1 in lung regeneration using a slow and sustained minipump infusion system. Similar experiments were conducted in both aged mice and young adult mice aged 2-3 months which can be compare to humans aged 18-25 years old [283].

The specific aims of this chapter were:

i) To examine the effects of sustained infusion of exogenous IGF-1 on total lung

volume, aerated lung volume, tissue volume and mean lung density in young and

old PNX mice as evaluated by longitudinal in vivo µCT.

ii) To determine the effects of sustained IGF-1 treatment on the cellular characteristics

of the lung and changes in the IGF-1 signalling pathway in young and old PNX

mice by immunohistochemical analysis.

124 4.2- Results

4.2.1- Sustained infusion of IGF-1 increases the aerated lung volume of old mice but not young mice

To examine the effect of IGF-1 on aging and PNX, the effect of IGF-1 administration on post-

PNX lung growth in young adult mice aged 8-12 weeks and old mice aged 22-24 months was compared. Recombinant human IGF-1 (rhIGF-1, 0.2mg/kg/day) or phosphate buffered saline

(PBS) solution control was continuously delivered subcutaneously via an osmotic pump implanted at the dorsum of the mice. Longitudinal measurements of total lung volume, air volume, tissue volume and lung density were obtained by in vivo μCT imaging of post-PNX lungs on days 4, 7, 14 and 21 after surgery. At the end of the longitudinal imaging timeline, lungs were harvested for morphometric and immunohistochemical analysis (Figure 4.1).

In both PBS and IGF-1 treated young PNX mice, total lung volume was restored to pre- operative levels by day seven post-PNX with the most rapid increase occurring in the first four days (p=0.0001, n=5/group) (Figure 4.2A). There was no significant difference in the mean total lung volume between IGF-1 and PBS treated mice at any of the sampled and comparison of the overall longitudinal trend revealed no significant difference in the total lung volume curves between the two groups (p=0.5593, Figure 4.2A).

Similar to young PNX mice, total lung volumes of both IGF-1 and PBS treated old PNX mice underwent a gradual and significant increase between day four and 21 post-PNX (Figure 4.2B).

Although IGF-1 treated mice showed consistently higher mean lung volumes at all timepoints sampled, overall there was no significant difference between the two groups (p=0.2107, n=5-7 for PBS; n=4-5 for IGF-1, Figure 4.2B). Interestingly, unlike young mice, both IGF-1-treated and PBS-treated old PNX mice did not restore their pre-operative lung volume by day 21 post surgery (Figure 4.2B).

125

Live in vivo µ-CT

(0.2 mg/kg/day)

Young mice (2-3 months old)

Old mice (22-24 months old)

Figure 4.1: Schematic of the IGF-1 supplementation experiments in young and old PNX mice.

Following left lung PNX, an osmotic pump supplying either 0.2mg/kg/day of rhIGF-1 or PBS was implanted subcutaneously in the dorsum of mice. Mice were imaged in vivo by µ-CT at days 4, 7, 14 and 21 post surgery to estimate lung volume and tissue density. Immunohistochemical analysis of markers associated with lung regeneration, indicated above, were analysed on day 23 post-PNX.

126

Figure 4.2: Longitudinal changes in total lung volume and air volume in young and old PNX mice supplemented with IGF-1 or PBS control.

(A-B) Total lung volume (TLV), and (C-D) aerated lung volume (ALV) were measured by in vivo µCT in young (2-3 months old) and old (22-24 months old) PNX mice supplemented with rhIGF-1 or PBS. TLV and ALV were also measured in unoperated age-matched mice as a control (blue bar). TLV included the volume of air and tissue measured in the lung. Values represented as mean±SEM. Statistical comparison between IGF-1-treated and PBS-treated age-matched PNX mice was performed using Two-Way ANOVA and Bonferroni’s post-hoc test. p-values from this analysis are shown in each graph where Δ denotes difference between treatment groups or timepoints, compared to age- matched PNX group. Statistical comparison between unoperated control mice and PNX lung volumes on day 21 was done by One-Way ANOVA followed by Bonferroni’s post-hoc test. #p<0.05 IGF-1 treated PNX compared to aged matched unoperated control. $$p<0.01 PBS treated PNX compared to aged matched unoperated control. ns= not significant. Old PNX PBS treated n=5-7, Old PNX IGF-1 treated n=4-5, Old unoperated control n=3, Young PNX mice n=5/group.

127 Total lung volume encompasses both air and tissue components in the lung. A µCT algorithm was employed to discriminate air and tissue components within the lung to examine the effect of IGF-1 treatment on the aerated lung volume post-PNX. The trends recorded in measurements of aerated lung volume closely mirrored those observed for total lung volume

(Figure 4.2C-D). Young PNX mice regardless of IGF-1 treatment underwent a gradual increase in aerated lung volume (p=0.0001) restoring the air content by day four post-PNX

(Figure 4.2C). IGF-1 treatment did not significantly alter the volume of air in young PNX mice

(p=0.6029, n=5/group, Figure 4.2C). In contrast, in old PNX mice, IGF-1 treatment significantly increased the volume of air in the lungs compared to PBS controls (p=0.0011, n=5-7 for PBS; 4-5 for IGF-1, Figure 4.2D).

4.2.2- IGF-1 treatment in old PNX mice transiently reduces the density of the lung

To further examine the differences in post-PNX lung growth between young and old mice, lung tissue volume and density was measured. In response to PNX, young mice gradually increased the volume of tissue in the lung (p=0.0006) reaching pre-operative levels by day 21 (Figure

4.3A). The increase in tissue volume in IGF-1 treated young PNX mice was not significantly different from that of PBS controls (p=0.2005, n=5/group, Figure 4.3A). In contrast, IGF-1 and PBS treated old PNX mice did not show any significant changes in tissue volume during the 21-day time course (p=0.6913, Figure 4.3B) and the tissue content in both groups of mice was similar overall (p=0.1264, Figure 4.3B). At day 21 post-PNX, the volume of tissue in IGF-

1 and PBS treated old PNX mice was significantly less than that of age-matched unoperated control mice (32.1 ±18.4% less for PBS, n=5, p<0.05 and 36.6 ±19.1% less for IGF-1, n=4, p<0.05, Figure 4.3B).

128 Although measurement of in vivo tissue volume is a good indicator of tissue density, the structural composition of the lung may vary despite similarities in tissue volume (for example a fibrotic lesion may be denser than a tumour of similar magnitude). Therefore, changes in mean lung density (measured in Hounsfield Units, HU) in response to PNX were also measured. In young PNX mice, the mean lung density was not significantly altered between day four and day 21 post-PNX (p=0.0899, n=5/group, Figure 4.3C). There was no difference in the mean lung density between IGF-1 treated and PBS control young PNX mice except on day seven post-PNX when PBS controls exhibited significantly higher density than IGF-1 treated animals (11.9 ±2.04% higher, p<0.05, n=5/group, Figure 4.3C).

In contrast to young mice, IGF-1 treatment significantly decreased the overall mean lung density in the group of older mice (p=0.0128, n=5/group, Figure 4.3.D) with the largest difference recorded on day 14 post-PNX (9.9 ±3.96% difference). However, by day 21, the mean lung density of IGF-1-treated old mice was comparable to that of age-matched PBS controls and not significantly different to the mean lung density of unoperated control old mice

(Figure 4.3D).

129

A Young mice B Old mice 100 100 PBS #,$ IGF-1

Unoperated Control

) 3 )

3 80  Treatment, p=0.2005 80  Time, p=0.0006 ns

60 60

40 40 PBS IGF-1 Unoperated Control

Tissue volume (mm volume Tissue 20 20 (mm volume Tissue  Treatment, p=0.1264  Time, p=0.6913

0 0 4 7 14 21 Pre-Op 4 7 14 21 Pre-Op Days after Pneumonectomy Days after pneumonectomy

Young mice Old mice C -250 D -250 PBS IGF-1 Unoperated Control -300 ns * -300  Treatment, p=0.0128  Time, p=0.2113

-350 -350 ns

PBS -400 IGF-1 -400

Unoperated Control Mean lung density (HU) density lung Mean

 Treatment, p=0.2075 (HU) density lung Mean  Time, p=0.0899 -450 -450

4 7 14 21 Pre-Op 4 7 14 21 Pre-Op Days after pneumonectomy Days after Pneumonectomy

Figure 4.3: IGF-1 supplementation transiently reduces mean lung density and has no significant effect on lung tissue volume in old PNX mice.

(A-B) Longitudinal evaluation of lung tissue volume and (C-D) mean lung density by µCT in pneumonectomised young mice (aged 2-3 months) and old mice (aged 22-24 months) in the presence or absence of sustained systemic infusion of rhIGF-1. Mean lung density was measured in Hounsfield Units (HU). Values represented as mean±SEM. Statistical comparison between IGF-1-treated and PBS-treated age-matched PNX mice performed by Two-Way ANOVA and Bonferroni’s post-hoc test. p-values from this analysis are shown in each graph where Δ denotes difference between treatment groups or timepoints, *p<0.05 compared to age-matched PNX group. Statistical comparison between unoperated control mice and PNX lung volumes on day 21 was done by One- Way ANOVA followed by Bonferroni’s post-hoc test. #p<0.05 IGF-1 treated PNX compared to aged matched unoperated control. $p<0.05 PBS treated PNX compared to aged matched unoperated control. ns= not significant. Old PNX PBS treated n=5-7, Old PNX IGF-1 treated n=4-5, Old unoperated control n=3, Young PNX mice n=5/group.

130 4.2.3 - Sustained infusion of IGF-1 induces IGF-1 signalling in young but not old PNX mice

IGF-1 promotes cell proliferation and differentiation by binding to its cognate receptor (IGF-

1R) triggering its phosphorylation and downstream activation of two principle signalling pathways, the pAKT and pERK pathways [284]. Though both pAKT and pERK signalling pathways contribute to cell proliferation, the pERK pathway is thought to be the primary driver of this process in response to IGF-1 stimulation [285]. Thus, to examine the effect of sustained systemic delivery of IGF-1 on post-PNX lung growth, mice were culled on day 23 post-PNX and their lungs prepared for immunohistochemical analysis of components of the IGF-1 signalling pathway including IGF-1, IGF-1R, pIGF-1R, pERK-1/2 and IGFBP-3 (the principle binding partner of IGF-1) [286]. The effect of IGF-1 supplementation on cell proliferation and lung tissue morphology was also assessed.

In young PNX mice, following 22 days of IGF-1 supplementation, IGF-1 staining was detected in various lung cell types including type one and type two pneumocytes, endothelial cells and mesenchymal cells, and the proportion of IGF-1+ cells in the lung was significantly increased

(~12-fold increase, p<0.05, n=5/group, Figure 4.4A-B). In contrast, IGF-1+ cells were rarely detected in lung sections from PBS control young PNX mice, IGF-1 treated and PBS treated old PNX mice and the proportion of IGF-1+ cells was similar among the three groups (3.65

±0.9 cells/field of view, n=4/group, Figure 4.4A). Consistent with an increased expression of

IGF-1, there was a significantly higher percentage of pIGF-1R+ cells in lung tissue from IGF-

1 treated young PNX mice compared to age-matched PBS treated PNX mice (~12.4-fold increase, p< 0.01, n=5/group, Figure 4.5A-B) whereas pIGF-1R staining was rarely detected on lung sections from old PNX mice (Figure 4.5A). Additionally, in IGF-1 treated young PNX mice, compared to age-matched PBS controls, there was significantly increased percentage of pERK-1/2+ cells (~3.65-fold increase, p< 0.05, n=5/group, Figure 4.6A-B). In old PNX mice,

131 approximately 5% of lung parenchymal cells were pERK-1/2+ and IGF-1 treatment did not significantly alter the percentage of pERK-1/2+ cells (Figure 4.6B).

A IGF-1 stain

Young Old

PBS

1 -

+rhIGF

B 60 * PBS 50 IGF-1 40

30 15

10 5

IGF-1+ve cells/field of view of cells/field IGF-1+ve 0 Young Old

Figure 4.4: Infusion of IGF-1 increases the number of IGF-1 positive cells in pneumonectomised young mice but not old mice.

(A) Shows representative IHC images of IGF-1 stained lung sections and (B) corresponding quantitative plots from young adult PNX mice (aged 2-3 months) and old PNX mice (aged 22-24 months) treated with rhIGF-1 or PBS control for 22 days. At least 20 non-overlapping random images (5/lobe) were captured for quantitative analysis using the Image J algorithm in Appendix C. Two sections at least 50µm apart were analysed per mouse. Isotype controls are shown in Appendix F. Statistical comparison between age-matched rhIGF-1 treated and PBS controls was done using the unpaired student’s t-test, values are represented as mean±SEM, *p<0.05, n=5/group for young PNX mice, n=4/group for old PNX mice. Scale bar in all images =20µm.

132

pIGF-1R stain

A Young Old

PBS

1 -

+rhIGF

B 20 ** PBS 15 IGF-1 10 5

2.0 1.5 1.0

pIGF-1R+ve cells (%) cells pIGF-1R+ve 0.5 0.0 Young Old

Figure 4.5: IGF-1 supplementation activates IGF-1R in young pneumonectomised mice but not old mice.

Following rhIGF-1 treatment or PBS treatment for 22 days, lung sections from young adult PNX mice (aged 2-3 months) and old PNX mice (aged 22-24 months) were stained for pIGF-1R to demonstrate the level of activation of IGF-1R. Representative pIGF-1R IHC images are shown in A. Panel B shows corresponding pIGF-1R quantitative plots derived from analysis of at least 20 non-overlapping images (5/lobe) using the Image J algorithm in Appendix C. Two sections at least 50µm apart were analysed per mouse. Isotype controls are shown in Appendix F. Statistical comparison between age-matched rhIGF-1 treated and PBS controls was done using the unpaired student’s t-test, values are represented as mean±SEM, **p<0.01, n=5/group for young PNX mice, n=4/group for old PNX mice. Scale bar in all images =20µm.

133

pERK stain A Young Old

PBS

1 -

+rhIGF

B 15 PBS * IGF-1

(%) +ve pERK cells 0 Young Old Figure 4.6: Activation of downstream pERK signalling in young pneumonectomised mice supplemented with IGF-1.

(A) Representative IHC images of pEKR-1/2 stained lung sections from young adult PNX mice (aged 2-3 months) and old PNX mice (aged 22-24 months) treated with rhIGF-1 or PBS for 22 days. (B) Quantitative analysis of pERK-1/2 staining generated from the analysis of at least 20 non-overlapping images (5/lobe) using the Image J algorithm in Appendix C. Two sections at least 50µm apart were analysed per mouse. Isotype controls are shown in Appendix F. Statistical comparison between age-matched rhIGF-1 treated and PBS controls was done using the unpaired student’s t-test, values are represented as mean±SEM, *p<0.05, n=5/group for young PNX mice, n=4/group for old PNX mice. Scale bar in all images =20µm.

134 IGF-1R and IGFBP-3 were prominently expressed in the airway and lung parenchyma of both young and old PNX mice (Figure 4.7, 4.8 respectively) and quantitative immunohistochemical analysis of these markers revealed that the percentage of IGF-1R+ and IGPBP-3+ cells was not significantly altered by IGF-1 treatment in both young and old PNX mice (Figure 4.7B, 4.8B respectively) although there was a trend towards increased IGFBP-3 expression in IGF-1 treated old PNX mice compared to age-matched PBS control mice (Figure 4.8B).

4.2.4- Sustained infusion of IGF-1 is associated with increased cell proliferation and lung septation in young but not old mice

In young PNX mice, IGF-1 treatment was associated with an increase in Ki67+ proliferating cells (49.4 ±14.02% increase, p<0.05, n=5/group, Figure 4.9A-C) and a marked increase in proliferation of alveolar type two progenitor cells (36 ±11.62% increase, p<0.05, n=5/group,

Figure 4.9A, C). Conversely, consistent with the lack of evidence of IGF-1R activation in old

PNX mice, the percentages of Ki-67+ and Ki67/SpC+ cells were not significantly altered by

IGF-1 treatment in old PNX mice (Figure 4.9B-C). Interestingly, the percentage of Ki-67+ cells in PBS treated young PNX mice was approximately five-fold more than that of old PNX mice (p<0.0001, Figure 4.9C) whereas the proportion of dual labelled Ki67/SpC+ cells was similar between young and aged mice (ranging from 13-17.7% in young PNX mice and 16.0-

20.4% in old PNX mice, Figure 4.9B-C).

135 IGF-1R stain A Young Old

PBS

1 -

+rhIGF

B 15 PBS IGF-1 10

IGF-1R+ve cells (%) cells IGF-1R+ve

0 Young Old

Figure 4.7: The number of IGF-1R expressing cells within the lungs of young and old pneumonectomised mice are similar and are not significantly altered by IGF-1 supplementation.

(A) Representative IHC images of IGF-1R stained lung sections from young adult PNX mice (aged 2-3 months) and old PNX mice (aged 22-24 months) treated with rhIGF-1 or PBS for 22 days. (B) Quantitative analysis of IGF1R cell numbers. At least 20 non- overlapping images (5 areas/lobe were analysed) using the Image J algorithm in Appendix C. Two independent lung tissue sections, at least 50µm apart were analysed per mouse. Isotype controls are shown in Appendix F. Statistical comparison between age-matched rhIGF-1 treated and PBS controls was performed using the unpaired student’s t-test.

Values are represented as mean±SEM. n=5/group for young PNX mice, n=4/group for old PNX mice. Scale bar in all images =20µm.

136 IGFBP-3 stain A Young Old

PBS

+rhIGF

B 10 PBS 8 IGF-1

2 IGFBP-3 +ve cells (%) +ve cells IGFBP-3 0 Young Old

Figure 4.8: IGF-1 supplementation does not significantly alter the number of IGFBP- 3 expressing cells in the lungs of young and old mice post-PNX.

(A) Representative IHC images of IGFBP-3 stained lung sections from young adult PNX mice (aged 2-3 months) and old PNX mice (aged 22-24 months) treated with rhIGF-1 or

PBS for 22 days. (B) Quantitative analysis of the % IGFBP3+ cell number was performed on at least 20 non-overlapping images (5 images/lobe) using the Image J algorithm in Appendix C. Two lungs sections at least 50µm apart were analysed per mouse. Isotype controls are shown in Appendix F. Statistical comparison between age-matched rhIGF-1 treated and PBS controls was performed using the unpaired student’s t-test, values are represented as mean±SEM. n=5/group for young PNX mice, n=4/group for old PNX mice. Scale bar in all images =20µm.

137

Ki67 (red)/SpC (brown) dual stain A Young Old

PBS

1 -

+rhIGF

B 25 C * * 20 PBS IGF-1 20 15 10 15 5 3 10

2 Ki-67-SpC +ve Ki-67-SpC

(% SpC+) total of 5 1

0 (%)the in lung cells Ki67+ 0 Young Old Young Old

Figure 4.9: Continuous infusion of IGF-1 enhances cell proliferation in the lungs of young but not old mice post-PNX.

(A) Lung sections from young adult PNX mice (2-3 months) and old PNX mice (aged 22- 24 months) treated with rhIGF-1 or PBS for 22 days were simultaneously stained for Ki67 to detect proliferating cells (red) and SpC to detect alveolar type two cells (brown). Inset in A shows a dual labelled Ki67/SpC+ cell. (B) Ki67/SpC+ cell quantification. (C) Ki67+ cell quantification. Quantitative IHC was performed on at least 20 non-overlapping images

(5 images/lung lobe) using the Image J algorithm in Appendix D. Two lung tissue sections at least 50µm apart were analysed per mouse. Isotype controls are shown in Appendix F. Statistical comparison between age-matched rhIGF-1 treated and PBS controls was performed using the unpaired student’s t-test. Values are represented as mean±SEM, *p<0.05, n=5/group for young PNX mice, n=4/group for old PNX mice. Scale bar in all images =20µm.

138 Given the significance of a supporting matrix and vasculature in the process of neoalveolarisation [151, 287], we assessed the effect of IGF-1 supplementation on α-SMA (a marker for myofibroblasts, matrix producing cells) and CD31 (a marker for vascular endothelial cells) expression in both young and old PNX mice. Parenchymal α-SMA+ cells were more prevalent in old PNX mice compared to young PNX mice (145.1 ±71.62% higher, p< 0.05, young PNX (n=10), old PNX (n=8), Figure 4.10A-B). Addition of IGF-1 did not significantly alter the proportion of α-SMA+ cells in either young or old PNX mice (Figure

4.10B). Similarly, the percentage of CD31+ endothelial cells was not significantly altered by

IGF-1 treatment in both young and old PNX mice although there appeared to be a moderate but non-significant increase in CD31+ cells in IGF-1 treated old PNX mice compared to age- matched PBS controls (Figure 4.10C-D).

Lastly, we evaluated the effect of IGF-1 treatment on the structure and alveolar density using

H&E stained sections of lungs from PNX treated mice. IGF-1 treatment did not induce any pathological changes in the lung such as inflammation, emphysema or fibrosis following 22 days of sustained delivery (Figure 4.11A). Compared to young PNX mice, lungs of old PNX mice appeared more emphysematous with thinner septa (Figure 4.11A). Interestingly, enumeration of the number of alveoli per field of view (FOV) revealed that IGF-1 treated young

PNX mice had a significantly higher number of alveoli/FOV compared to age-matched PBS controls (34.32 ±3.5% increase, p<0.01, n=5/group, Figure 4.11B). In contrast, there was no significant difference in alveolar density between IGF-1 treated and PBS treated old PNX mice

(Figure 4.11B). Overall, this histological and immunohistochemical data demonstrate that

IGF-1 infusion leads to increased activation of IGF-1 signalling which may contribute to enhanced cell proliferation and lung septation in young PNX mice. In contrast, analysis of older mice revealed an intrinsic deficit in the regenerative potential of the lung that could not be compensated by additional IGF-1 treatment.

139 α-SMA stain A Young Old

PBS B 15 PBS IGF-1 10

1 - 5

-SMA+ cells/field of view of cells/field -SMA+  +rhIGF 0 Young Old

CD31 stain C Young Old

D PBS 40 PBS IGF-1 30

1 - 10

CD31+ Endothelial cells (%) cells Endothelial CD31+ +rhIGF Young Old

Figure 4.10: Sustained administration of IGF-1 does not significantly alter α-SMA and

CD31 expressing cell numbers in the lungs of young and old mice post-PNX.

(A-B) Lung tissue sections from young adult PNX mice (2-3 months) and old PNX mice (aged 22-24 months) treated with rhIGF-1 or PBS for 22 days were stained for α-SMA to detect myofibroblasts in the lung parenchyma and (C-D) CD31 to detect endothelial cells. (C) Quantitative analysis of α-SMA+ cells and (D) CD31+ cell numbers in the lung were determined from at least 20 non-overlapping IHC images (five areas/lung lobe) using the Image J algorithm in Appendix C. Two sections at least 50µm apart were analysed per mouse. Isotype controls are shown in Appendix F. Statistical comparison between age- matched rhIGF-1 treated and PBS controls was performed using the unpaired student’s t- test, values are represented as mean±SEM, *p<0.05, n=5/group for young PNX mice, n=4/group for old PNX mice. Scale bar in all images =20µm.

140 H&E stain A Young Old

PBS

1 -

+rhIGF

** B 25 PBS IGF-1 20

10 5

Number of Alveoli /field Alveoli of Number 0 Young Old

Figure 4.11: IGF-1 supplementation enhances lung septation in young but not old pneumonectomised mice.

(A) Lung tissue sections from young adult PNX mice (2-3 months) and old PNX mice (aged 22-24 months) treated with rhIGF-1 or PBS for 22 days were stained with H&E to evaluate lung morphology and enumerate alveolar density. (B) The number of alveoli per field was counted at x600 magnification in areas devoid of major blood vessels and airways. At least 20 non-overlapping images (5 areas/lung lobe) were counted. Two sections at least 50µm apart were analysed per mouse. Statistical analysis between age- matched rhIGF-1 treated and PBS controls was performed using the unpaired student’s t- test. Values are represented as mean±SEM, *p<0.05, n=5/group for young PNX mice, n=4/group for old PNX mice. Scale bar in all images =20µm.

