Analysis of the Role of Crim1 in Heart Development Swati Iyer B.Sc, M.Sc
A thesis submitted for the degree of Doctor of Philosophy at The University of Queensland in 2016 School of Biomedical Sciences
i Abstract
To overcome the societal and financial burden of cardiovascular disease, a thorough understanding of the molecular and cellular programmes governing cardiac and coronary vascular development is vital. Cre-mediated genetic tracing in various models has provided valuable insight into cell-lineage contribution and inductive cues required for cardiac morphogenesis. Using mouse models, we have identified the transmembrane protein Cysteine-Rich Transmembrane BMP Regulator-1 (Crim1) as an essential protein for numerous aspects of heart development. The presence of six Cysteine-Rich Repeats (CRRs) in Crim1 render it to be a versatile molecule – these have been shown to bind a wide range of growth factors when Crim1 is co-expressed in the same cell as the growth factor (Wilkinson et al., 2003, Wilkinson et al., 2007). The main objective of this thesis is to examine the role that Crim1 plays both temporally and spatially during heart development, and how it functions in tandem with other factors to ensure the progression of this critical process.
Crim1 is strongly expressed in the proepicardium (PE), epicardium, coronary vascular smooth muscle cells and, to a weaker extent, in coronary endothelial cells (Pennisi et al., 2007). Interestingly, Crim1 mutant homozygotes die perinatally on a C57BL6 background (Chiu et al., 2012). Loss of Crim1 function leads to a reduced ventricular size and hypoplastic ventricular compact myocardium, and abnormal epicardial morphology including blebbing and a loss of the regular, cobblestone appearance of the epicardium. Furthermore, epicardium-restricted deletion of Crim1 resulted in increased epicardial epithelial-to-mesenchymal transition (EMT) and invasion of the myocardium, while primary epicardial cells lacking Crim1 displayed an increased migration.
Growth factors like TGFs are known to promote epicardial EMT; however, we observed a paradoxical reduction in epicardial TGF signalling in Crim1 mutant embryos. Interestingly, there appeared to be an accumulation of -catenin, a cell-adhesion molecule present at cadherin- dependent junctional complexes in epithelial cells, at epicardial cell-cell junctions, suggesting an interaction with Crim1 to promote stabilization of these junctions. There was also an increase in the level of phospho-ERK1/2 in the ventricular compact myocardium of mutants, though phospho-AKT levels remained unchanged. This indicates a cell-autonomous role for Crim1 in controlling epicardial migration, EMT and invasion, and a potential paracrine role in regulating myocardial development, likely through regulation of epicardium derived factors.
Another notable defect observed in the absence of Crim1 is that the coronary vasculature also appears to be malformed. We found a reduction in coronary vascular endothelial cell endowment in ii both ventricles, in Crim1Δflox/Δflox homozygotes and in endothelium-restricted mutants, identifying a cell-autonomous role for Crim1. To further study Crim1 function in vitro, lentiviral small hairpin ribonucleic acid (shRNA) targeting the CRIM1 gene was transduced into Human Cardiac Microvascular Endothelial Cells (HCMVECs), with a knockdown of more than 70% achieved at the transcript level. Trypan Blue viability assays in both HCMVECs and Human Umbilical Vein Endothelial Cells (HUVECs) indicate reduced viability in the absence of CRIM1. Protein lysates obtained from both Crim1-knockdown HCMVECs and HUVECs indicate a reduction in phospho- SMAD1/5 levels, but no change in phospho-ERK1/2 and phospho-AKT levels.
We also sought to understand the role of Crim1 in growth factor modulation in endothelial cells in vivo. There was a decrease in phospho-SMAD1/5 activity in endothelial cells from ventricles of Crim1Δflox/Δflox homozygotes, suggesting a dysregulation of TGFβ/BMP signalling in these cells, confirming our in vitro data. However, there was an increase in endothelial phospho-SMAD1/5 levels in ventricles of Crim1Flox/Flox; Tie2-Cre hearts, indicating a possible paracrine role for the other resident cells of the ventricular myocardium that retain Crim1 function. cDNA synthesised from HCMVECs that had lost Crim1 was then used in a human endothelial cell biology array. This revealed an upregulation of genes involved in increased inflammatory response and cell death, and, an increased probability of disease states including vascular lesions, infarction and heart failure, upon Ingenuity Pathway Analysis. Collectively, we illustrate a cell-autonomous role for Crim1 in the development of coronary vascular endothelial cells, in vivo and in vitro.
We further report for the first time, that CRIM1 is able to bind Insulin-like Growth Factors (IGFs). This was shown by means of co-immunoprecipitation experiments, and reveals that the IGFBP domain of CRIM1, and CRRs 3-6 are responsible for this binding. Moreover, in HCMVECs lacking CRIM1, phosphorylation of the IGF-1 receptor was decreased, indicating the ability of CRIM1 to regulate IGF signalling.
Our findings thus provide a platform for understanding how Crim1 functions in both the early and late stages of embryonic heart development in vivo. Many of these findings are mimicked in vitro, most significantly in the context of growth factor modulation in human cardiac microvascular endothelial cells, and aids in understanding the pathogenesis of disease states. This indicates translational relevance to humans, making the study of CRIM1 an exciting new avenue for future research in the field of cardiac development.
iii Declaration by author
This thesis is composed of my original work, and contains no material previously published or written by another person except where due reference has been made in the text. I have clearly stated the contribution by others to jointly-authored works that I have included in my thesis.
I have clearly stated the contribution of others to my thesis as a whole, including statistical assistance, survey design, data analysis, significant technical procedures, professional editorial advice, and any other original research work used or reported in my thesis. The content of my thesis is the result of work I have carried out since the commencement of my research higher degree candidature and does not include a substantial part of work that has been submitted to qualify for the award of any other degree or diploma in any university or other tertiary institution. I have clearly stated which parts of my thesis, if any, have been submitted to qualify for another award.
I acknowledge that an electronic copy of my thesis must be lodged with the University Library and, subject to the policy and procedures of The University of Queensland, the thesis be made available for research and study in accordance with the Copyright Act 1968 unless a period of embargo has been approved by the Dean of the Graduate School.
I acknowledge that copyright of all material contained in my thesis resides with the copyright holder(s) of that material. Where appropriate I have obtained copyright permission from the copyright holder to reproduce material in this thesis.
iv Publications during candidature
1. Crim1 has cell-autonomous and paracrine roles during embryonic heart development. Iyer S, Chou FY, Wang R, Chiu HS, Raju VK, Little MH, Thomas WG, Piper M, Pennisi DJ. Sci Rep. 2016 Jan 29; 6:19832.
2. Crim1 – a regulator of developmental organogenesis. Iyer S, Pennisi DJ, Piper M. Histol Histopathol. 2016 Apr 5:11766. Review.
3. CRIM1 is necessary for coronary vascular endothelial cell development and homeostasis. Iyer S, Chhabra Y, Harvey TJ, Wang R, Chiu HS, Smith AG, Thomas WG, Pennisi DJ, Piper M. Journal of Molecular Histology. 2016 Nov 1. Epub ahead of print.
4. Usp9x-deficiency disrupts the morphological development of the postnatal hippocampal dentate gyrus. Oishi S, Premarathne S, Harvey TJ, Iyer S, Dixon C, Alexander S, Burne TH, Wood SA, Piper M. Sci Rep. 2016 May 16; 6:25783.
Conferences attended
Oral presentations – Role of Crim1 in Epicardial EMT and Coronary Vasculature Development. Iyer S., Chiu H.S., Wang R., Chou F.Y., Chand R., Porter A. and Pennisi D.J. 5th International Postgraduate Symposium in Biomedical Sciences, Brisbane, Australia 2014 (Oral Presentation Award)
Analysis of the role of Crim1 in heart development. Iyer S., Chiu H.S., Wang R. and Pennisi D.J. Molecular Pathways of Diseases Seminar, SBMS, Brisbane, Australia, 2014.
v Poster presentations - Cell-autonomous and paracrine functions of Crim1 in epicardial EMT and coronary vascular development. Iyer S., Chiu H.S., Wang R., Porter A., Chand R., Chou F.Y. and Pennisi D.J. Brisbane Cell and Developmental Biology Meeting, Brisbane, Australia, 2014 (Best Poster Presentation Award)
Cell-autonomous and paracrine functions of Crim1 in epicardial EMT and development of the coronary vasculature Iyer S., Chiu, H.S., Wang, R., Porter A., Chand, R., Chou, F.Y. and Pennisi, D.J. ComBio 2014, Canberra, Australia, 2014.
The Role of Crim1 in Coronary Vasculature Development. Iyer S., Harvey T.J, Piper M., Pennisi D.J. 4th Meeting of the Australian Network of Cardiac and Vascular Developmental Biologists, Adelaide, Australia, 2015
Crim1 regulates EMT and myocardial development in a cell-autonomous and paracrine manner. Iyer S., Chiu H.S., Wang R., Chou F.Y., Harvey T.J., Piper M. and Pennisi D.J. Brisbane Cell and Developmental Biology Meeting, Brisbane, Australia, 2015.
Crim1 regulates EMT and myocardial development in a cell-autonomous and paracrine manner. Iyer S., Chou F.Y., Chiu H.S., Wang R., Piper M. and Pennisi D.J. 6th International Postgraduate Symposium in Biomedical Sciences, Brisbane, Australia 2015
vi Publications included in this thesis
1. Crim1 has cell-autonomous and paracrine roles during embryonic heart development. Iyer S, Chou FY, Wang R, Chiu HS, Raju VK, Little MH, Thomas WG, Piper M, Pennisi DJ. Scientific Reports. 2016 Jan 29; 6:19832.
Contributor Statement of contribution Swati Iyer (Candidate) Wrote the paper (80%) Designed experiments (50%) Performed experiments (50%) Edited the paper (25%) Dr. David Pennisi Wrote the paper (20%) Designed experiments (30%) Performed experiments (15%) Edited the paper (35%) Dr. Michael Piper Edited the paper (25%) Prof. Melissa Little Edited the paper (10%) Prof. Walter Thomas Edited the paper (5%)
Fang Yu Chou Performed experiments (15%) Han Chiu Performed experiments (10%) Richard Wang Performed experiments (10%)
2. Crim1 – a regulator of developmental organogenesis. Iyer S, Pennisi DJ, Piper M. Histology and Histopathology. 2016 Apr 5:11766. Review.
Contributor Statement of contribution Swati Iyer (Candidate) Wrote the paper (100%) Dr. Michael Piper Edited the paper (70%) Dr. David Pennisi Edited the paper (30%)
vii 3. CRIM1 is necessary for coronary vascular endothelial cell development and homeostasis. Iyer S, Chhabra Y, Harvey TJ, Wang R, Chiu HS, Smith AG, Thomas WG, Pennisi DJ, Piper M. Journal of Molecular Histology. 2016 Nov 1. Epub ahead of print.
Contributor Statement of contribution Swati Iyer (Candidate) Wrote the paper (80%) Designed experiments (60%) Performed experiments (80%) Edited the paper (32%)
Dr. Michael Piper Wrote the paper (20%) Edited the paper (30%) Dr. David Pennisi Designed experiments (20%) Edited the paper (20%) Dr. Yash Chhabra Designed experiments (10%) Performed experiments (10%) Edited the paper (2%) Dr. Tracey Harvey Designed experiments (10%) Performed experiments (10%) Edited the paper (5%) Richard Wang Edited the paper (2%) Han Chiu Edited the paper (2%) Aaron Smith Edited the paper (5%) Prof. Walter Thomas Edited the paper (2%)
viii Contributions by others to the thesis
Figures 3.2, 3.4, 3.5H, 3.7, 3.10: complete experiments were performed by Dr. David Pennisi Figures 3.5A-G, 3.6: complete experiments were performed by Han Chiu and Dr David Pennisi Figures 3.8, 3.9, 3.25, 3.27, and handling of WTI1-CreERT2 mice, sectioning and staining in figure 3.13: experiments were performed by Grace Chou (Honours student under supervision of Dr. David Pennisi) Figures 3.21, 3.22: complete experiments were performed by Richard Wang (Honours student under supervision of Dr. David Pennisi) Figure 4.10 B, C: quantification of viable cell number was done blinded by Dr. Yash Chhabra (B) and Dr. Tracey Harvey (C). Figures 4.14, 4.15, Table 4.2: complete experiment performed by Dr. Yash Chhabra. Western blotting was done in association with Dr. Yash Chhabra. Densitometry analyses were done blinded by Dr. Yash Chhabra. Human Endothelial Cell Biology qPCR array was performed under supervision of Dr. Tracey Harvey. Primers for generation of CRIM1 deletion constructs using InFusion cloning were designed by Dr. Aaron Smith
Statement of parts of the thesis submitted to qualify for the award of another degree None
ix Acknowledgements
I would like to acknowledge the people who have helped me both professionally and on a personal level, throughout the duration of my candidature.
Dr. David Pennisi, for introducing me to the world of heart development, your guidance with all aspects of my project, and for being a very approachable and encouraging supervisor. You have patiently provided me with the ability to think critically while teaching me the nuances of experimental techniques involving in vivo mouse work.
A/Prof. Michael Piper, for your generosity, motivation, and for just being the best supervisor any student could ask for. You provided an environment where I could grow as an individual while developing scientific skills, allowing me to come up with my own ideas and carry them through to fruition. I could not have completed this thesis without your support. Your focus, work ethic and commitment to Science are an inspiration for anyone wanting to excel in the field.
Prof. Wally Thomas, for being supportive and generous throughout my candidature.
Dr. Tracey Harvey, for your guidance and for being the beautiful person you are. You’ve been so amazing and approachable, and just talking to you, whether about Science or otherwise, has made life on level 7 wonderful in the past couple of years.
Dr. Yash Chhabra, for being the Western Blotting whiz that you are, and more importantly, for being such a great friend. You’ve helped me on so many occasions, and made it so much easier to finish experiments with your advice and expertise. Our desi connection has made our friendship even more wonderful!
Dr. Aaron Smith for always being willing to help, and being so generous with both your time and reagents.
Dr. Oressia Zalucki, for helpful discussions and genotyping my mice for me while I’ve been away.
A special thanks to my lab members and colleagues on level 7 – Diana Vidovic, Lachlan Harris, Kelvin Yin, Mitch Fane, James Fraser, Sabrina Oishi and Elise Horne. Whether you have helped me experimentally while in the lab, had troubleshooting discussions or just friendly chats during the
x day (or over lunch!), your friendship has meant so much to me over the past few years. Working with you and spending time with you has made this journey so much more enjoyable and fulfilling.
Sarah Piper, for all the last minute ordering, and for being so open and kind with equipment and reagent use.
Yvonne Yeap, for being a lovely person and so generous.
Past members of the Pennisi Lab, Han Chiu and Richard Wang for all your experimental help and otherwise, advice, positivity and friendship, and Grace Chou for performing useful experiments that have helped with publications.
I would like to thank my family – grandparents, aunts, uncles, cousins – for being the best in the world. I have no words to describe how much all of you mean to me. I would also like to thank my in-laws for their support, and the pick-ups and drop-offs in my first year at UQ.
I would like to thank my sister, for being the way she is, providing the perfect balance of both motivation and comic relief whenever required, inconspicuously making this journey so much easier (and also for being the most adorable little sister anyone could ask for).
I would like to thank my husband, for always knowing I could finish this, but for whom it didn’t matter if I did not. For putting up with my moods, cheering me up, encouraging me, proofreading documents and most importantly, for a whole lot of love. I couldn’t have done this without you.
Lastly, I would like to thank my parents, for your unconditional love and support, even though you have been miles away for the past few years. For everything you have done for me, and everything I know you will continue to do for me. For making me who I am today - everything I am, is because of you, everything I aspire to be, is because of you. I hope I have done you proud.
xi Keywords Heart development, epicardium, myocardium, coronary vasculature, growth factors, signal transduction, CRIM1.
Australian and New Zealand Standard Research Classifications (ANZSRC) 060111 Signal Transduction, 25% 060103 Cell Development, Proliferation and Death, 40% 060603 Animal Physiology – Systems, 35%
Fields of Research (FoR) Classifications FoR code: 0601, Biochemistry and Cell Biology, 60% FoR code: 0606, Physiology, 20% FoR code: 0699, Other Biological Sciences, 20%
xii Table of contents
Abstract...... ii Declaration by author...... iv Publications during candidature...... v Publications included in this thesis...... vii Contribution by others to the thesis...... ix Acknowledgements...... x Keywords...... xii Table of contents...... xiii List of figures and tables...... xviii List of abbreviations...... xxi Chapter 1. Introduction...... 2 Heart development – an overview………………...... ……..…………………………..…...2 Cellular contributions to heart structure…...... …………………………………...3 Growth factors involved in heart development….…………………………………….……..7 TGFβ…………………………………….…………………………………...…..…..8 BMP………………………………………………………………………………....10 FGF……………………………………………………………………………….…12 VEGF………………………………………………………………………….…….13 PDGF………………………………………………………………………….…….14 IGF…………………………………………………………………………….…….14 IGFBP………………………………………………………………………….……15 Wnt…………………………………………………………………………….……16 Cooperative interaction between growth factors directs development of the epicardium, myocardium, and coronary vasculature……………………………...... 17 CRIM1……………………………………………………………………………………....18 The role of Crim1 in organogenesis………………………………………………………...20 Kidney………………………………………………………………………………20 Placenta…………………………………………………………………….…….…21 Lens and retinal vasculature………………………………………………...……....21 Nervous system……………………………………………………………………..22 Crim1 in heart development?...... 22 Overall aims...... 24 Chapter 2. Materials and methods………………...………………………………………....…..25
xiii Ethics Statement………………..…………………………………………………….….….26 Mouse Lines……..……………..……………………………………………………..…….26 Animal Breeding...……………..……………………………………………………..…….26 Sample collection..……………..……………………………………………………..…….26 Genotyping……………..…………………………………………….………………….….27 Sample preparation..…………………………………..…………….………………...….…29 Immunohistochemistry.……………………………………….……………………….....…30 Immunofluorescence….……………………………………….…………………….…...…30 Antibodies……………...…………………………………………….…………………..…31 Bacterial transformation.…………………………………………………….…………...... 33 Mini and Midi preps for plasmid DNA purification..……………….…………………...... 33 DNA Sequencing.…………………………………………………………….…………...... 34 Cell lines.…………………………………………………………………….….…………..34 Immunoblotting..………………………………………………………….…….…….…….34 Protein Extraction…………………………………………………………..….……34 Sodium Dodecyl Sulphate Polyacrylamide Gel Electrophoresis (SDS- PAGE)...... 34 Electroblotting and Probing of Membranes…………………………………...... 35 Stripping of Polyvinylidene fluoride (PVDF) Membranes……………………...... 35 Data documentation and quantification..…………………………………………….…...... 35 Statistical analyses…………………………………………………………………….….....35 Chapter 3. Examining the cell autonomous and paracrine roles of Crim1 in the epicardium and compact myocardium...... 37 Author contributions...... 38 Introduction...... 39 Materials and Methods...... 40 Ethics Statement...... 40 Animal Breeding...... 40 Tamoxifen Administration...... 41 Sample preparation...... 41 Scanning Electron Microscopy...... 41 Antibodies and Lectin...... 41 Immunohistochemistry...... 42 Immunofluorescence...... 42 Primary Epicardial Explant Cultures and Migration Assays...... 42
xiv Cell culture and transfection………………………………………………………………...43 Protein Fragment Complementation Assay…………………………………………………43 Reverse Transcription, Quantitative Real-time PCR and Analysis………………………....44 Data documentation and quantification………………………….………………………….46 Statistical Analyses………………………………………………………………………….47 Results...... 47 Crim1 expression during heart development...... 47 Defective heart development in embryos lacking Crim1...... 50 Crim1 is dispensable for PE formation...... 52 Crim1 regulates compact myocardial development...... 53 Effect of Crim1 deletion on cardiomyocytes within the myocardium...... 54 Effect of epicardium-specific deletion of Crim1 on the compact myocardium...... 55 Cell-autonomous role of Crim1 in epicardial migration in vitro...... 56 Cell-autonomous role of Crim1 in epicardial migration and myocardial invasion in vivo....58 Crim1 may alter epicardial adhesion junctions...... 61 CRIM1 can bind to other CRIM1 molecules, and further confer junctional stability...... 62 Modulation of growth factor activity by Crim1...... 67 Loss of Crim1 results in decreased proliferation of intramyocardial cells...... 70 Crim1 may control fate specification of EPDCs, particularly into the fibroblast lineage...... 71 Discussion...... 73 Acknowledgements...... 77 Chapter 4. Examining the cell-autonomous role of Crim1 in the coronary vasculature, specifically in endothelial cells...... 78 Author contributions...... 79 Introduction...... 80 Materials and methods...... 83 Ethics Statement...... 83 Animal Breeding...... 83 Sample preparation...... 83 Antibodies and lectin...... 83 Cell Culture...... 84 Trypan Blue Viability Assay...... 84 Quantitative RT-PCR and q-PCR Array...... 84 Plasmid Constructs...... 85 CRIM1 overexpression...... 85
xv qPCR primers for analysis of effect of CRIM1 overexpression...... 86 Data documentation, quantification and analyses...... 87 Results...... 88 Crim1 deletion results in abnormal coronary vasculature development...... 88 Cell autonomous role for Crim1 in endothelial cell development...... 90 Growth factor signalling is perturbed in the absence of Crim1...... 96 Examination of CRIM1 function, in vitro...... 98 Dysregulation of canonical BMP signalling in the absence of CRIM1, in vitro...... 99 Misregulation of genes involved in endothelial stress response, apoptosis and cell adhesion in HCMVECs lacking CRIM1...... 100 Effect of CRIM1 overexpression on endothelial genes...... 106 Cell-autonomous role for Crim1 in regulating BMP signalling, in vivo...... 107 Increase in phospho-SMAD1/5 signalling may not directly be due to paracrine stimulation by BMP2/BMP4...... 109 Discussion...... 111 Acknowledgements...... 115 Chapter 5. Investigating whether the IGFBP domain of CRIM1 can serve to bind IGFs and modulate downstream signalling...... 116 Author Contributions...... 117 Introduction...... 118 Materials and methods...... 119 Cell Culture...... 119 Antibodies...... 119 Generation of deletion constructs...... 120 Co-Immunoprecipitation...... 121 Western Blotting...... 121 Data Documentation, Quantification and Analysis...... 121 Results...... 122 CRIM1 can bind IGF-1...... 122 Generation of CRIM1 deletion constructs...... 123 CRIM1 binds IGF-1 via its IGFBP-like domain, and CRRs 3-6...... 125 Interaction of CRIM1 with IGFs, and modulation of IGF signalling...... 126 Effect of IGF-1 stimulation in the absence of CRIM1...... 127 Discussion...... 129 Acknowledgements...... 131
xvi Chapter 6. General discussion and future directions...... 132 Crim1 in epicardial development...... 133 Crim1 in myocardial development...... 134 Crim1 in endothelial cell development and function...... 136 CRIM1 and IGF interactions in endothelial cells………………………………………………….137 Putative role of CRIM1 in other diseases...... 138 References...... 140 Appendix...... 163 Published research paper 1...... 164 Published research paper 2...... 203 Published review article 1...... 223
xvii List of figures and tables
Figure 1.1: Origin and lineage relationships of cardiac cell types Figure 1.2: Formation of the epicardium from the proepicardium Figure 1.3: Schematic representation of epithelial-to-mesenchymal transformation of epicardial cells Figure 1.4: Schematic representation of growth factor cross-talk during epicardial and myocardial development Figure 1.5: Schematic representation of the structure of Crim1 Figure 1.6: Schematic representation of the potential functions of Crim1 Figure 2.1: Schematic representation of Crim1 alleles detected by genotyping Figure 3.1: Schematic representation of protein complementation assay Figure 3.2: Crim1 expression during various stages of development Figure 3.3: Confirmation of Crim1 sites of expression Figure 3.4: Heart defects in Crim1∆flox/∆flox hearts Figure 3.5: Epicardial defects in Crim1∆flox/∆flox hearts Figure 3.6: Crim1 is dispensible for specification or early development of the proepicardium Figure 3.7: Crim1 is necessary for normal compact myocardial development Figure 3.8: Myocardial deletion of Crim1 does not affect cardiomyocyte proliferation and survival and thickness of compact myocardium Figure 3.9: Fidelity of the WT1-CreERT2 line Figure 3.10: Crim1 plays a cell-autonomous role in migration of epicardial cells, in vitro Figure 3.11: Fidelity of the WT1-Cre line Figure 3.10: Crim1 plays a cell-autonomous role in epicardial migration, in vitro Figure 3.12: Crim1 plays a cell-autonomous role in epicardial migration and myocardial invasion, in vivo, as observed using the WT1-Cre line Figure 3.13: Crim1 plays a cell-autonomous role in epicardial migration and myocardial invasion, in vivo, as observed using the WT1-CreERT2 line Figure 3.14: Changes in β-catenin accumulation at epicardial cell junctions in the absence of Crim1 Figure 3.15: No visible change in f-actin morphology in the epicardium or myocardium of Crim1∆flox/∆flox hearts at 13.5 dpc Figure 3.16: Protein complementation assay using AT1R-V1 and β-Arrestin-V2 Figure 3.17: Protein complementation assay using CRIM1-V1 and CRIM1-V2 Figure 3.18: Cellular localization of CRIM1 Figure 3.19: Magnified images showing CRIM1 localization
xviii Figure 3.20: Schematic representation of Crim1 stabilising epicardial cell junctions Figure 3.21: Reduction in canonical TGFβ signalling in the absence of Crim1 Figure 3.22: No change in canonical BMP signalling in the absence of Crim1 Figure 3.23: No change in pAKT levels in the absence of Crim1 Figure 3.24: Increase in myocardial pERK1/2 in the absence of Crim1 Figure 3.25: Decrease in the number of proliferating EPDCs in the absence of epicardial Crim1 Figure 3.26: Assessing expression levels of different cell lineage markers in the absence of Crim1 Figure 3.27: Decrease in number of myocardial periostin-positive EPDCs in the absence of epicardial Crim1 Figure 4.1: Loss of Crim1 results in malformed coronary vasculature at 17.5 dpc Figure 4.2: Reduced number of coronary vascular endothelial cells in Crim1∆flox/∆flox hearts Figure 4.3: Efficiency of the Tie2-Cre line Figure 4.4: Reduced number of coronary vascular endothelial cells in Crim1FLOX/FLOX; R26R; Tie2-Cre hearts Figure 4.5: Limited apoptosis of endothelial cells in Crim1FLOX/FLOX; R26R; Tie2-Cre hearts as assessed using CC3 Figure 4.6: No change in proliferation in Crim1FLOX/FLOX; R26R; Tie2-Cre hearts as assessed using PHH3 Figure 4.7: No change in proliferation of endothelial cells in Crim1FLOX/FLOX; R26R; Tie2-Cre hearts as assessed using Ki67 Figure 4.8: No change in the apoptosis and proliferation of non-endothelial cells in Crim1FLOX/FLOX; R26R; Tie2-Cre hearts as assessed with CC3 and PHH3 Figure 4.9: Decrease in canonical BMP signalling in cells of Crim1∆flox/∆flox hearts as assessed with phospho-SMAD1/5 Figure 4.10: CRIM1 knockdown in HCMVECs and its effect on cell viability Figure 4.11: CRIM1 knockdown in HCMVECs results in decreased phospho-SMAD1/5 activity Figure 4.12: Effect of CRIM1 knockdown on endothelial gene expression as assessed by a human endothelial cell biology array Figure 4.13: Ingenuity Pathway Analysis of misregulated endothelial genes in CRIM1 knockdown HCMVECs Figure 4.14: CRIM1 overexpression in HCMVECs Figure 4.15: Effect of CRIM1 overexpression in HCMVECs compared to CRIM1 knockdown HCMVECs Figure 4.16: Increase in canonical BMP signalling in cells of Crim1FLOX/FLOX; R26R; Tie2-Cre hearts as assessed with phospho-SMAD1/5
xix Figure 4.17: No change in canonical TGFβ signalling in endothelial cells of Crim1FLOX/FLOX; R26R; Tie2-Cre hearts as assessed with phospho-SMAD2 Figure 4.18: Addition of exogenous BMP-2 and BMP-4 does not exert a paracrine effect on CRIM1 knockdown HUVECs Figure 4.19: Schematic representation of endothelial proinflammatory activation Figure 5.1: COIP revealing that CRIM1 can bind IGF-1 Ea and IGF-1 Eb constructs Figure 5.2: Schematic representation of CRIM1 deletion constructs Figure 5.3: Generation of CRIM1 deletion constructs Figure 5.4A: Binding of IGF-1 Eb to CRIM1 deletion constructs Figure 5.4B: CRIM1 binds IGF-1 Eb via the IGFBP domain and CRRs 3-6 Figure 5.5: CRIM1 modulates IGF-1 signalling in HCMVECs Figure 5.6: Change in downstream IGF signalling upon treatment of CRIM1 knockdown HUVECs with exogenous IGF-1
xx List of abbreviations
Ab Antibody AGRF Australian Genome Research Facility AHF Anterior heart field AV Atrio-ventricular BAMBI BMP and Activin membrane bound inhibitor BCA Bicinchoninic Acid Assay BMP Bone Morphogenetic Protein BMPR BMP Receptor BSA Bovine Serum Albumin CC3 Cleaved Caspase 3 Crim1 Cysteine-Rich Transmembrane BMP Regulator-1 CNCC Cardiac Neural Crest Cells
CO2 Carbon dioxide CRR Cysteine-rich repeat CV-2 Crossveinless-2 DAB 3,3′-Diaminobenzidine DAPI 4',6-Diamidino-2-phenylindole DMEM Dulbecco’s Modified Eagle Medium DNA Deoxyribonucleic acid dNTP Deoxynucleotide triphosphate EDTA Ethylenediaminetetraacetic acid EMT Epithelial-to-Mesenchymal Transition EndoMT Endothelial-to-Mesenchymal Transition EPDCs Epicardium-Derived Cell ERK Extracellular signal-Regulated Kinase FAK Focal Adhesion Kinase FGF Fibroblast Growth Factors FGFR FGF Receptor HCl Hydrochloric acid HCMVEC Human Cardiac Microvascular Endothelial Cell HSPG Heparin sulphate proteoglycan HUVEC Human Umbilical Vein Endothelial Cell GFP Green Fluorescent Protein
xxi GDF Growth and differentiation factors IGF Insulin-like Growth Factor IGFBP IGF Binding Protein IGF-1R IGF-1 receptor LEF Lymphoid Enhancing Factor MAPK Mitogen-Activated Protein Kinases MGF Mechano Growth Factor NF1 Neurofibromin 1 NFR Nuclear Fast Red OFT Outflow tract PAGE Polyacrylamide Gel Electrophoresis PBS Phosphate Buffered Saline PBS-Tx PBS-TritonX PCR Polymerase Chain Reaction PDGF Platelet Derived Growth Factor PDGFR PDGF receptor PE Proepicardium PFA Paraformaldehyde PHH3 Phosphohistone H3 PRDC Protein Related to Dan and Cerberus PVDF Polyvinylidene fluoride RGD Arginyl-Glycyl-Aspartic acid RIPA Radioimmunoprecipitation RNA Ribonucleic acid SDS Sodium Dodecyl Sulphate Ser/Thr Serine/Threonine sFRPs Secreted frizzled-releated proteins SHF Secondary heart field ShRNA Short hairpin RNA TBS Tris Buffered Saline TBS-Tx TBS-TritonX TCF T-Cell transcription Factor TEMED N,N,N’-tetra-methylethylenediamine TGFβ Transforming Growth Factor β Tsg Twisted Gastrulation
xxii Wt1 Wilms Tumour 1 VCAM-1 Vascular Cell Adhesion Molecule-1 VEGF Vascular Endothelial Growth Factor VEGFR VEGF receptor VWFC Von Willebrand Factor C
xxiii
For my parents
xxiv
Chapter 1
Literature review
Parts of this chapter are adapted from the published review ‘Crim1 – a regulator of developmental organogenesis’, in the journal Histology and Histopathology.
Crim1 – a regulator of developmental organogenesis. Iyer S, Pennisi DJ, Piper M. Histol Histopathol. 2016 Apr 5:11766. Review.
1 INTRODUCTION
The heart is the first organ to form and function in the vertebrate embryo (Yutzey and Kirby 2002), and provides oxygen and nutrients to the body while also removing wastes. The process of cardiogenesis involves the tightly regulated, synergistic interaction and integration of various cell types and signalling molecules (Miquerol and Kelly 2013, Kelly, Buckingham et al. 2014) at varying stages. Congenital heart malformations are a major cause of foetal mortality and reflect the need for understanding the complexities of heart development. Furthermore, resolving damage to heart tissue due to disease and injury is also a burgeoning societal burden that requires urgent investigation. The limited regenerative potential of the heart has made it essential to uncover mechanisms to repair tissue following myocardial ischemia. Unsurprisingly, there is immense research in the context of heart regeneration, including therapies involving stem cells, genetic reprogramming of the epicardium and cells such as fibroblasts to cardiomyocytes, and tissue engineering, as strategies to restore myocardial function (Laflamme and Murry 2011).
In the following sections, we discuss the development of the heart, the interplay of growth factors involved in cardiogenesis, and introduce Crim1, a protein with the potential to regulate various aspects of developmental processes, including matters of the heart.
Heart development – an overview. Heart development begins with cardiac progenitor cells traversing the primitive streak during gastrulation (Garcia-Martinez and Schoenwolf 1993, Schoenwolf and Garcia-Martinez 1995). Angioblastic cords form in the cardiogenic region upon the aggregation of the principal cardiac precursors, splanchnic mesenchymal cells, which arise in the lateral plate mesoderm from primary heart fields (Waldo, Kumiski et al. 2001). Two thin-walled endocardial heart tubes form when the cords canalize, and subsequently fuse to form a single two-layered heart tube (Manasek 1969). This comprises the outer myocardium and the inner endothelial tube - the endocardium - separated by cardiac jelly (Waldo, Kumiski et al. 2001). The heart tube undergoes rightward looping and increases in length; cardiac progenitor cells that contribute the heart tube are mainly responsible for its elongation(Miquerol and Kelly 2013). Further investigations of cardiac lineage contributions have revealed the existence of two distinct lineages; a first lineage that gives rise to the left ventricle and atria, and a second lineage that contributes to the OFT myocardium, right ventricle and atria (Meilhac, Esner et al. 2004) The heart begins to grow due to cellular proliferation, and following cardiac chamber and cushion morphogenesis, cardiac septation and valve development ensue (Miquerol and Kelly 2013). The epicardium of the heart, which has
2 formed by the migration of propepicardial cells as a single sheet over the surface of the heart, via epithelial to mesenchymal transition (EMT) contributes various cell types to the myocardium, including coronary vascular smooth muscle cells, cardiac fibroblasts and a proportion of coronary vascular endothelial cells (Virágh and Challice 1981, Männer 1993, Virágh, Gittenberger-de Groot et al. 1993, Mikawa and Gourdie 1996, Gittenberger-de Groot, Vrancken Peeters et al. 1998). The myocardium is divided into two distinct components – the outer compact myocardium and the inner trabeculated myocardium. This is followed by development of the coronary vasculature, which supports the growing ventricular wall by providing blood supply (Miquerol and Kelly 2013) (Figure 1.1).
Figure 1.1: Origin and lineage relationship of cardiac cell types. A, Contribution of the three populations of embryonic heart progenitors, cardiogenic mesoderm (red), cardiac neural crest (purple) and proepicardial organ (yellow) to different heart compartments during cardiac morphogenesis in the mouse. Progenitors of the cardiogenic mesoderm are first recognizable under the head folds (HFs) of the embryo at E7.5, then move ventrally to the midline (ML) and form initially the linear heart tube and ultimately the four chambers of the heart. After the looping of the heart tube (E8.5), cardiac neural crest progenitors migrate from the dorsal neural tube to engulf the aortic arch arteries and contribute to vascular smooth muscle cells of the outflow tract (OFT) around E10.5. At the same time in mouse development, the proepicardial organ precursors contact the surface of the developing heart, give rise to the epicardial mantle (yellow) and contribute later to the coronary vasculature. In the fetal heart (∼E14), the chambers separate due to septation and are connected to the pulmonary trunk (PT) and aorta (Ao). Cranial (Cr)-caudal (Ca), right (R)-left (L), and dorsal (D)-ventral (V) axes are indicated. AA, aortic arch; IVS, interventricular septum; LA, left atrium; LV, left ventricle; PhA, pharyngeal arches; PLA, primitive left atrium; PRA, primitive right atrium; RA, right atrium; RV, right ventricle. Image adapted from (Laugwitz et al., 2008).
Cellular contributions to heart structure Cells that arise from the different heart fields as well as those from the cardic neural crest and proepicardium, all contribute to the vertebrate heart development (Torrent-Guasp, Kocica et al. 2005). Primary or first heart field (FHF) progenitors give rise to the nascent heart tube and form the left ventricle and atria (Stennard and Harvey 2005). The second heart field comprises progenitor cells of the arterial and venous poles of the heart tube (Dyer and Kirby 2009). The pharyngeal
3 mesoderm gives rise to part of the myocardium that forms the cardiac outflow tract (OFT) (Mjaatvedt, Nakaoka et al. 2001, Waldo, Kumiski et al. 2001), and part of the right ventricle (Zhou, Cashman et al. 2011), after the primary heart tube forms (de la Cruz, Sanchez Gomez et al. 1977) and as it elongates (Zaffran, Kelly et al. 2004). This is called the anterior heart field (AHF) region (Kelly, Brown et al. 2001, Mjaatvedt, Nakaoka et al. 2001). It is distinct from the secondary heart field (SHF) (Dyer and Kirby 2009) which generates the distal outflow tract myocardium and proximal aortic and pulmonary trunk smooth muscle (Waldo, Kumiski et al. 2001, Waldo, Hutson et al. 2005, Ward, Stadt et al. 2005, Sun, Liang et al. 2007).
Endocardial endothelial cells undergo Endothelial-to-Mesenchymal Transformation (EndoMT) and invade the cardiac jelly to form the atrioventricular (AV) canal endocardial cushions (Akiyama, Chaboissier et al. 2004). Cardiac neural crest cells (CNCCs) supply vascular smooth muscle cells in addition to the SHF–derived smooth muscle progenitors and play a role in cardiovascular patterning (Kirby and Waldo 1995, Waldo, Hutson et al. 2005). CNCCs are involved in OFT endocardial cushion formation (Akiyama Haruhiko et al, 2004). These later contribute to the formation of the valves and septa of the four-chambered heart (Markwald, Runyan et al. 1984, Akiyama, Chaboissier et al. 2004). Ecto-mesenchymal cardiac neural crest cells are involved in pharangeal arch patterning and form tissues that partition the vessels of the heart (Kirby and Waldo 1995). They are responsible for aorticopulmonary septation and division of the truncus arteriosus and bulbus cordis along with outflow tract endocardial cushion formation, aortic arch remodelling as well as contribution of smooth muscle to the arch arteries, aorta and pulmonary trunk as they mature (Kirby and Waldo 1990, Waldo, Miyagawa-Tomita et al. 1998, Li, Liu et al. 1999, Maschhoff and Baldwin 2000, Stoller and Epstein 2005, Brown and Baldwin 2006).
The PE is a transient structure located between the sinus horns and liver primordium (Virágh, Gittenberger-de Groot et al. 1993), and also derived from the lateral plate mesoderm (Serluca 2008). By attaching to the inner groove of the AV junction of the rudimentary heart (Männer , Ishii, Garriock et al. 2010), proepicardial cells form its outer layer, the epicardium (Virágh and Challice 1981, Männer 1993, Virágh, Gittenberger-de Groot et al. 1993, Mikawa and Gourdie 1996, Gittenberger-de Groot, Vrancken Peeters et al. 1998). The migration of these cells to the myocardium depends on the species, and occurs either by proepicardial vesicle budding or via the formation of an extracellular matrix bridge that possibly guides the cells to their destination (Nahirney, Mikawa et al. 2003) (Figure 1.2). Following the attachment, a population of epicardial cells undergo EMT and form subepicardial cells, whereas another subset of these cells migrate through the subepicardial space and into the myocardium (Virágh and Challice 1981).
4
Figure 1.2: Formation of the epicardium from the proepicardium. A, B, Species-specific migration of proepicardial cells over the myocardium, by vesicles budding off (mouse model, A) or via the formation of an extracellular matrix bridge guiding the proepicardial cells to the myocardium (avian model, B).C-E, Scanning electron micrographs showing the formation of the normal PE-derived tissue bridge to the heart in three-day old chick embryos (Hamburger and Hamilton (HH)- stages 16-18).Brown coloration marks the PE and PE- Derived primitive epicardium. C, at HH-stage 16, the PE appears as a cauliflower-shaped accumulation of mesothelial villous protrusions (*) covering the surface of the right horn of the sinus venosus. D, At HH-stage 17, the tips of some PE villi (*) adhere to the dorsal surface of the developing ventricles (Ven). E, At HH-stage 18, a secondary PE-derived tissue bridge (*) is formed, which bridges the free pericardial cavity between the ventral wall of the sinus venosus and the dorsal wall of the developing ventricles (sinu-ventricular ligament). From the point of its attachment to the heart, the primitive PE-derived epicardium spreads over the naked myocardial surface of the heart. AV, atrioventricular canal; LA, future left atrial appendage; Li, liver; RA, future right atrial appendage. C-E adapted from (Männer, 2013).
These epicardium-derived cells (EPDCs) are destined to form various cell types, including coronary vascular smooth muscle cells, cardiac fibroblasts and a proportion of coronary vascular endothelial cells (Virágh and Challice 1981, Männer 1993, Virágh, Gittenberger-de Groot et al. 1993, Mikawa and Gourdie 1996, Gittenberger-de Groot, Vrancken Peeters et al. 1998) (Figure 1.3). The contribution of the PE to the cardiomyocyte lineage is controversial. Van Wijk and colleagues have suggested that different growth factor cues can modulate the differentiation of proepicardial cells into cardiomyocytes (Kruithof, van Wijk et al. 2006, Schlueter, Männer et al. 2006, van Wijk, van den Berg et al. 2009). Some studies based on lineage tracing suggest that a small number of proepicardial cells are fated to become cardiomyocytes based on the expression of markers such as Tbx18 (Cai, Martin et al. 2008, Zhou, Ma et al. 2008). However, this interpretation has been refuted 5 based on the inherent expression of these markers during development, which skew lineage-tracing analyses (Christoffels, Grieskamp et al. 2009). Similarly, there has also been much speculation about the origins of coronary endothelial cells. It is a matter of contention whether the PE is the only contributor of endothelial cells to the coronary vasculature, as the liver bud and sinus venosus have also been demonstrated to supply these cells (Poelmann, Gittenberger-de Groot et al. 1993, Red-Horse, Ueno et al. 2010, Cossette 2011). Some studies indicate that endothelial cells originate in the PE and others suggest that the PE is used as a channel for endothelial cell migration from the adjacent liver bud and sinus venosus (Cossette and Misra 2011, Pennisi 2016).
Figure 1.3: Epicardial EMT and invasion of the myocardium by EPDCs A subset of epicardial cells undergo EMT and populate the subepicardial space. Another set of cells invades the myocardium; these EPDCs are precursors to cardiac fibroblasts, coronary smooth muscle cells and a small proportion of coronary endothelial cells.
6 The endothelial cells, by vasculogenesis, assemble de novo to form a primary vascular plexus (Wada, Willet et al. 2003). These structures undergo angiogenesis to form arteries, veins and capillaries (Bernanke and Velkey 2002, Ratajska, Ciszek et al. 2003, Wada, Willet et al. 2003). A venous connection arises between the vascular plexus and coronary sinus, and an arterial connection arises with aortic sinus developing coronary ostia which give rise to the right and left coronary arteries (Hutchins, Kessler-Hanna et al. 1988). A study by Tomanek et al provides evidence that a population of hemangioblast precursors that differentiate into erythrocytes are derived from the proepicardium (Tomanek, Ishii et al. 2006). These hemangioblasts form blood islands in the epicardium, which later penetrate the myocardium and coalesce into a primary capillary plexus (Risau 1991). From these pre-existing channels arise new vessels, by angiogenesis. The sinus venosus has also been studied in detail for its contribution to intramyocardial cardiac endothelial cells. Zhang and colleagues have confirmed, by employing the use of multiple transgenic mouse lines, that the sinus venosus is indeed a major contributor to myocardial endothelial cells (Zhang, Pu et al. 2016). Tian and colleagues have suggested that subepicardial cells that originate from the sinus venosus give rise to both arterial and venous endothelial cells (Tian, Hu et al. 2013). Red-Horse and colleagues have demonstrated that blood vessels sprout from the sinus venosus and to a lesser extent, the endocardium. Venous endothelial cells of the sinus venosus dedifferentiate into a coronary plexus and invade the myocardium, and consequently redifferentiate into arteries and capillaries. This implies that some venous cells are reprogrammed into becoming arterial cells while the other, mostly superficial ones, retain venous plasticity (Red- Horse, Ueno et al. 2010). Extensive remodelling of coronary vessels occurs during development, and the regulated mechanisms leading to the integration of the coronary vasculature to the systemic circulation are still incompletely understood (Reese, Mikawa et al. 2002).
Growth factors involved in heart development. The molecular and cellular events that orchestrate cardiac development are relatively well-studied, but not fully understood. This is partly due to multiple cellular populations from different origins, which respond to numerous growth factors at varying times. This, combined with the crosstalk between different signalling pathways that regulate cardiac differentiation and morphogenesis, and the pace at which cardiogenesis occurs, makes the heart a challenging organ to study.
The growth factors that direct cardiac organogenesis, and are intimately involved with heart development, include, but are not restricted to, TGFβ, BMPs, Fibroblast Growth Factors (FGFs), VEGFs, PDGFs, IGFs and members of the Wnt family. These are expressed in various stages of heart development and in different locations. Interaction between these ligands and their receptors
7 is vital for cardiogenesis to occur appropriately; the successful development of conditional knockout models has allowed for the roles of these molecules in target cells to be determined. Here, we summarize our current understanding of how these factors affect different aspects of cardiogenesis (Figure 1.4).
Figure 1.4: Epicardial-myocardial signalling pathways in coronary vascular development. Shown is a schematic of the crosstalk between the epicardium (blue) and myocardium (red) during the early stages of coronary development. An FGF-Shh-VEGF/Ang2-dependent pathway is important for the generation of coronary endothelial progenitors (right), whereas catenin, PDGF, TGFare important for the development of the coronary vascular smooth muscle (left). Image modified from Olivey and Svensson (2010).
Development and homeostasis are both regulated by the evolutionarily conserved TGFβ superfamily of growth factors (Wu and Hill, 2009). Members of this family include TGFβs, BMPs, growth and differentiation factors (GDFs) and Nodal, among others. TGFβ signalling can occur via two primary pathways, the SMAD-dependent or canonical pathway, and the SMAD-independent or non-canonical pathway. In the former, TGFβs signal via type I and II serine/threonine kinase receptors.Binding of the ligand to the type II receptors results in the formation of a heterotetrameric complex which then activates the type I receptors. The activated type I receptor phosphorylates the SMAD proteins which translocate into the nucleus,where they interact with transcription factors to modulate the transcription of target TGF-β-responsive genes (Wu and Hill, 2009). The non- canonical pathway entails activation of mitogen-activated protein kinases (MAPKs), AKT, among
8 others (Wang et al., 2014). Since TGFβs stimulate EMT, resulting in a loss of epithelial cell characteristics and induction of a mesenchymal phenotype, it follows that many of these genes should, in principle, include those involved in regulating cell adhesion, increased motility and expression of mesenchymal markers. Indeed, this has been demonstrated by way of microarray analysis of the genes upregulated and downregulated by TGFβs (Xie et al., 2004).
Some ligand-receptor interactions require TGFβ co-receptors such as the membrane-bound proteoglycan betaglycan, which is also referred to as TGFβ receptor type III, and endoglin. While betaglycan binds TGFβs 1, 2 and 3 with high affinity, endoglin recognizes only TGFβs 1 and 3. Endoglin is predominantly expressed on endothelial cells (Cheifetz et al., 1992). TGFβ-1 and BMP- 9 mediated phosphorylation of SMADs 1, 5 and 8 and consequent inhibition of endothelial cell migration is dependent on the phosphorylation of endoglin by the type 1 TGFβ receptors ALK-5 and ALK-1, as well as TβRII (Ray et al., 2010).
TGFβs 1, 2 and 3 are expressed in various stages of heart development and in different locations including endocardial-endothelial cells, the epicardium, the AV canal myocardium and the OFT (Arthur and Bamforth, 2011). The TGFβ receptors and co-receptors are expressed in overlapping sites, and the interaction between ligand and receptors is vital for different aspects of cardiogenesis; EMT, as previously mentioned, as well as coronary vessel and cushion development (Arthur and Bamforth, 2011). Formation of the cardiac cushions requires endothelial to mesenchymal stimulation in the endocardium, which follows the activation of the TGFβ signalling pathway by different TGFβ isoforms (Boyer et al., 1999). TGFβ-2 has been shown to stimulate the expression of Slug, a protein characterized by the zinc finger motif, which has been implicated in neural crest cell migration and transformation (Nieto et al., 1994) as well as the initial steps of EMT in the AV canal that lead to cushion formation (Romano and Runyan, 1999, Romano and Runyan, 2000). TGFβ-2 is also thought to increase the migration of endocardial cells that are undergoing EMT, through the cardiac jelly, to form cushion mesenchyme cells (Azhar et al., 2003). TGFβ signal transduction is also involved in the differentiation of vascular smooth muscle cells, generated upon the formation of EPDCs after epicardial EMT (Sridurongrit et al., 2008). PDGFR-β is one of the genes upregulated by TGFβ (Xie et al., 2004), and both are involved in EMT and vascular smooth muscle cell development in the heart (Arthur and Bamforth, 2011, Sridurongrit et al., 2008, Lu et al., 2001). Further studies will be required to correlate the time of expression of both factors as well as whether they work in synergy to mediate the aforementioned roles.
9 BMPs, a sub family of the TGFβ family (Ashe, 2005), are activated by the cleavage of inactive pre- protein into homo or heterodimers, in their secreted form (van Wijk et al., 2007). As in the case of TGFβs, downstream signalling occurs via both canonical and non-canonical pathways. In the case of canonical signalling, they bind to Type 1 and Type 2 BMP Receptors (BMPRs), showing greater affinity for the former. Formation of the ligand-receptor complex leads to the phosphorylation of Type 1 BMPRs, by Type 2 BMPRs, which in turn phosphorylate and activate their downstream mediators, the SMAD proteins. These positively and negatively regulate the transcription of target genes (Wang et al., 2014).
The BMPs largely involved in cardiac development are BMP-2, BMP-4, BMP-5, BMP-6, BMP-7 and BMP-10. BMP-2 signaling plays a role in proepicardial cell migration towards the AV junction, with the ligand being present in the AV myocardium and the receptor being present in the proepicardium (Ishii et al., 2010). BMP-2 also functions in the development of the endocardial cushions (heart valve precursors), atria and ventricles, as well as mesenchyme formation in the AV cushions and efficient EMT (Ma et al., 2005, Rivera-Feliciano and Tabin, 2006). BMP-4 is expressed in the lateral plate mesoderm (Fujiwara et al., 2002), sinus venosus, myocardial outflow tract (Yuasa and Fukuda, 2009) and in the PE proper (Kruithof et al., 2006). Its absence causes defective myocardial OFT septation, impaired formation of coronary vessels (Liu et al., 2004a) and other cardiac anomalies (Jiao et al., 2003). The absence of either BMP-5, -6 or -7 does not cause impaired development of the heart, but knocking down both BMP-6 and BMP-7, or BMP-5 and BMP-7, results in cardiac defects (van Wijk et al., 2007, Yuasa and Fukuda, 2009). The absence of BMP-10 leads to impaired ventricular trabeculation and formation of thin ventricular walls (Neuhaus et al., 1999). BMPs have also been shown to be involved in increasing the expression of IGF-1 but inhibiting that of IGFBPs in osteoblasts, indicating the possibility of a similar mechanism in the context of heart development(Canalis and Gabbitas, 1994, Gabbitas and Canalis, 1995).
BMP signal transduction is regulated by various proteins which exert a pro- or anti-BMP effect by sequestering BMPs and inhibiting BMP-receptor interactions, hindering BMP transport across cells and thus disturbing the BMP gradient required for downstream signaling and maintaining some BMPs in inactive form by preventing the cleavage of pre-BMPs (Umulis et al., 2009b). These include the intracellular SMAD6 (Galvin et al., 2000) and MAPKKK TAK1 (Monzen et al., 1999), membrane-bound co-receptors such as BAMBI (BMP and Activin membranebound inhibitor) (Balemans and Van Hul, 2002) and DRAGON (Samad et al., 2005), or extracellular secreted proteins such chordin (Piccolo et al., 1996), crossveinless-2 (CV2), twisted gastrulation (Tsg) (Zakin et al., 2008, Umulis et al., 2009b), noggin (Smith and Harland, 1992) and the DAN family of
10 BMP antagonists comprising members including Dan itself, cerebrus, gremlin and protein related to dan and cerberus (PRDC) among others (Rider, 2010, Pearce et al., 1999, Topol et al., 1997, Hsu et al., 1998, Minabe-Saegusa et al., 1998).
The pseudoreceptor BAMBI(BMP and activin membrane-bound inhibitor) is related to TGF-β family type I receptors but without the intracellular serine/threonine (ser/thr) phosphotransferase domain (Onichtchouk et al., 1999). It serves to inhibit BMPs by interacting with BMPRs via its intracellular domain to hinder receptor-complex formation (Onichtchouk et al., 1999). On the other hand, Lin et al have described that in Wnt Signalling, while BAMBI can interact with the Wnt receptor Frizzled 5 via both intracellular and extracellular domains, LRP6 and Dishevelled interaction requires the intracellular domain. Subsequent signalling leads to an increase in intracellular β-Catenin and expression of genes involved in stimulating cell proliferation (Lin et al., 2008). It has also been found that hBAMBI can act in synergy with Smad6 and Smad7 to negatively regulate BMP signal transduction (Yan et al., 2009).
Chordin, a BMP inhibitor, possesses four VWFC domains, of which VWFC1 and VWFC3 are implicated in BMP-2 binding and VWFC1 and VWFC4 are implicated in BMP-7 binding (Zhang et al., 2007). This is an example of reversible inhibition due to the chordin-cleaving ability of tolloid metalloproteinases (Zakin and De Robertis, 2010). Ont1, a member of the olfactomedin family assists in the breakdown of chordin and enhances the action of the tolloid proteinases (Inomata et al., 2008). Twisted gastrulation binds to both chordin and BMP to act as a negative regulator of BMP activity (De Robertis, 2009).
CV2, another VWFC-containing BMP-modulating protein, binds to and inhibits BMPs via VWC1 (Zhang et al., 2010, Zhang et al., 2008). On the other hand, to exert its pro-BMP effect, it is suggested that CV2 forms a ternary complex with both chordin and BMP, and reduces the affinity of chordin for BMP, thus allowing BMPs to activate their downstream effectors via the BMPRs (Zhang et al., 2010).
Noggin is related to the cystine-knot growth factors and specifically inhibits BMP-2, BMP-4 and, to a lesser extent, BMP-7 (Groppe et al., 2002, Monzen et al., 1999). It binds to its target and masks receptor binding sites present on BMPs, thereby preventing them from binding to type I and type II cell surface BMPRs (Groppe et al., 2002, Zimmerman et al., 1996). However, it has been suggested that the antagonistic effect of Noggin in blocking the BMP signal transduction pathway transiently during early cardiogenesis may be crucial for cardiomyocyte differentiation (Yuasa et al., 2005). As
11 previously mentioned, BMP signaling modulates EMT and endocardial cushion formation. The absence of Noggin increases BMP signal transduction, subsequently leading to an increase in myocardial cell proliferation and ultimately a thicker myocardium, and larger endocardial cushions (Choi et al., 2007).
Thereare 22 FGFs that bind to and activate FGF Receptors (FGFRs) or receptor tyrosine kinases (Vega-Hernandez et al., 2011). The presence of FGFs and FGFRs in cardiac and vascular mesoderm indicates they are involved in cardiac development and the coronary vasculogenesis (Vega-Hernandez et al., 2011). When an FGF binds to the extracellular domain of the FGFR, the receptor dimerizes. Dimerization leads to a rapid activation of the protein’s cytoplasmic kinase domains, the first substrate of which is the receptor itself. The activated receptor becomes autophosphorylated at intracellular tyrosine residues (McKeehan et al., 1998), which in turn are docking sites for Phosphotyrosine Binding Proteins or Src homology 2 proteins (Forman-Kay and Pawson, 1999, Pawson et al., 1993). The RAS-RAF-MAPK and phospholipase C-γ signaling pathways are activated(Szebenyi and Fallon, 1999), leading to the regulation of cellular processes such as cell proliferation, migration and differentiation (Lavine and Ornitz, 2008).
FGF-2 has been shown to influence proepicardial cell and hemangioblast proliferation as well as the formation of cardiomyocytes from cardiac precursors (Rosenblatt-Velin et al., 2005, Kruithof et al., 2006, Burger et al., 2002, Faloon et al., 2000). FGF-9 and FGF-16 expressed in the epicardium are involved in the regulation of myocardial proliferation along with other factors (Vega-Hernandez et al., 2011, Li et al., 2011). Myocardial FGF-10 signaling to its receptors in the epicardium appears to propel EPDCs to the fibroblast lineage (Vega-Hernandez et al., 2011). An FGF gradient across the myocardium has been shown to be essential in nascent blood vessel patterning (Pennisi and Mikawa, 2005). FGF signaling is implicated in EMT (Mikawa and Gourdie, 1996), activation of EPDC movement into the myocardium and its subsequent invasion, following which coronary vessel formation takes place (Pennisi and Mikawa, 2009). It is also involved in regulation of myocardial growth and differentiation (Lavine and Ornitz, 2008). Myocardial FGF signaling and Hedgehog (HH) activation are required for the formation of the coronary vascular plexus (Mikawa and Gourdie, 1996). BMP-2 and FGF-2 signaling are implicated in the separation of myocardial progenitors and proepicardial progenitors, with BMP-driven determination of myocardial cell fate and FGF-driven determination of proepicardial cell fate (van Wijk et al., 2009). VEGFs are a family of cystine-knot-containing (Wiesmann et al., 1997) growth factors which are critical regulators of angiogenesis and vasculogenesis. They comprise the isotypes VEGF-A, VEGF-B, VEGF-C, VEGF-D, VEGF-E and placenta growth factor PIGF (Holmes et al., 2007).
12 These are secreted in their homodimeric form and linked by intra and inter-chain DS bonds (Muller et al., 1997). Of the 6 isoforms of VEGF-A isoforms, some are secreted and freely diffusible, whereas some remain sequestered to the extracellular matrix or are membrane-bound (Robinson and Stringer 2001). VEGFs bind to VEGF receptor (VEGFR) tyrosine kinases VEGFR-1, VEGFR- 2 and VEGFR-3. VEGFR-1 binds to VEGF-A, VEGFB, PIGF, VEGFR-2 binds VEGF-A, VEGF- C, VEGFD, VEGF-E and VEGFR-3 binds VEGF-C and VEGF-D (Robinson and Stringer, 2001). VEGF receptor signaling is similar to FGF receptor signaling. Binding of VEGF to VEGFR leads to dimerization of the receptor and transphosphorylation (between the dimerized receptor molecules) at tyrosine residues in the VEGFR domains, activating Src homology proteins and the PLC-γ, Akt and Ras-Raf signaling pathways (Fuh et al., 1998, Semela and Dufour, 2004b, Semela and Dufour, 2004a).
VEGF-A plays an important role in endothelial cell proliferation, migration, vascular permeability and VEGF-B, which is primarily expressed in the heart, is implicated in myocardial blood vessel development (Claesson-Welsh, 2008, Li et al., 2008, Olofsson et al., 1996). VEGF expression in the myocardium is dependent on Hedgehog activation, which in turn depends on myocardial FGF signalling (Lavine et al., 2006); this may play a role in inducing epicardial EMT (Tomanek et al., 2001). In mice lacking VEGF-A in the ventricular myocardium, myocardial wall thickness was reduced (Giordano et al., 2001). Cardiac function was also not preserved, likely due to a decrease in microvessel recruitment in the absence of myocardial VEGF (Giordano et al., 2001). Embryonic lethality was observed in mouse embryos in which the VEGF allele was ablated, due to defective angiogenesis and vessel formation (Ferrara et al., 1996).
VEGF signal transduction can be regulated by the co-receptors heparin sulphate proteoglycan (HSPG) and Neuropilin (Olsson et al., 2006). VEGFs display varying affinities for Heparin and HSPGs and their activity can be modulated by exposure to both these molecules at different concentrations. The mechanism of regulation varies, from presenting the ligand to its receptor, stabilizing the ligand-receptor complex, preserving ligand integrity as well as sequestering VEGFs at the cell surface or within the matrix to render them unavailable for interaction with VEGFRs or their co-receptors (Robinson and Stringer, 2001). VEGFs also bind to Neuropilins 1 and 2, which are expressed on endothelial cells. Nrp-1 binds VEGF-A with high affinity (Parker et al., 2012) and potentiates its interaction with VEGFR-2, playing a role in VEGF-A-induced signaling. The VEGF- A-Nrp-1 complex has also been shown to induce endothelial cell migration (Wang et al., 2003, Herzog et al., 2011). Neuropilin-2, which is co-expressed with VEGFR-3, may function as a co-
13 receptor for VEGF-C (Karkkainen et al., 2001), enhance VEGF-C activity and also be essential for the formation of small lymphatic vessels and capillaries (Yuan et al., 2002).
The PDGF family comprises PDGF-A, PGDF-B, PGDF-C and PGDF-D. These exist either as homodimers, or the heterodimer PGDF-AB. They bind to their PGDF tyrosine kinase receptors (PDGFR) PDGFR- αα, PDGFR-ββ and PDGFR-αβ (Kang et al., 2008). PDGFR- ββ binds PDGF- BB and PDGF-AB, and PDGFR-αα binds all three PDGF isoforms. The interaction between PDGFR-αα and PDGF-AA is specifically via a PDGF-AA binding site located in the extracellular domain of its PDGF receptor (Heidaran et al., 1990). Thus, PDGF-BB can interact with both α and β receptors whereas PDGF-AA can interact with only PDGFR-αα (Heidaran et al., 1991). PDGF- CC can bind PDGFR-αα and PDGFR- αβ (Cao et al., 2002). PDGFRs, like VEGFRs, dimerize upon binding of their PDGF ligands, leading to a cascade of events that ultimately result in the regulation of cell growth and migration.
PDGFs have been shown to play a role in angiogenesis (Cao et al., 2002) but not specifically coronary angiogenesis. Epithelial and mesenchymal cells secrete PDGF-A, which may act as a mitogen that stimulates ventricular development and cardiomyocyte proliferation (Kang et al., 2008). Both PDGF receptors are implicated in efficient EMT, upon which EPDCs give rise to fibroblasts and vascular smooth muscle cells (Mellgren et al., 2008, Smith et al., 2011). PDFG-BB can stimulate proepicardial cells expressing smooth muscle markers to undergo EMT and subsequently commit to the coronary smooth muscle cell lineage. This differentiation is mediated through PDGFR-β (Lu et al., 2001). On the other hand, it has been shown that PDGFR-α plays an important role in the formation of cardiac fibroblasts (Smith et al., 2011). PDGFR-α signaling, and to a lesser extent PDGFR-β signaling, is involved in the migration and functioning of cardiac neural crest cells (Van den Akker et al., 2008). PDGFR-α has been shown to influence correct atrial and ventricular septation along with outflow tract formation (Van den Akker et al., 2008).
The IGFs share molecular similarity with insulin. There exist 2 IGF ligands, IGF-1 and IGF-2, which bind to the IGF-1R and IGF-2R receptors. Due to their sequence homology with insulin, they also bind the insulin receptor. The IGF receptor is a heterotetramer, with the 2 extracellular α subunits participating in ligand binding and the 2 intracellular β subunits possessing tyrosine kinase activity (Laustsen et al., 2007). Binding of IGF to its receptor triggers a signalling cascade that ultimately regulates cell growth, proliferation and apoptosis.
14 IGFs have been implicated as epicardial mitogentic factors functioning in a paracrine manner; IGF- 2 has been shown to be secreted from the epicardium and exert a mitogenic effect on the formation of the compact myocardium (Li et al., 2011). It has been shown by Hertig and colleagues that IGF-1 and neuregulin-1 act in synergy to stimulate the morphogenesis of cardiac chambers. This occurs via the phosphatidylinositol 3-kinase (PI 3-K) pathway; inhibiting PI 3-K by Wortmannin results in the suppression of IGF-1 and NRG-1-induced Deoxyribonucleic acid (DNA) synthesis during ventricular chamber and cushion development (Hertig et al., 1999).
