UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The pathogenesis of alopecia areata Wang, Eddy Hsi Chun 2014

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
24-ubc_2015_february_wang_eddyhsichun.pdf [ 7.66MB ]
Metadata
JSON: 24-1.0167061.json
JSON-LD: 24-1.0167061-ld.json
RDF/XML (Pretty): 24-1.0167061-rdf.xml
RDF/JSON: 24-1.0167061-rdf.json
Turtle: 24-1.0167061-turtle.txt
N-Triples: 24-1.0167061-rdf-ntriples.txt
Original Record: 24-1.0167061-source.json
Full Text
24-1.0167061-fulltext.txt
Citation
24-1.0167061.ris

Full Text

THE PATHOGENESIS OF ALOPECIA AREATA by  Eddy Hsi Chun Wang  B.Sc., Simon Fraser University, 2008  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES (Experimental Medicine)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)  December 2014  © Eddy Hsi Chun Wang, 2014 Abstract  The development of a hair loss disease, alopecia areata (AA), is believed to be associated with the initiation of an autoimmune response triggered by the activation of cytotoxic T-cells (CTLs) by unidentified autoantigen epitopes.  The long-term sequelae of AA also have not received investigation beyond clinical observations. I hypothesized that CTLs in AA can be activated by autoantigen epitopes derived from hair follicle (HF) cells and activated CTLs may affect the viability of HF keratinocytes. Furthermore, I hypothesized the development of AA can have adverse effects on heart health and induce apoptosis of cardiomyocytes mediated in part by stress hormones. To test these hypotheses, I stimulated AA human peripheral blood mononuclear cells (PBMCs) and mouse LNCs with HF autoantigen epitope peptides. I showed that trichohyalin (TCHH) peptides induced higher frequencies of AA PBMC activation and led to keratinocyte apoptosis, while cytokeratin 16 (KRT16) peptides activate more AA mouse LNCs; both indicating the importance of keratinocyte autoantigens in AA pathogenesis. We showed that AA mice displayed significantly heavier hearts and collagen deposition in hearts compared to controls. Exposure of heart tissues to stress hormone adrenocorticotrophic hormone (ACTH) resulted in differential expression of interleukin-18 (Il18) genes and increased secretion of cardiac disease marker cardic troponin-I (cTnI). AA patients showed highest levels of cTnI compared to androgenetic alopecia (AGA) and no-hair-loss (NHL) groups. Culturing of cardiomyocytes with plasma from AA subjects with higher levels of cTnI also resulted in higher levels of apoptosis compared to cultures with plasma expressing low levels of cTnI; suggesting presence of harmful factors in ii  the plasma. Additionally, to prove AA is a cell-mediated disease, we established a new mouse model of AA via injecting naïve C3H/HeJ mice with cultured LNCs isolated from spontaneously affected AA mice. In conclusion, AA pathogenesis is associated with higher frequencies of PBMC or LNC activation upon keratinocyte antigen epitope challenge and AA development resulted in changes in gene expression and heart morphology in mice and heart tissue remodeling markers in humans.  iii  Preface  This dissertation is the original intellectual property of Eddy Hsi Chun Wang where all experimental designs were conceived by both me and Dr. Kevin J. McElwee. Unless otherwise stated, I performed all the experiments, analysis of the data and preparation of the manuscripts. Dr. Kevin McElwee reviewed and revised all written works and the content of the thesis was approved by my supervisory committee, Drs, Aziz Ghahary, Jan P. Dutz and Megan K. Levings. Part of Chapter 1 has been published [Wang E, McElwee KJ. Etiopathogenesis of alopecia areata: Why do our patients get it? Dermatol Ther. 2011 May-Jun;24(3):337-47]. Figure 1.1 in Chapter 1 was obtained from a previously published article that I co-authored  [Alkhalifah A, Alsantali A, Wang E, McElwee KJ, Shapiro J. Alopecia areata update: part I. Clinical picture, histopathology, and pathogenesis. J Am Acad Dermatol. 2010 Feb;62(2):177-88, quiz 189-90].  Permission to reuse this figure was obtained from the publisher. A version of Chapter 2 is in preparation [Wang E, Khosravi-Maharlooei M, Jalili R, Yu R, Ghahary A, Shapiro J, McElwee KJ. Transfer of alopecia areata to C3H/HeJ mice using cultured lymph node derived cells]. I developed the protocol for induction of alopecia areata via culture cell transfer with the guidance of Dr. Kevin McElwee who also isolated the lymph nodes and performed the cell injections. Mohsen Khosravi-Maharlooei and Dr. Reza Jalili from Dr. Aziz Ghahary’s laboratory performed most of the flow cytometry analysis and iv  part of the histology. The research article was written by me and edited by Dr. Aziz Ghahary’s research group and Dr. Kevin McElwee. The detailed protocol in Appendix A is being submitted to a methodology journal [Wang E, McElwee KJ. The development of cultured cell transfer technique to induce AA development in C3H/HeJ mice]. A version of Chapter 3 is in preparation [Wang E, Breitkopf T, Akhoundsadegh N, Leung G, Wang X, Shi FT, Dutz JP, Shapiro J, McElwee KJ. The identification of autoantigen epitopes in alopecia areata.]. Dr. Jan Dutz collaborated in the experimental design, designed the preliminary peptide panels, and generated preliminary human ELISpot results. I performed all the rest of the experiments while Trisia Breitkopf, Noushin Akhoundsadegh, Gigi Leung, Dr. Feng Tao Shi and Dr.Xiaojie Wang assisted in the processing and culturing of hair follicle keratinocytes. Dr. Jerry Shapiro provided clinical knowledge and his Hair Clinic, and Vancouver Coastal Health Research Institute (VCHRI) Clinical Research Unit (CRU), provided blood samples used in this research. Dr. Kevin McElwee performed skin-grafting on the mice to induce alopecia areata. All mice used in this research were housed at Jack Bell Research Centre’s Animal Research Facility.  A version of Chapter 4 has been published [Wang E, Chong K, Yu M, Akhoundsadegh N, Granville DJ, Shapiro J, McElwee KJ. Development of autoimmune hair loss disease alopecia areata is associated with cardiac dysfunction in C3H/HeJ mice. PLoS One. 2013 Apr 26;8(4):e62935]. Preliminary qPCR was obtained by Dr. Mei Yu and Katie Chong. Noushin Akhoundsadegh assisted in performing mouse ELISA analysis. Drs. David Granville and Jerry Shapiro provided experimental and clinical advice.  v  All experiments in Chapter 5 was performed by me with assistance from summer students.  Human ethics was approved by UBC Clinical Research Ethics Board. Certificate number: H09-00649. Questionnaire forms are included in Appendix B and C. Animal research protocols were approved by UBC Department of Dermatology & Skin Science and Animal Care Committee. Protocol numbers: A10-0166, A10-0357. Biosafety certificates: B09-0172, B14-0066.   vi  Table of Contents  Abstract .................................................................................................................................... ii Preface ..................................................................................................................................... iv Table of Contents .................................................................................................................. vii List of Tables ........................................................................................................................ xiv List of Figures .........................................................................................................................xv List of Abbreviations ......................................................................................................... xviii Acknowledgements ............................................................................................................ xxiii Dedication .............................................................................................................................xxv Chapter 1: Introduction ..........................................................................................................1 1.1 Alopecia Areata – A Brief History ........................................................................... 1 1.2 Epidemiology and Clinical Characteristics ............................................................... 3 1.3 Pathology .................................................................................................................. 6 1.4 Current Hypotheses of AA Pathogenesis ................................................................ 10 1.4.1 Autoantibody activities ....................................................................................... 12 1.4.2 Immune cell activities ......................................................................................... 13 1.4.3 Collapse of hair follicle immune privilege ......................................................... 14 1.4.4 Abnormal antigen presentation ........................................................................... 15 1.4.5 Association with atopy and other autoimmune diseases..................................... 17 1.4.6 Potential contributing factors to AA ................................................................... 18 1.4.7 Genetic susceptibility of AA ............................................................................... 18 vii  1.4.8 Involvement of stress and psychiatric problems ................................................. 19 1.4.9 Diet ...................................................................................................................... 21 1.4.10 Other factors........................................................................................................ 22 1.5 Treatments............................................................................................................... 23 1.6 Outstanding Questions Regarding the Pathogenesis of Alopecia Areata ............... 25 1.6.1 Are animal models of AA satisfactory reflections of AA in humans and can the disease models be improved? ......................................................................................... 25 1.6.2 Where and what is the target of T-cell attack in AA affected hair follicles? ...... 27 1.6.3 Is AA really just a dermatological disorder? ...................................................... 29 1.7 Thesis Rationale, Hypothesis and Specific Aims ................................................... 29 Chapter 2: Transfer of Alopecia Areata to C3H/HeJ Mice via Cultured Cell Transfer: A Novel Animal Model to Study Human Alopecia Areata ................................................33 2.1 Introduction ............................................................................................................. 33 2.2 Materials and Methods ............................................................................................ 35 2.2.1 Mice .................................................................................................................... 35 2.2.2 Single cell suspension ......................................................................................... 35 2.2.3 In vitro stimulation and expansion of lymph node cells (LNCs) ........................ 35 2.2.4 Transfer of cultured LNCs into C3H/HeJ mice .................................................. 37 2.2.5 Labelling of the cultured LNC and fluorescent microscopy ............................... 37 2.2.6 Histology and image analysis ............................................................................. 38 2.2.7 Immunohistochemistry ....................................................................................... 39 2.2.8 Flow cytometry ................................................................................................... 40 2.3 Results ..................................................................................................................... 43 viii  2.3.1 Addition of IL7 and IL15 resulted in more rapid LNC expansion compared to IL2 alone ......................................................................................................................... 43 2.3.2 Cultured AA skin-draining lymph node cells induced AA development in healthy C3H/HeJ mice .................................................................................................... 45 2.3.3 The pattern of AA development was consistent and similar to the skin-grafting method ............................................................................................................................ 47 2.3.4 Cultured AA LNCs expressed similar markers compared to cultured control LNCs ............................................................................................................................. 49 2.3.5 Mice receiving either cultured AA or control LNCs showed an opposite trend of CD4+ and CD8+ T-cells localization in skin-draining lymph nodes over time ............... 51 2.3.6 Cultured AA LNCs did not directly participate in the inflammation of the hair follicles ............................................................................................................................ 51 2.4 Discussion ............................................................................................................... 54 Chapter 3: Identification of Autoantigen Epitopes in Humans and the C3H/HeJ Mouse Model of Alopecia Areata ......................................................................................................60 3.1 Introduction ............................................................................................................. 60 3.2 Materials and Methods ............................................................................................ 63 3.2.1 Study subject recruitment and blood cell isolation ............................................. 63 3.2.2 Generation of AA mice and isolation of lymph node cells ................................. 64 3.2.3 Subject sample haplotyping ................................................................................ 64 3.2.4 Peptide sequence prediction and synthesis ......................................................... 65 3.2.5 HLA-A*0201 stabilization assay ........................................................................ 71 3.2.6 Human PBMC IFNγ ELISpot assay ................................................................... 71 ix  3.2.7 Mouse LNC IFNγ ELISpot assay ....................................................................... 73 3.2.8 Intracellular cytokine stain .................................................................................. 74 3.2.9 Hair follicle micro-dissection and root sheath keratinocyte culture ................... 74 3.2.10 Apoptosis induction via peptide-activated PBMC conditioned media ............... 75 3.2.11 Human cytokine array on cell culture supernatant from PBMCs stimulated with epitope peptides .............................................................................................................. 76 3.3 Results ..................................................................................................................... 77 3.3.1 In silico designed peptides displayed varying degrees of affinity to HLA-A*0201 ............................................................................................................................. 77 3.3.2 Trichohyalin, tyrosinase and tyrosinase-related-protein 2 peptides induced a higher frequency of CTL activation in AA PBMCs ....................................................... 80 3.3.3 Antigen epitope peptide activation of PBMCs were affected by the extent and duration of AA as well as concurrent treatments ............................................................ 82 3.3.4 IFNγ Intracellular cytokine staining (ICS) confirmed CTL activation after stimulation by specific peptides ...................................................................................... 84 3.3.5 Trichohyalin peptides induced AA PBMCs to secrete pro-apoptotic factors harmful to primary HF keratinocytes .............................................................................. 86 3.3.6 Differential release of inflammatory cytokines into culture media by AA and control PBMCs after trichohyalin peptide stimulation ................................................... 86 3.3.7 Mouse homologues of cytokeratin 16 and melanoma antigen recognized by T cells 1 (MART1) induced higher frequencies of AA LNC activation ............................ 89 3.4 Discussion ............................................................................................................... 91 x  Chapter 4: The Association of Alopecia Areata Development with Cardiac Dysfunction: Evidence in C3H/HeJ Mice .................................................................................................102 4.1 Introduction ........................................................................................................... 102 4.2 Materials and Methods .......................................................................................... 104 4.2.1 Generation of AA Mice and Tissue Collection................................................. 104 4.2.2 RNA extraction, cDNA synthesis, and Quantitative Real-Time PCR (qPCR) . 105 4.2.3 Histology ........................................................................................................... 106 4.2.4 Immunohistochemistry ..................................................................................... 108 4.2.5 Heart Tissue Culture ......................................................................................... 109 4.2.6 Total protein extraction and quantification ....................................................... 109 4.2.7 Cardiac Troponin I, IL18, IL18R1, IL18BP and CASP1 ELISA ..................... 110 4.3 Results ................................................................................................................... 112 4.3.1 Preliminary gene screening showed significant elevation of the Il18 gene in skin and heart tissues of AA mice ........................................................................................ 112 4.3.2 AA mice displayed changes in heart morphology and had significantly heavier heart weights ................................................................................................................. 114 4.3.3 AA mouse hearts exhibited increased collagen deposition ............................... 117 4.3.4 Concentrations of cTnI in heart tissue and plasma were higher in AA-affected mice ........................................................................................................................... 117 4.3.5 AA mouse skin and heart tissues showed significantly higher pro-inflammatory cytokine gene expression .............................................................................................. 120 4.3.6 AA progression correlated to changes in the gene and protein expression of the Il18 family ..................................................................................................................... 122 xi  4.3.7 Localization of IL18 in the atria of AA mouse hearts ...................................... 125 4.3.8 ACTH treatment in tissue culture resulted in differential gene expression in AA and control mouse atria ................................................................................................. 127 4.3.9 ACTH exposure resulted in changes in collagen gene expression in the atria . 130 4.3.10 ACTH exposure resulted in increased release of cTnI by atria tissues............. 130 4.4 Discussion ............................................................................................................. 132 Chapter 5: The Effect of Alopecia Areata and Androgenetic Alopecia on the Expression of Heart Disease Markers in Humans and the Health of Cardiomyocytes ....................139 5.1 Introduction ........................................................................................................... 139 5.2 Materials and Methods .......................................................................................... 140 5.2.1 Study subject recruitment and blood plasma collection ................................... 140 5.2.2 cTnI and CRP ELISA analysis ......................................................................... 141 5.2.3 Human primary cardiomyocyte culture ............................................................ 142 5.2.4 Human primary cardiomyocyte apoptosis assay............................................... 143 5.2.5 Flow cytometry analysis ................................................................................... 144 5.3 Results ................................................................................................................... 144 5.3.1 AA affected patients showed highest level of cTnI and CRP compared to AGA and NHL subjects .......................................................................................................... 144 5.3.2 AA males had highest levels of cTnI while female subjects with any form of hair loss had higher levels of CRP ....................................................................................... 146 5.3.3 Plasma levels of cTnI were higher in younger subjects with AA or AGA while CRP levels were higher in older subjects ..................................................................... 148 xii  5.3.4 Recent onset of AGA resulted in higher cTnI release while CRP levels were higher in patients with chronic AA ............................................................................... 148 5.3.5 Patients with AA currently receiving treatments had lower levels of cTnI ...... 150 5.3.6 AA patients with patchy hair loss had highest levels of cTnI but CRP levels were highest in those with 25-75% hair loss ......................................................................... 150 5.3.7 AA patient plasma samples with higher levels of cTnI could induce higher rates of apoptosis in HCM ..................................................................................................... 152 5.4 Discussion ............................................................................................................. 154 Chapter 6: Major Findings, Conclusions and Future Directions ....................................163 6.1 Major Findings ...................................................................................................... 163 6.2 Conclusions ........................................................................................................... 169 6.3 Future Directions .................................................................................................. 170 Bibliography .........................................................................................................................173 Appendices ............................................................................................................................197 Appendix A Induction of Alopecia Areata in C3H/HeJ Mice via Cultured Cell Transfer 197 A.1 Introduction ....................................................................................................... 197 A.2 General Experimental Design ........................................................................... 197 A.3 Materials and Reagent Setup ............................................................................ 199 A.4 Detailed Procedures .......................................................................................... 202 A.5 Anticipated Results ........................................................................................... 209 Appendix B Questionnaire for Alopecia Areata Patients ................................................. 212 Appendix C Questionnaire for Control Subjects .............................................................. 215 Appendix D Primer Sequences ......................................................................................... 218 xiii  List of Tables  Table 3.1 List of human autoantigen epitope peptide sequences used. .................................. 67 Table 3.2 List of mouse autoantigen epitope peptide sequences used. ................................... 69 Table A.1 Troubleshooting suggestions. .............................................................................. 211 Table D.1 Primer sequences used in chapter 5 ..................................................................... 218  xiv  List of Figures  Figure 1.1 The development of hair loss in AA involves changes in the pattern of hair growth cycling. ...................................................................................................................................... 9 Figure 2.1 LNC size increased and morphology changed as cells proliferated exponentially over six days with the addition of IL7 and IL15. .................................................................... 44 Figure 2.2 Naïve C3H/HeJ mice that received cultured AA LNCs developed AA as early as two weeks................................................................................................................................ 46 Figure 2.3 The pathologic features of AA induced via cell injection showed typical swarming of lymphocytes around HFs. ................................................................................................... 48 Figure 2.4 Both AA and control LNC derived cell cultures after expansion showed elevated numbers of IFNγ producing cells; Treg frequency increased in controls but decreased in AA cultures. ................................................................................................................................... 