IDENTIFYING THE MOLECULAR MEDIATORS OF VN AND THE IGF:VN COMPLEX-STIMULATED BREAST CANCER CELL SURVIVAL
Ali Shokoohmand
School of Biomedical Sciences, Faculty of Health Queensland University of Technology, Australia
A thesis submitted for the degree of Doctor of Philosophy of the Queensland University of Technology 2015
QUT Verified Signature Acknowledgements
Commencing, pursuing and completing this dissertation like any other project, required abundant resources as well as strong motivation, which wouldn’t have been possible without the people who provided me with the much needed encouragement, support, scientific advice and help with experiments. Therefore, I would like to express my gratitude to the people below.
I would like to thank my principal supervisor ‘Dr Mr’ Hollier whose advice, guidance and encouragement has been always available for me through my PhD journey. Thanks for encouraging me to work hard at all times and keeping me motivated during my PhD. Your work ethic and your scientific expertise will always inspire me and I truly learned a lot from you. Abhi, I have always been appreciating to have you beside me. Your encouragement and support was a very great thing to me. I have learnt many deals from you. Your encouragement and support always helped me to work harder. Above all, I always enjoyed talking with you about our cultures, people and countries. I am sure we still have many things to talk about! Zee, thanks for the support you have given throughout my PhD. Without your support, this journey could have been harder for me. Derek, I would like to thank you for listening to me sometimes and being here for me.
I would like to thank all members of the TRR team for being helpful, considerate and supportive. It has been a pleasure working with you all. Special shout outs to Chen, Lipsa, Leo, Raju and Dominic. Dom, I would also thank you for helping me in generating heatmaps during my PhD studies, thank you ‘king of heatmaps’. I would also thank our laboratory technician Dod for providing me with reagents and equipment in the least possible time.
Good friends are the most important ingredient for any achievement. I would like to thank my great friends back home Morteza, Masoud, Mehdi and especially Amirhossien “Sardar” for their immense support and constant encouragement. Thank you for all those encouraging words and online sooshi breaks!
ii
Joony June, I don’t have words great enough to convey how much I have appreciated your love and care over the past three years. You have been my backbone and I was lucky enough to have you doing your PhD at the same time as mine. I am highly grateful for your support. I think this line summarises your impact on my life: “Thank you so much for all you do; you’re truly a delight; when my life overwhelms, you make everything all right”. You are one the most peaceful and down-to-earth human beings I have seen throughout my life!
Finally, to my family, I greatly appreciate your support throughout not only this entire PhD journey but the last 7 years that I have been away from home. You didn’t once complain about me being overseas and you always encouraged me to achieve my goals, even if that meant I was thousand miles away from you all. Mum and Dad, thanks for being great parents. You both are my biggest role models. Mum, you are just brilliant and the most amazing and big-hearted person I have ever known.
“Life is hard only for its first hundred years”
H.Karaminina
In loving memory of my dear grandmother
iii
Keywords
Attachment, breast, breast cancer, cancer progression, connective tissue growth factor, cysteine rich angiogenic inducer 61, extracellular matrix, gene knockdown, growth factors, growth factor binding proteins, insulin-like growth factor, integrin, integrin-mediated signalling, lentivirus, mediated down-regulation, metastasis, migration, molecular therapeutic targets, shRNA proliferation, three-dimensional cultures, survival, viability, vitronectin, vitronectin-binding integrins.
iv
Abstract
A key step in the cancer metastatic process is the ability of the cancer cells to survive in foreign tissue microenvironments. Two factors thought to be pivotal in cancer metastasis are altered cellular interactions with extracellular matrix (ECM) proteins within the tumour microenvironment and exposure to elevated levels of mitogenic hormones and growth factors. A compelling body of evidence suggests that the ECM protein, vitronectin (VN), can modulate tumour metastasis by providing traction and directionality to migratory cancer cells in addition to its effect on promoting cancer cell survival. While an increased deposition of VN has been reported at the leading edge of breast tumours, few studies have investigated the downstream VN-induced transcriptional changes responsible for its effect on cell survival and metastasis. In order to fill this knowledge gap, this project aimed to identify specific downstream mediators of VN induced in breast cancer cells with functional roles in cell survival and metastasis.
In addition to VN, insulin-like growth factors (IGFs) are known to play a critical role in the growth and development of the non-malignant cells as well as malignant cell types. The IGF system is complex and the biological effects of the IGFs are determined by their diverse interactions between many molecules, including their interactions with ECM proteins. Studies over the past decade have demonstrated that IGF ligands associate with VN, forming IGF:VN complexes that are capable of stimulating biological functions, namely survival and migration in both normal and malignant cell types. Considering the important role of IGF-I in the progression of breast cancer and its association with VN, this project also aimed to identify specific molecular mediators that are important in IGF-I:VN-mediated survival.
In order to identify the molecular mediators underpinning VN- and the IGF- I:VN complex-stimulated breast cancer cell survival, transcriptional profiling was performed using DNA microarrays to identify the transcriptional responses of MCF- 7 breast cancer cells stimulated with VN and the IGF-I:VN complex. Bioinformatics analysis identified a substantial number of differentially expressed transcripts in response to each treatment, whereby there was a considerable overlap observed between the transcripts differentially regulated by VN and the IGF:VN complexes.
v
To select transcripts for functional validation studies, we employed a strategy focused on selecting candidates to target not only VN, but potentially IGF-I:VN complex-mediated biological actions. Gene ontology and functional pathway enrichment were used to identify candidates that were regulated by both VN and the IGF-I:VN complex with a bias toward biological functions relevant to breast cancer cell survival. Based on published literatures, we identified a number of genes differentially expressed in common by VN and the IGF-I:VN complex with important roles in cell survival, such as TGFβ2, Smad6, S100A4, S100P, ANXA2, CCN1/Cyr61, CCN2/CTGF, RASGRP1, c-FOS, c-JUN, CAV1, snail 2 (slug), mir21 and LAMB1. Among the candidates identified, CCN family members, Cysteine-rich angiogenic inducer 61 (CCN1/Cyr61) and Connective Tissue Growth Factor (CCN2/CTGF), were selected to investigate their role in VN- and IGF-I:VN-induced breast cancer cell survival. It was demonstrated that VN and the IGF-I:VN complex increase Cyr61 and CTGF expression in a number of breast cancer cell types ranging from the weakly metastatic luminal-like to the highly metastatic basal-like breast cancer subtypes. Studies using functional blocking antibodies and chemical inhibitors confirmed the critical involvement of VN-binding αv-subunit containing integrins and the activation of the intracellular Src complex, PI3-K/AKT and ERK/MAPK signal transduction pathways. Furthermore, the functional role of Cyr61 and CTGF was assessed using shRNA-mediated gene suppression, which reduced VN- and IGF-I:VN-stimulated MCF-7 cell viability in both 2-dimentional and 3- dimentional cultures. Moreover, Cyr61 was observed to be important for the increased viability induced by VN and the IGF-I:VN complex in the highly metastatic MDA-MB-231 cell line, whereas CTGF had a more prominent role in the invasive and migratory phenotype.
In summary, this PhD project has identified for the first time the changes in global gene expression profiles underpinning VN- and the IGF:VN complex- mediated cancer cell survival. In addition, the outcomes of this study provide valuable new information on the mechanistic events underpinning VN- and IGF- I:VN-mediated effects on breast cell functions. Importantly, the up-regulation of Cyr61 and CTGF mediated by VN and the IGF:VN complex was determined to have an important functional role in breast cancer cell growth, survival and migration. Targeting Cyr61 and CTGF proteins or associated signalling pathways may disrupt
vi
IGF- and VN-mediated cancer progression, and with further development have the potential to be prognostic indicators and therapeutics for cancer patients.
vii
Table of Contents
Statement of Original Authorship ...... i Acknowledgements ...... ii Keywords ...... iv Abstract ...... v Table of Contents ...... viii List of Figures ...... xiii List of Tables ...... xvi List of Abbreviations ...... xvii List of Publications and Presentation……...………………………………………..xix CHAPTER 1: LITERATURE REVIEW...... 1 1.1 Introduction ...... 2 1.2 Breast cancer ...... 3 1.3 The extrecellular matrix (ECM) and breast cancer progression ...... 7 1.4 Vitronectin ...... 9 1.4.1 Structure of VN ...... 9 1.4.2 Function of VN ...... 11 1.5 Integrin signalling ...... 13 1.6 ECM-induced changes in gene expression ...... 15 1.7 VN in cancer progression ...... 15 1.8 Interactions between the ECM, integrins, growth factors and their receptors .. 17 1.8.1 Interactions between VN, VN-binding integrins and IGF system ...... 19 1.9 The IGF system ...... 20 1.9.1 IGF system ligands: IGF-I and IGF-II ...... 21 1.9.2 IGF binding proteins (IGFBPs)...... 22 1.9.3 Type-I IGF receptor (IGF-IR) ...... 23 1.9.4 IGF-IR Signalling ...... 23 1.9.5 Type-2 IGF receptor (IGF-IIR) ...... 23 1.10 The IGF system and breast cancer ...... 24 1.11 Evidence from the TRR laboratory supporting the role of VN and IGF interactions in breast cancer progression ...... 25 1.12 Conclusion and knowledge gaps ...... 27 1.13 Project outline ...... 28 1.13.1 Hypotheses ...... 28 1.13.2 Aims...... 28
viii
CHAPTER 2: MATERIALS AND METHODS ...... 29 2.1 Introduction ...... 30 2.2 Proteins and reagents ...... 30 2.3 Cell culture ...... 31 2.4 Treatament approach for in vitro investigations of VN- and IGF:VN complex- induced cell functions ...... 31 2.4.1 Pre-binding of VN, IGF-I, IGF-II and IGFBP-5 to 6- and 96-well plates ...... 32 2.4.2 Pre-binding of VN, IGF-I and IGFBP-5 to Transwells® ...... 33 2.5 MTS viability assay ...... 33 2.6 Transwells® migration assay ...... 33 2.7 Three-Dimensional (3D) Matrigel™ cultures ...... 34 2.7.1 AlamarBlue viability assay on cells grown in 3D Matrigel™ cell culture ...... 35 2.7.2 FDA staining of cells grown in 3D Matrigel™ cell culture...... 36 2.7.3 ImageJ analysis of FDA-stained cells ...... 36 2.8 RNA extraction from monolayer cell cultures ...... 37 2.8.1 Cell Lysis ...... 37 2.8.2 RNA isolation from cell lysate ...... 38 2.8.3 RNA concentration, purity and integrity check...... 38 2.9 Microarray analysis of differential gene expression ...... 38 2.9.1 The TargetAmp™-Nano single round biotin-aRNAbiotin amplification kit – overview ...... 39 2.9.2 In vitro transcription to synthesise biotin-labelled aRNA ...... 42 2.9.3 Biotin-labeled aRNA purification and quantification...... 42 2.9.4 BeadChip® Direct Hybridisation assay ...... 43 2.9.5 BeadChip® array scanning ...... 43 2.9.6 Samples and system Quality Control (QC) ...... 44 2.10 Microarray data analysis ...... 44 2.10.1 Data analysis and presentation using GeneSpring GX11 ...... 44 2.10.2 R program to generate Heatmap of gene expression data ...... 45 2.10.3 Gene ontology, functional network, and canonical pathway analyses using IPA ...... 45 2.11 qRT-PCR ANALYSES OF DIFFERENTIAL GENE EXPRESSION ...... 47 2.11.1 Primer design ...... 47 2.11.2 Reverse Transcription (RT) reaction to generate cDNA for qRT-PCR ...... 50 2.11.3 q-RT-PCR...... 50 2.12 Protein separation and immunoblotting ...... 51 2.12.1 Protein isolation from cell lines maintained in cell culture ...... 51 2.12.2 Immunoblotting ...... 51 2.13 Immunofluorescence staining...... 52
ix
2.14 Cloning ...... 53 2.14.1 Eukaryotic expression vectors...... 53 2.14.2 Preparing competent Stbl3 E.coli cells (calcium chloride method) ...... 53 2.14.3 Transformation of competent Stbl3 E.coli cells ...... 53 2.14.4 Plasmid purification – Miniprep ...... 54 2.15 Lentiviral vector-mediated gene manipulation ...... 55 2.15.1 shRNA lentiviral vector: production and design ...... 55 2.15.2 Virus production in HEK293T packaging cells ...... 58 2.15.3 Transfection of target cell lines ...... 58 2.16 Statistical analyses ...... 61 CHAPTER 3:IDENTIFYING CHANGES IN GENE EXPRESSION PROFILE ASSOCIATED WITH VN- AND IGF:VN COMPLEX-STIMULATED BREAST CANCER CELL SURVIVAL...... 63 3.1 Introduction ...... 64 3.2 Experimental methods ...... 66 3.2.1 Cell lines ...... 66 3.2.2 Treatment strategy for in vitro viability assay ...... 67 3.2.3 Cell viability measurement ...... 67 3.2.4 Optimisation of MCF-7 cell density for seeding onto 6-well plates ...... 67 3.2.5 Microarray analysis of differential gene expression ...... 68 3.2.6 Confirmation of differential gene expression using quantitative Real Time-PCR (qRT-PCR) ...... 70 3.3 Results ...... 70 3.3.1 VN, dimeric IGF-II:VN and trimeric IGF-I:IGFBP:VN complexes increase cellular viability of MCF-7 breast cancer cells ...... 70 3.3.2 Optimal MCF-7 cell seeding density for studies in 6-well plates ...... 73 3.3.3 RNA Quality Control (QC)...... 75 3.3.4 Global gene expression patterns induced by VN and IGF:VN complexes ...... 78 3.3.5 Identification of differential gene expression ...... 82 3.3.6 Gene ontology analysis of differentially regulated genes by VN and the IGF:VN complexes ...... 85 3.3.7 Pathway analysis ...... 89 3.3.8 Network analysis ...... 91 3.3.9 qRT-PCR validation of differentially expressed genes ...... 97 3.4 Discussion ...... 107 3.5 Conclusion ...... 114 CHAPTER 4:INVESTIGATION OF THE MECHANISMS UNDERLYING THE REGULATION OF CYR61 AND CTGF EXPRESSION INDUCED BY VN AND THE IGF-I:IGFBP:VN COMPLEX...... 116 4.1 Introduction ...... 117 4.2 Experimental methods ...... 118
x
4.2.1 Cell lines ...... 118 4.2.2 Heatmap analysis and presentation of target genes using R program ... 118 4.2.3 Gene-function interaction network presentation using Cytoscape 2.8.1 ...... 119 4.2.4 Pre-binding of proteins ...... 119 4.2.5 RNA and protein isolation to assess receptor involvement and signal transduction ...... 119 4.2.6 qRT-PCR analysis of target genes ...... 120 4.2.7 Immunoblotting ...... 120 4.2.8 Immunofluorescence staining of cells in 2D culture ...... 121 4.3 Results ...... 121 4.3.1 VN-induced genes are highly associated with biological processes relevant to cell survival ...... 121 4.3.2 Matricellular CCN family proteins Cyr61/CCN1 and CTGF/CCN2 associate with breast cancer cell survival ...... 124 4.3.3 VN- and IGF-I:IGFBP:VN complex-induce Cyr61 and CTGF expression ...... 126 4.3.4 Stimulated expression of Cyr61 and CTGF is specific to integrin ligation by ECM proteins ...... 128 4.3.5 VN and the IGF-I:IGFBP:VN complex regulate expression of Cyr61 and CTGF via αv-subunit containing integrin receptors ...... 131 4.3.6 VN and the IGF-I:IGFBP:VN complex regulate Cyr61 and CTGF expression via Src/PI3-K/AKT and Src/MAPK/ERK signalling pathways, respectively ...... 133 4.3.7 Basal-like breast cancer cells have increased levels of Cyr61 and CTGF ...... 135 4.3.8 Screening of breast cancer cell lines for expression of Cyr61 and CTGF in response to VN and the IGF-I:IGFBP:VN complex ...... 139 4.3.9 VN and the IGF-I:IGFBP:VN complex regulate expression of Cyr61 and CTGF via αv-integrin subunit receptor in MDA-MB-231 cells ...... 141 4.4 Discussion ...... 143 4.5 Conclusion ...... 150 CHAPTER 5:FUNCTIONAL VALIDATION OF CYR61 AND CTGF IN VN- AND IGF-I:IGFBP:VN-STIMULATED BREAST CANCER CELL SURVIVAL ...... 153 5.1 Introduction ...... 154 5.2 Experimental methods ...... 155 5.2.1 Cell lines ...... 155 5.2.2 Lentiviral shRNA mediated knockdown of target genes ...... 155 5.2.3 qRT-PCR to assess target gene expression ...... 156 5.2.4 Immunoblotting to assess protein expression...... 156 5.2.5 MTS assay for cell viability measurements ...... 157 5.2.6 CyQUANT for cell proliferation assay ...... 157 5.2.7 Transwell® migration assays ...... 158 5.2.8 3D Matrigel™ assay ...... 158
xi
5.3 Results ...... 159 5.3.1 Plasmid/viral delivery to packaging HEK 293T and target MDA- MB-231 cells ...... 159 5.3.2 Screening and selection of shRNA sequences resulting in optimal suppression of Cyr61 and CTGF in MDA-MB-231 cells ...... 162 5.3.3 Generation of constitutive MCF-7 shRNA knockdown models ...... 165 5.3.4 Suppression of Cyr61 and CTGF expression in MCF-7 cells reduces VN- and IGF-I:IGFBP-5:VN-stimulated cell viability and proliferation in 2D monolayer cultures ...... 167 5.3.5 Suppression of Cyr61 and CTGF decreases VN- and IGF-I-IGFBP- 5:VN-stimulated MCF-7 spheriod formation and long-term cell viability in 3D cultures ...... 169 5.3.6 Cyr61 knockdown reduces VN- and IGF-I:IGFBP-5:VN-stimulated MDA-MB-231 cell viability in 2D cultures ...... 172 5.3.7 Involvement of Cyr61 in regulating cell survival and CTGF in mediating spreding of MDA-MB-231 cells in response to VN and the IGF-I:IGFBP:VN complex in 3D cultures ...... 174 5.3.8 CTGF suppression decreases breast cancer cell migration in response to VN and the IGF-I:IGFBP:VN complex...... 177 5.4 Discussion ...... 180 5.5 Conclusion ...... 186 CHAPTER 6:GENERAL DISCUSSION ...... 188 APPENDICES ...... 201 BIBLIOGRAPHY...... 217
xii
List of Figures
Figure 1.1. Schematic depiction of normal and cancerous breast architecture...... 5 Figure 1.2. Schematic representation of the domain structure and key binding regions within VN...... 11 Figure 1.3. Schematic representations of major signalling pathways stimulated upon ligand-binding to integrin receptors...... 14 Figure 2.1: TargetAmp™-Nano single round biotin-aRNA amplification kit overview for the Illumina® system procedure...... 41 Figure 2.2: Map of pLKO.1 containing a shRNA insert...... 57 Figure 2.3: Schematic depiction of lentiviral shRNA-mediated suppression of target cells...... 60 Figure 3.1: Effect of VN, the IGF-II:VN and IGF-I:IGFBP:VN complexes on MCF-7 cellular viability...... 72 Figure 3.2: Optimisation of MCF-7 cell density for microarray analysis...... 74 Figure 3.3: Integrity and quality check of RNA samples extracted from MCF-7 cells...... 76 Figure 3.4: Heatmap of changes in global gene expression patterns...... 80 Figure 3.5: Euclidean distance map of changes in global gene expression patterns...... 81 Figure 3.6: Pair-wise comparisons of gene identifiers differentially expressed...... 84 Figure 3.7: Top ten molecular and cellular functions associated with differentially expressed genes at 12 hours...... 87 Figure 3.8: Top ten molecular and cellular functions associated with differentially expressed genes at 48 hours...... 88 Figure 3.9: Top two functional networks associated with differentially expressed genes at 12 hour...... 92 Figure 3.10: Top two functional networks associated with differentially expressed genes at 48 hours...... 95 Figure 3.11: FAK-containing networks associated with differentially expressed genes at 12 hours and 48 hours...... 96 Figure 3.12: qRT-PCR validation of up-regulated target genes identified by microarray analysis...... 102 Figure 3.13: qRT-PCR validation of down-regulated target genes identified by microarray analysis...... 104
xiii
Figure 3.14: Correlation of gene expression levels of 21 target genes determined by microarray analysis with qRT-PCR analysis...... 106 Figure 4.1: Top five biological functions associated with the VNGES...... 123 Figure 4.2: VNGES association with biological functions relevant to cell survival...... 125 Figure 4.3: VN- and the IGF-I:IGFBP:VN complex-induce expression of Cyr61 and CTGF in human breast cancer cells...... 127 Figure 4.4: ECM protein-induced expression of Cyr61 and CTGF...... 130 Figure 4.5: Involvement of VN-binding integrins and the IGF-IR in VN- and IGF-I:IGFBP:VN-induced expression of Cyr61 and CTGF in MCF-7 cells...... 132 Figure 4.6: Involvement of Src, PI3-K and ERK cell signalling mediators in VN- and IGF-I:IGFBP:VN-induced expression of Cyr61 and CTGF in MCF-7 cells...... 134 Figure 4.7: Association of the VNGES across breast cancer molecular subtypes...... 137 Figure 4.8: Differential expression of Cyr61 and CTGF across luminal, basal A, basal B and claudin-low breast cancer cell lines...... 138 Figure 4.9: Cyr61 and CTGF regulation by VN and the IGF-I:IGFBP:VN complex in human breast cancer cell lines...... 140 Figure 4.10: Involvement of VN-binding integrins and the IGF-IR in VN- and IGF-I:IGFBP:VN-induced expression of Cyr61 and CTGF in MDA- MB-231 cells...... 142 Figure 5.1:Visualisation of transfection and transduction efficiency ...... 161 Figure 5.2: Screening of gene-specific shCyr61 and shCTGF sequences in MDA-MB-231 cells using qRT-PCR and immunoblotting...... 164 Figure 5.3: Validation of Cyr61 and CTGF in MCF-7 breast cancer cells ...... 166 Figure 5.4: Effects of Cyr61 and CTGF knockdown on VN- and IGF-I:IGFBP- 5:VN-stimulated MCF-7 cell viability and proliferation...... 168 Figure 5.5: Effects of Cyr61 and CTGF suppression on MCF-7 cell growth in 3D in response to VN and the IGF-I:IGFBP:VN complex...... 171 Figure 5.6: Effects of Cyr61 and CTGF knockdown on VN- and IGF-I:IGFBP- 5:VN-stimulated MDA-MB-231cell viability...... 173 Figure 5.7: Effects of Cyr61 and CTGF suppression on growth and morphological appearance of MDA-MB-231 cells cultured in 3D Matrigel™ in response to VN and the IGF-I:IGFBP:VN complex ...... 176
xiv
Figure 5.8: The effects of Cyr61 and CTGF down-regulation on MCF-7 and MDA-MB-231 cell migration in response to VN and the IGF- I:IGFBP:VN complex...... 179 Figure 6.1:VN- and the IGF-I:IGFBP:VN complex-induced Cyr61 and CTGF signalling (identified by studies reported herein) in addition to signalling by secreted Cyr61 and CTGF (identified by others)...... 197
xv
List of Tables
Table 2.1. Target genes for qRT-PCR analysis...... 48 Table 2.2. Details of primers designed using NCBI Primer-BLAST ...... 49 Table 2.3. shRNA sequence information ...... 56 Table 3.1. QC of isolated total RNA samples for microarray analysis...... 77 Table 3.2. Number of gene identifiers differentially regulated in pair-wise comparisons...... 82 Table 3.3. Top ten canonical pathways associated with differentially expressed genes at 48 hours...... 90 Table 3.4. Target genes for qRT-PCR analysis...... 98
xvi
List of Abbreviations
α alpha β beta °C degrees Celsius Δ change μm micrometre μM micromolar
μmol micromoles μg/μL micrograms per microlitre μg/mL micrograms per millilitre < less than > more than 2D two-dimensional 3D three-dimensional 3D-GM 3D-growth medium AA amino acids Ab antibody AGRF Australian Genome Research Facility AIHW Australian Institute of Health and Welfare AKT akt/protein kinase B ALS acid-labile subunit ANOVA analysis of variance Anlg analogue Arg Arginine ADH atypical ductal hyperplasia ALH atypical lobular hyperplasia BHFDR Benjamini and Hochberg false discovery rate
xvii
BHLHB2 Basic helix-loop-helix B2 BP binding protein BSA bovine serum albumin BMP-2 bone morphogenetic Protein 2 CDK2 cyclin-dependent kinase 2 cDNA complimentary DNA CHO Chinese hamster ovarian CK2 kinase C and casein kinase II CNTRL control CO2 carbon dioxide DEPC diethyl pyrocarbonate DMEM Dulbecco‟s modified Eagles medium DMEM/F-12 Dulbecco‟s modified Eagles medium/Ham’s F-12 DCIS ductal carcinoma in situ LCIS lobular carcinoma in situ C-terminal carboxy-terminal CTGF connective tissue growth factor CXCL14 CXC chemokine ligand 14 CXCR4 CXC chemokine receptor 4 Cyr61 cysteine-rich angiogenic inducer 61 DCIS ductal carcinoma in situ DMSO Dimethyl sulfoxide DNA deoxyribonucleic acid ECL enhanced chemi-luminescence ECM extracellular matrix EDTA ethylenediamine tetraacetic acid
EFNB2 ephrin-B2 EGF epidermal growth factor receptor EIF4E eukaryotic translation initiation factor 4E
xviii
EMT epithelial-to-mesenchymal transition ER estrogen receptor ERCC External RNA Controls Consortium ERK1/2 extracellular signal-related kinase-1 and 2 F3 tissue factor FAK focal adhesion kinase FAS1 fasciclin 1 FBS fetal bovine serum FF3 firefly luciferase-3 gene FGFR fibroblast growth factor receptors FN fibronectin Fra fos-related antigen g grams GAPDH glyceraldehyde 3-phosphate dehydrogenase GEF Guanine nucleotide-exchange factor GEX-HYB Gene Expression Hybridisation Buffer GF growth factor GFP green fluorescent protein GH growth hormone Glu glutamic acid Grb2 growth-factor receptor bound protein-2 HBD heparin binding domain HEK human embryonic kidney HER2 human epidermal growth factor receptor-2 HS horse serum HSPG heparan-sulfate containing proteoglycan
HUVEC human umbilical vein endothelial cell HX hemopexin HYB Hybridisation Buffer
xix
I Iodine IGF insulin-like growth factor-I IGF-IR insulin-like growth factor type-1 receptor IGF-IIR insulin-like growth factor type II receptor IGFBP insulin-like growth factor binding protein IgG immunoglobulin G IHBI Institute of Health and Biomedical Innovation ILK Integrin-linked kinase IR insulin receptor IRS insulin-receptor substrate JNK c-Jun N-terminal kinase K8 cytokeratin 8 K14 cytokeratin 14 kb kilo bases kDa kilo Dalton LB lysogeny broth Leu Leucine Lck lymphocyte-specific tyrosine kinase LTR long terminal repeat MAPK mitogen-activated protein kinase MCF-7 Michigan cancer foundation-7 MDA-MB-231 M.D Anderson metastatic breast carcinoma MEK mitogen-activated protein kinase MGP matrix Gla protein MLCK myosin light chain kinase MSC mesenchymal stem cell mins minutes mL millilitre mm millimetre
xx
MMP matrix metalloproteinase mRNA messenger ribonucleic acid mTOR mammalian target of rapamycin MTS [3-(4,5-dimethylthiazol-2-yl)-5-(3- carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H- tetrazolium, inner salt MT1-MMP membrane type-1 MMP MLCK myosin light chain kinase mw molecular weight n subsample size NCBI National Centre Biotechnology Information NEG negative ng/mL nanogram per millilitre ng/μL nanogram per microlitre nM nanomolar nmol nano moles NTP nuceotide triphosphate N-terminal amino-terminal NF-κB Nuclear Factor kappa-light-chain-enhancer of activated B cells
NSCLC non-small cell lung carcinoma cell p p-value P8 peptide-8 P7 peptide-7 p38 MAPK p38 mitogen-activated protein kinase p70S6K p70 S6 kinase PAI-1 plasminogen activation inhibitor-1 PAK P21-acitvated kinase pAKT phosphorylated AKT PBS phosphate buffered saline
xxi
PCR polymerase chain reaction pERK1/2 phosphorylated extracellular signal-related kinase-1 and - 2 pH H+ ion concentration PI3-K phosphoinositol 3-kinase pIGF-IR phosphorylated IGF-IR PIP3 phosphatidylinositol (3,4,5)-trisphosphate PR progesterone receptor PTEN phosphatase and tensin homolog PTMA prothymosin-α qRT-PCR quantitative real-time PCR QUT Queensland University of Technology RGD Arginine-Glycine-Aspartic acid RMA Robust Multi-array Average RT reverse transcription RTK receptor tyrosine kinase SEM standard error mean Ser serine Sdc Syndecan SFM serum-free medium SFN stratifin SH2 src homology 2 domain SHARP-2 basic helix-loop-helix domain containing class B-2 Shc Src homology 2 domain containing SHP-2 SH2-containing phosphatase-2 shRNA short-hairpin RNA SMB somatomedin-B SMC smooth muscle cells SOS Son-of-sevenless
xxii
SSTN synstatin TAT thrombin-anti-thrombin TBS tris-buffered saline TBS-T tris-buffered saline with Tween-20 TCDD 2,3,7,8-tetrachlorodibenzo-[p]-dioxin TF tissue factor TFPI tissue factor pathway inhibitor TGFβ transforming growth factor β Thr threonine TK tyrosine kinase TMA tissue microarray TN Tenascin TNBC triple negative breast cancer TRI trimeric IGF-I:IGFBP:VN complex TRR Tissue Repair and Regeneration uPA urokinase-type plasminogen activator uPAR urokinase-type plasminogen activator receptor VEGF vascular endothelial growth factor VIM vimentin VN vitronectin VNGES Vitronectin-induced gene expression signature
xxiii
List of Publications and Presentations
Publications arising from this PhD project
Submitted
Kashyap AS., Shooter GK., Shokoohmand A, McGovern J., Sivaramakrishnan M.,
Cane G., Cane G., Mani SA., Leavesley DI., Söderberg O., Upton Z. and Hollier BG. (2014). Novel peptide-based inhibitors of IGF:Vitronectin interactions inhibit IGF-I- induced breast cancer cell function. Submitted to Cancer Discovery.
Leavesley DI., Kashyap AS., Croll T., Sivaramkrishan M., Shokoohmand A., Manton KJ., Hollier BG. and Upton Z. (2013). Vitronectin–Master Controller or Micromanager? IUBMB Life. Volume 65, Issue 10, 807–818.
To be submitted
Shokoohmand A., Kashyap AS., Guanzon D., Van Lonkhuyzen D., Leavesley DI., Shooter G, Upton Z. and Hollier BG. (2016). VN and the IGF-I:VN complex- induced expression of CCN1/Cyr61 and CCN2/CTGF regulate breast cancer survival and migration. To be submitted to The International Journal of Cancer. (Awaiting for compilation of final manuscript draft preparation).
Shokoohmand A., Kashyap AS., Leavesley DI. and Hollier BG. (2016). VN and VN-binding integrins in cancer progression. Review article to be featured in IJBCB (in preparation).
Other Publications
Dargaville T., Hollier BG., Shokoohmand A. and Hogenboom R. (2014). Poly (2- oxazoline) Hydrogels as Next Generation Three Dimensional Cell Supports. Cell Adhesion & Migration, Volume 8, Issue 2, 88-93.
Shokoohmand A. and Drew R. A. (2011). Micropropagation of Moringa oleifera. V International Symposium on Acclimatization and Establishment of Micropropagated Plants ISAEMP2012, International Society for Horticultural Science (ISHS), 988.
xxiv
Presentations
Shokoohmand A., Kashyap AS., Guanzon D., Leavesley DI., Van Lonkhuyzen D., Upton Z. and Hollier BG. (2014). Vitronectin-stimulated Cyr61 and CTGF expression via activation of αv-integrin receptors and FAK/AKT/ERK signalling pathways induces breast cancer cell survival. 26th Lorne Cancer Conference, Lorne, Victoria, Australia, Feb 13-15. (Poster presentation)
Shokoohmand A., Kashyap AS., Guanzon D., Leavesley DI., van Lonkhuyzen D., Upton Z. and Hollier BG. (2013). Identifying the molecular mediators of IGF:VN- stimulated cancer cell survival. IHBI Inspires Postgraduate Student Conference, Brisbane, Australia, Nov 24-25. (Oral presentation)
Shokoohmand A., Kashyap AS., Gupta R., Leavesley DI., van Lonkhuyzen D., Upton Z. and Hollier BG. (2013). Regulation of the matricellular proteins
CCN1/Cyr61 and CCN2/CTGF by vitronectin activation of αv integrin induces cancer cell survival. Australian Society of Medical Research (ASMR) Postgraduate Conference, Brisbane, QLD, Australia, May 29. (Poster presentation)
Shokoohmand A., Kashyap AS., Gupta R., Leavesley DI., Dayle LS., Van Lonkhuyzen D., Upton Z. and Hollier BG. (2013). Identifying the molecular mediators of Vitronectin and IGF-I:IGFBP:VN-stimulated cancer cell survival. Gordon research conference (GRC), IGF in physiology and disease, Ventura, California, United State of America, Mar 17-23. (Poster presentation)
Shokoohmand A., Kashyap AS., Gupta R., Leavesley DI., Dayle LS., van Lonkhuyzen D., Upton Z. and Hollier BG (2013). Regulation of the matricellular proteins CCN1/Cyr61 and CCN2/CTGF by vitronectin induces cancer cell survival. 25th Lorne Cancer Conference, Lorne, Victoria, Australia, Feb 14-16. (Poster presentation)
Shokoohmand A., Kashyap AS., Gupta R., Dayle LS., van Lonkhuyzen D., Leavesley D., Upton Z. and Hollier BG. (2012). Identifying the molecular mediators of Vitronectin and IGF-I:IGFBP:VN-stimulated cancer cell survival. IHBI Inspires Postgraduate Student Conference, Gold Coast, Australia, Nov 21-22. (Poster presentation)
Shokoohmand A., Kashyap AS., Van Lonkhuyzen D., Leavesley DI., Upton Z. and Hollier BG. (2011). Identifying the molecular mediators of IGF-I:IGFBP:VN- stimulated breast cancer cell survival. IHBI Inspires Postgraduate Student Conference, Brisbane, Australia, Nov 24-25. (Poster presentation)
xxv
Chapter 1: Literature Review
Chapter 1: Literature Review 1
1.1 INTRODUCTION
The metastatic spread of cancer cells to distant sites and their subsequent establishment in critical organs of the body is the major cause of death among patients with epithelial cancers (termed carcinomas), including breast cancer. The ability of cancer cells to invade and migrate away from the primary tumour site and survive within foreign tissue microenvironments is one of the key factors underlying cancer metastasis. Altered interactions between the malignant cells and the extracellular matrix (ECM) proteins within the tumour microenvironment are considered to be one of the most important factors contributing to cancer progression (Lochter & Bissell, 1995). It is well established that the interactions between cells and surrounding ECM proteins influences cell behaviour and function through adhesion receptors (Hynes, 1992, 2009). Less well established is the extent to which these ECM components directly mediate cancer progression. It is also unclear as to the identity and involvement of downstream molecular mediators in cell-ECM- mediated tumour progression.
One such ECM protein is the multifunctional adhesive glycoprotein vitronectin (VN), which has well documented roles in normal tissue development and progression of numerous malignancies, including breast cancer (Aaboe et al., 2003; Hurt et al., 2010; Leavesley et al., 2013). The biological effects of ECM proteins are known to be influenced by their diverse interactions with various growth factors (GFs), including the insulin-like growth factor (IGF) system (Beattie et al., 2010). The IGF system also has well documented roles in the development and progression of numerous malignancies, including breast cancer (Annunziata et al., 2011; Hartog et al., 2013). A comprehensive set of studies from the Tissue Repair and Regeneration (TRR) program have demonstrated that VN binds to the components of the IGF system to form IGF:VN complexes that are capable of stimulating breast cancer cell functions, namely cell attachment, migration and survival (Noble et al., 2003; Hollier et al., 2008; Kricker et al., 2010; Kashyap et al., 2011). Given that VN and IGFs are both implicated in breast cancer progression and the lack of information regarding the downstream molecular mediators of their biological functions in tumour cells, studies described herein aimed to address this void in our understanding. In view of this, the following review focuses on VN and its binding
Chapter 1: Literature Review 2
receptors, in particular integrins and their downstream signalling pathways with a focus on their implications in breast cancer. This review will also focus on interactions of the IGF system with VN and their implications in breast cancer and will further highlight the current knowledge gaps.
1.2 BREAST CANCER
Breast cancer is one of the most commonly diagnosed form of cancer and the leading cause of cancer related deaths in women worldwide (Globocan, WHO 2012). Breast cancer accounts for more than 25% (1.67 million) of newly diagnosed cases and 14% (522,000) of total cancer associated deaths annually (WHO, 2012). In Australia, it accounts for 28% of all diagnosed cancers with 14,181 newly diagnosed cases per annum and 16% of cancer-associated deaths among women (AIHW, 2010). Despite significant advances in diagnosing and treating breast cancer, several major unresolved clinical and scientific problems remain. These are related to: (1) prevention; (2) specific and sensitive methods for diagnosis; (3) tumour progression and recurrence (understanding the mechanism of recurrence and predicting its rate); (4) treatment (various molecular subtypes of breast cancer require different therapies); and (5) therapeutic resistance (developing reliable biomarkers predicting response to targeting agents to prevent, and overcome resistance). However, these problems are further exemplified as breast cancer is not a single disease but is highly heterogeneous at both the molecular and clinical levels (Perou et al., 2000; Sorlie et al., 2001).
Gene expression profiling studies from multiple independent groups have defined six major molecular subtypes of breast tumours: luminal A, luminal B, basal A, basal B, human epidermal growth factor 2 (HER2) positive and claudin-low subtypes (Perou et al., 2000; Sotiriou et al., 2003; Neve et al., 2006; Sorlie et al., 2006; Prat et al., 2010). Among these subtypes, the basal-like and claudin-low subtypes are found to be associated with higher risk of development of distant metastasis; poorer clinical outcomes and higher rate of mortality (Banerjee et al., 2006; Fulford et al., 2007). These six subtypes can also be distinguished by immunohistochemical analysis of estrogen receptor (ER), progesterone receptor (PR), and HER2 to yield three broad groups: (1) ER-positive tumours (when ER is positive and HER2 is not amplified), (2) HER2-possitive tumours (which may be ER
Chapter 1: Literature Review 3
positive or negative), and triple-negative breast cancer (TNBC) (when ER, PR and HER2 are all negative) (Nielsen et al., 2004; Hu et al., 2006; Foulkes et al., 2010).
Luminal tumours are estrogen and progesterone receptor positive (ER+ and PR+, respectively), with luminal A being low grade and luminal B being in general a higher grade (Voduc et al., 2010). ER+ tumours make up about 60% of breast cancers diagnosed in pre-menopausal woman and offer the best prognostic outcome (Jonat et al., 2006; Kulendran et al., 2009), with the majority responding to hormonal interventions, including Tamoxifen, Toremifene and Fulvestrant. HER2+ tumours have amplification and overexpression of HER2 oncogene (Ross et al., 2003) and can be controlled with a diverse array of anti-HER2 therapies available in the current market, such as Herceptin and Perjeta (Arteaga et al., 2012). Basal-like tumours lack hormone receptors and HER2; thus, the majority of these tumours are also called TN breast cancer. TNBC is linked to aggressive tumour development and has the worst prognosis among all forms of breast cancer, and cannot be treated with endocrine therapy or therapies targeted at HER2 (Dent et al., 2007; Foulkes et al., 2010). The claudin-low subtype is contained within the broader TNBC subtype and has characteristic enrichment for epithelial-to mesenchymal transition (EMT) and cancer stem cell-like features that most closely resemble the mammary epithelial stem cell showing the biologic heterogeneity of breast cancer (Prat et al., 2010).
Histologically, the breast tissue is composed of mammary glands, basement membrane and stroma. Mammary glands have milk-producing lobules linked by numerous ducts which are populated with luminal epithelial cells and transversely oriented myoepithelial cells (Figure 1.1 A). The surrounding stroma comprises ECM (composed of proteins such as laminin, collagen and VN), various cell types (for example, fibroblasts, immune cells, and adipocytes), and organised structures (namely, blood vessels, lymph vessels and lymph nodes) (Place et al., 2011). The architecture of the normal breast is maintained by the cross-talk between the stromal, epithelial and myoepithelial cells (Barsky & Karlin, 2006). As such interactions between the ECM proteins, growth factors, hormones, and the corresponding receptors present on the breast epithelial cells are crucial for the maintenance of the mammary gland architecture, and any perturbations in their interaction may lead to cancer development and progression (Howlett & Bissell, 1993).
Chapter 1: Literature Review 4
Breast cancer progression is characterised by multiple well-defined stages which initiates as flat epithelial atypia, progresses to ductal or lobular hyperproliferation (atypical ductal/lobular hyperplasia; ADH/ALH), evolves to in situ (ductal/lobular carcinoma in situ; DCIS/LCIS), invasive carcinomas, and eventually metastatic disease (Burstein et al., 2004; Bombonati & Sgroi, 2011). Factors involved in tumour progression and methods to selectively interfere in specific stages of progression, while not clear, are under intense investigation worldwide. Primary in situ carcinomas (DCIS and LCIS) are rarely the cause of high mortality associated with breast cancer (Barnes et al., 2012). Rather, the high mortality is caused by the metastatic dissemination of breast cancer cells to distant organs of the body (Weigelt et al., 2005). Genetic mutations influenced by the stromal and ECM interactions with primary tumour cells, can lead to changes in cell function that can facilitate these tumour cells to acquire proliferative, migratory and invasive capabilities. This in turn leads to the conversion of basement membrane- bound in situ lesions to escape from tissue boundaries (metastasise), survive and establish secondary tumours within alternate tissue sites (Kim et al., 2005; Bissell & Hines, 2011; Place et al., 2011) (Figure 1.1 B).
Figure 1. Schematic depiction of normal and cancerous breast architecture. The architecture of the normal breast (A) is made up of networks of alveoli containing the milk- producing lobules linked by numerous ducts, surrounded by intact basement membrane and extracellular matrix (ECM). This architecture is disrupted on the acquisition of tumorigenic properties by cells within the mammary tissue. A heterogeneous mass of primary tumour is depicted (B) along with the processes key to metastasis; ECM remodelling, acquisition of proliferative, migratory and invasive characteristics by the transformed cells, disorganisation of the basement membrane, infiltration of fibroblasts into to the tumour mass and intravasation of cancerous cells into adjacent blood and lymphatic vessels. Three dimensional illustration of the breast (left panel) is adapted from Patrick J. Lynch, medical illustrator.
Chapter 1: Literature Review 5
Multiple factors can influence the ability of cancer cells to metastasise. Microenvironmental factors have been reported to play a key role in the metastatic progression of breast cancer (Lochter & Bissell, 1995; Ronnov-Jessen & Bissell, 2009; Quail & Joyce, 2013). It is very well established that the interactions between breast cancer cells and the constituents of ECM within the tumour microenvironment play a central role in controlling the tumorigenic process (Lochter & Bissell, 1995; Bergamaschi et al., 2008; Muschler & Streuli, 2010). The ECM is a complex mixture of proteins, adhesive glycoproteins and proteoglycans. The interactions between the ECM components and cells through their cognate cell surface receptors coordinate downstream signals and subsequently gene expression that directly control cell behaviour and functions (Hynes, 2009). Besides the roles of the ECM proteins, growth factors (GFs), chemokines and cytokines present within the ECM have also been involved in breast cancer progression (Lin & Karin, 2007; Mantovani et al., 2010; Witsch et al., 2010).
For example, in xenograft models of breast cancer, co-injection of various fibroblasts or mesenchymal stem cells within the tumour stroma promotes breast tumour growth (Zhu et al., 2006) and metastasis (Karnoub et al., 2007). This may be attributed to paracrine factors such as chemokines (CXC ligand-12), cytokines (interleukin-11) and GFs (IGF, epidermal growth factor [EGF], transforming growth factor β [TGFβ]) released from these stromal cells (Kang et al., 2003; Orimo et al., 2005). This can in turn up-regulate cognate receptors present on the cancer cells (normally overexpressed), thereby favouring an invasive and proliferative phenotype. Since most tumour cells remain responsive to the extracellular signals mediated by these GFs (Hankinson et al., 1998), several studies have been directed at developing inhibitors to specific growth factor receptors (Eke & Cordes, 2011; Tognon & Sorensen, 2012; Marlow et al., 2013; Wolff et al., 2014). The IGF system is one such GF system that has been extensively studied in breast cancer due to its frequent role in mediating processes central to breast cancer progression, with several therapeutic strategies being designed to combat its downstream effects (Gao et al., 2012; Tognon & Sorensen, 2012; Ma et al., 2013). Together with cytokines, GFs and cell-cell interactions, the ECM determines whether cells survive, differentiae, proliferate or migrate (Streuli, 2009).
Chapter 1: Literature Review 6
1.3 THE EXTRECELLULAR MATRIX (ECM) AND BREAST CANCER PROGRESSION
The ECM helps to maintain normal tissue structure and functions and regulates cell behaviour by providing contextual information to the cells. During tissue development, remodelling, and maintenance, ECM components are synthesised and degraded in a tightly regulated manner to conserve its delicate balance (Page-McCaw et al., 2007). The ECM has significant effects on cell function through cell-ECM interactions that depend on specific ECM composition and organisation (Cox & Erler, 2011). Altered ECM dynamics and disruption of regulatory mechanisms over ECM maintenance can cause abnormal behaviour in surrounding cells, and are known hallmarks of cancer (Hanahan & Weinberg, 2011). For example, overexpression of laminin, one of the key components of the ECM, in certain contexts can be a poor diagnostic factor and has been related to tumour invasiveness in a number of cancers, including malignant glioma (Fukushima et al., 1998), lung (Moriya et al., 2001) and breast (Fujita et al., 2005) carcinomas. Correspondingly, increased deposition of collagen type I, III, IV and V is reported during tumour formation and cell invasion in breast and glioma cancers (Kauppila et al., 1998; Sangaletti et al., 2003; Huijbers et al., 2010). Similarly, in breast and glioma cancer cells, increased levels of VN can induce cell differentiation and tumour formation (Hurt et al., 2010), survival (Uhm et al., 1999), and invasion (Pola et al., 2013). Tenascin-C (TN-C) and FN were also found to be highly distributed during vascular organisation in several different carcinomas (Berndt et al., 2010). This leads to enhanced cell proliferation, vascular morphogenesis, tumour angiogenesis and growth, while promoting invasion and metastasis (Van Obberghen-Schilling et al., 2011).
In normal mammary tissue, epithelial, endothelial, and stromal cells produce and maintain the basement membrane ECM while stromal cells alone contribute to the interstitial matrix. These ECM layers in turn provide structural and mechanical support to the cells and tissues while also providing physical interactions between the ECM and the cells (Lu et al., 2012). The interactions between the ECM components and epithelial cells provide information that is necessary for almost all aspects of mammary development and function such as ductal formation, branching of ducts and alveoli as well as production of the milk-producing luminal epithelial cell layer
Chapter 1: Literature Review 7
during lactation (Muschler & Streuli, 2010; Polyak & Kalluri, 2010). The composition of the ECM and subsequently its mechanical properties are known to be among the most important factors contributing to the development and progression of breast cancer at every stage from pre-malignancy to metastasis (Bissell & Radisky, 2001). As outlined above, strong deposition of a number of integrin-binding matrix proteins, including laminin, collagens (types I, III, IV and V), FN, VN and TN-C have been found in the extracellular space around breast tumours (Kauppila et al., 1998; Ioachim et al., 2002; Kadar et al., 2002; Aaboe et al., 2003). Abnormal ECM composition can increase tissue stiffness to the levels that are detrimental to the mammary epithelial cells, altering their morphology and cell-specific characteristics (Paszek & Weaver, 2004). Indeed, differential expression of the ECM components delineates the breast cancer subtype likely to progress (Bergamaschi et al., 2008; Triulzi et al., 2013). In addition, during breast cancer development and progression, the abnormal composition of the ECM surrounding the breast tumour cells has a major impact in modulating a number of biological processes namely, cell survival, inhibition of apoptosis, migration and invasion that control the disease progression (Lu et al., 2012).
During tumorigenesis in the breast, primary tumors are characterised by increased deposition of FN (Christensen, 1992; Vasaturo et al., 2005). FN is known to stimulate breast cancer cell survival through activation and phosphorylation of Phosphoinositide 3-kinase (PI3-K)/protein kinase B (AKT) pathway (Xing et al., 2008). FN is also shown to increase expression of MMP-9 (Maity, Choudhury, et al., 2011) which is found to be associated with lymph node metastasis and higher tumour grades in breast cancer (Wu et al., 2008). Similarly, TN-C induces a number of MMPs, especially MMP-9 in breast cancer cells indicating the involvement of TN-C in breast cancer progression (Kalembeyi et al., 2003). Moreover, an increase in collagen deposition promotes cell survival and proliferation of breast cancer cells through up-regulation of integrin-mediated FAK signalling (Wozniak et al., 2003). Additionally, a growing body of experimental evidence has shown that multifunctional glycoprotein VN can also be involved in breast cancer progression and metastasis by promoting cell survival, migration and invasion (Carriero et al., 1999; Aaboe et al., 2003; Hurt et al., 2010; Pola et al., 2013). Indeed, immunohistochemical evidence indicated increased deposition of VN protein at the
Chapter 1: Literature Review 8
peritumoral stroma, blood vessel walls (Carriero et al., 1997) and in ECM surrounding breast cancer cells (Aaboe et al., 2003), with the VN-binding integrins and urokinase also being up-regulated in breast cancer cells (Andreasen et al., 1997; Meyer et al., 1998; Wong et al., 1998; Rezaeipoor et al., 2009).
1.4 VITRONECTIN
VN is a multifunctional adhesive glycoprotein and was first discovered as a protein that induced the immediate growth of unadapted cells in culture in vitro (Holmes, 1967). Initially, due to its cell attachment promoting activities, VN was described as ‘serum spreading factor’ (Hayman et al., 1983), but has also been named ‘S protein’ (Jenne & Stanley, 1985); and ‘heparopexin’ (Stanley, 1986). VN, with its high degree of conformational flexibility and multimerisation, can exist in both monomeric (native) and multimeric (denatured) forms, which along with its numerous binding domains can confer unique multivalent properties, especially within ECM sites (Stockmann et al., 1993; Seiffert & Loskutoff, 1996). In contrast to classical ECM proteins such as FN, collagen or laminin, VN acts as a matricellular protein and thereby does not contribute structurally to the ECM, but rather acts to coordinate cell-ECM interactions (Preissner & Reuning, 2011). Multimeric VN is a constituent of the ECM and platelet cells, while monomeric VN is found in the circulation. VN is synthesised predominantly by the liver and is present at high concentrations in the circulation (300 to 400 μg/mL) (Izumi et al., 1989; Boyd et al., 1993). Normal tissues generally lack VN immunoreactivity, however, VN deposition is enriched in areas of tissue injury, including membranous nephropathy, arteriosclerosis, degenerative central nervous system disorders and cancers (Reilly & Nash, 1988; Seiffert, 1997). Immobilised VN with high molecular weight and altered configuration is associated with abnormal ECM composition, and it is normally found in areas of sclerosis or fibrosis predominantly in the multimeric form of its structure (Preissner & Reuning, 2011).
1.4.1 Structure of VN VN is a 75 kDa protein containing 459 amino acids and is secreted from its site of production, hepatic cells, through a cleaved leader peptide of 19 residues upon processing (Jenne & Stanley, 1985). VN as a glycoprotein with adhesive properties contains three glycosylation sites, with its carbohydrate moiety accounting for
Chapter 1: Literature Review 9
approximately one-third of its molecular mass. While its three dimensional structure remains unknown, VN contains six domains and multiple binding sites for cells and other components of the ECM (Schvartz et al., 1999). These include, starting from the N-terminus, somatomedin-B domain (SMB), Arginine-Glycine-Aspartate (RGD) motif, a connecting region and two hemopexin-like (HX) domains, the second one of which contains the heparin binding domain (HBD) (Figure 1.2).
The SMB domain (residues 1 – 44) contains amino acid residues involved in binding plasminogen activator inhibitor-1 (PAI-1) and urokinase-type plasminogen activator receptor (uPAR) (Seiffert & Loskutoff, 1991), via which VN can regulate cell adhesion and migration (Wei et al., 1994). Located next to the SMB domain, VN contains an Arginine-Glycine-Aspartate (RGD) recognition sequence (residues 45 –
47) through which it binds a variety of integrin receptors, including αvβ1, αvβ3, αvβ5,
αvβ6, αvβ8 and αIIbβ3 (Felding-Habermann & Cheresh, 1993). Downstream of the RGD sequence there is a stretch of acidic amino acids (residues 53 – 64) sitting within the connecting linker domain (residues 48 – 130) which are involved in binding thrombin-antithrombin (TAT) complexes (Gechtman et al., 1997), and collagen (Izumi et al., 1988). Through an ionic interaction this acidic segment is also known to help neutralise the basic region presented at the carboxyterminal end (residues 348 – 379) (Stockmann et al., 1993). This may help stabilise VN’s three dimensional structure (Schvartz et al., 1999) and is also thought to be involved in its ability to form multimeric forms (Tomasini & Mosher, 1991). Located within the HX-2 at the carboxy-terminus edge (residues 332 – 348) is the binding site for plasminogen (Kost et al., 1992). Downstream of this domain also accommodates a cluster of basic amino acids (residues 348 – 376), containing binding sites for heparin, glycosaminoglycans (Liang et al., 1997) and plasminogen activator inhibitor 1 (PAI-1) (Gechtman et al., 1997). The relative binding positions of proteins interacting with VN are outlined in Figure 1.2.
VN also contains consensus sequences for phosphorylation by various protein kinases, including cAMP-dependent protein kinase, protein kinase C and casein kinase II (CK2), which modulate its conformation as well as function (Shaltiel et al., 1993; Seger et al., 1998). For example, phosphorylation of VN at its Thr50 and Thr57 by CK2 is shown to promote VN-induced cell adhesion and migration (Seger et al., 1998). Additionally, a number of matrix associated proteins are also found to bind to
Chapter 1: Literature Review 10
VN, however the precise effect of these interactions has not yet been identified. For instance, matrix Gla protein (MGP) is found to bind VN which is thought to be important for regulation of growth factors and morphogens, such as TGFβ and human bone morphogenetic protein 2 (BMP-2) to regulate cellular development through migration and differentiation (Nishimoto & Nishimoto, 2005). More recently, matricellular cysteine-rich angiogenic inducer 61 (Cyr61) was found to bind to the SMB domain of both monomeric or multimeric forms of VN with high affinity (Francischetti et al., 2010). Upon association with the ECM, VN from its original closed state becomes “denatured”, or unfolded, and exposes previously cryptic binding domains to form VN multimers, in turn facilitating the binding of VN to other ligands (Morris et al., 1994). Indeed, VN exists within the ECM in the multimeric form which can exert preferential binding to several ligands including integrins, collagen, PAI-1 and uPAR mediating its multifunctional effects (Seiffert, 1997; Felding-Habermann et al., 2001).
Figure 1.1. Schematic representation of the domain structure and key binding regions within VN.
VN includes the somatomedin-B domain (SMB), the integrin-binding RGD domain (RGD), the connecting region (CON) and two hemopexin-like domains (HX-1 and HX-2). The N-terminus (N-) binding sites within VN include those for PAI-I/uPAR (residues 1-44), integrins (45-47), TAT complex and collagen (48-64). Binding sites within the C-terminus (-C) include those for Plasminogen (332-342) and PAI-I/Heparin (343-376). A 19 amino acid signal peptide is cleaved to form the mature 459-amino acid protein (reviewed in Schvartz et al., 1999).
1.4.2 Function of VN VN is capable of modulating numerous biological functions. These include vascular remodelling, thrombosis, hemostasis, fibrinolysis and immune defense through coordination of cell adhesion, survival, and migration primarily by specific
Chapter 1: Literature Review 11
binding interactions with cell surface urokinase and integrin receptors, in particular
αv-subunit containing integrins (Preissner & Seiffert, 1998; Schvartz et al., 1999). Through interactions with uPAR, VN can mediate cell attachment (Wei et al., 1994) while also promoting cell migration through PAI-1 regulation of interactions between uPAR and VN (Waltz et al., 1997). It is also believed that the direct engagement of uPAR by VN can influence the activation state of integrins as well as integrin-mediated adhesion to VN (Madsen & Sidenius, 2008). This is supported by various studies showing that uPAR interacts with different integrins to promote VN signalling, leading to enhanced cell adhesion or migration in a variety of cell types (Degryse et al., 2001; Wei et al., 2001; Wei et al., 2007). Increased cellular migration and invasion is thought to be mediated by the co-localisation of VN-binding uPAR and the αvβ5 integrin within focal contacts of human keratinocytes (Reinartz et al., 1995) and breast cancer cells (Carriero et al., 1999).
VN can also mediates attachment and migration of cells within the ECM via ligation of VN-binding integrins, including αvβ1, αvβ3, αvβ5 and αvβ6, (Felding- Habermann & Cheresh, 1993). VN ligation of integrins causes integrins to become clustered in the plane of the cell membrane, which leads to the formation of focal adhesion contacts linking the cells to the ECM (Giancotti & Ruoslahti, 1999). These focal adhesions are essentially large clusters of integrins, cytoskeletal proteins, growth factors and signalling molecules that facilitate integrin signalling and are critical for cell survival and proliferation. In endothelial cells, VN/αvβ3 ligation is shown to induce survival via Nuclear Factor- kappa B (NF-қB) pathway activation
(Scatena et al., 1998). Likewise, VN engagement with β1, β3 and β5 integrin subunits has been observed to modulate neutrophil survival through integrin activated signalling pathways (Bae et al., 2012). Furthermore, treatment of human umbilical vein endothelial cell (HUVEC) cells with VN also induced their migration through a
αvβ3-dependent pathway (Naik & Naik, 2006). Taken together, it is no surprise that VN interactions with integrins, which are frequently overexpressed in a variety of cancers (Mizejewski, 1999), including breast cancer (Meyer et al., 1998) can modulate cellular processes important in cancer progression.. Indeed, in cancer biology, ECM-integrin engagement is known to not only be critical for coordinating key biological functions, such as adhesion, migration, survival or death responses (Stupack & Cheresh, 2002; Mitra & Schlaepfer, 2006), but can also directly
Chapter 1: Literature Review 12
influence the tumour growth as well as its microenvironment (Desgrosellier & Cheresh, 2010).
1.5 INTEGRIN SIGNALLING
Integrin signalling is complex as the cytoplasmic domains of these receptors are extremely short and lack kinase activity. However, ligation of integrins causes the phosphorylation of a number of cytoplasmic proteins via the association of adapter proteins. A majority of the signalling molecules implicated in ECM-integrin interactions appear to be rather ubiquitous mediators of signal transduction. For example, protein kinases including, focal adhesion kinase (FAK), Src family kinases, Rho family of GTPases, Raf kinases, PI3-K, and Mitogen-activated protein kinases (MAPKs) such as Extracellular signal-regulated kinase (ERK), can be recruited to the ECM-integrin binding site (Giancotti & Ruoslahti, 1999; Stupack & Cheresh, 2002) (Figure 1.3). However, a major hub of integrin signalling is the recruitment and activation of FAK, a cytoplasmic protein kinase that co-localises within focal adhesion sites (Mitra & Schlaepfer, 2006).
Upon ligation of integrins, FAK is auto-phosphorylated on Tyr397 causing a high- affinity binding site for the SH2 domain of Src-family kinases which subsequently lead to the recruitment and activation of Src through the formation of a dual kinase complex (Giancotti & Ruoslahti, 1999). Formation of the FAK-Src complex can phosphorylate various adaptor proteins such as paxillin (Bellis et al., 1995) and p130Cas (Cary et al., 1998) as well as a number of kinases, such as PI3-K, ERK and myosin light chain kinase (MLCK) (Schlaepfer et al., 2004; Webb et al., 2004). Phosphorylation of paxillin and p130Cas recruit the Crk-DOCK180 complex which subsequently activates Rac, a prototypical member of the Rho family, leading to activation of p21-activated kinase (PAK), Jun amino-terminal kinase (JNK), and NF- κB (Parsons & Parsons, 1997; Schlaepfer & Hunter, 1998; Cary et al., 1999). Additionally, FAK phosphorylation recruits the SH domain of PI3-K, allowing translocation of PI3-K-p110 complex (Fruman et al., 1998), leading to phosphorylation of lipid second messengers such as phosphatidylinositol-3,4,5- trisphosphate (PIP3) to bind and recruit AKT (Persad et al., 2001). Morever, FAK phosphorylation activates ERK/MAPK by recruiting the RAP1 guanine nucleotide- exchange factor (GEF) and subsequent activation of proto-oncogene B-Raf (Guo &
Chapter 1: Literature Review 13
Giancotti, 2004). Alternatively, FAK-Src complex can activate ERK/MAPK by requirement and phosphorylation of the growth factor receptor bound-2 (Grb2) and son-of-sevenless (SOS) complex (Wary et al., 1998) (Figure 1.3). These physical cues mediated by constituents of the ECM such as adhesive proteins can control cell behaviour and functions directly through regulation of gene expression programs (Kim et al., 2011).
Figure 1.2. Schematic representations of major signalling pathways stimulated upon ligand-binding to integrin receptors.
Binding of VN, present within the ECM, to its corresponding integrin receptors leads to the formation of FAK-Src complexes which activate downstream signalling pathways, including PI3-K/AKT, JNK,
NF-ҚB and ERK/MAPK pathways. Activation and recruitment of these pathways regulates changes in the gene expression, promoting key cellular functions (reviewed in Guo and Giancotti 2004).
Chapter 1: Literature Review 14
1.6 ECM-INDUCED CHANGES IN GENE EXPRESSION
Recruitment and activation of many aforementioned signalling pathways can influence cellular behaviour and function by mediating changes in downstream gene expression programs. For instance, collagen type I-induced β1 integrin clustering increases expression of the membrane type-1 MMP (MT1-MMP), a major transmembrane collagenase expressed by ovarian tumours as well as early growth response protein 1 (EGR1) through a Src kinase dependent pathway to regulate ovarian cancer cell invasion and metastasis (Barbolina et al., 2007). Collagen type
I/αvβ3 integrin engagement has also been shown to initiate EMT in lung cancer cells by up-regulation of transforming growth factor β (TGFβ)-1 and -3 via activation and involvement of PI3-K and ERK cascades (Shintani et al., 2008). VN ligation to integrin αvβ3 is reported to enhance ovarian cancer cell adhesion, proliferation and invasion by differential expression of a number of genes, including EGF receptor, integrin-linked kinase (ILK), c-myc, prothymosin-α (PTMA), the transcription factor fos-related antigen (Fra-1), and the cell adhesion molecule SQM-1 with key roles in these cellular processes (Hapke et al., 2003). Moreover, FN-binding integrin α5β1 induced down-regulation of the tumour suppressor genes LKB1 and EIF4EBP1 through activation of the Akt/mammalian target of rapamycin (mTOR)/p70S6K pathway can stimulate non-small cell lung carcinoma cell (NSCLC) cell proliferation, invasion and metastasis (Han et al., 2006). In breast cancer cells,
FN/α5β1 engagement induced MMP-9 expression and activity by recruiting ILK, FAK, PI3-K and NF-κB cascades (Maity, Choudhury, et al., 2011). Likewise, laminin, via binding to α2β1 integrin up-regulated MMP-9 mRNA expression in human cervical cancer cell via recruitment of similar pathways (Maity, Sen, et al., 2011). MMPs are proteolytic enzymes that are necessary for tissue remodelling and growth. The overall levels and the activation of the MMPs are known to be higher in tumours compared to their corresponding normal tissues (Egeblad & Werb, 2002). It is likely that altered ECM composition found in tumour tissues are contributing either directly or indirectly to the changes in MMP levels as well as its growth and development.
1.7 VN IN CANCER PROGRESSION
Immunohistochemical studies have reported increased accumulation of VN in a number of human tumours (Gladson & Cheresh, 1991; Tomasini-Johansson et al.,
Chapter 1: Literature Review 15
1994; Aaboe et al., 2003; Edwards et al., 2006). VN has been identified in breast carcinomas where it has been found in the peritumoral stroma, blood vessel walls (Carriero et al., 1997) and in the ECM surrounding cancer cells (Aaboe et al., 2003). VN can modulate cancer metastasis, by providing traction and direction to migrating tumour cells during invasion, and/or migrating endothelial cells during angiogenesis (Bello et al., 2001; Aaboe et al., 2003; Pola et al., 2013), while also conferring a survival advantage to advancing tumour cells (Uhm et al., 1999; Leavesley et al.,
2013). The overexpression of the VN receptor αvβ3 integrin is linked to increased lymphatic dissemination and metastasis in melanoma cells through enhanced binding ability to VN in the lymph node (Nip et al., 1992). Likewise, in ovarian cancer,
VN/αvβ3 complex is reported to result in the activation of FAK/pFAK pathway to ultimately promote MAPK-induced cell adhesion, proliferation and motility (Hapke et al., 2003). Moreover, VN is demonstrated to stimulate prostate cancer cell differentiation, tumour formation, invasion and metastasis through the αvβ3/FAK/Src pathway (Hurt et al., 2010; Franzen et al., 2012).