IGFBPs can be divided into 2 groups based on the level of affinity for IGF-1 and IGF-2. IGFBP- related Proteins (IGFBP-rP) 1-10 bind to IGFs with low affinity and IGFBPs 1-6 bind to IGFs with high affinity (Sun et al., 2011).The action of IGFBPs can be local, where they exert paracrine and autocrine control over regulating IGF function and signal transduction, or systemic, where there exists an endocrine control over IGFs in circulation (Yamada and Lee, 2009). IGFBPs consist of N- terminal and the C-terminal domains connected by a ‘hinge’ region in the middle. Both the N- and the C- terminals are essential for IGF binding. The limiting factor in the interaction between IGFs and IGFBPs seems to be the first 16 amino acids of the N-Terminal region of the IGFs (Bayne et al., 1988). IGFBPs regulate the availability of IGFs as well as reduce their activity by sequestering them so they cannot bind to and activate IGF receptors. They prolong the half-life of IGFs and are also vehicles for IGF trafficking.
In the later stages of heart development, IGFBP-4 interacts with the G-protein-coupled-receptor Frizzled, and coreceptor low-density lipoprotein receptor-related protein 6 and prevents Wnt- Receptor binding, thus promoting cardiogenesis by stimulating cardiomyocyte differentiation (Zhu et al., 2008). It has been demonstrated by Tamura et al that the expression and action of VEGFs in vascular endothelial cells is disturbed in the presence of IGFBP-rP1 (also called IGFBP-7 due to high sequence similarity to the IGFBPs) (Tamura et al., 2009). Contrastingly, IGFBP-3 has been shown to exert a pro-angiogenic effect on endothelial cells by upregulating VEGF (Granata R et al, 2007). The C-terminal of IGFBPs 1 and 2 contains an RGD motif for integrin binding (Jones et al., 1993). This may subsequently facilitate interactions with the extracellular matrix, via cell-surface integrins. Indeed, IGFBP-1 binds to the α5β1 integrin and stimulates cell migration in certain cells, thus exerting an IGF-independent effect (Jones et al., 1993, Gleeson et al., 2001). IGFBP-3 has also been shown to function in an IGF-independent manner by binding to the TGFβ type V receptor and inhibiting cellular growth (Leal et al., 1997).
15 Members of the Wnt family are glycoproteins involved in diverse cellular processes. There exist two categories of Wnt signalling based on the complex morphological changes mediated – the canonical pathway mediated by the Wnt1 group and the non-canonical pathway mediated by the Wnt5a group (Eisenberg and Eisenberg, 2006). An exquisite balance exists between activating and repressing canonical Wnt/β-Catenin and non-canonical Wnt signalling in the various stages of cardiogenesis.
Wnt1/ Wnt3a bind to the 7-transmembrane-domain receptor Frizzled, and along with its co- receptors result in the inhibition of GSK-3β kinase activity, a decrease in β-Catenin phosphorylation, and a concomitant increase in the amount of intracellular β-Catenin (the primary mediator of the canonical pathway). This in turn is required for the activation of T-cell Transcription Factor (TCF) and Lymphoid Enhancing Factor (LEF), which act upon target genes such as cyclinD1 to regulate cell cycle progression (Giles et al., 2003, Pal and Khanna, 2006). Canonical signalling promotes gastrulation and mesoderm formation before the cardiac fields form, as well as the expression of genes involved in early cardiac specification, but the inhibition of this signalling is required during the later stages of cardiogenesis (Zhu et al., 2008, Ueno et al., 2007). The canonical pathway also influences the correct migration of cardiac neural crest cells and their proliferation, the failure of which causes defects in outflow tract septation and endocardial cushion formation. In an elegant study, Zamora and colleagues have shown that β-Catenin is involved in formation of the subepicardial space, epicardium-derived cell invasion of the myocardium, as well as the differentiation of epicardial progenitors to form coronary smooth muscle cells (Zamora et al., 2007). Correspondingly, the compact myocardium in these mice lacking β-Catenin is also thin, likely due to a reduction in proliferation of cardiomyocytes (Zamora et al., 2007). Isl-1 and FGF-10 are expressed by progenitor cells of the anterior heart field (Cai et al., 2003, Cohen et al., 2007). Isl- 1 has been shown to be essential in the formation and development of the outflow tract, right ventricle and atria (Cai et al., 2003). In the absence of β-Catenin signalling, the expression of FGF- 10 is reduced, leading to a decrease in the expression of Isl-1 by AHF cells. This ultimately leads to impaired right heart morphogenesis (Cohen et al., 2007). Both Wnts and Wnt receptors are also expressed in abundance by endothelial cells, and Wnt signalling has been implicated in angiogenesis and vascular development (Cai et al., 2003).
Crescent and Dickkopf proteins are inhibitors of Canonical Wnt signalling (Brott and Sokol, 2005, Schneider and Mercola, 2001). Secreted frizzled-related proteins (sFRPs) compete with Wnts to bind the frizzled receptor, in another mechanism of canonical Wnt/β-Catenin inhibition (Eisenberg and Eisenberg, 2006). In the planar cell polarity pathway (a part of non-canonical Wnt signalling),
16 Wnt5a/ Wnt11 bind the Frizzled receptor and via RhoA and Rac, effect cell growth, proliferation, migration and polarity. The Wnt5a group is implicated in exerting an inhibitory effect on the Wnt1 group (Eisenberg and Eisenberg, 2006), in conjunction with the above mentioned repressors Crescent and Dickkopf, thus promoting cardiac development. Wnt11 and Wnt7a are differentially expressed in specific cardiac conduction tissues (Bond et al., 2003). Wnt11, in an autocrine or paracrine manner, may stimulate the recruitment of myocytes to differentiate into conduction cells. Wnt7 seems to be expressed at the same time as connexin40, a cardiac conduction marker expressed in the ventricular conduction system, and may serve to modulate its function (Bond et al., 2003) A direct association is observed between GATA transcription factors and the expression of the Wnt11 gene; inhibiting the expression of both GATA4 and GATA6 resulted in a marked change (reduction) in the levels of Wnt11 expressed, thus confirming that Wnt11 functions downstream of GATA factors to effect cardiogenesis (Afouda et al., 2008).
Cooperative interaction between growth factors directs development of the epicardium, myocardium, and coronary vasculature. During the early stages of heart development, there is significant crosstalk between the inner groove of the AV myocardium, that expresses BMPs, and proepicardial cells, which express BMPRs (Ishii et al., 2010). The extracellular matrix bridge contains matrix molecules like HSPGs and fibronectin (Nahirney et al., 2003); the former having the ability to interact with BMPs. Predictably, the experimental mis-expression of Noggin in the myocardium before proepicardial protrusion and migration impairs the attachment of proepicardial cells to the myocardium, possibly by interfering with proepicardial expression of signaling molecules (BMPs in particular) or adhesion factors (Ishii et al., 2010). The closeness of the liver bud to the proepicardial organ may provide an inductive cue for the expression of proepicardial marker genes such as Wt1, Capsulin and Tbx18 (Ishii et al., 2007). In addition, the regulation of Tbx18 by BMP4, Dkk1 and Noggin, and that of Wt1 by Noggin and Wnt3a, along with the role played by BMPs and FGFs, reveals that other paracrine factors are also involved in proepicardial induction and differentiation (Ishii et al., 2007). Co- operative interaction between FGFs and BMPs possibly mediates the differentiation of proepicardial cells into myocardial or epicardial lineages (Kruithof et al., 2006).
The subepicardial space also contains EPDCs (Pérez-Pomares et al., 1997, Dettman et al., 1998, Gittenberger-de Groot et al., 1998) and ECM proteins secreted by both epicardium and myocardium (Bouchey et al., 1996). Furthermore, it includes FGFs and VEGFs (Tomanek and Zheng, 2002) and BMPs (Yamagishi et al., 1999). FGF signaling from the epicardium to the myocardium and back may play a role in inducing EMT and in myocardial proliferation; moreover, FGFR1 signalling has
17 been implicated in the invasion of the myocardium by EPDCs (Pennisi and Mikawa, 2009). Cross- talk between the epicardium, subepicardium and myocardium is also essential in coronary and coronary vascular development (Kang and Sucov, 2005). Overexpression of FGFR1 in the proepicardium resulted in propelling EPDCs to the endothelial cell lineage (Pennisi and Mikawa, 2009). Furthermore, FGF-2 and VEGF may function cooperatively in order to simulate coronary vasculogenesis. Taken together, it is evident that multiple factors acting in tandem contribute to cardiogenesis in the developing embryo.
CRIM1 One such unique factor that could be at the nexus of growth factor signalling within the heart is Crim1. Crim1 is a novel transmembrane protein coded for by the Crim1 gene (Kolle et al., 2000, Glienke et al., 2002). Genes encoding CRIM1are evolutionarily conserved in vertebrates – this includes both humans and rodents (Kolle et al., 2000), the chicken (Kolle et al., 2003), zebrafish (Kinna et al., 2006) and Xenopus laevis(Ponferrada et al., 2012). In the nematode Caenorhabditis elegans, a Crim1 homolog called crm-1 has been found, although studies on its function are limited (Fung et al., 2007). Northern blot analyses initially revealed strong CRIM1 expression in the placenta, kidney, heart and skeletal muscle (Glienke et al., 2002). Further investigations using different transgenic models have revealed that this expression is defined and tightly regulated in the limbs, kidney, brain, spinal cord, lens, and vasculature (Kolle et al., 2000, Lovicu et al., 2000, Kolle et al., 2003, Pennisi et al., 2007, Fan et al., 2014, Phua et al., 2012, Glienke et al., 2002).
Figure 1.5: Schematic representation of the domain structure of CRIM1. CRIM1 possesses different domains such as 6 CRRs, an IGFBP-like domain, an RGD motif, an antistasin domain in the extracellular region. Adapted from original created by Dr. David Pennisi.
Structurally, the presence of six CRR motifs, an IGF Binding Protein like domain and an RGD motif (Kolle et al., 2000)(Figure 1.5) suggest Crim1 to be a versatile molecule, and while the CRRs have been shown to mediate the binding of Crim1 to TGFβ, BMP, VEGF and PDGF when Crim1 is co-expressed in the same cell as the growth factor (Wilkinson et al., 2003, Wilkinson et al., 2007), the functional significance of the IGFBP domain and the RGD motif remain unknown. The localization of Crim1 is also exciting – it has been found in the golgi and endoplasmic reticulum,
18 where proteins are modified post-translationally, and on the cell surface, where ligand-receptor interactions take place (Glienke et al., 2002, Wilkinson et al., 2003).
During development, the regulation of growth factor signalling is crucial, and proteins that can modulate the effect of growth factor activity have received much attention. This can occur by sequestering the growth factors and increasing their half-life, or tethering them so that they cannot bind their receptors. Conversely, they can also present growth factors to their receptors and increase downstream signalling, or, compete with receptors and hinder growth factor-receptor interactions, thereby decreasing downstream signalling. Diffusible growth factor-binding proteins can additionally act as facilitators of growth factor transport across cells through the extracellular space, which can cause changes in growth factor gradients resulting in either their positive or negative regulation (Umulis et al., 2009a). For example, proteins containing multiple cysteine-rich repeats such as chordin and noggin, are known to have antagonistic effects on BMP signal transduction. In case of the former, it binds BMPs directly (Larraín et al., 2000), and the latter, by masking binding sites on the receptor, prevents BMP binding to type I and type II cell surface BMPRs (Zimmerman et al., 1996, Groppe et al., 2002). Crossveinless-2, however, enhances BMP signaling by reducing the affinity of chordin for BMP and thus allowing downstream signal transduction to take place (Zhang et al., 2010). It is important to note, though, that whether these regulatory molecules are agnostic or otherwise, depends on their developmental context. This is exemplified in studies performed on mice lacking Crim1, which indicate that this protein can act as both an agonist and an antagonist during organogenesis.
To further explore its specific role in development, Crim1FLOX, a conditional loss-of-function allele was generated, by flanking exons 3 and 4 of Crim1 with LoxP sites (Chiu et al., 2012). The Crim1FLOX line, when crossed with the CMV-Cre deletor line, generated Crim1Δflox mice, intercrosses from which produced Crim1Δflox/Δfloxmutants (Chiu et al., 2012). Similarly, a genetrap line was also created by inserting a β-geo cassette into intron 1 of Crim1, creating a non-functional hypomorph (Pennisi et al., 2007). We have described perinatal lethality in mice homozygous for the Crim1KST264genetrap and embryonic lethality in Crim1∆flox null alleles (Pennisi et al., 2007, Chiu et al., 2012). These mice display defects in several organ systems (Glienke et al., 2002, Pennisi et al., 2007, Lovicu et al., 2000, Phua et al., 2012, Fan et al., 2014), which further substantiates the significance of Crim1 during development.
19 The role of Crim1 in organogenesis. Numerous studies in the past decade have taken us closer to understanding the role of Crim1 in organogenesis, by first mapping its expression patterns, and then analysing its roles in different organ systems. A summary of these findings is outlined below.
Kidney. The role of Crim1 in the kidney has been studied extensively. In the mouse kidney, during development, Crim1 expression is observed in smooth muscle cells lining the endothelium, within parietal epithelial cells, mesangial cells and glomerular podocytes from 15 days post coitum (dpc)(Pennisi et al., 2007, Georgas et al., 2000, Wilkinson et al., 2007). At the same age, analysis of kidneys from Crim1KST264/KST264 mice revealed smaller kidneys compared to their wild type littermates, suggesting that Crim1 plays a role in nephrogenesis (Pennisi et al., 2007). Glomerular lesions and capillary defects, along with podocyte effacement, were also observed in Crim1KST264/KST264 mice (Wilkinson et al., 2007). Crim1 is also expressed alongside VEGF-A in renal glomerular podocytes. When Crim1 was ablated, VEGF-A diffusion away from the podocytes leading to a disruption in the VEGF-A gradient was demonstrated, with a corresponding activation of the VEGF-A receptor Flk1 in adjacent vascular endothelial cells (Wilkinson et al., 2007). This indicates that Crim1 can modulate VEGF-A signalling by regulating the delivery of VEGF-A to endothelial cells within the glomerulus (Wilkinson et al., 2007).
Renal abnormalities have also been observed in the adult kidney. By using Crim1KST264/KST264 outbred mice, which show survival of a small number of homozygous mutants, renal defects such as multiple glomerular cysts, interstitial fibrosis and endothelial cell thickening were revealed (Wilkinson et al., 2007). Another study showed that endothelial anomalies, including a discontinuous endothelium with aberrant deposition of collagen, and an increase in vascular permeability, exist in Crim1KST264/KST264 adult mice, along with renal fibrosis (Phua et al., 2012). This is posited to be due to an association of Crim1 and TGFβ-1, which enhances EndoMT (Kim et al., 2001, Varga and Wrana, 2005, Phua et al., 2012). CRIM1 expression has also been demonstrated in the glomerular podocytes of the adult human kidney, as seen in vivo by performing immunohistochemical experiments on renal tissue (Nyström et al., 2009). BMPs and VEGFs are also expressed in this location (Godin et al., 1999, Simon et al., 1995), suggesting that CRIM1 could regulate the activity of these growth factors by sequestering them and controlling their release into the local environment (Nyström et al., 2009).
20 Placenta. An organ that is crucial during embryogenesis is the placenta. Placental development ensues after implantation, and consists of syncytiotrophoblast cells from the labyrinthine zone and spongiotrophoblast cells and glycogen trophoblast cells from the junctional zone. In murine models, Crim1 expression is observed in multiple placental cell types at different stages – at 9.5 dpc in chorionic trophoblasts, at 13.5 dpc and 15.5 dpc in syncytiotrophoblasts, and at 13.5 dpc in spongiotrophoblasts (Pennisi et al., 2012). In mice lacking Crim1, placental development is hindered; the size of the placenta is reduced from 13.5 dpc until 17.5 dpc, along with a decrease in the size of Crim1KST264/KST264 homozygous embryos themselves (Pennisi et al., 2012). Sinusoidal trophoblast cell endowment in reduced, and the number of glycogen cells is increased, in Crim1KST264/KST264 placentae at 15.5 dpc, possibly due to aberrations in growth factor signalling (Pennisi et al., 2012), including pathways such as such as the IGF and VEGF pathways (Randhawa and Cohen, 2005, Charnock-Jones et al., 1994).
Lens and retinal vasculature. Crim1 expression in the developing eye is observed from 9.5 dpc until at least day 21 postnatally (P) 21 (Lovicu et al., 2000). It starts in the lens placode, the precursor to the lens, and proceeds, by 11.5 dpc, to all lens cells. . Crim1 transcripts have also been observed in the corneal epithelium and endothelium by 15.5 dpc, and even later at 18.5 dpc in the retinal epithelium and retinal ganglion cells, until P21, when Crim1 is expressed only in the lens (Lovicu et al., 2000). In Crim1glcr11(glaucoma relevant 11) mutants, aptly named to describe their glaucoma and cataract phenotype, lens defects are observed from 16.5 dpc. These include junctional aberrations due to cell adhesion defects, abnormalities in polarity and a decrease in the number of lens epithelial cells that can proliferate, which ultimately results in a smaller, irregular lens (Zhang et al., 2016). True to their name, lens cataracts and atypical cell proliferation within the retina are also displayed by Crim1glcr11 mutant mice by P60. Defects in cell adhesion in these mutants is likely due to an interaction between Crim1 and β1 integrin, both of which are expressed in lens epithelial cells and lens fiber cells (Zhang et al., 2016). Further analysis revealed that Crim1 can indeed affect the phosphorylation of focal adhesion kinase (FAK) and ERK, downstream effectors of the β1 integrin signalling pathway, and ultimately results in membrane-bound Crim1 regulating lens morphogenesis (Zhang et al., 2016).
In vascular endothelial cells, Crim1 expression is observed, both in vivo and in vitro (Glienke et al., 2002). Its role in vascular development has been analysed in postnatal mouse retinas (Fan et al., 2014). This has been made possible by employing Pdgfrb-iCreER mice and crossing them with the
21 Crim1Flox allele, so as to enable inducible deletion of Crim1 specifically from endothelial cells of the vasculature (Fan et al., 2014). This resulted in defective retinal vascular development, and includes aberrations such as reduced vessel density and branching, as well as vessel regression visible in the first week of postnatal development (Fan et al., 2014). As previously mentioned, VEGF-A plays an important role in endothelial cell development, and indeed, analysis of VEGF-A and Crim1 mutants revealed angiogenesis defects and reduced expansion of the vasculature that was more severe in double mutants compared to individual knockouts, indicating that Crim1 and VEGF function along similar pathways during the development of the retinal vasculature (Fan et al., 2014). Furthermore, loss of Crim1 also resulted in abnormal VE-cadherin distribution at junctions of cells at the angiogenic front (Fan et al., 2014). This is congruent with endothelial tube formation defects observed in CRIM1-deficient HUVECs (Glienke et al., 2002), and indicates a cell adhesion anomaly in the absence of CRIM1. Interestingly, the cytoplasmic domain of CRIM1 has been shown to interact indirectly with β-catenin and N-cadherin in Xenopus(Ponferrada et al., 2012), and the RGD domain has been implicated in integrin signalling in the murine lens (Zhang et al., 2016).
Nervous system. During the development of the murine spinal cord, Crim1 expression is observed in the floor plate from 9.5 dpc, and in motor neurons at later stages (Kolle et al., 2000). Crim1 is also expressed in the forebrain and hindbrain from 11.5dpc, and in the midbrain at 13.5 dpc (Kolle et al., 2000). However, how Crim1 affects the formation and development of the nervous system is still unknown. Insights from Crim1 function in cell adhesion and polarity, arising from cooperative interaction between Crim1 and β1 integrins at the leading edges of lens epithelial cell projections could assist in our understanding neural development, as both cell adhesion and polarity are crucial in neural stem cell differentiation within the embryonic brain.
Crim1 in heart development? The heart expresses high levels of Crim1, but its role in cardiac development has not yet been examined in detail. There is evidence of Crim1 expression, owing to studies using the Crim1KST264 genetrap line, in the mouse proepicardium at 9.5 dpc, and the epicardium at all stages of development. In the later stages, strong Crim1 expression is also observed in coronary vascular smooth muscle cells, and also in coronary vascular endothelial cells, albeit at lower levels (Pennisi et al., 2007). Many epicardium-derived growth factors are also required for the development of the myocardium, along with other aspects of heart development. Much like its role in nervous system development, little is known about the signals that Crim1 could modulate during cardiac development. With its dynamic expression in different organs, Crim1 appears to be central in
22 governing key developmental processes, in a cell-autonomous manner as well as through regulation of paracrine factors and controlling the availability of growth factors to adjacent tissues. Identification of such signalling pathways will provide much insight into the complex role of Crim1 during heart development. The presence of six cysteine-rich repeats, the IGFBP-like domain and the RGD motif, alongside an intracellular domain whose biological significance has not been fully explored, renders Crim1 to be a multifaceted protein with the potential to modulate the activity of several growth factors, thus laying the foundation of my project.
Figure 1.6: Schematic representation of the established and potential roles of Crim1. Some potential functions of Crim1 in the context of development include tethering growth factors to the cell surface, presenting them to their receptors, affecting the production of growth factors and their delivery to the cell surface for signalling or secretion.
As such, we hypothesize that Crim1 plays both cell-autonomous and paracrine roles in epicardial and compact myocardial development within the nascent embryo via possible mechanisms as indicated in Figure 1.6. Furthermore, we postulate that coronary vasculature development and homeostasis is dependent on Crim1 function. We also hypothesize that a number of the functions mediated by Crim1 are dependent on its ability to bind growth factors and modulate their signalling. Finally, we posit that Crim1 can bind IGFs, likely via its IGFBP domain, and regulate downstream signalling.
23 Overall aim: To investigate the role of Crim1 in cardiac development
Aim 1: Examining the cell-autonomous and paracrine roles of Crim1 in the epicardium and compact myocardium.
Aim 2: Examining the cell-autonomous role of Crim1 in the coronary vasculature, specifically in endothelial cells, and growth factor signalling perturbed upon loss of Crim1 function.
Aim 3: Investigating whether the IGFBP domain of Crim1 can serve to bind IGFs and modulate downstream signalling
24
Chapter 2
Materials and general methods
25 Ethics Statement. This study conformed to The University of Queensland’s Animal Welfare Unit guidelines for animal use in research (AEC approval numbers: AIBN/274/08/NHMRC, SMBS/420/11/NHMRC), and followed the National Institutes of Health Guide for the Care and Use of Animals and the Australian Code of Practice for the Care and Use of Animals for Scientific Purposes. All experimental protocols were approved by The University of Queensland’s Animal Ethics Unit.
Mouse Lines. The mouse lines used in this study were the Crim1FLOX conditional mutant (Chiu et al., 2012), Crim1∆flox conditional null (Chiu et al., 2012), and Crim1KST264genetrap line (Pennisi et al., 2007), ROSA26 Cre-reporter line (R26R) (Soriano, 1999), WT1-Cre line (Wilm et al., 2005), WT1- CreERT2 line (Zhou and Pu, 2012), Mlc2v-Cre line (Chen et al., 1998) and Tie2-Cre line(Koni et al., 2001, Kisanuki et al., 2001). The Crim1KST264 mouse line was created as part of a genetrap screen (Leighton et al., 2001), resulting from insertion of a β-Geo cassette into intron 1 of Crim1, thereby resulting in the fusion of exon 1 and the β-Geo cassette within the Crim1 transcript, but creating a hypomorph with the expression of a minor, alternately spliced isoform of Crim1, instead of a complete null (Pennisi et al., 2007). The Crim1 conditional knockout mouse line, Crim1FLOX was generatedby flanking exons 3 and 4 with unidirectional LoxP sites (Chiu et al., 2012). These Crim1FLOX mice were crossed with a deletor line containing the ubiquitously expressed CMV-Cre to create the Crim1Δflox line (Chiu et al. 2012). All lines were maintained on a C57Bl6 genetic background.
Animal Breeding. Embryos were obtained from timed matings between Crim1+/Δfloxintercrosses, Crim1+/KST264 intercrosses, or Crim1FLOX/FLOX;R26R females mated with Crim1+/FLOX; WT1- Cre, Crim1+/FLOX; +/WT1-CreERT2 or Crim1+/FLOX; Mlc2v-Cre males.
Sample collection. Pregnant mice were sacrificed by cervical dislocation and dissected in order to obtain embryo and heart samples at appropriate embryonic (E) stages. Embryos were also staged according to published compendia (Theiler, 1989) by assessing morphological landmarks visible under the dissecting microscope. For all embryos, a small tail-tip sample was collected for genotyping.
26 Genotyping. Genotyping was performed according to published methods (Chiu et al., 2012). Tail-tip samples were immersed in 50 μL of DNA QuickExtract (Epicentre, EPQE09050), given a strong vortex and a brief centrifugation. Samples were placed in a heating block at 65°C for 20 minutes, after which they were vortexed again and centrifuged. The samples were then placed in a heating block at 98°C for 10 minutes and subsequently diluted with 150μL ultrapure water. Polymerase Chain Reactions (PCR) comprised 0.4mM primers, 2μLgDNA, and 0.15μLTaq-polymerase (BioLine) made up to 25μL with PCR myTaq red buffer (BioLine) and ultrapure water. PCR amplicons were detected using standard gel electrophoresis, with 1.5% agarose gels.
Primer pairs for genotyping 5’ – 3’ sequence Amplicon (size) F2 (forward) + B3 (reverse) F2: 389bp (wild-type allele) TTCTTGGGTTCACAGTTAGTCC 471bp B3: AATGGAATCTTCAGGGCAAC (Crim1FLOXallele) F2 (forward) + CKO (reverse) F2: ~600 bp TTCTTGGGTTCACAGTTAGTCC CKO: CTCTGAAGGTCATCAACAGC Cre (forward + reverse) Cre-For: ~400bp CCTGGAAAATGCTTCTGTCCG Cre-Rev: CAGGGTGTTATAAGCAATCCC LacZ (forward + reverse) LacZ-For: ~500bp CAGCACATCCCCCTTTCGCC LacZ-Rev: CCAACGCAGCACCATCACCG Wt1-CreERT2 - Common, Wt1-CreERT2 Common: 192 bp and 400 bp WT1-CreERT2 - Reverse ATCGCAGGAGCGGAGAAC Heterozygote Wt1-CreERT2 WT- Reverse: (+/Wt1CreERT2) GAAGGGTCCGTAGCGACA Wt1-CreERT2 Cre: GCAAACGGACAGAAGCATTT
Polymerase Chain Reaction (PCR) cycling conditions:
27 The PCR cycle conditions used for F2/B3 and Cre genotyping were: Initial denaturation: 95°C for 5 minutes 37 cycles of: Denaturation: 95°C for 30 seconds Annealing: 54°C for 30 seconds Extension: 72°C for 45 seconds Final extension: 72°C for 4 minutes Cooling: 4°C
The PCR cycle conditions used for F2/CKOR genotyping were: Initial denaturation: 95°C for 5 minutes 37 cycles of: Denaturation: 95°C for 30 seconds Annealing: 54°C for 30 seconds Extension: 72°C for 1 minute Final extension: 72°C for 4 minutes Cooling: 4°C
The PCR cycle conditions used for LacZ genotyping were: Initial denaturation: 95°C for 5 minutes 35 cycles of: Denaturation: 95°C for 30 seconds Annealing: 66°C for 30 seconds Extension: 72°C for 45 seconds Final extension: 72°C for 4 minutes Cooling: 4°C
The PCR cycle conditions used for WT1-CreERT2 genotyping were: Initial denaturation: 95°C for 3 minutes 36 cycles of: Denaturation: 94°C for 30 seconds Annealing: 51°C for 30 seconds Extension: 72°C for 45 seconds Final extension: 72°C for 4 minutes Cooling: 4°C
28
Figure 2.1: Schematic representation of the Crim1 alleles detected by genotyping. Black rectangles represent the coding regions of 17 exons that make up the fully functional wild-type Crim1 allele, and white rectangles represent the 5’ and 3’ untranslated regions (UTR). Arrows represent the location where the forward (F2) and reverse (B3, PGK, CKO) primers will amplify. The F2/B3 primer pair will detect and amplify the Crim1 WT and FLOX sequences. The Crim1FLOX allele has the addition of LoxP sites (red triangles). These are the recognition sites for Crerecombinase. Note the removal of exons 3 and 4 in the subsequent Crim1∆flox. However the Crim1 neomycin resistance cassette (blue rectangle) that was used as a selective marker during the initial generation of these mice, is retained in Crim1∆flox alleles. The removal of the neomycin cassette via FLPe recombination at the FRT sites (green ovals) has allowed for the generation of the Crim1∆flox∆frt allele. The process of eliminating the neomycin cassette occurred during the progress of this project. Hence, data in this thesis has utilised both mouse strains. Schematic not drawn to scale. Image created by Renee Chand.
Sample preparation. Embryonic hearts were dissected in Phosphate Buffered Saline (PBS) and fixed for 1-2 hours in 4% paraformaldehyde (PFA)/PBS at 4°C, washed in PBS, and photographed in whole-mount as required. Embryonic samples from the WT1-Cre, WT1-CreERT2 and Mlc2v-Cre (β-galactosidase activity being a readout of Cre activity based on the R26R reporter) and Crim1+/KST264 intercrosses (β-galactosidase activity being under the control of the endogenous Crim1 promoter) were X-Gal- stained in whole-mount as previously described (Pennisi et al., 2007). The hearts were processed for paraffin or cryosectioning. For paraffin processing, hearts were dehydrated through a series of graded ethanol baths (70%, 80%, 90%, 95% 100% Ethanol), and then infiltrated with wax. They were subsequently embedded in wax blocks and 7 µm sections cut on a microtome. For obtaining cryosections, samples were washed in PBS, cryoprotected in 30% sucrose/PBS at 4 °C for 4-6hours,
29 embedded in Tissue-Tek®O.C.T™ compound (Sakura Fine technical, Tokyo, Japan) and 10μm sections were cut on a cryostat and air-dried. Some X-Gal-stained sections were counterstained with nuclear fast red (NFR; Vector Laboratories), while others were subjected to immunohistochemical analyses as specified.
Immunohistochemistry. For paraffin-embedded sections, samples were de-waxed by submerging in xylene baths twice for 5 minutes, a series of graded ethanol (100%, 90%, 80%. 70%) once for 5 minutes each, and finally water for 5 minutes. They were then put through two PBS washes for 3 minutes each, one wash with PBS-TritonX 0.15% to permeabilize the sample for 3 minutes and three PBS washes for 5 minutes. The slides were then outlined with a PAP pen, and treated with 0.3% hydrogen peroxide solution for 20 minutes to quench endogenous peroxidase activity. They were then washed in PBS three times for 5 minutes each, blocked with 5% Bovine Serum Albumin (BSA) for 1 hour at room temperature, and incubated in primary antibody (Ab) overnight at 4°C. The samples were washed again with PBS three times for 5 minutes each, incubated in secondary antibody for 1 hour at room temperature, and after three more PBS washes, subjected to 1% 3,3′-Diaminobenzidine (DAB) staining. Finally, the slides were subjected to ethanol washes (70%, 80%, 90%, 100%) for 3 minutes, 2 washes in xylene for 5 minutes, and mounted with coverslips and DPX. For all immunohistochemical procedures, no-Ab and no-primary Ab controls were always included.
Immunofluorescence. Cryosectioned samples were allowed to equilibrate to room temperature by air-drying, and put through two PBS washes for 3 minutes, one wash with 0.15% TritonX-PBS (PBS-Tx) for 3 minutes, and two washes with PBS for 3 minutes, following which they were blocked in 5% BSA for 1 hour. They were then incubated with primary antibodies overnight at 4°C. They were then washed three times with PBS for 5 minutes, and incubated in secondary antibodies for 1 hour at room temperature. They were also incubated with 4',6-diamidino-2-phenylindole (DAPI) for 10 minutes at room temperature, after which they were washed with PBS three times for 5 minutes, mounted in PBS-glycerol 50% and coverslips sealed with nail polish. For immunofluorescence procedures, no-Ab and no-1°Ab controls were always included.
30 Antibodies. PRIMARY ANTIBODY SPECIES SPECIFICITY DILUTION CATALOG # Phosphohistone H3 Rabbit Polyclonal 1:300 Millipore #06-570 Cleaved Caspase 3 Rabbit Monoclonal 1:200 Cell Signalling #9664 Phospho-AKT Rabbit Monoclonal 1:1000 Cell Signalling #4060 Phospho-ERK1/2 Rabbit Monoclonal 1:200 (IF) Cell Signalling 1:2000 (WB) #4370 Phospho-SMAD1/5 Rabbit Monoclonal 1:100 Cell Signalling #9516 Phospho-SMAD1/5 Rabbit Monoclonal 1:150 Thermo Fisher Scientific#700047 Phospho-SMAD2 Rabbit Polyclonal 1:300 Millipore AB3849 Phospho-IGF-1R Rabbit Monoclonal 1:1000 Cell Signalling #3024S Total AKT Rabbit Polyclonal 1:1000 Cell Signalling #9272 Total ERK 1/2 Rabbit Polyclonal 1:1000 Santa Cruz sc-93 Total SMAD1/5/8/9 Rabbit Polyclonal 1:1000 Novus BiologicalsNB100- 56656
Total IGF-1R Rabbit Polyclonal 1:1000 Santa Cruz, sc-712 Smooth muscle α-actin Mouse Monoclonal 1:200 Santa Cruz, sc-3225 β-catenin, whole antiserum Rabbit Wh. antiserum 1:300 Sigma Aldrich #C2206 β-catenin (active) Mouse Monoclonal 1:100 Millipore #05-665 SM22 alpha Rabbit Polyclonal 1:300 Abcam Ab14106 Periostin Rabbit Polyclonal 1:200 Abcam Ab92460 Wilms’ Tumor 1 Mouse Monoclonal Dako #M3561 CD31 (PECAM-1) Rat Monoclonal 1:300 BD Pharmingen #550274 HA Tag Rabbit Monoclonal 1:1000 Cell Signalling
31 #3724 Myc Tag Mouse Monoclonal 1:1000 (WB) Cell Signalling 1:1000 (IP) #2276 Angiogenesis Sampler Kit: Rabbit Multiple multiple Cell Signalling Phospho-Akt (Ser473) #8696 Phospho-p44/42 MAPK (ERK1/2 Thr202/Tyr 204) Normal IgG Mouse 1:1000 Thermo Fisher Scientific #02-6200 GAPDH Rabbit Monoclonal 1:10000 Cell Signalling #2118 -tubulin Rabbit Polyclonal 1:10000 Cell Signalling #2146 Crim1 Mouse Monoclonal 1:200 R&D Systems #253511 Isolectin B4 1:100 Thermo Fisher Scientific #MP21410 MF20 (supernatant) Mouse Monoclonal Developmental Studies Hybridoma Bank
SECONDARY ANTIBODY SPECIES DILUTION CATALOG # AlexaFluor 594 Rat 1:500 Thermo Fisher Scientific #A-21209 Mouse 1:500 Thermo Fisher Scientific#R3712 1 Rabbit 1:500 Thermo Fisher Scientific#R3711 7
32 AlexaFluor 488 Rat 1:500 Thermo Fisher Scientific#A- 21208 Mouse 1:500 Thermo Fisher Scientific#R3712 0 Rabbit 1:500 Thermo Fisher Scientific#R3711 6 HRP conjugate Rabbit 1:500 (IHC) Promega; 1:10000 (WB) #W4011 Mouse 1:500 (IHC) Promega; 1:10000 (WB) #W4021 Anti-rabbit Biotin Rabbit 1:500 Sigma; # B7389 IgG mouse IRDYE Mouse 1:10000 Licor #926-32212 IgG rabbit IRDYE Rabbit 1:10000 Licor #926-68073
TSA plus Cyanine 3 system was used for signal amplification in the case of pERK1/2 and pAKT (PerkinElmer #NEL744001KT). The Vectastain Elite ABC kit (Standard) was also used (Vector Laboratories; #PK6100).
Bacterial transformation. Stellar competent cells (PT 5505-2) were used for bacterial transformation. Briefly, cells were thawed on ice, and 5-15 ng DNA was added to 20 L competent cells in eppendorf tubes. The tubes were placed on ice for 30 minutes, following which the cells were heat-shocked at 42°C for 1 minute. They were placed on ice for 2 minutes, after which 250 L SOC medium prewarmed to 37°C was added to each tube. The cells were then incubated on a shaker for 30 minutes at 37°C, after which appropriate amounts of culture were plated on Luria agar plates (10 g Tryptone, 5 g Yeast extract, 10 g Sodium Chloride, 15 g Agar dissolved in 1 L dH20) containing Ampicillin, and incubated overnight at 37°C.
Mini and Midi preps for plasmid DNA purification. LB broth (10 g Tryptone, 5 g Yeast extract, 10 g Sodium Chloride dissolved in 1 L dH20) was inoculated with bacteria containing plasmid of interest, and DNA purification was carried out according to manufacturers’ protocols (QiagenMiniprep, Machary Nagel Midiprep).
33
DNA Sequencing. The DNA to be sequenced (800-1000 ng) and each primer (0.64 µM) in a total volume of 12 µL were sent to Australian Genome Research Facility (AGRF, University of Queensland, QLD, Australia) for the ‘purified DNA’ sequencing service.
Cell lines. H-CMVECs and pooled HUVECs (Lonza, CC-7030 and CC-2519) were cultured in EGM-2MV and EGM-2 BulletKit media respectively (Lonza, CC-3202 and CC-3162). HEK293 cells and Lenti-X cells (Kind gift from Dr. Aaron Smith) were cultured in Dulbecco's Modified Eagle Medium (DMEM) media supplemented with Pen/Strep antibiotic and 10% FBS. All cell lines were maintained at 37°C in a humidified incubator containing 5% Carbon Dioxide (CO2).
Immunoblotting. Protein Extraction Cells were scraped in radioimmunoprecipitation (RIPA) buffer (150mM NaCl, 1% NP-40, 1% sodium deoxycholate, 0.1% SDS, 20 mMTris (pH8.0), 1 mM Na3VO4, 10mM Na4P2O7, 2 mMEthylenediaminetetraacetic acid (EDTA) plus Complete EDTA-free protease inhibitor cocktail (Roche, Mannhein, Germany). Extracts were cleared by centrifugation at full speed for 10minutes at 4°C. Protein concentration was determined by bicinchoninic acid assay (BCA) (Pierce, IL, USA).
Sodium Dodecyl Sulphate Polyacrylamide Gel Electrophoresis (SDS‐PAGE) Equal amounts of protein were incubated at 100°C for 5 minutes in sample buffer (2X - 0.125M Tris-hydrochloric acid HCl, pH 6.8, 4% SDS, 10% β-mercaptoethanol, 40% glycerol, 0.05% bromphenol blue), loaded into appropriate SDS-PAGE gels and separated by electrophoresis. Appropriate amounts of 40% acrylamide (BioRad, CA, USA), 1.5 M TrisHCl pH 8.0, 10% ammonium persulphate (APS), 10% SDS, N,N,N’-tetra-methylethylenediamine (TEMED (Roche)) and dH2O were used to prepare resolving gel mixes, which were then set in a Mini-PROTEAN® casting apparatus (BioRad). Isopropanol was overlaid onto the gel to ensure a straight interface. After polymerisation, the isopropanol was removed and stacking gel mix (40% Acrylamide, 05 M TrisHCl pH 6.8, 10% TEMED, 10% APS, dH20) was poured on top of the resolving gel, with a spacer comb inserted. Electrophoresis was performed in Mini-PROTEAN® Tetra cell tanks (BioRad), in running buffer (TrisHCl, Glycine, 10% SDS, dH20) at 180 V for 1 hour. Precision Plus Protein™ Dual Color Standards (BioRad, CA, USA) was used as a size reference.
34
Electroblotting and Probing of Membranes After electrophoresis, gels were transferred onto PVDF membranes (Immobilon-Fl, Merck Millipore), set up in transfer tanks (BioRad, CA, USA) and run in transfer buffer (TrisHCl, Glycine, Methanol, dH20) at 90 V for 1.5 hours. Membranes were blocked with 5% BSA or 5% skim milk in Tris Buffered Saline (TBS) containing 0.1% Tween (TBS-T) for 2 hours at room temperature. Blots were probed by incubating with specific primary antibody diluted in blocking buffer overnight at 4°C. Membranes were then washed with TBS-T thrice for 10 minutes, thrice. Blots were incubated with secondary antibody diluted in blocking buffer for 1-2 hours at room temperature. They were washed again and visualised either with a chemi-luminescent reagent (Merck Millipore) via exposure to X-ray film, or by using the Licor scanner. All densitometry quantitation was performed on immunoblots using ImageJ® software. All test values were normalised to respective loading control and represented relative to experimental control and graphed with Prism 6.
Stripping ofPolyvinylidene fluoride (PVDF) Membranes PVDF blots were stripped when required using Re-Blot Plus Mild Stripping Solution (Millipore, CA, USA) according to the manufacturer’s instructions and re-probed for other proteins of interest, or loading controls. Before re-blocking and re-probing, blots were washed with TBS-Tx for 10 minutes.
Data documentation and quantification. Bright-field images of tissue sections were captured using an Olympus BX-51 BF/DF slide microscope with Canon digital cameras with DP Controller software (Olympus). Whole-mount images were captured using either an Olympus SZX-12 stereo-microscope with DP Controller software, or a Zeiss Stemi 2000 stereo-microscope with an AxioCam digital camera with AxioVision software (Zeiss). Fluorescent images of tissue sections were captured on an Olympus IX-81 laser scanning confocal microscope and an Olympus FV1000 upright laser scanning confocal microscope with Olympus Fluoview software (Ver3.1a). Images were adjusted for colour levels, brightness and contrast, and figures compiled, using Adobe Photoshop software. The number of samples included in data averages is indicated in the figure legends where applicable, and quantification procedures are outlined in detail in individual chapters.
Statistical analyses. Statistical significance between pooled data from mutant and control groups was determined using a two-tailed, unpaired Student’s t-test with or without Welch’s correction, a one-way ANOVA with
35 Dunnett’s post-test, a two-way ANOVA with Bonferroni’s post-test, or a Z test as specified in the figure legends. Quantified data shown are mean ± standard deviation. GraphPad Prism v6.0 was used to conduct all analyses and generate graphs.
36
Chapter 3
Examining the cell-autonomous and paracrine roles of Crim1 in the epicardium and compact myocardium.
The data contained in this chapter was published in the journal Scientific Reports in January 2016.
Crim1 has cell-autonomous and paracrine roles during embryonic heart development. Iyer S, Chou FY, Wang R, Chiu HS, Raju VK, Little MH, Thomas WG, Piper M, Pennisi DJ. Sci Rep. 2016 Jan 29;6:19832. doi: 10.1038/srep19832.
37 Author Contributions: Figures 3.2, 3.4, 3.5H, 3.7, 3.10, 3.19C: complete experiments were performed by Dr. David Pennisi Figures 3.5A-G, 3.6:complete experiments were performed by Han Chiu and Dr David Pennisi Figures 3.8, 3.9, 3.25, 3.27, and handling of WTI1-CreERT2 mice, sectioning and staining in figure 3.13: complete experiments were performed by Grace Chou (Honours student under supervision of Dr. David Pennisi) Figures 3.21, 3.22: complete experiments were performed by Richard Wang (Honours student under supervision of Dr. David Pennisi)
These experiments were included in the thesis to allow a comprehensive story on the role of Crim1 in embryonic heart development.
38 Introduction. The proepicardium supplies cells that migrate to and encompass the myocardium of the rudimentary heart as a single sheet known as the epicardium. Epicardial cells, via a process called Epithelial to Mesenchymal Transition (EMT), give rise to cells of the coronary vasculature, cardiac fibroblasts and valve cells upon invading the myocardium, or reside in the subepicardium as subepicardial cells. This process requires the interplay of various factors that both promote and restrain epicardial EMT and invasion, including TGFs, BMPs, PDGFs, FGFs, Neurofibromin 1 (NF1) and Wilms Tumour 1 (WT1) (Morabito et al., 2001, Pennisi and Mikawa, 2009, Martínez-Estrada et al., 2010, Smith et al., 2011, Baek and Tallquist, 2012, Sánchez and Barnett, 2012). The myocardium is divided into two distinct components – the outer compact myocardium and the inner trabeculated myocardium, and complex morphogenetic patterning events both establish and regulate this process. Both the epicardium and the myocardium provide secreted factors that are essential not only to initiate an epicardial EMT and enable the process to reach completion, but also for myocardial growth, development and differentiation of epicardial lineages and the coronary vasculature.
This dual role of the epicardium – to contribute epicardium-derived lineages as well as to provide regulatory signals required for the development of the myocardium, and the molecular factors that govern cardiogenesis, are not completely understood. There are a number of animal models that have been studied that either have an incompletely formed epicardium or lack the epicardium entirely, and all these animals display defects in myocardial development (Gittenberger-de Groot et al., 2000, Pennisi et al., 2003, Yang et al., 1995, Moore et al., 1999). For example, in avian embryos, when epicardium outgrowth was inhibited, it resulted in the formation of thin compact myocardium containing disorganized cardiomyocyte arrangement(Gittenberger-de Groot et al., 2000). The compact myocardium developed normally in epicardium-rescue hearts (Gittenberger-de Groot et al., 2000), further corroborating a role for the epicardium in the formation of the myocardium. When epicardium formation was inhibited microsurgically by Pennisi and colleagues, again, the thickness of the compact myocardial zone was reduced, accompanied by a decrease in proliferation of cardiomyocytes (Pennisi et al., 2003).
In another study using WT1 null mice, where the epicardium either did not form, or was present only partially, the ventricular myocardial wall was thin, and the number of EPDCs was greatly reduced, or completely absent (Moore et al., 1999). This corresponds to a failure of the correct development of the myocardium, as these EPDCs give rise to resident vascular smooth muscle cells of the ventricle (Moore et al., 1999), among other cell types. In embryos deficient in integrin 4, an
39 epicardial cover overlying the myocardium was missing, and this was posited to lead to cardiac haemorrhage (Yang et al., 1995). Similarly, in the embryos of mice lacking Vascular Cell Adhesion Molecule-1 (VCAM-1), an epicardial sheath surrounding the myocardium was absent. Moreover, the myocardium itself showed aberrations such as reduced thickness and organization, when compared to control hearts (Kwee et al., 1995). Myocardial defects in observed in VCAM-1- deficient embryos and blood leakage from the heart in integrin 4-deficient embryos could be a secondary pathology, directly related to abnormal epicardium formation.
CRIM1 is important for the development of multiple organ systems including the kidney, eye and placenta. Interestingly, a study by Pennisi and colleagues in 2007 revealed that Crim1 is additionally expressed in the proepicardium as well as in the heart; in the epicardium, smooth muscle cells, the endothelium of coronary arteries and endocardial endothelial cells (Pennisi et al., 2007). While this implies a role for Crim1 in cardiac development, its precise role is unknown and will be evaluated in this chapter. This chapter will focus on the epicardium, and subsequent myocardial development, as well as the cell-autonomous and paracrine roles Crim1 may play in regulating the development of the myocardium.
Materials and Methods. Ethics Statement. Please refer to the Materials and General Methods chapter for the detailed ethics statement.
Animal Breeding. The mouse lines used in this study were the Crim1KST264 line (Pennisi et al., 2007), Crim1FLOXconditional knock-out (Chiu et al., 2012), the Crim1Δfloxline (Chiu et al., 2012), the ROSA26 Cre reporter line (R26R) (Soriano, 1999), the Mlc2v-Cre line (Chen et al., 1998)and the WT1-Cre and WT1-CreERT2 lines (Wilm et al., 2005, Zhou and Pu, 2012). Embryos were obtained from timed matings between Crim1+/Δfloxintercrosses, Crim1+/KST264 intercrosses, or Crim1FLOX/FLOX; R26R females mated with Crim1+/FLOX; WT1- Cre, Crim1+/FLOX; +/WT1-CreERT2 or Crim1+/FLOX; Mlc2v-Cre males. Timed matings were set up between Crim1+/flox; WT1-Cre males and Crim1FLOX/FLOX; R26R females to produce Crim1+/FLOX; WT1-Cre; R26R and Crim1FLOX/FLOX; WT1-Cre; R26R embryos collected at 15.5dpc (at this stage, the myocardium has expanded and cells derived from the epicardium have begun to differentiate into various lineages), Crim1+/FLOX; +/WT1-CreERT2 males and Crim1FLOX/FLOX; R26R females, for the production of Crim1+/FLOX; +/WT1-CreERT2; R26R and Crim1FLOX/FLOX; +/WT1-CreERT2;R26R embryos collected at 15.5dpc and 17.5dpc (at 17.5dpc, cells derived from the epicardium have
40 differentiated into cardiac fibroblasts, coronary vascular smooth muscle cells, and a small percentage of coronary vascular endothelial cells), and between Crim1+/FLOX; Mlc2v-Cre males and Crim1FLOX/FLOX; R26R females to produce Crim1+/FLOX; Mlc2v-Cre; R26R and Crim1FLOX/FLOX; Mlc2v-Cre; R26R embryos collected at 13.5dpc (at this stage, the myocardium has started to expand through proliferation of cardiomyocytes). The presence of a vaginal plug regarded as 0.5 dpc.
Tamoxifen Administration. Tamoxifen (MP Biomedicals, 02156738) was dissolved in corn oil and 100% ethanol at a concentration of 20mg/ml in a 65ºC incubator for 1 hour. The suspension was vortexed thoroughly. Tamoxifen was stored at -30°C until use. At 9.5dpc and 10.5dpc, prior to embryo harvestation, 2mg tamoxifen per dose (Chong et al., 2011) was injected intraperitoneally to pregnant dams.
Sample preparation. Embryonic hearts were dissected, photographed in whole-mount as required, and prepared by paraffin or cryoembedding and sectioning, for immunohistochemical analyses as detailed in the Materials and General Methods chapter.
Scanning Electron Microscopy. 13.5dpc embryos were collected from intercrosses of Crim1+/∆flox mice, and the hearts were dissected and fixed in 2.5% glutaraldehyde in PBS for 1 hour at room temperature. Samples were washed three times with PBS and stored at 4°C until processing for scanning electron microscopy as described (Ashe et al., 2012), using a CM-500 Benchtop Scanning Electron Neoscope.
Antibodies and Lectin. Primary antibodies/lectin used in the study were: rabbit polyclonal anti-phospho-Histone H3 (PHH3) (pSer10; Millipore #06-570); rabbit monoclonal anti-cleaved caspase 3 (CC3) (activated caspase 3; Cell Signalling Technology #9664); rabbit monoclonal anti-phospho-AKT (pSer473; D9E; Cell Signalling Technology #4060); rabbit monoclonal anti-phospho-ERK1/2 (pThr202/pTyr204; D13.14.4E; Cell Signalling Technology #4370); rabbit monoclonal anti- phospho-SMAD1/5 (pSer463/pSer465; Thermo Fisher Scientific #700047); rabbit polyclonal anti- phospho-SMAD2 (pSer465/pSer467; Millipore AB3849); mouse monoclonal anti-smooth muscle α-actin (clone 1A4, Santa Cruz Biotechnology, sc-3225); rabbit anti-β-catenin (whole antiserum; Sigma Aldrich #C2206); rabbit polyclonal SM22 alpha (Abcam Ab14106); rabbit polyclonal anti- periostin (Abcam Ab92460); mouse monoclonal anti-Wilms‟ Tumor 1 (WT1; Dako #M3561); MF20 (supernatant, Developmental Studies Hybridoma Bank) and Isolectin B4 (Thermo Fisher
41 Scientific). Secondary antibodies for immunofluorescence experiments were Alexa Fluor-488, −594 or −633-conjugated secondary antibodies (Thermo Fisher Scientific). TSA plus Cyanine 3 system was used for signal amplification in the case of pERK1/2 and pAKT (PerkinElmer #NEL744001KT) Secondary antibodies for immunohistochemistry experiments were anti-rabbit HRP (Promega; #W4011), anti-mouse HRP (Promega; #W4021), anti-rabbit biotin (Sigma; # B7389). The Vectastain Elite ABC kit (Standard) was also used (Vector Laboratories; #PK6100).
Immunohistochemistry. Immunohistochemical analyses were performed as outlined in the Materials and General Methods section. For experiments that required the use of mouse antibodies, the mouse-on-mouse kit (Vector MOM kit; BMK 2202) was utilised. Sections were blocked with blocking reagent for 1 hour, after which they were primed with MOM diluent for 5 minutes. Primary and secondary antibodies were diluted in MOM diluent. Immunohistochemistry using PHH3 was performed and sections counterstained with 1% DAB followed by NFR staining. Immunohistochemistry to detect periostin used anti-rabbit biotin and Vectastain ABC as secondary and tertiary antibodies, respectively. Vectastain ABC reagents were prepared as per manufacturer‟s protocols. It was added for 30 minutes at room temperature after incubation with anti-rabbit biotin, followed by hematoxylin staining.
Immunofluorescence. Heat-induced antigen retrieval with 10 mM citrate buffer (pH 6.0) was used on paraffin sections for detection of cleaved caspase 3 and phospho-histone H3. Tyramide amplification was used for signal amplification of phospho-ERK1/2 and phospho-AKT and performed according to manufacturer‟s protocol.
Primary Epicardial Explant Cultures and Migration Assays. 11.5 dpc embryos were collected from intercrosses between Crim1+/KST264 mice. The ventricles were dissected from the rest of the heart in sterile PBS and cultured in DMEM supplemented with 10% FCS and penicillin/streptomycin in four-well tissue culture plates. The ventricular mass was removed upon epicardial outgrowth from the primary culture. The epicardial cells were then grown to confluence and were passaged once and plated at a high confluence. The primary epicardial cells were used in a modified Boyden chamber assay after further culturing for two days. Epicardial cultures were collected after trypsin treatment to completely dissociate cells, washed and resuspended in DMEM (with 10% FCS). 1,000 cells in identical volumes were placed in the upper chamber of 6.5 mm transwell culture inserts (polycarbonate filter, 5.0 µm pore size, Corning). After
42 16 hours of culture, the cells in the upper part of the transwell inserts were removed using a cotton swab. The cells that had migrated through the polycarbonate filter were quantified after fixation in 4% PFA in PBS and staining with DAPI. Epicardial identity was confirmed after enrichment by immunofluorescence for the transcription factor WT-1 (Zhou and Pu, 2012, Rudat and Kispert, 2012), using standard techniques as described (Pennisi and Mikawa, 2009).
Cell culture and transfection. HEK293 cells were cultured in DMEM with 10% FBS and 0.1% Pen-Strep. Transfection was performed using Lipofectamine 2000 (Thermo Fisher Scientific) according to manufacturer‟s protocol.
Protein Fragment Complementation Assay. This assay consists of a „bait‟ protein, „prey‟ protein and, incomplete fragments of a third protein which acts as the reporter. On their own, reporter fragments are incomplete. Interaction between „bait‟ and „prey‟ proteins bring the reporter fragments in close proximity, which allows for the formation of a functional reporter protein whose activity can be measured (Remy and Michnick, 2003). When the Venus fluorescent protein is split into N-terminal (V1; amino acid residues 1-158) and C-terminal (V2; amino acid residues 159-239) amino acid fragments, neither displays fluorescence. However, when V1 and V2 are linked to 2 proteins that interact, it allows reformation of the Venus fragments and thus, fluorescence is generated (Porrello et al., 2011). Moreover, transient protein interactions can be identified, as once V1 and V2 recombine, they form a stable structure, thus capturing brief protein interactions.
Figure 3.1: Representation of Protein Fragment Complementation Assay (PCA) Schematic representation of the PCA which consists of a „bait‟ protein, „prey‟ protein and, incomplete fragments of a third reporter protein. Reporter fragments are incomplete on their own; interaction between „bait‟ and „prey‟ proteins
43 bring the reporter fragments in close proximity, allowing for the formation of a functional reporter protein whose activity can be measured (Remy and Michnick, 2003).
Leucine zipper transcription factors are well known to dimerize (Vinson et al., 2006). Zipper-V1 and Zipper-V2 constructs, were used to optimize the Protein Fragment Complementation Assay. These, and a full-length CRIM1 plasmid, were used to generate CRIM1-V1 and CRIM1-V2 constructs.
To generate CRIM1 containing NOT1 and CLA1 sites, primers were designed as follows: NOT1-CRIM1 FOR- 5‟-AGCGGCCGCACCATGTACTTGGT-3 CLA1-CRIM1 REV- 5‟-TTATCGATCACTGTTTGGTAGAAATTGTC-3‟
Phusion PCR was performed according to manufacturer‟s protocol, using full-length CRIM1 as a template. Cycling conditions were as follows: Initial denaturation: 98°C for 30 seconds 30 cycles of: Denaturation: 98°C for 10 seconds Annealing: 69°C for 25 seconds Extension: 72°C for 1 minute Final extension: 72°C for 7 minutes Cooling: 4°C
Zipper-V1 and Zipper-V2 were digested using Not1 and Cla1 restriction enzymes to release Zipper. All products were run on a gel to verify size, and then gel purified. A ligation using T4 DNA Ligase with the purified V1 (vector), V2 (vector), and CRIM1 containing Not1 and Cla1 sites, while maintaining a 3:1 insert:vector ratio, was performed at 4°C overnight. The ligation products were used to transform competent cells, with appropriate negative controls. Colonies were observed as expected, and 10 colonies from each plate were picked. A colony PCR was performed to confirm CRIM1-V1 and CRIM1-V2, using appropriate positive and negative controls. CRIM1-V1 and CRIM1-V2 were used to transfect HEK293 cells as described above.
Reverse Transcription, Quantitative Real-time PCR and Analysis. Ventricular samples from Crim1+/+ and Crim1Δflox/Δfloxhearts were miscrodissected, homogenized and frozen in RLT. Total ribonucleic acid (RNA) was extracted using an RNeasy Mini Kit (Qiagen, #74104). Reverse transcription was performed using SuperscriptIII (Thermo Fisher Scientific).
44 Briefly, 500ng total RNA was reverse-transcribed with random primers and deoxynucleotide triphosphates (dNTPs). cDNA was diluted 1/5 with RNase/DNase free water. qPCRs were carried out in a QuantStudio6 (Applied Biosystems) using SYBR green (Takara) and standard qPCR conditions. The QuantStudio6 software was used for analysis, with TFIID used as a relative standard. All samples were tested in triplicate. Relative transcript levels were assessed using the ΔCt method. Statistical analyses were performed using a two-tailed unpaired t-test.
The primer sequences used in this study were purchased from Sigma-Aldrich, Australia, and are detailed below:
Periostin:Forward 5‟-AAGCTGCGGCAAGACAAG-3‟ Periostin: Reverse 5‟-TCAAATCTGCAGCTTCAAGG-3‟ Collagen1a: Forward 5‟-AGACATGTTCAGCTTTGTGGAC-3‟ Collagen1a: Reverse 5‟-GCAGCTGACTTCAGGGATG-3‟ Collagen3a: Forward 5‟-ACGTAGATGAATTGGGATGCAG-3‟ Collagen3a: Reverse 5‟-GGGTTGGGGCAGTCTAGTG-3‟ Collagen5a: Forward 5‟-CTACATCCGTGCCCTGGT-3‟ Collagen5a: Reverse 5‟-CCAGCACCGTCTTCTGGTAG-3‟ Transgelin: Forward 5‟-CCTTCCAGTCCACAAACGAC-3‟ Transgelin: Reverse 5‟-CTAGGATGGACCCTTGTTGG-3‟ Alpha smooth muscle actin: Forward 5‟-ACTCTCTTCCAGCCATCTTTCA-3‟ Alpha smooth muscle actin: Reverse 5‟-ATAGGTGGTTTCGTGGATGC-3‟ Myosin heavy chain 11: Forward 5‟-ACGCCCTCAAGAGCAAACT-3‟ Myosin heavy chain 11: Reverse 5‟-CCCTTCTGGAAGGAACAATG-3‟ Calponin1: Forward 5‟-GAAGGTCAATGAGTCAACTCAGAA-3‟ Calponin1: Reverse 5‟-CCATACTTGGTAATGGCTTTGA-3‟ Myosin heavy chain 7: Forward 5‟-TGCTGAGGCCCAGAAACAAGTG-3‟ Myosin heavy chain 7: Reverse 5‟-CTGGATTTGAGTGTCCTTCAGCAG-3‟ Nkx2-5: Forward 5‟-ATTGACGTAGCCTGGTGTCTCG-3‟ Nkx2-5: Reverse 5‟-AGTGTGGAATCCGTCGAAAGTGC-3‟ Tie2: Forward 5‟-AGAGGCCGAACATTCCAAGTAGC-3‟ Tie2: Reverse 5‟-GGCCAAACAAACTGGCTTCTGC-3‟ Pdgfrα: Forward 5‟-TTGCACAGCCTCTGTTTGTTGC-3‟ Pdgfrα: Reverse 5‟-ATGGTGTCATGCAAGGCCCAAAG-3‟ Pdgfrβ: Forward 5‟-CAGTCCCGGCTACCCTATCT-3‟
45 Pdgfrβ: Reverse 5‟-GACCCAGAGTTTTCTCGGCA-3‟ Tcf21: Forward 5‟-CATTCACCCAGTCAACCTGA-3‟ Tcf21: Reverse 5‟-CCACTTCCTTCAGGTCATTCTC-3‟ TFIID: Forward 5‟-ACGGACAACTGCGTTGATTTT-3‟ TFIID: Reverse 5‟-ACTTAGCTGGGAAGCCCAAC-3‟
Data documentation and quantification. Bright-field and fluorescent images were captured as outlined in the Materials and General Methods chapter, and images were adjusted using Adobe Photoshop software.
For the analysis of 13.0 dpc compact myocardium thickness and cell density, sections were labelled with actin and Platelet Endothelial Cell Adhesion Molecule-1 (PECAM-1), and nuclei counter- stained with DAPI. Three-five non-consecutive sections for each sample were imaged at 40X field of view, on a laser scanning confocal microscope. The actin-positive area of the ventricle was considered as the compact myocardium. Olympus Fluoview or ImageJ software were used to determine the area from confocal images, and the number of DAPI-stained nuclei were quantified using ImageJ software to calculate cell density.
For the quantification of the pHH3+ and CC3+-cells at 13.0dpc, five-seven non-consecutive sections for each sample were imaged at 40X field of view, on a laser scanning confocal microscope. Cells were scored as epicardial or sub-epicardial if they were on the periphery of the section and were actin-negative.
For the analysis of epicardium-derived cells using the WT1-Cre line, 16 non-consecutive, coronal sections from each sample (X-Gal+; WT1-Cre) were quantified and the total number X-Gal+ cells in the ventricles was counted. For analysis using the WT1-CreERT2 line, two-four non-consecutive, coronal sections from each sample (X-Gal+; WT1-CreERT2) were quantified and the average of X- Gal+ cells in the mid-ventricular region per sample was measured. Cells were scored as epicardial, sub-epicardial if they were on the periphery of the section, or intramyocardial.
For quantification of intensity of phospho-SMAD levels, at least five sections for each sample were imaged using a laser scanning confocal microscope. For quantification of phospho-ERK1/2 and phospho-AKT levels, at least two sections for each sample were imaged. Signal intensity was determined using the Integration function of the Olympus Fluoview Software and the number of DAPI-positive nuclei in the demarcated area was quantified using ImageJ software.
46
For quantification of β-catenin, sections were co-immunolabelled for MF20, which recognizes the heavy chain of myosin II in cardiomyocytes, and 40x images were acquired using a laser scanning confocal microscope from at least two sections for each sample. Cells were considered to be epicardial or sub-epicardial if they were on the periphery of the section and were MF20-negative.
Statistical Analyses. To determine the statistical significance of the prevalence of phenotypes among mutant and control groups, a Z test was used. For the analysis of the location of epicardium-derived cells (X-Gal+, WT1-Cre; epicardial, sub-epicardial, or intramyocardial), a two-way ANOVA with Bonferroni‟s post-test was used. For all other analyses, statistical significance between pooled data from mutant and control groups was determined using a two-tailed, unpaired Student‟s t-test. Quantified data shown are mean ± standard deviation.
Results. Crim1 expression during heart development. A genetrap line for Crim1, called Crim1KST264(Pennisi et al., 2007),was used to assess Crim1 expression in the developing heart. The presence of a LacZ reporter working under the control of the Crim1 locus allowed us to observe Crim1-promoter driven reporter geneexpression at 9.5 dpc in the proepicardium and in the epicardium at all stages examined (12.5 dpc, 13.5 dpc, 15.5 dpc, 17.5 dpc and 18.5 dpc) (Figure 3.2A-J). As previously reported by Pennisi and colleagues, Crim1 expression was also seen in coronary vascular smooth muscle cells and coronary vascular endothelial cells (Pennisi et al., 2007)(Figure 3.2E-G, J). Smooth muscle cell expression was confirmed using immunohistochemical analyses against transgelin (Figure 3.3F). The expression of Crim1 in the outflow tract mesenchyme, endocardium and valve leaflets was confirmed using periostin double labelling (Figure 3.3A-E). This distinctive expression pattern indicates that Crim1 is likely involved in heart development.
47
Figure 3.2: Crim1 expression as monitored by the Crim1KST264 genetrap reporter reveals a dynamic pattern during heart development. A, micrograph of a histological section of a 9.5 dpc Crim1+/KST264 embryo that has been X-Gal-stained, paraffin- embedded, and counter-stained with nuclear fast red. B, higher magnification view of the boxed area in A. The proepicardial cells are X-Gal-positive (arrow, B). C–G, Whole-mount views of Crim1+/KST264 hearts after X-Gal staining (blue) at 12.5 dpc (C), 15.5 dpc (D), 17.5 dpc (E), and 18.5 dpc (F) Note Crim1 expression in the OFT mesenchyme (double arrows, C) and smooth muscle of the great vessels (arrows, D-F). G, Magnified left lateral view of the ventricle of the heart shown in F. Note the Crim1-LacZ expression in the developing coronary vasculature (arrows) and in the epicardium (double open arrowheads). H-J, histological sections Crim1-LacZ expression in the epicardium (arrows) and M, in the coronary vasculature (open arrowhead). my, myocardium; pe, proepicardium. Scale bars, A, 100 µm; B, 20 µm; C, 200 µm; D–F, 500 µm; G, 200 µm; H-J, 20 µm.