50 Figure 2.5 Small numbers of CM-DiI labelled LNCs were detected in single-cell suspensions extracted from the skin at the site of injection, peripheral skin and skin-draining lymph nodes, at all three time points. ................................................................................................ 52 Figure 2.6 Injected LNCs could be found localized at the site of injection even after four weeks, but did not participate in the inflammation of HFs in AA lesions. ............................. 53 Figure 3.1 HLA-A*0201 stabilization assay performed on candidate epitope peptides revealed different ability to stabilize the expression of HLA-A*0201 on T2 cells. ............... 79 Figure 3.2 TCHH, TYR and TYRP2 peptides significantly increased frequencies of AA PBMC activation. .................................................................................................................... 81 xv  Figure 3.3 AA patients not currently receiving treatments, have recent onset and more extensive body hair/nail involvements have higher frequencies of PBMC activation while PBMCs stimulated with individual trichohyalin peptides did not show significant difference.................................................................................................................................................. 83 Figure 3.4 Intracellular cytokine stain confirmed CD8+ T-cell specific activation by the synthesized peptides................................................................................................................ 85 Figure 3.5 CM derived from AA PBMCs stimulated with TCHH-G1 peptides can induce apoptosis in HF keratinocytes. ................................................................................................ 87 Figure 3.6 AA PBMCs secreted more IL-13 and RANTES upon trichohyalin peptide simulation while control PBMCs secreted more MIP1β and MMP9. .................................... 88 Figure 3.7 KRT16-G1 and MART peptides significantly increased the frequencies of LNC activation in AA mice compared to the controls. ................................................................... 90 Figure 4.1 Preliminary qPCR gene screening of chronic AA mice compared to the sham-grafted controls. .................................................................................................................... 113 Figure 4.2 Heart weight and heart to body weight ratio in AA and sham-grafted mice. ...... 115 Figure 4.3 AA mice had significantly fewer nuclei in atria tissue compared to sham-grafted mice. ...................................................................................................................................... 116 Figure 4.4 Evaluation of heart collagen deposition and calcinosis in AA and sham-grafted mice. ...................................................................................................................................... 118 Figure 4.5 Significantly higher level of cTnI was detected in the heart tissues of AA mice................................................................................................................................................ 119 Figure 4.6 qPCR analysis of selected genes in chronic stage AA and sham-grafted mice... 121 xvi  Figure 4.7 qPCR and ELISA analysis of Il18 and related genes and protein expression during the onset of AA compared to sham-grafted mice. ................................................................ 124 Figure 4.8 Immunohistochemistry of IL18 in AA and sham-grafted mice. ......................... 126 Figure 4.9 qPCR analysis of atria treated with ACTH for 72 hours. .................................... 129 Figure 4.10 Gene expression of collagens and the release of cardiac troponin from AA and sham-grafted mouse atria in response to ACTH treatment after 72 hours. .......................... 131 Figure 5.1 Plasma cTnI and CRP level was highest in AA patients and lowest in the NHL population. ............................................................................................................................ 145 Figure 5.2 AA males had highest levels of cTnI while females in general had highest levels of CRP. .................................................................................................................................. 147 Figure 5.3 Younger onset of AA and AGA resulted in higher cTnI levels while CRP levels showed an increasing trend with age. ................................................................................... 149 Figure 5.4 Recent onset of AGA resulted in higher cTnI release. ........................................ 149 Figure 5.5 AA patients not currently receiving treatments released higher levels of cTnI. . 151 Figure 5.6 The release of cTnI and CRP in AA patients with different extents of hair loss showed an inverse trend. ....................................................................................................... 151 Figure 5.7 AA patient plasma samples with high levels of cTnI induced early stage apoptosis in primary human cardiomyocytes compared to samples with low levels of cTnI. ............. 153  xvii  List of Abbreviations  2C   SIY-Kb 2C TCR AA   Alopecia areata ACTH   Adrenocorticotrophic hormone Ad   Adrenomedullin AGA   Androgenetic alopecia AIRE   Autoimmune regulator APC   Antigen resenting cell APC (fluorophore) Allophycocyanin AT   Alopecia totalis AU   Alopecia universalis Bcl2   B-cell lymphoma 2 Bclxl   B-cell lymphoma-extra large BDV   Borna disease virus  BIMAS  BioInformatics and Molecular Analysis Section CASP1  Caspase 1 CCR7   C-C chemokine receptor type 7  CFP10   M. tuberculosis Culture filtrate protein CK-MB  Creatin kinase  CM   Conditioned media  CMV   Cytomegalovirus xviii  Col1a1   Collagen Ia1 Col3a1   Collagen IIIa1 Col5a1   Collagen Va1 CORT   Corticosterone CRH-R2  Corticotropin releasing hormone receptor 2 CRP   C-reactive protein CSP   Circumsporozoite protein Cti   Cardiac tropinin I gene CTL   Cytotoxic T lymphocytes cTnI   Cardiac troponin I cTnT   Cardiac troponin T CVD   Cardiovascular disease DCM   Dilated cardiomyopathy  DEBR   Dundee Experimental Bald Rat DHT   Dihydrotestosterone DPCP   Diphenylcyclopropenone EBV   Epstein–Barr virus ECG   Electrocardiogram ECM   Extracellular matrix  ELISA   Enzyme-linked immunosorbent assay ELISpot  Enzyme-linked ImmunoSpot ESR1   Estrogen Receptor 1 Fasl   FAS ligand xix  FBS   Fetal bovine serum FITC   Fluorescein isothiocyanate FLU   Influenza GWAS  Genome-wide association study Gzmb   Granzyme B H&E   Hematoxylin and Eosin  HCM   Human cardiomyocyte HCV   Hepatitis C virus  HF   Hair follicle HIV   Human immunodeficiency virus HLA   Human leukocyte antigen HPA   Hypothalamic-pituitary-adrenal HRT   Hormone replacement therapies ICS   Intra-cellular cytokine stain IFNγ   Interferon-gamma IHC   Immunohistochemistry IL   Interleukin Il18bp   IL18 binding protein Il18r1   IL18 receptor-1 INS   Bovine insulin B chain IP   Immune privilege IRS   Inner root sheath iSkin   Injected skin xx  KRT16  Cytokeratin 16 LNC   Lymph node cell MART1  MelanA/MART1 analog-3  Mel-1   MelanA/MART1 analog-1 Mel-2   MelanA/MART1 analog-2 MFI   Mean fluorescence intensity MHC   Major histocompatibility complex MIP1β   Macrophage inflammatory protein 1-beta Mlik   Myosin light-chain kinase I Mmp   Matrix metalloproteinase MRI   Magnetic resonance imaging  Myh7   β-myosin heavy chain NHL   No hair loss Nppa   Atrial natriuretic factor Nppb   N-terminal pro-brain natriuretic OD   Optical density ORS   Outer root sheath PBMC   Peripheral blood mononuclear cell PE   R-Phycoerythrin PI   Propidium iodide POMC   Pro-opiomelanocortin pSkin   Peripheral skin PTPN22  Protein tyrosine phosphatase, non-receptor type 22 xxi  qPCR   Real-Time Quantitative PCR RANTES  Regulated on activation, normal T cell expressed and secreted SFC   Spot-forming cells SIS   Skin immune system SNP   Single nucleotide polymorphism Spn   Spleen St2   Interleukin 1 receptor-like 1 SV40   Simian virus 40  TB10   M. tuberculosis TB10.3/4 TCHH   Trichohyalin Th   Helper T-cell THB   Tyrosine Hydroxylase isoform B Timp2   Tissue inhibitor of metalloproteinases 2 TNFα   Tumor necrosis factor-alpha Treg   Regulatory T-cell TYR   Tyrosinase TYRP2  Tyrosinase Related Protein 2 ULBP   Cytomegalovirus UL16-binding protein xxii  Acknowledgements  I would like to thank many people who have helped me during my graduate career. In particular, I would like to express my gratitude to my supervisors Dr. Kevin McElwee and Dr. Jerry Shapiro for giving me the opportunity to pursue my graduate study under their mentorship.  Thank you for your all your generous support and encouragement throughout my study and for believing in me. Words cannot express my gratitude for your patience and the life-long lessons you have taught me. I would also like to thank my supervisory committee members Dr. Aziz Ghahary, Dr. Jan Dutz and Dr. Megan Levings as well as lab members in their respective labs for their mentorship, feedback on my work and their willingness to help me with different aspects of my graduate study. I would like to express my appreciation to Dr.Shapiro’s clinical fellows, specifically Dr. Lisa Chan and Dr. Assaf Monselise for taking time to collect blood samples during their clinic hours. I am also grateful for the support from the past and present colleagues of the Hair Research Laboratory who have also become dear friends. Without their willingness to help and morale support, many experiments seemed impossible. Specifically, I would like to thank Trisia Breitkopf, Noushin Akhoundsadegh and Gigi Leung who helped me with the most time consuming experiments. I would also like to thank the lab members from Dr.Aziz Ghahary’s lab: Mohsen Kohsravi, Reza Jalili and Richard Yu for their collaborative works that contributed to the completion of this thesis. xxiii  Throughout the years, I was honored to be the mentor of multiple summer students. I would like to thank Katrina Woo, Annie Tran, Sandra Kim, Angela Li and Mona Maleki for their hard work on their summer projects. I have learnt a lot by working with them as well. A majority of the stipend was provided by UBC-CIHR Skin Training Research Centre under the supervision of Dr. Harvey Lui who also provided me with academic and career advice. I am extremely grateful for the kind support. Much of the work was supported by grants granted by the National Alopecia Areata Foundation and the Canadian Dermatology Foundation. This work would also be impossible without my childhood friends, Albert Su, Alan Wu, John Lin, Kelly Liu and Winnie To. I thank them for their prompt response to my calls for help all these times. They have certainly helped me to get over many obstacles in life. Finally, I would like to express my deepest gratitude to my family for their endless love, support and their firm belief in the decisions I have made. My parents and sister, Amy, William and Sandy have supported me all these years with their love, wisdom and a sense of humour.  Also to my uncle, Stephen Chen, who secretly followed every step of my progress but is still too shy to admit it. xxiv  Dedication  To my parents, Amy and William. xxv  Chapter 1: Introduction  1.1 Alopecia Areata – A Brief History Alopecia areata (AA) is a dermatological disease that involves sudden loss of scalp hair, affecting men, women and children1. The first written record that described the morphology of AA was by Aulus Cornelius Celsus in 30 AD where he noted areas of circular and discrete patches of hair loss that later led to complete baldness, or a different form called “ophiasis” where hair loss is presented with a serpentine appearance2,3. The term “alopecia areata” was not used until 1760 by Sauvages in his "Nosologica Medica" in Lyons, France when it was, and still is, considered as a rare disease with debatable pathogenesis4,5. During the 1800s, the prevailing hypotheses for AA pathogenesis were parasite or fungal infections6,7. This hypothesis was supported by the clinical features of AA where slow expanding lesions resembled that of infections; epidemics of AA were reported to occur in schools and other localized settings as well, much like an infectious disease8-10. However, various attempts to isolate infective organisms and transfer AA via inoculation were unsuccessful6,11,12. The pathological features of AA also resemble that of patients who are affected with syphilis or ringworm, exhibiting sudden, rapid loss of hair in well-defined patches, as well as nail involvement in the secondary stage of syphilis12,13. Some dermatologists of the day suggested increased risk of AA occurred with the development of syphilis which itself was associated with mental stress and endocrine disorders14. More recent studies have raised the possibility of infection by cytomegalovirus infection as an 1  initiator of AA15,16, but further studies in human and animal models by other investigators have failed to support this finding17-20. Another hypothesis for AA pathogenesis at the time was the involvement of the nervous system and association with emotional stress; the neuropathic hypothesis21. This hypothesis was supported by various circumstantial evidence that indicated emotional and physical distress post trauma prior to the onset of AA22. The involvement of the nervous system with AA onset was also circumstantially supported by the anatomic distribution of certain nerves associated with human AA lesions23 as degeneration in neurofibrils in the nerves to the blood vessels and sweat glands was observed in AA patients24. In one animal study, the nerves in the neck of cats were cut and resulted in apparent AA-like symptoms25. However, the hair loss observed in cats with cut nerves were most likely due to scratching as sectioning of human sensory nerves supplying the scalp did not alter hair growth26. Even now, the involvement of the nervous system and neuropeptides in the onset of AA are still being investigated27-29. New hypotheses regarding AA development emerged as more observations were made on the disease. One such hypothesis was the association of nerve irritation due to dental foci of infection with AA30; several rare case reports also appeared to support this hypothesis, even to this day31,32. However, such observations are extremely rare and concordance was not always observed32,33. Additional observations also suggested eye strain as a cause of AA34.  The views on the onset of AA began to shift as evidence suggested involvement of disorders of the thyroid gland, put forward in the early 1900s35. Therefore, the dysregulation of hormones and neuropathic theories as an initiator for AA became predominant at the time. 2  At the same time, it was believed by some that AA could be a result of toxins in the bloodstream that accumulate in hair follicles (HFs), such as thallium acetate (rat poison)36. This hypothesis stemmed from the development of AA-like hair loss patterns in some people along with an AA diagnostic marker, the “exclamation mark” hairs, after injection with thallium acetate12,36-38.  In the later part of 1900s, with the advancement of experimental techniques, the idea that development of AA was due to infiltration of immune cells into HFs gained more traction. Interestingly, the hypothesis of inflammatory cell invasion of HFs causing AA is actually not a novel idea proposed in the past few decades, but rather, was first suggested in studies performed over 120 years ago39. Cellular activities in human AA scalp lesions was recognized (again) in 1958 when the “swarm of bees” infiltration of cells into HFs was (re)observed40,41 and later, immunohistochemical analysis identified multiple T-cell populations within and around the HFs42,43. With the abundance of supporting and circumstantial evidence from recent studies on animal models of AA such as the depletion of immune cells to restore AA44,45 as well as in humans where treatments often involve dampening or redirecting lesional immune cells46,47, the infiltration and activity of immune cells around the HF as the cause of AA became the most compelling hypothesis describing the mechanism of AA pathogenesis. 1.2 Epidemiology and Clinical Characteristics The distribution of AA across both genders and different races shows high variance between studies. In one recent study on North-American-Caucasian AA patients, the ratio of affected females to males is 2.3:1. However, more preadolescent males developed AA while 3  more adolescent females first developed AA; less difference was observed as both genders progressed into adulthood48. In a different, slightly dated, study based on a Turkish population, higher prevalence of AA was observed in males in the general population. Interestingly, early onset of AA was more common in females unlike the North American population49. Concomitant diseases are also more prevalent in females including thyroid diseases, atopy and nail abnormalities. AA males are more likely to report primary family members with history of AA48.  The largest epidemiological study ever performed indicates a 1.7% life time risk of developing AA in the USA50 with the newest update of the same study slightly adjusting the risk to 2.1%51; making AA one of the most common forms of hair loss disease encountered in the dermatology clinic52. The typical clinical presentation of AA is sudden, non-scarring, patchy hair loss, usually in an oval shape53. While the hair loss may be restored, even without any treatment, patients with AA can experience spontaneous exacerbation and reoccurrence of hair loss. There are several different clinical forms of AA and it can develop anywhere on the body; however, more than 90% patients initially develop hair loss on the scalp5. Multiple patches can coalesce to form bigger patches as AA progresses. It is estimated around 5% of patients with AA will progress into the more severe forms; alopecia totalis (AT), the loss of all scalp hair, and alopecia universalis (AU), involving the complete loss of scalp and body hair53,54. Nail abnormalities (pitting and longitudinal ridging) are a relatively common pathological feature associated with AA, observed in 17% of AA patients55.  In all patients suffering from AA, the lesional skin is smooth with well-defined borders adjacent to normally haired areas53. The skin tone of the area with hair loss is usually 4  natural, but may present with a slightly pink tone, and the hair fibers at the border of these patches may have an “exclamation mark” appearance; short, broken hair fibers with a broader distal end compared to the proximal end53. These features are often used as diagnostic features to differentiate between other types of inflammatory hair loss diseases such as cicatricial alopecia, where initial inflammatory skin lesions are later replaced with scar tissues and permanent loss of HFs56. Even though AA is not a life-threatening disease, the psychological burden it imposes onto the patients can be devastating in the current image-oriented society. Such psychosocial burdens are especially prevalent in women and children57-59. A relatively high association of AA with mental health problems, which includes depression and anxiety, has been observed (25.5%)60,61. In a different study on adults, it was also found 34.11% of AA male patients developed depression62. Association of AA with psychiatric problems was also observed in 78% of children63. However, despite the high prevalence of psychiatric disorders found associated with AA, it was reported that most of the patients were experiencing adjustment disorders and generalized anxiety; showing minor repercussions in overall quality of life60. Although association of psychological stress, such as depression, has long been debated as a cause of AA, dysregulation of stress hormones in AA subjects and a mouse model has already been identified64,65.  It is not uncommon that patients with AA may further augment the development of other more life-threatening comorbidities that could reduce life expectancy60. Hyperlipidemia (24.5%), hypertension (21.9%), and anemia (19.6%) is reported in association with AA60 and are known to contribute and modify cardiovascular disease severity as well as mortality66-68. 5  There is a complex association between type 1 diabetes and AA, it has been reported that relatives of AA patients have increased type 1 diabetes while AA patients are more protected69. However recent studies using genome wide association studies in AA patients revealed several risk loci that are known to be involved in type 1 diabetes70. Atopy has long been reported to associate with AA; current studies have reported the association of AA with atopy can be as high as 38.2%60. A recent large scale population study based in Taiwan showed AA patients have significantly increased risk of comorbid atopic dermatitis, allergic rhinitis but no significantly different risk for asthma. In all these cases, AA females had higher risk than males but not statistically significant71. The development of atopic diseases seems to be more prevalent in patients with early onset of AA72. 1.3 Pathology AA is a disease associated with HFs, the severity and duration of AA can result in several different abnormal hair cycle presentations. In the normal hair cycle there are three key stages; the active growth stage (anagen), the regressing stage (catagen) and resting stage (telogen). A controlled shedding stage (exogen) is a moveable event; in normal HFs the hair fibers shed as the new anagen stage begins. Overall hair coverage is maintained by normal cycling between these stages and the replacement of shed hair with new hair at a similar rate73.  AA is essentially a disease involving dysregulated hair growth cycles associated with inflammatory cell infiltration53,74. During the initial development of AA, the HFs are often stuck in a dystrophic anagen stage. Later, AA development involves increased numbers of 6  telogen stage follicles and a state of kenogen where no hair fibers are present in the HF; this is the end state of most HFs in AA affected skin lesions53,56. Histological analysis of AA lesions in both humans and rodent models usually shows a “swarm of bees” infiltration of lymphocytes into the peribulbar space around anagen stage HFs and some penetration of lymphocytes to intrafollicular locations53,75-78. However, the histological presentations can be very different as AA progresses from an early acute stage to a late chronic stage53. Diffuse infiltration of mast cells and eosinophils can be found in AA lesions in all stages, but the exact significance is still unknown79-81. Peribulbar infiltration in the acute phase of AA is mainly comprised of both CD8+ and CD4+ T-cells where a high density of the T-cells indicates active disease progression77. This infiltration of T-cells is believed to be the primary cause of the “dystrophic anagen” state where moderate inflammatory infiltrates affect the production of healthy hair fibers with enough size and integrity; producing weak hair fibers and results in breakage78,82,83. Alternatively, during severe inflammatory attack, HFs can be pushed into telogen prematurely resulting in increased hair shedding78,83. During the development of AA, both events can occur at the same time resulting in hair loss (Figure 1.1a,b)53. As AA progresses, an increase of inflammatory infiltrate is found within and surrounding the HFs. Over time and continued inflammation, miniaturization of HFs may occur and hair cycles can become truncated with rapid cycling of anagen and telogen stages (also known as nanogen HFs). As the disease progresses, more HFs go into a catagen stage which eventually progress into telogen leaving over 50% of HFs in AA skin lesions in this 7  stage78. When the majority of HFs go into a telogen stage, the amount of inflammatory cells actually decreases, as if the HFs are “burnt out”78. In the chronic stages of AA, the HFs are typically forced into a prolonged telogen stage where no new hair fibers are produced (Figure 1.1c); the hair cycle appears to be stalled78,82.  Any residual inflammation at this point is localized at the papillary dermis around miniaturized hair follicles and the number of terminal anagen scalp HFs has decreased to about the same number of vellus hair follicles78. These observations circumstantially suggest that AA progression is driven by inflammation where changes in intra- and peri-follicular inflammatory infiltrates controls the state of HFs; the effect of antigen epitope activated lymphocytes on the viability of HF keratinocytes is investigated in Chapter 3.   8   Figure 1.1 The development of hair loss in AA involves changes in the pattern of hair growth cycling. Moderate inflammatory insult of HFs results in dystrophic anagen and produces weak hair fibers (a). During severe inflammation, HFs can be pushed into premature telogen resulting in brief hair fiber growth followed by rapid shedding (b). In the chronic stage, HFs stay dormant in a prolonged telogen phase, no new hair fibers are produced (c).   9  1.4 Current Hypotheses of AA Pathogenesis Histological studies past and present show the hallmark “swarm of bees” clustering of inflammatory infiltrates around the hair bulbs. This observation led to the hypothesis that the intrafollicular and perifollicular infiltration of inflammatory cells is potentially a result of patient’s immune system mistakenly attack and destroy self-tissues that are not infected by foreign organisms. Closer histological analysis combined with clinical observations suggested AA is a disease of direct attack on differentiating HF cortical keratinocytes by the inflammatory infiltrates, causing dystrophic cycling between the growth and resting stages of HFs84. During this attack, the pigmentation mechanisms within the HF may also be targeted; senile white hairs are often spared at the early stage of AA84.  