In breast carcinomas, VN through interactions with αv integrins has been shown to induce migration in a number of breast cancer cell lines (Wong et al., 1998). Indeed, migration of the highly metastatic MDA-MB-231 and MDA-MB-435 breast cancer cell lines is inhibited when αvβ3 and αvβ5 VN-receptors are blocked with antagonists (Bartsch et al., 2003). More recently, interaction of VN and αvβ3 is found to mediate breast cancer cell invasion through a mechanism that involves ILK pathway, and mTOR activation of eukaryotic translation initiation factor 4E (EIF4E)- dependent mRNA (Pola et al., 2013). In addition, interactions between uPAR and the
αvβ5 VN receptor enhance breast cancer cell migration and invasion (Carriero et al., 1999). Similarly, a strong correlation has been observed between expression of both PAI-1and uPAR with the invasiveness of cells derived from normal and malignant mammary tissue (Tong et al., 1999; Zannetti et al., 2000). Recently, small molecules targeting the VN-binding site of uPAR are reported to block migration in response to VN in a number of cancer cell types, including breast cancer (Rea et al., 2013). VN is also shown to drive breast cancer stem cell differentiation and ultimately initiate tumour formation through an αvβ3/FAK/Src-dependent pathway by altering mRNA expression of a number of stem cell genes, including the basal cell maker cytokeratin 14 (K14), luminal marker cytokeratin 8 (K8), Nanog, SMO, Oct3/4 and BMI-1 (Hurt
Chapter 1: Literature Review 16
et al., 2010). Additionally, Ablation of VN or disruption of its uPA and integrin binding sites strongly impaired tumour formation by MDA-MB-231 mammary carcinoma cells xenografted into mice (Pirazzoli et al., 2013).
It is well established that stimulation of integrins and subsequent phosphorylation and activation of intracellular pathways by ECM components, such as VN, causes integrins to become clustered in the plane of the cell membrane. This in turn facilitates the formation of focal adhesion contacts that link the cells to the ECM (Giancotti & Ruoslahti, 1999). Interestingly, present within these focal adhesion contacts are growth factors (GFs). Indeed, integrin clustering and its association within focal adhesions in turn facilitate the formation of integrin-GF complexes that regulate cell behaviour. This in turn provides a mechanism for bi- directional communication or “crosstalk” between receptors and regulation of intersecting signalling pathways that control specific cellular responses to stimuli originating in the extracellular microenvironment (Eliceiri, 2001; Schwartz & Ginsberg, 2002; Beattie et al., 2010).
1.8 INTERACTIONS BETWEEN THE ECM, INTEGRINS, GROWTH FACTORS AND THEIR RECEPTORS
Integrins are reported to facilitate crosstalk with numerous GF receptors and to modulate various cellular functions and behaviours, suggesting a potentially important role for these interactions in cancer biology. For example, FN regulates migration of mesenchymal stem cells (MSCs) during vascular remodelling via αvβ5 integrin-mediated activation of platelet-derived growth factor receptor (PDGFR) and phosphorylation of FAK which subsequently mediates PDGFR-induced PI3-K activity (Veevers-Lowe et al., 2011). Interestingly, FAK is known to act as a receptor-proximal bridging protein that links integrin and GF receptor signalling pathways that activate the GF receptors through its N-terminal domain important for both integrin and GF stimulated effects (Sieg et al., 2000). Indeed, inhibition of both FAK and epidermal growth factor receptor (EGFR) pathways induces apoptosis in human breast cancer cells (Golubovskaya et al., 2002). Similarly, FAK and IGF-IR pathway association has been shown to be critical for human pancreatic adenocarcinoma cell survival (Liu et al., 2008). The αvβ5 integrin and EGFR crosstalk has been reported to promote pancreatic carcinoma cell invasion and metastasis through EGF-mediated recruitment of Src kinase and αvβ5 activation
Chapter 1: Literature Review 17
(Ricono et al., 2009). In addition to EGF, IGF-I is also known to interact with integrin system whereby the IGF-I-mediated adhesion and migration of human melanoma cells was observed to be dependent on β1 integrin activation and PI3-K phosphorylation (Tai et al., 2003). On the other hand, regulation of EGFR signalling by β1 integrins has roles in the maintenance of tumorigenic properties of lung cancer cells (Morello et al., 2011). Interactions between β3 integrin and the TGFβ type II receptor lead to enhanced TGFβ-stimulated MAPK activation, Smad-2/-3/-4- mediated gene transcription and subsequently induction of EMT in mammary epithelial cells (Galliher & Schiemann, 2006). Additionally, it has been suggested that a number of integrins such as αvβ3, αvβ6 and α6β4 cooperate with receptor tyrosine kinases (RTKs) to promote cancer cell invasion and metastasis (Guo & Giancotti, 2004). Indeed the association between integrins and growth factor receptors can lead to partial activation of the growth factor receptor, thereby bringing growth factor signalling closer to a threshold point of initiating activity (Moro et al., 1998).
In addition to the GF receptor-integrin crosstalk, several growth factors modulate their activity by directly binding to the integrins (Eliceiri 2001). For example, direct binding of αvβ3 integrin and TGFβ1 mediates EMT in mammary epithelial cells (Galliher & Schiemann, 2006). Likewise, it was found that
αvβ3/VEGF axis plays an important role in pathological angiogenesis via direct involvement of Src family kinases (Mahabeleshwar et al., 2006; Mahabeleshwar et al., 2007). In addition, the α9β1 integrin dependent binding of nerve growth factor (NGF) is also found to have pro-angiogenic activities in glial tumours (Walsh et al.,
2012). The VEGF ligation to α9β1 induced migration and invasion of primary HUVECs and human dermal microvascular endothelial cells (HMVECs) through the direct activation of the integrin signalling intermediates Src and FAK (Oommen et al., 2011). Interestingly, in recent studies fasciclin 1 (FAS1) domain-containing protein, a soluble αvβ3 integrin-binding molecule, was shown to effectively inhibit tumour angiogenesis by blocking not only αvβ3 integrin-, but also VEGF-mediated angiogenic pathways (Nam et al., 2012).
There have been increasing reports describing the specific and direct binding of GFs to ECM proteins (Hynes, 2009). These ECM:GF complexes are known to interact through direct binding and to activate synergistic or enhance signals
Chapter 1: Literature Review 18
mediating biological functions in normal and pathological conditions (Clark, 2008) such as wound repair (Schultz & Wysocki, 2009) and cancer development (Brizzi et al., 2012). Lin et al demonstrated the direct binding of FN to PDGF through its growth factor-binding domains (FN-GFBDs) and found this interaction to be required for fibroblast survival (Lin et al., 2011). Likewise, VEGF physically links to FN via the heparin-II domain of FN, forming FN:VEGF complex, capable of promoting VEGF-induced biological activities in endothelial cells such as proliferation and migration through strong phosphorylation of VEGF receptor and its downstream signalling ERK/MAPK that may be important for angiogenesis and tumour growth (Wijelath et al., 2006). Indeed, FN can bind to various GFs via multiple binding domains to synergistically enhance the biological activities of both integrin and GF receptors (Zhu & Clark, 2014). In addition, VN is found to bind IGF-I and IGF-II, forming IGF-I:VN and IGF-II:VN complexes (Kricker et al., 2003) that are capable of stimulating breast cancer cell migration and survival via activation of IGF-IR, PI3-K/AKT and ERK/MAPK (Hollier et al., 2008; Kashyap et al., 2011). Furthermore, antagonists to integrin signalling have also been shown to abrogate growth factor receptor-mediated signalling (Beattie et al., 2010), including the IGF-IR (Hollier et al., 2008). Altogether, this indicates an important role for the interaction between VN and its many binding partners, including integrin receptors, and their cross-talk with various signalling systems in processes essential to tissue remodelling and cancer progression.
1.8.1 Interactions between VN, VN-binding integrins and IGF system There is strong evidence for crosstalk being frequently observed between the different receptor-mediated signalling systems even when they are not known to be directly linked to each other. This is also the case for members of the IGF system, where the interplay between IGF signalling and integrin-mediated signalling can modulate IGF biology in normal health and disease states (Beattie et al., 2010). IGF-
I-dependent co-precipitation of IGF-IR clearly demonstrates the presence of αvβ3 integrin (Siwanowicz et al., 2005; Saegusa et al., 2009) and β1 integrin interaction with the IGF-IR (Kiely et al., 2006). The β3 integrin binding site on IGFBP-4 is also distinct from the IGF-I binding site (Siwanowicz et al., 2005). It is known that the ligand occupancy of the VN-binding αvβ3 integrin promotes phosphorylation and activity of the IGF-IR (Maile & Clemmons, 2002) and affects IGF and integrin
Chapter 1: Literature Review 19
signalling mediated by insulin receptor substrate 1 (IRS-1), SHC Src homology 2 domain containing (Shc), PI3K and MAPK (Clemmons & Maile, 2003, 2005).
Studies have shown that αvβ3-integrin occupancy by VN was required for IGF-I stimulated migration of smooth muscle cells (SMC) (Jones et al., 1996; Zheng &
Clemmons, 1998). Transfection of a mutant β3-integrin subunit into Chinese hamster ovarian (CHO) and SMC cells indicated that full length β3-integrin was required for subsequent IGF-I-induced cell migration in a wound assay model (Maile et al.,
2001). Specifically, the VN-binding domain (177CYDMKTTC184) in the β3-integrin is directly involved in the IGF-signalling cascade (Maile et al., 2006), whereas the C- loop region of β3-integrin is critical for VN-induced enhancement of IGF-I signalling and IGF-I-stimulated cell proliferation (Xi et al., 2008).
While the mechanisms outlined regarding SMCs may be specific for αvβ3 in modulating IGF-I responses, there is also evidence that other VN-binding integrins regulate cellular responses to IGFs. It has been reported that VN binding to αvβ5 integrin is required for IGF-I-stimulated cell migration of the MCF-7 breast carcinoma cell line (Doerr & Jones, 1996). Similarly, melanoma cells expressing
αvβ5 required insulin or IGF-I stimulation for migration on VN, whereas cells expressing αvβ3 supported IGF-independent migration (Brooks et al., 1997).
Furthermore, Syndecan-1 (Sdc1), a matrix receptor, is shown to modulate αvβ3 and
αvβ5 integrins activity by physically coupling them with the IGF-IR in human mammary carcinoma and endothelial cells (Beauvais et al., 2009). Furthermore, IGF-
IR docking and activation of the associated integrin is blocked by synstatin (SSTN92–
119), a peptide derived from the integrin engagement site in Sdc1 resulting in the inhibition of angiogenesis and impaired tumour growth in mice (Beauvais & Rapraeger, 2010; Rapraeger et al., 2013).
1.9 THE IGF SYSTEM
The IGF system can have prominent roles in a number of human malignancies and has prompted researchers to unravel its mechanism of action in several carcinomas, including breast cancer (Burroughs et al., 1999; Sachdev & Yee, 2007; Pola et al., 2013). The IGF system involves highly complex regulatory networks that operate at the whole organism level, including cellular and sub-cellular levels that act through endocrine, paracrine and autocrine mechanisms (Jones & Clemmons, 1995).
Chapter 1: Literature Review 20
It comprises a phylogenetically ancient family of peptides involved in mammalian growth and development, as well as a range of cellular processes such as cell proliferation, survival, migration and differentiation in normal as well as neoplastic conditions (De Meyts & Whittaker, 2002). The IGF family comprises the ligands IGF-I, IGF-II and insulin, six high affinity binding proteins (IGFBP-1 to 6) and their proteases, and the ligand binding cell surface receptors, namely the type-I IGF receptor (IGF-IR), cation independent mannose 6-phosphate/type-2 IGF receptor (CIMPR/IGF-IIR), the insulin receptor isoforms A and B (IR-A and IR-B) and the hybrid IR/IGF-IR receptors (HyRs) (reviewed in Annunziata et al., 2011; Martin & Baxter, 2011). Moreover, it includes proteins involved in intracellular signalling distal to IGF-IR, such as members of the IRS family, mTOR and AKT.
1.9.1 IGF system ligands: IGF-I and IGF-II IGF-I and IGF-II are single chain polypeptide growth factors (Wood, 1995). The majority of IGF-I and IGF-II peptides found in the circulation are produced by the liver, although extra-hepatic tissues also possess the ability to synthesize IGFs locally where they can exert autocrine or paracrine effects. The human IGF-I and IGF-II genes are located on the long arm of chromosome 12 (at position 12q22-23) and the short arm of chromosome 11 (at position 11p15.5), respectively (Brissenden et al., 1984; Tricoli et al., 1984). The IGF-I has a complex structure with multiple promoters and contains six exons, four of which are subjected to alternative splicing that generates different precursors without altering the structure of the mature peptide (Rotwein et al., 1986; Smith et al., 2002). On the other hand, IGF-II consists of only three promoters, five non-coding exons and three protein coding exons (Daughaday & Rotwein, 1989). The mature IGF-I is a single chain basic peptide with 70 amino acids and a molecular weight of 7.5 kDa, while IGF-II is a single chain neutral peptide with 67 amino acids and a molecular weight of 7.4 kDa (Daughaday & Rotwein, 1989).
IGF-I and IGF-II resemble insulin in both primary and tertiary structure. However, unlike insulin, IGFs retain the C-peptide of the pro-insulin, in addition to a small D-chain extension to the A-chain of the IGF molecule (Blundell et al., 1978; Rinderknecht & Humbel, 1978). Also, amino acid positions 3, 4, 15, and 16 within the IGF-I molecule facilitate selective binding of IGF-I to a family of six high affinity binding proteins (IGFBP-1 to -6) (Annunziata et al., 2011), further
Chapter 1: Literature Review 21
differentiating them from insulin, which does not bind to the IGFBPs. This allows the majority of the IGFs (~75%) to circulate as a 150 kDa complex with the acid labile subunit (ALS) and IGFBPs (Baxter et al., 1989; Leong et al., 1992), while insulin circulates unbound. IGFs are commonly found complexed with IGFBP-3 or IGFBP-5 (Clemmons, 1998; Martin & Baxter, 2011). Functionally, both IGF-I and IGF-II play central roles in stimulating tissue growth and maintenance especially during fetal development (Streck et al., 1992), normal postnatal growth (Yakar et al., 1999), tissue repair after injury (Jennische et al., 1987) and during tumorigenesis (Pollak, 2012). The stimulatory effect of the IGFs, in normal tissue growth and in disease progression, is related to their ability to induce angiogenesis, and their influence on cellular mechanisms that govern proliferation, differentiation, apoptosis and migration (Samani et al., 2007).
1.9.2 IGF binding proteins (IGFBPs) The IGFs in circulation are bound to a family of six high affinity binding proteins called the IGFBPs. The precursor forms of all IGFBPs contain a secretory signal peptide which is cleaved upon secretion into their mature forms (216-289 amino acids), with molecular masses ranging from ∼24 to 50 kDa (Firth & Baxter, 2002; Forbes et al., 2012). There are numerous reports indicating that IGFBPs prolong the half-life of bound IGFs to slow the clearance of IGFs and to maintain a large reservoir of IGF-I and IGF-II in the circulation (approximately 100 nM). Importantly, many IGFBPs, specifically IGFBP-2, -3 and -5, have defined heparin- binding domains that are known to associate with several matrix proteins (Russo et al., 2005; Beattie et al., 2009; Sureshbabu et al., 2009). These extracellular matrix- bound IGFBPs might serve as reservoirs of IGFs within tissues, concentrating IGFs in close proximity to the IGF-IR to mediate receptor activation (Jones et al., 1993; Beattie et al., 2010; Martin & Baxter, 2011). In addition, via its interaction with matrix proteins such as VN and FN, IGFBP-bound IGFs may enhance signalling by activating integrin and IGF-IR−integrin system cross-talk (Hollier et al., 2008). The increased proteolytic activities observed in the ECM of neoplastic tissues or tissues undergoing active remodelling, may lower the affinity of matrix-bound IGFBPs to IGFs, allowing for increased receptor activation (Yan et al., 2004).
Chapter 1: Literature Review 22
1.9.3 Type-I IGF receptor (IGF-IR) The IGF-IR is a transmembrane tyrosine kinase receptor that mediates the effects of IGF-I and IGF-II. IGF-IR is a heterotetrameric protein, consisting of two di-sulphide bond-linked α- and β-subunits (Ulrich et al., 1986). The cysteine-rich α- subunits are extracellular and form the ligand-binding domain, while the β-subunits, which contain short extracellular and transmembrane segments and a larger intracellular segment, transmit the ligand-induced signal (LeRoith et al., 1995). The IGF ligands bind to IGF-IR at the α-subunit, with a 4-fold higher affinity for IGF-I compared to IGF-II (Forbes et al., 2002).
1.9.4 IGF-IR: Signalling Ligand binding to IGF-IR triggers the activation of several receptor substrates, including the IRS and Shc (Butler et al., 1998; Samani & Brodt, 2001). Phosphorylated IRS-1 activates the p85 subunit of PI3-K, which in turn activating AKT cascade (Giorgetti et al., 1993) to stimulate protein synthesis and to prevent apoptosis (Gilmore et al., 2002). IRS-1 is also involved in signalling events mediated by other growth factors (Siddle, 2012) and integrins, such as αvβ3 (Maile &
Clemmons, 2002) and β1 (Wang et al., 2007), and may be an important part of the crosstalk between the different signalling systems. Phosphorylated IRS-1 can also recruit Grb2/SOS to activate the Raf-1/MEK/ERK pathway and other nuclear factors, resulting in cell proliferation (Hermanto et al., 2000; Grey et al., 2003). The dysregulation or hyperactivation of these pathways can have significant effects on the development of malignant tumours (Santarpia et al., 2012; Sheppard et al., 2012), including breast cancer (Samani et al., 2007).
1.9.5 Type-2 IGF receptor (IGF-IIR) The typeII-IGF receptor (IGF-IIR) is a monomeric receptor that binds with high affinity to IGF-II (Morgan et al., 1987). The IGF-IIR is not a tyrosine kinase receptor (El-Shewy et al., 2007), but it is involved in the lysosomal degradation of various proteins including IGF-II and mannose-6-phosphate containing proteins to control their extracellular availability (Pollak et al., 2004; Brown et al., 2009) Although IGF-IIR is not directly involved in tyrosine kinase dependent signalling pathways, evidence suggests that IGF-IIR can modulate the activities of G-proteins to influence downstream signal transduction. Supporting this view, IGF-II/IGF-IIR
Chapter 1: Literature Review 23
binding has been shown to activate ERK1/2 signalling to promote cell survival (El- Shewy et al., 2007).
1.10 THE IGF SYSTEM AND BREAST CANCER
The IGF system plays a key role in the biological functions of various tissues, with a highly controlled regulatory network to reflect its importance in maintaining the normal tissue phenotype. Likewise, the IGF system is also heavily involved in tumour development, especially that of breast cancer (Maki, 2010; Becker et al., 2011; Gallagher & LeRoith, 2011; Franks et al., 2012; Baxter, 2014). Epidemiological studies have suggested that elevated circulating levels of IGF-I can be associated with an increased risk of breast cancer among premenopausal women (Hankinson et al., 1998; Toniolo et al., 2000). IGF-I can rescue breast cancer cells from chemotherapy-induced cell death by induction of proliferation and inhibition of apoptosis (Gooch et al., 1999). Stimulation of breast cancer cells by IGF-I has been shown to stimulate a gene expression signature that most closely resembles basal-like (ER and HER2 negative) and the majority of luminal-B subtype tumours (ER positive but highly proliferative disease) which are highly correlated with poor clinical outcomes (Creighton et al., 2008).
In vitro and in vivo studies suggested that IGF-II plays an important role in promoting angiogenesis (Kim et al., 1998; Bid et al., 2012), a process critical to the growth of cancer cell. This is evidenced by highly vascularised tumours displaying increased IGF-II expression in the stroma during the transition of normal epithelium to dysplastic and neoplastic breast disease (Heffelfinger et al., 1999). This stromal- epithelial interaction is also thought to be critical for IGF action on breast cancer cells, as IGF-I and IGF-II are predominantly of stromal origin (Giani et al., 2002), with IGF-II also expressed by some breast epithelial cells (Yee et al., 1988).
In vitro and in vivo evidence strongly suggests that IGF-IR, the predominant receptor for the IGF system, plays a key role in breast cancer progression. IGF-IR is reported to be 14-fold higher in malignant breast cancer tissue compared to normal breast tissue (Resnik et al., 1998). Overexpression of IGF-IR in breast epithelial cells disrupts mammary ductal morphogenesis (Yanochko & Eckhart, 2006) and predominantly induces the formation of basal-like breast tumours (Franks et al., 2012). There is also evidence linking IGF-IR hyperphosphorylation with the early
Chapter 1: Literature Review 24
stages of breast cancer, and primary breast tumours, with high IGF-IR levels being associated with shorter disease-free survival (Surmacz, 2000; Samani et al., 2007; Yerushalmi et al., 2012). Indeed the direct targeting of IGF-IR mRNA using antisense oligonuclietides by either direct intra-tumour or systemic administration of phosphorothioate antisense oligonucleotide (AS[S]ODNs) reduced mammary tumour growth in mouse models (Salatino et al., 2004).
As noted earlier, the biological effects of IGF ligands and their signalling can be modulated by various ECM components and their binding receptors, including VN and VN-binding integrin receptors. Furthermore, integrin antagonists have also been shown to block IGF-IR-mediated signalling, suggesting the existence of crosstalk between growth factor and the integrin-mediated signalling pathways (Beattie et al., 2010). In addition, IGF-I co-precipitation of IGF-IR occurs in the presence of αvβ3 integrin (Siwanowicz et al., 2005; Saegusa et al., 2009) and β1 integrin interaction with the IGF-IR (Kiely et al., 2006). Moreover, the β3 integrin binding site on IGFBP-4 is reported to be distanced from the IGF-I binding site (Siwanowicz et al., 2005), suggesting that IGF-IGFBP complexes may also interact with β3-integrin subunits. Furthermore, intracellular signalling mediated by the IGF- I:IGFBP association with VN required the activation of both the IGF-IR and VN- binding integrins (Hollier et al., 2008). As the members of the IGF system and the VN/integrin axis lie in close proximity to each other on the cell surface the adhesion to VN and/or VN-containing IGF:VN complexe promote receptor cross-activation to synergistically enhance cell survival and migration. Indeed, the dual nature of activation of IGF-I and the integrin axes has recently been exploited to produce chimeric IGF-VN ligands (Van Lonkhuyzen et al., 2007; Upton et al., 2008). These chimeric molecules can simultaneously engage the IGF-IR and integrins to stimulate migration of MCF-7 breast cancer cells (Van Lonkhuyzen et al., 2007; Kricker et al., 2010).
1.11 EVIDENCE FROM THE TRR LABORATORY SUPPORTING THE ROLE OF VN AND IGF INTERACTIONS IN BREAST CANCER PROGRESSION
Research undertaken in the Tissue Repair and Regeneration (TRR) laboratory at QUT has led to the discovery that IGF-II binds directly to VN to form a dimeric IGF-II:VN complex (Upton et al., 1999). Furthemore, it was also found that IGF-I
Chapter 1: Literature Review 25
can form trimeric IGF-I:IGFBP:VN complexes through the association of IGF-I with VN via IGFBP-2, -3, -4 and -5 (Kricker et al., 2003). Moreover, functional studies revealed that the proliferation and migration of several epithelial cells, including HaCaT keratinocytes and corneal epithelial cells are enhanced in the presence of the dimeric IGF-II:VN (Upton and Kricker, 2002; Hyde et al., 2004) and trimeric IGF- I:IGFBP:VN complex (Hollier et al., 2005; Ainscough et al., 2006; Kricker et al., 2010).
As outlined in the previous section, given that there is a crosstalk between the IGF system and VN/VN-binding integrins, and that members of both these protein families have been shown to be overexpressed in breast cancer, the effects of these complexes on breast cell function is an important area of research in breast cancer biology. Indeed studies have shown that the IGF-II:VN complex (Noble et al., 2003) and the IGF-I:IGFBP-5:VN complex can stimulate the migration of MCF-7 cells in vitro, and that it is mediated through the IGF-IR only in the presence of the multimeric complexes (Hollier et al., 2008). Furthermore, results from previous studies indicated that both MCF-7 breast cancer cells and the non-tumorigenic MCF10A breast epithelial cell line show increased migration in response to the IGF- I:IGFBP:VN complex (Hollier et al., 2008). These effects were shown to be mediated through the sustained activation of the PI3-K/AKT pathway and the transient activation of the ERK/MAPK pathway (Hollier et al., 2008; Kashyap et al.,
2011). Furthermore, function blocking antibodies against the IGF-IR and αv integrins demonstrated the activation of both these receptors, in essence IGF-IR/integrin cross- talk, was critical for the IGF-I:IGFBP:VN-stimulated migratory effects (Hollier et al., 2008).
Global gene expression analysis of MCF-7 breast cancer cells stimulated to migrate through Transwells™ in response to IGF-I:IGFBP:VN complex has identified a number of genes, such as stratifin (SFN), tissue factor (TF), ephrin-B2 (EFNB2) and sharp-2, to be uniquely dysregulated in these migrated cells (Kashyap et al., 2011). These genes are known components of the IGF:VN/integrins axis and have well documented roles in cell migration and/or survival. Further functional studies revealed that TF and SFN play important roles in the IGF-I:IGFBP:VN complex- stimulated breast cancer cell migration (Dr Abhishek Kashyap, PhD thesis). Additionally, cell survival was also investigated wherein the trimeric IGF-
Chapter 1: Literature Review 26
I:IGFBP:VN complex stimulated growth and viability of breast cancer cells in 2D and 3D cultures (Dr Abhishek Kashyap, PhD thesis) and protected these cells from drug-induced apoptosis (Kashyap et al., 2011). However, further studies need to be undertaken to identify the downstream molecular mediators of IGF-I:IGFBP:VN- stimulated breast cancer cell survival and functionally validate the involvement of candidate target genes in this biological response.
1.12 CONCLUSION AND KNOWLEDGE GAPS
This literature review summarised the current knowledge supporting the critical role of ECM proteins and their integrin binding receptors in cell behaviour and function, and how its dysregulation is associated with cancer development and progression. VN is one such ECM protein which has been shown to play an important role in regulating a number of key biological functions, in particular, cell survival and migration during cancer progression. However, there remains a large gap in our understanding of specific mediators of VN-induced biological functions. Furthermore, this review highlighted the key and often overlooked influence of the ECM on the way cells respond to growth factors, specifically during breast cancer progression. Indeed, it was shown that VN, its αv integrin receptors and the IGF system were all frequently overexpressed in several types of breast cancer. It i also reported that interactions between these proteins could lead to increased breast cancer cell migration via dysregulation of a number of important downstream molecular mediators. These IGF:VN interactions were also found to be potent stimulators of cell viability and drug-induced apoptosis in breast cancer cells. Currently, no study has investigated the gene expression programs altered by VN and/or the IGF:VN complexes in the context of breast cancer cell survival. Indeed, identification of such changes in gene expression will not only increase our understanding of the mechanisms responsible for VN- and the IGF:VN complex- stimulated survival, but may also provide insights into the early molecular events promoting breast cancer metastasis.
Chapter 1: Literature Review 27
1.13 PROJECT OUTLINE
1.13.1 Hypotheses The underlying hypotheses explored in my PhD studies were:
1) Breast cancer cells that have survived in response to VN and the IGF:VN complexes differentially express genes with roles in cell survival;
2) Differential expression of the candidate genes cysteine rich angiogenic inducer 61 (Cyr61) and connective tissue growth factor (CTGF) by VN
and the IGF:VN complexes are mediated via the αv integrin and activation of downstream Src, MAPK/ERK and PI3-K/AKT signalling pathways; and
3) VN- and the IGF-I:VN complex-induced breast cancer cell survival will be inhibited upon the downregulation of either Cyr61 and/or CTGF genes.
1.13.2 Aims In order to address the aforementioned hypotheses, the aims of my PhD studies were to:
1) Use microarray expression profiling to screen for candidate genes involved in VN- and IGF:VN complex-induced breast cancer cell survival;
2) Identify differentially expressed candidate genes with important roles in breast cancer cell survival utilizing ingenuity pathway knowledge base and gene ontology analyses;
3) Investigate the possible mechanisms underlying regulation of candidate genes by VN and the IGF:VN complexes; and
4) Functionally validate the role of the candidate genes in VN- and IGF:VN-induced breast cancer cell survival and migration.
Chapter 1: Literature Review 28 Chapter 2: Materials and Methods
Chapter 2: Materials and Methods 29
2.1 INTRODUCTION
All materials and methods referred to in the following results chapters are outlined in this chapter. Any deviations from the standard protocols are described in the individual results chapters where they appear.
2.2 PROTEINS AND REAGENTS
Human VN was purchased from Promega (NSW, Australia). Human IGF-I, [L24][A31]-IGF-I and IGFBP-5 were obtained from GroPep (SA, Australia). Fraction V Bovine Serum Albumin (BSA) was purchased from HyClone (VIC, Australia). Recombinant human Fibronectin (FN), polyclonal sheep anti-human Cyr61/CCN1 and goat anti-human CTGF/CCN2 antibodies for immunoblot analyses were purchased from R&D Systems (MN, USA). Recombinant human collagen type I was obtained from FibroGen (CA, USA). Mouse monoclonal anti-human Cyr61 and CTGF antibodies for immunofluorescence staining were from Santa Cruz
Biotechnology (CA, USA). Mouse monoclonal antibodies against αv-integrin subunit
(AV1), αvβ3 (LM609), αvβ5 (P1F6), β1-subunit (P4G11) and the IgG matched control antibody were obtained from Chemicon (CA, USA). The pharmacological inhibitors LY294002, U0126 and PP2 were purchased from Merck Biosciences (Darmstadt, Germany). Alexa Fluor®-conjugated secondary antibodies, Fluorescein Diacetate (FDA) for staining viable cells, as well as phalloidin and DAPI (4',6-diamidino-2- phenylindole) fluorescent stains for immunofluorescence were purchased from Life Technologies (VIC, Australia). AlamarBlue and Trypan blue reagents for cell viability measurement and TRIzol® reagent for RNA extraction, SuperScript® III first-strand cDNA synthesis kit, SYBR green qPCR master mix and 96-well plates for RT-PCR analysis were also purchased from Life Technologies. Direct-Zol™ miniprep kit for RNA purification was obtained from Zymo Research (CA, USA). Gene specific short hairpin RNA (shRNA) against Cyr61 and CTGF target genes cloned into pLKO.1 lentiviral vectors were obtained from Open Biosystems (Thermo Fisher Scientific; VIC, Australia). Lentiviral packaging vectors pCMV-VSV-G and pCMV-delta-R8.2 were purchased from Addgene (MA, USA) and the FuGENE-6 transfection reagent was from Promega. Stbl3 E.coli competent cells were obtained from Life Technologies. Growth factor-reduced (GFR)-Matrigel™ and glass chamber slides were obtained from BD Biosciences (CA, USA). Nunclon™∆ surface
Chapter 2: Materials and Methods 30
multi-well cell culture plates were obtained from Nalge Nunc International (Roskilde, Denmark) and 8-μm pore 24-well Transwells® plates were purchased from Corning Inc. (NY, USA). All other reagents were purchased from Sigma-Aldrich (NSW, Australia) unless otherwise stated.
2.3 CELL CULTURE
The human breast adenocarcinoma MCF-7 (Cat no. HTB-22), T47D (Cat no. HTB-133), MDA-MB-231 (Cat no. HTB-26) and MDA-MB-468 (Cat no. HTB-133) cell lines were obtained from the American Type Culture Collection (ATCC; Manassas, VA). HEK293T human embryonic kidney cell lines were a gift from Dr Brett Hollier (Institute of Health Biomedical Innovation). Fetal Bovine Serum (FBS) growth supplement for cell culturing was purchased from HyClone (VIC, Australia). All other reagents used for cell culturing, including cell culture medium and penicillin + streptomycin mixture were purchased from Life Technologies. MCF-7 and T47D cell lines were maintained in Dulbecco's Modified Eagle’s Medium/Nutrient Mixture F-12 (DMEM/F12) medium (1:1) supplemented with 10% FBS whereas highly metastatic MDA-MB-231 and MDA-MB-468 cells were maintained in Roswell Park Memorial Institute 1640 (RPMI 1640) medium containing 5% FBS. HEK293T cells were maintained in DMEM high glucose medium (Life Technologies) containing 10% heat inactivated FBS (iFBS). All media were also supplemented with penicillin + streptomycin (100 µg/ml) and gentamicin (10 mg/mL). All cultures were passaged regularly every 2-3 days using trypsin- EDTA detachment and stocks of cells were maintained by freezing 2 x 106 cells suspended in liquid nitrogen cyropreservation medium, which was growth medium containing 40% FBS and 10% DMSO (Dimethyl sulfoxide).
2.4 TREATMENT APPROACH FOR IN VITRO INVESTIGATIONS OF VN AND IGF:VN COMPLEX-INDUCED CELL FUNCTIONS
A standard methodology employed in vitro to study the effects of growth factors on cells is to add the growth factors to the culture medium. This results in the growth factors being in a solution-phase, whereby they are constantly exposed to the cells. This is despite the fact that in vivo it is thought that cells within organised tissues respond to growth factors bound within the ECM. Therefore, in an effort to more accurately mimic the in vivo cellular environment this study adopted the
Chapter 2: Materials and Methods 31
“substrate-bound” strategy of pre-binding VN with or without growth factors to polystyrene multi-well cell culture plates prior to cell seeding. Growth factors and VN suspended in serum-free medium (SFM) were pre-bound to the tissue culture plastic for all viability assays as well as the lower chamber/under-side of the porous membrane of Transwells® in migration assays prior to seeding cells according to procedures previously described in our laboratory (Kricker et al., 2003; Noble et al., 2003; Hyde et al., 2004; Van Lonkhuyzen et al., 2007; Hollier et al., 2008; Kashyap et al., 2011).
2.4.1 Pre-binding of VN, IGF-I, IGF-II and IGFBP-5 to 6- and 96-well plates
Previous studies within our laboratory (Hollier et al., 2008; Kashyap et al., 2011) have shown that optimal IGF-I:IGFBP-5:VN-mediated attachment, survival and migration of MCF-7 cells in multi-well plates is obtained by pre-binding human native-IGF-I, IGFBP-5 and VN at concentrations of 30 ng/mL, 90 ng/mL and 1 μg/mL, respectively. The optimal IGF-II:VN concentrations have also been pre- optimised to be 100 ng/mL of human native IGF-II and 1 μg/mL of VN (Noble et al., 2003; Hyde et al., 2004). Hence, similar concentrations of these proteins were used for the experiments reported throughout this study. Nunclon™ 6- or 96-well plates were prepared by pre-binding VN either alone or in combination with IGF-II or IGF- I+IGFBP-5 following procedures previously described (Kricker et al., 2003; Noble et al., 2003; Hyde et al., 2004; Hollier et al., 2008; Kashyap et al., 2011). Briefly, DMEM/F12 serum-free phenol-red free medium (SFM) containing 1 μg/mL VN was added to culture plates and incubated for 3 hours at 37 °C. Unbound VN was removed and the wells were washed twice with 1 mL SFM + 0.5% BSA. Subsequently, growth factors suspended in 0.5 mL SFM + 0.05% BSA containing native IGF-I (30 ng/mL), IGFBP-5 (90 ng/mL) or IGF-II (100 ng/mL), either alone or in combinations, were added to the culture plates and incubated overnight at 4 °C. Following day the solution containing unbound growth factors was removed and the wells were washed twice with SFM + 0.05% BSA and replaced with SFM + 0.05% BSA prior to cell seeding. All volumes added were adjusted to compensate for the relative differences in the culture surface areas, with volumes being 2 mL for 6-well plates and 100 μL for 96-well plates.
Chapter 2: Materials and Methods 32
2.4.2 Pre-binding of VN, IGF-I and IGFBP-5 to Transwells®
Cell migration in response to VN and IGF-I:IGFBP:VN complexes was assessed using the 8-μm pore 24-well Transwell® plates (Corning Inc.). Pre-binding of VN, either alone or in combination with IGF-I and IGFBP-5 to the lower chamber and under-side of the Transwell® membrane was carried out according to the procedures used to pre-bind 6-well and 96-well plates (outlined in section 2.4.1). The volume added was adjusted to compensate for the relative differences in the culture surface areas, with volumes being 0.5 mL for Transwell®.
2.5 MTS VIABILITY ASSAY
The cells were split 1:1 24 hours prior to performing the viability assay to assist in synchronising the cells. The cells were serum-starved for 4 hours before seeding 5 x 103 cells suspended in 100 μL SFM + 0.05% BSA into 96-well plates pre-coated with VN, either alone or in combination with IGF-II or IGF-I+IGFBP-5.
The cells were incubated for 72 hours at 37 °C, 5% CO2 and cellular viability was assessed using the CellTitre 96® AQueous One Solution Cell Proliferation Assay (MTS) (Promega). The MTS Assay is a colorimetric method for determining the number of viable cells in proliferation, cytotoxicity or chemosensitivity assays. The CellTiter 96® AQueous One Solution Reagent contains a tetrazolium compound [3- (4,5-imethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H tetrazolium, inner salt; MTS] and an electron coupling reagent (phenazine ethosulfate; PES). After 72 hours, 20 μL of the MTS reagent was added into the spent medium in each well, incubated for 3 hours at 37 °C, 5% CO2 to allow for colour development and the absorbance was then recorded at 490 nm using a 96-well plate reader (Benchmark Plus, Bio-Rad). The quantity of formazan product as measured by the amount of 490 nm absorbance is directly proportional to the number of viable cells in culture (Cory et al., 1991). Wells containing SFM + 0.05% BSA medium without the cells were used as blanks.
2.6 TRANSWELLS® MIGRATION ASSAY
Migration assays were performed following protocols described previously by our laboratory (Kricker et al., 2003; Hollier et al., 2008). The cells in culture flasks were split 1:1 24 hours prior to performing the migration assay. The cells were serum-starved for 4 hours and a cell suspension containing 6 x 104 cells in 200 μL
Chapter 2: Materials and Methods 33
SFM + 0.05% BSA was seeded onto the upper-surface of the Transwell® inserts pre- coated on the underside with VN either alone or in combination with IGF-I+IGFBP-
5 and incubated for 15 hours at 37 °C, 5% CO2. Non-migrated cells from the upper- surface were removed using cotton-swabs. The cells that had migrated to the lower- surface of the Transwell® membranes were fixed in 3.7% para-formaldehyde (20 minutes) and stained with 0.01% Crystal Violet/PBS solution (20 minutes). The Transwell® inserts were washed to remove excess stain and the number of cells that migrated to the lower-surface of the membranes was quantified by extracting the crystal violet stain in 10% acetic acid (10 minutes) and determining the optical density of the extracted stain at 595 nm using a 96-well plate reader (Benchmark Plus, BioRad). Transwells® suspended without cells in SFM + 0.05% BSA were used as blanks.
2.7 THREE-DIMENSIONAL (3D) MATRIGEL™ CULTURES
A growing body of evidence suggests that the use of three-dimensional (3D) cell culture models, in contrast to the two-dimensional (2D) monolayer culture models, replicates more accurately depicts the microenvironment cells encounter within tissues; therefore the behaviour of cells cultured in 3D is more likely to exhibit responses observed in vivo (Abbott, 2003; Schmeichel & Bissell, 2003; Debnath & Brugge, 2005; Weigelt & Bissell, 2008). Hence, to confirm the findings from our 2D studies in a more relevant culture system, cells were grown in 3D using Matrigel™ as the scaffold. Matrigel™ basement membrane matrix (BD Biosciences; NSW, Australia) is a solubilised basement membrane preparation extracted from the Engelbroth-Holm-Swarm (EHS) mouse sarcoma, a tumour rich in extracellular matrix proteins including laminin (the major component of Matrigel™), collagen IV, heparan sulphate proteoglycans, and entactin/nidogen (Kleinman et al., 1986). Epithelial cells cultured in 3D Matrigel™ tend to grow in cell colonies resembling spheroids, whereas mesenchymal cells or highly invasive cancer cells tend to invade into the Matrigel™ and form spindle-like connections.
The approach chosen for these assays was “3D on top” wherein the cells, suspended in growth medium containing 2.5% Matrigel™, were gently layered onto a layer of 100% Matrigel™ and allowed to form spheroids on the Matrigel™ layer. In the studies reported in this thesis such 3D cultures were set up to assess cell
Chapter 2: Materials and Methods 34
proliferation, survival, invasive capabilities and the size and shape of the spheroids formed by the weakly metastatic MCF-7 and highly metastatic MDA-MB-231 breast cancer cells. Assays were performed in Nunc™ 96-well optical-bottom plates (Thermo Fisher Scientific; VIC, Australia) so that they were suitable for downstream image analyses using fluorescence or confocal microscopy. The bottom of each well was first layered with 30 μL of 100% Matrigel™ and 1,000 cells suspended in 100 μL of 3D growth medium (3D-GM; DMEM/F12; 2.5% Matrigel™ + 1% FBS) were then seeded onto the 100% Matrigel™ layer. As these studies were performed to investigate the effects of the VN and IGF-I on cell behaviour, the growth factor- reduced version of the Matrigel™ was used. The cells were then incubated at 37 °C,
5% CO2 for three days without medium change. Care was taken to avoid dislodging the cells layered onto the 100% Matrigel™. Once visible spheroids of cells were formed (Day 3) the spent medium from the treatment-wells was removed and replaced with 100 μL solutions of 3D growth medium (DMEM/F12; 2.5% Matrigel™ + 1%FBS) containing VN (1 μg/mL) either alone or in combination with IGF-I (100 ng/mL) and IGFBP-5 (300 ng/mL). Solutions containing 3D growth medium with 1% and 5% serum served as negative and positive controls, respectively. To facilitate the formation of the protein complexes, the protein + 3D growth medium solutions were incubated at 2 hours at 37 °C, 5% CO2 prior to transferring into their respective wells. Duplicate wells were used for each treatment.
The cells were then incubated at 37 °C, 5% CO2 for a further 6 days (for MDA-MB- 231 cells) or 9 days (for MCF-7 cells), with spent medium replaced every three days with fresh 3D growth medium containing the respective proteins. Cell morphology was observed every day with images recorded every three days using a phase contrast microscope (Nikon Eclipse TS100, Nikon; Libcombe, NSW, Australia). After Day 9 or 12, the cells were either subjected to AlamarBlue assay to assess cell survival/viability, or were processed for fluorescein diacetate (FDA) staining to be analysed using ImageJ software.
2.7.1 AlamarBlue viability assay on cells grown in 3D Matrigel™ cell culture AlamarBlue assay was used to assess the survival/viability of cells grown in 3D Matrigel™ cultures. AlamarBlue, also known as resazurin, is a water soluble and non-toxic dye that is transformed from blue colour to the highly fluorescent resorufin (pink) within a redox reaction process. Since the AlamarBlue assay is based on
Chapter 2: Materials and Methods 35
fluorescence, little or no interference is recorded by the 3D Matrigel™ mesh. The fluorescence intensity is directly proportional to the percentage of viable cells within the culture.
The AlamarBlue assay was performed following protocols reported by Loessner et al. (2010), with minor modifications. Briefly, spent medium was removed from cells grown in 3D and replaced with 100 μL of fresh 3D growth medium containing 5% (v/v) AlamarBlue solution (Life Technologies). AlamarBlue added to wells with 3D growth medium without cells served as blanks. The plates were incubated for 4 hours at 37 °C, 5% CO2 prior to being assayed. Fluorescence signals (excitation 544 nm, emission 590 nm) were detected using fluorescence plate reader (BMG PolarStar; BMG Labtech, Offenburg, Germany). Cell survival was calculated as a percentage of the fluorescence detected in untreated cells.
2.7.2 FDA staining of cells grown in 3D Matrigel™ cell culture To visualise live cells and to reduce the signal to noise ratio for analyses using the ImageJ quantification software, 3D Matrigel™ cell cultures growing in 96-well plates were stained using the fluorescein diacetate (FDA) (Life Technologies). FDA is metabolised to fluorescein using acetyl esterase within the living cells. This polar fluorescent compound is retained within the live cells (Calich et al., 1979).
Prior to staining, the AlamarBlue reagent that was added into the wells to measure cell viability was removed and cells were gently washed twice with 200 μL PBS. Cells were then stained with 100 μL FDA (20 µg/Ml) (1:500 in PBS) followed by 10 minutes incubation before imaging by fluorescence microscopy (Nikon Eclipse TS100, Nikon).
2.7.3 ImageJ analysis of FDA-stained cells ImageJ, a publically available Java-based image processing software developed by the National Institutes of Health (NIH), was used to quantify the size and the shape of breast cancer cells grown in 3D Matrigel™ cultures. In order for ImageJ analyses to be performed on 3D cultured cells phase-contrast images were deemed sub-optimal due to the high signal to noise ratio. In order to remove the background the cells within the Matrigel™ cultures were fluorescently stained (using FDA), resulting in a dark background with only the live cells emitting fluorescence. Images were then recorded and were subjected to analyses via ImageJ. ImageJ calculates
Chapter 2: Materials and Methods 36
area and pixel value statistics of user-defined selections with which the relative size or circularity of the cell colonies can be calculated. Once opened in ImageJ the image was converted to mono-chrome black and white image with the background being white and the spheroids being detected as dark opaque circles. The threshold value above/below which the spheroids are considered for further calculations can be set. The scale can be set in units of length (here in μm) which then converts pixels into length (here μm). Once these values were set the images were processed, with the software reporting the number of spheroids/colonies, the size of each spheroid (in μm2) and also its circularity score (ranging from 0 to 1; 0 being least circular) for each image. Once the spheroids identified by the software were manually checked, the results were exported to an Excel spreadsheet (.xls file) for data analysis. Three such images from 3 different quadrants within each well were recorded for each treatment, with treatments tested in duplicate in each experiment. The values for all such treatments from duplicate experiments were averaged, the standard error calculated and data plotted.
2.8 RNA EXTRACTION FROM MONOLAYER CELL CULTURES
2.8.1 Cell Lysis To identify the specific induced gene expression responses in cells treated with VN alone or in combination with IGF-I+IGFBP or IGF-II, all experiments were prformed in serume-free medium (SFM) to minimise binding of any inducer compound to albumins contained in serum. Due to the potential complications in affecting downstren gene expression and subsequently the data interpretation when serum-containig medium is used, cell grown in untreated control cell (SFM only) were used for comparative puposes. Since, MCF-7 cells attach weakly to the tissue culture plate in the absence of any adhesive ECM proteins, in all experiments reported throughout this study both attached and floating cells were collected for RNA and protein extraction to have more fair comparison between treated and untreated cells. For RNA extraction floating cells were first collected and the attached cells were harvested by adding 250 µL cell dissociation buffer (TrypLE™) to each well. For each treatment attached and floating cells were combined and collected as a cell pellet prior to lysis. The supernatant was carefully removed and the cell pellet was lysed by adding 1 mL TRIzol® reagent (Life Technologies) to
Chapter 2: Materials and Methods 37
isolate total RNA from cells lysates were homogenised carefully by pipetting and stored at -80 °C until further use.
2.8.2 RNA isolation from cell lysate RNA was extracted from cell lysates using the Direct-zol™ RNA miniprep Kit (Zymo Research) following the manufacturer’s protocol. Briefly, one volume ethanol (95-100%) was added directly to the samples homogenised in TRIzol® reagent and mixed well by vortexing. Mixtures were then loaded into Zymo-Spin™ IIC column in a collection tube and centrifuged at ~12,000g for one minute. Spin columns were then transferred into a new collection tube and treated with DNase cocktail for 15 minutes at room temperature. The columns were washed with RNA pre-wash and RNA wash buffers to remove any cell debris and other solvents. Finally, RNA was eluted using 30-50 μL RNase-free MilliQ water and stored at -80 °C until further use.
2.8.3 RNA concentration, purity and integrity check RNA concentration was measured using the Nanodrop® ND-1000 Spectrophotometer. The RNA suspension (2 μL) was loaded onto the Nanodrop® and RNA concentrations were measured using A260. The purity of samples indicated ® using A260/A280 ratios identified by Nanodrop . Only RNA samples with A260/A280 ratios of 2.05-2.15 were considered as having valid purity (ideal A260/A280 ratio for RNA is 2). To assess the integrity as well as the purity of the extracted RNA, 1 μL of each sample was separated on 2% agarose/tris-borate-EDTA gel electrophoresis for 30 minutes at 110 volts. Quality of the RNA within the sample was assessed by the presence of two distinct 28S rRNA and 18S rRNA bands with an intensity ratio of 2:1, respectively. RNA integrity was assessed by the absence of smearing or laddering in the 2% agarose gel sample wells, which usually indicates presence of degraded RNA or RNA with genomic DNA contamination.
2.9 MICROARRAY ANALYSIS OF DIFFERENTIAL GENE EXPRESSION
The HumanHT-12 v4 BeadChip® array (Illumina®; VIC, AU) was used to identify changes in gene expression patterns induced by VN alone, dimeric IGF- II:VN and trimeric IGF-I:IGFBP:VN complexes in MCF-7 breast cancer cells. The Illumina® HumanHT-12 v4 expression BeadChip® array technology provides
Chapter 2: Materials and Methods 38
genome-wide transcriptional coverage of well-characterised genes, including both coding and non-coding genes. Each array on the HumanHT-12 v4 Expression BeadChip contains more than 47,000 probes derived from the National Centre for Biotechnology Information Reference Sequence (NCBI) RefSeq Release 38 (November 7, 2009). Sequence clusters are created from UniGene and are refined by analyses and compared with a number of publicly available human genome databases.
Each Illumina® BeadChip array holds tens of thousands of different oligonucleotide probes synthesised in situ onto microscopic beads held (covalently attached) in microwells on the surface of an array substrate improving robustness and measurement precision. The Illumina® BeadChip array-based expression profiling uses biotin-labeled aRNA (also called complementary RNA, or cRNA) as the target that has been optimised for high signal intensity on BeadChip arrays, enabling sensitive gene detection and reproducible differential expression results. The labelling of aRNA was produced using TargetAmp™-Nano Labeling Kit (Epicentre Biotechnologies; WI, USA). Labelled probe-targets presented on each bead are further used for a hybridisation-based procedure to map the array, identifying the location of each bead. In addition, probe-target hybridisation is quantified by detection of fluorophore-labelled target signals to determine the relative abundance of nucleic acid sequences in the target bead. Total strength of the signal, from a bead, depends upon the amount of target sample binding to the probes present on that bead. The intensity of a gene probe-target is compared to the intensity of the same gene probe-target under a different condition (control), and the identity of the feature is known by its position. Additional feature of the Illumina® HumanHT-12 v4.0 array includes hybridisation-based quality control of each array feature, allowing consistent production of high-quality, reproducible arrays. For full details see www.illumina.com.
2.9.1 The TargetAmp™-Nano single round Biotin-aRNAbiotin amplification kit – overview The in vitro transcription reaction produces biotin-labelled anti-sense (aRNA) by incorporating biotin-UTP into the RNA transcripts to map the array using a hybridisation-based procedure. All steps from cDNA synthesis through to the fragmentation of the biotin-labelled aRNA were performed by Australian Genome
Chapter 2: Materials and Methods 39
Research Facility (AGRF) (Melbourne, VIC, Australia) using the TargetAmp™- Nano Labeling kit for Illumina® Expression BeadChip (Epicentre Biotechnologies). This method is based on the RNA amplification protocol developed by James Eberwine (Van Gelder et al., 1990). The procedure consists of reverse transcription (RT) with an oligo(dT) primer bearing a T7 promoter using a SuperScript® III reverse transcriptase (Life Technologies). The cDNA then undergoes 2nd strand synthesis and clean-up to become a template for in vitro transcription in a reaction containing biotin-modified UTP and T7 RNA polymerase. To maximise the yields of biotin-labelled aRNA, this kit uses an optimised mixture of biotin-labelled and unlabelled nucleotide tri-phosphates (NTPs) and the T7 transcription buffers to generate hundreds to thousands of aRNA copies from each existing copy of mRNA in a sample. The biotin-labelled aRNA is then purified for hybridisation to Illumina® BeadChip® arrays (VIC, Australia) (Figure 2.1).
Chapter 2: Materials and Methods 40
Figure 2.1. TargetAmp™-Nano single round Biotin-aRNA Amplification kit overview for the Illumina® System Procedure. Adopted from epibio.com.
Chapter 2: Materials and Methods 41
2.9.2 In vitro transcription to synthesise biotin-labelled aRNA To prepare single stranded cDNA, one microliter of oligo(dT) primer containing a phage T7 RNA polymerase promoter sequence at its 5′ end was added to each tube containing 1 µg of total RNA sample diluted in a maximum volume of 9 μL and incubated at 65 °C for 5 minutes. The reaction was cooled on ice for one minute and reverse transcription was performed by incubation at 50 °C for 30 minutes with each reaction containing 0.25 μL of SuperScript® in a final volume of 2 μL of the reverse transcription master mix (1.5 μL 1st strand cDNA PreMix + 0.25 μL DTT). Single stranded RNA was then converted to double-stranded cDNA containing a T7 transcription promoter in an orientation that will generate aRNA during the subsequent in vitro transcription reaction without the need for purification. Double-stranded DNA was synthesised by adding 5 μl of the 2nd strand cDNA master mix, containing 2nd strand cDNA PreMix (4.5 μl) and 2nd strand cDNA polymerase (0.5 μl) to each reaction. Reactions were gently mixed and incubated at 65 °C for 10 minutes followed by incubation at 80 °C for 3 minutes. Reactions were performed at elevated temperatures (50-80 °C) to reduce the formation of RNA secondary structure. The samples were placed on ice before proceeding to in vitro transcription procedure. The in vitro transcription reaction was performed using in vitro transcription master mix to produce biotin-aRNA by incorporating biotin-UTP into the newly synthesised aRNA. The master mix was prepared in a final volume of 20 μl containing biotin-UTP (3 µl), NTP (10 µl), T7 transcription buffer (2 µl), T7 RNA polymerase (2 µl) and DTT (3 µl) and added to each cDNA sample. The tubes were mixed gently by pipetting, spun down briefly and then incubated at 42 °C for 4 hours to achieve the optimal yield and quality (length) of biotin-labelled aRNA.
2.9.3 Biotin-labelled aRNA purification and quantification Biotin-labelled aRNA was purified using the RNeasy® miniprep Kit (Qiagen; Doncaster, VIC, Australia) following the manufacturer’s protocol. To purify aRNA, 350 μL of RNA binding buffer was added to each sample and mixed briefly before adding 250 μL of 100% ethanol and pipette mixing. Each sample was then added to the purification columns and centrifuged at 10,000g for 15 seconds. The flow- through was discarded and the columns were then washed with 650 μL of wash buffer, centrifuged again and the flow-through discarded. An additional centrifugation at full speed for 5 minutes was performed to remove any residual wash
Chapter 2: Materials and Methods 42
buffer or solvents and the filters were transferred to clean elution tubes. Finally, RNA was eluted using 50 μL RNase-free MilliQ water and stored at -80 °C until further use. The concentrations of the biotin-aRNA samples were determined using a NanoDrop® ND-1000 UV-Vis Spectrophotometer (NanoDrop Technologies). The quality of the samples was further assessed by running 2 μL of purified aRNA on a 2% agarose/tris-EDTA gel and electrophoresed for 30 minutes at 110 volts. The samples were then separated into the smaller aliquots to avoid freeze thawing of aRNA and stored at -80 °C.
2.9.4 BeadChip® Direct Hybridisation assay Tubes containing aRNA samples were incubated at 65 °C for 5 minutes, vortexed and allowed to cool at room temperature. For each sample 750 ng of aRNA was suspended in a total of volume of 15 µL of Gene Expression Hybridisation Buffer (GEX-HYB) (100 mM MES, 1 M NaCl, 20 mM EDTA and 0.01% Tween- 20) and loaded into a single array on the Illumina® HumanHT-12.v4.0 BeadChip®. The BeadChips® were then sealed in the hybridisation chamber, placed on the rocking platform and incubated at 58 °C for 16 hours in the Illumina® hybridization oven. Following hybridisation, the BeadChips® were removed from the chambers and placed into the staining dish containing 250 mL of the wash E1BC solution. The chips were then placed into the Hybex® water bath insert containing Illumina® high- temperature wash buffer and incubated at 55 °C for 10 minutes. Chips were immediately washed at room temperature with fresh wash E1BC buffer for 10 minutes and 100% ethanol for 2 minutes prior to scanning.
2.9.5 BeadChip® array scanning The hybridised arrays were treated with fluorescently labelled Streptavidin, a tetrameric protein with high binding affinity to biotin and the differential detection of florescence signals was performed using the Illumina® iScan reader. Briefly, Streptavidin was purified from the bacterium Streptomyces avidinii using Affinity chromatography and then conjugated with Cy3 fluorescent dye (Streptavidin-Cy3). Following washing steps, BeadChips® were transferred into Cy3-Streptavidin solution, incubated at room temperature for 10 minutes, and dried prior to scanning. The iScan reader uses a laser to excite the fluor of the hybridised single-stranded product on the beads of the BeadChip® sections. The scanner operating software, GenomeStudio®, then converts the signal on the array into a ‘txt’ file for analysis.
Chapter 2: Materials and Methods 43
2.9.6 Samples and system Quality Control (QC) The quality and quantity of total RNA samples was ascertained prior to microarray analysis on the Agilent Bioanalyser 2100 using the NanoChip protocol. The Illumina® Whole-Genome Gene Expression Direct Hybridisation assay control system covers every aspect of the array experiment, from the biological specimen, sample integrity to sample labelling, hybridisation, and signal generation. The intactness of the biological specimen was monitored by expression of housekeeping gene controls. These controls consist of two probes per housekeeping gene that should be expressed in each sample. Sample labelling was also controlled using the Ambion® External RNA Controls Consortium (ERCC) RNA Spike-In Control Mixes (Life Technologies) pre-formulated with sets of 92 polyadenylated transcripts form ERCC plasmid reference library (for details, please see Ambion’s ERCC RNA Spike-In Control Mixes protocol). This kit uses control biotin-tagged oligonucleotides to assess the success of the labelling process. Secondary staining of these control biotin-tagged oligonucleotides with a detected signal was considered as the positive labelling for target probes. Negative control probes with no corresponding targets in the genomes were also used to detect any signal from non- specific binding of dye or cross-hybridisation. To monitor the quality of the hybridisation assays, three oligonucleotides at low, medium, and high concentrations of Cy3 labelling with matches to the control oligonucleotides were also hybridised in each BeadChip array. An array with signal intensities from these concentrations of Cy3-labelling matching the control oligonucleotide signals were passed the hybridisation assay quality control (for details, please see Whole-Genome Gene Expression Direct Hybridisation assay manual). GenomeStudio® application automatically tracks the performance of these controls and generates a report for each array in a cell summary probe control profile report. The integrity of total RNA was checked before the processing and is summarised in Appendix 1. All RNA samples passed the performance parameters and were used in subsequent processes.
2.10 MICROARRAY DATA ANALYSIS
2.10.1 Data analysis and presentation using GeneSpring GX11 The data files containing individual raw array data (probe intensities) were imported to the GeneSpring GX 11 software and pre-processed using Robust Multi- array Average with GC-content background correction (GC-RMA). Data which has
Chapter 2: Materials and Methods 44
been normalised with GC-RMA was then further normalised using the ‘per gene normalisation’ step in which all the samples were normalised against the median of the control samples (serum-free medium; SFM). Therefore, the expression value for one gene across the different condition was centred on one, by dividing the expression value by the median expression value for that gene across the condition (treatments). This ensured that genes which did not change across conditions received a normalised expression value of one, allowing for easier visual detection of differentially expressed genes. Statistically significant differentially expressed gene probe setswere then identified from pair-wise comparisons using two-way ANOVA (parametric test, assuming unequal variances) with ‘Benjamini and Hochberg false discovery rate’ (BHFDR) as the multiple testing correction (p = 0.05). Additionally, a fold change threshold of 1.5 was applied to these statistically significant gene probe sets. Together, this generated a list of gene probe sets with expression either significantly increased or decreased by at least 1.5-fold. The selection criteria of ±1.5-fold was used to ensure critical genes which may not have a large magnitude of change were identified, while also keeping data sets to a manageable size.
2.10.2 R program to generate Heatmap of gene expression data The data containing normalised gene values were imported into R 3.0.2 language to generate and plot the heatmaps using the “gplots” version 2.12.1 package. The expression of each individual gene presented in the heatmap was compared across all the conditions and indicated using different colour keys. The genes and treatments were clustered using an algorithm based on similarity in expression using algorithms applying Ward's minimum variance method within the R package “gplots” and results were added to the heatmap as dendrograms (hierarchical tree). The Euclidean distance map demonstrating Euclidean distance between treatments was also generated using the Ward's method within the “gplots” package.
2.10.3 Gene ontology, functional network, and canonical pathway analyses using IPA Gene ontology, functional networks and canonical pathways analyses were undertaken using the Ingenuity Pathway Analysis (IPA) tool (Ingenuity® Systems, www.ingenuity.com). The program is a web-based application that enables the discovery, visualisation, and exploration of molecular interaction networks in gene expression or proteomic data sets. Data sets containing gene identifiers and
Chapter 2: Materials and Methods 45
corresponding expression values of gene prob sets determined to be significantly expressed by at least ±1.5-fold in response to VN alone, the dimeric IGF-II:VN and trimeric IGF-I:IGFBP:VN complexes compared to SFM (control) were uploaded into the application. Each gene identifier was mapped to its corresponding gene object in the Ingenuity Pathways Knowledge Base. The structure of network knowledge base approach to analyse genome-wide transcriptional data in the context of known functional interrelationships among proteins, small molecules and phenotype has been described in details (Calvano et al., 2005). Functions of genes and their interaction with associated networks and pathways are mined from the peer-reviewed literature and encoded into ontology. Moreover, biological functions are assigned to whole data sets by using IPA knowledge base as a reference set and a propriety ontology representing over 300,000 classes of biological objects and consisting of millions of individual modelled relationships between proteins, genes, complexes, cells, tissues, small molecules and diseases. Genes which map to the IPA knowledge base or so-called “focus genes” are then used as the starting point for generating biological networks. Networks of focus genes were then algorithmically generated based on their connectivity. IPA computes a score for each network according to the fit of the original set of significant genes. This score reflects the negative algorithm of p-value that indicates the likelihood of the focus genes in a network being found together due to random chance. A score of 3 indicates a 1 in 1000 chance that the focus genes are together in a network by random chance. Therefore, scores of 3 or higher have a 99.9% chance of not being generated by random chance alone. This score was used as the cut-off for identifying gene networks significantly affected by VN alone, IGF- II:VN and IGF-I:IGFBP:VN complexes. Networks are ranked and a graphical representation of the molecular relationships between genes/gene products is generated. The functional analysis also identified biological functions and/or diseases that were most significant to the data set. Furthermore, Fisher’s exact test score below 0.05 was used to calculate the p-value to determining the probability that each biological function and/or disease assigned to that data set is due to chance alone. The same statistical approach was used to determine the Canonical pathways that were most significant to the data set.
Chapter 2: Materials and Methods 46
2.11 QRT-PCR ANALYSES OF DIFFERENTIAL GENE EXPRESSION
2.11.1 Primer design The forward and reverse primers for qRT-PCR analyses of a number of genes (Table 2.1) were designed using NCBI Primer-BLAST. The parameters used to generate acceptable primer sets included: melting temperature of 57 °C, maximum of
63 °C and optimal melting temperature (Tm) of 60 °C, with no more than 3 °C difference in Tm between forward and reverse primers; 55-70% GC content; primer length between 20-23 nucleotides and amplicon length between 100-200 nucleotides (Table 2.2). Refer to Appendix 2 for full detailed sequence information on all primers used.
Chapter 2: Materials and Methods 47
Table 2.1. Target genes for qRT-PCR analysis.
Gene Full name Symbol
JUN Cellular oncogene JUN or c-JUN FOS Cellular oncogene FOS or c-FOS CD24 Cluster of differentiation 24 TGFβ2 Transforming growth factor beta 2 VAV2 vav2 guanine nucleotide exchange factor LAMB1 Laminin beta 1 CTGF Connective tissue growth factor GABRP Gamma-aminobutyric acid receptor subunit pi NBPF20 Neuroblastoma breakpoint family, member 20 FZD8 Frizzled class receptor 8 RASGRP1 RAS guanyl releasing protein 1 ZIC Zinc finger of the cerebellum HSPB8 Heat shock protein beta Cyr61 Cysteine-rich angiogenic inducer 61 ANXA2 Annexin A2 NFIB Nuclear factor 1 B-type H19 Imprinted maternally expressed transcript H19 SPDEF SAM pointed domain-containing ETS transcription factor MTSS1 Metastasis suppressor 1 XBP1 X-box binding protein 1 PAK4 p21 protein (Cdc42/Rac)-activated kinase 4
Chapter 2: Materials and Methods 48
Table 2.2. Details of primers designed using NCBI Primer-BLAST.