141 4.3- Discussion

Previous studies have shown that in response to PNX surgery, young adult mice readily regenerate their lung volume whereas old mice show a reduced capacity to regenerate lung tissue [4, 80]. However, no study has examined the dynamics of post-PNX lung growth in vivo and tested whether growth factor supplementation can rejuvenate lung growth in pneumonectomised aged mice. This chapter examined the effect of sustained exogenous IGF-

1 supplementation on longitudinal changes in lung volume using in vivo µCT imaging, as well as assessing the lung’s cellular composition following left lung PNX in both young adult mice

(aged 2-3 months) and old mice (aged 22-24 months). Sustained subcutaneous infusion of IGF-

1 using an osmotic pump did not enhance the recovery of lung volume in young mice but significantly increased the extent of lung cell proliferation and lung septation. In contrast, addition of IGF-1 in aged mice increased the aerated lung volume and transiently reduced the density of the lung but had no significant effect on the cellular characteristics of the lung after

22 days of treatment, suggesting that despite increasing lung volume, IGF-1 treatment did not lead to enhanced post-PNX tissue growth in these mice.

Consistent with previous studies and as previously reported in Chapter three of this thesis, young adult mice demonstrated a robust regenerative response post-PNX with lung volume and tissue content recovering to pre-operative levels by day 14 [3, 52, 80] (Figure 4.2-4.3).

Longitudinal evaluation of lung volume by µCT demonstrated that the majority of the lung volume recovery occurs within the first four days after surgery (Figure 4.2A). This early response is likely to be predominated by hyper-inflation of the lung rather than growth of new tissue, an observation supported by our lung tissue volume data which demonstrated that lung volume did not reach levels of the unoperated control mice until day 21 post-PNX (Figure 4.3

A). Subcutaneous (s.c.) infusion of IGF-1 using an osmotic pump did not increase lung volume

142 in young PNX mice (Figure 4.2A, C). This observation is contrary to earlier findings (Chapter three, Figure 3.14) in which we demonstrated that intraperitoneal (i.p.) delivery of a similar calculated dose of IGF-1 significantly induced higher total lung volume on day seven post-

PNX. This difference may be explained by the fact that IGF-1 delivered by s.c. osmotic pump infusion is done at a slow and controlled rate while IGF-1 delivered by i.p. has a faster rate of absorption [288]. Furthermore, given that lung volume recovery in post-PNX young adult mice is a rapid process (Figure 4.2A, C) and that IGF-1 expression peaks early (day three post-PNX,

Chapter three, Figure 3.7), it is possible that the window of opportunity for enhancing lung volume is quite narrow and that the maximum dose of IGF-1 required for regenerative lung growth may have already been supplied by endogenous sources of IGF-1. It is also worth noting that the size of the lung is constrained by the size of the thoracic cavity [289], therefore even though IGF-1 was continuously delivered in circulation, it could not increase the size of the lung beyond the maximum space allowed within the thoracic cavity.

While PNX in young mice resulted in a rapid increase in lung volume in the first four days after surgery, the opposite was observed in aged mice which demonstrated a more gradual and steady increase in lung volume post-PNX (Figure 4.2B, D). The reason for this steady increase in lung volume may be attributed to the reduced elasticity of the aged lung that makes it less compliant to mechanical forces generated by PNX [278]. It was also interesting to note that, in contrast to young adult mice, both IGF-1 treated and PBS treated old PNX mice did not restore their pre-operative total lung volume by day 21 post-PNX, consistent with earlier findings by

Paxson and colleagues [4]. Further examination of the lungs by µCT revealed that IGF-1 treatment significantly increased the air content in the lung (Figure 4.2D) and transiently reduced the volume of tissue and mean lung density between day four and day 14 post-PNX

(Figure 4.3B, D) suggesting that the IGF-1-mediated increase in total lung volume was largely driven by increased aerated lung volume as opposed to an anticipated increase in tissue volume.

143 The mechanisms by which IGF-1 mediates this effect are not clear. IGF-1 has been shown to enhance collagen and elastin synthesis in vivo and in vitro [248, 290]. Conceivably, IGF-1 could influence the elasticity of the lung enabling it to expand more than PBS treated old mice in response to PNX surgery. It would be important to determine whether this moderate but significant increase in lung volume resulting from IGF-1 treatment improves lung function in aged mice.

Having explored the effect of IGF-1 treatment on longitudinal changes in lung volume in vivo, mice were culled on day 23 to analyse effects of sustained IGF-1 infusion on components of the IGF-1 signalling pathway and changes in cellularity within the lung. Similar to changes in lung volume, we observed differential effects of IGF-1 treatment in young and aged mice. In young PNX mice, IGF-1 treatment increased the proportion of IGF-1+ cells, activated IGF-1R

(as demonstrated by increased pIGF-1R+ cell numbers) and enhanced activation of pERK, a major signalling pathway downstream of IGF-1R (Figure 4.4-4.6) [178]. Increased IGF-1R signalling likely contributed to the observed significant increase in proliferation of alveolar type-two (AECII) progenitor cells and non-AECII cell proliferation (Figure 4.9). Importantly,

IGF-1 supplementation in young PNX mice increased the number of alveoli per field, suggesting formation of new alveoli by neoseptation despite a lack of a significant increase in lung volume in these mice (Figure 4.11).

In contrast, in old PNX mice, IGF-1 treatment did not alter the number of lung cells expressing

IGF-1, pIGF-1R or pERK, suggesting a reduced sensitivity of cells within the lungs of these older animals to IGF-1 (Figure 4.4-4.6). This lack of response is unlikely to be due to altered endogenous regulation of IGF-1 signalling as the number of both IGFBP-3 and IGF-1R expressing cells were similar to those detected in the young PNX mice lungs (Figure 4.7-4.8).

Rather, it is possible that activation of the IGF-1R (pIGF-1R) in aged mice lungs may be impaired resulting in failure to respond to IGF-1 treatment. Consistent with lack of evidence

144 of IGF-1R activation in response to IGF-1 treatment, aged lungs did not undergo significant cell proliferation and alveolarisation, although it must be noted the proportion of AECII progenitors was similar in young and aged mice suggesting that AECIIs are present in the aged lung but do not undergo significant proliferation (Figure 4.9). This data is congruent with findings by Paxson and others [4] who showed that proliferation of non-AECIIs but not AECIIs was diminished in 24-month old PNX mice. This finding also highlights a potential complexity one may have to overcome when considering cell-based therapies in aged individuals. It may not be sufficient to supply stem cells or growth factor alone; there is also a need to consider means of promoting growth factor signalling in aged lung cells to promote integration of exogenous cells into the local environment and stimulate growth and differentiation of resident cells.

At present, we do not understand the mechanism(s) that explains failure of exogenous IGF-1 to stimulate IGF-1R signalling in aged mice lungs. It could be that the inability of aged lungs to expand early after PNX (Figure 4.2B, D), diminishes their ability to respond to IGF-1 treatment. Lack of stretch may diminish the ability of IGFBP proteases such as matrix metalloproteinase (MMP)-1-to-3 and MMP-7 [291, 292], that breakdown IGF-1-IGFBP bonds to increase availability of IGF-1 to the cell. Interestingly, Liu et.al [293] demonstrated that stretch promotes MMP-2 activity and increased MAPK and JNK signalling. It may also be that the aged matrix has a diminished capacity for binding IGF-1 supplied from circulation [249].

The importance of fibroblasts and the extracellular matrix (ECM) in lung regeneration post-

PNX has also been suggested in two studies by Paxson and colleagues examining post-PNX lung growth in old mice [4, 287]. Fibroblasts synthesise elements of the ECM such as elastin and collagen which transmit mechanical forces resulting from PNX. These studies showed that with age, fibroblasts are less proliferative, biased towards a senescent myofibroblastic signature

(a trend replicated in this study, Figure 4.10A-B) and thus have a diminished capacity to

145 synthesise ECM elements. Therefore, this results in an age-related loss of ECM elasticity and diminished capacity to transmit mechanical forces necessary for lung growth post-PNX.

Alternatively, lack of IGF-1R signalling and the associated lack of a regenerative response in aged mice lungs may suggest that supplementation with IGF-1 will not be able to overcome phenotypes that are associated with the aging lung including increased cell turnover, cellular senescence, DNA damage, oxidative stress, increased apoptosis and loss of stem cells and stem cell niche components, changes that progressively result in diminished lung function [280].

These findings also highlight the complexity of rhIGF-1 treatment in aged lungs. It may not be as simple as supplying extra IGF-1 but rather other extra signals, for example increased mechanical force or enhanced activity of early response mechanosensitive transcription factors such as EGR-1 [87] may be required for successful growth factor signalling and improved regenerative lung growth.

Nonetheless, at this stage, we cannot dismiss the ability of IGF-1 to influence growth of aged lung cells. Firstly, we have only tested a single dose of IGF-1. It is possible that higher doses might have a more potent effect. As indicated by our studies in young post-PNX mice, the route of administration of IGF-1 may influence lung growth and regeneration. Taken together these data suggest that while IGF-1 enhances lung cell proliferation and alveolarisation in young

PNX mice without affecting overall lung volume, IGF-1 in old PNX mice has no overt effect on lung tissue volume and cellular composition. We conclude from these observations that supplemental IGF-1 can play a role in regeneration of the lung in our model by modulating post-natal lung alveolarisation in young mice but not old mice.

146

Chapter 5

Investigating EGR-1 and HIF-1α in post-PNX lung growth

147 5.1- Introduction

The regenerative growth response in the residual lung following left lung PNX is a complex process, influenced by many factors including mechanical stress, growth factors and transcription factors [46, 62] (reviewed in Chapter One, Section 1.7). Increased cardiac and respiratory load experienced by the remaining lung following PNX surgery results in increased mechanical stress that is believed to be one of the most important factors initiating regenerative lung growth [42]. Increased airflow stretches the alveolar parenchyma initiating signals associated with compensatory lung growth such as DNA synthesis, protein synthesis and cell proliferation [42]. Similarly, increased blood flow following PNX stretches the pulmonary vasculature and increases endothelial cell permeability, resulting in proliferative signals that promote vascular and alveolar proliferation [67]. Limiting expansion of the lung post-PNX through the insertion of a space-occupying plombage in the thorax, retards lung growth, signifying the importance of mechanical stretch in post-PNX lung growth [88, 294].

The mechanisms that link mechano-transduction to lung cell proliferation post-PNX have not been well characterised, however a number of studies support the prominent role for two immediate and early transcription factors in post-PNX lung growth, namely Early Growth

Response protein 1 (EGR-1) and Hypoxia-Inducible Factor 1α (HIF-1α) (Chapter One,

Section 1.7.4). Microarray analysis of PNX lungs within the first 24 hrs post-PNX has revealed a transient four-fold upregulation of EGR-1 expression at 2 hrs post-PNX [87]. Additionally, some studies have shown that post-PNX lung growth is associated with increased expression of HIF-1α and its principle downstream target VEGF has been shown to be critical in endothelial cell proliferation and alveolarisation post-PNX [72, 74, 151]. Interestingly, EGR-1 knockout mice develop normally, although these mice have reduced regenerative growth in the liver following partial hepatectomy [295], whilst HIF-1α knockout is embryonically lethal

148 [296]. However, modulation of either of these transcription factors has not previously been investigated in the PNX model.

In Chapter three, we demonstrated that PNX-induced lung growth is associated with increased expression of IGF-1 and that exogenous administration of IGF-1 enhances post-PNX lung growth in young adult mice, whilst blocking IGF-1R signalling attenuates this response. The mechanisms that drive the involvement of IGF-1 during post-PNX lung growth are yet to be determined. We hypothesised that PNX-induced stretch activates the mechano-sensitive transcription factors EGR-1 and HIF-1α which contribute to IGF-1 expression and IGF-1 signalling (Figure 5.1).

Neoalveolarisation post-PNX involves formation of new alveolar septa within existing intra- alveolar spaces. This process is guided by α-SMA+ myofibroblasts that deposit ECM components including elastin and collagen at the septal tips to support growing septa [297].

Proliferation and migration of fibroblasts is therefore essential for neoalveolarisation. Studies have shown that platelet-derived growth factor (PDGF)-AA and IGF-1, produced by the alveolar epithelium, are an important source of trophic factors for fibroblasts’ neoseptation function [297]. Therefore, a fibroblast cell line was used to study the effect of IGF-1 and stretch on cultured fibroblasts.

The specific aims of this chapter were:

i) To examine the contribution of EGR-1 and HIF-1α activity to IGF-1-induced lung

regeneration.

ii) To determine effects of IGF-1 supplementation on post-PNX lung growth in the

presence or absence of pharmacological inhibitors of HIF-1α or EGR-1.

149

PNX

Stretch Acriflavine Mithramycin-A

HIF-1α EGR-1

IGF-1

pIGF-1R BMS-536294 pERK pAkt

Cell differentiation, proliferation and migration

Figure 5.1: Proposed mechanism driving the IGF-1 signalling pathway post-PNX.

Following PNX surgery, mechanical tension is created in the remaining lung activating many intracellular signalling pathways that lead to cell proliferation, differentiation and migration to promote lung growth. IGF-1 acting through the IGF-1R activates the pERK and pAkt signalling pathways. However, the mechanisms that link IGF-1 to the initiating stretch stimulus remain unknown. We proposed that activation of immediate early response transcription factors EGR-1 and HIF-1α in response to stretch contributes to IGF-1 expression. To demonstrate the link between stretch activated EGR-1, HIF-1α and IGF-1, cells were treated with the pharmacological inhibitors mithramycin-A (EGR-1 inhibitor) and acriflavine (HIF-1α inhibitor). Furthermore, by blocking IGF-1R phosphorylation using BMS-536294, the role of IGF-1R signalling in cell proliferation was determined in vitro.

150 5.2- Results

5.2.1- Stretch induces Egr-1, Igf-1r and Igfbp-3 mRNA expression in cultured lung fibroblasts

To determine the role of EGR-1 and HIF-1α in post-PNX lung growth, we first performed proof of concept studies using lung fibroblasts to demonstrate a mechanistic link between stretch, EGR-1, HIF-1α and IGF-1. As discussed above (Section 5.1), mechanical stretch is a primary driver of cellular responses associated with regenerative lung growth after PNX [42].

To model stretch in vitro, we employed a stretch device that manually applied static homogenous equibiaxial stretch of up to 10% to 3T3 mouse embryo fibroblasts cultured on an elastic membrane [242]. Cells were stretched for 24 hrs to assess whether IGF-1 is secreted in culture and assess changes in mRNA expression of the IGF-1 signalling pathway (Igf-1, Igf-1r and Igfbp-3) as well as Egr-1 and Hif-1α, two transcription factors essential for mediating the cellular response to stretch [298, 299]. Egr-1 mRNA was significantly upregulated in stretched fibroblasts (~2.9-fold increase, p<0.001, n=3, Figure 5.2) whereas Hif-1α mRNA expression was not different between unstretched and stretched 3T3 fibroblasts. Neither Igf-1 mRNA nor

IGF-1 protein in supernatants was detected in this cell line. Interestingly, Igfbp-3 and Igf-1r mRNA expression was significantly up-regulated in stretched fibroblasts (~1.3-fold increase, p<0.01; ~1.1-fold increase, p<0.05, respectively, n=3, Figure 5.2).

151

0.8 0.7 Unstretched * 0.6 Stretched 0.5

Pgk-1 0.4 0.3 **

0.2

relative to relative *** expression mRNA 0.1 0.0 EGR-1 HIF-1 IGFBP-3 IGF1R

Figure 5.2: Stretch induces mRNA expression of Egr-1, Igfbp-3 and Igf-1r.

Comparison of Egr-1, Hif-1α, Igfbp-3 and Igf-1r mRNA expression between stretched and unstretched 3T3 cultured lung fibroblasts. Stretched fibroblasts were subjected to 10% homogenous equibiaxial mechanical stretch using a custom-made device for 24 hrs prior to RNA harvest and subsequent analysis by quantitative real-time PCR relative to the house- keeping gene pgk-1. *p<0.05, **p<0.01, ***p<0.001, Data is representative of two independent experiments with three replicates per condition.

152 5.2.2– IGF-1 signalling and fibroblast proliferation in vitro can be effectively modulated by the IGF-1R antagonist, BMS-536294

In Chapter three, we demonstrated that treatment of PNX mice with the IGF-1R antagonist

BMS-536294 delayed regenerative lung growth. To better understand the mechanisms that drive IGF-1 signalling in post-PNX lung growth, we evaluated effectiveness of BMS-536294 in inhibiting IGF-1-induced signalling and cell proliferation in vitro using 3T3 fibroblasts. It is well established that binding of the IGF-1 ligand to the IGF-1R leads to phosphorylation of its cytoplasmic tyrosine residues and consequently activation of downstream signalling pathways most notably the pERK and pAkt pathways [285] (Figures 1.9 and 5.1). Here we demonstrate that BMS-536294 treatment attenuated IGF-1-induced IGF-1R phosphorylation and subsequent downstream pERK activation 15 minutes after IGF-1 stimulation (Figure 5.3).

Importantly, treatment of serum-starved fibroblasts with IGF-1 led to a 5.3-fold increase in cell proliferation at 48 hours post-stimulation (p<0.001, n=6, Figure 5.4). In the presence of BMS-

536294, IGF-1-induced cell proliferation was significantly reduced in a dose-dependent manner (Figure 5.4).

5.2.3- IGF-1 induces EGR-1 and HIF-1α protein expression in cultured lung fibroblasts

To further examine the link between EGR-1, HIF-1α and IGF-1, we studied the kinetics of

EGR-1 and HIF-1α protein expression in lung fibroblasts stimulated with IGF-1 over an eight- hour time course. EGR-1 protein was induced early, 30 minutes after IGF-1 treatment. Its expression remained elevated at one hour but declined after two hours and remained undetected throughout the rest of the time course. In contrast, HIF-α protein expression was not expressed until after four hours of stimulation and was maintained at eight hours post-stimulation (Figure

5.5).

153

rhIGF-1: - - + + IGF-1R antagonist: - + - +

pIGF-1R 95 kDa

IGF-1R 95 kDa

pERK-1/2 41/42 kDa

ERK-2 42 kDa

α-tubulin 50 kDa

Figure 5.3: The IGF-1R antagonist BMS-536924 attenuates activation of the IGF-1 signalling pathway.

Following overnight serum starvation, 3T3 fibroblasts grown to confluency were stimulated with rhIGF-1 (100ng/mL) for 15 minutes or left unstimulated in the presence or absence of 1μM of BMS-536924, an IGF-1R antagonist which was pre-incubated for 1hr prior to addition of rhIGF-1. Protein lysates were harvested to probe for pIGF-1R, IGF-1R, pERK- 1/2, ERK-2 and α-tubulin loading control. IGF-1 induced pIGF-1R and pERK expression which was decreased in the presence of BMS-536924. Data is representative of one experiment with three biological replicates.

154

2 ns *** 1 ** 0.20 *** *** 0.15

0.10

0.05 Absorbance at 650nm Absorbance 0.00 1 2 5 0.1 0.2 0.5

Vehicle 10% FCS IGF-1R antagnonist (M) IGF-1 only IGF-1 (100ng/mL) IGF-1R antg. only

Figure 5.4: IGF-1R inhibition reduces IGF-1-mediated fibroblast proliferation.

3T3 fibroblast cell proliferation was measured by methylene blue assay at 48hrs post- stimulation with either 10% FCS, vehicle control (0.1% DMSO), rhIGF-1 only (100ng/mL), IGF-1R antagonist (BMS-536294, 0.1µM) only, or 100ng/mL rhIGF-1 with increasing concentrations of IGF-1R antagonist (0.1-5µM BMS-536294), following overnight serum starvation of subconfluent cells. rhIGF-1 induced a five-fold increase in cell density relative to vehicle control, which was gradually attenuated with increasing concentrations of the IGF-1R antagonist. **p<0.01, ***p<0.001, ns=not significant, values represented as mean±SEM. These data are representative of three independent experiments with 6 replicates per condition.

155

+rhIGF-1 (100ng/mL)

0 15’ 30’ 1hr 2hr 4hr 8hr

EGR-1 60 kDa

100 kDa HIF-1α 75 kDa

α-tubulin 50 kDa

Figure 5.5: The kinetics of IGF-1-induced EGR-1 and HIF-1α expression in 3T3 fibroblasts.

3T3 fibroblasts were treated with 100 ng/mL of rhIGF-1 for 15 min, 30 min, 1hr, 2hrs, 4hrs and 8hrs or left unstimulated and the relative expression of. EGR-1 and HIF-1α determined by western blot. Protein expression levels were normalised to the loading control α-tubulin. EGR-1 was activated early within 30 min post-stimulation and its expression detected until 2hrs post-stimulation. HIF-1α expression was not detected until 4hrs after stimulation. Blots shown are from one experiment.

156 5.2.4- Pneumonectomy induces EGR-1 and HIF-1α expression in the lung

Although prior studies have suggested a role for EGR-1 and HIF-1α in regenerative lung growth post-PNX [74, 87], the precise mechanism by which these transcription factors contribute to this process has not been described. We hypothesised that activation of EGR-1 and HIF-1α contributes to IGF-1-mediated post-PNX lung growth. To explore this hypothesis in vivo, we first longitudinally analysed expression of EGR-1 and HIF-1α in post-PNX lung tissue compared to sham operated control lung tissue on days 0, 1, 3 and 7 post-surgery, by immunohistochemistry. EGR-1 and HIF-1α was detected at all time points in both sham and

PNX lung tissue and expressed by cells located the lung parenchyma and airways. In response to PNX surgery, EGR-1 and HIF-1α staining was more prominent around the subpleural edges of the lung (Figure 5.6A). In unoperated and sham operated control lung tissue, the percentage of EGR-1 and HIF-1α positive was ~5% at all the time points examined (Figure 5.6B, C).

EGR-1 and HIF-1α expression increased at day one post-PNX with peak expression occurring on day three post-PNX (~2.9-fold increase for EGR-1, p<0.05; ~4.7-fold increase for HIF-1α, p<0.001, n=3/group, Figure 5.6B, C). By day seven, EGR-1 and HIF-1α levels had returned to baseline levels (Figure 5.6B, C).

157 A EGR-1 HIF-1α

D3 SHAM

D3 PNX

B C 25 Unp. Cont. *** 20 SHAM *** PNX 15

+ve cells cells (%)+ve 10 

5 HIF-1 0 0 1 3 7 Days after surgery

Figure 5.6: PNX induces a transient increase in EGR-1 and HIF-1α expression in mouse lung tissue. (A) Representative images of EGR-1 and HIF-1α IHC stains taken from lung sections of PNX and left sham thoracotomised (SHAM) control mice on post-operative day three. Scale bar

=100µm. (B) IHC quantification of EGR-1 and (C) HIF-1α staining of lung sections from unoperated (unp.) control mice, PNX and SHAM mice on day one, three and seven post- surgery was performed by sampling at least 20 non-overlapping images (5/lobe) and quantifying positive cells using the Image J algorithm in Appendix C. EGR-1 and HIF-1α expression in PNX lungs gradually increased peaking at day three then significantly declined approaching baseline levels by seven. EGR-1 and HIF-1α expression levels in SHAM lungs were no different to that of unoperated control lungs and remained unchanged. *p<0.05, ***p<0.001, values represented as mean ± SEM. n=3/group.

158 5.2.5- Pharmacological inhibition of EGR-1 or HIF-1α did not significantly alter lung weight but increased lung volume at day 15 post-PNX

Using the pharmacological inhibitors mithramycin-A (inhibits EGR-1) or acriflavine (HIF-1α inhibitor), we determined the effect of EGR-1 and HIF-1α inhibition on lung volume and lung weight following PNX, as depicted in Figure 5.7A. There was no significant difference in lung wet weight measurements between acriflavine, mithramycin-A and PBS treated control PNX mice (Figure 5.7B). However, measurement of lung volume by the water displacement method revealed that pharmacological inhibition of EGR-1 and HIF-1α significantly increased lung volume compared to PBS treated PNX mice (335.6 ±20.38 mm3 for PBS controls vs 424.3

±10.14 mm3 and 429.1 ±26.82 mm3 for EGR-1 and HIF-1α antagonist treated mice respectively, p<0.05, n=3/group, Figure 5.7C). Collectively, these results suggested that PNX mice treated with either mithramycin-A or acriflavine exhibited lower lung density compared to PBS control PNX mice. Furthermore, morphometric analysis of hematoxylin stained sections revealed that inhibition of EGR-1 but not HIF-1α in PNX mice expanded the diameter of alveoli compared to PBS treated PNX controls (22.17 ±0.63µm for PBS vs 25.53 ±0.63µm for mithramycin, p<0.001, n=3/group, Figure 5.8) suggesting a shift towards an emphysematous phenotype in these mice.