Genetic analysis has revealed abnormal expression of various genes involved in antigen presentation providing the first evidence suggesting abnormal antigen-presenting and/or the potential for presence of autoantigens in AA85,86. Several HF keratinocyte and melanocyte self-antigens have been proposed to be potentially pathogenic, but the exact antigen epitopes are still debated87,88. Recent studies on AA patient skin lesions with genome-wide association studies (GWAS) also identified polymorphisms in genes associated with activation and proliferation of regulatory T-cells, cytotoxic T lymphocytes (CTLs) and interleukin-270.  Compared to AA skin lesions, there are very few lymphocytes residing within healthy skin. This difference provides circumstantial support for the important role of the immune system in the development of AA. Several animal studies and observation from AA patients receiving treatments, provides more functional evidence in support of a role for inflammation 10  in AA. T-cell depletion in rat models of AA partially restored hair loss44,45; the blockage of activities of certain T-cell clones also prevented disease development (see below)89. In humans, immunomodulatory therapies such as corticosteroids and contact sensitizers improve hair loss albeit with variable effects46,74.  Inflammatory cells that comprise the skin immune system (SIS) are transitory and antigen presenting cells (APCs) usually migrate into secondary lymphoid organs (such as skin-draining lymph nodes) upon antigen challenge to activate antigen specific T-cells90,91. With studies performed on AA affected mice, it has been confirmed that the blockage of lymphocyte homing receptors92,93 or disrupting APC migration to skin-draining lymph nodes, can prevent AA development and modulate disease progression94,95. Even though the AA phenotype is restricted to the skin and its appendages, the development of AA involves the complete immune system as a whole, not just restricted to the SIS. Activities in peripheral immune sites are likely equally important for modulating AA.  Throughout history, many hypotheses have been put forward to explain the pathogenesis of AA; though most of them were later disproved (as highlighted above). Currently, AA is generally accepted as an autoimmune disease initiated by the abnormal interaction of a patient’s immune system with self-antigens in conjunction with environmental factors and genetics that could influence degree of susceptibility to AA, the rate of onset and patterns of hair loss. However, the identity of the primary self-antigen target(s) in the inflammatory cell attack, derived from the HF unit, remains unproven. Therefore, the research data on AA is best described as “consistent with” an autoimmune disease hypothesis21. Nevertheless, an autoimmune disease is still thus far the best explanation for the pathological features observed in AA patients. With animal models, 11  functional evidence has demonstrated the critical role of an active immune system in the AA disease mechanism (see below), whereby AA does not develop if there is a lack of a functional immune system44,45,96-99. 1.4.1 Autoantibody activities The nature, level and specificity of autoantibodies in AA subjects compared to healthy subjects has been a controversial topic as several groups have reported different findings100-102. Early studies reported a high prevalence of AA patients with serum antibodies against thyroid antigens such as thyroid microsomal and thyroglobulin as well as antibodies against smooth muscle antigens103,104, but some studies also reported presence of these antibodies in normal subjects and showed similar levels as AA patients100,105. In addition, B-cell autoantigen tyrosine hydroxylase antibody is elevated in AA patients106. In more recent studies, higher titers of antibodies against multiple antigens unique to HFs, but not the adjacent epidermis or dermis, have been identified107. A majority of these antigens were autoantigens as they reacted with donor’s own serum antibodies and more specifically, directed to HF-specific keratinocytes and melanocytes108,109. An increase of HF-specific IgG antibody concentration in the peripheral blood of AA patients was found. Direct immunofluorescence labeling indicates autoantibodies are localized around the HFs at the edges of the AA lesions110. The specificity of HF antigen specific autoantibodies can vary between different patients111. But recently, autoantibodies isolated from AA patient sera have been most commonly been found to bind to cytokeratin 16 and HF-specific trichohyalin88,112.  In the DEBR rodent model of AA, there is an elevated level of serum IgG against differentiating HF root sheath antigens but not against thyroid and smooth muscle antigens. 12  However, these antibodies were not observed prior to the inflammatory cell infiltration into HFs113.  Further and more significantly, investigations on autoantibodies revealed that AA is not transferrable to the human scalp explants grafted on immune deficient nude mice by injection of patient sera114. These results indicate that AA is less likely a consequence of a primary humoral response. Nevertheless, the presence of autoantibodies implies an active autoimmune response specific against HF antigens and may provide significant clues as to the antigenic targets for T-cells. 1.4.2 Immune cell activities The AA skin lesions in both animal models of AA and human patients both present with follicular inflammatory infiltrates comprised of mostly CD4+ and CD8+ T-cells. Immunohistochemical analysis of T-cell populations revealed intrafollicular infiltration of activated CD8+ T-cells while CD4+ T-cells are almost exclusively found in the perifollicular regions42,43,77. Moreover, it has been found that a failure to recruit CD4+ and CD8+ T-cells, and an increase in regulatory cytokines, leads to resistance to AA development in the mouse model of AA115. Due to the cytotoxic nature of CD8+ T-cells (also known as cytotoxic T lymphocytes; CTLs), their presence in the close vicinity within the HF could easily disrupt the growth of hair. Activated CTLs are known to be able to produce a large variety of harmful/proinflammatory cytokines and pro-apoptotic factors such as interferon-γ (IFNγ), tumor necrosis factor-α (TNFα) and FAS ligand 115-120. These cytokines and mediators all have the potential to disrupt the hair growth cycle, promote degeneration of HFs, and potentially to induce apoptosis of HF cells in AA121,122. 13  In the rodent model of AA, the depletion of CD8+ T-cells successfully inhibited the development of AA44. Compared to partial hair regrowth after depletion of CD4+ T-cells, depletion of CD8+ T-cells resulted in full regrowth; but such effect was soon reversed when the cell depletion regimen was stopped44,45. In addition, the blockage of activities from a specific CD8+/NKG2D+ T-cell population in AA-affected mice with JAK inhibitor prevented disease development89. Injection of skin-draining lymph node cells (LNCs) isolated from AA affected mice intra-dermally into healthy, histocompatible recipients can result in the development of AA (Will be discussed further in Chapter 2). In a different study, the development of localized hair loss exclusively at the site of injection occurred when AA activated, skin-draining lymph node derived CD8+ T-cells were injected subcutaneously while CD4+/CD25- T-cells promoted a systemic hair loss with multiple alopecia patches distant from the injection site98. These functional studies confirm the immune cell-mediated nature of AA development and that CD8+ T-cells are required as promoters of hair follicle disruption and hair loss while CD4+ T-cells likely provide a “helper” role in promoting AA. 1.4.3 Collapse of hair follicle immune privilege A healthy HF is characterized by the absence of immune cells and hence it is considered as an immune privileged (IP) site like the anterior chamber of the eye, the brain and testes123,124. Under normal conditions, the epithelium of the lower HF does not express major histocompatibility complex (MHC) class I and II, but instead expresses high levels of various immunosuppressive molecules such as α-MSH, TGF-β, and, IGF-1125-127. However, in the diseased AA state, there is an increase of MHC-I and II in both humans and the mouse model of AA as well as a decrease in the expression of important immunosuppressive 14  molecules125. Furthermore, expression of adhesion molecules such as ICAM-2 and ELAM-1 are increased in the perivascular and peribulbar area during the advancing stage of AA, indicating higher leukocyte trafficking into the dermis via adhesion molecules on the endothelial cells128. The complete reversal of antigen presenting and immunosuppressive molecule expression in the AA skin lesions provides compelling evidence of collapsing HF IP which provides opportunity for the infiltration of inflammatory lymphocytes and targeting of HF cells in AA patients. However, the loss of IP is not the only event required to initiate AA development. Changes in expression of immunoregulatory genes can be observed in advance of overt hair loss; one example is the significant downregulation of red/IK found in patients before hair loss begins125. This supports the collapse of IP as a contributing factor to AA development129. As deficiencies of IP in HF occur before the onset of AA and inflammatory attack, it is also a strong indication that such deficiency alone is not sufficient to elicit AA. Such an hypothesis is supported by C3H/HeJ AA mouse model induced via skin-graft96. In sham-grafted control groups, the mice experience the same HF damage and induction of MHC-I and II in HF adjacent to the site of injury as in the AA graft, but the sham-grafted mice do not develop AA96,115. This supports the logic that HF IP disruption must occur before AA can develop, but additional triggers are likely required to initiate disease progression. 1.4.4 Abnormal antigen presentation The collapse of IP hypothesis provides a compelling rationale as to how AA may initiate. However, the loss of IP is also a part of natural course of events during the hair growth cycle. During catagen, the rapid regression of HF involves apoptosis and tissue 15  remodeling and can lead to infiltration of a small amount of lymphocytes130-133. Lymphocyte mediated inflammation conventionally requires activation after antigen epitope presentation as well as appropriate co-stimulatory signals134. Therefore, during catagen when HF antigens are exposed to the immune system, the infiltrated lymphocytes have the opportunity to interact with inappropriately presented antigenic peptides derived from within the HFs and elicit an inflammatory response. However, autoreactivity is not an “all or nothing” event, an accumulation of autoimmune responses that surpasses a threshold is needed to observe pathological features of an autoimmune disease135. This is supported by the observation of low level HF autoantibodies found in some healthy people and rodents in the absence of overt AA111,113,136. Professional antigen presenting cells (APCs) like dendritic cells and Langerhans cells are capable of promoting autoimmunity via presenting antigens derived from apoptotic events to lymphocytes90,91,137. Inflammatory events are observed in the skin-draining lymph nodes of mice several weeks in advance of AA development or even lymphocyte infiltration in the skin138. Therefore, even though AA is a disease in the skin, the initial event most likely occurs in the lymph nodes rather than in the skin. Potentially, an AA patient could experience disordered catagen where HF antigen epitopes are inappropriately presented to lymphocytes and eventually elicit overt AA. There have been several studies that aimed to identify crucial autoantigen epitopes that could trigger AA development but the exact T-cell antigen epitopes are still not yet identified87,139. The identification of autoantigen epitopes in mouse and human AA is discussed in Chapter 3. 16  1.4.5 Association with atopy and other autoimmune diseases Epidemiological studies have suggested increased risk of AA development in patients with atopy (allergy) and other autoimmune diseases60,140. Genetic predisposition could be an important factor that results in increased incidence of other autoimmune diseases. Gene alleles coding for higher levels of immune-surveillance, or increased levels of co-stimulation in association with antigen presentation to lymphocytes, could make an individual more susceptible to the development of aberrant immune responses that leads to disease. Potentially, it is also possible that atopy and other autoimmune diseases act as a causal factor to disrupt the immune system and initiate a cascade of events leading to AA onset73.  The higher severity of atopy seems to have an association with increased frequencies of AA140. A recent epidemiological study that compared the comorbidities in AA with healthy controls revealed a significantly higher prevalence of allergic rhinitis (14.3%), thyroid disease (7.2%), and atopic dermatitis (5.0%) in AA patients. Relatively lower prevalence of psoriasis (1.9%) and vitiligo (0.3%) was found in AA patients, but was still significantly higher than the controls71. In a different epidemiological study, the authors used psoriasis patients as controls because it is also a T-cell mediated autoimmune disease and comorbidity profiles have been defined60. In this study, individuals with AA showed significantly higher prevalence of atopy (38.2%) compared to psoriasis patients; other diseases such as dermatitis (35.9%), hyperlipidemia (24.5%), hypertension (21.9%), anemia (19.6%), thyroid disease (14.6%), and diabetes mellitus (11.1%) were also higher than the general population60. 17    A potential interrelationship between AA and the above comorbidities is that the development of one or more of these diseases may share similar high-risk loci, immune cell populations, or cytokine profiles as AA, leading to multiple diseases60,71,141. A recent study using ruxolitinib, a JAK inhibitor used as a treatment for myelofibrosis and rheumatoid arthritis, showed positive effect on AA89. As JAK inhibitors inhibit important cytokines relevant to multiple autoimmune diseases, it strengthens the hypothesis that AA could share similar mechanisms to certain other autoimmune diseases like rheumatoid arthritis. 1.4.6 Potential contributing factors to AA AA is a disease that likely requires the input of multiple variables before overt hair loss becomes apparent. These factors may include, genetics135, stress64,65,142, hormones143, diet144, infectious agents145, vaccination146,147, and various other inputs. Potentially, these factors can each have varying effects on the progression, duration, and patterns of AA; they may also affect how each individual responds to treatments. As humans are a heterogeneous population, different factors may be prevalent in each individual and the risk factors for AA may be dominated by genetics for some people rather than environmental factors. Certain genetic susceptibilities have been identified, but the specific environmental risk factors and their relative contributions to AA are still largely to be determined. 1.4.7 Genetic susceptibility of AA The heritability of AA has been observed in first-degree relatives148, twin studies149 as well as animal models150. Several important genes residing in the human leukocyte antigen (HLA) gene region (HLA-DQB1, HLA-DRB1, HLA-A, HLA-B, HLA-C, NOTCH4, MICA) were implicated to associate with AA development in early candidate-gene association 18  studies; genes residing outside of HLA regions (AIRE and PTPN22) were also found29. These results suggest that abnormal antigen presentation and regulation of autoimmunity could be important for initiation of AA. A genome-wide association study (GWAS) surveyed a large panel of single nucleotide polymorphisms (SNPs) to count the number of alleles with abnormal SNPs in AA patients compared to controls. GWAS identified several genes controlling regulatory T-cells (Treg), CTLA4, IL2RA, IL2/IL21, Eos and several HLA genes, associated with AA70. Genes implicated with a function in HFs were also identified (PRDX5 and STX17)70. Among the genes identified by GWAS, the ULBP (cytomegalovirus UL16-binding protein) gene, which encodes activating ligands of the natural killer cell receptor NKG2D, is a novel susceptibility gene that has not so far been identified in other autoimmune diseases. The exact mechanisms of ULBP function in the development of AA is being investigated. A recent update to the GWAS study also identified IL-13 and KIAA0350/CLEC16A are susceptibility loci for AA and other autoimmune diseases, suggesting shared autoimmune pathways151. 1.4.8 Involvement of stress and psychiatric problems Psychological stress is one of the most commonly cited causes of AA and has been a topic of debate for many years. Various controlled and uncontrolled clinical studies have failed to reach consensus as to whether stress is causal, a consequence, or unrelated to the development of AA152-158. In a more recent study, there was a high prevalence of anxiety and depression in AA patients60. Psychological stress is a variable that is more difficult to quantify as stress is perceived differently in each person and it may exacerbate the progression of certain dermatological disorders while paradoxically benefiting another 19  disease159,160. Overall, current evidence supports the association of AA with increased psychological stress, but maybe at a more molecular level rather than as a direct cause of AA161.  In recent studies, a high prevalence of psychiatric diseases are also recorded in AA patients57,60,161. The prevalence of mental health problems can reach as high as 25.5% in AA patients, but was significantly less than the prevalence in psoriasis patients60. Notably, anxiety (5.0%), depression (2.0%) and patients with at least one type of psychiatric condition (8.1%) were observed in AA patients and were significantly higher than healthy controls161. These results emphasize the possibility of AA being a psychosomatic disease162,163 and proper stress-coping training may be beneficial for AA patients164-166.  In animal studies, the mouse model of AA showed altered hypothalamic-pituitary-adrenal (HPA) activity65. AA mice displayed significantly higher active central and peripheral HPA tone compared to unaffected controls after AA induction. They also experienced a decreased habituation response to chronic psychological stress as well as a blunted systemic HPA response to acute physiological stress65. The brains of AA affected mice also had differential expression of genes related to stress. The gene and protein expression of local HPA hormone receptors such as corticotropin releasing hormone receptor 2 (CRH-R2) are increased in the skin of both humans and mice with AA65,167. Furthermore, HFs in AA affected mice also show increased expression of Estrogen receptor 1 (esr1) expression65. Both CRH-R2 and Esr1 are important receptors that regulate the local HPA axis and response to inflammation upon stressful stimulus167,168. Potentially, the immune system activity in AA may be the cause of the 20  changes in local skin HPA and the aberrant central HPA activity, and this may result in an inability to cope with stress.  The differentiation of mast cells can be modulated by CRH169,170 and this in turn may be a pathway of how stress can modulate the progression of AA since CRH/receptor activity is high in AA skin. Altered expression of neuropeptide substance P is also associated with AA and application of substance P onto the skin resulted in accelerated HF regression, induced mast cell degranulation, and increased numbers of CD8+ T-cells expressing granzyme B171-173. These investigations suggest that there is a positive feedback loop in which inflammatory activities in AA can disrupt the HPA axis, leading to secretion of stress hormones and changes in responses to stress, and these events exacerbate the situation by accentuating inflammatory activity. The dysregulated stress hormones in AA can affect sites away from the skin as with psoriasis174-176; the potential for AA development to affect cardiovascular health in AA affected mice and humans is discussed in Chapter 4 and 5. 1.4.9 Diet Iron deficiency is reported as one of the features associated with various forms of hair loss including AA as revealed in numerous cross-sectional studies177. It is observed mostly in females and between 24-71% of AA female patients suffer iron deficiency. It is suspected that the lack of iron hinders a rate-limiting enzyme for DNA synthesis which may in turn decrease the proliferative capacity of HF matrix cells177; but the exact functional mechanism behind this apparent correlation is still unknown178,179. Even though there is still a debate to whether dietary intake of iron can improve hair loss180, many dermatologists value the 21  importance of iron-deficiency and still recommend using iron as a supplementary treatment to help women with hair loss177. Different diets in different geographical settings may also have an effect on the lifetime risk of AA. From very limited studies, it has been shown that, the lifetime risk of AA in the UK and the US is about 1.7%50, but the lifetime risk in Japan has been estimated at less than 1%144. Interestingly, Japanese who live on Hawaii (with a predominantly Western non-soy diet) have higher AA incidence compared to mainland Japan181. Whether these observational studies have any significance for dietary input on AA development still need to be further investigated.  Food with high dietary soy oil content seems to increase resistance to AA in mice. Mice that received AA-affected skin grafts showed hair regrowth compared to similarly grafted mice that received a normal diet144. A potential explanation for this may be the antioxidant enzyme enhancement and estrogen receptor binding properties of soy derivatives might block AA development in mice182. 1.4.10 Other factors Vaccination has been implicated as an inducer of AA146, but a large scale study with mice was unable to determine a correlation between AA and hepatitis B vaccination147. Studies on hormones also reveal that estrogens may accelerate AA progression while testosterone might reduce AA susceptibility143. The involvement of Cytomegalovirus (CMV) in AA has been debated15. Although many subsequent studies disproved this hypothesis18,19,183,184, the recent GWAS study suggests a possible role for a CMV binding protein in NK cell activation, leading to AA70. 22  1.5 Treatments In many AA patients, their hair loss waxes and wanes or they may experience multiple episodes of remission and expression73. It is estimated that up to 50% of AA patients experience spontaneous hair regrowth within 1 year47. However, there is no universally proven treatment or cure for AA that could permanently reverse hair loss47,185. Various therapies are available to control the disease and they are prescribed based on the disease severity, duration and responses to other treatments46.  Immunotherapies, including intralesional injection and topical application of corticosteroids, as well as diphenylcyclopropenone (DPCP), are some of the most commonly used treatments46. Corticosteroids are chemicals with steroid hormone-like properties that have immunosuppressive effects, therefore they nonspecifically dampen immune responses in the applied area and help to clear the infiltration of inflammatory lymphocytes.  The mode of mechanism behind DPCP is less clear, it is a contact sensitizer that is believed to have an immune-deviating effect, directing immune cells to attack other targets instead of HF cells, or inducing apoptosis in perifollicular lymphocytes46. Other topical treatments include anthralin, which is a treatment that should not be used with corticosteroids46. The exact mechanism behind anthralin is unknown, animal studies showed anthralin can suppress inflammatory cytokine TNFα and TNFβ expression as well as slow down cell proliferation46.  Minoxidil is used as an adjuvant treatment and may function as a vasodilator, inducer of angiogenesis, enhancer of cell proliferation and potassium channel opener; it was originally developed as a treatment for hypertension46. The exact mechanism behind 23  minoxidil is also unknown, but it is suggested that it has the ability to stimulate HF cell DNA synthesis when used as a treatment for AA186. Minoxidil is not specific to AA, it is also part of treatment regimen for androgenetic alopecia187. Systemic therapies include the use of systemic corticosteroids or photochemotherapy (ie. PUVA). These systemic treatments are associated with many undesirable side effects therefore they are not the first line of treatment for AA46. Other experimental types of treatments include the use of cyclosporine, methotrexate as well as biologics such as sulfasalazine (inhibition of T cell activity) and prostaglandin analogues (stimulation of HF melanocytes and conversion to anagen), however these treatment modalities also have very limited effectiveness46.   One of the most recent experimental treatments for AA is with a JAK inhibitor, ruxolitinib, which was originally a FDA approved treatment for rheumatoid arthritis89. The JAK inhibitor interferes with cytokine production such as IFNγ, IL-2 and IL-15, hence potentially reducing the accumulation of CD8+/NKG2D+ T-cells. This treatment has showed a positive effect in AA-affected mice as well as in humans (in the initial clinical trial)89.  