Primes Gene RefSeq ID Sequence (5’3’) Start Product Symbol (NCBI) size F: TTGCCAGAGCCCTGTTGCGG 967 JUN NM_002228.3 125 R: TCGGACGGGAGGAACGAGGC 1108
F: CCTGTCAACGCGCAGGACTTCT 332 FOS NM_005252.3 174 R: GGCGGGGACTCCGAAAGGGT 505
F: ACAGCACCAGGGACTTGCTCCA 1583 TGFβ2 NM_003238.3 147 R: TGGGCGGGATGGCATTTTCGG 1729
F: CCGGTCCCCAGTGTTCACGC 2353 VAV2 NM_003371.3 167 R: TCCGTCCGTTGGTCTCGCCC 2519
F: GTGAGCCTCGTGCTGGACGG 345 CTGF NM_005080.3 168 R: GGGAGCACCATCTTTGGCGGT 512
F: GCTGGGAGCCTGTCAGTTTATGA 1730 HSBP8 NM_014365.2 179 R: ACACACGTCCCGCCCATATACA 1908
F: CCCGGAGAGTTTCCCGCTTGG 82 ANXA2 NM_001002857 179 R: TGTTCAAAGCATCCCGCTCAGCA 260
F: GACAGCGCTTGGGGATGGATGC 449 XBP1 NM_005080.3 125 R: CTGCTGCAGAGGTGCACGTAGT 573
F: GCACTGCTCCTACCCACGCA 159 CD24A NM_013230.2 146 R: CCCAGGTACCAGGACTGGCTGG 304
F: CCTCCACCTTCACCGTTGCCA 1616 NFIB NM_005596.3 179 R: CCCAGGTACCAGGACTGGCTGG 1794
F: AAGACGCCAGGTCCGGTGGA 1365 H19 NR_002196.1 160 R: TGTGCCGACTCCGTGGAGGA 1543
F: TCTGACCCCAGCATTTCCAGAGCA 1692 SPDEF NM_012391.1 185 R: ATTATCCATTCCCGGGGGCACT 1876
F: GAACCTATGTTCAGGTTGCGGGT 2720 GABRP NM_014211.1 128 R: AACTGGAAGCCTTTACCCTGGCA 2847
F: ACAGTCCAGCAAGGGCCAAAGA 4554 RASGRP1 NM_005739.2 129 R: ACTAGGCTTGATAAGCCAAAGCC 4683
F: ATGCTACCTGGGTTTCCAGGGCA 1354 Cyr61 NM_001554.3 190 R: TTAGTGTCCATCCGCACCAGCA 1543
F: AGGCTCAACGGCGTGCTGAT 2661 NBPF20 NM_0010376 164 R: AGGGCGAAGCTGATGTGCTGT 2824
F: AACCTGCACAAGGTGTCGCCAT 1844 PAK4 NM_005884.3 113 R: TTGGCCAGGAATGGGTGCTTCA 1956
F: AGCAAGCAAAAGTGCAACAGATG 5073 LAMB1 NM_002291.1 200 R: TGGGACGCGTTGAACAAGGTTT 5272
F: GGGCACATTTCTGACCGAACCCT 4203 MTSS1 NM_014751.4 200 R: CCTTGAGCAATCGCAGCAGTGT 4402
F: AGTGCTTGCTAAGGACGCATGG 2870 FZD8 NM_031866.1 177 R: ACCGTCTTTGGGGAAGGTGCAAA 3046
Chapter 2: Materials and Methods 49
2.11.2 Reverse Transcription (RT) reaction to generate cDNA for qRT-PCR First strand DNA synthesis was performed using the SuperScript® III First- Strand Synthesis SuperMix for qRT-PCR kit (Life Technologies) to reverse transcribe total RNA. One microgram of total RNA was added into 0.5 mL low binding tubes containing 10 μL 2X RT Reaction Mix (containing dNTPs, random hexamers and reaction buffer), 2 μL RT enzyme mix [Moloney Murine Leukemia Virus (M-MLV) RT] and made up to 20 μL with RNase-free MilliQ water. Tubes were then incubated at 25 °C for 10 minutes, 55 °C for 30 minutes followed by 5 minutes incubation at 85 °C to terminate the reaction. Each reaction tube was left to chill on ice for 5 minutes before adding 1 μL (2 U) of E.coli RNase H and incubating at 37 °C for 20 minutes. The resulting cDNA was either stored at -20 °C or diluted 1:10 with nuclease-free water to obtain a concentration of 5 ng/μL cDNA for qRT- PCR analysis
2.11.3 q-RT-PCR qRT-PCR was performed to measure relative fold expression levels of target genes in cells treated with VN alone, IGF-II:VN and IGF-I:IGFBP-5:VN compared to the cells in control treatment (SFM). All reactions were performed in triplicate in 20 μL reaction volumes in a 96-well format using SYBR green (Life Technologies) chemistry and ABI Prism 7500 Sequence Detection System (Life Technologies). Reaction contained 1X SYBR-green PCR mix, 0.4 μM forward and reverse primers each and 2 μL of the cDNA dilutions (10 ng). PCR amplification followed a two-step protocol with an initial 10 minute denaturation step at 95 °C followed by 40 cycles of 95 °C for 15 seconds and 60 °C for 1 minute. In addition to target genes, Glyceraldehyde 3-phosphate dehydrogenase (GAPDH) was used as an internal control for normalisation purposes. The cycle threshold (Ct) of the gene of interest for each reaction was obtained and directly normalised to GAPDH expression levels
(ΔCt) and then compared with the control (ΔΔCt). The relative fold-change in expression of the gene of interest between the two samples was then calculated using the 2^(-ΔΔCt) method. To confirm the specificity of the primers and the expected product size the amplified PCR products were separated on a 2% agarose gel at 110 V for 30 minutes and visualised using SYBR safe (Life Technologies) fluorescence probe under UV illumination (Appendix 3).
Chapter 2: Materials and Methods 50
2.12 PROTEIN SEPARATION AND IMMUNOBLOTTING
2.12.1 Protein isolation from cell lines maintained in cell culture Following desired incubation of cells, the growth medium including unattached cells was removed and the attached cells were washed with PBS to remove traces of spent medium that could interfere with any downstream applications. The cells were directly lysed by adding 300 µL of Radio Immuno-Precipitation Assay (RIPA) buffer (150 mM NaCl, 1.0% Igepal CA-630, 0.5% sodium deoxycholate, 0.1% SDS, 50 mM Tris, pH 8.0) (Sigma-Aldrich). In order to make the complete protein lysis buffer phosphatase (Thermo Fisher Scientific) (1:10) and protease (Roche Diagnostics) (1:10) inhibitors (inhibiting serine, cysteine, threonine, tyrosine and aspartic acid proteases activity) were also added to RIPA buffer. The plates were incubated on ice for 5 minutes and the cells were dislodged into suspension using a rubber policeman. The lysed cell suspension was then mixed with pellet collected from unattached cells. Lysates were then homogenised by passing through a 26- gauge needle at least five times before being centrifuged at 14,000g for 20 minutes at 4 °C. The supernatants were collected and their protein concentrations determined using Coomassie Plus® (Thermo Fisher Scientific; VIC, Australia). Samples were transferred to microfuge tubes, and stored at -80 °C until further use.
2.13 IMMUNOBLOTTING
Proteins were resolved on pre-made NuPAGE® gradient SDS-PAGE gels (4%- 12%) (Life Technologies) and transferred to BioTrace™ NT nitrocellulose transfer membranes (Pall Corporation; VIC, Australia) at 200 mA in transfer buffer (25mM Tris, 40 mM glycine and 10% methanol) for 2 hours at 4 °C. Membranes were then blocked with 5% non-fat skim milk powder in Tris-buffered saline (10 mM Tris, 0.5 M NaCl; pH 7.6) with 0.1% Tween-20 (TBS-T) for 1 hour at room temperature to avoid non-specific interaction of primary antibody with the membranes. The blocked membranes were then incubated with primary antibodies, namely, polyclonal anti- Cyr61 (1:1000) and anti-CTGF (1:1000) (R&D Systems; MN, USA) overnight at 4 ºC in 5% non-fat skim milk powder in TBS-T or 5% BSA in TBS-T. The membranes were then washed in TBS-T at least six times for 5 minutes each and incubated with horseradish peroxidise (HRP)-conjugated secondary IgG antibody in TBS-T+5% non-fat skim milk powder (1:5000) for 1 hour at room temperature. Following a
Chapter 2: Materials and Methods 51
further six washes of 5 minutes each, the protein bands were visualised using enhanced chemiluminescence (ECL) as per the manufacturer’s protocol (GE Healthcare; Melbourne, VIC, Australia). The membranes were either exposed onto x- ray films (Fujifilm; QLD, Australia) and developed using a film processor (AGFA CP 1000, Superior Radiographics; Madison, WI, USA) or were imaged directly using the ChemiDoc™ XRS system (BioRad; CA, USA). The same membranes were then stripped using restore western blot stripping buffer (Pierce) and the total levels of GAPDH were detected as outlined above using anti-GAPDH antibodies, respectively to validate equal loading of proteins onto the gel.
2.14 IMMUNOFLUORESCENCE STAINING
To visualise VN- and IGF-I:IGFBP:VN complex-induced expression of Cyr61 and CTGF target proteins, MCF-7 cells grown in Nunc™ 96-well optical-bottom plates (Thermo Fisher Scientific; VIC, Australia) pre-bound with VN and IGF- I+IGFBP-5+VN were stained using fluorescently labelled dyes and antibodies. All reactions were performed at room temperature. Cells were fixed after 48 hours with 4% formaldehyde in phosphate buffered saline (PBS; Life Technologies) for 30 minutes and then carefully washed twice with 400 μL PBS:Glycine (0.1M Glycine in PBS, pH 7.4). Cells were then permeabilised with 200 μL of 0.4% Triton-X in PBS for 10 minutes followed by one hour incubation with 200 μL of 5% BSA fraction-V (Merck-Calbiochem; VIC, Australia) to block non-specific binding. Cells were washed with 0.1% PBS-Tween-20 (PBS-T) and incubated for one hour with 100 µL primary mouse anti-Cyr61 and anti-CTGF antibodies (Santa Cruz Biotechnology; CA, USA) diluted in 5% BSA + PBS-T (1:100). Subsequently, cells were washed twice with PBS-T and incubated with 100 μL fluorophore-conjugated secondary AlexaFluor® 594 (red) or 488 (green) goat anti-mouse IgG antibodies diluted in PBS- T (1:250) and incubated for one hour. Cells were then washed three times with 0.1% PBS-T and incubated for 40 minutes in the dark with 200 μL fluorescently conjugated AlexaFluor® 594 (red) or 488 (green) phalloidin (1:40, in PBS) and DAPI (1:500) dyes (Life Technologies). The plates were stored at 4 °C and in the dark until imaging by DeltaVision microscopy.
Chapter 2: Materials and Methods 52
2.15 CLONING
2.15.1 Eukaryotic expression vectors pLKO.1 lentiviral vectors with a gene specific short hairpin RNA (shRNA) already cloned into it (for genes Cyr61 and CTGF) were obtained from Open Biosystems (Thermo Fisher Scientific). pLKO.1 vector with a shRNA for firefly luciferase gene cloned into it (pLKO.shFF3), pLKO.1 vector containing cDNA for GFP (pLKO.3G) and the lentiviral packaging vectors pCMV-VSV-G (encoding the Vesicular stomatitis virus G-protein capsule), pCMV delta R8.2 (encoding the HIV-1 GAG/POL, Tat and Rev elements) were from Addgene.org. For vector diagrams of pCMV-delta-R8.2 and pCMV-VSV-G refer to Appendix 4 and Appendix 5.
2.16 PREPARING COMPETENT STBL3 E.COLI CELLS (CALCIUM CHLORIDE METHOD)
One Shot® Stbl3 competent cells (Life Technologies) were used to maintain lentiviral vectors as these cells reduce the frequency of unwanted homologous recombination of long terminal repeats (LTRs) found in lentiviral and other retroviral vectors. Owing to the large costs associated with purchasing these cells, contents from one 50 μL vial of the One Shot® Stbl3 competent cells (Life Technologies) were expanded to produce a working stock following methods outlined below. The contents from one vial of the cells (50 μL) were transferred to 100 mL antibiotic-free Lysogeny Broth (LB) medium. The cells were allowed to grow for 6 hours or until their optical density reached 0.4 to 0.6 (measured at 600 nm). The cell culture was centrifuged at 2,500g for 15 minutes at 4 °C and the supernatant was discarded. The cell pellet was resuspended, without vortexing or pipetting, in 5 mL of sterile 0.1M
CaCl2 and incubated at 4 °C for 2 hours to obtain chemically competent Stbl3 cells. The cells were then mixed with 1.5 mL of sterile 80% glycerol, which was subsequently aliquoted and stored at -80 °C until further use.
2.16.1 Transformation of competent Stbl3 E.coli cells One 50 μL aliquot of competent Stbl3 cells was thawed on ice for each transformation. Upon thawing, ~50 ng of each vector DNA was added to the competent cells, mixed gently, and incubated on ice for 30 minutes. Cells were then heat shocked for 45 seconds at 42 °C and then returned to ice for 5 minutes. Two hundred and fifty microlitres of antibiotic-free LB medium (5 g/L bactotryptone, 2.5
Chapter 2: Materials and Methods 53
g/L yeast extract and 2.5 g/L NaCl) were added to each tube and the cells were allowed to recover at 37 °C for 1 hour with constant shaking at 225 rpm. For all vectors, 50 μL was then spread onto pre-warmed LB agar plates containing 100 ng/mL ampicillin and incubated overnight at 37 °C. Up to 4 colonies were picked from each plate using a heat-sterilised inoculation loop, which was then placed into 5 mL of LB medium containing 100 μg/mL ampicillin and incubated overnight at 37 °C with constant shaking at 225 rpm. The following day, glycerol stocks were made by mixing 800 μL of each overnight culture with 200 μL of sterile 80% glycerol and stored at -80 °C. The remaining volume of the culture was then used to extract DNA using plasmid purification miniprep procedures detailed in section 2.14.4. For cloning that started directly from the bacterial glycerol stocks (for all pre-purchased vectors obtained from Open Biosystems) the inoculation loop was dipped into a partially thawed vial of the glycerol stock which was then streaked onto LB agar plates containing 100 μg/mL ampicillin and was then incubated overnight at 37 °C. The following day, the colony was subsequently transferred into 5 mL LB medium containing 100 μg/mL ampicillin and DNA extracted using plasmid miniprep procedures detailed below.
2.16.2 Plasmid purification – Miniprep Plasmid DNA was purified from the Stbl3 bacterial cells using the QIAprep miniprep kit (Qiagen; VIC, AU) which contained all the solutions and spin columns required for plasmid purification. All steps were performed at room temperature following the manufacturer’s instructions. Briefly, 5 mL overnight cultures were centrifuged at 3000 rpm for 5 minutes to pellet the cells. The pelleted cells were lysed using 250 μL Buffer P1. Subsequently, two hundred and fifty microlitres of buffer P2 were added into each tube and mixed several times by inversion. After no longer than 5 minutes, 350 μL of buffer N3 was added and again mixed by inversion. The tubes were then centrifuged for 10 minutes at 13,000 rpm (~18,000g) in a bench- top centrifuge and the resulting supernatants were applied to the QIAprep spin columns which bind the plasmid DNA. The spin columns were then centrifuged for 1 minute (13,000 rpm) and the flow-through was discarded. The QIAprep columns were washed with 500 μL Buffer PB, followed by a subsequent wash with 750 μL Buffer PE. To elute the DNA, 30 μL of elution buffer (10 mM Tris-HCl, pH 8.5) was added to the centre of each column and incubated for 1 minute. The eluted DNA was
Chapter 2: Materials and Methods 54
then collected by centrifugation for 1 minute. The quality and quantity of the extracted plasmid DNA was then assessed by agarose gel electrophoresis and by measuring the UV spectrophotometric absorbance at 260 and 280 nm, respectively.
2.17 LENTIVIRAL VECTOR-MEDIATED GENE MANIPULATION
To assess the functional roles of the candidate genes, namely cysteine-rich, angiogenic inducer 61 (Cyr61) and connective tissue growth factor (CTGF), lentiviral-mediated transduction of breast cells was performed to stably express the gene specific short hairpin RNA (shRNA) sequences against Cyr61 and CTGF to suppress their gene expression (gene knockdown).
2.17.1 shRNA lentiviral vector: production and design pLKO.1 was the lentiviral vector of choice to perform shRNA mediated knockdown of Cyr61 and CTGF mRNA and protein expression. shRNA sequences already cloned into pLKO.1 vectors were purchased from Open Biosystems. For each gene five sets of shRNAs were obtained resulting in five pLKO-shRNA vectors for each gene. The shRNA resulting in the maximal down-regulation of its respective gene was chosen for further knockdown studies. The pLKO.1 vector containing the shRNA targeting the firefly luciferase gene (pLKO.shFF3), which is absent in mammalian cells, was used as a non-targeting negative control and the pLKO.1 vector with GFP cDNA inserted (pLKO.3G) was used as a positive control for transfection to visually inspect the transfection efficiency under a fluorescence microscope. Nucleotide sequences of shRNAs targeting each gene are detailed in Table 2.3. For pLKO.1 vector map refer to Figure 2.2. Vectors obtained from Open Biosystems came in glycerol stocks of DH5α cells. To isolate the vector DNA the bacterial cells were streaked onto LB-agar ampicillin plates, incubated overnight, and single colonies were isolated and transferred to liquid culture following protocols outlined in the latter half of section 2.16.1. The vector DNA was isolated using the miniprep procedures described in section 2.16.2.
Chapter 2: Materials and Methods 55
Table 2.3. shRNA sequence information. shRNA sets cloned into pLKO.1 lentiviral vectors for genes Cyr61 and CTGF were purchased from Open Biosystems. The sequence for each shRNA (sense-loop-antisense) within the pLKO.1 vector is depicted in this table.
Description Hairpin Sequence
Cyr61-1 5'CCGGGCAAACAGAAATCAGGTGTTTCTCGAGAAACACCTGATTTCTGTTTGCTTTTTG3'
Cyr61-2 5'CCGGCGCATCCTATACAACCCTTTACTCGAGTAAAGGGTTGTATAGGATGCGTTTTTG3'
Cyr61-3 5'CCGGCCGAACCAGTCAGGTTTACTTCTCGAGAAGTAAACCTGACTGGTTCGGTTTTTG3'
Cyr61-4 5'CCGGCCCTTCTACAGGCTGTTCAATCTCGAGATTGAACAGCCTGTAGAAGGGTTTTTG3'
Cyr61-5 5'CCGGCGAACCAGTCAGGTTTACTTACTCGAGTAAGTAAACCTGACTGGTTCGTTTTTG3'
CTGF-1 5'CCGGGAGAACATTAAGAAGGGCAAACTCGAGTTTGCCCTTCTTAATGTTCTCTTTTTG3
CTGF-2 5'CCGGCGCATCCTATACAACCCTTTACTCGAGTAAAGGGTTGTATAGGATGCGTTTTTG3'
CTGF-3 5'CCGGCATCTTTGAATCGCTGTACTACTCGAGTAGTACAGCGATTCAAAGATGTTTTTG3'
CTGF-4 5'CCGGGCATGAAGACATACCGAGCTACTCGAGTAGCTCGGTATGTCTTCATGCTTTTTG3'
CTGF-5 5'CCGGCCCAAGGACCAAACCGTGGTTCTCGAGAACCACGGTTTGGTCCTTGGGTTTTTG3'
Colour Codes: sense (target sequence)-loop-antisense
Chapter 2: Materials and Methods 56
Description Vector Element
Human U6 promoter drives RNA Polymerase III transcription for U6 generation of shRNA transcripts.
Central polypurine tract, cPPT, improves transduction efficiency by cPPT facilitating nuclear import of the vector’s preintegration complex in the transduced cells.
hPGK Human phosphoglycerate kinase promoter drives expression of puromycin.
Puromycin resistance gene for selection of pLKO.1 plasmid in mammalian Puro R cells.
sin 3’LTR 3’ Self-inactivating long terminal repeat.
f1 ori f1 bacterial origin of replication.
Ampicillin resistance gene for selection of pLKO.1 plasmid in bacterial Amp R cells
pUC ori pUC bacterial origin of replication.
5’LTR 5’ long terminal repeat.
RRE Rev response element.
Figure 2.2: Map of pLKO.1 containing a shRNA insert. Adapted from addegene.org
Chapter 2: Materials and Methods 57
2.17.2 Virus production in HEK293T packaging cells The pLKO.1 vector containing the shRNA of interest and the viral packaging plasmids, pCMV-VSV-G and pCMV delta R8.2, were transiently transfected into HEK293T packaging cells to produce lentiviral particles. The day before the transfection, 1.5 x 106 cells suspended in 10 mL DMEM+10% iFBS were seeded onto 10 cm dishes. The following day, a transfection mix was prepared in DNase/RNase free tubes containing 185 μL serum-free and antibiotics-free DMEM, 12 μL FuGENE-6 transfection reagent (Promega), 1.8 μg pCMV delta R8.2, 0.2 μg pCMV-VSV-G and 2 μg of the shRNA-containing pLKO.1 vector. This mix was incubated for 15 minutes at room temperature, following which the mixture was added drop wise to the pre-plated HEK293T cells (50-60% confluent at the time of transfection). These plates were incubated overnight at 37 °C, 5% CO2. The following day, the transfection mix-containing medium was replaced with 7 mL of fresh DMEM+10% iFBS medium or with medium used for the target cells and incubated overnight at 37 °C, 5% CO2. The viral supernatants were harvested at 48 and 72 hours post-transfection. These viral supernatants from both 48 and 72 hours were pooled, centrifuged at 1000 rpm for 5 minutes to remove any cellular debris, filtered using a 0.45 μm cut-off filter and stored at -80 °C until further use. Prior to harvesting, the viral supernatants from HEK293T cells transfected with the pLKO.3G control were visually inspected for green fluorescence under a fluorescence microscope to confirm DNA transfection.
2.17.3 Transfection of target cell lines Target cell lines used in this project for transduction of lentiviruses were MCF- 7 and MDA-MB-231. Target cells were seeded onto a 6-well plate at a density such that they were 30-50% confluent at the time of infection. Two hundred thousand cells for MCF-7 and 150,000 cells for MDA-MB-231 suspended in 2 mL target cell medium were seeded onto each well of a 6-well plate. The following day the spent medium was removed and 2 mL of the lentiviral particles-containing medium obtained in section 2.15.2 was added onto the target cells. Simultaneously, 6 μg/mL protamine sulphate was also added to the well to facilitate virus uptake by the cells.
After overnight incubation at 37 °C, 5% CO2, the viral supernatant was removed and the cells were allowed to recover overnight in the target cell medium. The following day, target cells containing the viral DNA inserted within their genome were selected
Chapter 2: Materials and Methods 58
using 2 μg/mL puromycin. This high concentration of puromycin was maintained until all the un-transfected negative control cells died, after which the concentration of puromycin was halved to 1 μg/mL, with transfected cells constantly being maintained in puromycin-containing medium. The cells were observed for any changes in morphology compared to the shFF3-containing control cells (shCntrl). RNA and protein was isolated following the protocols outlined in sections 2.8.2 and sections 2.12.1, respectively to analyse shRNA-mediated knockdown of the gene of interest through qRT-PCR and western immunoblotting. Figure 2.3 depicts the overall schema for lentiviral production and transduction of target cells.
Chapter 2: Materials and Methods 59
Packaging vectors
shRNA Delta VSV-G vector R8.2
+FuGENE 6
transfection
293T Packaging cells
Virus produced By 293T cells
Overnight infection
Target cells
Figure 2.3. Schematic depiction of lentiviral shRNA-mediated suppression of target cells. pLKO-shRNA vector along with delta R8.2 and VSV-G packaging vectors are transfected into HEK293T cells. These packaging cells secrete into the medium the lentiviral particles harbouring the shRNA. Incubation of the target cells with the viral supernatant results in the shRNA being expressed by the target cells.
Chapter 2: Materials and Methods 60
2.17.4 Statistical Analyses Data were pooled from multiple experiments (two being the minimum) with each treatment tested in three wells (two wells for Matrigel™ experiments) in each replicate experiment. The only exceptions were the studies screening for shRNA with maximal down-regulation of target gens as well as cell signalling and receptor studies where the data were pooled only from one biological replicate with each treatment tested in three wells. Data are expressed as percentage of the response observed in wells treated with SFM (control) or as fold change in gene expression observed in wells treated with SFM, unless otherwise stated. For functional assays and qRT-PCR analysis the data analyses was performed using two-tailed student’s t- test utilising InerSTAT-a version 1.3. For microarray analysis the data were analysed using two-way ANOVA (parametric test, assuming unequal variances) with ‘Benjamini and Hochberg false discovery rate’ (BHFDR) tests. Differences of p<0.05 (95% confidence interval) were considered to be statistically significant.
Chapter 2: Materials and Methods 61
Chapter 3: Identifying changes in gene expression profile associated with VN- and IGF:VN complex-stimulated breast cancer cell survival
Chapter 3: Identifying changes in gene expression profile associated with VN- and IGF:VN complex-stimulated breast cancer cell survival 63
3.1 INTRODUCTION
The metastatic spread of cancer cells to distant sites of the body and the subsequent establishment of secondary tumours in critical organs of the body is the major cause of death among breast cancer patients. A lack of a clear understanding of how cancer cells attain the ability to survive and promote metastasis within surrounding tissue microenvironments is one of the most challenging aspects in the treatment of patients with aggressive disease. Altered interactions between the malignant cells and extracellular matrix (ECM) proteins within the tumour microenvironment are considered to be a major driver of cancer progression (Lochter & Bissell, 1995; Muschler & Streuli, 2010; Quail & Joyce, 2013). These interactions have been reported to alter biological phenotypes manifested at most stages during cancer progression, from pre-malignancy to proliferative dysregulation, dissemination, survival in foreign tissues and subsequent metastasis formation.
As outlined in chapter one, the multi-functional ECM glycoprotein vitronectin (VN) plays a key role in the regulation of diverse cellular functions during normal and pathological conditions (Preissner & Reuning, 2011). In breast carcinoma, VN was identified to be present in the peritumoral stroma, blood vessel walls and in ECM surrounding cancer cells (Carriero et al., 1997; Aaboe et al., 2003; Kadowaki et al., 2011). VN has been shown to modulate cancer metastasis by providing traction and direction to migrating tumor cells during invasion (Aaboe et al., 2003), while also conferring a survival advantage to advancing tumour cells (Uhm et al., 1999; Leavesley et al., 2013). The effects of VN are predominantly mediated through its specific interactions with cell surface integrins and the urokinase plasminogen activator system (Felding-Habermann & Cheresh, 1993), both of which are frequently overexpressed in tumour cells (Andreasen et al., 1997; Mizejewski, 1999). Upon integrin ligation, VN triggers intracellular signalling cascades, controlling critical cellular events contributing to cancer progression (Hapke et al., 2003; Hurt et al., 2010).
In addition, VN ligation of integrins causes integrins to become clustered in the plane of the cell membrane, to facilitate the formation of focal adhesion contacts that link the cells to the ECM (Giancotti & Ruoslahti, 1999). Interestingly, present within these focal adhesion contacts are growth factors (GFs) (Taipale & Keski-Oja, 1997). Integrin clustering and its association with focal adhesions in turn facilitate the Chapter 3: Identifying changes in gene expression profile associated with VN- and IGF:VN complex-stimulated breast cancer cell survival 64
formation of integrin-GF complexes that regulate cell behaviour. In this regard, studies from the last two decades indicate that several GFs and cytokines interact with VN-binding integrins, and this integrin-GF cooperation regulates diverse biological processes (Eliceiri, 2001; Schwartz & Ginsberg, 2002; Beattie et al., 2010). In view of this, studies within our laboratory are only among the few that have investigated the effects of such GF-ECM interactions, in particular, interactions between VN and IGF ligands on cellular functions (Upton et al., 1999; Noble et al., 2003; Hollier et al., 2005; Hollier et al., 2008; Kricker et al., 2010; Kashyap et al., 2011). These studies have revealed that IGF ligands bind to VN forming dimeric IGF-II:VN and trimeric IGF-I:IGFBP:VN complexes (Upton et al., 1999; Kricker et al., 2003) stimulating breast cancer cell migration (Noble et al., 2003; Hollier et al., 2008) and inhibiting drug-induced apoptosis (Kashyap et al., 2011). Moreover, microarray analysis of non-tumorigenic MCF10A breast cells that migrated in response to the trimeric IGF-I:IGFBP-5:VN complex revealed genes specific to cell movement to be uniquely deregulated in cells that migrated in response to the complex. These studies provide mechanistic as well as molecular insight into the functional effects of the IGF:VN complexes and provide further evidence for the important role these novel complexes play in regulating biological functions relevant to cancer progression.
The emergence of global gene expression profiling using DNA microarrays has added valuable new information to our understanding of the molecular events underpinning ECM component-mediated biological functions. The studies to date have focused on identifying gene expression changes across a variety of cell types, largely in response to the ECM proteins FN and collagen (Thornton et al., 2002; Chen et al., 2004; Charoenpanich et al., 2011). However, there exists no global expression studies investigating the changes in gene expression associated with VN- stimulated breast cell survival and metastasis, especially since we and others have shown that VN is able to stimulate cell survival in variety of cell types (Uhm et al., 1999; Kashyap et al., 2011; Hazawa et al., 2012). Additionally, considering the important roles of the IGF in the transformation and progression of breast cancer and its association with VN, the molecular mechanisms underlying the IGF:VN-induced responses in breast cancer cells remains to be investigated.
Chapter 3: Identifying changes in gene expression profile associated with VN- and IGF:VN complex-stimulated breast cancer cell survival 65
Based on the evidence outlined above it was hypothesised that VN and the IGF:VN complexes would induce differential expression of genes critical to breast cell survival. Identification of these transcriptional programs will not only increase our understanding of the mechanisms responsible for VN- and IGF:VN-stimulated survival, but may also provide insights into the early molecular events promoting breast cell metastasis. To identify changes in gene expression patterns induced by VN and the IGF:VN complexes, DNA microarrays were performed on RNA samples isolated from cells which had survived in response to VN with or without IGF- I+IGFBP-5 and IGF-II pre-bound to polystyrene multi-well cell culture plates. This “pre-bound” strategy was adopted in an effort to more accurately mimic the in vivo cellular environment in which cells are not constantly exposed to growth factors in solution but rather interact with components of the ECM. Hence, the genes identified would more accurately represent the actual cellular and molecular processes involved in cell survival in comparison to the more common and relatively static pre-plated culture methodologies previously adopted by others.
In summary, this chapter builds on the previous studies performed within our laboratory and reveals for the first time the global transcription programs induced by VN and the IGF:VN complexes relevant to breast cancer cell survival. In particular, we have identified enriched biological networks and intracellular signalling pathways highly enriched in these transcriptional responses. These datasets not only shed new light on ECM-induced cancer cell survival, but will also guide the selection of candidate genes for functional validation studies in subsequent chapters.
3.2 EXPERIMENTAL METHODS
The following is a brief summary of the materials and procedures used to generate the data presented in this chapter. Full details of materials and methods used herein are described in Chapter 2.
3.2.1 Cell lines MCF-7 adenocarcinoma breast cancer cell line was the model of choice for the studies performed in this chapter. MCF-7 are epithelial breast cancer cells that express relatively high levels of the IGF-IR, display a number of VN-binding integrins and are poorly metastatic (Meyer et al., 1998; Bartucci et al., 2001), making them an ideal cell line to investigate the transcriptional changes in response to VN Chapter 3: Identifying changes in gene expression profile associated with VN- and IGF:VN complex-stimulated breast cancer cell survival 66
alone and the IGF:VN complexes. For details on maintenance of these cells please refer section 2.3.
3.2.2 Treatment strategy for in vitro viability assay In an effort to more accurately mimic the in vivo cellular environment this study adopted the strategy of “pre-binding” VN with or without growth factors to polystyrene multi-well cell culture plates prior to seeding cells. These “pre-bound” studies have been implemented to reflect the in vivo microenvironment in which these breast cancer cells normally encounter to maintain their survival. The protocol to pre-bind VN and IGFs to the tissue culture plastic and the full details of this method is outline in section 2.4.
3.2.3 Cell viability measurement To measure the relative viability, cells were incubated for 72 hours and cellular metabolic activity was assessed using the CellTitre 96® AQueous One Solution Cell Proliferation Assay (MTS). After 72 hours, 20 µL MTS reagent was added into each well and the absorbance recorded at 490 nm using a 96-well plate reader (Benchmark Plus, Bio-Rad). Furthermore, as a complementary approach Trypan blue exclusion assay was also employed to determine cell viability. MCF-7 cells were serum-starved for 4 hours before seeding 5 x 104 cells suspended in 2 mL SFM + 0.05% BSA into 6-well plates pre-coated with proteins treatments outlined above (section 3.2.2). Treatments with only SFM served as negative controls. After 72 hours incubation attached cells were harvested, centrifuged at 1000 rpm for 5 minutes and the resulting pellet was resuspended in SFM. Trypan blue 4% solution was then mixed with cell suspension by preparing a 1:10 dilution and the number of viable cells was assessed using a hemocytometer.
3.2.4 Optimisation of MCF-7 cell density for seeding onto 6-well plates To optimise seeding density for time course microarray experiments, MCF-7 cells were seeded onto 6-well plates pre-bound with either VN alone or the IGF:VN complex over a period of 48 hours. Three different cell densities including 2.5 x 105, 5 x 105 and 7.5 x 105 cells per well were tested for confluency using phase contrast microscopy (Eclipse TE 2000-U, Nikon).
Chapter 3: Identifying changes in gene expression profile associated with VN- and IGF:VN complex-stimulated breast cancer cell survival 67
3.2.5 Microarray analysis of differential gene expression Experimental set up
MCF-7 cells were spilt 1:1 24 hours prior to performing the experiment. Cells were serum-starved for 4 hours and a cell suspension containing 5 x 105 cells in 2.5 mL SFM + 0.05% BSA was seeded onto each well of the 6-well plates pre-bound with either VN only, IGF-II:VN or IGF-I:IGFBP-5:VN treatments. Cells seeded in wells containing only SFM served as the control treatment. Plates were incubated for
12 and 48 hours at 37 °C, 5% CO2. These time points were selected to account for changes in the gene expression patterns induced by VN and the IGF:VN complexes at both the early (12 hours) as well as the late (48 hours) stages. For each treatment and time point three wells were used for RNA and three wells for protein extraction according to protocols outlined in sections 2.8.1 and 2.12.1, respectively. MCF-7 cells attach weakly to the tissue culture plate in the absence of any adhesive ECM proteins. This was also the case for MCF-7 cells grown in untreated wells (control wells) in this study. Hence, to more fairly compare chenges in gene expression profiles induced in MCF-7 cells in response to VN alone or in combination with IGF-I+IGFBP or IGF-II tretaments compare to the control treatment (untreated cells) in all experiments reported throughout this study both attached and floating cells were collected for RNA and protein extraction.
Total RNA isolation and concentration, purity and integrity check
For full details on this method refer to section 2.8.2. RNA was extracted from cell lysates using the Direct-zol™ MiniPrep Kit (Zymo Research) following the manufacturer’s protocol and concentration, purity and integrity of samples were checked as outlined in section 2.8.3.
Microarray analysis
The Illumina® HumanHT-12 v4.0 BeadChip array technology (Australian Genome Research Facility; VIC, AU) used in this study to identify changes in gene expression patterns induced by VN alone, dimeric IGF-II:VN and trimeric IGF- I:IGFBP:VN complexes in MCF-7 breast cancer cells. This technology provides genome-wide transcriptional coverage of well-characterised genes, including both coding and non-coding genes. The majority of probes used in the array were designed to cover content form National Centre Biotechnology Information (NCBI)
Chapter 3: Identifying changes in gene expression profile associated with VN- and IGF:VN complex-stimulated breast cancer cell survival 68
Reference Sequence (RefSeq) release 38 (November 7, 2009). Sequence clusters are created from UniGene and are refined by analysis and comparison with a number of publicly available human genome databases. Oligonucleotide probes are synthesised in situ complementary to each corresponding sequence Additional feature of the Illumina® HumanHT-12 v4.0 array includes hybridisation-based quality control of each array feature, allowing consistent production of high-quality, reproducible arrays.
RNA samples were used for in vitro transcription to synthesise biotin-labelled anti-sense RNA (aRNA), aRNA purification and quantification using procedures outlined in sections 2.9.1 to 2.9.3. The biotin-labelled aRNA was further used for hybridisation to BeadChip® Arrays using direct hybridisation assay and then scanned. The scanner operating software, GenomeStudio®, converts the signal on the array into a TXT file for analysis. For full details of hybridisation assay and scanning procedures refer to sections 2.9.4 and 2.9.5. The Illumina® Whole-Genome Gene Expression Direct Hybridisation assay control system covers every aspect of the array experiment, from the biological specimen, sample integrity to sample labelling, hybridisation, and signal generation. For full details on samples and system Quality Control (QC) refer to section 2.9.6.
Microarray data analysis The data files containing individual raw array data (probe intensities) were imported to GeneSpring GX 11 and pre-processed using Robust Multi-array Average with GC-content background correction. This (GC-RMA) normalise data were then further normalised using the ‘per gene normalisation’ step in which all the samples were normalised against the median of the control samples (SFM being the control). Statistically significant differentially expressed genes were then identified using a fold change threshold of 1.5 and two-way ANOVA (parametric test, assuming unequal variances) with Benjamini and Hochberg false discovery rate (BHFDR) as the multiple testing correction (p<0.05). Together, this generated a list of gene probe sets whose expression was significantly increased or decreased by at least 1.5-fold in response to VN, IGF-I:IGFBP:VN and IGF-II:VN when compared to the SFM treatment. To get a visual representation of the global changes in the gene expression patterns induced by VN and the IGF:VN complexes, the data containing normalised gene values were imported into R 3.0.2 language to generate and plot the heatmaps,
Chapter 3: Identifying changes in gene expression profile associated with VN- and IGF:VN complex-stimulated breast cancer cell survival 69
using the “gplots” version 2.12.1 package. The genes and treatments were clustered using hierarchical cluster analysis based on ‘similarity in expression’ using algorithms applying Ward's minimum variance method within the R package “gplots” and results were added to the heatmap as dendrograms (hierarchical tree). Furthermore, Ingenuity Pathway Analysis tool (Ingenuity® Systems, www.ingenuity.com) was used to investigate gene ontology, canonical pathway, and network enrichment associated with differentially expressed genes in response to each treatment. For full details on microarray data analysis refer to section 2.10.
3.2.6 Confirmation of differential gene expression using quantitative Real-Time PCR (qRT-PCR) qRT-PCR was used to validate microarray data by measuring absolute expression levels of selected differentially expressed genes of interest. The total RNA used for qRT-PCR was from the original samples isolated and used for microarray analysis. For full details of primer design for target genes, Reverse transcription of RNA to produce cDNA and qRT-PCR analysis refer to section 2.11.
3.3 RESULTS
3.3.1 VN, dimeric IGF-II:VN and trimeric IGF-I:IGFBP:VN complexes increase cellular viability of MCF-7 breast cancer cells Previous studies have shown that VN and the IGF:VN complexes stimulate breast cancer cell survival in 2D in vitro assays (Kashyap et al., 2011). To confirm these results, the viability of MCF-7 breast cancer cells in response to the dimeric IGF-II:VN and trimeric IGF-I:IGFBP:VN complexes and VN alone was assessed using MTS and trypan blue viability assays (Figure 3.1). The responses of serum- starved cells were investigated over 72 hours after seeding the cells into culture plates pre-coated with either 1 μg/mL of VN alone or in combination with 30 ng/mL of IGF-I, 90 ng/mL of IGFBP-5 and 100 ng/mL of IGF-II.
The treatment of MCF-7 cells with VN alone, dimeric and trimeric complexes showed significant increases (p<0.01) in viability compared to SFM alone (control treatment) (Figure 3.1 A). The trimeric complex-stimulated cell viability was higher (p<0.05) than that recorded for VN only treatment. Similar findings were observed when the actual number of viable cells was counted using the trypan blue assay (Figure 3.1 B). All treatments showed a significant increase (p<0.01) in the number
Chapter 3: Identifying changes in gene expression profile associated with VN- and IGF:VN complex-stimulated breast cancer cell survival 70
of viable cells compared to the SFM control treatment. The number of viable cells in wells treated with the trimeric complex was also significantly (p<0.05) greater than the viable cell count from VN only treated wells. These results confirm those our laboratory has reported previously and validates our experimental set-up for subsequent RNA isolation and microarray analysis. Taken together, these data indicate that substrate-bound VN and the IGF:VN complexes are capable of stimulating MCF-7 breast cancer cell viability over 72 hours. These data also show that presence of IGF-I and IGFBP-5 in combination with VN (IGF-I:IGFBP-5:VN) is providing substantially greater viability responses than those obtained with VN. This data fits well with our previous findings showing that substrate-bound IGF-I:IGFBP- 5:VN complex stimulates breast cancer cell survival significantly greater than cells exposed to VN alone (Kashyap et al., 2011).
Chapter 3: Identifying changes in gene expression profile associated with VN- and IGF:VN complex-stimulated breast cancer cell survival 71
A MTS
B Trypan blue
Figure 3.1: Effect of VN, the IGF-II:VN and IGF-I:IGFBP:VN complexes on MCF-7 cellular viability.
Viability of MCF-7 cells seeded for 72 hours into 96-well (for MTS assay) and 6-well (for trypan blue assay) plates pre-coated with VN alone or dimeric IGF-II:VN and trimeric IGF-I:IGFBP:VN complexes (TRI) was assessed using MTS (A) and trypan blue (B) viability assays. The data are pooled from three experiments with treatments tested in at least six and two wells in each replicate experiment for MTS and trypan blue assays, respectively. The data for MTS assay are expressed as the average percent stimulation of control wells containing only SFM (A) and for trypan blue assay the data are presented as the number of viable cells number for each treatment (B). The asterisks indicate treatments which significantly increased viability compared to the respective control treatments (for TRI vs VN (*p<0.05) and for VN vs SFM, TRI vs SFM and IGF-II+VN vs VN (**p<0.01), two-tailed student’s t-test). Both VN alone and the IGF:VN complexes significantly (p<0.01) increased MCF-7 cell viability compared to SFM control treatment. TRI complex-stimulated MCF-7 viability was significantly (p<0.05) greater than VN-stimulated viability. Error bars indicate SEM. SFM = Serum Free Media, VN = Vitronectin and TRI = trimeric IGF-I:IGFBP-5:VN complex.
Chapter 3: Identifying changes in gene expression profile associated with VN- and IGF:VN complex-stimulated breast cancer cell survival 72
3.3.2 Optimal MCF-7 cell seeding density for studies in 6-well plates Previous studies have shown that high-density cell cultures result in morphological changes and influences endogenous gene expression (Wang & Adamo, 2000; Schmitt-Ney & Habener, 2004). Prior to performing microarray analysis, MCF-7 cell seeding density was optimised to avoid changes associated with cell confluency while allowing the cells to multiply to a sufficient number to facilitate efficient RNA extraction and yields. Since the maximum cell viability observed was in response to the IGF-I:IGFBP-5:VN complex, this treatment was used as the determinant to optimise the cell density. Cells treated with VN alone were also used in parallel for comparative purposes. Comparison of different cell densities grown in 6-well plates pre-coated with the IGF-I:IGFBP-5:VN complex over a 48 hour period showed that 5 x 105 cells per well is the optimal seeding density for MCF-7 cells (Figure 3.2). This was clearly demonstrated in microscopic images wherein wells with 5 x 105 cells reached an optimal sub-confluent density whereas wells with 7.5 x 105 seeded cells reached an undesirable overly confluent state. In addition, 2.5 x 105 cells per wells were not sufficient to obtain reliable amounts of RNA for downstream molecular analyses.
Chapter 3: Identifying changes in gene expression profile associated with VN- and IGF:VN complex-stimulated breast cancer cell survival 73
Figure 3.2: Optimisation of MCF-7 cell density for microarray analysis. Three different densities of MCF-7 cells, including 2.5 x 105, 5 x 105 and 7.5 x 105 cells per well were seeded into each well of 6-well plates pre-coated either with VN only or trimeric IGF-I:IGFBP:VN complex and allowed to grow for 48 hours. Cell confluency was observed and recorded using light microscopy. Scale bars indicate 100 μm. TRI = trimeric IGF-I:IGFBP:VN complex.
Chapter 3: Identifying changes in gene expression profile associated with VN- and IGF:VN complex-stimulated breast cancer cell survival 74
3.3.3 RNA Quality Control (QC) The quality and integrity of RNA can significantly impact the performance of microarray experiments. To confirm whether the total RNA samples extracted from MCF-7 breast cancer cells were intact and free of DNA contamination, 1 μL of the RNA sample was loaded onto a 2% agarose gel. The total RNA separated into two distinct 28S rRNA and 18S rRNA bands with an intensity ratio of 2:1, respectively (Figure 3.3). No smearing within the samples was observed, indicating the RNA was intact and free of any contaminating DNA.
In addition, integrity, purity and the quantity of extracted RNA samples was determined by A260/A280 ratios (ideal A260/A280 and A230/A260 ratio for RNA is ~2.0) using UV spectroscopy. All isolated total RNA samples had A260/A280 and A230/A260 ratios of 2.00-2.15, indicating purity of the samples and yielded a sufficient quantity of RNA (Table 3.1). In addition, all samples demonstrated
A260/A230 ratios above 2.0 indicating minimal solvent carry over. Taken together, these results indicated that all isolated total RNA samples were of sufficient quality to be used for downstream applications.
Chapter 3: Identifying changes in gene expression profile associated with VN- and IGF:VN complex-stimulated breast cancer cell survival 75
Figure 3.3: Integrity and quality check of RNA samples extracted from MCF-7 cells. MCF-7 cells were seeded into 6-well plates pre-coated with either VN alone or in combination with IGF-I+IGFBP-5 (TRI) and IGF-II. Cells were incubated for 12 and 48 hours, lysed and RNA was extracted using the Direct-zolTM RNA MiniPrepkit (Zymo Research). One µL RNA sample was loaded onto a 2% agarose gel, run for 30 minutes at 110 volts and visualised under UV light. Hundred base pair DNA molecular marker XIV (Roche) was run in parallel. SFM = Serum Free Media and TRI = trimeric IGF-I:IGFBP:VN complex.
Chapter 3: Identifying changes in gene expression profile associated with VN- and IGF:VN complex-stimulated breast cancer cell survival 76
Table 3.1. QC of isolated total RNA samples for microarray analysis.
Purity and concentration of extracted RNA samples was measured using Nanodrop® ND-1000 spectrophotometer. RNA samples with A260/A230 and A260/A280 ratios of 2.00 - 2.2 were considered valid for further downstream applications.
RNA Conc (ng/µL) A260/A280 A260/A230
SFM 12hrs (Exp-1) 166.6 2.03 2.04
VN 12hrs (Exp-1) 267.9 2.02 2.10
TRI 12hrs (Exp-1) 254.7 2.03 2.10
IGF-II+VN 12hrs (Exp-1) 313.6 2.03 2.08
SFM 48hrs (Exp-1) 157.0 2.03 2.01
VN 48hrs (Exp-1) 347.5 2.00 2.12
TRI 48hrs (Exp-1) 773.1 2.04 2.14
IGF-II+VN 48hrs (Exp-1) 337.1 2.02 2.12
SFM 12hrs (Exp-2) 377.5 2.03 2.15
VN 12hrs (Exp-2) 404.4 2.00 2.15
TRI 12hrs (Exp-2) 567.1 2.02 2.18
IGF-II+VN 12hrs (Exp-2) 426.0 2.04 2.13
SFM 48hrs (Exp-2) 508.8 2.07 2.13
VN 48hrs (Exp-2) 444.4 2.03 2.13
TRI 48hrs (Exp-2) 737.1 2.08 2.17
IGF-II+VN 48hrs (Exp-2) 678.2 2.08 2.16
SFM 12hrs (Exp-3) 250.4 2.04 2.06
VN 12hrs (Exp-3) 546.0 2.13 2.07
TRI 12hrs (Exp-3) 422.4 2.03 2.09
IGF-II+VN 12hrs (Exp-3) 822.3 2.12 2.10
SFM 48hrs (Exp-3) 475.0 2.01 2.09
VN 48hrs (Exp-3) 532.1 2.14 2.05
TRI 48hrs (Exp-3) 646.0 2.13 2.05
IGF-II+VN 48hrs (Exp-3) 457.5 2.03 2.10
SFM = Serum Free Media and TRI = trimeric IGF-I:IGFBP-5:VN complex.
Chapter 3: Identifying changes in gene expression profile associated with VN- and IGF:VN complex-stimulated breast cancer cell survival 77
3.3.4 Global gene expression patterns induced by VN and IGF:VN complexes Text files containing the individual raw array data (probe intensities) were generated using Illumina’s GenomeStudio® data analysis software (AGRF), imported into GeneSpring GX 11 software and pre-processed using Robust Multi-chip Average (RMA). Statistically significantly expressed gene across treatment groups, taking into account the variability within the three biological replicates per group were then identified using two-way ANOVA with BHFDR as the multiple testing correction (p<0.05). A data set containing probe intensities of 3650 gene identifiersas being significantly expressed (across the conditions for all biological replicates) in cells were then imported into R program to generate and plot a heatmap, using the “gplots” package. The heatmap provided a graphical representation of changes in global gene expression patterns in MCF-7 cells in response to VN alone, the IGF- II:VN and IGF-I:IGFBP-5:VN in addition to control treatment (cells in SFM) (Figure 3.4). The genes and treatments were clustered using hierarchical cluster analysis based on ‘similarity in expression’ using algorithms (Ward's minimum variance method) within the R package “gplots”. This method was also used to measure Euclidean distance between two treatments. The resulting analysis was graphed as dendrograms within the heatmaps and Euclidean distance maps.
The clustering evident within the upper dendrogram (hierarchical tree) demonstrated that the largest difference in gene expression patterns is evident between the 12 and 48 hour time point samples (Figure 3.4). It is also seen that VN and the IGF-I:IGFBP:VN treatments among all treatments shared the most common pattern of gene expression changes at both 12 and 48 hour time points indicated by the homogenous shading and the strong clustering observed for these treatments. Moreover, very similar expression patterns were observed for VN, IGF-II:VN and IGF-I:IGFBP:VN treatments for the 48 hour time point. Furthermore, changes in gene expression induced by VN, IGF-II:VN and IGF-I:IGFBP:VN treatments in comparison with control SFM treatment were observed to be larger at 48 hours compared to those seen at the earlier 12 hour time point (Figure 3.4).
These results can also be deduced from Euclidean distance map analysis (Figure 3.5). VN and IGF-I:IGFBP:VN treatments induced the most similar patterns of gene expression changes at 12 and 48 hour time points, indicated by the similar shading and the smaller Euclidean distances (between 16 to 18) and the late
Chapter 3: Identifying changes in gene expression profile associated with VN- and IGF:VN complex-stimulated breast cancer cell survival 78
branching of the dendrogram compared to other treatments (Figure 3.5). The IGF- II:VN treatment was also found to have the most similar patterns of gene expression to VN and IGF-I:IGFBP:VN treatments with 18.9 and 19.3 Euclidean distance at 48 hour time point, respectively.
Taken together, these data suggest that changes in global gene expression patterns induced by VN alone, dimeric IGF-II:VN and trimeric IGF-I:IGFBP:VN complexes are greater at the longer (48 hours) time point compared to the early (12 hours) time point when compared to cells grown in SFM control treatment. The clustering analysis also suggests that changes in gene expression patterns are strongly influenced by VN which is presented in all three treatments.
Chapter 3: Identifying changes in gene expression profile associated with VN- and IGF:VN complex-stimulated breast cancer cell survival 79
Figure 3.4: Heatmap of changes in global gene expression patterns. The heatmap was generated using the heatmap function in the “gplots” package of the R program. Statistically significant (p<0.05) gene identifiers across all three biological replicates in response to each tretament and time point were identified using GeneSpring GX 11 software. The clusters (hierarchical tree) were generated using hierarchical cluster algorithms based on Ward's minimum variance method within the R package “gplots”. The left and top dendrograms are clustering of genes and treatments, respectively based on similarity in expression levels. The expression of the gene is compared across all treatments, where low expression is indicated by green, while high expression is indicated in red. SFM = Serum Free Media and TRI = trimeric IGF-I:IGFBP:VN complex.
Chapter 3: Identifying changes in gene expression profile associated with VN- and IGF:VN complex-stimulated breast cancer cell survival 80
Figure 3.5: Euclidean distance map of changes in global gene expression patterns. Euclidean distance map produced by the R package “gplots”. The numerals signify the Euclidean distance between the treatments, with regards to their gene expression profiles. Statistically significant gene identifiersacross all three biological replicates were identified using GeneSpring GX 11 prior to Euclidean distance analysis. The Euclidean distance between two treatments and clustering was analysed using hierarchical cluster algorithms based on Ward's minimum variance method within the R package “gplots”. The closer the Euclidean distance is to zero indicates increasing similarity between the two treatments, with regards to their profiles. This is also demonstrated by the shading of the coloured square, where an increase in shading is proportional to increased similarity between the two treatments. Likewise, the dendrogram (hierarchical tree) on either side of the map clusters the treatments based on their similarity, with late branching indicating treatments which are more similar to each other. For example, the largest difference when comparing the treatments is IGF-II:VN (12 hours) and TRI (48 hours) with Euclidean distance value of 59.54. SFM = Serum Free Media and TRI = trimeric IGF-I:IGFBP:VN complex.
Chapter 3: Identifying changes in gene expression profile associated with VN- and IGF:VN complex-stimulated breast cancer cell survival 81
3.3.5 Identification of differential gene expression Using GeneSpring GX 11 software, a number genes were identified to be significantly (p<0.05) expressed in response to each treatment compared to the control treatment at 12 and 48 hour time points. An additional ‘per gene normalisation’ step was performed in which the expression of each gene identifier across treatments was normalised against the median expression of the corresponding gene identifier on the control arrays. This ensured that genes which did not change across the treatments received a normalised expression value of one. Replicates for each sample array were obtained by “Treatment type”, effectively forming eight treatment groups, each containing the combined normalised data generated from the hybridisation of the 3 biological replicates. The 8 treatment groups were as follow 1) Control or no treatment at 12 hours (SFM at 12 hours), 2) VN alone at 12 hours, 3) IGF-I+IGFBP-5+VN at 12 hours, 4) IGF-II+VN at 12 hours, 5) no treatment at 48 hours (SFM at 48 hours), 6) VN alone at 48 hours, 7) IGF-I+IGFBP-5+VN at 48 hours and 8) IGF-II+VN at 48 hours.
In order to identify differentially expressed genes in response to each treatment at 12 and 48 hours, pair-wise comparisons between each “treatment” group and the “control” group were performed. In addition, pair-wise comparisons between each “treatment” group and “VN alone” were also performed to identify IGF-I+IGFBP-5- and IGF-II-induced changes from those induced by VN. Statistically significant differentially expressed genes were then identified using two-way ANOVA with BHFDR as the multiple testing correction (p<0.05). The selection criteria of ±1.5- fold were used to ensure critical genes which may not have a large magnitude of change was identified, while also keeping the data set to a manageable size.
Table 3.2. Number of gene identifiers differentially regulated in pair-wise comparisons. The table shows the number of gene identifiers identified to be at least ±1.5-fold differentially expressed in each pairwise comparison (p<0.05). SFM = Serum Free Media and TRI = trimeric IGF- I:IGFBP:VN complex.
12 hours VN TRI IGF-II+VN Control (SFM) 245 263 362 VN - 104 216 48 hours Control (SFM) 568 594 595 VN - 147 164
Chapter 3: Identifying changes in gene expression profile associated with VN- and IGF:VN complex-stimulated breast cancer cell survival 82
Since the control arrays (SFM treatments) represented RNA isolated from cells grown in un-coated wells, the number of transcripts differentially regulated by VN only, IGF-II:VN and IGF-I:IGFBP:VN treatments were compared to the control arrays. A number of transcripts were identified to be differentially expressed in MCF-7 cells by VN alone (245 at 12 hours and 568 at 48 hours), IGF-I:IGFBP:VN (263 at 12 hours and 594 at 48 hours) and IGF-II:VN (362 at 12 hours and 595 at 48 hours) treatments when compared to the control arrays (Table 3.2). As expected in view of enhanced survival observed in MCF-7 cells in response to IGF-I:IGFBP:VN compared to VN alone treatments (Figure 3.1), a substantial number of IGF- I:IGFBP:VN-unique transcripts were identified. Cells treated with IGF-I:IGFBP:VN were seen to have 104 and 147 uniquely regulated genes at the 12 and 48 hour time points compared to cells treated with VN only, respectively. Similarly, 216 genes at 12 hours and 164 genes at 48 hours were uniquely regulated in these cells by the IGF-II:VN treatment when compared to cells treated with VN alone.
Direct pair-wise comparison of treatments also demonstrated a considerable overlap in probes differentially expressed in response to VN alone, IGF-I:IGFBP:VN and IGF-II:VN treatments compared to the control treatment with 97 and 378 genes at 12 and 48 hour time points, respectively (Figure 3.6). Specifically, 40% (146 out of 362) and 60% (159 out of 263) of differentially expressed genes in response to IGF-II:VN and IGF-I:IGFBP:VN treatments were in common with VN treatment at 12 hour time point, respectively. Similarly, the IGF-II:VN and IGF-I:IGFBP:VN treatment groups were shared 72 % (428 out of 595) and 80% (447 out of 594) of differentially expressed genes at 48 hours with VN treatment group (Figure 3.6). This indicates that the majority of transcripts differentially regulated by these three treatments, in comparison to the control cells, are most likely due to the effect of VN which is present in all treatments. This implies that VN is potentially responsible for the majority of changes in the gene expression identified in these treatments. Since VN was observed to be a significant contributor to the functional responses induced by IGF:VN treatments, it was no surprise that VN was observed to induce the bulk of the changes in gene expression stimulated by IGF-II:VN and IGF-I:IGFBP:VN treatments. This will be discussed in further detail in the following chapter.
Chapter 3: Identifying changes in gene expression profile associated with VN- and IGF:VN complex-stimulated breast cancer cell survival 83
12 hour 48 hour
hour VN vs SFM TRI vs SFM VN vs SFM TRI vs SFM (245 probes) (263 probes) (568 probes) (594 probes)
IGF-II+VN vs SFM IGF-II+VN vs SFM (362 probes) (595 probes)
Figure 3.6: Pair-wise comparisons of gene identifiers differentially expressed. All gene identifiers passed a ±1.5-fold change threshold compared to the control samples and two-way ANOVA with BHFDR, p<0.05. The venn diagram shows the number of gene identifiers in common or uniquely expressed by each treatment when compared to the expression of the control group (SFM). SFM = Serum Free Media and TRI = trimeric IGF-I:IGFBP-5:VN complex.
Chapter 3: Identifying changes in gene expression profile associated with VN- and IGF:VN complex-stimulated breast cancer cell survival 84
3.3.6 Gene ontology analysis of differentially regulated genes by VN and the IGF:VN complexes The microarray analysis produced large data sets of genes differentially expressed in response to VN alone, dimeric IGF-II:VN and trimeric IGF-I:IGFBP- 5:VN complexes at 12 and 48 hour time points. Such differentially expressed genes are often from a diverse range of gene families associated with a variety of biological functions. Ingenuity Pathway Analysis (IPA) software was used to interpret and place differentially regulated genes in the context of biological processes and networks. Data sets containing genes identified as being significantly differentially expressed in MCF-7 cells by at least ±1.5-fold compared to the control SFM treatment at both 12 and 48 hours were uploaded into IPA. Out of the probe identifiers uploaded for each data set, > 90% were identified and mapped to corresponding gene objects in the Ingenuity Knowledge Base (termed focus genes).
Biological functions were assigned to the whole data set using IPA knowledge base reference database. A propriety ontology representing over 300,000 classes of biological objects and millions of individual modelled relationships between proteins, genes, complexes, cells, tissues, small molecules and diseases was used to identify enriched biological functions associated with differentially expressed genes in response to each treatment. Molecular and cellular functions with a Fischer’s exact test score below 0.05 and containing at least 3 focus genes were considered significant. IPA analysis identified a number of biological functions associated with genes differentially regulated. For full details of biological processes associated with genes differentially regulated at 12 hour and 48 hour time points, refer to (Appendices 6 to 8). Ranked in order of significance, Figures 3.7 and 3.8 display the top ten biological functions associated with genes differentially expressed in response to VN, IGF-II:VN and IGF-I:IGFBP:VN treatments at 12 and 48 hour time points.
Biological processes involved in “cellular development” followed by “cellular growth and proliferation” were found to be the most significant functions associated with genes differentially expressed at 12 hours among all three treatments (Figure 3.7). Similarly, “cellular growth and proliferation” was identified as one of the highest ranking functional categories associated with differentially expressed genes at 48 hours in response to VN, IGF-II:VN and IGF-I:IGFBP:VN treatments,
Chapter 3: Identifying changes in gene expression profile associated with VN- and IGF:VN complex-stimulated breast cancer cell survival 85
containing 153, 113 and 171 focused genes, respectively (Figure 3.8). “Cellular movement” was also found as one of the most significant functional categories associated with differentially expressed genes at 48 hours. Importantly, “cell death and survival” was identified as one of the most significant functional categories at both 12 hour (42, 83 and 57 focus genes) and 48 hour time points (130, 102 and 149 focus genes) for cells treated with VN, IGF-II:VN and IGF-I:IGFBP:VN, respectively (Figure 3.8).
Chapter 3: Identifying changes in gene expression profile associated with VN- and IGF:VN complex-stimulated breast cancer cell survival 86
2 2 8 3 2 0 2 3 7
4
4 4 1 1 2 1 1 1 4
4 9 7 3 8 3 1
6 6 4
6 8 4 3 1 8 4
5 1 4
3 9 7 0 4 5 2
7 4 7
4 2 5 2 1 1 3
Figure 3.7: Top ten molecular and cellular functions associated with differentially expressed genes at 12 hours. Out of all differentially expressed genes at 12 hours in response to VN, IGF-II:VN and IGF-I:IGFBP- 5:VN treatments, 196, 304 and 220 genes, respectively, were identified and mapped as gene objects in Ingenuity Pathway Knowledge for molecular and cellular function analysis. Molecular and cellular functions were ranked in the order of significance with a Fischer’s exact test score below 0.05 and containing at least 4 focus genes. Number of involved genes associated with each biological function is also represented within the columns of the bar graph. SFM = Serum Free Media. Chapter 3: Identifying changes in gene expression profile associated with VN- and IGF:VN complex-stimulated breast cancer cell survival 87
0 3 9 0
4 3 0 6 8 8
4 5 1 3
9 9 9 2 8 5
1 1 1 1
3 2
5 8 5 8 9 3 3 8
1 0
7 8 5 3 3 2 3 2
1 1
0 1 6 0 9
4 9 0 2 9
1 7 5 3 4
9 9 4 3 5
1 1 1 1 1
Figure 3.8: Top ten molecular and cellular functions associated with differentially expressed genes at 48 hours. Out of all differentially expressed genes at 48 hours in response to VN, IGF-II:VN and IGF-I:IGFBP- 5:VN treatments, 507, 530 and 519 genes, respectively, were identified and mapped as gene objects in Ingenuity Pathway Knowledge for molecular and cellular function analysis. Molecular and cellular functions were ranked in the order of significance with a Fischer’s exact test score below 0.05 and containing at least 4 focus genes. Number of involved genes associated with each biological function is also represented within the columns of the bar graph. SFM = Serum Free Media.
Chapter 3: Identifying changes in gene expression profile associated with VN- and IGF:VN complex-stimulated breast cancer cell survival 88
3.3.7 Pathway analysis Canonical pathway analysis was performed to relate the differentially expressed genes to well documented biological pathways. This analysis identified the pathways from the IPA library of canonical pathways that were most significant to the genes expressed differentially in cells treated with VN, IGF-II:VN and IGF- I:IGFBP-5:VN treatments. To assign canonical pathways to the dataset the significance p-value as well as the ratio of focus genes to overall genes in each pathway was used. Since the number of differentially expressed genes with roles in cell survival at 48 hours was substantially more than those detected at12 hours, the 48 time point was chosen to report the data obtained from the canonical pathway analysis.
Table 3.3 depicts the top ten canonical pathways associated with differential gene expression for each treatment at 48 hours. When ranked according to p-value, the Aryl Hydrocarbon Receptor (AHR) pathway was identified as the most significant canonical pathway in response to VN (p = 1.37 x 10-3), IGF-II:VN (p = 2.03 x 10-3) and IGF-I:IGFBP-5:VN (p = 9.82 x 10-5) treatments (Table 3.3). The transforming Growth Factor beta (TGFβ) signalling pathway with previously reported roles in breast cancer progression (Buck & Knabbe, 2006) was ranked as the second highest (P = 1.25 x 10-3) canonical pathway associated with IGF-I:IGFBP- 5:VN treatment. This pathway was also found to be one of the most significant pathways in response to VN alone (P = 1.81 x 10-2) and IGF-II:VN (P = 6.44 x 10-3) treatments. Interestingly, molecular mechanism of cancer was one the highest ranking canonical pathways associated with IGF-II:VN (P = 5.22 x 10-3) and IGF- I:IGFBP-5:VN (P = 1.48 x 10-3) complexes at 48 hours (Table 3.3).
Chapter 3: Identifying changes in gene expression profile associated with VN- and IGF:VN complex-stimulated breast cancer cell survival 89
Table 3: Top ten canonical pathways associated with differentially expressed genes at 48 hours. Top ten canonical pathways identified by IPA as being associated with genes differentially expressed (±1.5-fold) at 48 hours in response to VN, IGF-II:VN and IGF-I:IGFBP-5:VN treatments (p<0.05).