48
Figure 3.3: Crim1 is expressed in the outflow tract, mitral valve leaflets and coronary vascular smooth muscle cells. A, Periostin immunohistochemistry on an X-Gal stained Crim1+/KST264 transverse section at 12.5 dpc. B, higher magnification view of the boxed area in A, showing periostin- and X-Galpositive outflow tract mesenchymal cells (arrows). C, Periostin immunohistochemistry on an X-Gal stained Crim1+/KST264 coronal section at 13.5 dpc. D, higher magnification view of boxed area in C, showing periostin- and X-Gal-positive mesenchymal cells (arrows). E, 3 Periostin immunohistochemistry on an X-Gal stained Crim1+/KST264 coronal section at 18.5 dpc, showing periostin- and X-Gal-positive mitral valve leaflet cells (arrows). F, Transgelin immunohistochemistry on an X-Gal stained Crim1+/KST264 coronal section at 18.5 dpc, showing transgelin- and X-Gal-positive coronary vascular smooth muscle cells (arrows). Scale bars, A, C, 100 µm; B, D, E, 40 µm; F, 10 µm.
49 Defective heart development in embryos lacking Crim1. Crim1∆flox/∆floxdisplayed a reduced ventricular size at 14.5 dpc (Figure 3.4A, B, D, E), and also at 17.0 dpc accompanied by distended atria (Figure 3.4C, F). Analysis of the epicardium of Crim1∆flox/∆flox hearts at 13.5 dpc revealed abnormal epicardial morphology compared to wildtype controls (Figure 3.5A-H). This included epicardial blistering as visualised with histological analyses (Figure 3.5A-C), and irregular appearance along with loss of squamous morphology as observed performing electron scanning microscopy on Crim1∆flox/∆flox hearts (Figure 3.5 D-G).
Figure 3.4: Heart defects in Crim1Δflox/Δfloxembryos. A–B, D-E Representative whole-mount views of 14.5 dpc embryonic hearts (A, B, Crim1+/+; D, E, Crim1Δflox/Δflox). B, E, Magnified, left lateral views of A and D, respectively. C,F, Representative whole-mount views of 17 dpc embryonic hearts (C, Crim1+/+; F, Crim1Δflox/Δflox). Note the reduced size of the ventricles in the Crim1Δflox/Δfloxheart (D, E, F). Scale bars, A–F, 500 µm.
50
Figure 3.5: Epicardial defects in hearts of Crim1Δflox/Δfloxembryos. A-C, Hematoxylin and eosin-stained sections of the ventricular region of 13.5 dpc Crim1+/+ (A) and two different Crim1Δflox/Δflox(B, C) hearts. Note the irregular appearance of the epicardium in Crim1Δflox/Δfloxembryos, with one example showing blebbing (B). D-G, Scanning electron micrographs of the ventricular surface of 13.5 dpc Crim1+/+ (D, F) and Crim1Δflox/Δflox(E, G) hearts. E, Note the loss of the even spacing of cells of the mesothelial epicardium, and G, diffusely arranged microvilli. H, Quantification of penetrance of epicardial defects between Crim1+/+ and Crim1Δflox/Δfloxhearts at
51 13.5 dpc. Z-score -3.8801. *, P<0.0001. cm, compact myocardium; ep, epicardium. Scale bars, A-C, 50 µm; D, E, 20 µm; F, G, 2 µm.
Crim1 is dispensable for PE formation. To address whether these anomalies were due to changes in formation of the PE, we used in situ hybridisation for PE markers Gata4 and Tbx18 on hearts of Crim1∆flox/∆flox embryos. There were no gross changes in marker gene expression between mutant and wildtype control embryos (Figure 3.6A-F), suggesting that while not required for early PE specification, Crim1 may regulate epicardial migration or adhesion at a later time point during development.
Figure 3.6: Crim1 is dispensible for specification or early development of the proepicardium. A-F, Micrographs of 9.5 dpc embryos from Crim1+/+(A, C, E) and Crim1Δflox/Δflox(B, D, F) embryos stained for Gata4 mRNA expression (purple stain, A and B) and Tbx18 mRNA expression (purple stain, C and D). E, F, Micrographs of histological sections of wholemount in situ hybridization-stained embryos showing Tbx18 expression. The PE is indicated (arrows). Scale bars, A and E, 50 µm.
52 Crim1 regulates compact myocardial development. A decrease in the size of Crim1∆flox/∆flox ventricles at 14.5 dpc prompted us to investigate changes in thickness of the compact myocardium. At 13.0 dpc, when myocardial development has started, we observed a reduction in compact myocardial thickness in Crim1∆flox/∆flox hearts (Figure 3.7A, B, D). However, this reduction was not due to a decrease in cell density (Figure 3.7C). It is noteworthy that different mutants lacking either the epicardium itself, or epicardium-derived factors, share a similar phenotype (Yang et al., 1995, Moore et al., 1999, Kwee et al., 1995, Chen et al., 2002, Pennisi et al., 2003, Männer, 1993, Gittenberger-de Groot et al., 2000). We went on to analyse whether there were any changes in proliferation or apoptosis of epicardial and myocardial cells, using markers PHH3 and CC3 respectively (Figure 3.7E-L). While epicardial cell proliferation and apoptosis was unchanged between conditional nulls and controls, there was an almost two-fold increase in proliferation and a four-fold increase in apoptosis of compact myocardial cells at 13.0 dpc (Figure 3.7E-L). This indicates that loss of Crim1 results in increased cell death in the compact myocardium, but it is not known whether this is due to compact myocardial Crim1 acting in an autocrine way or a paracrine effect of epicardial Crim1.
Figure 3.7: Crim1 is necessary for normal compact myocardial development. A and B, Merged confocal images of left ventricular sections of 13.0 dpc hearts from Crim1+/+ (A) and Crim1Δflox/Δflox (B) hearts stained with DAPI (blue), PECAM-1 (red), and actin (green). C, Compact myocardium cell density was unchanged in Crim1Δflox/Δfloxhearts. D, Reduced ventricular compact myocardial thickness in Crim1Δflox/Δfloxhearts at 13.0 dpc (n=6–8). E and F, Merged confocal images of ventricular sections of 13.0 dpc hearts stained with Phospho-Histone H3 (PHH3, white), DAPI (blue), and actin (green). An epicardial cell (arrowhead, E) and myocardial cell (arrowhead, 53 F) immuno-positive for PHH3 are shown. G, No difference in proliferation in the epicardium of Crim1Δflox/Δfloxhearts.H, An almost two-fold increase in proliferation in the compact myocardium of Crim1Δflox/Δfloxhearts.I and J, Examples of merged confocal images of ventricular sections of 13.0 dpc hearts stained with cleaved caspase 3 (CC3, white), DAPI (blue), and actin (green). An epicardial cell (arrowhead, I) and myocardial cell (arrowhead, J) immuno-positive for CC3 are shown. K, No difference in apoptosis in the epicardium of Crim1Δflox/Δflox hearts.H, A four-fold increase in apoptosis in the compact myocardium of Crim1Δflox/Δfloxhearts.cm, compact myocardium; ep, epicardium; n.s., not significant. (n=4–5), *, P<0.05; **, P<0.01. Scale bars A, B, 50 μm; E, F, I, J, 20 μm.
Effect of Crim1 deletion on cardiomyocytes within the myocardium. We first used the Mlc2v-Cre line for the cardiomyocyte-specific deletion of Crim1 from the compact myocardium, so as to assess the posited autocrine role of Crim1. This line has been previously used and has been shown to be both efficient and specific to cardiomyocytes of the ventricle (Chen et al., 1998, O'Brien et al., 1993, Minamisawa et al., 1999). We crossed Crim1FLOX/FLOX; R26R mice with the Crim1+/FLOX; Mlc2v-Cre line to generate Crim1+/FLOX; R26R; Mlc2v-Cre controls and Crim1FLOX/FLOX; R26R; Mlc2v-Cre mutant mice. Embryonic hearts were collected at 13.5dpc and stained for X-Gal. We assessed proliferation and apoptosis of cardiomyocytes at 13.5 dpc using PHH3 and CC3 but found no significant changes between conditional null and control hearts (Figure 3.8A-H). Additionally, there were no differences in the thickness of the compact myocardium in these hearts at this age (Figure 3.8I). It is important to note that the phenotype demonstrating a reduction in compact myocardium thickness is observed before epicardial EMT and invasion of the myocardium occurs, and thus, we can infer that while myocardial depletion of Crim1 may not be responsible for the observed defect, epicardial Crim1 could regulate myocardial development in a paracrine manner.
54
Figure 3.8: Myocardial deletion of Crim1 does not affect cardiomyocyte proliferation and survival and thickness of compact myocardium. A, D representative whole mount images of X-Gal stained Crim1+/FLOX; Mlc2v- Cre; R26R and Crim1FLOX/FLOX; Mlc2v- Cre; R26R hearts at 13.5 dpc. B, E merged confocal images of DAPI (blue) and pHH3 (red), with a pHH3-positive nucleus (arrow). C, F merged confocal images of DAPI (blue) and CC3 (red), with a CC3-positive nucleus (arrow). G, No significant difference in percentage of pHH3-positive cells in the compact myocardium of left and right ventricles in control and mutant hearts (n=4-6). H, No significant difference in percentage of CC3-positive cells in the compact myocardium of left and right ventricles in control and mutant hearts (n=4-5). I, No significant difference in thickness of compact myocardium in the left ventricle of control and mutant hearts (n=3).cm, compact myocardium; ep, epicardium; n.s, not significant. Scale bars, B, C, E, F 20 µm.
Effect of epicardium-specific deletion of Crim1 on the compact myocardium. To further assess the role of epicardial Crim1 in the development of the myocardium, we used the Crim1FLOX line and an inducible transgenic line for the epicardium-specific deletion of Crim1, the WT1-CreERT2 line,which has been shown to be restricted to the epicardium and epicardium- derived lineages, (Zhou and Pu, 2012). We crossed Crim1FLOX/FLOX; R26R mice with the Crim1+/FLOX; WT1-CreERT2 line to generate Crim1+/FLOX; R26R; +/WT1-CreERT2 controls and Crim1FLOX/FLOX; R26R; +/WT1-CreERT2 mutant mice. Two doses of Tamoxifen were administered prior to embryo collection, at 9.5 dpc and 10.5 dpc, and the LacZ reporter was used to confirm that Cre activity in embryonic hearts from 12.5 dpc was restricted to the epicardium, upon X-Gal 55 staining (data not shown). Its specificity was confirmed using immunohistochemical analyses against the cardiomyocyte marker MF20, which showed limited overlap between X-Gal positive cells that had lost Crim1 function, and cardiomyocytes (Figure 3.9A, B). However, decreased thickness of the myocardial wall was not observed in Crim1FLOX/FLOX; R26R; +/WT1- CreERT2hearts at 15.5 dpc (data not shown). We hypothesise that this is due to incomplete penetrance and variable expressivity of the WT1-CreERT2 line, as previously reported(Rudat and Kispert, 2012). This occurrence was confirmed upon quantifying the percentage of epicardial cells lacking Crim1 in the left and right ventricles (only 37.5% of cells had Crim1 ablated), indicating that the majority of epicardial cells stillexpressed Crim1 in this line (pooled genotypes, n=6). This could potentially compensate for any myocardial defects that could have been caused due to a complete loss of epicardial Crim1.
Figure 3.9: Fidelity of the WT1-CreERT2 line. A,B MF20 immunohistochemistry performed on 13.5 dpc WT1-CreERT2 X-Gal stained sections showing MF20- negative epicardium (arrows) and MF20-negative EPDCs (open arrowheads), A, WT1-CreERT2 13.5 dpc X-gal stained section showing that most epicardial cells are X-Gal-positive (arrow). B, WT1-CreERT2 17.5 dpc X-Gal stained section showing MF20-negative EPDCs (open arrowheads), confirming that the EPDCs are not cardiomyocytes. ep; epicardium. Scale bar, A, B, 50 µm.
Cell-autonomous role of Crim1 in epicardial migration in vitro. The epicardial defects observed in the conditional null embryos, as well as the potential role of epicardial Crim1 in modulating myocardial development, prompted us to investigate the role of Crim1 in the development of the epicardium. First, we cultured primary embryonic epicardial cells
56 isolated from the Crim1KST264 line and used these in a transwell migration assay, a modified Boyden chamber assay. As mentioned previously, the Crim1KST264 allele allowed us to monitor Crim1 expression using X-Gal staining in the representative samples of explant cultures, which revealed ubiquitous Crim1-LacZ expression in primary epicardial cells (data not shown). Furthermore, WT1 labelling of the primary epicardial cultures from Crim1KST264/KST264 hearts confirmed epicardial identity after enrichment (Figure 3.10A-C). These cells exhibited an increase in the rate of migration compared to heterozygote controls, indicating that the loss of Crim1 from the epicardium resulted in an increase in epicardial migration (Figure 3.10D-E).
Figure 3.10: Crim1 plays a cell-autonomous role in epicardial migration, in vitro. D, Schematic of the transwell assay used for the in vitro migration assay. A, Representative micrograph of a primary 11.5 dpc ventricular explant after attachment and migration of epicardial cells. The front of the epicardial monolayer is indicated (arrows). The primary explant is denoted by an asterisk. B, Primary epicardial cells after removal of the ventricular explant, passage and subsequent culture. Note maintenance of a “cobblestone” appearance of the cellular monolayer. C, Epicardial identity after enrichment was confirmed by immunostaining for WT1 (green nuclear signal). The no-primary antibody control showed no signal (inset). D, Schematic of the transwell assay used for the in vitro migration assay. E, Crim1KST264/KST264primary epicardial cells showed increased migration in vitro relative to controls (n=3–5 and representative of three independent experiments) *, P<0.05.Scale bar, A, 100 µm; B, C, 50 µm.
57 Cell-autonomous role of Crim1 in epicardial migration and myocardial invasion,in vivo. To confirm our observations in vivo, and also to assess whether Crim1 regulated epicardial migration, we used transgenic lines for the epicardium-specific deletion of Crim1, namely Crim1FLOX, WT1-Cre(Wilm et al., 2005) and WT1-CreERT2(Zhou and Pu, 2012). We first crossed Crim1+/FLOX; R26R mice with the Crim1+/FLOX; WT1-Cre line to generate Crim1+/FLOX; R26R; WT1- Cre controls and Crim1FLOX/FLOX; R26R; WT1-Cre mutant mice with Crim1 deleted from epicardial cells and their derivatives. The fidelity of this line was also confirmed by MF20 immunohistochemistry (Figure 3.11A-C).
Figure 3.11: Fidelity of the WT1-Cre line. A-C MF20 immunohistochemistry performed on 15.5 dpc WT1-Cre X-Gal stained sections showing MF20-negative epicardium (arrows) and MF20-negative EPDCs (open arrowheads), ep; epicardium. Scale bar, A-C, 25 µm.
Epicardium-restricted Cre activity was assessed by the R26R LacZ reporter. Embryonic hearts were collected at 15.5dpc and X-Gal stained. We assessed the location of X-Gal-positive cells at 15.5 dpc in Crim1FLOX/FLOX; R26R; WT1-Cre and littermate controls, determining whether tagged cells remained epicardial, or had undergone epicardial EMT and migrated to the sub-epicardium or to an intramural myocardial location. Section data showed a distinct staining of epicardial cells, and epicardium-derived cells in the subepicardium and myocardium. Quantification of X-Gal-positive cells in the right ventricle of Crim1+/FLOX; R26R; WT1-Cre hearts revealed a greater number of epicardial cells and a lower number of intramyocardial cells compared to their Crim1FLOX/FLOX; R26R; WT1-Cre littermates (Figure 3.12A-J). There was a significant decrease in the proportion of X-Gal-positive epicardial cells, and an associated increase in intramyocardial X-Gal-positive cells in the right ventricle of Crim1FLOX/FLOX; R26R; WT1-Cre hearts compared to controls. However, it is not known why this phenotype was observed only in one ventricle.Thus, this suggests that with a loss of Crim1 function, there is an increase in the number of epicardial cells that have undergone EMT and invaded the myocardium (Figure 3.12I, J).
58
Figure 3.12: Crim1 plays a cell-autonomous role in epicardial migration and invasion of the myocardium, in vivo, as observedby using the WT1-Cre line. A, E, Whole-mount dorsal views of 15.5dpc embryonic hearts after X-Gal-staining (blue) of Crim1+/FLOX; R26R; WT1- Cre and Crim1FLOX/FLOX; R26R; WT1-Cre genotype, respectively. B, F, Higher magnification views of the boxed areas in F and K, respectively. C-H Histological sections of hearts from Crim1+/FLOX; R26R; WT1-Cre (C, D) and Crim1FLOX/FLOX; R26R; WT1-Cre (G, H) 15.5dpc embryos, showing X-Gal+ epicardial cells (arrows), subepicardial cells (double arrowheads), and intramyocardial cells (open arrowheads). I, J, The proportion of X-Gal+ cells that are epicardial, sub-epicardial, or intramyocardial in the left (I) and right (J) ventricles. Note the decrease in epicardial X- Gal+ cells, and increase of intramyocardial X-Gal+ cells in the right ventricles of Crim1FLOX/FLOX; R26R; WT1-Cre hearts, indicating increased epicardial EMT and invasion of the myocardium by epicardial cells with loss of Crim1 function (n=7–8; *, P<0.01). Ep, epicardial; S.E., subepicardial; My, intramyocardial; LV, left ventricle; RV, right ventricle. Scale bars, 20 µm.
It is important to note, however, that the Cre was active only in a very small minority of epicardial cells and EPDCs. Although the WT1-Cre line allowed us to examine a cell-autonomous role for Crim1, a drawback of this line is that complete activation of Cre is not achieved in all epicardial cells (unpublished data, (Rudat and Kispert, 2012)), thus requiring an alternate Cre line that is specific to the epicardium and its derivatives, to examine the effect of a more global epicardial loss of Crim1 function.
59 Therefore, the WT1-CreERT2 line was used to circumvent this problem (Zhou and Pu, 2012). Again, we crossed Crim1FLOX/FLOX; R26R mice with the Crim1+/FLOX; WT1-CreERT2 line to generate Crim1+/FLOX; R26R; +/WT1-CreERT2 controls and Crim1FLOX/FLOX; R26R; +/WT1- CreERT2 mutant mice. Upon X-Gal staining, we observed that like in the case of WT1-Cre mice, absence of Crim1 in the epicardium resulted in increased epicardial EMT and invasion of the myocardium of right ventricles compared to controls (Figure 3.13A-H). These in vivo data are consistent with primary epicardial culture migration studies conducted, and suggest that in the absence of Crim1, epicardial EMT and migration of EPDCs into the myocardium is increased.
Figure 3.13: Crim1 plays a cell-autonomous role in epicardial migration and invasion of the myocardium, in vivo,as observedby using the WT1-CreERT2 line. A, B, Whole-mount dorsal views of 15.5 dpc embryonic hearts after X-Gal staining (blue) of Crim1+/FLOX; WT1- CreERT2; R26R and Crim1FLOX/FLOX; WT1-CreERT2; R26R genotype, respectively. Two doses of Tamoxifen were
60 administered at 9.5 dpc and 10.5 dpc prior to harvesting the embryos. C-D, E-F, Histological sections of hearts from Crim1+/FLOX; WT1-CreERT2; R26R and Crim1FLOX/FLOX; WT1-CreERT2; R26R 15.5 dpc embryos, respectively, and D, F, magnified views of boxed regions showing X-Gal-positive epicardial, sub-epicardial and intramural cells. G, H, Quantification of X-Gal-positive cells showing increased myocardial EPDCs in the right ventricles of mutant hearts (n=4-6; *P<0.05). ep, epicardial; se, sub-epicardial; my, intramyocardial; LV, left ventricle; RV, right ventricle. Scale bars, C, E, 20 μm; D, F, 10 μm.
Crim1 may alter epicardial adhesion junctions. The observed epicardial defects, namely epicardial blistering and a non-uniform epicardium, and the increase in epicardial EMT, may be suggestive of defects in cell adhesion. β-catenin is a crucial component of intercellular adhesion complexes, and has been shown to interact indirectly with Crim1 (Ponferrada et al., 2012). Therefore, we assessed localisation of β-catenin (Wu et al., 2010) in Crim1+/+ wildtype and Crim1∆flox/∆flox mutant hearts at 13.5 dpc. We found that there was an increase in the percentage of epicardial cells that displayed a concentrated accumulation of β- catenin at epicardial cell-cell junctions in Crim1+/+ hearts (Figure 3.14A-G), suggesting that Crim1 may play a role in stabilising intercellular junctions. We also looked for changes in f-actin in Crim1∆flox/∆flox hearts, but f-actin morphology in the myocardium appeared comparable between mutant and wildtype (Figure 3.15A-F).
Figure 3.14: Changes in β-catenin accumulation at epicardial cell junctions in the absence of Crim1. A-F, Confocal images of ventricular sections of 13.5 dpc hearts from Crim1+/+ (A, B, C) and Crim1Δflox/Δflox(D, E, F) embryos stained for β-catenin (red).A, B, merged images of DAPI (blue) and MF20 (green) to delineate the epicardium with the β-catenin (red) from B, C and E, F respectively. C, F, magnified views of boxed regions in B and E. Note β- catenin distribution at epicardial cell-cell junctions of Crim1+/+ hearts (arrows, C, F). G, the percentage of epicardial cells with an accumulation of β-catenin at cell junctions (n=5-6). cm, compact myocardium; ep, epicardium; s.ep, sub epicardium.n.s., not significant. *, P<0.05. Scale bars, A,D, 10 μm.
61
Figure 3.15: No visible change in filamentous actin morphology in the epicardium or myocardium of Crim1Δflox/Δflox hearts at 13.5 dpc. A-F, Representative confocal images of ventricular sections of 13.5 dpc hearts from Crim1+/+(A, B, C) and Crim1Δflox/Δflox (D, E, F) hearts at 13.5 dpc.A, D merged images of DAPI (blue) and MF20 (green) to delineate the myocardium with Phalloidin (red) from B and E, respectively (n=4-5). C, F, magnified views of boxed regions in B and E respectively. Scale bars A – F, 20 µm.
CRIM1 can bind to other CRIM1 molecules, and further confer junctional stability. CRIM1 has been shown, via immunoprecipitation, to self-associate and form complexes where multiple CRIM1 molecules are present (Wilkinson et al., 2003, Ponferrada et al., 2012). We further examined this using a Protein Fragment Complementation Assay. HEK293 cells were transfected with Zipper-V1 only, Zipper-V2 only, AT1R(Angiotensin II Type 1 Receptor)-V1 only and β- Arrestin-V2 only, and, both Zipper-V1 + Zipper-V2 and AT1R-V1 + β-Arrestin-V2. Only cells with both constructs displayed strong fluorescence (Figure 3.16G-I), indicating stable complementation and function of V1 and V2 fragments.
62
Figure 3.16: Protein Complementation Assay as tested with AT1-V1 and β-Arrestin-V2 A-C, HEK293 cells transfected with AT1-V1 (no fluorescence). D-F HEK293 cells transfected with β-Arrestin-V2 (no fluorescence). G-I HEK293 cells transfected with both AT1-V1 and β-Arrestin-V2 (fluorescence observed). A-F images were captured using a 10x objective, zoomed 4 times. G-I images used a 10x objective. Scale bars A-F, 50 μm; G-1, 160 μm.
We then generated CRIM1-V1 and CRIM1-V2 clones and used these to transfect HEK293 cells. Fluorescence was observed, confirming the interaction (Figure 3.17A-F). It is known from other studies that CRIM1 localizes in the Golgi and ER, as well as on the plasma membrane (Glienke, Sturz et al. 2002, Wilkinson, Kolle et al. 2003). To confirm the cellular localization of CRIM1, immunofluorescence was performed on the transfected samples using anti-Myc and anti-CRIM1 antibodies, and compared with the control N-terminal Myc-tagged CRIM1-Green Fluorescent Protein (GFP) clone. The constructs CRIM1-GFP, CRIM1-V1 and CRIM1-V2, both when co- transfected and transfected individually, have a similar localization pattern compared to the control (Figure 3.18A-E, Figure 3.19A-C). In another experiment, live CHO cells were transfected with N- terminal Myc-tagged CRIM1-GFP. Since the GFP is cytoplasmic and the Myc tag at the N- 63 terminus; a proportion of CRIM1 molecules are at the cell surface, and the Myc-tag will be extracellular. As the experiment was performed on chilled, live cells, the plasma membrane is impermeable to antibodies; therefore only the cell surface, extracellular Myc tag is detected. While no double staining was performed to confirm the localization of CRIM1, the staining pattern visualized by GFP expression indicates that it is expressed in these compartments. This implies that if CRIM1 molecules are present at epicardial junctions, theymay confer stability at these junctions via associating with other CRIM1 moleculeson adjacent epicardial cells (Figure 3.20).
Figure 3.17: Protein Complementation Assay as tested with CRIM1-V1 and CRIM1-V2 A, B HEK293 cells transfected with CRIM1-V1 (no fluorescence). C, D HEK293 cells transfected with CRIM1-V2 (no fluorescence). E, F HEK293 cells transfected with both CRIM1-V1 and CRIM1-V2 (fluorescence observed). Scale bars A-F, 50 μm.
64
Figure 3.18: Cellular localization of CRIM1 A-T, Cellular localization of Crim1. A-D, E-H, No fluorescence observed when CRIM1-V1 and CRIM1-V2 are transfected individually. I-P, Fluorescence observed when CRIM1-V1 and CRIM1-V2 are co-transfected, and also stained for α-Myc and α-CRIM1 antibodies. Observe localization of CRIM1 using the α-CRIM1 and α-Myc antibodies compared to control, across panels J, N, R. Q-T, HEK293 cells transfected with CRIM1-GFPtagged control. Scale bars A-T, 50 μm.
65
Figure 3.19: Magnified images showing CRIM1 localization A-C, Confocal images showing that CRIM1 is localized to the endoplasmic reticulum (ER) and golgi. A, B, magnified images of PCA with CRIM1-V1 and CRIM1-V2 showing CRIM1 localization. C, Immunofluorescence performed on live, chilled CHO cells under non-permeabilising conditions. Cells were transfected with N-terminal Myc-tagged CRIM1-GFPand subjected to immunofluorescence to the Myc epitope (green, GFP; more prominent in golgi and ER, red, Myc tag; extracellular), indicating CRIM1 localization in the ER and golgi. Scale bars A,B, 50 μm; C, 4 μm.
66
Figure 3.20: CRIM1 may confer stability at epicardial cell junctions Schematic representation of epicardial cell junctions with adhesion and bridging molecules along with CRIM1, which may serve as a stabilization mechanism at these junctions and restrain epicardial EMT.
Modulation of growth factor activity by Crim1. To assess a role for Crim1 in regulating the activity of growth factors, indirect immunofluorescence for phospho-SMAD2 and phospho-SMAD1/5 to assess canonical TGFβ and BMP signalling, respectively, was performed. At 13.0dpc, Crim1∆flox/∆flox hearts showed comparable levels of ventricular compact mycoardial phospho-SMAD2 relative to controls (Figure 3.21A, B, E-H). However, there was a reduced level of phospho-SMAD2 in the epicardium of Crim1∆flox/∆flox hearts at 13.0dpc (Figure 3.21A-D, E, F). At this stage, there was no change in the levels of phospho- SMAD1/5 in the ventricular compact myocardium or the epicardium of Crim1∆flox/∆flox hearts relative to controls, suggesting canonical BMP signalling was unchanged in Crim1∆flox/∆flox hearts at 13.0 dpc (Figure 3.22A-H). Despite the increased epicardial EMT and invasion resulting from the epicardial-restricted loss of Crim1 function, these results suggested that canonical TGFβ signalling was reduced in the epicardium of Crim1∆flox/∆flox hearts.
67
Figure 3.21: Reduction in canonical TGFβ signalling in the absence of Crim1 A-F, Confocal images of ventricular sections of 13.0 dpc hearts from Crim1+/Δflox (A, B) and Crim1Δflox/Δflox(E, F) embryos stained for phospho-SMAD2/3 (white).A, B, merged images of DAPI (blue) and actin (green) to delineate the myocardium with the phospho-SMAD2/3 (white) from B and F, respectively. C, D, G, and H, There was a reduction in phospho-SMAD2 signal intensity in epicardial cells of left and right ventricles of Crim1Δflox/Δfloxhearts, but the levels in compact myocardial cells was unchanged (n=4–5). cm, compact myocardium; ep, epicardium; s.ep, subepicardium.n.s., not significant. *, P<0.05. Scale bars, A-B, E-F, 10 μm.
Figure 3.22: No change in canonical BMP signalling in the absence of Crim1 A-F, Confocal images of ventricular sections of 13.0 dpc hearts from Crim1+/Δflox (A, B) and Crim1Δflox/Δflox (E, F) embryos stained for phospho-SMAD1/5 (white). A and E, merged images of DAPI (blue) and actin (green) to delineate the myocardium with the phosphoSMAD1/5 (white) from A and E, respectively. C, D, G, H, Quantification of the average signal intensity/cell after indirect immunofluorescence to detect phospho-SMAD1/5 showed there was no change in the levels in the epicardium or compact myocardium of left and right ventricles of Crim1Δflox/Δflox hearts in comparison to controls (n=5). cm, compact myocardium; ep, epicardium; n.s., not significant. Scale bars, A, B, E, F 10 µm.
We also assessed whether there were changes in the effectors phospho-AKT and phospho-ERK1/2 that are known to act downstream of a number of growth factors as another representation of
68 misregulated signalling. Levels of phospho-AKT in the epicardium and myocardium were comparable (Figure 3.23A-H) between mutants and controls, as were the levels of epicardial phospho-ERK1/2 (Figure 3.24A, B, E-H). However, we observed that there was an increase in levels of phospho-ERK1/2 in the ventricular compact myocardium of Crim1∆flox/∆flox hearts relative to controls (Figure 3.24A-D, E, F). This indicates that Crim1 does not play a role in the regulation of growth factors within the epicardium, but that it could serve a paracrine function in modulating the development of the myocardium by controlling the activity of epicardium-derived molecules. However, further analysis needs to be performed by using cardiac tissue homogenates at relevant embryonic ages in Western blot experiments, which will helps us quantify and confirm the growth factor signalling changes observed upon immuno-staining.
Figure 3.23: No change in phospho-AKT levels in the absence of Crim1 A-D, Confocal images of ventricular sections of 13.5 dpc hearts from Crim1+/+ (A, B) and Crim1Δflox/Δflox (C, D) embryos stained for phospho-AKT (red). Quantification of signal intensity per cell to detect phospho-AKT levels revealed no change in the levels in the epicardium or compact myocardium of left and right ventricles of Crim1Δflox/Δflox hearts in comparison to controls (n=5-6). cm, compact myocardium; ep, epicardium; n.s., not significant. Scale bars, A, B, C, D, 50 µm.
69
Figure 3.24: Increase in myocardial phospho-ERK1/2 levels in the absence of Crim1 A-F, Confocal images of ventricular sections of 13.5 dpc hearts from Crim1+/+ (A, B) and Crim1Δflox/Δflox(E, F) embryos stained for phospho-ERK1/2 (red). C, D, G and H, There was an increase in phospho-ERK1/2 signal intensity in compact myocardial cells of left and right ventricles of Crim1Δflox/Δfloxhearts, but the levels in epicardial cells was unchanged (n=8-6). cm, compact myocardium; ep, epicardium; n.s., not significant. *, P<0.05. Scale bars, A-B, E-F, 50 μm.
Loss of Crim1 results in decreased proliferation of intramyocardial cells. Proliferation of EPDCs was assessed in control Crim1+/FLOX; +/Wt1-CreERT2; R26R and mutant Crim1FLOX/FLOX; +/Wt1CreERT2; R26R heart sections at 15.5 dpc using pHH3 (Figure 3.25A, B). Quantification of pHH3-positive cells that were X-Gal-positive in the myocardium of left and right ventricles revealed a significant reduction in proliferation of intramyocardial cells in the mutant (Figure 3.25A, B, D). Quantification of pHH3-positive cells among X-Gal-positive epicardial cells, however, revealed comparable results (Figure 3.25A-C). This demonstrates that deleting Crim1 from the epicardium can perturb proliferation of EPDCs, and confirms the cell-autonomous role of Crim1.
70
Figure 3.25: Decrease in the number of proliferating EPDCs in the absence of epicardial Crim1 A, B, Phospho-Histone H3 immunohistochemistry on X-Gal stained Crim1+/FLOX; + /WT1-CreERT2; R26R and Crim1FLOX/FLOX; + /WT1-CreERT2; R26R sections at 15.5 dpc, counter stained with 1% DAB followed by NFR staining for visualisation of nuclei, showing (A) pHH3- and X-Gal-positive nuclei (arrow) and (B), a PHH3-positive nucleus (open arrowhead). C, D Quantification of proliferating EPDCs showing no change in the epicardium, but a significant reduction in the number of pHH3-positive EPDCs in mutant hearts compared to control in the myocardium (n = 6–4). *P < 0.05. n.s., not significant; LV, left ventricle. Scale bars, A,B 50 μm.
Crim1 may control fate specification of EPDCs, particularly into the fibroblast lineage. Having observed a reduction in proliferation of EPDCs residing in the myocardium, we next investigated whether there were any changes in epicardial cell fate into vascular smooth muscle, cardiomyocyte or fibroblast lineages. We performed immunofluorescence experiments on Crim1∆flox/∆flox and Crim1+/+ hearts at 17.5 dpc using periostin and transgelin as markers but found no observable changes between the two genotypes (Figure 3.26A-H). We next performed qPCR analyses on ventricular tissue extracted from Crim1∆flox/∆flox and Crim1+/+ hearts at 17.5 dpc, using markers for these cell types. There were no significant changes observed, although a trend for the fibroblast marker Collagen1a reduction in Crim1∆flox/∆flox samples was observed (Figure3.26 I, Trend P = 0.0556).
71
Figure 3.26: Assessing expression levels of different cell lineage markersin the absence of Crim1 A-I, No change in cardiac fibroblasts, coronary vascular smooth muscle cells or cardiomyocytes in Crim1Δflox/Δflox hearts at 17.5dpc. A–H, Representative confocal images of whole-heart sections (A, E, C, G) and magnified views of the myocardium (B, F, D, H) of Crim1+/+ (A-D) and Crim1Δflox/Δflox (E-H) heart stained with DAPI (blue), and Periostin (red) (A-B, E-F) and Transgelin (red) (C-D, G-H). I, qPCR with fibroblast, smooth muscle and cardiomyocyte markers showing no significant difference in fold change between Crim1+/+ and Crim1Δflox/Δflox ventricular samples at 17.5 dpc (n=4-5).
To further analyse this, we performed immunohistochemical analyses on control Crim1+/FLOX; +/Wt1-CreERT2; R26R and mutant Crim1FLOX/FLOX; +/Wt1CreERT2; R26R heart sections at 17.5 dpc, using periostin, a fibroblast marker (Figure 3.27A-C). We observed a reduction in periostin expression upon quantification of X-Gal- and periostin-positive cells in the myocardium of mutant hearts, which indicates that the number of fibroblasts is decreased in the absence of epicardial
72 Crim1 (Figure 3.27A-E). This implies a potential role for Crim1 in determining the fate of EPDCs, specifically committed to the fibroblast lineage.
Figure 3.27: Decrease in the number of periostin-positive EPDCs in the absence of epicardial Crim1 A-C, Periostin immunohistochemistry on X-Gal stained Crim1+/FLOX; + /WT1-CreERT2; R26R and Crim1FLOX/FLOX; + /WT1-CreERT2; R26R sections at 17.5 dpc, counter stained with 1%DAB followed by Hematoxylin staining, showing (A), a periostin- and X-Gal-positive cardiac fibroblast (arrow), (B), an X-Gal-positive EPDC (open arrowhead) and (C) a periostin-positive cardiac fibroblast (double open arrowheads). D, E, quantification of % periostin-positive cells among X-Gal-positive myocardial cells in left and right ventricles, showing a decrease in number of cardiac fibroblasts (n = 3–5). *P < 0.05. n.s., not significant; LV, left ventricle; RV, right ventricle. Scale bars, A–C, 10 μm.
Discussion. Crim1 is a type-I transmembrane protein found to be expressed in a spatially and temporally restricted manner in a number of cell types during development and homeostasis, and important for different aspects of organogenesis (Kolle et al., 2000, Pennisi et al., 2007, Wilkinson et al., 2007, Lovicu et al., 2000, Kinna et al., 2006). By employing the use of various transgenic mouse lines,
73 including tissue-restricted mutants, we sought to determine the role of Crim1 in heart development. Crim1 expression begins in the proepicardium, and although it is dispensable for proepicardial specification and outgrowth, its expression continues in the epicardium and epicardium-derived lineages.
There is much effort being put into how to harness the potential of the largely quiescent adult mammalian epicardium, and activate and mobilize it in response to myocardial damage, so as to introduce therapies to reduce the societal burden of cardiovascular disease. This includes re- expression of embryonic epicardial genes and the release of paracrine and autocrine factors from the epicardium or epicardium-derived cells (Zhou et al., 2011a, Smart et al., 2007, Smart and Riley, 2012).
The epicardium is known to contribute important components of the developing heart, and activated epicardial cells have been demonstrated to commit to the fibroblast and smooth muscle cell lineages following myocardial damage (van Wijk et al., 2012, Duan et al., 2012, Limana et al., 2011, Huang et al., 2012, Zhou et al., 2011a). Additionally, the epicardium provides instructive molecules to the developing myocardium, and appears to aid its growth and expansion in a paracrine manner.
Hearts of Crim1∆flox/∆flox mutants displayed abnormal epicardial morphology at 13.5 dpc, indicative of epicardial EMT and adhesion defects. Further investigations involving epicardium-specific Crim1 deletion confirmed increased epicardial EMT and invasion of EPDCs into the myocardium. Moreover, primary epicardial cultures showed increased migration in the absence of Crim1. The distribution of β-catenin at epicardial cell junctions was also altered. Furthermore, CRIM1 has also been shown to be localized to junctional complexes at points of cell-to-cell contact of microvascular endothelial cells when stimulated with lipopolysaccharide (Glienke, Sturz et al. 2002).Taken together with previous studies implicating the interaction of the cytoplasmic domain of Xenopus Crim1 with β-catenin and N-cadherin in the formation and maintenance of cadherin-dependent complexes at epithelial cell junctions, we can infer a role for Crim1 in stabilizing junctional complexes (Ponferrada et al., 2012), not only in epicardial cells, but also in endothelial cells. The binding of Crim1 to β-catenin at junctional complexes could also impede Wnt signalling, and the link between the two molecules especially in the context of developmental interactions between the epicardium and myocardium, is an attractive avenue that is yet to be explored. Additionally, the fact that multiple CRIM1 molecules can self-associate, lends further credibility to its role in increasing junctional stability.
74 Some cystine-knot growth factors that promote epicardial EMT and invasion (such as PDGFs, BMPs, and TGF) (Morabito et al., 2001, Smith et al., 2011, Sánchez and Barnett, 2012) can be bound by CRIM1 (Wilkinson et al., 2003). The lack of changes in canonical BMP signalling via the SMAD1/5 pathway, along with no significant differences in AKT and ERK1/2 signalling in the epicardium of Crim1∆flox/∆flox hearts, indicates that Crim1 may play a role other than epicardial growth factor regulation. Furthermore, the paradoxical reduction in TGF signalling in the epicardium could imply an uncoupling of these pathways and epicardial EMT.
The underlying mechanisms involved in regulation of myocardial development via modulation of epicardial and myocardial signalling molecules is unclear. Mouse mutants where epicardial formation is perturbed, such as VCAM-1, 4 integrin, WT1 knockouts, and mouse lines with altered retinoic acid signalling, show secondary myocardial defects during development (Yang et al., 1995, Moore et al., 1999, Kwee et al., 1995, Chen et al., 2002). Additionally, hypoplastic ventricles with abnormal compact myocardium are also observed in avian embryos in which epicardium formation was obstructed microsurgically (Pennisi et al., 2003, Männer, 1993, Gittenberger-de Groot et al., 2000).
We observed a reduced compact myocardial thickness in Crim1 null hearts at 13.0 dpc. Further investigations revealed that myocardial loss of Crim1 did not affect compact myocardium development through changes in proliferation and apoptosis of cardiomyocytes, which is consistent with the finding that the thickness of compact myocardium was not significantly altered. This indirectly implies the importance of epicardial Crim1 for compact myocardium development and suggests non-cell-autonomous requirements for epicardial Crim1 in myocardial development. However, epicardial loss of Crim1 also did not recapitulate this phenotype, and this could be attributed to incomplete penetrance of the WT1Cre-ERT2 line during development. Taken together with the fact that reciprocal signalling between both the epicardium and the myocardium is required for the development of the myocardium, this suggests a requirement for both epicardial and myocardial Crim1 for myocardial development. Cardiomyocytes comprise an important part of the myocardium and are responsive to a number of signalling molecules via expression of surface receptors that are able to bind mitogens (Sucov et al., 2009). Coincident with this, epicardial IGFs and PDGFs have indeed been shown to act as epicardial mitogenic factors that stimulate the proliferation of cardiomyocytes and subsequently ventricular development (Kang et al., 2008, Li et al., 2011).
75 Crim1 possesses an IGFBP domain with highest sequence similarity to IGFBP-7 (Kolle et al., 2000). Since IGFBPs are known to regulate the activity of IGFs (Hwa et al., 1999), Crim1 could also possibly serve to bind IGFs and regulate downstream signalling. This would be consistent with our observation that ERK1/2 signalling in the myocardium of Crim1∆flox/∆floxhearts is increased. Thus Crim1 could mediate epicardial growth factor sequestration, and this is in accordance with our observation demonstrating increased ERK1/2 signalling in the compact myocardium of Crim1∆Flox/∆Flox hearts. The lack of Crim1 would mean that untethered growth factors are free to signal to the adjacent myocardium. A number of growth factors, however, activate downstream ERK1/2 signalling, and whether this activation is transient or prolonged can lead to different outcomes (Mebratu and Tesfaigzi, 2009). Therefore, the increase in ERK1/2 signalling could either have a pro- or anti-apoptotic effect, and much depends on both the triggering ligand and the downstream effectors. Crim1 may be required to modulate growth factor activity in a timely manner for proper myocardial maturation to ensue.
This hypothesis is exemplified in a study by Wilkinson et al. who showed that Crim1 is co- expressed with VEGF-A in the podocytes of the renal glomerulus (Wilkinson et al., 2007), consistent with a role in tethering epicardium-derived growth factors. In mice lacking Crim1, there was an increased diffusion of VEGF-A away from the podocytes and a concomitant activation of the VEGF-A receptor Flk1 in adjacent vascular endothelial cells, further substantiating the role of Crim1 in growth factor binding and regulation via its CRRs (Wilkinson et al., 2007) and potentially the IGFBP domain as well. The processing and secretion of BMPs has also been shown to be reduced in the absence of CRIM1, and thus, in Crim1∆Flox/∆Flox hearts, there could be increased growth factor production and concomitant signalling. This could subsequently cause a co-activation of apoptosis and proliferation in the myocardium of mutant hearts and result in a high myocardial cell turnover. However, the context-dependent agonistic or antagonistic functions of Crim1 must be noted, which could be dependent on both the cell type within the heart and also the stage of development concerned.
The growth factors that Crim1 could bind and modulate during the process of myocardial maturation can be assessed by immunoprecipitation of endogenous Crim1 and associated growth factors at the specific stages of cardiogenesis when the myocardium is developing, for example, between 10.5 and 15.5 dpc. This would help us understand the temporal regulation of growth factors by Crim1, in the context of myocardial maturation.
76 Although the underlying mechanisms that controls the development of the myocardium is not transparent, regulation of signalling molecules from the epicardium appears to be crucial for myocardial development. Furthermore, though there was an increased migration of EPDCs into the myocardium in the case of Crim1FLOX/FLOX; WT1-Cre and Crim1FLOX/FLOX;WT1-CreERT2 hearts, their proliferative ability was diminished. Alongside, there also exists a potential role for Crim1 in determining specification of EPDCs into the fibroblast lineage. Cardiac fibroblasts are important residents of the myocardium, especially in the case of cardiac repair and performance following myocardial damage (Souders et al., 2009, Camelliti et al., 2005). While we observe a reduction in the number of fibroblasts, it is not known whether the EPDCs that have migrated into the myocardium remain undifferentiated or commit to the coronary vascular smooth muscle cell lineage, altering the balance between cardiac fibroblasts and vascular smooth muscle cells; thus this remains an attractive avenue to be explored.
In summary, the data collectively validates essential roles for Crim1 in heart development, in both autocrine and paracrine ways. These include epicardial EMT and invasion, proliferation and fate specification of EPDCs, and modulation of growth factor activity. Another critical role identified in heart development includes regulating compact myocardial development, likely through regulation of epicardium-derived paracrine factors. Though previously thought to be a quiescent lining that surrounds the myocardium, recent experiments demonstrate that epicardial genes can be reactivated and regulated to release factors from the epicardium or EPDCs in response to injury or disease (Smart and Riley, 2012). Thus, Crim1 may likely provide a potential avenue for future molecular and cellular therapeutic interventions for cardiac regeneration and repair.
Acknowledgements We are grateful to Professor Richard Harvey for providing us with the Mlc2v-Cre and WT1- CreERT2 mouse lines. We thank Dr. Tracey Harvey and Dr. Aaron Smith for helpful discussions, Ms. Sabrina Oishi for assistance with bright field microscopy and Mr. Lachlan Harris for advice on statistical analyses. We thank the staff of UQBR animal facilities and The Centre for Microscopy and Microanalysis at The Universityof Queensland for support.
77
Chapter 4
Examining the cell-autonomous role of Crim1 in the coronary vasculature, specifically in endothelial cells.
78 Author Contributions: Figure 4.10B, C: quantification of viable cell number done blinded by Dr. Yash Chhabra (B) and Dr. Tracey Harvey (C). Figures 4.14, 4.15, Table 4.2: complete experiment performed by Dr. Yash Chhabra. Western blotting done in association with Dr. Yash Chhabra. Densitometry analyses done blinded by Dr. Yash Chhabra. Human Endothelial Cell Biology qPCR array performed under supervision of Dr. Tracey Harvey.
79 Introduction. Cardiac ischemia, which occurs due to reduced blood supplyto the heart, is a feature of coronary artery or ischemic heart disease. Any irregularities in blood vessel function, usually caused due to endothelial cell abnormalities, can ultimately lead to myocardial infarction and impairment of heart muscle function (Hansson 2005). In order to understand the pathogenesis of cardiovascular disease and overcome its burden, we need to examine the functional biology of the network of factors that regulate it at the tissue, molecular and cellular level.
Endothelial cells have not received muchattention in the past due to the misconception that they are an inactive monolayer lining blood vessel walls. However, they have recently been shown to have crucial roles in angiogenesis and vasculogenesis, regulation of vessel tone and permeability(Aird 2007, Aird 2012). Detailed studies on endothelial cells that have their origins in different tissue and vascular beds have made it possible for us to understand endothelial cell heterogeneity (Aird 2012). Within the heart, there are a number of endothelial cell subtypes that display differential gene expression and, subsequently, function (Hendrickx, Doggen et al. 2004). For example, the inner lining of the heart chambers and valves comprise endocardial endothelial cells, microvascular endothelial cells line smaller myocardial vessels, and macrovascular endothelial cells line larger coronary vessels.
The source of cardiac endothelial cells is somewhat controversial. The use of lineage tracing has revealed that the proepicardial organ in avian models is a source of endothelial cells (Mikawa and Gourdie 1996, Pérez-Pomares, Carmona et al. 2002), but in murine models, is not a significant contributor of endothelial cells to the coronary vasculature (Cai, Martin et al. 2008, Zhou, von Gise et al. 2008). However, by using fate-mapping studies in mice, Katz and colleagues have shown that different subcompartments within the proepicardium do give rise to the coronary endothelium (Katz, Singh et al. 2012). Another study presents the possibility of endothelial cells migrating to the proepicardium from the sinus venosus and liver bud, with the proepicardium facilitating the passage of these cells to the developing heart (Cossette 2011). Coronary endothelial cells have also been shown to have epicardial and subepicardial origins (Riley 2012, Tian, Hu et al. 2013). Coronary venous endothelial cells have been shown to arise from the sinus venosus (Red-Horse, Ueno et al. 2010), as well has have endocardial origins (Wu, Zhang et al. 2012). A small population of these cells, by de-differentiation, also form coronary artery endothelial cells (Red-Horse, Ueno et al. 2010), though Wu and colleagues suggest that the major source for coronary arterial endothelial cells appears to be the endocardium (Wu, Zhang et al. 2012). However, it has most recently been confirmed in an elegant study that employed genetic labelling and lineage tracing using a number of
80 transgenic mouse lines that a large population of intramyocardial endothelial cells derives from the sinus venosus in the developing mouse heart, and that the endocardium contributes only minimally to coronary arterial endothelial cells (Zhang, Pu et al. 2016).
Origins notwithstanding, the role that endothelial cells play in vasculature development and homeostasis (Coultas, Chawengsaksophak et al. 2005, Aird 2012), and regulating responses to inflammation, has had widespread acceptance (Pate, Damarla et al. 2010). These cells are known to produce a large number of cytokines, growth factors and other molecules, while also being responsive to the same.
What are the growth factors that play an important role in endothelial cell biology? BMP-2, -4 and - 9 have crucial roles in endothelial cell development (Dyer, Pi et al. 2014), and signal predominantly via binding to type I BMP receptors ALK2, ALK3 and ALK6 (BMPRs), which in turn phosphorylate and activate their downstream mediators, SMAD1, -5 and -8 (SMAD1/5/8) (Kawabata, Imamura et al. 1998). Signalling can also occur via or in combination with other receptors, such as the TGF receptor ALK1 (Brown, Zhao et al. 2005, Scharpfenecker, van Dinther et al. 2007). Some ligand-receptor interactions also require the presence of co-receptors; endoglin is one such glycoprotein that is present predominantly on endothelial cells (Cheifetz, Bellon et al. 1992, Brown, Zhao et al. 2005). Absence of each of these components – the ligands, receptors or downstream mediators – results in defective cardiac and vasculature development (Mishina, Suzuki et al. 1995, Yang, Castilla et al. 1999, Goumans and Mummery 2000, Lechleider, Ryan et al. 2001, Liu, Selever et al. 2004).
As with the case for most growth factors, BMPs exert dual roles depending on the site and time of expression, and also the cell type concerned. For example, smooth muscle cell proliferation in both rat aortic smooth muscle cells and pulmonary artery smooth muscle cells is hindered by BMP-2 (Nakaoka, Gonda et al. 1997, Morrell, Yang et al. 2001). BMP-9 has also been shown to impede both proliferation and angiogenesis in bovine aortic endothelial cells (Scharpfenecker, van Dinther et al. 2007). Conversely, BMP-2 has been shown to have a positive effect on the survival of pulmonary artery endothelial cells (Teichert-Kuliszewska, Kutryk et al. 2006). In human aortic endothelial cells, BMP-2 also stimulates proliferation and tube formation, mediating an angiogenic response (Langenfeld and Langenfeld 2004). Absence of BMP-4 causes impaired formation of branchial arch arteries (Liu, Selever et al. 2004). Furthermore, BMP-4 has also been shown to increase cell number and proliferation of human microvascular endothelial cells while suppressing apoptosis, and also to stimulate angiogenesis via VEGF receptors (Suzuki, Montagne et al. 2008).
81 Similarly, BMP-9 has been found to increase endothelial cell proliferation via phospho-SMAD1/5 in mouse embryonic-stem-cell-derived endothelial cells in vitro, and angiogenesis in vivo(Suzuki, Ohga et al. 2010). Regulation of BMP signal transduction occurs at a number of levels, and in multiple ways (Wilkinson, Kolle et al. 2003, Umulis, O'Connor et al. 2009). Well-known antagonists include the cysteine-rich domain containing chordin (Larraín, Bachiller et al. 2000)and noggin (Zimmerman, De Jesus-Escobar et al. 1996, Groppe, Greenwald et al. 2002), and the agonist crossveinless-2 (Zhang, Patterson et al. 2010).
How does Crim1 function in relation to cardiac and endothelial cell development? It contains six CRRs, via which it potentially binds growth factors such as TGFβ, BMP and VEGF when both are co-expressed in the same cell (Wilkinson, Kolle et al. 2003, Wilkinson, Gilbert et al. 2007). The association of CRIM1 and BMPs has been studied the most, although the exact mechanisms through which a downstream effect is mediated, is not fully understood. Also, whether the binding leads to growth factor signalling being positively or negatively regulated is also not transparent. For example, crm-1, a Crim1 homolog in C.elegans, plays an agonistic role in the BMP-like DBL-1 pathway (Fung, Chi Fat et al. 2007). However, Crimpy, another BMP homolog in Drosophila, impedes BMP signalling at neuromuscular junctions of motorneurons (James and Broihier 2011). Similarly, in vitro studies have revealed that CRIM1 antagonizes both the production and secretion of mature BMPs (Wilkinson, Kolle et al. 2003).
Another growth factor that has received attention in the context of Crim1 binding is VEGF. This growth factor has a pervasive role in endothelial cell function, and its interaction with Crim1 in cultured endothelial cells as well as in the murine retina has been studied (Glienke, Sturz et al. 2002, Fan, Ponferrada et al. 2014). Initially it was shown that in HUVECs, CRIM1 has a crucial role in endothelial cell capillary network formation, indicating a likely involvement in cell-cell contact formation (Glienke, Sturz et al. 2002). Coincident to this, CRIM1 was shown to localize at points of cell contact at the plasma membrane of microvascular endothelial cells (Glienke, Sturz et al. 2002). In vivo, loss of Crim1 has been shown to impede retinal vascular development, with a reduction in vascularized area alongside reduced endothelial cell proliferation (Fan, Ponferrada et al. 2014). CRIM1 has been previously shown to bind VEGF-A (Wilkinson, Gilbert et al. 2007), an indispensible factor for angiogenesis, and conditional deletion of Vegfa or Crim1 alone resulted in impaired angiogenesis, but deletion of both Vegfa and Crim1 caused a more severe phenotypic defect, indicating that Crim1 and VEGF-A function along a similar pathway (Fan, Ponferrada et al. 2014). Furthermore, it has also been demonstrated that CRIM1 is necessary for HUVEC viability, and that it can regulate the autocrine activity of VEGF-A (Fan, Ponferrada et al. 2014).
82
In this chapter, we explore the modulation of BMPs by Crim1 in cardiac endothelial cell function. In the heart, Crim1 is expressed in the epicardium at all stages of development, and also in coronary vascular smooth muscle and endothelial cells (Iyer, Chou et al. 2016). In the current study, we reveal a role for Crim1 in cardiac microvascular endothelial cell development in vivo and in vitro; specifically, a dysregulation of BMP signalling is evident in the absence of Crim1. We also report the misexpression of a number disease-related endothelial genes in the absence of CRIM1, collectively defining a cell-autonomous role in normal endothelial cell function in the heart.
Materials and methods. Ethics Statement. Please refer to the Materials and General Methods chapter for the detailed ethics statement.
Animal Breeding. The mouse lines used in this study were the Crim1FLOXconditional knock-out (Chiu, York et al. 2012), the Crim1Δfloxline (Chiu, York et al. 2012), the ROSA26 Cre reporter line (R26R;(Soriano 1999), and Tie2-Cre line (Kisanuki, Hammer et al. 2001). Embryos were obtained from timed matings between Crim1+/Δfloxintercrosses to produce Crim1Δflox/Δflox, Crim1+/Δflox andCrim1+/+mice or Crim1FLOX/FLOX; R26R females mated with Crim1+/FLOX; Tie2-Cre males to produce Crim1FLOX/FLOX; R26R; Tie2-Cre and Crim1+/FLOX; R26R; Tie2-Cre mice.The presence of a vaginal plug was regarded as 0.5 dpc.
Sample preparation. Embryonic hearts were dissected, photographed in whole-mount as required, and prepared by paraffin or cryo-embedding and sectioning, for immunohistochemical analyses as detailed in the Materials and General Methods chapter.
Antibodies and lectin. Primary antibodies/lectin used in the study were: rabbit polyclonal anti-phospho-Histone H3 (pSer10; #06-570, Millipore); rabbit monoclonal anti-cleaved caspase 3 (activated caspase 3; #9664, Cell Signaling Technology); rabbit monoclonal anti-phospho-Smad1/5 (pSer463/pSer465; Invitrogen #700047); rabbit polyclonal anti-phospho-Smad2 (pSer465/pSer467; Millipore AB3849); total SMAD1/5/8/9 (NB100-56656;Novus Biologicals); rat monoclonal anti-CD31 (PECAM-1; BD Pharmingen #550274); Angiogenesis Sampler kit (#8696; Cell Signalling); total AKT (#9272; Cell Signalling); total ERK1/2 (sc-93; Santa Cruz); rabbit monoclonal HA tag
83 (#3724; Cell Signalling); mouse monocolonal myc tag (#2276; Cell Signalling); GAPDH (#2118; Cell Signalling); -tubulin (#2146; Cell Signalling) and Isolectin B4 (#MP21410; Invitrogen). Secondary antibodies used for immunofluorescence experiments were Alexa Fluor-488, and 594- conjugated secondary antibodies (Invitrogen). Secondary antibodies used for immunohistochemistry experiments were anti-rabbit HRP (Promega; #W4011), anti-mouse HRP (Promega; #W4021), anti-rabbit biotin (Sigma; # B7389).
Cell Culture. HCMVECs (Lonza, CC-7030) were cultured in EGM-2MV BulletKit media (Lonza, CC-3202) and pooled HUVECs (CC-2519) were cultured in EGM-2 BulletKit media CC-3162). Cells before passage six were used in all experiments. shRNA targeting CRIM1 mRNA (Fan et al, 2014) was acquired from TRC Mission Library (TRCN0000063862, Invitrogen). Lentiviral particles were synthesized by transfecting CRIM1 shRNA, scrambled shRNA and the lentiviral packaging vectors into HEK293T cells using Lipofectamine 2000. The supernatant containing lentiviral particles was collected at 24 and 48 hours after transfection. Cells at 80% confluence were then incubated with lentiviral particles along with 8 μg/ml polybrene for 48 hours and then treated with 2 μg/ml puromycin for 48 hours to enrich the infected cell population. After puromycin removal, cells were used in subsequent experiments.
For experiments not involving external stimulation with growth factors (BMP-2/4), HCMVECs were cultured in EGM-2MV and EGM-2 complete medium. To examine the effect of exogenous growth factor stimulation and inhibitor treatment, cells were starved in serum-free basal media without growth factors for 2 hours, and treated with LDN193189 (1mM)(Yu, Deng et al. 2008) for 30 minutes. This was done prior to stimulation with BMP-2 and BMP-4 (100 ng/ml) for ten minutes (Star, Giovinazzo et al. 2009).
Trypan Blue Viability Assay. 5,000 scrambled shRNA Control and CRIM1-knockdown CMVECs and 16,000 scrambled shRNA Control and CRIM1-knockdown HUVECs were plated into wells of a 24-well plate at time zero, in EGM-2MV and EGM-2 complete medium. The number of viable cells was counted at two time points namely 24 hours and 48 hours based on Trypan Blue exclusion.
Quantitative RT-PCR and q-PCR Array. RNA was extracted using the RNeasy Mini Kit (Qiagen) and stored at -80°C. cDNA was prepared using RT2 First Strand Kit (Qiagen) according to manufacturer’s protocol. Quantitative PCR to
84 assess percentage CRIM1 knockdown was performed with QuantiFast SYBR Green PCR kit (Qiagen). The RT2 SYBR Green ROX qPCR Mastermix was used for Profiler PCR Array (Plate E). Signals were detected using the Roche LightCycler and Applied Biosystems QuantStudio 7 system and gene expression levels were calculated using the standard curve method. Gene analyses were performed using the Qiagen Gene Globe Data Analysis Center and the Ingenuity Pathway Analysis.
Plasmid Constructs. GFP and CRIM1 open reading frames were cloned into the pLVX-Puro lentiviral construct as per manufacturer’s instructions (Clontech).
CRIM1 overexpression. The 2A peptide sequence is a self-processing viral peptide bridge that can separate different protein coding sequences in a single ORF transcription unit (Ryan, King et al. 1991). Using these 2A peptide constructs it has been shown that quantitative co-expression of heterologous proteins, driven by one promoter is possible for both in vitro and in vivo analysis (Tang, Ehrlich et al. 2009).
For creating CRIM1 lentiviral constructs using 2A peptide, two individual PCRs were carried out. The first fragment was generated by attB1 sequence linked to Crim1 CDS and with the reverse complement of half of a 2A peptide sequence linked to reverse complement of GHR C-terminus without the stop codon. The second fragment was generated by using half of 2A peptide linked with GFP start codon and attB2 sequence linked with reverse complement of GFP C-terminus sequence with a stop codon between attB2 and GFP.
Primer sequences were as follows: attB1-CRIM1 Fwd 1 5'-GGGGACAAGTTTGTACAAAAAAGCAGGCTATGTACTTGGTGGCGGGGGACAG
CRIM1-HAtag-2Cc1 Rev 1 5’- GTTTTCCTCCACGTCGCCGCAGGTGAGCAGGCTTCCCCTGCCCTCAGCGTAATCTGGAA CATCGTATGGGTACACTGTTTGGTAGAAATTGTCTGCCTGTAGATGG
2Ac1-GFP Fwd 2 5'- ACCTGCGGCGACGTGGAGGAAAACCCTGGCCCCGCCCCAGGATCCATGGTGAGCAAG GGCGAGGAGCTG
GFP-attB2 Rev 2 5'- GGGGACCACTTTGTACAAGAAAGCTGGGTTTACTTGTACAGCTCGTCCATGCCGAGAG
85
The portion of DNA sequence of 2A peptide was kept overlapping between the two PCRs while designing the Rev primer1 and Fwd primer2. This common region between the two PCR products was on the same principle as overlap extension PCR cloning (Bryksin and Matsumura 2013). The final PCR fragment containing CRIM1-2A-GFP flanked by attB sites was cloned in to pLenti-PGK- Puro DEST (plasmid# 19068, Addgene) by Gateway Cloning (as per manufacturer's guidelines, Thermofisher). pLenti-PGK-GFP-puro (Plasmid#19070, Addgene) was used as control plasmid. qPCR primers for analysis of effect of CRIM1 overexpression. hCasp1-Fwd 5’-TGTTCCTGTGATGTGGAGGAAA-3’ hCasp1-Rvs 5’-CATCTGGCTGCTCAAATGAAAA-3’ hCCL5-Fwd 5’-AGCCTCTCCCACAGGTACCA-3’ hCCL5-Rvs 5’-GGCAGTAGCAATGAGGATGACA-3’ hCx3cl-Fwd 5’-CCCGGAGCTGTGGTAGTAATTC-3’ hCx3cl-Rvs 5’-ACTAGACACAGGCCAGAGGAGTTC-3’ hIL1B-Fwd 5’-TTAAAGCCCGCCTGACAGA-3’ hIL1B-Rvs 5’-GCGAATGACAGAGGGTTTCTTAG-3’ hKIT-Fwd 5’-GCTTTTATACTACCGACCTGGTTTTT-3’ hKIT-Rvs 5’-GGGAGGCCTTCTCTCCAATT-3’ hSELE-Fwd 5’-AAGCCTAAACCTTTGGGTGAAAA-3’ hSELE-Rvs AAGTGAGAGCTGAGAGAAACTGTGAA hSerpine1-Fwd 5’-GCTTTTGTGTGCCTGGTAGAAA-3’ hSerpine1-Rvs 5’-GGCAGGCAGTACAAGAGTGATG-3’ hTNSF10-Fwd 5’-CATCAAGGAGTGGGCATTCA-3’ hTNSF10-Rvs 5’-TGACCAGTTCACCATTCCTCAA-3’ hTYMP-Fwd 5’-GCGGACGGAATCCTATATGC-3’ hTYMP-Rvs 5’-TGAGAATGGAGGCTGTGATGAG-3’
86 hVCAM-Fwd 5’-CACTCCGCGGTATCTGCAT-3’ hVCAM-Rvs 5’-TGTGCCTGGGAGGGTATTCA-3’
Data documentation, quantification and analyses. Bright-field and fluorescent images were captured as outlined in the Materials and General Methods chapter, and images were adjusted using Adobe Photoshop software.
Signal intensity was determined using the Integration function of the Olympus Fluoview Software and the number of DAPI-positive nuclei in the demarcated area was quantified using the ImageJ software.
For quantification of coronary endothelial cell numbers, 40x images were acquired of the left and right mid-ventricular myocardial free wall from at least three non-consecutive sections of each sample on a laser scanning confocal microscope. PECAM-1 positive cells were counted and divided by total cell number to determine the PECAM-1 positive percentage. Endocardial and trabecular endothelial cells were not included in the quantification.