159

B 6.0

5.5

5.0

4.5 Right lung: Right 4.0

(mg/g) weight Body 3.5 Saline Acriflavine Mithramycin

C 500 *

ns )

3 450

400

350

300

by water displacement water by Total lung volume (mm volume lung Total 250 Saline Acriflavine Mithramycin

Figure 5.7: Pharmacological inhibition of EGR-1 and HIF-1α did not alter lung mass but significantly increased lung volume at day 15 post-PNX.

(A) Treatment regimen of PNX mice to test effect of pharmacological inhibition of EGR-1 (using mithramycin-A) and HIF-1α (using acriflavine). (B) Ratio of lung wet weight to body weight measured after 14 days of drug or PBS treatment, each dot represents an individual animal. (C) Lungs were inflated with 4% paraformaldehyde and their volume estimated by the water displacement method. *p<0.05, ns=non-significant.

160

A- PBS B- Acriflavine

C- Mithramycin D

Figure 5.8: Inhibition of EGR-1 increases alveolar diameter in PNX mice.

(A-C) Representative images of hematoxylin stained lung sections from mice treated with acriflavine, a HIF-1α antagonist, mithramycin-A, an EGR-1 antagonist or PBS control solution for 15 days post-PNX. (D) Quantification plot comparing mean alveolar diameter among the three treatment groups. Mithramycin-A but not acriflavine treatment significantly increased the mean alveolar diameter compared to PBS controls. Values represented as mean±SEM, n=3/group, ***p<0.001, scale bar= 30µm.

161 5.2.6- Sustained inhibition of HIF-1α does not alter lung volume but significantly reduces lung cell proliferation at day seven post-PNX.

Given that both IGF-1 and PNX induced HIF-1α expression (Figures 5.5 and 5.6, respectively), we next performed experiments to examine the effect of in vivo inhibition of

HIF-1α on lung regeneration post-PNX in the presence or absence of IGF-1. The treatment protocol, similar to that depicted in Figure 5.7A, involved single daily intraperitoneal injections with either PBS alone, acriflavine (HIF-1α antagonist), rhIGF-1 alone or acriflavine in combination with rhIGF-1 for seven days post-PNX. On day seven, mice lungs were imaged by µCT to determine tissue volume and total lung volume (which included air plus tissue volume) before sacrifice for immunohistochemical analysis of Ki-67+ proliferating cells. Mice were treated for seven days because this coincides with peak lung cell proliferation in untreated

PNX mice (Chapter three, Figure 3.3).

There was no statistically significant difference in lung volume between PBS and acriflavine treated mice on day seven after PNX surgery (358.6 ±1.41mm3 for PBS vs 342.1 ±33.9mm3 for acriflavine, n=3/group, Figures 5.9A, 3.15B). Similarly, there was no statistically significant difference in tissue volume between the two groups (12.63 ±0.54mm3 vs 11.1

±1.9mm3 respectively, n=3/group) although there appeared to be a trend towards reduced tissue volume in acriflavine treated PNX mice (Figures 5.9B, 3.15C). Addition of rhIGF-1 in acriflavine treated PNX mice significantly increased total lung volume by 24.4 ±3.4% (p<0.05, n=4, Figures 5.9A) and tissue volume by 121.3 ±19.4% (p<0.01, n=4, Figure 5.9B) compared to acriflavine treated PNX mice. Supplementation with rhIGF-1 alone increased lung and tissue volume by a similar magnitude (Figure 5.9B) indicating that inhibition of HIF-1α could not inhibit the pro-regenerative growth induced by IGF-1 treatment of PNX treated mice.

162 A

500 **

) 3 m *

(

m 400 u

g n

u 300

l a

o T 200 7 7 7 7 D D D D e 1 S n -1 - B i F F P v G la IG I f + ri . c lv A rf c A B **

) 30 3

(mmTissue volume 0

PBS D7 IGF-1 D7

Acriflavine D7 Acrflv. +IGF-1 D7

Figure 5.9: Inhibition of HIF-1α has no effect on total lung volume and lung tissue volume in the presence or absence of IGF-1 at day seven post PNX.

(A) Comparison of total lung volume and (B) tissue volume on post-PNX day seven, measured by in vivo µCT, in mice administered with once daily i.p injections of either PBS, acriflavine (a HIF-1α antagonist, 2mg/kg/day), rhIGF-1 (0.2mg/kg/day) or rhIGF-1 combined with acriflavine for seven consecutive days after PNX surgery. Acriflavine treatment did not alter total lung volume or tissue content compared to PBS controls. IGF-

1 alone or in combination with acriflavine significantly increased both total lung and tissue volume. PBS and IGF-1 alone data are the same as that represented in Figure 3.15. Values represented as mean ± SEM, *p<0.05, **p<0.01, PBS and acriflavine treatment groups n=3, IGF-1 only n=5, IGF-1+Acriflavine n=4.

163 Next, lung sections from IGF-1 and acriflavine treated PNX mice were stained for Ki-67 to determine the effect of HIF-1α inhibition on cell proliferation in vivo. Acriflavine treatment alone reduced cell proliferation by 53.2 ±5.7% with respect to PBS control PNX mice (p<0.05, n=3/group, Figures 5.10, 3.15D, E). Conversely, rhIGF-1 treatment increased cell proliferation by 72.8 ±16.7% relative to PBS control PNX mice (p<0.001, n=3 for PBS, n=5 for IGF-1,

Figures 5.10, 3.15D, E). In the presence of acriflavine, IGF-1-induced increase in cell proliferation was significantly reduced (IGF-1 alone=19.53 ±1.9% vs Acriflavine+ IGF-1=

9.52 ±1.53%, p<0.01, n=5 and 4 respectively, Figures 5.10, 3.15D, E). However, it was interesting to note that IGF-1 effectively restored levels of cell proliferation to PBS levels in acriflavine treated PNX mice (Figures 5.10, 3.15D, E). Collectively, this data suggested that

IGF-1 treatment was able to counteract the anti-proliferative effect of acriflavine treatment in post-PNX lung tissue.

164

Figure 5.10: Lung cell proliferation at day seven post-PNX is significantly inhibited by HIF-1α inhibition but restored to baseline by IGF-1 co-treatment.

(A) Representative images of Ki-67 stained lung tissue sections from mice treated with

PBS, rhIGF-1, Acriflavine or rhIGF-1 plus acriflavine for seven days after PNX surgery. Ki67 (green), nuclei (blue). (B) The number of Ki67+ nuclei were quantified in 20 non- overlapping images at x400 magnification using the Image J macro in Appendix C. PBS and IGF-1 alone data are the same as that represented in Figure 3.15. Values represented as mean±SEM, *p<0.05, **p<0.01, PBS and Acriflavine n=3/group, IGF-1 only n=5, IGF- 1+Acriflavine n=4. Scale bar = 100µm.

165 5.2.7- Pharmacological inhibition of EGR-1 or HIF-1α alters distribution of CD31+ cells in the lung parenchyma post-PNX.

It was observed that following lung fixation, acriflavine (anti-HIF-1α) and mithramycin-A

(anti-EGR-1) treated PNX mice lungs appeared paler compared to PBS control PNX mice suggestive of a disturbance of the vasculature (Figure 5.11A). Therefore drug-associated effects on the lung vasculature were examined by immunohistochemical staining of lung sections for CD31, a marker for endothelial cells. Lung sections examined from acriflavine and mithramycin-A treated PNX mice displayed an altered cellular morphology with enumeration of CD31+ cells by quantitative immunohistochemistry indicating a significant reduction in the percentage of CD31+ cells in the lung parenchyma (11.87 ±1.6% for PBS vs 6.18 ±0.7% for acriflavine, p<0.01 and 7.95 ±0.93% for mithramycin-A, p<0.05, n=3/group, Figure 5.11C).

This finding was further corroborated by western blot analysis for CD31 protein which revealed a similar reduction in CD31 expression in acriflavine and mithramycin-A treated PNX mice compared to PBS control PNX mice (PBS vs acriflavine ~1.7-fold decrease, p<0.01; PBS vs mithramycin, ~1.7-fold decrease, p<0.05, n=3/group, Figure 5.12A-B).

166

PBS Acriflavine PBS Mithramycin

B Nuclei CD31 Merged

PBS

Acriflavine

Mithramycin

C * 15 **

5 CD31+ cells (%)

0 PBS Acriflavine Mithramycin

Figure 5.11: Inhibition of EGR-1 or HIF-1α reduces the percentage of CD31+ cells in post-PNX lung tissue.

(A) Physical appearance of 4% PFA inflated PNX lungs from mice that had been treated either with the EGR-1 antagonist, mithramycin-A (150µg/kg/day), acriflavine a HIF-1α antagonist (2mg/kg/day) or PBS control solution for 15 days. (B) Representative CD31- stained lung tissue sections from PBS, mithramycin-A and acriflavine treated mice at day 15 post-PNX. CD31 is shown in red) and nuclear staining in blue. Scale bar= 50µm. (C) IHC quantification of CD31+ cells was performed by capturing at least 20-non-overlapping random images and analysed using the Image J algorithm in Appendix C. All values represented as mean±SEM, *p<0.05, **p<0.01, n=3/group.

167

A PBS Acriflavine

CD31 100 kDa

α-tubulin 50 kDa

1.5 **

1.0 -tubulin ratio -tubulin

 0.5

CD-31/ 0.0 PBS Acriflavine

B PBS Mithramycin

CD31 100 kDa

α-tubulin 50 kDa

1.0 *

0.8

0.6

-tubulin ratio -tubulin 0.4  0.2

CD-31/ 0.0 PBS MIthramycin

Figure 5.12: Reduction of CD31 protein levels in PNX mice treated with acriflavine or mithramycin.

(A) Comparison of CD31 protein levels by western blot between acriflavine and PBS treated PNX mice and (B) mithramycin-A and PBS treated PNX mice on day 15 post-PNX. Semi-quantitative densitometry was performed and CD31 levels expressed as a ratio of the loading control α-tubulin. All values represented as mean ± SEM, *p<0.05, **p<0.01.

168 5.3- Discussion

The increased mechanical tension observed following lung resection is essential for post-PNX lung growth as evidenced by plombage studies [88, 294]. Data presented in Chapter three and

Chapter four of this thesis has shown that IGF-1 production and IGF-1 signalling plays an important role in post-PNX lung growth. The precise mechanism by which mechanical force induces IGF-1 expression and cell proliferation is not completely understood. This chapter began to explore the underlying mechanisms regulating IGF-1-mediated post-PNX lung expansion with the hypothesis that the transcription factors EGR-1 and HIF-1 contribute to

IGF-1 activity post-PNX (Figure 5.1).

The results described in this chapter demonstrated that stretch induced mRNA expression of

EGR-1, IGF-1R and IGFBP-3 in cultured fibroblasts, whilst IGF-1 could induce EGR-1 and

HIF-1 protein expression in a time-dependent manner. Pharmacological inhibition studies showed that crosstalk between EGR-1, HIF-1α and IGF-1R is essential for IGF-1-mediated cell proliferation post-PNX. Furthermore, a potential role for EGR-1 and HIF-1α in promoting vascularisation post-PNX was demonstrated.

Blocking IGF-1R signalling impaired fibroblast proliferation consistent with previous studies that have demonstrated a crucial role for IGF-1 in wound healing and repair (Figure 5.3-5.4)

[300, 301]. Given the significance of IGF-1 in neoseptation and regenerative lung growth

[297], the in vitro data presented here further supports in vivo data presented in Chapter three

(Figure 3.16) indicating that a reduction in cell proliferation in part accounts for the observed reduction in tissue volume and total lung volume mediated via the IGF-1R. Demonstrating similar findings in a murine epithelial cell line would supplement the validity of this in vitro assay.

169 Having illustrated the significance of IGF-1 in promoting fibroblast proliferation, I next wanted to determine how mechanical stretch activates IGF-1 signalling using an equibiaxial stretch device that applied static homogenous stretch to lung fibroblasts grown on an elastic membrane

[242]. Secretion of endocrine IGF-1 from the liver in response to GH is known to be regulated by STAT-5b [302]. However, the liver is not the only source of IGF-1; several extrahepatic tissues including the lung secrete local IGF-1 that acts via autocrine and paracrine signalling to influence various cellular processes including proliferation, growth and survival. The mechanisms that regulate local expression of IGF-1 are less well defined. I proposed that the transcription factors EGR-1 and HIF-1α regulate stretch-induced activation of IGF-1 based on data obtained from previous studies that have identified induction of these transcription factors in post-PNX lungs [74, 87]. In the present study, stretch induced Egr-1 mRNA (Figures 5.2), similar to previous studies [303], [304], [305]. Additionally, I showed that IGF-1 induced

EGR-1 protein expression in unstretched 3T3 fibroblasts as early as 30 minutes post- stimulation (Figure 5.5), consistent with findings by Youreva et.al [306] who demonstrated a similar phenomenon in vascular smooth muscle cells. However, I was unable to demonstrate the combined effect of stretch and IGF-1 stimulation.

In both stretched and unstretched 3T3 fibroblasts, Igf-1 mRNA was not detected nor was IGF-

1 protein detected in cell lysates and conditioned medium from this cell line suggesting that fibroblasts may not be a prominent source of this growth factor. Contrary to these findings, skeletal muscle fibroblasts [307] and ventricular fibroblasts [308] have been reported to produce IGF-1 in response to stretch, although it must be noted that in the latter study, cyclic stretch but not static stretch resulted in IGF-1 secretion. Thus, it is likely that the ability of fibroblasts to secrete IGF-1 may not only be organ dependent but also dependent on the intensity, frequency and duration of the stretch stimulus.

170 Interestingly, stretch induced significant expression of Igf-1r and Igfbp-3 mRNA (Figure 5.2) in cultured lung fibroblasts. This may suggest that although lung fibroblasts may not be a prominent source of IGF-1, in response to stretch they become primed for paracrine IGF-1R signalling by increasing IGF-1R and IGFBP-3 expression to enhance their affinity for IGF-1.

Although I was unable to demonstrate the link between stretch-induced EGR-1 and IGF-1R signalling, Wu et.al. [303] showed that stretch-induced EGR-1 stimulated IGF-1R transcription in venous smooth muscle cells and that EGR-1 knockout mice have reduced IGF-1R expression. Similar findings were reported by Ma et.a.l [309] in prostate cancer cells. These findings raise the possibility that the IGF-1 ligand may not be a direct target of EGR-1 but rather EGR-1 only serves to promote activation of IGF-1 signalling in response to stretch.

The mechanism by which mechanical stretch activates EGR-1 transcription is not well understood. Some of the mechanisms suggested include activation of G-protein coupled receptors and changes in intracellular calcium fluxes that activate the Akt, ERK and JNK pathways [303, 308]. A recent study by Liu et.al. [88] suggested that PNX-induced mechanical stretch induces F-actin polymerisation that activates Cdc42, which in turn activates YAP-TAZ transcription factors via the JNK-MAPK pathway. However, YAP-TAZ require transcriptional co-activators as they lack DNA binding activity [310]. EGR-1 has been shown to complex with

YAP in irradiated cancer cells [311]. It would be interesting to determine whether PNX- induced mechanical tension drives formation of YAP-TAZ-EGR-1 complexes.

I also investigated whether HIF-1α can regulate IGF-1 activity in response to stretch. No significant difference in Hif-1α mRNA was detected between stretched and unstretched 3T3 fibroblasts (Figure 5.2), although other studies have shown that mechanical stretch can induce

Hif-1α mRNA and HIF-1α protein [99, 299]. Failure to detect HIF-1α may be due to inappropriate timing and duration of the stretch stimulus. However, I showed that IGF-1 treatment induced HIF-1α protein expression in unstretched cultured 3T3 fibroblasts, four

171 hours after stimulation (Figure 5.5) similar to kinetics reported by Treins et.al. [100] who further demonstrated that activation of HIF-1α by IGF-1 leads to downstream activation of the principle HIF-1α target, VEGF, which is crucial for endothelial cell proliferation. Importantly,

I showed that in post-PNX lung tissue, HIF-1α as well as EGR-1 protein expression was significantly upregulated between day one and day three after surgery (Figure 5.6). The interactions between EGR-1, HIF-1α and IGF-1 are not uni-directional and there is evidence suggesting that reciprocal feedback loops exist [100, 298]. Therefore, it is possible that IGF-1 produced after PNX induces EGR-1 and HIF-1α, which helps sustain IGF-1 activity.

I next explored the importance of EGR-1 and HIF-1α in post-PNX lung growth by blocking their activity using mithramycin-A and acriflavine. In acriflavine and mithramycin-A treated

PNX mice, total lung volume was significantly increased with no difference in lung weight observed between PBS-treated control PNX mice and drug treated PNX mice after 14 days of treatment (Figure 5.7). In this experiment, total lung volume was measured by the water displacement method as the µCT instrument was non-operational. This method involves inflation of the lung with a fixative to estimate its volume thus the larger lung volumes recorded in acriflavine and mithramycin-A treated PNX mice could indicate that the lungs of these mice were relatively more compliant and thus take up more fixative. This may explain the discrepancies in the effect of acriflavine on lung volume shown in Figure 5.7 (where the water displacement method was used) and Figure 5.9 (where end of-exhalation lung volume was measured by micro-CT). It could also indicate that effects of acriflavine on lung tissue, for example a reduction in cell proliferation and increase in lung compliance are largely evident after day seven post-surgery.

Furthermore, morphometric analysis of these lungs revealed increased airspace enlargement in

EGR-1 inhibited PNX mice (Figure 5.8). A study by Rauschkolb and colleagues [312] reported similar findings in EGR-1 knockout mice which exhibited larger lung volume and airspace

172 enlargement. HIF-1α knockdown has also been reported to cause airspace enlargement [313].

However, this was not witnessed in the present study, perhaps suggesting that acriflavine, compared to mithramycin has a less potent effect on reducing cell proliferation and/or matrix synthesis following PNX. (Figure 5.8).

An unexpected effect resulting from pharmacological inhibition of EGR-1 and HIF-1α in PNX lungs was the significant reduction in CD31+ capillary endothelial cells, that likely contributed to the pale physical appearance of these lungs ex vivo (Figures 5.11 and 5.12). The role of HIF-

1α in neovascularisation has been extensively characterised. HIF-1α has several target genes including erythropoietin receptor (EPO-R) and VEGF which are involved in endothelial cell proliferation, survival, angiogenesis and vascular remodelling [100, 313]. HIF-1α is critical for lung development and is elevated during post-PNX lung growth [74]. VEGF supplementation enhances post-PNX lung growth [151, 164]. In an experimental model of bronchopulmonary dysplasia, inhibition of HIF-1α using a dominant negative adenovirus and the drug chetomin, which works similarly to acriflavine, impaired vascular development and caused airspace enlargement [313]. To our knowledge, this is the first study to demonstrate the effect of HIF-

1α inhibition in the PNX model. Furthermore, we illustrated for the first time that IGF-1 could restore lung volume and cell proliferation in acriflavine (HIF-1α inhibited) treated PNX mice

(Figures 5.9 and 5.10). Given that IGF-1 is a transcriptional target of HIF-1α [100] and that inhibition of HIF-1α diminishes cell proliferation post-PNX which can partially be restored by

IGF-1 supplementation (Figure 5.10), it is reasonable to speculate that induction of IGF-1 by

HIF-1α probably makes a significant contribution to post-PNX lung growth and vascularisation.

EGR-1 activity has also been shown to modulate vascularisation although its role is less well appreciated [85, 298]. It was therefore surprising to note that the reduction in distribution of

CD31+ endothelial cells resulting from EGR-1 inhibition was similar to that seen in HIF-1α

173 inhibited PNX mice (Figures 5.11 and 5.12). Mechanical strain induces EGR-1 and proangiogenic factors such as PDGF-AA and fibroblast growth factor 2 (FGF2) in endothelial cells [85, 298]. Therefore, blocking EGR-1 when cells are experiencing an enhanced stretch force, can result in severe consequences on the vasculature as indicated by the loss of CD31+ cells following PNX described in the present study.

Finally, although the drugs used in this study were used at a relatively low dose, the effect of toxicity and off-target effects cannot be excluded. For example, acriflavine not only binds HIF-

1α but is also known to disrupt DNA synthesis by intercalation hence its use as an antiseptic agent [314]. Mithramycin-A on the other hand is an antibiotic anti-cancer drug that can cause hepatic injury [315]. Furthermore, the specific activity of these drugs on their intended targets in vivo was not confirmed in the present study. Thus, results need to be interpreted with caution.

Use of gene-specific knockout mice may aid in further understanding the specific contribution of these transcription factors in regenerative lung growth post-PNX. Nonetheless, these results underscore the importance of these transcription factors in vascularisation and overall lung growth following lung resection.

To summarise, this chapter has demonstrated that PNX induces expression of EGR-1 and HIF-

1α in the lung and blocking either transcription factor diminishes IGF-1-induced proliferative activity and CD31+ endothelial cell distribution, suggesting that there is an interdependent relationship between EGR-1, HIF-1α and IGF-1 that is crucial in maintaining and intact vasculature and sustaining IGF-1 signalling post-PNX.

174

Chapter 6

General Discussion, Conclusions and Future Directions

175 6.1-General Discussion

In young humans and young animals, the lung has a capacity to initiate parenchymal growth following partial lung resection or pneumonectomy [46]. Subsequent to sexual maturity, the lungs regenerative potential declines and indeed over a life time, normal lung function diminishes due an accumulation of tissue damage caused by injury or disease [42]. Studying mechanisms of lung re-growth may yield strategies for reinitiating lung growth in adult human lungs for treatment and management of chronic destructive lung diseases.

Using a mouse model of pneumonectomy, I was able to contrast the effect of age on regenerative lung growth and use interventionist strategies to assess the involvement of key molecules such as IGF-1, EGR-1 and HIF-1α, which have highly conserved biological functions in mammals.

Supporting the hypothesis that IGF-1 contributes to regenerative lung growth, I detected a transient but significantly higher percentage of IGF-1+ lung cells, mainly localised to the lung parenchyma, following PNX but not SHAM surgery at day three post-surgery. There was also a significantly higher percentage of IGF-1R+, pIGF-1R+ and pERK-1/2+ lung cells, which is consistent with an IGF-1-mediated response (Chapter three). Examination of IGF-1 mRNA and protein expression from the whole lung homogenates, revealed that peak expression coincided with the peak of cell proliferation which occurs at day seven post-PNX. This result is consistent with previous studies that have shown increased levels of IGF-1 in post-PNX animals [7-10], although Price and colleagues did not show an increase in IGF-1 in a rat PNX model [156]. Importantly, this study is the first to demonstrate evidence of phosphorylation of

IGF-1R and the down-stream intracellular signalling molecule ERK-1/2, following PNX.

I tested the ability of IGF-1 to specifically initiate regenerative lung growth by conducting pneumonectomies and then assessing regenerative lung growth with or without IGF-1

176 treatment, which was administered continuously via an osmotic pump. Young adult mice supplemented with IGF-1 demonstrated an enhancement of lung cell proliferation, including that of SpC+ AECIIs, the main progenitor cells of AECIs [116], at day 23 post-surgery

(Chapter four). IGF-1 supplementation was also associated with significantly higher numbers of IGF-1+ lung cells, significantly higher numbers of phosphorylated IGF-1R and ERK-1/2 positive cells and with a higher number of alveoli counted per microscopic field, suggestive of enhanced neo-septation. Longitudinal evaluation of lung morphological parameters using µCT did not detect an IGF-1-associated enhancement in the total lung volume or tissue volumes as both treatment groups reached preoperative levels, by day 21 post-PNX, and this is consistent with the lung volumes being constrained by the size of thoracic space [42].

In a separate experiment, I examined the effect of daily i.p. administration of IGF-1 for seven days post-PNX, which corresponded with the peak of lung cell proliferation (Chapter three).

Over this time frame, longitudinal measurement of the lungs revealed that IGF-1 treatment was associated with significant increase in total lung volume, and sections of lung examined at the end of the experiment contained significantly higher numbers of proliferating lung cells. These two studies both suggested that supplemental IGF-1 has the potential to enhance regenerative lung growth following PNX.

When the continuous IGF-1 administration studies were performed on very old mice (aged 22-

24 month) the biological response was markedly different (Chapter four). Firstly, unlike the lung of young mice, the lungs of old mice did not increase in volume back to pre-operative levels, as has been previously reported by Paxson et.al [4]. However, IGF-1 treatment was associated with an insignificant increase in total lung volume, related to an increase in the volume of air within the lung. The lungs of IGF-1 treated mice in the early weeks following

PNX, appeared to be transiently more stretched, indicated by reduced lung density, relative to

PBS treated controls. It is not clear whether the transient increase in lung volume following

177 IGF-1 treatment might result in better lung function or whether it exacerbated the emphysematous phenotype of aged mice, and this can only be resolved by physiological measurements of lung function.