VN vs SFM
* Canonical pathway Significance Ratio p-value AHR signalling 1.37 x 10-3 10/141 (0.071) RAR activation 2.93 x 10-3 11/179 (0.061) NRF2-mediated oxidative stress response 3.65 x 10-3 11/187 (0.059) GDP-glucose biosynthesis 5.45 x 10-3 2/5 (0.4) GDNF family ligand-receptor interactions 5.7 x 10-3 6/70 (0.086) Glucose and glucose-1-phosphate degradation 8.05 x 10-3 2/6 (0.333) Thyroid cancer signalling 1.5 x 10-2 4/4 (0.098) UVC-Induced MAPK Signalling 1.72 x 10-2 4/42 (0.095) TGF-β signalling 1.81 x 10-2 6/93 (0.065) Glycogen degradation II 1.84 x 10-2 2/9 (0.222) IGF-II+VN vs SFM AHR signalling 2.03 x 10-3 10/141 (0.071) Hereditary breast cancer signalling 2.13 x 10-3 9/119 (0.076) Role of CHK proteins in cell cycle checkpoint control 2.56 x 10-3 6/57 (0.105) ATM signalling 3.65 x 10-3 6/61 (0.098) Cell cycle: G2/M DNA damage checkpoint regulation 4.39 x 10-3 5/48 (0.104) Mitotic roles of polo-like kinase 4.68 x 10-3 6/64 (0.094) Molecular mechanisms of cancer 5.22 x 10-3 17/363 (0.047) NRF2-mediated oxidative stress response 5.42 x 10-3 11/187 (0.059) GPD-glucose biosynthesis 6.05 x 10-3 2/5 (0.4) TGF-β signalling 6.44 x 10-3 7/93 (0.075) IGF-I+IGFBP-5+VN vs SFM AHR signalling 9.82x 10-5 12/141 (0.085) TGF-β signalling 1.25 x 10-3 8/93 (0.086) Molecular mechanisms of cancer 1.48 x 10-3 18/363 (0.05) LPS/IL-1 mediated inhibition of RXR function 1.93 x 10-3 13/222 (0.083) Regulation of IL-2 expression in activated and anergic 2.47 x 10-3 7/84 (0.069) T lymphocytes Epithelial adherens junction signalling 2.75 x 10-3 10/144 (0.119) UVC-induced MAPK signalling 3.36 x 10-3 5/42 (0.5) Tight junction signalling 4.06 x 10-3 10/157 (0.064) NRF2-mediated oxidative stress response 4.07 x 10-3 11/187 (0.059) ILK signalling 4.82 x 10-3 11/185 (0.059)
* Ratio of focus genes present in each pathway
Chapter 3: Identifying changes in gene expression profile associated with VN- and IGF:VN complex-stimulated breast cancer cell survival 90
3.3.8 Network analysis The IPA network analysis was further performed to investigate regulatory networks associated with the differentially expressed genes in cells treated with VN, IGF-II:VN and IGF-I:IGFBP-5:VN. To start generating networks, IPA queries the IPA Knowledge Base for interactions between focus genes and all other gene objects stored in the knowledge base, and generates a set of networks to describe the functional relationships between gene products based on known interactions in the literature. The output, displayed graphically as nodes (genes) and edges (the biological relationship between the nodes), provides a detailed representation of a number of biological pathways and functions implicated by the data set. The tool then associates these networks with known biological pathways. Using a 99.9% confidence level (score ≥ 3) that the genes present in a particular network are not due to random chance, the tool identified and mapped 15 and 25 networks significantly associated with differential gene expression in response to each treatment at 12 and 48 hour time points, respectively. The networks generated ranged in score depending of their significance and number of focus genes. Differentially expressed genes associated with the top five functional networks are reported in (Appendices 9 to 11).
Network analysis indicated that a number of networks, including the top two significant networks associated with genes differentially expressed at 12 and 48 hours in response to VN and the IGF:VN complexes, contained PI3-K, AKT, ERK and MAPK signalling pathways primarily as central nodes in these networks. This is in alignment with previous reports from our laboratory and others showing activation of PI3/K/AKT and MAPK/ERK pathways by VN and the IGF:VN complexes (Salasznyk et al., 2004; Hollier et al., 2008; Kashyap et al., 2011; Bae et al., 2012). Specifically, differentially expressed genes at 12 hours in response to each treatment were found to be significantly associated with 5 networks out of 15 mapped networks containing PI3K, AKT and ERK signalling pathways (data not shown). Although not differentially regulated, these pathways mapped primarily to the core of associated networks. Interestingly, all three pathways were found to be involved in the top two ranking networks associated with VN, IGF-II:VN and IGF-I:IGFBP:VN treatments at 12 hours (Figure 3.9). This indicates the possibility that VN and the IGF:VN complexes may regulate the activity of a number of genes present in these networks via PI3K, AKT and ERK pathways. Chapter 3: Identifying changes in gene expression profile associated with VN- and IGF:VN complex-stimulated breast cancer cell survival 91
Figure 3.9: Top two functional networks associated with differentially expressed genes at 12 hour.
The top two ranking networks significantly associated with genes differentially expressed by VN, IGF-II:VN and IGF-I:IGFBP-5:VN treatments at 12 hours. These pathways were found to contain key survival signalling pathways including, PI3K, AKT and ERK pathways (bold letters and colour coded in yellow). Nodes are colour coded according to their gene expression determined by the microarray analysis (red, up-regulated; green, down-regulated). Unshaded nodes or presented in yellow colour were either not on the expression array or not significantly regulated. Number of focus genes in each network and the score are also shown for each network. The score reflects the negative algorithm of p- value that indicates the likelihood of the focus genes in a network being found together due to random chance. Scores of 3 or higher have a 99.9% chance of not being generated by random chance alone.
Chapter 3: Identifying changes in gene expression profile associated with VN- and IGF:VN complex-stimulated breast cancer cell survival 92
Functional network analysis of genes at 48 hours also identified 5 networks containing PI3K, AKT and ERK pathways out of the 25 mapped networks. Similar to networks identified at 12 hours, the two highest ranking networks associated with genes at 48 hours were found to contain PI3-K and AKT within the networks (Figure 3.10). The ERK pathway also mapped to the third most significant network (Appendices 9 to 11). Furthermore, members of Nuclear Factor kappa-light-chain- enhancer of activated B cells (NF-κB) family were found to map to the core of the top two networks in response to all treatments tested at 48 hours (Figure 3.10). In addition, this pathway mapped to the core of the highest ranking pathways for both 12 and 48 hours in response to the IGF-II:VN treatment (Figures 3.9 & Figure 3.10). This suggests that NF-κB pathway may play an important role in IGF-II:VN complex-induced changes in gene expression. With regards to the IGF-I:IGFBP- 5:VN treatment, the oncogene FOS was seen to map to the core of the most significant network at 48 hours (Figure 3.10).
VN via ligation of cell surface integrins can also trigger intracellular signalling cascades via FAK and the Src tyrosine kinase pathways, which affect critical cellular events contributing to cancer progression, including cell attachment, cell proliferation and cell survival (Hapke et al., 2003; Hurt et al., 2010). Interestingly, genes differentially expressed in cells treated with VN or the IGF:VN complexes at both time points were found to be associated with networks containing FAK and Src signalling pathways. Networks containing the FAK pathway were found to include other associated intracellular signalling intermediates including, Src, MAPK, P38 MAPK, MAP2K 1/2, PI3K and ERK pathways in addition to the integrin receptors being mapped as the major core nodes to these FAK containing networks (Figure 3.11). This is likely due to the effect of VN-binding integrins activating FAK phosphorylation (Hurt et al., 2010) which subsequently triggers other signalling pathways downstream, such as AKT and MAPK (McLean et al., 2005). Taken together, these data suggest the involvement of intracellular MAPK/ERK, PI3K/AKT, FAK, Src and NF-κB pathways in VN- and IGF:VN complex-stimulated MCF-7 breast cancer cell survival. In addition, these results suggest the possibility that the differential expression of genes involved in these networks may play an important role in VN- and IGF:VN-induced cell survival. These data also provide
Chapter 3: Identifying changes in gene expression profile associated with VN- and IGF:VN complex-stimulated breast cancer cell survival 93
further insights into the selection of critical pathways for identification of the mechanisms underlying VN- and IGF:VN- complex-induced cell survival.
Chapter 3: Identifying changes in gene expression profile associated with VN- and IGF:VN complex-stimulated breast cancer cell survival 94
Figure 3.10: Top two functional networks associated with differentially expressed genes at 48 hours. The top two ranking networks significantly associated with genes differentially expressed in response to VN, IGF-II:VN and IGF-I:IGFBP-5:VN at 48 hours. These pathways were found to contain key survival signalling pathways including, PI3K, AKT, ERK and NF-κB pathways (bold letters and colour coded in yellow). Nodes are colour coded according to their gene expression determined by the microarray analysis (red, up-regulated; green, down-regulated). Unshaded nodes or presented in yellow colour were either not on the expression array or not significantly regulated. Number of focus genes in each network and the score are also shown for each network. The score reflects the negative algorithm of p-value that indicates the likelihood of the focus genes in a network being found together due to random chance. Scores of 3 or higher have a 99.9% chance of not being generated by random chance alone. Chapter 3: Identifying changes in gene expression profile associated with VN- and IGF:VN complex-stimulated breast cancer cell survival 95
Figure 3.11: FAK-containing networks associated with differentially expressed genes at 12 hours and 48 hours. Networks containing FAK signalling pathways significantly associated with genes differentially expressed in response to VN, IGF-II:VN and IGF-I:IGFBP-5:VN treatments at 12 hour and 48 hour time points. These pathways were found to contain other key members of intracellular signals including, MAPK, P38 MAPK, MAP2K 1/2, ERK and PI3K pathways as well as integrin receptors (colour coded in orange). Nodes are colour coded according to their gene expression determined by the microarray analysis (red, up-regulated; green, down-regulated). Unshaded nodes or presented in yellow and orange colours (except integrin which are up-regulated) were either not on the expression array or not significantly regulated. Number of focus genes in each network and the score are also shown for each network.
Chapter 3: Identifying changes in gene expression profile associated with VN- and IGF:VN complex-stimulated breast cancer cell survival 96
3.3.9 qRT-PCR validation of differentially expressed genes A number of genes determined by microarray analysis to be differentially regulated and deemed biologically interesting because of their IPA classification into relevant cell survival categories were selected for validation using qRT-PCR. In addition, among these genes, several genes were observed to be regulated by members of key intracellular singling pathways such as FAK, Src, MAPK/ERK and PI3-K/AKT pathways. Hence, to validate the microarray data, the expression of 21 genes with roles in cell survival and/or involvement in key signalling pathways was assessed using qRT-PCR analysis (Table 3.4).
The differential expression of 16 up-regulated (Figure 3.12 A, B, C & D) and 5 down regulated (Figure 3.13 A & B) target genes were confirmed. The IGF- I:IGFBP-5:VN treatment was found to change the expression of genes namely, LAMB1, GABRP, NBPF20, FZD8, NFIB, HSPB8, PAK4 and MTSS1 above the IGF-II:VN- or VN-induced levels, although the expression of the majority of target genes remained relatively unchanged across all three treatments. This likely suggests that although the IGF-I:IGFBP:VN complex increases the expression of a few candidate genes, overall the gene expression profiles of target genes are strongly influenced by VN which is present in all treatment groups. This was also suggested by the microarray data wherein the majority of genes differentially expressed in these treatments were seen to be commonly regulated by VN.
Chapter 3: Identifying changes in gene expression profile associated with VN- and IGF:VN complex-stimulated breast cancer cell survival 97
Table 3.3. Target genes for qRT-PCR analysis.
Fold change identified by Gene Microarray analysis Full name Regulation Symbol VN IGF- TRI II:VN
JUN Cellular oncogene JUN or c-JUN Up 2.00 1.83 1.68
FOS Cellular oncogene FOS or c-FOS Up 3.40 2.70 2.52
CD24 Cluster of Differentiation 24 Up 1.40 1.64 1.52
TGFβ2 Transforming Growth Factor beta 2 Up 5.30 5.32 6.60
VAV2 vav2 guanine nucleotide Up 1.28 1.11 1.59 exchange factor LAMB1 Laminin beta Up 1.90 2.2 2.87
CTGF Connective Tissue Growth Factor Up 7.96 6.055 6.32
GABRP Gamma-aminobutyric acid Up 21.59 27.70 28.76 receptor subunit pi NBPF20 Neuroblastoma Breakpoint Family, Up 1.01 1.04 1.95 Member 20 FZD8 Frizzled class receptor 8 Up 1.30 1.51 1.65
RASGRP1 RAS guanyl releasing protein 1 Up 5.76 5.04 5.33
ZIC Zinc finger of the cerebellum Up 1.24 1.52 1.86
HSPB8 Heat shock protein beta Up 1.98 1.76 1.62
Cyr61 Cysteine-rich angiogenic inducer 61 Up 5.76 4.54 5.44
ANXA2 Annexin A2 Up 1.34 1.62 1.79
NFIB Nuclear factor 1 B-type Up 1.35 1.21 1.56
H19 Imprinted maternally expressed Down 4.98 2.40 2.58 transcript H19 SPDEF SAM pointed domain-containing Down 1.60 1.52 1.811 ETS transcription factor MTSS1 Metastasis suppressor 1 Down 1.60 1.55 1.77
XBP1 X-box binding protein 1 Down 1.59 1.59 1.51
PAK4 p21 protein (Cdc42/Rac)- Down 1.30 1.57 1.78 activated kinase 4 VN = Vitronectin and TRI = trimeric IGF-I:IGFBP-5:VN complex.
Chapter 3: Identifying changes in gene expression profile associated with VN- and IGF:VN complex-stimulated breast cancer cell survival 98
A
Chapter 3: Identifying changes in gene expression profile associated with VN- and IGF:VN complex-stimulated breast cancer cell survival 99
B
Chapter 3: Identifying changes in gene expression profile associated with VN- and IGF:VN complex-stimulated breast cancer cell survival 100
C
Chapter 3: Identifying changes in gene expression profile associated with VN- and IGF:VN complex-stimulated breast cancer cell survival 101
D
Figure 3.12: qRT-PCR validation of up-regulated target genes identified by microarray analysis. (A, B, C & D) mRNA expression of each target gene was normalised to GAPDH and data from each treatment were expressed as the fold change to control treatment (SFM). The asterisks indicate treatments with significant increase in gene expression compared to SFM (*p<0.05, **p<0.01, two- tailed student’s t-test). The data are pooled from three experiments with treatments tested in at least three wells in each replicate experiment. Error bars indicate SEM. SFM = Serum Free Media and TRI = trimeric IGF-I:IGFBP-5:VN complex.
Chapter 3: Identifying changes in gene expression profile associated with VN- and IGF:VN complex-stimulated breast cancer cell survival 102
A
Chapter 3: Identifying changes in gene expression profile associated with VN- and IGF:VN complex-stimulated breast cancer cell survival 103
B
Figure 3.13: qRT-PCR validation of down-regulated target genes identified by microarray analysis. (A & B) mRNA expression of each target gene was normalised to GAPDH and data from each treatment were expressed as the fold change to control treatment (SFM). The asterisks indicate treatments with significant decrease in expression compared to SFM (*p<0.05, **p<0.01, two-tailed student’s t-test). The data are pooled from three experiments with treatments tested in at least three wells in each replicate experiment. Error bars indicate SEM. SFM = Serum Free Media and TRI = trimeric IGF-I:IGFBP-5:VN complex.
Chapter 3: Identifying changes in gene expression profile associated with VN- and IGF:VN complex-stimulated breast cancer cell survival 104
To fully validate the microarray data, the relative mRNA levels for target genes were assessed across SFM (control), VN alone, IGF-I+IGFBP-5+VN and IGF- II+VN treatments. The expression levels for each candidate gene was then normalised to GAPDH and fold expression relative to control samples determined. This was then compared to the fold change values identified by microarray analysis in response to each treatment compared to the control SFM treatment. Correlation of relative fold change values were then analysed using the Pearson product-moment correlation coefficient in GraphPad Prism 6 software. The results indicated a significant (p<0.0001) correlation between the differential expression determined by microarray analysis and the relative gene expression assessed by qRT-PCR (Figure 3.14). This, therefore aids in validating the data set of genes identified to be differentially regulated in response to VN alone, the dimeric IGF-II:VN and trimeric IGF-I:IGFBP-5:VN complexes.
Chapter 3: Identifying changes in gene expression profile associated with VN- and IGF:VN complex-stimulated breast cancer cell survival 105
Microarray vs qRT-PCR
3 0 Coefficient correlation: 0.8997
P<0.0001
2 0 qRT-PCR
1 0
-1 0 1 0 2 0 3 0
Microarray -1 0
Figure 3.14: Correlation of gene expression levels of 21 target genes determined by microarray analysis with qRT-PCR analysis. Correlations of relative fold change of 21 genes differentially expressed in response to VN alone, the dimeric IGF-II:VN and trimeric IGF-I:IGFBP-5:VN complexes compared to the control treatment (SFM) for 48 hour time point samples as determined by microarray analysis vs qRT-PCR analysis using the Pearson product-moment correlation coefficient analysis in GraphPad Prism 6 software.
Chapter 3: Identifying changes in gene expression profile associated with VN- and IGF:VN complex-stimulated breast cancer cell survival 106
3.4 DISCUSSION
It is widely accepted that the development, progression and subsequent metastasis of breast cancer is largely influenced by constituents of the ECM (Lochter & Bissell, 1995; Bissell & Radisky, 2001; Muschler & Streuli, 2010). Less well established is the extent to which these ECM components actually enhance cancer, and the involvement of the downstream mediators of such functions. In this regard, VN, an ECM glycoprotein, has been shown to induce cell survival, differentiation and tumour formation in a number of cancers, including breast cancer (Aaboe et al., 2003; Hurt et al., 2010). Moreover, altered interactions between mitogenic hormones and growth factors (GFs) such as IGFs are amongst the most important factors promoting breast cancer progression (Beattie et al., 2010). In view of this, previous work within our laboratory has demonstrated that IGF-I can associate with the ECM protein VN through IGFBP-2, -3, -4 and -5, forming multiprotein IGF-I:IGFBP:VN complexes (Kricker et al., 2003) that are capable of stimulating increased breast cancer cell migration and survival and inhibiting drug-induced apoptosis (Hollier et al., 2008; Kashyap et al., 2011). IGF-II was also shown to bind directly to VN forming dimeric IGF-II:VN complexes (Upton et al., 1999), which were shown to stimulate increased breast cancer cell proliferation and migration (Noble et al., 2003).
Previous studies within our laboratory have largely focused on complex- induced migration. Using microarray analysis, a number of unique differentially expressed genes were identified in non-tumorigenic MCF10A breast cells that had migrated in response to the trimeric IGF-I:IGFBP:VN complex. Some of these genes are known to be involved in cell movement, growth and survival, including Tissue Factor (TF), Stratifin (SFN), Ephrin-B2 (EFNB2), Plasminogen Activator Inhibitor-1 (PAI-I), and Basic helix-loop-helix B2 (BHLHB2). The role of TF and SFN in IGF:VN-induced migration were also investigated and confirmed using gene knock- down and overexpression techniques. However, such downstream molecular mediators of IGF-I:IGFBP:VN- and IGF-II:VN-stimulated cell survival have not yet been investigated. In addition, the identity of molecular mediators responsible for VN-induced cell survival remains to be resolved. Hence, the studies reported in this chapter attempted to identify differentially expressed genes involved in cell survival
Chapter 3: Identifying changes in gene expression profile associated with VN- and IGF:VN complex-stimulated breast cancer cell survival 107
stimulated by VN alone, the dimeric IGF-II:VN and trimeric IGF-I:IGFBP-5:VN complexes.
To study the changes in gene expression accompanying the VN and IGF:VN complex-induced cell survival, the strategy adopted was to culture cells in tissue culture plastic wells pre-coated with VN bound to IGF-I+IGFBP-5 or IGF-II. This substrate-bound strategy was adopted to identify genes more accurately representing the molecular and cellular mediators involved in cell survival stimulated by VN and the IGF:VN complexes. In vivo, ECM proteins bind soluble growth factors and regulate their activation, distribution and presentation to cells (Hynes, 2009). In addition, the binding of growth factors and ECM proteins can result in maintenance of local growth factor concentrations, modulate growth factor receptor interactions and signalling, and also alter the diffusion of growth factors through the matrix (Taipale & Keski-Oja, 1997). Indeed, ECM-bound IGFBP-5 has been shown to enhance the effects of IGF-I on smooth muscle cell DNA synthesis, whereas IGFBP- 5 added in solution had no effect on IGF-I-mediated effects (Jones et al., 1993). To accommodate the influence of ECM on growth factor responses, substrate-bound approaches were adopted in the hope that the genes identified would more accurately represent the actual cellular and molecular processes involved in stimulated cell survival in comparison to the more common and relatively static pre-plated culture methodologies previously adopted by others. It was envisioned our approach would also reduce the number of genes identified that may be irrelevant to our phenotype of interest, thus making the data sets obtained more manageable for further analysis.
Using genome-wide microarray analysis the changes in gene expression patterns were identified in MCF-7 breast cancer cells treated with substrate-bound VN alone, the IGF-II:VN and IGF-I:IGFBP-5:VN complexes when compared to the cells in the control SFM treatment. Furthermore, using a threshold of ±1.5-fold (two- way ANOVA with BHFDR, p<0.05), a substantial number of genes were identified to be up- or down-regulated in response to each treatment (Figure 3.6). It was also found that VN is the component responsible for the majority of changes in global gene expression patterns induced by the IGF:VN complex. This was represented by the high degree of commonality in genes differentially regulated by VN alone, IGF- II:VN and IGF-I:IGFBP-5:VN treatments. IGF-I and IGF-II were seen in our current and previous studies (Noble et al., 2003; Hollier et al., 2008; Kashyap et al., 2011) to
Chapter 3: Identifying changes in gene expression profile associated with VN- and IGF:VN complex-stimulated breast cancer cell survival 108
stimulate cell migration and survival responses above those responses observed in VN alone. However, cell migration and survival responses induced by VN were still significantly higher than that observed in SFM controls (Upton et al., 2008; Kashyap et al., 2011). This suggests that VN was the key component in IGF-mediated cell responses, including migration and survival. Thus, it was no surprise that VN was also the component which induced the majority of differential gene expression stimulated by IGF-II:VN and IGF-I:IGFBP-5:VN treatments. Moreover, there were a significantly larger number of differentially expressed genes at 48 hours compared to 12 hours, suggesting a more predominant effect of VN and the IGF:VN complexes at longer time points. Nevertheless, we have also identified a subset of differentially expressed genes uniquely regulated by IGF-II:VN and IGF-I:IGFBP-5:VN complexes, most likely acting via direct action of IGF-II and IGF-I presented in these complexes, respectively. Further studies using non-receptor binding IGF-I and IGF-II analogues, such as [L24][A31]-IGF-I and [Leu27]IGF-II would be required to confirm this direct action on cells.
The microarray analysis allowed for a global view of treatment induced transcriptional responses across time points. Providing us with the first ever insight into downstream molecular events underpinning these ECM:GF complexes. However, to place the differentially expressed genes in a biological context, pathway and network analysis was performed using the IPA knowledge base. Overall, the entire data set of differentially expressed genes were determined to be highly associated with a number of important molecular and cellular functions, the most significant of which included functions such as “cell development”, “cell growth and proliferation”, “cell assembly and organisation”, “cell cycle”, “cell movement” and “cell death and survival” (Figures 3.7 and 3.8). This is in accordance with well documented effects of VN and IGF family members on cell growth, proliferation, migration, inhibition of cell cycle arrest and cell survival (Annunziata et al., 2011; Preissner & Reuning, 2011; Hazawa et al., 2012; Leavesley et al., 2013). Importantly, biological functions related to cell survival, namely “cell growth and proliferation” and “cell death and survival” were ranked as one of the most significant biological functions, adding further evidence for prosurvival effects on VN and IGF:VN complexes on MCF-7 breast cancer cells confirming the prosurvival effect of VN and the IGF:VN complexes on MCF-7 breast cancer cells.
Chapter 3: Identifying changes in gene expression profile associated with VN- and IGF:VN complex-stimulated breast cancer cell survival 109
Comparing and contrasting treatments and time points, there were a substantially larger number of differentially expressed genes with roles in the “cell growth and proliferation” and “cell death and survival” at the 48 hour compared to the 12 hour time-point. This suggests that VN and the IGF:VN complexes, in particular IGF- I:IGFBP:VN complex augment their prosurvival effects at longer time points by stimulating cell survival genes at later time points. Moreover, further VN- and IGF:VN-induced induction and expression of a larger number of genes with roles in cell growth and proliferation at 48 hours. This suggests that differential expression of these genes may be critical for cells progressing through the lag phase of growth/proliferative and subsequently entering the exponential phase. This in turn may be potentially responsible for the concomitant increase in cell survival as measured over a longer 72 hour time period (Figure 3.1). This also fits in well with our previous observations wherein cell viability assessed over 24, 48 and 72 hours in response to these treatments demonstrated significant changes only at 48 and 72 hours, with no changes observed at 24 hours (Kashyap et al., 2011). Moreover, differentially expressed genes in response to the IGF-I:IGFBP-5:VN complex at 48 hours were found to contain more genes involved in the “cell growth and proliferation” biological function (171 focus genes) compared to VN (153 focus gene) and IGF-II:VN (113 focus gene) treatments. A similar result was also observed for differentially expressed genes involved in the “cell death and survival” biological function (Figures 3.7 & 3.8). This can also be reflected in the greater levels of cell viability stimulated by IGF-I:IGFBP-5:VN treatment compared to VN and IGF- II:VN treatments (Figure 3.1). In addition, “cellular movement” was identified as one of the highest ranking biological functions associated with differentially expressed genes at 48 hours in response to VN, IGF-II:VN and IGF-I:IGBP-5:VN treatments. This is in concordance with our previous reports of these treatments stimulating MCF-7 transwell migration (Kricker et al., 2003; Noble et al., 2003; Hollier et al., 2008; Kashyap et al., 2011).
It is well documented that ECM proteins through binding cell surface integrin receptors trigger FAK, Src, PI3-K/AKT, MAPK/ERK and NF-κB signalling pathways promoting survival (Stupack & Cheresh, 2002). Using the IPA knowledge base a number of networks and functional pathways were identified to be significantly associated with the differentially expressed genes within our dataset.
Chapter 3: Identifying changes in gene expression profile associated with VN- and IGF:VN complex-stimulated breast cancer cell survival 110
Interestingly, out of all mapped networks at each time point, 5 networks were identified to include the intracellular PI3-K/AKT and MAPK/ERK pathways mapped to the core of the network. Importantly, these networks were found to be among the highest ranking networks associated with genes differentially expressed in response to VN and the IGF:VN complexes (Figures 3.9 and 3.10). This is in accordance with our previous results demonstrating that VN- and IGF:VN complex-stimulated breast cancer cell survival is mediated via the PI3-K/AKT and MAPK/ERK pathways (Hollier et al., 2008; Kashyap et al., 2011).
Genes differentially expressed in cells treated with VN and the IGF:VN complexes were also identified to be associated with networks containing FAK and Src signalling pathway intermediates. This is consistent with previous findings indicating that FAK and Src are involved in VN stimulated cell adhesion, cell migration and cell survival (Hapke et al., 2003; Hurt et al., 2010). FAK also acts as a receptor-proximal bridging protein that links growth factor receptor and integrin signalling pathways to promote both growth factor and integrin-mediated downstream effects (Sieg et al., 2000). Upon ligation of integrins with VN, FAK is autophosphorylated, creating a high affinity binding site for proteins containing the Src homology 2 (SH2) domain, namely proteins including the upstream Src kinase itself, proto-oncogene Crk (p38) and lymphocyte-specific tyrosine kinase (Lck) (McLean et al., 2005; Hurt et al., 2010). The association of Src and FAK leads to conformational change and activation of the kinase activity of Src which subsequently activates other critical downstream signalling pathways such as AKT and MAPK to promote cellular functions such as cell motility, cell cycle progression and cell survival (Mitra & Schlaepfer, 2006). Indeed, FAK-Src complex activation has been shown to promote cell survival by enhancing activation of both PI3-K/AKT and MAPK/ERK pathways (Schlaepfer et al., 2004; Benlimame et al., 2005). These findings can also be observed in the network analysis data, wherein networks containing FAK also displayed interactions with other key signalling pathways including Src (not shown in the figure), PI3K, AKT, MAPK/ERK, and P38 MAPK pathways as well as integrin receptors (Figure 3.11). This suggest that VN and the IGF:VN complexes via activation of the FAK-Src complex, PI3-K/AKT and MAPK/ERK pathways may regulate a number of genes presented in these networks capable of mediating the observed cell survival responses.
Chapter 3: Identifying changes in gene expression profile associated with VN- and IGF:VN complex-stimulated breast cancer cell survival 111
Network analysis also demonstrated that the NF-κB pathway mapped to the core of the most significant networks associated with differentially expressed genes at 48 hours (Figure 3.10). NF-κB protein is one of the most investigated transcription factors found to control transcription of a number of genes that regulate cancer progression (Chaturvedi et al., 2011). Involvement of NF-κB signalling has been implicated in maintaining self-sufficiency in growth signals, evasion of apoptosis, promoting survival, tissue invasion and metastasis (Naugler & Karin, 2008). Moreover, it has been shown that the activation of NF-κB occurs via the PI3-K/AKT and /MAPK/ERK signalling pathways (Takahashi et al., 2005; Sun et al., 2010). This suggests the possibility that VN- and IGF:VN complex-stimulated activation of the PI3-K-AKT and MAPK/ERK pathways may also lead to the subsequent activation of the NF-κB signalling pathway. Furthermore, the NF-κB pathway was found to be included in the highest ranking networks associated with differentially expressed genes in response to IGF-II:VN complexes at both 12 hour and 48 hour time points (Figures 3.9 and 3.10). This is most likely due to the presence of IGF-II which is known to induce NF-κB activity downstream of the PI3-K/AKT pathway (Kaliman et al., 1999).
The FOS oncogene was found to be mapped to the core of the highest ranking network associated with genes differentially expressed in response to the IGF- I:IGFBP-5:VN complex at 48 hours (Figure 3.10). It is well established that the MAPK/ERK pathway, in the presence of growth factors, is involved in the activation of the members of the FOS family and also mediates its downstream signalling and transcriptional activity (Murphy et al., 2002; Murphy et al., 2004). Moreover, GF activation of the MAPK/ERK pathway regulates cell cycle and proliferation via FOS-mediated signalling (Chambard et al., 2007). The ERK/FOS pathways have been shown to be involved in regulating IGF-I/IGF-IR-mediated vascular smooth muscle cell (VSMC) growth and proliferation (Li et al., 2008). Recent studies within our laboratory have demonstrated that the IGF-I:IGFBP:VN complex prolongs cell survival through activation of the MAPK/ERK pathway (Kashyap et al., 2011). Thus, it is possible that activation of the MAPK/ERK pathway by IGF-I:IGFBP-5:VN complex may also influence the activation of the FOS pathway, in turn promoting cell survival.
Chapter 3: Identifying changes in gene expression profile associated with VN- and IGF:VN complex-stimulated breast cancer cell survival 112
In addition to network analysis, canonical pathway analysis was performed to relate the differentially expressed genes to well documented biological pathways. Amongst the signalling pathways identified, the AHR signalling pathway was the most significant pathway associated with differentially expressed genes at 48 hours (Table 3.3). This pathway was also one of the highest ranking pathways associated with genes differentially expressed at 12 hours. The proposed underlying mechanisms of AHR signalling in promoting cancer cell survival are multifaceted. AHR signalling has been reported to inhibit the expression of key tumour suppressor genes as well as altering proliferation and differentiation by influencing the physiologic processes of cell-cycle, apoptosis and cell–matrix interactions (reviewed in Feng et al., 2013). Importantly, the implication of AHR in cell plasticity and motitility is also supported in studies using xenibiotic ligands such as TCDD (2,3,7,8-tetrachlorodibenzo-[p]-dioxin) for this receptor where treatment of MCF-7 breast cancer cells with TCDD increrased cell surface, appearance of lamellipodia- like morphology and subsequent higher cell adherence and motility (Diry et al., 2006). This sugessts the possibility that VN may also stimulates cell attachemnt and subesequently migration and poliferation through activation of AHR and its downstream signalling. Furthermore, TGFβ signalling, one of the most important pathways in cancer progression was also identified to be highly associated with genes differentially expressed in response to VN and the IGF:VN complexes. Differentially expressed genes namely, c-FOS, c-JUN, RRAS, Smad6, TGIF1, TGFβ2 and TGFβ2 receptor were found to be involved in this pathway. The TGFβ receptor along with its transcriptional mediators, the Smad proteins are known to enhance bone metastasis of breast cancer through regulation of genes such as CTGF (Kang et al., 2003). These data suggest the possibility that the components of AHR signalling and TGFβ canonical pathways and/or their downstream target genes whose deregulation was observed in our data, may be important in VN- and IGF:VN- mediated actions. Amongst the differentially expressed genes associated with pathways identified in network and canonical pathway analyses, we observed genes with previously reported roles in cancer cell survival such as TGFβ2, Cyr61, CTGF, RASGRP1, c- FOS, c-JUN, CAV1, and LAMB1. The presence of these proteins in the top functional networks and signalling pathways highlights the potential role of these genes in VN- and IGF:VN complex-stimulated survival. These genes and several Chapter 3: Identifying changes in gene expression profile associated with VN- and IGF:VN complex-stimulated breast cancer cell survival 113
others either commonly or uniquely regulated in response to the treatments were assessed using qRT-PCR to aid the validation of our microarray data. Overall, there was a high concordance between the qRT-PCR results and the differential expression determined via microarray analysis, whereby a strong correlation between the two was identified (Figures 3.12 and 3.13). This gives us confidence in our microarray data and the sets of genes identified to be differentially regulated by cells in response to VN alone, the IGF-II:VN and IGF-I:IGFBP-5:VN complexes.
The most important finding of microarray studies was the identification of a number of genes differentially expressed in cells survived in response to VN and the IGF:VN complexes that are associated with biological processes highly relevant to cell survival. It was also shown that VN and the IGF:VN complexes also regulate differential expression of genes with roles in critical cellular functions including, cell growth, proliferation and migration (Figure 3.7 and 3.8). The studies reported in this chapter have added valuable knowledge to our understanding of the mechanisms responsible for VN- and IGF:VN complex-stimulated cancer cell survival. This might provide insights into the early molecular events promoting breast cell metastasis, whereby tumour cells proliferating within the primary tumour site, gain the ability to migrate and invade away from their original tissue of origin and acquire the ability to survive within their new tissue microenvironment.
3.5 CONCLUSION
In summary, data acquired within this chapter has identified for the first time unique gene expression patterns underlying VN- and IGF:VN complex-stimulated breast cancer cell survival. Moreover, functional analysis found a number of these genes to be highly relevant to biological processes involved in cell growth, migration and survival. Herein, it is also reported that the FAK-Src complex, PI3-K/AKT MAPK/ERK, and NF-κB intracellular signalling pathways as well as AHR and TGFβ canonical pathways may play central roles in VN- and IGF:VN complex- mediated actions. The following chapter (Chapter 4) uses this dataset to identify candidate genes involved in mediating the cell survival observed in response to VN and the IGF:VN complexes as well as the biochemical mechanisms underlying their regulation. Findings within this chapter suggest that VN is likely the key component driving the differential gene expression profile, with the IGF family members
Chapter 3: Identifying changes in gene expression profile associated with VN- and IGF:VN complex-stimulated breast cancer cell survival 114
stimulating additional subsets of transcriptional programs possibly contributing to the increased cell survival observed above that of VN alone. Therefore, we decided to further investigate genes commonly regulated by VN and the IGF:VN complexes. This strategy aimed at selecting candidate genes to target not only VN- but potentially IGF:VN complex-mediated actions. Since the IGF-I containing complexes are thought to play more significant roles not only in mediating cell survival responses but also in other processes involved in breast cancer metastasis, further studies focused only on trimeric IGF-I:IGFBP-5:VN complexes. However, given the role of IGF-II:VN complex on cell proliferation and migration, it would indeed be worthwhile in the future to investigate the functional role of genes differentially expressed by the dimeric IGF-II:VN complexes and will add valuable knowledge to the molecular mechanisms underlying these responses.
Chapter 3: Identifying changes in gene expression profile associated with VN- and IGF:VN complex-stimulated breast cancer cell survival 115
Chapter 4: Investigation of the mechanisms underlying the regulation of Cyr61 and CTGF expression induced by VN and the IGF:IGFBP:VN complex.
Chapter 4: Investigation of the mechanisms underlying the regulation of Cyr61 and CTGF expression induced by VN and the IGF:IGFBP:VN complex. 116
4.1 INTRODUCTION
It is widely accepted that cellular behaviour and functions in vivo is largely influenced by its interactions with the ECM components, neighbouring cells, and soluble local and systemic cues presented within the cellular microenvironment, namely hormones, cytokines, and growth factors (Nelson & Bissell, 2006; Rondeau et al., 2014). However, the components of the ECM are known to be active contributors to cellular behaviour and biological responses through mediating changes in downstream gene expression programs in normal and malignant cells (Roskelley et al., 1995; Spencer et al., 2007; Birnie et al., 2008). As outlined in the previous chapter, we have used microarray analysis to identify for the first time the transcriptional responses of MCF-7 breast cancer cells to VN and the trimeric IGF- I:IGFBP:VN complex at the early (12 hour) and the late (48 hour) time points (reported in Chapter 3). This revealed a number of genes were identified to be differentially expressed by at least ±1.5-fold in response to VN and the trimeric complex compared to the control treatment (SFM) at these time points (Chapter 3, Figure 3.6). The majority of these genes were observed to be commonly regulated by both VN and the trimeric complex, suggesting that VN is the key component responsible for inducing the majority of differential gene expression stimulated by the trimeric complex. Hence, the studies reported in this chapter aim to identify candidate genes to target for inhibiting not only VN-, but potentially also trimeric complex-mediated downstream functional responses.
Our previous gene ontology and functional pathway enrichment analyses of differentially expressed genes in common by VN and the IGF-I:IGFBP:VN complex identified a number of genes that are highly associated with biological functions relevant to cell survival and proliferation. Among these highly up-regulated genes were members of the ECM-associated CCN matricellular family, namely Cyr61/CCN1 and CTGF/CCN2. Based on previously published literature of their role in breast cancer progression and metastasis (Tsai et al., 2000; Xie, Nakachi, et al., 2001; Kang et al., 2003), Cyr61 and CTGF were selected as our lead candidates for further functional validation. To understand the mechanisms underlying the VN- and IGF-I:IGFBP:VN-stimulated expression of Cyr61 and CTGF, the studies reported in this chapter also investigate the involvement of key cell surface receptors,
Chapter 4: Investigation of the mechanisms underlying the regulation of Cyr61 and CTGF expression induced by VN and the IGF:IGFBP:VN complex. 117
namely the IGF-IR and VN-binding αv-integrins, and the contribution of downstream signalling intermediates (Src/PI3-K/ERK) to this regulation.
4.2 EXPERIMENTAL METHODS
The following is a brief summary of the materials and procedures used to generate the data presented in this chapter. Full details of materials and methods used in experimental procedures are described in Chapter 2.
4.2.1 Cell lines The human breast cancer cell lines used in this chapter were MCF-7, MDA- MB-231, MDA-MB-468 and T47D cells. Cell lines were maintained following procedures described in section 2.3. Since microarray studies reported in Chapter 3 were performed using MCF-7 cells, this cell line was used for the majority of studies implemented in this chapter. Cells from luminal (MCF-7 and T47D), basal (MDA- MB-468) and claudin-low (MDA-MB-231) cell lines were used to represent major breast cancer subtypes were also chosen to screen VN- and the trimeric IGF- I:IGFBP-5:VN complex-regulated Cyr61 and CTGF expression. The mechanistic studies reported in this chapter were performed using MCF-7 (weakly metastatic/luminal) and MDA-MB-231 (highly metastatic/basal B/claudin-low) cell lines.
4.2.2 Heatmap analysis and presentation of target genes using the R program To get a visual representation of the global changes in the gene expression patterns induced by VN and the IGF:VN complexes, the data containing normalised gene values were imported into R 3.0.2 language to generate and plot the heatmaps, using the “gplots” version 2.12.1 package. The expression of each individual gene presented in the heatmap was compared across all the conditions and indicated using different colour keys. The genes and treatments were clustered using hierarchical cluster analysis based on ‘similarity in expression’ using algorithms based on Ward's minimum variance method within the R package “gplots” and results were added to the heatmap as dendrograms (hierarchical tree).
Chapter 4: Investigation of the mechanisms underlying the regulation of Cyr61 and CTGF expression induced by VN and the IGF:IGFBP:VN complex. 118
4.2.3 Gene-function interaction network presentation using Cytoscape 2.8.1 Cytoscape is an open source software platform for visualising molecular interaction networks and integrating these networks with gene expression profiles (Cline et al., 2007). Data from the IPA containing the top five biological functions significantly (p<0.05) associated with genes differentially expressed in common by VN and IGF-I:IGFBP:VN treatments were imported into the open source software platform for molecular network interaction visualisation. Gene-network interactions were represented as node and edges where the nodes were the biological function and the associated genes being the edges.
4.2.4 Pre-binding of proteins The approach used to pre-bind proteins to 6-well plates for use in viability assays was according to protocols in section 2.4. Nunclon™ surface 6-well plates were prepared by pre-binding with proteins as described in section 2.4.1 For all 2D functional assays described throughout this chapter, including receptor and signalling studies, treatment combinations for the pre-binding steps were VN alone and the trimeric IGF-I:IGFBP-5:VN complex with concentrations of proteins added into the culture wells being as follows: VN, 1 μg/mL; IGF-I, 30 ng/mL; and IGFBP-5, 90 ng/mL. For attachment study, in addition to VN, cell culture plates were coated with Poly-l-Lysine, 5 μg/mL; FN, 1 μg/mL; and type I collagen, 1 μg/mL. These concentrations were chosen to coat polystyrene culture plates since visual inspection of cells 1 hour post seeding demonstrated relatively similar levels of overall cell attachment when compared to wells treated with VN.
4.2.5 RNA and protein isolation to assess receptor involvement and signal transduction Sub-confluent cells were serum-starved for 4 hours, detached using a cell dissociation buffer (TrypLE™) and resuspended in DMEM/F12-SFM + 0.05% BSA. To assess the signal transduction pathways involved in VN- and IGF-I:IGFBP:VN- mediated regulation of CTGF and Cyr61, MCF-7 cells were pre-treated with pharmacological inhibitors LY294002 (10 µM), U0126 (10 µM) or PP2 (20 µM) for
60 minutes at 37°C, 5% CO2 prior to seeding. The LY294002 and PP2 inhibitors were used to specifically block PI3-K and Src-family kinases pathways. The MEK inhibitor U0126 was also used to inhibit activation of MAPK (ERK 1/2) by inhibiting the kinase activity of MAP Kinase (Davies et al., 2000). To investigate the
Chapter 4: Investigation of the mechanisms underlying the regulation of Cyr61 and CTGF expression induced by VN and the IGF:IGFBP:VN complex. 119
involvement of possible integrin receptors in CTGF and Cyr61 regulation, MCF-7 and MDA-MB-231 cells were also pre-incubated for 30 minutes at room temperature with monoclonal antibodies against αv-integrin subunit (1:10), αvβ3 (10 μg/mL), αvβ5
(10 μg/mL), β1-subunit (10 μg/mL) receptors and IGF-IR (10 μg/mL). Cells were then seeded onto 6-well plates pre-coated with either VN alone or the IGF-I:IGFBP-
5:VN complex and incubated for a total of 48 hours at 37 °C, 5% CO2 (refer to section 4.2.6 for proteins pre-bound and concentrations used). During this incubation spent medium was replaced after 24 hours with fresh medium containing the respective pharmacological inhibitors and receptor-blocking antibodies. Following the complete 48 hour incubation both floating and attached cells were lysed for RNA and protein solation using TRIzol® reagent and completed protein lysis buffer, respectively according to the protocols outlined in section 2.8.2 and 2.12.1.
4.2.6 qRT-PCR analysis of target genes
RNA was extracted from cell lysates using the Direct-zol™ MiniPrep Kit (Zymo Research) following the manufacture’s protocol and stored at -80 °C until further use (for full details on this method please refer to section 2.8.2.). RNA concentration was measured using the Nanodrop® ND-1000 Spectrophotometer.
Only RNA samples with A260/A280 and A230/A260 ratios of 2.00-2.15 were considered as having valid purity (ideal A260/A280 and A230/A260 ratio for RNA is ~2). RNA samples which passed the purity check were then reverse transcribed into cDNA using the SuperScript® III First-Strand Synthesis SuperMix kit (Life Technologies) following the manufacture’s protocol as briefly outlined in section 2.11.2. The relative fold expression levels of target genes were then measured using qRT-PCR analysis as outlined in section 2.11.3. The cycle threshold (Ct) value of the gene of interest for each reaction was obtained and directly normalised to the housekeeping gene and then compared, using the ΔΔCt method of qPCR data analyses. The relative fold-change in the expression of the gene of interest compared to the control sample (SFM) was then calculated using the 2^(-ΔΔCt) method.
4.2.7 Immunoblotting
The protein concentration was determined using the colorimetric Coomassie Plus® kit (Thermo Fisher Scientific), with BSA used as standards. Proteins (20-60 μg per sample) were resolved on pre-made NuPAGE® gradient SDS-PAGE gels (4%-
Chapter 4: Investigation of the mechanisms underlying the regulation of Cyr61 and CTGF expression induced by VN and the IGF:IGFBP:VN complex. 120
12%) and transferred onto nitrocellulose membranes. Western immunoblotting was performed following procedures outlined in section 2.13. The primary antibodies used in this chapter were anti-Cyr61 (1:1000) and anti-CTGF (1:1000) (R&D system). To detect the total levels of these proteins, the same membranes were stripped using Restore® Western blot stripping buffer and probed with anti-GAPDH to validate equal loading of proteins onto the gel.
4.2.8 Immunofluorescence staining of cells in 2D culture
For full details on this method please refer to section 2.14. The 2D cell cultures were stained using fluorescently labelled dyes and antibodies. All incubations were performed at room temperature. MCF-7 cells growing in 96-well optical-bottom plates pre-coated with VN, IGF:IGFBP-5:VN and poly-l-lysine (control), were fixed after 48 hours with 4% paraformaldehyde, washed twice with PBS:Glycine and then permeabilised with 0.4% Triton-X. Cells were then blocked with 5% BSA, washed with 0.1% PBS-Tween-20 (PBS-T) and incubated for 1 hour either with primary mouse anti-Cyr61 or anti-CTGF antibody (Santa Cruz Biotechnology) diluted 1:100 in 5% BSA + PBS-T. Subsequently, cells were washed twice with PBS-T and incubated with fluorophore conjugated secondary AlexaFluor® 594 (red) or 488 (green) goat anti-mouse IgG antibodies (1:250). Cells were also stained for actin cytoskeleton and nucleus using fluorescently conjugated AlexaFluor® 633 (red) or 488 (green) phalloidin (1:40, in PBS) and DAPI (1:10,000 in PBS), respectively. The plates were stored at 4 °C in the dark until imaging with the DeltaVision microscope (Applied Precision - GE Healthcare).
4.3 RESULTS
4.3.1 VN-induced genes are highly associated with biological processes relevant to cell survival
Using the Illumina® HumanHT-12 v4.0 BeadChip® array technology a number of genes were identified to be differentially expressed by at least ±1.5-fold in response to VN and the trimeric IGF-I:IGFBP:VN complex compared to the control treatment (SFM) at 48 hour (Chapter 3, Figure 3.6). Cells treated with VN and the trimeric complex over 48 hours were seen to uniquely express 121 out of 568 and 147 out of 594 differentially genes, respectively (Figure 4.1 A). Moreover, as previously observed (Chapter 3, Figure 3.6) there was a considerable overlap of
Chapter 4: Investigation of the mechanisms underlying the regulation of Cyr61 and CTGF expression induced by VN and the IGF:IGFBP:VN complex. 121
differentially regulated genes amongst both the treatments. Specifically, 447 genes were regulated in common by VN and the trimeric complex (Figure 4.1 A). Differentially expressed genes from this list of 447 were used to select for candidates to potentially target not only VN-mediated but also the trimeric complex-mediated actions. For ease of reference we will refer to these 447 genes as the VN-induced Gene Expression Signature (or VNGES).
To place the VNGES in a biological context, the IPA knowledge base was utilised. The biological functions assigned to the VNGES were ranked according to the significance of that biological function to the data set. The top five biological functions (in order of significance) associated with these commonly regulated genes that were identified by IPA are represented in Figure 4.1 B. Of particular relevance to this study, “cell growth and proliferation” was identified as the most significant functional category, containing 125 genes (p = 2.22 x 10-8 to 8.81 x 10-2). In addition, the “cell death and survival” function was amongst the highest-ranking functional categories (p = 1.31 10-8 to 8.52 x 10-2) with 106 genes present in our VNGES (Figure 4.1 B). These data suggest that the VNGES is highly enriched in genes with previously reported roles in cell growth and survival providing a suit of genes to target for inhibition of VN- and trimeric IGF-I:IGFBPVN complex-induced cell survival effects.
Chapter 4: Investigation of the mechanisms underlying the regulation of Cyr61 and CTGF expression induced by VN and the IGF:IGFBP:VN complex. 122
A V N v s S F M T R I v s S F M 5 6 8 g e n e s 5 9 4 g e n e s
1 2 1 4 4 7 1 4 7
B
Figure 4.1: Top five biological functions associated with the VNGES. Microarray analyses identified a number of differentially expressed genes in MCF-7 cells in response to VN alone and the IGF-I:IGFBP-5:VN complex (TRI) compared to the control (SFM) treatment. (A) Statistically significant (p<0.05) differentially expressed genes with a ±1.5-fold change threshold were identified using one-way ANOVA with BHFDR. The Venn diagram demonstrates the number of genes in common or uniquely expressed by VN or the TRI complex when compared to the expression in cells treated with SFM. (B) IPA was utilised to investigate genes commonly regulated by VN and the TRI complex (VNGES) with more specific functions within the broader cell survival category. Top five significant biological functions (p<0.05) associated with VNGES were identified and mapped (using Cytoscape) to corresponding gene objects (Red: Up-regulated/Green: Down- regulated). Size of the function represents number of involved genes and increasing darkness of colour represents increasing significance (p values) of the function. VNGES = VN-induced Gene Expression Signature, SFM = Serum Free Media and TRI = trimeric IGF-I:IGFBP:VN complex.
Chapter 4: Investigation of the mechanisms underlying the regulation of Cyr61 and CTGF expression induced by VN and the IGF:IGFBP:VN complex. 123
4.3.2 Matricellular CCN family proteins Cyr61/CCN1 and CTGF/CCN2 associate with breast cancer cell survival
To identify gene clusters that may be involved in cancer cell survival, those associated with both the “cellular growth and proliferation” and the “cell death and survival” biological functions (80 genes) were further sub-categorised. The association of these genes with biological processes relevant to cell growth, survival, proliferation and cell death in a variety of tumour cell types, including breast cancer was identified (As presented in a gene:function interaction network map using the Cytoscape software platform; Figure 4.2). Relative expression levels and the relationship between these genes was then analysed using hierarchical clustering (As represented in the heatmap and associated dendrogram; across VN, the trimeric IGF:IGFBP:VN complex and SFM treatments over 12 and 48 hour time points; Figure 4.2). Several genes were identified with multiple roles in cell proliferation, growth, apoptosis and survival in a range of tumour cells. prominent among these genes were, TGFβ1/2, S100A4, CCN1/Cyr61, CCN2/CTGF, c-JUN, c-FOS, ANXA2, SMAD6, CAV1 and mir21, whose deregulation have been reported to correlate with breast cancer progression and shown to play critical roles in breast cancer cell survival (Figure 4.2; heatmap).
The further analysis of this gene subset revealed the ECM-associated matricellular CCN family members, Cyr61/CCN1 and CTGF/CCN2, to be dramatically up-regulated by VN and the trimeric IGF-I:IGFBP:VN complex (identified by microarray analysis; Table 3.4). Cyr61 and CTGF were also found to be involved in the most significant networks associated with differentially expressed genes (identified by IPA) in response to VN and the IGF-I:IGFBP:VN complex (Chapter 3; Figures 3.9 & 3.10). Previous published literatures indicated the important role of Cyr61 and CTGF overexpression in breast tumorigenesis and metastasis (Tsai et al., 2000; Xie, Nakachi, et al., 2001; Kang et al., 2003). Considering their roles in cell survival and breast cancer progression and their association with key survival pathways commonly implicated in VN- and IGF- I:IGFBP:VN-mediated functions (identified by IPA), Cyr61 and CTGF were selected as our top lead candidates for further functional validation.
Chapter 4: Investigation of the mechanisms underlying the regulation of Cyr61 and CTGF expression induced by VN and the IGF:IGFBP:VN complex. 124
Figure 4.2: VNGES association with biological functions relevant to cell survival. The cluster heat map was generated for 80 genes identified to be associated with the “cell growth and proliferation” and the “cell death and survival” biological functions using R program the “gplots” package. This heatmap represents the changes in gene expression between treatments (SFM, VN and TRI) and time points (12 and 48 hours) (Red: Up-regulated/Green: Down-regulated). The clusters (hierarchical tree) were generated using hierarchical cluster analysis algorithms (Ward's minimum variance method) within the R package “gplots”. The right and top dendrogram are clustering of genes and treatments, respectively based on similarity in expression levels. The IPA tool was used to determine the association these genes to the sub-functions (circles coloured blue). The Cytoscape software was then used to map these associations (grey lines between gene name and sub-function). VNGES = VN-induced Gene Expression Sinature; SFM = Serum Free Media and TRI = trimeric IGF-I:IGFBP-5:VN complex.
Chapter 4: Investigation of the mechanisms underlying the regulation of Cyr61 and CTGF expression induced by VN and the IGF:IGFBP:VN complex. 125 4.3.3 VN- and IGF-I:IGFBP:VN complex-induce Cyr61 and CTGF expression
The differential regulation of Cyr61 and CTGF that was identified by microarray analysis was validated using SYBR-green based qRT-PCR. qRT-PCR analysis confirmed that Cyr61 and CTGF expression in MCF-7 cells was significantly increased (p<0.01) in response to VN at 12 and 48 hour time points compared to the SFM only (control), respectively (Figure 4.3 A). Similarly, trimeric IGF-I:IGFBP-5:VN complex increased the expression of Cyr61 and CTGF at both time points, whereby both genes were elevated at 48 hours as compared to the 12 hour time point (Figure 4.3 A). The increase in Cyr61 and CTGF gene expression was also confirmed at the protein level wherein western blot analysis demonstrated a substantial increase in Cyr61 and CTGF protein expression in response to VN and the trimeric complex over 12 and 48 hours compared to SFM controls (Figure 4.3 B). As reflected in the microarray data, no significant differences in the degree of increased expression of either target genes were observed between VN alone and the trimeric complex trreaments.
To examine whether VN was the major component of the trimeric complex mediating increased expression of Cyr61 and CTGF, we substituted native IGF-I with the [L24][A31]-IGF-I analogue to form the trimeric complex. This analogue retains its affinity for IGFBPs, thus forms heterotrimeric complexes (Kricker et al., 2003), but has a very low affinity for the IGF-IR (Milner et al., 1995; Forbes et al., 2002). No appreciable differences in Cyr61 and CTGF protein expression were evident in cells treated with VN alone or trimeric complex containing either native IGF-I or the [L24][A31]-IGF-I analogue (Figure 4.3 C). This suggests that VN mediates the increased expression of Cyr61 and CTGF induced by the trimeric IGF- complex. Taken together these results clearly show that VN and the trimeric IGF- I:IGFBP-5:VN complex can strongly induce Cyr61 and CTGF expression, whereby VN is the major driver of this increased expression.
Chapter 4: Investigation of the mechanisms underlying the regulation of Cyr61 and CTGF expression induced by VN and the IGF:IGFBP:VN complex. 126
A
B C
Figure 4.3: VN- and the IGF-I:IGFBP:VN complex-induce expression of Cyr61 and CTGF in human breast cancer cells.
Serum starved MCF-7 cells were seeded onto 6-well plates coated with VN (1 μg/mL) either alone (VN) or with IGF-I (30 ng/mL) and IGFBP-5 (90 ng/mL) combination (TRI). (A) RNA was harvested at 12 and 48 hours and expression levels of Cyr61 and CTGF were measured by qRT-PCR analysis. mRNA expression of each target gene was normalised to GAPDH and data from each treatment were expressed as the fold change to control treatment (SFM).The data are pooled from three experiments with treatments tested in at least three wells in each replicate experiment. The asterisks indicate treatments with significant differences compared to their respective control treatment (*p<0.05, **p<0.01, two-tailed student’s t-test). Error bars indicate SEM. (B) Western blotting was used to determine the protein expression levels of Cyr61 and CTGF. The membranes were stripped and re- probed to determine GAPDH levels to confirm equal loading. (C) Cyr61 and CTGF protein expression was also determined in response to TRI complex containing the [L24][A31]-IGF-I analogue and compared against TRI complex containing native IGF-I. SFM = Serum Free Media, TRI = trimeric IGF-I:IGFBP:VN complex and Analog = [L24][A31]-IGF-I.
Chapter 4: Investigation of the mechanisms underlying the regulation of Cyr61 and CTGF expression induced by VN and the IGF:IGFBP:VN complex. 127
4.3.4 Stimulated expression of Cyr61 and CTGF is specific to integrin ligation by ECM proteins Cyr61 and CTGF expression were both up-regulated in MCF-7 cells cultured in VN-coated plates compared to those cultured on non-coated plates. To confirm that the up-regulation of Cyr61 and CTGF expression is mediated through a VN-cell interaction and not via a non-specific cell adhesion effect (as MCF-7 cells adhere poorly to non-coated vessels in SFM), MCF-7 cells were seeded onto plates coated with poly-l-lysine (Figure 4.4 A). In addition, to investigate if this was a VN-specific effect or wether additional ECM proteins could also increase Cyr61 and CTGF expression, cells were also seeded in wells coated with Fibronectin (FN) and Collagen type-I (Figures 4.4 B & C). Cells treated with only SFM and 10% FBS were used as controls.
Immunofluorescent staining of cells with anti-human Cyr61 and CTGF antibodies following a 48 hour treatment with VN and the trimeric complex resulted in substantial immunoreactivity compared to cells treated with poly-l-lysine (Figure 4.4 A). This was also confirmed using qRT-PCR and western blotting (Figure 4.4 B and C). Importantly, no differences were observed in Cyr61 and CTGF expression in cells grown in wells pre-bound with poly-l-Lysine compared to cells in SFM control treatment wells. This suggests that the increased expression of Cyr61 and CTGF is mediated by VN-engagement of cells via integrin receptors, and is not due to non- specific cell adhesion and spreading.
Analysis of cells treated with FN and collagen type-I also showed significant increases (p<0.01) in Cyr61 and CTGF mRNA expression compared to cells treated with SFM only (Figure 4.4 B). This was also confirmed at the protein level wherein cells treated with FN and collagen type-I substantially increased Cyr61 and CTGF protein levels compared to SFM-treated cells (Figure 4.4 C). Moreover, mRNA expression of Cyr61 in cells treated with VN was greater compared to cells treated with FN and collagen type-I (Figure 4.4 B). However, at the protein level no considerable differences were observed in Cyr61 expression in cells treated with VN, FN and collagen type-I (Figure 4.4 C). Moreover, VN and FN similarly induced the highest expression of CTGF among all treatments, followed by collagen type-I, suggesting the important role of ECM proteins in CTGF regulation.
Chapter 4: Investigation of the mechanisms underlying the regulation of Cyr61 and CTGF expression induced by VN and the IGF:IGFBP:VN complex. 128
Taken together, these results demonstrate that non-specific attachment of cells to tissue culture surface is not sufficient to induce Cyr61 and CTGF expression. These results also suggest that while VN is a potent stimulator of Cyr61and CTGF expression in MCF-7 cells is not a specific effect and they can also be regulated via additional ECM proteins via an integrin-mediated mechanisms.
Chapter 4: Investigation of the mechanisms underlying the regulation of Cyr61 and CTGF expression induced by VN and the IGF:IGFBP:VN complex. 129
A
B C
Figure 4.4: ECM protein-induced expression of Cyr61 and CTGF.
(A) Serum starved MCF-7 cells were seeded onto 96-well optical-bottom plates pre-bound with poly- l-lysine, VN and the IGFI:IGFBP-5:VN complex andwere fixed with 4% paraformaldehyde at 48 hours, stained with phalloidin for actin (green for Cyr61 and red for CTGF), DAPI for nucleus (blue) and antibodies against Cyr61 (red) and CTGF (green). Cells were imaged using DeltaVision microscopy. Scale bars indicate 10 μm. To examine the effect of other integrin-responsive ECM proteins, serum starved MCF-7 cells were seeded onto 6-well plates coated with either VN (1 μg/mL), FN (1 μg/mL) and collagen type I (1 μg/mL). Cells treated with SFM and 10% FBS were used as negative and positive controls, respectively. To investigate the effect of cell attachment of Cyr61 and CTGF expression, plates were also coated with poly-l-lysine (5 μg/mL) to provide attachment of cell in the absence of any ECM proteins. (B) RNA was harvested at 48 hours and mRNA expression of Cyr61 and CTGF was measured using qRT-PCR. Data were expressed as fold change of gene expression in control SFM. The asterisks indicate treatments with significant increases compared to the SFM control (**p<0.01). Error bars indicate SEM. (C) The protein expression of Cyr61 and CTGF was also determined using immunoblot analysis. The membranes were stripped and re-probed to determine the GAPDH protein levels to confirm equal loading in all wells. SFM = Serum Free Media and TRI = trimeric IGF-I:IGFBP:VN complex.
Chapter 4: Investigation of the mechanisms underlying the regulation of Cyr61 and CTGF expression induced by VN and the IGF:IGFBP:VN complex. 130
4.3.5 VN and the IGF-I:IGFBP:VN complex regulate expression of Cyr61 and CTGF via αv-subunit of containing integrin receptors Since VN is known to mediate the majority of its biological effects through the ligation of cell surface integrin receptors (Felding-Habermann & Cheresh, 1993), the involvement of specific integrin subunits in VN- and trimeric IGF-I:IGFBP:VN complex-mediated expression of Cyr61 and CTGF was investigated. Meyer and co- workers (1998) have shown that MCF-7 cells express the αv and β1 subunits (forming
αvβ1 integrin heterodimer) and the αvβ5 integrin (Meyer et al., 1998). Therefore, the contribution of these integrins, as well as the IGF-IR, in VN- and trimeric-induced expression of Cyr61 and CTGF was determined by pre-incubating MCF-7 cells with function blocking monoclonal antibodies specific for these integrins as well as the IGF-IR.
Treatment of cells with the antibody against the αv-integrin subunit significantly down-regulated the mRNA expression of Cyr61and CTGF in response to both VN and the trimeric complex when compared to the cells treated with the IgG isotype control antibody (Figure 4.5 A). Similarly, a dramatic reduction in the protein expression of Cyr61 and CTGF was observed in cells pre-incubated with the αv- integrin blocking antibody (Figure 4.5 B). Although, no appreciable changes were observed in Cyr61 expression at mRNA levels, immunoblot analyses demonstrated a slight reduction in VN- and trimeric complex-induced Cyr61 expression in cells pre- incubated with blocking antibodies against the αvβ5 heterodimer and the β1-integrin subunit, which effectively blocks the αvβ1 integrin heterodimer. This suggests the possibility that αvβ5 and the αvβ1 integrin expressed on the surface of MCF-7 cells may have redundant roles in mediating the expression of Cyr61 and CTGF in MCF-7 cells as the general αv-integrin blocking antibody (inhibiting both αvβ5 and αvβ1 integrins and possibly other αv-containing integrins) leads to a significant reduction in Cyr61 and CTGF expression (Figures 4.5 A & B). As expected from the previous study using the [L24][A31]-IGF-I analogue (Figure 4.3 D), no appreciable changes were observed in Cyr61 and CTGF expression at both the mRNA and protein levels in cells treated with the IGF-IR blocking antibody (Figures 4.5 A & B). From these data it is clear that the αv–integrin subunit is critical for VN- and IGF-I:IGFBP:VN complex-mediated regulation of Cyr61 and CTGF.
Chapter 4: Investigation of the mechanisms underlying the regulation of Cyr61 and CTGF expression induced by VN and the IGF:IGFBP:VN complex. 131
A
B
Figure 4.5: Involvement of VN-binding integrins and the IGF-IR in VN and IGF-I:IGFBP:VN- induced expression of Cyr61 and CTGF in MCF-7 cells.
VN- and TRI-mediated expression of Cyr61 and CTGF in MCF-7 cells was determined in the presence of monoclonal function-blocking antibodies against αv (1:10), αvβ5 (10 µg/mL) and β1 (10 µg/mL) integrins, IGF-IR (10 µg/mL) or control mouse isotype matched IgG antibody (10 µg/mL). Serum-starved MCF-7 cells were treated with blocking antibodies for 30 minutes at room temperature prior to seeding onto 6-well plates coated with VN or the IGF-I:IGFBP-5:VN complex. RNA and protein were harvested at 48 hours and the expression levels of Cyr61 and CTGF were measured using qRT-PCR (A) and immunoblot (B) analyses. The data are expressed as the fold change of gene expression in control wells containing mouse isotype matched IgG antibody. The asterisks indicate treatments with significant differences compared to the control IgG treatment (**p<0.01). Error bars indicate SEM. TRI = trimeric IGF-I:IGFBP:VN complex.