For quantification of proliferating and dying endothelial cells, 40x images were acquired of the left and right mid-ventricular myocardial free wall from at least three non-consecutive sections of each sample on a laser scanning confocal microscope. PHH3-and-CC3 PECAM-1-positive cells were counted and divided by PECAM-1-positive cell number to determine percentages of proliferating and dying endothelial cells respectively. PHH3- and CC3-positive and PECAM-1-negative cells were also counted to investigate a paracrine effect.
For quantification of intensity of phospho-SMAD2 and phospho-SMAD1/5 levels, at least two non- consecutive sections for each sample were imaged using a laser scanning confocal microscope. To determine total signal intensity in, and area of, the specified region of interest, Olympus Fluoview software was used.
Signal intensity within specifically endothelial cells was determined using the IMARIS software (ver 8.2.0). Channel 1 was set to DAPI, channel 2 to PECAM-1, and channel 3 to phospho-SMAD 1/5 or phospho-SMAD 2, based on the fluorescent dyes. A surface for channel 2 was created. Channel 1 for that surface was masked. This created ‘Channel 4’, which was PECAM-1 plus the DAPI associated with that PECAM-1. A surface for channel 4 was created. Channel 3 for that
87 surface was masked. This created ‘Channel 5’, which was PECAM-1 + DAPI associated with that PECAM-1, + phospho-SMAD associated with that DAPI. A surface for Channel 5 was created. All non-endothelial cells were deselected and intensity of phospho-SMAD within only endothelial cells was obtained.
Experiments were repeated three times, and each time in duplicate, for CRIM1 knockdown quantitation and the results were averaged and compared to controls. For the endothelial biology qPCR array, 4 separate biological samples were used in duplicate (as a technical replicate).
Overexpression experiments were performed in triplicate, and relative expression was normalised to β2-microglobulin. All P-values were evaluated by a two-tailed, unpaired t-test.
To determine statistical significance between pooled data from mutant and control groups a two- tailed, unpaired Student’s t-test was used. Quantified data shown are mean ± standard deviation. GraphPad Prism v6.0 was used to conduct all analyses and generate graphs.
Results. Crim1 deletion results in abnormal coronary vasculature development. That Crim1 is necessary for normal heart development has been established previously, with Crim1Δflox/Δflox hearts exhibiting a range of defects including reduced ventricular size and irregular epicardia (Iyer, Chou et al. 2016). Crossing Crim1+/Δflox mice to produce Crim1Δflox/Δflox mutants did not reveal abnormal vasculature development upon whole mount PECAM staining at 13.5 dpc (Figure 4.1A,B, D, E); however, vascular malformations at 17.5 dpc were observed (Figure 4.1C,F).
88
Figure 4.1: Loss of Crim1 results in malformed coronary vasculature at 17.5 dpc A, B, D, E, Whole mount PECAM staining showing no significant differences in vasculature formation between Crim1+/+ (A, B) and Crim1Δflox/Δflox (D, E) hearts at 13.5 dpc. B, E, higher magnifications of boxed areas in (A) and (D). (n=3). C, F, Whole-mount views of 17.5 dpc embryonic hearts fromCrim1+/+ (C) and Crim1Δflox/Δflox (F) embryos. Note the malformed coronary vessels in the Crim1∆flox/∆flox heart (F). (n=3). Scale bars, C, F, 400 μm.
To analyse any defects arising from endothelial cell anomalies, we assessed whether there were any changes in endothelial cell endowment in these mutant hearts. To do this we immunostained cryosections from Crim1+/+,Crim1+/ΔfloxandCrim1Δflox/Δfloxhearts with PECAM-1, a marker for
89 endothelial cells.We observed a reduction in the number of endothelial cells in the ventricular compact myocardium of Crim1Δflox/Δfloxhearts at 17.5 dpc (Figure 4.2A-F).
Figure 4.2: Reduced number of coronary vascular endothelial cells in Crim1∆flox/∆flox hearts A-D, Confocal images of sections of 17.5 dpc hearts from Crim1+/+ and Crim1+/∆flox (A, B) and Crim1∆flox/∆flox (C, D) embryos stained with PECAM-1 (red). A, C, merged confocal images of DAPI (blue) and the PECAM-1 stain (red).E, F, There was a reduced percentage of PECAM-1+ endothelial cells (y-axis) in the ventricular myocardium of Crim1∆flox/∆flox hearts (n=4). RV, right ventricle; ep, epicardium; en, endocardium. *P<0.05. Scale bars A, C, 50 μm.
Cell autonomous role for Crim1 in endothelial cell development. To address whether this was due to a cell-autonomous role of Crim1, we crossed Crim1FLOX/FLOX; R26R mice with the Crim1+/FLOX; Tie2-Cre line for the deletion of Crim1 from specifically the endothelial cells. This line has been previously used and restricted to endothelial cells (Kisanuki, Hammer et al. 2001, Koni, Joshi et al. 2001). We confirmed this by both X-Gal staining and also histochemical analyses using Isolectin B4 as a marker of endothelial cells (Figure 4.3A-G). There was extensive overlap between the blue cells that had lost Crim1, and Isolectin-B4- positive endothelial cells (Figure 4.3F, G). At 17.5 dpc, we also observed a decrease in endothelial cell endowment in Crim1FLOX/FLOX; R26R; Tie2-Cre mutant hearts compared to Crim1+/FLOX; R26R; Tie2-Cre controls (Figure 4.4A-F).
90
Figure 4.3: Efficiency of the Tie-2Cre line A-G, Efficiency of the Tie2-Cre line. A, X-Gal-stained Crim1+/FLOX; Tie2-Cre; R26R heart at 17.5dpc displaying lacZ expression in the vasculature upon Cre-mediated excision of Crim1.B, C, Higher magnification of vasculature in (A). D, E, Histological sections after nuclear fast red staining showing Cre-mediated loss of Crim1 in coronary vascular endothelial cells at 15.5 dpc. F, G, Histological sections showing overlap between endothelial marker Isolectin B4 and X-Gal at 15.5 dpc, confirming efficiency of Tie2Cre line. Scale bars, A, 500 μm; B, 300 μm; C, 150 μm; D, F, 200 μm; E, G, 20 μm.
91
Figure 4.4: Reduced number of coronary vascular endothelial cells in Crim1FLOX/FLOX; Tie2- Cre hearts A-D, Confocal images of sections of 17.5 dpc hearts from Crim1+/FLOX; Tie2-Cre; R26R (A, B) and Crim1FLOX/FLOX; Tie2-Cre; R26R (C, D) embryos stained with PECAM-1 (red). A, C, merged confocal images of DAPI (blue) and the PECAM-1 stain, which is specific for endothelial cells (red).E, F, There was a reduced percentage of PECAM-1+ endothelial cells (y-axis) in the ventricular compact myocardium of both ventricles of Crim1FLOX/FLOX; Tie2-Cre; R26R hearts. (n=9-7). RV, right ventricle; ep, epicardium; en, endocardium. *P<0.05 Scale bars A, C, 50 μm.
This decrease could be either due to a decrease in proliferation or an increase in apoptosis, and thus we investigated both, using PHH3 (Yang 2011)and CC3(Barbosky, Lawrence et al. 2006) as corresponding markers for proliferation and apoptosis respectively. While there was a limitedincrease in apoptosis in the left ventricles of Crim1FLOX/FLOX; R26R; Tie2-Cre mutant hearts (Figure 4.5A-F), there was no change in the number of proliferating endothelial cells (Figure 4.6A- D). This could be because PHH3 marks only a small population of proliferating cells. To address this concern, we also assessed proliferation using the marker Ki67(Yang 2011), which marks all phases of the cell cycle except for the resting phase (Figure 4.7A-D) and can thus be used to completely rule out that the observed reduction in number of endothelial cells is not due to a proliferation defect. There was still no change in the number of proliferating endothelial cells. There was also no change in the apoptosis or proliferation of adjacent non-endothelial cells in experiments employing CC3 (Figure 4.8A, B) and PHH3 (Figure 4.8C, D). Interestingly, while Crim1Δflox/Δflox mice displayed embryonic lethality, Crim1FLOX/FLOX; R26R; Tie2-Cre mice
92 survived into adulthood without any apparent external phenotypic anomalies (data not shown; n=9).
Figure 4.5: Limited apoptosis of endothelial cells in Crim1FLOX/FLOX; Tie2-Cre heartsas assessed using CC3 A-D, Confocal images of ventricular sections of 17.5 dpc hearts from Crim1+/FLOX; Tie2-Cre (A, B)andCrim1FLOX/FLOX; Tie2-Cre (C, D) embryos immunostained with PECAM-1 and CC3. A-D, Merged images of DAPI (blue), CC3 (red) and PECAM-1 (green), with higher magnification images showing a non endothelial cell undergoing apoptosis (B), and an endothelial cell undergoing apoptosis (D). E, F, Increase in endothelial cell apoptosis in the left ventricles of Crim1FLOX/FLOX; Tie2-Cre hearts. (n=6-7). LV, left ventricle; en, endocardium; ep, epicardium. *P<0.05. Scale bars A, C, 50 μm, B, D 10 μm.
93
Figure 4.6: No change in proliferation of endothelial cells in Crim1FLOX/FLOX; Tie2-Cre heartsas assessed with PHH3 A-D, No change in proliferation of endothelial cells in ventricles of Crim1+/FLOX; Tie2-Cre hearts. A, B, Confocal images of ventricular sections of 17.5 dpc hearts from Crim1+/FLOX; Tie2-Cre (A)andCrim1FLOX/FLOX; Tie2-Cre (B) embryos immunostained with PECAM-1 (green) and PHH3 (red). C, D, No change in percentage of PHH3- positive endothelial cells (y-axis) in both ventricles of Crim1FLOX/FLOX; Tie2-Cre hearts (n=9-7). LV, left ventricle; en, endocardium; ep, epicardium; n.s, not significant. Scale bars, A, B, 50 μm.
94
Figure 4.7: No change in proliferation of endothelial cells in Crim1FLOX/FLOX; Tie2-Cre hearts as assessed with Ki67 A-D, No change in proliferation of endothelial cells in ventricles of Crim1+/FLOX; Tie2-Cre hearts. A, B, Confocal images of ventricular sections of 17.5 dpc hearts from Crim1+/FLOX; Tie2-Cre (A)andCrim1FLOX/FLOX; Tie2-Cre (B) embryos immunostained with PECAM-1 (red) and Ki67 (green). C, D, No change in percentage of Ki67-positive endothelial cells (y-axis) in both ventricles of Crim1FLOX/FLOX; Tie2-Cre hearts (n=5-6). LV, left ventricle; en, endocardium; ep, epicardium; n.s, not significant. Scale bars, A, B, 50 μm.
95
Figure 4.8: No change in apoptosis and proliferation of non-endothelial cells in Crim1FLOX/FLOX; Tie2-Cre hearts as assessed with CC3 and PHH3 A, B, No change in apoptosis of CC3-positive non-endothelial cells in both ventricles of Crim1FLOX/FLOX; Tie2-Cre hearts compared to controls (n=6-7). C, D, No change in proliferation of PHH3-positive non-endothelial cells in both ventricles of Crim1FLOX/FLOX; Tie2-Cre hearts compared to controls (n=9-7). n.s., non significant.
Growth factor signalling is perturbed in the absence of Crim1. Growth factors are known to regulate cell development and survival in both normal and diseased states in a number of ways, although most of these are complex and poorly understood. CRIM1 is a versatile protein that can bind a number of growth factors such as TGFβ, BMPs and VEGF (Wilkinson, Kolle et al. 2003, Wilkinson, Gilbert et al. 2007). The interactions and function of Crim1 and VEGF in retinal vascular endothelial cells as been studied (Fan, Ponferrada et al. 2014). Thus we next assessed whether loss of Crim1 would impact BMP signal transduction within coronary vascular endothelial cells. We performed indirect immunofluorescence for PECAM-1 and phospho-SMAD1/5 on Crim1Δflox/Δflox hearts at 17.5 dpc, as a proxy to assess canonical BMP signalling (Figure 4.9A-D). This revealed a reduction in phospho-SMAD1/5 in all cells of the ventricular compact myocardium as well as in specifically endothelial cells of the left ventricle, although a trend is visible in the right ventricle. (Figure 4.9E-H).
96
Figure 4.9: Decrease in canonical BMP signalling in cells of Crim1∆flox/∆floxhearts as assessed with phospho-SMAD1/5 A-D, Confocal images of ventricular sections of 17.5 dpc hearts from Crim1+/+, Crim1 +/Δflox (A, B) and Crim1Δflox/Δflox (C, D) embryos, immunostained for PECAM-1 and phospho-SMAD1/5. (A, C) Merged images of DAPI (blue), PECAM-1 (red) and phospho-SMAD1/5 (green). (E, F) Reduction in phospho-SMAD1/5 levels in all
97 cells of both ventricles of Crim1Δflox/Δflox hearts and (G H) endothelial cells of the left ventricles of Crim1Δflox/Δflox hearts (n=4). RV, right ventricle; ep, epicardium; en, endocardium. *P<0.05, Scale bars A-D, 50 μm.
Examination of CRIM1 function in vitro. Our findings to date using the Crim1+/Δflox, Crim1+/FLOX; Tie2-Cre, and Crim1FLOX/FLOX; R26R mouse lines suggested that Crim1 plays is essential for coronary endothelial cell development and BMP signalling, in vivo. To further probe the role of Crim1 in endothelial cells, we employed in vitro assays using both HUVECs and HCMVECs. First, we mediated the knockdown of CRIM1 using lentiviral shRNA, in HCMVECs (Figure 4.10A), using an shRNA targeting CRIM1 that has been previously used for the efficient knockdown of CRIM1 in HUVECs (Fan, Ponferrada et al. 2014). We were able to achieve CRIM1 knockdown by approximately 75% at the transcript level, compared to scrambled shRNA controls (Figure 4.10A). Our hypothesis was that loss of CRIM1 would lead to a decrease in cell viability, and also impact growth factor signalling within endothelial cells. To examine this, we assessed the viability of HCMVECs and HUVECs lacking CRIM1, and found a reduction in viable cell number as well as percentage viability in HCMVECs, and a reduction in viable cell number but not percentage viability in HUVECs, as measured by the Trypan Blue Viability Assay (Figure 4.10B, C, D, E). This suggests a functional requirement of CRIM1 in maintaining the ability of HCMVECs to survive.
98
Figure 4.10: CRIM1 knockdown in HCMVECs and its effect on cell viability A, Approximately 75% CRIM1 knockdown in HCMVECs (n=3). B, C, Trypan Blue Viability Assay demonstrating decrease in viablecell number and percentage viability in CRIM1 knockdown HCMVECs compared to control after 24 and 48 hours respectively, after plating 5000 cells per well of a 24 well plate (n = 3, and each n represents an average of 3 replicates). D, E, Trypan Blue Viability Assay demonstrating decrease in viable cell number but not percentage viability in CRIM1 knockdown HUVECs compared to control after 24 and 48 hours respectively, after plating 16000 cells per well of a 24 well plate (n = 3, and each n represents an average of 3 replicates). *P<0.05, **P<0.01, ****P<0.0001.
Dysregulation of canonical BMP signalling in the absence of CRIM1 in vitro. We have observed a decrease in endothelial phospho-SMAD1/5 signalling in left ventricles of Crim1Δflox/Δflox mutants (Figure 4.9G). To investigate the effect that loss of CRIM1 function has on canonical BMP signalling in HCMVECs, we assessed the phosphorylation of downstream effectors SMAD1/5, AKT and ERK1/2 as proxies for downstream signal transduction (Figure 4.11A). While there was no change in either phospho-ERK1/2 or phospho-AKT levels, there was a significant decrease in phospho-SMAD1/5 levels in CRIM1 knockdown lysates compared to
99 scrambled shRNA controls (Figure 4.11B). This is consistent with our in vivo observations on Crim1Δflox/Δflox mutants (Figure 4.9).
Figure 4.11: CRIM1 knockdown in HCMVECs results in decreased phospho-SMAD1/5 activity A, B, Reduction in pSMAD1/5 levels as a proxy to assess BMP signalling in CRIM1 knockdown HCMVECs, but no change in pERK1/2 and pAKT levels, compared to scrambled shRNA control HCMVECs (n=4-7). Representative blot chosen*P<0.05
Misregulation of genes involved in endothelial stress response, apoptosis and cell adhesion in HCMVECs lacking CRIM1. Certain growth factors, such as BMPs, are known to have an impact on endothelial disease states, including inflammatory responses and responses to stress and injury (Lowery and deCaestecker 2010). CRIM1 has been shown to interact with many of these growth factors (Wilkinson, Kolle et al. 2003, Wilkinson, Gilbert et al. 2007). Furthermore, we have observed aberrant BMP signalling in endothelial cells in the absence of Crim1. Additionally, vascular malformations are also a potential consequence of chronic inflammation and stress. The effect of CRIM1 ablation on HCMVEC genes has not been studied, and hence, I hypothesised that loss of CRIM1 would result in misregulation of endothelial genes that are critical to endothelial cell development and function. Using a Human Endothelial Cell Biology qPCR array, I performed a quantitative assay testing a panel of 84 genes involved in endothelial disease progression. This revealed that 29 genes involved in endothelial cell function are greater than two-fold up or downregulated (Figure 4.12, Table 4.1).
100
Figure 4.12: Effect of CRIM1 knockdown on endothelial gene expression as assessed by a human endothelial cell biology array A, Scatterplot displaying misregulation of 29 genes out of 84, involved in endothelial cell function, with a fold-change cut-off of 2. The scatter plot compares the normalized expression of every gene on the array between the two selected groups by plotting them against one another to quickly visualize large gene expression changes. The central line indicates unchanged gene expression. The dotted lines indicate the selected fold regulation threshold. Data points beyond the dotted lines in the upper left and lower right sections meet the selected fold regulation threshold. (n = 4, with 2 technical repeats i.e 4 independent experiments performed in duplicate).
101 Table 4.1: List of all 84 endothelial genes assessed in the human endothelial cell biology array, with fold regulation and P values. Significant genes are highlighted in bold font.
NUMBER GENE REFSEQ FOLD REGULATION P-VALUE 1 ACE NM_000789 1.0909 0.290151 2 ADAM17 NM_003183 -2.5219 0.00018 3 AGT NM_000029 -1.4969 0.076943 4 AGTR1 NM_031850 -1.1724 0.522084 5 ALOX5 NM_000698 1.0231 0.969908 6 ANGPT1 NM_001146 -1.8019 0.007017 7 ANXA5 NM_001154 -2.2338 0.000006 8 APOE NM_000041 -1.6211 0.044427 9 BAX NM_004324 -1.0566 0.281637 10 BCL2 NM_000633 -1.5905 0.030518 11 BCL2L1 NM_138578 -1.3774 0.000041 12 CALCA NM_001741 1.0374 0.488782 13 CASP1 NM_033292 5.1266 0.00003 14 CASP3 NM_004346 1.0501 0.424219 15 CAV1 NM_001753 -2.0448 0.000345 16 CCL2 NM_002982 1.1958 0.04106 17 CCL5 NM_002985 15.0637 0.000131 18 CDH5 NM_001795 -1.4064 0.000322 19 CFLAR NM_003879 -1.0154 0.820888 20 COL18A1 NM_030582 -1.3561 0.007195 21 CX3CL1 NM_002996 6.139 0.026342 22 EDN1 NM_001955 -1.8019 0.000243 23 EDN2 NM_001956 1.0073 0.925998 24 EDNRA NM_001957 -1.431 0.154058 25 ENG NM_000118 -1.2393 0.007858 26 F2R NM_001992 1.32 0.01127 27 F3 NM_001993 3.7529 0.000074 28 FAS NM_000043 1.8797 0.000744 29 FASLG NM_000639 1.0374 0.488782 30 FGF1 NM_000800 1.32 0.277886 31 FGF2 NM_002006 -1.0566 0.535619 32 FLT1 NM_002019 -1.1266 0.145134 33 FN1 NM_002026 -2.2611 0.0004 34 HIF1A NM_001530 -1.0457 0.107155 35 HMOX1 NM_002133 -1.1247 0.142892 36 ICAM1 NM_000201 1.7207 0.000004 37 IL11 NM_000641 -2.117 0.013737 38 IL1B NM_000576 7.0153 0.001666 39 IL3 NM_000588 1.0374 0.488782 40 IL6 NM_000600 1.7752 0.002456 41 IL7 NM_000880 1.6621 0.124883 42 ITGA5 NM_002205 -1.2414 0.001696 43 ITGAV NM_002210 -2.3983 0.00012 44 ITGB1 NM_002211 -2.8669 0.000001
102 45 ITGB3 NM_000212 -1.5099 0.003451 46 KDR NM_002253 -1.0996 0.224994 47 KIT NM_000222 -3.7373 0.000736 48 KLK3 NM_001648 -1.0845 0.479634 49 MMP1 NM_002421 -1.5204 0.000071 50 MMP2 NM_004530 -1.2986 0.012208 51 MMP9 NM_004994 2.247 0.008745 52 NOS3 NM_000603 1.0629 0.378124 53 NPPB NM_002521 -3.0042 0.007349 54 NPR1 NM_000906 1.8188 0.005741 55 OCLN NM_002538 -2.3367 0 56 PDGFRA NM_006206 -1.456 0.104707 57 PECAM1 NM_000442 1.0091 0.766073 58 PF4 NM_002619 -4.2047 0.030872 59 PGF NM_002632 -1.7255 0.000027 60 PLAT NM_000930 -1.4459 0.126347 61 PLAU NM_002658 1.0556 0.288138 62 PLG NM_000301 1.0374 0.488782 63 PROCR NM_006404 -2.3327 0.000002 64 PTGIS NM_000961 -2.0028 0.008857 65 PTGS2 NM_000963 2.6126 0.000009 66 PTK2 NM_005607 -1.3967 0.000021 67 SELE NM_000450 7.0641 0.000001 68 SELL NM_000655 1.6364 0.000056 69 SELPLG NM_003006 -2.1803 0.029765 70 SERPINE1 NM_000602 -4.6251 0.000001 71 SOD1 NM_000454 -1.7863 0.000009 72 SPHK1 NM_021972 -2.1391 0.000192 73 TEK NM_000459 -2.2532 0 74 TFPI NM_006287 -1.2986 0.000011 75 TGFB1 NM_000660 -1.1092 0.093609 76 THBD NM_000361 -2.4487 0.000072 77 THBS1 NM_003246 -1.5878 0.000133 78 TIMP1 NM_003254 1.3269 0.004462 79 TNF NM_000594 1.7752 0.0015 80 TNFSF10 NM_003810 9.5666 0.000001 81 TYMP NM_001953 4.494 0.000058 82 VCAM1 NM_001078 5.056 0 83 VEGFA NM_003376 -1.0677 0.210602 84 VWF NM_000552 -1.2117 0.025954
When these genes were further analysed using Ingenuity Pathway Analysis, we observed that those involved in inflammation such as pro-inflammatory cytokines produced in response to stress, such as CCL5, CX3CL1, IL11 and IL1B, adhesion molecules such as VCAM1 and SELE which are induced after endothelial cell activation, and molecules involved in cell death, such as CASP1, TNFSF10, PF4, CAV1 and IL11, were enriched (Figure 4.13B, Table 4.1). Chronic inflammatory conditions can lead tovascular malformations as seen in Crim1Δflox/Δflox mice, and
103 eventually lead to irreversible multiple organ defects(Pennisi, Wilkinson et al. 2007). This could be one of the causes for embryonic lethality observed in Crim1 conditional null mice(Chiu, York et al. 2012).
In the case of genes associated with cellular infiltration, ILIB, VCAM1, PTGS2, CCL5, SELE, F3, CASP1, which increase infiltration, were upregulated, and CAV1, SERPINE1, FN1, ADAM17 and THBD, which decrease cellular infiltration, were downregulated. Genes associated with inflammation were predicted to be enriched based on upregulation of IL1B and VCAM1, and downregulation of OCLN, THBD and IL11. Specifically, there was an increase in expression of endothelial genes involved in recruitment of phagocytes, monocytes and leukocytes, and infiltration by cells such as monocytes, neutrophils, leukocytes and granulocytes (Figure 4.13B, Table 4.1), which is characteristic of cellular recruitment to sites of injury and stress and an inflammatory response. Similarly, genes associated with cytolysis were predicted to be increased, due to an upregulation of IL1B, TNFSF10, CASP1 and CX3CL1, and downregulation of PF4, CAV1 and IL11. This is consistent with invivo data indicating a reduction in endothelial cell numbers in the absence of Crim1as observed using endothelium-specific mutants and conditional null mice, as well as in vitro data which indicates reduced HCMVEC viability.
Genes associated with vascular lesions were predicted to be enriched (Fig. 4.13B, Table 3), consistent with an upregulated expression of endothelial genes such as IL1B, VCAM1, PTGS2, CX3CL1 and SELE. Genes associated with size of vascular lesions were also predicted to be increased based on increase in endothelial expression of IL1B, VCAM1, PTGS2, SELE, CX3CL1, MMP9 and CASP1, and downregulation of CAV1, ANXA5 and THBD. Pertinent to other cardiovascular diseases, genes associated with heart failure and infarction were predicted to be increased based on misregulation of endothelial genes such as CAV1, ANXA5, PTGS2, THBD, ILIB, F3 and MMP9 (Fig. 4.13C, Table 4.1).
Additionally, upstream factors such as TNF and TNF family members, CD40, IL17 and RELA (of the NFB family) which have been shown to stimulate inflammation (Galkina and Ley 2009, Lawrence 2009) were predicted to be activated based on changes in downstream effector genes (Figure 4.13A). Specifically, in the case of TNF, 16 genes had an expression direction consistent with activation of TNF. VCAM1, SELE, IL1B, CCL5, TYMP, PTGS2, F3, TNFSF10, MMP9 and CX3CL1 which are known to be upregulated by TNF were upregulated in the dataset, and thus TNF was predicted to be activated. THBD, SPHK1, OCLN, KIT andCAV1 which are known to be downregulated by TNF are downregulated in the dataset, and thus, TNF is predicted to be activated.
104 RELA is predicted to be activated based on upregulation of VCAM1, SELE, PTGS2, MMP9, IL1B and CCL5. IL17A is predicted to be activated based on upregulation of VCAM1, SELE, PTGS2, MMP9, IL1B, CX3CL1 and CCL5, and downregulation of THBD and OCLN (Fig. 4.13A, Table 4.1).
4.13: Ingenuity Pathway Analysis of misregulated endothelial genes in CRIM1 knockdown HCMVECs A-C, IPA analysis of genes misregulated in the human endothelial cell biology qPCR array, indicating predicted activity by upstream regulators such as TNF, RELA, IL17A (A), a heightened inflammatory response and cell death (B), along with a predicted increase in heart failure, infarction and vascular lesions (C) based on endothelial gene expression. Z scores > 2 are significant.
105 Effect of CRIM1 overexpression on endothelial genes. Having observed that CRIM1 ablation causesgene expression changes that have been correlated with heightened inflammatory response and cell death, we hypothesized that CRIM1 overexpression could ameliorate these results by having an opposite effect on the expression of these endothelial genes. To address this hypothesis, we cloned CRIM1 in HCMVECs using Gateway Technology,to overexpress CRIM1, confirming that there was greater than a 2-fold increase in CRIM1 expression compared to GFP-only controls (Figure 4.14A).
4.14: CRIM1 overexpression in HCMVECs A, CRIM1 overexpression in HCMVECs as compared to GFP-only control. (n=3). **P<0.01.
We then tested the expression of 10 genes that were more than three-fold up or downregulated in the absence of CRIM1, including CASP1, CCL5, CX3CL1, IL1B, KIT, SELE, SERPINE1, TNFSF10, TYMP and VCAM1 (Table 4.2).We found that the presence of CRIM1 had an attenuating effect on the expression of many of these genes compared to CRIM1 knockdown (Figure 4.15A), and could likely abrogate inflammation and promote survival of endothelial cells.For example, the expression of pro-inflammatory and pro-apoptotic markers like CCL5, IL1-B, SELE, TNFSF10 and VCAM-1, which is increased in CRIM1 knockdown HCMVECs, is reduced in HCMVECs in which CRIM1 was overexpressed.The limited ability of the endothelium to repair itself has made it critical to uncover molecules that can counter or prevent endothelial dysfunction. Thus, the presence of Crim1 could counter the pathophysiology of various diseases that are directly involved with endothelial abnormalities, such as atherothrombosis, hypertension and kidney failure, while maintaining homeostasis of the cardiovascular system and heart function on the whole(Pennisi, Wilkinson et al. 2007, Rajendran, Rengarajan et al. 2013, Morbidelli, Donnini et al. 2016).
106 Table 4.2: 10 endothelial genes had more than a 3-fold regulation in the CRIM1 knockdown assay. These 10 genes were assessed again in HMCVECs that had CRIM1 overexpressed, revealing a change in fold regulation upon CRIM1 overexpression that is mostly beneficial in the context of endothelial cell inflammation and death.
NUMBER GENE FOLD REGULATION P-VALUE 1 SERPINE1 1.315 0.101 2 VCAM1 1.46 0.0039 3 TYMP 1.90 0.0018 4 KIT 2.529 0.0005 5 CX3CL1 14.407 0.0086 6 CCL5 3.46 0.0011 7 CASP1 1.64 0.1340 8 IL1B -3.07 0.0028 9 SELE -2.614 0.003 10 TNFSF10 2.566 0.0001
4.15: Effect of CRIM1 overexpression in HCMVECs compared to CRIM1 knockdown HCMVECs A, Overexpression of CRIM1 ameliorated the effect of CRIM1 knockdown in HCMVECs as assessed using 10 endothelial genes that had more than a 3-fold change upon initial CRIM1 knockdown in HCMVECs. (n=3).
Cell-autonomous role for Crim1 in regulating BMP signalling, in vivo. Having observed a reduction in BMP signalling in endothelial cells of Crim1 conditional null mutants and HCMVECs that have lost CRIM1, we further investigated whether this was a cell-
107 autonomous defect by using Crim1FLOX/FLOX; R26R; Tie2-Cre mutant hearts. However, contrary to expectations, we found that there was a marked increase in phospho-SMAD1/5 in mutants compared to Crim1+/FLOX; R26R; Tie2-Cre controls (Figure 4.16A-J), in all cells (Figure 4.16G, H) as well as exclusively endothelial cells (Figure 4.16I, J).
Figure 4.16: Increase in canonical BMP signalling in cells of Crim1FLOX/FLOX; Tie2-Cre hearts as assessed with phospho-SMAD1/5 A-F, Confocal images of ventricular sections of 17.5 dpc hearts from ventricles of Crim1+/FLOX; Tie2-Cre (A- C)andCrim1FLOX/FLOX; Tie2-Cre (D-F)hearts, immunostained for PECAM-1 and phospho-SMAD1/5.A, C, D,
108 F,Merged images of DAPI (blue), PECAM-1 (green) and pSMAD1/5 (red). C, F, Magnified images showing vessel with increased phospho-SMAD1/5 levels within endothelial cells in Crim1FLOX/FLOX; Tie2-Cre hearts (F) compared to control (C). G, H, Increase in phospho-SMAD1/5 signalling in Crim1FLOX/FLOX; Tie2-Crehearts in all cells. I, J, Increase in phospho-SMAD1/5 signalling in Crim1FLOX/FLOX; Tie2-Crehearts in endothelial cells. (n=6). RV, right ventricle; ep, epicardium; en, endocardium. *P<0.05. Scale bars A, B, D, E, 50 μm, C, F, 10 μm.
We also examined phospho-SMAD2 levels in mutant hearts compared to controls but there was no significant change, indicating that canonical TGFβ signalling is largely independent of Crim1 (Figure 4.17A-F).
Figure 4.17: No change in canonical TGF signalling in endothelial cells of Crim1FLOX/FLOX; Tie2-Cre hearts as assessed with phospho-SMAD2 A-F, No change in phospho-SMAD2 levels as a proxy to assess canonical TGF signalling in ventricles of Crim1+/FLOX; Tie2-Cre andCrim1FLOX/FLOX; Tie2-Cre hearts. A-D, Confocal images of ventricular sections of 17.5 dpc hearts fromCrim1+/FLOX; Tie2-Cre (A, B) and Crim1FLOX/FLOX; Tie2-Cre(C, D) embryos, immunostained for PECAM-1 and phospho-SMAD2.A, C, Merged images of DAPI (blue), PECAM-1 (green) and pSMAD1/5 (red).E, F No change in phospho-SMAD2 levels in Crim1FLOX/FLOX; Tie2-Creendothelial cells (n=4-6).LV, left ventricle; en, endocardium; ep, epicardium. Scale bars, A, C, 50 μm.
Increase in phospho-SMAD1/5 signalling may not directly be due to paracrine stimulation by BMP2/BMP4. In light of the fact that there was an increase in phospho-SMAD1/5 levels in Crim1FLOX/FLOX; R26R; Tie2-Cre hearts at 17.5 dpc (Figure 4.16A-J), it could indicate that in these mutants, the
109 ventricular cells that retain Crim1 function could provide BMP stimulation so as to overcompensate for endothelial loss of Crim1. To address this hypothesis further, we stimulated HUVECs that had CRIM1 ablated with exogenous BMP2 and BMP4. Addition of BMP2 and BMP4 did not cause an increase in phospho-SMAD1/5 levels CRIM1 knockdown HUVECs compared to scrambled shRNA controls (Figure 4.18A, B). We also treated the cells with LDN193189, an inhibitor of BMP receptors ALK2 and ALK3 that mediate downstream BMP signalling, along with BMP2 and BMP4. Addition of LDN alone suppressed phospho-SMAD1/5 signalling in only CRIM1 knockdown HUVECs (Figure 4.18A, B). Addition of exogenous BMP2 and BMP4 to LDN-treated cells did not significantly increase phospho-SMAD1/5 levels between CRIM1 knockdown and control cells, indicating that LDN is efficient in inhibiting downstream signalling via ALK2/3 (Figure 4.18B). Thus, our data indicates that there could be another member of the BMP family that is responsible for the endothelial increase in phospho- SMAD1/5 that was observed in Crim1FLOX/FLOX; R26R; Tie2-Cre hearts.
Figure 4.18: Addition of exogenous BMP-2/4 does not exert a paracrine effect on CRIM1 knockdown HUVECs A, Change in pSMAD1/5 levels as a proxy to assess BMP signalling in the presence of inhibitor LDN193189 and exogenous BMP-2/4. B, Compared to PBS-treated scrambled shRNA control HUVECs, there is an increase in pSMAD1/5 upon treatment of scrambled shRNA control HUVECs with exogenous BMP-2 and decrease in pSMAD1/5 in PBS-treated, LDN-treated, LDN + BMP-2 treated, and LDN + BMP-4 treated CRIM1 knockdown HUVECs, by way of a 1-way ANOVA (significance represented by #). Using t-tests, we see a decrease in pSMAD1/5 in PBS-treated CRIM1 knockdown HUVECs, and also in LDN-treated, LDN + BMP-2 treated, LDN + BMP-4 treated, BMP-2 CRIM1 knockdown HUVECs, compared to their respective scrambled shRNA control HUVECs (significance represented by *) (n=3).Representative blot chosen.*/#P<0.05, **P<0.01, ***/###P<0.001.
110 Discussion. We have investigated the function of Crim1 in coronary endothelial cell development in vivo, using both constitutive nulls and tissue-restricted knockouts. Specifically, the number of coronary vascular endothelial cells was reduced, and this was also observed in vitro in HCMVECs lacking CRIM1, which display decreased viability. However, further experiments are required to discern whether the loss of CRIM1 promotes apoptotic or necrotic endothelial cell death. We also provide evidence that in coronary endothelial cells, Crim1 modulates BMP signalling, as demonstrated by changes in downstream phospho-SMAD1/5 levels. Finally, we describe changes in endothelial gene expression, characteristic of endothelial dysfunction and disease, which may either cause, or result in, diminished endothelial cell numbers and disrupted growth factor signalling collectively. While we have described both a knockdown and overexpression of CRIM1 in HCMVECs, both these will have to be confirmed at the protein level by Western blotting in future experiments.
Cardiac endothelial cells are in close contact with other cells of the heart, importantly the cardiomyocytes and vascular smooth muscle cells (Brutsaert, Fransen et al. 1998). Essential factors produced by both endocardial and myocardial vessel endothelial cells have been shown to influence cardiomyocyte survival (Narmoneva, Vukmirovic et al. 2004), and also smooth muscle and cardiomyocyte performance (Brutsaert, Fransen et al. 1998, Brutsaert 2003). Reciprocal signalling between these cells is required for proper cardiac development (Brutsaert 2003, Hsieh, Davis et al. 2006). The regenerative potential of the heart has received considerable focus in recent years and there exist a variety of strategies aimed at engineering cardiac tissue (Narmoneva, Vukmirovic et al. 2004). Neovascularization is a critical process that aids myocardial regeneration following injury (Porrello, Mahmoud et al. 2013), and studies have shown that the interaction of cardiomyocytes and endothelial cells leads to enhanced cardiomyocyte network formation, coordinated beating, and overall spatial organization (Narmoneva, Vukmirovic et al. 2004). We reveal a reduction in endothelial cell number in the compact myocardium of constitutive and tissue-restricted mutants,in vivo. This is further confirmed in vitro, where there is a decrease in the number of viable HUVECs and HCMVECs in the absence of CRIM1. Moreover, in HCMVECs, inducers of apoptosis, such as CASP1 and TNFSF10 are upregulated, and molecules such as PF4, CAV1 and IL11, which decrease cell death, are downregulated.In light of this, a reduction in the number of endothelial cells in the absence of Crim1, as seen in both constitutive nulls as well as endothelial-restricted mutants, could imply sub-optimal conditions for the functioning of surrounding coronary smooth muscle cells and cardiomyocytes. It would be of interest to assess endothelial cell death in conditional nulls, as well as in surrounding cells, to further investigate this hypothesis. Additionally,
111 it would be worthwhile to assess vessel length and number of branchpoints in both embryonic and adult hearts lacking Crim1, to further examine the role of Crim1 in causing abnormal vessel morphology. While we did not find any discernable differences in the vasculature of Crim1Δflox/Δfloxhearts at earlier time points, more detailed analysis needs to be performed using a larger sample size, while also assessing apoptosis and proliferation of these cells.
The VWC domains in Crim1 are similar to those present in secreted proteins such as chordin (Sasai, Lu et al. 1994) that has been shown to function as BMP antagonists. Though Crim1 has been shown to antagonize BMP-7 and BMP-4 (Wilkinson, Kolle et al. 2003), its role in the heart appears to be different. Growth factor modulation by CRIM1 can occur via its various domains, but whether it acts as an agonist or antagonist is thus likely dependent on its environment and developmental context (Wilkinson, Kolle et al. 2003, Fung, Chi Fat et al. 2007, James and Broihier 2011).BMPs 2 and 4 are produced by endothelial cells, smooth muscle cells and cardiomyocytes and have both autocrine and paracrine functions on the same targets, via activation of smad1/5/8 (Morrell, Bloch et al. 2016). Inflammation and atherogenesis are stimulated due to heightened BMP signalling, especially by BMPs 2 and 4 (Jo, Song et al. 2006, Derwall, Malhotra et al. 2012, Morrell, Bloch et al. 2016). This implies that when BMP signalling is inhibited, both inflammation and atherogenesis are reduced. However, we observe an increase inexpression of markers associated with inflammation, accompanying reduced BMP signalling via the phospho-SMAD1/5 pathway in CRIM1 knockdown HCMVECs. Moreover, in conditional nulls, BMP signalling is also decreased. This is coincident with other studies that imply a protective effect of BMP-2 in atherosclerotic plaques(Nakaoka, Gonda et al. 1997), and BMP-9 in promoting vascular stability and preventing the death of pulmonary artery endothelial cells (Long, Ormiston et al. 2015). Contrary to these two models, in Crim1FLOX/FLOX;Tie2-Cre mice, BMP signaling via the phospho-SMAD1/5 pathway is increased. A caveat of this study, however, is that inflammation in Tie2-Cre mutants was not assessed. Irrespective of this, it may be likely that other cells in the myocardium that retain Crim1, could produce BMPs which would bind to receptors on endothelial cells, and activate downstream signaling. Furthermore, any differences observed in our studies and others could be a result of different characteristics displayed by endothelial cells arising from distinctive vascular networks from different areas (Aird 2007). Another explanation could be that because BMPs are required for endothelial cell development (Langenfeld and Langenfeld 2004, Liu, Selever et al. 2004, Suzuki, Montagne et al. 2008), any decrease in BMP signaling, could lead to a manifestation of endothelial stress, and any inflammatory responses and gene changes associated with this could be a secondary pathology.
112 There was no obvious external phenotypic defect, or change in animal health, in adult Tie2-Cre mutants. Thus, it would be interesting to subject Crim1FLOX/FLOX;Tie2-Cre mice to stress, injury or diet-dependent load to the metabolic system, and assess their response. We hypothesize that these mutants would be more prone to inflammation and subsequently, artherosclerotic lesions, based on our current observations. Another path to explore is the crossbreeding of Crim1 mutant mice with ApoE-/- or LDLr-/-mice, which are the most extensively used in atherosclerosis and hyperlipidemia research,to directly examine the role of Crim1 in vascular inflammation and atherogenesis. As previously mentioned, other growth factors and molecules secreted by surrounding cell types may compensate for the loss of Crim1 function in specifically endothelial cells, leading to the lack of an outward phenotype in the adulttissue-restricted mice, exacerbated only upon environmental changes.
The role that endothelial cells play in the development of atherosclerosis, which is a leading cause of coronary artery disease and risk factor for myocardial infarction, is firmly established, and dependent on endothelium activation (Hansson 2005). This activation of the endothelium is a result of both a response to, and a secretion of a large number of factors that in turn lead to the recruitment of leukocytes, neutrophils and granulocytes, eliciting an inflammatory response (Figure 4.19). It is evident from our results that there is a misregulation of these factors, including cell adhesion molecules, interleukins and other chemokines, when CRIM1 is ablated. CX3CL1, specifically expressed in endothelial cells, may be a marker for artherosclerosis (Umehara, Bloom et al. 2004), and is upregulated in CRIM1-deficient endothelial cells. Coincident with a chronic inflammatory state that often precedes atherogenesis, pro-inflammatory mediators like IL1-B, SELE, CCL5 and VCAM-1 are also upregulated, and this likely by regulators such as TNFs and RELA (of the NFB) family that function upstream of these genes (Gearing and Newman 1993, Hajra, Evans et al. 2000, Aird 2007, Galkina and Ley 2009, Frostegård 2013, Gimbrone and García-Cardeña 2016). It has been previously established that a pro-inflammatory response and secretion of cytokines by endothelial cells stimulates vascular smooth muscle cell proliferation, and that aberrant proliferation promotes plaque and lesion formation (Rudijanto 2007). CRIM1 knockdown HCMVECs may also possessthe potential to increase smooth muscle cell proliferation based on an increased expression of endothelial genes such as CCL5, CX3CL1(Perros, Dorfmuller et al. 2007), ILIB and TNFSF10; this proliferation is characteristic of atherosclerosis progression (Rudijanto 2007), and could implicate CRIM1 in this disease. Furthermore, CRIM1 overexpression could have a therapeutic effect in attenuating a pro-inflammatory state based on the comparison of the expression of target genes examined in our study, making it an exciting avenue for future research.
113
Figure 4.19: Schematic representation of endothelial proinflammatory activation In lesion-prone regions of the arterial vasculature, the actions of proinflammatory agonists (eg, interleukin [IL]-1, tumor necrosis factor [TNF], and endotoxin), oxidized lipoproteins (ox-LDL) and advanced glycation end products (AGE), as well as biomechanical stimulation by disturbed blood flow, leads to endothelial activation. These biochemical and biomechanical stimuli signal predominantly via the pleiotropic transcription factor nuclear factor-κB (NF-κB), resulting in a coordinated program of genetic regulation within the endothelial cell. This includes the cell surface expression of adhesion molecules (eg, vascular cell adhesion molecule-1 [VCAM-1), secreted and membrane-associated chemokines (eg, monocyte chemotractant protein [MCP]-1 and fractalkine (CX3CL1), and prothrombotic mediators (eg, tissue factor [TF], vonWillebrand Factor [vWF], and plasminogen activator inhibitor [PAI]-1 (called SERPINE1 in humans)). These events foster the selective recruitment of monocytes and various types of T lymphocytes, which become resident in the subendothelial space. The concerted actions of activated endothelial cells, smooth muscle cells, monocyte/macrophages, and lymphocytes result in the production of a complex paracrine milieu of cytokines, growth factors, and reactive oxygen species (ROS) within the vessel wall, which perpetuates a chronic proinflammatory state and fosters atherosclerotic lesion progression. IL-R, TNF-R indicates receptor(s) for IL-1, TNF; Ox-LDL-R, receptor for oxidized LDL; RAGE, receptor for AGE; and TLRs, Toll-like receptors (Illustration Credit: Ben Smith). (Adapted from (Gimbrone and García-Cardeña 2016)).
In conclusion, in Crim1-mutant mice we observed a reduced number of endothelial cells and dysregulation of the phospho-SMAD1/5 pathway involved in BMP signal transduction. Subsequently, we have elucidated that the BMP pathway is aberrant in CRIM1 mutant human cardiac endothelial cells; this will help provide novel information on the specific roles played by
114 different growth factors in coronary vascular endothelial cell development in the future. The absence of CRIM1 also perturbs the expression of a large number of genes involved in progression of endothelial disease states. Collectively, our findings provide a basis for understanding the molecular mechanisms of Crim1 dysfunction in endothelial cells of the heart and identify Crim1 to be a critical regulator of endothelial cell function.
Acknowledgements. Tie2-Cre mice were a kind gift from Prof. Richard Harvey at the Victor Chang Cardiac Research Institute.
115
Chapter 5
Investigating whether the IGFBP domain of CRIM1 can serve to bind IGFs and modulate downstream signalling
116 Author contributions Primers for generation of CRIM1 deletion constructs using InFusion cloning were designed by Dr. Aaron Smith. Western blotting was performed in association with Dr. Yash Chhabra. All densitometry quantifications were performed blinded by Dr. Yash Chhabra.
117 Introduction A large number of growth factors function cooperatively to mediate a variety of cellular processes, and their timely regulation is crucial for proper development to ensue. This makes it essential for us to investigate novel modulators of these growth factors, to uncover mechanisms by which developmental aberrations that occur due to impaired growth factor signalling can be alleviated. CRIM1 possesses six CRRs, via which it has been shown to bind growth factors such as BMPs, TGFβ and VEGF (Wilkinson et al., 2003, Wilkinson et al., 2007), along with an IGFBP-like domain whose functional relevance is not yet known (Kolle et al., 2000). In light of this, we hypothesize that CRIM1 can bind another family of growth factors, the IGFs, via its CRRs and its IGFBP domain.
While VEGFs and BMPs receive maximum attention in the context of endothelial cell biology, another essential growth factor that is required for normal endothelial cell function is the IGF. Both autocrine and paracrine functions for IGFs are known, and their cellular effects are widespread (Jones and Clemmons, 1995). Two IGF ligands exist, IGF-1 and IGF-2, which bind to the IGF-1 and IGF-2 receptors. These also bind to the insulin receptor, based on similar sequence homology (Jones and Clemmons, 1995). The IGF receptor is a heterotetramer, with the 2 extracellular α subunits participating in ligand binding and the two intracellular β subunits possessing tyrosine kinase activity (Laustsen et al., 2007), and triggers a signalling cascade upon phosphorylation, leading to the activation of the MAPK, ERK1/2 and AKT pathways (Baserga, 1999) and consequently, multiple cellular responses including cell proliferation, death, differentiation and overall function (Jones and Clemmons, 1995).
Differential alternative splicing of the IGF-1 gene produces transcripts coding for different IGF-1 precursor polypeptides; these Ea, Eb and Ec variants of IGF-1 add to the diversity of this growth factor (Delafontaine et al., 2004). Most of the actions mediated by IGF-1 are a result of the binding of the mature IGF-1 peptide to the IGF receptor; the bioactive IGF-1 peptide is generated upon post-translational modifications involving removal of the C-terminal E peptides (Hede et al., 2012). The significance of each of these peptides is unclear; however, their existence in different tissues implies the complex role of these IGF-1 isoforms under varying conditions (Philippou et al., 2014). Interestingly, the IGF-1 Eb peptide, also called mechano-growth factor (MGF), has been shown to improve cardiac function and preserve blood supply to heart muscle (Carpenter et al., 2008).
The amount of IGF-1 and IGF-2 produced by endothelial cells is low, but their IGF-1 receptor expression is abundant (Delafontaine et al., 2004, Back et al., 2012). Endothelial cells also express
118 IGF Binding proteins, which modulate IGF activity (Hwa et al., 1999). This occurs locally, or is systemic, based on endocrine control over circulating IGFs (Yamada and Lee, 2009). The bioavailability of IGFs is controlled by IGFBPs, as is their half-life. IGFBPs can also sequester IGFs, thereby limiting the activation of IGF receptors, or act as vehicles for IGF trafficking (Hwa et al., 1999). It is important to note that both the IGF receptor and IGFBPs contain cysteine-rich regions that are required for binding IGFs (Lee et al., 1997,Keyhanfar et al., 2007).
Thus, the CRRs are a common thread that likely link the binding of IGF-1 to the IGFR, IGFBPs, and potentially, CRIM1. Furthermore, IGFs secreted from the epicardium have been shown to influence development of the ventricular chambers in the embryo (Li et al., 2011). In light of the fact that Crim1 is strongly expressed in the epicardium, and that it has both CRRs and an IGFBP domain, led us to hypothesize that it can bind IGFs, and modulate their activity.
In the current chapter, we reveal for the first time, that CRIM1 can indeed bind IGF-1. Moreover, the examination of the phosphorylation status of the IGF-1 receptor in the absence of CRIM1 also shows that CRIM1 modulates IGF-1 signalling in human cardiac endothelial cells. Taken together, our studies demonstrate aputative role for CRIM1 in regulating the activity of yet another family of growth factors, the IGFs.
Materials and methods Cell Culture H-CMVECs and pooled HUVECs were cultured in EGM-2MV and EGM-2 BulletKit media respectively. Cells before passage six were used in all experiments. Lentiviral particles containing shRNA targetingCRIM1 mRNA were synthesized, and used to infect cell populations as described previously.Prior to examining the effect of exogenous IGF-1 stimulation, cells were starved in serum-free basal media without growth factors for 2 hours.
Antibodies Primary antibodies used in the study were: phospho-IGF-1Rβ (#3024S; Cell Signalling), phospho- AKT (#4060; Cell Signalling); phospho-ERK1/2 (#4370; Cell Signalling); total AKT (#9272; Cell Signalling); total ERK1/2 (sc-93; Santa Cruz); total IGF-1R (sc-712; Santa Cruz); GAPDH (#2118; Cell Signalling); -tubulin (#2146, Cell Signalling); rabbit monoclonal HA tag (#3724; Cell Signalling); mouse monocolonal myc tag (#2276; Cell Signalling); normal mouse IgG2a isotype control (#02-6200; Thermo Fisher Scientific). Secondary antibodies were rabbit HRP (#W4011;
119 Promega), mouse HRP (#W4021; Promega), IgG mouse IRDYE (#926-32212; Licor) and IgG rabbit IRDYE (#926-68073; Licor).
Generation of deletion constructs Full-length myc-tagged CRIM1(Wilkinson et al., 2003) was used as a template to generate myc- tagged CRIM1 deletion constructs. Primers were designed as follows:
CRIM1-3.1-InF-RV-F 5’-TGGAATTCTGCAGATATGTACTTGGTGGCGGGG-3’
CRIM1-3.1-InF-RV-R 5’-GCCACTGTGCTGGATTCATAGGTCTTCTTCTGA-3’
CRIM1-IGFBPsplice-InF-F 5’-GAGAACTGGACTGATGAC-3’
CRIM1-IGFBPsplice-InF-R 5’-ATCAGTGCAGTTCTCCGCCCGGGTGCCGGAGCGCGC-3’
CRIM1-D1-2-IF-F 5’-GAAGAACCAACCATCATC-3’
CRIM1-D1-2-IF-R 5’-GATGGTTGGTTCTTCGGCTGGCTTTGTATCATTAAC-3’
CRIM1-D3-6-IF-F 5’-CCAGAAATGTATGTCCCA-3’
CRIM1-D3-6-IF-R 5’-GACATACATTTCTGGAGTGCCCGACAGGATGGGTGG-3’
CRIM1-D1-6-IF-R 5’-GACATACATTTCTGGGGCTGGCTTTGTATCATTAAC-3’
Cloning was performed using the In-Fusion Kit (#638910). Ensembl transcript ID CRIM1-001 ENST00000280527 was used as a reference for analysis. For CRIM1ΔIGFBP, amino acids 34-111 (bp 102 – 333) were deleted, for CRIM1Δ1-2, amino acids 335 – 456 (bp 1005 – 1367) were deleted. For CRIM1Δ3-6, amino acids 607 – 873 (bp 1821 – 2619) were deleted. For CRIM1Δ1-6, amino acids 335 – 873 (bp 1005 – 2619) were deleted. For CRIM1ΔIGFBPΔ1-6, amino acids 34 – 111 (bp 102 – 333) and amino acids 335 – 873 (bp 1005 – 2619) were deleted. Constructs were sequenced to confirm deletion.
Briefly, the vector pcDNA3.1 was linearized by ECORV digestion, gel purified and run on a gel to check for complete digestion. Wildtype CRIM1 was generated using InFusion primers, run on a gel
120 to check the size of product, and ligated using the InFusion kit. The Ligation was performed at 50°C for 15 minutes and placed on ice. The ligated mixture was used to transform competent cells, and plated on agar plates containing Ampicillin. Colonies were observed the next day, and picked for plasmid mini-prep experiments. The colonies were also digested with restriction enzymes HindIII and XhoI, which are non-cutters of CRIM1 and single, unique cutters of pcDNA3.1. Bands were observed at the right sizes. The protocol was repeated for generation of all deletion constructs.
Co-Immunoprecipitation HEK293 at 80% confluence were transfected with full length myc-tagged CRIM1(Wilkinson et al., 2003), myc-tagged CRIM1 deletion constructs, and HA-tagged IGF-1 Ea and IGF-1 Eb constructs (Hede et al., 2012) using Lipofectamine 2000 (Invitrogen). After 24 hours, the cells were lysed in RIPA containing protease inhibitor and phosphatase inhibitor on ice. Immunoprecipitation was performed by the addition of α-myc antibody for 2 hours at 4°C, followed by addition of 30 μL of Protein A Agarose (Roche, # 11719408001). The mix was further incubated for 1 hour at 4°C. The beads were pelleted and washed five times with cold PBS 0.1% Triton X-100. Protein was eluted by addition of SDS-PAGE loading buffer and boiled for 4 minutes.
Western Blotting Control and CRIM1-knockdown HCMVECs were lysed in RIPA buffer containing protease inhibitor and phosphatase inhibitor. The lysates were heated at 95°C for 4 minutes with SDS-PAGE loading sample buffer, run on a 10% SDS-polyacrylamide gel and transferred onto PVDF membranes. Membranes were blocked in 5% BSA, probed with primary and secondary antibodies and detected using ECL.
Data Documentation, Quantification and Analysis Immunoblots were visualised either with a chemi-luminescent reagent (Merck Millipore) via exposure to X-ray film, or by using the Licor scanner. All densitometry quantitation was performed on immunoblots using ImageJ software. All test values were normalised to respective loading and experimental controls and graphed with Prism 6.
Specifically for COIP quantification, IP-myc intensity = x. IB-HA intensity = y. y/x = z. total cell lysate-CRIM1/GAPDH =a. total cell lysate-IGF/GAPDH = b. a/b=c. z/c = plotted values for assessing binding of IGF-1 Eb to different CRIM1 deletion constructs. IP refers to CRIM1 pulldown with Myc, and IB refers to probing for IGF-HA.
121 Statistical significance between pooled data from mutant and control groups was determined using a two-tailed, unpaired Student’s t-test or a standard one-way ANOVA with Dunnett’s post-test, as specified in the figure legend. For coimmunoprecipitation quantification, a standard one-way ANOVA with Dunnett’s post-test was used. Quantified data shown are mean ± standard deviation. GraphPad Prism v6.0 was used to conduct all analyses and generate graphs.
Results CRIM1 can bind IGF-1 Although evidence suggests that CRIM1 can bind IGFs, based on the presence of the IGFBP domain of CRIM1 as well as its CRRs, such an interaction has never been reported. Thus,we hypothesised that CRIM1 could bind IGFs and modulate their activity. To address this, we first investigated whether CRIM1 could interact with IGF-1 intracellularly when transiently co- transfected in HEK293 cells. Two splice variants of IGF-1, IGF-1 Ea-HA and IGF-1 Eb-HA, were co-immunoprecipitated with CRIM1-myc (Figure 5.1A). This suggests that full-length CRIM1 can bind to IGFs.
Figure 5.1: COIP revealing that CRIM1 can bind IGF-1 Ea and IGF-1 Eb constructs A, HEK293 cells transiently transfected with full length CRIM1-myc and IGF-1 Ea-HA and IGF-1 Eb-HA. Immunoprecipitation performed with anti-myc and immunoblotting with anti-HA. IB bands were detected only when both CRIM1 and IGF-1 Ea/Eb were co-transfected. (n=3).
122
Generation of CRIM1 deletion constructs CRIM1 possesses six CR motifs, in addition to an IGFBP-like domain, and could thus bind IGFs via these domains as well. To investigate this, myc-tagged CRIM1 WT and deletion constructs were generated as outlined in the Materials and Methods section. Specifically, deletion constructs lacking the IGFBP domain, CRRs 1-2, 3-6, 1-6 and 1-6 along with the IGFBP domain were made (Figure 5.2). Product sizes were confirmed on a gel (Figure 5.3A-C), and constructs were sequenced to confirm deletions.
Figure 5.2: Schematic representation of CRIM1 deletion constructs A, Representation of CRIM1 deletion constructs as generated using InFusion cloning. 1. Full length CRIM1, 2. CRIM1IGFBP lacking the IGFBP domain, 3.CRIM1lacking CRRs 1 and 2, 4. CRIM1lacking CRRs 3, 4, 5 and 6, 5. CRIM1lacking all six CRRs. 6. .CRIM1IGFBPlacking the IGFBP domain and all six CRRs. Key: Green box – IGFBP domain Orange box – RGD motif Grey box – VWFC domain Red box – antistasin domain Black double line – trannsmembrane domain
123
Figure 5.3: Generation of CRIM1 deletion constructs A-C, Gel images of CRIM1 deletion constructs after restriction digestion with appropriate enzymes to confirm correct ligation and sizes of the constructs after InFusion cloning.
124 CRIM1 binds IGF-1 via its IGFBP-like domain, and CRRs 3-6 Full length CRIM1-myc and all CRIM1-myc-tagged deletion constructs were co-transfected in HEK293 cells with IGF-1Eb-HA, due to its suggested role in the heart (Carpenter et al., 2008). In each case, IGF-1 Eb-HA was immunoprecipitated with CRIM1, but at varying levels. There was a significant decrease in IGF binding in the CRIM1 construct lacking the IGFBP domain, and CRRs 3-6 (Figure 5.4A, B). In the construct lacking all six CRRs, binding was significantly reduced, and almost absent in the construct lacking the IGFBP domain in addition to the six CRRs (Figure 5.4 A,B). This confirmed that as in the case of BMPs, VEGF and PDGF, the CRRs are required to mediate the binding of CRIM1 to IGF-1, along with the IGFBP domain.
Figure 5.4A:Binding of IGF-1 Eb to CRIM1 deletion constructs A, HEK293 cells transiently co-transfected with full length CRIM1-myc, CRIM1IGFBP- myc,CRIM1mycCRIM1mycCRIM1mycand CRIM1IGFBPmyc, and IGF-1 Eb-HA. Control cells transiently co-transfected with only CRIM1-myc and only IGF-1 Eb-HA. Immunoprecipitation performed with anti-myc and immunoblotting with anti-HA. IB bands were detected only when both CRIM1 and IGF-1 Ea/Eb were co-transfected. (Three independent IPs performed. Representative blot chosen).IgG controls were utilized in the first out of the three independent experiments performed, along with ‘no antibody’ controls and CRIM1-Myc-only and IGF-HA-only transfection controls, which were used in all three experiments.
125
Figure 5.4B: CRIM1 binds IGF-1 Eb via the IGFBP domain and CRRs 3-6 B, CRIM1 binds IGF-1 Eb via the IGFBP domain and CRRs 3-6, as indicated by 1-way ANOVA analysis, which reveals that deletion of the IGFBP domain and CRRs 3-6 of CRIM1 significantly reduces binding of CRIM1 to IGF-1 Eb. Deletion of CRRs 1-2 does not affect binding to IGF-1 Eb. (n=3). **P<0.01, ***P<0.001
Interaction of CRIM1 with IGFs, and modulation of IGF signalling Using IGF-1 receptor phosphorylation as a proxy for IGF signal transduction, we examined whether CRIM1 could regulate the same. The level of phospho-IGF-1Rβ was significantly reduced in CRIM1 knockdown HCMVEC protein lysates (Figure 5.5 A,B), implying that CRIM1 could serve to bind and present IGFs to their receptors, thereby increasing downstream signalling. Levels of phospho-AKT were unchanged (Figure 4.13A, B), indicating that endogenous IGF-1 may signal via another downstream pathway.
126
Figure 5.5: CRIM1 modulates IGF-1 signalling in HCMVECs Reduction in phospho-IGF-1 Rlevels as a proxy to assess IGF-1 signalling in CRIM1 knockdown H-CMVECs, indicating misregulation of the IGF-1 pathway in the absence of CRIM1 (n=3). **P<0.01
Effect of IGF-1 stimulation in the absence of CRIM1 The level of endogenous IGF-1 expression in endothelial cells is low (Tucci et al., 1998. Therefore, we sought to determine whether addition of exogenous IGF-1 would have an effect on IGF-1 signal transduction in the presence and absence of CRIM1. This would allow us to assess whether CRIM1 can regulate IGF-1 signalling when the growth factor is supplied by an external source. Therefore we stimulated HUVECs with recombinant IGF-1 to assess phosphorylation of the IGF receptor (Figure 5.6A-C). We also examined AKT phosphorylation in IGF-1 stimulated HUVECs, in the presence and absence of CRIM1 (Figure 5.6A, D, E). We found that phospho- IGF-1Rβ levels continued to be low in both unstimulated and stimulated CRIM1 knockdown cells (Figure 5.6 B, C), but interestingly, levels of phospho-AKT, which were unchanged in unstimulated CRIM1 knockdown and control HCMVECs (Figure 4.13A, B) and HUVEC lysates (Figure 5.6D), were also decreased in IGF-1 stimulated CRIM1 knockdown HUVECs (Figure 5.6E). Furthermore, addition of IGF1 to scrambled shRNA control cells also increased phospho- AKT levels compared to untreated cells (data not shown), implying that the AKT pathway is employed when a sufficient amount of IGF-1 is present.
127
Figure 5.6: Change in downstream IGF signalling upon treatment of CRIM1 knockdown HUVECs with exogenous IGF-1 A, CRIM1 knockdown and control scrambled shRNA HUVECs treated with exogenous IGF-1. B, In unstimulated CRIM1 knockdown HUVECs, levels of phospho-IGF-1Rwere reduced. C, In both CRIM1 knockdown and control scrambled shRNA HUVECs, level of phospho-IGF-1Rwere increased compared to unstimulated conditions, but CRIM1 knockdown HUVECs still displayed lower levels of phospho-IGF-1RD, In unstimulated CRIM1 knockdown
128 HUVECs, levels of phospho-AKT were unchanged compared controls. E, In IGF-1 stimulated CRIM1 knockdown and control scrambled shRNA HUVECs, levels of phospho-AKT were increased compared to unstimulated conditions, accompanied by a decrease in phospho-AKT in CRIM1 knockdown HUVECs compared to control. (n=2). Representative blot chosen*P<0.05, **P<0.01
Discussion IGFBPs mainly function as regulators of IGFs (Hwa et al., 1999); thus, the presence of an IGFBP domain in CRIM1 suggested that it would solely mediate binding to IGFs. CRRs are known to mediate binding of TGFβ superfamily members to other proteins that modulate their function (Larraín et al., 2000). Furthermore, it has also been revealed that cysteine-rich regions, which are present in the IGF receptor and IGFBPs as well, are essential for IGF binding (Lee et al., 1997,Keyhanfar et al., 2007). This is the first study revealing that CRIM1 can bind IGF-1, via both its IGFBP domain and CRRs 3-6. It is noteworthy that IGFBP-7 of the IGFBP family has the highest sequence similarity to CRIM1 (Kolle et al, 2000). We have demonstrated that like IGFBPs, CRIM1 modulates IGF signalling, but the mechanism by which it does so, is not yet established. This is a novel finding, which will have implications in most tissues and organs, due to the involvement of IGFs in governing development and regulation of normal physiology, homeostasis and disease states.
How does our finding impact vascular development and disease? Neovascularization in murine retinas was reduced upon IGF-1 receptor inhibition, and this occurred in part due to a reduction in VEGF-induced MAPK activation (Smith et al., 1999). Moreover, there was a decrease in endothelial cells in human fetal lung explants when IGF signalling was disrupted by blocking the IGF receptor (Han et al., 2003). IGF-1 has previously been known to enhance angiogenesis, likely through enhancing the expression of other angiogenic factors such as VEGFs and TGF(Punglia et al., 1997, Miele et al., 2000), and exogenously administered IGF-1 was found to have a protective function against myocardial ischemia (Lisa et al., 2011). Interestingly, it has been found that low levels of IGF in circulation are associated with coronary artery disease (Spallarossa et al., 1996), and an increased risk of ischemic heart disease (Juul et al., 2002).