Consistent with the failure of aged lungs to recover lung volume following PNX, lung sections showed five-fold lower levels of proliferating lung cells at day 23 post-surgery, with no difference detected between lungs supplemented with or without IGF-1. Similarly, there was no evidence of increased numbers of IGF-1+ cells or enhanced phosphorylation of IGF-1R or

ERK-1/2 due to IGF-1 treatment, and no difference in the number of alveoli per field between treatment groups. Unresponsiveness to IGF-1 treatment was not due to an absence of IGF-1R expression as we detected similar levels of IGF-1R in lung sections from PNX young and aged mice.

IGF-1R signalling is critical for lung development with Igf-1R-/- knockout mice dying shortly after birth due to respiratory failure [216, 218]. Given that IGF-1 mediates its effect primarily through the IGF-1R [174], the significance of IGF-1R signalling in post-PNX lung growth was tested by blocking the IGF-1R using the chemical antagonist BMS-536294 [255], following

PNX surgery. It was established that BMS-536294 could attenuate IGF-1-mediated fibroblast cell proliferation and downstream pERK-1/2 signalling in vitro (Chapter five). When the drug was administered to young adult mice, just prior to and daily up until the peak rate of lung cell proliferation at day seven post PNX, there was a delayed increase in total lung volume and tissue volume, relative to control treated PNX mice. Similar experiments will be needed to confirm a decrease in phosphorylation of IGF-1R and ERK1/2 in BMS-536294 mice, as seen in both the untreated and IGF-1 supplemented PNX response (Chapter three).

An important caveat to pharmacological inhibition of IGF-1R by BMS-536294 is its potential inhibition of the insulin receptor [174]. A more specific inhibition of IGF-1R expression might

178 involve the breeding of conditional SpC targeted IGF-1R knock out mice [218]. Taken together, these results are indicative of a potential defect in responsiveness to IGF-1 in the aged lung, which warrants further investigations to confirm.

The loss of responsiveness to IGF-1 in mice is not without precedent. It has been very recently shown in a murine model of rheumatoid arthritis that IGF-1R+ cells present in the hippocampus are unresponsive to IGF-1 signalling, with this unresponsiveness associated with a significantly increased frequency of hippocampal cells positive for the phosphorylated inhibitory serine-612

(S612) of the signalling adaptor protein IRS-1 [316]. It would be interesting to investigate whether similar alterations in inhibitory phosphorylation of signalling adaptor molecules, such as IRS-1 or IRS-2, are present.

Aging lungs, as reflected in the µCT evaluation, fail to stretch in response to PNX to the same degree as young lungs (Chapter four). This age related loss of elasticity and reduced lung compliance may result in diminished capacity to translate mechanical force generated by PNX to proliferative signals [278]. Furthermore, an accumulation of senescent cells in aged lungs potentially reduces to number of critical progenitor cells able to re-enter the cell cycle [281].

Perhaps reducing the burden of senescent cells with any of the many FDA-approved senolytics

(senescence-depleting drugs) may improve the proliferative capacity of aged mice in response to PNX and growth factor treatment [317].

An equally fascinating question that I endeavoured to address was defining which cell types predominantly express IGF-1, IGF-1R/pIGF-1R and IGFBP-3 (the principle binding partner of

IGF-1) in PNX lungs. IHC analysis combined with cell morphology and localisation indicated that post-PNX, IGF-1 was abundantly expressed by alveolar macrophages, SpC+ AECIIs and airway club cells. IGF-1R/pIGF-1R and IGFBP-3 were more widely expressed suggesting that activation of IGF-1 signalling is a preeminent event in multiple cell types participating in post-

179 PNX lung growth. This assertion is supported by several studies that have shown that IGF-1 is a multifactorial pulmotrophic factor for multiple cell types in the lung and is indispensable for lung development, growth and repair [6, 218, 318-320]. In a rat hyperoxia model, IGF-1, IGF-

2, IGF-1R and IGFBP-2, 3 and 5 were found to be upregulated in lung tissue particularly in

AECIIs [301]. In culture, IGF-1 has been shown to promote proliferation and differentiation of epithelial cells [301, 321, 322] and fibroblasts [323-325] whilst targeted over expression of

IGF-1R in mice enhanced endothelial cell regeneration after injury [326] signifying a role for the IGF system in alveolarisation, matrix formation and vascularisation, all important facets of lung regeneration.

Although our IHC analysis highlighted some of the key cell types involved in IGF-1 production post-PNX, further studies are required to confirm the identity of these cell types by co-labelling

IHC, to compare relative expression of IGF-1 amongst different cell populations and to determine which cell type(s) constitute(s) a major source of IGF-1 in post-PNX lungs. In particular, the assessment of the contribution of alveolar macrophages to IGF-1 levels may be important as our observations and other studies have associated them with post-PNX lung growth [7, 145, 147].

It has been established that mechanical stretching of the remaining lung following PNX initiates growth and remodelling in the remaining lung [42]. As activation of the transcription factors EGR-1 and HIF-1α are also associated with mechanical stretch and cellular responses to stress [298, 299], the potential interactions of these transcription factors and IGF-1 in post-

PNX regenerative lung growth was investigated (Chapter five).

I demonstrated in vitro that IGF-1 treatment induces EGR-1 and HIF-1α protein expression in cultured 3T3 fibroblasts. Furthermore, it was identified that both EGR-1 and HIF-1α are transiently increased in the lung parenchyma following PNX at post-operative day three, which

180 coincides with the peak of IGF-1+ cells (Chapter three). Pharmacological inhibition of either

EGR-1 or HIF-1α activity revealed that the density of CD31+ cells and the amount of CD31 protein was significantly reduced in the lung. There is some support for the possibility that this reflects an inhibitor-associated defect in neo-vascularisation following PNX. HIF-1α has been shown to play a vital role in neovascularisation and the HIF-1α inhibitor used in our study, acriflavine, has been shown to inhibit neo-vascularisation in tumours [98]. Similarly, EGR-1 has been implicated to play a role in neo-vascularisation, with the application of DNAzymes, targeting a specific motif in the 5' untranslated region of Egr-1 mRNA, inhibiting angiogenesis in subcutaneous Matrigel plugs in mice, and impaired Matrigel plug angiogenesis demonstrated using EGR-1 deficient mice [327].

By measuring lung cell proliferation in mice receiving a PNX and simultaneously treated with or without acriflavine, I demonstrated that a proportion of regenerative lung growth is dependent on HIF-1α activity. Importantly, I demonstrated that normal levels of lung cell proliferation in pneumonectomised and acriflavine treated mice could be restored by co- administrating IGF-1. Additional experiments to determine an IGF-1-mediated HIF-1α or

EGR-1 dependent stimulation of regenerative lung growth will need to be performed by looking for indirect evidence of a reduction in IGF-1R phosphorylation in the presence of inhibitors. Alternately, inhibition of HIF-1α or EGR-1 may be resulting in a failure to drive other growth factors such as VEGF or FGF-2, and quantitation of these molecules with or without inhibitor could resolve this question [327].

The finding that EGR-1 and HIF-1α expression post PNX is predominantly located towards the subpleural regions of the lung is consistent with the hypothesis that these transcription factors are required for regenerative growth activity in this zone. This is the zone that is reported to undergo the most active neoalveolarisation [3]. Given that the spatio-temporal pattern of expression of these transcription factors coincides with that of IGF-1, it is reasonable

181 to suggest an inter-dependent relationship between EGR-1/HIF-1α and IGF-1 that is crucial in maintaining an intact vasculature and ensuring successful regeneration post-PNX.

Furthermore, this proposition is supported by an elegant study by Pais et.al [6] that employed transcriptomic microarray analysis to show that knockout of the Igf-1 gene in mice perturbs, amongst other important physiological pathways, genes related to the vasculature, cell growth and immunological function. Interestingly, Pais et.al [6] also reported that in these mice EGR-

1 and connective tissue growth factor (CTGF) expression is downregulated, suggesting an intricate relationship between EGR-1 and IGF-1 that is integral in vascularisation and lung development in general. It remains to be determined whether effects of EGR-1/HIF-1α pharmacological blockade on the vasculature are exclusive to PNX lungs or whether a similar phenotype can be recapitulated in unoperated lung. In a separate experiment, it was observed that lungs from unoperated control mice treated with acriflavine are not blanched following perfusion and fixation (data not shown) in contrast to acriflavine treated lungs from PNX.

However, CD31 expression is yet to be assessed in acriflavine treated unoperated control lung tissue.

182 6.2- Conclusions

Based on data obtained from this thesis, the following can be concluded:

i) IGF-1 plays a significant role in post-PNX lung growth with PNX-inducing a robust

but transient expression of IGF-1 in the lung of young adult mice. Furthermore,

blocking IGF-1R signalling retards lung growth and supplemental IGF-1 augments

lung volume, cell proliferation and alveolar septation.

ii) In aged mice, supplemental IGF-1 has no effect on cell proliferation, alveolar

septation or IGF-1R signalling pathway activation.

iii) IGF-1 supplementation in aged PNX mice increases the aerated lung volume causing

a transient decline in tissue volume and mean lung density.

iv) IGF-1-induced post-PNX lung growth appears partly dependent on the activity of

EGR-1 and HIF-1α transcription factors.

183 6.3- Implications of findings

6.3.1- Clinical relevance of findings

Recombinant human (rh) IGF-1 (Mecasermin) and rhIGF-1-rhIGFBP-3 complex

(Mercasermin-reinfabate, iPLEX) are two FDA-approved drugs that have been available on the market for over 10 years for treatment of growth disorders, burns, osteopenia, diabetes and several other endocrine disorders [12]. Additionally, there are a number of on-going or just completed clinical trials of rhIGF-1 replacement therapy conducted in the context of inflammatory and lung development disorders including acute respiratory distress syndrome

(ARDS), asthma, cystic fibrosis, COPD and IPF where dysregulation of IGF-1 signalling is implicated, as well as age-associated neurodegenerative and muscular atrophy diseases [328].

The results of this study provide further insights into the role of IGF-1 plays in lungs capable of regenerative lung growth following lung resection. Data from the current study suggests a defect in IGF-1 signalling in aged lungs and cautions that supplemental IGF-1 therapy might not be an effective adjunct driver of regenerative lung growth in this setting. Further work to confirm and define the nature of any IGF-1 signalling defect in humans may lead to strategies to re-establish IGF-1 responsiveness and lead to new treatments to support the improvement of lung function in patients following pulmonary resection surgery. Given that some positive effects on regenerative lung growth in the lungs of IGF-1 treated young adult mice were evident, the use of IGF-1 might be more effective at enhancing lung regeneration post-PNX in young adults and children. In particular, young patients with lower than normal serum levels of IGF-1 are more likely to benefit from such a therapy, similar to a study by Hardin et.al [329] where administration of GH in children with cystic fibrosis normalised IGF-1 and IGFBP-3 levels and improved lung function.

184 Despite lack of evidence of IGF-1 enhancing regenerative lung growth in the aged mouse lung following PNX, we did demonstrate that the treatment did enhance aerated lung volume. A critical examination on any effects on lung function in IGF-1 treated aged lungs will be necessary to assess whether this might be a translatable effect of IGF-1 to improve post-PNX outcomes in patients.

Understanding how mechano-transduction pathways are linked to the IGF-1 signalling pathway could potentially improve the efficacy of rhIGF-1-based therapies for treatment of pulmonary disorders. The finding that EGR-1 and HIF-1α and IGF-1 contribute to post-PNX lung growth highlights their potential as interventional targets for modulating regenerative lung growth. The use of EGR-1/HIF-1α agonists in the setting of less compliant older lungs, potentially with defective IGF-1R signalling, would be an interesting area to explore.

Furthermore, EGR-1/HIF-1α agonists may be useful in patients undergoing extensive pulmonary resection with insertion of a plombage that helps ameliorate symptoms of dyspnea and adverse post-surgical events such as mediastinal shift, hypertension and pulmonary edema

[1]. The use of plombage has been associated with reduced compensatory lung growth in multiple animal models [67, 88]. Finding ways to molecularly manipulate lung growth whilst preventing adverse cardio-pulmonary events could yield tremendous benefits for patients undergoing lung resection with plombage.

185 6.3.2- Potential challenges of translating research findings to the clinic

There are many significant challenges and caveats which concern advocacy for rhIGF-1-based therapies in treating lung diseases. Firstly, IGF-1 like many other growth-promoting factors, if given in excess, can contribute to neoplastic transformation as well as acute hypoglycaemia

[330]. Monoclonal antibody therapies against IGF-1 and IGF-1R are currently being tested in combination with tyrosine kinase inhibitors for treatment of non-cell lung carcinoma and other types of lung cancer [331]. If IGF-1 is to be used for treatment post lung surgery, it will be important to carefully work out the best dose, mode of delivery and timing of the regimen to avoid any deleterious effects. A probable strategy would be to target IGF-1 specifically to the lung and perhaps co-administer it with IGFBPs or in the form of IGF1-Eb peptides, which would prolong the half-life of IGF-1 and minimise spill over in serum [252]. Our work suggests that the timing of IGF-1 administration might be limited to a short window after surgery, coinciding with the presence of a robust enough mechanical force to promote optimum cell growth.

Secondly, findings from the murine PNX model have to be interpreted with caution and some aspects may not be directly translatable to humans because of the inherent differences in the morphology and physiology between the murine and human lung. Whilst the human lung ceases to grow after sexual maturity, the murine lung retains the potential for somatic growth and thoracic expansion well into adulthood [4]. Mouse lungs are less stratified with fewer respiratory bronchioles, and have a thin pleura making their lungs more complaint and susceptible to mechanical stress and ventilatory signals [51]. The cellular hierarchy of the mouse lung is also different from human, including the distribution and presence of progenitor cells (reviewed in Chapter one, Section 1.2.2). Additionally, some progenitors cells reported in mice such as BASCs are yet to be reported in humans [112]. Thus, although mice provide useful insights into the potential mechanisms regulating post-natal lung growth, it is important

186 to remember that findings in rodents may not exactly recapitulate what transpires in the human lung.

All the above highlighted factors, however challenging, will have to be considered before recommending IGF-1 therapy for lung regeneration in patients. They also point to a need for more comprehensive preclinical studies aimed at unravelling the complexity of the IGF-1 signalling network in regenerative lung growth.

6.4- Future directions

This study has provided many novel findings concerning the role of IGF-1 in regenerative lung growth post-PNX. However, there are many areas that could be improved and future experiments that can be performed to further our understanding of the PNX model and the role of IGF-1 in regenerative lung growth.

To further clarify the kinetics of IGF-1 signalling post-PNX, future studies should consider assaying levels of IGF-1 in serum, lung protein lysates and BALF by ELISA with increased numbers of animals. To more accurately investigate the role of different cell types involved, lung digests could be performed, combined with flow cytometry analysis and cell sorting, to aid in the identification of cell populations actively involved in IGF-1-associated signalling.

Furthermore, characterisation of the IGF-1 pathway gene expression as well as the hypothesis that EGR-1 and/or HIF-1α drive IGF-1 expression was based on selection of specific targets.

An unbiased approach like RNA-seq could more broadly identify regulation of other IGF-1- related pathways [332, 333].

187 Pharmacological inhibition of the IGF-1R indicated that IGF-1R signalling was only able to reduce or delay post-PNX lung growth (Chapter three). However, as with many pharmacological inhibitors, despite being clinically relevant, data might be confounded by issues of non-specificity and toxic side effects. The reliance of post-PNX lung growth on IGF-

1 and IGF-1R signalling could be further tested using IGF-1/IGF1R gene knockout mice respectively. Homozygous IGF-1/IGF1R gene knock out in mice has severe consequences on lung development and pups rarely make it past a few days after birth due to respiratory failure

[334]. Conditional knockouts where depletion of IGF-1 or IGF1R is targeted only to a particular lung cell type could be used to determine the effect of post-natal deletion of IGF-

1/IGF-1R. For example, Lopez et.al [200] generated Nkx2-1-Cre;Igfrfl/fl and Scgb1a1;Igf1r fl/fl double transgenic mouse lines to study the effect of IGF-1R deletion in lung epithelial cells and club cells respectively, following naphthalene challenge. Alternatively, partial knockout mice for example the Igf1m/m midi mice (Jackson Laboratory) which have 60% less serum levels of IGF-1 could be used. It would also be interesting to dissect the contribution of endocrine vs paracrine/autocrine IGF-1 using a liver specific IGF-1 knockout line such as that used by

Sjögren and colleagues [335]. Furthermore, with the advent of lineage tracing technologies, it would be interesting to follow the fate of various alveolar progenitors with IGF-1/IGF-1R gene deletion in response to PNX.

It was shown that rhIGF-1 supplementation via an osmotic pump enhanced total lung volume but had no effect on the cellular characteristics of aged mice (Chapter four). To further test the potential of IGF-1 to influence PNX-induced regenerative lung growth in aged mice, it may be worth considering performing a rhIGF-1 dose-response experiment delivered via i.p. injection. Unlike young adult mice, the response of aged mice to PNX has not been characterised longitudinally with respect to IGF-1 expression. The ability of aged mice to

188 spontaneously induce IGF-1 expression in response to PNX needs to be tested by analysing them on day three or seven post-PNX.

To further understand the differential responses to IGF-1 treatment between young and old mice, alveolar type two cells or primary lung fibroblasts could be isolated from both groups of mice to study their response to IGF-1 treatment and stretch in vitro. Cell proliferation and migration assays could provide important insight in these proposed studies. Preclinical studies could also include the use of alveolar organoid assays to dissect molecular pathways involved in IGF-1 signalling in vitro [336]. An established alveolar epithelial cell line such as A549, co- cultured with a 3T3-feeder cell line could be used to optimise culture conditions and test various compounds (for example EGR-1 and HIF-1α antagonists) before pursuing primary cells.

The role of EGR-1 and HIF-1α transcription factors in driving IGF-1 signalling was investigated using pharmacological inhibitors (Chapter five). To determine the effect of EGR-

1/ HIF-1α blockade on IGF-1 signalling, similar studies could be conducted with mice analysed at either day three or day seven post-PNX to examine effects on IGF-1 expression by

IHC/Western blot and cell proliferation by IHC. Additionally, the extension of EGR-1/HIF-1α blockade experiments by including IGF-1 only and IGF-1 plus EGR-1/HIF-1α inhibited treatment groups taken to day 15 post-PNX could help determine whether IGF-1 can reverse or prevent the loss of CD31 expression in these mice. Use of EGR-1 or HIF-1α gene knockout mice would further the understanding of these transcription factors in post-PNX lung growth and how they relate to the IGF-1 signalling pathway.

Furthermore, future studies would be improved by inclusion of lung function measurements as additional parameters in assessing the success of regeneration post-PNX. The most common method for measuring lung function in PNX mice is the forced oscillation technique which

189 measures pressure volume changes using the FlexiVent (SCIREQ) to assess pulmonary compliance, total lung capacity, functional residual capacity and vital capacity in anaesthetised tracheostomised animals [337]. Other measures of lung function which assess efficiency of gas change such as diffusion capacity for carbon monoxide (DLCO) and pulse oximetry have been used in the PNX model [65, 140, 338].

Lastly, one of the strengths of this thesis was the use µCT to characterise regenerative lung growth. There are many potential exciting avenues that can be further explored with this imaging approach. Fixed/dehydrated lungs can be prepared and imaged ex vivo at higher resolution to provide stereological estimates of alveolar number, size and surface area in 3D as opposed to conventional 2D histological techniques [339-341]. µCT can be combined with

PET imaging to highlight areas of high metabolism/growth [56]. Contrast dyes such as barium sulphate may be used to highlight the vasculature or improve image definition [342]. Working out these protocols might allow for similar non-invasive studies to be conducted on humans.

Whilst additional studies suggested above might provide a clear direction for the translation of current and future findings to the clinic, it is hoped that findings from this study have provided an important contribution on the role of IGF-1 in regenerative lung growth following PNX.

190

Chapter 7

Appendices

191 Appendix A- Materials

A.1- Buffers and Solutions

Buffer Composition and Preparation Purpose

Acid Ethanol 100% ethanol Elution of 0.1M HCl methylene blue dye for cell proliferation Make a 1:1 v/v solution of the above analysis components

Alkaline phosphatase buffer 10mM TBS, pH8 Used to dissolve 1mM MgCl2 Streptavidin alkaline 150mM NaCl phosphatase for chromogenic Add components at indicated molar substrate concentrations to dH2O and adjust development during pH to 8. Store at buffer at room IHC temperature (R.T). Dissolved Streptavidin alkaline phosphatase should be stored in aliquots at -20°C.

50x Borate buffer 0.5M Boric acid Washing buffer for 0.135M NaOH methylene blue assay Components were dissolved in dH2O at stated molar concentrations and the pH of the solution adjusted to 8.5. Buffer was stored at R.T and diluted 1 in 50 with dH2O prior to use.

Citrate buffer 10mM Tri-sodium citrate Heat-mediated 1L dH2O antigen retrieval of paraffin embedded Tri-sodium citrate was dissolved in tissue sections 600mL of dH2O at indicated molar concentration, mixed, the pH adjusted to 6 with HCl and the volume brought up to 1L with dH2O. Buffer was stored at 4°C.

192 Buffer Composition and Preparation Purpose

Eosin Y solution Eosin Y powder H&E stain 95% Ethanol 80% Ethanol Glacial acetic acid

Mix 2g of Eosin Y powder with 40mL of dH2O. Add 160mL of 95% ethanol then 600mL of 80% ethanol and mix. Finally add 4mL of glacial acetic acid. Store air tight at R.T.

Ethylenediaminetetraacetic 1mM EDTA Heat-mediated acid (EDTA) buffer 0.05% Tween-20 antigen retrieval of 2L dH2O paraffin embedded tissue sections 1mM EDTA was added to 900mL of dH2O, mixed, the pH adjusted to 8 with NaOH solution and the volume brought up to 2L with dH2O and finally the buffer was supplemented with 0.05% Tween-20. Buffer was stored at 4°C.

Glycine stripping buffer 50mM Glycine Stripping proteins on 35mM SDS Western blot 1% Tween-20 membranes for re- probe. Add 50mM Glycine and 35mM SDS to 600mL of dH2O with gentle stir. Adjust pH to 2.5 with HCl. Bring the solution to 1L and store at 4°C.

Hypertonic buffer 0.1mM EDTA Nuclear protein 50mM HEPES, pH 7.8 extraction 50mM KCl 300mM NaCl

Components were added at the indicated molar concentration in dH2O, mixed and stirred. Buffer was stored at 4ºC. To make 1mL of protein lysis buffer, 843µL of hypertonic buffer was mixed with 143µL of 7x protease inhibitors, 100mM PMSF, 2mM and DTT reducing agent.

193 Buffer Composition and Preparation Purpose

Hypotonic buffer 0.1mM EDTA Nuclear protein 10mM HEPES, pH 7.8 extraction 10mM KCl 2mM MgCl2

Components were added at the indicated molar concentration in dH2O, mixed and stirred. Buffer was stored at 4ºC. To make 1mL of protein lysis buffer, 833µL of hypertonic buffer was mixed with 143µL of 7x protease inhibitors, 100mM PMSF, 2mM, DTT reducing agent and 10µL of NP-40 detergent. 2% Methylene blue 2% w/v Methylene blue For methylene blue dH2O binding cell proliferation assay Dissolve 4g of methylene blue in 200mL of 0.01M borate buffer (pH 8.5) and filter prior to use. Keep at R.T.

4% Paraformaldehyde 4% w/v PFA Lung inflation and cell (4% PFA) dH2O fixation during cell culture To make a 500mL solution, heat 400mL of dH2O to 60°C in a fume cabinet, add 20g of PFA and stir while adding 10N NaOH solution dropwise until solution clears. Cool solution to R.T, add 25mL of 20x PBS and adjust pH to 7.4. Top up solution to 500mL with dH2O and store solution at -20°C in 20-50mL aliquots

1x Phosphate buffered saline One PBS tablet containing 0.01M General purpose (PBS) phosphate buffer, 0.0027M washes potassium chloride and 0.137M sodium chloride, pH 7.4 was dissolved in 200mL dH2O, autoclaved and stored at R.T

194 Buffer Composition and Preparation Purpose

4x Protein Loading buffer 100% Glycerol Mixed with protein 1M Tris-HCl pH 6.8 lysates prior to 8% SDS loading on the gel 0.5% Bromophenol blue 5% β-mercaptoethanol dH20

10mL of loading buffer was made up inside a fume cabinet, aliquoted and stored at -20°C

Scott’s tap water 42mM NaHCO3 Blueing solution 164mM MgSO4 used to enhance 1L dH2O haematoxylin counterstain Components were added at indicated molar concentrations in dH2O, mixed and the solution stored at R.T.