At the moment, there is no cure for AA, only different modalities to manage hair loss1,46,47. Many of the current therapies or experimental drugs, such as minoxidil, DPCP and JAK inhibitors, stemmed from treatments being used on other diseases. How exactly these modalities improve AA is still largely unknown, a broad range of cellular or cytokine pathways are often targeted46,89,186. However, as the goal of many new emerging treatments generally involves targeting immune cell activity and/or reduction of inflammation and also 24  the observation of inflammatory cell regression following successful treatments47; it further strengthens the hypothesis that AA is a cell-mediated inflammatory disease. 1.6 Outstanding Questions Regarding the Pathogenesis of Alopecia Areata Even though AA has been well recognized for many decades, there is still much we do not understand regarding its pathogenesis and the long term consequences of the disease. There are several important questions regarding the etiopathogenesis of AA that still need to be answered and they may provide further understanding of the different pathologic features of AA. Addressing these questions will in turn lead to further investigations and development of more effective treatment modalities. 1.6.1 Are animal models of AA satisfactory reflections of AA in humans and can the disease models be improved? The best model to study human disease is another human, however there are too many variables in the human population to study disease effectively without using large cohorts. Furthermore, ethical issues prevent the involvement of patients in many forms of research, particularly those requiring proactive interventions with unknown consequences, such as genetic modification procedures. Although animal models may not represent human diseases completely, they are in a controlled environment and can provide more insight to the disease via genetic and cellular manipulations. In any investigation of disease pathogenesis, information from in vitro and in vivo studies can be equally important as clinical and retrospective studies. Therefore, various animal models have been developed to study the pathogenesis of AA as well as to evaluate the efficiency of potential new treatments. 25  However, each animal model is also associated with certain limitations and needs to be selected carefully depending on what research questions are being addressed. The Dundee Experimental Bald Rat (DEBR) was the first validated rodent model for AA188. It was with this model that the first knowledge of the roles of CD4+ and CD8+ T-cells in AA development was deciphered, it has also been used to test different immune-modulatory drugs as potential new treatments44,45,189. However, the relatively larger size of the rats makes them not as easily accessible compared to mouse models.  A humanized AA mouse model was also developed by grafting human scalp biopsies onto beige SCID mice followed by transfer of enriched NKG2D+/CD56+ cells from cultured human peripheral mononuclear cells (PBMCs)190. This method has the advantage of mimicking the activities of the human skin immune system in mice. However, it is limited by the complicated and specialized procedures required to graft the skin, as well as the prolonged time required to generate these mice; not all laboratories have access to human scalp biopsies too. The C3H/HeJ AA mouse model has been a very popular model for studying AA for the past decade189. These mice can develop AA like hair loss spontaneously, albeit unpredictably, in old age191, or be induced to develop AA at a higher rate via a full-thickness skin grafts taken from an AA-affected donor mouse96. Because of its resemblance to human AA and its relative availability, it is one of the most well-defined AA models so far192,193. It has also been found that AA can be transferred between C3H/HeJ mice by cells isolated from skin-draining lymph nodes as well, providing evidence that AA is a cell-mediated disease. However, to generate C3H/HeJ AA mice via skin graft, a relatively invasive surgical 26  procedure is also needed. Even though it does not require human scalp biopsies, it also does not reflect human skin immune cell activities completely. Several other strains of mice develop AA spontaneously as well, and are useful for studying complex genetic associations, but they are not as well-defined as the current C3H/HeJ mouse model and are used to a lesser extent194. There are also larger mammalian models of AA available such as horses and dogs; they are very useful in terms of acquiring large sample volumes for studies such as investigating autoantibodies in the serum195-197. However, these models are extremely limited in their availability and are not practical models for studying AA compared to mice.  All of the animal models for AA have their own strengths and limitations, and none of them will be a complete reflection of human AA. To help investigators to study the biology of AA, an animal model that can consistently be induced to develop AA without the complications of invasive surgeries, while maintaining the same resemblance to human AA, is needed. A novel technique to generate a large number of AA C3H/HeJ mice via cultured cell injection, developed to supplement the currently available models, is discussed in detail in Chapter 2. 1.6.2 Where and what is the target of T-cell attack in AA affected hair follicles? The infiltration of inflammatory cells is localized around the transient region of the anagen HFs84. Therefore it is logical to assume the antigen targets are located in the transient region rather than the permanent region (bulge region) of the HF. Keratinocytes constitute the HF matrix and root sheaths; dermal papilla and dermal sheath cells are mesenchyme 27  derived, while melanocytes in the hair bulbs, are all capable of expressing candidate antigens potentially targeted by inflammatory cells.  Melanocytes are believed to be the source of antigen epitopes, based on several studies and clinical observations, because selective pigmented hair loss occurs during the development of AA in humans87,198-200. The DEBR rats can sometimes lose the pigmented hair around the neck before the onset of AA in non-pigmented pelage areas on their trunk189. As such, proteins produced by melanocytes that are associated with pigmentation are believed to be potential antigen targets for AA development. Melanoma-associated antigens such as GP100 and MLANA/MART1 were used to stimulate T-cells and successfully induced AA-like symptoms in previously AA affected human skin biopsies grafted to SCID mice; strengthening the hypothesis of target antigen epitopes with melanocyte origin139. However, from histological analysis, the intrafollicular infiltration of lymphocytes in AA is not focused on melanocytes as one might anticipate if they were the primary target, but instead localized among root sheath and matrix keratinocytes78,138. In addition, autoantibodies against keratinocyte proteins like cytokeratin 16 and trichohyalin were found in AA patients88,111,136, suggesting autoantigen epitopes from multiple origins can contribute to the development of AA. In this thesis, a panel of candidate autoantigen epitope peptides are analyzed for their ability to trigger AA CD8+ T-cell activation as well as induction of apoptosis in primary HF keratinocytes compared to healthy controls; this is discussed in Chapter 3.  28  1.6.3 Is AA really just a dermatological disorder? AA is known to have several different phenotypic presentations from the classic circular patchy hair loss to diffuse AA, exclusive ophiasis AA, nevoid AA as well as totalis and universalis53,54. All these phenotypes could potentially indicate different underlying disease development mechanisms or variations thereof. With the diversity of potential genetic and environmental inputs, it is also possible that individuals presenting with same phenotypes have different causal disease pathogenesis mechanisms. In addition, AA can be associated with many other types of diseases and dermatological disorders, mental health problems, metabolic diseases, cardiovascular diseases and autoimmune diseases60. All of these may be part of the trigger, or sequelae of AA development. In recent years, associations between cardiovascular diseases and stress to psoriasis have been made174-176, highlighting the potential for psoriasis to have systemic effects, rather than just local effects limited to the skin. In Chapter 4 and 5 of this thesis, the potential association of cardiovascular disease with AA development is investigated. 1.7 Thesis Rationale, Hypothesis and Specific Aims Despite all the recent advances in the research on AA, a critical component of evidence that can classify AA as a true autoimmune disease is still lacking; the identification of autoantigen epitopes. There is a paucity of research with the aim to identify autoantigen epitopes that elicit the activation of autoreactive T-cells even though a cell-mediated mechanism is apparent in both human and animal models of AA. As such, many of the current treatments for AA are unspecifically targeting the immune system. With the identification of epitope peptides or autoreactive T-cell clones, more specific treatment 29  modalities can be derived. Furthermore, comorbidities associated with AA have only been observed in epidemiological analyses and reported in retrospective studies, more functional studies on potential disease association with AA is needed. For this dissertation, there are two main hypotheses: 1) Autoreactive CD8+ T-cell (CTL) activation in AA is elicited by HF autoantigen derived epitope peptide sequences; the activation of autoreactive CTLs can have an adverse effect on the viability of HF keratinocytes. 2) The dysregulated stress hormones found in AA, or the inflammatory cytokine expression related to AA development, can have adverse effects on the health of the heart. Consequently, the development of AA can elevate the expression of heart disease markers in association with increased apoptosis in cardiomyocytes. In Chapter 2, in the process of analyzing skin-draining lymph node cells (LNCs) of AA-affected mice compared to controls, I have developed a novel technique to induce AA in C3H/HeJ mice via cultured cell transfer. I have found only the LNCs isolated from an AA affected mouse can be cultured to induce AA in the recipients; cultured control LNCs have no effect. This led me to investigate the mechanism behind this AA induction method as well as the changes in cell markers before and after cell culture.  In Chapter 3, I isolated peripheral blood mononuclear cells (PBMCs) from AA affected human subjects and control subjects without inflammatory hair loss. I cultured the PBMCs with a panel of autoantigen epitope peptides to identify peptides that can specifically activate CD8+ T-cells (CTLs) in AA patient samples. I have also evaluated the ability of AA 30  epitope peptide-specific CTLs to induce apoptosis in HF keratinocytes via soluble factors compared to controls. The activation of AA-affected C3H/HeJ mouse CD8+ T-cells from skin-draining lymph nodes by mouse homologues of human HF antigen epitope peptides was also investigated.  In Chapter 4, I investigated the association of AA development in C3H/HeJ mice with abnormal heart morphologies. I have measured the changes in gene expression of heart disease markers, inflammatory genes, and apoptosis related genes in the skin and heart tissue of AA mice compared to controls. Furthermore, the role of stress hormone adrenocorticotrophic hormone (ACTH) in the change of gene expression and heart disease marker secretion of AA mouse hearts was investigated. The positive correlation between heart tissue damage and AA in mice encouraged me to measure heart disease markers in the serum of patients with AA, androgenetic alopecia (AGA), versus subjects without any hair loss, to identify potential correlations of both hair loss diseases on the health of the heart in humans. Additionally, the effect of sera with high heart disease marker presence on the viability of primary human cardiac myocytes was investigated using cell cultures (Chapter 5).  The overarching aim in this thesis is to identify and confirm autoantigen epitope peptides that may be important for the pathogenesis of AA as well as investigate the long term effect of AA on the health of the heart. The specific goals for the present thesis are: 31  1) Confirm AA is a cell-mediated disease by transferring disease development via cultured skin-draining LNCs. 2) Identify autoantigen epitope peptides that can specifically trigger activation of AA CTLs (both human and mouse); confirm the effect of such activation on HF keratinocytes. 3) Elucidate the effect of AA on the health of the heart using a mouse model; to evaluate the effect of AA development in humans on the expression of heart disease markers. 32  Chapter 2: Transfer of Alopecia Areata to C3H/HeJ Mice via Cultured Cell Transfer: A Novel Animal Model to Study Human Alopecia Areata  2.1  Introduction Alopecia areata (AA) is believed to be a cell-mediated, inflammatory, non-scarring, autoimmune hair loss disease. In humans, it is presented as patchy hair loss in the acute phase but can progress to the loss of all scalp hair or even body hair during the chronic phase1. The exact pathogenesis of AA is still under investigation; much of what we know is derived from various animal models currently available such as the C3H/HeJ mice96, DEBR rats201 and more recently, humanized SCID mice202. The C3H/HeJ mouse model is one of the most popular and well defined rodent models for AA research203. It is an inbred strain with individual mice spontaneously develop AA-like hair loss at a low frequency when they reach an older age191. Previously, a surgical method was developed to transfer the AA phenotype from AA affected mice to healthy histocompatible recipients as a way to control disease onset and increase numbers96. This was achieved by full-thickness skin grafting from spontaneous AA affected C3H/HeJ donors to naïve recipients of the same strain. The skin-grafted AA model has proven useful with its similarity to human AA as well as being relatively easy to generate. The model has been used to test several experimental treatments for human AA189,204.  However, there are caveats associated with the skin-grafting method. The major draw-back is the need to perform a relatively long, invasive surgery on the mice. The 33  investigator must have good surgical skills and must fulfill additional requirements enforced by institutional and local authorities to ensure animal wellbeing even though it is a relatively simple procedure with limited adverse impact on the health of the mice. In some jurisdictions (eg. Canada), the use of controlled drugs for anaesthesia require government certification; involving a potentially lengthy approval process. Additionally, the time it takes for the mice to recover from the skin-graft surgery is relatively long and there is a risk of infection and graft rejection96. Usually, 10-20 recipient mice can be grafted with skin from a single AA donor and it takes about 10 weeks before AA starts to develop. Therefore, the time and effort involved, as well as the skills required for inducing AA via skin-grafting, may not be feasible for many laboratories that may wish to do AA research with mice. Previously, McElwee et al. demonstrated that freshly isolated cells from skin-draining lymph nodes of C3H/HeJ mice affected with AA could be injected into histocompatible recipients to transfer the disease98. The transfer of CD8+ T-cells alone into naïve hosts can result in localized hair loss while the transfer of CD4+/CD25- T-cells resulted in systemic AA, not just at the injection site. The data  suggested that CD8+ T-cells may play a cytotoxic role in AA while CD4+/CD25- T-cells provide a classic “helper” role45. A cell-injection method to transfer AA is potentially an easier alternative to skin-grafting. However, successful transfer of AA required large numbers of cells and the number of LNCs obtained from one donor is relatively limited98.  In this study, we aimed to develop a novel AA rodent model that closely resembles human AA, but without the need to perform invasive surgery, by injecting normal haired mice with cultured skin-draining LNCs derived from AA affected C3H/HeJ mice. 34  2.2 Materials and Methods 2.2.1 Mice Animal studies were approved by the University of British Columbia Animal Care Committee. Normal haired, female C3H/HeJ mice (The Jackson Laboratory, Bar Harbor, ME) were induced to express AA by skin grafting as described previously96. Age matched, sham-grafted littermates were used as controls. The skin draining lymph nodes were collected for cell expansion. 2.2.2 Single cell suspension The skin draining lymph nodes were collected into complete medium (AR10) comprised of Advanced RPMI 1640 (catalog #12633-012) with 10% FBS (catalog #16000-044), 2 mM Glutamax (catalog #35050-061), and 100 μg/mL Streptomycin with 100 U/mL Penicillin (catalog #15140-148); all from Invitrogen (Burlington, ON). Prior to processing, the lymph nodes were washed by dipping into sterile DPBS (Invitrogen, catalog number: 14190-250) and transferred to a 70 μm cell strainer (Fisher Scientific, Ottawa, ON. #08-771-2). 1 mL of fresh AR10 was added to the lymph nodes and, with the back end of a syringe plunger, the lymph nodes were crushed against the cell strainer membrane for 5 minutes or until all tissues disintegrated. The resulting cell suspension was washed with AR10, counted and resuspended at 2 million cells/mL. 2.2.3 In vitro stimulation and expansion of lymph node cells (LNCs) In a non-tissue culture treated 24-well plate (BD Bioscience, Mississauga, ON. catalog #170-340-06), 2 million LNCs (1 mL) were seeded into each well and set aside in a 35  37 ºC incubator with 5% CO2. The stimulation of LNCs was achieved using Dynabeads Mouse T-Activator CD3/CD28 kit (Invitrogen, catalog #114.52D). The preparation of the Dynabeads were performed by resuspending the desired amount (25 μL/million LNCs) of beads with 1 mL sterile PBS with 2 mM EDTA and 0.1% FBS in a sterile tube. The Dynabeads were washed carefully by pipetting up and down 10 times without generating bubbles, and placed into a magnet for 1 minute. The Dynabeads were retained on the side of the tube and wash buffer can be removed by pipetting without touching the side of the tube. After removing the magnet the Dynabeads were resuspended in AR10 (500 μL AR10/25 μL beads) and 500 μL of Dynabeads distributed to each well of LNCs for a final volume of 1.5 mL per well. To each well, 30 U/mL of human recombinant IL-2 (Roche Life Science, Laval, QC. catalog #11011456001), 25 ng/mL mouse recombinant IL-7 (R&D Systems, Minneapolis, MN. catalog #407-ML-005) and 50 ng/mL mouse recombinant IL-15 (R&D Systems, catalog #447-ML-010) was added. The Dynabead/LNCs culture was gently mixed and then incubated for 3 days without disturbing the beads.  On the third day, when the morphology of the LNCs should be irregular and should cover the bottom of the wells completely, the culture medium at this point should also be light yellow, indicating rapid cell proliferation. At this point, each well was split into two and fresh AR10 and cytokines were added to make 1.5 mL and cultured for another 24 hours. On the fourth day, the LNCs should reach the peak of proliferation and two wells were combined into 1 T25 suspension cell flask (Sarstedt, Montreal, QC. catalog #83-1810-502) and fresh AR10 and cytokines added to total 6 mL. On the fifth day, two T25 flasks were combined into one T75 flask, topped up to 20mL with fresh AR10 and cytokines, or 6mL fresh AR10 with cytokines added to each T25 flask. On the sixth day, all LNCs with Dynabeads were 36  transferred into tubes in a magnet for one minute to separate the LNCs from the beads. After all the Dynabeads were removed, the number of LNCs were counted and centrifuged for 10 minutes at 350x g at room temperature. The LNCs were resuspended in sterile PBS at 10 million cells/100 μL and drawn into an insulin syringe. 2.2.4 Transfer of cultured LNCs into C3H/HeJ mice The mice that were at least eight weeks old (previous skin graft studies indicate mice younger than 6 weeks are less likely to develop AA) were anesthetized with isofluorane and shaved on the lower-dorsal area to expose the skin. In the initial group, cells were injected intra-dermally (preliminary studies indicated subcutaneous injection was not successful in transferring AA – data not shown) into the skin regardless of the growth cycle stage of hair follicles at the time of transfer. With a pair of forceps, a small portion of the skin was pinched and lifted up ready for intra-dermal injection. The syringe was inserted into the dermis almost horizontally relative to the skin surface of the skin and 100 µL of LNCs (10 million LNCs) were intra-dermally injected into each mouse. Hair grows back at the site of injection, the hair loss typically begins at places away from the injection site. In a different group of mice, cultured AA LNCs were injected into mice with telogen skin (confirmed visually) or with anagen skin. The mice with anagen skin were induced by plucking hair in a small 1 cm diameter area one week prior to cell-transfer. 2.2.5 Labelling of the cultured LNC and fluorescent microscopy Following cell expansion, the LNCs were resuspended in serum-free medium at 1 million cells/mL. CellTracker CM-DiI fluorescent dye (Invitrogen, catalog number: C-7000) was prepared by diluting the stock to 1 mg/mL with sterile DMSO (Sigma Aldrich, Oakville, 37  ON. catalog number: D2438). 5 µM of CM-DiI was added per 1 million LNCs and incubated for 30 minutes at 37 ºC, while mixed well by gentle pipetting every 10 minutes. The LNCs were washed twice with PBS, resuspended in PBS at 10 million cells/100 µL, and transferred into mice as described above.  As soon as the mice that received the cultured AA LNCs developed AA, they were sacrificed along with comparable mice that received cultured control LNCs. Skin at the site of injection, peripheral skin, skin-draining lymph nodes, spleen and liver were collected and fixed in 10% formalin for three days and mounted into paraffin blocks for routine histology by University of British Columbia Department of Pathology. The paraffin processing steps were performed by an automatic processor Miles/Tissue Tek VIP 3000 (IMEB INC, San Marcos, CA). The tissues were incubated in 70% ethanol bath overnight until the process begins. All steps took 45 minutes with alternating pressure and vacuum every few minutes: 1x in 80% ethanol, 2x in 95% ethanol, 3x in 100% ethanol, 3x in xylene, 3x in molten paraffin (59 ºC). At the end of last molten paraffin bath, the tissues were embedded into block molds and let cool.  2.2.6 Histology and image analysis The paraffin blocks were cut into 5 µM sections and deparaffinised by incubating 2x 5 minutes in 100% Citrasolv, 1x 5 minute in 100% Citrasolv/ethanol. Followed by 2x 3 minutes in 100% ethanol, 1x 1 minute in 95% ethanol and 1x 1 minute in 75% ethanol.  To visualize the CM-DiI labelled cells in the tissues. The tissue sections were mounted with Vectashield DAPI (Vector Laboratories, Burlington, ON. Catalog #H-1200) and the cover slide was sealed with nail polish. After drying, the tissue sections were 38  visualized with a Zeiss Axiovert 200M inverted fluorescence microscope (Zeiss, Toronto, ON). Image processing were performed with AxioVision Rel. 4.6 software (Zeiss, Toronto, ON) with red colour set to CM-DiI and DAPI for blue colour.  Hematoxylin and Eosin (H&E) staining was performed following standard protocols. After deparaffinization, the slides were stained 3 minutes with Accustain Harris Hematoxylin Solution (Sigma Aldrich. Catalog #HHS). After staining, the slides were rinsed with deionized water by dipping a few times and rinsed again in tap water 2x 1 minute to allow the color to develop. The slides were blotted to remove excess water and stained with acidified Aqueous Eosin Y Solution (Sigma Aldrich. Catalog #HT1102) for 30 seconds; the aqueous eosin solution was acidified with glacial acetic acid (0.5 mL glacial acetic acid per 100 mL of stain). Dehydrate, clear and mount with DPX Mountant (Sigma Aldrich. Catalog #06522). Dehydration process was performed in the same way as the deparaffinization steps but in reverse order. 2.2.7 Immunohistochemistry The paraffin blocks were cut into 5 µM sections and deparaffinised by incubating 2x 3 minutes in 100% Xylene, 1x 3 minute in 100% xylene/ethanol. Followed by 2x 3 minutes in 100% ethanol, 1x 3 minute in 95% ethanol, 1x 3 minute in 70% ethanol, 1x 3 minute in 50% ethanol and rinse with running tap water (keep in water). Heat-induced epitope retrieval (HIER) was performed by first heating up the antigen retrieval buffer (Tris-EDTA Buffer: 10 mM Tris Base, 1 mM EDTA Solution, 0.05% Tween 20, pH 9.0) to boiling point. Slides were transferred into hot antigen retrieval buffer and keep in the boiling steamer for 20 minutes then cool with running cold tap water for 10 minutes. 39  Following antigen retrieval, slides were washed 2x 5 minutes with TBS 0.025% with Triton-X (0.25ml of Triton-X to 1L of 1xTBS) and blocked in 10% normal goat serum (Vector Labs, Burlington, ON. Catalog #S-1000) in TBS for 30 minutes at room temperature. Primary antibody, Rabbit polyclonal anti-CD3 (Abcam, Toronto, ON. Catalog #ab828) diluted 50x in TBS with 5% normal goat serum was applied to the slides and incubated overnight at 4 ºC.  