HCMVECS have been shown to be IGF-1-responsive, but not sensitive to insulin, possibly due to the abundance of IGF-1Rs on these cells (Chisalita and Arnqvist, 2004). The role of IGF-1 in restraining atherosclerosis has been revealed, by reducing senescence in vascular endothelial cells, caused by oxidative stress (Higashi et al., 2013). Similarly, IGF-1 infusion was demonstrated to counter inflammation and atherosclerosis progression in mouse models, further confirming its role as an anti-atherogenic factor (Sukhanov et al., 2007). IGF-1 has also been shown to reduce the
129 expression of IL1B and TNFconsequently abrogating a pro-inflammatory response (Spies et al., 2001).
Similarly, IGF-1 was found to decrease levels of proteins implicated in cellular ageing (Torella et al., 2004). The senescence of cardiac stem cells, which leads to them being unable to form new myocyte populations, impedes myocardial regeneration. However, when the expression of IGF-1 was increased, cardiac stem cells re-entered the cell cycle and displayed improved telomerase activity and delayed senescence (Torella et al., 2004). Furthermore, this is seemingly caused by a reduction in phospho-AKT, which is also rescued in part by IGF-1 overexpression (Torella et al., 2004). Therefore, the IGF-1 pathway in cardiac stem cells is likely an effective route to counter cardiac ageing and enhance myocardial regeneration (Torella et al., 2004). Other studies also implicate IGF-1 in cardiomyocyte survival after infarction, by preventing further cell death in the myocardium (Li et al., 1997). We have demonstrated that an absence of CRIM1 leads to reduced IGF-1 signalling; this implies that CRIM1 serves as an agonist of IGF-1, and furthermore, the presence of CRIM1 also enhances AKT levels upon IGF-1 stimulation.
The role of Crim1 in the epicardium and myocardium of nascent hearts in relation to IGF modulation has not been studied. This can be further explored by using Crim1-and-IGF double knockouts and assessing myocardial development to determine whether Crim1 and IGFs function along similar pathways. Additionally, we can investigate whether CRIM1 modulates IGF-1 in cultured epicardial cells.
We have demonstrated that an absence of CRIM1 leads to reduced IGF-1 signalling; this implies that CRIM1 serves as an agonist of IGF-1. A decrease in IGF-1R phosphorylation as seen in CRIM1 knockdown H-CMVECs does not necessarily indicate a decrease in levels of secreted or circulating IGFs, but might mean that IGF-1 requires to be presented to its receptor effectively to mediate downstream signalling. Another avenue to explore in the future would be to assess whether absence of CRIM1 reduces the amount of secreted, mature IGF-1. Also, it would be interesting to assess whether overexpression of CRIM1 increases IGF-1 signalling.
A caveat of this study and other studies performed examining the binding of CRIM1 to various growth factors is that both were co-expressed by the same cell (Wilkinson et al., 2003, Wilkinson et al., 2007). Thus, it is essential to examine the interaction of CRIM1 and growth factors produced by different cells. This will heavily influence any interpretations, and provide insight into the role of CRIM1 in growth factor modulation, as well its biological relevance in development and disease. In
130 light of all evidences suggesting that IGF-1 protects against ventricular and endothelial dysfunction, and progression of atherosclerosis(Higashi et al., 2014), our data posits a novel mechanism by which CRIM1 and IGF-1 could function in tandem as cardioprotective factors. This could constitute a unique approach to developing new therapies to prevent vascular disease or improve outcomes after ischemic myocardial damage.
Acknowledgements We are grateful to Prof. Melissa Little and Prof. Nadia Rosenthal for providing us with the CRIM1 and IGF-1 Ea /Eb expression constructs, respectively.
131
Chapter 6
General discussion and future directions
132 The role of Crim1 in developmental organogenesis is well-established. However, the mechanism through which it acts remains unclear, most critically with respect to its role in mediating growth factor signalling in different developmental contexts. For example, in C. elegans, crm-1 acts as an non-cell-autonomous agonist of a BMP-like pathway, specifically the DBL-1 pathway (Fung et al., 2007). It is not known whether this occurs through a ligand- or receptor-based interaction. In contrast, in Drosophila, the homolog of Crim1, CRIMPY, antagonises the function of the BMP ligand, Glass bottom boat (Gbb), in motorneurons at the neuromuscular junction (NMJ), restraining its expansion (James and Broihier, 2011). Synaptic development appears to be regulated by a preferential association of the full-length Gbb precursor with CRIMPY, and Gbb is then secreted from the motorneuron terminal (James and Broihier, 2011). Similarly, in transfected COS7 cells, CRIM1 binds to and antagonistically modulates the processing of pre-BMPs and the secretion of mature BMPs(Wilkinson et al., 2003). Thus, the function of Crim1 in organogenesis is context- dependent; prominent variables influencing its overall biological function include the timing and location of Crim1 expression and interactions with local growth factors.We have sought to determine the role played by Crim1 in cardiac development by employing various transgenic mouse lines, including tissue-restricted mutants.
Crim1 in epicardial development There is much interest in harnessing the potential of the mostly quiescent adult mammalian epicardium. Its activation and mobilisation in response to myocardial and ischemic damage, including through re-expression of embryonic epicardial genes, releasing paracrine and autocrine factors from the epicardium or epicardium-derived cells, and through therapeutic revascularization that could contribute to the coronary vasculature, is crucial to repair following injury or stress (Smart et al., 2007, Zhou et al., 2011a). The expression of Crim1 in the epicardium and epicardium- derived lineages is noteworthy. There is significant contribution from epicardium-derived cells to the developing heart, as well as to the fibroblast and smooth muscle cell lineages following myocardial damage (van Wijk et al., 2012, Duan et al., 2012, Limana et al., 2011, Huang et al., 2012, Zhou et al., 2011a). Furthermore, the epicardium aids the growth and expansion of the developing myocardium in a paracrine manner through provision of signalling cues and reciprocal signalling.
Aberrations in epicardial morphology during the early stages of heart development in Crim1∆flox/∆floxmutants imply defective epicardialEMT as well as adhesion. This was confirmed by employing the use of transgenic models displaying epicardium-specific Crim1 deletion, which showed increased epicardial EMT and EPDC invasion into the myocardium. Moreover, in the
133 absence of Crim1, an increased migration of primary epicardial cultures was demonstrated. Additionally, βcatenin distribution at epicardial cell junctions was altered. Combined with the results of previous studies implicating the indirect interaction of the cytoplasmic domain of Xenopus Crim1 β-catenin and N-cadherin in the formation and maintenance of cadherin-dependent complexes at epithelial cell junctions (Ponferrada et al., 2012), it may be inferred that Crim1 plays a role in stabilising junctional complexes. The link between these two molecules in the context of interactions between the epicardium and myocardium during development is an attractive avenue of research that has not yet been explored. Using primary epicardial cells and ablating Crim1, β- catenin localisation at junctions can be assessed. Crim1 binding to β-catenin at junctional complexes may also impede Wnt signalling. Levels of both phosphorylated and non-phosphorylated β-catenin require to be assessed by western blotting - upon binding of Wnt ligands to their receptors and triggering a signalling cascade, the level of non-phosphorylated β-catenin increases, which then translocates into the nucleus to mediate transcription of regulatory genes. An increased amount of β-catenin tethered at cell junctions by Crim1 might reduce available protein for nuclear translocation, thus impeding Wnt signalling. Furthermore, levels of Wnt-target genes such as Axin2,Nanog,Sox2,Sox9and others can also be assessed using qPCR, to determine the impact of loss of Crim1 on Wnt signal transduction.
The increased migration of EPDCs into the myocardium at 15.5 dpc, however, does not translate to an increase number of endothelial cells in the ventricular myocardium of mutant hearts at 17.5 dpc. This is because only a small proportion of EPDCs give rise to endothelial cells (Mikawa and Gourdie, 1996, Gittenberger-de Groot et al., 1998). Furthermore, the EPDCs that have migrated into the myocardium could also remain in an undifferentiated state. Stringent analyses of numbers of differentiated derivaties from the epicardium, vascular smooth muscle cells and fibroblasts, will also need to be performed, so as to confirm this hypothesis.
Crim1 in myocardial development The underlying mechanisms involved in regulation of myocardial development via modulation of epicardial and myocardial signaling molecules remain uncertain. Epicardial IGFs and PDGFs have been shown to act as epicardialmitogenic factors that stimulate the proliferation of cardiomyocytes and ventricular development (Kang et al., 2008, Li et al., 2011). The effect of loss of Crim1 in the context of PDGF signalling has not yet been assessed; this could be determined by analyzing levels of phosphorylation of the PDGF receptor, and also downstream effectors such as ERK1/2 and AKT. Crim1 could act as a co-receptor in the cells expressing it. Alternately, Crim1 could mediate epicardial growth factor sequestration, and the absence of Crim1 could mean that the growth factors
134 that are not tethered to the cell surface are free to signal to the adjacent myocardium. Wilkinson and colleagues supported this hypothesis with their finding that in mice lacking Crim1, VEGF-A diffused away from the podocytes, where Crim1 and VEGF-A are co-expressed, and resulted in a simultaneous increased activation of the VEGF-A receptor in adjacent vascular endothelial cells (Wilkinson et al., 2007).
While the underlying controlling mechanisms of myocardial development remain unclear, epicardial signalling molecule regulation appears to be crucial in this process. Furthermore, although an increased migration of EPDCs into the myocardial was observed in cases of Crim1FLOX/FLOX;WT1-Cre and Crim1FLOX/FLOX;WT1-CreERT2 hearts, overall proliferative ability of EPDCs was reduced, indicating an important role for Crim1 in the development of the myocardium. In cardiac remodelling and function following myocardial damage, cardiac fibroblasts play an important role (Duan et al., 2012). Thus, another attractive future research avenue involves investigation of the mechanisms by which Crim1 may potentially reactivate the adult myocardium following myocardial damage, particularly through modulation of growth factors such as BMPs, TGFβ, VEGF, IGF and PDGF. Future research into therapeutic interventions to improve cardiac function after myocardial damage is aided by our recent observation that the number of cardiac fibroblasts is reduced in the absence of Crim1, as epicardial cells are stimulated to produce fibroblasts and smooth muscle cells by myocardial injury (Limana et al., 2011, Zhou et al., 2011a, Duan et al., 2012, Huang et al., 2012, van Wijk et al., 2012). While we observed a reduction in the number of fibroblasts, it is not known whether the EPDCs that have migrated into the myocardium remain undifferentiated or commit to the coronary vascular smooth muscle cell lineage; thus this remains an interesting prospect to be explored, making it essential to assess the number of coronary vascular smooth muscle cells in the same WT1-CreERT2 mice used for assessing fibroblast number. Using in vitro studies involving primary epicardial cells that have lost Crim1, and assessing cell fate, will also aid in our understanding of whether it contributes to lineage specification.
The limited regenerative capacity of the adult myocardium has led to research into new technologies investigating the possibility of generating cardiomyocytes from induced stem cells. However, in order to make this process feasible, clarification of the pertinent growth factors and signalling molecules is essential. Molecules such as BMPs regulate stem cell renewal and differentiation into cardiomyocytes, and also provide links to other signalling pathways to further modulate gene expression of transcription factors (van Wijk et al., 2007, Varga and Wrana, 2005a). Since Crim1 can regulate BMP signalling, the interactions between CRIM1 and BMP signalling during in vitro development of cardiomyocytes could be a valuable area of investigation.
135
Crim1 in endothelial cell development and function In the absence of Crim1, the coronary vasculature is malformed. Looking forward, it will be important to perform morphometric investigations of the vascular network for patterning defects, by assessing vessel length and vascular diameter, as well as branch point number and number of blind- ending vessels. In vivo, we show an overall reduction in endothelial cell numbers and vascularized area. In vitro, this is further confirmed, with a decrease in viable HUVEC and HCMVEC numbers when CRIM1 is ablated. However, in vivo studies were performed only at 17.5 dpc embryos, making it essential to analyse whether endothelial cell endowment, accompanied by apoptosis and proliferation, are affected at earlier stages as well. The number of apoptotic endothelial cells in our immunofluorescence analyses was extremely low; this could be due to CC3 being an unsuitable apoptosis marker in our experiments. A better detection of apoptosis in tissue sections as well as in HCMVECs using TUNEL assay would provide more insight into how CRIM1 affects viability of endothelial cells. Furthermore, it would be crucial to distinguish between necrotic and apoptotic endothelial cell populations by using methods such as Propidium Iodide staining, to elucidate the extent to which Crim1 influences each of these phenomena.
Smooth muscle cells play an important role in vascular function and pathophysiology. Endothelial cells supply PDGF-B, required for the endothelial-cell-mediated recruitment of vascular smooth muscle cells to developing vessels (Van den Akker et al., 2008). It would be of interest to perform immunofluorescence labelling with Transgelin, a vascular smooth muscle cell marker, to assess the integrity of vessel walls, as well as degree of pericyte coverage of coronary vessels, based on vessel contact with pericytes, in the hearts of mice with an endothelial-restricted deletion of Crim1. Alongside this analysis, activation of the PDGF receptor on adjacent smooth muscle cells can also be investigated. Additionally, the presence of Crim1 at endothelial cell junctions makes it worthwhile to investigate endothelial cell permeability in vivo by using published protocols such as the intravenous injection of Evans Blue in mice so as to assess the possible extravasation of the dye in nearby vascularized tissue due to increased endothelial permeability (Radu and Chernoff 2013)
Adult Tie2-Cre mutants did not display any obvious external phenotypic changes or health deficiencies, suggesting that decrease in number of endothelial cells does not impede survival. However, a more stringent phenotypic analysis will have to be performed. It would therefore be of interest to investigate the response of Crim1FLOX/FLOX;Tie2-Cre mice to stress, injury or diet- dependent metabolic loading, as other growth factors and molecules from surrounding cell types may compensate for deficient Crim1 function in endothelial cells in these mice. Endothelial cells
136 play a firmly established role in the development of atherosclerosis. This is the leading cause of coronary artery disease and therefore myocardial infarction, and is dependent on activation of the endothelium (Hansson, 2005). This activation is a response to a large number of factors, some of which are secreted by the endothelium itself, which in turn results in leukocyte, neutrophil and granulocyte recruitment, leading to an inflammatory response. It is also important to confirm the relevant gene changes observed upon CRIM1 knockdown in HCMVECs, at the protein level by Western blotting. It would also be interesting to assess inflammation in adult Tie2-Cre mutants, given that loss of CRIM1 in HCMVECs caused an upregulation of genes involved in the inflammatory response. The limited ability of the endothelium to repair itself has made it critical to uncover molecules that can counter or prevent endothelial dysfunction. Thus, the presence of CRIM1 could counter the pathophysiology of various diseases that are directly involved with endothelial abnormalities.
CRIM1 and IGF interactions in endothelial cells. This study is the first to show that CRIM1 can bind IGF-1 via both the IGFBP domain and CRRs 3- 6, as performed in HEK293 cells. We have demonstrated that IGF signalling in HUVECs is modulated by CRIM1, and thus that CRIM1 potentially functions as an agonist of IGF-1, but the mechanism by which this occurs remains to be established. Decreased IGF-1R phosphorylation, seen in CRIM1-knockdown HCMVECs may not necessarily indicate reduced IGF levels, but may in fact indicate specific and effective presentation of IGF-1 to the receptor is required to mediate downstream signalling. The impact of this on vascular development and disease is currently unknown. IGF-1 plays a role in limiting atherosclerosis by reducing oxidative stress and thus senescence in vascular endothelial cell (Higashi et al., 2013). Similarly, infusion of IGF-1 was shown to counter the progress of inflammation and atherosclerosis in murine models, confirming it as an anti-atherosclerotic factor (Sukhanov et al., 2007). Furthemore, an IGF-1-mediated reduction in IL1B and TNFα expression has been demonstrated with a resultant abrogation of pro- inflammatory response (Spies et al., 2001). IGF-1 receptor inhibition also reduced murine retinal neovascularisation, partly due to reduced VEGF-induced MAPK activation (Smith et al., 1999). A decrease in endothelial cells was also seen upon disruption of IGF signalling through receptor blockade in human fetal lung explants (Han et al., 2003). Thus, based on existing evidences, there is a likelihood that CRIM1 and IGF-1 function together in endothelial cells to restrain inflammation and atherosclerosis, and maintain endothelial cell numbers as well. This can be further investigated by using IGF-1 ablated HCMVECs, and CRIM1 and IGF-1 double knockdown HCMVECs, and assessing whether they both function along a similar atheroprotective pathway by assessing severity of phenotype and endothelial gene expression. Furthermore, similar double knockdown in vivo
137 studies employing Tie2-Cre mice could also be a powerful tool in assessing whether Crim1 and IGF-1 function in tandem to mediate their protective function in the context of inflammation and atherosclerosis.
The ectodomain of Crim1 has been shown to be cleaved(Wilkinson et al., 2003)by matrix metalloproteinases(Butler et al., 2008); thus it could exist in circulation and mediate growth factor regulation in a paracrine manner. Further, it would be interesting to assess the relationship between CRIM1 and secreted mature IGF-1 levels and function, specifically, whether or not overexpression of CRIM1 results in increased IGF-1 signalling. Limitations of this study, as with others that have investigated the binding of CRIM1 to growth factors, include the fact that both are co-expressed by the same cell (Wilkinson et al., 2003, Wilkinson et al., 2007). It is therefore essential to examine how CRIM1 interacts with growth factors produced by different cells, as this will provide insight into CRIM1’s role in modulating these factors as well as its biological relevance in development and pathological conditions. Our data posits a novel mechanism by which CRIM1 and IGF-1 function synergistically as cardioprotective factors. Novel therapies to prevent cardiovascular disease or restore myocardial function after ischaemic damage may be developed using these unique findings.
Putative role of CRIM1 in other diseases In the context of other disorders highlighting potential roles for CRIM1 in humans, analysis of serum from patients with chronic heart failure has shown increased CRIM1 levels and fibrosis- associated secreted factors compared to other pathological disorders. This indicates a positive correlation between CRIM1 and pro-fibrotic activity (Eleuteri et al., 2014). Furthermore, increased CRIM1 expression has been demonstrated in drug-resistant leukaemia cells, implicating CRIM1 as a potential marker for drug resistance (Prenkert et al., 2010). Intronic regions of CRIM1 and ZEB2 are downregulated in breast cancer epithelial cells (Kim et al., 2015), and CRIM1 overexpression is seen in gastric cancer tissues (Lim et al., 2014). It is also a suggested target of the Hippo pathway (Lim et al., 2014). These glimpses of the overarching role of CRIM1 in cancer biology and disease may provide further impetus for investigating the role of this gene in pathological conditions. Developmental studies have suggested that CRIM1 may be the fulcrum for many critical signalling pathways, and as such, manipulating its expression may reveal targets for modulation of cellular proliferation, differentiation and repair following injury and tumorigenesis.
Thus, while our findings take us a step closer to understanding the role of CRIM1 in the heart and identify CRIM1 to be a critical regulator of epicardial and myocardial development and coronary
138 endothelial cell function, there are further questions pertaining to its role that have been raised. Its extensive crosstalk with other major signalling pathways in the same cell or tissue type needs to be further explored, as a better understanding of the molecular mechanisms of CRIM1 dysfunction in within the various cells of the heart may facilitate the development of innovative therapies, especially in the context of myocardial repair and coronary vascular disease.
139
References
140 Afouda, B. A., J. Martin, F. Liu, A. Ciau-Uitz, R. Patient and S. Hoppler (2008). "GATA transcription factors integrate Wnt signalling during heart development." Development135(19): 3185-3190. Aird, W. C. (2007). "Phenotypic Heterogeneity of the Endothelium: I. Structure, Function, and Mechanisms." Circulation Research100(2): 158-173. Aird, W. C. (2007). "Phenotypic Heterogeneity of the Endothelium: II. Representative Vascular Beds." Circulation Research100(2): 174-190. Aird, W. C. (2012). "Endothelial Cell Heterogeneity." Cold Spring Harbor Perspectives in Medicine2(1): a006429. Akiyama, H., M.-C. Chaboissier, R. R. Behringer, D. H. Rowitch, A. Schedl, J. A. Epstein and B. de Crombrugghe (2004). "Essential role of Sox9 in the pathway that controls formation of cardiac valves and septa." Proceedings of the National Academy of Sciences of the United States of America101(17): 6502-6507. Arthur, H. M. and S. D. Bamforth (2011). "TGFbeta signaling and congenital heart disease: Insights from mouse studies." Birth Defects Research A Clinical and Molecular Teratology91(6): 423-434. Ashe, A., N. C. Butterfield, L. Town, A. D. Courtney, A. N. Cooper, C. Ferguson, R. Barry, F. Olsson, K. F. Liem, Jr., R. G. Parton, B. J. Wainwright, K. V. Anderson, E. Whitelaw and C. Wicking (2012). "Mutations in mouse Ift144 model the craniofacial, limb and rib defects in skeletal ciliopathies." Human Molecular Genetics21(8): 1808-1823. Ashe, H. L. (2005). "BMP Signalling: Synergy and Feedback Create a Step Gradient." Current Biology15(10): R375-R377. Azhar, M., J. Schultz Jel, I. Grupp, G. W. Dorn, 2nd, P. Meneton, D. G. Molin, A. C. Gittenberger- de Groot and T. Doetschman (2003). "Transforming growth factor beta in cardiovascular development and function." Cytokine and Growth Factor Reviews14(5): 391-407. Back, K., R. Islam, G. S. Johansson, S. I. Chisalita and H. J. Arnqvist (2012). "Insulin and IGF1 receptors in human cardiac microvascular endothelial cells: metabolic, mitogenic and anti- inflammatory effects." Journal of Endocrinology215(1): 89-96. Baek, S. T. and M. D. Tallquist (2012). "Nf1 limits epicardial derivative expansion by regulating epithelial to mesenchymal transition and proliferation." Development139(11): 2040-2049. Balemans, W. and W. Van Hul (2002). "Extracellular regulation of BMP signaling in vertebrates: a cocktail of modulators." Developmental Biology250(2): 231-250. Barbosky, L., D. K. Lawrence, G. Karunamuni, J. C. Wikenheiser, Y. Q. Doughman, R. P. Visconti, J. B. Burch and M. Watanabe (2006). "Apoptosis in the developing mouse heart." Developmental Dynamics235(9): 2592-2602. Baserga, R. (1999). "The IGF-I receptor in cancer research." Expermintal Cell Research253(1): 1-6. Bayne, M. L., J. Applebaum, G. G. Chicchi, N. S. Hayes, B. G. Green and M. A. Cascieri (1988). "Structural analogs of human insulin-like growth factor I with reduced affinity for serum binding proteins and the type 2 insulin-like growth factor receptor." Journal of Biological Chemistry263(13): 6233-6239. Bernanke, D. H. and J. M. Velkey (2002). "Development of the coronary blood supply: changing concepts and current ideas." The Anatomical Record269(4): 198-208.
141 Bond, J., D. Sedmera, J. Jourdan, Y. Zhang, C. A. Eisenberg, L. M. Eisenberg and R. G. Gourdie (2003). "Wnt11 and Wnt7a are up-regulated in association with differentiation of cardiac conduction cells in vitro and in vivo." Developmental Dynamics227(4): 536-543. Bouchey, D., C. J. Drake, A. M. Wunsch and C. D. Little (1996). "Distribution of connective tissue proteins during development and neovascularization of the epicardium." Cardiovascular Research31 Spec No: E104-E115. Boyer, A. S., Ayerinskas, II, E. B. Vincent, L. A. McKinney, D. L. Weeks and R. B. Runyan (1999). "TGFbeta2 and TGFbeta3 have separate and sequential activities during epithelial- mesenchymal cell transformation in the embryonic heart." Developmental Biology208(2): 530-545. Brott, B. K. and S. Y. Sokol (2005). "A vertebrate homolog of the cell cycle regulator Dbf4 is an inhibitor of Wnt signaling required for heart development." Developmental Cell8(5): 703-715. Brown, C. B. and H. S. Baldwin (2006). "Neural crest contribution to the cardiovascular system." Advances in Experimental Medicine and Biology589: 134-154. Brown, M. A., Q. Zhao, K. A. Baker, C. Naik, C. Chen, L. Pukac, M. Singh, T. Tsareva, Y. Parice, A. Mahoney, V. Roschke, I. Sanyal and S. Choe (2005). "Crystal structure of BMP-9 and functional interactions with pro-region and receptors." Journal of Biological Chemistry280(26): 25111-25118. Brutsaert, D. L. (2003). "Cardiac endothelial-myocardial signaling: its role in cardiac growth, contractile performance, and rhythmicity." Physiological Reviews83(1): 59-115. Brutsaert, D. L., P. Fransen, L. J. Andries, G. W. De Keulenaer and S. U. Sys (1998). "Cardiac endothelium and myocardial function." Cardiovascular Research38(2): 281-290. Bryksin, A. and I. Matsumura (2013). "Overlap extension PCR cloning." Methods in Molecular Biology1073: 31-42. Burger, P. E., S. Coetzee, W. L. McKeehan, M. Kan, P. Cook, Y. Fan, T. Suda, R. P. Hebbel, N. Novitzky, W. A. Muller and E. L. Wilson (2002). "Fibroblast growth factor receptor-1 is expressed by endothelial progenitor cells." Blood100(10): 3527-3535. Butler, G. S., R. A. Dean, E. M. Tam and C. M. Overall (2008). "Pharmacoproteomics of a metalloproteinase hydroxamate inhibitor in breast cancer cells: dynamics of membrane type 1 matrix metalloproteinase-mediated membrane protein shedding." Molecular and Cellular Biology28(15): 4896-4914. Cai, C. L., X. Liang, Y. Shi, P. H. Chu, S. L. Pfaff, J. Chen and S. Evans (2003). "Isl1 identifies a cardiac progenitor population that proliferates prior to differentiation and contributes a majority of cells to the heart." Developmental Cell5(6): 877-889. Cai, C. L., J. C. Martin, Y. Sun, L. Cui, L. Wang, K. Ouyang, L. Yang, L. Bu, X. Liang, X. Zhang, W. B. Stallcup, C. P. Denton, A. McCulloch, J. Chen and S. M. Evans (2008). "A myocardial lineage derives from Tbx18 epicardial cells." Nature454(7200): 104-108. Camelliti, P., T. K. Borg and P. Kohl (2005). "Structural and functional characterisation of cardiac fibroblasts." Cardiovascular Research65(1): 40-51. Canalis, E. and B. Gabbitas (1994). "Bone morphogenetic protein 2 increases insulin-like growth factor I and II transcripts and polypeptide levels in bone cell cultures." Journal of Bone and Mineral Research9(12): 1999-2005. Cao, R., E. Brakenhielm, X. Li, K. Pietras, J. Widenfalk, A. Ostman, U. Eriksson and Y. Cao (2002). "Angiogenesis stimulated by PDGF-CC, a novel member in the PDGF family, involves activation of PDGFR-alphaalpha and -alphabeta receptors." FASEB Journal16(12): 1575-1583. 142 Carpenter, V., K. Matthews, G. Devlin, S. Stuart, J. Jensen, J. Conaglen, F. Jeanplong, P. Goldspink, S.-Y. Yang, G. Goldspink, J. Bass and C. McMahon (2008). "Mechano-Growth Factor Reduces Loss of Cardiac Function in Acute Myocardial Infarction." Heart, Lung and Circulation17(1): 33-39. Charnock-Jones, D. S., A. M. Sharkey, C. A. Boocock, A. Ahmed, R. Plevin, N. Ferrara and S. K. Smith (1994). "Vascular endothelial growth factor receptor localization and activation in human trophoblast and choriocarcinoma cells." Biology of Reproduction51(3): 524-530. Cheifetz, S., T. Bellon, C. Cales, S. Vera, C. Bernabeu, J. Massague and M. Letarte (1992). "Endoglin is a component of the transforming growth factor-beta receptor system in human endothelial cells." J Biol Chem267(27): 19027-19030. Chen, J., S. W. Kubalak and K. R. Chien (1998). "Ventricular muscle-restricted targeting of the RXRa gene reveals a non-cell-autonomous requirement in cardiac chamber morphogenesis." Development125(10): 1943-1949. Chen, T. H. P., T. C. Chang, J. O. Kang, B. Choudhary, T. Makita, C. M. Tran, J. B. E. Burch, H. Eid and H. M. Sucov (2002). "Epicardial induction of fetal cardiomyocyte proliferation via a retinoic acid-inducible trophic factor." Developmental Biology250: 198-207. Chisalita, S. I. and H. J. Arnqvist (2004). "Insulin-like growth factor I receptors are more abundant than insulin receptors in human micro- and macrovascular endothelial cells." American Journal of Physiology Endocrinology and Metabolism286(6): E896-901. Chiu, H. S., J. P. York, L. Wilkinson, P. Zhang, M. H. Little and D. J. Pennisi (2012). "Production of a mouse line with a conditional Crim1 mutant allele." Genesis50(9): 711-716. Choi, M., R. W. Stottmann, Y.-P. Yang, E. N. Meyers and J. Klingensmith (2007). "The Bone Morphogenetic Protein Antagonist Noggin Regulates Mammalian Cardiac Morphogenesis." Circulation Research100(2): 220-228. Chong, J. J., V. Chandrakanthan, M. Xaymardan, N. S. Asli, J. Li, I. Ahmed, C. Heffernan, M. K. Menon, C. J. Scarlett, A. Rashidianfar, C. Biben, H. Zoellner, E. K. Colvin, J. E. Pimanda, A. V. Biankin, B. Zhou, W. T. Pu, O. W. Prall and R. P. Harvey (2011). "Adult cardiac-resident MSC- like stem cells with a proepicardial origin." Cell Stem Cell9(6): 527-540. Christoffels, V. M., T. Grieskamp, J. Norden, M. T. Mommersteeg, C. Rudat and A. Kispert (2009). "Tbx18 and the fate of epicardial progenitors." Nature458(7240): E8-9; discussion E9-10. Claesson-Welsh, L. (2008). "VEGF-B taken to our hearts: specific effect of VEGF-B in myocardial ischemia." Arteriosclerosis, Thrombosis and Vascular Biology28(9): 1575-1576. Cohen, E. D., Z. Wang, J. J. Lepore, M. M. Lu, M. M. Taketo, D. J. Epstein and E. E. Morrisey (2007). "Wnt/beta-catenin signaling promotes expansion of Isl-1-positive cardiac progenitor cells through regulation of FGF signaling." Journal of Clinical Investigations117(7): 1794-1804. Cossette, S. and R. Misra (2011). "The identification of different endothelial cell populations within the mouse proepicardium." Developmental Dynamics240(10): 2344-2353. Cossette, S., Misra, Ravi. (2011). "The identification of different endothelial cell populations within the mouse proepicardium. ." Developmental Dynamics240(10): 2344–2353. Coultas, L., K. Chawengsaksophak and J. Rossant (2005). "Endothelial cells and VEGF in vascular development." Nature438(7070): 937-945.
143 de la Cruz, M. V., C. Sanchez Gomez, M. M. Arteaga and C. Arguello (1977). "Experimental study of the development of the truncus and the conus in the chick embryo." Journal of Anatomy123(Pt 3): 661-686. De Robertis, E. M. (2009). "Spemann’s organizer and the self-regulation of embryonic fields." Mechanisms of Development126(11–12): 925-941. Delafontaine, P., Y.-H. Song and Y. Li (2004). "Expression, Regulation, and Function of IGF-1, IGF-1R, and IGF-1 Binding Proteins in Blood Vessels." Arteriosclerosis, Thrombosis, and Vascular Biology24(3): 435-444. Derwall, M., R. Malhotra, C. S. Lai, Y. Beppu, E. Aikawa, J. S. Seehra, W. M. Zapol, K. D. Bloch and P. B. Yu (2012). "Inhibition of bone morphogenetic protein signaling reduces vascular calcification and atherosclerosis." Arteriosclerosis, Thrombosis and Vascular Biology32(3): 613- 622. Dettman, R. W., W. Denetclaw, Jr., C. P. Ordahl and J. Bristow (1998). "Common epicardial origin of coronary vascular smooth muscle, perivascular fibroblasts, and intermyocardial fibroblasts in the avian heart." Developmental Biology193(2): 169–181. Duan, J., C. Gherghe, D. Liu, E. Hamlett, L. Srikantha, L. Rodgers, J. N. Regan, M. Rojas, M. Willis, A. Leask, M. Majesky and A. Deb (2012). "Wnt1/βcatenin injury response activates the epicardium and cardiac fibroblasts to promote cardiac repair." The EMBO Journal31(2): 429-442. Dyer, L. A. and M. L. Kirby (2009). "The role of secondary heart field in cardiac development." Developmental Biology336(2): 137-144. Dyer, L. A., X. Pi and C. Patterson (2014). "The role of BMPs in endothelial cell function and dysfunction." Trends in endocrinology and metabolism: TEM25(9): 472-480. Eisenberg, L. M. and C. A. Eisenberg (2006). "Wnt signal transduction and the formation of the myocardium." Developmental Biology293(2): 305-315. Eleuteri, E., A. Di Stefano, D. Vallese, I. Gnemmi, A. Pitruzzella, F. Tarro Genta, L. Delle Donne, F. Cappello, F. L. Ricciardolo and P. Giannuzzi (2014). "Fibrosis markers and CRIM1 increase in chronic heart failure of increasing severity." Biomarkers19(3): 214-221. Faloon, P., E. Arentson, A. Kazarov, C. X. Deng, C. Porcher, S. Orkin and K. Choi (2000). "Basic fibroblast growth factor positively regulates hematopoietic development." Development127(9): 1931-1941. Fan, J., V. G. Ponferrada, T. Sato, S. Vemaraju, M. Fruttiger, H. Gerhardt, N. Ferrara and R. A. Lang (2014). "Crim1 maintains retinal vascular stability during development by regulating endothelial cell Vegfa autocrine signaling." Development141(2): 448-459. Ferrara, N., K. Carver-Moore, H. Chen, M. Dowd, L. Lu, K. S. O'Shea, L. Powell-Braxton, K. J. Hillan and M. W. Moore (1996). "Heterozygous embryonic lethality induced by targeted inactivation of the VEGF gene." Nature380(6573): 439-442. Forman-Kay, J. D. and T. Pawson (1999). "Diversity in protein recognition by PTB domains." Current Opinions in Structural Biology9(6): 690-695. Frostegård, J. (2013). "Immunity, atherosclerosis and cardiovascular disease." BMC Medicine11(1): 1-13. Fuh, G., B. Li, C. Crowley, B. Cunningham and J. A. Wells (1998). "Requirements for binding and signaling of the kinase domain receptor for vascular endothelial growth factor." Journal of Biological Chemistry273(18): 11197-11204. 144 Fujiwara, T., D. B. Dehart, K. K. Sulik and B. L. Hogan (2002). "Distinct requirements for extra- embryonic and embryonic bone morphogenetic protein 4 in the formation of the node and primitive streak and coordination of left-right asymmetry in the mouse." Development129(20): 4685-4696. Fung, W. Y., K. F. Chi Fat, C. K. Song Eng and C. King Lau (2007). "crm-1 facilitates BMP signaling to control body size in Caenorhabditis elegans." Developmental Biology311(1): 95-105. Gabbitas, B. and E. Canalis (1995). "Bone morphogenetic protein-2 inhibits the synthesis of insulin-like growth factor-binding protein-5 in bone cell cultures." Endocrinology136(6): 2397- 2403. Galkina, E. and K. Ley (2009). "Immune and Inflammatory Mechanisms of Atherosclerosis." Annual review of immunology27: 165-197. Galvin, K. M., M. J. Donovan, C. A. Lynch, R. I. Meyer, R. J. Paul, J. N. Lorenz, V. Fairchild- Huntress, K. L. Dixon, J. H. Dunmore, M. A. Gimbrone, Jr., D. Falb and D. Huszar (2000). "A role for smad6 in development and homeostasis of the cardiovascular system." Nature Genetics24(2): 171-174. Garcia-Martinez, V. and G. C. Schoenwolf (1993). "Primitive-streak origin of the cardiovascular system in avian embryos." Developmental Biology159(2): 706-719. Gearing, A. J. and W. Newman (1993). "Circulating adhesion molecules in disease." Immunology Today14(10): 506-512. Georgas, K., J. Bowles, T. Yamada, P. Koopman and M. H. Little (2000). "Characterisation of Crim1 expression in the developing mouse urogenital tract reveals a sexually dimorphic gonadal expression pattern." Developmental Dynamics219(4): 582-587. Giles, R. H., J. H. van Es and H. Clevers (2003). "Caught up in a Wnt storm: Wnt signaling in cancer." Biochimica et Biophysica Acta1653(1): 1-24. Gimbrone, M. A. and G. García-Cardeña (2016). "Endothelial Cell Dysfunction and the Pathobiology of Atherosclerosis." Circulation Research118(4): 620-636. Giordano, F. J., H. P. Gerber, S. P. Williams, N. VanBruggen, S. Bunting, P. Ruiz-Lozano, Y. Gu, A. K. Nath, Y. Huang, R. Hickey, N. Dalton, K. L. Peterson, J. Ross, Jr., K. R. Chien and N. Ferrara (2001). "A cardiac myocyte vascular endothelial growth factor paracrine pathway is required to maintain cardiac function." Proceedings of the National Academy of Sciences U S A98(10): 5780-5785. Gittenberger-de Groot, A. C., M. P. Vrancken Peeters, M. Bergwerff, M. M. Mentink and R. E. Poelmann (2000). "Epicardial outgrowth inhibition leads to compensatory mesothelial outflow tract collar and abnormal cardiac septation and coronary formation." Circulation Research87(11): 969- 971. Gittenberger-de Groot, A. C., M. P. Vrancken Peeters, M. M. Mentink, R. G. Gourdie and R. E. Poelmann (1998). "Epicardium-derived cells contribute a novel population to the myocardial wall and the atrioventricular cushions." Circulation Research82(10): 1043-1052. Gleeson, L. M., C. Chakraborty, T. McKinnon and P. K. Lala (2001). "Insulin-like growth factor- binding protein 1 stimulates human trophoblast migration by signaling through alpha 5 beta 1 integrin via mitogen-activated protein Kinase pathway." Journal of Clinical Endocrinology and Metabolism86(6): 2484-2493.
145 Glienke, J., A. Sturz, A. Menrad and K. H. Thierauch (2002). "CRIM1 is involved in endothelial cell capillary formation in vitro and is expressed in blood vessels in vivo." Mechanisms of Development119(2): 165-175. Godin, R. E., E. J. Robertson and A. T. Dudley (1999). "Role of BMP family members during kidney development." International Journal of Developmental Biology43(5): 405-411. Goumans, M. J. and C. Mummery (2000). "Functional analysis of the TGFbeta receptor/Smad pathway through gene ablation in mice." International Journal of Developmental Biology44(3): 253-265. Groppe, J., J. Greenwald, E. Wiater, J. Rodriguez-Leon, A. N. Economides, W. Kwiatkowski, M. Affolter, W. W. Vale, J. C. Belmonte and S. Choe (2002). "Structural basis of BMP signalling inhibition by the cystine knot protein Noggin." Nature420(6916): 636-642. Hajra, L., A. I. Evans, M. Chen, S. J. Hyduk, T. Collins and M. I. Cybulsky (2000). "The NF-κB signal transduction pathway in aortic endothelial cells is primed for activation in regions predisposed to atherosclerotic lesion formation." Proceedings of the National Academy of Sciences97(16): 9052-9057. Han, R. N. N., M. Post, A. K. Tanswell and S. J. Lye (2003). "Insulin-Like Growth Factor-I Receptor–Mediated Vasculogenesis/Angiogenesis in Human Lung Development." American Journal of Respiratory Cell and Molecular Biology28(2): 159-169. Hansson, G. K. M. D. P. (2005). "MECHANISMS OF DISEASE: Inflammation, Atherosclerosis, and Coronary Artery Disease." The New England Journal of Medicine352(16): 1685-1695. Hede, M. S., E. Salimova, A. Piszczek, E. Perlas, N. Winn, T. Nastasi and N. Rosenthal (2012). "E- peptides control bioavailability of IGF-1." PLoS One7(12): e51152. Heidaran, M. A., J. H. Pierce, R. A. Jensen, T. Matsui and S. A. Aaronson (1990). "Chimeric alpha- and beta-platelet-derived growth factor (PDGF) receptors define three immunoglobulin-like domains of the alpha-PDGF receptor that determine PDGF-AA binding specificity." Journal of Biological Chemistry265(31): 18741-18744. Heidaran, M. A., J. H. Pierce, J. C. Yu, D. Lombardi, J. E. Artrip, T. P. Fleming, A. Thomason and S. A. Aaronson (1991). "Role of alpha beta receptor heterodimer formation in beta platelet-derived growth factor (PDGF) receptor activation by PDGF-AB." Journal of Biological Chemistry266(30): 20232-20237. Hendrickx, J., K. Doggen, E. O. Weinberg, P. Van Tongelen, P. Fransen and G. W. De Keulenaer (2004). "Molecular diversity of cardiac endothelial cells in vitro and in vivo." Physiological Genomics19(2): 198-206. Hertig, C. M., S. W. Kubalak, Y. Wang and K. R. Chien (1999). "Synergistic roles of neuregulin-1 and insulin-like growth factor-I in activation of the phosphatidylinositol 3-kinase pathway and cardiac chamber morphogenesis." Journal of Biological Chemistry274(52): 37362-37369. Herzog, B., C. Pellet-Many, G. Britton, B. Hartzoulakis and I. C. Zachary (2011). "VEGF binding to NRP1 is essential for VEGF stimulation of endothelial cell migration, complex formation between NRP1 and VEGFR2, and signaling via FAK Tyr407 phosphorylation." Molecular Biology of the Cell22(15): 2766-2776. Higashi, Y., A. Pandey, B. Goodwin and P. Delafontaine (2013). "Insulin-like growth factor-1 regulates glutathione peroxidase expression and activity in vascular endothelial cells: Implications for atheroprotective actions of insulin-like growth factor-1." Biochimica et Biophysica Acta1832(3): 391-399.
146 Higashi, Y., H. C. Quevedo, S. Tiwari, S. Sukhanov, S. Y. Shai, A. Anwar and P. Delafontaine (2014). "Interaction between insulin-like growth factor-1 and atherosclerosis and vascular aging." Frontiers of Hormone Research43: 107-124. Holmes, K., O. L. Roberts, A. M. Thomas and M. J. Cross (2007). "Vascular endothelial growth factor receptor-2: structure, function, intracellular signalling and therapeutic inhibition." Cell Signal19(10): 2003-2012. Hsieh, P. C. H., M. E. Davis, L. K. Lisowski and R. T. Lee (2006). "Endothelial-Cardiomyocyte Interactions in Cardiac Development and Repair." Annual Review of Physiology68: 51-66. Hsu, D. R., A. N. Economides, X. Wang, P. M. Eimon and R. M. Harland (1998). "The Xenopus Dorsalizing Factor Gremlin Identifies a Novel Family of Secreted Proteins that Antagonize BMP Activities." Molecular Cell1(5): 673-683. Huang, G. N., J. E. Thatcher, J. McAnally, Y. Kong, X. Qi, W. Tan, J. M. DiMaio, J. F. Amatruda, R. D. Gerard, J. A. Hill, R. Bassel-Duby and E. N. Olson (2012). "C/EBP transcription factors mediate epicardial activation during heart development and injury." Science338(6114): 1599-1603. Hutchins, G. M., A. Kessler-Hanna and G. W. Moore (1988). "Development of the coronary arteries in the embryonic human heart." Circulation77(6): 1250-1257. Hwa, V., Y. Oh and R. G. Rosenfeld (1999). "The Insulin-Like Growth Factor-Binding Protein (IGFBP) Superfamily." Endocrine Reviews20(6): 761-787. Inomata, H., T. Haraguchi and Y. Sasai (2008). "Robust Stability of the Embryonic Axial Pattern Requires a Secreted Scaffold for Chordin Degradation." Cell134(5): 854-865. Ishii, Y., R. J. Garriock, A. M. Navetta, L. E. Coughlin and T. Mikawa (2010). "BMP signals promote proepicardial protrusion necessary for recruitment of coronary vessel and epicardial progenitors to the heart." Developmental Cell19(2): 307-316. Ishii, Y., J. D. Langberg, R. Hurtado, S. Lee and T. Mikawa (2007). "Induction of proepicardial marker gene expression by the liver bud." Development134(20): 3627–3637. Iyer, S., F. Y. Chou, R. Wang, H. S. Chiu, V. K. S. Raju, M. H. Little, W. G. Thomas, M. Piper and D. J. Pennisi (2016). "Crim1 has cell-autonomous and paracrine roles during embryonic heart development." Scientific Reports6: 19832. James, R. E. and H. T. Broihier (2011). "Crimpy inhibits the BMP homolog Gbb in motoneurons to enable proper growth control at the Drosophila neuromuscular junction." Development138(15): 3273-3286. Jiao, K., H. Kulessa, K. Tompkins, Y. Zhou, L. Batts, H. S. Baldwin and B. L. Hogan (2003). "An essential role of Bmp4 in the atrioventricular septation of the mouse heart." Genes & Development17(19): 2362-2367. Jo, H., H. Song and A. Mowbray (2006). "Role of NADPH oxidases in disturbed flow- and BMP4- induced inflammation and atherosclerosis." Antioxidiants and Redox Signalling8(9-10): 1609-1619. Jones, J. I. and D. R. Clemmons (1995). "Insulin-like growth factors and their binding proteins: biological actions." Endocrine Reviews16(1): 3-34. Jones, J. I., A. Gockerman, W. H. Busby, Jr., G. Wright and D. R. Clemmons (1993). "Insulin-like growth factor binding protein 1 stimulates cell migration and binds to the alpha 5 beta 1 integrin by means of its Arg-Gly-Asp sequence." Proceedings of the National Academy of Sciences USA90(22): 10553-10557.
147 Juul, A., T. Scheike, M. Davidsen, J. Gyllenborg and T. Jorgensen (2002). "Low serum insulin-like growth factor I is associated with increased risk of ischemic heart disease: a population-based case- control study." Circulation106(8): 939-944. Kang, J., Y. Gu, P. Li, B. L. Johnson, H. M. Sucov and P. S. Thomas (2008). "PDGF-A as an epicardial mitogen during heart development." Developmental Dynamics237(3): 692-701. Kang, J.-O. and H. M. Sucov (2005). "Convergent proliferative response and divergent morphogenic pathways induced by epicardial and endocardial signaling in fetal heart development." Mechanisms of Development122(1): 57-65. Karkkainen, M. J., A. Saaristo, L. Jussila, K. A. Karila, E. C. Lawrence, K. Pajusola, H. Bueler, A. Eichmann, R. Kauppinen, M. I. Kettunen, S. Ylä-Herttuala, D. N. Finegold, R. E. Ferrell and K. Alitalo (2001). "A model for gene therapy of human hereditary lymphedema." Proceedings of the National Academy of Sciences98(22): 12677-12682. Katz, T. C., M. K. Singh, K. Degenhardt, J. Rivera-Feliciano, R. L. Johnson, J. A. Epstein and C. J. Tabin (2012). "Distinct compartments of the proepicardial organ give rise to coronary vascular endothelial cells." Developmental Cell22(3): 639-650. Kawabata, M., T. Imamura and K. Miyazono (1998). "Signal transduction by bone morphogenetic proteins." Cytokine & Growth Factor Reviews9(1): 49-61. Kelly, R. G., N. A. Brown and M. E. Buckingham (2001). "The arterial pole of the mouse heart forms from Fgf10-expressing cells in pharyngeal mesoderm." Developmental Cell1(3): 435-440. Kelly, R. G., M. E. Buckingham and A. F. Moorman (2014). "Heart Fields and Cardiac Morphogenesis." Cold Spring Harbor Perspectives in Medicine4(10). Keyhanfar, M., Grant W. Booker, J. Whittaker, John C. Wallace and Briony E. Forbes (2007). "Precise mapping of an IGF-I-binding site on the IGF-1R." Biochemical Journal401(Pt 1): 269-277. Kim, R. Y., E. J. Robertson and M. J. Solloway (2001). "Bmp6 and Bmp7 Are Required for Cushion Formation and Septation in the Developing Mouse Heart." Developmental Biology235(2): 449-466. Kim, S. W., E. Fishilevich, G. Arango-Argoty, Y. Lin, G. Liu, Z. Li, A. P. Monaghan, M. Nichols and B. John (2015). "Genome-wide transcript profiling reveals novel breast cancer-associated intronic sense RNAs." PLoS One10(3): e0120296. Kinna, G., G. Kolle, A. Carter, B. Key, G. J. Lieschke, A. Perkins and M. H. Little (2006). "Knockdown of zebrafish crim1 results in a bent tail phenotype with defects in somite and vascular development." Mechanisms of Development123(4): 277-287. Kirby, M. and K. Waldo (1990). "Role of neural crest in congenital heart disease." Circulation82(2): 332-340. Kirby, M. L. and K. L. Waldo (1995). "Neural crest and cardiovascular patterning." (0009-7330 (Print)). Kisanuki, Y. Y., R. E. Hammer, J.-i. Miyazaki, S. C. Williams, J. A. Richardson and M. Yanagisawa (2001). "Tie2-Cre Transgenic Mice: A New Model for Endothelial Cell-Lineage Analysis in Vivo." Developmental Biology230(2): 230-242. Kolle, G., K. Georgas, G. P. Holmes, M. H. Little and T. Yamada (2000). "CRIM1, a novel gene encoding a cysteine-rich repeat protein, is developmentally regulated and implicated in vertebrate CNS development and organogenesis." Mechanisms of Development90(2): 181-193.
148 Kolle, G., A. Jansen, T. Yamada and M. Little (2003). "In ovo electroporation of Crim1 in the developing chick spinal cord." Developmental Dynamics226(1): 107-111. Koni, P. A., S. K. Joshi, U. A. Temann, D. Olson, L. Burkly and R. A. Flavell (2001). "Conditional vascular cell adhesion molecule 1 deletion in mice: impaired lymphocyte migration to bone marrow." Journal of Experimental Medicine193(6): 741-754. Kruithof, B. P., B. van Wijk, S. Somi, M. Kruithof-de Julio, J. M. Perez Pomares, F. Weesie, A. Wessels, A. F. Moorman and M. J. van den Hoff (2006). "BMP and FGF regulate the differentiation of multipotential pericardial mesoderm into the myocardial or epicardial lineage." Developmental Biology295(2): 507-522. Kruithof, B. P., B. van Wijk, S. Somi, M. Kruithof-de Julio, J. M. Pérez Pomares, F. Weesie, A. Wessels, A. F. Moorman and M. J. van den Hoff (2006). "BMP and FGF regulate the differentiation of multipotential pericardial mesoderm into the myocardial or epicardial lineage." Developmental Biology295(2): 507–522. Kwee, L., H. S. Baldwin, H. M. Shen, C. L. Stewart, C. Buck, C. A. Buck and M. A. Labow (1995). "Defective development of the embryonic and extraembryonic circulatory systems in vascular cell adhesion molecule (VCAM-1) deficient mice." Development121(2): 489-503. Laflamme, M. A. and C. E. Murry (2011). "Heart Regeneration." Nature473(7347): 326-335. Langenfeld, E. M. and J. Langenfeld (2004). "Bone morphogenetic protein-2 stimulates angiogenesis in developing tumors." Molecular Cancer Research2(3): 141-149. Larraín, J., D. Bachiller, B. Lu, E. Agius, S. Piccolo and E. M. De Robertis (2000). "BMP-binding modules in chordin: a model for signalling regulation in the extracellular space." Development (Cambridge, England)127(4): 821-830. Laugwitz, K.-L., A. Moretti, L. Caron, A. Nakano and K. R. Chien (2008). "Islet1 cardiovascular progenitors: a single source for heart lineages?" Development135(2): 193-205. Laustsen, P. G., S. J. Russell, L. Cui, A. Entingh-Pearsall, M. Holzenberger, R. Liao and C. R. Kahn (2007). "Essential role of insulin and insulin-like growth factor 1 receptor signaling in cardiac development and function." Molecular And Cellular Biology27(5): 1649-1664. Lavine, K. J. and D. M. Ornitz (2008). "Fibroblast growth factors and Hedgehogs: at the heart of the epicardial signaling center." Trends in Genetics24(1): 33-40. Lavine, K. J., A. C. White, C. Park, C. S. Smith, K. Choi, F. Long, C. C. Hui and D. M. Ornitz (2006). "Fibroblast growth factor signals regulate a wave of Hedgehog activation that is essential for coronary vascular development." Genes and Development20(12): 1651–1666. Lawrence, T. (2009). "The Nuclear Factor NF-κB Pathway in Inflammation." Cold Spring Harbor Perspectives in Biology1(6). Leal, S. M., Q. Liu, S. S. Huang and J. S. Huang (1997). "The type V transforming growth factor beta receptor is the putative insulin-like growth factor-binding protein 3 receptor." Journal of Biological Chemistry272(33): 20572-20576. Lechleider, R. J., J. L. Ryan, L. Garrett, C. Eng, C.-x. Deng, A. Wynshaw-Boris and A. B. Roberts (2001). "Targeted Mutagenesis of Smad1 Reveals an Essential Role in Chorioallantoic Fusion." Developmental Biology240(1): 157-167. Lee, P. D., L. C. Giudice, C. A. Conover and D. R. Powell (1997). "Insulin-like growth factor binding protein-1: recent findings and new directions." Proceedings of the Society for Experimental Biology and Medicine216(3): 319-357. 149 Leighton, P. A., K. J. Mitchell, L. V. Goodrich, X. Lu, K. Pinson, P. Scherz, W. C. Skarnes and M. Tessier-Lavigne (2001). "Defining brain wiring patterns and mechanisms through gene trapping in mice." Nature410(6825): 174-179. Li, J., K. C. Liu, F. Jin, M. M. Lu and J. A. Epstein (1999). "Transgenic rescue of congenital heart disease and spina bifida in Splotch mice." Development126(11): 2495-2503. Li, P., S. Cavellero, Y. Gu, T. H. P. Chen, J. Hughes, A. Bassim Hassan, J. C. Bruning, M. Pashmforoush and H. M. Sucov (2011). "IGF signaling directs ventricular cardiomyocyte proliferation during embyronic heart development." Development138: 1795-1805. Li, Q., B. Li, X. Wang, A. Leri, K. P. Jana, Y. Liu, J. Kajstura, R. Baserga and P. Anversa (1997). "Overexpression of insulin-like growth factor-1 in mice protects from myocyte death after infarction, attenuating ventricular dilation, wall stress, and cardiac hypertrophy." Journal of Clinical Investigation100(8): 1991-1999. Li, X., M. Tjwa, I. Van Hove, B. Enholm, E. Neven, K. Paavonen, M. Jeltsch, T. D. Juan, R. E. Sievers, E. Chorianopoulos, H. Wada, M. Vanwildemeersch, A. Noel, J. M. Foidart, M. L. Springer, G. von Degenfeld, M. Dewerchin, H. M. Blau, K. Alitalo, U. Eriksson, P. Carmeliet and L. Moons (2008). "Reevaluation of the role of VEGF-B suggests a restricted role in the revascularization of the ischemic myocardium." Arteriosclerosis, Thrombosis and Vascular Biology28(9): 1614-1620. Lim, B., J.-L. Park, H.-J. Kim, Y.-K. Park, J.-H. Kim, H. A. Sohn, S.-M. Noh, K.-S. Song, W.-H. Kim, Y. S. Kim and S.-Y. Kim (2014). "Integrative genomics analysis reveals the multilevel dysregulation and oncogenic characteristics of TEAD4 in gastric cancer." Carcinogenesis35(5): 1020-1027. Limana, F., M. C. Capogrossi and A. Germani (2011). "The epicardium in cardiac repair: from the stem cell view." Pharmacology & Therapeutics129(1): 82-96. Lin, Z., C. Gao, Y. Ning, X. He, W. Wu and Y.-G. Chen (2008). "The Pseudoreceptor BMP and Activin Membrane-bound Inhibitor Positively Modulates Wnt/β-Catenin Signaling." The Journal of Biological Chemistry283(48): 33053-33058. Lisa, M., N. Haleagrahara and S. Chakravarthi (2011). "Insulin-Like Growth Factor-1 (IGF-1) Reduces ischemic changes and increases circulating angiogenic factors in experimentally - induced myocardial infarction in rats." Vascular Cell3(1): 1-6. Liu, W., J. Selever, D. Wang, M. F. Lu, K. A. Moses, R. J. Schwartz and J. F. Martin (2004). "Bmp4 signaling is required for outflow-tract septation and branchial-arch artery remodeling." Proceedings of the National Academy of Sciences of the United States of America101(13): 4489- 4494. Liu, W., J. Selever, D. Wang, M. F. Lu, K. A. Moses, R. J. Schwartz and J. F. Martin (2004). "Bmp4 signaling is required for outflow-tract septation and branchial-arch artery remodeling." Proceedings Of The National Academy Of Sciences Of The United States Of America101(13): 4489-4494. Long, L., M. L. Ormiston, X. Yang, M. Southwood, S. Gräf, R. D. Machado, M. Mueller, B. Kinzel, L. M. Yung, J. M. Wilkinson, S. D. Moore, K. M. Drake, M. A. Aldred, P. Yu, P. D. Upton and N. W. Morrell (2015). "Selective enhancement of endothelial BMPR-II with BMP9 reverses pulmonary arterial hypertension." Nature medicine21(7): 777-785. Lovicu, F. J., G. Kolle, T. Yamada, M. H. Little and J. W. McAvoy (2000). "Expression of Crim1 during murine ocular development." Mechanisms of Development94(1-2): 261-265.
150 Lu, J., T. E. Landerholm, J. S. Wei, X. R. Dong, S. P. Wu, X. Liu, K. Nagata, M. Inagaki and M. W. Majesky (2001). "Coronary smooth muscle differentiation from proepicardial cells requires rhoA-mediated actin reorganization and p160 rho-kinase activity." Developmental Biology240(2): 404-418. Ma, L., M. F. Lu, R. J. Schwartz and J. F. Martin (2005). "Bmp2 is essential for cardiac cushion epithelial-mesenchymal transition and myocardial patterning." Development132(24): 5601-5611. Manasek, F. J. (1969). "Embryonic development of the heart. II. Formation of the epicardium." Journal of Embryology and Experimental Morphology22(3): 333-348. Männer, J. (1992). "The development of pericardial villi in the chick embryo." Anatomy and Embryology186(4): 379-385. Männer, J. (1993). "Experimental study on the formation of the epicardium in chick embryos." Anatomy and Embryology187(3): 281-289. Männer, J. (2013). "Microsurgical Procedures for Studying the Developmental Significance of the Proepicardium and Epicardium in Avian Embryos: PE-Blocking, PE-Photoablation, and PE- Grafting." Journal of Developmental Biology1(1): 47. Markwald, R. R., R. B. Runyan, G. T. Kitten, F. M. Funderburg, D. H. Bernanke and P. R. Brauer (1984). Use of collagen gel cultures to study heart development: proteoglycan and glycoprotein interactions during formation of endocardial cushion tissue. . The Role of Extracellular Matrix in Development. R. L. Trelstad. New York, Alan R. Liss, Inc.: 323-350. Martínez-Estrada, O. M., L. A. Lettice, A. Essafi, J. A. Guadix, J. Slight, V. Velecela, E. Hall, J. Reichmann, P. S. Devenney, P. Hohenstein, N. Hosen, R. E. Hill, R. Muñoz-Chapuli and N. D. Hastie (2010). "Wt1 is required for cardiovascular progenitor cell formation through transcriptional control of Snail and E-cadherin." Nature Genetics42(1): 89-93. Maschhoff, K. L. and H. S. Baldwin (2000). "Molecular determinants of neural crest migration." American Journal of Medical Genetics97(4): 280-288. McKeehan, W. L., F. Wang and M. Kan (1998). "The heparan sulfate-fibroblast growth factor family: diversity of structure and function." Progress in Nucleic Acid Research and Molecular Biology59: 135-176. Mebratu, Y. and Y. Tesfaigzi (2009). "How ERK1/2 Activation Controls Cell Proliferation and Cell Death Is Subcellular Localization the Answer?" Cell cycle (Georgetown, Tex.)8(8): 1168-1175. Meilhac, S. M., M. Esner, R. G. Kelly, J. F. Nicolas and M. E. Buckingham (2004). "The clonal origin of myocardial cells in different regions of the embryonic mouse heart." Devleopmental Cell6(5): 685-698. Mellgren, A. M., C. L. Smith, G. S. Olsen, B. Eskiocak, B. Zhou, M. N. Kazi, F. R. Ruiz, W. T. Pu and M. D. Tallquist (2008). "Platelet-derived growth factor receptor beta signaling is required for efficient epicardial cell migration and development of two distinct coronary vascular smooth muscle cell populations." Circulation Research103(12): 1393-1401. Miele, C., J. J. Rochford, N. Filippa, P. S. Giorgetti and O. E. Van (2000). "Insulin and insulin-like growth factor-1 induced vascular growth factor mRNA expression via different signaling pathways." Journal of Biological Chemistry275. Mikawa, T. and R. G. Gourdie (1996). "Pericardial mesoderm generates a population of coronary smooth muscle cells migrating into the heart along with ingrowth of the epicardial organ." Developmental Biology174(2): 221-232.
151 Minabe-Saegusa, C., H. Saegusa, M. Tsukahara and S. Noguchi (1998). "Sequence and expression of a novel mouse gene PRDC (protein related to DAN and cerberus) identified by a gene trap approach." Development, Growth & Differentiation40(3): 343-353. Minamisawa, S., Y. Gu, J. Ross, Jr., K. R. Chien and J. Chen (1999). "A post-transcriptional compensatory pathway in heterozygous ventricular myosin light chain 2-deficient mice results in lack of gene dosage effect during normal cardiac growth or hypertrophy." Journal of Biological Chemistry274(15): 10066-10070. Miquerol, L. and R. G. Kelly (2013). "Organogenesis of the vertebrate heart." Wiley Interdisciplinary Reviews: Developmental Biology2(1): 17-29. Miquerol, L. and R. G. Kelly (2013). "Organogenesis of the vertebrate heart." Wiley Interdiscip Rev Dev Biol2(1): 17-29. Mishina, Y., A. Suzuki, N. Ueno and R. R. Behringer (1995). "Bmpr encodes a type I bone morphogenetic protein receptor that is essential for gastrulation during mouse embryogenesis." Genes & Development9(24): 3027-3037. Mjaatvedt, C. H., T. Nakaoka, R. Moreno-Rodriguez, R. A. Norris, M. J. Kern, C. A. Eisenberg, D. Turner and R. R. Markwald (2001). "The outflow tract of the heart is recruited from a novel heart- forming field." Developmental Biology238(1): 97-109. Monzen, K., I. Shiojima, Y. Hiroi, S. Kudoh, T. Oka, E. Takimoto, D. Hayashi, T. Hosoda, A. Habara-Ohkubo, T. Nakaoka, T. Fujita, Y. Yazaki and I. Komuro (1999). "Bone morphogenetic proteins induce cardiomyocyte differentiation through the mitogen-activated protein kinase kinase kinase TAK1 and cardiac transcription factors Csx/Nkx-2.5 and GATA-4." Molecular And Cellular Biology19(10): 7096-7105. Moore, A. W., L. McInnes, J. Kreidberg, N. D. Hastie and A. Schedl (1999). "YAC complementation shows a requirement for Wt1 in the development of epicardium, adrenal gland and throughout nephrogenesis." Development126(9): 1845-1857. Morabito, C. J., R. W. Dettman, J. Kattan, J. M. Collier and J. Bristow (2001). "Positive and negative regulation of epicardial-mesenchymal transformation during avian heart development." Developmental Biology234(1): 204-215. Morbidelli, L., S. Donnini and M. Ziche (2016). "Targeting endothelial cell metabolism for cardio- protection from the toxicity of antitumor agents." Cardio-Oncology2(1): 1-12. Morrell, N. W., D. B. Bloch, P. ten Dijke, M. J. Goumans, A. Hata, J. Smith, P. B. Yu and K. D. Bloch (2016). "Targeting BMP signalling in cardiovascular disease and anaemia." Nature Reviews Cardiology13(2): 106-120. Morrell, N. W., X. Yang, P. D. Upton, K. B. Jourdan, N. Morgan, K. K. Sheares and R. C. Trembath (2001). "Altered Growth Responses of Pulmonary Artery Smooth Muscle Cells From Patients With Primary Pulmonary Hypertension to Transforming Growth Factor-β1 and Bone Morphogenetic Proteins." Circulation104(7): 790-795. Muller, Y. A., B. Li, H. W. Christinger, J. A. Wells, B. C. Cunningham and A. M. de Vos (1997). "Vascular endothelial growth factor: crystal structure and functional mapping of the kinase domain receptor binding site." Proceedings of the National Academy of Sciences U S A94(14): 7192-7197. Nahirney, P. C., T. Mikawa and D. A. Fischman (2003). "Evidence for an extracellular matrix bridge guiding proepicardial cell migration to the myocardium of chick embryos." Developmental Dynamics227(4): 511-523.