Chapter 4: Investigation of the mechanisms underlying the regulation of Cyr61 and CTGF expression induced by VN and the IGF:IGFBP:VN complex. 132
4.3.6 VN and the IGF-I:IGFBP:VN complex regulate Cyr61 and CTGF expression via Src/PI3-K/AKT and Src/MAPK/ERK signalling pathways, respectively To identify the downstream signalling pathways involved in the integrin- mediated regulation of Cyr61 and CTGF, pharmacological inhibitors of the FAK/Src kinase, PI3-K/AKT and MAPK/ERK signal transduction pathways were utilised. These pathways are commonly activated following integrin ligation and have reported roles in mediating the downstream biological effects of ECM proteins, including VN (Giancotti & Ruoslahti, 1999; Hapke et al., 2003; Playford & Schaller, 2004; Hurt et al., 2010). We have previously reported PI3-K/AKT and ERK/MAPK to be activated by VN and the trimeric IGF-I:IGFBP:VN complex (Hollier et al., 2008; Kashyap et al., 2011). Moreover, all three signal transduction pathways were identified to have central roles in gene networks associated with differential gene expression induced by VN and the trimeric complex (Chapter 3, Figures 3.9, 3.10 & 3.11). The relative contributions of these signalling pathways to the increased expression of Cyr61 and CTGF was determined using pharmacological inhibitors (LY294002, U0126 and PP2 directed against PI3-K, MEK/ERK and Src-family kinases, respectively).
Inhibition of the Src-family kinases via the PP2 inhibitor significantly down- regulated VN-induced and trimeric IGF-I:IGFBP:VN complex-induced expression of Cyr61 and CTGF mRNA levels as compared to DMSO vehicle control treated cells (Figure 4.6 A). Similar results were also observed at the protein level where western blot analysis of cells treated with PP2 demonstrated substantial reductions in Cyr61 and CTGF expression in response to VN and the trimeric complex (Figure 4.6 B). Treatment of cells with LY294002 directed against the PI3-K/AKT pathway led to decreased VN- and trimeric complex-stimulated Cyr61 expression at both the mRNA and protein level. Interestingly, CTGF mRNA and protein expression was not altered by inhibition of the PI3-K/AKT pathway (Figures 4.6 A & B). Conversely, inhibition of the ERK/MAPK pathway using the U0126 inhibitor resulted in reduced mRNA and protein levels of CTGF, but not Cyr61 as compared to DMSO vehicle control treated cells. This effect was common to both VN and the trimeric complex (Figures 4.6 A & B). Altogether, these results have revealed a novel bifurcation of signalling downstream of FAK/Src activation to mediate Cyr61 and CTGF expression via PI3- K/AKT and ERK/MAPK pathways, respectively.
Chapter 4: Investigation of the mechanisms underlying the regulation of Cyr61 and CTGF expression induced by VN and the IGF:IGFBP:VN complex. 133
A
B
Figure 4.6: Involvement of PI3-K, ERK and Src cell signalling mediators in VN- and IGF- I:IGFBP:VN-induced expression of Cyr61 and CTGF in MCF-7 cells.
VN and TRI regulation of Cyr61 and CTGF in MCF-7 cells was determined in the presence of pharmacological inhibitors LY294002 (10 µM), U0126 (10 µM) and PP2 (20 µM) against PI3- K/AKT, ERK/MAPK and Src intracellular signalling pathways, respectively. Cells treated with DMSO (20 µM) were used as the control. Serum-starved MCF-7 cells were treated with pharmacological inhibitors for 1 hour prior to seeding onto 6-well plates coated with VN or the IGF- I:IGFBP-5:VN complex. RNA and protein were harvested at 48 hours and the expression levels of Cyr61 and CTGF were measured using qRT-PCR (A) and immunoblot (B) analyses. The data are expressed as the fold-change of gene expression in control wells containing DMSO. The asterisks indicate treatments with significant differences compared to the DMSO control treatment (*p<0.05, **p<0.01). Error bars indicate SEM. TRI = trimeric IGF-I:IGFBP:VN complex.
Chapter 4: Investigation of the mechanisms underlying the regulation of Cyr61 and CTGF expression induced by VN and the IGF:IGFBP:VN complex. 134
4.3.7 Basal-like breast cancer cells have increased levels of Cyr61 and CTGF Breast cancer cell lines can be classified into luminal, basal A, basal B and claudin-low molecular subtypes. (Neve et al., 2006; Prat et al., 2010). The basal-like breast cancer subtype is associated with poor clinical outcomes, development of distant metastasis and a relatively higher rate of mortality (Banerjee et al., 2006; Fulford et al., 2007). In contrast, the luminal-like subtype is correlated with lower tumour grades, longer survival and a better prognosis (Sotiriou et al., 2003). The recently identified claudin-low subtype is known to be enriched in markers of EMT and cancer stem cell-like features (Prat et al., 2010). In addition to these characteristics, claudin-low tumours/cells are also triple negative (PR-, ER- & Her2-), making them highly associated with therapeutic resistance and poor survival outcomes (X. Li et al., 2008; Creighton et al., 2009; Gupta et al., 2009).
To investigate the broader relevance of VN-induced Gene Expression Signature VNGES across breast cancer subtypes, hierarchical clustering analysis was performed using the VNGES across 51 cell lines representing the major breast cancer subtypes using gene expression data published by Neve et al. (2006) (Figure 4.7). The results revealed the VNGES to be highly associated with Basal B cell lines, whereby the VNGES could cluster together the majority of the basal B cell lines apart from the luminal and basal A cell models. Further examination of these cell lines revealed them to be predominantly claudin-low cell lines, as recently classified by Prat and colleagues, 2010 (Figure 4.7). Interestingly, based on the VNGES expression, all 9 known claudin-low cell lines, as the originally published claudin- low predictor signature (Prat et al., 2010) were clustered together in their own distinct group. The results suggest that expression patterns observed in the VNGES are highly associated with expression patterns of these genes in claudin-low breast cancer subtypes.
In addition, using the Neve et al., (2006) data sets it was identified that Cyr61 and CTGF are up-regulated in basal-like (A and B) and claudin-low breast cancer cell sub-types (Figure 4.8). Interestingly, 6 out of 9 claudin-low cell lines were found to have high expression of Cyr61 and CTGF. Among these, MDA-MB-231, HS578T and SUM1315 cell lines were found to have the highest expression of Cyr61 and CTGF (Figure 4.8). These data suggest that Cyr61 and CTGF may play roles in
Chapter 4: Investigation of the mechanisms underlying the regulation of Cyr61 and CTGF expression induced by VN and the IGF:IGFBP:VN complex. 135
maintaining or stimulating the aggressive phenotype of basal-like breast cancers, in particular claudin-low subtypes.
Chapter 4: Investigation of the mechanisms underlying the regulation of Cyr61 and CTGF expression induced by VN and the IGF:IGFBP:VN complex. 136 Figure 4.7: Association of the VNGES across breast cancer molecular subtypes.
Gene expression data from the profiling of 51 breast cancer cell lines representing the major intrinsic subtypes reported by Neve et al. 2006 was used to investigate the expression of the VNGES across multiple representative cell models. The VNGES was observed to be associated with the Basal B and Claudin-low cell lines. The colour bars indicate the subtype classification for individual cell lines according to Neve et al. 2006 (bottom) and Prat et al. 2010 (top) (blue: luminal; red: basal A; orange: basal B and yellow: claudin-low). The heatmap was generated using the heatmap function in the “gplots” package of R program. The left and top dendrograms (hierarchical tree) are clustering of genes presented in VNGES and the similar genes in breast cancer cell lines, respectively based on similarity in expression levels. The clusters were generated using hierarchical cluster algorithms based on Ward's minimum variance method within the R package “gplots”. The expression of the gene is compared across all treatments, where low expression is indicated by green, while high expression is indicated in red. VNGES = VN-induced Gene Expression Signature.
Chapter 4: Investigation of the mechanisms underlying the regulation of Cyr61 and CTGF expression induced by VN and the IGF:IGFBP:VN complex. 137
CTGF
Cyr61
Figure 4.8: Differential expression of Cyr61 and CTGF across luminal, basal A, basal B and claudin-low breast cancer cell lines.
Gene expression data across 51 cell lines representing the major breast cancer subtypes used in studies performed by Neve et al. 2006 was used to investigate Cyr61 and CTGF gene expression pattern among luminal, basal A, basal B and cludin-low breast cancer cell types. Data are presented in a form of heat map using the heatmap function in the “gplots” package of R program. Target genes and cell lines were clustered (top dendrogram) based on similarity in expression using hierarchical cluster algorithms based on Ward's minimum variance method within the R package “gplots” Low expression is indicated by green, while high expression is indicated in red. Colour keys indicate the subtype classification of each cell line (blue: luminal; red: basal A; orange: basal B and yellow: claudin-low) as reported by Prat et al. 2010.
Chapter 4: Investigation of the mechanisms underlying the regulation of Cyr61 and CTGF expression induced by VN and the IGF:IGFBP:VN complex. 138 4.3.8 Screening of breast cancer cell lines for expression of Cyr61 and CTGF in response to VN and the IGF-I:IGFBP:VN complex In order to validate the data outlined above and to also confirm the increased expression of Cyr61 and CTGF by VN and the IGF-I:IGFBP:VN trimeric complex in a broader range of breast cancer cell types, additional cell lines representing the luminal (T47D), basal-like (MDA-MB-468) and claudin-low (MDA-MB-231) subtypes were used. Figure 4.9 A, represents a heatmap view of the cycle threshold (Ct) gene expression values of target genes measured by qRT-PCR analysis in cell lines treated with VN, the trimeric complex and SFM alone (control). We observed VN and the trimeric complex to stimulate expression of both Cyr61 and CTGF in all cell lines tested compared to their respective SFM control treatments, with exception to CTGF expression in MDA-MB-231 cells where no significant increase was observed in response to VN and the trimeric complex. This is most likely due to very high endogenous levels of CTGF in MDA-MB-231 cells. For quantitative comparisons, relative fold-changes in expression of Cyr61 and CTGF in response to VN and the trimeric complex were also normalised to their respective SFM treatment using the 2^(-ΔΔCt) method (Figure 4.9 B). Cyr61 and CTGF were found to be up- regulated in all cell lines (except CTGF expression in MDA-MB-231 cell line) treated with VN and the trimeric complex compared with cells treated with SFM only. Furthermore, the endogenous expression of Cyr61 and CTGF is generally higher in basal-like (MDA-MB-468) and claudin-low (MDA-MB-231) cell lines compared to luminal cells (MCF-7 and T47D). Since MDA-MB-231 cells are highly aggressive and metastatic cells with a claudin-low phenotype enriched in our VNGES with high levels of endogenous Cyr61 and CTGF, as compared to MCF-7 cells (Figure 4.9 C), they were selected for further studies as an additional cell model to investigate the role of these two proteins.
Chapter 4: Investigation of the mechanisms underlying the regulation of Cyr61 and CTGF expression induced by VN and the IGF:IGFBP:VN complex. 139
A
B
C
Figure 4.9: Cyr61 and CTGF regulation by VN and the IGF-I:IGFBP:VN complex in human breast cancer cell lines. Serum-starved MCF-7, T47D, MDA-MB-231 and MDA-MB-468 cell lines were seeded onto 6-well plates coated with VN and the IGF-I:IGFBP-5:VN complex. (A) RNA was harvested at 48 hours and the cycle threshold (Ct) values of target genes were measured using qRT-PCR analysis. Ct values were then visualised in a form of heatmap using the “gplots” package of R program. (B) The fold- change in expression of Cyr61 and CTGF in response to VN and the TRI complex were calculated using the formula 2^(-ΔΔCt) method. The data are expressed as the relative fold change in gene expression compared to the control (cells in SFM). (C) Regulation of Cyr61 and CTGF by VN and the TRI complex in MDA-MB-231 cells. The data are expressed as the relative fold change in gene expression compared to the control (MCF-7 cells in SFM). The data are pooled from three experiments with treatments tested in at least three wells in each replicate experiment. The asterisks indicate treatments with significant differences compared to the SFM control treatment (**p<0.01).The data are pooled from three experiments with treatments tested in triplicate wells in each replicate experiment. Error bars indicate SD. SFM = Serum Free Media, TRI = trimeric IGF- I:IGFBP:VN complex. Chapter 4: Investigation of the mechanisms underlying the regulation of Cyr61 and CTGF expression induced by VN and the IGF:IGFBP:VN complex. 140
4.3.9 VN and the IGF-I:IGFBP:VN complex regulate expression of Cyr61 and CTGF via αv-integrin subunit receptor in MDA-MB-231 cells
To confirm the key involvement of the αv-integrin in VN- and trimeric IGF- I:IGFBP:VN complex-mediated regulation of Cyr61 and CTGF observed in the MCF-7 cell line, we performed similar studies (as presented in section 4.3.5) using the MDA-MB-231 cell line. MDA-MB-231 are reported to express high levels of the
αv-integrin subunit (Meyer et al., 1998) and are well known for their increased expression of the “classical” VN-receptor, the αvβ3 integrin heterodimer (Menendez et al., 2005). Both αv and αvβ3 are reported to have functional roles in the aggressive behaviour of MDA-MB-231 cell model (Chen et al., 2001; Zhao et al., 2014). Hence the relative contribution of the αv-subunit and αvβ3 integrin, as well as the IGF-IR, for VN- and trimeric complex-induced expression of Cyr61 and CTGF were determined (as outlined in section 4.3.5).
As shown in Figure 4.10 A, inhibition of the αv-integrin subunit in MDA-MB- 231 cells with its respective function blocking antibody caused a significant down- regulation of Cyr61 and CTGF mRNA expression (p<0.01) in response to VN and the trimeric complex, as compared to the IgG control. This was also seen in immunoblot results where Cyr61 and CTGF protein levels were considerably reduced in cells following αv-integrin blockade (Figure 4.10 B). These results are in accordance with those we observed for MCF-7 cells (Figure 4.5). An antibody specific for inhibition of the αvβ3 integrin led to a significant (p<0.05) reduction in Cyr61 mRNA levels, however, no alteration to Cyr61 protein level was observed (Figures 4.10 A & B). Moreover, no alteration to CTGF expression in MDA-MB-231 cells treated with the anti-αvβ3 antibody was observed. Similarly, treatment of cells with antibody against IGF-IR did not cause any significant alternation in Cyr61 and CTGF expression. (Figure 4.10 A & B).
Chapter 4: Investigation of the mechanisms underlying the regulation of Cyr61 and CTGF expression induced by VN and the IGF:IGFBP:VN complex. 141
A
B
Figure 4.10: Involvement of VN-binding integrins and the IGF-IR in VN- and IGF-I:IGFBP:VN- induced expression of Cyr61 and CTGF in MDA-MB-231 cells.
VN- and IGF-I:IGFBP:VN-mediated expression of Cyr61 and CTGF in MDA-MB-231cells were determined in the presence of monoclonal blocking antibodies against αv (1:10) and αvβ3 (10 µg/mL) integrins, IGF-IR (10 µg/mL) and the αv + IGF-IR combination or control mouse isotype matched IgG antibody (10 µg/mL). Serum- starved MDA-MB-231 cells were treated with blocking antibodies for 1 hour prior to seeding onto 6-well plates coated with either VN alone or IGF-I+IGFBP-5 combination (TRI). RNA and protein were harvested at 48 hours and the expression levels of Cyr61 and CTGF were measured using qRT-PCR (A) and immunoblot (B) analyses. The data for qRT-PCR analysis are expressed as the fold change of gene expression in control wells containing mouse isotype matched IgG antibody. The asterisks indicate treatments with significant differences compared to the control IgG antibody treatment (*p<0.05, **p<0.01). Error bars indicate SEM. TRI = trimeric IGF- I:IGFBP:VN complex.
Chapter 4: Investigation of the mechanisms underlying the regulation of Cyr61 and CTGF expression induced by VN and the IGF:IGFBP:VN complex. 142
4.4 DISCUSSION
ECM-cell interactions recruit pathways that transmit signals to promote cell survival which is one of the key processes contributing to cancer development and progression (Jacks & Weinberg, 2002; Stupack & Cheresh, 2002). Compelling experimental evidences from our laboratory and others demonstrate that the ECM glycoprotein VN is able to promote cell survival in a variety of cells including breast cancer cells (Uhm et al., 1999; Kashyap et al., 2011; Hazawa et al., 2012). However, there exist no studies investigating the changes in gene expression that are associated with VN-stimulated breast cancer cell survival. Indeed, the data presented in the previous chapter (Chapter 3) identified a number of genes differentially regulated in MCF-7 breast cancer cells that have survived in response to VN. The results reported in Chapter 3 also indicated a number of genes differentially expressed in MCF-7 cells in response to the trimeric IGF-I:IGFBP:VN complex. This complex was shown to provide greater cell viability responses in breast cancer cells than those obtained with VN alone. However, the majority of genes differentially expressed were found to be in common between VN and the IGF-I:IGFBP:VN treatments. This suggests that VN is the key component responsible for inducing the majority of differential gene expression stimulated by the IGF-I:IGFBP:VN complex. Hence, the studies reported in this Chapter focussed on assessing the genes commonly regulated by VN and the IGF-I:IGFBP:VN complex for further functional studies. This strategy aimed at selecting candidate genes to target not only VN but potentially IGF:VN complex- mediated actions
Comparison of genes differentially regulated by VN and the IGF-I:IGFBP:VN complex indicated a considerable overlap between both treatments. Specifically, 447 genes were regulated in common by VN and the IGF-I:IGFBP:VN complex as opposed to those treated with SFM only (Figure 4.2.1). Importantly, this VN-induced gene expression signature (VNGES) was found to be enriched in genes with biological roles important in cell survival, including the “cell growth and proliferation” and the “cell death and survival”. This suggests that the differential expression of genes involved in these cell survival related biological functions may be responsible for the observed VN- and IGF-I:IGFBP:VN-induced increases in cell survival responses. Interestingly, the gene expression patterns observed in the VNGES are highly associated with expression patterns of these genes in basal-like
Chapter 4: Investigation of the mechanisms underlying the regulation of Cyr61 and CTGF expression induced by VN and the IGF:IGFBP:VN complex. 143
and claudin-low breast cancer subtypes. This suggests that the gene expression profile induced by VN may contribute to these phenotypes in certain breast cancer subtypes. Furthermore, to identify the genes that may be involved in cancer cell survival, genes with reported roles in both “cell growth and proliferation” and the “cell death and survival” biological functions (80 genes) were further sub- categorised (Figure 4.2). A number of candidates within these 80 genes, including TGFβ1/2, S100A4, Smad6, S100P, ANXA2, RAPGAP, Cyr61/CCN1, CTGF/CCN2, Slug, Caveolin-1, mir-21, c-JUN and c-FOS were deemed to be biologically interesting due to their importance in regulating cell survival while their deregulation has been reported to correlate with breast cancer progression. Among these genes, multifunctional secreted ECM-associated CCN family proteins, Cysteine-rich 61 (Cyr61/CCN1) and Connective Tissue Growth Factor (CTGF/CCN2) were selected as top candidates in this study to investigate the mechanisms underpinning their regulation by VN and the IGF:VN complex and further functionally validate their role in VN- and IGF:IGFBP:VN-induced breast cancer cell survival. Indeed, qRT- PCR and western blot analyses confirmed that breast cancer cells treated with VN and the IGF-I:IGFBP:VN complex dramatically induced the mRNA and protein expression of Cyr61 and CTGF (Figure 4.3).
The CCN family, in addition to Cyr61/CCN1 and CTGF/CCN2, consists of the nephroblastoma overexpressed (NOV/CCN3), Wnt-induced secreted proteins-1 (WISP-1/CCN4), -2 (WISP-2/CCN5) and -3 (WISP-3/CCN6) proteins. These proteins regulate key biological functions, including ECM production (Frazier et al., 1996; Quan et al., 2006), stimulation of cell adhesion (Babic et al., 1999), survival (Menendez et al., 2005; Wang et al., 2009), differentiation (Zuo et al., 2010), angiogenesis (Brigstock, 2002), and tumorigenesis (Xie et al., 2004) as well as wound healing (Chen et al., 2001) in a variety of cell types. CCN proteins are a subset of ECM proteins known as ‘matricellular proteins’ which are dynamically expressed and serve as regulatory proteins more so than structural proteins (Bornstein & Sage, 2002; Chen & Lau, 2009).
Once secreted into the extracellular microenvironment, Cyr61 and CTGF promote adhesion, predominantly through their interactions with integrins (Leu et al., 2002; Gao & Brigstock, 2004) and heparan-sulfate containing proteoglycans (HSPGs), including syndecan 4 and perlecan (Grzeszkiewicz et al., 2002; Y. Chen et
Chapter 4: Investigation of the mechanisms underlying the regulation of Cyr61 and CTGF expression induced by VN and the IGF:IGFBP:VN complex. 144
al., 2004). It is also known that CTGF can bind to growth factor receptors, including fibroblast growth factor receptor (FGFR) (Aoyama et al., 2012) and vascular endothelial growth factor receptor (VEGFR) (Inoki et al., 2002) to possibly enhance its biological activities through the activation of signalling pathways downstream of these receptors. For example, CTGF binding to FGFRs is reported to affect the phosphorylation of ERK and the differentiation of osteoblastic cells (Aoyama et al., 2012). Interestingly, growth factors such as TGFβ and VEGF are reported to induce expression of both Cyr61 (Parisi et al., 2006; Athanasopoulos et al., 2007) and CTGF (Suzuma et al., 2000; Leivonen et al., 2005). In particular, TGFβ-induced expression of Cyr61 and CTGF through its transcriptional mediators Smad family proteins (mainly Smad 2, 3 and or 4) is known to be a key mechanism underlying the initiation of bone metastasis in breast cancer (Kang et al., 2003; Bartholin et al., 2007). Additionally CTGF directly binds TGFβ to activate its downstream signals by enhancement of receptor binding and indirectly potentiating the phosphorylation of Smad-responsive promoters induced by TGFβ1 (Abreu et al., 2002). Interestingly, in the study reported herein TGFβ2 was significantly up-regulated in excess of 10-fold in response to VN and the IGF-I:IGFBP:VN complex (Figure 3.13 A). This suggests the possibility that TGFβ may act in a concerted fashion with Cyr61 and CTGF to drive biological functions such as cell migration and survival in response to VN and the IGF:VN complexes.
In addition to their ability to bind integrins, HSPGs and growth factor receptors, Cyr61 and CTGF can interact with, and bind to VN (Francischetti et al., 2010) and FN (Hoshijima et al., 2006) with high affinity. Thereby, potentially functioning as a local scaffold that coordinates the interaction of ECM proteins with the tumour cells (Jun & Lau, 2011). Furthermore, Cyr61 and CTGF expression is reported to stimulate the production of ECM proteins such as FN and collagen type-I (Frazier et al., 1996; Quan et al., 2006). No studies to date have reported the involvement of ECM proteins in the regulation of Cyr61 and CTGF expression.
Overexpression of Cyr61 and CTGF has been reported to contribute to breast cancer progression and metastasis and is often associated with the more advanced clinical stage of the disease (Tsai et al., 2000; Xie, Nakachi, et al., 2001; Kang et al., 2003). Overexpression of Cyr61 in MCF-7 breast cancer cells was found to be sufficient to promote growth and colony formation in vitro and to induce
Chapter 4: Investigation of the mechanisms underlying the regulation of Cyr61 and CTGF expression induced by VN and the IGF:IGFBP:VN complex. 145
tumorigenesis in vivo (Tsai et al., 2002). Induced expression of Cyr61 in MCF-7 cells was also found to stimulate cell proliferation and cell survival, and significantly hinder chemotherapy drug-induced apoptosis, with more than 10-fold resistance to Taxol, doxorubicin, paclitaxel and β-Lapachone being observed (Lin et al., 2004; Menendez et al., 2005). In addition, blocking the effects of Cyr61 using a function- blocking monoclonal antibody significantly reduced cell proliferation, migration and invasion in vitro in the highly metastatic breast cancer cell line MDA-MB-231, and also inhibited tumour growth and node metastasis in vivo (Lin et al., 2012).
In a similar fashion, up-regulation of CTGF is found to be critically involved in bone metastasis in breast cancer (Kang et al., 2003; Shimo et al., 2006). Overexpression of CTGF has been shown to enhance the migratory ability of both weakly metastatic MCF-7 and highly metastatic MDA-MB-231 breast cancer cell lines (Chen et al., 2007). Moreover, CTGF expression enhances MCF-7 and MDA- MB-231 cell viability and confers resistance to apoptosis on exposure to doxorubicin and paclitaxel (Wang et al., 2009). Furthermore, TAZ (transcriptional co-activator with PDZ-binding motif), a protein involved in drug resistance, has been shown to induce Taxol resistance in breast cancer cells through the activation of Cyr61 and CTGF promoters and stimulation of their gene and protein expression (Lai et al., 2011). Cyr61 and CTGF have also been found to activate critical intracellular cascades, specifically the cell survival pathways, including FAK-Src, PI3-K/AKT, ERK/MAPK and NF-κB pathways (Chen & Lau, 2009; Lau, 2011). These findings suggest the possibility that up-regulation of Cyr61 and CTGF by VN and the IGF- I:IGFBP:VN complex may play an important role in cell survival, cell migration and drug-induced apoptosis in breast cancer cells exposed to VN and the IGF-I:VN complex. These phenotypes have been reported previously and are in alignment with the previous findings by our group (Hollier et al., 2008; Kashyap et al., 2011).
Taken together, considering the up-regulation of Cyr61 and CTGF by VN and the IGF-I:IGFBP:VN complex and their reported overexpression in mediating breast cancer cell survival, drug-induced apoptosis and migration, these genes were selected as our top candidates to investigate the mechanisms underlying their expression. More importantly, these genes were functionally validated for their roles in VN- and IGF-I:IGFBP:VN-induced breast cancer cell survival. Interestingly, Cyr61 and CTGF have similar biological activities to VN in many aspects, including interaction
Chapter 4: Investigation of the mechanisms underlying the regulation of Cyr61 and CTGF expression induced by VN and the IGF:IGFBP:VN complex. 146
with integrin receptors, supporting cell attachment and inducing adhesive signalling as well as pro-angiogenic and survival properties (Brigstock, 2002) in addition to their interactions with growth factors.
To investigate the broader relevance of Cyr61 and CTGF across breast cancer subtypes, hierarchical clustering was used to analyse the expression of Cyr61 and CTGF among 51 cell lines representing the major breast cancer subtypes using gene expression data published by Neve et al. (2006) and Part et al. (2010). The results revealed that the claudin-low and basal B-like breast cancer cell lines generally have higher Cyr61 and CTGF endogenous levels compared to luminal- and basal A-like breast cancer cell subtype models (Figure 4.7 B). Since claudin-low and basal B breast tumours are metastatic and associated with an aggressive clinical behaviour (Banerjee et al., 2006; Fulford et al., 2007; Taube et al., 2010), it is possible that Cyr61 and CTGF expression may correlate with increasing aggressiveness of metastatic breast cancer cells. Importantly, VN and the IGF-I:IGFBP:VN complex were found to increase the expression of Cyr61 and CTGF in all cell lines tested ranging from luminal-like MCF-7 and T47D cell lines to basal-like MDA-MB-468 and claudin-low MDA-MB-231 cell lines (Figure 4.8). This suggests that VN and the IGF-I:IGFBP:VN complex are not only capable of stimulating Cyr61 and CTGF expression in luminal-like breast cancer cells with low endogenous expression levels, but also in cells highly expressing these proteins. Thus, we further extended our investigation into the mechanisms underlying VN- and IGF-:IGFBP:VN-induced Cyr61 and CTGF expression in the highly aggressive claudin-low MDA-MB-231 breast cancer cell line, in addition to weakly-metastatic MCF-7 cells.
The studies reported in this chapter demonstrated that Cyr61 and CTGF expression is specific to VN-engagement of cell adhesion and not due to non-specific cell attachment and spreading (Figure 4.4) Furthermore, it was also found that in addition to VN other integrin-associated ECM constituents such as FN and collagen type I are capable of stimulating Cyr61 and CTGF expression in MCF-7 cells (Figure 4.4). This suggests the possibility that integrin-mediated cell:ECM interactions are a key mechanism underlying VN- and IGF-I:IGFBP:VN-induced Cyr61 and CTGF expression.
ECM:cell interactions in breast cancer cells are known to be predominantly facilitated via the αv and β1 integrin subunits (Wong et al., 1998; Park et al., 2008) as
Chapter 4: Investigation of the mechanisms underlying the regulation of Cyr61 and CTGF expression induced by VN and the IGF:IGFBP:VN complex. 147
well as the αvβ3 and αvβ5 heterodimeric integrins (Maeshima et al., 2001; Bartsch et al., 2003). MCF-7 cells with high expression of αv, β1 and αvβ5 integrins and MDA-
MB-231 cells with high expression of αvβ3 (Meyer et al., 1998; Bartucci et al., 2001) were used to investigate the relative involvement of these receptors in VN- and IGF- I:IGFBP:VN-induced expression of Cyr61 and CTGF. Incidentally, these cell line models also have a high endogenous expression of the IGF-IR making these ideal cell models to investigate both integrin- as well as IGF-IR-mediated cell signalling events. Blocking the αv-integrin subunit caused substantial reduction in VN- as well as IGF-I:IGFBP:VN-induced expression of both Cyr61 and CTGF (Figures 4.5 & 4.10). This suggests that the VN- and IGF-I:IGFBP:VN-induced expression of Cyr61 and CTGF is extensively dependent on the αv-integrin subunit. This can also be seen in studies demonstrating the blockade of VN ligation to αv integrin reduces VN- mediated activation of p38 MAPK and uPA in MDA-MB-231 cells (Han et al., 2002) which possibly effect the regulation of genes downstream these signals.
In order to investigate the specific heterodimeric integrin involved, the αvβ5 heterodimer was inhibited using an anti-αvβ5 antibody. Although, no appreciable changes were observed in Cyr61 expression at mRNA levels, only a slight reduction was observed in the VN- and IGF-I:IGFBP:VN-induced Cyr61 expression at the protein level (Figure 4.5). Similarly, only a slight reduction in Cyr61 protein expression was observed in MCF-7 cells treated with the anti-β1 integrin antibody which effectively inhibit αvβ1 integrin (Figure 4.5). This suggests a redundancy between the αv-containing integrins, whereby blocking specific αv-containing integrin heterodimeric integrins gets compensated for the other one. Thus, blocking all αv-containing integrins with αv blocking antibody is required for the maximal inhibition of VN and the IGF-I:IGFBP:VN-induced expression of Cyr61. Moreover, inhibition of the “classical” VN-receptor αvβ3 heterodimer integrins in MDA-MB- 231 cells did not alter Cyr61 and CTGF expression (Figure 4.10). Such additional function blocking studies are required to investigate the involvement of other heterodimer receptors such as α2β1, α3β1 α5β1 which are also highly expressed in MDA-MB-231 cells (Morini et al., 2000). Furthermore, No appreciable changes were observed in Cyr61 and CTGF expression in cells treated with antibodies directed against IGF-IR compared to cells treated with IgG control (Figure 4.5 & 4.10 B).
Chapter 4: Investigation of the mechanisms underlying the regulation of Cyr61 and CTGF expression induced by VN and the IGF:IGFBP:VN complex. 148
Upon ECM ligation to integrins FAK is phosphorylated and binds to the N- terminus of Src homology 2 (SH2) domain forming the FAK-Src complexes which subsequently activate the Src kinase signalling mediator and its downstream intermediates such as PI3-K, AKT, ERK and MAPK. (Stupack & Cheresh, 2002; Guo & Giancotti, 2004; Mitra & Schlaepfer, 2006). In view of this, phosphorylation and recruitment of the FAK-Src signalling downstream of VN/αvβ integrin engagement is reported to be critical for differential expression of genes known to drive breast and prostate cancer stem cell differentiation such as, K14, K8, Nanog, SMO, Oct3/4, BMI and cytokeratins (Hurt et al., 2010). In a similar fashion, VN has been shown to induce changes in gene expression promoting cell proliferation and migration in ovarian cancer cells in a αvβ3/FAK/Src/MAPK pathway-dependent mechanism (Hapke et al., 2003). Considering the important role of FAK-Src signalling in regulation of gene expression induced by VN-binding integrins, the studies reported herein investigated the involvement of this Src protein in VN- and IGF-I:IGFBP:VN-induced expression of Cyr61 and CTGF. However, further studies are still required to exclusively investigate the involvement of FAK in this process. In addition, PI3-K/AKT and ERK/MAPK, two of the known signalling pathways downstream of FAK-Src signal, were also investigated. These pathways are known to be critical for a range of cellular functions, in particular cell survival and migration (Datta et al., 1997; Bouchard et al., 2007). More importantly, these pathways were found to map within the centre of the top-ranking networks identified in this project associated with differentially expressed genes in response to VN and the IGF- I:IGFBP:VN complex (Figures 3.9, 3.10 & 3.11).
Experiments using pharmacological inhibitors of these pathways in MCF-7 cells revealed a critical role for the recruitment and activation (via phosphorylation) of FAK-Src in stimulating the expression of both Cyr61 and CTGF by VN and the IGF-I:IGFBP:VN complex (Figure 4.6). Cells treated with pharmacological inhibitor against PI3-K demonstrated a considerable decrease in Cyr61 expression at both the mRNA and protein level whereas CTGF expression was substantially reduced in cells treated with an ERK inhibitor. Taken together, these results propose a mechanism in which VN and the IGF-I:IGFBP:VN complex regulate Cyr61 and
CTGF expression through interactions with αv-subunit containing integrins and
Chapter 4: Investigation of the mechanisms underlying the regulation of Cyr61 and CTGF expression induced by VN and the IGF:IGFBP:VN complex. 149
recruitment of FAK-Src signalling, which then bifurcates downstream to regulate Cyr61 via PI3-K and CTGF via ERK pathways.
In summary, the studies reported in this chapter have revealed that VN and the IGF-I:IGFBP:VN complex can induce expression of Cyr61 and CTGF in multiple breast cancer cell lines ranging from luminal to the basal/claudin-low subtypes. The studies reported throughout this Chapter also highlight that the regulation of CCN family members is through the engagement of cells with the components of the ECM and not due to the non-specific cell attachment and spreading to the culture plastic. Furthermore, it was found that VN-induced expression of Cyr61 and CTGF is mediated through VN:αv engagement and recruitment of Src/PI3-K and Src/ERK pathways, respectively. Indeed, these findings add new valuable knowledge to our current understanding of the mechanisms underlying VN and the IGF:VN complex- mediated gene regulation. However, the functional role of these molecular targets still needs to be investigated. Considering the important role of Cyr61 and CTGF in mediating various biological functions it is plausible that VN, and potentially other ECM elements such as FN and collagen, stimulate the expression as well as the secretion of Cyr61 and CTGF to sustain a positive feedback loop in an autocrine fashion to regulate biological functions critical to cancer metastasis.
4.5 CONCLUSION
The studies reported in this chapter identified a number of genes differentially regulated in common by VN and the IGF-I:IGFBP:VN complex which were significantly associated with biological processes related to cell survival. Among these genes, the ECM-associated Cyr61 and CTGF proteins (found to be up- regulated), with previously reported roles in breast cancer cell survival, cell invasion, and drug-induced apoptosis, were selected as the top candidates to investigate their functional roles in VN- and IGF-I:IGFBP:VN-stimulated breast cancer cell survival. The studies reported throughout this chapter for the first time support a new mechanism, in which VN and the IGF-I:IGFBP:VN complex stimulate Cyr61 and
CTGF expression through αv-integrin receptors and the recruitment of the Src signalling mediator after which the signal bifurcates to uniquely regulate Cyr61 through PI3-K and CTGF through ERK. The involvement of Cyr61 and CTGF in
Chapter 4: Investigation of the mechanisms underlying the regulation of Cyr61 and CTGF expression induced by VN and the IGF:IGFBP:VN complex. 150
mediating VN- and IGF-I:IGFBP:VN-stimulated breast cancer cell survival is unknown and is the objective of the following chapter (Chapter 5).
Chapter 4: Investigation of the mechanisms underlying the regulation of Cyr61 and CTGF expression induced by VN and the IGF:IGFBP:VN complex. 151
Chapter 5: Functional validation of Cyr61 and CTGF in VN- and IGF-I:IGFBP:VN- stimulated breast cancer cell survival
Chapter 5: Functional validation of Cyr61 and CTGF in VN- and IGF-I:IGFBP:VN-stimulated breast cancer cell survival 153
5.1 INTRODUCTION
Using microarray analysis, we have identified for the first time a VN-induced gene expression signature (VNGES) in breast cancer cells that have survived in response to VN. More importantly, gene ontology analysis has elucidated a number of genes within the VNGES with important roles in cell survival. Furthermore, the studies performed within the previous chapter (Chapter 4) demonstrated that the ECM-associated matricellular proteins Cyr61 and CTGF with previously reported roles in breast cancer progression are highly up-regulated at both mRNA and protein levels in cells treated with both VN and the IGF-I:IGFBP:VN complex. In addition, cell signalling studies demonstrated that VN and the IGF-I:IGFBP:VN complex induced the expression of Cyr61 and CTGF through the recruitment of αv-integrin and the activation of the Src signalling axis and recruitment of its bifurcated downstream pathways of PI3-K/AKT and ERK/MAPK, respectively. Cyr61 and CTGF both have documented roles contributing to the enhancement of a number of key biological functions, including growth, survival, migration invasion and angiogenesis (reviewd in Brigstock, 2002, 2003; Perbal & Takigawa, 2005; Leask & Abraham, 2006), to contribute to breast cancer progression and have been associated ate be often associated with more advanced stages of disease (Xie, Nakachi, et al., 2001; Tsai et al., 2002; Kang et al., 2003). However, the functional role and the extent of involvement of these genes in mediating VN- and IGF-I:IGFBP:VN complex-induced effects, namely cell survival and cell migration remain to be validated. In view of this, the studies reported in this chapter are directed at addressing this knowledge gap.
Briefly, the gene knockdown studies performed within this chapter provide the first evidence highlighting the importance of CTGF in VN- and IGF-I:IGFBP:VN- induced cell survival of the poorly metastatic MCF-7 cell lines in both 2D and 3D culture models. In contrast, the highly invasive and metastatic MDA-MB-231 cells genetically engineered to have reduced expression of CTGF did not change their cell viability in 3D cultures. However, these cells were observed to be less-invasive in 3D Matrigel™ cultures and adopt a more rounded morphology, reminiscent of MCF-7 cell colonies. Cyr61 was also demonstrated to be important for VN- and IGF- I:IGFBP:VN -stimulated cell survival of MCF-7 and MDA-MB-231 cells. Taken together, results from this chapter provide the first functional evidence for the role of
Chapter 5: Functional validation of Cyr61 and CTGF in VN- and IGF-I:IGFBP:VN-stimulated breast cancer cell survival 154
Cyr61 and CTGF in VN- and IGF-I:IGFBP:VN complex-mediated biological effects, including the growth, survival and invasion of breast cancer cells.
5.2 EXPERIMENTAL METHODS
The following is a brief summary of the materials and procedures used to generate the data presented in this chapter. Full details of materials and methods used herein are described in Chapter 2.
5.2.1 Cell lines MCF-7 and MDA-MB-231 cells were utilised throughout this chapter. Cell lines were maintained following procedures described in section 2.3. The highly metastatic MDA-MB-231 breast cancer cell line was chosen to screen and select the best shRNA owing to the high expression of the target genes detected within these cells. The functional studies reported in this chapter were performed using both MCF-7 and MDA-MB-231 cells. Validation experiments assessing the effect of shRNA knockdown on cell survival and migration, were cross-validated with the parental (MOCK, non-transfected) as well as the non-targeting shFF3-containing control cells (shCntrl).
5.2.2 Lentiviral shRNA mediated knockdown of target genes To assess the putative functional role of candidate genes; namely Cyr61 and CTGF, lentiviral-mediated transduction of gene specific shRNA sequences was performed to stably suppress target gene expression (knockdown). For details on the lentiviral vector system used refer to section 2.15.1. For all the cloning protocols applied to produce these vectors refer to section 2.14.
Virus production in HEK293T packaging cells The pLKO.1 vector containing the target shRNA of interest and the viral packaging plasmids, pCMV-VSV-G and pCMV delta R8.2, were transiently transfected into HEK293T cells to produce lentiviral particles as per the procedures described in section 2.15.2. Virus-rich supernatants were harvested at 48 and 72 hours post-transfection. The viral supernatant was clarified by centrifugation at 1000 rpm for 5 minutes, filtered through a 0.45 μm cut-off filter, pooled together and stored at -80 °C until use. Prior to harvesting viral supernatants, HEK293T cells
Chapter 5: Functional validation of Cyr61 and CTGF in VN- and IGF-I:IGFBP:VN-stimulated breast cancer cell survival 155
transfected with the GFP-containing control vector pLKO.3G were visually inspected for green fluorescence under a fluorescence microscope to confirm HEK293T transfection efficiency.
Transduction of target cell lines For full details on the transduction protocols refer to section 2.15.3. Target cells at 30-40% confluency were transduced with 2 mL of the lentivirus-containing supernatant medium obtained above. Protamine sulphate (6 μg/mL) was added to facilitate virus uptake. Following overnight incubation, the viral supernatant was removed and cells were allowed to recover for 24 hours in regular growth medium. The following day, transduced cells were subjected to selection with 2 μg/mL puromycin. This high concentration of puromycin was maintained until all the untransfected negative control cells died, after which the transduced cells were maintained in 1 μg/mL puromycin-containing growth medium. Cells transduced with the pLKO.3G-lentiviral particles were visually inspected for green fluorescence to confirm and judge the transduction efficiency of target cells. Positively selected target cells were observed for any changes in morphology compared to the shFF3- containing control cells (shCntrl). RNA and proteins were isolated to analyse shRNA-mediated knockdown of the gene of interest through qRT-PCR and immunoblotting, respectively.
5.2.3 qRT-PCR to assess target gene expression RNA was extracted from cell lysates using the Direct-zol™ MiniPrep Kit (Zymo Research) following the manufacture’s protocol (for full details on this method please refer to section 2.8.2.). RNA samples were then reverse transcribed into cDNA using the SuperScript® III First-Strand Synthesis SuperMix kit (Life Technologies) following the manufacture’s protocol as briefly outlined in section 2.11.2.
5.2.4 Immunoblotting to assess protein expression To assess down-regulation of protein expression MCF-7 and MDA-MB-231 cells containing either the control or target gene shRNA were seeded in 10 cm culture dishes, washed with PBS, and lysed directly by adding 600 μL of Radio Immuno-Precipitation Assay (RIPA) buffer containing (Sigma-Aldrich) phosphatase (Thermo Fisher Scientific) (1:10) and protease (Roche Diagnostics) (1:10) inhibitors
Chapter 5: Functional validation of Cyr61 and CTGF in VN- and IGF-I:IGFBP:VN-stimulated breast cancer cell survival 156
to each dish. The plates were then incubated on ice for 5 minutes and the cells were dislodged into suspension using a rubber policeman. For full details on this method refer to section 2.12.1.
The protein concentrations were determined using Coomassie Plus® Protein Quantitation kit. Proteins (20-60 μg per sample) were then resolved on pre-made NuPAGE® gradient SDS-PAGE gels (4%-12%) and transferred onto nitrocellulose membranes. Western immunoblotting was performed following procedures outlined in section 2.12.2. The primary antibodies used in this chapter were anti-Cyr61 (1:1000) and anti-CTGF (1:1000) (R&D system). To detect the total levels of these proteins, the same membranes were stripped using Restore® Western blot stripping buffer and probed with anti-GAPDH to validate equal loading of proteins onto the gel.
5.2.5 MTS assay for cell viability measurements For full details on this method refer to section 2.5. Briefly, MCF-7 and MDA- MB-231 cells containing either non-targeting control shRNA or shRNA specific for the target gene were serum-starved for 4 hours before seeding 5 x 103 cells suspended in 100 μL serum-free DMEM/F12 medium (SFM) + 0.05% BSA into 96- well plates pre-coated with VN and IGF-I:IGFBP-5:VN combinations (refer to section 2.4.1 for proteins pre-bound and concentrations used). Treatments with medium containing 10% FBS and only SFM served as positive and negative controls, respectively. After 72 hours, 20 µL MTS reagent was added into each well and the absorbance was recorded after 2 hours and 30 minutes at 490 nm using a 96-well plate reader (Benchmark Plus, Bio-Rad).
5.2.6 CyQUANT for cell proliferation assay Similar to MTS assay, for CyQUANT assay, serum-starved MCF-7 cells seeded onto 96-well plates per-coated with VN either alone or in combination with
IGF-I+IGFBP-5 and allowed to grow for 72 hours at 37 °C, 5% CO2. After incubation duration, media was aspirated from wells and plated were wrapped in parafilm and stored at -80 °C overnight. The following day plates were thawed at room temperature and 100 µL of lysis buffer and CyQuant GR fluorescent dye mixture [provided by the CyQUANT cell proliferation assay kit (Life Technologies)] was added to each well and mix thoroughly. Plates were incubated at room
Chapter 5: Functional validation of Cyr61 and CTGF in VN- and IGF-I:IGFBP:VN-stimulated breast cancer cell survival 157
temperature in the dark for 5 minutes and the absorbance was recorded after 5 minutes with excitation at 485 nm and emission detection at 530 nm at using a 96- well plate reader (Benchmark Plus, Bio-Rad). To convert sample fluorescence values into cell numbers, 1 x 106 MCF-7 cells were collected as a cell pellet, supernatant removed and the pellet stored at -80 °C overnight. The following day, cell pellet was lysed by adding 1 mL of the CyQUANT GR dye/cell-lysis buffer and serial dilution generated corresponding to cell numbers ranging from 50 to 50,000 cells per 100 µL adding to each well. Wells with no cells containing 100 µL of the CyQUANT GR dye/cell-lysis buffer, also used as a control. Fluorescent intensity was measured and standard curve was obtained using linear regression analysis.
5.2.7 Transwell® migration assays For full details on this method refer to section 2.6.
5.2.8 3D Matrigel™ assay For full details on this method refer to section 2.7. Briefly, assays were performed in 96-well plates. MCF-7 and MDA-MB-231 cells containing either non- targeting control shRNA or shRNA specific for the target gene were suspended in 3D growth medium (DMEM/F12 for MCF-7 cells and RPMI 1640 for MDA-MB-231 cells; 2.5% Matrigel™ + 1% FBS) and were seeded onto a 100% Matrigel™ layer. Cells were incubated for three days without medium change. On Day 3 the spent medium from the treatment-wells was removed and replaced with solutions of 3D growth medium containing VN (1 μg/mL) alone or complexed with IGFBP-5 (300 ng/mL), and IGF-I (100 ng/mL). To facilitate protein complexes being formed, these protein + 3D growth medium solutions were incubated for 2 hours at 37 °C, 5% CO2 prior to adding them into the respective wells. Solutions containing 3D growth medium with 1% serum served as the negative control. Cells were incubated for further 6 days (for MDA-MB-231 cells) or 9 days (for MCF-7 cells), with spent medium being replaced every three days with fresh 3D growth medium containing the respective protein solutions.
Analysis of 3D cell survival using AlamarBlue assay For full details on this method refer to section 2.7.1.
Chapter 5: Functional validation of Cyr61 and CTGF in VN- and IGF-I:IGFBP:VN-stimulated breast cancer cell survival 158
ImageJ analysis for spheroid size calculations For full details on this method refer to section 2.7.3. The 3D cultures were fluorescently stained, images recorded and subjected to analysis via ImageJ. The threshold value above/below which the spheroids were considered viable was set to include small colonies but to omit debris accumulated within the 3D culture (these appear as small dots within the image). For MCF-7 which form round colonies the size (area of pixels occupied in μm2) of the spheroids detected within the image was calculated and results were exported to Microsoft Excel for data analysis. Treatments were tested in duplicate in each experiment. When cultured as 3D, the MCF-7 cells grow as a mass resembling an irregular “spheroid” (Kenny et al., 2007), closely representing an in vivo primary tumour (Chishima et al., 1997; Ravizzini et al., 2009). As opposed to the spheroid-like colonies of MCF-7 cells, the highly metastatic cells have a spindle-like morphology and invade the 3D culture in different directions resembling invasive tumours in vivo (Benton et al., 2014). Thus, the studies reported herein used ImageJ to measure the size of spheroids in MCF-7 cells to relatively quantify the colony formation ability in this cell line. ImageJ was also used to measure the relative circularity in MDA-MB-231 cells to quantify the change in cell appearance in response to various treatments.
5.3 RESULTS
5.3.1 Plasmid/viral delivery to packaging HEK 293T and target MDA-MB-231 cells Lentiviral transduction was the method chosen to deliver shRNA against Cyr61 and CTGF into the target cells. Lentiviruses were produced within HEK293T packaging cells (refer to section 2.4.2) which had the pLKO vector containing either shCyr61 or shCTGF. For use as controls, lentiviruses were also produced to contain an shRNA sequence targeting firefly luciferase (pLKO.shFF3). To enable coincident assessment of transfection efficiency, cells were transfected in parallel with GFP- containing vector, pLKO.3G. Using the pLKO.3G-transfected HEK293T cells, visual inspection via fluorescence microscopy revealed that the majority of cells to be positive for GFP. (Figure 5.1).
Chapter 5: Functional validation of Cyr61 and CTGF in VN- and IGF-I:IGFBP:VN-stimulated breast cancer cell survival 159
Once it was confirmed that the pLKO vectors were transfected efficiently into HEK293T cells, the transduction efficiency of the viral particles was confirmed using the target MDA-MB231 cell line. Similarly, it was observed that the majority of MDA-MB-231 cells were effectively transduced with pLKO.3G lentiviral particles as observed by GFP fluorescence (Figure 5.1). This confirmed that our pLKO lentiviral system for the delivery of shRNA into human cells was working effectively. In addition, the majority of the pLKO.1 infected target cells remained viable upon selection with the antibiotic puromycin (2 μg/mL), indicating the successful stable incorporation and expression of the pLKO-shRNA sequences in target cells.
Chapter 5: Functional validation of Cyr61 and CTGF in VN- and IGF-I:IGFBP:VN-stimulated breast cancer cell survival 160
Packaging vectors
shRNA Delta VSV-G
vector R8.2
+ FuGENE 6 GFP positive Packaging cells
transfection (HEK 293T)
Virus produced
by 293T cells
GFP positive Target cells
Overnight (MDA-MB-231) infection
Figure 5.1: Visualisation of transfection and transduction efficiency
A schematic flow-chart is depicted of the steps involved in the production of lentiviral particles using HEK293T packaging cells and its subsequent infection into the target cells (here MDA-MB-231). Fluorescence microscopy images of HEK293T and MDA-MB-231 cells emitting GFP fluorescence post transfection and transduction, respectively, are also depicted. Scale bars indicate 100 μm.
Chapter 5: Functional validation of Cyr61 and CTGF in VN- and IGF-I:IGFBP:VN-stimulated breast cancer cell survival 161
5.3.2 Screening and selection of shRNA sequences resulting in optimal suppression of Cyr61 and CTGF in MDA-MB-231 cells The MDA-MB-231 cell line with high endogenous Cyr61 and CTGF expression (Chapter 4; Figure 3.9) was used to screen five individual shRNA sequences cloned into the pLKO.1 vector for the optimal suppression of endogenous Cyr61 and CTGF expression. MDA-MB-231 cells were transduced with pLKO.1 containing viral particles containing individual shRNA sequences and RNA collected following puromycin selection. qRT-PCR was performed to assess the expression level of Cyr61 and CTGF transduced with specific shRNA sequences. MDA-MB- 231 parental (MOCK) and MDA-MB-231 cells containing non-targeting control shFF3 shRNA (shCntrl) were used as controls. Gene expression in MDA-MB-231 clones containing gene-specific shRNAs was compared to shCntrl cells and the relative levels of gene expression were assessed (Figures 5.2 A & B). Expression of the target genes was also confirmed at the protein level using immublotting (Figure 5.2 C).
The mRNA expression of Cyr61 was found to be significantly down-regulated (p<0.01) by three of the five candidate shRNAs, including shCyr61-1, -2 and -4 compared to the shCntrl cells (Figure 5.2 A). Maximal reduction in the expression of Cyr61 was observed in shCyr61-4 cells, with 75 ± 9% down-regulation observed compared to the shCntrl cells. Similarly, immunoblot analysis demonstrated a substantial down-regulation of Cyr61 at the protein level in cells transfected with Cyr61-4 sequence (Figure 5.2 C). Thus, the shCyr61-4 clone was used to down- regulate Cyr61 in MCF-7 cell line in the subsequent experiments. Furthermore, out of five different shRNAs tested for CTGF, four shRNAs resulted in a significant down-regulation of mRNA CTGF expression in MDA-MB-231 cells (Figure 5.2 B). However, the shCTGF-3 clone was chosen for subsequent experiments since it resulted in the maximal down-regulation of CTGF, with 84 ± 7% down-regulation observed compared to the shCntrl cells. This was also confirmed by western blot analysis wherein cells transfected with the shCTGF-3 sequence demonstrated a substantial reduction in the CTGF protein expression level (Figure 5.2 C). Importantly, no significant differences were observed in the expression of either Cyr61 or CTGF between the MDA-MB-231 parental (MOCK) and the shCntrl cells (Figures 5.2 A, B & C). This suggests that the pLKO vector backbone as such had no
Chapter 5: Functional validation of Cyr61 and CTGF in VN- and IGF-I:IGFBP:VN-stimulated breast cancer cell survival 162
effects on the gene expression and that shCntrl was suitable to be used as a negative control as it did not induce any off-target effects. Moreover, the morphology of shCyr61 and shCTGF cells remained unchanged and resembled control shCntrl cells (Figure 5.2. D).
In summary, among the various shRNA sequences for Cyr61 and CTGF the shCyr61-4 and shCTGF-3 were selected as the best shRNA sequences for the maximal suppression of Cyr61 and CTGF, respectively. These shRNAs were then tested in MCF-7 cells for which the results are reported in the following section.
Chapter 5: Functional validation of Cyr61 and CTGF in VN- and IGF-I:IGFBP:VN-stimulated breast cancer cell survival 163
A B
C
D
Figure 5.2: Screening of gene-specific shCyr61 and shCTGF sequences in MDA-MB-231 cells using qRT-PCR and immunoblotting.
(A & B) Five shRNA sequences were designed each against Cyr61 (A) and CTGF (B) and were transduced into MDA-MB-231 cells generating 5 clones per target gene. Expression of the target genes was assessed through qRT-PCR and data analysed was performed using the ΔΔCt approach. Data are expressed in log scale as relative fold expression of the target genes in each shRNA- containing cell line compared to that of the control non-targeting shRNA cell line (shCntrl). Target gene expression of un-transduced parental MDA-MB-231 cells (MOCK) is also included for control purposes. The asterisks indicate shRNAs-containing cells with significant down-regulation (*p<0.05, **p<0.01, two-tailed student’s t-test) in target gene expression compared to the shCntrl cells. Error bars indicate SEM. (C) Down-regulation of Cyr61 and CTGF was confirmed using immunoblott analysis in cells containing the best-targeting shCyr61-4 and shCTGF-3 shRNAs, respectively. (D) Phase contrast images of monolayer cultures of MDA-MB 231 cells were used to compare the morphology of the control cells (shCntrl) versus the knockdown cells (shCyr61-3 and shCTGF-4). Scale bar indicates 100 μm. MOCK = parental MDA-MB-231 cells; shCntrl = MDA-MB-231 cells containing the non-targeting shRNA.
Chapter 5: Functional validation of Cyr61 and CTGF in VN- and IGF-I:IGFBP:VN-stimulated breast cancer cell survival 164
5.3.3 Generation of constitutive MCF-7 shRNA knockdown models MCF-7 cells with reduced expression of Cyr61 and CTGF were generated using shCyr61-4 and shCTGF-3 as outlined above for the MDA-MB-231 cell line (as observed in Figure 5.2). The shCyr61-4 and shCTGF-3 sequence constructs will be henceforth simply called shCyr61 and shCTGF. Monolayer cultures of MCF-7- shCyr61 and MCF-7-shCTGFcells were lysed and RNA and protein was extracted. qRT-PCR analysis confirmed a significant reduction (p<0.01) of Cyr61 gene expression in MCF-7-shCyr61 cells, with 68 ± 5% down-regulation observed compared to shCntrl cells (Figure 5.3 A). Likewise, expression of CTGF at mRNA levels was found to be significantly (p<0.01) decreased in MCF-7-shCTGFcells, with 83 ± 1% down-regulation in gene expression observed compared to shCntrl cells (Figure 5.3 A). Similar results were observed at protein levels wherein immunoblot analysis detected a considerable reduction in Cyr61 and CTGF protein expression (Figure 5.3 B). Down-regulation of Cyr61 did not have cross-targeting effect in regulation of CTGF. Likewise, no change in Cyr61 expression was observed in cells containing shCTGF. In addition, no differences in the mRNA and protein expression of target genes were observed in either MCF-7-shCntrl control or MOCK parental MCF-7 cells. Furthermore, the morphology of shCyr61 and shCTGF cells remained unchanged and resembled shCntrl (Figure 5.3 C) and parental (MOCK) cells (data not shown).
Taken together, these data demonstrate the successful generation of MCF-7 breast cancer cell models with knockdown of endogenous Cyr61 and CTGF expression.
Chapter 5: Functional validation of Cyr61 and CTGF in VN- and IGF-I:IGFBP:VN-stimulated breast cancer cell survival 165
A B
C
Figure 5.3: Validation of Cyr61 and CTGF in MCF-7 breast cancer cells.
(A & B) The best shRNA sequences with maximal knockdown for Cyr61 and CTGF genes, shCyr61- 4 and shCTGF-3, were used to produce MCF-7 cells with reduced Cyr61 and CTGF expression. Down-regulation of Cyr61 and CTGF was confirmed in MCF-7-shCyr61and MCF-7-shCTGF cells using qRT-PCR analysis (A) and western immunoblotting (B). The asterisks indicate shRNAs- containing cells with significant down-regulation (**p<0.01, two-tailed student’s t-test) in target gene expression compared to the shCntrl control MCF-7 cells. Error bars indicate SEM. (C) Phase contrast images of monolayer cultures of MCF-7 cells were used to compare the morphology of the control cells (shCntrl) versus the knockdown cells (shCyr61-3 and shCTGF-4). Scale bar indicates 100 μm. MOCK = parental MDA-MB-231 cells; shCntrl = MCF-7 cells containing the non-targeting shRNA.
Chapter 5: Functional validation of Cyr61 and CTGF in VN- and IGF-I:IGFBP:VN-stimulated breast cancer cell survival 166
5.3.4 Suppression of Cyr61 and CTGF expression in MCF-7 cells reduces VN- and IGF-I:IGFBP-5:VN-stimulated cell viability and proliferation in 2D monolayer cultures The main purpose of performing the gene knockdown studies was to investigate the potential involvement of Cyr61 and CTGF in VN- and trimeric IGF- I:IGFBP:VN complex-stimulated breast cancer cell survival. In view of this, MCF-7 cells either with reduced Cyr61 (MCF-7-shCyr61) or CTGF (MCF-7-shCTGF) were seeded along with non-transfected parental (MOCK) and the control non-targeting shRNA-containing (MCF-7-shCntrl) cells into 96-well plates pre-coated with either VN alone or the trimeric complex. Cell viability and cell proliferation were then assessed after 72 hours using MTS and CyQUANT reagents, respectively. Cells suspended in SFM only and medium containing 10% FBS were used as the negative and positive controls, respectively.
MCF-7-ShCyr61cells showed significant decreases (p<0.01) in VN- and trimeric complex-induced viability compared to that observed for parental MCF-7 cells (Figure 5.4 A). Likewise, MCF-7-shCTGF cells were observed to have a significant reduction (p<0.05) in cell viability in response to VN and the trimeric complex compared to the parental cells. The cell viability of shCyr61 cells was also seen to be significantly reduced in response to the SFM control treatment. Similar results were observed for the proliferation assay wherein cell numbers calculated over 72 hours in response to VN and the trimeric complex were significantly lower in MCF-7-shCyr61 and -shCTGF cells compared to the parental controls (Figure 5.4 B). Additionally, shCyr61 and shCTGF containing cells significantly decreased (p<0.05) proliferation in response to the 10% serum treatment. This suggests the possibility that other growth-promoting components present in the serum might also induce MCF-7 cell proliferation through induction and regulation of Cyr61 and CTGF expression. Importantly, no changes were observed in viability and proliferation of shCntrl cells compared to the parental cells, suggesting that shCntrl did not induce any additional effects on cell function.
Taken together, these data suggest that Cyr61 and CTGF play roles in VN and the trimeric IGF-I:IGFBP:VN complex-induced MCF-7 cell survival.
Chapter 5: Functional validation of Cyr61 and CTGF in VN- and IGF-I:IGFBP:VN-stimulated breast cancer cell survival 167
A Cell viability
* *
B Cell proliferation
Figure 5.4: Effects of Cyr61 and CTGF knockdown on VN- and IGF-I:IGFBP-5:VN-stimulated MCF- 7 cell viability and proliferation.
Viability (A) and proliferation (B) of MCF-7 parental (MOCK), shCntrl, shCyr61 and shCTGF cells seeded for 72 hours into 96-well plates pre-coated with VN alone or the trimeric IGF-I:IGFBP-5:VN complex (TRI) were assessed using the MTS and CyQUANT assays, respectively. Cells suspended in SFM or SFM supplemented with 10% FBS were used as the negative and positive controls, respectively. For the viability assay the data are expressed as average percent stimulation of the MOCK cells treated with SFM, whereas for the proliferation assay fluorescent intensity was measured after 72 hours and the relative cell number was obtained and depicted using standard serial dilution and linear regression analysis. The data were pooled either from three experiments (for viability) or two experiments (for proliferation), with treatments tested in at least six wells in each replicate experiment. The asterisks indicate treatments with significant differences compared to their respective MOCK (parental) controls (*p<0.05, **p<0.01, two-tailed student’s t-test). Error bars indicate SEM. SFM = Serum Free Media; TRI = trimeric IGF-I:IGFBP-5:VN complex; 10% FBS = growth medium containing 10% FBS. Fluorescent intensity was measured and standard curve was obtained using linear regression analysis.
Chapter 5: Functional validation of Cyr61 and CTGF in VN- and IGF-I:IGFBP:VN-stimulated breast cancer cell survival 168
5.3.5 Suppression of Cyr61 and CTGF decreases VN- and IGF-I-IGFBP-5:VN- stimulated MCF-7 spheroid formation and long-term cell viability in 3D cultures As outlined in Chapter 1, cells behave differentially when grown in 3D matrices as opposed to those grown in 2D monolayer cultures on plastic. Studies performed in our laboratory (unpublished data) and by others (Weigelt & Bissell, 2008) have demonstrated that the weakly metastatic and poorly-invasive MCF-7 breast cancer cells grown in a 3D matrix such as Matrigel™ form round spheroids resembling in vivo primary breast tumours. To investigate the effects of Cyr61 and CTGF down-regulation on MCF-7 spheroid formation and growth in a 3D setting and to also explore effects on VN- and IGF-I:IGFBP:VN-stimulated spheroid growth, MCF-7 cells containing either shCyr61 or shCTGF (along with the shCntrl cells) were grown in 3D Matrigel™ cultures as per the protocols described in section 2.7. Cells were exposed to either VN alone or the trimeric complex. Cells suspended in the base 3D growth medium (3D-GM; comprising DMEM/F12 + 1% FBS + 2.5 % Matrigel™) served as the negative control. On day 12, the cultures were stained for viable cells using fluorescein diacetate (FDA) and either visually assessed via fluorescent microscopy (Figure 5.5 A) or the size of spheroids was quantified using ImageJ software (Figure 5.5 B). The whole-culture cell viability was measured using the AlamarBlue assay (Figure 5.5 C).
Fluorescence microscopy images clearly demonstrated that in the VN and 3D- GM treatment the size of spheroids formed by shCyr61 and shCTGF cells remained relatively small compared to the control shCntrl cells (Figure 5.5 A). The trimeric complex treatment caused a substantial increase in the size of the spheroids in the shCntrl cells compared to cells treated with VN or 3D-GM alone. These observations are in accordance with previous findings from our laboratory demonstrated by Dr Abhishek Kashyap during his PhD (unpublished data). Interestingly, no such increase in spheroid size was observed in shCyr61and shCTGF cells treated with the trimeric complex (Figure 5.5 A). Based on the visual inspection of the spheroid colonies these results suggest that Cyr61 and CTGF may play a role in VN- and trimeric complex-induced spheroid formation and growth ability of MCF-7 cells in a 3D culture model. To quantify the changes in spheroid size, ImageJ software was used to calculate the amount of pixels occupied by each spheroid to eventually obtain the surface area of each spheroid (in µm2). The box-and-whisker plot of these data
Chapter 5: Functional validation of Cyr61 and CTGF in VN- and IGF-I:IGFBP:VN-stimulated breast cancer cell survival 169
(Figure 5.5 B) clearly demonstrate a dramatic downward shift in the size of VN- and trimeric complex-treated spheroids both the shCyr61 and shCTGF cells compared to shCntrl control cells. A similar result was also observed in cells treated with 3D-GM only.