On the following day, the slides were washed 3x 5 minutes in TBS 0.025% Triton-X with gentle agitation. The slides were block with 0.3% H2O2 in TBS for 15 minutes at room temperature and then rinsed 3x 5 minutes in TBS. Secondary antibody, biotinylated goat anti-rabbit (Vector Labs. Catalog #BA-1000) diluted 200x in TBS with 5% normal goat serum was applied to the slides. Vectastain Elite Universal ABC kit (Vector Labs. Catalog #PK-6200) was used as the immunoperoxidase detection system to detect CD3 via avidin/biotin interaction. DAB Peroxidase (HRP) Substrate Kit (Vector Labs. Catalog #SK-4100) was added to the slides and color was left to develop until desired intensity is achieved. The slides was counterstained with haematoxylin and washed with running cold tap water. After washing, the slides were dehydrated, and cleared by following the deparaffinization steps in reverse; slides were mounted with DPX Mountant. Slides were visualized with a Nikon ECLIPSE LV150N microscope and analyzed with a NIS-Elements Imaging software (Nikon, Mississauga, ON). 2.2.8 Flow cytometry To evaluate the changes in surface and intracellular markers before and after cell expansion, a protocol for intracellular cytokine staining was employed. LNCs before 40  expansion were stained right away after isolation while LNCs after expansion were passed through a magnet to remove the Dynabeads. All antibodies and reagents were purchased from Affymetrix eBioscience (San Diego, CA) unless otherwise indicated. To investigate the change in the population of CD4+, CD8+ and IL-17+ LNCs before and after expansion, endoplasmic reticulum inhibitor brefeldin A (5 µg/mL. Catalog #00-4506-51) was added to the culture 16 hours prior to the staining to inhibit the secretion of IL-17 into the culture media. The LNCs were stained with Anti-Mouse CD4 FITC (Clone: GK1.5, 1:100 dilution. Catalog #11-0041-86), Anti-Mouse CD8a PE (Clone: 53-6.7, 1:100 dilution. Catalog #12-0081-85) and Anti-Mouse/Rat IL-17A eFluor 660 (Clone: eBio17B7, 1:100 dilution. Catalog #50-7177-82) for 30 minutes in the dark. To investigate the changes in the level of IFNγ+ CD4+ or CD8+ T-cells. Endoplasmic reticulum inhibitor brefeldin A (5 µg/mL) was added to the culture 16 hours prior to the staining to inhibit the secretion of IFNγ into the culture media. The LNCs were first stained with Anti-Mouse CD4 FITC (Clone: GK1.5, 1:100 dilution), Anti-Mouse CD8a APC (Clone: 53-6.7, 1:100 dilution. Catalog #17-0081-81). The LNCs were washed with FACS staining buffer (PBS with 1% FBS) once and fixed. LNCs were washed with 1x Foxp3 permeabilization buffer after fixation and stained with Anti-Mouse IFN gamma PerCP-Cyanine5.5 (Clone: XMG1.2, 1:100 dilution in Foxp3 permeabilization buffer. Catalog #45-7311-80) for 30 minutes in the dark on ice.  To investigate the changes in regulatory T-cells (Tregs; CD4+/CD25+/Foxp3+). The LNCs were first stained with Anti-Mouse CD4 FITC (Clone: GK1.5, 1:100 dilution) and Anti-Mouse CD25 APC (Clone: PC61.5, 1:100 dilution. Catalog #17-0251-82) for 30 41  minutes in the dark. After fixation and permeabilization steps LNCs were stained with Anti-Mouse/Rat Foxp3 PE (Clone: FJK-16s, 1:50 dilution in Foxp3 permeabilization buffer. Catalog #12-5773-80). In a different group of mice, the LNCs labelled with CM-DiI (red fluorescence) were injected intra-dermally. The level of CM-DiI labelled cells in the skin at the site of injection, peripheral skin, skin-draining lymph nodes, and the spleen was also investigated with flow cytometry in addition to immunohistochemistry. Mice received expanded, labelled AA or control LNCs after 3, 7 and 11 days were analyzed. To isolate lymphocytes, skin piece (1 x 1 cm) was washed with PBS and placed in a 60 mm x 15mm petri dish and 3 mL of type I collagenase solution (1 mg/mL in PBS; Sigma Aldrich, catalog #C0130-1G) was added. Skin was chopped into small pieces with a blade and incubated for 1 hour at 37 ºC. Cells were further dissociated by pipetting up and down. Cells were washed twice with RPMI 1640 medium (Invitrogen, Catalog #11875093) with 10% FBS before resuspending with FACS staining buffer for flow cytometry staining. Single cell suspension from skin-draining lymph nodes and spleens were obtained as outlined above. Isolated cells from the tissues at different time points were stained with Anti-Mouse CD3 APC (Clone: 17A2, 1:100 dilution. Catalog #17-0032-80) or Anti-Mouse CD4 PerCP-Cyanine5.5 (Clone: RM4-5, 1:100 dilution. Catalog #45-0042-80) or Anti-Mouse CD8a APC (Clone: 53-6.7, 1:100 dilution. Catalog #17-0081-81). During flow cytometry analysis, the PE channel was used to identify CM-DiI labelled LNCs co-expressing CD3, CD4 or CD8. In all different flow cytometry analyses, the LNCs were fixed and permeabilized using the same method. The LNCs were fixed with Foxp3/Transcription Factor Staining 42  Buffer Set (catalog #00-5523-00). Fixation was carried out by incubating LNCs with Foxp3 fixation/permeabilization buffer for 30 minutes in the dark on ice and then washed with 1x Foxp3 permeabilization buffer (both included in the kit). After fixation and permeabilization, intracellular markers were stained if applicable and then washed with FACS staining buffer. The labelled LNCs were resuspended in 400 µL FACS staining buffer and analyzed with a BD Accuri C6 Flow Cytometer (BD Bioscience). 2.3 Results 2.3.1 Addition of IL7 and IL15 resulted in more rapid LNC expansion compared to IL2 alone The addition of IL7 and IL15 on top of IL2 and Dynabeads provided LNCs with an optimal condition for rapid activation and expansion. The morphology of LNCs started to change after 24 hours, typically increase in cell size and irregular shape (Figure 2.1a). The cultures initially developed clumps of Dynabeads and LNCs, but these clumps dispersed at subsequent sub-cultures and did not redevelop. The rate of growth from LNCs isolated from both AA and healthy C3H/HeJ mice was similar. The number of cells expand around 8-10 fold after six days of culture compared to day 1 (Figure 2.1b); freshly isolated LNCs from a single donor can be expanded to inject into at least 50 recipients (10 million LNCs/mouse), if no LNCs were spared. 43   Figure 2.1 LNC size increased and morphology changed as cells proliferated exponentially over six days with the addition of IL7 and IL15. At around day 2 and 3, clumps of cells could be observed in combination with anti-CD3 and anti-CD28 antibody coated Dynabeads, however, clumping did not affect cell expansion (a). The clumps disappeared at subsequent days after passaging. The rate of LNC growth was faster when the culture was supplemented with IL7 and IL15 instead of just IL2 as recommended by the manufacturer’s protocol. At the end of culture, around 8-10 fold expansion of LNCs, compared to day 1, could usually be obtained (b). Bar = 50 µm.   44  2.3.2 Cultured AA skin-draining lymph node cells induced AA development in healthy C3H/HeJ mice The cell-transfer method using the cultured skin-draining LNCs isolated from an AA donor was able to consistently induce AA development in healthy C3H/HeJ recipients with over 90% success rate. Out of all the mice that received cultured LNCs from a control donor without AA, only one developed AA within the 20 week observation period (Figure 2.2). The time required for the first observable AA symptoms ranged from two weeks to up to 19 weeks post injection. The time it took for the mice to progress from patchy to extensive hair loss varied between each individual as observed with skin graft induced AA. The injection of LNCs into anagen or telogen skin did not seem to affect the frequency of AA development. Kaplan–Meier plot was employed to represent the fraction of mice that developed AA after injection with cultured AA or control LNCs.    45   Figure 2.2 Naïve C3H/HeJ mice that received cultured AA LNCs developed AA as early as two weeks. The majority of mice that received cultured AA LNCs developed AA-like symptoms around 7-10 weeks. However, it was observed from different individual mice that AA could develop as early as two weeks after LNC injections. In the control group, only one mouse developed AA within the 20 week observation period. The survival curve was plotted using Kaplan-Meier method and significance was assessed with the log-rank test.   46  2.3.3 The pattern of AA development was consistent and similar to the skin-grafting method In mice that received cultured AA LNCs, the hair loss did not begin at the site of injection. Instead, the hair loss usually began on the ventral side of the mice and progressed towards the dorsal side; the site of injection was often the last place to lose hair (Figure 2.3a). In mice that received control LNCs, no hair loss was observed within the 20 week observation period (Figure 2.3b). Histologically, typical swarming of lymphocytes and dystrophic hair follicles were found in the AA lesions (Figure 2.3c) while in mice that received cultured control LNCs, there was an absence of any apparent inflammatory infiltrate around the hair follicles (Figure 2.3d). Immunohistochemistry (IHC) further confirmed high density of CD3+ T-cells around the HF of AA mice compared to the control mice (Figure 2.3e,f).    47   Figure 2.3 The pathologic features of AA induced via cell injection showed typical swarming of lymphocytes around HFs. The AA-like symptoms from the mice that received cultured AA LNCs usually initially expanded from one or more patches on the ventrum towards the dorsum of the mice (a) while the mice that received cultured control LNCs remained fully haired even after 20 weeks (b). H&E staining of AA mice showed typical inflammatory cell clustering around dystrophic anagen HFs (c) while control mice had healthy HFs with an absence of inflammation (d). IHC analysis showed clustering of CD3+ T-cells (brown) around HFs in AA lesions (e) while control mice showed very limited numbers of CD3+ T-cells in the skin (f). Figure 2.2c, 2.2d scale bar = 100 µm. Figure 2.2e, 2.2f scale bar = 400 µm.   48  2.3.4 Cultured AA LNCs expressed similar markers compared to cultured control LNCs Analyses on the expression of surface and intracellular markers of LNCs before and after cell expansion with flow cytometry revealed similar pattern of expansion of cell populations in AA compared to control LNCs. Comparing cell populations before and after expansion, there was an increase of CD4+/IL17+ T-cells in both AA and control LNCs after expansion (Figure 2.4a). However, CD8+/IL17+ T-cells was decreased in both AA and control LNCs after expansion. Both CD4+ and CD8+ expressing IFNγ T-cells were increased after expansion and statistical significance was reached in CD4+/IFNγ+ T-cells in control LNCs (Figure 2.4a). As expected, CD4+ T-cells in both AA and control LNCs were significantly activated after expansion as indicated by T-cells expressing CD4 and CD25 but not FOXP3. The level of Treg in AA LNCs decreased after expansion while it was increased in control LNCs (Figure 2.4b). The percentage of Tregs in AA LNCs after culture was significantly lower than control LNCs.    49   Figure 2.4 Both AA and control LNC derived cell cultures after expansion showed elevated numbers of IFNγ producing cells; Treg frequency increased in controls but decreased in AA cultures. After six days of culture, both AA and control LNCs showed higher levels of IFNγ producing CD4+ and CD8+ T-cells (a). There was an increase of CD4+/IL17+ T-cells but decrease in CD8+/IL17+ T-cells in both AA and control LNC derived cell cultures (a). In the cultured AA LNCs, there was a significantly lower percentage of Tregs (CD4+/CD25+/FOXP3+) compared to control LNCs after culture. n=3 for both AA and control LNC samples. Significance was determined by Student’s t-test where * denotes significant difference (p<0.05).   50  2.3.5 Mice receiving either cultured AA or control LNCs showed an opposite trend of CD4+ and CD8+ T-cells localization in skin-draining lymph nodes over time Within the skin-draining lymph nodes of injected mice, over 95% of labelled cells expressed CD3 (Figure 2.5a). Among the labelled population, both AA and control mice showed decreasing trend of CD4+ T-cells from 3 to 11 days while CD8+ T-cells showed an increase during this time (Figure 2.5a). However, only a very low level of labelled LNCs were detected in the skin tissues at the site of injection (iSkin) and the spleen; no labelled LNCs were detected in the peripheral skin (pSkin) (Figure 2.5b).  2.3.6 Cultured AA LNCs did not directly participate in the inflammation of the hair follicles Fluorescent microscopy revealed clustering of CM-DiI (red) labelled LNCs at the site of injection in both AA and control LNC-injected mice near the adipose layer of the skin four weeks post cell injection. However, the density of labelled LNCs in the injection site of AA mice was relatively lower compared to control mice (Figure 2.6a,b). In the peripheral skin of AA mice beyond the immediate site of injection, there was an absence of labelled LNCs among the inflammatory infiltrates around anagen stage hair follicles in AA lesions and in the control skin (Figure 2.6c,d). More of the labelled AA LNCs were localized in the skin-draining lymph nodes of mice with active AA compared to control mice without AA (Figure 2.6e,f). Conversely, more of the labelled control LNCs were found in the spleen of control mice without AA (Figure 2.6g,h).   51   Figure 2.5 Small numbers of CM-DiI labelled LNCs were detected in single-cell suspensions extracted from the skin at the site of injection, peripheral skin and skin-draining lymph nodes, at all three time points. Only around 0.2% of total LNCs had CM-DiI labelling but over 95% of the labelled LNCs also expressed CD3. No labelled cells could be detcted from the spleens of the same mice (Spn)  by flow cytometry (a). In the skin at the site of injection (iSkin) and peripheral skin away from the injection site (pSkin), a small percentage of cells were labelled but did not show a significant difference between time points (b). n=1 for each time point.  52   Figure 2.6 Injected LNCs could be found localized at the site of injection even after four weeks, but did not participate in the inflammation of HFs in AA lesions. CM-DiI labelled LNCs were found at the site of injection (iSkin) for both AA and control mice (a,b) but not in the peripheral skin (pSkin) away from the injection (c,d). A small number of labelled LNCs were found in the skin draining lymph nodes (LN) of the AA LNC injected mice (e,f) while more labelled LNCs were found in the spleen (Spn) of control LNC injected mice (g,h).   53  2.4 Discussion The use of a cultured cell-transfer technique alleviates many of the complications associated with skin-grafting not only by minimizing procedures performed on the mice, reducing technical training required, and shortening the recovering time, but also by reducing the number of donor mice required for the procedure. The cell-transfer method also delivered very consistent result such that at least 90% of the mice developed AA within 20 weeks over multiple trials. Using cultured LNCs has advantages over using freshly isolated LNCs by being more cost-effective; we obtained 8-10 fold increase in cell number over 6-days of culture. With this, the number of cells from one donor after a single round of cell expansion was enough to potentially induce AA development in around 50 naïve mice; a significant improvement over the skin-grafting AA transfer method. Among the control mice that received cultured LNCs from a donor without AA, one developed AA. As C3H/HeJ mice spontaneously develop AA at a low rate191, it is not surprising that some control mice developed AA. In previous studies, some mice sham-grafted with control skin have also developed AA (unpublished) but the incidence was significantly lower compared to AA skin grafted mice and in line with spontaneous AA development expectations. Therefore, the development of AA in the mouse that received cultured control LNCs may be coincidental. In mice that have developed AA, the hair loss in general began on the ventral side and later progressed to the site of injection. This pattern of hair loss could indicate that CD4+ T-cells may be driving the systemic hair loss as supported by previous observations with the injection of CD4+/CD25- T-cells98. From our observation, there seemed to be no difference in 54  the length of the latent period before AA onset with injection of LNCs into either anagen or telogen skin.  The H&E staining of the skin lesion from AA affected mice showed a typical “swarm of bees” infiltration of lymphocytes around the hair follicles, the hair fibres in these lesions were also in a dystrophic state similar to AA mice induced via skin-grafting as well as to human AA lesions. This observation confirmed the phenotype similarity of AA mice induced by cell-transfer compared to skin-grafting and therefore can be used similarly to study human AA. The viability of AA specific T-cell populations are potentially enhanced by the addition of IL7 and IL15. While IL2 is necessary for overall T-cell growth, IL7 is an important growth factor for lymphoid cell survival as it supports the growth of both memory CD8+ T-cells and CD4+/CD25+ helper T-cells205,206. In addition, IL15 is also an important growth factor for CD8+ memory T-cells as well as CD4+ effector memory T-cells207,208.  The inclusion of IL7 and IL15 on top of IL2 and the stimulant Dynabeads showed more rapid growth of LNCs and yielded higher cell number compared to just IL2 and Dynabeads alone. The result we obtained is similar to another group who attempted to use this combination of cytokines and stimulants to expand memory T-cells209 where T-cell cultures with IL2, IL7 and IL15 showed better growth than just with single cytokines. As expected, the level of CD4+/CD25+ T-cells were significantly elevated after expansion in both AA and control LNCs.  The rate of CD4+ and CD8+ T-cells expressing IFNγ after expansion was similar between AA and control mice LNCs but significant elevation of CD4+/IFNγ after expansion 55  was observed in control mice LNCs. This result indicates that there is no difference in terms of the abilities of T-cells respond to stimulation and proliferation in control mice compared to AA mice. This potentially suggest that additional sub-population(s) of T-cells are required to drive AA development44,45,98. There were no significant difference in the pattern of CD4+ and CD8+ T-cells expressing IL17 in AA and control mice LNCs. Since both AA and control mice LNCs displayed similar profile of CD4+ and CD8+ T-cells expressing IFNγ, CCR7 and IL17 before and after expansion, additional markers may need to be evaluated to determine which population(s) of T-cells are important for AA development. Potentially, the culturing process could promote survival and enrichment of special T-cell subset(s) like the CD8+/NKG2D+ T-cells, enabling the transfer of AA89. The cell expansion protocol employed in this study does not target Treg cells, different polarizing cytokines or factors are required to selectively and effectively expand Tregs213,214. However, control LNCs after expansion showed significantly higher percentage of Tregs compared to AA LNCs. Due to the unspecific activation mechanism of the protocol, it is possible the level of Tregs in the control mice LNCs before and after expansion were maintained rather than expanded (as no significant elevation compared to before expansion was observed). The Tregs present in the cultured LNC population may potentially maintain a certain level of regulatory cytokines215 that help C3H/HeJ mice stay resistant to AA development115. Quantification of regulatory cytokines in the conditioned media of AA and control LNCs cultures may be able to confirm whether control LNCs display a more Treg profile compare to AA LNCs. This was not performed as we did not expect to see significant changes in Treg levels with this procedure. 56  We investigated the fate of injected cells as well as any potential difference in the proportion of CD4+ and CD8+ T-cells after 3, 7, and 11 days. With flow cytometry, we found a very small number of cells that were labelled with CM-DiI in the skin draining lymph nodes (0.1-0.2 %). In the LNCs, over 95% of the labelled cells in both AA and control LNCs were CD3+ as expected. Interestingly, labelled CD4+ T-cells decreased and CD8+ T-cell increased as time progressed, however significance cannot be determined because only one mouse was studied per group in each time point. In the skin at the site of injection 0.5-2.5% of lymphocyte population was CM-DiI labelled; higher than in the lymph nodes. However, the attempts to stain CD3, CD4 and CD8 in the skin were unsuccessful possibly due to the destruction of the surface marker during tissue preparation. There were no obvious trends with the number of CM-DiI labelled cells in the skin at the site of injection over time. No labelled cells were detected in the peripheral skin away from the site of injection as well as in the spleen.  A mouse that received CM-DiI labelled AA LNCs were sacrificed at 2 weeks and 4 weeks when they first started to develop AA to compare with mice that received labelled control LNCs and did not develop AA. In both groups of mice, the injected LNCs could easily be spotted at the site of injection even after four weeks. However, there appeared to be less labelled AA LNCs at the site of injection compared to labelled control LNCs. Potentially, the AA LNCs may have higher activity and undergo higher rates of proliferation compared to the controls; this would result in dye dilution as CM-DiI intercalates into the lipid bilayer of cells216. More interestingly, in the peripheral skin away from the injection site, no labelled LNCs were observed for both AA and controls; consistent with flow cytometry results. Even though inflammatory infiltrates were observed around the hair 57  follicles in the H&E staining and IHC of AA lesions, it appeared that the injected LNCs did not directly participate in the inflammation of hair follicles. Therefore, taking the current flow cytometry and fluorescent microscopy results together, the cultured AA LNCs may be interacting with the host immune system and result in a cascade of events that lead to onset of AA mediated by host cells. By comparing the skin-draining lymph nodes and spleen of AA and control mice, it was found that relatively more labelled AA LNCs were found in the skin-draining lymph nodes while labelled control LNCs were mostly found in the spleen. This suggested that there were populations of activated antigen-specific T-cells in the cultured AA LNCs that migrated to the host skin-draining lymph nodes (also resulting in swelling of the lymph nodes). Conversely, there may be less activated antigen-specific T-cells within the control LNCs, causing them to go into the blood stream and to eventually be filtered into the spleen. In this study, we successfully cultured skin-draining LNCs isolated from AA affected and control C3H/HeJ mice. The cell-transfer method allowed us to induce AA development in more mice with a single donor in a relatively simple and consistent manner compared to previous AA induction methods.  In addition, new applications were made possible with this method. We have identified that only cultured AA LNCs can induce AA development consistently, it is therefore possible to stimulate these AA LNCs with specific antigen epitopes instead of Dynabeads to obtain specific antigen-specific T-cell clones. These clones of T-cells can then be expanded further to evaluate their ability to induce AA development. More studies will still need to be done to elucidate the mechanisms of AA induction by cultured LNCs after transfer but this method potentially provides applications beyond AA research. The method may be incorporated into a protocol to study the response of immune-58  modulatory treatments in other cell-mediated autoimmune disease mouse models. Potentially, it can be utilized as an alternative method to generate rodent models for other cell-mediated autoimmune diseases such as vitiligo217, rheumatoid arthritis218 or even multiple sclerosis219,220. Current rodent models for these diseases may be developed by complicated surgical procedures, transgenic modifications, or induction using chemicals. By culturing and injecting immune cells into histocompatible hosts, it may minimize the labours needed to generate these models, making them more economical and accessible to more laboratories. Successful transfer of autoimmune diseases with a cell injection method could also improve the resemblance of pathogenesis to the natural progression and biological properties of the diseases in humans.    59  Chapter 3: Identification of Autoantigen Epitopes in Humans and the C3H/HeJ Mouse Model of Alopecia Areata  3.1 Introduction Alopecia areata (AA) is considered as one of the most common autoimmune diseases in humans with a with a lifetime risk of 1.7% in the United States50. It is characterized by the shedding of small, circular patches of scalp hair in the acute phase, but can develop into more severe forms such as alopecia totalis (loss of all scalp hair) or even alopecia universalis (loss of all body hair)54. AA skin lesions are characterized by the clustering of primarily CD4+ (helper T-cells) and CD8+ (cytotoxic T lymphocytes; CTLs) around the hair follicles (HFs)77,78. CTLs were found to be able to infiltrate into the HFs based on previous observations with immunohistology and electron microscopy43,138,221. The functional importance of T-cells in the induction of hair loss in humans is inferred from several observations such as the recovery of hair growth for some patients when treated with immuno-suppressants and the reduction of the “swarm of bees” infiltration of CTLs in the recovered HFs1,46,53.  