10% Sodium Dodecyl 5g SDS 10% SDS solution Sulphate (SDS) solution 50mL dH2O was used as a reducing agent for making 5g SDS was added to 40mL of polyacrylamide gels dH2O, mixed with a magnetic stirring rod until dissolved and then the volume adjusted to 50mL.

10x SDS-PAGE Running 2.5M Tris-base Immersion buffer buffer 1.9M Glycine used to run western 170mM SDS blot gels 500mL dH2O

Chemicals were added at indicated molar concentrations to 400mL dH2O and mixed using a magnetic stirrer for 16-18 hours. The volume was then adjusted to 500mL with dH2O and stored at room temperature.

195 Buffer Composition and Purpose Preparation

20x Tris buffered saline 200mM Tris To make 1x TBS for IHC (TBS) 3M NaCl and Western blot washes

Chemicals were added added at indicated molar concentrations to 600mL dH2O and mixed using a magnetic stirrer for one to two hours. The pH was adjusted to 7.6 with HCl and the volume adjusted to 1L with dH2O. The solution was autoclaved and stored at room temperature.

1x TBS/T 1x TBS For IHC and western blot 0.05% Tween 20 washes

10x Transfer buffer 390mM Glycine To make 1x Transfer buffer 480mM Tris base for Western blots

Chemicals were added at indicated molar concentrations to 400mL dH2O and mixed using a magnetic stirrer for 16- 18 hours. The volume was then adjusted to 500mL with dH2O and stored at 4 °C.

1x Transfer buffer 45mL 10x Transfer Buffer Immersion buffer for 50mL 100% Methanol electrophoretic transfer of 405mL dH2O proteins from polyacrylamide gel to PVDF/Nitrocellulose membrane

Table 7.1- General purpose buffers and solutions made in the laboratory.

All chemicals were purchased from Sigma-Aldrich (Melbourne, Australia) unless otherwise specified.

196 A.2- Western Blot Gels

Running DI Water 30% 1.5M 1M 10% 10% TEMED* gel % (mL) Acrylamide Tris-HCl Tris-HCl SDS APS (mL) (pH 8.8) (pH 6.8) (mL) (mL) 10% 4 3.3 2.5 ---- 0.1 0.1 0.004

12% 3.3 4 2.5 ---- 0.1 0.1 0.004

15% 2.3 5 2.5 ---- 0.1 0.1 0.004

5% 6.8 1.7 ---- 1.25 0.1 0.1 0.01 Stacking gel

Table 7.2: SDS-PAGE gel composition. The volumes above were used to make 10mL of the running gel or staking gel. *APS was always added last.

A.3- Tissue Culture Reagents

Reagent Company

Dulbecco’s Modified Eagle’s Medium Life Technologies (Mulgrave, Australia) (DMEM)

Foetal Calf Serum (FCS) SERENA (Bunbury, Australia)

L-glutamine Life Technologies

Penicillin/Streptomycin Life Technologies

0.25% Trypsin-EDTA Life Technologies

Trypan blue Sigma-Aldrich (St. Louis, MO)

197 A.4- Real-time PCR reagents

A.4.1- Complementary DNA (cDNA) Mastermix

Component Volume per Reaction (μL) 10x Reverse Transcriptase buffer 2 10x Deoxyribonucleoside triphosphates 2 (dNTPs) Random Hexamer primers (50μM) 0.7 RNase Inhibitor (4U/μL) 0.5 Reverse transcriptase (4U/μL) 0.5

Nuclease free H20 0.8

Table 7.3: Mastermix constituents for synthesising cDNA. All cDNA mastermix components were purchased from Qiagen, Chadstone, Australia except for Random Hexamer primers (Invitrogen Life Technologies, Mulgrave, Australia).

A.4.2- Real-time PCR mastermix

Component Volume per reaction (μL) 2x Quantifast Mastermix 5 20x Target probe-FAM labelled 0.5 Control probe (PGK-1)-VIC labelled 0.5 RNase-free water 1.5

Table 7.4: Mastermix composition for real time PCR. All Real-time PCR components were purchased from Invitrogen Life Technologies, Mulgrave, Australia except for 2x Quantifast Mastermix (Qiagen).

198 A.4.3- Taqman probe sets

Gene Taqman Reference Sequence

Insulin-like growth factor (IGF)-1 Mm01228180_m1

IGF-1 Eb transcript Mm00710307_m1

IGF-1 receptor (IGF-1R) Mm01318466_m1

IGF binding protein (IGFBP)-2 Mm00492632_m1

IGFBP-3 Mm01187817_m1

IGFBP-6 Mm00599696_m1

Phosphoglycerate kinase (PGK)-1 Mm00435617_m1 (Housekeeping gene)

Table 7.5: TaqMan primer probes used for real time PCR. All TaqMan primer probe sets were purchased from Applied Biosystems by Life Technologies, Australia. All probes were FAM-labelled except for PGK-1 which was VIC-labelled.

A.5- Kits Kit Company (Location) Purpose

ReliaPrepTM RNA Cell Promega RNA extraction Miniprep System (Alexendria, Australia)

Omniscript Reverse Qiagen (Germantown, PA) cDNA synthesis Transcriptase Kit

Liquid Fast-Red Subtrate Kit Abcam Red chromogen for IHC (Melbourne, Australia) antigen detection

Stay Green/AP Plus Abcam Green chromogen for IHC (Melbourne, Australia) antigen detection

Human IGF-1 ELISA kit Abcam To detect IGF-1 in mouse (Melbourne, Australia) serum

199 A.6- Immunohistochemistry (IHC) and Western blot (WB) Antibodies

Primary Stock Source Secondary Source Antibody Concentration (Location) Antibody (Location)

(Species, Purpose; Concentration; clonality) (Dilution) (Dilution)

1mg/mL Abcam Polyclonal rabbit Agilent Dako α-Smooth (Melbourne, anti-mouse (Santa Clara, CA) muscle actin IHC; Australia) biotinylated [A14] (1 in 1000) F(ab’)2

(Mouse mc) 0.95g/L; (1in 200)

0.25µg/mL Pierce Goat anti- Santa Cruz α-Tubulin Biotechnology mouseHRP Biotechnology WB; (Rockford, IL) (Santa Cruz, CA) (Mouse mc) (1 in 10,000) 0.01µg/mL; (1 in 80,000)

1mg/mL Abcam Polyclonal swine Agilent Dako (Melbourne, anti-rabbit (Santa Clara, CA) IHC; Australia) biotinylated IgG (1 in 200) 0.92g/L; CD31 (1 in 200)

(Rabbit pc) WB; Goat anti- Santa Cruz (1 in 1000) RabbitHRP Biotechnology (Santa Cruz, CA) 0.02µg/mL; (1 in 50,000)

0.1mg/mL Santa Cruz Polyclonal swine Agilent Dako Biotechnology anti-rabbit (Santa Clara, CA) IHC; (Santa Cruz, CA) biotinylated IgG (1 in 1000) 0.92g/L; EGR-1 (1 in 200) [588]

WB; Goat anti- Cell Signaling (Rabbit pc) (1 in 1000) rabbitHRP Technology polyclonal IgG (Danvers, MA)

65.7µg/mL; (1 in 10,000)

200 Primary Stock Source Secondary Source Antibody Concentration (Location) Antibody (Location)

(Species, Purpose; Concentration; clonality) (Dilution) (Dilution)

0.2mg/mL Santa Cruz Goat anti- Santa Cruz ERK-2 Biotechnology RabbitHRP Biotechnology [C-14] WB; (Santa Cruz, CA) (Santa Cruz, CA) (1 in 2000) 0.02µg/mL; (Rabbit pc) (1 in 50,000)

1mg/mL Novus Goat anti- Santa Cruz HIF-1α Biologicals MouseHRP Biotechnology [H1alpha67] WB; (Littleton, CO) (Santa Cruz, CA) (1 in 500) 0.01µg/mL; (Mouse mc) (1 in 10,000)

1mg/mL Thermo Fisher Polyclonal swine Agilent Dako Scientific anti-rabbit (Santa Clara, CA) HIF-1α IHC; (Rockford, IL) biotinylated IgG (1in 500) (Rabbit pc) 0.92g/L; (1 in 200)

IGF-1 1mg/mL Peprotech n/a (Rocky Hill, NJ) (Biotinylated IHC; Rabbit pc) (1 in 100)

1mg/mL Abcam Goat anti- Santa Cruz IGF-1 (Melbourne, MouseHRP Biotechnology

WB; Australia) (Santa Cruz, CA) (Mouse mc) (1 in 500) 0.01µg/mL;

(1 in 15,000)

1mg/mL Sigma Aldrich Polyclonal swine Agilent Dako IGF-1R (St. Louis, MO) anti-rabbit (Santa Clara, CA) [AB-1161] IHC; biotinylated IgG (1 in 1000) (Rabbit pc) 0.92g/L; (1 in 200)

1mg/mL Thermo Fisher Goat anti- Santa Cruz IGF-IR Scientific MouseHRP Biotechnology [194Q13] WB; (Rockford, IL) (Santa Cruz, CA)

(1 in 1000) 0.01µg/mL; (Mouse mc) (1 in 50,000)

201 Primary Stock Source Secondary Source Antibody Concentration (Location) Antibody (Location)

(Species, Purpose; Concentration; clonality) (Dilution) (Dilution)

1mg/mL GroPep Polyclonal Agilent Dako Bioreagents swine anti-rabbit (Santa Clara, CA) IHC; Pty Ltd biotinylated IgG (1in 400) (Thebarton, Australia) 0.92g/L; IGFBP-3 (1 in 200)

(Rabbit pc) WB; Goat anti- Santa Cruz

(1 in 500) RabbitHRP Biotechnology (Santa Cruz, CA) 0.02µg/mL; (1 in 10,000)

0.2mg/mL Santa Cruz Polyclonal Agilent Dako Pro-SPC Biotechnology rabbit anit- (Santa Clara, CA) [M20] IHC (Santa Cruz, goatHRP (1in 1000) CA) (Goat pc) 0.95g/L; (1in 200)

1mg/mL Merck Polyclonal Agilent Dako Millipore swine anti-rabbit (Santa Clara, CA) IHC; (Temecula, biotinylated IgG (1 in 50,000) CA) pIGF-1R 0.92g/L; [Tyr: (1 in 200) 1161/1165/

1166] WB; Goat anti- Cell Signaling

(1 in 2000) rabbitHRP Technology (Rabbit pc) polyclonal IgG (Danvers, MA)

65.7µg/mL; (1 in 10,000)

202

Primary Stock Source Secondary Source Antibody Concentration (Location) Antibody (Location)

(Species, Purpose; Concentration; clonality) (Dilution) (Dilution)

90µg/mL Cell Signaling Goat anti- Cell Signaling pAKT Technology rabbitHRP Technology [Ser473 (D9E) WB; (Danvers, MA) polyclonal IgG (Danvers, MA) XP®] (1in 1000) 65.7µg/mL; (Rabbit mc) (1 in 2,000)

35µg/mL Cell Signaling Goat anti- Cell Signaling Technology rabbitHRP Technology pan AKT WB; (Danvers, MA) polyclonal IgG (Danvers, MA) [C67E7] (1 in 2000)

65.7µg/mL; (Rabbit mc) (1 in 2,000)

177µg/mL Cell Signaling Polyclonal Agilent Dako Technology swine anti- (Santa Clara, CA) IHC; (Danvers, MA) rabbit (1 in 1000) biotinylated IgG

p-ERK1/2 0.92g/L; [Tyr202/204] (1 in 200)

(Rabbit mc) WB; Goat anti- Cell Signaling (1 in 1000) rabbitHRP Technology polyclonal IgG (Danvers, MA)

65.7µg/mL; (1 in 10,000)

Table 7.6: Primary and secondary antibodies used for immunohistochemistry and western blot analysis.

Primary antibodies for Western blot (WB) analysis were diluted as described in the table above. WB secondary antibodies were diluted in 5% milk/TBST. IHC primary and secondary antibodies were diluted in 0.1% Tx-100/1% BSA/TBS. Biotinylated secondary antibody conjugates were detected by further incubation in streptavidin horseradish peroxidase (s-HRP) solution (0.82g/L [1in 200]; Agilent Dako, CA) or streptavidin alkaline phosphatase (s-AP) solution (1mg/mL [1in 50]; Sigma Aldrich, St. Louis, MO) prior to addition of the chromogen. mc= monoclonal antibody; pc= polyclonal antibody; p=phosphorylated.

203 A.7- Equipment and Instruments used

INSTRUMENT MANUFACTURER (LOCATION)

AGFA CP1000 X-ray Processor AGFA Healthcare (Mortsel, Belgium)

ALZET® Mini-osmotic pump 2004 model ALZET®, DURECT ™ Corporation (Cupertino,CA)

Aperio AT2 slide scanner Leica Biosystems (Vista, CA)

Bioflex 6-well plates- Collagen I coated Flexcell® Int. Corp. (Burlington, NC)

Bolt™ Mini Gel tank ThermoFisher Scientific (Rockford, IL)

CL-XPosure Film (18x24cm) Thermo Scientific (Erembodegem, Belgium)

Cuisinart Electric Pressure Cooker, Cuisinart® (East Windsor, NJ) CPC-600A Falcon® Clear flat-bottomed IncuCyte cell Corning (Noble Park, NY) proliferation 96-well plates

HURST Adhesive glass slides HURST Scientific Pty Ltd (Canning Vale, Australia)

HyperCassetteTM Amersham Pharmacia Biotech (Buckinghamshire, England) iBlot® Gel transfer stacks Nitrocellose ThermoFisher Scientific (Rockford, IL) regular iBlot® Transfer system ThermoFisher Scientific (Rockford, IL)

Image Lock 96-well microplates Essen BioScience Inc. (Ann Arbor, MI)

IncuCyte ZOOM® Live Cell Imaging Essen BioScience Inc. (Ann Arbor, MI) system

ISOTEC3 Anaesthetic chamber Advanced Anaesthetic Specialists (Sydney, Australia)

Leica automatic tissue processor Leica Biosystems (Nussloch, Germany) model TP 1020

Leica microtome model RM2245 Leica Biosystems (Nussioch, Germany)

Leica M651 dissection microscope Leica Microsystems Pty Ltd. (Macquarie Park, Australia)

204 INSTRUMENT MANUFACTURER (LOCATION)

Leica tissue embedder model EG1150C Leica Biosystems (Nussloch, Germany)

Mastercycler® thermo cycler Eppendorf (Hauppauge, NY)

MiniVent respirator, Type 845 Hugo Sachs Elektronik (March, Germany)

NanoDrop 2000 U.V-Vis spectrophotometer Thermoscientific (Wilmington, DE)

Neubauer Improved Bright-line Hirschmann® (Eberstadt, Germany) Haemocytometer

Nikon Diaphot Inverted Tissue Culture Nikon Instruments Inc. (Tokyo, Japan) Microscope

Nikon Eclipse Ni-E motorised Nikon Instruments Inc. (Tokyo, Japan) brightfield/fluorescent microscope

ROTINA 420R Centrifuge VWR International Pty Ltd (Murarrie, Australia) SkyScan 1176 small animal µ-CT scanner Bruker micro-CT (Kontich, Belgium)

StepOne Plus™ Real-Time PCR system Applied Biosystems (Carlsbad, CA)

2 Wallac Victor V™ Plate Reader Perkin Elmer Life Sciences (Waltham, MA)

Western Blot electrophoresis and transfer BIO-RAD (Hercules, CA) tanks

Woundmaker™ device Essen BioScience Inc. (Ann Arbor, MI) pH 700 pH/mV/ 0C/ 0F Bench meter EUTECH INSTRUMENTS (Ayer Rajah Crescent, Singapore)

205 A.8- Components of the Equibiaxial Stretch Device

Number Component Length Width Height Radius required/plate Acrylic plate 155 130 4.5 2

PVC discs 10 32 1-6

Bolts 65 6 4

Nylon lock nuts 6 4

Butterfly 6 4 nuts

Table 7.7: Specifications of the Equibiaxial stretch device. All dimensions are in millimetres (mm). Acrylic plates and polyvinyl chloride (PVC) discs were custom made at a plastic shop (The Plastic Display People, Perth, Australia). Bolts and nuts were purchased at Bunnings Warehouse (Perth, Australia).

A.9- Drugs used

Drug Manufacturer Dose Vehicle Mode of Target (Location) used Delivery In vivo Acriflavine Sigma Aldrich 2mg/kg PBS i.p. injection HIF-1α hydrochloride (St. Louis, MO)

BMS-536924 Sigma Aldrich 20mg/kg 0.2% Oral gavage IGF-1R (St. Louis, MO) DMSO/PBS

Mithramycin-A Sigma Aldrich 150µg/kg 0.2% i.p. injection EGR-1 (St. Louis, MO) DMSO/PBS rhIGF-1 PeproTech, 500ng/mL PBS Subcutaneous All (Rocky Hill, NJ) delivery by tissues osmotic pump

i.p. injection

Table 7.8: Drugs administered in in vivo experiments PBS= Phosphate buffered saline, DMSO= Dimethyl Sulfoxide, i.p.= Intraperitoneal, rh= Recombinant human.

206 A.10- Other Reagents

ITEM MANUFACTURER (LOCATION) Albumin from Bovine Serum Sigma Aldrich (Australia)

Ammonium Persulphate (NH4)2S2O8 (APS) BIO-RAD (Gladesville, Australia)

Bolt™ 4-12% Bis-Tris Plus gels ThermoFisher Scientific (Rockford, IL)

Bolt™ Sample loading buffer (4X) Novex by Life Technologies (Carlsbad, CA)

Bolt™ Sample reducing agent (10X) Novex by Life Technologies (Carlsbad, CA)

Bovine Serum Albumin Standard Ampules Thermo Scientific (Scoresby, Australia) (2mg/mL)

Chemiluminescent HRP Substrate Millipore (Billerica, MA) cOmplete Tablets Mini EDTA-free Easy Roche (Indianapolis, IN) pack

DePex Mounting medium VMR® (Murarrie, Australia)

3,3’-Diaminobenzidine Tablets Sigma Aldrich (Australia)

Dimethyl Sulfoxide (DMSO) Sigma Aldrich (Australia)

100% Ethanol ROWE Scientific (Wangara, Australia)

Ficoll-PaqueTM PLUS GE Healthcare (Perth, Australia)

Gill’s hematoxylin, Grade 2 Polysciences Inc. (Warrington, PA)

Human Placental Collagen Sigma –Aldrich (St. Louis, MO)

30% Hydrogen Peroxide Solution Riedel-de Häen® (Seelze, Germany)

Methanol Chem Supply (Port Adelaide, Austalia)

NuPAGE MES SDS Running buffer (20x) Novex by Life Technologies (Carlsbad, CA) pH 4/7/10 buffer solutions Perth Scientific (Perth, Australia)

Phenylmethylsulfonyl fluoride (PMSF) Sigma Aldrich (Australia)

PierceTM Coomassie PlusTM (Bradford) Thermo Scientific (Scoresby, Australia) Protein Assay

207 ITEM MANUFACTURER (LOCATION)

Pierce® RIPA lysis buffer ThermoFisher Scientific (Rockford, IL)

Ponceau S Sigma Aldrich (Australia)

Precision Plus ProteinTM Dual Colour BIO-RAD (Hercules, CA) Standards

Propan-2-ol (iso-propyl alcohol) VWR® (Murarrie, Australia)

Skim Milk Powder IGA (Perth, Australia)

Tergitol® solution Type NP-40 Sigma Aldrich (St. Louis, MO)

Triton X-100 Sigma Aldrich (Australia)

Tris-base Sigma Aldrich (St. Louis, MO) Tris-HCl Sigma Aldrich (St. Louis, MO)

Tween-20® Sigma Aldrich (St. Louis, MO)

Xylene Chem Supply (Port Adelaide, Australia)

208 Appendix B- µ-CT Image acquisition, Sorting and Reconstruction settings [System] Scanner=SkyScan1176 Instrument S/N=10B08015 Hardware version=A Software=Version 1. 1 (build 11) Home directory=C:\SkyScan Source Type=PANalytical's Microfocus Tube Camera=Princeton Instruments Camera Pixel Size (um)=12.47 CameraXYRatio=0.9840 Bed shift in X per 1000 lines=-1.740000 Bed shift in Y per 1000 lines=2.320000 [Acquisition] Data directory=D:\Joe Yasa\PNX D7 IGF-1 only and IGF-1 plus Acriflavine treated-02-03- 2018\IGF-1 plus Acriflavine\P317 Filename Prefix=p317_02032018_ Filename Index Length=4 Number of Files=283 Source Voltage (kV)=50 Source Current (uA)= 500 Number of Rows=668 Number of Columns=1000 Image crop origin X= 0 Image crop origin Y=0 Camera binning=4x4 Image Rotation=0.7940 Gantry direction=CC Number of connected scans=1 Image Pixel Size (um)= 35.55

209 Object to Source (mm)=121.000 Camera to Source (mm)=169.000 System Matrix Calibration=NO Vertical Object Position (mm)=140.425 Optical Axis (line)= 337 Filter=Al 0.5mm Image Format=TIFF Data Offset (bytes)= 264 Horizontal overlap (pixel)=32 Camera horizontal position=Center Depth (bits)=16 Visual Camera=ON Synchronised Scan=OFF Delay for external event (ms)=(300) List mode=ON (8) Screen LUT=0 Exposure (ms)= 85 Rotation Step (deg)=0.700 Frame Averaging=OFF (1) Use 360 Rotation=NO Geometrical Correction=ON Camera Offset=OFF Scanning Start Angle=-0.044 Median Filtering=ON Flat Field Correction=ON Rotation Direction=CC Scanning Trajectory=ROUND User Name= User Rights=full access

210 Type Of Motion=STEP AND SHOOT Source Temperature=35.86 °C Study Date and Time=Mar 02, 2018 11:56:15 Scan duration=00:09:43 [Sorting] Sorted by=DV Sorted at=Mar 03, 2018 19:24:33 Number of bins=4 PM channel=Software-based [Reconstruction] Reconstruction Program=NRecon Program Version=Version: 1.7.1.0 Program Home Directory=C:\Skyscan Reconstruction engine=GPUReconServer Engine version=Version: 1.7.1 Reconstruction from batch=Yes Postalignment=-1.00 Reconstruction servers= SCI-CMCA-W-D039 Dataset Origin=SkyScan1176 Dataset Prefix=p317_02032018_~#0004~ Dataset Directory=E:\Joe Yasa\PNX D7 IGF-1 only and IGF-1 plus Acriflavine treated-02- 03-2018\IGF-1 plus Acriflavine\P317\SB_Sorted Output Directory=E:\Joe Yasa\PNX D7 IGF-1 only and IGF-1 plus Acriflavine treated-02- 03-2018\IGF-1 plus Acriflavine\P317\SB_Sorted\p317_02032018_Rec Time and Date=Mar 03, 2018 19:55:14 First Section=23 Last Section=639 Reconstruction duration per slice (seconds)=0.019449 Total reconstruction time (617 slices) in seconds=12.000000 Section to Section Step=1

211 Sections Count=617 Result File Type=BMP Result File Header Length (bytes)=1134 Result Image Width (pixels)=1000 Result Image Height (pixels)=1000 Pixel Size (um)=35.57453 Reconstruction Angular Range (deg)=198.10 Use 180+=OFF Angular Step (deg)=0.7000 Smoothing=4 Smoothing kernel=0 (Asymmetrical boxcar) Ring Artifact Correction=4 Draw Scales=OFF Object Bigger than FOV=ON Reconstruction from ROI=OFF Filter cutoff relative to Nyquist frequency=100 Filter type=0 Filter type description=Hamming (Alpha=0.54) Undersampling factor=1 Threshold for defect pixel mask (%)=0 Beam Hardening Correction (%)=40 CS Static Rotation (deg)=0.00 Minimum for CS to Image Conversion=0.000000 Maximum for CS to Image Conversion=0.030000 HU Calibration=OFF BMP LUT=0 Cone-beam Angle Horiz.(deg)=16.725420 Cone-beam Angle Vert.(deg)=11.216640 [File name convention]