152 Nakaoka, T., K. Gonda, T. Ogita, Y. Otawara-Hamamoto, F. Okabe, Y. Kira, K. Harii, K. Miyazono, Y. Takuwa and T. Fujita (1997). "Inhibition of rat vascular smooth muscle proliferation in vitro and in vivo by bone morphogenetic protein-2." Journal of Clinical Investigation100(11): 2824-2832. Narmoneva, D. A., R. Vukmirovic, M. E. Davis, R. D. Kamm and R. T. Lee (2004). "Endothelial Cells Promote Cardiac Myocyte Survival and Spatial Reorganization: Implications for Cardiac Regeneration." Circulation110(8): 962-968. Neuhaus, H., V. Rosen and R. S. Thies (1999). "Heart specific expression of mouse BMP-10 a novel member of the TGF-beta superfamily." Mechanisms of Development80(2): 181-184. Nieto, M. A., M. G. Sargent, D. G. Wilkinson and J. Cooke (1994). "Control of cell behavior during vertebrate development by Slug, a zinc finger gene." Science264(5160): 835-839. Nyström, J., K. Hultenby, S. Ek, J. Sjölund, H. Axelson, K. Jirström, M. A. Saleem, K. Nilsson and M. E. Johansson (2009). "CRIM1 is localized to the podocyte filtration slit diaphragm of the adult human kidney." Nephrology Dialysis Transplantation24(7): 2038-2044. O'Brien, T. X., K. J. Lee and K. R. Chien (1993). "Positional specification of ventricular myosin light chain 2 expression in the primitive murine heart tube." Proceedings of the National Academy of Sciences U S A90(11): 5157-5161. Olivey, H. E. and E. C. Svensson (2010). "Epicardial-myocardial signaling directing coronary vasculogenesis." Circulation Research106(5): 818-832. Olofsson, B., K. Pajusola, A. Kaipainen, G. von Euler, V. Joukov, O. Saksela, A. Orpana, R. F. Pettersson, K. Alitalo and U. Eriksson (1996). "Vascular endothelial growth factor B, a novel growth factor for endothelial cells." Proceedings of the National Academy of Sciences USA93(6): 2576-2581. Olsson, A. K., A. Dimberg, J. Kreuger and L. Claesson-Welsh (2006). "VEGF receptor signalling - in control of vascular function." Nature Reviews Molecular Cell Biology7(5): 359-371. Onichtchouk, D., Y. G. Chen, R. Dosch, V. Gawantka, H. Delius, J. Massague and C. Niehrs (1999). "Silencing of TGF-beta signalling by the pseudoreceptor BAMBI." Nature401(6752): 480- 485. Pal, R. and A. Khanna (2006). "Role of smad- and wnt-dependent pathways in embryonic cardiac development." Stem Cells Development15(1): 29-39. Parker, M. W., P. Xu, H. F. Guo and C. W. Vander Kooi (2012). "Mechanism of selective VEGF-A binding by neuropilin-1 reveals a basis for specific ligand inhibition." PLoS One7(11): e49177. Pate, M., V. Damarla, D. S. Chi, S. Negi and G. Krishnaswamy (2010). Chapter 5 - Endothelial Cell Biology: Role in the Inflammatory Response. Advances in Clinical Chemistry. S. M. Gregory, Elsevier. Volume 52: 109-130. Pawson, T., P. Olivier, M. Rozakis-Adcock, J. McGlade and M. Henkemeyer (1993). "Proteins with SH2 and SH3 domains couple receptor tyrosine kinases to intracellular signalling pathways." Philosophical Transactions of the Royal Society London B Biological Sciences340(1293): 279-285. Pearce, J. J. H., G. Penny and J. Rossant (1999). "A Mouse Cerberus/Dan-Related Gene Family." Developmental Biology209(1): 98-110. Pennisi, D. J. (2016). "Towards Consensus on Coronary Vessel Development: Coronary Arterial Endothelial Cells Derive Primarily From the Sinus venosus During Embryogenesis." Circulation Research118(12): 1861-1862. 153 Pennisi, D. J., V. L. Ballard and T. Mikawa (2003). "Epicardium is required for the full rate of myocyte proliferation and levels of expression of myocyte mitogenic factors FGF2 and its receptor, FGFR-1, but not for transmural myocardial patterning in the embryonic chick heart." Developmental Dynamics228(2): 161-172. Pennisi, D. J., G. Kinna, H. S. Chiu, D. G. Simmons, L. Wilkinson and M. H. Little (2012). "Crim1 has an essential role in glycogen trophoblast cell and sinusoidal-trophoblast giant cell development in the placenta." Placenta33(3): 175-182. Pennisi, D. J. and T. Mikawa (2005). "Normal patterning of the coronary capillary plexus is dependent on the correct transmural gradient of FGF expression in the myocardium." Developmental Biology279(2): 378-390. Pennisi, D. J. and T. Mikawa (2009). "FGFR-1 is required by epicardium-derived cells for myocardial invasion and correct coronary vascular lineage differentiation." Developmental Biology328(1): 148-159. Pennisi, D. J., L. Wilkinson, G. Kolle, M. L. Sohaskey, K. Gillinder, M. J. Piper, J. W. McAvoy, F. J. Lovicu and M. H. Little (2007). "Crim1KST264/KST264 mice display a disruption of the Crim1 gene resulting in perinatal lethality with defects in multiple organ systems." Developmental Dynamics236(2): 502–511. Pérez-Pomares, J. M., R. Carmona, M. González-Iriarte, G. Atencia, A. Wessels and R. Muñoz- Chápuli (2002). "Origin of coronary endothelial cells from epicardial mesothelium in avian embryos." International Journal of Developmental Biology46(8): 1005-1013. Pérez-Pomares, J. M., D. Macías, L. García-Garrido and R. Muñoz-Chápuli (1997). "Contribution of the primitive epicardium to the subepicardial mesenchyme in hamster and chick embryos." Developmental Dynamics210(2): 96-105. Perros, F., P. Dorfmuller, R. Souza, I. Durand-Gasselin, V. Godot, F. Capel, S. Adnot, S. Eddahibi, M. Mazmanian, E. Fadel, P. Herve, G. Simonneau, D. Emilie and M. Humbert (2007). "Fractalkine- induced smooth muscle cell proliferation in pulmonary hypertension." European Respiratory Journal29(5): 937-943. Philippou, A., M. Maridaki, S. Pneumaticos and M. Koutsilieris (2014). "The Complexity of the IGF1 Gene Splicing, Posttranslational Modification and Bioactivity." Molecular Medicine20(1): 202-214. Phua, Y. L., N. Martel, D. J. Pennisi, M. H. Little and L. Wilkinson (2012). "Distinct sites of renal fibrosis in Crim1 mutant mice arise from multiple cellular origins." Journal of Pathology. Piccolo, S., Y. Sasai, B. Lu and E. M. De Robertis (1996). "Dorsoventral patterning in Xenopus: inhibition of ventral signals by direct binding of chordin to BMP-4." Cell86(4): 589-598. Poelmann, R. E., A. C. Gittenberger-de Groot, M. M. Mentink, R. Bökenkamp and B. Hogers (1993). "Development of the cardiac coronary vascular endothelium, studied with antiendothelial antibodies, in chicken-quail chimeras." Circulation Research73(3): 559-568. Ponferrada, V. G., J. Fan, J. E. Vallance, S. Hu, A. Mamedova, S. A. Rankin, M. Kofron, A. M. Zorn, R. S. Hegde and R. A. Lang (2012). "CRIM1 complexes with β-catenin and cadherins, stabilizes cell-cell junctions and is critical for neural morphogenesis." PLoS One7(3): e32635. Porrello, E. R., A. I. Mahmoud, E. Simpson, B. A. Johnson, D. Grinsfelder, D. Canseco, P. P. Mammen, B. A. Rothermel, E. N. Olson and H. A. Sadek (2013). "Regulation of neonatal and adult mammalian heart regeneration by the miR-15 family." Proceedings of the National Academy of Sciences USA110(1): 187-192.
154 Porrello, E. R., K. D. Pfleger, R. M. Seeber, H. Qian, C. Oro, F. Abogadie, L. M. Delbridge and W. G. Thomas (2011). "Heteromerization of angiotensin receptors changes trafficking and arrestin recruitment profiles." Cell Signal23(11): 1767-1776. Prenkert, M., B. Uggla, U. Tidefelt and H. Strid (2010). "CRIM1 is expressed at higher levels in drug-resistant than in drug-sensitive myeloid leukemia HL60 cells." Anticancer Research30(10): 4157-4161. Punglia, R., M. Lu, J. Hsu, M. Kuroki, M. Tolentino, K. Keough, A. P. Levy, N. S. Levy, M. A. Goldberg, R. J. D'Amato and A. P. Adamis (1997). "Regulation of vascular endothelial growth factor expression by insulin-like growth factor 1." Diabetes46. Radu, M. and J. Chernoff (2013). "An in vivo Assay to Test Blood Vessel Permeability." Journal of Visualized Experiments : JoVE(73): 50062. Rajendran, P., T. Rengarajan, J. Thangavel, Y. Nishigaki, D. Sakthisekaran, G. Sethi and I. Nishigaki (2013). "The Vascular Endothelium and Human Diseases." International Journal of Biological Sciences9(10): 1057-1069. Randhawa, R. and P. Cohen (2005). "The role of the insulin-like growth factor system in prenatal growth." Molecular Genetics and Metabolism86(1-2): 84-90. Ratajska, A., B. Ciszek and A. Sowinska (2003). "Embryonic development of coronary vasculature in rats: corrosion casting studies." Anat Records A Discovery in Molecular, Cellular and Evolutionary Biology270(2): 109-116. Ray, B. N., N. Y. Lee, T. How and G. C. Blobe (2010). "ALK5 phosphorylation of the endoglin cytoplasmic domain regulates Smad1/5/8 signaling and endothelial cell migration." Carcinogenesis31(3): 435-441. Red-Horse, K., H. Ueno, I. L. Weissman and M. A. Krasnow (2010). "Coronary arteries form by developmental reprogramming of venous cells." Nature464(7288): 549-553. Reese, D. E., T. Mikawa and D. M. Bader (2002). "Development of the Coronary Vessel System." Circulation Research91(9): 761. Remy, I. and S. W. Michnick (2003). "Dynamic visualization of expressed gene networks." Journal of Cellular Physiology196(3): 419-429. Rider, C. C., Mulloy, Barbara (2010). "Bone morphogenetic protein and growth differentiation factor cytokine families and their protein antagonists " Biochemical Journal 429(1): 1-12. Riley, P. R. (2012). "An epicardial floor plan for building and rebuilding the mammalian heart." Current Topics in Developmental Biology100: 233-251. Risau, W. (1991). "Embryonic angiogenesis factors." Pharmacology & Therapeutics51(3): 371-376. Rivera-Feliciano, J. and C. J. Tabin (2006). "Bmp2 instructs cardiac progenitors to form the heart- valve-inducing field." Developmental Biology295(2): 580-588. Robinson, C. J. and S. E. Stringer (2001). "The splice variants of vascular endothelial growth factor (VEGF) and their receptors." Journal of Cell Science114(Pt 5): 853-865. Romano, L. A. and R. B. Runyan (1999). "Slug is a mediator of epithelial-mesenchymal cell transformation in the developing chicken heart." Developmental Biology212(1): 243-254. Romano, L. A. and R. B. Runyan (2000). "Slug is an essential target of TGFbeta2 signaling in the developing chicken heart." Developmental Biology223(1): 91-102.
155 Rosenblatt-Velin, N., M. G. Lepore, C. Cartoni, F. Beermann and T. Pedrazzini (2005). "FGF-2 controls the differentiation of resident cardiac precursors into functional cardiomyocytes." The Journal Of Clinical Investigation115(7): 1724-1733. Rudat, C. and A. Kispert (2012). "Wt1 and epicardial fate mapping." Circulation Research111(2): 165-169. Rudijanto, A. (2007). "The role of vascular smooth muscle cells on the pathogenesis of atherosclerosis." Acta Medica Indonesiana39(2): 86-93. Ryan, M. D., A. M. King and G. P. Thomas (1991). "Cleavage of foot-and-mouth disease virus polyprotein is mediated by residues located within a 19 amino acid sequence." Journal of General Virology72 ( Pt 11): 2727-2732. Samad, T. A., A. Rebbapragada, E. Bell, Y. Zhang, Y. Sidis, S. J. Jeong, J. A. Campagna, S. Perusini, D. A. Fabrizio, A. L. Schneyer, H. Y. Lin, A. H. Brivanlou, L. Attisano and C. J. Woolf (2005). "DRAGON, a bone morphogenetic protein co-receptor." Journal of Biological Chemistry280(14): 14122-14129. Sánchez, N. S. and J. V. Barnett (2012). "TGFbeta and BMP-2 regulate epicardial cell invasion via TGFbetaR3 activation of the Par6/Smurf1/RhoA pathway." Cell Signal24(2): 539-548. Sasai, Y., B. Lu, H. Steinbeisser, D. Geissert, L. K. Gont and E. M. De Robertis (1994). "Xenopus chordin: a novel dorsalizing factor activated by organizer-specific homeobox genes." Cell79(5): 779-790. Scharpfenecker, M., M. van Dinther, Z. Liu, R. L. van Bezooijen, Q. Zhao, L. Pukac, C. W. Lowik and P. ten Dijke (2007). "BMP-9 signals via ALK1 and inhibits bFGF-induced endothelial cell proliferation and VEGF-stimulated angiogenesis." Journal of Cell Science120(Pt 6): 964-972. Schlueter, J., J. Männer and T. Brand (2006). "BMP is an important regulator of proepicardial identity in the chick embryo." Developmental Biology295(2): 546-558. Schneider, V. A. and M. Mercola (2001). "Wnt antagonism initiates cardiogenesis in Xenopus laevis." Genes and Development15(3): 304-315. Schoenwolf, G. C. and V. Garcia-Martinez (1995). "Primitive-streak origin and state of commitment of cells of the cardiovascular system in avian and mammalian embryos." Cellular and Molecular Biology Research41(4): 233-240. Semela, D. and J.-F. o. Dufour (2004). "Angiogenesis and hepatocellular carcinoma." Journal of Hepatology41(5): 864-880. Semela, D. and J. F. Dufour (2004). "Angiogenesis and hepatocellular carcinoma." Journal of Hepatology41(5): 864-880. Serluca, F. C. (2008). "Development of the proepicardial organ in the zebrafish." Developmental Biology315(1): 18-27. Simon, M., H. J. Grone, O. Johren, J. Kullmer, K. H. Plate, W. Risau and E. Fuchs (1995). "Expression of vascular endothelial growth factor and its receptors in human renal ontogenesis and in adult kidney." American Journal of Physiology268(2 Pt 2): F240-250. Smart, N. and P. R. Riley (2012). "The epicardium as a candidate for heart regeneration." Future Cardiology8(1): 53-69.
156 Smart, N., C. A. Risebro, A. A. Melville, K. Moses, R. J. Schwartz, K. R. Chien and P. R. Riley (2007). "Thymosin b4 induces adult epicardial progenitor mobilization and neovascularization." Nature445(7124): 177–182. Smith, C. L., S. T. Baek, C. Y. Sung and M. D. Tallquist (2011). "Epicardial-derived cell epithelial- to-mesenchymal transition and fate specification require PDGF receptor signaling." Circulation Research108(12): e15-e26. Smith, L. E., W. Shen, C. Perruzzi, S. Soker, F. Kinose, X. Xu, G. Robinson, S. Driver, J. Bischoff, B. Zhang, J. M. Schaeffer and D. R. Senger (1999). "Regulation of vascular endothelial growth factor-dependent retinal neovascularization by insulin-like growth factor-1 receptor." Nature Medicine 5(12): 1390-1395. Smith, W. C. and R. M. Harland (1992). "Expression cloning of noggin, a new dorsalizing factor localized to the Spemann organizer in Xenopus embryos." Cell70(5): 829-840. Soriano, P. (1999). "Generalized lacZ expression with the ROSA26 Cre reporter strain." Nature Genetics21(1): 70-71. Souders, C. A., S. L. K. Bowers and T. A. Baudino (2009). "Cardiac Fibroblast: The Renaissance Cell." Circulation Research105(12): 1164-1176. Spallarossa, P., C. Brunelli, F. Minuto, D. Caruso, M. Battistini, S. Caponnetto and R. Cordera (1996). "Insulin-like growth factor-I and angiographically documented coronary artery disease." American Journal of Cardiology77(2): 200-202. Spies, M., O. Nesic, R. E. Barrow, J. R. Perez-Polo and D. N. Herndon (2001). "Liposomal IGF-1 gene transfer modulates pro- and anti-inflammatory cytokine mRNA expression in the burn wound." Gene Therapy8(18): 1409-1415. Sridurongrit, S., J. Larsson, R. Schwartz, P. Ruiz-Lozano and V. Kaartinen (2008). "Signaling via the Tgf-beta type I receptor Alk5 in heart development." Developmental Biology322(1): 208-218. Star, G. P., M. Giovinazzo and D. Langleben (2009). "Effects of bone morphogenic proteins and transforming growth factor-beta on In-vitro production of endothelin-1 by human pulmonary microvascular endothelial cells." Vascular Pharmacology50(1-2): 45-50. Stennard, F. A. and R. P. Harvey (2005). "T-box transcription factors and their roles in regulatory hierarchies in the developing heart." Development132(22): 4897-4910. Stoller, J. Z. and J. A. Epstein (2005). "Cardiac neural crest." Seminars in Cell and Developmental Biology16(6): 704-715. Sucov, H. M., Y. Gu, S. Thomas, P. Li and M. Pashmforoush (2009). "Epicardial control of myocardial proliferation and morphogenesis." Pediatric Cardiology30(5): 617-625. Sukhanov, S., Y. Higashi, S. Y. Shai, C. Vaughn, J. Mohler, Y. Li, Y. H. Song, J. Titterington and P. Delafontaine (2007). "IGF-1 reduces inflammatory responses, suppresses oxidative stress, and decreases atherosclerosis progression in ApoE-deficient mice." Arteriosclerosis, Thrombosis and Vascular Biology27(12): 2684-2690. Sun, T., H. Cao, L. Xu, B. Zhu, Q. Gu and X. Xu (2011). "Insulin-Like Growth Factor Binding Protein-Related Protein 1 Mediates VEGF-Induced Proliferation, Migration and Tube Formation of Retinal Endothelial Cells." Current Eye Research36(4): 341-349. Sun, Y., X. Liang, N. Najafi, M. Cass, L. Lin, C. L. Cai, J. Chen and S. M. Evans (2007). "Islet 1 is expressed in distinct cardiovascular lineages, including pacemaker and coronary vascular cells." Developmental Biology304(1): 286-296. 157 Suzuki, Y., K. Montagne, A. Nishihara, T. Watabe and K. Miyazono (2008). "BMPs Promote Proliferation and Migration of Endothelial Cells via Stimulation of VEGF-A/VEGFR2 and Angiopoietin-1/Tie2 Signalling." Journal of Biochemistry143(2): 199-206. Suzuki, Y., N. Ohga, Y. Morishita, K. Hida, K. Miyazono and T. Watabe (2010). "BMP-9 induces proliferation of multiple types of endothelial cells in vitro and in vivo." Journal of Cell Science123(Pt 10): 1684-1692. Szebenyi, G. and J. F. Fallon (1999). "Fibroblast growth factors as multifunctional signaling factors." International Review of Cytology185: 45-106. Tamura, K., K. Hashimoto, K. Suzuki, M. Yoshie, M. Kutsukake and T. Sakurai (2009). "Insulin- like growth factor binding protein-7 (IGFBP7) blocks vascular endothelial cell growth factor (VEGF)-induced angiogenesis in human vascular endothelial cells." European Journal of Pharmacology610(1–3): 61-67. Tang, W., I. Ehrlich, S. B. Wolff, A. M. Michalski, S. Wolfl, M. T. Hasan, A. Luthi and R. Sprengel (2009). "Faithful expression of multiple proteins via 2A-peptide self-processing: a versatile and reliable method for manipulating brain circuits." Journal of Neuroscience29(27): 8621-8629. Teichert-Kuliszewska, K., M. J. Kutryk, M. A. Kuliszewski, G. Karoubi, D. W. Courtman, L. Zucco, J. Granton and D. J. Stewart (2006). "Bone morphogenetic protein receptor-2 signaling promotes pulmonary arterial endothelial cell survival: implications for loss-of-function mutations in the pathogenesis of pulmonary hypertension." Circulation Research98(2): 209-217. Theiler, K. (1989). The house mouse : atlas of embryonic development. New York. Berlin : Springer-Verlag. Tian, X., T. Hu, H. Zhang, L. He, X. Huang, Q. Liu, W. Yu, Z. Yang, Z. Zhang, T. P. Zhong, X. Yang, Y. Yan, A. Baldini, Y. Sun, J. Lu, R. J. Schwartz, S. M. Evans, A. C. Gittenberger-de Groot, K. Red-Horse and B. Zhou (2013). "Subepicardial endothelial cells invade the embryonic ventricle wall to form coronary arteries." Cell Research23(9): 1075-1090. Tomanek, R. J., Y. Ishii, J. S. Holifield, C. L. Sjogren, H. K. Hansen and T. Mikawa (2006). "VEGF family members regulate myocardial tubulogenesis and coronary artery formation in the embryo." Circulation Research98(7): 947–953. Tomanek, R. J., A. Sandra, W. Zheng, T. Brock, R. J. Bjercke and J. S. Holifield (2001). "Vascular endothelial growth factor and basic fibroblast growth factor differentially modulate early postnatal coronary angiogenesis." Circulation Research88(11): 1135-1141. Tomanek, R. J. and W. Zheng (2002). "Role of growth factors in coronary morphogenesis." Texas Heart Institute Journal29(4): 250-254. Topol, L. Z., M. Marx, D. Laugier, N. N. Bogdanova, N. V. Boubnov, P. A. Clausen, G. Calothy and D. G. Blair (1997). "Identification of drm, a novel gene whose expression is suppressed in transformed cells and which can inhibit growth of normal but not transformed cells in culture." Molecular and Cellular Biology17(8): 4801-4810. Torella, D., M. Rota, D. Nurzynska, E. Musso, A. Monsen, I. Shiraishi, E. Zias, K. Walsh, A. Rosenzweig, M. A. Sussman, K. Urbanek, B. Nadal-Ginard, J. Kajstura, P. Anversa and A. Leri (2004). "Cardiac Stem Cell and Myocyte Aging, Heart Failure, and Insulin-Like Growth Factor-1 Overexpression." Circulation Research94(4): 514-524.
158 Torrent-Guasp, F., M. J. Kocica, A. F. Corno, M. Komeda, F. Carreras-Costa, A. Flotats, J. Cosin- Aguillar and H. Wen (2005). "Towards new understanding of the heart structure and function." European Journal of Cardio-Thoracic Surgery27(2): 191-201. Ueno, S., G. Weidinger, T. Osugi, A. D. Kohn, J. L. Golob, L. Pabon, H. Reinecke, R. T. Moon and C. E. Murry (2007). "Biphasic role for Wnt/β-catenin signaling in cardiac specification in zebrafish and embryonic stem cells." Proceedings of the National Academy of Sciences104(23): 9685-9690. Umehara, H., E. T. Bloom, T. Okazaki, Y. Nagano, O. Yoshie and T. Imai (2004). "Fractalkine in vascular biology: from basic research to clinical disease." Arteriosclerosis, Thrombosis and Vascular Biology24(1): 34-40. Umulis, D., M. B. O'Connor and S. S. Blair (2009). "The extracellular regulation of bone morphogenetic protein signaling." Development136(22): 3715-3728. Umulis, D., M. B. O'Connor and S. S. Blair (2009). "The extracellular regulation of bone morphogenetic protein signaling." Development136(22): 3715-3728. Van den Akker, N. M., L. C. Winkel, M. H. Nisancioglu, S. Maas, L. J. Wisse, A. Armulik, R. E. Poelmann, H. Lie-Venema, C. Betsholtz and A. C. Gittenberger-de Groot (2008). "PDGF-B signaling is important for murine cardiac development: its role in developing atrioventricular valves, coronaries, and cardiac innervation." Developmental Dynamics237(2): 494-503. van Wijk, B., Q. D. Gunst, A. F. M. Moorman and M. J. B. van den Hoff (2012). "Cardiac Regeneration from Activated Epicardium." PLoS ONE7(9): e44692. van Wijk, B., A. F. Moorman and M. J. van den Hoff (2007). "Role of bone morphogenetic proteins in cardiac differentiation." Cardiovascular Research74(2): 244-255. van Wijk, B., G. van den Berg, R. Abu-Issa, P. Barnett, S. van der Velden, M. Schmidt, J. M. Ruijter, M. L. Kirby, A. F. Moorman and M. J. van den Hoff (2009). "Epicardium and myocardium separate from a common precursor pool by crosstalk between bone morphogenetic protein- and fibroblast growth factor-signaling pathways." Circulation Research105(5): 431-441. Varga, A. C. and J. L. Wrana (2005). "The disparate role of BMP in stem cell biology." Oncogene24(37): 5713-5721. Varga, A. C. and J. L. Wrana (2005). "The disparate role of BMP in stem cell biology." Oncogene24(37): 5713-5721. Vega-Hernandez, M., A. Kovacs, S. De Langhe and D. M. Ornitz (2011). "FGF10/FGFR2b signaling is essential for cardiac fibroblast development and growth of the myocardium." Development138(15): 3331-3340. Vinson, C., A. Acharya and E. J. Taparowsky (2006). "Deciphering B-ZIP transcription factor interactions in vitro and in vivo." Biochimica Biophysica Acta1759(1-2): 4-12. Virágh, S. and C. E. Challice (1981). "The origin of the epicardium and the embryonic myocardial circulation in the mouse." The Anatomical Record201(1): 157-168. Virágh, S., A. C. Gittenberger-de Groot, R. E. Poelmann and F. Kálmán (1993). "Early development of quail heart epicardium and associated vascular and glandular structures." Anatomy and Embryology188(4): 381-393. Wada, A. M., S. G. Willet and D. Bader (2003). "Coronary vessel development: a unique form of vasculogenesis." Arteriosclerosis, Thrombosis and Vascular Biology23(12): 2138-2145.
159 Waldo, K., S. Miyagawa-Tomita, D. Kumiski and M. L. Kirby (1998). "Cardiac neural crest cells provide new insight into septation of the cardiac outflow tract: aortic sac to ventricular septal closure." Developmental Biology196(2): 129-144. Waldo, K. L., M. R. Hutson, C. C. Ward, M. Zdanowicz, H. A. Stadt, D. Kumiski, R. Abu-Issa and M. L. Kirby (2005). "Secondary heart field contributes myocardium and smooth muscle to the arterial pole of the developing heart." Developmental Biology281(1): 78-90. Waldo, K. L., D. H. Kumiski, K. T. Wallis, H. A. Stadt, M. R. Hutson, D. H. Platt and M. L. Kirby (2001). "Conotruncal myocardium arises from a secondary heart field." Development128(16): 3179-3188. Wang, L., H. Zeng, P. Wang, S. Soker and D. Mukhopadhyay (2003). "Neuropilin-1-mediated vascular permeability factor/vascular endothelial growth factor-dependent endothelial cell migration." Journal of Biological Chemistry278(49): 48848-48860. Wang, R. N., J. Green, Z. Wang, Y. Deng, M. Qiao, M. Peabody, Q. Zhang, J. Ye, Z. Yan, S. Denduluri, O. Idowu, M. Li, C. Shen, A. Hu, R. C. Haydon, R. Kang, J. Mok, M. J. Lee, H. L. Luu and L. L. Shi (2014). "Bone Morphogenetic Protein (BMP) signaling in development and human diseases." Genes & Diseases1(1): 87-105. Ward, C., H. Stadt, M. Hutson and M. L. Kirby (2005). "Ablation of the secondary heart field leads to tetralogy of Fallot and pulmonary atresia." Developmental Biology284(1): 72-83. Wiesmann, C., G. Fuh, H. W. Christinger, C. Eigenbrot, J. A. Wells and A. M. de Vos (1997). "Crystal Structure at 1.7 Ã… Resolution of VEGF in Complex with Domain 2 of the Flt-1 Receptor." Cell91(5): 695-704. Wilkinson, L., T. Gilbert, G. Kinna, L. A. Ruta, D. Pennisi, M. Kett and M. H. Little (2007). "Crim1KST264/KST264 mice implicate Crim1 in the regulation of vascular endothelial growth factor-A activity during glomerular vascular development." Journal of the American Society of Nephrology18(6): 1697–1708. Wilkinson, L., G. Kolle, D. Wen, M. Piper, J. Scott and M. Little (2003). "CRIM1 regulates the rate of processing and delivery of bone morphogenetic proteins to the cell surface." Journal of Biological Chemistry278(36): 34181-34188. Wilm, B., A. Ipenberg, N. D. Hastie, J. B. Burch and D. M. Bader (2005). "The serosal mesothelium is a major source of smooth muscle cells of the gut vasculature." Development132(23): 5317-5328. Wu, B., Z. Zhang, W. Lui, X. Chen, Y. Wang, A. Chamberlain, R. A. Moreno-Rodriquez, R. R. Markwald, B. P. O’Rourke, D. J. Sharp, D. Zheng, J. Lenz, H. S. Baldwin, C.-P. Chang and B. Zhou (2012). "Endocardial Cells Form the Coronary Arteries by Angiogenesis through Myocardial- Endocardial VEGF Signaling." Cell151(5): 1083-1096. Wu, M., C. L. Smith, J. A. Hall, I. Lee, K. Luby-Phelps and M. D. Tallquist (2010). "Epicardial spindle orientation controls cell entry into the myocardium." Developmental Cell19(1): 114-125. Wu, M. Y. and C. S. Hill (2009). "Tgf-beta superfamily signaling in embryonic development and homeostasis." Developmental Cell16(3): 329-343. Xie, L., B. K. Law, A. M. Chytil, K. A. Brown, M. E. Aakre and H. L. Moses (2004). "Activation of the Erk pathway is required for TGF-beta1-induced EMT in vitro." Neoplasia6(5): 603-610. Yamada, P. M. and K. W. Lee (2009). "Perspectives in mammalian IGFBP-3 biology: local vs. systemic action." American Journal of Physiology Cell Physiology296(5): C954-976.
160 Yamagishi, T., Y. Nakajima, K. Miyazono and H. Nakamura (1999). "Bone morphogenetic protein- 2 acts synergistically with transforming growth factor-beta3 during endothelial-mesenchymal transformation in the developing chick heart." Journal of Cell Physiology180(1): 35-45. Yan, X., Z. Lin, F. Chen, X. Zhao, H. Chen, Y. Ning and Y. G. Chen (2009). "Human BAMBI cooperates with Smad7 to inhibit transforming growth factor-beta signaling." Journal of Biological Chemistry284(44): 30097-30104. Yang, G. G. (2011). "Phosphohistone H3: new standard IHC marker for grading tumors via mitotic indexing." MLO Medical Laboratory Observer43(7): 20, 23.
Yang, J. T., H. Rayburn and R. O. Hynes (1995). "Cell adhesion events mediated by a4 integrins are essential in placental and cardiac development." Development121(2): 549-560. Yang, X., L. H. Castilla, X. Xu, C. Li, J. Gotay, M. Weinstein, P. P. Liu and C. X. Deng (1999). "Angiogenesis defects and mesenchymal apoptosis in mice lacking SMAD5." Development126(8): 1571-1580. Yu, P. B., D. Y. Deng, C. S. Lai, C. C. Hong, G. D. Cuny, M. L. Bouxsein, D. W. Hong, P. M. McManus, T. Katagiri, C. Sachidanandan, N. Kamiya, T. Fukuda, Y. Mishina, R. T. Peterson and K. D. Bloch (2008). "BMP type I receptor inhibition reduces heterotopic ossification." Nature Medicine14(12): 1363-1369. Yuan, L., D. Moyon, L. Pardanaud, C. Bréant, M. J. Karkkainen, K. Alitalo and A. Eichmann (2002). "Abnormal lymphatic vessel development in neuropilin 2 mutant mice." Development129(20): 4797-4806. Yuasa, S. and K. Fukuda (2009). "Multiple roles for BMP signaling in cardiac development." Drug discovery today: therapeutic strategies; Cardiovascular stem cellsIn Press DOI: 10.1016/j.ddstr.2008.12.001. Yuasa, S., Y. Itabashi, U. Koshimizu, T. Tanaka, K. Sugimura, M. Kinoshita, F. Hattori, S. Fukami, T. Shimazaki, S. Ogawa, H. Okano and K. Fukuda (2005). "Transient inhibition of BMP signaling by Noggin induces cardiomyocyte differentiation of mouse embryonic stem cells." Nature Biotechnology23(5): 607-611. Yutzey, K. E. and M. L. Kirby (2002). "Wherefore heart thou? Embryonic origins of cardiogenic mesoderm." Developmental Dynamics223(3): 307-320. Zaffran, S., R. G. Kelly, S. M. Meilhac, M. E. Buckingham and N. A. Brown (2004). "Right ventricular myocardium derives from the anterior heart field." Circulation Research95(3): 261-268. Zakin, L. and E. M. De Robertis (2010). "Extracellular regulation of BMP signaling." Current Biology20(3): R89-R92. Zakin, L., C. A. Metzinger, E. Y. Chang, C. Coffinier and E. M. De Robertis (2008). "Development of the vertebral morphogenetic field in the mouse: Interactions between Crossveinless-2 and Twisted Gastrulation." Developmental Biology323(1): 6-18. Zamora, M., J. Männer and P. Ruiz-Lozano (2007). "Epicardium-derived progenitor cells require beta-catenin for coronary artery formation." Proceedings Of The National Academy Of Sciences Of The United States Of America104(46): 18109-18114. Zhang, H., W. Pu, G. Li, X. Huang, L. He, X. Tian, Q. Liu, L. Zhang, S. M. Wu, H. M. Sucov and B. Zhou (2016). "Endocardium Minimally Contributes to Coronary Endothelium in the Embryonic Ventricular Free Walls." Circulation Research118(12): 1880-1893.
161 Zhang, J. L., Y. Huang, L. Y. Qiu, J. Nickel and W. Sebald (2007). "von Willebrand factor type C domain-containing proteins regulate bone morphogenetic protein signaling through different recognition mechanisms." Journal of Biological Chemistry282(27): 20002-20014. Zhang, J. L., L. J. Patterson, L. Y. Qiu, D. Graziussi, W. Sebald and M. Hammerschmidt (2010). "Binding between Crossveinless-2 and Chordin von Willebrand factor type C domains promotes BMP signaling by blocking Chordin activity." PLoS One5(9): e12846. Zhang, J. L., L. Y. Qiu, A. Kotzsch, S. Weidauer, L. Patterson, M. Hammerschmidt, W. Sebald and T. D. Mueller (2008). "Crystal structure analysis reveals how the Chordin family member crossveinless 2 blocks BMP-2 receptor binding." Developmental Cell14(5): 739-750. Zhang, Y., J. Fan, J. W. K. Ho, T. Hu, S. C. Kneeland, X. Fan, Q. Xi, M. A. Sellarole, W. N. de Vries, W. Lu, S. A. Lachke, R. A. Lang, S. W. M. John and R. L. Maas (2016). "Crim1 regulates integrin signaling in murine lens development." Development143(2): 356-366. Zhou, B., L. B. Honor, H. He, Q. Ma, J. H. Oh, C. Butterfield, R. Z. Lin, J. M. Melero-Martin, E. Dolmatova, H. S. Duffy, A. Gise, P. Zhou, Y. W. Hu, G. Wang, B. Zhang, L. Wang, J. L. Hall, M. A. Moses, F. X. McGowan and W. T. Pu (2011). "Adult mouse epicardium modulates myocardial injury by secreting paracrine factors." Journal of Clinical Investigation121(5): 1894-1904. Zhou, B., Q. Ma, S. Rajagopal, S. M. Wu, I. Domian, J. Rivera-Feliciano, D. Jiang, A. von Gise, S. Ikeda, K. R. Chien and W. T. Pu (2008). "Epicardial progenitors contribute to the cardiomyocyte lineage in the developing heart." Nature454(7200): 109-113. Zhou, B. and W. T. Pu (2012). "Genetic Cre-loxP assessment of epicardial cell fate using Wt1- driven Cre alleles." Circulation Research111(11): e276-280. Zhou, B., A. von Gise, Q. Ma, J. Rivera-Feliciano and W. T. Pu (2008). "Nkx2-5- and Isl1- expressing cardiac progenitors contribute to proepicardium." Biochemical And Biophysical Research Communications375(3): 450-453. Zhou, Y., T. J. Cashman, K. R. Nevis, P. Obregon, S. A. Carney, Y. Liu, A. Gu, C. Mosimann, S. Sondalle, R. E. Peterson, W. Heideman, C. E. Burns and C. G. Burns (2011). "Latent TGF-beta binding protein 3 identifies a second heart field in zebrafish." Nature474(7353): 645-648. Zhu, W., I. Shiojima, Y. Ito, Z. Li, H. Ikeda, M. Yoshida, A. T. Naito, J.-i. Nishi, H. Ueno, A. Umezawa, T. Minamino, T. Nagai, A. Kikuchi, M. Asashima and I. Komuro (2008). "IGFBP-4 is an inhibitor of canonical Wnt signalling required for cardiogenesis." Nature454(7202): 345-349. Zimmerman, L. B., J. M. De Jesus-Escobar and R. M. Harland (1996). "The Spemann organizer signal noggin binds and inactivates bone morphogenetic protein 4." Cell86(4): 599-606.
162
Appendices
163 Published research paper 1. Crim1 has cell-autonomous and paracrine roles during embryonic heart development.
Swati Iyer1, Fang Yu Chou1, Richard Wang1, Han Sheng Chiu1, Vinay K. Sundar Raju1, Melissa H. Little2,3,4, Walter G. Thomas1, Michael Piper1,5#, David J. Pennisi1#.
1School of Biomedical Sciences, The University of Queensland, Brisbane, 4072, Australia, 2Institute for Molecular Bioscience, The University of Queensland, Brisbane, 4072, Australia, 3Murdoch Children‟s Research Institute, Melbourne, 4Department of Pediatrics, University of Melbourne and 5Queensland Brain Institute, The University of Queensland, Brisbane, 4072, Australia.
#Address correspondence to: Dr. David Pennisi School of Biomedical Sciences, The University of Queensland, Brisbane, 4072, Australia Email: [email protected]
Dr. Michael Piper School of Biomedical Sciences and Queensland Brain Institute, The University of Queensland, Brisbane, 4072, Australia. Email: [email protected]
Abstract The epicardium has a critical role during embryonic development, contributing epicardium-derived lineages to the heart, as well as providing regulatory and trophic signals necessary for myocardial development. Crim1 is a unique trans-membrane protein expressed by epicardial and epicardially- derived cells but its role in cardiogenesis is unknown. Using knockout mouse models, we observe that loss of Crim1 leads to congenital heart defects including epicardial defects and hypoplastic ventricular compact myocardium. Epicardium-restricted deletion of Crim1 results in increased epithelial-to-mesenchymal transition and invasion of the myocardium in vivo, and an increased migration of primary epicardial cells. Furthermore, Crim1 appears to be necessary for the proliferation of epicardium-derived cells (EPDCs) and for their subsequent differentiation into 164 cardiac fibroblasts. It is also required for normal levels of cardiomyocyte proliferation and apoptosis, consistent with a role in regulating epicardium-derived trophic factors that act on the myocardium. Mechanistically, Crim1 may also modulate key developmentally expressed growth factors such as TGFβs, as changes in the downstream effectors phospho-SMAD2 and phospho- ERK1/2 are observed in the absence of Crim1. Collectively, our data demonstrates that Crim1 is essential for cell-autonomous and paracrine aspects of heart development.
165 Introduction During development, the heart is invested with cells of the proepicardium (PE), a structure that will give rise to the epicardium, myocardial fibroblasts, cells of the coronary vasculature and some cells of the valves 1-4. A critical step in the formation of these epicardium-derived lineages is the epithelial-to-mesenchymal transition (EMT) that epicardial cells must undergo to occupy the sub- epicardial space or to invade the myocardium. Some factors implicated in promoting epicardial EMT and invasion include FGFs, PDGF-A/B, TGFβ-1/2/3, and BMP-2 5-8, while these processes are restrained by NF1 9 and WT1 10. However, the molecular factors that control epicardial EMT and myocardial invasion, and the differentiation of epicardium-derived lineages, are incompletely understood.
Crim1 is a type-I transmembrane protein that is broadly expressed during development and which has been shown to regulate embryonic development of multiple organ systems 11-14. We have previously demonstrated embryonic lethality in mice homozygous for the Crim1KST264genetrap and Crim1∆flox null alleles 11,15, and that homozygous Crim1KST264/KST264mice display defects in organ systems including the kidney, limb, eye and placenta. These mice die perinatally on a C57BL6 background 11. The Crim1∆flox/∆flox mice display similar defects but there exist variations in severity and penetrance of the phenotypes, likely due to differences in the type of mutation used to generate each mouse line 15. The Crim1∆flox homozygotes are also embryonic lethal on the C57BL6 background 15. The underlying role of this molecule in vivo is complex, and likely involves interactions with growth factors and the extracellular milieu. Crim1 is homologous to the BMP- regulating protein Chordin, and contains six von Willebrand factor, type C-like cysteine-rich repeats (CRRs) through which it binds a broad range growth factors including TGFβ, BMP, VEGF and PDGF, when Crim1 is co-expressed in the same cell as the growth factor 12,13. Moreover, Crim1 can variably have an agonistic or antagonistic effect on BMP signaling 13,16,17, although this activity may be context-dependent. In the heart, it is expressed in PE-derived lineages, including the epicardium, and coronary vasculature 11, however, its role in cardiogenesis, along with the growth factors it interacts with at different stages, is unclear.
In this study, we demonstrate essential autocrine and paracrine roles for Crim1 in heart development. Crim1 loss-of-function and genetrap mice display congenital heart defects such as ventricular septal defects, epicardial defects and hypoplastic ventricular myocardium. We reveal an involvement of Crim1 in regulating epicardial cell migration into the myocardium both in vivo and in vitro. Crim1 also limits EMT in a manner independent of canonical TGFβ signaling, but possibly via an interaction with β-catenin, a crucial component of adherens junctions, which has been 166 previously shown to complex indirectly with Crim1 18. Interestingly, the cells that have lost Crim1 and have taken up residence in the myocardium, proliferate less, and give rise to a reduced number of cardiac fibroblasts, suggesting a further cell-autonomous role for Crim1 in fate specification of cells derived from the epicardium. We also provide evidence that upon loss of Crim1 function, compact myocardial development is aberrant, likely due to epicardial loss of Crim1, indicative of a paracrine role. Collectively, these findings illustrate how Crim1 and other factors function in tandem to regulate heart development.
Results Expression of Crim1 during heart development using a genetrap reporter To investigate Crim1 expression in the developing heart, we used a genetrap line for Crim1, Crim1KST26411 that allowed the use of a LacZ reporter under the control of the Crim1 locus. We observed Crim1 expression in the proepicardium at 9.5 days post coitum (dpc), and in the epicardium at all stages during cardiac development (Fig. 1A-M). Crim1 was also expressed in coronary vascular smooth muscle and coronary endothelial cells, as reported previously 11. Crim1 expression in the endocardium and valve leaflets was confirmed by performing X-Gal staining coupled with immunohistochemical analyses against periostin, a mesenchymal cell marker (Supplementary Fig. 1A-E), and Crim1 expression in coronary vascular smooth muscle cells was confirmed using transgelin double labelling (Supplementary Fig. 1F). The dynamic expression pattern observed suggests that Crim1 plays a role in heart development.
Developmental defects in hearts of Crim1 mutant embryos. The expression pattern of Crim1 led us to investigate whether there were heart defects exhibited by Crim1 mutant mice. Examination of hearts from embryos homozygous for the Crim1KST264 genetrap mutant (Supplementary Fig. 2A-L) revealed epicardial malformations and blebbing (Supplementary Fig. 2J-L), as well as a reduced ventricular size (Supplementary Fig. 2A-C) and muscular ventricular septal defects (Supplementary Fig. 2D-I). Similarly, Crim1∆flox/∆flox embryonic hearts had reduced ventricle size at 14.5 dpc (Fig. 2A-B, D-E), and distended atria and reduced ventricular size at 17.0 dpc (Fig. 2C, F). Histological analysis of Crim1∆flox/∆flox hearts at 13.5 dpc revealed abnormal epicardial morphology, including epicardial blistering (Fig. 2G-I, N), findings supported by scanning electron microscopic analysis of Crim1∆flox/∆flox hearts, which revealed a loss of the regular, cobblestone appearance of the epicardium (Fig. 2J-M).
These phenotypes may have arisen from deficits in early PE formation. To address this, we used in situ hybridization on Crim1∆flox/∆flox embryos for the PE markers Gata4 and Tbx18. We found that 167 PE formation and morphology, and PE marker expression, was similar between Crim1∆flox/∆flox and control embryos (Supplementary Fig. 3), indicating that Crim1 is not necessary for initial outgrowth of the PE. From this we infer that Crim1 may regulate adhesion or migration at a later time point within the developing epicardium.
Crim1 is necessary for normal compact myocardial development. As we had observed a reduction in ventricular size at 14.5 dpc in Crim1∆flox/∆flox hearts, we investigated whether there were any changes in thickness of the compact myocardium at 13.0 dpc (when the myocardium has started to develop) that could underlie this phenotype. We observed a reduced compact myocardial thickness in Crim1∆flox/∆flox hearts at 13.0 dpc, a finding that was not due to changes in cell density (Fig. 3A-D). Interestingly, this phenotype is shared by other mutants that lack epicardium or epicardium-derived factors 19. To further examine the nature of this defect, we analyzed the proliferation marker phospho-histone H3 (pHH3; Fig. 3E-F) and the apoptotic marker cleaved caspase 3 (CC3; Fig. 3I-J). We found an almost two-fold increase in proliferation, and a four-fold increase in apoptosis in the compact myocardium of Crim1∆flox/∆flox hearts (Fig. 3H, L). No change was observed in proliferation or apoptosis in epicardial or sub-epicardial cells of Crim1∆flox/∆flox hearts at this age (Fig. 3G-, K). This suggests that the loss of Crim1 culminates in elevated cell death within the compact myocardium during development, but whether it is an autocrine effect of Crim1 expressed within the compact myocardium, or epicardially-expressed Crim1 acting in a paracrine manner, is unknown.
To address this question, we analyzed the potential autocrine role of Crim1 in compact myocardial development by deleting Crim1 from the cardiomyocytes in the myocardium of the developing heart using the Mlc2v-Cre line. This line is efficient and restricted to ventricular cardiomyocytes 20- 23. Cardiomyocyte proliferation and survival upon myocardial loss of Crim1 in the ventricular myocardium were examined using pHH3 and CC3 immunocytochemistry, respectively. Quantification of cells expressing both X-Gal and pHH3, or cells expressing X-Gal and CC3 in the compact myocardium at 13.5 dpc revealed no significant differences in the number of either pHH3- and CC3-positive cells between control and mutant hearts (Supplementary Fig. 4). Moreover, no significant change in compact myocardium thickness was observed in these mice (Supplementary Fig. 4). From this, and the fact that the compact myocardium phenotype is observed before the epicardial cells have undergone EMT and invaded the compact myocardium, we infer that the myocardial deletion of Crim1 does not give rise to deficits in the compact myocardium, but rather that epicardial Crim1 regulates the development of the myocardium in a paracrine fashion. In order to substantiate this, we crossed the conditional Crim1FLOX line to a strain that would enable 168 inducible ablation of Crim1 from the epicardium (WT1-CreERT2). The WT1-CreERT2 line has previously been shown to be specific to the epicardium and epicardium-derived lineages 24. We further corroborated this by immunohistochemical analysis using the cardiomyocyte marker MF20, which revealed limited overlap between the X-Gal staining and MF20 immunoreactivity (Supplementary Fig. 5D-E). Two doses of Tamoxifen were administered at 9.5 dpc and 10.5 dpc prior to harvesting the embryos, and epicardial-restricted Cre activity as assessed by the R26R LacZ reporter was confirmed in embryonic hearts from 12.5 dpc (Fig. 4F-G, J-K, and data not shown). Unlike the hearts of Crim1∆flox/∆flox mice,the phenotype of reduced myocardial wall thickness was not recapitulated in the WT1-CreERT2 hearts at 15.5 dpc (data not shown). This could be due to incomplete penetrance and variable expressivity; a common occurrence with Cre lines, including the WT1-CreERT2 line 24. Indeed, upon quantification of the percentage of epicardial cells that had lost Crim1 (pooled genotypes, n=6; data not shown), we saw that in the left and right ventricles, there were still 62.5% of epicardial cells in which Crim1 was not ablated, which could compensate for any myocardial defects that may have arisen with a complete lack of Crim1.
Cell-autonomous requirement for Crim1 to control epicardial migration and myocardial invasion in vitro and in vivo. Given the potential role for epicardially-derived Crim1 in regulating myocardial development, as well as the epicardial defects observed in the hearts of conditional null and genetrap mutant embryos, we next examined the role of Crim1 in epicardial development. We first sought to determine whether Crim1 mutant epicardial cells exhibited migration defects. To do this we cultured primary embryonic epicardial cells from the Crim1KST264 genetrap line in a transwell migration assay (modified Boyden chamber assay). The Crim1KST264 allele in these experiments allowed us to monitor the continued expression of Crim1 by X-Gal staining in representative examples of explant cultures. In all cases, ubiquitous Crim1-LacZ expression was observed in primary epicardial cells (data not shown). We found that primary epicardial cultures from Crim1KST264/KST264 hearts, immunostained with WT1 as an epicardial marker to confirm epicardial identity after enrichment (Fig. 4D), exhibited an increased rate of migration relative to heterozygote controls (Fig. 4A–E), suggesting that the loss of Crim1 from the developing epicardium resulted in increased epicardial migration.
To confirm these observations in vivo, and to also investigate changes in EMT, we used the WT1- CreERT2 line and assessed the location of X-Gal-positive cells at 15.5 dpc in Crim1FLOX/FLOX;WT1- CreERT2; R26R and littermate heterozygous controls, determining whether labelled cells remained epicardial, or had undergone EMT and migrated to the sub-epicardium or to an intramural 169 myocardial location. We found that loss of Crim1 in epicardial cells resulted in an increase in the number of epicardial cells that had undergone EMT and invaded the myocardium of the right ventricle relative to controls (Fig. 4F-M). This finding was confirmed using a second Cre driver, namely WT1-Cre25 (data not shown). These in vivo data indicate that the loss of Crim1 in the epicardium results in increased epicardial EMT, and are consistent with the primary epicardial culture migration experiments indicating that the loss of Crim1 function results in increased epicardial cell migration.
To gain insight into the mechanism by which Crim1 may regulate EMT and the migration of epicardial cells, we next investigated whether growth factor signalling in both the epicardium and the myocardium is perturbed in Crim1-deficient mice. Numerous growth factors have been implicated in epicardial EMT, including TGFβ and BMPs (acting via TGFβ receptors 5-8). As Crim1 has been shown to bind such growth factors in cell culture binding experiments 12,13,26, and potentially regulate their activity in vivo12,26, we examined whether there were changes in canonical TGFβ or BMP signalling at the time of epicardial EMT and invasion, using the Crim1Δflox line. To do this we performed indirect immunofluorescence for phospho-SMAD2 and phospho-SMAD1/5 as proxies to assess canonical TGFβ and BMP signalling, respectively. We also investigated whether there were changes in downstream effectors such as ERK1/2 and AKT as further read-outs of aberrant signal transduction. There was a reduced level of phospho-SMAD2 in the epicardium of Crim1∆flox/∆flox hearts at 13.0 dpc (77% and 69% of control values for the left and right ventricular epicardium, respectively) (Fig. 5A-B, E-F, C-D). Despite the increased epicardial EMT and invasion suggested by the epicardial-restricted loss of Crim1 function, these results suggest that canonical TGFβ signalling was reduced in the epicardium of Crim1∆flox/∆flox hearts. There was no change in levels of phospho-SMAD1/5 (Supplementary Fig. 6A-H), phospho-AKT (Supplementary Fig. 6I-P) and phospho-ERK1/2 (Fig. 5I-J, M-N, K-L) in the epicardium of Crim1∆flox/∆flox hearts relative to wildtype controls. At 13.0 dpc, Crim1∆flox/∆flox hearts showed comparable levels of ventricular compact myocardial phospho-SMAD2 (Fig. 5A-B, E-F, G-H), phospho-SMAD1/5 (Supplementary Fig. 6A-H) and phospho-AKT (Supplementary Fig. 6I-P) relative to controls. However, we observed that there was an increase in both the number of phospho-ERK1/2 positive cells (for example, 29% of cells were phospho-Erk1/2-positive in the left ventricle of the mutant, compared to 6% in the left ventricle of the control; P < 0.05, t-test), as well as an increase in the phospho-ERK1/2 signal intensity per cell in the ventricular compact myocardium of mutant hearts relative to controls (Fig. 5I-J, M-N, O-P). There was also no significant difference in the distance that pERK1/2 positive cells were from the epicardium in both control and mutant hearts (data not shown). Together with the lack of compact myocardial changes observed in the myocardium- 170 restricted Crim1knock-out, this indicates that Crim1 could play a paracrine role in regulating myocardial development, likely through the regulation of epicardium-derived factors.
The adhesion molecule β-catenin is an important regulator of epithelial stability and EMT, and Crim1 has been shown to interact indirectly with β-catenin 18. To investigate the role of Crim1 in altering EMT and migration, we analyzed the localisation of β-catenin 27 in Crim1+/+ and Crim1∆flox/∆flox hearts at 13.5 dpc. We found that there was a decrease in the percentage of epicardial cells that displayed an accumulation of β-catenin at epicardial cell-cell junctions in mutant hearts (Fig. 5Q-W). When considered in light of the findings that increased EMT and myocardial invasion occurs in the hearts of Crim1 mutant mice, this finding suggests that Crim1 normally regulates the stability of intercellular junctions within the epicardium. Analysis of filamentous actin to assess the cytoskeletal remodelling that accompanies EMT and migration revealed no visible changes in stress fibre morphology between mutant and wildtype hearts at 13.5 dpc (Supplementary Fig. 7). The data reveal a cell-autonomous role for Crim1 in controlling epicardial migration, EMT and invasion in primary cell culture and in vivo, which appears to rely on interactions with β-catenin.
Cell-autonomous requirement for Crim1 to control proliferation and fate specification of EPDCs. The increase in the number of EPDCs that had migrated into the myocardium prompted us to further investigate the proliferative ability and differentiation potential of these EPDCs. We used the WT1-CreERT2 and Crim1FLOX; R26R lines and assessed EPDC proliferation at 15.5 dpc using pHH3 (Fig. 6A-B). Quantification of X-Gal expressing cells that were pHH3-positive in the myocardium of left and right ventricles revealed a significant reduction in proliferation of intramyocardial cells in Crim1FLOX/FLOX; +/WT1-CreERT2;R26R hearts (Fig. 6D). Quantification of pHH3-positive cells among X-Gal-positive epicardial cells, however, revealed comparable results (Fig. 6C). This demonstrates that deleting Crim1 from the epicardium can perturb proliferation of epicardium-derived cells, even though there is increased migration into the myocardium.
To address whether changes in epicardial cell fate occurred due to loss of Crim1, we performed qPCR analyses on ventricular tissue from 17.5 dpc Crim1∆flox/∆flox and Crim1+/+ hearts using vascular smooth muscle, fibroblast and cardiomyocyte markers. No significant changes in these markers were observed, although there was a trend for a reduction in the fibroblast marker Collagen1a (P=0.0556; Supplementary Fig. 8A). To specifically examine the effect of epicardial Crim1 deletion on the differentiation of EPDCs into cardiac fibroblasts, we used the WT1-CreERT2 line to perform IHC using periostin, another fibroblast marker, on heterozygote control and mutant 171 WT1-CreERT2 heart sections at 17.5 dpc. Quantification of the proportion of periostin- and X-Gal- positive cells in the myocardium revealed a significant decrease in the expression of periostin in the left and right ventricles of the mutant (Fig. 6E-I). This suggests that the number of cardiac fibroblasts were reduced, and the secretory phenotype of cardiac fibroblasts was perturbed upon epicardial Crim1 loss-of-function, indicating a possible cell-autonomous requirement for Crim1 to regulate fate specification of EPDCs into fibroblasts.
Discussion The epicardium gives rise to crucial components of the developing heart, and while the differentiation of EPDCs into cardiomyocytes and coronary endothelial cells remains a controversial topic, reactivated epicardial cells after myocardial damage have been shown to give rise to fibroblast and smooth muscle cells28-32. The epicardium also has a critical role in providing instructive cues and trophic factors for the developing myocardium. Although certain epicardium- derived factors are believed to act on the myocardium in a paracrine manner, the nature of this interaction is not entirely understood. Here, we demonstrate an essential role for Crim1 in the epicardium during heart development. Crim1 appears to regulate epicardial EMT and invasion, fate specification of EPDCs into cardiac fibroblasts, and modulation of growth factor activity. As a result, we infer that Crim1 controls compact myocardial development via regulation of epicardium- derived paracrine factors.
Crim1 is expressed in a large number of tissues in a spatially and temporally regulated manner and has been shown to have diverse extracellular and intracellular functions 11-14,18,33,34. This study focused determining the role of Crim1 in the cardiogenesis by using different transgenic mouse lines. We have observed that Crim1 expression in the heart begins in the proepicardium, although it is not necessary for PE specification, and continues in the epicardium and epicardium-derived lineages. Crim1KST264/KST264homozygotes showed a disrupted epicardial layer, reduced ventricular size, and prominent ventricular septal defects. Hearts of both Crim1∆flox/∆flox and Crim1KST264/KST264homozygotes at 13.5 dpc showed irregular epicardial morphology indicative of EMT and adhesion defects. Indeed, epicardial deletion of Crim1 resulted in an increase in epicardial cell EMT and myocardial invasion. Similarly, primary epicardial cultures showed increased migration upon loss of Crim1. However, epicardial cells lacking Crim1 also showed reduced proliferation and an overall reduction in cardiac fibroblast marker expression. Overall this suggests that loss of Crim1 in epicardial cells affects cell phenotype, proliferation and fate.
172 Growth factors like BMPs and TGFβs that can be bound by Crim1 promote epicardial EMT and invasion 5,7,8. Canonical BMP signalling via the SMAD1/5 pathway, AKT signalling and ERK1/2 in the epicardium appear to be largely independent of Crim1 in Crim1∆flox/∆flox hearts. Indeed, we observed a paradoxical reduction in epicardial TGFβ signalling in Crim1∆flox/∆flox embryos. This suggests that, in the epicardium, Crim1 may play a role other than growth factor regulation with Crim1 mutants representing an uncoupling of these growth factor signalling pathways and epicardial EMT. β-catenin distribution, however, at epicardial cell-cell junctions was altered in the absence of Crim1. Interestingly, XenopusCrim1 has been shown to interact via its cytoplasmic domain with β-catenin and N-cadherin, identifying a role for Crim1 in the formation or stabilization of cadherin-dependent junctional complexes in epithelial cells 18. Though Crim1 was not always present at sites of cadherin expression, it may interact with cadherins and β-catenin within the endoplasmic reticulum, where complex formation between the 3 interacting partners would occur, possibly displacing an existing stabilization mechanism 18. Crim1-mediated sequestration of β- catenin from junctional complexes could also limit Wnt signalling. Hence, the link between Crim1 and the β-catenin pathway, particularly in the developmental interactions between the epicardium and myocardium, remains to be explored.
The altered myocardial thickness in the absence of Crim1 suggests a role for Crim1 in the regulation of signalling molecules from the epicardium, and possibly the myocardium as well, crucial for normal myocardial development. The underlying mechanism that controls the developing myocardium is not clear. Mouse mutants, including knockouts of VCAM-1, α4 integrin and WT-1 and mouse lines with altered retinoic acid signalling, show perturbation of epicardial formation with secondary myocardial defects during development35-38. Furthermore, microsurgical inhibition of epicardium formation in avian embryos results in hypoplastic ventricular compact myocardium39,40. Similarly, we observed a reduced compact myocardial thickness in Crim1∆flox/∆flox hearts at 13.0 dpc. However, epicardium-specific deletion of Crim1 did not recapitulate this phenotype, possibly due to incomplete penetrance within the WT1-CreERT2line during embryonic stages. Similarly, the myocardial depletion of Crim1 did not culminate in a thinner myocardium, nor did it reveal an effect on cardiomyocyte proliferation or apoptosis. In light of the fact that reciprocal signalling between the epicardium and myocardium is required for proper myocardial development to ensue, it is plausible that both epicardial and myocardial Crim1 are required for normal formation of the myocardium. An alternative interpretation of these results is that there is a non-cell-autonomous requirement for epicardial Crim1 in myocardial development. Embryonic cardiomyocytes respond to a variety of signalling molecules and express a broad range of receptors which can bind mitogenic factors 19. Indeed, IGFs, PDGFs and VEGFs secreted by the epicardium 173 may act as epicardial mitogens that stimulate ventricular development and cardiomyocyte proliferation 41,42. IGFBPs are known to regulate the activity of IGFs 43. The IGFBP domain of Crim1 has been shown to possess sequence similarity to IGFBP-7 14 potentially allowing it to bind IGFs and insulin to mediate the regulation of their downstream signalling. This is in accordance with our observation that ERK1/2 signalling in the myocardium of Crim1∆flox/∆flox hearts is increased. Crim1 may serve to sequester growth factors secreted by the epicardium, without necessarily affecting the range of growth factor action in Crim1∆flox/∆flox hearts, where these may be free to signal to the adjacent myocardium. This has been reported previously with respect to a role for Crim1 in VEGF sequestration in the glomerulus 12. Crim1 and the growth factors it binds have been reported to be co-expressed in the same cell 12,13, which would be consistent with a role in tethering epicardially-derived growth factors. Prolonged ERK1/2 phosphorylation has been shown to lead to cell death, whereas transient activation drives proliferation44. Thus, an increase in phospho-ERK1/2 activity in the compact myocardium could have a pro- or anti-apoptotic effect, depending on the activation of downstream effectors. Crim1 has also been shown to reduce the production and secretion of BMPs13. A lack of Crim1 could lead to increased growth factor production and signalling, and result in a high cell turnover in the myocardium of mutant hearts as a result of co-activation of proliferation and apoptosis. It is important to note that the antagonistic or agonistic functions of Crim1 are context-dependent and could change for different cell types within the heart and at different stages of development.
We also observed a decrease in the proliferative ability of EPDCs in the myocardium of Crim1FLOX/FLOX; WT1-CreERT2; R26R hearts, in spite of an increased migration of cells derived from the epicardium as seen in the Crim1FLOX/FLOX; WT1-CreERT2; R26R and Crim1FLOX/FLOX; WT1-Cre; R26R hearts. This was observed alongside a reduction in the number of cardiac fibroblasts, one of the major derivatives of the epicardium 1,3,45. Cardiac fibroblasts represent a crucial component of the myocardium, secreting a large number of ECM molecules required to support other resident cells of the heart, as well as growth factors and cytokines for the proper functioning of these cells 46. However, there is little known about the factors that contribute to cardiac fibroblast fate specification. These EPDCs that have lost Crim1 could remain undifferentiated cells residing in the compact myocardium, or preferentially „switch‟ to coronary vascular smooth muscle cells rather than cardiac fibroblasts. Cardiac fibroblasts generated after injury are crucial to cardiac repair in terms of both the ECM molecules secreted as well as overall cardiac performance 29,47. Our data indicate a cell-autonomous requirement for Crim1 in the production of cardiac fibroblasts. Further studies are required to validate the role of Crim1 in
174 lineage specification into fibroblasts and to also address whether it plays a cell-autonomous role in the differentiation of EPDCs into coronary vascular smooth muscle cells.
In summary, this study reveals a role for epicardial Crim1 in normal heart development. Although thought to be a largely quiescent lining surrounding the myocardium, there has been significant recent focus on ways to reactivate the adult epicardium in response to injury and disease 48. This revolves around re-expression of embryonic epicardial genes and the release of paracrine and autocrine factors from the epicardium or EPDCs 48. An active participant in epicardially-regulated heart development, Crim1 modulation may therefore provide a potential avenue for future molecular and cellular therapeutic interventions in cardiac regeneration and repair. Moreover, whether or not abnormal CRIM1 expression plays a role in congenital human heart disease will be an important topic for future investigations.
Methods Ethics statement The work performed in this study conformed to The University of Queensland‟s Animal Welfare Unit guidelines for animal use in research, and followed the National Institutes of Health Guide for the Care and Use of Animals and the Australian Code of Practice for the Care and Use of Animals for Scientific Purposes. All experimental protocols were approved by The University of Queensland‟s Animal Ethics Unit (Animal Ethics Committee approval numbers: AIBN/274/08/NHMRC, SMBS/420/11/NHMRC).