The cytotoxic nature of infiltrating CTLs may cause and/or perpetuate hair follicle disruption observed in AA lesions. In the DEBR rodent model of AA, in vivo depletion of CTLs can restore hair growth44 while depletion of CD4+ T-cells resulted in partial regrowth of hair45 indicating the importance of both CTLs and CD4+ helper T-cells (Th) cells in the persistence of AA. The induction of AA in C3H/HeJ mice can be achieved via transfer of 60  syngeneic T-cells as demonstrated in Chapter 2. Furthermore, transfer of human PBMCs to scalp explants on SCID mice can also induce AA in the humanized AA mouse model190. However, while the role of CTLs is evident in AA rodent models, with a strong indication of autoreactive T cell activation by specific antigens139,222, how and why these cells are involved in the progression of human AA is still unclear. In addition, the exact autoantigen epitopes that are involved in eliciting autoreactive CTL responses in AA are still debated.  Various studies have implied that certain HF melanocyte and keratinocyte derived antigens may be involved in the pathogenesis of AA84,87,223. It has been reported that AA patients with a mixture of white and pigmented hair shed primarily pigmented hairs during the onset of AA and new hair fibers that develop after the resolution of AA tend to be non-pigmented224. Studies conducted by Gilhar et al showed the development of an AA-phenotype in SCID mice with human AA scalp explants by stimulating scalp T-cells with melanocyte-associated T-cell epitopes139. However it still remains unclear whether the T-cells stimulated by the epitopes are the same clones as those normally involved in spontaneous human AA, and whether the melanocyte antigen specific T-cells present in the scalp explants have direct contact with HF melanocytes in the diseased lesions.  Electron microscopy of AA lesions have revealed a close association of infiltrating T-cells with the root sheaths of the HFs where keratinocytes are present84,138. This circumstantially suggests that keratinocyte-associated T-cell epitopes could potentially be targeted by autoreactive T-cells. Further, it has been found that there is an elevation of auto-antibodies against trichohyalin and cytokeratin 16, in the serum of AA patients88. Sera from AA affected human subjects did not alter hair growth in scalp transplants from the respective 61  AA human subjects when grafted onto nude mice114. However, the identification of auto-antibodies to keratinocyte derived antigens suggests these are stimulatory to the immune system and they might also be targeted by CTLs. In this study, we utilized the collective knowledge of potential AA-associated antigens presented in previous studies to identify candidate human autoantigen epitopes that may be able to induce high frequency CTL responses in AA-affected human subjects as compared to subjects without inflammatory hair loss diseases. Furthermore, we investigated the potential cytotoxicity effect on primary HF root sheath keratinocyte cells associated with this CTL activation. We also analyzed a panel of AA autoantigen epitopes in C3H/HeJ mouse model of AA, using mouse homologues of the antigens used in human study, to compare whether AA-affected mice show similar antigen targets as in humans.  Successful identification of antigen epitopes in humans that are able to induce significant responses from AA subjects’ CTLs could lead to the development of screening procedures and/or specific therapeutic modalities targeting antigen specific CTLs instead of using unspecific immunosuppressive therapies. If similar antigen epitope targets are identified in AA-affected C3H/HeJ mice, further applications could potentially be developed such as investigating the effects of expansion or depletion of autoantigen epitope-specific CTLs on AA progression. 62  3.2 Materials and Methods 3.2.1 Study subject recruitment and blood cell isolation Study participants gave their consent for the study according to University of British Columbia Clinical Research Ethics Board approved protocols. AA affected subjects and controls were recruited from Dr. Jerry Shapiro’s Hair Clinic in the Skin Care Centre of Vancouver General Hospital and Vancouver General Hospital - Gordon and Leslie Diamond Health Care Centre Clinical Research Unit. AA subjects ranged from those with less than 25%, patchy hair loss to subjects with alopecia totalis (AT) and/or alopecia universalis (AU). Control subjects were screened to exclude those with inflammatory hair loss such as cicatricial alopecia and comprised individuals with either no hair loss or androgenetic alopecia.  Investigators completed a questionnaire (Appendix A, B) for each individual regarding their hair loss and background information at the time of blood collection. Peripheral blood (20 mL) was collected by venous puncture into EDTA-coated vacutainers and processed. Peripheral blood mononuclear cells (PBMCs) were obtained by the Ficoll-paque density gradient centrifugation method. Whole blood was diluted 50% with complete medium R10 comprised of RPMI 1640 (catalog #11875-093) 10% FBS (catalog #16000-044), 2 mM Glutamax (catalog #35050-061), and 100 U/mL Streptomycin with 100 μg/mL Penicillin (catalog #15140-148); all from Invitrogen (Burlington, ON), and overlaid on an equal volume of Ficoll-Paque Plus (GE Health Care, Burnaby, BC. Catalog #17-1440-03) and centrifuged at 600x g for 20 minutes with the break set to off to prevent disruption of the buffy coat interphase. The buffy coat layer containing the PBMCs was pipetted into clean 63  centrifuge tubes and washed twice with R10 at 350x g for 15 minutes. Isolated PBMCs were aliquoted into cryo-vials and slowly frozen in storage medium (90% FBS, 10% DMSO) at -80 °C overnight and transferred to liquid nitrogen until use. 3.2.2 Generation of AA mice and isolation of lymph node cells The C3H/HeJ mouse model for AA was utilized with the approval of the University of British Columbia Animal Care Committee. All female C3H/HeJ mice were supplied from colonies at The Jackson Laboratory (Bar Harbor, ME, USA) specific pathogen-free production facility. Normal haired C3H/HeJ mice were induced with AA by skin grafting as described previously96. Briefly, AA affected mice were completely shaved, sacrificed and full-thickness skin sections about 1 cm diameter were excised. The recipient mice were shaved on their back and a piece of full-thickness skin (1 cm diameter) was removed; the AA affected skin was grafted on this site. The mice were observed for the development of AA. Normal haired, age matched, sham-grafted C3H/HeJ littermates were used as controls. Skin draining lymph nodes were obtained from the AA mice with >80% hair loss and control mice without AA. The lymph node cells (LNCs) were isolated using a single-cell suspension method as described in Chapter 2. 3.2.3 Subject sample haplotyping Both AA and control subjects with HLA-A*0201 were identified for further analysis. The HLA-A*0201 haplotype was selected because it is the most predominant allele present over 50% of the general population across all ethnic groups225. PBMC samples were stained with FITC conjugated monoclonal antibody against human HLA-A*0201 (Clone BB7.2; BD Pharmagen, Mississauga, ON. Catalog #551285), following standard single-color flow 64  cytometry protocols, with a FACSCanto II flow cytometer (BD Bioscience, Mississauga, ON). PBMCs were resuspended in FACS staining buffer (PBS with 0.1% FBS) at a density of 1 million cells/mL. The cells were centrifuged for 5 minutes at 300x g and supernatant removed (washing). HLA-A*0201 antibody was diluted 25x and 100 µL was used to resuspend the PBMC pellets. The PBMCs were stained for 30 minutes in the dark, on ice, and washed again with staining buffer. Fixation of PBMCs was performed with 100 µL of BD CytoFix/CytoPerm Buffer (BD Pharmagen. Catalog #554714) for 15 minutes in the dark, on ice. After fixation, the PBMCs were washed and resuspended again in 400 uL staining buffer until read under a FACSCanto II flow cytometer. 3.2.4 Peptide sequence prediction and synthesis Computer-based online software programs were used to predict the nonamer peptide sequences of antigens that would bind to a common MHC-I allele, HLA-A*0201. The combined results from two algorithms, SYFPEITHI from the University of Tuebingen (http://www.syfpeithi.de)226 and Matrix-assisted algorithms from the National Institutes of Health, Center for Information Technology, BioInformatics and Molecular Analysis Section (BIMAS; http://www-bimas.cit.nih.gov/molbio/hla_bind/)227, were used to design a panel of candidate peptide sequences. The score of each peptide from both prediction algorithms were multiplied together to give rise to a list of peptide sequences with highest avidity. All peptides were synthesized (Kinexus Bioinformatics, Vancouver, BC) in 2 mg, crude quality (between 60-80% purity) for screening purposes. The quality of peptides was controlled by high-performance liquid chromatography and mass spectrometry, the purity of the peptide was defined by the percent of correct target peptide in the total content. The list of peptides 65  and their sequences are listed in Table 3.1. Epitope peptide prediction for C3H/HeJ mice was performed using the same method except the protein sequence of mouse homologues were used with reactivity restricted to H-2Kk (C3H/HeJ mouse MHC-I) only, the list of peptides and their sequences are listed in Table 3.2.  66  Table 3.A.1 List of human autoantigen epitope peptide sequences used. Identity Position Sequence BIMAS SYFPEITHI Score Trichohyalin-1 (TCHH-1) 35 NLLEREFGA 437.114 19 8305.17 Trichohyalin-2 (TCHH-2) 835 LLQEEEEEL 72.959 26 1896.93 Trichohyalin-3 (TCHH-3) 4 FLLFIFKVA 65.869 15 988.04 Trichohyalin-4 (TCHH-4) 778 QLQEEEDGL 42.917 23 987.09 Trichohyalin-5 (TCHH-5) 895 KLQQKEEQL 36.637 22 806.01 Trichohyalin-6 (TCHH-6) 52 KTVDLILEL 9.793 24 235.03 Trichohyalin-7 (TCHH-7) 182 VLRKEEEKL 1.352 23 31.10 Trichohyalin-8 (TCHH-8) 55 DLILELLDL 3.685 26 95.81 MelanA/MART1 analog-1 (Ala26) (MEL-1) 26 AAAGIGILTV - - - MelanA/MART1 analog-2 (MEL-2) 27 AAGIGLTV - - - MelanA/MART1 analog-3 (Leu27) (MART1)228 26 ELAGIGILT - - - Tyrosinase-1229 (TYR-1) 1 MLLAVLYCL 309.050 27 8344.35 Tyrosinase-2229 (TYR-2) 369 YMNGTMSQV 531.455 23 12223.47 Tyrosinase Related Protein 2 (TYRP2)230 180 SVYDFFVWL 973.849 21 20450.83 Modified GP100-1 209m231 (GP100-1) 209 IMDQVPSFV - - - Modified GP100-2 280.9v231 (GP100-2) 280 YLEPGPVTV - - - SIY-Kb 2C TCR (2C)  SIYRYYGL - - - Cytokeratin 16-1 (KRT16-1) 402 ILLDVKTRL 550.915 25 13772.875 Cytokeratin 16-2 (KRT16-2) 111 GLLVGSEKV 126.098 27 3404.646 67  Identity Position Sequence BIMAS SYFPEITHI Score Cytokeratin 16-3 (KRT16-3) 127 RLASYLDKV 78.385 29 2273.165 Melanoma-Associated-Antigen 3 (MAGE3) 195 IMPKAGLLI 12.809 19 243.371 Pro-opiomelanocortin-1 (POMC-1) 16 LLLQASMEV 437.482 26 11374.532 Pro-opiomelanocortin-2 (POMC-2) 12 LLLALLLQA 71.872 24 1724.928 Tyrosine Hydroxylase isoform B-1 (THB-1) 299 FLASLAFRV 1855.647 24 44535.528 Tyrosine Hydroxylase isoform B-2 (THB-2) 365 KLSTLYWFT 1613.824 16 25821.584 Tyrosine Hydroxylase isoform B-3 (THB-3) 292 GLLSARDFL 213.015 23 4899.345 Cytomegalovirus pp65 (CMV)232 495 NLVPMVATV - - - Epstein–Barr virus BMFL1 (EBV)232 259 GLCTLVAML - - - Influenza Matrix 1 (FLU)232 58 GILGFVFTL - - - Human Immunodeficiency Virus p17 (HIV)233  77 SLYNTVATL - - - Hepatitis C Virus (HCV)234 132 DLMGYIPLV - - -    68  Table 3.A.2 List of mouse autoantigen epitope peptide sequences used. Identity Position Sequence BIMAS SYFPEITHI Score Trichohyalin-1 (TCHH-1) 1232 EERLRDRKI 1000 25 25000 Trichohyalin-2 (TCHH-2) 1074 EERLRDSKI 1000 24 24000 Trichohyalin-3 (TCHH-3) 1520 EEQFARDTI 1000 23 23000 Trichohyalin-4 (TCHH-4) 363 AEEDELTRI 500 21 10500 Trichohyalin-5 (TCHH-5) 1560 QEEQRRRQI 500 21 10500 Trichohyalin-6 (TCHH-6) 523 EELREERLL 80 15 1200 Trichohyalin-7 (TCHH-7) 1411 REELLHRQV 50 20 1000 Cytokeratin 16-1 (KRT16-1) 167 IEDLKSKII 500 24 12000 Cytokeratin 16-2 (KRT16-2) 168 EDLKSKIII 400 21 8400 Cytokeratin 16-3 (KRT16-3) 204 NELFLRQSV 100 21 2100 Cytokeratin 16-4 (KRT16-4) 135 EEANRDLEV 100 20 2000 Cytokeratin 16-5 (KRT16-5) 394 QEYNILLDV 100 19 1900 MelanA/MART1-1 (MART-1) 22 AEEAAGIGI 500 22 11000 MelanA/MART1-2 (MART-2) 56 MDKRRHIGI 100 18 1800 Tyrosinase-1 (TYR-1) 422 RDSYMVPFI 100 20 2000 Tyrosinase-2 (TYR-2) 279 SEEYNSHQV 50 21 1050 Tyrosinase related protein-1 (TYRP-1) 471 QEFTVSEII 1000 22 22000 Tyrosinase related protein-2 (TYRP-2) 476 SEIITIAVV 50 20 1000 GP100 (GP100) 362 SEQMLTSAV 100 19 1900   69  Identity Position Sequence BIMAS SYFPEITHI Score M. tuberculosis Culture filtrate protein (CFP10)235 32 VESTAGSL - - - Borna disease virus (BDV)236 129 TELEISSI - - - Circumsporozoite protein precursor (CSP)237 375 YENDIEKKI - - - Influenza A virus (FLU)236 259 FEANGNLI - - - HIV-1 reverse transcriptase (HIV)238 33 TEMEKEGKI - - - Simian virus 40 (SV40)239 560 SEFLLEKRI - - - M. tuberculosis TB10.3/4 (TB10)240 20 GYAGTLQSL - - - Bovine insulin B chain (INS)241 7 CGSHLVEAL - - -    70  3.2.5 HLA-A*0201 stabilization assay T2 cells (174xCEM.T2) from the American Type Culture Collection (ATCC, Manassas, VA. Catalog #CRL-1992) were used as an initial screening to confirm stable presentation of each peptide by HLA-A*0201 complex in vitro. T2 cells are TAP-1 and 2 (transporter associated with antigen processing) deficient and only express stable HLA-A2*0201 in the presence of high-affinity peptides, therefore HLA-A*0201 stability correlates to peptide biological activity and affinity in vitro. T2 cells (500,000 cells/600 uL) were cultured in R10 with 1 μg/μL β2-Microglobulin (Sigma-Aldrich, Oakville, ON. Catalog #M4890) as well as 50 μg/mL of each of the peptides for 18 hours at 37 °C, 5% CO2. After 18 hours, brefeldin A (5 µg/mL; Affymetrix eBioscience, San Diego, CA. Catalog #00-4506-51) was added and incubated for another 3 hours at 37 °C. The cells were then washed with staining buffer, incubated with HLA-A*0201 antibody (Clone BB7.2), and flow cytometry was performed as outlined above. HLA-A*0201 restrictive Influenza M1 58-66 peptide (GILGFVFTL) was used as a positive control to indicate maximum HLA-A*0201 expression levels. The stabilization values (Mean Fluorescence Intensity, MFI) for each of the epitope peptides were normalized with reference to maximum HLA-A*0201 expression induced by the high-affinity positive control peptide. 3.2.6 Human PBMC IFNγ ELISpot assay The numbers of auto-antigen specific CTLs from frozen human PBMC samples were quantified with IFNγ ELISpot assays. ELISpot plates (Millipore, Billerica, MA. Catalog #MAIPN4550) were coated with IFNγ capture antibody (15 μg/mL) from Mabtech (Clone 1-D1K; Cincinnati, OH. Catalog #3420-2A) overnight at 4 °C. The coated ELISpot plates were 71  washed five times with PBS and blocked with AR10 (Advanced RPMI 1640 with the same supplements as R10, Invitrogen. catalog #12633-012) for two hours at 37 °C, 5% CO2. At the same time, PBMC samples were thawed and washed once with AR10 (350 x g for 7 minutes) and resuspended into 2 million cells/mL with AR10. 100 μL of cell suspension (200,000 cells) were seeded into each well on the ELISpot plate. Different peptides were added to each well at the final concentration of 5 μg/mL and incubated at 37 °C 5% CO2 for 48 hours. The ELISpot plate was then washed with PBS to remove the cells and incubated with biotinylated anti-IFNγ antibody (Mabtech; clone 7-B6-1) for 2 hours at room temperature. The plate was washed again with PBS and incubated with streptavidin-alkaline phosphatase (Mabtech) for 1 hour at room temperature. Finally, the plate was washed again with PBS and developed with AP conjugate substrate kit (BCIP/NBT) solution (Bio-Rad; Hercules, CA. Catalog #170-6432); the reaction was stopped by rinsing with tap water after spots developed (around 20 minutes). The number of spots in each well were quantified with an AID iSpot FluoroSpot Reader System (Autoimmun Diagnostika GmbH, Strassberg, Germany. Catalog #ELR07IFL). The number of spots denoted the number of CTLs activated against the respective peptide treatments. The cut-off for positive responses against each epitope peptide was defined by 3x standard error above the mean number of the spot-forming cells (SFCs) of negative control epitopes (HIV and HCV) after correcting for the background control (no peptide stimulation). As the samples have non-normal distributions, a more powerful nonparametric test, the Mann-Whitney U-test, was used to determine significance. 72  3.2.7 Mouse LNC IFNγ ELISpot assay The numbers of auto-antigen specific CTLs from mouse LNC samples were quantified with Mouse IFNγ ELISpot assays. ELISpot plates (Millipore. Catalog #MAIPN4550) were coated with IFNγ capture antibody (15 μg/mL) from Mabtech (Clone AN18. Catalog #3321-2A) overnight at 4 °C. The coated ELISpot plates were washed five times with PBS and blocked with AR10 for two hours at 37 °C, 5% CO2. At the same time, frozen LNC samples were thawed and washed once with AR10 (350 x g for 7 minutes) and resuspended into 2 million cells/mL with AR10. 100 μL of cell suspension (200,000 cells) were seeded into each well on the ELISpot plate. Different peptides were added to each well at the final concentration of 5 μg/mL and incubated at 37 °C 5% CO2 for 48 hours. The ELISpot plate was then washed with PBS to remove the cells and incubated with biotinylated anti-IFNγ antibody (Mabtech; clone R4-6A2) for 2 hours at room temperature. The plate was washed again with PBS and incubated with streptavidin-alkaline phosphatase (Mabtech) for 1 hour at room temperature. Finally, the plate was washed again with PBS and developed with AP conjugate substrate kit (BCIP/NBT) solution; the reaction was stopped by rinsing with tap water after spots developed (around 20 minutes). The number of spots in each well was quantified with an AID iSpot FluoroSpot Reader System. The number of spots denoted the number of CTLs activated against the respective peptide treatments. The cut-off for positive responses against each epitope peptide was defined by 3x standard error above the mean number of the spot-forming cells (SFCs) of negative control epitopes (H-2Kb restrictive peptides; TB10.3/4 from Mycobacterium tuberculosis and bovine insulin B chain) after correcting for the background control (no peptide stimulation). Mann-Whitney U-test was used to determine significance. 73  3.2.8 Intracellular cytokine stain The production of IFNγ by the CTLs upon stimulation with various peptides was confirmed with intracellular cytokine stain (ICS) with additional surface staining for CD8+ and CD4+ cells. As with ELISpot assays, 100 μL of cell suspension (200,000 cells) were seeded into each well of 96-well U-bottom plates. Different peptides were added to each well at the final concentration of 5 μg/mL and incubated at 37 °C for 16 hours. Endoplasmic reticulum inhibitor brefeldin A (Affymetrix eBioscience) was added to the culture in the last 5 hours of stimulation to inhibit the secretion of IFNγ into the culture media. Following the stimulation, the PBMCs were stained with anti-human CD4 FITC (Clone: RPA-T4, Catalogue #11-0049) and anti-human CD8a APC (Clone: OKT8, Catalogue #17-0086) with corresponding isotype controls, Mouse IgG1 K FITC (Clone: P3.6.2.8.1, Catalogue #11-4714) and Mouse IgG2a K APC (Clone: eBM2a, Catalogue #17-4724). The PBMCs were washed and fixed with IC Fixation Buffer (Catalogue #00-8222-49) and stained with anti-human IFNγ PE (Clone: 4S.B3, Catalogue #12-7319) with Mouse IgG1 K PE (Clone: P3.6.2.8.1, Catalogue #12-4714) as isotype control. All reagents and antibodies were purchased from Affymetrix eBioscience. The labelled cells were analyzed with a BD FACSCanto II flow cytometer. Mann-Whitney U-test was used to determine significance. 3.2.9 Hair follicle micro-dissection and root sheath keratinocyte culture Fresh human scalp tissues were obtained as waste material from subjects undergoing hair transplants. Single hair follicle (HF) units were obtained by removing the fat layer; hair bulb and dermal sheath were removed subsequently to expose the outer root sheath (ORS) and inner root sheath (IRS) of the HFs. Each hair fiber with attached ORS/IRS was then 74  treated with TrypLE Express (Invitrogen. Catalog #12604-13) for 15 minutes to digest the tissue and allow the HF keratinocytes to dissociate into a single-cell suspension for culturing. After digestion, ORS/IRS cells from five HFs were combined together and cultured in each well of BioCoat collagen-I coated 24-well plates (BD Bioscience. Catalog #356408) in 50% Defined Keratinocyte-SFM (Catalog #10785012) combined with 50% EpiLife plus a HKGS kit without calcium (Catalog #MEPICF500 and S-001-K, respectively); all culture media and supplements were purchased from Invitrogen.   Colonies of HF keratinocytes formed after about a week, during which fresh media was added every two days to replenish the culture, and passaged into T25 or T75 BioCoat Collagen I Coated Vented flasks (BD Bioscience, Catalog #354485 and 354484 respectively) once cultures reached about 90% confluence. Every time the culture reached 90% confluence, the cultures were passaged into a bigger flask. At the end of passage five, the HF keratinocytes were transferred to BioCoat collagen-I coated 24-well plates again with a density of 30,000 cells/500 μL per well and cultured for three days prior to use (below). 3.2.10 Apoptosis induction via peptide-activated PBMC conditioned media PBMCs (2 million cells/mL) from AA and control subjects were cultured with 5 μg/mL of peptides that showed significant ability to activate AA PBMCs (TCHH cocktails, MART1, TYRP2) for three days. During the first one and half days, AR10 was used as the culture medium to provide optimal conditions to activate the PBMCs. The latter one and half days, AR10 was switched with K-SFM/EpiLife and treated with the same concentration of peptides. K-SFM/EpiLife is serum free which maintained HF keratinocyte viability and reduced apoptosis induced by the presence of FBS in AR10 (data not shown). At the end of 75  the third day, conditioned media (CM) from PBMCs stimulated with different peptides was transferred to the cultured HF keratinocytes and incubated for eight hours at 37 °C with 5% CO2. After eight hours, the culture supernatants were collected and HF keratinocytes were also collected after TrypLE Express treatment. The HF keratinocytes were combined together, washed (600x g for 5 minutes) and resuspended in 200 uL PBS. The keratinocytes were washed further and stained with Fixable Viability Dye eFluor 780 (1 µL/mL. Catalog #65-0865-14) for 30 minutes in the dark on ice. The keratinocytes were washed again with staining buffer and once more with binding buffer as provided from an Annexin V apoptosis detection kit (5 uL/100uL. Catalog #88-8007-72). Following washing, the keratinocytes were labelled with Annexin V for 15 minutes in the dark on ice. After staining, the cells were washed once with binding buffer and once with staining buffer and then fixed with Foxp3 fixation/permeabilization buffer (Catalog #00-5523-00) for 30 minutes in the dark on ice. After fixation, the cells were washed with 1x Foxp3 permeabilization buffer (Catalog #00-8333) and stained with anti-human Ki67 PE (Clone: 20Raj1, 1:20 dilution. Catalog #12-5699) for 30 minutes in the dark on ice. All antibodies were purchased from Affymetrix eBioscience. FACS analysis was performed on a BD FACSCanto II flow cytometer. Mann-Whitney U-test was used to determine significance. 3.2.11 Human cytokine array on cell culture supernatant from PBMCs stimulated with epitope peptides A portion of cell culture supernatant conditioned media (CM; 100 µL) from the above culture was frozen at -80 °C. The frozen CM samples were sent to RayBiotech Inc (Norcross, GA) for a quantitative analysis of inflammatory cytokine content in the CM. Quantibody 76  Human Cytokine Array Q1 (RayBiotech. Catalog #QAH-CYT-1) was used to analyze the CMs. Quantibody is a multiplex sandwich-based ELISA kit where an antibody array chip is coated with different capture antibodies specific for the cytokines of interest. After the samples and control standards were incubated on the array chip and washed to remove unspecific bindings, a cocktail of biotinylated detection antibodies were added to each wells. To initiate enzymatic reaction, streptavidin-conjugated fluor was added and visualized using a RayBiotech’s fluorescence laser scanner. To determine cytokine concentration, an 8-point standard curve for each target cytokine was generated from the array-specific protein standards, whose concentrations have been predetermined. The signals from the unknown samples were compared to the standard curve and concentration determined automatically by the Q-Analyzer Software and a report was generated. When the report was received by us, statistical calculations was performed by using GraphPad Prism 6 software (La Jolla, CA). The cytokines that can be detected with this array are: IL-1α, IL-1β, IL-2, IL-4, IL-5, IL-6, IL-8, IL-10, IL-12p70, IL-13, GM-CSF, GRO, IFNγ, MCP1, MIP1α, MIP1β, MMP-9, RANTES, TNFα, and VEGF-A. Fisher’s Least Significant Difference test was used to calculate the significance of difference in each pairwise comparison between the levels of cytokines expressed by AA and control PBMCs cultured with TCHH peptide cocktails. 3.3 Results 3.3.1 In silico designed peptides displayed varying degrees of affinity to HLA-A*0201 The panel of peptides used for screening was compiled with matrix-assisted algorithms BIMAS and SYFPEITHI. Highest ranking peptides predicted by both of the algorithms indicate the likelihood of them being potential epitopes to HLA-A*0201 complex 77  and were selected for further screening. As expected, some of the predicted peptide sequences did not have strong affinity to the HLA-A*0201 complex (as reflected by their low Mean Fluorescence Intensity; MFI) on T2 cells, despite their high computer algorithm scores, and vice versa. From the peptides tested, three of the eight peptides (2, 6, 8) from TCHH stabilized the expression of HLA-A*0201 at ≥50% of the maximum level stabilized by positive control FLU (influenza) peptide, one of the two melanocyte expressed GP100 peptides had over 90% stabilization while a single MART1 peptide was able to stabilize HLA-A*0201 better than the FLU peptide (>100%) as shown in Figure 3.1a and b. Conversely, several peptides derived from cytokeratin 16, POMC and tyrosine hydroxylase were eliminated due to their overall inability to stabilize HLA-A*0201 in vitro.  