212 Filename Index Length=4 Filename Prefix=p317_02032018_~#0004~_rec

Appendix C- Image J Macro for analysing single coloured IHC stains title = getTitle(); run("Colour Deconvolution", "vectors=[User values] [r1]=0.45796692 [g1]=0.7366096 [b1]=0.49766722 [r2]=0.09787784 [g2]=0.8166144 [b2]=0.56882405 [r3]=0.268 [g3]=0.570 [b3]=0.776"); selectWindow(title + "-(Colour_1)"); run("Invert"); run("Blue"); rename(title+" -Bluechannel"); BlueImage = getTitle(); rename(title+" -Nuclei"); setThreshold(100, 255); //setThreshold(120, 255); setOption("BlackBackground", true); run("Convert to Mask"); run("Make Binary"); run("Watershed"); run("Analyze Particles...", "size=1-Infinity exclude summarize"); wait(100); selectWindow(title + "-(Colour_3)"); run("Invert"); run("Green"); rename(title+" -Greenchannel"); GreenImage = getTitle(); run("Duplicate...", " "); rename(title+" -CD31 "); setThreshold(100, 255);

213 //setThreshold(120, 255); setOption("BlackBackground", true); run("Convert to Mask"); run("Make Binary"); run("Fill Holes"); run("Dilate"); run("Close-"); run("Analyze Particles...", "size=5-Infinity exclude summarize"); run("Close All");

Appendix D- Image J Macro for analysing dual coloured IHC stains title = getTitle(); run("Colour Deconvolution", "vectors=[User values] [r1]=0.536 [g1]=0.712 [b1]=0.453 [r2]=0.110 [g2]=0.788 [b2]=0.606 [r3]=0.268 [g3]=0.570 [b3]=0.776"); selectWindow(title + "-(Colour_1)"); run("Invert"); run("Blue"); rename(title+" -Bluechannel"); BlueImage = getTitle(); run("Duplicate...", " "); rename(title+" -Nuclei analsysis"); run("Make Binary"); run("Fill Holes"); run("Dilate"); run("Close-"); run("Analyze Particles...", "size=20-Infinity exclude summarize"); wait(100); selectWindow(title + "-(Colour_2)"); run("Invert");

214 run("Red"); rename(title+" -Redchannel"); RedImage = getTitle(); selectWindow(title + "-(Colour_3)"); run("Invert"); run("Green"); rename(title+" -Greenchannel"); GreenImage = getTitle(); run("Duplicate...", " "); rename(title+" -SpC analsysis"); run("Make Binary"); run("Fill Holes"); run("Dilate"); run("Close-"); run("Analyze Particles...", "size=20-Infinity exclude summarize"); wait(100); run("Merge Channels...", "c1=["+RedImage+"] c2=["+GreenImage+"] create ignore"); selectWindow("Composite"); run("RGB Color"); // Color Thresholder 2.0.0-rc-49/1.51d // Autogenerated macro, single images only! min=newArray(3); max=newArray(3); filter=newArray(3); a=getTitle(); run("HSB Stack"); run("Convert Stack to Images"); selectWindow("Hue"); rename("0");

215 selectWindow("Saturation"); rename("1"); selectWindow("Brightness"); rename("2"); min[0]=20; max[0]=65; filter[0]="pass"; min[1]=0; max[1]=255; filter[1]="pass"; min[2]=80; max[2]=255; filter[2]="pass"; for (i=0;i<3;i++){ selectWindow(""+i); setThreshold(min[i], max[i]); run("Convert to Mask"); if (filter[i]=="stop") run("Invert"); } imageCalculator("AND create", "0","1"); imageCalculator("AND create", "Result of 0","2"); for (i=0;i<3;i++){ selectWindow(""+i); close(); } selectWindow("Result of 0"); close(); selectWindow("Result of Result of 0"); rename(a);

216 // Colour Thresholding------//setThreshold(255, 255); setOption("BlackBackground", true); run("Make Binary"); run("Fill Holes"); run("Dilate"); run("Close-"); // Rename binary image rename(title+" -Ki67-SpC analysis"); run("Analyze Particles...", "size=20-Infinity exclude summarize"); run("Close All");

Appendix E- Image J Macro for generating pseudocoloured fluorescent images title = getTitle(); run("Colour Deconvolution", "vectors=[User values] [r1]=0.536 [g1]=0.712 [b1]=0.453 [r2]=0.110 [g2]=0.788 [b2]=0.606 [r3]=0.268 [g3]=0.570 [b3]=0.776"); selectWindow(title + "-(Colour_1)"); run("Invert"); run("Blue"); rename(title+" -Bluechannel"); BlueImage = getTitle(); setMinAndMax(50, 255); //run("Brightness/Contrast..."); selectWindow(title + "-(Colour_2)"); run("Invert"); run("Red"); rename(title+" -Redchannel"); RedImage = getTitle(); setMinAndMax(30, 255); //run("Brightness/Contrast...");

217

Ms. α-SMA Ms. IgG2a Isotype control

Rb. CD-31 Rb. IgG Isotype control

Ki-67/SpC No. 1ºAb Control

Rb. IGF-1 Anti-IGF-1+ rhIGF-1 Control

Appendix F- Primary Antibody controls used in IHC Experiments.

In all IHC experiments, species-matched isotype negative controls were included to assess background staining caused by primary antibodies. This panel of figures illustrates examples of primary antibody stained sections and their corresponding negative controls imaged in the same region as the primary antibody stain. The Rabbit (Rb.) IgG isotype control was used to control for primary antibodies directed against CD-31, IGF-1, IGF-1R, IGFBP-3, pERK-1/2 and Ki-67. Mouse (Ms.) IgG2a isotype control was used to control for α-SMA actin staining. Where no suitable isotype control was available, a no primary antibody (1ºAb) negative control was used. In the case of IGF-1 stain, an additional negative control constituting a mixture of anti-IGF-1 adsorbed in 10x rhIGF-1 peptide was employed to assess anti-IGF-1 specificity. Scale bar in all images= 50µm.

218

Chapter 8

References

219 References

1. Butler, J.P., et al., Evidence for adult lung growth in humans. N Engl J Med, 2012.

367(3): p. 244-7.

2. Laros, C.D. and C.J. Westermann, Dilatation, compensatory growth, or both after

pneumonectomy during childhood and adolescence. A thirty-year follow-up study. J

Thorac Cardiovasc Surg, 1987. 93(4): p. 570-6.

3. Fehrenbach, H., et al., Neoalveolarisation contributes to compensatory lung growth

following pneumonectomy in mice. European Respiratory Journal, 2008. 31(3): p. 515-

522.

4. Paxson, J.A., et al., Age-dependent decline in mouse lung regeneration with loss of lung

fibroblast clonogenicity and increased myofibroblastic differentiation. PLoS One,

2011. 6(8): p. e23232.

5. Puche, J.E. and I. Castilla-Cortázar, Human conditions of insulin-like growth factor-I

(IGF-I) deficiency. Journal of Translational Medicine, 2012. 10: p. 224-224.

6. Pais, R.S., et al., Transcriptome analysis in prenatal IGF1-deficient mice identifies

molecular pathways and target genes involved in distal lung differentiation. PLoS One,

2013. 8(12): p. e83028.

7. McAnulty, R.J., et al., Growth factor activity in the lung during compensatory growth

after pneumonectomy: evidence of a role for IGF-1. Eur Respir J, 1992. 5(6): p. 739-

47.

8. Khadempour, M.H., et al., Changes of growth hormone, somatomedin C, and bombesin

following pneumonectomy. Exp Lung Res, 1992. 18(3): p. 421-32.

9. Nobuhara, K.K., et al., Insulin-like growth factor-I gene expression in three models of

accelerated lung growth. J Pediatr Surg, 1998. 33(7): p. 1057-60; discussion 1061.

220 10. Paxson, J.A., et al., Global gene expression patterns in the post-pneumonectomy lung

of adult mice. Respir Res, 2009. 10(1): p. 92.

11. Bidlingmaier, M., et al., Reference intervals for insulin-like growth factor-1 (igf-i) from

birth to senescence: results from a multicenter study using a new automated

chemiluminescence IGF-I immunoassay conforming to recent international

recommendations. J Clin Endocrinol Metab, 2014. 99(5): p. 1712-21.

12. Williams, R.M., et al., Mecasermin rinfabate: rhIGF-I/rhIGFBP-3 complex: iPLEX.

Expert Opin Drug Metab Toxicol, 2008. 4(3): p. 311-24.

13. Institute, W. Lung Diseases in Australia. 2014; Available from:

http://lungfoundation.com.au/wp-content/uploads/2014/10/LUNG-DISEASE-IN-

AUSTRALIA-REPORT_Final_22October14.pdf.

14. Sun, Z., et al., Stem cell therapies for chronic obstructive pulmonary disease: current

status of pre-clinical studies and clinical trials. Journal of Thoracic Disease, 2018.

10(2): p. 1084-1098.

15. Cao, J., et al., Survival Rates After Lobectomy, Segmentectomy, and Wedge Resection

for Non-Small Cell Lung Cancer. Ann Thorac Surg, 2018. 105(5): p. 1483-1491.

16. Paraskeva, M.A., et al., Lung transplantation in Australia, 1986-2018: more than 30

years in the making. Med J Aust, 2018. 208(10): p. 445-450.

17. Hogan, B.L., et al., Repair and regeneration of the respiratory system: complexity,

plasticity, and mechanisms of lung stem cell function. Cell Stem Cell, 2014. 15(2): p.

123-38.

18. Garcia, O., et al., Cell-based therapies for lung disease. British Medical Bulletin, 2012.

101(1): p. 147-161.

19. Petersen, T.H., E.A. Calle, and L.E. Niklason, Strategies for lung regeneration.

Materials Today, 2011. 14(5): p. 196-201.

221 20. Yang, C., et al., Stem/progenitor cells in endogenous repairing responses: new toolbox

for the treatment of acute lung injury. J Transl Med, 2016. 14: p. 47.

21. Yang, J. and Z. Jia, Cell-based therapy in lung regenerative medicine. Regen Med Res,

2014. 2(1): p. 7.

22. Cao, X.-Y. and S.-Y. Xiao, Chronic lung disease, lung regeneration and future

therapeutic strategies. Chronic Diseases and Translational Medicine, 2018. 4(2): p.

103-108.

23. Calle, E.A., et al., Strategies for Whole Lung Tissue Engineering. IEEE transactions on

bio-medical engineering, 2014. 61(5): p. 1482-1496.

24. Cargnoni, A., et al., Transplantation of allogeneic and xenogeneic placenta-derived

cells reduces bleomycin-induced lung fibrosis. Cell Transplant, 2009. 18(4): p. 405-22.

25. Rojas, M., et al., Bone marrow-derived mesenchymal stem cells in repair of the injured

lung. Am J Respir Cell Mol Biol, 2005. 33(2): p. 145-52.

26. Serikov, V.B., et al., Evidence of Temporary Airway Epithelial Repopulation and Rare

Clonal Formation by BM-derived Cells Following Naphthalene Injury in Mice. The

Anatomical Record, 2007. 290(9): p. 1033-1045.

27. Chan, M.C.W., et al., Human mesenchymal stromal cells reduce influenza A H5N1-

associated acute lung injury in vitro and in vivo. Proceedings of the National Academy

of Sciences of the United States of America, 2016. 113(13): p. 3621-3626.

28. Wang, D., et al., Transplantation of human embryonic stem cell-derived alveolar

epithelial type II cells abrogates acute lung injury in mice. Mol Ther, 2010. 18(3): p.

625-34.

29. Lim, R., et al., First-In-Human Administration of Allogeneic Amnion Cells in

Premature Infants With Bronchopulmonary Dysplasia: A Safety Study. Stem Cells

Transl Med, 2018. 7(9): p. 628-635.

222 30. Ott, H.C., et al., Regeneration and orthotopic transplantation of a bioartificial lung.

Nat Med, 2010. 16(8): p. 927-33.

31. Song, J.J., et al., Enhanced in vivo function of bioartificial lungs in rats. Ann Thorac

Surg, 2011. 92(3): p. 998-1005; discussion 1005-6.

32. Doi, R., et al., Transplantation of bioengineered rat lungs recellularized with

endothelial and adipose-derived stromal cells. Scientific Reports, 2017. 7: p. 8447.

33. Macchiarini, P., et al., Clinical transplantation of a tissue-engineered airway. Lancet,

2008. 372(9655): p. 2023-30.

34. Grillo, H.C., Tracheal replacement: a critical review. Ann Thorac Surg, 2002. 73(6):

p. 1995-2004.

35. OpenStax, Anatomy & Physiology, O. CNX, Editor. Feb 26, 2016

36. Fernández, L.G., et.al, Compensatory Lung Growth After Pneumonectomy, in Topics in

Thoracic Surgery, P.F.G. Cardoso, Editor. 2012, IntechOpen. p. 415-432.

37. Harding R, P.K.E.a.P.C.G., The Lung Development, Aging and the Environment. 2004,

London, U.K: Elsevier Academic Press.

38. Peer, M., et al., Morbidity and mortality after major pulmonary resections in patients

with locally advanced stage IIIA non-small cell lung carcinoma who underwent

induction therapy. Heart Lung Circ, 2015. 24(1): p. 69-76.

39. Shepherd, M.P., Plombage in the 1980s. Thorax, 1985. 40(5): p. 328-40.

40. Hoffman, A.M., et al., Matrix modulation of compensatory lung regrowth and

progenitor cell proliferation in mice. Am J Physiol Lung Cell Mol Physiol, 2010.

298(2): p. L158-68.

41. Hsia, C.C.W., et al., Preventing mediastinal shift after pneumonectomy impairs

regenerative alveolar tissue growth. American Journal of Physiology - Lung Cellular

and Molecular Physiology, 2001. 281(5 25-5): p. L1279-L1287.

223 42. Hsia, C.C., Signals and mechanisms of compensatory lung growth. J Appl Physiol

(1985), 2004. 97(5): p. 1992-8.

43. Hsia, C.C., et al., Compensatory lung growth occurs in adult dogs after right

pneumonectomy. J Clin Invest, 1994. 94(1): p. 405-12.

44. McBride, J.T., Lung volumes after an increase in lung distension in pneumonectomized

ferrets. J Appl Physiol (1985), 1989. 67(4): p. 1418-21.

45. Langston, C., et al., Alveolar multiplication in the contralateral lung after unilateral

pneumonectomy in the rabbit. Am Rev Respir Dis, 1977. 115(1): p. 7-13.

46. Thane, K., E.P. Ingenito, and A.M. Hoffman, Lung regeneration and translational

implications of the postpneumonectomy model. Transl Res, 2014. 163(4): p. 363-76.

47. Ackermann, M., et al., Sprouting and intussusceptive angiogenesis in

postpneumonectomy lung growth: mechanisms of alveolar neovascularization.

Angiogenesis, 2014. 17(3): p. 541-51.

48. Hsia, C.C., Comparative analysis of the mechanical signals in lung development and

compensatory growth. Cell Tissue Res, 2017. 367(3): p. 687-705.

49. Schittny, J.C., Development of the lung. Cell Tissue Res, 2017. 367(3): p. 427-444.

50. Burri, P.H., Structural aspects of postnatal lung development - alveolar formation and

growth. Biol Neonate, 2006. 89(4): p. 313-22.

51. Irvin, C.G. and J.H.T. Bates, Measuring the lung function in the mouse: the challenge

of size. Respiratory Research, 2003. 4(1): p. 4-4.

52. Voswinckel, R., et al., Characterisation of post-pneumonectomy lung growth in adult

mice. Eur Respir J, 2004. 24(4): p. 524-32.

53. Wang, W., et al., Longitudinal, noninvasive monitoring of compensatory lung growth

in mice after pneumonectomy via (3)He and (1)H magnetic resonance imaging. Am J

Respir Cell Mol Biol, 2013. 49(5): p. 697-703.

224 54. Fehrenbach, H., et al., Neoalveolarisation contributes to compensatory lung growth

following pneumonectomy in mice. Eur Respir J, 2008. 31(3): p. 515-22.

55. Ysasi, A.B., et al., Remodeling of alveolar septa after murine pneumonectomy.

American journal of physiology. Lung cellular and molecular physiology, 2015.

308(12): p. L1237-L1244.

56. Gibney, B.C., et al., Detection of murine post-pneumonectomy lung regeneration by

18FDG PET imaging. EJNMMI Res, 2012. 2(1): p. 48.

57. Gibney, B.C., et al., Mechanostructural adaptations preceding postpneumonectomy

lung growth. Exp Lung Res, 2012. 38(8): p. 396-405.

58. Liu, S., J. Cimprich, and B.M. Varisco, Mouse pneumonectomy model of compensatory

lung growth. J Vis Exp, 2014(94).

59. Tata, P.R. and J. Rajagopal, Plasticity in the lung: making and breaking cell identity.

Development, 2017. 144(5): p. 755-766.

60. ad hoc Statement Committee, A.T.S., Mechanisms and limits of induced postnatal lung

growth. Am J Respir Crit Care Med, 2004. 170(3): p. 319-43.

61. Carlin, J.I., et al., Recruitment of lung diffusing capacity with exercise before and after

pneumonectomy in dogs. J Appl Physiol (1985), 1991. 70(1): p. 135-42.

62. Paisley, D., et al., The pneumonectomy model of compensatory lung growth: insights

into lung regeneration. Pharmacol Ther, 2014. 142(2): p. 196-205.

63. Ravikumar, P., et al., Separating in vivo mechanical stimuli for postpneumonectomy

compensation: imaging and ultrastructural assessment. J Appl Physiol (1985), 2013.

114(8): p. 961-70.

64. Hsia, C.C. and R.L. Johnson, Jr., Further examination of alveolar septal adaptation to

left pneumonectomy in the adult lung. Respir Physiol Neurobiol, 2006. 151(2-3): p.

167-77.

225 65. Dane, D.M., et al., Separating in vivo mechanical stimuli for postpneumonectomy

compensation: physiological assessment. J Appl Physiol (1985), 2013. 114(1): p. 99-

106.

66. Yilmaz, C., et al., Progressive adaptation in regional parenchyma mechanics following

extensive lung resection assessed by functional computed tomography. J Appl Physiol

(1985), 2011. 111(4): p. 1150-8.

67. Dane, D.M., et al., Perfusion-related stimuli for compensatory lung growth following

pneumonectomy. J Appl Physiol (1985), 2016. 121(1): p. 312-23.

68. Hsia, C.C., et al., Preventing mediastinal shift after pneumonectomy impairs

regenerative alveolar tissue growth. Am J Physiol Lung Cell Mol Physiol, 2001.

281(5): p. L1279-87.

69. Ysasi, A.B., et al., Effect of unilateral diaphragmatic paralysis on postpneumonectomy

lung growth. Am J Physiol Lung Cell Mol Physiol, 2013. 305(6): p. L439-45.

70. Nobuhara, K.K., et al., Continuous intrapulmonary distension with perfluorocarbon

accelerates neonatal (but not adult) lung growth. J Pediatr Surg, 1998. 33(2): p. 292-8.

71. Sekhon, H.S., C. Smith, and W.M. Thurlbeck, Effect of hypoxia and hyperoxia on

postpneumonectomy compensatory lung growth. Exp Lung Res, 1993. 19(5): p. 519-

32.

72. Li, D., et al., Upregulation of hypoxia-induced mitogenic factor in compensatory lung

growth after pneumonectomy. Am J Respir Cell Mol Biol, 2005. 32(3): p. 185-91.

73. Vadivel, A., et al., Hypoxia-inducible factors promote alveolar development and

regeneration. Am J Respir Cell Mol Biol, 2014. 50(1): p. 96-105.

74. Zhang, Q., et al., Postpneumonectomy lung expansion elicits hypoxia-inducible factor-

1alpha signaling. Am J Physiol Lung Cell Mol Physiol, 2007. 293(2): p. L497-504.

226 75. Foster, D.J., O.W. Moe, and C.C. Hsia, Upregulation of erythropoietin receptor during

postnatal and postpneumonectomy lung growth. Am J Physiol Lung Cell Mol Physiol,

2004. 287(6): p. L1107-15.

76. McBride, J.T., et al., Lung Growth and Airway Function after Lobectomy in Infancy for

Congenital Lobar Emphysema. Journal of Clinical Investigation, 1980. 66(5): p. 962-

970.

77. Nakajima, C., et al., Longitudinal follow-up of pulmonary function after lobectomy in

childhood - factors affecting lung growth. Pediatr Surg Int, 1998. 13(5-6): p. 341-5.

78. Yoshitomi, A., et al., [Compensatory growth of residual lung after pneumonectomy in

childhood]. Nihon Kokyuki Gakkai Zasshi, 2000. 38(8): p. 642-4.

79. Wolff, J.-C., et al., Comparative gene expression profiling of post-natal and post-

pneumonectomy lung growth. European Respiratory Journal, 2010. 35(3): p. 655-666.

80. Paxson, J.A., et al., Age dependence of lung mesenchymal stromal cell dynamics

following pneumonectomy. Stem Cells Dev, 2013. 22(24): p. 3214-25.

81. Jackson, S.-R., et al., Partial pneumonectomy of telomerase null mice carrying

shortened telomeres initiates cell growth arrest resulting in a limited compensatory

growth response. American Journal of Physiology - Lung Cellular and Molecular

Physiology, 2011. 300(6): p. L898-L909.

82. Brody, J.S., et al., Acromegalic pneumonomegaly: lung growth in the adult. J Clin

Invest, 1970. 49(6): p. 1051-60.

83. Ghammad, K., P. Fievez, and P. Remy, Compensatory Lung Growth after

Pneumonectomy: Case Report. Acta Chir Belg, 2015. 115(3): p. 234-6.

84. Lucas G. Fernández, J.M.I. and D.R.J.a.V.E. Laubach,

pneumonectomy compensatory lung growth in pulmonary physiology and disease.pdf>.

Topics in Thoracic Surgery, 2012.

227 85. Silverman, E.S. and T. Collins, Pathways of Egr-1-Mediated Gene Transcription in

Vascular Biology. The American Journal of Pathology, 1999. 154(3): p. 665-670.

86. Gilbert, K.A. and D.E. Rannels, Increased lung inflation induces gene expression after

pneumonectomy. Am J Physiol, 1998. 275(1 Pt 1): p. L21-9.

87. Landesberg, L.J., et al., Upregulation of transcription factors in lung in the early phase

of postpneumonectomy lung growth. Am J Physiol Lung Cell Mol Physiol, 2001.

281(5): p. L1138-49.

88. Liu, Z., et al., MAPK-Mediated YAP Activation Controls Mechanical-Tension-Induced

Pulmonary Alveolar Regeneration. Cell Rep, 2016. 16(7): p. 1810-9.

89. Takahashi, Y., et al., Thyroid transcription factor-1 influences the early phase of

compensatory lung growth in adult mice. Am J Respir Crit Care Med, 2010. 181(12):

p. 1397-406.

90. Takahashi, Y., et al., Nuclear factor-kappa B influences early phase of compensatory

lung growth after pneumonectomy in mice. J Biomed Sci, 2017. 24(1): p. 41.

91. Pagel, J.I. and E. Deindl, Early growth response 1--a transcription factor in the

crossfire of signal transduction cascades. Indian J Biochem Biophys, 2011. 48(4): p.

226-35.

92. Abdulkadir, S.A., et al., Impaired prostate tumorigenesis in Egr1-deficient mice. Nat

Med, 2001. 7(1): p. 101-7.

93. Bae, S.K., et al., Egr-1 mediates transcriptional activation of IGF-II gene in response

to hypoxia. Cancer Res, 1999. 59(23): p. 5989-94.

94. Santiago, F.S., et al., Early growth response factor-1 induction by injury is triggered

by release and paracrine activation by fibroblast growth factor-2. Am J Pathol, 1999.

154(3): p. 937-44.

228 95. Wung, B.S., et al., Modulation of Ras/Raf/extracellular signal-regulated kinase

pathway by reactive oxygen species is involved in cyclic strain-induced early growth

response-1 gene expression in endothelial cells. Circ Res, 1999. 84(7): p. 804-12.

96. Chiu, J.J., et al., Nitric oxide regulates shear stress-induced early growth response-1.

Expression via the extracellular signal-regulated kinase pathway in endothelial cells.

Circ Res, 1999. 85(3): p. 238-46.

97. Fei, J., et al., Atherogenic omega-6 Lipids Modulate PPAR- EGR-1 Crosstalk in

Vascular Cells. PPAR Res, 2011. 2011: p. 753917.

98. Lee, K., et al., Acriflavine inhibits HIF-1 dimerization, tumor growth, and

vascularization. Proceedings of the National Academy of Sciences of the United States

of America, 2009. 106(42): p. 17910-17915.