Animal breeding The mouse lines used in this study were the Crim1FLOX conditional mutant 15, Crim1∆floxknock-out 15 and Crim1KST264 genetrap line 11, ROSA26 Cre-reporter line (R26R) 49, WT1-Cre line 25, WT1- CreERT2 line 24,50 and Mlc2v-Cre line 20. The Crim1KST264 mouse line was created as part of a genetrap screen 51, with insertion of the β-Geo cassette into intron 1 of Crim1 resulting in the fusion of exon 1 and the β-Geo cassette within the Crim1 transcript, thereby creating a hypomorphic allele which expresses a minor, alternately spliced isoform of CRIM111. The Crim1FLOX conditional mutant mouse line was generated, using Cre/LoxP recombination, by flanking exons 3 and 4 with unidirectional LoxP sites 15. These Crim1FLOX mice were crossed with a line that expressed Cre recombinase ubiquitously (CMV-Cre) to produce the Crim1∆flox line. This deletion created an out- of-frame transcript and a predicted non-functional protein and as such, Crim1 is deleted from all tissues in these mice 15. All lines were maintained on a C57Bl6 genetic background. Embryos were obtained from timed matings between Crim1+/∆flox intercrosses, Crim1+/KST264 intercrosses, or 175 Crim1FLOX/FLOX; R26R females mated with Crim1+/FLOX; WT1-Cre, Crim1+/FLOX; +/WT1-CreERT2 or Crim1+/FLOX; Mlc2v-Cre males.
Sample preparation Mouse embryos were obtained from timed matings between Crim1+/∆flox intercrosses or Crim1+/KST264 intercrosses, with the presence of a vaginal plug regarded as 0.5 days post coitum (dpc). Embryonic hearts were dissected in PBS and fixed for 2 hours in 4% PFA/PBS at 4°C, washed in PBS, and photographed in whole-mount. Samples were then dehydrated and processed for paraffin infusion and embedding, and 7 µm sections were cut on a microtome. Hematoxylin and eosin staining was performed using standard protocols. Embryonic hearts collected from matings between Crim1+/∆flox intercrosses collected at E13.0 and E13.5 were also fixed in 4% PFA/PBS at 4°C, washed in PBS,cryo-protected in 30% sucrose/PBS at 4°C, embedded in OCT (Tissue-Tek) and 10 µm sections were cut on a cryostat and air-dried before proceeding with the immunofluorescence procedure.
Timed matings were set up between Crim1+/FLOX; WT1-Cre males and Crim1FLOX/FLOX; R26R females to produce Crim1+/FLOX; WT1-Cre; R26R and Crim1FLOX/FLOX; WT1-Cre; R26R embryos collected at 15.5 dpc (at this stage, the myocardium has expanded and cells derived from the epicardium have started to differentiate into various lineages), Crim1+/FLOX; +/WT1-CreERT2 males and Crim1FLOX/FLOX; R26R females, for the production of Crim1+/FLOX; +/WT1-CreERT2; R26R and CrimFLOX/FLOX; +/WT1-CreERT2;R26R embryos collected at 15.5 dpc and 17.5 dpc (at 17.5 dpc, cells derived from the epicardium have differentiated into cardiac fibroblasts, coronary vascular smooth muscle cells, and a small proportion of coronary vascular endothelial cells), and between Crim1+/FLOX; Mlc2v-Cre males and Crim1FLOX/FLOX; R26R females to produce Crim1+/FLOX; Mlc2v- Cre; R26R and Crim1FLOX/FLOX; Mlc2v-Cre; R26R embryos collected at 13.5 dpc (the myocardium has begun to expand at this stage through proliferation of cardiomyocytes).
Embryonic samples from the WT1-Cre, WT1-CreERT2 and Mlc2v-Cre (β-galactosidase activity being a readout of Cre activity based on the R26R reporter) and Crim1+/KST264 intercrosses (β- galactosidase activity being under the control of the endogenous Crim1 promoter) were X-Gal- stained in whole-mount as previously described 6 before processing for paraffin embedding and sectioning as described above. Some X-Gal-stained sections were counterstained with nuclear fast red (NFR; Vector Laboratories). Others were subjected to cell biological and immunohistochemical analyses as detailed below.
176 Tamoxifen administration Tamoxifen (MP Biomedicals, 02156738) was dissolved in corn oil at a concentration of 20 mg/ml. To induce WT1-CreERT224,50, two doses of 2 mg tamoxifen 52 were injected intraperitoneally to pregnant dams (at 9.5 dpc and 10.5 dpc), prior to harvesting the mouse embryos.
Scanning electron microscopy 13.5 dpc embryos were collected from intercrosses of Crim1+/∆flox mice, and the hearts were dissected and fixed in 2.5% glutaraldehyde in PBS for 1 hour at room temperature. Samples were then washed 3 times with PBS and stored at 4°C until processing for scanning electron microscopy as previously described 53. Scanning electron microscopy was performed using a CM-500 Benchtop Scanning Electron Neoscope.
Antibodies Primary antibodies/lectin used in the study were: rabbit polyclonal anti-phospho-Histone H3 (pSer10; Millipore #06-570); rabbit monoclonal anti-cleaved caspase 3 (activated caspase 3; Cell Signalling Technology #9664); rabbit monoclonal anti-phospho-AKT (pSer473; D9E; Cell Signalling Technology #4060); rabbit monoclonal anti-phospho-ERK1/2 (pThr202/pTyr204; D13.14.4E; Cell Signalling Technology # 4370); rabbit monoclonal anti-phospho-SMAD1/5 (pSer463/pSer465; Invitrogen #700047); rabbit polyclonal anti-phospho-SMAD2 (pSer465/pSer467; Millipore AB3849); mouse monoclonal anti-smooth muscle α-actin (clone 1A4, Santa Cruz Biotechnology, sc-3225); rabbit anti-β-catenin (whole antiserum; Sigma Aldrich #C2206); rabbit polyclonal SM22 alpha (Abcam Ab14106); rabbit polyclonal anti-periostin (Abcam Ab92460); mouse monoclonal anti-Wilms‟ Tumor 1 (WT1; Dako #M3561); MF20 (supernatant, Developmental Studies Hybridoma Bank) and Isolectin B4 (Invitrogen). Secondary antibodies for immunofluorescence experiments were Alexa Fluor-488, -594 or -633-conjugated secondary antibodies (Invitrogen). TSA plus Cyanine 3 system was used for signal amplification in the case of pERK1/2 and pAKT (PerkinElmer #NEL744001KT) Secondary antibodies for immunohistochemistry experiments were anti-rabbit HRP (Promega; #W4011), anti-mouse HRP (Promega; #W4021), anti-rabbit biotin (Sigma; # B7389). The Vectastain Elite ABC kit (Standard) was also used (Vector Laboratories; #PK6100).
Immunohistochemistry Experiments using mouse antibodies were blocked with mouse-on-mouse (Vector MOM kit; BMK 2202) blocking reagent for one hour. Sections were primed with MOM diluent for five minutes after washes in 1xPBS twice. Mouse primary and secondary antibodies were diluted in MOM protein 177 diluent. Immunohistochemistry using pHH3 was performed and sections counterstained with 1% DAB followed by NFR staining. Immunohistochemistry to detect periostin used anti-rabbit biotin and Vectastain ABC as secondary and tertiary antibodies. Vectastain ABC reagents were prepared as per manufacturer‟s protocols. After one-hour incubation with anti-rabbit biotin, Vectastain ABC was added for 30 minutes in room temperature, followed by hematoxylin staining. Indirect immunofluorescence on tissue sections was performed on paraformaldehyde-fixed, paraffin- embedded samples as described 6. For immunofluorescence involving detection of cleaved caspase 3 or PHH3, paraffin sections were subject to heat-induced antigen retrieval in a citrate buffer (pH 6.0). Immunofluorescence on cryosectioned samples was performed to detect β-Catenin, MF20, phospho-SMAD1/5, phospho-SMAD2, phospho-AKT and phospho-ERK1/2. Tyramide amplification according to manufacturer‟s methods was used for signal amplification of phospho- ERK1/2 and phospho-AKT.
Primary epicardial explant cultures and migration assays 11.5 dpc embryos were collected from intercrosses of Crim1+/KST264 mice, and the ventricular component was dissected from the rest of the heart in sterile PBS and cultured in DMEM supplemented with 10% FCS and penicillin/streptomycin in 4-well tissue culture plates. Once the epicardium had grown out from the primary culture, the ventricular mass was removed. The epicardial cells were grown to confluence and were passaged once and plated at a high confluence. After two days of further culture, the primary epicardial cells were used in a modified Boyden chamber assay. Epicardial cultures were collected after trypsin treatment to completely dissociate cells, washed and resuspended in DMEM (with 10% FCS). 1,000 cells in identical volumes were placed in the upper chamber of 6.5 mm transwell culture inserts (polycarbonate filter, 5.0 µm pore size, Corning). After 16 hours of culture, the cells in the upper part of the transwell inserts were removed using a cotton swab, and the cells that had migrated through the polycarbonate filter were quantified after fixation in 4% PFA in PBS and staining with DAPI. Epicardial identity was confirmed after enrichment by immunofluorescence for the transcription factor WT-1 24,54, using standard techniques as described 6.
In situ hybridization Section in situ hybridization was performed as previously described 55. Whole-mount in situ hybridization was performed as previously described 56 with minor changes. Hybridization was performed with a riboprobe concentration of 0.4 µg/mL in pre-hybridization solution. Color detection was performed with the chromogenic substrate NBT/BCIP (Roche). Samples were then washed, post-fixed in 4% PFA/PBS, and photographed. At least three embryos of each 178 genotype/gene probe were analyzed. Some whole-mount samples were processed for paraffin infusion and sectioned as described above.
Quantitative real-time PCR analysis Ventricular samples from Crim1+/+ and Crim1Δflox/Δfloxhearts were microdissected, homogenized and total RNA was extracted using an RNeasy Mini Kit (Qiagen, #74104). Reverse transcription was performed using SuperscriptIII (Invitrogen) and qPCR was performed. Briefly, 500ng total RNA was reverse-transcribed with random primers and dNTPs. cDNA was diluted 1/5 with RNase/DNase-free water. qPCRs were carried out in a QuantStudio6 (Applied Biosystems) using SYBR green (Takara) and standard qPCR conditions. The data were analyzed with the QuantStudio6 software, with TFIID used as a relative standard. All samples were tested in triplicate. Relative transcript levels were assessed using the ΔCt method. Statistical analyses were performed using a two-tailed unpaired t-test. The primer sequences used in this study were purchased from Sigma-Aldrich, and are detailed below:
Periostin: Forward 5‟-AAGCTGCGGCAAGACAAG-3‟ Periostin: Reverse 5‟-TCAAATCTGCAGCTTCAAGG-3‟ Collagen1a: Forward 5‟-AGACATGTTCAGCTTTGTGGAC-3‟ Collagen1a: Reverse 5‟-GCAGCTGACTTCAGGGATG-3‟ Collagen3a: Forward 5‟-ACGTAGATGAATTGGGATGCAG-3‟ Collagen3a: Reverse 5‟-GGGTTGGGGCAGTCTAGTG-3‟ Collagen5a: Forward 5‟-CTACATCCGTGCCCTGGT-3‟ Collagen5a: Reverse 5‟-CCAGCACCGTCTTCTGGTAG-3‟ Transgelin: Forward 5‟-CCTTCCAGTCCACAAACGAC-3‟ Transgelin: Reverse 5‟-CTAGGATGGACCCTTGTTGG-3‟ Alpha smooth muscle actin: Forward 5‟-ACTCTCTTCCAGCCATCTTTCA-3‟ Alpha smooth muscle actin: Reverse 5‟-ATAGGTGGTTTCGTGGATGC-3‟ Myosin heavy chain 11: Forward 5‟-ACGCCCTCAAGAGCAAACT-3‟ Myosin heavy chain 11: Reverse 5‟-CCCTTCTGGAAGGAACAATG-3‟ Calponin1: Forward 5‟-GAAGGTCAATGAGTCAACTCAGAA-3‟ Calponin1: Reverse 5‟-CCATACTTGGTAATGGCTTTGA-3‟ Myosin heavy chain 7: Forward 5‟-TGCTGAGGCCCAGAAACAAGTG-3‟ Myosin heavy chain 7: Reverse 5‟-CTGGATTTGAGTGTCCTTCAGCAG-3‟ Nkx2-5: Forward 5‟-ATTGACGTAGCCTGGTGTCTCG-3‟ Nkx2-5: Reverse 5‟-AGTGTGGAATCCGTCGAAAGTGC-3‟ 179 Tie2: Forward 5‟-AGAGGCCGAACATTCCAAGTAGC-3‟ Tie2: Reverse 5‟-GGCCAAACAAACTGGCTTCTGC-3‟ PDGFRα: Forward 5‟-TTGCACAGCCTCTGTTTGTTGC-3‟ PDGFRα: Reverse 5‟-ATGGTGTCATGCAAGGCCCAAAG-3‟ PDGFRβ: Forward 5‟-CAGTCCCGGCTACCCTATCT-3‟ PDGFRβ: Reverse 5‟-GACCCAGAGTTTTCTCGGCA-3‟ Tcf21: Forward 5‟-CATTCACCCAGTCAACCTGA-3‟ Tcf21: Reverse 5‟-CCACTTCCTTCAGGTCATTCTC-3‟ TFIID: Forward 5‟-ACGGACAACTGCGTTGATTTT-3‟ TFIID: Reverse 5‟-ACTTAGCTGGGAAGCCCAAC-3‟
Imaging and data analysis Bright-field images of tissue sections were captured using an Olympus BX-51 BF/DF slide microscope with Canon digital cameras with DP Controller software (Olympus). Whole-mount images were captured using either an Olympus SZX-12 stereo-microscope with DP Controller software, or a Zeiss Stemi 2000 stereo-microscope with an AxioCam digital camera with AxioVision software (Zeiss). Fluorescent images of tissue sections were captured on an Olympus IX-81 laser scanning confocal microscope and an Olympus FV1000 upright laser scanning confocal microscope with Olympus Fluoview software (Ver3.1a). Images were adjusted for colour levels, brightness and contrast, and figures compiled, using Adobe Photoshop software. The number of samples included in data averages is indicated in the figure legends where applicable. To determine the statistical significance of the prevalence of phenotypes among mutant and control groups, a Z- test was used (Fig. 2N). Epicardial defects examined and quantified include either a loss of squamous morphology or blebbing. For the analysis of 13.0 dpc compact myocardium thickness and cell density (Fig. 3), sections were immunolabelled for Acta2 and PECAM-1, and nuclei counter-stained with DAPI as described above. 40x fields of view of the ventricular compact myocardium were imaged on a laser scanning confocal microscope on each of three–five non- consecutive sections for each sample analyzed. The compact myocardium was considered as the actin-positive area of the ventricular chambers, bounded on the endocardial side by the base of the trabeculae (delineated by PECAM-1 staining). To determine cell density, the area was determined from confocal images using either Olympus Fluoview or ImageJ software, and the number of DAPI-stained nuclei quantified using ImageJ software. For the quantification of the pHH3-positive and CC3-positive cells at 13.0 dpc, immunolabelling was performed as described above. The ventricular compact myocardium was imaged on a laser scanning confocal microscope (multiple images at 40x objective) on five–seven non-consecutive sections for each sample analyzed. Cells 180 were scored as epicardial or sub-epicardial if they were on the periphery of the section and were actin-negative. Statistical significance between pooled data from mutant and control groups was determined using a two-tailed, unpaired Student‟s t-test.
For the analysis of epicardium-derived cells using the using the WT1-CreERT2 line, 2-4 non- consecutive, coronal sections from each sample (X-Gal-positive; WT1-CreERT2) (Fig. 4I, M) were quantified and the average of X-Gal-positive cells in the mid-ventricular region per sample was measured. For analysis using the WT1-Cre line, 16 non-consecutive, coronal sections from each sample (X-Gal-positive; WT1-Cre) were quantified and the total number X-Gal-positive cells in the ventricles was counted (data not shown). Cells were scored as epicardial, sub-epicardial if they were on the periphery of the section, or intramyocardial. For the analysis of the location of epicardium- derived cells (X-Gal-positive, WT2-CreERT2/ WT1-Cre; epicardial, sub-epicardial, or intramyocardial), a two-way ANOVA with Bonferroni‟s post-test was used.
For quantification of β-Catenin (Fig. 5Q-U), sections were co-immunolabelled for MF20 and 40x images were acquired using a laser scanning confocal microscope from at least 2 sections for each sample. Cells were scored as epicardial or sub-epicardial if they were on the periphery of the section and were MF20-negative. β-Catenin accumulation at adherens junctions between epicardial cells27 was quantified, with the junctions being defined as points of contact between two epicardial cells. Statistical significance between pooled data from mutant and control groups was determined using a two-tailed, unpaired Student‟s t-test.
For quantification of cardiac fibroblasts and proliferating EPDCs (Fig. 6), 20X images were acquired using a slide microscope from a minimum of three non-consecutive sections for each sample. The number of X-Gal- and periostin-positive and X-Gal- and pHH3-positive cells were counted using ImageJ software and data analyzed using two-tailed, unpaired Student‟s t-test and two-way ANOVA respectively. Positive cells were identified as those that possessed both blue β- gal and a corresponding brown nucleus denoted by DAB staining.
For quantification of phospho-SMAD levels (Fig. 5A-H, Supplementary Fig. 6A-H), 40x images were acquired with a laser scanning confocal microscope from at least five sections for each sample and for phospho-ERK1/2 and phospho-AKT levels (Fig. 5I-P, Supplementary Fig. 6I-P), from at least 2 sections for each sample. To determine average signal intensity/cell, unsaturated confocal images were used, and signal intensity determined using the Integration function in Olympus Fluoview software. The number of DAPI-stained nuclei in the demarcated area was quantified using 181 ImageJ software. Average signal per cell determined was divided by 1000 for graphical representation. Statistical significance between pooled data from mutant and control groups was determined using a two-tailed, unpaired Student‟s t-test. Quantified data shown are mean ± standard deviation.
References
1 Gittenberger-de Groot, A. C., Vrancken Peeters, M. P., Mentink, M. M., Gourdie, R. G. & Poelmann, R. E. Epicardium-derived cells contribute a novel population to the myocardial wall and the atrioventricular cushions. Circ Res82, 1043-1052 (1998). 2 Mikawa, T. & Fischman, D. A. Retroviral analysis of cardiac morphogenesis: discontinuous formation of coronary vessels. Proc Natl Acad Sci U S A89, 9504-9508 (1992). 3 Dettman, R. W., Denetclaw, W., Jr., Ordahl, C. P. & Bristow, J. Common epicardial origin of coronary vascular smooth muscle, perivascular fibroblasts, and intermyocardial fibroblasts in the avian heart. Dev Biol193, 169–181 (1998). 4 Mikawa, T. & Gourdie, R. G. Pericardial mesoderm generates a population of coronary smooth muscle cells migrating into the heart along with ingrowth of the epicardial organ. Dev Biol174, 221-232 (1996). 5 Morabito, C. J., Dettman, R. W., Kattan, J., Collier, J. M. & Bristow, J. Positive and negative regulation of epicardial-mesenchymal transformation during avian heart development. Dev Biol234, 204-215 (2001). 6 Pennisi, D. J. & Mikawa, T. FGFR-1 is required by epicardium-derived cells for myocardial invasion and correct coronary vascular lineage differentiation. Dev Biol328, 148-159 (2009). 7 Smith, C. L., Baek, S. T., Sung, C. Y. & Tallquist, M. D. Epicardial-derived cell epithelial- to-mesenchymal transition and fate specification require PDGF receptor signaling. Circ Res108, e15-e26 (2011). 8 Sánchez, N. S. & Barnett, J. V. TGFbeta and BMP-2 regulate epicardial cell invasion via TGFbetaR3 activation of the Par6/Smurf1/RhoA pathway. Cell Signal24, 539-548 (2012). 9 Baek, S. T. & Tallquist, M. D. Nf1 limits epicardial derivative expansion by regulating epithelial to mesenchymal transition and proliferation. Development139, 2040-2049 (2012). 10 Martínez-Estrada, O. M. et al. Wt1 is required for cardiovascular progenitor cell formation through transcriptional control of Snail and E-cadherin. Nature Genetics42, 89-93 (2010). 11 Pennisi, D. J. et al. Crim1KST264/KST264 mice display a disruption of the Crim1 gene resulting in perinatal lethality with defects in multiple organ systems. Dev Dyn236, 502–511 (2007). 12 Wilkinson, L. et al. Crim1KST264/KST264 mice implicate Crim1 in the regulation of vascular endothelial growth factor-A activity during glomerular vascular development. Journal of the American Society of Nephrology18, 1697–1708 (2007). 13 Wilkinson, L. et al. CRIM1 regulates the rate of processing and delivery of bone morphogenetic proteins to the cell surface. J Biol Chem278, 34181-34188 (2003).
182 14 Kolle, G., Georgas, K., Holmes, G. P., Little, M. H. & Yamada, T. CRIM1, a novel gene encoding a cysteine-rich repeat protein, is developmentally regulated and implicated in vertebrate CNS development and organogenesis. Mech Dev90, 181-193 (2000). 15 Chiu, H. S. et al. Production of a mouse line with a conditional Crim1 mutant allele. Genesis50, 711-716 (2012). 16 Fung, W. Y., Fat, K. F., Eng, C. K. & Lau, C. K. crm-1 facilitates BMP signaling to control body size in Caenorhabditis elegans. Dev Biol311, 95-105 (2007). 17 James, R. E. & Broihier, H. T. Crimpy inhibits the BMP homolog Gbb in motoneurons to enable proper growth control at the Drosophila neuromuscular junction. Development138, 3273-3286 (2011). 18 Ponferrada, V. G. et al. CRIM1 complexes with β-catenin and cadherins, stabilizes cell-cell junctions and is critical for neural morphogenesis. PLoS One7, e32635 (2012). 19 Sucov, H. M., Gu, Y., Thomas, S., Li, P. & Pashmforoush, M. Epicardial control of myocardial proliferation and morphogenesis. Pediatr Cardiol30, 617-625 (2009). 20 Chen, J., Kubalak, S. W. & Chien, K. R. Ventricular muscle-restricted targeting of the RXRa gene reveals a non-cell-autonomous requirement in cardiac chamber morphogenesis. Development125, 1943-1949 (1998). 21 O'Brien, T. X., Lee, K. J. & Chien, K. R. Positional specification of ventricular myosin light chain 2 expression in the primitive murine heart tube. Proc Natl Acad Sci U S A90, 5157- 5161 (1993). 22 Minamisawa, S., Gu, Y., Ross, J., Jr., Chien, K. R. & Chen, J. A post-transcriptional compensatory pathway in heterozygous ventricular myosin light chain 2-deficient mice results in lack of gene dosage effect during normal cardiac growth or hypertrophy. J Biol Chem274, 10066-10070 (1999). 23 Pashmforoush, M. et al. Nkx2-5 pathways and congenital heart disease; loss of ventricular myocyte lineage specification leads to progressive cardiomyopathy and complete heart block. Cell117, 373-386 (2004). 24 Zhou, B. & Pu, W. T. Genetic Cre-loxP assessment of epicardial cell fate using Wt1-driven Cre alleles. Circ Res111, e276-280 (2012). 25 Wilm, B., Ipenberg, A., Hastie, N. D., Burch, J. B. & Bader, D. M. The serosal mesothelium is a major source of smooth muscle cells of the gut vasculature. Development132, 5317- 5328 (2005). 26 Fan, J. et al. Crim1 maintains retinal vascular stability during development by regulating endothelial cell Vegfa autocrine signaling. Development141, 448-459, (2014). 27 Wu, M. et al. Epicardial spindle orientation controls cell entry into the myocardium. Dev Cell19, 114-125 (2010). 28 van Wijk, B., Gunst, Q. D., Moorman, A. F. M. & van den Hoff, M. J. B. Cardiac Regeneration from Activated Epicardium. PLoS ONE7, e44692 (2012). 29 Duan, J. et al. Wnt1/βcatenin injury response activates the epicardium and cardiac fibroblasts to promote cardiac repair. The EMBO Journal31, 429-442 (2012). 30 Limana, F., Capogrossi, M. C. & Germani, A. The epicardium in cardiac repair: from the stem cell view. Pharmacol Ther129, 82-96 (2011). 183 31 Zhou, B. et al. Adult mouse epicardium modulates myocardial injury by secreting paracrine factors. J Clin Invest121, 1894-1904 (2011). 32 Huang, G. N. et al. C/EBP transcription factors mediate epicardial activation during heart development and injury. Science338, 1599-1603 (2012). 33 Lovicu, F. J., Kolle, G., Yamada, T., Little, M. H. & McAvoy, J. W. Expression of Crim1 during murine ocular development. Mech Dev94, 261-265 (2000). 34 Georgas, K., Bowles, J., Yamada, T., Koopman, P. & Little, M. H. Characterisation of Crim1 expression in the developing mouse urogenital tract reveals a sexually dimorphic gonadal expression pattern. Dev Dyn219, 582-587 (2000). 35 Chen, T. H. P. et al. Epicardial induction of fetal cardiomyocyte proliferation via a retinoic acid-inducible trophic factor. Dev Biol250, 198-207 (2002). 36 Kwee, L. et al. Defective development of the embryonic and extraembryonic circulatory systems in vascular cell adhesion molecule (VCAM-1) deficient mice. Development121, 489-503 (1995). 37 Moore, A. W., McInnes, L., Kreidberg, J., Hastie, N. D. & Schedl, A. YAC complementation shows a requirement for Wt1 in the development of epicardium, adrenal gland and throughout nephrogenesis. Development126, 1845-1857 (1999).
38 Yang, J. T., Rayburn, H. & Hynes, R. O. Cell adhesion events mediated by a4 integrins are essential in placental and cardiac development. Development121, 549-560 (1995). 39 Gittenberger-de Groot, A. C., Vrancken Peeters, M. P., Bergwerff, M., Mentink, M. M. & Poelmann, R. E. Epicardial outgrowth inhibition leads to compensatory mesothelial outflow tract collar and abnormal cardiac septation and coronary formation. Circ Res87, 969-971 (2000). 40 Pennisi, D. J., Ballard, V. L. & Mikawa, T. Epicardium is required for the full rate of myocyte proliferation and levels of expression of myocyte mitogenic factors FGF2 and its receptor, FGFR-1, but not for transmural myocardial patterning in the embryonic chick heart. Dev Dyn228, 161-172 (2003). 41 Kang, J. et al. PDGF-A as an epicardial mitogen during heart development. Dev Dyn237, 692-701 (2008). 42 Li, P. et al. IGF signaling directs ventricular cardiomyocyte proliferation during embyronic heart development. Development138, 1795-1805 (2011). 43 Hwa, V., Oh, Y. & Rosenfeld, R. G. The insulin-like growth factor-binding protein (IGFBP) superfamily. Endocrine Reviews20, 761-787 (1999). 44 Mebratu, Y. & Tesfaigzi, Y. How ERK1/2 Activation Controls Cell Proliferation and Cell Death Is Subcellular Localization the Answer? Cell cycle (Georgetown, Tex.)8, 1168-1175 (2009). 45 Muñoz-Chápuli, R. et al. The Epicardium and Epicardial-Derived Cells: Multiple Functions in Cardiac Development. Revista Española de Cardiología (English Version)55, 1070-1082 (2002). 46 Souders, C. A., Bowers, S. L. K. & Baudino, T. A. Cardiac Fibroblast: The Renaissance Cell. Circ Res105, 1164-1176 (2009).
184 47 Camelliti, P., Borg, T. K. & Kohl, P. Structural and functional characterisation of cardiac fibroblasts. Cardiovascular Research65, 40-51 (2005). 48 Smart, N. & Riley, P. R. The epicardium as a candidate for heart regeneration. Future Cardiol8, 53-69 (2012). 49 Soriano, P. Generalized lacZ expression with the ROSA26 Cre reporter strain. Nature Genetics21, 70-71 (1999). 50 Zhou, B. et al. Epicardial progenitors contribute to the cardiomyocyte lineage in the developing heart. Nature454, 109-113 (2008). 51 Leighton, P. A. et al. Defining brain wiring patterns and mechanisms through gene trapping in mice. Nature410, 174-179 (2001). 52 Chong, J. J. et al. Adult cardiac-resident MSC-like stem cells with a proepicardial origin. Cell Stem Cell9, 527-540 (2011). 53 Ashe, A. et al. Mutations in mouse Ift144 model the craniofacial, limb and rib defects in skeletal ciliopathies. Human Molecular Genetics21, 1808-1823 (2012). 54 Rudat, C. & Kispert, A. Wt1 and epicardial fate mapping. Circ Res111, 165-169 (2012). 55 Wilhelm, D. et al. SOX9 regulates prostaglandin D synthase gene transcription in vivo to ensure testis development. J Biol Chem282, 10553-10560 (2007). 56 Chiu, H. S. et al. Comparative gene expression analysis of genital tubercle development reveals a putative appendicular Wnt7 network for the epidermal differentiation. Dev Biol344, 1071-1087 (2010).
Acknowledgements This work was funded by grants from the National Health and Medical Research Council (NHMRC) to DJP (631658), MP (1057751), MHL (301056 and 455972). MHL is a Senior Principal Research Fellow of the NHMRC, and MP holds an Australian Research Council Future Fellowship (FT120100170). SI was supported by a UQ Research Scholarship. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. We are grateful to Professor Richard Harvey for providing us with the Mlc2v-Cre and WT1-CreERT2 mouse lines. We thank Dr. Tracey Harvey and Dr. Aaron Smith for helpful discussions, Ms. Sabrina Oishi for assistance with bright field microscopy and Mr. Lachlan Harris for advice on statistical analyses. We thank the staff of UQBR animal facilities and The Centre for Microscopy and Microanalysis at The University of Queensland for support.
Author Contributions Conceived and designed the experiments: SI, MHL, DJP. Performed the experiments: SI, FYC, RW, HSC, VKSR, DJP. Analysed the data: SI, WGT, MP, DJP. Contributed
185 reagents/materials/analysis tools and manuscript editing: SI, MHL, WGT, MP, DJP. Wrote the paper: SI, MP, DJP.
Additional Information The authors declare no competing financial interests.
186 Figures and figure legends
Figure 1. Crim1 expression as monitored by the Crim1KST264 genetrap reporter reveals a dynamic pattern during heart development. A, micrograph of a histological section of a 9.5 dpc Crim1+/KST264 embryo that has been X-Gal-stained, paraffin-embedded, and counter-stained with nuclear fast red. B, higher magnification view of the boxed area in A. The proepicardial cells are X-Gal-positive (arrow, B). Note that X-Gal staining is localised close to 187 nuclei due to the targeting of the β-geo to cell bodies that was typical of the secretory trap vector used in the genetrap screen that generated the Crim1KST264 allele [49]. C–G, Whole-mount views of Crim1+/KST264 hearts after X-Gal staining (blue) at 12.5 dpc (C), 15.5 dpc (D), 16.5 dpc (E), 17.5 dpc (F), and 18.5 dpc (G). In addition to expression in the epicardium at these stages, note the Crim1-LacZ expression in the OFT mesenchyme (arrows) at 12.5 dpc, and the smooth muscle of the great vessels (arrows, D–G). H–J, Magnified left lateral views of the ventricles of the hearts shown in E–G, respectively. Note the Crim1-LacZ expression in the developing coronary vasculature (arrows) and in the epicardium (double open arrowheads). K-M, histological sections Crim1-LacZ expression in the epicardium (arrows) and M, in the coronary vasculature (open arrowhead). my, myocardium; pe, proepicardium. Scale bar A, 100 µm; B, 20 µm; C, 200 µm; D–G, 500 µm; H–J, 200 µm; K-M, 20 µm.
188
Figure 2. Developmental defects in hearts of Crim1Δflox/Δfloxembryos. A–B, D-E Representative whole-mount views of 14.5 dpc embryonic hearts (A, B, Crim1+/+; D, E, Crim1Δflox/Δflox). B, E, Magnified, left lateral views of A and D, respectively. C,F, Representative whole- mount views of 17 dpc embryonic hearts (C, Crim1+/+; F, Crim1Δflox/Δflox). Note the reduced size of the ventricles in the Crim1Δflox/Δfloxheart (D, E, F). Epicardial defects in Crim1Δflox/Δfloxhearts. G-I, Hematoxylin 189 and eosin-stained sections of the ventricular region of 13.5 dpc G, Crim1+/+ and H,I, two different Crim1Δflox/Δfloxhearts. Note the irregular appearance of the epicardium in Crim1Δflox/Δfloxembryos, with one example showing blebbing (H). J–M, Scanning electron micrographs of the ventricular surface of 13.5 dpc Crim1+/+ (J, L) and Crim1Δflox/Δflox(K, M) hearts. K, Note the loss of the even spacing of cells of the mesothelial epicardium, and M, diffusely arranged microvilli. N, Quantification of penetrance of epicardial defects, specifically either epicardial blebbing or a loss of squamous morphology, between Crim1+/+ and Crim1Δflox/Δflox hearts at 13.5 dpc. Z-score -3.8801. *, P<0.0001. cm, compact myocardium; ep, epicardium. Scale bars; A–F, 500 µm; G-I, 50 µm; J-K, 20 µm; L-M, 2 µm.
Figure 3. Crim1 is necessary for normal compact myocardial development. A and B, Merged confocal images of left ventricular sections of 13.0 dpc hearts from Crim1+/+ (A) and Crim1Δflox/Δflox (B) hearts stained with DAPI (blue), PECAM-1 (red), and actin (green). C, Compact myocardium cell density was unchanged in Crim1Δflox/Δfloxhearts. D, Reduced ventricular compact myocardial thickness in Crim1Δflox/Δfloxhearts at 13.0 dpc (n=6–8). E and F, Merged confocal images of ventricular sections of 13.0 dpc hearts stained with phospho- Histone H3 (pHH3, white), DAPI (blue), and actin (green). An epicardial cell (arrowhead, E) and myocardial cell (arrowhead, F) immuno-positive for pHH3 are shown. G, No difference in proliferation in the epicardium of Crim1Δflox/Δfloxhearts. H, An almost two-fold increase in proliferation in the compact myocardium of Crim1Δflox/Δfloxhearts. I and J, Examples of merged confocal images of ventricular sections of 13.0 dpc hearts stained with cleaved caspase 3 (CC3, white), DAPI (blue), and actin (green). An epicardial cell (arrowhead, I) and myocardial cell (arrowhead, J) immuno-positive for CC3 are shown. K, No difference in apoptosis in the epicardium of Crim1Δflox/Δflox hearts. H, A four-fold
190 increase in apoptosis in the compact myocardium of Crim1Δflox/Δfloxhearts. cm, compact myocardium; ep, epicardium; n.s., not significant. (n=4–5), *, P<0.05; **, P<0.01. Scale bars A, B, 50 μm; E, F, I, J, 20 μm.
Figure 4. A cell-autonomous requirement for Crim1 to control epicardial migration and myocardial invasion in vitro and in vivo. A, Schematic of the transwell assay used for the in vitro migration assay. B, Representative micrograph of a primary 11.5 dpc ventricular explant after attachment and migration of epicardial cells. The front of the epicardial monolayer is indicated (arrows). The primary explant is denoted by an asterisk. C, Primary epicardial cells after removal of the ventricular explant, passage and subsequent culture. Note maintenance of a “cobblestone” appearance of the cellular monolayer. D, Epicardial identity after enrichment was confirmed by immunostaining for WT1 (green nuclear signal). The no-primary antibody control showed no signal (inset). E, Crim1KST264/KST264primary epicardial cells showed increased migration in vitro relative to controls (n=3–5 and representative of three independent experiments; *, P<0.05). F, J, Whole-mount dorsal views of 15.5 dpc embryonic hearts after X-Gal staining (blue) of Crim1+/FLOX; WT1-CreERT2; R26R and Crim1FLOX/FLOX; WT1-CreERT2; R26R genotype, respectively. Two doses of Tamoxifen were administered at 9.5 dpc and 10.5 dpc prior to harvesting the embryos. G, K, Histological sections of hearts from Crim1+/FLOX; WT1-CreERT2; R26R and Crim1FLOX/FLOX; WT1-CreERT2; R26R 15.5 dpc embryos, respectively, and H, L, magnified views of boxed regions showing X-Gal-positive epicardial, sub-epicardial and intramural cells. I, M, Quantification of X-Gal-positive cells showing increased myocardial EPDCs in the right ventricles of mutant hearts (n=4-6; *P<0.05). ep, epicardial; se, sub-epicardial; my, intramyocardial; LV, left ventricle; RV, right ventricle. Scale bars, G, K, 20 μm; H, L, 10 μm.
191
Figure 5. Changes in epicardial and myocardial signalling, alongside localisation of β-catenin in Crim1Δflox/Δfloxhearts. A-B, E-F, Confocal images of ventricular sections of 13.0 dpc hearts from Crim1+/Δflox (A, B) and Crim1Δflox/Δflox(E, F) embryos stained for phospho-SMAD2/3 (white). A, B, merged images of DAPI (blue) and actin (green) to delineate the myocardium with the phospho-SMAD2/3 (white) from B and F, respectively. C, D, G, and H, There was a reduction in phospho-SMAD2 signal intensity in epicardial cells of left and right ventricles of Crim1Δflox/Δfloxhearts, but the levels in compact myocardial cells was unchanged (n=4–5). Confocal images of ventricular sections of 13.5 dpc hearts from Crim1+/+ (I, J) and Crim1Δflox/Δflox(M and N) embryos stained for phospho-ERK1/2 (red). I, M, merged images of DAPI (blue) and phospho- ERK1/2 (red). K, L, O and P, There was an increase in phospho-ERK1/2 signal intensity in compact
192 myocardial cells of left and right ventricles of Crim1Δflox/Δfloxhearts, but the levels in epicardial cells was unchanged (n=8-6). Q-R, T-U, Confocal images of ventricular sections of 13.5 dpc hearts from Crim1+/+ (Q, R, S) and Crim1Δflox/Δflox(T, U, V) embryos stained for β-catenin (red). Q, T, merged images of DAPI (blue) and MF20 (green) to delineate the epicardium with the β-catenin (red) from R, S and U, V respectively. S, V, magnified views of boxed regions in R and U. Note β-catenin distribution at epicardial cell-cell junctions of Crim1+/+ hearts (arrows, R, S), and a decreased accumulation at epicardial cell junctions in mutants (double arrowheads, V) . W, the percentage of epicardial cells with an accumulation of β-catenin at cell junctions (n=5-6). cm, compact myocardium; ep, epicardium; s.ep, sub epicardium. n.s., not significant. *, P<0.05. Scale bars, A-B, E-F, 10 μm; I-J, M-N, 50 μm; Q-R, T-U; 10 μm.
193
Figure 6. Epicardium-restricted Crim1 loss-of-function results in decreased proliferation of EPDCs and cardiac fibroblasts in both left and right ventricles. A,B, phospho-Histone H3 immunohistochemistry on X-Gal stained Crim1+/FLOX; +/WT1-CreERT2; R26R and Crim1FLOX/FLOX; +/WT1-CreERT2; R26R sections at 15.5 dpc, counter stained with 1% DAB followed by NFR staining for visualisation of nuclei, showing A, pHH3- and X-Gal-positive nuclei (arrow) and B, a pHH3-positive nucleus (open arrowhead). C,D, Quantification of proliferating EPDCs showing no change in the epicardium, but a significant reduction in the number of pHH3-positive EPDCs in mutant hearts
194 compared to control in the myocardium (n=6-4). E-G, Periostin immunohistochemistry on X-Gal stained Crim1+/FLOX; +/WT1-CreERT2; R26R and Crim1FLOX/FLOX; +/WT1-CreERT2; R26R sections at 17.5 dpc, counter stained with 1%DAB followed by Hematoxylin staining, showing E, a periostin- and X-Gal-positive cardiac fibroblast (arrow), F, an X-Gal-positive EPDC (open arrowhead) and G, a periostin-positive cardiac fibroblast (double open arrowheads). H,I, quantification of % periostin-positive cells among X-Gal-positive myocardial cells in left and right ventricles, showing a decrease in number of cardiac fibroblasts (n=3-5). *, P<0.05. n.s., not significant; LV, left ventricle; RV, right ventricle. Scale bars A,B 50 µm; E-G, 10 µm.
195 Supplementary Information
Supplementary figure 1. Crim1 is expressed in the outflow tract, mitral valve leaflets and coronary vascular smooth muscle cells. A, Periostin immunohistochemistry on an X-Gal stained Crim1+/KST264 transverse section at 12.5 dpc. B, higher magnification view of the boxed area in A, showing periostin- and X-Galpositive outflow tract mesenchymal cells (arrows). C, Periostin immunohistochemistry on an X-Gal stained Crim1+/KST264 coronal section at 13.5 dpc. D, higher magnification view of boxed area in C, showing periostin- and X-Gal-positive mesenchymal cells (arrows). E, 3 Periostin immunohistochemistry on an X-Gal stained Crim1+/KST264 coronal section at 18.5 dpc, showing periostin- and X-Gal-positive mitral valve leaflet cells (arrows). F, Transgelin immunohistochemistry on an X-Gal stained Crim1+/KST264 coronal section at 18.5 dpc, showing transgelin-
196 and X-Gal-positive coronary vascular smooth muscle cells (arrows). Scale bars, A, C, 100 µm; B, D, E, 40 µm; F, 10 µm.
Supplementary figure 2. Congenital heart defects in Crim1KST264/KST264 embryos. A–C, 18.5 dpc littermate hearts stained with X-Gal to reveal Crim1-LacZ expression (blue). A, Crim1+/KST264 18.5 dpc heart. B, Crim1KST264/KST264 18.5 dpc heart. C, Quantification of the 5 size of the ventricular component at 18.5 dpc, showing a significant reduction in the width of Crim1KST264/KST264 hearts (n=7–9). D-
197 F, Coronal sections of 18.5 dpc hearts stained with XGal and counter-stained with nuclear fast red. The right atrium (RA) is labelled. D, Crim1+/KST264 18.5 dpc heart showing normal morphology. E,F, Crim1KST264/KST264 hearts displaying a prominent VSD, a communication between left and right ventricles (arrow). G, Another section of the heart in F, showing the VSD. H, View of the inset in G. Note the Crim1-LacZ-positive cells (arrows) in the IVS either side of the communication between the ventricles. I, Quantification of prevalence of VSDs in Crim1+/KST264 and Crim1KST264/KST264 hearts (n=5). Z-Score -2.582. *, P<0.01. LV, left ventricle; RV, right ventricle. Scale bars D, 500 μm; H, 50 μm; K, 10μm.
Supplementary figure 3. Crim1 is not necessary for specification or early development of the proepicardium. A-F, Micrographs of 9.5 dpc embryos from Crim1+/+ (A, C, E) and Crim1Δflox/Δflox(B, D, F) embryos stained for Gata4 mRNA expression (purple stain, A and B) and Tbx18 mRNA expression (purple stain, C and D).
198 E, F, Micrographs of histological sections of wholemount in situ hybridization-stained embryos showing Tbx18 expression. The PE is indicated (arrows). Scale bars A and E, 50 µm.
Supplementary figure 4. Myocardial deletion of Crim1 does not affect cardiomyocyte proliferation and survival and thickness of compact myocardium. A, D representative whole mount images of X-Gal stained Crim1+/FLOX; Mlc2v- Cre; R26R and Crim1FLOX/FLOX; Mlc2v-Cre; R26R hearts at 13.5 dpc. B, E merged confocal images of DAPI (blue) and pHH3 (red), with a pHH3-positive nucleus (arrow). C, F merged confocal images of DAPI (blue) and CC3 (red), with a CC3-positive nucleus (arrow). G, No significant difference in percentage of pHH3-positive cells in the compact myocardium of left and right ventricles in control and mutant hearts (n=4-6). H, No significant difference in percentage of CC3-positive cells in the compact myocardium of left and right ventricles in control and mutant hearts (n=4-5). I, No significant difference in thickness of compact myocardium in the left ventricle of control and mutant hearts (n=3). cm, compact myocardium; ep, epicardium; n.s, not significant. Scale bars B, C, E, F 20 µm.
199
Supplementary figure 5. Fidelity of the WT1-Cre and WT1-CreERT2 lines. A-E, MF20 immunohistochemistry performed on A-C, 15.5 dpc WT1-Cre X-Gal stained sections showing MF20-negative epicardium (arrows) and MF20-negative EPDCs (open arrowheads), D, WT1-CreERT2 13.5 dpc X-gal stained section showing that most epicardial cells are X-Gal-positive (arrow), and E, WT1- CreERT2 17.5 dpc X-Gal stained section showing MF20-negative EPDCs (open arrowheads), confirming that the EPDCs are not cardiomyocytes. ep; epicardium. Scale bars A-C, 25µm; D-E, 50 µm.
200
Supplementary figure 6. Phospho-SMAD1/5 and phospho-AKT levels are not significantly changed in the epicardium or myocardium of Crim1Δflox/Δfloxhearts. Confocal images of ventricular sections of 13.0 dpc hearts from Crim1+/Δflox(A, B) and Crim1Δflox/Δflox(E, F) embryos stained for phospho-SMAD1/5 (white). A and E, merged images of DAPI (blue) and actin (green) to delineate the myocardium with the phosphoSMAD1/5 (white) from A and E, respectively. C, D, G, and H, Quantification of the average signal intensity/cell after indirect immunofluorescence to detect phospho- SMAD1/5 showed there was no change in the levels in the epicardium or compact myocardium of left and right ventricles of Crim1Δflox/Δfloxhearts in comparison to controls (n=5). Confocal images of ventricular sections of 13.5 dpc hearts from Crim1+/+(I, J) and Crim1Δflox/Δflox(M, N) embryos stained for phospho-AKT (red). Quantification of signal intensity/ cell after 10 immunofluorescence to detect phospho-AKT revealed no change in the levels in the epicardium or compact myocardium of left and right ventricles of Crim1Δflox/Δfloxhearts in comparison to controls (n=5-6). cm, compact myocardium; ep, epicardium; n.s., not significant. Scale bars, A, B, E, F 10µm; I, J, M, N, 50 µm.
201
Supplementary figure 7. No visible change in filamentous actin morphology in the epicardium or myocardium of Crim1Δflox/Δfloxhearts at 13.5 dpc. A-F, Representative confocal images of ventricular sections of 13.5 dpc hearts from Crim1+/+(A, B, C) and Crim1Δflox/Δflox(D, E, F) hearts at 13.5 dpc. A, D merged images of DAPI (blue) and MF20 (green) to delineate the myocardium with Phalloidin (red) from B and E, respectively (n=4-5). C, F, magnified views of boxed regions in B and E respectively. Scale bars A – F, 20 µm.
Supplementary figure 8. No change in the expression of key cardiac genes in Crim1Δflox/Δfloxhearts at 17.5 dpc. A, qPCR with fibroblast, smooth muscle and cardiomyocyte markers showing no significant difference in fold change between Crim1+/+and Crim1Δflox/Δfloxventricular samples at 17.5 dpc. Note that for Collagen1a, P=0.0556. (n=4-5). 202 Published research article 2.
CRIM1 is necessary for coronary vascular endothelial cell development and homeostasis.
Iyer, Swati1; Chhabra, Yash1; Harvey, Tracey J1; Wang, Richard1; Chiu, Han Sheng1; Smith, A.G1; Thomas, Walter G1; Pennisi, David J1*#; Piper, Michael1,2*# 1School of Biomedical Sciences, and 2Queensland Brain Institute, The University of Queensland, Brisbane, 4072, Australia.
# These authors contributed equally to this work. *Address correspondence to: A/Prof. Michael Piper / Dr. David Pennisi School of Biomedical Sciences, The University of Queensland, Brisbane, 4072, Australia Email: [email protected] / [email protected]
Abstract Endothelial cells form a critical component of the coronary vasculature, yet the factors regulating their development remain poorly defined. Here we reveal a novel role for the transmembrane protein CRIM1 in mediating cardiac endothelial cell development. In the absence of Crim1 in vivo, the coronary vasculature is malformed, the number of endothelial cells reduced, and the canonical BMP pathway dysregulated. Moreover, we reveal that CRIM1 can bind IGFs, and regulate IGF signalling within endothelial cells. Finally, loss of CRIM1 from human cardiac endothelial cells results in misregulation of endothelial genes involved in an increased inflammatory response and cytolysis, reminiscent of endothelial cell dysfunction in cardiovascular disease pathogenesis. Collectively, these findings implicate Crim1 in endothelial cell development and homeostasis in the coronary vasculature.
203 Introduction Endothelial cells play an important role in angiogenesis and vasculogenesis, and regulate vessel permeability and vascular tone (Aird, 2012). The heart contains several subtypes of endothelial cells - the endocardial lining of the heart chambers and valves, myocardial microvessels, and coronary macrovessels - which differ considerably in their function and gene expression (Hendrickx et al., 2004). Endothelial cells both produce and respond to a variety of cytokines and growth factors, and their misregulation contributes to cardiac disease progression (Pate et al., 2010).
Bone Morphogenetic Proteins (BMPs) (Dyer et al., 2014) and Insulin-like Growth Factor (IGF) (Delafontaine et al., 2004) are crucial to endothelial cell biology, and as such, these growth factors undergo regulation at multiple levels.Modulators such as chordin, noggin and crossveinless-2 contain cysteine-rich repeats (CRRs), which potentiate binding to BMPs (Larraín et al., 2000, Zimmerman et al., 1996, Groppe et al., 2002, Zhang et al., 2010). Endothelial cells also express IGF-binding proteins (IGFBPs), CRR-containing proteins which act to mediate the effect of IGF binding by controlling the availability of IGFs as well as reducing their activity by sequestering them such that they cannot bind to and activate IGF receptors (Hwa et al., 1999).
Another protein implicated in the regulation of growth factor activity during development is cysteine rich transmembrane BMP regulator 1 (CRIM1) (Kolle et al., 2000, Wilkinson et al., 2003, Wilkinson et al., 2007). It contains six CRRs via which it binds growth factors such as TGFβ, BMP and VEGF when both are co-expressed in the same cell (Kolle et al., 2000, Wilkinson et al., 2003, Wilkinson et al., 2007), as well as an IGFBP domain (Kolle et al., 2000), though the significance of the latter remains to be determined. Crim1 has been previously implicated in cardiac development (Iyer et al., 2016), but how this factor mediates cardiac endothelial cell biology is unknown. Here we reveal a role for CRIM1 in cardiac microvascular endothelial cell development, demonstrating that canonical BMP signalling is dysregulated in the absence of Crim1, that CRIM1 can bind IGF-1, and that the IGF signalling pathway is altered upon loss of CRIM1 function. We also report the misexpression of a number disease-related endothelial genes in the absence of CRIM1, collectively defining a role for CRIM1 in normal endothelial cell function in the heart.
Results Absence of Crim1 affects coronary endothelial cell endowment. Crim1 mutant embryos display developmental defects including reduced ventricular size, abnormal epicardial morphology and malformed vasculature (Iyer et al., 2016). However, the effect of Crim1-deficiency in coronary endothelial cells is unknown. Analysis ofCrim1Δflox/Δflox 204 embryos revealed a decrease in endothelial cell endowment in the compact myocardium of the ventricular free walls at 17.5 dpc (Figure 1A-F). To address whether Crim1 is required cell- autonomously for coronary vascular endothelial cell development, we conditionally ablated Crim1 from endothelial cells using aTie2-Cre line(Kisanuki et al., 2001, Li et al., 2005). As with the constitutive knockout, we detected a significant reduction in endothelial cell endowment in the ventricular myocardium of conditional mutants in comparison to controls at 17.5 dpc (Figure 1G-L). This phenotype was not due to apoptosis at this stage, given the limited death of endothelial cells (Figure S1A-F), nor did we observe marked alterations in endothelial cell proliferation in mutant embryos (Figure S1G-L). As such we hypothesised that the lack of Crim1 led to alterations in growth factor signalling that led to reduced endothelial cell endowment.
Canonical BMP signalling is perturbed in the absence of Crim1. CRIM1 binds to growth factors, including BMPs, which are known to drive cardiac development. To investigate whether BMP signalling was abnormal within coronary vascular endothelial cells of Crim1-deficient embryos, we assessed phosphorylated SMAD1/5 (pSMAD1/5) expression as a proxy for BMP activity, as phosphorylation of these intracellular effectors is part of the canonical signalling pathway downstream of BMPs (Dyer et al., 2014). Quantification of fluorescent labelling revealed a reduction in pSMAD1/5 signal intensity in Crim1Δflox/Δflox hearts at 17.5 dpc in all cells, as well as specifically within endothelial cells (Figure 1M-T).
CRIM1 regulates cell viability and growth factor signalling. Having implicated Crim1 in the modulation of BMP signalling in cardiac endothelial cells in vivo, we next sought to further characterise signalling pathways perturbed upon reduction of CRIM1 expression. Using shRNA-mediated knockdown of CRIM1 (Fan et al., 2014) in Human Cardiac Microvascular Endothelial Cells (HCMVECs) we achieved approximately a 75% reduction in CRIM1 expression, a finding correlated with reduced cellular viability (Figure 2A, B). Next, analysis of protein lysates from CRIM1 knockdown HCMVECsrevealed significant decreases in pSMAD1/5 and phosphorylated IGF-1 receptor (pIGF-1Rβ) expression in CRIM1-knockdown lysates compared to scrambled shRNA controls, whereas no change in pERK1/2 and pAKT was observed (Figure 2C, D). These findings suggest that the activity of BMP and IGF signalling pathways is aberrant when CRIM1 expression is perturbed. Interestingly, an interaction between CRIM1 and IGF signalling has not previously been reported. To probe this further, 2 splice variants of IGF-1, IGF-1 Ea-HA and IGF-1 Eb-HA (Hede et al., 2012), were co-immunoprecipitated with CRIM1-myc (Figure S2A). Using a series of CRIM1 expression constructs (Figure 3A), coupled 205 with co-immunoprecipitation, we found that CRIM1 binds to IGF-1 via CRRs 3-6 and its IGFBP domain (Figure 3B, C). Collectively, these findings suggest that CRIM1 may regulate multiple signalling pathways to mediate cardiac endothelial development and homeostasis.
Misregulation of endothelial genes in the absence of CRIM1 Lastly, we investigated the mechanisms through which CRIM1 regulates endothelial cell biology by assessing the transcriptional consequences of CRIM1 knockdown in HCMVECs using a Human Endothelial Cell Biology qPCR array. Of the 84 genes analysed, 29 were significantly misregulated upon Crim1 knockdown (>2-fold up/downregulation) (Figure 4A, Table S1). Ingenuity Pathway Analysis further revealed enrichment of genes involved in cellular functions including inflammation, cytolysis and immune cell infiltration, many of which are also implicated in cardiovascular disease (Figure 4B, C). Together, these data provide an initial mechanistic insight into the cascade of perturbed signalling events arising from aberrant CRIM1 expression in endothelial cells.
Discussion Our findings provide a platform for understanding the role that CRIM1 plays in the development and function of cardiac endothelial cells. Our data demonstrate that Crim1 regulates endothelial cell endowment, and that CRIM1 can bind IGF, and can modulate BMP and IGF downstream signalling. Given that CRIM1 deficiency leads to reduced IGF-1 and BMP signalling, we posit that in the heart CRIM1 serves as an agonist of IGF-1 and BMP signalling. BMPs and IGFs are required for the development of endothelial cells (Suzuki et al., 2008, Langenfeld and Langenfeld, 2004, Higashi et al., 2014), and thus it is plausible that a reduction in BMP and IGF signalling due to loss of CRIM1 could lead to endothelial stress and dysfunction, which leads to reduced endothelial cell endowment. Interestingly, Crim1FLOX/FLOX; Tie2 Cre mice survive to adulthood (data not shown), suggesting that the reduced endowment of endothelial cells during development does not preclude survival. However, given our qPCR array findings, further analysis of the response of these adult mice to cardiac stress and injury, as well as their inflammatory responses, would serve to deepen our understanding of the role of CRIM1- mediated endothelial cell development and how this affects cardiac function later in life.
One caveat of this study is that the binding of CRIM1 to IGF-1 was assessed only when both CRIM1 and the growth factor were co-expressed by the same cell. Examination of interactions between CRIM1 and growth factors expressed by paracrine sources may provide a greater mechanistic understanding of CRIM1 function within the complex milieu of the nascent heart. 206 Finally, these data also suggest that aberrant CRIM1 expression may culminate in abnormal signalling correlated with inflammation and heart disease. The development of atherosclerosis, a leading cause of coronary artery disease and risk factor for myocardial infarction, has been linked to activation of the endothelium and orchestration of an inflammatory response (Hansson, 2005). Initiation of atherogenesis involves this recruitment of inflammatory cells to the endothelium, and is further exacerbated by stimuli such as hyperlipidemia, high blood pressure and a consequently heightened pro-inflammatory cascade (Libby et al., 2011). Furthermore, the limited ability of the endothelium to repair itself has made it critical to uncover molecules that can counter or prevent endothelial dysfunction. Thus, the presence of CRIM1 could counter the pathophysiology of various diseases that are directly involved with endothelial abnormalities. Given these findings, understanding how CRIM1 may mediate endothelial cell biology within the mature heart is an exciting field of future research.
Materials and Methods Ethics Statement All experimental protocols were approved by The University of Queensland‟s Animal Ethics Committee (AIBN/274/08/NHMRC, SMBS/420/11/NHMRC), and followed the National Institutes of Health Guide for the Care and Use of Animals, and the Australian Code of Practice for the Care and Use of Animals for Scientific Purposes.
Mouse lines The mouse lines used in this study were the Crim1FLOXconditional knock-out (Chiu et al., 2012); Crim1Δfloxconstitutive knock-out (Chiu et al., 2012), and Tie2-Cre mouse lines (Kisanuki et al., 2001). Embryos were used at 17.5 dpc.
Antibodies Primary antibodies used were: rabbit anti-phosphohistone H3 (pSer10, #06-570; Millipore); rabbit anti-cleaved caspase 3 (#9664; Cell Signaling Technology); rabbit anti-pSMAD1/5 (pSer463/pSer465, #700047; Invitrogen); total SMAD1/5/8/9 (NB100-56656; Novus Biologicals); rat anti-CD31 (PECAM-1, #550274; BD Pharmingen); pIGF-1Rβ (#3024S; Cell Signalling); total AKT (#9272; Cell Signalling); total ERK1/2 (sc-93; Santa Cruz); total IGF-1R (sc-712; Santa Cruz); GAPDH (#2118; Cell Signalling); β-tubulin (#2146; Cell Signalling); Angiogenesis sampler kit (#8696; Cell Signalling); rabbit HA tag (#3724; Cell Signalling); mouse myc tag (#2276; Cell Signalling); normal mouse IgG2a isotype control (#02-6200; Thermofisher Scientific). Secondary antibodies were Alexa Fluor-488, and 594-conjugated secondary antibodies (Invitrogen) rabbit HRP 207 (#W4011; Promega), mouse HRP (#W4021; Promega), IgG mouse IRDYE (#926-32212; Licor) and IgG rabbit IRDYE (#926-68073; Licor).
Cell Culture H-CMVECs (Lonza, CC-7030) were cultured in EGM-2MV BulletKit media (Lonza, CC-3202). The cell line was authenticated and tested for contamination by Lonza Walkersville/Lonza Australia. Cells before passage six were used in all experiments. shRNA targeting CRIM1 mRNA (Fan et al, 2014) was acquired from TRC Mission Library (TRCN0000063862, Invitrogen). Lentiviral particles were synthesized by transfecting CRIM1 shRNA, scrambled shRNA and the lentiviral packaging vectors into HEK293T cells using Lipofectamine 2000 and collecting the supernatant containing lentiviral particles at 48 hours after transfection. Cells at 80% confluence were incubated with lentiviral particles together with 8 μg/mL polybrene for 2 days and then treated with 2 μg/mL puromycin for 2 days to enrich the infected cell population. After puromycin removal, cells were used in subsequent experiments.
Trypan Blue Viability Assay Control and CRIM1-knockdown CMVECs (5000 cells) were plated into wells of a 24-well plate at time zero. The number of viable cells at 24 hours and 48 hours was counted based on Trypan Blue exclusion. q-PCR Array RNA was extracted using the RNeasy Mini Kit (Qiagen). cDNA was prepared using RT2 First Strand Kit (Qiagen). Quantitative PCR to assess percentage CRIM1 knockdown was performed with QuantiFast SYBR Green PCR kit (Qiagen). The RT2 SYBR Green ROX qPCR Mastermix was used for Human Endothelial cell Biology Profiler PCR Array (Plate E). Signals were detected using the Roche LightCycler and Applied Biosystems QuantStudio 7 system and gene expression levels were calculated using the standard curve method. Gene analyses were performed using the Qiagen Gene Globe Data Analysis Center and the Ingenuity Pathway Analysis.
Generation of deletion constructs Full-length myc-tagged CRIM1(Wilkinson et al., 2003) was used as a template to generate myc- tagged CRIM1 deletion constructs. Primers were designed as follows:
CRIM1-3.1-InF-RV-F 5‟-TGGAATTCTGCAGATATGTACTTGGTGGCGGGG 208
CRIM1-3.1-InF-RV-R 5‟-GCCACTGTGCTGGATTCATAGGTCTTCTTCTGA-3‟
CRIM1-IGFBPsplice-InF-F 5‟-GAGAACTGGACTGATGAC-3‟
CRIM1-IGFBPsplice-InF-R 5‟-ATCAGTGCAGTTCTCCGCCCGGGTGCCGGAGCGCGC-3‟
CRIM1-D1-2-IF-F 5‟-GAAGAACCAACCATCATC-3‟
CRIM1-D1-2-IF-R 5‟-GATGGTTGGTTCTTCGGCTGGCTTTGTATCATTAAC-3‟
CRIM1-D3-6-IF-F 5‟-CCAGAAATGTATGTCCCA-3‟
CRIM1-D3-6-IF-R 5‟-GACATACATTTCTGGAGTGCCCGACAGGATGGGTGG-3‟
CRIM1-D1-6-IF-R 5‟-GACATACATTTCTGGGGCTGGCTTTGTATCATTAAC-3‟
Cloning was performed using the In-Fusion Kit (Clontech, #638910). Constructs were sequenced to confirm deletion.
Co-Immunoprecipitation HEK293 cells were grown to 80% confluence and transfected with full length myc-tagged CRIM1(Wilkinson et al., 2003), myc-tagged CRIM1 deletion constructs, and HA-tagged IGF-1 Ea and IGF-1 Eb constructs (Hede et al., 2012) using Lipofectamine 2000 (Invitrogen). After 24 hours, the cells were lysed in RIPA containing protease inhibitor and phosphatase inhibitor on ice. Immunoprecipitation was performed by the addition of α-myc antibody for 2 hours at 4 degrees, followed by addition of 30 μL of Protein A Agarose (Roche, #11719408001) and a further 1 209 hourincubation at 4 degrees. The beads were pelleted and washed five times with cold PBS 0.1% Triton X-100. Protein was eluted by addition of SDS-PAGE loading buffer and boiled for 4 minutes. Western Blotting was performed according to standard methods.
Data Analysis Fluorescent images of tissue sections were captured on an Olympus FV1000 upright laser scanning confocal microscope. For all quantification, 40x images were acquired of the left and right mid- ventricular myocardial free wall from at least two-three non-consecutive sections of each sample. The number of PECAM-1 positive cells were counted and divided by total cell number to determine the PECAM-1-positive percentage of endothelial cells. For quantification of pSMAD1/5 levels, Olympus Fluoview software (ver3.1a) was used to determine total signal intensity in, and area of, the specified region of interest. The number of DAPI-stained nuclei in the demarcated area was quantified using ImageJ software. Signal intensity within specifically endothelial cells was determined using the IMARIS software (ver8.2.0). Endocardial / trabecular endothelial cells were not included in the quantifications.
Immunoblots were visualised either with a chemi-luminescent reagent (Merck Millipore) via exposure to X-ray film, or by using the Licor Odyssey scanner. All densitometry quantitation on immunoblots was performed using ImageJ software. All test values were normalised to respective loading and experimental controls and graphed with Prism 6.
Statistical significance between pooled data from mutant and control groups was determined using a two-tailed, unpaired Student‟s t-test or a standard one-way ANOVA with Dunnett‟s post-test, as specified in the figure legend. Quantified data shown are mean ± standard deviation. GraphPad Prism v6.0 was used to conduct all analyses and generate graphs.