AlamarBlue assay was performed in order to quantify the viability of the cells cultured in 3D Matrigel™. In addition to the reduced sizes of the spheroids, the overall cell viability in shCyr61 cell cultures was significantly reduced (p<0.01) in response to VN and the trimeric complex compared to cultures containing shCntrl cells (Figure 5.5 C). In a similar fashion, the suppression of CTGF led to significantly reduced viability in response to VN (p<0.05) and the trimeric complex (p<0.01). This effect was also observed for shCyr61 and shCTGF cells treated with 3D-GM only.
Altogether the evidence obtained from the visual inspection of cultures combined with their quantitative assessment (Figure 5.5) suggest that Cyr61 and CTGF are essential for the VN and trimeric IGF-I:IGFBP:VN complex-stimulated MCF-7 breast cancer cell growth, long-term survival and colony formation in a 3D environment.
Chapter 5: Functional validation of Cyr61 and CTGF in VN- and IGF-I:IGFBP:VN-stimulated breast cancer cell survival 170
A
B C 3D viability
Figure 5.5: Effects of Cyr61 and CTGF suppression on MCF-7 cell growth in 3D in response to VN and the IGF-I:IGFBP:VN complex.
One thousand MCF-7 shCntrl, shCyr61 and shCTGF cells suspended in 3D growth medium with 1% serum were seeded onto 100% Matrigel™ and allowed to grow for 3 days and form spheroids. Spent medium was replaced with 3D growth medium containing 1 μg/mL of VN either alone (VN) or pre- complexed with IGF-I (100 ng/mL) and IGFBP-5 (300 ng/mL) (TRI). (A) Cells were fluorescently stained using FDA and images were recorded on Day 12. Box-and-whisker plot (B) depicts mean spheroid size (μm2) at Day 12 for each treatment, calculated using ImageJ software. Data in the graph are the average size of all spheroids present in a single field of view at 40x magnification from eight fields of view (boxes), where the lower bars and the upper bars (whiskers) indicate the smallest and the largest size of spheroids presented in all images, respectively. (C) AlamarBlue assay was performed on day 12 of the 3D culture, with data expressed as average percent cell viability compared to the VN treatment of shCntrl cells. Data are pooled from three replicate experiments with treatments tested in duplicate wells in each replicate experiment. The asterisks indicate significant differences in cell viability compared to their respective shCntrl cells (*p<0.05, **p<0.01). Scale bars indicate 100 μm. Error bars indicate SEM. 3D-GM = 3D growth medium containing 1% FBS; TRI = trimeric IGF- I:IGFBP-5:VN complex; shCntrl = non-targeting control shRNA-containing cells.
Chapter 5: Functional validation of Cyr61 and CTGF in VN- and IGF-I:IGFBP:VN-stimulated breast cancer cell survival 171
5.3.6 Cyr61 knockdown reduces VN- and IGF-I:IGFBP-5:VN-stimulated MDA- MB-231 cell viability in 2D cultures Having shown the involvement of Cyr61 and CTGF in the weakly metastatic MCF-7 cell line, the involvement of these genes in the highly aggressive MDA-MB- 231 cells was further investigated. More specifically, to determine whether Cyr61 and CTGF are crucial determinants in MDA-MB-231 cell viability in response to VN and the trimeric IGF-I:IGFBP:VN complex, non-transfected MDA-MB-231 parental (MOCK) and the shCntrl cells were seeded along with cells with reduced expression of Cyr61 and CTGF onto 96-well plates pre-coated with either VN and the trimeric IGF-I:IGFBP-5:VN complex. Cell viability was then assessed after 72 hours using MTS reagent. Cells suspended in SFM only and medium containing 5% FBS were used as the negative and positive controls, respectively.
Treatment of MDA-MB-231 cells with VN and trimeric complex caused significant increases (p<0.01) in cell viability in MDA-MB-231-parental and MDA- MB-231-shCnrtl cells compared to that observed in cells treated with SFM only (Figure 5.6). Interestingly, the parental MDA-MB-231 and MDA-MB-231-shCntrl cells treated with the trimeric complex significantly increased (p<0.05) viability compared to cells treated with VN alone. This suggests that VN and trimeric complex can promote cell survival in the highly aggressive MDA-MB-231 cell models. Importantly, the MDA-MB-231-shCyr61 cells had significantly reduced cell viability (p<0.05) in response to VN and the trimeric complex compared to the parental and shCntrl cells (Figure 5.6). No appreciable alternation in cell viability was observed in shCTGF cells in response to VN and the trimeric complex.
In summary, this data demonstrate for the first time that VN and the trimeric IGF-I:IGFBP:VN complex are capable of stimulating increased MDA-MB-231 cell viability. The data also suggest that Cyr61 plays and important role in VN- and the trimeric IGF-I:IGFBP:VN-induced viability in the metastatic MDA-MB-231 cell line.
Chapter 5: Functional validation of Cyr61 and CTGF in VN- and IGF-I:IGFBP:VN-stimulated breast cancer cell survival 172
2D viability
Figure 5.6: Effects of Cyr61 and CTGF knockdown on VN- and IGF-I:IGFBP-5:VN-stimulated MDA-MB-231cell viability.
Viability of MDA-MB-231 parental (MOCK), shCntrl, shCyr61 and shCTGF cells seeded for 72 hours into 96-well plates pre-coated with VN alone or the trimeric IGF-I:IGFBP-5:VN complex (TRI) was assessed using the MTS assay. The data are expressed as average percent stimulation of the respective MOCK cells in the SFM negative control. The data are pooled from two experiments with treatments tested in at least six wells in each replicate experiment. The asterisks indicate significant differences between treatments (* p<0.05). Error bars indicate SEM. SFM = Serum Free Media; TRI = trimeric IGF-I:IGFBP-5:VN complex; 5% FBS = growth medium containing 5% FBS.
Chapter 5: Functional validation of Cyr61 and CTGF in VN- and IGF-I:IGFBP:VN-stimulated breast cancer cell survival 173
5.3.7 Involvement of Cyr61 in regulating cell survival and CTGF in mediating spreding of MDA-MB-231 cells in response to VN and the IGF- I:IGFBP:VN complex in 3D cultures To examine the role of Cyr61 and CTGF in cell growth, survival and invasive properties of MDA-MB-231 cells in a 3D setting in response to VN and the IGF- I:IGFBP:VN complex, MDA-MB-231 cells with reduced expression of either Cyr61 or CTGF were grown along with shCntrl cells in 3D Matrigel™ cultures and treated with either VN alone or the IGF-I:IGFBP:VN complex. Cells suspended in the base 3D-GM again served as negative controls. On day 9, viable cells were fluorescently stained using FDA and the morphology was assessed visually via fluorescence microscopy (Figure 5.7 A). Since MDA-MB-231 cells are highly invasive in this culture system (reflecting their in vivo metastatic ability) and do not form defined colonies, ImageJ was used to assess the effect of candidate gene suppression on cell invasiveness and morphology, rather than colony size as for MCF-7 cultures. Thus, the morphological changes in MDA-MB-231 3D cultures were analysed by measuring cell circularity and the relative frequency distribution using ImageJ (Figure 5.7 B). As previously, the whole-culture cell viability was also measured using the AlamarBlue assay (Figure 5.7 C).
As expected, MDA-MD-231 cells formed meshworks of spindle-like mesenchymal colonies where cells were seen to invade in different directions within the Matrigel™ matrix (Figure 5.7A). Visual inspection of cultures indicated no apparent differences in growth and morphology were observed in MDA-MD-231- shCntrl cells cultured with VN, trimeric complex or the 3D-GM control treatments (Figure 5.7 A). Interestingly, however MDA-MB 231-shCTGF cells were observed to have lost their spindle-like morphology (Figure 5.7 A), and instead formed more rounded epithelial-like spheroids resembling the MCF-7 3D culture spheroids (Figure 5.5 A). These results were also reflected in the ImageJ analyses of the circularity of the cell colonies wherein round colonies were over represented in the shCTGF cultures as compared to the shCntrl cultures (Figure 5.7 B).
Upon assessing the viability of 3D cultures using AlamarBlue it was observed that unlike the responses observed for MCF-7 cultures, there was a non-significant increase in shCntrl cell viability observed in cultures treated with the trimeric complex compared to the cells treated with VN or 3D-GM alone. Importantly, the MDA-MB-231-shCyr61 cells had a significantly reduced (p<0.05) viability in Chapter 5: Functional validation of Cyr61 and CTGF in VN- and IGF-I:IGFBP:VN-stimulated breast cancer cell survival 174
response to VN and the trimeric complex compared to shCntrl cells (Figure 5.7). In contrast to the 2D viability results, no appreciable changes in cell viability were observed in cultures containing MDA-MB-231-shCTGF cells as compared to MDA- MB-231-shCntrl cultures.
Chapter 5: Functional validation of Cyr61 and CTGF in VN- and IGF-I:IGFBP:VN-stimulated breast cancer cell survival 175
A
B C
Figure 5.7: Effects of Cyr61 and CTGF suppression on growth and morphological appearance of MDA-MB-231 cells cultured in 3D Matrigel™ in response to VN and the IGF-I:IGFBP:VN complex
Eight hundred MDA-MB-231-shCntrl, -shCyr61 and -shCTGF cells suspended in 3D growth medium with 1% serum were seeded onto 100% Matrigel™ and allowed to grow for 3 days and form spheroids. Spent medium was replaced with 3D growth medium containing 1 μg/mL of VN either alone (VN) or pre-complexed with IGF-I (100 ng/mL) and IGFBP-5 (300 ng/mL) (TRI). (A) Cells were fluorescently stained using FDA and images were recorded on day 9 using Nikon Eclipse TS100 microscope. (B) ImageJ was then used to measure the circularity of the cell colonies (ranging from 0 to 1) cultured in Matrigel™ and the results were plotted as a frequency distribution graph. 0 = not circular/more invasive to 1 = circular/less invasive. (C) AlamarBlue assay was performed on Day 9 on these cell cultures and the data were expressed as average percent cell viability compared to the shCntrl cells of the VN treatment. Data are pooled from two replicate experiments with treatments tested in duplicate wells in each replicate experiment. The asterisks indicate significant differences compared to the MDA-MB-231-shCntrl cells within the respective treatment group (*p<0.05, two- tailed student’s t-test). Scale bars indicate 100 μm. Error bars indicate SEM. 3D-GM = 3D growth medium containing 1% FBS; TRI = trimeric IGF-I:IGFBP-5:VN complex.
Chapter 5: Functional validation of Cyr61 and CTGF in VN- and IGF-I:IGFBP:VN-stimulated breast cancer cell survival 176
5.3.8 CTGF suppression decreases breast cancer cell migration in response to VN and the IGF-I:IGFBP:VN complex. It is well documented that VN promotes migration in a variety of epithelial cell types, including breast cancer cells (Carriero et al., 1999; Madsen et al., 2007; Pola et al., 2013). Additionally, studies within our laboratory have demonstrated that the trimeric IGF-I:IGFBP:VN complex can mediate breast cancer cell migration above the response observed with VN alone (Hollier et al., 2008; Kashyap et al., 2011). Importantly, the studies reported herein demonstrated that suppression of endogenous CTGF expression can alter the morphological appearance of MDA-MB- 231 cells in 3D cultures (Figure 5.7 A). Thus, the effect of CTGF down-regulation in addition to Cyr61 was further investigated in VN and IGF-I:IGFBP:VN-stimulated migration of MCF-7 and MDA-MB-231 cells. Transwell® migration plates were coated on the under-surface with VN or the IGF-I:IGFBP-5:VN complex. MCF-7 and MDA-MB-231 cells containing shRNA against either Cyr61 or CTGF or the non-targeting control shRNA (shCntrl) were seeded onto the upper-surface of the inserts. Cells suspended in SFM only and medium containing 10% FBS (for MCF-7 cells) or 5% FBS (for MDA-MB-231 cells) were used as the negative and positive controls, respectively.
Treatment of the MCF-7-shCntrl cells with VN and the trimeric complex stimulated a significant increase (p<0.01) in cell migration compared to the SFM control treatment cells (Figure 5.8 A). As previously reported by our laboratory (Hollier et al., 2008), the trimeric complex significantly enhanced (p<0.01) the migration of MCF-7 cells compared to the VN treatment. MCF-7 cell migration was seen to be significantly reduced in the shCTGF cells in response to both VN (p<0.05) and the trimeric complex (p<0.01) compared to shCntrl cells (Figure 5.8 A). This suggests a role for CTGF in VN as well as trimeric complex-induced MCF-7 cell migration. No significant differences were observed in migration of cells with reduced Cyr61.
In a similar fashion to MCF-7-shCntrl cells, treatment of the control MDA- MB-231 cells (MDA-MB-231-shCntrl) with VN and the trimeric complex significantly increased (p<0.01) cell migration compared to cells treated with SFM only (Figure 5.8 B). Again, the migration of MDA-MB-231-shCntrl cells treated with the trimeric complex was significantly higher (p<0.01) than cells treated with
Chapter 5: Functional validation of Cyr61 and CTGF in VN- and IGF-I:IGFBP:VN-stimulated breast cancer cell survival 177
VN alone. Importantly, VN- and trimeric complex-induced migration in MDA-MB- 231 cells containing shCTGF (MDA-MB-231-sh-CTGF) was observed to be significantly reduced (p<0.01) compared to that observed in MDA-MB-231-shCntrl cells (Figure 5.8 B).This suggests that suppression of CTGF not only reduces the migratory responses induced by VN, but also can influence IGF-I-stimulated migration in breast cancer cells. MDA-MB-231-shCyr61cells did not alter migration in response to either VN or the trimeric complex.
Taken together, these results indicate a potential role for CTGF but not Cyr61 in the migratory abilities of both MCF-7 (weakly metastatic) and MDA-MB-231 (highly metastatic) breast cancer cell lines in response to VN- and the trimeric IGF- I:IGFBP:VN complex.
Chapter 5: Functional validation of Cyr61 and CTGF in VN- and IGF-I:IGFBP:VN-stimulated breast cancer cell survival 178
A MCF-7
B MDA-MB-231
Figure 5.8: The effects of Cyr61 and CTGF down-regulation on MCF-7 and MDA-MB-231 cell migration in response to VN and the IGF-I:IGFBP:VN complex. shCyr61 and shCTGF containing MCF-7 (A) and MDA-MB-231 (B) cells along with shCntrl were seeded into Transwells® pre-coated with VN alone or the trimeric IGF-I:IGFBP-5:VN complex (TRI) were allowed to migrate for 14 hours and the levels of migration were assessed using crystal violet staining. The data are pooled from two independent experiments with treatments tested in duplicate wells in each replicate experiment. All treatments significantly increased (p<0.01) migration compared to the control SFM treatment. The asterisks indicate the shCntrl cell treated with the IGF- I:IGFBP:VN complex which significantly increased (p<0.01) migration compared to cells treated with VN alone. The asterisks also indicate shCTGF cells which significantly decreased migration compared to shCntrl cells (*p<0.05; **p<0.01). Error bars indicate SEM. SFM = Serum Free Media; TRI = trimeric IGF-I:IGFBP-5:VN complex.
Chapter 5: Functional validation of Cyr61 and CTGF in VN- and IGF-I:IGFBP:VN-stimulated breast cancer cell survival 179
5.4 DISCUSSION
Microarray analysis of the MCF-7 breast cancer cells that remained viable in response to VN and the IGF-I:IGFBP:VN complex revealed a number of differentially expressed genes with key roles in breast cancer cell survival. Among these genes, Cyr61 and CTGF genes were selected as top candidates to functionally validate their roles in VN- and trimeric IGF-I:IGFBP:VN complex-induced cell survival. Indeed, the results presented in Chapter 4 confirmed that VN and the IGF- I:IGFBP:VN complex dramatically up-regulated the expression of Cyr61 and CTGF at both mRNA and protein levels in MCF-7 breast cancer cells. It was also shown that VN and the IGF-I:IGFBP:VN complex are capable of stimulating Cyr61 and CTGF expression in a range of breast cancer cells from weakly metastatic luminal- like to the highly metastatic basal-like and claudin-low breast cancer cell types. However, the extent of the involvement of these two genes in regulating the functions of VN and the trimeric complex remain to be validated. In view of this, studies performed in this chapter comprise the first investigations into the involvement of Cyr61 and CTGF in VN- and IGF-I:IGFBP:VN-induced cell functions, namely cell survival and cell migration, using in vitro 2D and 3D culture assays.
A loss-of-function approach was employed to functionally validate the involvement of Cyr61 and CTGF in VN- and IGF-I:IGFBP:VN-stimulated cell responses. Lentiviral gene delivery was used to deliver gene specific shRNA to target Cyr61 and CTGF. The target cells chosen were the weakly metastatic MCF-7 and the highly metastatic MDA-MB-231 breast cancer cells. The MCF-7 cell line was the obvious choice given that these genes were previously reported (Chapter 4) to be dramatically up-regulated in MCF-7 cells in response to VN and the IGF- I:IGFBP:VN complex. Nevertheless, the basal expression levels of Cyr61 and CTGF in MCF-7 cells are quite low making them sub-optimal for testing shRNA sequences for maximal gene suppression. This was apparent from studies presented in Chapter 4 wherein basal expression of Cyr61 and CTGF protein is barely detectable in MCF- 7 cells cultured under short-term serum depleted conditions. This is consistent with findings from others, namely Menendez et al. (2005), Chen et al. (2007) and Lin et al (2012) who reported a low expression of Cyr61 and CTGF at both mRNA and protein levels in MCF-7 cells grown under serum-free conditions. Hence, given their
Chapter 5: Functional validation of Cyr61 and CTGF in VN- and IGF-I:IGFBP:VN-stimulated breast cancer cell survival 180
high endogenous levels of both Cyr61 and CTGF (Xie, Miller, et al., 2001; Chen et al., 2007), MDA-MB-231 cells were used to screen multiple shRNA sequences for each target gene. This would also serve the dual purpose of functional validation in an additional breast cancer cell line with metastatic properties. The shRNA construct with maximal down-regulation of Cyr61 and CTGF in MDA-MB-231 cells identified at both mRNA and protein levels was selected (Figure 5.2 A & B) and consequently were used to further supress these candidates in MCF-7 cells (Figure 5.3 A & B).
The overexpression of Cyr61 has been reported to enhance cell survival and proliferation in response to serum (Sampath et al., 2001; Tsai et al., 2002; Lin et al., 2004; Menendez et al., 2005), and conversely its inhibition has been shown to attenuate the viability of MCF-7 cells in 2D monolayer cultures (Espinoza et al., 2011). In view of this, experiments were performed to assess the impact of Cyr61 down-regulation on VN- and IGF-I:IGFBP:VN complex-induced MCF-7 cell viability as well as proliferation using conventional 2D monolayer assays. Cells with down-regulated Cyr61 demonstrated significant decreases in VN-mediated cell viability and proliferation (Figure 5.4). Interestingly, these cells were also found to significantly reduce viability and proliferation induced by the IGF-I:IGFBP:VN complex. This suggests that targeting the key downstream mediators of VN, which is the major component of the IGF-I:IGFBP:VN complex, can also influence IGF-I- mediated biological functions. In addition to its effects on VN- and IGF- I:IGFBP:VN-induced cell survival, Cyr61 suppression also resulted in a significant decrease in serum-stimulated cell viability and proliferation. As reported previously (Sampath et al., 2001; Tsai et al., 2002; Lin et al., 2004; Menendez et al., 2005). Serum is a complex mixture of growth factors, hormones, proteins, lipids and minerals that are required for successful in vitro cell culture. Therefore, it is possible that growth-promoting and ECM elements present within the serum may support cell viability and proliferation through the induction and regulation of Cyr61. Importantly, serum contains high levels of VN, which in turn may be the component responsible for serum-induced Cyr61 effects.
In addition to MCF-7 cells, the studies performed in this chapter also functionally validated the effect of Cyr61 and CTGF down-regulation in VN- and the IGF-I:IGFBP:VN complex-induced viability of MDA-MB-231 cells. Similar to those responses observed for MCF-7 cells, VN and the IGF-I:IGFBP:VN complex
Chapter 5: Functional validation of Cyr61 and CTGF in VN- and IGF-I:IGFBP:VN-stimulated breast cancer cell survival 181
substantially increased viability of MDA-MB-231 cells compared to those cells in serum-free medium devoid of any added protein treatments (Figure 5.6). Importantly, MDA-MB-231 cells with down-regulation of Cyr61 had reduced VN- and IGF- I:IGFBP:VN-induced viability (Figure 5.6). This is in accordance with studies performed by others demonstrating that blocking Cyr61 expression using Zoledronic acid, a bisphosphonate with anti-cancer properties, considerably reduced MDA-MB- 231 cell viability in 2D cultures by inhibiting cell growth and inducing apoptosis (Espinoza et al., 2011). Likewise, treatment of MDA-MB-231 cells with humanised anti-Cyr61 monoclonal antibody (093G9) inhibited cell proliferation in vitro and in vivo (Lin et al., 2012). Considering the dramatic up-regulation of Cyr61 by VN and the IGF-I:IGFBP:VN complex in MCF-7 and MDA-MB-231 cells (Figure 4.8) and the effect of its down-regulation in inhibition of survival induced by VN and the IGF-I:IGFBP:VN complex (Figures 5.4 & 5.6), it is clear that Cyr61 is critically involved in VN- and IGF-I:IGFBP:VN complex-induced cell survival effects.
A growing body of evidence suggests that the use of in vitro 3D cell culture models, in contrast to 2D monolayer cultures on plastic, replicate more accurately the actual microenvironment where cells reside in tissue. Therefore, the behaviour of cells cultured in 3D is likely to reflect more closely in vivo responses (Debnath & Brugge, 2005; Weigelt & Bissell, 2008; Muthuswamy, 2011; Weigelt et al., 2014). In view of this, in addition to the 2D studies reported herein, the impact of Cyr61 and CTGF suppression on VN- and IGF-I:IGFBP:VN-induced cell growth/survival of breast cancer cells in 3D Matrigel™ cultures was assessed. The “3D on-top” approach was employed to culture cells on a 3D matrix where cells suspended in Growth Factor Reduced (GFR)-Matrigel™ (2.5%) + 1% FBS were layered on top of a 100% GFR-Matrigel™ layer. This approach was deemed more amenable to our experimental systems than the “3D embedded” approach where the cells are grown within the 100% Matrigel™ matrix making it difficult to study the cell-protein interactions in such a dense matrix.
On visual assessment it was observed that the size of spheroids formed by MCF-7 cells with reduced Cyr61 expression in response to both VN as well as the IGF-I:IGBFP:VN complex was substantially smaller compared to spheroids formed by shCntrl control cells (Figure 5.5 A). The ImageJ calculations for spheroid size and the AlamarBlue results also correlated well with fluorescent microscopy images of
Chapter 5: Functional validation of Cyr61 and CTGF in VN- and IGF-I:IGFBP:VN-stimulated breast cancer cell survival 182
MCF-7 cell spheroids (Figure 5.5 B & C). The results indicated that cells with suppression of Cyr61 had significantly decreased long-term viability/proliferation in response to VN and the IGF-I:IGFBP:VN complex compared to the control cells. Hence, reducing viability/proliferation of cells within the spheroids leading to formation of smaller sized spheroids. Interestingly, the long-term viability and the size of spheroids in response to the IGF-I:IGFBP:VN complex was reduced to the levels as observed for the VN alone treatment with suppression of Cyr61. This suggests that targeting Cyr61 not only influences the VN- but also the IGF-I-induced long-term viability and spheroid formation of MCF-7 cells in a 3D environment. These results are in concordance with previous studies showing the expression of Cyr61 to influence the MCF-7 outgrowth and long-term viability in 3D cultures, including Matrigel™ (Tsai et al., 2002) and soft agar cultures (Lin et al., 2004; Menendez et al., 2005). This is likely due to the stimulatory effect of Cyr61 expression on cell proliferation in the MCF-7 cell line (Sampath et al., 2001; O'Kelly et al., 2008). Similar effects have also been reported in epidemiological studies where a high level of Cyr61 expression was found to positively correlate with tumour size in breast cancer patients (O'Kelly et al., 2008). These findings suggest the possibility that Cyr61 may be a promising therapeutic target for the inhibition of tumour growth and progression in breast cancer patients where VN and IGF-I are implicated.
In contrast to MCF-7 cells, analysis of fluorescence images demonstrated no apparent changes in overall spheroid size or invasive morphology of MDA-MB-231 cells with reduced Cyr61 levels grown in 3D Matrigel™ cultures (Figure 5.7 A). However, AlamarBlue analysis showed a significant reduction in long-term viability/proliferation of MDA-MB-231-shCyr61 cells compared to the control MDA-MB-231-shCntrl cells (Figure 5.7 C). Similar results were reported by Espinoza et al. (2011), whereby zoledronic acid-mediated down-regulation of Cyr61 reduced outgrowth and cell viability of MDA-MB-231 cells grown in 3D Matrigel™ cultures. Moreover, treatment of nude mice bearing MDA-MB-231 cells with anti- Cyr61 monoclonal antibody 093G9 significantly reduced the tumour growth (Lin et al., 2012). This suggests that targeting Cyr61 may not only hold promise for inhibition of the growth properties of luminal-like breast cancer subtypes (such as
Chapter 5: Functional validation of Cyr61 and CTGF in VN- and IGF-I:IGFBP:VN-stimulated breast cancer cell survival 183
MCF-7 cells), but also for treatment of aggressive tumours with a basal-like/claudin- low phenotype.
In a similar fashion, the suppression of CTGF significantly reduced VN- and IGF-I:IGFBP:VN-induced MCF-7 cell viability and proliferation in 2D cultures (Figure 5.4). These results were also observed in 3D cultures wherein MCF-7 cells with reduced CTGF appreciably decreased VN- and the IGF-I:IGFBP:VN complex- stimulated spheroid formation and long term viability of MCF-7 cells in 3D Matrigel™ cultures (Figure 5.5). It has been reported that CTGF can promote cell survival through stimulation of cell cycle (Kothapalli et al., 1998), cell proliferation and protection of cells from apoptosis (Babic et al., 1999). Indeed, it has been shown that overexpression of CTGF in MCF-7 cells increased their colony formation ability and viability in response to chemotrophic agents (Wang et al., 2009). Consequently, it is possible that CTGF down-regulation reduces cell survival in MCF-7 cells in response to VN and the trimeric complex via cell cycle arrest and/or less protection of cells from apoptosis. However, further biological assays specifically assessing cell cycle using flow cytometry and apoptosis will be necessary to confirm these mechanisms in our model. In contrast, the suppression of CTGF levels showed no appreciable inhibition of MDA-MB-231 cell viability in 2D monolayer cultures (Figure 5.6). This result was also reported in findings by others wherein down- regulation of CTGF in MDA-MB-231 cells did not alter growth, proliferation or viability in this cell line (Chen et al., 2007; Wang et al., 2009).
Visual assessment of 3D cultures also revealed the suppression of CTGF to reduce the invasive morphology of MDA-MB-231 cells. Reduced CTGF levels were observed to revert MDA-MB-231 cells and changed their morphology from their characteristic spindle-like invasive phenotype to more rounded non-invasive spheroid morphology (Figure 5.7 A), resembling MCF-7 cell spheroids (Figure 5.5 A). We attempted to quantify this switch in morphology using ImageJ analysis, whereby MDA-MB-231-shCTGF cell cultures contained more “circular” spheroid- like colonies compared to MDA-MB-231-shCntrl cells that contained predominantly “non-circular” invasive colonies (Figure 5.7 B). This finding is consistent with other observations where suppression of CTGF in MDA-MB-231 cells led to morphological alterations from fibroblast-like to epithelial-like morphology in 2D cultures (Chen et al., 2007). However, our study has demonstrated this phenotypic
Chapter 5: Functional validation of Cyr61 and CTGF in VN- and IGF-I:IGFBP:VN-stimulated breast cancer cell survival 184
reversion for the first time in a more complex 3D culture which more accurately mimics the in vivo cellular environment. These results suggest CTGF may have a role in maintaining the invasive and migratory phenotype of breast cancer cells, particularly in tumours where components of the VN and/or IGF axis are dysregulated.
To more directly assess if CTGF has a role in the migratory phenotype of breast cancer cells in response to VN and the IGF-I:IGFBP:VN complex, Transwell® migration assays were performed on MCF-7-shCTGF and MDA-MB-231-CTGF cells. Cells with reduced Cyr61 were also used in these assays to investigate the effect of Cyr61 down-regulation in VN- and IGF-I:IGFBP:VN-induced migration of target cell lines. Pre-binding Transwells® with VN was sufficient to significantly increase migration of the control MCF-7-shCntrl cells compare to control SFM treatment (Figure 5.8 A). In addition, migration of shCntrl cells in response to the IGF-I:IGFBP:VN complex was significantly higher than those observed in response to VN. This corroborates the data observed in studies previously performed within our laboratory showing that the IGF-I:IGFBP:VN complex can induce migration to significantly higher levels than those observed for cells treated with VN alone (Hollier et al., 2008; Kashyap et al., 2011). Importantly, down-regulation of CTGF attenuated both VN- and IGF-I:IGFBP:VN complex-induced MCF-7 cell migration (Figure 5.8 A). This fits well with studies showing that MCF-7 cells stably transfected with full-length CTGF increased migration in both FN- and VN-coated Boyden chambers, with more significant increase on VN-coated chambers (Chien et al., 2011). This result again highlights the fact that targeting VN-induced CTGF expression can also influence IGF-I:IGBFP:VN complex-mediated effects.
Similar results were also observed for MDA-MB-231 cells, wherein, down- regulation of CTGF significantly reduced both VN- and IGF-I:IGBFP:VN complex- stimulated migration (Figure 5.8 B). This can also been seen in the studies performed by Chen et al. (2007) showing the reduction of CTGF expression in MDA-MB-231 by antisense CTGF cDNA impaired cellular migration in a wound scratch assay. Recently, overexpression of bone morphogenic protein 9 (BMP-9) in MDA-MB-231 cells was shown to inhibit breast cancer bone metastasis through down-regulation of CTGF (Ren et al., 2014). In addition, enhanced expression of CTGF through activation of ERK1/2 signalling by parathyroid hormone-related peptide (PTHrP)
Chapter 5: Functional validation of Cyr61 and CTGF in VN- and IGF-I:IGFBP:VN-stimulated breast cancer cell survival 185
was also found to be one of the key mechanisms involved in osteolytic metastasis of breast cancer (Shimo et al., 2006). This correlates well with findings reported in Chapter 4 showing that VN and the IGF-I:IGFBP:VN complex induce expression of CTGF through the Src/ERK signalling axis (Figure 4.6). Taken together, these findings suggest that increased expression of CTGF by VN and the IGF-I:IGFBP:VN complex plays an important role in their stimulatory effects on breast cancer cell migration which may possibly be regulated through activation and recruitment of the ERK signalling pathway. Moreover, no correlation was found between Cyr61 regulation and the migratory responses induced by VN and the IGF-I:IGFBP:VN complex in MCF-7 and MDA-MB-231 cells, suggesting that Cyr61 is more predominantly involved in cell survival whereas CTGF is more involved in cell migration.
In summary, the studies reported in this chapter demonstrate for the first time the critical roles of Cyr61 and CTGF in VN- and the IGF-I:IGFBP:VN complex- induced breast cell viability and proliferation in 2D cultures. These results were also confirmed in 3D Matrigel™ cultures where down-regulation of Cyr61 and CTGF reduced long-term cell viability and colony formating ability of MCF-7 cells induced by VN and the IGF-I:IGFBP:VN complex. Cyr61 was also found to be important for VN- and IGF-I:IGFBP:VN -stimulated MDA-MB-231 cell viability in 2D and 3D cultures. The most exciting part of the studies presented in this chapter was the data demonstrating the effect of CTGF suppression on transforming the invasive spindle- like morphological appearance of the highly metastatic MDA-MB-231 cells to more round epithelial-like colonies. The data presented herein also provide the first evidence for the key functional role of CTGF in VN- and IGF-I:IGFBP:VN- enhanced MCF-7 and MDA-MB-231 cell migration.
5.5 CONCLUSION
The data acquired in this chapter for the first time reveals the functional roles of Cyr61 and CTGF in VN- and the trimeric IGF-I:IGBFP:VN complex-stimulated breast cancer cell survival and migration. These results suggest that Cyr61 plays a more predominant role in cell survival, whereas CTGF is more involved in cell migration responses induced by VN and the IGF-I:IGFBP:VN complex. Additionally, the studies reported in the previous chapter (Chapter 4) demonstrated
Chapter 5: Functional validation of Cyr61 and CTGF in VN- and IGF-I:IGFBP:VN-stimulated breast cancer cell survival 186
that VN and the IGF-I:IGBFP:VN complex dramatically induced Cyr61 and CTGF expression through interaction with αv-integrins and activation of the Src/PI3-K/AKT and Src/ERK/MAPK pathways, respectively. Since AKT signalling is known to play a crucial role in cell survival (Datta et al., 1997) and MAPK in cell migration (Huang et al., 2004), we propose that VN and the IGF-I:IGFBP:VN complex induce expression of Cyr61 through PI3-K/AKT to mediate cell survival and CTGF though ERK/MAPK pathway to enhance cell migration.
The functional studies reported throughout this chapter provides the rationale for the use of Cyr61 and CTGF as diagnostic and prognostic biomarkers predicting the severity of breast cancer, in particular in patients with overexpression or increased deposition of VN and/or IGF system components specifically within the tumour microenvironment. Targeting Cyr61 and CTGF proteins or associated signalling pathways also holds promise in the development of specific therapeutic targets for inhibiting the biological functions of VN and possibly other integrin- mediated ECM components relevant to breast cancer progression.
Chapter 5: Functional validation of Cyr61 and CTGF in VN- and IGF-I:IGFBP:VN-stimulated breast cancer cell survival 187 Chapter 6: General Discussion
Chapter 6: General Discussion 188
It is well established that alteration in the dynamics of ECM components within the mammary gland is among the most important factors contributing to breast cancer progression, from pre-malignancy to the proliferation, invasion, and survival of the malignant tumour and its subsequent metastasis (Lochter & Bissell, 1995; Hansen & Bissell, 2000; Weigelt & Bissell, 2008; Place et al., 2011). However the daunting task is to determine which of these ECM factors actually confers causative effects and/or promotes cancer, and to identify the downstream mediators of such biological functions. In this regard, a growing body of evidence indicates that the ECM glycoprotein VN with its multifunctional roles can modulate breast cancer progression and metastasis by promoting cell differentiation, adhesion, and migration while also providing tumour cells with survival advantages (Carriero et al., 1999; Aaboe et al., 2003; Hurt et al., 2010; Leavesley et al., 2013). Given the correlation between VN and the progression/metastasis potential of breast cancer and the localisation of VN and VN-binding integrins at the leading edges of invasive tumours (Carriero et al., 1997; Aaboe et al., 2003; Kadowaki et al., 2011), the underlying aim of this PhD project was to identify for the first time the molecular mediators responsible for VN-stimulated cellular functions relevant to breast cancer progression, in particular cellular survival.
Previous work within our laboratory has also demonstrated that IGF ligands can associate with VN forming multiprotein complexes of IGF-I:IGFBP:VN and IGF-II:VN (Upton et al., 1999; Kricker et al., 2003) that are capable of stimulating breast cancer cell migration, survival and inhibiting drug-induced apoptosis (Hollier et al., 2008; Kashyap et al., 2011). As such, the studies reported herein also aimed at identifying the molecular mediators underpinning IGF:VN complex-mediated breast cancer cell survival. Identification of molecular mediators underlying VN and the IGF:VN complexes may help to identify the molecular targets to inhibit the effects of VN- as well as the IGF:VN complex-induced responses, including promoting cancer cell survival.
In order to identify the candidate molecular mediators, DNA microarrays were performed on RNA samples isolated from cells which had survived in response to VN with or without growth factors pre-bound to polystyrene cell culture plates. This “pre-bound” approach was implemented to more accurately mimic the in vivo microenvironment which these breast cancer cells normally encounter where cells
Chapter 6: General Discussion 189
are more likely to be exposed to growth factors that are present in association with ECM proteins (Hynes, 2009). Moreover, in vivo the binding of growth factors to ECM proteins can result in the maintenance of a uniform growth factor concentration, alter diffusion rates through the matrix regulating its local concentration, and also modulate growth factor receptor interactions (Taipale & Keski-Oja, 1997). Hence, this approach was adopted in an attempt to more accurately depict the situation cells encounter in vivo and identify genes that represent the actual cellular and molecular processes involved in cell survival induced by these proteins and growth factors.
DNA microarray analysis studies reported in Chapter 3 identified a number of genes to be differentially expressed by at least ± 1.5-fold in MCF-7 breast cancer cells that survived in response to substrate-bound VN and the IGF:VN complexes. Furthermore, these genes were found to be significantly associated with a range of biological functions such as “cellular growth and proliferation”, “cellular movement”, “cellular development”, “cell cycle”, “cellular assembly and organisation”, “cell death and survival”, “cell morphology”, “cell-cell signalling and interactions” and “molecular transport”. The results from these studies have provided valuable new knowledge regarding the gene expression profile underlying the functional impact of VN and the IGF:VN complexes, which can also be used by others with interest in any of these areas. Furthermore, the analysis outlined in Chapter 3 revealed that the vast majority of the differential gene expression induced by the IGF:VN complexes highly overlapped with those regulated by VN alone. This suggests that in terms of breast cancer cell survival VN is the key component driving differential gene expression observed in response to the IGF:VN complexes.
The behaviour of a cell in vivo is largely determined by its interactions with the ECM, neighbouring cells, and soluble local and systemic cues presented within the cellular microenvironment, namely hormones, cytokines, and growth factors (Nelson & Bissell, 2006; Rondeau et al., 2014). However, the components of the ECM are known to be active participants in these biological responses that can induce or supress gene expression in normal and malignant cells (Roskelley et al., 1995; Spencer et al., 2007; Birnie et al., 2008). Thus, are not simply just passive bystanders that maintain tissue structure and integrity. Hence, we hypothesised that the identification and subsequent targeting of candidate genes differentially
Chapter 6: General Discussion 190
expressed in common by VN and the IGF:VN complex (simply called the VN- induced gene expression signature or VNGES) will inhibit not only VN, but potentially also IGF:VN complex-mediated survival. This was supported via gene ontology analysis revealing VNGES to be enriched by genes with previously reported roles in cell growth and survival. Thus, VNGES likely contains drivers of VN- and trimeric IGF-I:IGFBP:VN complex-mediated cell survival effects.
The ability of cancer cells to survive during invasion and migration through the blood and lymphatic vessels and subsequently within distant inhospitable tissue microenvironments is a key factor in cancer progression and metastasis in vivo (Scheel & Weinberg, 2012). The interactions between malignant epithelial cells and the surrounding ECM involves the transmission of anti-apoptotic as well as proliferative signals critical for survival and proliferation of cancer cells during tumour development, metastasis and resistance against chemotherapeutic agents (Jacks & Weinberg, 2002). Earlier experimental evidence has established that disruption of interactions between epithelial cells and the ECM can induce apoptosis (Frisch & Francis, 1994). Likewise, the blockade of ECM-binding cell surface receptors such as integrins is well known to inhibit cell growth and induce apoptosis in a number of tumours, including the tumours of the lung (Sethi et al., 1999), breast (Weaver et al., 2002; Park et al., 2006) and prostate (Bisanz et al., 2005). As previously discussed, components of the ECM can control cell behaviour and function by regulating the changes in gene expression patterns. To date, a number of genes have been found to be regulated by specific components of the ECM to promote cancer progression and metastasis. However, there is still a lack of studies investigating the direct effect of specific ECM proteins on the global transcriptional responses in cancer cells and whether these transcriptional alterations are critical for biological phenotypes relevant to metastasis.
For instance, regulation of cellular proto-oncogene c-JUN and mir-21 by components of the ECM such as collagen type I and hyaluronan has been reported to inhibit apoptosis and promote the survival of breast cancer cells (Li et al., 2011; Chen & Bourguignon, 2014). Moreover, coordination of MMP-9 and CD44 expression (receptor for hyaluronan) can activate latent TGFβ promoting tumour cell growth and invasion (Yu & Stamenkovic, 2000). Furthermore, an increased autocrine TGFβ loop can protect breast cancer cells from apoptosis and enhance their
Chapter 6: General Discussion 191
proliferation (Hoshino et al., 2011), while also providing resistance to paclitaxel in triple negative breast cancers (Bhola et al., 2013). Interestingly, Smad proteins, the downstream transcriptional mediators of TGFβ are also found to be up-regulated by ECM components such as type I collagen and are involved in mediating the growth of pancreatic cancer cells (Imamichi et al., 2007).
In a similar fashion, studies reported in this PhD thesis have added significant new information to our understanding of how the ECM can instruct the transcriptional and biological responses of tumour cells. Based on previous experimental evidences we have identified that the VNGES contains numerous genes relevant to survival, including TGFβ1/2, Smad6, S100A4, S100P, ANXA2, RAPGAP, Cyr61/CCN1, CTGF/CCN2, Snail 2 (Slug), Caveolin-1, mir-21, c-JUN and c-FOS. Deregulation of these genes is reported to correlate with breast cancer progression and metastasis, and their expression is also found to be directly or indirectly associated with other ECM components. Since these genes were found to be differentially expressed by VN and their regulation is shown to correlate with breast cancer progression and metastasis, we propose these genes could form a prognostic and/or predictive biomarker signature for breast cancer progression in patients with increased intra-tumoral deposition of VN. This will need to be confirmed in human breast cancer patient cohorts using either IHC staining of these target genes in Tissue Microarrays (TMA) or meta-analysis of publicly available gene expression data available via data repositories such as the Gene Expression Omnibus (GEO; www.ncbi.nlm.nih.gov/geo/) and Oncomine (www.oncomine.org). Correlation analysis can also be performed to determine the correlation of VN expression with VNGES target genes within these cohorts and their relationship to clinicopathological parameters, such as tumour recurrence, time to distant metastasis and overall or disease specific survival outcomes.
Although the genes outlined above are found to be critically involved in breast cancer progression, the extent of their contribution in mediating VN- and the IGF:VN complex-induced biological effects was unknown. Hence, a loss-of-function approach was utilised to identify their role in breast cancer progression. Select target genes were suppressed using shRNA and their effect on VN- and IGF:VN complex- induced cell survival and migration were analysed. To this end, secreted ECM- associated matricellular CCN family proteins, Cyr61/CCN1 and CTGF/CCN2, were
Chapter 6: General Discussion 192
selected as our lead candidates to investigate their functional role in VN- and IGF- I:IGFBP:VN complex-induced breast cancer cell survival. Cyr61 and CTGF both have roles documented in the enhancement of breast cancer cell growth and survival as well as migration (Menendez et al., 2005; Chen et al., 2007; Pandey et al., 2009; Wang et al., 2009). In addition, the overexpression of Cyr61 and CTGF is found to strongly correlate with breast cancer progression and tumorigenesis, and is often associated with advanced clinical stage of the disease (Xie, Nakachi, et al., 2001; Tsai et al., 2002; Kang et al., 2003). Indeed, the inhibition of Cyr61 using a novel monoclonal antibody was reported to significantly supress the in vivo tumour size and proliferation of the highly metastatic MDA-MB-231 breast xenograft models (Lin et al., 2012). Overexpression of the CTGF protein has also been strongly associated with resistance against drug-induced apoptosis in both MCF-7 and MDA- MB-231 breast cancer cells (Wang et al., 2009; Lai et al., 2011). Additionally, induced expression of CTGF by hydroxytamoxifen, an ER antagonist, has been shown to enhance the proliferation of poorly differentiated SKBr3 breast tumour cells (Pandey et al., 2009).
Once the candidate genes Cyr61 and CTGF were selected, 2D monolayer and 3D MatrigelTM assays were used to assess the impact of Cyr61 and CTGF down- regulation on VN- and IGF-I:IGFBP:VN-induced MCF-7 and MDA-MB-231 breast cancer cell growth/survival. For Cyr61, our results clearly indicate that it is involved in VN- and IGF-I:IGFBP:VN complex-induced MCF-7 and MDA-MB-231 breast cell survival in both 2D monolayer and 3D cultures. This was shown by a significant reduction in the relative size and growth of the MCF-7 cell spheroids in which levels of Cyr61 were attenuated by shRNA. Cyr61 is known to play an important role in cell proliferation and inhibition of apoptosis (Gery et al., 2005; Todorovic et al., 2005), and the expression of Cyr61 in MCF-7 cells is thought to be crucial in maintaining cell proliferation (Sampath et al., 2001) and resistance to drug-induced apoptosis (Lin et al., 2004). Our results are consistent with these findings, and we demonstrate that down-regulation of Cyr61 can effectively reduce VN- and the IGF- I:IGFBP:VN-stimulated MCF-7 cell proliferation. Altogether, it is possible that Cyr61 expression can influence key pathways involved in cell cycle progression and anti-apoptosis, thereby stimulating cell proliferation and survival induced by VN and the IGF-I:IGFBP:VN complex. To validate this, cell signalling studies investigating
Chapter 6: General Discussion 193
the involvement of Cyr61 in the apoptotic pathways and cell cycle pathways such as the PI3-K/AKT, MEK/ERK, and small GTPase (Rho and Rac) will be required in the future.
Similar to Cyr61, down-regulation of CTGF also significantly decreased MCF-7 cell growth and viability. However, in contrast to MCF-7 cells, growth and viability of the MDA-MB-231 cells was relatively unaffected by CTGF down- regulation. Instead, in MDA-MB-231 cells reduced expression of CTGF reversed their invasive phenotype in 3D MatrigelTM cultures. The characteristic MDA-MB- 231 invasive spindle-like morphology was transformed to a less invasive and rounded morphology. This finding highlights that CTGF plays a role in the invasive morphology of MDA-MB-231 cells, which in turn may be responsible for its metastatic ability in vivo. Indeed, CTGF is known to play a critical role in breast cancer bone metastasis (Kang et al., 2003; Kang et al., 2005) and its down-regulation has been recently reported to inhibit the metastasis of MDA-MB-231 cells to bone in mouse xenograft breast tumours (Ren et al., 2014).
Based on these results demonstrating MDA-MB-231 cells had a less invasive and migratory phenotype in 3D cultures upon CTGF suppression, we shifted our focus in subsequent studies to identify whether or not the down-regulation of Cyr61 and CTGF has an effect on VN- and IGF-I:IGFBP:VN-stimulated cell migration. Results from the migration assays revealed that CTGF down-regulation significantly decreased VN- and IGF-I:IGFBP:VN-induced migration of both the MCF-7 (relatively low migratory phenotype) and MDA-MB-231 (highly motile phenotype) cell lines. In contrast, the migration of these cells with reduced Cyr61 expression remained relatively unaffected. These results, in combination with previous findings, suggest that Cyr61 may be more predominantly involved in the growth and viability of breast cancer cells, whereas CTGF is more involved in the invasive/migratory behaviour of breast cancer cells in response to VN and the IGF-I:IGFBP:VN complex.
The role of CTGF in mediating the invasive morphology and conferring migratory abilities may be related to its reported role in epithelial-to-mesenchymal transition (EMT), a process associated with the increase in the ability of epithelial cells to become more migratory and invasive. CTGF is known to reduce cytokeratin expression and trigger the expression of mesenchymal markers Vimentin and FN
Chapter 6: General Discussion 194
(Burns et al., 2006; Liu et al., 2006), characteristic markers of EMT-induction. While beyond the time and scope of the current project, future studies are planned to investigate classical markers of EMT, including E-cadherin, Epcam, Vimentin and FN in RNA and protein samples we have already collected following suppression of CTGF in the mesenchymal MDA-MB-231 cell line. In support for this our microarray study revealed Slug, a well-known EMT-inducing transcription factor (Bolos et al., 2003) to be up-regulated in our microarray analysis by 12.3- and 14.8- fold at the 48 hour time point by VN and the IGF-I:IGFBP:VN complex, respectively (Figure 4.2). Hence, future research will be directed towards delineating the role of Slug in the VN/IGF/CTGF axis.
In addition to CTGF, VN and the IGF-I:IGFBP:VN complex increased the gene expression of TGFβ2 (~10 fold, identified by qRT-PCR) and Smad proteins (Smad 4 and 6). TGFβ is known to be one of the key regulatory elements that induce the expression of CTGF through Smad and Ras/MEK/ERK signalling axis in a number of cell types (Chen et al., 2002; Leivonen et al., 2005; Secker et al., 2008), including breast cancer (Giampieri et al., 2009). Interestingly, studies performed by Kang et al. (2003 and 2005) revealed that the transcriptional activation of CTGF and its elevated expression by TGFβ/Smad-dependent pathways can contribute to the bone metastasis of breast cancer in vivo. Based on this result, it may be postulated that VN- and the IGF-I:IGBFP:VN complex-induced expression of CTGF and resultant migratory responses may also be a consequence of the up-regulation of TGFβ and Smad genes that were observed in our study. To explore this, further studies involving function blocking antibodies and small molecule inhibitors targeting the TGFβ/Smad pathway will be necessary to elucidate its role in the expression of CTGF and breast cancer cell migration.
In the broader context of cancer progression, Cyr61 and CTGF also have well established roles in regulating angiogenesis and endothelial cell function (reviewed in Brigstock, 2002). This is observed in breast cancer wherein Cyr61 and CTGF expression was associated with increased angiogenesis (Tsai et al., 2002; Chien et al., 2011), through the stimulation of VEGF expression and secretion in breast cancer cells (Espinoza et al., 2014). Recent studies indicate that VN also promotes angiogenesis through the regulation of VEGF (Li et al., 2014), suggesting the possibility that Cyr61 and CTGF up-regulation by VN may also play an important
Chapter 6: General Discussion 195
role in stimulating angiogenesis, and thereby increasing the survival of breast tumour cells. As there is a concomitant increase in expression of VN in the sub-endothelial area of blood vessel walls surrounding malignant breast tumours (Abbott, 2003; Kadowaki et al., 2011), it is plausible that VN-induced regulation of Cyr61 and CTGF may have roles in tumour angiogenesis to facilitate tumour growth and progression. In future studies, these questions may be better answered through angiogenesis assays such as in vitro tube formation assays using endothelial cells cultured in Matrigel™ or in vivo chick chorioallantoic membrane assays where the chorioallantoic membranes are treated with conditioned media from normal breast cancer cells and those that overexpress Cyr61 and CTGF in response to VN and the IGF-I:IGFBP:VN complex.
In addition to unravelling the functional importance of Cyr61 and CTGF in mediating biological processes induced by VN and the IGF-I:VN complex, studies presented in this thesis have revealed novel mechanisms underpinning their regulation in breast cancer cells. Cell signalling studies presented in Chapter 4 demonstrate that VN and the IGF-I:IGFBP:VN complex stimulate the expression of
Cyr61 and CTGF via ligation of αv-subunit containing-integrins and subsequent recruitment and activation of the Src/FAK signalling axis. Furthermore, the regulation of Cyr61 and CTGF was observed to bifurcate at PI3K/AKT and ERK/MAPK signalling pathways, respectively, downstream of Src. Previous reports have demonstrated that Cyr61 and CTGF can mediate their downstream effects through interaction with cell surface integrins and heparan sulfate proteoglycans (HSPG) (Syndecan 4, perlecan receptors) as well as growth factor receptors (Leask & Abraham, 2006; Holbourn et al., 2008; Jun & Lau, 2011). These interactions can lead to activation and recruitment of downstream signalling cascades, including the FAK/Src complex, Rac/JNK, PI3-K/AKT, MEK/ERK/MAPK, and NF-κB signalling pathways (Secker et al., 2008; Wang et al., 2009; Goodwin et al., 2010; Chuang et al., 2012).
Taking the evidence gathered together, it is proposed that the mechanism by which VN and the IGF-I:IGFBP:VN complex enhance expression and secretion of
Cyr61 and CTGF is through the αv/Src/PI3-K/ERK signalling pathways, with ERK predominantly mediating the expression of CTGF and PI3-K mediating Cyr61 expression. Furthermore, we postulate that upon stimulation of Cyr61 and CTGF,
Chapter 6: General Discussion 196
they are secreted and can act in a positive feedback loop in an autocrine fashion through their interactions with integrins, HSPG and growth factor receptors to modulate cellular functions (Figure 6.1).
Figure 6.1:VN and the IGF-I:IGFBP:VN complex-induced Cyr61 and CTGF signalling (identified by studies reported herein) in addition to signalling by secreted Cyr61 and CTGF (identified by others).
VN and the IGF:IGFBP:VN complex mediate expression of Cyr61 and CTGF through interactions with αv-integrin and activation and recruitment of Src signalling axis. Their regulation is then bifurcated downstream of Src where PI3K/AKT stimulates the expression of Cyr61 and ERK/MAPK regulates the expression of CTGF. Studies by others (Chuang et al., 2012; Goodwin et al., 2010; Secker et al., 2008; Wang et al., 2009) demonstrated that the secreted Cyr61 and CTGF proteins bind cell surface integrins, the heparan sulfate proteoglycan (HSPG) and growth factor receptors (Holbourn et al., 2008; Jun & Lau, 2011; Leask & Abraham, 2006). This would then activates and recruits downstream signalling mediators, including FAK/Src complex, Rac/JNK, PI3-K/AKT, MEK/ERK/MAPK, and NF-κB to mediate the biological effects such as cell proliferation, survival, migration and angiogenesis.
Chapter 6: General Discussion 197
In addition to elucidating the mechanisms of CTGF and Cyr61 expression and function, network analyses performed in Chapter 3 identified a number of additional genes and pathways that may also contribute to VN- and the IGF:VN- mediated functions. For instance, the NF-κB pathway was identified to be one of the top functional networks associated with our VNGES (established from VN and IGF:VN regulated genes at the 48 hour time point). This is in accordance with published reports, which indicate that adhesion and migration of pancreatic carcinoma cells on VN is due to the latent activation (24 hours) of signals leading to the induction of NF-κB-mediated gene transcription (Yebra et al., 1995). This suggests the possibility that in addition to Src/AKT/ERK pathways identified in this study, NF-κB may also be involved in regulation of Cyr61 and CTGF expression. As such, JNK is known to be activated downstream of the integrin/FAK/NF-κB pathway. Indeed, in our dataset, the JNK pathway overlapped with networks containing the FAK pathway. Moreover, c-JUN, the transcription factor downstream of the JNK pathway, was also identified to be significantly up-regulated by VN and the IGF:VN complexes, providing evidence for the potential involvement of the NF- κB/JNK pathway. Nonetheless, future studies involving specific inhibition of these pathways will be critical to determine their contribution to VN- and IGF- I:IGFBP:VN-induced expression of Cyr61 and CTGF and resultant biological effects.
Drug resistance remains one of the major challenges in designing effective cancer therapeutics. In addition to the effect of the ECM in promoting cancer cell survival, studies suggest that integrin-mediated attachment of malignant cells to ECM components can strongly enhance their resistance to chemotherapeutic agents (reviewed in Elliott & Sethi, 2002; Correia & Bissell, 2012). It has been reported that expression of Cyr61 and CTGF enhances the resistance of breast cancer cells to various chemotherapeutic drugs, in particular Taxols (Lin et al., 2004; Menendez et al., 2005; Wang et al., 2009). TAZ (transcriptional co-activator with PDZ-binding motif), a novel gene implicated in breast cancer drug resistance, was found to induce Taxol resistance through activation of Cyr61 and CTGF gene promoters (Lai et al., 2011). Conversely, Cyr61 and CTGF suppression completely reversed TAZ-induced Taxol resistance in breast cancer cells. Studies reported herein demonstrate that integrin-binding ECM components such as VN, FN and type I collagen can further
Chapter 6: General Discussion 198
stimulate the expression of Cyr61 and CTGF in breast cancer cells. Collectively, it is possible that ECM components such as VN can induce the expression and secretion of Cyr61 and CTGF through interactions with integrins, and further contribute to drug resistance. Importantly, cell studies performed herein reveal that blockade of αv- integrin inhibits the VN-stimulated expression of Cyr61 and CTGF. Therefore, using an αv-integrin inhibitor alone or in combination with Taxol may provide better treatment outcomes for breast cancer patients with high expression of Cyr61 and
CTGF. Future studies directed at testing the effectiveness of αv-integrin blockade alone or in combination with Taxol in Cyr61/CTGF over-expressing breast cancer xenografts will be important. Outcomes from such a study may reveal new therapeutic combinations, either in the adjuvant setting or to prevent relapse post primary therapy in patients with increased levels of VN, Cyr61and CTGF.
Over the past decade, gene expression profiling has become increasingly utilised to predict and characterise breast cancer molecular subtypes and their clinical outcome (Perou et al., 2000; van 't Veer et al., 2002; Sorlie et al., 2003; Neve et al., 2006; Prat et al., 2010). To investigate the broader relevance of the VNGES across relevant breast cancer subtypes, hierarchical clustering was employed to analyse the expression of the VNGES across 51 cell lines representing the major breast cancer subtypes using gene expression data published by Neve et al, 2006. This revealed the VNGES to be highly associated with basal B (Mesenchymal) cell lines, whereby the VNGES could cluster together the majority of the highly metastatic basal B cell lines apart from the less-metastatic luminal and basal A cell models. Further examination of these cell lines revealed them to be predominantly claudin-low cell lines (as described by Prat et al., 2010). The claudin-low subtype is reported to be enriched in markers of EMT and cancer stem cell-like features (Prat et al., 2010). Given these characteristics, combined with claudin-low tumours/cell lines being triple negative (PR-, ER- & Her2-), they are associated with therapeutic resistance and poor survival outcomes (X. Li et al., 2008; Creighton et al., 2009; Gupta et al., 2009). Based on the VNGES expression, all 9 known claudin-low cell lines clustered together in their own distinct group. Interestingly, VNGES was just as efficient in clustering together the claudin-low cell lines as the originally published claudin-low predictor signature generated by comparing the intrinsic molecular differences between the claudin-low subtype with the basal-like and luminal sub-types (Prat et al., 2010).
Chapter 6: General Discussion 199
A number of VNGES genes, including Cyr61, CTGF, MBNL2, LAMB3, ACTG2, MAOA, slug and IGFBP6 are also featured within the 9-cell line claudin- low predictor signature that was developed by Prat and colleagues to characterise and predict the claudin-low phenotype in human breast cancer samples (Part et al., 2010). Moreover, using the Prat et al. 2010 data sets, Cyr61 and CTGF were identified to be frequently up-regulated in claudin-low breast cancer sub-types. The association of the VNGES with the claudin-low subtype is a novel and significant finding, given that it was developed using genes commonly regulated by VN in the luminal MCF-7 cell model. To our knowledge this is the first time a component of the ECM has been reported to induce luminal breast cancer cells to transition to a claudin-low-like phenotype. While an array of studies will be required to confirm this result, targeting the VN/CTGF/Cyr61 axis may provide a targeted treatment option for claudin-low tumours. Currently patients diagnosed with claudin-low tumours have no such targeted therapies available. Supporting this rationale are the findings in the MDA- MB-231 cell line following suppression of Cyr61 and CTGF reported above. MDA- MB-231 is a prototypical claudin-low cell model and supports the potential therapeutic targeting of the VN/CTGF/Cyr61 axis in this breast cancer subtype.
In summary, this PhD project has identified a molecular signature involved in VN- and the IGF-I:IGFBP:VN complex-mediated functions critical to breast cancer progression and metastasis. In particular, the findings from this project have identified Cyr61 and CTGF as key mediators involved in VN- and IGF-I:IGFBP:VN- induced breast cancer cell survival and migration. By identifying the molecular mediators responsible for the functional effects of VN and the IGF:VN complex, we have revealed potential targets for inhibiting the biological functions of VN and the IGF-I:VN complex. Further refinement of the panel of biomarkers (VNGES) identified in this process may help provide prognostic and predictive indicators to assist the clinical management of breast cancer, which may be particularly relevant to the claudin-low subtype of human breast cancers. In addition, the inhibition of key molecular targets identified, namely Cyr61 and CTGF, that are important in VN- and IGF-I:VN complex-induced biological effects, may have significant implications in designing future therapeutic targets for patient with tumours overexpressing VN and/or the components of the IGF:VN axis.
Chapter 6: General Discussion 200 Appendices
Appendix 1: Integrity and quality check of RNA samples prior to array analysis on the Agilent Bioanalyser 2100 using the NanoChip protocol.
Appendices 201
Appendix 2: Details of primers designed using NCBI Primer-BLAST.
Primes Gene RefSeq ID Symbol (NCBI) Sequence (5’3’) Length Start Stop Tm GC% Self 5’ Self 3’ Product complementary complementary size JUN NM_002228.3 F: TTGCCAGAGCCCTGTTGCGG 20 967 986 59.90 65.00 6.00 2.00 125 R: TCGGACGGGAGGAACGAGGC 20 1108 1089 60.87 70.00 7.00 2.00
FOS NM_005252.3 F: CCTGTCAACGCGCAGGACTTCT 22 332 353 59.42 59.09 4.00 3.00 174 R: GGCGGGGACTCCGAAAGGGT 20 505 486 60.53 70.00 5.00 0.00
TGFβ2 NM_003238.3 F: ACAGCACCAGGGACTTGCTCCA 22 1583 1604 60.05 59.09 4.00 1.00 147 R: TGGGCGGGATGGCATTTTCGG 21 1729 1709 60.50 61.90 5.00 1.00
VAV2 NM_003371.3 F: CCGGTCCCCAGTGTTCACGC 20 2353 2372 59.97 70.00 4.00 2.00 167 R: TCCGTCCGTTGGTCTCGCCC 21 2519 2500 60.87 70.00 2.00 2.00
CTGF NM_005080.3 F: CCTGGTCCAGACCACAGAGT 20 345 364 59.77 70.00 5.00 3.00 168 R: TGGAGATTTTGGGAGTACGG 20 512 492 59.37 61.90 4.00 2.00
HSBP8 NM_014365.2 F: GCTGGGAGCCTGTCAGTTTATGA 24 1730 1753 58.77 54.17 5.00 2.00 179 R: ACACACGTCCCGCCCATATACA 22 1908 1887 57.80 54.55 4.00 2.00
ANXA2 NM_001002857 F: CCCGGAGAGTTTCCCGCTTGG 21 82 102 59.45 66.67 5.00 0.00 179 R: TGTTCAAAGCATCCCGCTCAGCA 23 260 238 59.38 52.17 5.00 3.00
XBP1 NM_005080.3 F: GACAGCGCTTGGGGATGGATGC 22 449 470 60.43 63.64 6.00 2.00 125 R: CTGCTGCAGAGGTGCACGTAGT 22 573 552 59.17 59.09 7.00 2.00
CD24A NM_013230.2 F: GCACTGCTCCTACCCACGCA 20 159 178 59.07 65.00 5.00 0.00 146 R: CCCAGGTACCAGGACTGGCTGG 21 304 284 59.51 61.90 6.00 2.00
NFIB NM_005596.3 F: CCTCCACCTTCACCGTTGCCA 21 1616 1636 58.97 61.90 4.00 3.00 179 R: CCCAGGTACCAGGACTGGCTGG 22 1794 1773 60.24 68.18 8.00 3.00
Appendices 202
Primes Gene RefSeq ID Sequence (5’3’) Length Start Stop Tm GC% Self 5’ Self 3’ Product Symbol (NCBI) complementary complementary size H19 NR_002196.1 F: AAGACGCCAGGTCCGGTGGA 20 1365 1384 59.83 65.00 4.00 3.00 160 R: TGTGCCGACTCCGTGGAGGA 20 1543 1522 59.55 65.00 6.00 2.00
SPDEF NM_012391.1 F: TCTGACCCCAGCATTTCCAGAGCA 24 1692 1715 59.87 54.17 4.00 0.00 185 R: ATTATCCATTCCCGGGGGCACT 22 1876 1855 57.46 54.55 6.00 2.00
GABRP NM_014211.1 F: GAACCTATGTTCAGGTTGCGGGT 23 2720 2742 57.04 52.17 8.00 0.00 128 R: AACTGGAAGCCTTTACCCTGGCA 23 2847 2825 58.16 52.17 5.00 3.00
RASGRP1 NM_005739.2 F: ACAGTCCAGCAAGGGCCAAAGA 22 4554 4575 58.35 54.55 4.00 0.00 129 R: ACTAGGCTTGATAAGCCAAAGCC 25 4683 4659 57.33 48.00 8.00 3.00
Cyr61 NM_001554.3 F: ATGCTACCTGGGTTTCCAGGGCA 24 1354 1377 60.11 54.17 6.00 2.00 190 R: TTAGTGTCCATCCGCACCAGCA 24 1543 1522 58.25 54.55 3.00 0.00
NBPF20 NM_0010376 F: AGGCTCAACGGCGTGCTGAT 20 2661 2680 59.07 60.00 5.00 2.00 164 R: AGGGCGAAGCTGATGTGCTGT 21 2824 2804 58.77 57.14 4.00 2.00
PAK4 NM_005884.3 F: AACCTGCACAAGGTGTCGCCAT 22 1844 1856 59.28 54.55 4.00 3.00 113 R: TTGGCCAGGAATGGGTGCTTCA 22 1956 1935 58.67 54.55 6.00 1.00
LAMB1 NM_002291.1 F: AGCAAGCAAAAGTGCAACAGATG 25 5073 5097 57.04 44.00 4.00 3.00 200 R: TGGGACGCGTTGAACAAGGTTT 22 5272 5252 57.57 50.00 6.00 2.00
MTSS1 NM_014751.4 F: GGGCACATTTCTGACCGAACCCT 24 4203 4226 60.11 54.17 5.00 1.00 200 R: CCTTGAGCAATCGCAGCAGTGT 22 4402 4381 58.10 54.55 4.00 1.00
FZD8 NM_031866.1 F: AGTGCTTGCTAAGGACGCATGG 22 2870 2891 57.58 54.55 7.00 2.00 177 R: ACCGTCTTTGGGGAAGGTGCAAA 23 3046 3024 58.93 52.17 4.00 0.00
Appendices 203 Appendix 3: The correct size of the primers was confirmed by running 5 µl of PCR products on a 2% agarose gel at 110V for 30 minutes and visualizing with SYBR safe fluorescence staining under UV illumination. For each product two wells were used to confirm the correct size of the product. DNA molecular marker XIV (Roche) was also run in parallel.