78   Figure 3.1 HLA-A*0201 stabilization assay performed on candidate epitope peptides revealed different ability to stabilize the expression of HLA-A*0201 on T2 cells. TCHH peptides 2, 6, and 8 had the ability to stabilize >50% of the binding affinity of the positive control FLU peptide (a). While the MART1 peptide had stronger affinity to HLA-A*0201 compared to the FLU peptide (b). High affinity peptides were most likely able to induce response in vitro and in vivo.   79  3.3.2 Trichohyalin, tyrosinase and tyrosinase-related-protein 2 peptides induced a higher frequency of CTL activation in AA PBMCs ELISpot assays were employed to quantify IFNγ secretion by PBMCs in response to epitope peptide groups (Figure 3.2a). The threshold for denoting positive responses against certain epitopes was defined as 3x SEs above the mean Spot Forming Cells (SFC) value for the negative control epitopes (HIV and HCV) after correcting for the background control (no peptide stimulation). Both negative control peptides and background control wells had very low numbers of SFCs produced by AA and control PBMCs indicating a low frequency of activation. The number of SFCs from AA PBMCs that were above the threshold was in general higher than control subject PBMCs (Figure 3.2b, c). However, peptide group TCHH-G1 (trichohyalin peptide 1 to 4) and TCHH-G2 (trichohyalin peptide 5 to 8) induced significantly higher frequencies of activated CTLs in AA PBMCs, resulting in higher numbers of SFCs (Figure 3.2b). In addition, melanocyte antigen derived epitopes TYR and TYRP2 were also able to induce higher numbers of SFCs in AA PBMCs versus control PBMCs (Figure 3.2c). In the control subjects, the number of SFCs induced by TYR was relatively higher compared to other peptides despite it still being significantly lower than the number of SFCs in AA patients. This could imply that TYR may not be a specific target for autoreactive CTLs involved in AA pathogenesis.   80   Figure 3.2 TCHH, TYR and TYRP2 peptides significantly increased frequencies of AA PBMC activation. A representative picture from an ELISpot plate is shown, each single spot was a result of IFNγ production and represents one single activated PBMC (a). Both groups of TCHH peptides, TCHH-G1 and TCHH-G2, were able to induce significantly higher frequencies of PBMC activation compared to similarly stimulated control PBMCs (b). Melanocyte derived antigen epitopes TYR and TYRP2 were also able to induce significantly higher frequencies of AA PBMC activation (c). The threshold line denotes true positive activation and was determined by 3x SE of SFCs from the negative controls. Significance of difference was determined by nonparametric Mann-Whitney U-Test where *=p<0.05.   81  3.3.3 Antigen epitope peptide activation of PBMCs were affected by the extent and duration of AA as well as concurrent treatments ELISpot assay results for AA subjects were stratified based whether the subject was undergoing treatment for AA at the time of blood collection. Subjects who were receiving treatments revealed overall lower frequencies of PBMC activation compared to those who were not on any treatments at the time of blood collection (Figure 3.3a). When the subjects were grouped based on the length of the current episode of AA, the trend showed that the patients whose most current AA episode was less than two years had higher frequencies of PBMC activation compared to those with greater than two years (Figure 3.3b). Furthermore, AA subjects with body hair and nail involvements revealed overall higher frequencies of PBMC activation compared to those who only had AA on the scalp (Figure 3.3c). Investigation into the ability of individual trichohyalin peptides to induce activation of PBMCs yielded variable results and no significant difference was observed between AA and control PBMCs (Figure 3.3d).   82   Figure 3.3 AA patients not currently receiving treatments, have recent onset and more extensive body hair/nail involvements have higher frequencies of PBMC activation while PBMCs stimulated with individual trichohyalin peptides did not show significant difference. AA patients who were not receiving any treatments at the time of blood collection showed overall higher frequencies of PBMC activation after peptide stimulation compared to those who were being treated (a). In addition, AA patients who developed AA within the past 2 years showed overall higher PBMC activation compared to those with more long-standing AA (b). Furthermore, patients with more chronic forms of AA such that body hair loss and nail abnormalities were observed showed overall higher frequencies of PBMC activation (c). However, stimulation of AA and control subject PBMCs with individual trichohyalin peptides did not yield significant difference unlike when stimulating as peptide cocktails (d). Significance of difference was determined by nonparametric Mann-Whitney U-Test where *=p<0.05.   83  3.3.4 IFNγ Intracellular cytokine staining (ICS) confirmed CTL activation after stimulation by specific peptides ICS performed on randomly selected PBMC samples confirmed the results obtained from ELISpot assays. The number of CD8+/IFNγ+ cells increased after stimulation with specific peptides in AA subjects (Figure 3.4a). In the subjects tested, TCHH-G1 treated AA PBMCs resulted in significantly higher numbers of CD8+/IFNγ+ cells over baseline (Figure 3.4b). While other peptides were not able induce statistically significant increases of CD8+/IFNγ+ cells in AA PBMCs, the pattern of response levels was similar to that of ELISpot assays. TYR peptides stimulation was able to induce very high levels of CD8+/IFNγ+ cells in both AA and control subject PBMCs similar to ELISpot assay results which suggested responses to TYR epitopes are not AA specific.   84   Figure 3.4 Intracellular cytokine stain confirmed CD8+ T-cell specific activation by the synthesized peptides. Representative flow cytometry scatter plots of ICS showed the CD8+/IFNγ+ population of cells in AA and control PBMCs after peptide stimulation (a). The percentage of CD8+/IFNγ+ cells in AA PBMC populations after stimulation with peptides were higher than the control PBMCs (b). TYR peptides showed stimulation of CD8+ (CTLs) to express IFNγ in both AA and control subjects. Significance of difference compared to baseline (no treatment) was determined by nonparametric Mann-Whitney U-Test where *=p<0.05.   85  3.3.5 Trichohyalin peptides induced AA PBMCs to secrete pro-apoptotic factors harmful to primary HF keratinocytes Conditioned media (CM) derived from peptide stimulation of CTLs from AA or control subjects had no significant differential effect on the expression of proliferation marker, Ki67, in keratinocytes treated with either AA or control subject CTL CM (Figure 3.5a, b). However, AA PBMC cell derived CM was able to induce higher expression of early (Annexin V) and late stage (Fixable viability dye) apoptosis markers in HF keratinocytes in general compared to control subject PBMC cell derived CM as shown in the representative flow cytometry results (Figure 3.5a). TCHH-G1 (peptide 1-4) and TCHH-G2 (peptide 5-8) were able stimulate AA CTLs to secrete soluble factors into CM that were able to induce significantly higher expression of Annexin V in HF keratinocytes (Figure 3.5c). A similar trend was observed with the level of fixable viability dye (indication of late stage apoptosis) in HF keratinocytes but statistical significance was not achieved (Figure 3.5d). 3.3.6 Differential release of inflammatory cytokines into culture media by AA and control PBMCs after trichohyalin peptide stimulation A portion of CM from the cultures of PBMC stimulated with trichohyalin peptides were collected to perform a cytokine array to determine the level of various inflammatory cytokines. Signature inflammatory cytokines such as IFNγ and TNFα were not detected however, IL-13 and CCL5 (RANTES) was generally elevated in the CM of AA PBMCs stimulated with both TCHH-G1 and TCHH-G2 (Figure 3.6a). Conversely, CCL4 (MIP1β) and MMP9 were generally higher in the CM of control PBMCs stimulated with TCHH peptides (Figure 3.6b). 86   Figure 3.5 CM derived from AA PBMCs stimulated with TCHH-G1 peptides can induce apoptosis in HF keratinocytes. The evaluation of the expression of proliferation (Ki67), early stage apoptosis (Annexin V), and late stage apoptosis (Fixable Viability Dye, FVD) markers on HF keratinocytes by flow cytometry (a). The effect of CM from both AA and control PBMCs stimulated with different peptides were variable and showed no significant difference between the two subject groups (b). CM from TCHH-G1 and TCHH-G2 stimulated AA PBMCs increased Annexin V+/FVD- HF keratinocyte numbers significantly compared to CM from control PBMCs; indicating higher ability to induce early stage apoptosis (c). Keratinocytes in late stage apoptosis (Annexin V+/FVD+) after culture with either AA or control CM showed no statistically significant difference, but exhibited similar trend as observed for early stage apoptosis (d). Significance of difference was determined by Fisher’s Least Significant Difference test where *=p<0.05.   87   Figure 3.6 AA PBMCs secreted more IL-13 and RANTES upon trichohyalin peptide simulation while control PBMCs secreted more MIP1β and MMP9. Analysis of CM after PBMCs were stimulated by trichohyalin peptides via cytokine array revealed AA PBMCs secreted more IL-13 and RANTES after stimulated by both TCHH-G1 and G2 (a). However, control PBMCs also secreted certain inflammatory cytokines such as MIP1β and MMP9 at a higher degree compared to AA PBMCs (b). Significance of difference was determined by Fisher’s Least Significant Difference test where *=p<0.05.   88  3.3.7 Mouse homologues of cytokeratin 16 and melanoma antigen recognized by T cells 1 (MART1) induced higher frequencies of AA LNC activation Mouse IFNγ ELISpot assays revealed higher frequencies of CD8+ T-cell activation within the LNCs isolated from AA-affected C3H/HeJ mice after stimulation with cytokeratin 16 peptide group 1 (KRT16-G1; peptide 1-4) and melanoma antigen recognized by T cells 1 (MART1) peptides (Figure 3.7). LNCs from AA-affected mice generally showed higher frequencies of activation compared to the sham-grafted control mice. Similar to human IFNγ ELISpot assays, TCHH peptides also induced higher frequencies of activation but the difference was not statistically significant (Figure 3.7a). Similar to AA human patients, TYR peptides stimulated LNCs showed higher frequencies of activation albeit not significant in mice (Figure 3.7b). Unlike human cells, there were higher frequencies of activation from GP100 peptide stimulated AA mouse LNCs, but the difference was not significant (Figure 3.7b).   89   Figure 3.7 KRT16-G1 and MART peptides significantly increased the frequencies of LNC activation in AA mice compared to the controls. One of the two groups of KRT16 peptides was able to significantly increase the frequencies of activation from AA LNCs when compared to similarly stimulated control sham-grafted mice LNCs (a). TCHH-G1, G2 and KRT16-G2 also showed higher frequencies of activation in AA LNCs but were not statistically significant compared to the controls. MART1 peptide also showed significant difference compared to the control (b). The threshold line denotes true positive activation and was determined by 3x SE of SFCs from the negative controls. Significance of difference was determined by nonparametric Mann-Whitney U-Test where *=p<0.05.   90  3.4 Discussion AA is an incurable, autoimmune, inflammatory, hair loss disease that waxes and wanes unpredictably in each patient1. The genetics of a person may determine the basal risk to AA development while many environmental factors could potentially contribute as triggers to the onset of hair loss. Even though the pathogenesis of AA is multifactorial, the activation of autoreactive T-cells, more specifically CTLs, is widely believed to be the mechanism that is in control of the progression of AA203. However, as with many other autoimmune diseases, the exact antigen(s) and epitopes that can trigger the activation of autoreactive CTLs are still not yet elucidated. Previously, various groups have suggested several different autoantigens that may be the sources of epitope peptide targets for CTLs in AA patients88,106,139.  In this study, we used a step-wise method to design antigenic epitope peptides from selected antigens, followed by evaluating their ability to bind to HLA-A*0201 and finally assessing the frequency of CTLs in AA patient PBMCs that can be activated and induce apoptosis in HF keratinocytes compared to control subjects. Two main groups of antigens were evaluated in this study. The first group was melanocyte-associated antigens like melanoma antigen recognized by T-cells 1 (MART1), variation of MART1 (MEL1), GP100, tyrosinase (TYR), and tyrosinase related protein 2 (TYRP2). Many of them have been implicated in the development of vitiligo and some of them have been shown to induce the AA phenotype in human skin grafted to SCID mice139. The second group of peptides was derived from HF keratinocyte specific antigens such as trichohyalin (TCHH) and cytokeratin 91  16 (KRT16) that provide important structural scaffolds in cells242,243, and both of which are targeted by autoantibodies in AA patients88. The use of online matrix-assisted algorithms, BIMAS and SYFPEITHI, provided the initial prediction of HLA-A*0201 binding peptide sequences within the antigens listed in Table 3.1. However, in silico computer algorithm designed peptides may not be functional and hence may not actually induce CTL responses in vitro or in vivo. A direct correlation between the predicted HLA-A*0201 binding scores of the peptide in silico and the degree of HLA-A*0201 stabilization in vitro was not observed. Such observation was expected as there must be an inherent difference between an in silico system, where the peptide sequence affinity is calculated, versus in vitro stabilization in the presence of live cells. However, peptide sequence prediction via computer algorithms was still a logical starting point to narrow down the number of candidate peptides based on their likelihood to bind to HLA-A*0201. This approach has been used to screen for antigenic epitope sequences for other autoimmune diseases244,245 and cancers246,247. Such data is always integrated with in vitro assays to assess the binding affinity for each of the peptides which result in a streamlined protocol that allowed us to screen for a panel of peptides for further validation.  The ability of each peptide to induce the activation of CTLs in the PBMCs of AA patients and control subjects was evaluated with IFNγ ELISpot assays and confirmed via IFNγ ICS. The release of IFNγ after stimulation with the candidate peptides was more similar to the results obtained with the T2 stabilization assay. Interestingly, some the peptides that were able to stabilize the highest levels of HLA-A*0201, like MART1 and GP100, only induced moderate levels of IFNγ in ELISpot assays and ICS. Meanwhile, the peptide 92  sequences with modest ability to stabilize HLA-A*0201, like TCHH-G1 and TCHH-G2, and TYRP2, induced much higher frequencies of CTL activation from AA PBMCs compared to their respective controls. This phenomenon might be explained by the mechanism of positive and negative selection during the maturation of naïve T-cells. Conventional understanding of negative selection during T-cell maturation involves the survival of T-cells with fast on and off-rate self-peptides; this process could enable the survival of small populations of autoreactive CTLs248. In other autoimmune diseases, peptides derived from self-antigens with moderate or low affinity to HLA*0201 molecules have been reported to be immunodominant epitopes that can trigger the onset of disease, as in multiple sclerosis249 and type I diabetes mellitus250. Therefore, it is reasonable to anticipate that the lower affinity of HLA-A*0201 to TCHH-G1, TCHH-G2, and TYRP2 might allow for the escape of antigen specific CTLs duration negative selection processes, eventually enabling activation and proliferation in the periphery upon epitope challenge. Of note in our study, we found TYR peptides were able to induce relatively high frequencies of CTL activation in both AA and control PBMCs, suggesting it was a relatively non-specific activator of CTLs and not unique to AA.  We used a smaller subset of AA patients and controls to analyze PBMC responsiveness to each of the eight individual TCHH peptides. We found CTL response to individual TCHH peptides were highly variable and relatively weaker compared to cocktails of four peptides (Figure 3.3d), suggesting variability in primary or preferential epitope within TCHH was different in different AA subjects. Potentially, the state of AA, extent of hair loss, age and gender or even the type of treatments the subjects were on while their blood was 93  collected could have significant impact on the activation of CTLs. Due to the limitation of population size, we were unable to stratify subjects equally in these categories to perform statistical analysis that show significant differences (Figure 3.3). A more well-defined subject recruitment process and larger population size is required investigate the impact of hair loss status and treatment regimens on CTL response to different antigen epitopes. The ELISpot assays and ICS confirmed the activation of antigen-specific CTLs against certain autoantigen epitopes via quantification of CD8+ IFNγ+ T-cells. While injection of IFNγ has been shown to be able to induce AA-like phenotypes via promoting MHC expression in C3H/HeJ mice251, such action was likely systemic and would involve unspecific activation of all T-cells. Whether IFNγ is the key inducer of AA is still debatable as there are multiple proinflammatory cytokines secreted by activated CTLs. In addition, since injection of IFNγ alone to induce AA could not be replicated by other investigators252, it implies that other cytokines or events also need to be present for the onset of AA.  A significant number of HF infiltrating leukocytes are identified as CTLs and disruption of root sheath cells is observed in advance of overt hair loss in AA affected C3H/HeJ mice138. As CTLs are active secretors of multiple pro-inflammatory cytokines, their close proximity with HF keratinocytes in AA lesions is likely harmful to the health of keratinocytes. We stimulated PBMCs of AA and control subjects with the peptide sequences that were able to activate higher frequencies of response in AA CTLs, as seen in ELISpot assays, and investigate whether the resulting conditioned media (CM) could induce apoptosis in HF keratinocytes. TCHH-G1, TCHH-G2, TYRP2 and MART1 were selected to perform the apoptosis assays with a subset of AA and control subject PBMC samples. MART1 was 94  selected even though it was not able to induce significantly higher frequencies of AA CTL activation because previously MART1 epitope peptides were shown to induce the AA phenotype in human skin grafted to SCID mice139. The proliferation of HF keratinocytes after treatment with AA PBMC CM without peptide stimulation showed a slight decrease compared to control subject PBMC CM (Figure 3.5b). As this experiment was conducted without cell-to-cell contact, it is possible a large population of keratinocytes were still undergoing proliferation despite the presence of potentially harmful factors secreted by activated PBMCs. From the apoptosis assay, the culture media conditioned by AA PBMCs were in general able to induce higher expression of Annexin V (early stage apoptosis) at basal level without stimulation with peptides (Figure 3.5c). AA PBMC simulated with TCHH-G1 or G2 peptides released factors into culture media that significantly increased the expression of Annexin V in HF keratinocytes. This is consistent with the results obtained with ELISpot assays such that TCHH-G1 and G2 peptides were able to significantly increase the frequency of activated CTLs in AA PBMCs. Even though our ELISpot assays only quantified the production of IFNγ, the activated CTLs might also produce several other inflammatory or apoptotic cytokines such as TNFα and granzymes253 that could contribute to death of HF keratinocytes. Trichohyalin is an intermediate filament-associated protein that is found in the HF keratinocyte inner root sheath (IRS), as well as nails and epithelium of the tongue; it is believed to be an integral scaffolding protein242,254. Trichohyalin is a relatively large protein compared to the other antigens in our panel with about 250 kDa molecular weight. Therefore, there are potentially many HLA-A*0201 specific, epitope peptide sequences that can be derived from this protein. Although tongue abnormalities have not been reported by AA 95  patients, a “pitted nail” presentation is commonly associated with the progression of AA53. It is probable that multiple epitope sequences can contribute to the activation of AA CTLs, and the primary target may vary between subjects with different extents and stages of AA, similar to observations with CTL epitopes in type I diabetes255,256. By comparing AA patient backgrounds, our results also revealed that AA subjects with body hair loss and nail abnormalities tended to have higher frequencies of CTL activation across multiple antigen epitopes (Figure 3.3c). This feature could be a result of epitope spreading as is often observed in the chronic phase of autoimmune diseases such as Type 1 diabetes257,258.  To a lesser extent, stimulation of AA PBMCs with MART1 epitope also resulted higher expression of Annexin V in HF keratinocytes compared to stimulation of control subject PBMCs. Even though MART1 is not an epitope derived from keratinocyte autoantigens, its ability to drive keratinocyte apoptosis could indicate the production of harmful cytokines from melanocyte antigen specific CTLs that may act in a non-specific paracrine manner (extrinsic apoptosis pathways)259. Granzyme B expressing mononuclear cells were found in close contact with hair follicles in the lesions of chronic AA patients116. In our apoptosis assays, the presence of these factors in CM could have equivalent ability to induce apoptosis in HF keratinocytes despite being produced by melanocyte antigen specific CTLs. We anticipate if the assays were carried out using actual activated CTLs (apoptosis via cell-to-cell contact) instead of CMs (apoptosis via secretory factors), there would be differential levels of apoptosis of HF keratinocytes compared to melanocytes by CTLs activated by melanocyte or keratinocyte antigen epitopes. However, this cannot be done in our study as we were unable to obtain CTLs and histocompatible HF keratinocytes and melanocytes from the subjects.  96  Previous observation of white hair development before onset of hair loss in AA suggested proteins/antigens involved in melanogenesis may be important targets for autoreactive CTLs139. It has been reported that there are abnormalities observed with melanosomes in acute AA and a potential association with another autoimmune disease, vitiligo, which results in loss of pigmentation260,261. Analogs of peptide sequences derived from MART1 were used based on published literature indicating highly immunogenic epitopes compared to the natural HLA-A*0201 restrictive peptide228.  Interestingly, despite it being a highly immunogenic peptide, it was unable to stimulate AA CTLs to induce a significantly higher level of apoptosis in HF keratinocytes. MART1 is an antigen recognized by T-cells, it is used as a marker for melanoma and its expression is found in normal melanocytes in the skin and retina but not in other normal tissues (UniProt Q16655). HF melanocytes possess different properties than epidermal melanocytes262, which may contribute to the difference in the activation of AA specific autoreactive CTLs. However, as AA is a dynamic disease, and there was a distinct population of CTLs specific to MART1 in AA PBMCs as revealed by ELISpot and ICS, but similar to TCHH epitopes, the frequency of such MART1 responsive CTLs may predominate only in certain stages or specific forms of AA.  Rates of HF keratinocyte apoptosis elicited by CM from TYRP2 stimulated PBMCs were not different between AA and control subjects, at least in the subset of AA subjects’ PBMCs evaluated. This further implies that the high production of IFNγ, as observed in ELISpot assays, may not be the sole factor responsible for the apoptosis or disruption of HF keratinocytes. Certainly, IFNγ has very different roles under different conditions such that it can have immunosuppressive effects via induction of indoleamine 2,3-dioxygenase 97  expression for example263. We were unable to see statistically significant differences in late stage apoptosis rates exhibited by the HF keratinocytes, however, the trend was very similar to observations for Annexin V. It is likely that significance could be achieved with longer culture times with the CM from CTLs stimulated with peptides.  