99. Kim, C.H., et al., Early expression of myocardial HIF-1alpha in response to

mechanical stresses: regulation by stretch-activated channels and the

phosphatidylinositol 3-kinase signaling pathway. Circ Res, 2002. 90(2): p. E25-33.

100. Treins, C., et al., Regulation of hypoxia-inducible factor (HIF)-1 activity and

expression of HIF hydroxylases in response to insulin-like growth factor I. Mol

Endocrinol, 2005. 19(5): p. 1304-17.

101. Haematology, A.o.G.a.C.i.O.a. EGR1 (Early Growth Response 1). 03rd September

2018]; Available from: http://atlasgeneticsoncology.org/Genes/GC_EGR1.html.

102. Morrisey, E.E. and B.L. Hogan, Preparing for the first breath: genetic and cellular

mechanisms in lung development. Dev Cell, 2010. 18(1): p. 8-23.

103. Kotton, D.N. and E.E. Morrisey, Lung regeneration: mechanisms, applications and

emerging stem cell populations. Nat Med, 2014. 20(8): p. 822-32.

104. Beers, M.F. and E.E. Morrisey, The three R's of lung health and disease: repair,

remodeling, and regeneration. J Clin Invest, 2011. 121(6): p. 2065-73.

229 105. Giangreco, A., S.D. Reynolds, and B.R. Stripp, Terminal bronchioles harbor a unique

airway stem cell population that localizes to the bronchoalveolar duct junction. Am J

Pathol, 2002. 161(1): p. 173-82.

106. Hong, K.U., et al., Clara cell secretory protein-expressing cells of the airway

neuroepithelial body microenvironment include a label-retaining subset and are

critical for epithelial renewal after progenitor cell depletion. Am J Respir Cell Mol

Biol, 2001. 24(6): p. 671-81.

107. Reynolds, S.D., et al., Neuroepithelial bodies of pulmonary airways serve as a reservoir

of progenitor cells capable of epithelial regeneration. Am J Pathol, 2000. 156(1): p.

269-78.

108. Rawlins, E.L., et al., The role of Scgb1a1+ Clara cells in the long-term maintenance

and repair of lung airway, but not alveolar, epithelium. Cell Stem Cell, 2009. 4(6): p.

525-34.

109. Rock, J.R., et al., Multiple stromal populations contribute to pulmonary fibrosis without

evidence for epithelial to mesenchymal transition. Proc Natl Acad Sci U S A, 2011.

108(52): p. E1475-83.

110. Zheng, D., et al., Regeneration of alveolar type I and II cells from Scgb1a1-expressing

cells following severe pulmonary damage induced by bleomycin and influenza. PLoS

One, 2012. 7(10): p. e48451.

111. Zheng, D., et al., A cellular pathway involved in Clara cell to alveolar type II cell

differentiation after severe lung injury. PLoS One, 2013. 8(8): p. e71028.

112. Kim, C.F., et al., Identification of bronchioalveolar stem cells in normal lung and lung

cancer. Cell, 2005. 121(6): p. 823-35.

230 113. Nolen-Walston, R.D., et al., Cellular kinetics and modeling of bronchioalveolar stem

cell response during lung regeneration. Am J Physiol Lung Cell Mol Physiol, 2008.

294(6): p. L1158-65.

114. Evans, M.J., et al., Transformation of alveolar type 2 cells to type 1 cells following

exposure to NO2. Exp Mol Pathol, 1975. 22(1): p. 142-50.

115. Kaplan, H.P., et al., Pathogenesis and reversibility of the pulmonary lesions of oxygen

toxicity in monkeys. I. Clinical and light microscopic studies. Lab Invest, 1969. 20(1):

p. 94-100.

116. Barkauskas, C.E., et al., Type 2 alveolar cells are stem cells in adult lung. J Clin Invest,

2013. 123(7): p. 3025-36.

117. Chamoto, K., et al., Alveolar Epithelial Dynamics in Postpneumonectomy Lung

Growth. The Anatomical Record, 2013. 296(3): p. 495-503.

118. Wada, H., et al., Transplantation of alveolar type II cells stimulates lung regeneration

during compensatory lung growth in adult rats. J Thorac Cardiovasc Surg, 2012.

143(3): p. 711-719 e2.

119. Chapman, H.A., et al., Integrin alpha6beta4 identifies an adult distal lung epithelial

population with regenerative potential in mice. J Clin Invest, 2011. 121(7): p. 2855-62.

120. Jain, R., et al., Plasticity of Hopx(+) type I alveolar cells to regenerate type II cells in

the lung. Nat Commun, 2015. 6: p. 6727.

121. Takeda, N., et al., Hopx expression defines a subset of multipotent hair follicle stem

cells and a progenitor population primed to give rise to K6+ niche cells. Development,

2013. 140(8): p. 1655-64.

122. Treutlein, B., et al., Reconstructing lineage hierarchies of the distal lung epithelium

using single-cell RNA-seq. Nature, 2014. 509(7500): p. 371-5.

231 123. Wang, Y., et al., Pulmonary alveolar type I cell population consists of two distinct

subtypes that differ in cell fate. Proceedings of the National Academy of Sciences,

2018. 115(10): p. 2407-2412.

124. Kumar, P.A., et al., Distal airway stem cells yield alveoli in vitro and during lung

regeneration following H1N1 influenza infection. Cell, 2011. 147(3): p. 525-38.

125. Vaughan, A.E., et al., Lineage-negative progenitors mobilize to regenerate lung

epithelium after major injury. Nature, 2015. 517(7536): p. 621-5.

126. Ray, S., et al., Rare SOX2+ Airway Progenitor Cells Generate KRT5+ Cells that

Repopulate Damaged Alveolar Parenchyma following Influenza Virus Infection. Stem

Cell Reports, 2016. 7(5): p. 817-825.

127. Zuo, W., et al., p63(+)Krt5(+) distal airway stem cells are essential for lung

regeneration. Nature, 2015. 517(7536): p. 616-20.

128. McGowan, S.E., et al., Platelet-derived growth factor receptor-alpha-expressing cells

localize to the alveolar entry ring and have characteristics of myofibroblasts during

pulmonary alveolar septal formation. Anat Rec (Hoboken), 2008. 291(12): p. 1649-61.

129. Chen, L., et al., Dynamic regulation of platelet-derived growth factor receptor alpha

expression in alveolar fibroblasts during realveolarization. Am J Respir Cell Mol Biol,

2012. 47(4): p. 517-27.

130. Perl, A.K. and E. Gale, FGF signaling is required for myofibroblast differentiation

during alveolar regeneration. Am J Physiol Lung Cell Mol Physiol, 2009. 297(2): p.

L299-308.

131. Burgess, H.A., et al., PPARgamma agonists inhibit TGF-beta induced pulmonary

myofibroblast differentiation and collagen production: implications for therapy of lung

fibrosis. Am J Physiol Lung Cell Mol Physiol, 2005. 288(6): p. L1146-53.

232 132. Milam, J.E., et al., PPAR-gamma agonists inhibit profibrotic phenotypes in human lung

fibroblasts and bleomycin-induced pulmonary fibrosis. Am J Physiol Lung Cell Mol

Physiol, 2008. 294(5): p. L891-901.

133. Wu, M., et al., Rosiglitazone abrogates bleomycin-induced scleroderma and blocks

profibrotic responses through peroxisome proliferator-activated receptor-gamma. Am

J Pathol, 2009. 174(2): p. 519-33.

134. Chamoto, K., et al., CD34+ progenitor to endothelial cell transition in post-

pneumonectomy angiogenesis. Am J Respir Cell Mol Biol, 2012. 46(3): p. 283-9.

135. Konerding, M.A., et al., Spatial dependence of alveolar angiogenesis in post-

pneumonectomy lung growth. Angiogenesis, 2012. 15(1): p. 23-32.

136. Jancelewicz, T., et al., Vascular endothelial growth factor isoform and receptor

expression during compensatory lung growth. J Surg Res, 2010. 160(1): p. 107-13.

137. Lin, M., et al., Angiogenesis gene expression in murine endothelial cells during post-

pneumonectomy lung growth. Respir Res, 2011. 12: p. 98.

138. Leuwerke, S.M., et al., Inhibition of compensatory lung growth in endothelial nitric

oxide synthase-deficient mice. Am J Physiol Lung Cell Mol Physiol, 2002. 282(6): p.

L1272-8.

139. Ding, B.S., et al., Endothelial-derived angiocrine signals induce and sustain

regenerative lung alveolarization. Cell, 2011. 147(3): p. 539-53.

140. Rafii, S., et al., Platelet-derived SDF-1 primes the pulmonary capillary vascular niche

to drive lung alveolar regeneration. Nat Cell Biol, 2015. 17(2): p. 123-36.

141. Tonkin, J., et al., Monocyte/Macrophage-derived IGF-1 Orchestrates Murine Skeletal

Muscle Regeneration and Modulates Autocrine Polarization. Mol Ther, 2015. 23(7): p.

1189-1200.

233 142. Elsegood, C.L., et al., Kupffer cell-monocyte communication is essential for initiating

murine liver progenitor cell-mediated liver regeneration. Hepatology, 2015.

143. Cao, Q., D.C. Harris, and Y. Wang, Macrophages in kidney injury, inflammation, and

fibrosis. Physiology (Bethesda), 2015. 30(3): p. 183-94.

144. Gow, D.J., D.P. Sester, and D.A. Hume, CSF-1, IGF-1, and the control of postnatal

growth and development. J Leukoc Biol, 2010. 88(3): p. 475-81.

145. Chamoto, K., et al., Alveolar macrophage dynamics in murine lung regeneration. J Cell

Physiol, 2012. 227(9): p. 3208-15.

146. Chamoto, K., et al., Migration of CD11b+ accessory cells during murine lung

regeneration. Stem Cell Res, 2013. 10(3): p. 267-77.

147. Lechner, A.J., et al., Recruited Monocytes and Type 2 Immunity Promote Lung

Regeneration following Pneumonectomy. Cell Stem Cell, 2017.

148. Kaza, A.K., et al., Epidermal growth factor augments postpneumonectomy lung

growth. J Thorac Cardiovasc Surg, 2000. 120(5): p. 916-21.

149. Kaza, A.K., et al., Keratinocyte growth factor enhances post-pneumonectomy lung

growth by alveolar proliferation. Circulation, 2002. 106(12 Suppl 1): p. I120-4.

150. Karapolat, S., et al., Effects of retinoic acid on compensatory lung growth. Journal of

Cardiothoracic Surgery, 2008. 3: p. 37-37.

151. Sakurai, M.K., et al., Vascular endothelial growth factor accelerates compensatory

lung growth after unilateral pneumonectomy. Am J Physiol Lung Cell Mol Physiol,

2007. 292(3): p. L742-7.

152. Sakamaki, Y., et al., Hepatocyte growth factor stimulates proliferation of respiratory

epithelial cells during postpneumonectomy compensatory lung growth in mice. Am J

Respir Cell Mol Biol, 2002. 26(5): p. 525-33.

234 153. Yuan, S., et al., A role for platelet-derived growth factor-BB in rat postpneumonectomy

compensatory lung growth. Pediatr Res, 2002. 52(1): p. 25-33.

154. Philippou, A., et al., Expression of IGF-1 isoforms after exercise-induced muscle

damage in humans: characterization of the MGF E peptide actions in vitro. In Vivo,

2009. 23(4): p. 567-75.

155. Kaza, A.K., et al., Epidermal growth factor receptor up-regulation is associated with

lung growth after lobectomy. Ann Thorac Surg, 2001. 72(2): p. 380-5.

156. Price, W.A., et al., Expression of the insulin-like growth factor system in

postpneumonectomy lung growth. Exp Lung Res, 1998. 24(2): p. 203-17.

157. Thurlbeck, W.M., D.E. AJ, and B.T. Smith, Serum somatomedin C concentrations

following pneumonectomy. Am Rev Respir Dis, 1984. 130(3): p. 499-500.

158. Matsumoto, K., et al., Keratinocyte growth factor accelerates compensatory growth in

the remaining lung after trilobectomy in rats. J Thorac Cardiovasc Surg, 2009. 137(6):

p. 1499-507.

159. Green, J., et al., Diversity of Interstitial Lung Fibroblasts Is Regulated by Platelet-

Derived Growth Factor Receptor alpha Kinase Activity. Am J Respir Cell Mol Biol,

2016. 54(4): p. 532-45.

160. Kaza, A.K., et al., Retinoic acid enhances lung growth after pneumonectomy. Ann

Thorac Surg, 2001. 71(5): p. 1645-50.

161. Yan, X., et al., Lack of response to all-trans retinoic acid supplementation in adult dogs

following left pneumonectomy. J Appl Physiol (1985), 2005. 99(5): p. 1681-8.

162. Yan, X., et al., Retinoic acid induces nonuniform alveolar septal growth after right

pneumonectomy. J Appl Physiol (1985), 2004. 96(3): p. 1080-9.

235 163. Ravikumar, P., et al., Long-term post-pneumonectomy pulmonary adaptation following

all-trans-retinoic acid supplementation. J Appl Physiol (1985), 2011. 110(3): p. 764-

73.

164. Dao, D.T., et al., Intranasal delivery of VEGF enhances compensatory lung growth in

mice. PLoS One, 2018. 13(6): p. e0198700.

165. Matsui, Y., et al., The role of vascular endothelial growth factor receptor-1 signaling

in compensatory contralateral lung growth following unilateral pneumonectomy. Lab

Invest, 2015. 95(5): p. 456-68.

166. Kajimoto, Y. and P. Rotwein, Evolution of insulin-like growth factor I (IGF-I):

structure and expression of an IGF-I precursor from Xenopus laevis. Mol Endocrinol,

1990. 4(2): p. 217-26.

167. Salmon, W.D., Jr. and W.H. Daughaday, A hormonally controlled serum factor which

stimulates sulfate incorporation by cartilage in vitro. J Lab Clin Med, 1957. 49(6): p.

825-36.

168. Daughaday, W.H., et al., On the nomenclature of the somatomedins and insulin-like

growth factors. J Clin Endocrinol Metab, 1987. 65(5): p. 1075-6.

169. El-Shewy, H.M., et al., The Insulin-like Growth Factor Type 1 and Insulin-like Growth

Factor Type 2/Mannose-6-phosphate Receptors Independently Regulate ERK1/2

Activity in HEK293 Cells. Journal of Biological Chemistry, 2007. 282(36): p. 26150-

26157.

170. Denley, A., et al., Molecular interactions of the IGF system. Cytokine & Growth Factor

Reviews, 2005. 16(4–5): p. 421-439.

171. Hawkes, C. and S. Kar, The insulin-like growth factor-II/mannose-6-phosphate

receptor: structure, distribution and function in the central nervous system. Brain

Research Reviews, 2004. 44(2–3): p. 117-140.

236 172. Gilmour, R.S., et al., From animal to molecule: aspects of the biology of insulin-like

growth factors. The British Journal of Cancer. Supplement, 1988. 9: p. 23-30.

173. O'Dell, S.D. and I.N.M. Day, Molecules in focus Insulin-like growth factor II (IGF-II).

The International Journal of Biochemistry & Cell Biology, 1998. 30(7): p. 767-771.

174. Adams, T.E., et al., Structure and function of the type 1 insulin-like growth factor

receptor. Cell Mol Life Sci, 2000. 57(7): p. 1050-93.

175. Dalcik, H., et al., Expression of insulin-like growth factor in the placenta of intrauterine

growth-retarded human fetuses. Acta Histochem, 2001. 103(2): p. 195-207.

176. Lupu, F., et al., Roles of growth hormone and insulin-like growth factor 1 in mouse

postnatal growth. Dev Biol, 2001. 229(1): p. 141-62.

177. Liu, J.-P., et al., Mice carrying null mutations of the genes encoding insulin-like growth

factor I (Igf-1) and type 1 IGF receptor (Igf1r). Cell, 1993. 75(1): p. 59-72.

178. Butler, A.A., et al., Insulin-like growth factor-I receptor signal transduction: at the

interface between physiology and cell biology. Comp Biochem Physiol B Biochem Mol

Biol, 1998. 121(1): p. 19-26.

179. Adams, T.E., et al., Structure and function of the type 1 insulin-like growth factor

receptor. Cellular and Molecular Life Sciences, 2000. 57(7): p. 1050-93.

180. Stewart, C.E. and P. Rotwein, Growth, differentiation, and survival: multiple

physiological functions for insulin-like growth factors. Physiological Reviews, 1996.

76(4): p. 1005-1026.

181. Kavran, J.M., et al., How IGF-1 activates its receptor, ed. J. Kuriyan. Vol. 3. 2014.

182. Zha, J. and M.R. Lackner, Targeting the Insulin-like Growth Factor Receptor-1R

Pathway for Cancer Therapy. Clinical Cancer Research, 2010. 16(9): p. 2512-2517.

237 183. Ludwig, T., et al., Mouse mutants lacking the type 2 IGF receptor (IGF2R) are rescued

from perinatal lethality in Igf2 and Igf1r null backgrounds. Dev Biol, 1996. 177(2): p.

517-35.

184. Duan, C. and Q. Xu, Roles of insulin-like growth factor (IGF) binding proteins in

regulating IGF actions. General and Comparative Endocrinology, 2005. 142(1–2): p.

44-52.

185. Yamada, P.M. and K.-W. Lee, Perspectives in mammalian IGFBP-3 biology: local vs.

systemic action. American Journal of Physiology - Cell Physiology, 2009. 296(5): p.

C954-C976.

186. Zapf, J., et al., Inhibition of the action of nonsuppressible insulin-like activity on

isolated rat fat cells by binding to its carrier protein. J Clin Invest, 1979. 63(5): p. 1077-

84.

187. Laron, Z., Insulin-like growth factor 1 (IGF-1): a growth hormone. Molecular

Pathology, 2001. 54(5): p. 311-316.

188. Miyakoshi, N., et al., Evidence that IGF-binding protein-5 functions as a growth factor.

J Clin Invest, 2001. 107(1): p. 73-81.

189. Andress, D.L. and R.S. Birnbaum, Human osteoblast-derived insulin-like growth factor

(IGF) binding protein-5 stimulates osteoblast mitogenesis and potentiates IGF action.

J Biol Chem, 1992. 267(31): p. 22467-72.

190. Huang, S.S., et al., Cellular growth inhibition by IGFBP-3 and TGF-beta1 requires

LRP-1. FASEB J, 2003. 17(14): p. 2068-81.

191. Schedlich, L.J., et al., Nuclear import of insulin-like growth factor-binding protein-3

and -5 is mediated by the importin beta subunit. J Biol Chem, 2000. 275(31): p. 23462-

70.

238 192. Jones, J.I., et al., 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, 1993. 90(22): p. 10553-7.

193. Ohlsson, C., et al., The role of liver-derived insulin-like growth factor-I. Endocr Rev,

2009. 30(5): p. 494-535.

194. Uchiyama, T., et al., Negative regulation of hypothalamic growth hormone-releasing

factor messenger ribonucleic acid by growth hormone and insulin-like growth factor I.

Neuroendocrinology, 1994. 59(5): p. 441-50.

195. Wood, T.L., et al., Hormonal regulation of rat hypothalamic neuropeptide mRNAs:

effect of hypophysectomy and hormone replacement on growth-hormone-releasing

factor, somatostatin and the insulin-like growth factors. Neuroendocrinology, 1991.

53(3): p. 298-305.

196. Ji, S., et al., Insulin inhibits growth hormone signaling via the growth hormone

receptor/JAK2/STAT5B pathway. J Biol Chem, 1999. 274(19): p. 13434-42.

197. Yakar, S., et al., Normal growth and development in the absence of hepatic insulin-like

growth factor I. Proceedings of the National Academy of Sciences, 1999. 96(13): p.

7324-7329.

198. Allen, J.T., et al., Expression of growth hormone-releasing factor, growth hormone,

insulin-like growth factor-1 and its binding proteins in human lung. Neuropeptides,

2000. 34(2): p. 98-107.

199. Wang, Z., et al., Insulin-Like Growth Factor-1 Signaling in Lung Development and

Inflammatory Lung Diseases. Biomed Res Int, 2018. 2018: p. 6057589.

200. Lopez, I.P., et al., Involvement of Igf1r in Bronchiolar Epithelial Regeneration: Role

during Repair Kinetics after Selective Club Cell Ablation. PLoS One, 2016. 11(11): p.

e0166388.

239 201. Westbrook, R., et al., Metabolic alterations due to caloric restriction and every other

day feeding in normal and growth hormone receptor knockout mice. J Gerontol A Biol

Sci Med Sci, 2014. 69(1): p. 25-33.

202. Bonkowski, M.S., et al., Targeted disruption of growth hormone receptor interferes

with the beneficial actions of calorie restriction. Proc Natl Acad Sci U S A, 2006.

103(20): p. 7901-5.

203. Fontana, L., et al., Long-term effects of calorie or protein restriction on serum IGF-1

and IGFBP-3 concentration in humans. Aging cell, 2008. 7(5): p. 681-687.

204. Hartman, M.L., et al., Evaluation of pulsatile patterns of growth hormone release in

humans: A brief review. Am J Hum Biol, 1993. 5(6): p. 603-614.

205. Yakar, S., et al., Serum IGF-1 Determines Skeletal Strength by Regulating

Subperiosteal Expansion and Trait Interactions. Journal of Bone and Mineral Research,

2009. 24(8): p. 1481-1492.

206. Elis, S., et al., Unbound (bioavailable) IGF1 enhances somatic growth. Disease Models

& Mechanisms, 2011. 4(5): p. 649-658.

207. Gong, Z., et al., Reductions in serum IGF-1 during aging impair health span. Aging

Cell, 2014. 13(3): p. 408-18.

208. Elis, S., et al., Elevated serum levels of IGF-1 are sufficient to establish normal body

size and skeletal properties even in the absence of tissue IGF-1. J Bone Miner Res,

2010. 25(6): p. 1257-66.

209. Hoppener, J.W., et al., The human gene encoding insulin-like growth factor I is located

on chromosome 12. Hum Genet, 1985. 69(2): p. 157-60.

210. Oberbauer, A.M., The Regulation of IGF-1 Gene Transcription and Splicing during

Development and Aging. Front Endocrinol (Lausanne), 2013. 4: p. 39.

240 211. Adamo, M.L., et al., Transcription initiation in the two leader exons of the rat IGF-I

gene occurs from disperse versus localized sites. Biochem Biophys Res Commun,

1991. 176(2): p. 887-93.

212. D'Ercole, A.J., A.D. Stiles, and L.E. Underwood, Tissue concentrations of

somatomedin C: further evidence for multiple sites of synthesis and paracrine or

autocrine mechanisms of action. Proc Natl Acad Sci U S A, 1984. 81(3): p. 935-9.

213. Wang, X., Y. Yang, and M.L. Adamo, Characterization of the rat insulin-like growth

factor I gene promoters and identification of a minimal exon 2 promoter.

Endocrinology, 1997. 138(4): p. 1528-36.

214. Kho, A.T., et al., Identification of dedifferentiation and redevelopment phases during

postpneumonectomy lung growth. Am J Physiol Lung Cell Mol Physiol, 2013. 305(8):

p. L542-54.

215. Park, K.S., et al., Transdifferentiation of ciliated cells during repair of the respiratory

epithelium. Am J Respir Cell Mol Biol, 2006. 34(2): p. 151-7.

216. Liu, J.L. and D. LeRoith, Insulin-like growth factor I is essential for postnatal growth

in response to growth hormone. Endocrinology, 1999. 140(11): p. 5178-84.

217. Wang, J., et al., Effects of Igf1 gene deletion on postnatal growth patterns.

Endocrinology, 1999. 140(7): p. 3391-4.

218. Epaud, R., et al., Knockout of insulin-like growth factor-1 receptor impairs distal lung

morphogenesis. PLoS One, 2012. 7(11): p. e48071.

219. Galvis, L.A., et al., Repression of Igf1 expression by Ezh2 prevents basal cell

differentiation in the developing lung. Development, 2015. 142(8): p. 1458-1469.

220. Raile, K., et al., Clinical and functional characteristics of the human Arg59Ter insulin-

like growth factor i receptor (IGF1R) mutation: implications for a gene dosage effect

of the human IGF1R. J Clin Endocrinol Metab, 2006. 91(6): p. 2264-71.

241 221. Walenkamp, M.J., et al., A variable degree of intrauterine and postnatal growth

retardation in a family with a missense mutation in the insulin-like growth factor I

receptor. J Clin Endocrinol Metab, 2006. 91(8): p. 3062-70.