Appendices 204
Appendix 4: Vector map for the lentiviral packaging plasmid pCMV-delta R8.2. Obtained from addgene.org.
Appendices 205
Appendix 5: Vector map for the lentiviral packaging plasmid pCMV-VSV-G. Obtained from addgene.org.
Appendices 206 Appendix 6: Molecular and cellular functions associated with differentially expressed genes in response to VN at 12 and 48 hour time points.
VN vs SFM (12 hours)
VN vs SFM (48 hours)
Appendices 207
Appendix 7: Molecular and cellular functions associated with differentially expressed genes in response to IGF-II:VN at 12 and 48 hour time points.
IGF-II:VN vs SFM (12 hours)
IGF-II:VN vs SFM (48 hours)
Appendices 208
Appendix 8: Molecular and cellular functions associated with differentially expressed genes in response to IGF-I:IGFBP-5:VN at 12 and 48 hour time points.
IGF-I:IGFBP-5:VN vs SFM (12 hours)
IGF-I:IGFBP-5:VN vs SFM (48 hours)
Appendices 209
Appendix 9: Top five functional networks associated with differentially expressed genes in response to VN at 12 hours. Genes are colour coded according to their gene expression determined by the microarray analysis (red, up-regulated; green, down-regulated). Genes in black colour were either not on the expression array or not significantly regulated. Focus ID Molecules in Network Score Top Diseases and Functions Molecules
ACHE, Akt, Alp, ANKRD1, CAMK2B, CDH16, COL12A1, collagen, Collagen Alpha1, CTGF, CYR61, Connective Tissue Development and 1 estrogen receptor, FABP5, GPIIB-IIIA, Growth hormone, GST,GSTA1, GSTM2, JUN/JUNB/JUND, 36 21 Function, Embryonic Development, KLF10, KRT20, Laminin, mir-21, NT5E, PARM1, PIK3IP1, PLAGL1, PLAUR, PRODH, Rock, Rxr, RXRG, Organ Development Smad, Tgf beta,TGFB2
ACTG2, Ap1, ARPC4, BCR (complex), CDKN2B, CEACAM5, Cg, Collagen type I, Collagen type II, Developmental Disorder, Hereditary 2 29 18 CXCR4, CYP26B1, ELF5,ERK1/2, FSH,HHAT, KCNN4, Lh, MAP2K1/2, MATK, Notch, PI3K (family) ,Pki Disorder, Skeletal and Muscular ,PKIA , Rap1, RASGRP1, RUNX2, Sapk, SLC12A4, Smad1/5/8, SNAI2, SOX9, TCF, TIPIN, Wnt, YY2 Disorders
ALDH3B2, ATRX, COLEC12, Creb, Cyclin A, FOS, Hdac, HIST1H4A (includes others), HIST2H3C DNA Replication, Recombination, 3 (includes others), HISTONE, Histone h3, Histone h4, HSPB8, IgG, Igm, IL12 (complex), 29 18 and Repair, Gene Expression, Immunoglobulin, KMT2A, KRT13, LCAT, LDHA, LDL, Mek, MYCT1, NFkB (complex) ,PAK1IP1 Dermatological Diseases and ,PSG5,QKI ,Rb ,SH3RF1 ,SLURP1, Sos, TAF1A, TSH, Ubiquitin Conditions
ABHD17B, AGRP, ARID3A, C5orf34, CCP110, CEP290, CETN1, CFL2, CKAP4, CYR61, DICER1, Cellular Development, Nervous 4 ENDOV, FAM216A, FAM81A, KLHL7, MAG, MTFP1, MXRA5, NANOS1, PIWIL1, PRPSAP2, PUM2, 25 16 System Development and Function, RBM38, RCCD1, RREB1, SERPINF1, SLC25A30, SLC39A3, SPG20, STAU2, TMEM64, TRIM71, UBC, Hereditary Disorder USP49, ZNF493
APP, ASPH, BHMT2, C1QL1, CCDC25, CDK9, CLUAP1, CYP26B1, DMD, DNAJC14, ELAVL1, ERMP1, Nervous System Development and 5 25 16 FAM110A, FAM134B, FYN, GNAT2, GPM6B, HSP90AA1, KCNS3, KLHL3, MALL, MATK, NAPG, Function, Organ Morphology, Tissue NR2E3, NUAK1, OLFM1, PGM5, PI4K2B, RN7SK, SATB1, TBP, TMTC4, TSPAN12, VKORC1L1, WNK2 Morphology
Appendices 210
Appendix 10: Top five functional networks associated with differentially expressed genes in response to VN at 48 hours. Genes are colour coded according to their gene expression determined by the microarray analysis (red, up-regulated; green, down-regulated). Genes in black colour were either not on the expression array or not significantly regulated. Focus ID Molecules in Network Score Top Diseases and Functions Molecules
Embryonic Development, ACTG2, Akt, ANGPTL4, ANKRD1, ARPC4, ATP2B1, BMF, CADM1, CGA, FAM162A, FSH, GFRA2, GK, HK2, HPGD, Nervous System 1 IGFBP6, ITGB5, KRT20, Lh,LIN7A, MPP6, MTSS1, NUCB2, PARM1, PLAGL1, PRODH, PROK1, PTPase, PTPN21, Development and Function, PVRL3, RAP1GAP, ROBO1, SLIT2, STARD13, Vdac 47 27 Organ Development
ABLIM3, AJUBA, AMOT, AMPH, Calcineurin A, CDC14A, CEP70, DHODH, DLEU1, DNAH1, E2f, EPS8L1, F Actin, 2 FANCD2, GAD1, Gamma tubulin, GTPase, KIF23, LYNX1, MLKL, MT2A, MYH9, Myosin, NFkB (complex), Cell Cycle, Cellular Nicotinic acetylcholine receptor, PDLIM1, PKP1, PRMT2, RACGAP1, RFC3, RGS11, S100A4, S100P, TMOD2, 39 27 TRAIP Movement, Cancer
Connective Tissue ACHE, AChR, ALDOC, ARL4A, ASAP1, CD68, COL11A2, COL18A1, COL4A5, COL5A1, Collagen type II, Collagen Development and Function, 3 type IV, Collagen(s), CRAT, CYR61, DCLK1, ERK, GPIIB-IIIA, GPX2, LAMB1, LAMB3 ,Laminin, Laminin1, MATK, Embryonic Development, MT1F, NT5E, PPP1R15A, SERPINA3, SERPINA5, SNAI2, SPDEF, TACC1, TCF, TSKU, UACA 37 25 Organ Development
Cellular Assembly and BAG1, C1QL1, CAV1, CAV2, Caveolin, COL12A1, CRABP1, Dynamin, estrogen receptor, FOS, GAL, Growth Organization, Cellular 4 hormone, GST, GSTM3, HBA1/HBA2, Hdac, HHEX, HSPB8, KLF6, KLF10, KRT13, LDL-cholesterol, MALL, MT1A, Function and Maintenance, MT1H, PKIB, PREX1, PSG5, QKI, SLURP1, Sos, SPRY2, STAT5a/b, STX11, TGFBI 37 25 Infectious Disease
ABCC3, ABCC4, Alp, CASC5, COCH, Collagen Alpha1, CPE, Ctbp, CTGF, CYP26B1, DKK1, DLC1, ELF5, FABP5, H19, Drug Metabolism, Lipid 5 Hedgehog, HOPX, Igfbp, ISG20, KCNN4, LRP, MAOA, MBNL2, NPNT, PI3K (complex), PLAUR, Rb, SALL4, 35 25 Metabolism, Molecular SCNN1A, Sox, SOX2, SOX7, SOX9, UVRAG, Wnt Transport
Appendices 211
Appendix 11: Top five functional networks associated with differentially expressed genes in response to IGF-II:VN at 12 hours. Genes are colour coded according to their gene expression determined by the microarray analysis (red, up-regulated; green, down-regulated). Genes in black colour were either not on the expression array or not significantly regulated. Focus ID Molecules in Network Score Top Diseases and Functions Molecules
ACOT11, Alpha catenin, ARPC4,BAG1, CBR3, CDC25C, CDKN2B, Cg, CGA, CXCR4, FABP5, FILIP1L, FSH, Endocrine System Disorders, 1 HDL-cholesterol, IER3,ISG20, KLF10, LDHA, Lh, NFkB (complex), Notch, Nr1h, NR1H3, NR5A2, PAK1IP1, 41 23 Organismal Injury and Abnormalities, PI3K (family), PKP1, PTPN14, Rxr, RXRG, SH3RF1, TDRD5, WDR33, WWOX, YY2 Reproductive System Disease
ADCY5, Akt, ANKRD1, APC (complex), BCR (complex), CADM1, CD79B, CDC20, Cdk, Cyclin A, Cyclin B, Cell Cycle, Cellular Assembly and 2 Cyclin D, Cyclin E, E2f, ESPL1, HPGD, IGSF1, KCNN4, KNSTRN, KRT13, LETMD1, MPP6, PARM1, PIK3IP1, 36 20 Organization, DNA Replication, Pki, PKIA, PLAGL1, PRODH, PRR11, PTTG1, PVRL3, RAP1GAP, SLURP1, Sos, Vla-4 Recombination, and Repair
ABCC9, Alp, CDH16, CDK2AP2, CEACAM5, COL11A2, Collagen Alpha1, Collagen type I Collagen type II, CTGF, CYR61, ELF5, ERK1/2, FOXP4, GPIIB-IIIA, H19, HBA1/HBA2, HMMR, HSF4, Laminin, Lefty, LYPD3, Skeletal and Muscular System 3 mediator, mir-24, RASGRP1, RUNX2, SLC12A4, Smad, Smad1/5/8, Smad2/3, SNAI2, SOX9, Tgf beta, 34 18 Development and Function, Tissue TGFB2,TGFBI Development, Connective Tissue
26s Proteasome, caspase, CD3, CENPA, CENPM, CENPP, CLTCL1, DHRS4,ENSA, GPM6B, GSG2, GTF2A1, 4 Hdac, HIRIP3, Histone h3, Histone h4, Hsp90, IgG1, Igm, IL12 (complex), Immunoglobulin, Interferon Development and Function alpha, KLF6, LCAT, LDL-cholesterol, MHC Class II (complex), NUDT1, PER3, PSEN1, PSMD9, RBPJ, 29 16 SLC30A1, SOX7, TOM1L2, Ubiquitin
ABCE1, Actin, ALDH3B2, AMPK, ANG, ASPH, CaMKII, COTL1, CRAT, Ctbp, DNASE1, ERK, F Actin, GPX2, 5 HISTONE, IgG, ITPR2, MECOM, Pdgf (complex), PDGF BB, PHF19, PLAUR, Pld, PLD1, PP2A, Rap1, Rock, 25 15 Antimicrobial Response, Inflammatory Ryr, SRI, STAT2, TCF, TGIF1, TH, TOB1, TSH Response, Cancer
Appendices 212
Appendix 12: Top five functional networks associated with differentially expressed genes in response to IGF-II:VN at 48 hours. Genes are colour coded according to their gene expression determined by the microarray analysis (red, up-regulated; green, down-regulated). Genes in black colour were either not on the expression array or not significantly regulated. Focus ID Molecules in Network Score Top Diseases and Functions Molecules
Cell Cycle, Cell Death and AMPH, ANXA2, CA12, CDC25C, CDCA2, CDK6, CDKN1A, CDKN2B, CHEK2, Cyclin A, Cyclin D, Cyclin E, DLEU1, 1 Survival, Post-Translational E2f, ERCC6L , HAUS8, HIST1H4A (includes others), IER3, MBNL2, MELK, NFkB (complex), Notch, OPTN, 37 28 Modification PDLIM1, PKMYT1, PLK1, RAB31, RELB, RNA polymerase II, S1PR3, SLC27A2, SMAD6, SOD2, ST3GAL1
ACTA2, AKR1C1/AKR1C2, B3GNT1, CAV1, CAV2, CITED2, COL4A5, COL5A1, Collagen Alpha1, CTGF, CXCR4, Cellular Movement, Cancer, 2 DCLK1, DLC1, ELF5, Focal adhesion kinase, FZD8, GPX2, H2AFX, HISTONE, HPGD, IGFBP6, MALL, MAP2K1/2, 37 26 Cellular Development P38 MAPK, PBK,PI3K (family), PPARG, PRODH, RUNX2, SCNN1A, Smad2/3, SNAI2, SPDEF, TAGLN, TP63 Cellular Assembly and ACTG2, ADAM15, APOBEC3F, ARPC4, ATP2B1, AUTS2, CD24, CLTCL1, Creb, CTSC, FABP5, FILIP1L, FSH, Organization, Gene Expression, 3 HIST2H3C (includes others), Histone h3, Histone h4, IgG, IL15, IL12 (complex), Interferon alpha, ISG20, DNA Replication, KIF18A, KIF4A, MT1H, NCAPG2, PDE3B, Pkc(s), PPIF, PTPN21, RND2, RUNX1T1, SMC2, TCR, TFF1, TRAIP 33 26 Recombination, and Repair
Ap1, CEACAM5, Cg, Ctbp, CYR61, DAB2, DKK1, ENC1, ERK, ERK1/2, FOS, FZD2, H19, Hdac, ITGB5, JUN, KLF6, Cancer, Cellular Development, 4 Cellular Growth and Mek, mir-24, MT2A, PALLD, PDGF BB, Pka, PKIB, PLAUR, PPP2R2A, PREX1, PSG5, PTGER2, Rac, RASGRP1, 31 23 Proliferation SFTPD, SOX9, SPRY2, TH Cell Death and Survival, Organ Akt, ALOXE3, ANG, ANKRD1, ASB9, BAG1, BCR (complex), BMF, CD3, CD68, COL12A1, estrogen receptor, 5 Morphology, Nervous System GSTA1, Hsp27, Hsp70, Hsp90, IGSF1, Jnk, LDL, Mapk, mir-21, NEDD9, p85 (pik3r), PCDH7, PI3K (complex), 24 21 Development and Function PIK3IP1, PLIN2, PRKCA, PSRC1, RAP1GAP, Ras, RNF13, S100A4, TGFB2, Vegf
Appendices 213
Appendix 13: Top five functional networks associated with differentially expressed genes in response to IGF-I:IGFBP-5:VN at 12 hours. Genes are colour coded according to their gene expression determined by the microarray analysis (red, up-regulated; green, down-regulated). Genes in black colour were either not on the expression array or not significantly regulated. Focus Molecules in Network ID Score Molecules Top Diseases and Functions
ACTG2, Actin, Akt, ALDH3B2, ANG, ANKRD1, ARPC4, BCR (complex), CAMK2B, CD79B, CORO1A, Drug Metabolism, Glutathione 1 COTL1, CYP2E1 , EPS8L1, F Actin , FABP5, GST, GSTA1, GSTM2, Ige, IgG, Immunoglobulin, MAP1B, 37 22 Depletion In Liver, Cell Death and Nfat (family), Notch, PARM1, PIK3IP1, Rap1, RASGRP1, RELB, Rock, Rxr, RXRG, SAMD4A, SMOX Survival
Alp, ARHGAP4, CDH16, CDKN2B, CEACAM5, COL11A2, collagen, Collagen Alpha1, Collagen type I, Skeletal and Muscular System Collagen type II, CRAT, CTGF, CYR61, DAB2, ELF5,E RK1/2,ETS, HHAT, KLF10,Laminin, mir-24, NT5E, Development and Function, Tissue 35 21 2 PDGF BB, RUNX2, SLC12A4, Smad, Smad1/5/8, SNAI2, SOX9, TCF, Tgf beta, TGFB2, TGIF1, WDR33, Development, Connective Tissue Wnt Development and Function
Cellular Development, Cellular ADHFE1, AMT, ANKRD35, BCL2L2, BDKRB1, BDKRB2, C10orf10, C11orf82, C19orf66, CDH10, CLTCL1, Growth and Proliferation, DNA CTSF, CYR61, ESR1, FAM183A, FZD8, GABRP, GNAQ, GPER1, GUSB, HNF4A, KCNJ13, KCNK4, MMD, Replication, Recombination, and 3 NFkB (family), RNASE4, RPL14, SEMA7A, SEPP1, SFTPA1, SP2, TEF,T NF, ZFP64, ZNF366 26 17 Repair
ATL1, CENPV, CEP290, CLIP4, CTAGE5, DCAF6, DHODH, ERCC6L, ERP29, FGD3, FZD2, FZD4, FZD6, 4 GGT7, GLP1R, GSTM1, HTR2A, KIAA1683, KLHL7, MAP7D1, NAT1, NEK7, NMNAT1, NUCKS1, ODF2L, 24 17 Cellular Assembly and Organization, PCNX, PRICKLE1, RAB34, RAMP1, RAMP2, RAMP3, RMI2, SLC31A2, SLC7A5, UBC Cell Signalling, Molecular Transport
AK4, BDNF, C5orf34, CASC5, CEBPB, ELAVL1, ERMP1, FAM107B, FAM83F, FOXP4, GABRA5, GTF2A1, Connective Tissue Disorders, 5 GTF2A2, HEXIM1, HIAT1, HNMT, LYPD1, MBNL2, METTL7A, MKX, NUAK1, NUF2, PI4K2B, RASGEF1C, 24 17 Developmental Disorder, Cell RN7SK, SCAMP4, SLC43A2, SLC7A5, TERF1,TERF2, TMEM50B, TMTC4, UBC, VKORC1L1, ZNF200 Morphology
Appendices 214
Appendix 14: Top five functional networks associated with differentially expressed genes in response to IGF-I:IGFBP-5:VN at 48 hours.Genes are colour coded according to their gene expression determined by the microarray analysis (red, up-regulated; green, down-regulated). Genes in black colour were either not on the expression array or not significantly regulated.
Focus ID Molecules in Network Score Top Diseases and Functions Molecules
ABCC9, AKR1C1/AKR1C2, ARL4A, C1QL1, CRABP1, CRAT, CXCR4, DAB2, FAM189A2, FOS, GAD1 Neurological Disease, Cancer, Cellular Development 1 , GOT, GPX2, HBA1/HBA2, hemoglobin, HIPK2, LDHA, MAP1B, MLF1, MT1A, NBPF10 (includes others), 41 27 NQO1, OPTN, PI3K (family), Pias, PKIB, QKI, RPS7, SERPINA5, STAT5a/b, STX11, TCF, TGFBI, Ubiquitin, YY2
ADAM15, Akt, ANGPTL4, ANKRD1, ARPC4, CADM1, CDK4/6, CGA, CYR61, FAM162A, Fascin, FTO, Cell-To-Cell Signalling and Interaction, Tissue Development, 2 GAL, GFRA2, Growth hormone, HK2, HPGD, IER3, Igfbp, IGFBP6, ITGB5, Lh, MTSS1, NFkB (family), 37 26 Cardiovascular System Development NUCB2, PARM1, PRODH, PROK1, PVRL3, RAP1GAP, ROBO1, SLIT2, STARD13,Vdac, Vla-4 and Function
ACHE ,AChR, ALDOC, Alp, ASAP1, CAV2, Caveolin, COL18A1, COL4A5, Collagen type IV, DCLK1, ERK, GK, Cell Morphology, Tissue KLF10, LAMB1, LAMB3, Laminin, Laminin1, MALL, MATK, MGP, MT1F, NFIB, Morphology, Tissue Development 3 35 24 NREP, NT5E, Nuclear factor 1, PDGF BB, SERPINA3, SNAI2, SORCS1, SWI-SNF, TACC1, TAGLN, TGFB2, UACA
ACTG2, Adaptor protein 2, AJUBA, AKR1C4, Alpha catenin, AMOT, AMPH, ANKRD35, Ap2 alpha, Cellular Assembly and Organization, ATP2A3, ATP2B1, CD24, Cg, Ck2, Clathrin, CLTCL1, CPLX1, Dynamin, FILIP1L, FSH, GTPase, Cellular Function and Maintenance, 4 33 24 Infectious Disease ISG20, JUN, KLF6, MPDZ, MT1X, NR5A2, PTPase, PTPN14, PTPN21,SERCA, SFTPD, SNAP25, TDRD5, UNC13D
ALOXE3, BCL3, BCR (complex), C1q, CD79B, CEACAM5, Collagen type II, ENC1, ERK1/2, ETS, FCRLB, Cancer, Hematological System 5 FUT8, GPIIB-IIIA, Iga, IgG1, Igg3, Igm, IGSF1, KCNN4, KRT13, mir-24, MT1H, NCK, NCKAP5, NEDD9, PIR, 27 21 Development and Function, Rap1, RASGRP1, SEMA3F, SLC12A4, SLURP1, Smad2/3, Sos, SUSD4, VAV2 Humoral Immune Response
Appendices 215
Appendix 15: Images of MCF-7 (A) and MDA-MB-231 (B) cells grown in Matrigel™ cultures. Top panel: Images obtained through fluorescence microscopy. Bottom panel: same images being converted to monochrome for the analysis of spheroid size (for MCF-7 cells) and circularity (for MDA-MB-231 cells) via ImageJ. TRI = IGF-I:IGFBP-5:VN.
A
B
Appendices 216
Bibliography
Aaboe, M., Offersen, B. V., Christensen, A., & Andreasen, P. A. (2003). Vitronectin in human breast carcinomas. Biochim Biophys Acta, 1638(1), 72-82.
Abbott, A. (2003). Cell culture: biology's new dimension. Nature, 424(6951), 870-872.
Abreu, J. G., Ketpura, N. I., Reversade, B., & De Robertis, E. M. (2002). Connective- tissue growth factor (CTGF) modulates cell signalling by BMP and TGF-beta. Nat Cell Biol, 4(8), 599-604.
Ainscough, S. L., Barnard, Z., Upton, Z., & Harkin, D. G. (2006). Vitronectin supports migratory responses of corneal epithelial cells to substrate bound IGF-I and HGF, and facilitates serum-free cultivation. Exp Eye Res, 83(6), 1505-1514.
Andreasen, P. A., Kjoller, L., Christensen, L., & Duffy, M. J. (1997). The urokinase-type plasminogen activator system in cancer metastasis: a review. Int J Cancer, 72(1), 1-22.
Annunziata, M., Granata, R., & Ghigo, E. (2011). The IGF system. Acta Diabetol, 48(1), 1-9.
Aoyama, E., Kubota, S., & Takigawa, M. (2012). CCN2/CTGF binds to fibroblast growth factor receptor 2 and modulates its signalling. FEBS Lett, 586(24), 4270- 4275.
Arteaga, C. L., Sliwkowski, M. X., Osborne, C. K., Perez, E. A., Puglisi, F., & Gianni, L. (2012). Treatment of HER2-positive breast cancer: current status and future perspectives. Nat Rev Clin Oncol, 9(1), 16-32.
Athanasopoulos, A. N., Schneider, D., Keiper, T., Alt, V., Pendurthi, U. R., Liegibel, U. M., Sommer, U., Nawroth, P. P., Kasperk, C., & Chavakis, T. (2007). Vascular endothelial growth factor (VEGF)-induced up-regulation of CCN1 in osteoblasts mediates proangiogenic activities in endothelial cells and promotes fracture healing. J Biol Chem, 282(37), 26746-26753.
Babic, A. M., Chen, C. C., & Lau, L. F. (1999). Fisp12/mouse connective tissue growth factor mediates endothelial cell adhesion and migration through integrin alphavbeta3, promotes endothelial cell survival, and induces angiogenesis in vivo. Mol Cell Biol, 19(4), 2958-2966.
Bae, H. B., Zmijewski, J. W., Deshane, J. S., Zhi, D., Thompson, L. C., Peterson, C. B., Chaplin, D. D., & Abraham, E. (2012). Vitronectin inhibits neutrophil apoptosis through activation of integrin-associated signalling pathways. Am J Respir Cell Mol Biol, 46(6), 790-796.
Bibliography 217
Banerjee, S., Reis-Filho, J. S., Ashley, S., Steele, D., Ashworth, A., Lakhani, S. R., & Smith, I. E. (2006). Basal-like breast carcinomas: clinical outcome and response to chemotherapy. J Clin Pathol, 59(7), 729-735.
Barbolina, M. V., Adley, B. P., Ariztia, E. V., Liu, Y., & Stack, M. S. (2007). Microenvironmental regulation of membrane type 1 matrix metalloproteinase activity in ovarian carcinoma cells via collagen-induced EGR1 expression. J Biol Chem, 282(7), 4924-4931.
Barnes, N. L., Ooi, J. L., Yarnold, J. R., & Bundred, N. J. (2012). Ductal carcinoma in situ of the breast. BMJ, 344, e797.
Barsky, S. H., & Karlin, N. J. (2006). Mechanisms of disease: breast tumor pathogenesis and the role of the myoepithelial cell. Nat Clin Pract Oncol, 3(3), 138-151.
Bartholin, L., Wessner, L. L., Chirgwin, J. M., & Guise, T. A. (2007). The human Cyr61 gene is a transcriptional target of transforming growth factor beta in cancer cells. Cancer Lett, 246(1-2), 230-236.
Bartsch, J. E., Staren, E. D., & Appert, H. E. (2003). Adhesion and migration of extracellular matrix-stimulated breast cancer. J Surg Res, 110(1), 287-294.
Bartucci, M., Morelli, C., Mauro, L., Ando, S., & Surmacz, E. (2001). Differential insulin-like growth factor I receptor signalling and function in estrogen receptor (ER)-positive MCF-7 and ER-negative MDA-MB-231 breast cancer cells. Cancer Res, 61(18), 6747-6754.
Baxter, R. C. (2014). IGF binding proteins in cancer: mechanistic and clinical insights. Nat Rev Cancer, 14(5), 329-341.
Baxter, R. C., Martin, J. L., & Beniac, V. A. (1989). High molecular weight insulin-like growth factor binding protein complex. Purification and properties of the acid- labile subunit from human serum. J Biol Chem, 264(20), 11843-11848.
Beattie, J., Kreiner, M., Allan, G. J., Flint, D. J., Domingues, D., & van der Walle, C. F. (2009). IGFBP-3 and IGFBP-5 associate with the cell binding domain (CBD) of fibronectin. Biochem Biophys Res Commun, 381(4), 572-576.
Beattie, J., McIntosh, L., & van der Walle, C. F. (2010). Cross-talk between the insulin- like growth factor (IGF) axis and membrane integrins to regulate cell physiology. J Cell Physiol, 224(3), 605-611.
Beauvais, D. M., Ell, B. J., McWhorter, A. R., & Rapraeger, A. C. (2009). Syndecan-1 regulates alphavbeta3 and alphavbeta5 integrin activation during angiogenesis and is blocked by synstatin, a novel peptide inhibitor. J Exp Med, 206(3), 691- 705.
Beauvais, D. M., & Rapraeger, A. C. (2010). Syndecan-1 couples the insulin-like growth factor-1 receptor to inside-out integrin activation. J Cell Sci, 123(Pt 21), 3796- 3807.
Bibliography 218
Becker, M. A., Ibrahim, Y. H., Cui, X., Lee, A. V., & Yee, D. (2011). The IGF pathway regulates ERalpha through a S6K1-dependent mechanism in breast cancer cells. Mol Endocrinol, 25(3), 516-528.
Bellis, S. L., Miller, J. T., & Turner, C. E. (1995). Characterization of tyrosine phosphorylation of paxillin in vitro by focal adhesion kinase. J Biol Chem, 270(29), 17437-17441.
Bello, L., Francolini, M., Marthyn, P., Zhang, J., Carroll, R. S., Nikas, D. C., Strasser, J. F., Villani, R., Cheresh, D. A., & Black, P. M. (2001). Alpha(v)beta3 and alpha(v)beta5 integrin expression in glioma periphery. Neurosurgery, 49(2), 380- 389; discussion 390.
Benlimame, N., He, Q., Jie, S., Xiao, D., Xu, Y. J., Loignon, M., Schlaepfer, D. D., & Alaoui-Jamali, M. A. (2005). FAK signalling is critical for ErbB-2/ErbB-3 receptor cooperation for oncogenic transformation and invasion. J Cell Biol, 171(3), 505-516.
Benton, G., Arnaoutova, I., George, J., Kleinman, H. K., & Koblinski, J. (2014). Matrigel: From discovery and ECM mimicry to assays and models for cancer research. Adv Drug Deliv Rev.
Bergamaschi, A., Tagliabue, E., Sorlie, T., Naume, B., Triulzi, T., Orlandi, R., Russnes, H. G., Nesland, J. M., Tammi, R., Auvinen, P., Kosma, V. M., Menard, S., & Borresen-Dale, A. L. (2008). Extracellular matrix signature identifies breast cancer subgroups with different clinical outcome. J Pathol, 214(3), 357-367.
Berndt, A., Kollner, R., Richter, P., Franz, M., Voigt, A., Berndt, A., Borsi, L., Giavazzi, R., Neri, D., & Kosmehl, H. (2010). A comparative analysis of oncofetal fibronectin and tenascin-C incorporation in tumour vessels using human recombinant SIP format antibodies. Histochem Cell Biol, 133(4), 467-475.
Bhola, N. E., Balko, J. M., Dugger, T. C., Kuba, M. G., Sanchez, V., Sanders, M., Stanford, J., Cook, R. S., & Arteaga, C. L. (2013). TGF-beta inhibition enhances chemotherapy action against triple-negative breast cancer. J Clin Invest, 123(3), 1348-1358.
Bid, H. K., Zhan, J., Phelps, D. A., Kurmasheva, R. T., & Houghton, P. J. (2012). Potent inhibition of angiogenesis by the IGF-1 receptor-targeting antibody SCH717454 is reversed by IGF-2. Mol Cancer Ther, 11(3), 649-659.
Birnie, R., Bryce, S. D., Roome, C., Dussupt, V., Droop, A., Lang, S. H., Berry, P. A., Hyde, C. F., Lewis, J. L., Stower, M. J., Maitland, N. J., & Collins, A. T. (2008). Gene expression profiling of human prostate cancer stem cells reveals a pro- inflammatory phenotype and the importance of extracellular matrix interactions. Genome Biol, 9(5), R83.
Bisanz, K., Yu, J., Edlund, M., Spohn, B., Hung, M. C., Chung, L. W., & Hsieh, C. L. (2005). Targeting ECM-integrin interaction with liposome-encapsulated small
Bibliography 219
interfering RNAs inhibits the growth of human prostate cancer in a bone xenograft imaging model. Mol Ther, 12(4), 634-643.
Bissell, M. J., & Hines, W. C. (2011). Why don't we get more cancer? A proposed role of the microenvironment in restraining cancer progression. Nat Med, 17(3), 320- 329.
Bissell, M. J., & Radisky, D. (2001). Putting tumours in context. Nat Rev Cancer, 1(1), 46-54.
Blundell, T. L., Bedarkar, S., Rinderknecht, E., & Humbel, R. E. (1978). Insulin-like growth factor: a model for tertiary structure accounting for immunoreactivity and receptor binding. Proc Natl Acad Sci U S A, 75(1), 180-184.
Bolos, V., Peinado, H., Perez-Moreno, M. A., Fraga, M. F., Esteller, M., & Cano, A. (2003). The transcription factor Slug represses E-cadherin expression and induces epithelial to mesenchymal transitions: a comparison with Snail and E47 repressors. J Cell Sci, 116(Pt 3), 499-511.
Bombonati, A., & Sgroi, D. C. (2011). The molecular pathology of breast cancer progression. J Pathol, 223(2), 307-317.
Bornstein, P., & Sage, E. H. (2002). Matricellular proteins: extracellular modulators of cell function. Curr Opin Cell Biol, 14(5), 608-616.
Bouchard, V., Demers, M. J., Thibodeau, S., Laquerre, V., Fujita, N., Tsuruo, T., Beaulieu, J. F., Gauthier, R., Vezina, A., Villeneuve, L., & Vachon, P. H. (2007). Fak/Src signalling in human intestinal epithelial cell survival and anoikis: differentiation state-specific uncoupling with the PI3-K/Akt-1 and MEK/Erk pathways. J Cell Physiol, 212(3), 717-728.
Boyd, N. A., Bradwell, A. R., & Thompson, R. A. (1993). Quantitation of vitronectin in serum: evaluation of its usefulness in routine clinical practice. J Clin Pathol, 46(11), 1042-1045.
Brigstock, D. R. (2002). Regulation of angiogenesis and endothelial cell function by connective tissue growth factor (CTGF) and cysteine-rich 61 (CYR61). Angiogenesis, 5(3), 153-165.
Brigstock, D. R. (2003). The CCN family: a new stimulus package. J Endocrinol, 178(2), 169-175.
Brissenden, J. E., Ullrich, A., & Francke, U. (1984). Human chromosomal mapping of genes for insulin-like growth factors I and II and epidermal growth factor. Nature, 310(5980), 781-784.
Brizzi, M. F., Tarone, G., & Defilippi, P. (2012). Extracellular matrix, integrins, and growth factors as tailors of the stem cell niche. Curr Opin Cell Biol, 24(5), 645- 651.
Bibliography 220
Brooks, P. C., Klemke, R. L., Schon, S., Lewis, J. M., Schwartz, M. A., & Cheresh, D. A. (1997). Insulin-like growth factor receptor cooperates with integrin alpha v beta 5 to promote tumor cell dissemination in vivo. J Clin Invest, 99(6), 1390- 1398.
Brown, J., Jones, E. Y., & Forbes, B. E. (2009). Interactions of IGF-II with the IGF2R/cation-independent mannose-6-phosphate receptor mechanism and biological outcomes. Vitam Horm, 80, 699-719.
Buck, M. B., & Knabbe, C. (2006). TGF-beta signalling in breast cancer. Ann N Y Acad Sci, 1089, 119-126.
Burns, W. C., Twigg, S. M., Forbes, J. M., Pete, J., Tikellis, C., Thallas-Bonke, V., Thomas, M. C., Cooper, M. E., & Kantharidis, P. (2006). Connective tissue growth factor plays an important role in advanced glycation end product-induced tubular epithelial-to-mesenchymal transition: implications for diabetic renal disease. J Am Soc Nephrol, 17(9), 2484-2494.
Burroughs, K. D., Dunn, S. E., Barrett, J. C., & Taylor, J. A. (1999). Insulin-like growth factor-I: a key regulator of human cancer risk? J Natl Cancer Inst, 91(7), 579- 581.
Burstein, H. J., Polyak, K., Wong, J. S., Lester, S. C., & Kaelin, C. M. (2004). Ductal carcinoma in situ of the breast. N Engl J Med, 350(14), 1430-1441.
Butler, A. A., Yakar, S., Gewolb, I. H., Karas, M., Okubo, Y., & LeRoith, D. (1998). Insulin-like growth factor-I receptor signal transduction: at the interface between physiology and cell biology. Comp Biochem Physiol B Biochem Mol Biol, 121(1), 19-26.
Calich, V. L., Purchio, A., & Paula, C. R. (1979). A new fluorescent viability test for fungi cells. Mycopathologia, 66(3), 175-177.
Calvano, S. E., Xiao, W., Richards, D. R., Felciano, R. M., Baker, H. V., Cho, R. J., Chen, R. O., Brownstein, B. H., Cobb, J. P., Tschoeke, S. K., Miller-Graziano, C., Moldawer, L. L., Mindrinos, M. N., Davis, R. W., Tompkins, R. G., & Lowry, S. F. (2005). A network-based analysis of systemic inflammation in humans. Nature, 437(7061), 1032-1037.
Carriero, M. V., Del Vecchio, S., Capozzoli, M., Franco, P., Fontana, L., Zannetti, A., Botti, G., D'Aiuto, G., Salvatore, M., & Stoppelli, M. P. (1999). Urokinase receptor interacts with alpha(v)beta5 vitronectin receptor, promoting urokinase- dependent cell migration in breast cancer. Cancer Res, 59(20), 5307-5314.
Carriero, M. V., Del Vecchio, S., Franco, P., Potena, M. I., Chiaradonna, F., Botti, G., Stoppelli, M. P., & Salvatore, M. (1997). Vitronectin binding to urokinase receptor in human breast cancer. Clin Cancer Res, 3(8), 1299-1308.
Cary, L. A., Han, D. C., & Guan, J. L. (1999). Integrin-mediated signal transduction pathways. Histol Histopathol, 14(3), 1001-1009.
Bibliography 221
Cary, L. A., Han, D. C., Polte, T. R., Hanks, S. K., & Guan, J. L. (1998). Identification of p130Cas as a mediator of focal adhesion kinase-promoted cell migration. J Cell Biol, 140(1), 211-221.
Chambard, J. C., Lefloch, R., Pouyssegur, J., & Lenormand, P. (2007). ERK implication in cell cycle regulation. Biochim Biophys Acta, 1773(8), 1299-1310.
Charoenpanich, A., Wall, M. E., Tucker, C. J., Andrews, D. M., Lalush, D. S., & Loboa, E. G. (2011). Microarray analysis of human adipose-derived stem cells in three- dimensional collagen culture: osteogenesis inhibits bone morphogenic protein and Wnt signalling pathways, and cyclic tensile strain causes upregulation of proinflammatory cytokine regulators and angiogenic factors. Tissue Eng Part A, 17(21-22), 2615-2627.
Chaturvedi, M. M., Sung, B., Yadav, V. R., Kannappan, R., & Aggarwal, B. B. (2011). NF-kappaB addiction and its role in cancer: 'one size does not fit all'. Oncogene, 30(14), 1615-1630.
Chen, & Lau, L. F. (2009). Functions and mechanisms of action of CCN matricellular proteins. Int J Biochem Cell Biol, 41(4), 771-783.
Chen, C. C., Mo, F. E., & Lau, L. F. (2001). The angiogenic factor Cyr61 activates a genetic program for wound healing in human skin fibroblasts. J Biol Chem, 276(50), 47329-47337.
Chen, H., Huang, X. N., Stewart, A. F., & Sepulveda, J. L. (2004). Gene expression changes associated with fibronectin-induced cardiac myocyte hypertrophy. Physiol Genomics, 18(3), 273-283.
Chen, J., Baskerville, C., Han, Q., Pan, Z. K., & Huang, S. (2001). Alpha(v) integrin, p38 mitogen-activated protein kinase, and urokinase plasminogen activator are functionally linked in invasive breast cancer cells. J Biol Chem, 276(51), 47901- 47905.
Chen, L., & Bourguignon, L. Y. (2014). Hyaluronan-CD44 interaction promotes c-Jun signalling and miRNA21 expression leading to Bcl-2 expression and chemoresistance in breast cancer cells. Mol Cancer, 13, 52.
Chen, P. S., Wang, M. Y., Wu, S. N., Su, J. L., Hong, C. C., Chuang, S. E., Chen, M. W., Hua, K. T., Wu, Y. L., Cha, S. T., Babu, M. S., Chen, C. N., Lee, P. H., Chang, K. J., & Kuo, M. L. (2007). CTGF enhances the motility of breast cancer cells via an integrin-alphavbeta3-ERK1/2-dependent S100A4-upregulated pathway. J Cell Sci, 120(Pt 12), 2053-2065.
Chen, Y., Abraham, D. J., Shi-Wen, X., Pearson, J. D., Black, C. M., Lyons, K. M., & Leask, A. (2004). CCN2 (connective tissue growth factor) promotes fibroblast adhesion to fibronectin. Mol Biol Cell, 15(12), 5635-5646.
Bibliography 222
Chen, Y., Blom, I. E., Sa, S., Goldschmeding, R., Abraham, D. J., & Leask, A. (2002). CTGF expression in mesangial cells: involvement of SMADs, MAP kinase, and PKC. Kidney Int, 62(4), 1149-1159.
Chien, W., O'Kelly, J., Lu, D., Leiter, A., Sohn, J., Yin, D., Karlan, B., Vadgama, J., Lyons, K. M., & Koeffler, H. P. (2011). Expression of connective tissue growth factor (CTGF/CCN2) in breast cancer cells is associated with increased migration and angiogenesis. Int J Oncol, 38(6), 1741-1747.
Chishima, T., Miyagi, Y., Wang, X., Yamaoka, H., Shimada, H., Moossa, A. R., & Hoffman, R. M. (1997). Cancer invasion and micrometastasis visualized in live tissue by green fluorescent protein expression. Cancer Res, 57(10), 2042-2047.
Christensen, L. (1992). The distribution of fibronectin, laminin and tetranectin in human breast cancer with special attention to the extracellular matrix. APMIS Suppl, 26, 1-39.
Chuang, J. Y., Yu, N. Y., Chiang, I. P., Lai, C. H., Lin, C. D., & Tang, C. H. (2012). Cyr61 increases matrix metalloproteinase-3 expression and cell motility in human oral squamous cell carcinoma cells. J Cell Biochem, 113(6), 1977-1986.
Clark, R. A. (2008). Synergistic signalling from extracellular matrix-growth factor complexes. J Invest Dermatol, 128(6), 1354-1355.
Clemmons, D. R. (1998). Role of insulin-like growth factor binding proteins in controlling IGF actions. Mol Cell Endocrinol, 140(1-2), 19-24.
Clemmons, D. R., & Maile, L. A. (2003). Minireview: Integral membrane proteins that function coordinately with the insulin-like growth factor I receptor to regulate intracellular signalling. Endocrinology, 144(5), 1664-1670.
Clemmons, D. R., & Maile, L. A. (2005). Interaction between insulin-like growth factor- I receptor and alphaVbeta3 integrin linked signalling pathways: cellular responses to changes in multiple signalling inputs. Mol Endocrinol, 19(1), 1-11.
Cline, M. S., Smoot, M., Cerami, E., Kuchinsky, A., Landys, N., Workman, C., Christmas, R., Avila-Campilo, I., Creech, M., Gross, B., Hanspers, K., Isserlin, R., Kelley, R., Killcoyne, S., Lotia, S., Maere, S., Morris, J., Ono, K., Pavlovic, V., Pico, A. R., Vailaya, A., Wang, P. L., Adler, A., Conklin, B. R., Hood, L., Kuiper, M., Sander, C., Schmulevich, I., Schwikowski, B., Warner, G. J., Ideker, T., & Bader, G. D. (2007). Integration of biological networks and gene expression data using Cytoscape. Nat Protoc, 2(10), 2366-2382.
Correia, A. L., & Bissell, M. J. (2012). The tumor microenvironment is a dominant force in multidrug resistance. Drug Resist Updat, 15(1-2), 39-49.
Cory, A. H., Owen, T. C., Barltrop, J. A., & Cory, J. G. (1991). Use of an aqueous soluble tetrazolium/formazan assay for cell growth assays in culture. Cancer Commun, 3(7), 207-212.
Bibliography 223
Cox, T. R., & Erler, J. T. (2011). Remodeling and homeostasis of the extracellular matrix: implications for fibrotic diseases and cancer. Dis Model Mech, 4(2), 165- 178.
Creighton, C. J., Casa, A., Lazard, Z., Huang, S., Tsimelzon, A., Hilsenbeck, S. G., Osborne, C. K., & Lee, A. V. (2008). Insulin-like growth factor-I activates gene transcription programs strongly associated with poor breast cancer prognosis. J Clin Oncol, 26(25), 4078-4085.
Creighton, C. J., Li, X., Landis, M., Dixon, J. M., Neumeister, V. M., Sjolund, A., Rimm, D. L., Wong, H., Rodriguez, A., Herschkowitz, J. I., Fan, C., Zhang, X., He, X., Pavlick, A., Gutierrez, M. C., Renshaw, L., Larionov, A. A., Faratian, D., Hilsenbeck, S. G., Perou, C. M., Lewis, M. T., Rosen, J. M., & Chang, J. C. (2009). Residual breast cancers after conventional therapy display mesenchymal as well as tumor-initiating features. Proc Natl Acad Sci U S A, 106(33), 13820- 13825.
Datta, S. R., Dudek, H., Tao, X., Masters, S., Fu, H., Gotoh, Y., & Greenberg, M. E. (1997). Akt phosphorylation of BAD couples survival signals to the cell-intrinsic death machinery. Cell, 91(2), 231-241.
Daughaday, W. H., & Rotwein, P. (1989). Insulin-like growth factors I and II. Peptide, messenger ribonucleic acid and gene structures, serum, and tissue concentrations. Endocr Rev, 10(1), 68-91.
Davies, S. P., Reddy, H., Caivano, M., & Cohen, P. (2000). Specificity and mechanism of action of some commonly used protein kinase inhibitors. Biochem J, 351(Pt 1), 95-105.
De Meyts, P., & Whittaker, J. (2002). Structural biology of insulin and IGF1 receptors: implications for drug design. Nat Rev Drug Discov, 1(10), 769-783.
Debnath, J., & Brugge, J. S. (2005). Modelling glandular epithelial cancers in three- dimensional cultures. Nat Rev Cancer, 5(9), 675-688.
Degryse, B., Orlando, S., Resnati, M., Rabbani, S. A., & Blasi, F. (2001). Urokinase/urokinase receptor and vitronectin/alpha(v)beta(3) integrin induce chemotaxis and cytoskeleton reorganization through different signalling pathways. Oncogene, 20(16), 2032-2043.
Dent, R., Trudeau, M., Pritchard, K. I., Hanna, W. M., Kahn, H. K., Sawka, C. A., Lickley, L. A., Rawlinson, E., Sun, P., & Narod, S. A. (2007). Triple-negative breast cancer: clinical features and patterns of recurrence. Clin Cancer Res, 13(15 Pt 1), 4429-4434.
Desgrosellier, J. S., & Cheresh, D. A. (2010). Integrins in cancer: biological implications and therapeutic opportunities. Nat Rev Cancer, 10(1), 9-22.
Diry, M., Tomkiewicz, C., Koehle, C., Coumoul, X., Walter B. K., Barouli, R., & Transy. C. (2006). Activation of the doxin/aryl hydrocarbon receptor (AhR)
Bibliography 224
modulates cell plasticity through a JNK-dependent mechanism. Oncogene, 25, 5570-5574.
Doerr, M. E., & Jones, J. I. (1996). The roles of integrins and extracellular matrix proteins in the insulin-like growth factor I-stimulated chemotaxis of human breast cancer cells. J Biol Chem, 271(5), 2443-2447.
Edwards, S., Lalor, P. F., Tuncer, C., & Adams, D. H. (2006). Vitronectin in human hepatic tumours contributes to the recruitment of lymphocytes in an alpha v beta3-independent manner. Br J Cancer, 95(11), 1545-1554.
Egeblad, M., & Werb, Z. (2002). New functions for the matrix metalloproteinases in cancer progression. Nat Rev Cancer, 2(3), 161-174.
Eke, I., & Cordes, N. (2011). Dual targeting of EGFR and focal adhesion kinase in 3D grown HNSCC cell cultures. Radiother Oncol, 99(3), 279-286.
El-Shewy, H. M., Lee, M. H., Obeid, L. M., Jaffa, A. A., & Luttrell, L. M. (2007). 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. J Biol Chem, 282(36), 26150-26157.
Eliceiri, B. P. (2001). Integrin and growth factor receptor crosstalk. Circ Res, 89(12), 1104-1110.
Elliott, T., & Sethi, T. (2002). Integrins and extracellular matrix: a novel mechanism of multidrug resistance. Expert Rev Anticancer Ther, 2(4), 449-459.
Espinoza, I., Liu, H., Busby, R., & Lupu, R. (2011). CCN1, a candidate target for zoledronic acid treatment in breast cancer. Mol Cancer Ther, 10(5), 732-741.
Espinoza, I., Menendez, J. A., Kvp, C. M., & Lupu, R. (2014). CCN1 promotes vascular endothelial growth factor secretion through alphavbeta 3 integrin receptors in breast cancer. J Cell Commun Signal, 8(1), 23-27.
Felding-Habermann, B., & Cheresh, D. A. (1993). Vitronectin and its receptors. Curr Opin Cell Biol, 5(5), 864-868.
Felding-Habermann, B., O'Toole, T. E., Smith, J. W., Fransvea, E., Ruggeri, Z. M., Ginsberg, M. H., Hughes, P. E., Pampori, N., Shattil, S. J., Saven, A., & Mueller, B. M. (2001). Integrin activation controls metastasis in human breast cancer. Proc Natl Acad Sci U S A, 98(4), 1853-1858.
Feng, S., Cao, Z., & Wang, X. (2013). Role of aryl hydrocarbon receptor in cancer. Biochim Biophys Acta, 1836(2), 197-210.
Firth, S. M., & Baxter, R. C. (2002). Cellular actions of the insulin-like growth factor binding proteins. Endocr Rev, 23(6), 824-854.
Bibliography 225
Forbes, B. E., Hartfield, P. J., McNeil, K. A., Surinya, K. H., Milner, S. J., Cosgrove, L. J., & Wallace, J. C. (2002). Characteristics of binding of insulin-like growth factor (IGF)-I and IGF-II analogues to the type 1 IGF receptor determined by BIAcore analysis. Eur J Biochem, 269(3), 961-968.
Forbes, B. E., McCarthy, P., & Norton, R. S. (2012). Insulin-like growth factor binding proteins: a structural perspective. Front Endocrinol (Lausanne), 3, 38.
Foulkes, W. D., Smith, I. E., & Reis-Filho, J. S. (2010). Triple-negative breast cancer. N Engl J Med, 363(20), 1938-1948.
Francischetti, I. M., Kotsyfakis, M., Andersen, J. F., & Lukszo, J. (2010). Cyr61/CCN1 displays high-affinity binding to the somatomedin B(1-44) domain of vitronectin. PLoS One, 5(2), e9356.
Franks, S. E., Campbell, C. I., Barnett, E. F., Siwicky, M. D., Livingstone, J., Cory, S., & Moorehead, R. A. (2012). Transgenic IGF-IR overexpression induces mammary tumors with basal-like characteristics, whereas IGF-IR-independent mammary tumors express a claudin-low gene signature. Oncogene, 31(27), 3298-3309.
Franzen, C. A., Todorovic, V., Desai, B. V., Mirzoeva, S., Yang, X. J., Green, K. J., & Pelling, J. C. (2012). The desmosomal armadillo protein plakoglobin regulates prostate cancer cell adhesion and motility through vitronectin-dependent Src signalling. PLoS One, 7(7), e42132.
Frazier, K., Williams, S., Kothapalli, D., Klapper, H., & Grotendorst, G. R. (1996). Stimulation of fibroblast cell growth, matrix production, and granulation tissue formation by connective tissue growth factor. J Invest Dermatol, 107(3), 404- 411.
Frisch, S. M., & Francis, H. (1994). Disruption of epithelial cell-matrix interactions induces apoptosis. J Cell Biol, 124(4), 619-626.
Fruman, D. A., Meyers, R. E., & Cantley, L. C. (1998). Phosphoinositide kinases. Annu Rev Biochem, 67, 481-507.
Fujita, M., Khazenzon, N. M., Bose, S., Sekiguchi, K., Sasaki, T., Carter, W. G., Ljubimov, A. V., Black, K. L., & Ljubimova, J. Y. (2005). Overexpression of beta1-chain-containing laminins in capillary basement membranes of human breast cancer and its metastases. Breast Cancer Res, 7(4), R411-421.
Fukushima, Y., Ohnishi, T., Arita, N., Hayakawa, T., & Sekiguchi, K. (1998). Integrin alpha3beta1-mediated interaction with laminin-5 stimulates adhesion, migration and invasion of malignant glioma cells. Int J Cancer, 76(1), 63-72.
Fulford, L. G., Reis-Filho, J. S., Ryder, K., Jones, C., Gillett, C. E., Hanby, A., Easton, D., & Lakhani, S. R. (2007). Basal-like grade III invasive ductal carcinoma of the breast: patterns of metastasis and long-term survival. Breast Cancer Res, 9(1), R4.
Bibliography 226
Gallagher, E. J., & LeRoith, D. (2011). Minireview: IGF, Insulin, and Cancer. Endocrinology, 152(7), 2546-2551.
Galliher, A. J., & Schiemann, W. P. (2006). Beta3 integrin and Src facilitate transforming growth factor-beta mediated induction of epithelial-mesenchymal transition in mammary epithelial cells. Breast Cancer Res, 8(4), R42.
Gao, J., Chang, Y. S., Jallal, B., & Viner, J. (2012). Targeting the insulin-like growth factor axis for the development of novel therapeutics in oncology. Cancer Res, 72(1), 3-12.
Gao, R., & Brigstock, D. R. (2004). Connective tissue growth factor (CCN2) induces adhesion of rat activated hepatic stellate cells by binding of its C-terminal domain to integrin alpha(v)beta(3) and heparan sulfate proteoglycan. J Biol Chem, 279(10), 8848-8855.
Gechtman, Z., Belleli, A., Lechpammer, S., & Shaltiel, S. (1997). The cluster of basic amino acids in vitronectin contributes to its binding of plasminogen activator inhibitor-1: evidence from thrombin-, elastase- and plasmin-cleaved vitronectins and anti-peptide antibodies. Biochem J, 325 ( Pt 2), 339-349.
Gery, S., Xie, D., Yin, D., Gabra, H., Miller, C., Wang, H., Scott, D., Yi, W. S., Popoviciu, M. L., Said, J. W., & Koeffler, H. P. (2005). Ovarian carcinomas: CCN genes are aberrantly expressed and CCN1 promotes proliferation of these cells. Clin Cancer Res, 11(20), 7243-7254.
Giampieri, S., Manning, C., Hooper, S., Jones, L., Hill, C. S., & Sahai, E. (2009). Localized and reversible TGFbeta signalling switches breast cancer cells from cohesive to single cell motility. Nat Cell Biol, 11(11), 1287-1296.
Giancotti, F. G., & Ruoslahti, E. (1999). Integrin signalling. Science, 285(5430), 1028- 1032.
Giani, C., Campani, D., Rasmussen, A., Fierabracci, P., Miccoli, P., Bevilacqua, G., Pinchera, A., & Cullen, K. J. (2002). Insulin-like growth factor II (IGF-II) immunohistochemistry in breast cancer: relationship with the most important morphological and biochemical prognostic parameters. Int J Biol Markers, 17(2), 90-95.
Gilmore, A. P., Valentijn, A. J., Wang, P., Ranger, A. M., Bundred, N., O'Hare, M. J., Wakeling, A., Korsmeyer, S. J., & Streuli, C. H. (2002). Activation of BAD by therapeutic inhibition of epidermal growth factor receptor and transactivation by insulin-like growth factor receptor. J Biol Chem, 277(31), 27643-27650.
Giorgetti, S., Ballotti, R., Kowalski-Chauvel, A., Tartare, S., & Van Obberghen, E. (1993). The insulin and insulin-like growth factor-I receptor substrate IRS-1 associates with and activates phosphatidylinositol 3-kinase in vitro. J Biol Chem, 268(10), 7358-7364.
Bibliography 227
Gladson, C. L., & Cheresh, D. A. (1991). Glioblastoma expression of vitronectin and the alpha v beta 3 integrin. Adhesion mechanism for transformed glial cells. J Clin Invest, 88(6), 1924-1932.
Golubovskaya, V., Beviglia, L., Xu, L. H., Earp, H. S., 3rd, Craven, R., & Cance, W. (2002). Dual inhibition of focal adhesion kinase and epidermal growth factor receptor pathways cooperatively induces death receptor-mediated apoptosis in human breast cancer cells. J Biol Chem, 277(41), 38978-38987.
Gooch, J. L., Van Den Berg, C. L., & Yee, D. (1999). Insulin-like growth factor (IGF)-I rescues breast cancer cells from chemotherapy-induced cell death--proliferative and anti-apoptotic effects. Breast Cancer Res Treat, 56(1), 1-10.
Goodwin, C. R., Lal, B., Zhou, X., Ho, S., Xia, S., Taeger, A., Murray, J., & Laterra, J. (2010). Cyr61 mediates hepatocyte growth factor-dependent tumor cell growth, migration, and Akt activation. Cancer Res, 70(7), 2932-2941.
Grey, A., Chen, Q., Xu, X., Callon, K., & Cornish, J. (2003). Parallel phosphatidylinositol-3 kinase and p42/44 mitogen-activated protein kinase signalling pathways subserve the mitogenic and antiapoptotic actions of insulin- like growth factor I in osteoblastic cells. Endocrinology, 144(11), 4886-4893.
Grzeszkiewicz, T. M., Lindner, V., Chen, N., Lam, S. C., & Lau, L. F. (2002). The angiogenic factor cysteine-rich 61 (CYR61, CCN1) supports vascular smooth muscle cell adhesion and stimulates chemotaxis through integrin alpha(6)beta(1) and cell surface heparan sulfate proteoglycans. Endocrinology, 143(4), 1441- 1450.
Guo, W., & Giancotti, F. G. (2004). Integrin signalling during tumour progression. Nat Rev Mol Cell Biol, 5(10), 816-826.
Gupta, P. B., Onder, T. T., Jiang, G., Tao, K., Kuperwasser, C., Weinberg, R. A., & Lander, E. S. (2009). Identification of selective inhibitors of cancer stem cells by high-throughput screening. Cell, 138(4), 645-659.
Han, Q., Leng, J., Bian, D., Mahanivong, C., Carpenter, K. A., Pan, Z. K., Han, J., & Huang, S. (2002). Rac1-MKK3-p38-MAPKAPK2 pathway promotes urokinase plasminogen activator mRNA stability in invasive breast cancer cells. J Biol Chem, 277(50), 48379-48385.
Han, S., Khuri, F. R., & Roman, J. (2006). Fibronectin stimulates non-small cell lung carcinoma cell growth through activation of Akt/mammalian target of rapamycin/S6 kinase and inactivation of LKB1/AMP-activated protein kinase signal pathways. Cancer Res, 66(1), 315-323.
Hanahan, D., & Weinberg, R. A. (2011). Hallmarks of cancer: the next generation. Cell, 144(5), 646-674.
Hankinson, S. E., Willett, W. C., Colditz, G. A., Hunter, D. J., Michaud, D. S., Deroo, B., Rosner, B., Speizer, F. E., & Pollak, M. (1998). Circulating concentrations of
Bibliography 228
insulin-like growth factor-I and risk of breast cancer. Lancet, 351(9113), 1393- 1396.
Hansen, R. K., & Bissell, M. J. (2000). Tissue architecture and breast cancer: the role of extracellular matrix and steroid hormones. Endocr Relat Cancer, 7(2), 95-113.
Hapke, S., Kessler, H., Luber, B., Benge, A., Hutzler, P., Hofler, H., Schmitt, M., & Reuning, U. (2003). Ovarian cancer cell proliferation and motility is induced by engagement of integrin alpha(v)beta3/Vitronectin interaction. Biol Chem, 384(7), 1073-1083.
Hartog, H., Boezen, H. M., de Jong, M. M., Schaapveld, M., Wesseling, J., & van der Graaf, W. T. (2013). Prognostic value of insulin-like growth factor 1 and insulin- like growth factor binding protein 3 blood levels in breast cancer. Breast, 22(6), 1155-1160.
Hayman, E. G., Pierschbacher, M. D., Ohgren, Y., & Ruoslahti, E. (1983). Serum spreading factor (vitronectin) is present at the cell surface and in tissues. Proc Natl Acad Sci U S A, 80(13), 4003-4007.
Hazawa, M., Yasuda, T., Noshiro, K., Saotome-Nakamura, A., Fukuzaki, T., Michikawa, Y., Gotoh, T., & Tajima, K. (2012). Vitronectin improves cell survival after radiation injury in human umbilical vein endothelial cells. FEBS Open Bio, 2, 334-338.
Heffelfinger, S. C., Miller, M. A., Yassin, R., & Gear, R. (1999). Angiogenic growth factors in preinvasive breast disease. Clin Cancer Res, 5(10), 2867-2876.
Hermanto, U., Zong, C. S., & Wang, L. H. (2000). Inhibition of mitogen-activated protein kinase kinase selectively inhibits cell proliferation in human breast cancer cells displaying enhanced insulin-like growth factor I-mediated mitogen- activated protein kinase activation. Cell Growth Differ, 11(12), 655-664.
Holbourn, K. P., Acharya, K. R., & Perbal, B. (2008). The CCN family of proteins: structure-function relationships. Trends Biochem Sci, 33(10), 461-473.
Hollier, B. G., Harkin, D. G., Leavesley, D., & Upton, Z. (2005). Responses of keratinocytes to substrate-bound vitronectin: growth factor complexes. Exp Cell Res, 305(1), 221-232.
Hollier, B. G., Kricker, J. A., Van Lonkhuyzen, D. R., Leavesley, D. I., & Upton, Z. (2008). Substrate-bound insulin-like growth factor (IGF)-I-IGF binding protein- vitronectin-stimulated breast cell migration is enhanced by coactivation of the phosphatidylinositide 3-Kinase/AKT pathway by alphav-integrins and the IGF-I receptor. Endocrinology, 149(3), 1075-1090.
Holmes, R. (1967). Preparation from human serum of an alpha-one protein which induces the immediate growth of unadapted cells in vitro. J Cell Biol, 32(2), 297- 308.
Bibliography 229
Hoshijima, M., Hattori, T., Inoue, M., Araki, D., Hanagata, H., Miyauchi, A., & Takigawa, M. (2006). CT domain of CCN2/CTGF directly interacts with fibronectin and enhances cell adhesion of chondrocytes through integrin alpha5beta1. FEBS Lett, 580(5), 1376-1382.
Hoshino, Y., Katsuno, Y., Ehata, S., & Miyazono, K. (2011). Autocrine TGF-beta protects breast cancer cells from apoptosis through reduction of BH3-only protein, Bim. J Biochem, 149(1), 55-65.
Howlett, A. R., & Bissell, M. J. (1993). The influence of tissue microenvironment (stroma and extracellular matrix) on the development and function of mammary epithelium. Epithelial Cell Biol, 2(2), 79-89.
Hu, Z., Fan, C., Oh, D. S., Marron, J. S., He, X., Qaqish, B. F., Livasy, C., Carey, L. A., Reynolds, E., Dressler, L., Nobel, A., Parker, J., Ewend, M. G., Sawyer, L. R., Wu, J., Liu, Y., Nanda, R., Tretiakova, M., Ruiz Orrico, A., Dreher, D., Palazzo, J. P., Perreard, L., Nelson, E., Mone, M., Hansen, H., Mullins, M., Quackenbush, J. F., Ellis, M. J., Olopade, O. I., Bernard, P. S., & Perou, C. M. (2006). The molecular portraits of breast tumors are conserved across microarray platforms. BMC Genomics, 7, 96.
Huang, C., Jacobson, K., & Schaller, M. D. (2004). MAP kinases and cell migration. J Cell Sci, 117(Pt 20), 4619-4628.
Huijbers, I. J., Iravani, M., Popov, S., Robertson, D., Al-Sarraj, S., Jones, C., & Isacke, C. M. (2010). A role for fibrillar collagen deposition and the collagen internalization receptor endo180 in glioma invasion. PLoS One, 5(3), e9808.
Hurt, E. M., Chan, K., Serrat, M. A., Thomas, S. B., Veenstra, T. D., & Farrar, W. L. (2010). Identification of vitronectin as an extrinsic inducer of cancer stem cell differentiation and tumor formation. Stem Cells, 28(3), 390-398.
Hyde, C., Hollier, B., Anderson, A., Harkin, D., & Upton, Z. (2004). Insulin-like growth factors (IGF) and IGF-binding proteins bound to vitronectin enhance keratinocyte protein synthesis and migration. J Invest Dermatol, 122(5), 1198- 1206.
Hynes, R. O. (1992). Integrins: versatility, modulation, and signalling in cell adhesion. Cell, 69(1), 11-25.
Hynes, R. O. (2009). The extracellular matrix: not just pretty fibrils. Science, 326(5957), 1216-1219.
Imamichi, Y., Konig, A., Gress, T., & Menke, A. (2007). Collagen type I-induced Smad- interacting protein 1 expression downregulates E-cadherin in pancreatic cancer. Oncogene, 26(16), 2381-2385.
Inoki, I., Shiomi, T., Hashimoto, G., Enomoto, H., Nakamura, H., Makino, K., Ikeda, E., Takata, S., Kobayashi, K., & Okada, Y. (2002). Connective tissue growth factor
Bibliography 230
binds vascular endothelial growth factor (VEGF) and inhibits VEGF-induced angiogenesis. FASEB J, 16(2), 219-221.
Ioachim, E., Charchanti, A., Briasoulis, E., Karavasilis, V., Tsanou, H., Arvanitis, D. L., Agnantis, N. J., & Pavlidis, N. (2002). Immunohistochemical expression of extracellular matrix components tenascin, fibronectin, collagen type IV and laminin in breast cancer: their prognostic value and role in tumour invasion and progression. Eur J Cancer, 38(18), 2362-2370.
Izumi, M., Shimo-Oka, T., Morishita, N., Ii, I., & Hayashi, M. (1988). Identification of the collagen-binding domain of vitronectin using monoclonal antibodies. Cell Struct Funct, 13(3), 217-225.