Acknowledgements We are grateful to Prof. Melissa Little and Prof. Nadia Rosenthal for providing us with the CRIM1 and IGF-1 Ea /Eb expression constructs, respectively. Tie2-Cre mice were a kind gift from Prof. Richard Harvey at the Victor Chang Cardiac Research Institute. This work was funded by grants from the National Health and Medical Research Council to MP (1057751) and DJP (631658). MP holds an Australian Research Council Future Fellowship (FT120100170). SI was supported by a UQ Research Scholarship. Performed experiments: SI, YC, TJH. Analysed data: SI, MP, DJP. Wrote paper: SI, MP. Edited paper: SI, WGT, AGS, MP, DJP. 210 References
AIRD, W. C. 2012. Endothelial Cell Heterogeneity. Cold Spring Harbor Perspectives in Medicine, 2, a006429. CHIU, H. S., YORK, J. P., WILKINSON, L., ZHANG, P., LITTLE, M. H. & PENNISI, D. J. 2012. Production of a mouse line with a conditional Crim1 mutant allele. Genesis, 50, 711-716. DELAFONTAINE, P., SONG, Y.-H. & LI, Y. 2004. Expression, Regulation, and Function of IGF- 1, IGF-1R, and IGF-1 Binding Proteins in Blood Vessels. Arteriosclerosis, Thrombosis, and Vascular Biology, 24, 435-444. DYER, L. A., PI, X. & PATTERSON, C. 2014. The role of BMPs in endothelial cell function and dysfunction. Trends in endocrinology and metabolism: TEM, 25, 472-480. FAN, J., PONFERRADA, V. G., SATO, T., VEMARAJU, S., FRUTTIGER, M., GERHARDT, H., FERRARA, N. & LANG, R. A. 2014. Crim1 maintains retinal vascular stability during development by regulating endothelial cell Vegfa autocrine signaling. Development, 141, 448-459. GROPPE, J., GREENWALD, J., WIATER, E., RODRIGUEZ-LEON, J., ECONOMIDES, A. N., KWIATKOWSKI, W., AFFOLTER, M., VALE, W. W., BELMONTE, J. C. & CHOE, S. 2002. Structural basis of BMP signalling inhibition by the cystine knot protein Noggin. Nature, 420, 636-42. HANSSON, G. K. M. D. P. 2005. MECHANISMS OF DISEASE: Inflammation, Atherosclerosis, and Coronary Artery Disease. The New England Journal of Medicine, 352, 1685-95. HEDE, M. S., SALIMOVA, E., PISZCZEK, A., PERLAS, E., WINN, N., NASTASI, T. & ROSENTHAL, N. 2012. E-peptides control bioavailability of IGF-1. PLoS One, 7, e51152. HENDRICKX, J., DOGGEN, K., WEINBERG, E. O., VAN TONGELEN, P., FRANSEN, P. & DE KEULENAER, G. W. 2004. Molecular diversity of cardiac endothelial cells in vitro and in vivo. Physiological Genomics, 19, 198-206. HIGASHI, Y., QUEVEDO, H. C., TIWARI, S., SUKHANOV, S., SHAI, S. Y., ANWAR, A. & DELAFONTAINE, P. 2014. Interaction between insulin-like growth factor-1 and atherosclerosis and vascular aging. Front Horm Res, 43, 107-24. HWA, V., OH, Y. & ROSENFELD, R. G. 1999. The insulin-like growth factor-binding protein (IGFBP) superfamily. Endocrine Reviews, 20, 761-787. IYER, S., CHOU, F. Y., WANG, R., CHIU, H. S., RAJU, V. K. S., LITTLE, M. H., THOMAS, W. G., PIPER, M. & PENNISI, D. J. 2016. Crim1 has cell-autonomous and paracrine roles during embryonic heart development. Scientific Reports, 6, 19832. KISANUKI, Y. Y., HAMMER, R. E., MIYAZAKI, J.-I., WILLIAMS, S. C., RICHARDSON, J. A. & YANAGISAWA, M. 2001. Tie2-Cre Transgenic Mice: A New Model for Endothelial Cell-Lineage Analysis in Vivo. Developmental Biology, 230, 230-242. KOLLE, G., GEORGAS, K., HOLMES, G. P., LITTLE, M. H. & YAMADA, T. 2000. CRIM1, a novel gene encoding a cysteine-rich repeat protein, is developmentally regulated and implicated in vertebrate CNS development and organogenesis. Mechanisms of Development, 90, 181-193. LANGENFELD, E. M. & LANGENFELD, J. 2004. Bone morphogenetic protein-2 stimulates angiogenesis in developing tumors. Mol Cancer Res, 2, 141-9. 211 LARRAÍN, J., BACHILLER, D., LU, B., AGIUS, E., PICCOLO, S. & DE ROBERTIS, E. M. 2000. BMP-binding modules in chordin: a model for signalling regulation in the extracellular space. Development (Cambridge, England), 127, 821-830. LI, W. L., CHENG, X., TAN, X. H., ZHANG, J. S., SUN, Y. S., CHEN, L. & YANG, X. 2005. Endothelial cell-specific expression of Cre recombinase in transgenic mice. Yi Chuan Xue Bao, 32, 909-15. LIBBY, P., RIDKER, P. M. & HANSSON, G. K. 2011. Progress and challenges in translating the biology of atherosclerosis. Nature, 473. PATE, M., DAMARLA, V., CHI, D. S., NEGI, S. & KRISHNASWAMY, G. 2010. Chapter 5 - Endothelial Cell Biology: Role in the Inflammatory Response. In: GREGORY, S. M. (ed.) Advances in Clinical Chemistry. Elsevier. SUZUKI, Y., MONTAGNE, K., NISHIHARA, A., WATABE, T. & MIYAZONO, K. 2008. BMPs Promote Proliferation and Migration of Endothelial Cells via Stimulation of VEGF- A/VEGFR2 and Angiopoietin-1/Tie2 Signalling. Journal of Biochemistry, 143, 199-206. WILKINSON, L., GILBERT, T., KINNA, G., RUTA, L. A., PENNISI, D., KETT, M. & LITTLE, M. H. 2007. Crim1KST264/KST264 mice implicate Crim1 in the regulation of vascular endothelial growth factor-A activity during glomerular vascular development. Journal of the American Society of Nephrology, 18, 1697–1708. WILKINSON, L., KOLLE, G., WEN, D., PIPER, M., SCOTT, J. & LITTLE, M. 2003. CRIM1 regulates the rate of processing and delivery of bone morphogenetic proteins to the cell surface. Journal of Biological Chemistry, 278, 34181-34188. ZHANG, J.-L., PATTERSON, L. J., QIU, L.-Y., GRAZIUSSI, D., SEBALD, W. & HAMMERSCHMIDT, M. 2010. Binding between Crossveinless-2 and Chordin Von Willebrand Factor Type C Domains Promotes BMP Signaling by Blocking Chordin Activity. PLoS ONE, 5, e12846. ZIMMERMAN, L. B., DE JESUS-ESCOBAR, J. M. & HARLAND, R. M. 1996. The Spemann organizer signal noggin binds and inactivates bone morphogenetic protein 4. Cell, 86, 599- 606.
212 Figures and Figure legends
Figure 1: Loss of Crim1 results in reduced coronary endothelial cell endowment, and dysregulation of canonical BMP signalling, in vivo. A-D, M-P, Confocal images of ventricular sections of 17.5 dpc hearts from Crim1+/+, Crim1 +/Δflox (A, B, M, N) and Crim1Δflox/Δflox (C, D, O, P) embryos, immunostained with PECAM-1 (A-D), and PECAM-1 213 and pSMAD1/5 (M-P). E, F, Reduced endothelial cell endowment as observed by a reduced percentage of PECAM-1+ve endothelial cells in the ventricular compact myocardium of Crim1Δflox/Δflox hearts (n=4). Q, R, Reduction in pSMAD1/5 levels in all cells of both ventricles of Crim1Δflox/Δflox hearts and S, T, endothelial cells of the left ventricle (n=4). G-J, Confocal images of ventricular sections of 17.5 dpc hearts from Crim1+/FLOX; Tie2-Cre (G, H) and Crim1FLOX/FLOX; Tie2-Cre (I, J) embryos, immunostained with PECAM- 1 (G-J). K, L, Reduced endothelial cell endowment as observed by a reduced percentage of PECAM-1+ve endothelial cells in the ventricular compact myocardium ofCrim1FLOX/FLOX; Tie2-Cre hearts (n=9-7). RV, right ventricle; LV, left ventricle; ep, epicardium; en, endocardium; n.s, not significant.Two-tailed, unpaired Student‟s t-test; *P<0.05. Scale bars A-D, G-J, M-P, 50 μm. Error bars, SD
Figure 2: CRIM1 ablationresults in decreased endothelial cell viability and reduced pSMAD1/5 levels A, Approx. 75% CRIM1 knockdown in HCMVECs (n=3). B, Trypan Blue Viability Assay demonstrating decrease in cell viability in CRIM1 knockdown HCMVECs compared to scrambled shRNA controls after 24 and 48 hours respectively, after plating 5000 cells per well of a 24 well plate (n = 3, and each n represents an average of 3 replicates).C, D, Reduction in pSMAD1/5 and pIGF-1Rβ levels as a proxy to assess BMP and IGF-1 signalling in CRIM1 knockdown HCMVECs, but no change in pERK1/2 and pAKT levels, compared to scrambled shRNA control HCMVECs (n = 4-7).Two-tailed, unpaired Student‟s t-test; *P < 0.05, **P < 0.01, ****P < 0.0001.Error bars, SD.
214 Figure 3: CRIM1 binds IGF-1 via the IGFBP domain and CRRs 3-6. A, Representation of CRIM1 deletion constructs as generated using InFusion cloning. 1. Full length CRIM1- myc, 2. CRIM1ΔIGFBP-myc lacking the IGFBP domain, 3.CRIM11-2-myc lacking CRRs 1 and 2, 4.CRIM13-6-myc lacking CRRs 3, 4, 5 and 6, 5.CRIM11-6-myc lacking all 6 CRRs. 6. CRIM1ΔIGFBP1-6-myc lacking the IGFBP domain and all six CRRs. B, C, CRIM1 binds IGF-1 Eb via the IGFBP domain and CRRs 3-6. A, HEK293 cells transiently co-transfected with full length CRIM1-myc, CRIM1ΔIGFBP-myc, CRIM1Δ1-2-myc, CRIM1Δ3-6-myc, CRIM1Δ1-6-myc, CRIM1ΔIGFBPΔ1-6-myc, and IGF-1 Eb-HA. Control cells transiently co-transfected with only CRIM1-myc and only IGF-1 Eb-HA. Immunoprecipitation performed with anti-myc and immunoblotting with anti-HA. IB bands were detected only when both CRIM1 and IGF-1 Eb were co-transfected. B, CRIM1 binds IGF-1 Eb via the IGFBP domain and CRRs 3-6, as indicated by 1-way ANOVA analysis with Dunnett‟s post-test, which reveals that deletion of the IGFBP domain and CRRs 3-6 of CRIM1 significantly reduces binding of CRIM1 to IGF-1 Eb. Deletion of CRRs 1-2 does not affect binding to IGF-1 Eb. (3 independent IPs performed. Representative blot chosen).**P < 0.01, ***P < 0.001. Error bars, SD.
215
Figure 4. Misregulation of genes involved in endothelial cell function in the absence of CRIM1. A, Scatterplot displaying misregulation of 29 genes out of 84, involved in endothelial cell function, with a fold-change cut-off of 2. The scatter plot compares the normalized expression of every gene on the array between the two selected groups by plotting them against one another to visualize gene expression changes. The central line indicates unchanged gene expression. The dotted lines indicate the selected fold regulation 216 threshold.B, C, IPA analysis of genes misregulated in the human endothelial cell biology qPCR array, indicating an increase in genes predicted to be associated with a heightened inflammatory response and cytolysis (B, Table S1), along with heart failure, infarction and vascular lesions (C, Table S1). Z scores > 2 are significant. (n = 4, with 2 technical repeats).
217 Supplementary Information
Supplementary figure 1: Increase in apoptosis but no change in proliferation of endothelial cells that have lost Crim1. A-D, G, H, Confocal images of ventricular sections of 17.5 dpc hearts from Crim1+/FLOX; Tie2-Cre (A, B, G) and Crim1FLOX/FLOX; Tie2-Cre (C, D, H) embryos, immunostained with PECAM-1 and Cleaved Caspase-3 (CC3) (A-D) and PECAM-1 and Phosphohistone H3 (PHH3) (G, H). E, F, Limited apoptosis in the left ventricles of Crim1FLOX/FLOX; Tie2-Cre hearts, immunostained with CC3 (n=6-7). I-L, No change in
218 proliferation of endothelial cells in both ventricles of Crim1FLOX/FLOX; Tie2-Cre hearts as assessed with PHH3 (I, J) and Ki67 (K, L). RV, right ventricle; LV, left ventricle; ep, epicardium; en, endocardium. *P<0.05. Scale bars A, C, G, H, 50 μm. B, D, 10 μm. Error bars, SD
Supplementary figure 2: COIP revealing that CRIM1 can bind IGF-1 Ea and IGF-1 Eb A, HEK293 cells transiently co-transfected with full length CRIM1-myc and IGF-1 Ea-HA and IGF-1 Eb- HA. Immunoprecipitation performed with anti-myc and immunoblotting with anti-HA. IB bands were detected only when both CRIM1 and IGF-1 Ea/Eb were co-transfected. (n=3).
Table S1: List of all 84 endothelial genes assessed in the human endothelial cell biology array, with fold regulation and P values, after CRIM1 knockdown. Significantly misregulated genes (>2-fold up/downregulation) are highlighted in bold font.
NUMBER GENE REFSEQ FOLD REGULATION P-VALUE 1 ACE NM_000789 1.0909 0.290151 2 ADAM17 NM_003183 -2.5219 0.00018 3 AGT NM_000029 -1.4969 0.076943 4 AGTR1 NM_031850 -1.1724 0.522084 5 ALOX5 NM_000698 1.0231 0.969908 6 ANGPT1 NM_001146 -1.8019 0.007017 7 ANXA5 NM_001154 -2.2338 0.000006
219 8 APOE NM_000041 -1.6211 0.044427 9 BAX NM_004324 -1.0566 0.281637 10 BCL2 NM_000633 -1.5905 0.030518 11 BCL2L1 NM_138578 -1.3774 0.000041 12 CALCA NM_001741 1.0374 0.488782 13 CASP1 NM_033292 5.1266 0.00003 14 CASP3 NM_004346 1.0501 0.424219 15 CAV1 NM_001753 -2.0448 0.000345 16 CCL2 NM_002982 1.1958 0.04106 17 CCL5 NM_002985 15.0637 0.000131 18 CDH5 NM_001795 -1.4064 0.000322 19 CFLAR NM_003879 -1.0154 0.820888 20 COL18A1 NM_030582 -1.3561 0.007195 21 CX3CL1 NM_002996 6.139 0.026342 22 EDN1 NM_001955 -1.8019 0.000243 23 EDN2 NM_001956 1.0073 0.925998 24 EDNRA NM_001957 -1.431 0.154058 25 ENG NM_000118 -1.2393 0.007858 26 F2R NM_001992 1.32 0.01127 27 F3 NM_001993 3.7529 0.000074 28 FAS NM_000043 1.8797 0.000744 29 FASLG NM_000639 1.0374 0.488782 30 FGF1 NM_000800 1.32 0.277886 31 FGF2 NM_002006 -1.0566 0.535619 32 FLT1 NM_002019 -1.1266 0.145134 33 FN1 NM_002026 -2.2611 0.0004 34 HIF1A NM_001530 -1.0457 0.107155 35 HMOX1 NM_002133 -1.1247 0.142892 36 ICAM1 NM_000201 1.7207 0.000004 37 IL11 NM_000641 -2.117 0.013737 38 IL1B NM_000576 7.0153 0.001666 39 IL3 NM_000588 1.0374 0.488782 40 IL6 NM_000600 1.7752 0.002456 41 IL7 NM_000880 1.6621 0.124883
220 42 ITGA5 NM_002205 -1.2414 0.001696 43 ITGAV NM_002210 -2.3983 0.00012 44 ITGB1 NM_002211 -2.8669 0.000001 45 ITGB3 NM_000212 -1.5099 0.003451 46 KDR NM_002253 -1.0996 0.224994 47 KIT NM_000222 -3.7373 0.000736 48 KLK3 NM_001648 -1.0845 0.479634 49 MMP1 NM_002421 -1.5204 0.000071 50 MMP2 NM_004530 -1.2986 0.012208 51 MMP9 NM_004994 2.247 0.008745 52 NOS3 NM_000603 1.0629 0.378124 53 NPPB NM_002521 -3.0042 0.007349 54 NPR1 NM_000906 1.8188 0.005741 55 OCLN NM_002538 -2.3367 0 56 PDGFRA NM_006206 -1.456 0.104707 57 PECAM1 NM_000442 1.0091 0.766073 58 PF4 NM_002619 -4.2047 0.030872 59 PGF NM_002632 -1.7255 0.000027 60 PLAT NM_000930 -1.4459 0.126347 61 PLAU NM_002658 1.0556 0.288138 62 PLG NM_000301 1.0374 0.488782 63 PROCR NM_006404 -2.3327 0.000002 64 PTGIS NM_000961 -2.0028 0.008857 65 PTGS2 NM_000963 2.6126 0.000009 66 PTK2 NM_005607 -1.3967 0.000021 67 SELE NM_000450 7.0641 0.000001 68 SELL NM_000655 1.6364 0.000056 69 SELPLG NM_003006 -2.1803 0.029765 70 SERPINE1 NM_000602 -4.6251 0.000001 71 SOD1 NM_000454 -1.7863 0.000009 72 SPHK1 NM_021972 -2.1391 0.000192 73 TEK NM_000459 -2.2532 0 74 TFPI NM_006287 -1.2986 0.000011 75 TGFB1 NM_000660 -1.1092 0.093609
221 76 THBD NM_000361 -2.4487 0.000072 77 THBS1 NM_003246 -1.5878 0.000133 78 TIMP1 NM_003254 1.3269 0.004462 79 TNF NM_000594 1.7752 0.0015 80 TNFSF10 NM_003810 9.5666 0.000001 81 TYMP NM_001953 4.494 0.000058 82 VCAM1 NM_001078 5.056 0 83 VEGFA NM_003376 -1.0677 0.210602 84 VWF NM_000552 -1.2117 0.025954
222 Published review article 1. Crim1– a regulator of developmental organogenesis.
Swati Iyer1, David J. Pennisi1, Michael Piper1,2,# 1School of Biomedical Sciences, and 2Queensland Brain Institute, The University of Queensland, Brisbane, 4072, Australia.
#Address correspondence to: Dr. Michael Piper School of Biomedical Sciences, The University of Queensland, Brisbane, 4072, Australia Email: [email protected]
Abstract The regulation of growth factor localization, availability and activity is critical during embryogenesis to ensure appropriate organogenesis. This process is regulated through the coordinated expression of growth factors and their cognate receptors, as well as via proteins that can bind, sequester or localize growth factors to distinct locations. One such protein is the transmembrane protein Crim1. This protein has been shown to be expressed broadly within the developing embryo, and to regulate organogenesis within the eye, kidney and placenta. Mechanistically, Crim1 has been revealed to mediate organogenesis via its interaction with growth factors including TGFβs, BMPs, VEGFs and PDFGs. More recently, Crim1 has been shown to influence cardiac development, providing further insights into the function of this protein. This review will provide an overview of the role of Crim1 in organogenesis, largely focusing on how this protein regulates growth factor signaling in the nascent heart. Moreover, we will address the challenges ahead relating to further elucidating how Crim1 functions during development.
223 List of Abbreviations BMP – Bone Morphogenetic Protein Crim1 – Cysteine-Rich Transmembrane BMP Regulator-1 CRR – Cysteine-rich Repeat EMT – Epithelial-to-Mesenchymal Transition EndoMT – Endothelial-to-Mesenchymal Transition EPDC – Epicardium-Derived Cells ERK – Extracellular signal-Regulated Kinase FAK – Focal Adhesion Kinase Gbb – Glass bottom boat HUVEC – Human Umbilical Vein Endothelial Cell IGF – Insulin-like Growth Factor IGFBP – Insulin-like Growth Factor Binding Protein NMJ – Neuromuscular Junction PDGF – Platelet-Derived Growth Factor PE – Proepicardium RGD – Arginyl-glycyl-aspartic acid motif TGFβ – Transforming Growth Factor β VEGF – Vascular Endothelial Growth Factor
224 Introduction
Cysteine-Rich Transmembrane BMP Regulator-1 (Crim1) is a novel,N-glycosylated transmembrane protein encoded by the Crim1gene (Kolle et al., 2000; Glienke et al., 2002). Crim1 genes are evolutionarily conserved in vertebrates including rodents and humans(Kolle et al., 2000), as well as zebrafish (Kinna et al., 2006), the chicken (Kolle et al., 2003) and Xenopus(Ponferrada et al., 2012). Interestingly, a Crim1 homolog called crm-1 has been described in Caenorhabditis elegans as well, although the role of this factor in the nematode has only received limited attention (Fung et al., 2007). Crim1is expressed in various organs during embryogenesis, including the spinal cord, lens, kidney, vasculature and placenta (Kolle et al., 2000; Lovicu et al., 2000; Glienke et al., 2002; Kolle et al., 2003; Pennisi et al., 2007; Pennisi et al., 2012; Phua et al., 2012; Fan et al., 2014), indicative of a role in their development. Structurally, the presence of six cysteine-rich repeat (CRR) motifs, an Insulin-like Growth Factor (IGF) Binding Protein (IGFBP) like domain and an
Arginyl-glycyl-aspartic acid motif(RGD)(Kolle et al., 2000)suggests thatCrim1 can bind a variety of different proteins. In support of this, the CRRs have been shown to mediate the binding of CRIM1 to TGFβ, BMP, VEGF and PDGF when Crim1 is co-expressed in the same cell as the growth factor (Wilkinson et al., 2003; Wilkinson et al., 2007). Recently, Crim1 has also been shown to interact with β1 Integrin via its RGD domain (Zhang et al., 2016). However, the functional significance of the IGFBP motif remains unknown. Interestingly, Crim1 has been demonstrated to localize in the endoplasmic reticulum and golgi, where post translational modification of proteins are known to occur, and also the cell surface, where ligand-receptor interactions occur (Glienke et al., 2002; Wilkinson et al., 2003), Pennisi lab, unpublished data).
The importance of proteins that regulate growth factor signalling has been highlighted by recent findings that implicate molecules containing multiple cysteine-rich regions, such as chordin and noggin, during development. For example, BMP signal transduction is regulated by various proteins that exert either a pro- or anti-BMP effect. Mediation of this pathway can occur at multiple levels, such as by sequestering BMPs and either facilitating or inhibiting BMP-receptor interactions, hindering BMP transport across cells and thus disturbing the BMP gradient required for downstream signaling, or by maintaining some BMPs in inactive form by preventing the cleavage of pre-BMPs and reducing the amount of mature, secreted BMPs (Wilkinson et al., 2003; Umulis et al., 2009).Chordin is an antagonist that directly binds BMPs (Larraín et al., 2000), and noggin, a cysteine-knot protein that binds BMPs and prevents them from binding to type I and type II cell surface BMPRs via masking of receptor binding sites (Zimmerman et al., 1996; Groppe et al., 2002). On the other hand, crossveinless-2 potentiates BMP signaling by forming a ternary complex
225 with both chordin and BMP, and reduces the affinity of chordin for BMP, thus allowing BMPs to activate their downstream effectors via the BMPRs (Zhang et al., 2010).Importantly, the action of these proteins is dependent of the developmental context in which they, and the molecules they interact with, are expressed. This has been highlighted by Crim1-deficient mice, which exhibit phenotypes that indicate Crim1 can perform both agonistic and antagonistic functions on growth factor signaling during organogenesis.
To explore the role of Crim1 in development, a number of transgenic mice have been generated. Firstly, a genetrap line was created by random insertion of a β-geo cassette into intron 1 of the Crim1 gene (called Crim1KST264)(Leighton et al., 2001; Pennisi et al., 2007). Secondly, a conditional loss-of-function allele was generated (Crim1FLOX), by flanking exons 3 and 4 of the Crim1gene with LoxP sites(Chiu et al., 2012). The Crim1FLOX line, when crossed with the CMV-Cre deletor line, generated Crim1Δflox mice, intercrosses from which produced mice lacking a functional Crim1 gene(called Crim1Δflox/Δflox) (Chiu et al., 2012). We have previously described perinatal lethality in mice homozygous for the Crim1KST264genetrap and in Crim1∆flox/∆flox mice(Pennisi et al., 2007; Chiu et al., 2012). Both strains display defects in multiple organ systems including the kidney, eye and placenta(Lovicu et al., 2000; Glienke et al., 2002; Pennisi et al., 2007; Pennisi et al., 2012; Phua et al., 2012; Fan et al., 2014), highlighting the importance of Crim1 during development. In this review we discuss the role of Crim1 in organogenesis, with a specific focus on the developing heart, as well as providing mechanistic insight into how it can regulate the activity of growth factors.
The role of Crim1 in organogenesis In the past 15 years a number of studies have begun to map the expression of Crim1 within the developing embryo, and have used the different transgenic mouse lines described above to decipher the role for Crim1 in organogenesis. A summary of these findings is detailed below.
Kidney Perhaps the most widely studied organ in the context of Crim1 function is the kidney. In the embryonic murine kidney, Crim1 is expressed in pericytes lining the endothelium and within the parietal epithelial cells, mesangial cells and podocytes of the glomeruli from 15.5 days post coitum (dpc) (Georgas et al., 2000; Pennisi et al., 2007; Wilkinson et al., 2007). Studies of transgenic animals have provided significant insights into the role of Crim1 during kidney development. Kidneys from Crim1KST264/KST264mice at 15.5 dpc were significantly smaller than their wild type littermates, indicating a role for Crim1 in nephrogenesis(Pennisi et al., 2007). Further analysis revealed multiple lesions in the glomerulus and glomerular capillary defects, and podocyte
226 effacement (Wilkinson et al., 2007). In their elegant study, Wilkinson et al. showed Crim1 to be co- expressed with VEGF-A in the podocytes of the renal glomerulus. In mice lacking Crim1, there was an increased diffusion of VEGF-A away from the podocytes at 17.5 dpc, and a concomitant activation of the VEGF-A receptor Flk1 in adjacent vascular endothelial cells, supporting the observation of glomerular defects in these homozygotic mice(Wilkinson et al., 2007). To assess whether renal abnormalities were present in the adult kidney, Wilkinson et al. made use of Crim1KST264/KST264 outbred mice, as a proportion of these homozygous animals survive to adulthood (Wilkinson et al., 2007). The kidneys of these adult mice displayed multiple glomerular cysts, interstitial fibrosis and endothelial cell thickening (Wilkinson et al., 2007), accompanied by further evidence of increased vascular leakiness and compromised extraglomerular vasculature (Wilkinson et al., 2009).Furthermore,a later study revealed renal fibrosis in Crim1KST264/KST264 adult mice along with endothelial aberrations, including an increase in vascular permeability and a discontinuous endothelium displaying abnormal collagen deposits (Phua et al., 2012). This could be due to the association of Crim1 with TGFβ-1, which is known stimulate endothelial-to-mesenchymal transformation (EndoMT)(Kim et al., 2001; Varga and Wrana, 2005; Phua et al., 2012). In the adult human kidney, CRIM1 has been shown to be localized in the podocytes, both qualitatively, using renal tissue in immunohistochemical experiments, and quantitatively, using immortalised human podocytes (Nyström et al., 2009), where BMPs and VEGFs are also expressed(Simon et al., 1995; Godin et al., 1999), indicating a possible role for CRIM1 in tethering these growth factors and releasing them into the local environment in a controlled manner.
Placenta The placenta plays a central role during embryogenesis. It comprises a labyrinthine zone consisting of syncytiotrophoblast cells and junctional zone consisting of spongiotrophoblast and glycogen trophoblast cells. In mice, Crim1 is expressedin various placental cell types including chorionic trophoblasts at 9.5 dpc, synctiotrophoblasts at 13.5 and 15.5 dpc, and spongiotrophoblasts from 13.5 dpc(Pennisi et al., 2012).Crim1 is important for placental development as, in the absence of this gene, placental size is reduced from 13.5 dpc until 17.5 dpc, with a consequent reduction in the size of Crim1KST264/KST264 embryos at the later stage(Pennisi et al., 2012). There is also a decrease in the number of sinusoidal-trophoblast giant cells and an increase in glycogen cells of Crim1KST264/KST264 placentae at 15.5 dpc, hypothesized to be due to a possible dysregulation of multiple signalling pathways (Pennisi et al., 2012), such as the IGF and VEGF pathways (Charnock-Jones et al., 1994; Randhawa and Cohen, 2005).
Lens and retinal vasculature
227 In the developing murine embryonic eye, Crim1 expression is observed from 9.5 dpc until at least day postnatal day (P) 21 (Lovicu et al., 2000). Initially, Crim1 is detected in the precursor to the lens, the lens placode, and subsequently is expressed by all lens cells by 11.5 dpc.Crim1 transcripts are also present in the corneal epithelium and endothelium by 15.5 dpc, as well as the retinal epithelium and retinal ganglion cells at 18.5 dpc. At P21, Crim1 expression persists only in the lens (Lovicu et al., 2000). Analysis ofCrim1glcr11(glaucoma relevant 11)mutants, termed so because of their glaucoma and cataract phenotype, revealsmultiple lens defects that are evident from 16.5 dpc, including abnormal cell adhesion at epithelial adhesion junctions, disrupted polarity and a reduction in the number of proliferating lens epithelial cells, which collectively culminate in a smaller, atypical lens(Zhang et al., 2016). By P60, Crim1 mutant mice also display lens cataracts and abnormal cellular proliferation within the retina. Mechanistically, the adhesion defects in these Crim1 mutants are consistent with an interaction observed between Crim1 and β1 integrin, which is also expressed in both lens epithelial cells and lens fiber cells (Zhang et al., 2016). Indeed, analysis of β1 integrin signalling reveals that Crim1 regulatesthe phosphorylation status of its downstream effectors focal adhesion kinase (FAK) and extracellular signal-regulated kinase (ERK), resulting in modulation of lens morphogenesis by membrane-bound Crim1 (Zhang et al., 2016).
As Crim1 is expressed in vivo and in vitro in vascular endothelial cells (Glienke et al., 2002), and in the vasculature of the embryonic mouse hindbrain and postnatal retinas (Fan et al., 2014), its role in the retinal vasculature has also been analysed. Crossing the Crim1Flox allele to Pdgfrb-iCreERmice to enable inducible deletion of Crim1 from endothelial cells reveals thatdefective retinal vascular development occurs in the absence of Crim1 from the vasculature (Fan et al., 2014). The phenotypes observed included reduced vessel density, length and branchpoint number, and vessel regression in the first week of postnatal development (Fan et al., 2014). Indeed, these authors revealed modulation of the autocrine activity of VEGF-A by Crim1, indicating that it has an important regulatory role in the formation and development of the vasculature(Fan et al., 2014). The distribution of the cell adhesion molecule VE-Cadherin at the angiogenic front of Crim1flox/flox; Pdgfb-iCreERretinal vasculature preparations was also altered (Fan et al., 2014), a finding consistent with the impaired endothelial tube formation evident in HUVECs in the absence of CRIM1(Glienke et al., 2002). This is suggestive of a cell adhesion anomaly. A possible role for the conserved RGD motif in Crim1 in this context is possible, as this domain of Crim1 potentially binds integrins and so modulates cellular attachment(Kolle et al., 2000). The intracellular domain of Crim1 could also play a role in cell adhesion, as the cytoplasmic domain of Crim1 has been shown to indirectly bind β-catenin and N-cadherin in Xenopus(Ponferrada et al., 2012). Nervous system
228 Preliminary investigations have also shown that Crim1 may be important for the development of the nervous system. For instance, in the developing mouse spinal cord, Crim1 is expressed from 9.5 dpc in the floor plate, and in pools of motor neurons at later stages of development (Kolle et al., 2000). Moreover, Crim1 expression is observed in other regions of the nascent mouse neuraxis, including the forebrain and hindbrain from 11.5 dpc, and in the midbrain at 13.5dpc, (Kolle et al., 2000). Despite this, studies have yet to elucidate the mechanistic function of Crim1 in the developing nervous system. Looking forwards, using our understanding of the role of Crim1 in other developmental contexts may provide insights into the role of this factor within the developing nervous system. For example, the colocalization of Crim1 and β1 integrinsat the leading edges of lens epithelial cell projections regulates cell adhesion and polarity (Zhang et al., 2016). As cellular adhesion and polarity are critical components that underlie the proliferation and subsequent differentiation of neural stem cells within the embryonic brain, a role for Crim1 in mediating these aspects during neural development is plausible.
The role of Crim1 in cardiac development Overview of organogenesis of the heart Another organ in which Crim1 plays an important role during development is the heart. The heart is the first organ to form and function in the vertebrate embryo (Yutzey and Kirby, 2002), and cardiac progenitor cells are among the earliest to migrate through the primitive streak during gastrulation (Garcia-Martinez and Schoenwolf, 1993; Schoenwolf and Garcia-Martinez, 1995). Splanchnic mesenchymal cells, the principal cardiac precursors, arise from primary heart fields in the lateral plate mesoderm (Waldo et al., 2001) and aggregate in the cardiogenic region to form angioblastic cords. These canalize to form two thin-walled endocardial heart tubes. Their subsequent fusion forms a single two-layered heart tube (Manasek, 1969) comprising the outer mesenchymal myocardial mantle which forms the myocardium, and the inner endothelial tube which forms the endocardium, separated by myocardium-produced cardiac jelly (Waldo et al., 1999).
The epicardium of the heart develops from a transient structure, the proepicardium (PE) (Virágh and Challice, 1981; Männer, 1993; Virágh et al., 1993; Gittenberger-de Groot et al., 1998), that is located between the sinus horns and liver primordium (Virágh et al., 1993), and which is derived from the lateral plate mesoderm (Serluca, 2008). The PE contains smooth muscle, fibroblast and endothelial progenitors (Mikawa and Gourdie, 1996; Dettman et al., 1998; Gittenberger-de Groot et al., 1998; Männer, 1999; Pérez-Pomares et al., 2002), but whether it is the sole contributor of endothelial cells to the coronary vasculature is contentious, as the liver bud and sinus venosus have
229 also beensuggested as sources of the same (Poelmann et al., 1993; Ishii et al., 2007; Red-Horse et al., 2010; Cossette and Misra, 2011).
Proepicardial cells attach to the inner curvature of the atrioventricular junction of the rudimentary heart (Männer, 1993; Ishii et al., 2010) to form its outermost layer – the epicardium. Species- specific migration of proepicardial cells to the myocardium occurs either by proepicardial vesicle budding or via an extracellular matrix bridge to potentially guide the translocation (Nahirney et al., 2003). It has been suggested that the proximity between the liver bud and the PE affects proepicardial attachment to the heart and differentiation via an associated effect on proepicardial marker genes (Ishii et al., 2007). Following the attachment, a population of epicardial cells undergo EMT to form subepicardial cells, including those contributing to the subepicardial coronary vasculature, whereas another subset of these cells traverse the subepicardial space and migrate into the myocardium (Virágh and Challice, 1981) to give rise to various cell types, including coronary vascular smooth muscle cells, coronary vascular endothelial cells and cardiac fibroblasts (Virágh and Challice, 1981; Männer, 1993; Virágh et al., 1993; Mikawa and Gourdie, 1996; Gittenberger-de Groot et al., 1998).
Crim1 and cardiac development What evidence is there that Crim1 plays a role in cardiac development? Firstly, studies on the Crim1KST264 genetrap line (which carries a LacZ reporter), have shown that Crim1-promoter mediated LacZ expression is evident in the murine proepicardium at 9.5 dpc, and within the epicardium throughout cardiac development. It is also observed in coronary vascular smooth muscle cells and, to a weaker extent, in coronary vascular endothelial cells, at later stages of heart development (Pennisi et al., 2007). LacZ expression is also observed in the outflow tract mesenchyme, bicuspid and tricuspid valve leaflets and atrial septum at 18.5 dpc(Iyer et al., 2016). These sites of expression imply that Crim1 may regulate many aspects of cardiac development. Interestingly, there aremany cardiac phenotypes, such as chamber septation and valve defects, hypoplastic ventricular walls and coronary vasculature defects, that arise as a result of dysregulation of growth factors such including TGFβs, BMPs, VEGFs and IGFs (Kim et al., 2001; Chen et al., 2004; Goldman et al., 2009; Uchimura et al., 2009; Li et al., 2011; Wu et al., 2012). This indicates that growth factor activity is normally tightly controlled during cardiogenesis. Given the interaction of Crim1 with many of these factors in other organ systems, and preliminary data indicating thatCrim1 mutant cardiac phenotypesare reminiscent of these phenotypes, we posit that Crim1 mediates cardiogenesis, at least in part, via the regulation of growth factor signalling.
230 For example, epicardial EMT is a vital process that occurs during normal heart development, and several growth factors have been implicated in both EMT, and the differentiation of epicardial cells into their correct lineages. A number of distinct molecular processes work cooperatively in order to initiate and promoteepicardial EMT. As the heart develops, there is significant cross-talk between the epicardium and myocardium, with epicardial signalling via its secreted factors and epicardium- derived cells, as well as signalling from the myocardium, being essential for myocardial growth and differentiation (Sucov et al., 2009) and coronary vascular development (Kang and Sucov, 2005). Thus, it is this reciprocal signalling between the epicardium and myocardium which provides cues to ensure the proper and timely differentiation of epicardial lineages, maturation of the myocardium and the coronary vasculature, and, ultimately, the development and functioning of the heart. For instance, TGFβs stimulate epicardial EMT (Dokic and Dettman, 2006; Olivey et al., 2006; Sánchez and Barnett, 2012). Binding of TGFβ2 and TGFβ3 to TGFβR2, and the subsequent activation of TGFβR1, leads to the phosphorylation of SMAD2/3 proteins and upregulation of transcription factors such as snail1 and slug. These factors repress the expression of E-cadherin, while promoting the expression of Vimentin, RhoA and various ECM molecules (Xu et al., 2009), thus facilitating a transition away from epithelial characteristics. Using the WT1-Cre and WT1-CreERT2 cell lines, we have shown that EMT and epicardial migration are increased in the absence of Crim1(Iyer et al., 2016). The epicardium of Crim1 null mutant hearts surprisingly shows a reduced phospho-SMAD2 level, indicative of reduced TGFβ signalling(Iyer et al., 2016), at 13.5 dpc, despite enhanced EMT. This indicates that there could be a role for Crim1 in the formation or stabilization of cadherin- dependent junctional complexes in epithelial cells, via which it could serve to normally restrain epicardial EMT. β-catenin is a crucial component of adherens junctions, and has been previously shown to complex indirectly with Crim1 (Ponferrada et al., 2012). Indeed, assessment of β-Catenin distribution at epicardial cell-cell junctions is altered in Crim1-deficient mice, indicating a loss of stability at these contact points within the developing heart (Iyer et al., 2016).
Epithelial and mesenchymal cells secrete PDGF-A, which may act as a mitogen that stimulates ventricular development and cardiomyocyte proliferation(Kang et al., 2008). Moreover, PDGF-B can stimulate proepicardial cells expressing smooth muscle markers to undergo epicardial EMT and subsequently commit to the coronary smooth muscle cell lineage, mediated through PDGFR-β (Lu et al., 2001), while PDGFR-α plays an important role in the formation of cardiac fibroblasts(Smith et al., 2011). Both PDGF receptors are implicated in EMT, whereby epicardium-derived cells (EPDCs) give rise to myocardial fibroblasts and vascular smooth muscle cells(Mellgren et al., 2008; Smith et al., 2011). We have recently reported a reduction in the number of EPDC-derived myocardial fibroblasts withinCrim1mutant mice (Iyer et al., 2016). Although Crim1 has been
231 shown to capable of binding PDGF-B (Wilkinson et al., 2007), direct evidence to support a role of Crim1 in the modulation of this pathway in the heart is currently lacking. Future work aimed at investigating this exciting prospect will undoubtedly advance our understanding of the mechanism through which Crim1 regulates the biology of EPDC-derived cells. Moreover, PDGF-B is also expressed by endothelial cells is required for the endothelial-cell-mediated recruitment for coronary vascular smooth muscle cells to the developing coronary vessels (Van den Akker et al., 2008). Given the nature of the Crim1 protein, this cysteine-knot protein could also potentially antagonise this aspect of cardiac development by tethering PDGFs to the cell surface and limiting their action, another fruitful avenue of future research.
Interestingly, in support of this hypothesis, Wilkinson et al.previously identified that Crim1potentially functions as an antagonistic regulator of certain members of the BMP family. Crim1 interacts intracellularly with both BMP4 and BMP7 (Wilkinson et al., 2003), and co- localizes with their respective pre-BMPs within the golgi via its CRRs, ultimately reducing the secretion of mature BMPs. Furthermore, a proportion of the BMPs released remain bound to Crim1 (Wilkinson et al., 2003). Crim1 has also been implicated in tethering BMPs to the cell, which may serve to restrict their functional potential, and, since BMPs are known to act across a restricted distance (Jones et al., 1996), possibly to act in the presentation of BMP ligands to neighbouring cells. BMPs are well known for their role in cardiac development. For instance, BMP2 increases epicardial EMT via TGFβR3 activation in epicardial cell lines (Sánchez and Barnett, 2012), andBMP4 has been shown to play an important role in both atrioventricular and outflow tract septation (Jiao et al., 2003; Liu et al., 2004). The absence of Bmp10 leads to impaired ventricular trabeculation and formation of thin ventricular walls (Neuhaus et al., 1999), a phenotype recapitulated in mice lacking both Bmp 6 and Bmp7(Kim et al., 2001). Given the need for exquisite spatial and temporal modulation if BMP signaling, it is likely that Crim1 also regulates this family of molecules during cardiac development, Indeed, hypoplastic ventricles are observed in Crim1 null mice, alongside a concomitant increase in apoptosis of intramural cells, indicating that Crim1 is necessary for the formation of the myocardium, potentially via the modulation of BMP signalling (Iyer et al., 2016). The use of next generation sequencing in Crim1-deficient mice, coupled with proteomic approaches, could provide a future avenue to determine the role Crim1 plays in the modulation of BMP biology during cardiac development.
IGFs have been also implicated as epicardial mitogenic factors during heart formation. For instance, IGF-2 is secreted from the epicardium and exerts a mitogenic effect on the formation of the compact myocardium(Li et al., 2011). Could the IGFBP motif of Crim1, along with the CRR
232 domains, bind IGFs and regulate their activity?Interestingly, there is an increase in ERK1/2 signalling in the myocardium of Crim1–deficient hearts at 13.5 dpc (Iyer et al., 2016). This indicates that Crim1 regulates signalling molecules secreted by the epicardium, or by the myocardium itself, and is thus essential for myocardial maturation in the early stages of heart development. A large number of growth factors are known to activate the ERK pathway, including IGFs, and downstream ERK signalling can be both pro- and anti-apoptotic, making it important to identify which growth factors are specifically involved in this process, and exactly how this augmented ERK1/2 signalling affects the development of the myocardium. It would thus be useful to assess whether Crim1 can indeed bind IGFs, and further whether Crim1 can modulate this important growth factor in the context of the developing heart.
Concluding remarks While the broad role of Crim1 in developmental organogenesis is now well established, much remains unclear as to how this transmembrane protein exerts its biological influence. Critically, the mechanisms by which Crim1 mediates growth factor signalling in different developmental contexts are still poorly defined. For example, in C. elegans, crm-1 acts as an agonist of a BMP-like pathway, the DBL-1 pathway, in a non-cell-autonomous fashion(Fung et al., 2007), although it is not known whether this occurs through an interaction with the ligand or its receptor. In contrast to this agonistic role with regards to BMP signalling, the Drosophila homolog of Crim1, CRIMPY, antagonizes the function of the BMP ligand Glass bottom boat (Gbb) in motorneurons at the neuromuscular junction (NMJ), and restrains the expansion of the NMJ(James and Broihier, 2011). The full-length Gbb precursor associates preferentially with the extracellular domain of CRIMPY to regulate synaptic development, before Gbb is secreted from the motorneuron terminal(James and Broihier, 2011). Similarly, Crim1 binds to and antagonistically modulates the processing of pre- BMPs and the secretion of mature BMPs in COS7 cells (Wilkinson et al., 2003). These findings highlight the fact that the function of Crim1 in organogenesis is context-dependent, and that the timing and site of Crim1 expression, coupled with that of the multitude of growth factors it can potentially interact with, will influence its biological function.
Looking to the future, investigations into Crim1 function during development will enable us to probe the mechanisms underlying a variety of pathological disorders. For instance, the adult myocardium has limited regenerative capacity. New technologies have made the generation of cardiomyocytes from induced stem cells a possibility, but it remains essential to clarify the growth factors and signalling molecules germane to this process to make this feasible. BMPs, for instance, regulate stem cell renewal and differentiation into cardiomyocytes, and cooperate with other
233 signalling pathways to further modulate gene expression of transcription factors(Varga and Wrana, 2005; van Wijk et al., 2007).As Crim1 can regulate BMP processing, investigating the intersection between Crim1 biology and BMP signalling during the generation of cardiomyocytes in vitro could be a valuable approach. Similarly, research on the reactivation of the adult epicardium following myocardial damage has increased in recent years, and the modulation of growth factors such as BMPs, TGFβ, VEGF and PDGF by Crim1 is an attractive avenue of research that remains to be explored. Moreover, as myocardial injury stimulates epicardial cells to give rise to fibroblasts and smooth muscle cells(Limana et al., 2011; Zhou et al., 2011; Duan et al., 2012; Huang et al., 2012; van Wijk et al., 2012), our recent observation that, in the absence of Crim1, the number of cardiac fibroblasts is reduced(Iyer et al., 2016), is another step towards deciphering its role not only in lineage specification but also potential therapeutic interventions to improve cardiac performance after damage.
With relation to other pathological disorders, the analysis of serum from chronic heart failure patients shows an increase in CRIM1 levels, along with an increase in secreted factors involved in fibrosis, indicating a positive correlation between CRIM1 and pro-fibrotic activity (Eleuteri et al., 2014), though whether these increases are reflected within cardiac tissue is not clear. Moreover, CRIM1 is expressed at higher levels in drug-resistant leukaemia cells, implicating it as a potential drug resistance marker(Prenkert et al., 2010). The intronic regions of CRIM1 and ZEB2have been demonstrated to be downregulated in breast cancer epithelial cells(Kim et al., 2015), andCRIM1 has also been suggested to be a target of the Hippo pathway, and to be overexpressed in gastric cancer tissues(Lim et al., 2014). These tantalising vignettes into the role of CRIM1 in cancer biology and disease provide a platform on which to further investigate the role of this gene in pathological conditions. Indeed, studies from development have illustrated that Crim1 may act at the nexus of many critical signalling pathways, and so the manipulation of Crim1 expression may provide a parsimonious mechanism by which cellular functions such as proliferation, differentiation and repair can be efficiently manipulated following injury and tumorigenesis.
Acknowledgements This work was funded by grants from the National Health and Medical Research Council to MP (1057751) and DJP (631658). MP holds an Australian Research Council Future Fellowship (FT120100170). SI was supported by a UQ Research Scholarship. We gratefully acknowledge Dr. Diane Maresco-Pennisi for reading preliminary drafts of the review.
234 References
Charnock-Jones D.S., Sharkey A.M., Boocock C.A., Ahmed A., Plevin R., Ferrara N. and Smith S.K. (1994). Vascular endothelial growth factor receptor localization and activation in human trophoblast and choriocarcinoma cells. Biol. Reprod. 51, 524-530. Chen H., Shi S., Acosta L., Li W., Lu J., Bao S., Chen Z., Yang Z., Schneider M.D., Chien K.R., Conway S.J., Yoder M.C., Haneline L.S., Franco D. and Shou W. (2004). Bmp10 is essential for maintaining cardiac growth during murine cardiogenesis. Development 131, 2219-2231. Chiu H.S., York J.P., Wilkinson L., Zhang P., Little M.H. and Pennisi D.J. (2012). Production of a mouse line with a conditional crim1 mutant allele. Genesis 50, 711-716. Cossette S. and Misra R. (2011). The identification of different endothelial cell populations within the mouse proepicardium. Dev. Dyn. 240, 2344-2353. Dettman R.W., Denetclaw W., Jr., Ordahl C.P. and Bristow J. (1998). Common epicardial origin of coronary vascular smooth muscle, perivascular fibroblasts, and intermyocardial fibroblasts in the avian heart. Dev. Biol. 193, 169–181. Dokic D. and Dettman R.W. (2006). Vcam-1 inhibits tgfbeta stimulated epithelial-mesenchymal transformation by modulating rho activity and stabilizing intercellular adhesion in epicardial mesothelial cells. Dev. Biol. 299, 489-504. Duan J., Gherghe C., Liu D., Hamlett E., Srikantha L., Rodgers L., Regan J.N., Rojas M., Willis M., Leask A., Majesky M. and Deb A. (2012). Wnt1/βcatenin injury response activates the epicardium and cardiac fibroblasts to promote cardiac repair. EMBO J. 31, 429-442. Eleuteri E., Di Stefano A., Vallese D., Gnemmi I., Pitruzzella A., Tarro Genta F., Delle Donne L., Cappello F., Ricciardolo F.L. and Giannuzzi P. (2014). Fibrosis markers and crim1 increase in chronic heart failure of increasing severity. Biomarkers 19, 214-221. Fan J., Ponferrada V.G., Sato T., Vemaraju S., Fruttiger M., Gerhardt H., Ferrara N. and Lang R.A. (2014). Crim1 maintains retinal vascular stability during development by regulating endothelial cell vegfa autocrine signaling. Development 141, 448-459. Fung W.Y., Fat K.F., Eng C.K. and Lau C.K. (2007). Crm-1 facilitates bmp signaling to control body size in caenorhabditis elegans. Dev. Biol. 311, 95-105. Garcia-Martinez V. and Schoenwolf G.C. (1993). Primitive-streak origin of the cardiovascular system in avian embryos. Dev. Biol. 159, 706-719. Georgas K., Bowles J., Yamada T., Koopman P. and Little M.H. (2000). Characterisation of crim1 expression in the developing mouse urogenital tract reveals a sexually dimorphic gonadal expression pattern. Dev. Dyn. 219, 582-587. Gittenberger-de Groot A.C., Vrancken Peeters M.P., Mentink M.M., Gourdie R.G. and Poelmann R.E. (1998). Epicardium-derived cells contribute a novel population to the myocardial wall and the atrioventricular cushions. Circ. Res. 82, 1043-1052. Glienke J., Sturz A., Menrad A. and Thierauch K.H. (2002). Crim1 is involved in endothelial cell capillary formation in vitro and is expressed in blood vessels in vivo. Mech. Dev. 119, 165- 175. Godin R.E., Robertson E.J. and Dudley A.T. (1999). Role of bmp family members during kidney development. Int. J. Dev. Biol. 43, 405-411.
235 Goldman D.C., Donley N. and Christian J.L. (2009). Genetic interaction between bmp2 and bmp4 reveals shared functions during multiple aspects of mouse organogenesis. Mechanisms of Development 126, 117-127. Groppe J., Greenwald J., Wiater E., Rodriguez-Leon J., Economides A.N., Kwiatkowski W., Affolter M., Vale W.W., Belmonte J.C. and Choe S. (2002). Structural basis of bmp signalling inhibition by the cystine knot protein noggin. Nature 420, 636-642. Huang G.N., Thatcher J.E., McAnally J., Kong Y., Qi X., Tan W., DiMaio J.M., Amatruda J.F., Gerard R.D., Hill J.A., Bassel-Duby R. and Olson E.N. (2012). C/ebp transcription factors mediate epicardial activation during heart development and injury. Science 338, 1599-1603. Ishii Y., Langberg J.D., Hurtado R., Lee S. and Mikawa T. (2007). Induction of proepicardial marker gene expression by the liver bud. Development 134, 3627–3637. Ishii Y., Garriock R.J., Navetta A.M., Coughlin L.E. and Mikawa T. (2010). Bmp signals promote proepicardial protrusion necessary for recruitment of coronary vessel and epicardial progenitors to the heart. Dev. Cell 19, 307-316. Iyer S., Chou F.Y., Wang R., Chiu H.S., Raju V.K.S., Little M.H., Thomas W.G., Piper M. and Pennisi D.J. (2016). Crim1 has cell-autonomous and paracrine roles during embryonic heart development. Sci. Rep. 6, 19832. James R.E. and Broihier H.T. (2011). Crimpy inhibits the bmp homolog gbb in motoneurons to enable proper growth control at the drosophila neuromuscular junction. Development 138, 3273-3286. Jiao K., Kulessa H., Tompkins K., Zhou Y., Batts L., Baldwin H.S. and Hogan B.L. (2003). An essential role of bmp4 in the atrioventricular septation of the mouse heart. Genes Dev. 17, 2362-2367. Jones C.M., Armes N. and Smith J.C. (1996). Signalling by tgf-β family members: Short-range effects of xnr-2 and bmp-4 contrast with the long-range effects of activin. Curr. Biol. 6, 1468-1475. Kang J., Gu Y., Li P., Johnson B.L., Sucov H.M. and Thomas P.S. (2008). Pdgf-a as an epicardial mitogen during heart development. Dev. Dyn. 237, 692-701. Kang J.O. and Sucov H.M. (2005). Convergent proliferative response and divergent morphogenic pathways induced by epicardial and endocardial signaling in fetal heart development. Mech. Dev. 122, 57-65. Kim R.Y., Robertson E.J. and Solloway M.J. (2001). Bmp6 and bmp7 are required for cushion formation and septation in the developing mouse heart. Dev. Biol. 235, 449-466. Kim S.W., Fishilevich E., Arango-Argoty G., Lin Y., Liu G., Li Z., Monaghan A.P., Nichols M. and John B. (2015). Genome-wide transcript profiling reveals novel breast cancer-associated intronic sense rnas. PLoS One 10, e0120296. Kinna G., Kolle G., Carter A., Key B., Lieschke G.J., Perkins A. and Little M.H. (2006). Knockdown of zebrafish crim1 results in a bent tail phenotype with defects in somite and vascular development. Mech. Dev. 123, 277-287. Kolle G., Jansen A., Yamada T. and Little M. (2003). In ovo electroporation of crim1 in the developing chick spinal cord. Dev. Dyn. 226, 107-111.
236 Kolle G., Georgas K., Holmes G.P., Little M.H. and Yamada T. (2000). Crim1, a novel gene encoding a cysteine-rich repeat protein, is developmentally regulated and implicated in vertebrate cns development and organogenesis. Mech. Dev. 90, 181-193. Larraín J., Bachiller D., Lu B., Agius E., Piccolo S. and De Robertis E.M. (2000). Bmp-binding modules in chordin: A model for signalling regulation in the extracellular space. Development 127, 821-830. Leighton P.A., Mitchell K.J., Goodrich L.V., Lu X., Pinson K., Scherz P., Skarnes W.C. and Tessier-Lavigne M. (2001). Defining brain wiring patterns and mechanisms through gene trapping in mice. Nature 410, 174-179. Li P., Cavellero S., Gu Y., Chen T.H.P., Hughes J., Bassim Hassan A., Bruning J.C., Pashmforoush M. and Sucov H.M. (2011). Igf signaling directs ventricular cardiomyocyte proliferation during embyronic heart development. Development 138, 1795-1805. Lim B., Park J.L., Kim H.J., Park Y.K., Kim J.H., Sohn H.A., Noh S.M., Song K.S., Kim W.H., Kim Y.S. and Kim S.Y. (2014). Integrative genomics analysis reveals the multilevel dysregulation and oncogenic characteristics of tead4 in gastric cancer. Carcinogenesis 35, 1020-1027. Limana F., Capogrossi M.C. and Germani A. (2011). The epicardium in cardiac repair: From the stem cell view. Pharmacol. Therapuet. 129, 82-96. Liu W., Selever J., Wang D., Lu M.F., Moses K.A., Schwartz R.J. and Martin J.F. (2004). Bmp4 signaling is required for outflow-tract septation and branchial-arch artery remodeling. PNAS 101, 4489-4494. Lovicu F.J., Kolle G., Yamada T., Little M.H. and McAvoy J.W. (2000). Expression of crim1 during murine ocular development. Mech. Dev. 94, 261-265. Lu J., Landerholm T.E., Wei J.S., Dong X.R., Wu S.P., Liu X., Nagata K., Inagaki M. and Majesky M.W. (2001). Coronary smooth muscle differentiation from proepicardial cells requires rhoa-mediated actin reorganization and p160 rho-kinase activity. Dev. Biol. 240, 404-418. Manasek F.J. (1969). Embryonic development of the heart. Ii. Formation of the epicardium. JEEM 22, 333-348. Männer J. (1993). Experimental study on the formation of the epicardium in chick embryos. Anat. Embryol. 187, 281-289. Männer J. (1999). Does the subepicardial mesenchyme contribute myocardioblasts to the myocardium of the chick embryo heart? A quail-chick chimera study tracing the fate of the epicardial primordium. Anat. Rec. 255, 212-226. Mellgren A.M., Smith C.L., Olsen G.S., Eskiocak B., Zhou B., Kazi M.N., Ruiz F.R., Pu W.T. and Tallquist M.D. (2008). Platelet-derived growth factor receptor beta signaling is required for efficient epicardial cell migration and development of two distinct coronary vascular smooth muscle cell populations. Circ. Res. 103, 1393-1401. Mikawa T. and Gourdie R.G. (1996). Pericardial mesoderm generates a population of coronary smooth muscle cells migrating into the heart along with ingrowth of the epicardial organ. Dev. Biol. 174, 221-232. Nahirney P.C., Mikawa T. and Fischman D.A. (2003). Evidence for an extracellular matrix bridge guiding proepicardial cell migration to the myocardium of chick embryos. Dev. Dyn. 227, 511-523.
237 Neuhaus H., Rosen V. and Thies R.S. (1999). Heart specific expression of mouse bmp-10 a novel member of the tgf-beta superfamily. Mech. Dev. 80, 181-184. Nyström J., Hultenby K., Ek S., Sjölund J., Axelson H., Jirström K., Saleem M.A., Nilsson K. and Johansson M.E. (2009). Crim1 is localized to the podocyte filtration slit diaphragm of the adult human kidney. Neph. Dial. Transplant. 24, 2038-2044. Olivey H.E., Mundell N.A., Austin A.F. and Barnett J.V. (2006). Transforming growth factor-β stimulates epithelial-mesenchymal transformation in the proepicardium. Dev. Dyn. 235, 50- 59. Pennisi D.J., Kinna G., Chiu H.S., Simmons D.G., Wilkinson L. and Little M.H. (2012). Crim1 has an essential role in glycogen trophoblast cell and sinusoidal-trophoblast giant cell development in the placenta. Placenta 33, 175-182. Pennisi D.J., Wilkinson L., Kolle G., Sohaskey M.L., Gillinder K., Piper M.J., McAvoy J.W., Lovicu F.J. and Little M.H. (2007). Crim1kst264/kst264 mice display a disruption of the crim1 gene resulting in perinatal lethality with defects in multiple organ systems. Dev. Dyn. 236, 502–511. Pérez-Pomares J.M., Carmona R., González-Iriarte M., Atencia G., Wessels A. and Muñoz-Chápuli R. (2002). Origin of coronary endothelial cells from epicardial mesothelium in avian embryos. Int. J. Dev. Biol. 46, 1005-1013. Phua Y.L., Martel N., Pennisi D.J., Little M.H. and Wilkinson L. (2012). Distinct sites of renal fibrosis in crim1 mutant mice arise from multiple cellular origins. J. Pathol. 229, 685-696. Poelmann R.E., Gittenberger-de Groot A.C., Mentink M.M., Bökenkamp R. and Hogers B. (1993). Development of the cardiac coronary vascular endothelium, studied with antiendothelial antibodies, in chicken-quail chimeras. Circ. Res. 73, 559-568. Ponferrada V.G., Fan J., Vallance J.E., Hu S., Mamedova A., Rankin S.A., Kofron M., Zorn A.M., Hegde R.S. and Lang R.A. (2012). Crim1 complexes with β-catenin and cadherins, stabilizes cell-cell junctions and is critical for neural morphogenesis. PLoS One 7, e32635. Prenkert M., Uggla B., Tidefelt U. and Strid H. (2010). Crim1 is expressed at higher levels in drug- resistant than in drug-sensitive myeloid leukemia hl60 cells. Anticancer Res 30, 4157-4161. Randhawa R. and Cohen P. (2005). The role of the insulin-like growth factor system in prenatal growth. Mol. Genet. Metab. 86, 84-90. Red-Horse K., Ueno H., Weissman I.L. and Krasnow M.A. (2010). Coronary arteries form by developmental reprogramming of venous cells. Nature 464, 549-553. Sánchez N.S. and Barnett J.V. (2012). Tgfbeta and bmp-2 regulate epicardial cell invasion via tgfbetar3 activation of the par6/smurf1/rhoa pathway. Cell Signal. 24, 539-548. Schoenwolf G.C. and Garcia-Martinez V. (1995). Primitive-streak origin and state of commitment of cells of the cardiovascular system in avian and mammalian embryos. Cell. Mol. Biol. Res. 41, 233-240. Serluca F.C. (2008). Development of the proepicardial organ in the zebrafish. Dev. Biol. 315, 18- 27. Simon M., Grone H.J., Johren O., Kullmer J., Plate K.H., Risau W. and Fuchs E. (1995). Expression of vascular endothelial growth factor and its receptors in human renal ontogenesis and in adult kidney. Am. J. Physiol. 268, F240-250.
238 Smith C.L., Baek S.T., Sung C.Y. and Tallquist M.D. (2011). Epicardial-derived cell epithelial-to- mesenchymal transition and fate specification require pdgf receptor signaling. Circ. Res. 108, e15-e26. Sucov H.M., Gu Y., Thomas S., Li P. and Pashmforoush M. (2009). Epicardial control of myocardial proliferation and morphogenesis. Pediat. Cardiol. 30, 617-625. Uchimura T., Komatsu Y., Tanaka M., McCann K.L. and Mishina Y. (2009). Bmp2 and bmp4 genetically interact to support multiple aspects of mouse development including functional heart development. Genesis 47, 374-384. Umulis D., O'Connor M.B. and Blair S.S. (2009). The extracellular regulation of bone morphogenetic protein signaling. Development 136, 3715-3728. Van den Akker N.M., Winkel L.C., Nisancioglu M.H., Maas S., Wisse L.J., Armulik A., Poelmann R.E., Lie-Venema H., Betsholtz C. and Gittenberger-de Groot A.C. (2008). Pdgf-b signaling is important for murine cardiac development: Its role in developing atrioventricular valves, coronaries, and cardiac innervation. Dev. Dyn. 237, 494-503. van Wijk B., Moorman A.F. and van den Hoff M.J. (2007). Role of bone morphogenetic proteins in cardiac differentiation. Cardiovasc. Res. 74, 244-255. van Wijk B., Gunst Q.D., Moorman A.F.M. and van den Hoff M.J.B. (2012). Cardiac regeneration from activated epicardium. PLoS One 7, e44692. Varga A.C. and Wrana J.L. (2005). The disparate role of bmp in stem cell biology. Oncogene 24, 5713-5721. Virágh S. and Challice C.E. (1981). The origin of the epicardium and the embryonic myocardial circulation in the mouse. Anat. Rec. 201, 157-168. Virágh S., Gittenberger-de Groot A.C., Poelmann R.E. and Kálmán F. (1993). Early development of quail heart epicardium and associated vascular and glandular structures. Anat.Embryol. 188, 381-393. Waldo K., Zdanowicz M., Burch J., Kumiski D.H., Stadt H.A., Godt R.E., Creazzo T.L. and Kirby M.L. (1999). A novel role for cardiac neural crest in heart development. J. Clin. Invest. 103, 1499-1507. Waldo K.L., Kumiski D.H., Wallis K.T., Stadt H.A., Hutson M.R., Platt D.H. and Kirby M.L. (2001). Conotruncal myocardium arises from a secondary heart field. Development 128, 3179-3188. Wilkinson L., Kolle G., Wen D., Piper M., Scott J. and Little M. (2003). Crim1 regulates the rate of processing and delivery of bone morphogenetic proteins to the cell surface. J. Biol. Chem. 278, 34181-34188. Wilkinson L., Gilbert T., Kinna G., Ruta L.A., Pennisi D., Kett M. and Little M.H. (2007). Crim1kst264/kst264 mice implicate crim1 in the regulation of vascular endothelial growth factor-a activity during glomerular vascular development. J. Amer. Soc. Nephrol. 18, 1697– 1708. Wilkinson L., Gilbert T., Sipos A., Toma I., Pennisi D.J., Peti-Peterdi J. and Little M.H. (2009). Loss of renal microvascular integrity in postnatal crim1 hypomorphic transgenic mice. Kidney Int. 76, 1161-1171.
239 Wu B., Zhang Z., Lui W., Chen X., Wang Y., Chamberlain A., Moreno-Rodriquez R.A., Markwald R.R., O’Rourke B.P., Sharp D.J., Zheng D., Lenz J., Baldwin H.S., Chang C.-P. and Zhou B. (2012). Endocardial cells form the coronary arteries by angiogenesis through myocardial- endocardial vegf signaling. Cell 151, 1083-1096. Xu, J., S. Lamouille and R. Derynck (2009). TGF-β-induced epithelial to mesenchymal transition. Cell Res. 19, 156-172. Yutzey K.E. and Kirby M.L. (2002). Wherefore heart thou? Embryonic origins of cardiogenic mesoderm. Dev. Dyn. 223, 307-320. Zhang J.L., Patterson L.J., Qiu L.Y., Graziussi D., Sebald W. and Hammerschmidt M. (2010). Binding between crossveinless-2 and chordin von willebrand factor type c domains promotes bmp signaling by blocking chordin activity. PLoS One 5, e12846. Zhang Y., Fan J., Ho J.W.K., Hu T., Kneeland S.C., Fan X., Xi Q., Sellarole M.A., de Vries W.N., Lu W., Lachke S.A., Lang R.A., John S.W.M. and Maas R.L. (2016). Crim1 regulates integrin signaling in murine lens development. Development 143, 356-366. Zhou B., Honor L.B., He H., Ma Q., Oh J.H., Butterfield C., Lin R.Z., Melero-Martin J.M., Dolmatova E., Duffy H.S., Gise A., Zhou P., Hu Y.W., Wang G., Zhang B., Wang L., Hall J.L., Moses M.A., McGowan F.X. and Pu W.T. (2011). Adult mouse epicardium modulates myocardial injury by secreting paracrine factors. J. Clin. Invest. 121, 1894-1904. Zimmerman L.B., De Jesus-Escobar J.M. and Harland R.M. (1996). The spemann organizer signal noggin binds and inactivates bone morphogenetic protein 4. Cell 86, 599-606.
240