222. Choi, J.H., et al., Clinical and functional characteristics of a novel heterozygous

mutation of the IGF1R gene and IGF1R haploinsufficiency due to terminal 15q26.2-

>qter deletion in patients with intrauterine growth retardation and postnatal catch-up

growth failure. J Clin Endocrinol Metab, 2011. 96(1): p. E130-4.

223. Kruis, T., et al., Heterozygous mutation within a kinase-conserved motif of the insulin-

like growth factor I receptor causes intrauterine and postnatal growth retardation. J

Clin Endocrinol Metab, 2010. 95(3): p. 1137-42.

224. Labarta, J.I., et al., Familial short stature and intrauterine growth retardation

associated with a novel mutation in the IGF-I receptor (IGF1R) gene. Clin Endocrinol

(Oxf), 2013. 78(2): p. 255-62.

225. Roback, E.W., et al., An infant with deletion of the distal long arm of chromosome 15

(q26.1----qter) and loss of insulin-like growth factor 1 receptor gene. Am J Med Genet,

1991. 38(1): p. 74-9.

226. Miyazaki, E., et al., Altered insulin-like growth factor I mRNA expression in human

hypoplastic lung in congenital diaphragmatic hernia. J Pediatr Surg, 1998. 33(10): p.

1476-9.

227. Chetty, A., et al., Insulin-like growth factor-1 (IGF-1) and IGF-1 receptor (IGF-1R)

expression in human lung in RDS and BPD. Pediatr Pulmonol, 2004. 37(2): p. 128-36.

228. Stormann, S., et al., Assessment of lung function in a large cohort of patients with

acromegaly. Eur J Endocrinol, 2017. 177(1): p. 15-23.

242 229. Smith, B.T., W. Galaugher, and W.M. Thurlbeck, Serum from pneumonectomized

rabbits stimulates alveolar type II cell proliferation in vitro. Am Rev Respir Dis, 1980.

121(4): p. 701-7.

230. Vande Velde, G., et al., Longitudinal in vivo microcomputed tomography of mouse

lungs: No evidence for radiotoxicity. Am J Physiol Lung Cell Mol Physiol, 2015.

309(3): p. L271-9.

231. Vande Velde, G., et al., Longitudinal micro-CT provides biomarkers of lung disease

that can be used to assess the effect of therapy in preclinical mouse models, and reveal

compensatory changes in lung volume. Dis Model Mech, 2016. 9(1): p. 91-8.

232. De Langhe, E., et al., Quantification of lung fibrosis and emphysema in mice using

automated micro-computed tomography. PLoS One, 2012. 7(8): p. e43123.

233. microCT, B., NRecon User Manual. 2016.

234. Birnbaum, B.A., et al., Multi–Detector Row CT Attenuation Measurements: Assessment

of Intra- and Interscanner Variability with an Anthropomorphic Body CT Phantom.

Radiology, 2007. 242(1): p. 109-119.

235. Johnson, K.A., Imaging Techniques for Small Animal Imaging Models of Pulmonary

Disease: Micro-CT. Toxicologic Pathology, 2007. 35(1): p. 59-64.

236. microCT, B., Applying Hounsfield unit density calibration in Skyscan CT-analyser.

2005.

237. microCT, B., How to analyze the lung from an in vivo scan.

238. Scherle, W., A simple method for volumetry of organs in quantitative stereology.

Mikroskopie, 1970. 26(1): p. 57-60.

239. Livak, K.J. and T.D. Schmittgen, Analysis of relative gene expression data using real-

time quantitative PCR and the 2(-Delta Delta C(T)) Method. Methods, 2001. 25(4): p.

402-8.

243 240. Schneider, C.A., W.S. Rasband, and K.W. Eliceiri, NIH Image to ImageJ: 25 years of

image analysis. Nat Methods, 2012. 9(7): p. 671-5.

241. Young, L., et al., Detection of Mycoplasma in cell cultures. Nat. Protocols, 2010. 5(5):

p. 929-934.

242. Rana, O.R., et al., A simple device to apply equibiaxial strain to cells cultured on

flexible membranes. Am J Physiol Heart Circ Physiol, 2008. 294(1): p. H532-40.

243. Lee, A.A., et al., An equibiaxial strain system for cultured cells. Am J Physiol, 1996.

271(4 Pt 1): p. C1400-8.

244. Feng, H., et al., Paracrine communication between mechanically stretched myocytes

and fibroblasts. Methods Mol Biol, 2013. 1066: p. 57-66.

245. Wang, Y., et al., An experimental system to study mechanotransduction in fetal lung

cells. J Vis Exp, 2012(60).

246. Motulsky, H.J., Prism 4 Statistics Guide- Statistical analyses for laboratory and

clinical researchers. 2003, GraphPad Software Inc.: San Diego CA.

247. Fernandez, L.G., et al., Reinitiation of compensatory lung growth after subsequent lung

resection. J Thorac Cardiovasc Surg, 2007. 134(5): p. 1300-5.

248. Li, J., et al., The IGF-I/IGF-R1 pathway regulates postnatal lung growth and is a

nonspecific regulator of alveologenesis in the neonatal rat. American Journal of

Physiology - Lung Cellular and Molecular Physiology, 2013. 304(9): p. L626.

249. Balestrini, J.L. and L.E. Niklason, Extracellular Matrix as a Driver for Lung

Regeneration. Annals of Biomedical Engineering, 2015. 43(3): p. 568-576.

250. Bennett, R.D., et al., Deformation-induced transitional myofibroblasts contribute to

compensatory lung growth. Am J Physiol Lung Cell Mol Physiol, 2017. 312(1): p. L79-

L88.

244 251. McGowan, S., Understanding the developmental pathways pulmonary fibroblasts may

follow during alveolar regeneration. Cell Tissue Res, 2017. 367(3): p. 707-719.

252. Hede, M.S., et al., E-peptides control bioavailability of IGF-1. PLoS One, 2012. 7(12):

p. e51152.

253. Duan, C. and Q. Xu, Roles of insulin-like growth factor (IGF) binding proteins in

regulating IGF actions. Gen Comp Endocrinol, 2005. 142(1-2): p. 44-52.

254. Rotwein, P., Diversification of the insulin-like growth factor 1 gene in mammals. PloS

one, 2017. 12(12): p. e0189642-e0189642.

255. Vaquero, J., et al., The IGF2/IR/IGF1R Pathway in Tumor Cells and Myofibroblasts

Mediates Resistance to EGFR Inhibition in Cholangiocarcinoma. Clin Cancer Res,

2018.

256. Wolff, J.C., et al., Comparative gene expression profiling of post-natal and post-

pneumonectomy lung growth. Eur Respir J, 2010. 35(3): p. 655-66.

257. Ito, T., et al., Concordant pattern of radiologic, morphologic, and genomic changes

during compensatory lung growth. Journal of Surgical Research, 2017. 212: p. 60-67.

258. O'Donoghue, R.J., et al., Genetic partitioning of interleukin-6 signalling in mice

dissociates Stat3 from Smad3-mediated lung fibrosis. EMBO Mol Med, 2012. 4(9): p.

939-51.

259. Hsia, C.C.W., Signals and mechanisms of compensatory lung growth. Journal of

Applied Physiology, 2004. 97(5): p. 1992-1998.

260. Eisenhauer, P., et al., Endogenous distal airway progenitor cells, lung mechanics, and

disproportionate lobar growth following long-term postpneumonectomy in mice. Stem

Cells, 2013. 31(7): p. 1330-9.

261. Sun, R., et al., A change in the number of CCSP(pos)/SPC(pos) cells in mouse lung

during development, growth, and repair. Respir Investig, 2013. 51(4): p. 229-40.

245 262. Haber, S., et al., Alveolar septal patterning during compensatory lung growth: Part II

the effect of parenchymal pressure gradients. Journal of Theoretical Biology, 2017.

421: p. 168-178.

263. Li, J., et al., Csk/Src/EGFR signaling regulates migration of myofibroblasts and

alveolarization. Am J Physiol Lung Cell Mol Physiol, 2016. 310(6): p. L562-71.

264. Ysasi, A.B., et al., Evidence for pleural epithelial-mesenchymal transition in murine

compensatory lung growth. PLoS One, 2017. 12(5): p. e0177921.

265. Burri, P.H., R. Hlushchuk, and V. Djonov, Intussusceptive angiogenesis: its

emergence, its characteristics, and its significance. Dev Dyn, 2004. 231(3): p. 474-88.

266. Schittny, J.C., S.I. Mund, and M. Stampanoni, Evidence and structural mechanism for

late lung alveolarization. Am J Physiol Lung Cell Mol Physiol, 2008. 294(2): p. L246-

54.

267. Kim, N.T., et al., An original approach for quantification of blood vessels on the whole

tumour section. Anal Cell Pathol, 2003. 25(2): p. 63-75.

268. López, I.P., et al., Involvement of Igf1r in Bronchiolar Epithelial Regeneration: Role

during Repair Kinetics after Selective Club Cell Ablation. PLoS ONE, 2016. 11(11): p.

e0166388.

269. Jones, C.V., et al., M2 macrophage polarisation is associated with alveolar formation

during postnatal lung development. Respiratory Research, 2013. 14(1): p. 41-41.

270. Paxson, J.A., et al., Global gene expression patterns in the post-pneumonectomy lung

of adult mice. Respir Res, 2009. 10: p. 92.

271. Moody, G., et al., IGF1R blockade with ganitumab results in systemic effects on the

GH-IGF axis in mice. J Endocrinol, 2014. 221(1): p. 145-55.

272. Siddle, K., Signalling by insulin and IGF receptors: supporting acts and new players.

J Mol Endocrinol, 2011. 47(1): p. R1-10.

246 273. Fontana, L., et al., Long-term effects of calorie or protein restriction on serum IGF-1

and IGFBP-3 concentration in humans. Aging Cell, 2008. 7(5): p. 681-7.

274. Philippou, A., et al., The complexity of the IGF1 gene splicing, posttranslational

modification and bioactivity. Mol Med, 2014. 20: p. 202-14.

275. Brisson, B.K. and E.R. Barton, Insulin-like growth factor-I E-peptide activity is

dependent on the IGF-I receptor. PLoS One, 2012. 7(9): p. e45588.

276. Fu, S., et al., Expression and clinical significance of insulin-like growth factor 1 in lung

cancer tissues and perioperative circulation from patients with non-small-cell lung

cancer. Curr Oncol, 2016. 23(1): p. 12-9.

277. Welfare, A.I.o.H.a., Ageing and the health system: challenges, opportunities and

adaptations. 2014.

278. Lowery, E.M., et al., The aging lung. Clin Interv Aging, 2013. 8: p. 1489-96.

279. Navarro, S. and B. Driscoll, Regeneration of the Aging Lung: A Mini-Review.

Gerontology, 2017. 63(3): p. 270-280.

280. Navarro, S. and B. Driscoll, Regeneration of the Aging Lung: A Mini-Review.

Gerontology, 2016.

281. Meiners, S., O. Eickelberg, and M. Konigshoff, Hallmarks of the ageing lung. Eur

Respir J, 2015. 45(3): p. 807-27.

282. Sonntag, W.E., et al., Diverse Roles of Growth Hormone and Insulin-Like Growth

Factor-1 in Mammalian Aging: Progress and Controversies. The Journals of

Gerontology Series A: Biological Sciences and Medical Sciences, 2012. 67A(6): p.

587-598.

283. Flurkey, K., J. M. Currer, and D.E. Harrison, Chapter 20 - Mouse Models in Aging

Research, in The Mouse in Biomedical Research (Second Edition), J.G. Fox, et al.,

Editors. 2007, Academic Press: Burlington. p. 637-672.

247 284. Zha, J. and M.R. Lackner, Targeting the insulin-like growth factor receptor-1R

pathway for cancer therapy. Clin Cancer Res, 2010. 16(9): p. 2512-7.

285. Stewart, C.E. and P. Rotwein, Growth, differentiation, and survival: multiple

physiological functions for insulin-like growth factors. Physiol Rev, 1996. 76(4): p.

1005-26.

286. Yamada, P.M. and K.W. Lee, Perspectives in mammalian IGFBP-3 biology: local vs.

systemic action. Am J Physiol Cell Physiol, 2009. 296(5): p. C954-76.

287. Hoffman, A.M., et al., Matrix modulation of compensatory lung regrowth and

progenitor cell proliferation in mice. American Journal of Physiology - Lung Cellular

and Molecular Physiology, 2010. 298(2): p. L158-L168.

288. Hirota, J. and S. Shimizu, Chapter 5.2 - Routes of Administration, in The Laboratory

Mouse (Second Edition), H.J. Hedrich, Editor. 2012, Academic Press: Boston. p. 709-

725.

289. Hsia, C.C., D.M. Hyde, and E.R. Weibel, Lung Structure and the Intrinsic Challenges

of Gas Exchange. Compr Physiol, 2016. 6(2): p. 827-95.

290. Chetty, A., G.J. Cao, and H.C. Nielsen, Insulin-like Growth Factor-I signaling

mechanisms, type I collagen and alpha smooth muscle actin in human fetal lung

fibroblasts. Pediatr Res, 2006. 60(4): p. 389-94.

291. Fowlkes, J.L., et al., Matrix metalloproteinases as insulin-like growth factor binding

protein-degrading proteinases. Prog Growth Factor Res, 1995. 6(2-4): p. 255-63.

292. Miyamoto, S., et al., Matrix metalloproteinase-7 facilitates insulin-like growth factor

bioavailability through its proteinase activity on insulin-like growth factor binding

protein 3. Cancer Res, 2004. 64(2): p. 665-71.

248 293. Liu, X., et al., Mechanical stretch promotes matrix metalloproteinase-2 and prolyl-4-

hydroxylase alpha1 production in human aortic smooth muscle cells via Akt-p38

MAPK-JNK signaling. Int J Biochem Cell Biol, 2015. 62: p. 15-23.

294. Hsia, C.C.W., et al., Preventing mediastinal shift after pneumonectomy impairs

regenerative alveolar tissue growth. American Journal of Physiology-Lung Cellular

and Molecular Physiology, 2001. 281(5): p. L1279-L1287.

295. Liao, Y., et al., Delayed hepatocellular mitotic progression and impaired liver

regeneration in early growth response-1-deficient mice. J Biol Chem, 2004. 279(41):

p. 43107-16.

296. Iyer, N.V., et al., Cellular and developmental control of O2 homeostasis by hypoxia-

inducible factor 1 alpha. Genes Dev, 1998. 12(2): p. 149-62.

297. Bourbon, J., et al., Control mechanisms of lung alveolar development and their

disorders in bronchopulmonary dysplasia. Pediatr Res, 2005. 57(5 Pt 2): p. 38R-46R.

298. Khachigian, L.M., et al., Egr-1 is activated in endothelial cells exposed to fluid shear

stress and interacts with a novel shear-stress-response element in the PDGF A-chain

promoter. Arterioscler Thromb Vasc Biol, 1997. 17(10): p. 2280-6.

299. Milkiewicz, M. and T.L. Haas, Effect of mechanical stretch on HIF-1{alpha} and

MMP-2 expression in capillaries isolated from overloaded skeletal muscles: laser

capture microdissection study. Am J Physiol Heart Circ Physiol, 2005. 289(3): p.

H1315-20.

300. Hung, C.F., et al., Role of IGF-1 pathway in lung fibroblast activation. Respir Res,

2013. 14: p. 102.

301. Narasaraju, T.A., et al., Expression profile of IGF system during lung injury and

recovery in rats exposed to hyperoxia: a possible role of IGF-1 in alveolar epithelial

cell proliferation and differentiation. J Cell Biochem, 2006. 97(5): p. 984-98.

249 302. Davey, H.W., et al., STAT5b Is Required for GH-Induced Liver Igf-I Gene Expression.

Endocrinology, 2001. 142(9): p. 3836-3841.

303. Wu, X., et al., Mechano-sensitive transcriptional factor Egr-1 regulates insulin-like

growth factor-1 receptor expression and contributes to neointima formation in vein

grafts. Arterioscler Thromb Vasc Biol, 2010. 30(3): p. 471-6.

304. Ma, Y., et al., Induction of IGF-1R expression by EGR-1 facilitates the growth of

prostate cancer cells. Cancer Lett, 2012. 317(2): p. 150-6.

305. Pardo, P.S., et al., Induction of Sirt1 by Mechanical Stretch of Skeletal Muscle through

the Early Response Factor EGR1 Triggers an Antioxidative Response. Journal of

Biological Chemistry, 2011. 286(4): p. 2559-2566.

306. Youreva, V. and A.K. Srivastava, Early Growth Response Protein-1 Expression by

Insulin-Like Growth Factor-1 Requires ROS-Dependent Activation of ERK1/2 and

PKB Pathways in Vascular Smooth Muscle Cells. J Cell Biochem, 2016. 117(1): p. 152-

62.

307. Perrone, C.E., D. Fenwick-Smith, and H.H. Vandenburgh, Collagen and Stretch

Modulate Autocrine Secretion of Insulin-like Growth Factor-1 and Insulin-like Growth

Factor Binding Proteins from Differentiated Skeletal Muscle Cells. Journal of

Biological Chemistry, 1995. 270(5): p. 2099-2106.

308. Hu, B.S., et al., An analysis of the effects of stretch on IGF-I secretion from rat

ventricular fibroblasts. American Journal of Physiology-Heart and Circulatory

Physiology, 2007. 293(1): p. H677-H683.

309. Ma, Y., et al., Induction of IGF-1R expression by EGR-1 facilitates the growth of

prostate cancer cells. Cancer Letters, 2012. 317(2): p. 150-156.

310. Kim, M.-K., J.-W. Jang, and S.-C. Bae, DNA binding partners of YAP/TAZ. BMB

Reports, 2018. 51(3): p. 126-133.

250 311. Zagurovskaya, M., et al., EGR-1 forms a complex with YAP-1 and upregulates Bax

expression in irradiated prostate carcinoma cells. Oncogene, 2009. 28(8): p. 1121-31.

312. Rauschkolb, P.M., et al., Influence of Early Growth Response 1 (Egr1) and Tenascin C

(Tnc) on compensatory lung growth. The FASEB Journal, 2013. 27(1_supplement): p.

723.1-723.1.

313. Vadivel, A., et al., Hypoxia-Inducible Factors Promote Alveolar Development and

Regeneration. American Journal of Respiratory Cell and Molecular Biology, 2014.

50(1): p. 96-105.

314. Gatasheh, M.K., et al., Proflavine an acridine DNA intercalating agent and strong

antimicrobial possessing potential properties of carcinogen. Karbala International

Journal of Modern Science, 2017. 3(4): p. 272-278.

315. King, P.D. and M.C. Perry, Hepatotoxicity of Chemotherapy. The Oncologist, 2001.

6(2): p. 162-176.

316. Andersson, K.M.E., et al., Inflammation in the hippocampus affects IGF1 receptor

signaling and contributes to neurological sequelae in rheumatoid arthritis. Proc Natl

Acad Sci U S A, 2018. 115(51): p. E12063-E12072.

317. Gurău, F., et al., Anti-senescence compounds: A potential nutraceutical approach to

healthy aging. Ageing Research Reviews, 2018. 46: p. 14-31.

318. Chetty, A. and H.C. Nielsen, Regulation of cell proliferation by insulin-like growth

factor 1 in hyperoxia-exposed neonatal rat lung. Mol Genet Metab, 2002. 75(3): p. 265-

75.

319. Jung, H.J. and Y. Suh, Regulation of IGF -1 signaling by microRNAs. Front Genet,

2014. 5: p. 472.

320. Narasaraju, T.A., et al., Expression profile of IGF system during lung injury and

recovery in rats exposed to hyperoxia: A possible role of IGF-1 in alveolar epithelial

251 cell proliferation and differentiation. Journal of Cellular Biochemistry, 2006. 97(5): p.

984-998.

321. Ghosh, M.C., et al., Insulin-like growth factor-I stimulates differentiation of ATII cells

to ATI-like cells through activation of Wnt5a. Am J Physiol Lung Cell Mol Physiol,

2013. 305(3): p. L222-8.

322. Oyamada, H., et al., An optimal condition of bronchial cell proliferation stimulated by

insulin-like growth factor-I. Int Arch Allergy Immunol, 2000. 122 Suppl 1: p. 59-62.

323. Hung, C., et al., Role of IGF-1 pathway in lung fibroblast activation. Respiratory

Research, 2013. 14(1): p. 102.

324. Choi, J.-E., et al., Insulin-like Growth Factor-I Receptor Blockade Improves Outcome

in Mouse Model of Lung Injury. American Journal of Respiratory and Critical Care

Medicine, 2009. 179(3): p. 212-219.

325. Ghahary, A., et al., Induction of transforming growth factor beta 1 by insulin-like

growth factor-1 in dermal fibroblasts. J Cell Physiol, 1998. 174(3): p. 301-9.

326. Imrie, H., et al., Novel Role of the IGF-1 Receptor in Endothelial Function and Repair:

Studies in Endothelium-Targeted IGF-1 Receptor Transgenic Mice. Diabetes, 2012.

61(9): p. 2359-2368.

327. Fahmy, R.G., et al., Transcription factor Egr-1 supports FGF-dependent angiogenesis

during neovascularization and tumor growth. Nat Med, 2003. 9(8): p. 1026-32.

328. Medicine, N.U.N.L.o. ClinicalTrials.gov. 05th September 2018; Available from:

https://clinicaltrials.gov/ct2/results?cond=IGF-

1&term=&cntry=&state=&city=&dist=.

329. Hardin, D.S. and J.P. Sy, Effects of growth hormone treatment in children with cystic

fibrosis: the National Cooperative Growth Study experience. J Pediatr, 1997. 131(1 Pt

2): p. S65-9.

252 330. Grimberg, A. and P. Cohen, Role of Insulin-Like Growth Factors and Their Binding

Proteins in Growth Control and Carcinogenesis. Journal of cellular physiology, 2000.

183(1): p. 1-9.

331. Medicine, N.U.N.L.o., ClinicalTrials.gov.

332. Al-Rekabi, Z., et al., Activation of the IGF1 pathway mediates changes in cellular

contractility and motility in single-suture craniosynostosis. J Cell Sci, 2016. 129(3): p.

483-91.

333. Ho, K.H., et al., Identification of IGF-1-enhanced cytokine expressions targeted by

miR-181d in glioblastomas via an integrative miRNA/mRNA regulatory network

analysis. Sci Rep, 2017. 7(1): p. 732.

334. Lopez, I.P., et al., Differential organ phenotypes after postnatal Igf1r gene conditional

deletion induced by tamoxifen in UBC-CreERT2; Igf1r fl/fl double transgenic mice.

Transgenic Res, 2015. 24(2): p. 279-94.

335. Sjogren, K., et al., A model for tissue-specific inducible insulin-like growth factor-I

(IGF-I) inactivation to determine the physiological role of liver-derived IGF-I.

Endocrine, 2002. 19(3): p. 249-56.

336. Hegab, A.E., et al., Mimicking the niche of lung epithelial stem cells and

characterization of several effectors of their in vitro behavior. Stem Cell Research,

2015. 15(1): p. 109-121.

337. Vanoirbeek, J.A.J., et al., Noninvasive and Invasive Pulmonary Function in Mouse

Models of Obstructive and Restrictive Respiratory Diseases. American Journal of

Respiratory Cell and Molecular Biology, 2010. 42(1): p. 96-104.

338. Mammoto, T., et al., Acceleration of Lung Regeneration by Platelet-Rich Plasma

Extract through the Low-Density Lipoprotein Receptor-Related Protein 5-Tie2

Pathway. Am J Respir Cell Mol Biol, 2016. 54(1): p. 103-13.

253 339. Vasilescu, D.M., et al., Assessment of morphometry of pulmonary acini in mouse lungs

by nondestructive imaging using multiscale microcomputed tomography. Proc Natl

Acad Sci U S A, 2012. 109(42): p. 17105-10.

340. Vasilescu, D.M., et al., Stereological assessment of mouse lung parenchyma via

nondestructive, multiscale micro-CT imaging validated by light microscopic histology.

J Appl Physiol (1985), 2013. 114(6): p. 716-24.

341. Vasilescu, D.M., et al., Optimized murine lung preparation for detailed structural

evaluation via micro-computed tomography. J Appl Physiol (1985), 2012. 112(1): p.

159-66.

342. Blery, P., et al., Vascular imaging with contrast agent in hard and soft tissues using

microcomputed-tomography. Journal of Microscopy, 2015. 262(1): p. 40-49.

254