Izumi, M., Yamada, K. M., & Hayashi, M. (1989). Vitronectin exists in two structurally and functionally distinct forms in human plasma. Biochim Biophys Acta, 990(2), 101-108.
Jacks, T., & Weinberg, R. A. (2002). Taking the study of cancer cell survival to a new dimension. Cell, 111(7), 923-925.
Jenne, D., & Stanley, K. K. (1985). Molecular cloning of S-protein, a link between complement, coagulation and cell-substrate adhesion. EMBO J, 4(12), 3153- 3157.
Jennische, E., Skottner, A., & Hansson, H. A. (1987). Dynamic changes in insulin-like growth factor I immunoreactivity correlate to repair events in rat ear after freeze- thaw injury. Exp Mol Pathol, 47(2), 193-201.
Jonat, W., Pritchard, K. I., Sainsbury, R., & Klijn, J. G. (2006). Trends in endocrine therapy and chemotherapy for early breast cancer: a focus on the premenopausal patient. J Cancer Res Clin Oncol, 132(5), 275-286.
Jones, J. I., & Clemmons, D. R. (1995). Insulin-like growth factors and their binding proteins: biological actions. Endocr Rev, 16(1), 3-34.
Jones, J. I., Gockerman, A., Busby, W. H., Jr., Camacho-Hubner, C., & Clemmons, D. R. (1993). Extracellular matrix contains insulin-like growth factor binding protein-5: potentiation of the effects of IGF-I. J Cell Biol, 121(3), 679-687.
Jones, J. I., Prevette, T., Gockerman, A., & Clemmons, D. R. (1996). Ligand occupancy of the alpha-V-beta3 integrin is necessary for smooth muscle cells to migrate in response to insulin-like growth factor. Proc Natl Acad Sci U S A, 93(6), 2482- 2487.
Jun, J. I., & Lau, L. F. (2011). Taking aim at the extracellular matrix: CCN proteins as emerging therapeutic targets. Nat Rev Drug Discov, 10(12), 945-963.
Kadar, A., Tokes, A. M., Kulka, J., & Robert, L. (2002). Extracellular matrix components in breast carcinomas. Semin Cancer Biol, 12(3), 243-257.
Bibliography 231
Kadowaki, M., Sangai, T., Nagashima, T., Sakakibara, M., Yoshitomi, H., Takano, S., Sogawa, K., Umemura, H., Fushimi, K., Nakatani, Y., Nomura, F., & Miyazaki, M. (2011). Identification of vitronectin as a novel serum marker for early breast cancer detection using a new proteomic approach. J Cancer Res Clin Oncol, 137(7), 1105-1115.
Kalembeyi, I., Inada, H., Nishiura, R., Imanaka-Yoshida, K., Sakakura, T., & Yoshida, T. (2003). Tenascin-C upregulates matrix metalloproteinase-9 in breast cancer cells: direct and synergistic effects with transforming growth factor beta1. Int J Cancer, 105(1), 53-60.
Kaliman, P., Canicio, J., Testar, X., Palacin, M., & Zorzano, A. (1999). Insulin-like growth factor-II, phosphatidylinositol 3-kinase, nuclear factor-kappaB and inducible nitric-oxide synthase define a common myogenic signalling pathway. J Biol Chem, 274(25), 17437-17444.
Kang, Y., He, W., Tulley, S., Gupta, G. P., Serganova, I., Chen, C. R., Manova- Todorova, K., Blasberg, R., Gerald, W. L., & Massague, J. (2005). Breast cancer bone metastasis mediated by the Smad tumor suppressor pathway. Proc Natl Acad Sci U S A, 102(39), 13909-13914.
Kang, Y., Siegel, P. M., Shu, W., Drobnjak, M., Kakonen, S. M., Cordon-Cardo, C., Guise, T. A., & Massague, J. (2003). A multigenic program mediating breast cancer metastasis to bone. Cancer Cell, 3(6), 537-549.
Kashyap, A. S., Hollier, B. G., Manton, K. J., Satyamoorthy, K., Leavesley, D. I., & Upton, Z. (2011). Insulin-like growth factor-I:vitronectin complex-induced changes in gene expression effect breast cell survival and migration. Endocrinology, 152(4), 1388-1401.
Kauppila, S., Stenback, F., Risteli, J., Jukkola, A., & Risteli, L. (1998). Aberrant type I and type III collagen gene expression in human breast cancer in vivo. J Pathol, 186(3), 262-268.
Kenny, P. A., Lee, G. Y., Myers, C. A., Neve, R. M., Semeiks, J. R., Spellman, P. T., Lorenz, K., Lee, E. H., Barcellos-Hoff, M. H., Petersen, O. W., Gray, J. W., & Bissell, M. J. (2007). The morphologies of breast cancer cell lines in three- dimensional assays correlate with their profiles of gene expression. Mol Oncol, 1(1), 84-96.
Kiely, P. A., O'Gorman, D., Luong, K., Ron, D., & O'Connor, R. (2006). Insulin-like growth factor I controls a mutually exclusive association of RACK1 with protein phosphatase 2A and beta1 integrin to promote cell migration. Mol Cell Biol, 26(11), 4041-4051.
Kim, J. B., Stein, R., & O'Hare, M. J. (2005). Tumour-stromal interactions in breast cancer: the role of stroma in tumourigenesis. Tumour Biol, 26(4), 173-185.
Bibliography 232
Kim, K. W., Bae, S. K., Lee, O. H., Bae, M. H., Lee, M. J., & Park, B. C. (1998). Insulin-like growth factor II induced by hypoxia may contribute to angiogenesis of human hepatocellular carcinoma. Cancer Res, 58(2), 348-351.
Kim, S. H., Turnbull, J., & Guimond, S. (2011). Extracellular matrix and cell signalling: the dynamic cooperation of integrin, proteoglycan and growth factor receptor. J Endocrinol, 209(2), 139-151.
Kleinman, H. K., McGarvey, M. L., Hassell, J. R., Star, V. L., Cannon, F. B., Laurie, G. W., & Martin, G. R. (1986). Basement membrane complexes with biological activity. Biochemistry, 25(2), 312-318.
Kost, C., Stuber, W., Ehrlich, H. J., Pannekoek, H., & Preissner, K. T. (1992). Mapping of binding sites for heparin, plasminogen activator inhibitor-1, and plasminogen to vitronectin's heparin-binding region reveals a novel vitronectin-dependent feedback mechanism for the control of plasmin formation. J Biol Chem, 267(17), 12098-12105.
Kothapalli, D., Hayashi, N., & Grotendorst, G. R. (1998). Inhibition of TGF-beta- stimulated CTGF gene expression and anchorage-independent growth by cAMP identifies a CTGF-dependent restriction point in the cell cycle. FASEB J, 12(12), 1151-1161.
Kricker, J. A., Hyde, C. E., Van Lonkhuyzen, D. R., Hollier, B. G., Shooter, G. K., Leavesley, D. I., Herington, A. C., & Upton, Z. (2010). Mechanistic investigations into interactions between IGF-I and IGFBPs and their impact on facilitating cell migration on vitronectin. Growth Factors, 28(5), 359-369.
Kricker, J. A., Towne, C. L., Firth, S. M., Herington, A. C., & Upton, Z. (2003). Structural and functional evidence for the interaction of insulin-like growth factors (IGFs) and IGF binding proteins with vitronectin. Endocrinology, 144(7), 2807-2815.
Kulendran, M., Salhab, M., & Mokbel, K. (2009). Oestrogen-synthesising enzymes and breast cancer. Anticancer Res, 29(4), 1095-1109.
Lai, D., Ho, K. C., Hao, Y., & Yang, X. (2011). Taxol resistance in breast cancer cells is mediated by the hippo pathway component TAZ and its downstream transcriptional targets Cyr61 and CTGF. Cancer Res, 71(7), 2728-2738.
Lau, L. F. (2011). CCN1/CYR61: the very model of a modern matricellular protein. Cell Mol Life Sci, 68(19), 3149-3163.
Leask, A., & Abraham, D. J. (2006). All in the CCN family: essential matricellular signalling modulators emerge from the bunker. J Cell Sci, 119(Pt 23), 4803- 4810.
Leavesley, D. I., Kashyap, A. S., Croll, T., Sivaramakrishnan, M., Shokoohmand, A., Hollier, B. G., & Upton, Z. (2013). Vitronectin--master controller or micromanager? IUBMB Life, 65(10), 807-818.
Bibliography 233
Leivonen, S. K., Hakkinen, L., Liu, D., & Kahari, V. M. (2005). Smad3 and extracellular signal-regulated kinase 1/2 coordinately mediate transforming growth factor- beta-induced expression of connective tissue growth factor in human fibroblasts. J Invest Dermatol, 124(6), 1162-1169.
Leong, S. R., Baxter, R. C., Camerato, T., Dai, J., & Wood, W. I. (1992). Structure and functional expression of the acid-labile subunit of the insulin-like growth factor- binding protein complex. Mol Endocrinol, 6(6), 870-876.
LeRoith, D., Werner, H., Beitner-Johnson, D., & Roberts, C. T., Jr. (1995). Molecular and cellular aspects of the insulin-like growth factor I receptor. Endocr Rev, 16(2), 143-163.
Leu, S. J., Lam, S. C., & Lau, L. F. (2002). Pro-angiogenic activities of CYR61 (CCN1) mediated through integrins alphavbeta3 and alpha6beta1 in human umbilical vein endothelial cells. J Biol Chem, 277(48), 46248-46255.
Li, C., Nguyen, H. T., Zhuang, Y., Lin, Y., Flemington, E. K., Guo, W., Guenther, J., Burow, M. E., Morris, G. F., Sullivan, D., & Shan, B. (2011). Post- transcriptional up-regulation of miR-21 by type I collagen. Mol Carcinog, 50(7), 563-570.
Li, M., Chiu, J. F., Gagne, J., & Fukagawa, N. K. (2008). Age-related differences in insulin-like growth factor-1 receptor signalling regulates Akt/FOXO3a and ERK/Fos pathways in vascular smooth muscle cells. J Cell Physiol, 217(2), 377- 387.
Li, R., Luo, M., Ren, M., Chen, N., Xia, J., Deng, X., Zeng, M., Yan, K., Luo, T., & Wu, J. (2014). Vitronectin regulation of vascular endothelial growth factor-mediated angiogenesis. J Vasc Res, 51(2), 110-117.
Li, X., Lewis, M. T., Huang, J., Gutierrez, C., Osborne, C. K., Wu, M. F., Hilsenbeck, S. G., Pavlick, A., Zhang, X., Chamness, G. C., Wong, H., Rosen, J., & Chang, J. C. (2008). Intrinsic resistance of tumorigenic breast cancer cells to chemotherapy. J Natl Cancer Inst, 100(9), 672-679.
Liang, O. D., Rosenblatt, S., Chhatwal, G. S., & Preissner, K. T. (1997). Identification of novel heparin-binding domains of vitronectin. FEBS Lett, 407(2), 169-172.
Lin, F., Ren, X. D., Pan, Z., Macri, L., Zong, W. X., Tonnesen, M. G., Rafailovich, M., Bar-Sagi, D., & Clark, R. A. (2011). Fibronectin growth factor-binding domains are required for fibroblast survival. J Invest Dermatol, 131(1), 84-98.
Lin, J., Huo, R., Wang, L., Zhou, Z., Sun, Y., Shen, B., Wang, R., & Li, N. (2012). A novel anti-Cyr61 antibody inhibits breast cancer growth and metastasis in vivo. Cancer Immunol Immunother, 61(5), 677-687.
Lin, M. T., Chang, C. C., Chen, S. T., Chang, H. L., Su, J. L., Chau, Y. P., & Kuo, M. L. (2004). Cyr61 expression confers resistance to apoptosis in breast cancer MCF-7
Bibliography 234
cells by a mechanism of NF-kappaB-dependent XIAP up-regulation. J Biol Chem, 279(23), 24015-24023.
Lin, W. W., & Karin, M. (2007). A cytokine-mediated link between innate immunity, inflammation, and cancer. J Clin Invest, 117(5), 1175-1183.
Liu, B. C., Zhang, J. D., Zhang, X. L., Wu, G. Q., & Li, M. X. (2006). Role of connective tissue growth factor (CTGF) module 4 in regulating epithelial mesenchymal transition (EMT) in HK-2 cells. Clin Chim Acta, 373(1-2), 144- 150.
Liu, W., Bloom, D. A., Cance, W. G., Kurenova, E. V., Golubovskaya, V. M., & Hochwald, S. N. (2008). FAK and IGF-IR interact to provide survival signals in human pancreatic adenocarcinoma cells. Carcinogenesis, 29(6), 1096-1107.
Lochter, A., & Bissell, M. J. (1995). Involvement of extracellular matrix constituents in breast cancer. Semin Cancer Biol, 6(3), 165-173.
Loessner, D., Stok, K. S., Lutolf, M. P., Hutmacher, D. W., Clements, J. A., & Rizzi, S. C. (2010). Bioengineered 3D platform to explore cell-ECM interactions and drug resistance of epithelial ovarian cancer cells. Biomaterials, 31(32), 8494-8506.
Lu, P., Weaver, V. M., & Werb, Z. (2012). The extracellular matrix: a dynamic niche in cancer progression. J Cell Biol, 196(4), 395-406.
Ma, C. X., Suman, V. J., Goetz, M., Haluska, P., Moynihan, T., Nanda, R., Olopade, O., Pluard, T., Guo, Z., Chen, H. X., Erlichman, C., Ellis, M. J., & Fleming, G. F. (2013). A phase I trial of the IGF-1R antibody Cixutumumab in combination with temsirolimus in patients with metastatic breast cancer. Breast Cancer Res Treat, 139(1), 145-153.
Madsen, C. D., Ferraris, G. M., Andolfo, A., Cunningham, O., & Sidenius, N. (2007). uPAR-induced cell adhesion and migration: vitronectin provides the key. J Cell Biol, 177(5), 927-939.
Madsen, C. D., & Sidenius, N. (2008). The interaction between urokinase receptor and vitronectin in cell adhesion and signalling. Eur J Cell Biol, 87(8-9), 617-629.
Maeshima, Y., Yerramalla, U. L., Dhanabal, M., Holthaus, K. A., Barbashov, S., Kharbanda, S., Reimer, C., Manfredi, M., Dickerson, W. M., & Kalluri, R. (2001). Extracellular matrix-derived peptide binds to alpha(v)beta(3) integrin and inhibits angiogenesis. J Biol Chem, 276(34), 31959-31968.
Mahabeleshwar, G. H., Feng, W., Phillips, D. R., & Byzova, T. V. (2006). Integrin signalling is critical for pathological angiogenesis. J Exp Med, 203(11), 2495- 2507.
Mahabeleshwar, G. H., Feng, W., Reddy, K., Plow, E. F., & Byzova, T. V. (2007). Mechanisms of integrin-vascular endothelial growth factor receptor cross- activation in angiogenesis. Circ Res, 101(6), 570-580.
Bibliography 235
Maile, L. A., Badley-Clarke, J., & Clemmons, D. R. (2001). Structural analysis of the role of the beta 3 subunit of the alpha V beta 3 integrin in IGF-I signalling. J Cell Sci, 114(Pt 7), 1417-1425.
Maile, L. A., Busby, W. H., Sitko, K., Capps, B. E., Sergent, T., Badley-Clarke, J., & Clemmons, D. R. (2006). Insulin-like growth factor-I signalling in smooth muscle cells is regulated by ligand binding to the 177CYDMKTTC184 sequence of the beta3-subunit of alphaVbeta3. Mol Endocrinol, 20(2), 405-413.
Maile, L. A., & Clemmons, D. R. (2002). The alphaVbeta3 integrin regulates insulin- like growth factor I (IGF-I) receptor phosphorylation by altering the rate of recruitment of the Src-homology 2-containing phosphotyrosine phosphatase-2 to the activated IGF-I receptor. Endocrinology, 143(11), 4259-4264.
Maity, G., Choudhury, P. R., Sen, T., Ganguly, K. K., Sil, H., & Chatterjee, A. (2011). Culture of human breast cancer cell line (MDA-MB-231) on fibronectin-coated surface induces pro-matrix metalloproteinase-9 expression and activity. Tumour Biol, 32(1), 129-138.
Maity, G., Sen, T., & Chatterjee, A. (2011). Laminin induces matrix metalloproteinase-9 expression and activation in human cervical cancer cell line (SiHa). J Cancer Res Clin Oncol, 137(2), 347-357.
Maki, R. G. (2010). Small is beautiful: insulin-like growth factors and their role in growth, development, and cancer. J Clin Oncol, 28(33), 4985-4995.
Mantovani, A., Savino, B., Locati, M., Zammataro, L., Allavena, P., & Bonecchi, R. (2010). The chemokine system in cancer biology and therapy. Cytokine Growth Factor Rev, 21(1), 27-39.
Marlow, R., Honeth, G., Lombardi, S., Cariati, M., Hessey, S., Pipili, A., Mariotti, V., Buchupalli, B., Foster, K., Bonnet, D., Grigoriadis, A., Rameshwar, P., Purushotham, A., Tutt, A., & Dontu, G. (2013). A novel model of dormancy for bone metastatic breast cancer cells. Cancer Res, 73(23), 6886-6899.
Martin, J. L., & Baxter, R. C. (2011). Signalling pathways of insulin-like growth factors (IGFs) and IGF binding protein-3. Growth Factors, 29(6), 235-244.
McLean, G. W., Carragher, N. O., Avizienyte, E., Evans, J., Brunton, V. G., & Frame, M. C. (2005). The role of focal-adhesion kinase in cancer - a new therapeutic opportunity. Nat Rev Cancer, 5(7), 505-515.
Menendez, J. A., Vellon, L., Mehmi, I., Teng, P. K., Griggs, D. W., & Lupu, R. (2005). A novel CYR61-triggered 'CYR61-alphavbeta3 integrin loop' regulates breast cancer cell survival and chemosensitivity through activation of ERK1/ERK2 MAPK signalling pathway. Oncogene, 24(5), 761-779.
Meyer, T., Marshall, J. F., & Hart, I. R. (1998). Expression of alphav integrins and vitronectin receptor identity in breast cancer cells. Br J Cancer, 77(4), 530-536.
Bibliography 236
Milner, S. J., Francis, G. L., Wallace, J. C., Magee, B. A., & Ballard, F. J. (1995). Mutations in the B-domain of insulin-like growth factor-I influence the oxidative folding to yield products with modified biological properties. Biochem J, 308 ( Pt 3), 865-871.
Mitra, S. K., & Schlaepfer, D. D. (2006). Integrin-regulated FAK-Src signalling in normal and cancer cells. Curr Opin Cell Biol, 18(5), 516-523.
Mizejewski, G. J. (1999). Role of integrins in cancer: survey of expression patterns. Proc Soc Exp Biol Med, 222(2), 124-138.
Morello, V., Cabodi, S., Sigismund, S., Camacho-Leal, M. P., Repetto, D., Volante, M., Papotti, M., Turco, E., & Defilippi, P. (2011). beta1 integrin controls EGFR signalling and tumorigenic properties of lung cancer cells. Oncogene, 30(39), 4087-4096.
Morgan, D. O., Edman, J. C., Standring, D. N., Fried, V. A., Smith, M. C., Roth, R. A., & Rutter, W. J. (1987). Insulin-like growth factor II receptor as a multifunctional binding protein. Nature, 329(6137), 301-307.
Morini, M., Mottolese, M., Ferrari, N., Ghiorzo, F., Buglioni, S., Mortarini, R., Noonan, D. M., Natali, P. G., & Albini, A. (2000). The alpha 3 beta 1 integrin is associated with mammary carcinoma cell metastasis, invasion, and gelatinase B (MMP-9) activity. Int J Cancer, 87(3), 336-342.
Moriya, Y., Niki, T., Yamada, T., Matsuno, Y., Kondo, H., & Hirohashi, S. (2001). Increased expression of laminin-5 and its prognostic significance in lung adenocarcinomas of small size. An immunohistochemical analysis of 102 cases. Cancer, 91(6), 1129-1141.
Moro, L., Venturino, M., Bozzo, C., Silengo, L., Altruda, F., Beguinot, L., Tarone, G., & Defilippi, P. (1998). Integrins induce activation of EGF receptor: role in MAP kinase induction and adhesion-dependent cell survival. EMBO J, 17(22), 6622- 6632.
Morris, C. A., Underwood, P. A., Bean, P. A., Sheehan, M., & Charlesworth, J. A. (1994). Relative topography of biologically active domains of human vitronectin. Evidence from monoclonal antibody epitope and denaturation studies. J Biol Chem, 269(38), 23845-23852.
Murphy, L. O., MacKeigan, J. P., & Blenis, J. (2004). A network of immediate early gene products propagates subtle differences in mitogen-activated protein kinase signal amplitude and duration. Mol Cell Biol, 24(1), 144-153.
Murphy, L. O., Smith, S., Chen, R. H., Fingar, D. C., & Blenis, J. (2002). Molecular interpretation of ERK signal duration by immediate early gene products. Nat Cell Biol, 4(8), 556-564.
Bibliography 237
Muschler, J., & Streuli, C. H. (2010). Cell-matrix interactions in mammary gland development and breast cancer. Cold Spring Harb Perspect Biol, 2(10), a003202.
Muthuswamy, S. K. (2011). 3D culture reveals a signalling network. Breast Cancer Res, 13(1), 103.
Naik, M. U., & Naik, U. P. (2006). Junctional adhesion molecule-A-induced endothelial cell migration on vitronectin is integrin alpha v beta 3 specific. J Cell Sci, 119(Pt 3), 490-499.
Nam, J. O., Son, H. N., Jun, E., Cha, K., Lee, B. H., Park, R. W., & Kim, I. S. (2012). FAS1 domain protein inhibits VEGF165-induced angiogenesis by targeting the interaction between VEGFR-2 and alphavbeta3 integrin. Mol Cancer Res, 10(8), 1010-1020.
Naugler, W. E., & Karin, M. (2008). NF-kappaB and cancer-identifying targets and mechanisms. Curr Opin Genet Dev, 18(1), 19-26.
Nelson, C. M., & Bissell, M. J. (2006). Of extracellular matrix, scaffolds, and signalling: tissue architecture regulates development, homeostasis, and cancer. Annu Rev Cell Dev Biol, 22, 287-309.
Neve, R. M., Chin, K., Fridlyand, J., Yeh, J., Baehner, F. L., Fevr, T., Clark, L., Bayani, N., Coppe, J. P., Tong, F., Speed, T., Spellman, P. T., DeVries, S., Lapuk, A., Wang, N. J., Kuo, W. L., Stilwell, J. L., Pinkel, D., Albertson, D. G., Waldman, F. M., McCormick, F., Dickson, R. B., Johnson, M. D., Lippman, M., Ethier, S., Gazdar, A., & Gray, J. W. (2006). A collection of breast cancer cell lines for the study of functionally distinct cancer subtypes. Cancer Cell, 10(6), 515-527.
Nielsen, T. O., Hsu, F. D., Jensen, K., Cheang, M., Karaca, G., Hu, Z., Hernandez- Boussard, T., Livasy, C., Cowan, D., Dressler, L., Akslen, L. A., Ragaz, J., Gown, A. M., Gilks, C. B., van de Rijn, M., & Perou, C. M. (2004). Immunohistochemical and clinical characterization of the basal-like subtype of invasive breast carcinoma. Clin Cancer Res, 10(16), 5367-5374.
Nip, J., Shibata, H., Loskutoff, D. J., Cheresh, D. A., & Brodt, P. (1992). Human melanoma cells derived from lymphatic metastases use integrin alpha v beta 3 to adhere to lymph node vitronectin. J Clin Invest, 90(4), 1406-1413.
Nishimoto, S. K., & Nishimoto, M. (2005). Matrix Gla protein C-terminal region binds to vitronectin. Co-localization suggests binding occurs during tissue development. Matrix Biol, 24(5), 353-361.
Noble, A., Towne, C., Chopin, L., Leavesley, D., & Upton, Z. (2003). Insulin-like growth factor-II bound to vitronectin enhances MCF-7 breast cancer cell migration. Endocrinology, 144(6), 2417-2424.
Bibliography 238
O'Kelly, J., Chung, A., Lemp, N., Chumakova, K., Yin, D., Wang, H. J., Said, J., Gui, D., Miller, C. W., Karlan, B. Y., & Koeffler, H. P. (2008). Functional domains of CCN1 (Cyr61) regulate breast cancer progression. Int J Oncol, 33(1), 59-67.
Oommen, S., Gupta, S. K., & Vlahakis, N. E. (2011). Vascular endothelial growth factor A (VEGF-A) induces endothelial and cancer cell migration through direct binding to integrin {alpha}9{beta}1: identification of a specific {alpha}9{beta}1 binding site. J Biol Chem, 286(2), 1083-1092.
Orimo, A., Gupta, P. B., Sgroi, D. C., Arenzana-Seisdedos, F., Delaunay, T., Naeem, R., Carey, V. J., Richardson, A. L., & Weinberg, R. A. (2005). Stromal fibroblasts present in invasive human breast carcinomas promote tumor growth and angiogenesis through elevated SDF-1/CXCL12 secretion. Cell, 121(3), 335-348.
Page-McCaw, A., Ewald, A. J., & Werb, Z. (2007). Matrix metalloproteinases and the regulation of tissue remodelling. Nat Rev Mol Cell Biol, 8(3), 221-233.
Pandey, D. P., Lappano, R., Albanito, L., Madeo, A., Maggiolini, M., & Picard, D. (2009). Estrogenic GPR30 signalling induces proliferation and migration of breast cancer cells through CTGF. EMBO J, 28(5), 523-532.
Parisi, M. S., Gazzerro, E., Rydziel, S., & Canalis, E. (2006). Expression and regulation of CCN genes in murine osteoblasts. Bone, 38(5), 671-677.
Park, C. C., Zhang, H., Pallavicini, M., Gray, J. W., Baehner, F., Park, C. J., & Bissell, M. J. (2006). Beta1 integrin inhibitory antibody induces apoptosis of breast cancer cells, inhibits growth, and distinguishes malignant from normal phenotype in three dimensional cultures and in vivo. Cancer Res, 66(3), 1526- 1535.
Park, C. C., Zhang, H. J., Yao, E. S., Park, C. J., & Bissell, M. J. (2008). Beta1 integrin inhibition dramatically enhances radiotherapy efficacy in human breast cancer xenografts. Cancer Res, 68(11), 4398-4405.
Parsons, J. T., & Parsons, S. J. (1997). Src family protein tyrosine kinases: cooperating with growth factor and adhesion signalling pathways. Curr Opin Cell Biol, 9(2), 187-192.
Paszek, M. J., & Weaver, V. M. (2004). The tension mounts: mechanics meets morphogenesis and malignancy. J Mammary Gland Biol Neoplasia, 9(4), 325- 342.
Perbal, B. V., & Takigawa, M. (2005). CCN proteins: a new family of cell growth and differentiation regulators: Imperial College Press. Perou, C. M., Sorlie, T., Eisen, M. B., van de Rijn, M., Jeffrey, S. S., Rees, C. A., Pollack, J. R., Ross, D. T., Johnsen, H., Akslen, L. A., Fluge, O., Pergamenschikov, A., Williams, C., Zhu, S. X., Lonning, P. E., Borresen-Dale, A. L., Brown, P. O., & Botstein, D. (2000). Molecular portraits of human breast tumours. Nature, 406(6797), 747-752.
Bibliography 239
Persad, S., Attwell, S., Gray, V., Mawji, N., Deng, J. T., Leung, D., Yan, J., Sanghera, J., Walsh, M. P., & Dedhar, S. (2001). Regulation of protein kinase B/Akt-serine 473 phosphorylation by integrin-linked kinase: critical roles for kinase activity and amino acids arginine 211 and serine 343. J Biol Chem, 276(29), 27462- 27469.
Pirazzoli, V., Ferraris, G. M., & Sidenius, N. (2013). Direct evidence of the importance of vitronectin and its interaction with the urokinase receptor in tumor growth. Blood, 121(12), 2316-2323.
Place, A. E., Jin Huh, S., & Polyak, K. (2011). The microenvironment in breast cancer progression: biology and implications for treatment. Breast Cancer Res, 13(6), 227.
Playford, M. P., & Schaller, M. D. (2004). The interplay between Src and integrins in normal and tumor biology. Oncogene, 23(48), 7928-7946.
Pola, C., Formenti, S. C., & Schneider, R. J. (2013). Vitronectin-alphavbeta3 integrin engagement directs hypoxia-resistant mTOR activity and sustained protein synthesis linked to invasion by breast cancer cells. Cancer Res, 73(14), 4571- 4578.
Pollak, M. (2012). The insulin and insulin-like growth factor receptor family in neoplasia: an update. Nat Rev Cancer, 12(3), 159-169.
Pollak, M. N., Schernhammer, E. S., & Hankinson, S. E. (2004). Insulin-like growth factors and neoplasia. Nat Rev Cancer, 4(7), 505-518.
Polyak, K., & Kalluri, R. (2010). The role of the microenvironment in mammary gland development and cancer. Cold Spring Harb Perspect Biol, 2(11), a003244.
Prat, A., Parker, J. S., Karginova, O., Fan, C., Livasy, C., Herschkowitz, J. I., He, X., & Perou, C. M. (2010). Phenotypic and molecular characterization of the claudin- low intrinsic subtype of breast cancer. Breast Cancer Res, 12(5), R68.
Preissner, K. T., & Reuning, U. (2011). Vitronectin in vascular context: facets of a multitalented matricellular protein. Semin Thromb Hemost, 37(4), 408-424.
Preissner, K. T., & Seiffert, D. (1998). Role of vitronectin and its receptors in haemostasis and vascular remodeling. Thromb Res, 89(1), 1-21.
Quail, D. F., & Joyce, J. A. (2013). Microenvironmental regulation of tumor progression and metastasis. Nat Med, 19(11), 1423-1437.
Quan, T., He, T., Shao, Y., Lin, L., Kang, S., Voorhees, J. J., & Fisher, G. J. (2006). Elevated cysteine-rich 61 mediates aberrant collagen homeostasis in chronologically aged and photoaged human skin. Am J Pathol, 169(2), 482-490.
Rapraeger, A. C., Ell, B. J., Roy, M., Li, X., Morrison, O. R., Thomas, G. M., & Beauvais, D. M. (2013). Vascular endothelial-cadherin stimulates syndecan-1-
Bibliography 240
coupled insulin-like growth factor-1 receptor and cross-talk between alphaVbeta3 integrin and vascular endothelial growth factor receptor 2 at the onset of endothelial cell dissemination during angiogenesis. FEBS J, 280(10), 2194-2206.
Ravizzini, G., Turkbey, B., Barrett, T., Kobayashi, H., & Choyke, P. L. (2009). Nanoparticles in sentinel lymph node mapping. Wiley Interdiscip Rev Nanomed Nanobiotechnol, 1(6), 610-623.
Rea, V. E., Lavecchia, A., Di Giovanni, C., Rossi, F. W., Gorrasi, A., Pesapane, A., de Paulis, A., Ragno, P., & Montuori, N. (2013). Discovery of new small molecules targeting the vitronectin-binding site of the urokinase receptor that block cancer cell invasion. Mol Cancer Ther, 12(8), 1402-1416.
Reilly, J. T., & Nash, J. R. (1988). Vitronectin (serum spreading factor): its localisation in normal and fibrotic tissue. J Clin Pathol, 41(12), 1269-1272.
Reinartz, J., Schafer, B., Batrla, R., Klein, C. E., & Kramer, M. D. (1995). Plasmin abrogates alpha v beta 5-mediated adhesion of a human keratinocyte cell line (HaCaT) to vitronectin. Exp Cell Res, 220(2), 274-282.
Ren, W., Sun, X., Wang, K., Feng, H., Liu, Y., Fei, C., Wan, S., Wang, W., Luo, J., Shi, Q., Tang, M., Zuo, G., Weng, Y., He, T., & Zhang, Y. (2014). BMP9 inhibits the bone metastasis of breast cancer cells by downregulating CCN2 (connective tissue growth factor, CTGF) expression. Mol Biol Rep, 41(3), 1373-1383.
Resnik, J. L., Reichart, D. B., Huey, K., Webster, N. J., & Seely, B. L. (1998). Elevated insulin-like growth factor I receptor autophosphorylation and kinase activity in human breast cancer. Cancer Res, 58(6), 1159-1164.
Rezaeipoor, R., Chaney, E. J., Oldenburg, A. L., & Boppart, S. A. (2009). Expression order of alpha-v and beta-3 integrin subunits in the N-methyl-N-nitrosourea- induced rat mammary tumor model. Cancer Invest, 27(5), 496-503.
Ricono, J. M., Huang, M., Barnes, L. A., Lau, S. K., Weis, S. M., Schlaepfer, D. D., Hanks, S. K., & Cheresh, D. A. (2009). Specific cross-talk between epidermal growth factor receptor and integrin alphavbeta5 promotes carcinoma cell invasion and metastasis. Cancer Res, 69(4), 1383-1391.
Rinderknecht, E., & Humbel, R. E. (1978). The amino acid sequence of human insulin- like growth factor I and its structural homology with proinsulin. J Biol Chem, 253(8), 2769-2776.
Rondeau, G., Abedinpour, P., Desai, P., Baron, V. T., Borgstrom, P., & Welsh, J. (2014). Effects of different tissue microenvironments on gene expression in breast cancer cells. PLoS One, 9(7), e101160.
Ronnov-Jessen, L., & Bissell, M. J. (2009). Breast cancer by proxy: can the microenvironment be both the cause and consequence? Trends Mol Med, 15(1), 5-13.
Bibliography 241
Roskelley, C. D., Srebrow, A., & Bissell, M. J. (1995). A hierarchy of ECM-mediated signalling regulates tissue-specific gene expression. Curr Opin Cell Biol, 7(5), 736-747.
Ross, J. S., Fletcher, J. A., Linette, G. P., Stec, J., Clark, E., Ayers, M., Symmans, W. F., Pusztai, L., & Bloom, K. J. (2003). The Her-2/neu gene and protein in breast cancer 2003: biomarker and target of therapy. Oncologist, 8(4), 307-325.
Rotwein, P., Pollock, K. M., Didier, D. K., & Krivi, G. G. (1986). Organization and sequence of the human insulin-like growth factor I gene. Alternative RNA processing produces two insulin-like growth factor I precursor peptides. J Biol Chem, 261(11), 4828-4832.
Russo, V. C., Schutt, B. S., Andaloro, E., Ymer, S. I., Hoeflich, A., Ranke, M. B., Bach, L. A., & Werther, G. A. (2005). Insulin-like growth factor binding protein-2 binding to extracellular matrix plays a critical role in neuroblastoma cell proliferation, migration, and invasion. Endocrinology, 146(10), 4445-4455.
Sachdev, D., & Yee, D. (2007). Disrupting insulin-like growth factor signalling as a potential cancer therapy. Mol Cancer Ther, 6(1), 1-12.
Saegusa, J., Yamaji, S., Ieguchi, K., Wu, C. Y., Lam, K. S., Liu, F. T., Takada, Y. K., & Takada, Y. (2009). The direct binding of insulin-like growth factor-1 (IGF-1) to integrin alphavbeta3 is involved in IGF-1 signalling. J Biol Chem, 284(36), 24106-24114.
Salasznyk, R. M., Klees, R. F., Hughlock, M. K., & Plopper, G. E. (2004). ERK signalling pathways regulate the osteogenic differentiation of human mesenchymal stem cells on collagen I and vitronectin. Cell Commun Adhes, 11(5-6), 137-153.
Salatino, M., Schillaci, R., Proietti, C. J., Carnevale, R., Frahm, I., Molinolo, A. A., Iribarren, A., Charreau, E. H., & Elizalde, P. V. (2004). Inhibition of in vivo breast cancer growth by antisense oligodeoxynucleotides to type I insulin-like growth factor receptor mRNA involves inactivation of ErbBs, PI-3K/Akt and p42/p44 MAPK signalling pathways but not modulation of progesterone receptor activity. Oncogene, 23(30), 5161-5174.
Samani, A. A., & Brodt, P. (2001). The receptor for the type I insulin-like growth factor and its ligands regulate multiple cellular functions that impact on metastasis. Surg Oncol Clin N Am, 10(2), 289-312, viii.
Samani, A. A., Yakar, S., LeRoith, D., & Brodt, P. (2007). The role of the IGF system in cancer growth and metastasis: overview and recent insights. Endocr Rev, 28(1), 20-47.
Sampath, D., Winneker, R. C., & Zhang, Z. (2001). Cyr61, a member of the CCN family, is required for MCF-7 cell proliferation: regulation by 17beta-estradiol and overexpression in human breast cancer. Endocrinology, 142(6), 2540-2548.
Bibliography 242
Sangaletti, S., Stoppacciaro, A., Guiducci, C., Torrisi, M. R., & Colombo, M. P. (2003). Leukocyte, rather than tumor-produced SPARC, determines stroma and collagen type IV deposition in mammary carcinoma. J Exp Med, 198(10), 1475-1485.
Santarpia, L., Lippman, S. M., & El-Naggar, A. K. (2012). Targeting the MAPK-RAS- RAF signalling pathway in cancer therapy. Expert Opin Ther Targets, 16(1), 103-119.
Scatena, M., Almeida, M., Chaisson, M. L., Fausto, N., Nicosia, R. F., & Giachelli, C. M. (1998). NF-kappaB mediates alphavbeta3 integrin-induced endothelial cell survival. J Cell Biol, 141(4), 1083-1093.
Scheel, C., & Weinberg, R. A. (2012). Cancer stem cells and epithelial-mesenchymal transition: concepts and molecular links. Semin Cancer Biol, 22(5-6), 396-403.
Schlaepfer, D. D., & Hunter, T. (1998). Integrin signalling and tyrosine phosphorylation: just the FAKs? Trends Cell Biol, 8(4), 151-157.
Schlaepfer, D. D., Mitra, S. K., & Ilic, D. (2004). Control of motile and invasive cell phenotypes by focal adhesion kinase. Biochim Biophys Acta, 1692(2-3), 77-102.
Schmeichel, K. L., & Bissell, M. J. (2003). Modeling tissue-specific signalling and organ function in three dimensions. J Cell Sci, 116(Pt 12), 2377-2388.
Schmitt-Ney, M., & Habener, J. F. (2004). Cell-density-dependent regulation of actin gene expression due to changes in actin treadmilling. Exp Cell Res, 295(1), 236- 244.
Schultz, G. S., & Wysocki, A. (2009). Interactions between extracellular matrix and growth factors in wound healing. Wound Repair Regen, 17(2), 153-162.
Schvartz, I., Seger, D., & Shaltiel, S. (1999). Vitronectin. Int J Biochem Cell Biol, 31(5), 539-544.
Schwartz, M. A., & Ginsberg, M. H. (2002). Networks and crosstalk: integrin signalling spreads. Nat Cell Biol, 4(4), E65-68.
Secker, G. A., Shortt, A. J., Sampson, E., Schwarz, Q. P., Schultz, G. S., & Daniels, J. T. (2008). TGFbeta stimulated re-epithelialisation is regulated by CTGF and Ras/MEK/ERK signalling. Exp Cell Res, 314(1), 131-142.
Seger, D., Gechtman, Z., & Shaltiel, S. (1998). Phosphorylation of vitronectin by casein kinase II. Identification of the sites and their promotion of cell adhesion and spreading. J Biol Chem, 273(38), 24805-24813.
Seiffert, D. (1997). Constitutive and regulated expression of vitronectin. Histol Histopathol, 12(3), 787-797.
Bibliography 243
Seiffert, D., & Loskutoff, D. J. (1991). Evidence that type 1 plasminogen activator inhibitor binds to the somatomedin B domain of vitronectin. J Biol Chem, 266(5), 2824-2830.
Seiffert, D., & Loskutoff, D. J. (1996). Type 1 plasminogen activator inhibitor induces multimerization of plasma vitronectin. A suggested mechanism for the generation of the tissue form of vitronectin in vivo. J Biol Chem, 271(47), 29644-29651.
Sethi, T., Rintoul, R. C., Moore, S. M., MacKinnon, A. C., Salter, D., Choo, C., Chilvers, E. R., Dransfield, I., Donnelly, S. C., Strieter, R., & Haslett, C. (1999). Extracellular matrix proteins protect small cell lung cancer cells against apoptosis: a mechanism for small cell lung cancer growth and drug resistance in vivo. Nat Med, 5(6), 662-668.
Shaltiel, S., Schvartz, I., Korc-Grodzicki, B., & Kreizman, T. (1993). Evidence for an extra-cellular function for protein kinase A. Mol Cell Biochem, 127-128, 283- 291.
Sheppard, K., Kinross, K. M., Solomon, B., Pearson, R. B., & Phillips, W. A. (2012). Targeting PI3 kinase/AKT/mTOR signalling in cancer. Crit Rev Oncog, 17(1), 69-95.
Shimo, T., Kubota, S., Yoshioka, N., Ibaragi, S., Isowa, S., Eguchi, T., Sasaki, A., & Takigawa, M. (2006). Pathogenic role of connective tissue growth factor (CTGF/CCN2) in osteolytic metastasis of breast cancer. J Bone Miner Res, 21(7), 1045-1059.
Shintani, Y., Maeda, M., Chaika, N., Johnson, K. R., & Wheelock, M. J. (2008). Collagen I promotes epithelial-to-mesenchymal transition in lung cancer cells via transforming growth factor-beta signalling. Am J Respir Cell Mol Biol, 38(1), 95- 104.
Siddle, K. (2012). Molecular basis of signalling specificity of insulin and IGF receptors: neglected corners and recent advances. Front Endocrinol (Lausanne), 3, 34.
Sieg, D. J., Hauck, C. R., Ilic, D., Klingbeil, C. K., Schaefer, E., Damsky, C. H., & Schlaepfer, D. D. (2000). FAK integrates growth-factor and integrin signals to promote cell migration. Nat Cell Biol, 2(5), 249-256.
Siwanowicz, I., Popowicz, G. M., Wisniewska, M., Huber, R., Kuenkele, K. P., Lang, K., Engh, R. A., & Holak, T. A. (2005). Structural basis for the regulation of insulin-like growth factors by IGF binding proteins. Structure, 13(1), 155-167.
Smith, P. J., Spurrell, E. L., Coakley, J., Hinds, C. J., Ross, R. J., Krainer, A. R., & Chew, S. L. (2002). An exonic splicing enhancer in human IGF-I pre-mRNA mediates recognition of alternative exon 5 by the serine-arginine protein splicing factor-2/alternative splicing factor. Endocrinology, 143(1), 146-154.
Bibliography 244
Sorlie, T., Perou, C. M., Tibshirani, R., Aas, T., Geisler, S., Johnsen, H., Hastie, T., Eisen, M. B., van de Rijn, M., Jeffrey, S. S., Thorsen, T., Quist, H., Matese, J. C., Brown, P. O., Botstein, D., Lonning, P. E., & Borresen-Dale, A. L. (2001). Gene expression patterns of breast carcinomas distinguish tumor subclasses with clinical implications. Proc Natl Acad Sci U S A, 98(19), 10869-10874.
Sorlie, T., Tibshirani, R., Parker, J., Hastie, T., Marron, J. S., Nobel, A., Deng, S., Johnsen, H., Pesich, R., Geisler, S., Demeter, J., Perou, C. M., Lonning, P. E., Brown, P. O., Borresen-Dale, A. L., & Botstein, D. (2003). Repeated observation of breast tumor subtypes in independent gene expression data sets. Proc Natl Acad Sci U S A, 100(14), 8418-8423.
Sorlie, T., Wang, Y., Xiao, C., Johnsen, H., Naume, B., Samaha, R. R., & Borresen- Dale, A. L. (2006). Distinct molecular mechanisms underlying clinically relevant subtypes of breast cancer: gene expression analyses across three different platforms. BMC Genomics, 7, 127.
Sotiriou, C., Neo, S. Y., McShane, L. M., Korn, E. L., Long, P. M., Jazaeri, A., Martiat, P., Fox, S. B., Harris, A. L., & Liu, E. T. (2003). Breast cancer classification and prognosis based on gene expression profiles from a population-based study. Proc Natl Acad Sci U S A, 100(18), 10393-10398.
Spencer, V. A., Xu, R., & Bissell, M. J. (2007). Extracellular matrix, nuclear and chromatin structure, and gene expression in normal tissues and malignant tumors: a work in progress. Adv Cancer Res, 97, 275-294.
Stanley, K. K. (1986). Homology with hemopexin suggests a possible scavenging function for S-protein/vitronectin. FEBS Lett, 199(2), 249-253.
Stockmann, A., Hess, S., Declerck, P., Timpl, R., & Preissner, K. T. (1993). Multimeric vitronectin. Identification and characterization of conformation-dependent self- association of the adhesive protein. J Biol Chem, 268(30), 22874-22882.
Streck, R. D., Wood, T. L., Hsu, M. S., & Pintar, J. E. (1992). Insulin-like growth factor I and II and insulin-like growth factor binding protein-2 RNAs are expressed in adjacent tissues within rat embryonic and fetal limbs. Dev Biol, 151(2), 586-596.
Streuli, C. H. (2009). Integrins and cell-fate determination. J Cell Sci, 122(Pt 2), 171- 177.
Stupack, D. G., & Cheresh, D. A. (2002). Get a ligand, get a life: integrins, signalling and cell survival. J Cell Sci, 115(Pt 19), 3729-3738.
Sun, H. Z., Yang, T. W., Zang, W. J., & Wu, S. F. (2010). Dehydroepiandrosterone- induced proliferation of prostatic epithelial cell is mediated by NFKB via PI3K/AKT signalling pathway. J Endocrinol, 204(3), 311-318.
Sureshbabu, A., Okajima, H., Yamanaka, D., Shastri, S., Tonner, E., Rae, C., Szymanowska, M., Shand, J. H., Takahashi, S., Beattie, J., Allan, G. J., & Flint,
Bibliography 245
D. J. (2009). IGFBP-5 induces epithelial and fibroblast responses consistent with the fibrotic response. Biochem Soc Trans, 37(Pt 4), 882-885.
Surmacz, E. (2000). Function of the IGF-I receptor in breast cancer. J Mammary Gland Biol Neoplasia, 5(1), 95-105.
Suzuma, K., Naruse, K., Suzuma, I., Takahara, N., Ueki, K., Aiello, L. P., & King, G. L. (2000). Vascular endothelial growth factor induces expression of connective tissue growth factor via KDR, Flt1, and phosphatidylinositol 3-kinase-akt- dependent pathways in retinal vascular cells. J Biol Chem, 275(52), 40725- 40731.
Tai, Y. T., Podar, K., Catley, L., Tseng, Y. H., Akiyama, M., Shringarpure, R., Burger, R., Hideshima, T., Chauhan, D., Mitsiades, N., Richardson, P., Munshi, N. C., Kahn, C. R., Mitsiades, C., & Anderson, K. C. (2003). Insulin-like growth factor-1 induces adhesion and migration in human multiple myeloma cells via activation of beta1-integrin and phosphatidylinositol 3'-kinase/AKT signalling. Cancer Res, 63(18), 5850-5858.
Taipale, J., & Keski-Oja, J. (1997). Growth factors in the extracellular matrix. FASEB J, 11(1), 51-59.
Takahashi, S., Harigae, H., Ishii, K. K., Inomata, M., Fujiwara, T., Yokoyama, H., Ishizawa, K., Kameoka, J., Licht, J. D., Sasaki, T., & Kaku, M. (2005). Over- expression of Flt3 induces NF-kappaB pathway and increases the expression of IL-6. Leuk Res, 29(8), 893-899.
Taube, J. H., Herschkowitz, J. I., Komurov, K., Zhou, A. Y., Gupta, S., Yang, J., Hartwell, K., Onder, T. T., Gupta, P. B., Evans, K. W., Hollier, B. G., Ram, P. T., Lander, E. S., Rosen, J. M., Weinberg, R. A., & Mani, S. A. (2010). Core epithelial-to-mesenchymal transition interactome gene-expression signature is associated with claudin-low and metaplastic breast cancer subtypes. Proc Natl Acad Sci U S A, 107(35), 15449-15454.
Thornton, S., Sowders, D., Aronow, B., Witte, D. P., Brunner, H. I., Giannini, E. H., & Hirsch, R. (2002). DNA microarray analysis reveals novel gene expression profiles in collagen-induced arthritis. Clin Immunol, 105(2), 155-168.
Todorovic, V., Chen, C. C., Hay, N., & Lau, L. F. (2005). The matrix protein CCN1 (CYR61) induces apoptosis in fibroblasts. J Cell Biol, 171(3), 559-568.
Tognon, C. E., & Sorensen, P. H. (2012). Targeting the insulin-like growth factor 1 receptor (IGF1R) signalling pathway for cancer therapy. Expert Opin Ther Targets, 16(1), 33-48.
Tomasini-Johansson, B. R., Sundberg, C., Lindmark, G., Gailit, J. O., & Rubin, K. (1994). Vitronectin in colorectal adenocarcinoma--synthesis by stromal cells in culture. Exp Cell Res, 214(1), 303-312.
Bibliography 246
Tomasini, B. R., & Mosher, D. F. (1991). Vitronectin. Prog Hemost Thromb, 10, 269- 305.
Tong, D., Czerwenka, K., Sedlak, J., Schneeberger, C., Schiebel, I., Concin, N., Leodolter, S., & Zeillinger, R. (1999). Association of in vitro invasiveness and gene expression of estrogen receptor, progesterone receptor, pS2 and plasminogen activator inhibitor-1 in human breast cancer cell lines. Breast Cancer Res Treat, 56(1), 91-97.
Toniolo, P., Bruning, P. F., Akhmedkhanov, A., Bonfrer, J. M., Koenig, K. L., Lukanova, A., Shore, R. E., & Zeleniuch-Jacquotte, A. (2000). Serum insulin- like growth factor-I and breast cancer. Int J Cancer, 88(5), 828-832.
Tricoli, J. V., Rall, L. B., Scott, J., Bell, G. I., & Shows, T. B. (1984). Localization of insulin-like growth factor genes to human chromosomes 11 and 12. Nature, 310(5980), 784-786.
Triulzi, T., Casalini, P., Sandri, M., Ratti, M., Carcangiu, M. L., Colombo, M. P., Balsari, A., Menard, S., Orlandi, R., & Tagliabue, E. (2013). Neoplastic and stromal cells contribute to an extracellular matrix gene expression profile defining a breast cancer subtype likely to progress. PLoS One, 8(2), e56761.
Tsai, M. S., Bogart, D. F., Castaneda, J. M., Li, P., & Lupu, R. (2002). Cyr61 promotes breast tumorigenesis and cancer progression. Oncogene, 21(53), 8178-8185.
Tsai, M. S., Hornby, A. E., Lakins, J., & Lupu, R. (2000). Expression and function of CYR61, an angiogenic factor, in breast cancer cell lines and tumor biopsies. Cancer Res, 60(20), 5603-5607.
Uhm, J. H., Dooley, N. P., Kyritsis, A. P., Rao, J. S., & Gladson, C. L. (1999). Vitronectin, a glioma-derived extracellular matrix protein, protects tumor cells from apoptotic death. Clin Cancer Res, 5(6), 1587-1594.
Ulrich, J. T., Schenck, J. R., Rittenhouse, H. G., Shaper, N. L., & Shaper, J. H. (1986). Monoclonal antibodies to bovine UDP-galactosyltransferase. Characterization, cross-reactivity, and utilization as structural probes. J Biol Chem, 261(17), 7975- 7981.
Upton, Z., Cuttle, L., Noble, A., Kempf, M., Topping, G., Malda, J., Xie, Y., Mill, J., Harkin, D. G., Kravchuk, O., Leavesley, D. I., & Kimble, R. M. (2008). Vitronectin: growth factor complexes hold potential as a wound therapy approach. J Invest Dermatol, 128(6), 1535-1544.
Upton, Z., Webb, H., Hale, K., Yandell, C. A., McMurtry, J. P., Francis, G. L., & Ballard, F. J. (1999). Identification of vitronectin as a novel insulin-like growth factor-II binding protein. Endocrinology, 140(6), 2928-2931. van 't Veer, L. J., Dai, H., van de Vijver, M. J., He, Y. D., Hart, A. A., Mao, M., Peterse, H. L., van der Kooy, K., Marton, M. J., Witteveen, A. T., Schreiber, G. J., Kerkhoven, R. M., Roberts, C., Linsley, P. S., Bernards, R., & Friend, S. H.
Bibliography 247
(2002). Gene expression profiling predicts clinical outcome of breast cancer. Nature, 415(6871), 530-536.
Van Gelder, R. N., von Zastrow, M. E., Yool, A., Dement, W. C., Barchas, J. D., & Eberwine, J. H. (1990). Amplified RNA synthesized from limited quantities of heterogeneous cDNA. Proc Natl Acad Sci U S A, 87(5), 1663-1667.
Van Lonkhuyzen, D. R., Hollier, B. G., Shooter, G. K., Leavesley, D. I., & Upton, Z. (2007). Chimeric vitronectin:insulin-like growth factor proteins enhance cell growth and migration through co-activation of receptors. Growth Factors, 25(5), 295-308.
Van Obberghen-Schilling, E., Tucker, R. P., Saupe, F., Gasser, I., Cseh, B., & Orend, G. (2011). Fibronectin and tenascin-C: accomplices in vascular morphogenesis during development and tumor growth. Int J Dev Biol, 55(4-5), 511-525.
Vasaturo, F., Sallusti, E., Gradilone, A., Malacrino, C., Nardo, T., Avagnano, G., Agliano, A. M., Granato, T., De Vincenzi, B., Coppotelli, G., Marzullo, A., Soda, G., Simonelli, L., Modesti, M., & Scarpa, S. (2005). Comparison of extracellular matrix and apoptotic markers between benign lesions and carcinomas in human breast. Int J Oncol, 27(4), 1005-1011.
Veevers-Lowe, J., Ball, S. G., Shuttleworth, A., & Kielty, C. M. (2011). Mesenchymal stem cell migration is regulated by fibronectin through alpha5beta1-integrin- mediated activation of PDGFR-beta and potentiation of growth factor signals. J Cell Sci, 124(Pt 8), 1288-1300.
Voduc, K. D., Cheang, M. C., Tyldesley, S., Gelmon, K., Nielsen, T. O., & Kennecke, H. (2010). Breast cancer subtypes and the risk of local and regional relapse. J Clin Oncol, 28(10), 1684-1691.
Walsh, E. M., Kim, R., Del Valle, L., Weaver, M., Sheffield, J., Lazarovici, P., & Marcinkiewicz, C. (2012). Importance of interaction between nerve growth factor and alpha9beta1 integrin in glial tumor angiogenesis. Neuro Oncol, 14(7), 890-901.
Waltz, D. A., Natkin, L. R., Fujita, R. M., Wei, Y., & Chapman, H. A. (1997). Plasmin and plasminogen activator inhibitor type 1 promote cellular motility by regulating the interaction between the urokinase receptor and vitronectin. J Clin Invest, 100(1), 58-67.
Wang, J. Y., Gualco, E., Peruzzi, F., Sawaya, B. E., Passiatore, G., Marcinkiewicz, C., Staniszewska, I., Ferrante, P., Amini, S., Khalili, K., & Reiss, K. (2007). Interaction between serine phosphorylated IRS-1 and beta1-integrin affects the stability of neuronal processes. J Neurosci Res, 85(11), 2360-2373.
Wang, L., & Adamo, M. L. (2000). Cell density influences insulin-like growth factor I gene expression in a cell type-specific manner. Endocrinology, 141(7), 2481- 2489.
Bibliography 248
Wang, M. Y., Chen, P. S., Prakash, E., Hsu, H. C., Huang, H. Y., Lin, M. T., Chang, K. J., & Kuo, M. L. (2009). Connective tissue growth factor confers drug resistance in breast cancer through concomitant up-regulation of Bcl-xL and cIAP1. Cancer Res, 69(8), 3482-3491.
Wary, K. K., Mariotti, A., Zurzolo, C., & Giancotti, F. G. (1998). A requirement for caveolin-1 and associated kinase Fyn in integrin signalling and anchorage- dependent cell growth. Cell, 94(5), 625-634.
Weaver, V. M., Lelievre, S., Lakins, J. N., Chrenek, M. A., Jones, J. C., Giancotti, F., Werb, Z., & Bissell, M. J. (2002). beta4 integrin-dependent formation of polarized three-dimensional architecture confers resistance to apoptosis in normal and malignant mammary epithelium. Cancer Cell, 2(3), 205-216.
Webb, D. J., Donais, K., Whitmore, L. A., Thomas, S. M., Turner, C. E., Parsons, J. T., & Horwitz, A. F. (2004). FAK-Src signalling through paxillin, ERK and MLCK regulates adhesion disassembly. Nat Cell Biol, 6(2), 154-161.
Wei, Y., Eble, J. A., Wang, Z., Kreidberg, J. A., & Chapman, H. A. (2001). Urokinase receptors promote beta1 integrin function through interactions with integrin alpha3beta1. Mol Biol Cell, 12(10), 2975-2986.
Wei, Y., Tang, C. H., Kim, Y., Robillard, L., Zhang, F., Kugler, M. C., & Chapman, H. A. (2007). Urokinase receptors are required for alpha 5 beta 1 integrin-mediated signalling in tumor cells. J Biol Chem, 282(6), 3929-3939.
Wei, Y., Waltz, D. A., Rao, N., Drummond, R. J., Rosenberg, S., & Chapman, H. A. (1994). Identification of the urokinase receptor as an adhesion receptor for vitronectin. J Biol Chem, 269(51), 32380-32388.
Weigelt, B., & Bissell, M. J. (2008). Unraveling the microenvironmental influences on the normal mammary gland and breast cancer. Semin Cancer Biol, 18(5), 311- 321.
Weigelt, B., Ghajar, C. M., & Bissell, M. J. (2014). The need for complex 3D culture models to unravel novel pathways and identify accurate biomarkers in breast cancer. Adv Drug Deliv Rev, 69-70, 42-51.
Weigelt, B., Peterse, J. L., & van 't Veer, L. J. (2005). Breast cancer metastasis: markers and models. Nat Rev Cancer, 5(8), 591-602.
Wijelath, E. S., Rahman, S., Namekata, M., Murray, J., Nishimura, T., Mostafavi-Pour, Z., Patel, Y., Suda, Y., Humphries, M. J., & Sobel, M. (2006). Heparin-II domain of fibronectin is a vascular endothelial growth factor-binding domain: enhancement of VEGF biological activity by a singular growth factor/matrix protein synergism. Circ Res, 99(8), 853-860.
Witsch, E., Sela, M., & Yarden, Y. (2010). Roles for growth factors in cancer progression. Physiology (Bethesda), 25(2), 85-101.
Bibliography 249
Wolff, A. C., Hammond, M. E., Hicks, D. G., Dowsett, M., McShane, L. M., Allison, K. H., Allred, D. C., Bartlett, J. M., Bilous, M., Fitzgibbons, P., Hanna, W., Jenkins, R. B., Mangu, P. B., Paik, S., Perez, E. A., Press, M. F., Spears, P. A., Vance, G. H., Viale, G., & Hayes, D. F. (2014). Recommendations for human epidermal growth factor receptor 2 testing in breast cancer: American Society of Clinical Oncology/College of American Pathologists clinical practice guideline update. Arch Pathol Lab Med, 138(2), 241-256.
Wong, N. C., Mueller, B. M., Barbas, C. F., Ruminski, P., Quaranta, V., Lin, E. C., & Smith, J. W. (1998). Alphav integrins mediate adhesion and migration of breast carcinoma cell lines. Clin Exp Metastasis, 16(1), 50-61.
Wood, T. L. (1995). Gene-targeting and transgenic approaches to IGF and IGF binding protein function. Am J Physiol, 269(4 Pt 1), E613-622.
Wozniak, M. A., Desai, R., Solski, P. A., Der, C. J., & Keely, P. J. (2003). ROCK- generated contractility regulates breast epithelial cell differentiation in response to the physical properties of a three-dimensional collagen matrix. J Cell Biol, 163(3), 583-595.
Wu, Z. S., Wu, Q., Yang, J. H., Wang, H. Q., Ding, X. D., Yang, F., & Xu, X. C. (2008). Prognostic significance of MMP-9 and TIMP-1 serum and tissue expression in breast cancer. Int J Cancer, 122(9), 2050-2056.
Xi, G., Maile, L. A., Yoo, S. E., & Clemmons, D. R. (2008). Expression of the human beta3 integrin subunit in mouse smooth muscle cells enhances IGF-I-stimulated signalling and proliferation. J Cell Physiol, 214(2), 306-315.
Xie, D., Miller, C. W., O'Kelly, J., Nakachi, K., Sakashita, A., Said, J. W., Gornbein, J., & Koeffler, H. P. (2001). Breast cancer. Cyr61 is overexpressed, estrogen- inducible, and associated with more advanced disease. J Biol Chem, 276(17), 14187-14194.
Xie, D., Nakachi, K., Wang, H., Elashoff, R., & Koeffler, H. P. (2001). Elevated levels of connective tissue growth factor, WISP-1, and CYR61 in primary breast cancers associated with more advanced features. Cancer Res, 61(24), 8917-8923.
Xie, D., Yin, D., Wang, H. J., Liu, G. T., Elashoff, R., Black, K., & Koeffler, H. P. (2004). Levels of expression of CYR61 and CTGF are prognostic for tumor progression and survival of individuals with gliomas. Clin Cancer Res, 10(6), 2072-2081.
Xing, H., Weng, D., Chen, G., Tao, W., Zhu, T., Yang, X., Meng, L., Wang, S., Lu, Y., & Ma, D. (2008). Activation of fibronectin/PI-3K/Akt2 leads to chemoresistance to docetaxel by regulating survivin protein expression in ovarian and breast cancer cells. Cancer Lett, 261(1), 108-119.
Yakar, S., Liu, J. L., Stannard, B., Butler, A., Accili, D., Sauer, B., & LeRoith, D. (1999). Normal growth and development in the absence of hepatic insulin-like growth factor I. Proc Natl Acad Sci U S A, 96(13), 7324-7329.
Bibliography 250
Yan, X., Forbes, B. E., McNeil, K. A., Baxter, R. C., & Firth, S. M. (2004). Role of N- and C-terminal residues of insulin-like growth factor (IGF)-binding protein-3 in regulating IGF complex formation and receptor activation. J Biol Chem, 279(51), 53232-53240.
Yanochko, G. M., & Eckhart, W. (2006). Type I insulin-like growth factor receptor over-expression induces proliferation and anti-apoptotic signalling in a three- dimensional culture model of breast epithelial cells. Breast Cancer Res, 8(2), R18.
Yebra, M., Filardo, E. J., Bayna, E. M., Kawahara, E., Becker, J. C., & Cheresh, D. A. (1995). Induction of carcinoma cell migration on vitronectin by NF-kappa B- dependent gene expression. Mol Biol Cell, 6(7), 841-850.
Yee, D., Cullen, K. J., Paik, S., Perdue, J. F., Hampton, B., Schwartz, A., Lippman, M. E., & Rosen, N. (1988). Insulin-like growth factor II mRNA expression in human breast cancer. Cancer Res, 48(23), 6691-6696.
Yerushalmi, R., Gelmon, K. A., Leung, S., Gao, D., Cheang, M., Pollak, M., Turashvili, G., Gilks, B. C., & Kennecke, H. (2012). Insulin-like growth factor receptor (IGF-1R) in breast cancer subtypes. Breast Cancer Res Treat, 132(1), 131-142.
Yu, Q., & Stamenkovic, I. (2000). Cell surface-localized matrix metalloproteinase-9 proteolytically activates TGF-beta and promotes tumor invasion and angiogenesis. Genes Dev, 14(2), 163-176.
Zannetti, A., Del Vecchio, S., Carriero, M. V., Fonti, R., Franco, P., Botti, G., D'Aiuto, G., Stoppelli, M. P., & Salvatore, M. (2000). Coordinate up-regulation of Sp1 DNA-binding activity and urokinase receptor expression in breast carcinoma. Cancer Res, 60(6), 1546-1551.
Zhao, F., Li, L., Guan, L., Yang, H., Wu, C., & Liu, Y. (2014). Roles for GP IIb/IIIa and alphavbeta3 integrins in MDA-MB-231 cell invasion and shear flow-induced cancer cell mechanotransduction. Cancer Lett, 344(1), 62-73.
Zheng, B., & Clemmons, D. R. (1998). Blocking ligand occupancy of the alphaVbeta3 integrin inhibits insulin-like growth factor I signalling in vascular smooth muscle cells. Proc Natl Acad Sci U S A, 95(19), 11217-11222.
Zhu, J., & Clark, R. A. (2014). Fibronectin at select sites binds multiple growth factors and enhances their activity: expansion of the collaborative ECM-GF paradigm. J Invest Dermatol, 134(4), 895-901.
Zhu, W., Xu, W., Jiang, R., Qian, H., Chen, M., Hu, J., Cao, W., Han, C., & Chen, Y. (2006). Mesenchymal stem cells derived from bone marrow favor tumor cell growth in vivo. Exp Mol Pathol, 80(3), 267-274.
Zuo, G. W., Kohls, C. D., He, B. C., Chen, L., Zhang, W., Shi, Q., Zhang, B. Q., Kang, Q., Luo, J., Luo, X., Wagner, E. R., Kim, S. H., Restegar, F., Haydon, R. C.,
Bibliography 251
Deng, Z. L., Luu, H. H., He, T. C., & Luo, Q. (2010). The CCN proteins: important signalling mediators in stem cell differentiation and tumorigenesis. Histol Histopathol, 25(6), 795-806.
Bibliography 252