We also investigated possible cytokines or secretory factors that could induce apoptosis in HF keratinocytes. By using a multiplex cytokine array, we found a differential production of various inflammation cytokines between AA and control PBMCs stimulated with trichohyalin epitope peptides such as the elevation of IL-13 and RANTES production by AA PBMCs as well as elevated MIP1β and MMP9 by control PBMCs. However, many other inflammatory cytokines were below the minimal detection point of the array. In addition, due to the small sample size and large variations between each subjects’ PBMC cytokine expression, statistical significance cannot be determined. As the multiplex cytokine array is an ELISA-based assay, it may not be sensitive enough to detect small amounts of IFNγ or other cytokines such as TNFα as well as ELISpot and flow cytometry assays as performed here. As described in Chapter 2, the C3H/HeJ mouse model of AA has been widely used to study the pathogenesis of AA because it shares similar pathologic features with human AA203. We investigated the potential of mouse homologues of the various autoantigens used in the human ELISpot assays to activate mouse skin-draining lymph node derived LNCs. The frequencies of AA LNC activation after stimulation with various peptides were in general higher than control LNCs; similar to the trend observed from human ELISpot assays with autoantigen homologues. Unlike AA human PBMC responses, stimulation of AA mouse 98  LNCs yielded higher, but not statistically significant frequencies of activation with trichohyalin peptides (TCHH-G1, G2) compared to the controls. However, significant difference was observed with groups of cytokeratin 16 (KRT16-G1) and MART1 peptides, KRT16 was not tested with human ELISpot assay due to low binding affinity to HLA-A*0201 (Figure 3.1a) for all the KRT16 peptides we tested. Autoantibodies against cytokeratin 16 has been found in AA human patients in addition to trichohyalin88.  While MART1 was not able to induce significantly higher frequencies of PBMC activation in AA human patients, it was able to act as an autoantigen (in addition to GP100 peptides) to transfer AA to SCID mice139. This suggest that there could be some variations between the primary autoantigen targets in humans compared to the C3H/HeJ mouse model for AA, but keratinocyte derived autoantigen epitopes seemed to consistently induce higher frequencies of PBMC or LNC activation in both humans and mice. Even though all AA mice were sacrificed when they reached over 80% hair loss, as with human AA, the development of AA in C3H/HeJ mice was not uniform with different times of onset and rates of hair loss. Therefore, variable epitope spreading could also result in differential frequencies of LNC activation with different epitopes in different AA mice. Furthermore, human subjects may possess more heterogeneity in disease development than mouse models264 such as the C3H/HeJ inbred strain used here. It is not surprising that a difference between primary autoantigen target(s) was observed between mice and humans. In the identification of autoantigen epitopes in other autoimmune diseases such as type 1 diabetes, not all of the T-cell autoantigen epitopes identified in the mouse model were targeted by human patients’ cells265,266.  99  One of the challenges of identifying autoantigen epitopes in C3H/HeJ mouse is the lack of equivalent methods to verify H-2Kk binding affinity of the synthetic peptides; and thus prevented pre-screening of the peptides. It is possible to transfect the T2 cells used for human HLA-A*0201 stabilization assay with H-2Kk and then perform the same assay, but such an approach was beyond our expertise. A time-course assay may be useful to identify primary autoantigen targets in different stages of AA and potentially identify the “AA initiating” antigen epitope(s) but a large amount of mice would be needed for this study and it is beyond the scope of the current investigation; it is also difficult to define when AA actually initiates as immune responses can occur several weeks before overt hair loss is apparent138. The diversity of autoantigens suspected to be involved in AA development has increased significantly in recent years due to advances in relevant analytical technologies available. In our study, we compiled and evaluated some of the current candidate autoantigens believed to be involved in AA development. The epitope sequences of these antigens were designed, synthesized and evaluated with the same methods. Our data demonstrated that TCHH peptides consistently stood out with an ability to induce higher frequencies of AA PBMC activation and HF keratinocyte apoptosis via CM compared to control subject PBMCs. While MART1 showed strong activation in some AA subjects’ PBMCs, the cumulative effect was at a lower degree compared to TCHH. In the investigation on autoantigen epitopes in AA-affected mice, different primary targets were identified, but stronger frequencies of activations were still observed from HF keratinocyte derived antigen epitope (KRT16 and TCHH) stimulated AA LNCs in general.  100  At the moment, the effects of CTLs stimulated by these epitope sequences on HF melanocytes have not yet been investigated. In addition, the cytokines involved in the apoptosis and the degree of HF cell apoptosis induced via cell-to-cell contact will still need to be explored further. Nevertheless, the success of the methods employed here to narrow down a large panel of candidate antigen epitopes indicates that further AA epitopes could be identified using similar approaches. Knowledge of inciting antigen epitopes might also be used as a potential diagnostic tool for understanding variations in disease pathogenesis in AA patients. Ultimately, knowing the key antigen targets in AA may allow new targeted treatments to be developed.   101  Chapter 4: The Association of Alopecia Areata Development with Cardiac Dysfunction: Evidence in C3H/HeJ Mice  4.1 Introduction The development of autoimmune disease is often accompanied by diseases that affect sites elsewhere from the original site under attack by the autoimmune response. The autoimmune disease activated immune system can easily disrupt other important molecular pathways involved in modulating normal functions in the body. The development of AA can be associated with other inflammatory diseases such as thyroiditis71,267, psoriasis71,268, and more recently vitiligo also showed significantly higher prevalence in AA71.  AA may also potentially be associated with other diseases such as hypertension269 and mental health disorders60.  Recent studies have found that psoriasis is highly correlated with cardiovascular diseases (CVDs)175, depression270,271 as well as dysregulated stress hormones in humans174,272,273. Among many other CVDs, ischemic heart disease, angina and myocardial infarction are significantly related to psoriasis as compared to the general population175. It is believed that damage in the heart may be a result of inflammation induced by the upregulation of psoriasis associated inflammatory cytokines274,275. These findings raised the question whether the development of AA is associated with some forms of CVDs as well since AA pathogenesis shares many similarities to psoriasis. 102  Ischemic heart disease, hypertension and atherosclerosis are among some of the most common types of CVD. Dilated cardiomyopathy (DCM) is the most common form of non-ischemic, heart muscle disease, comprising 60% of cases of identified cardiomyopathy276. As the name suggests, in DCM a portion of the heart is dilated; the enlargement of the heart can impair the normal functions of the heart such as left and right ventricular pumping. The impairment or deregulation of chamber contractility can ultimately lead to heart failure. Although in many cases the DCM is idiopathic, coronary thrombosis after heart attack, viral infection such as coxsackievirus B, or inflammatory diseases, can all lead to DCM277-279. Recent studies also implicated the presence of autoantibodies suggesting that DCM might also involve an autoimmune disease mechanism in certain circumstances280-282. Stress is also a major contributor to the development of DCM where both β-adrenergic receptors and the elevation of stress hormones can adversely affect the contractility and remodelling of the heart283-285. Brief commentary in passing has suggested a possible epidemiological link between AA and heart disease, but this has not been actively investigated269,286. It was noted from previous studies with the C3H/HeJ mouse model for AA, the release of stress hormones, such as corticosterone (CORT) and adrenocorticotropic hormone (ACTH), was found to be elevated in AA affected mice and AA mice displayed heightened central and peripheral hypothalamic-pituitary-adrenal (HPA) tone65. The AA mice also showed a decreased in the ability to cope with physiological stress and a deficit in habituation to repeated psychological stress65.  103  In this study, we investigated the potential structural and molecular changes of hearts in AA affected mice. Real-Time Quantitative PCR (qPCR) of C3H/HeJ AA mouse hearts revealed select genes associated with heart tissue damage, particularly IL18, were significantly differentially expressed. With chronic AA development, heart enlargement and other morphological changes occurred. Whole tissue culture studies suggested exposure to ACTH may modulate gene expression and promote release of cardiac troponin I (cTnI), a marker of heart tissue damage. 4.2 Materials and Methods 4.2.1 Generation of AA Mice and Tissue Collection The C3H/HeJ mouse model for AA was utilized with the approval of the University of British Columbia Animal Care Committee. All female C3H/HeJ mice were supplied from colonies at The Jackson Laboratory (Bar Harbor, ME, USA) specific pathogen-free production facility. Normal haired C3H/HeJ mice were induced with AA by skin grafting as described previously96. Briefly, AA affected mice were completely shaved, sacrificed and full-thickness skin sections about 1 cm diameter were excised. The recipient mice were shaved on their back and a piece of full-thickness skin (1 cm diameter) was removed; the AA affected skin was grafted on this site. The mice were observed for the development of AA. Normal haired, age matched, sham-grafted C3H/HeJ littermates were used as controls. The mice were sacrificed 18 months after surgery, in which, the hearts and skin were collected for gene screening and immunohistochemistry. The weight of mice was measured before euthanasia and their hearts were measured after blood-collection. The tissue samples were 104  divided equally such that half of the tissues collected from each mouse was used for RNA extraction while the other half was used for histology or protein extraction. 4.2.2 RNA extraction, cDNA synthesis, and Quantitative Real-Time PCR (qPCR) RNA extraction from skin and hearts was performed using Qiagen RNeasy Fibrous Tissue Mini Kit (Qiagen, Toronto, ON. Catalog #74704) with manufacturer’s protocols except for the following changes: Double the amount of buffer RLT and Proteinase K (supplied in the kit) were used and QIAshredder (Qiagen. Catalog #79654) was used to homogenize the tissue after grinding with a pestle and mortar. First strand cDNA was synthesized by using 200 ng of total RNA from each sample and subjected to reverse transcription using the Superscript first-strand cDNA synthesis kit (Invitrogen, Burlington, ON. Catalog #18080-400) according to manufacturer’s protocols using a MJ Research PTC-150 MiniCycler (MJ Research, MA).  The qPCR reactions were conducted using 20.1 μl of mixture (5 μl of 1 ng/μl cDNA, 10 μl of DyNAmo HS SYBR Green qPCR Kit with 0.1 μl passive reference dye (Fisher Scientific, Ottawa, ON. Catalog #F-410L), and 5 μl of 5 μmol/L forward and reverse primers. The qPCR reaction mixtures were loaded into an Opticon DNA Engine (MJ Research) and programmed to run for 10 minutes at 95 °C before 41 thermal cycles, each of 15 seconds at 94 °C, 30 seconds at 60 °C and 30 seconds at 72°C. Housekeeping gene 18S was used as internal control (Invitrogen. Catalog #AM1718M) and the primers were designed with Primer3287 and purchased from Invitrogen. Primer sequences are listed in Appendix D. Relative quantification was used to determine the fold change of the expression of selected target genes in the AA tissue compared to the normal sham-grafted group as 105  described by Livak et al288. Briefly, a normalized threshold cycle number, ∆C(t), was calculated by normalizing the sample cycle number of the targeted gene with that of the internal control reference gene 18S. The ∆∆C(t) was then determined using the following formula: ∆∆C(t) = ∆C(t) sample (AA)-∆C(t)calibrator (normal). Finally the gene expression fold change in AA tissue relative to the normal controls was calculated by 2-∆∆C(t). Statistical significance (P-value <0.05) was calculated with Student's T test; error bars represented the range of possible values and were derived from the standard error. 4.2.3 Histology Hearts were fixed in Telly-fekete’s acid alcohol, embedded in paraffin in random orientations, and cut into 6 μm thick sections. For histology on all paraffin-embedded tissue samples, the de-paraffin step involved submerging the slides 2x for 5 minutes in 100% Citrasolv Clearing Agent (Fisher Scientific, Ottawa, ON. Catalog # 22-143-975), 1x 5 minutes in 100% Citrasolv:100% Ethanol 1:1. Follow by a hydration step: 2x 3 minutes in 100% Ethanol, 1x 1 minute in 95% Ethanol and 1x 1 minute in 70% Ethanol. Masson's Trichrome Stain (Sigma-Aldrich, Oakville, ON. Catalog #HT10-5-16) was used to stain for collagen deposition in heart sections of AA and normal mice (n=4 each group). The staining protocol was adopted from the manufacturer’s protocol: First by preheating Bouin’s Solution (Sigma-Aldrich. Catalog #HT10-1-32) to 56 ºC and then incubating the deparaffinised slides for 15 minutes. The slides were cooled in tap water and washed with running tap water to remove yellow color from the sections. The slides were stained in Working Weigert’s Iron Haematoxylin Solution for 5 minutes and washed again in running tap water for 5 minutes. After rinsing with deionized water, slides were stained with 106  Bierbrich Scarlet-Acid Fucshin for 5 minutes, rinsed with deionized water and incubated with Phosphotungstic/Phosphomolybdic Acid Solution for 5 minutes. Finally, the slides were placed in Aniline Blue Solution for 15 minutes and then 1% Acetic acid for 2 minutes.  Dehydration and clearing of the slides involved the same steps as de-parraffin steps but working in reverse order and finally the slides were mounted with Fisher Chemical Permount (Fisher. Catalog #SP15-100). Four hearts each from AA and normal control groups were selected randomly for staining; for each heart, three random sections were analysed and collagen deposition calculated for random blood vessels from each section. The areas encompassed by the collagen, blood vessel lumen, and endothelium were measured in pixels with ImageJ289 (National Institute of Health, Bethesda, MD). To quantify collagen within whole heart sections, blue pixel (Masson's Trichrome Stain) frequencies were counted with Adobe Photoshop. The hearts were measured randomly to avoid bias. Alizarin Red S (Sigma-Aldrich) was used to stain for calcium deposits in heart sections of AA and normal mice. Staining was performed after the de-paraffin steps and then rinsed rapidly in distilled water. The Alizarin red S Solution was prepared by adding 2.0 g of Alizarin red S powder to 100 mL of distilled water and the pH adjusted to 4.1-4.3 with 0.5% ammonium hydroxide. The tissue sections were stained with the freshly prepared Alizarin red S solution for 30 seconds to 5 minutes until an orange-red color developed. Excess dye was removed by blotting and the slides dipped in acetone 20 times and then in acetone/Citrasolv solution for another 20 times. Slides were dehydrated with alcohol and cleared in Citrasolv and mount as described above. 107  Haematoxylin and Eosin (H&E) staining was performed following standard protocols. Heart sections from AA and control mice (n=3 per group) were stained and the numbers of nuclei in the atria and ventricles per 100 µm2 were quantified and compared with two separate counts for each of the images with atrial and ventricle heart sections; the average of counts from two atrial and two ventricle images were taken for calculation. One-way ANOVA was used when performing statistical calculations involving multiple comparisons; Student’s t-test was used when comparing between two groups. 4.2.4 Immunohistochemistry Immunohistochemistry (IHC) was performed on heart sections from AA and control mice (n=4 per group) to investigate protein expression and localization. An antigen retrieval step was performed after de-paraffinization; slides were washed 3 minutes for 3x in Tris-buffered Saline (TBS) buffer and immersed at 90 ºC 1x in Target Retrieval Solution (Dako, Burlington, ON. Catalog #S1699). The target proteins tested (and respective antibodies) were IL18 (goat polyclonal IgG, M-19, Catalog #sc-6179), IL18R (rabbit polyclonal IgG, M-307, Catalog #sc-50520) and IL18BP (goat polyclonal IgG, A-17, Catalog #sc-9463) (all Santa Cruz Biotechnologies, Santa Cruz, CA). The staining method used was the Vector avidin:biotinylated enzyme complex (ABC) staining system, with alkaline phosphatase as the enzyme and Vector Red Alkaline Phosphatase Substrate Kit (Vector Laboratories, Burlington, ON. Catalog #SK-5100).  IHC staining followed a two-day protocol where primary antibody (IL18, IL18R and IL18BP) was diluted (1:1000, 1:200 and 1:75, respectively) and allowed to label the tissue sections overnight; the negative controls had no primary antibody. Biotinylated secondary antibody was diluted to 0.5% and incubated with 108  the slides on the second day followed by development with alkaline phosphatase based Vector Red substrate kit. The slides were counter-stained with haematoxylin, dehydrated and mounted with Permount. 4.2.5 Heart Tissue Culture Short term cultures of fresh mouse heart tissues (n=6 per group) with different concentrations of full length human ACTH (Sigma. Catalog #A0423) were performed. Atrial and ventricle heart tissues were separated and treated with four different concentrations of ACTH (0 µM, 0.1 µM, 1 µM, 2 µM); concentrations ranged from low to high and were determined from published literature290-292. Cultures were set up with serum-free medium, DMEM F12/Glutamax (Gibco, Burlington, Ontario. Catalog # 10565-018) was used as described by another group293, and changed every 24 hours with fresh ACTH peptide292 and media until the 72 hour endpoint. Culture media at each 24 hour period were collected for ELISA analysis (below), heart tissues were collected at the end of culture for qPCR analysis as described above. 4.2.6 Total protein extraction and quantification Total protein extraction was performed on one portion of each heart (n=4 per group) using a Total Protein Extraction Kit (Millipore, Billerica, MA. Catalog #2140) following the manufacturer’s protocol. Protease inhibitor (PI) was diluted with TM buffer and added to the heart tissues and kept on dry ice. The tissues were homogenized using a homogenizer with 20 second intervals and cooled on ice for 5 minutes in between. Once homogenized, homogenates were rotated for 20 minutes at 4 ºC and then centrifuged at maximum speed (at 109  least 500x g) for another 20 minutes at 4 ºC. Supernatant from each samples was collected to determine total protein concentration. Pierce BCA Protein Assay Kit (Pierce Biotechnology, Rockford, IL. Catalog #23225) was used to determine the total protein concentration. Procedures were supplied by the manufacturer, and a microplate method was used to determine protein concentration. Standards with known concentrations were prepared by serial dilution from the most concentrated aliquot. BCA standards or protein solutions extracted from the heart tissues were transferred to a microplate (25 µL per well). Working Reagent (WR; 200 µL) was added to each well (1:8 dilution) and resuspended. The microplates were covered and incubated at 37 ºC for 30 minutes, cooled to room temperature and read under 560 nm with a spectrophotometer to obtain absorbance values.  Standard curves were generated to calculate protein concentration. This is calculated by first subtracting the absorbance reading of the samples with the absorbance from the blank standard. Standard curves were generated by plotting the average of absorbance values from each BCA standard corrected with a blank against its known concentration in μg/mL. The equation of the best-fit line (y=mx+b; where m=slope and b=y-intercept) was determined by drawing a linear line that passed through the data points. The concentration of total protein from each sample was determined by substituting y (absorbance) into the equation to determine x (concentration) and multiplied with the dilution factor. 4.2.7 Cardiac Troponin I, IL18, IL18R1, IL18BP and CASP1 ELISA Cardiac troponin I (cTnI), a marker for heart tissue remodelling, is released into the blood stream during heart tissue modelling processes as a result of tissue damage294-296. Its 110  expression is increased in those whose heart is undergoing rapid remodelling associated with dysfunction 297. Mouse cTnI ELISA kits (Life Diagnostics, West Chester, PA. Catalog #2010-1-HS) were used to test serum samples of AA and control mice (n=5 each group). A different cTnI ELISA (Kamiya Biomedical, Seattle, WA. Catalog #KT-469) with higher detectable range was used to measure cTnI in the heart tissue AA and control mice (n=4 for each group); 50 ng of total protein from each sample, as determined by BCA protein assay (above), was subjected to cTnI ELISA. For mouse IL-18 ELISA (eBioscience San Diego, CA. Catalog # BMS618/2), IL18R1, IL18BP and CASP1 ELISA (MyBioSource Inc, San Diego, CA. Catalog # MBS931556, MBS904753 and MBS165060), 1 mg of heart tissue homogenate protein was used to perform the assays (n=3 for each group and time points from 4-12 weeks).  After ACTH stimulation, the supernatant from AA and control mouse atria culture was collected at 72 hours (AA n=5, control n=6) to measure cTnI released as a response to increasing concentration of ACTH, using the higher sensitivity cTnI ELISA kit (Kamiya Biomedical). For all ELISA assays, a standard sandwich-ELISA protocol was followed with standard dilutions specified by each individual kit. Standards or samples were added to microplates pre-coated with antigen specific primary antibody. Then a biotinylated detection antibody specific for each antigen and Avidin-Horseradish Peroxidase (HRP) conjugate was added to each well and incubated. The excess antibody is washed away and the substrate solution was added to each well. Only those wells that contain target antigen, biotinylated detection antibody and Avidin-HRP conjugate will appear blue in color. The enzyme-111  substrate reaction was stopped with sulphuric acid solution and the color turns yellow. The optical density (OD) or absorbance at 450 nm (405 nm if the OD was out of range) was read with a spectrophotometer.  A standard curve was generated by drawing a best fit line (4-parameter non-linear curve fit) through all standard absorbance readings with GraphPad Prism 6 Software (GraphPad Software, La Jolla, CA) which also extrapolates the concentration of samples based on the standard curve. One-way ANOVA was used when performing statistical calculations involving multiple comparisons. 4.3 Results 4.3.1 Preliminary gene screening showed significant elevation of the Il18 gene in skin and heart tissues of AA mice Preliminary gene screening on both mice that had AA for over 18 months, and control age and sex-matched littermates, showed significant increase of Il18 in the skin and heart tissues of AA mice. Conversely, there was also a significant decrease of cardiac troponin I (Cti) in both skin and heart tissues of AA mice. In AA skin, there was also a very high expression of Gzmb, Fasl, Nppb and Tnfa; genes involved in apoptosis and inflammation. While in the AA heart, there was also a significant increase of Fas (Figure 4.1a,b).   112   Figure 4.1 Preliminary qPCR gene screening of chronic AA mice compared to the sham-grafted controls. In both the skin (a) and heart (b), there was a significant increase of Il18 and significant decrease of Cti in the AA mice (n=6) compared to the healthy sham-grafted controls (n=6). There was an over 1,000 fold increase in granzyme B (Gzmb) activity in the skin of AA mice but such increase was not observed in the heart. qPCR analyses of relative fold change in gene expression were calculated using 2−ΔΔCt; average fold change is presented. Error bars represent the range factor difference (2−ΔΔCt±ΔCtSD). Statistical significance determined with Student’s t-test; * denotes p<0.05. 113  4.3.2 AA mice displayed changes in heart morphology and had significantly heavier heart weights There was a significant difference between the heart weights of AA mice that had AA for 18 months compared to sham-grafted control mice (Figure 4.2a). The heart-to-body weight ratio (determined by dividing the wet weight of the heart over the body weight of the mouse) of AA mice was also significantly greater than the control mice (Figure 4.2a). The changes in heart morphology were identified with H&E staining (Figure 4.3a-d). The ventricles of AA mice had significantly higher frequencies of nuclei compared to the controls (Figure 4.3e). But there were significantly lower frequencies of nuclei in the atria of AA mouse hearts compared to the controls despite the increase in heart size (Figure 4.3e). While there was no significant difference between the average nuclei size in atria of AA mice (measured in pixels) compared to the controls, there was a significant difference between the average area covered by individual cells in the AA atria compared to the control as well as the nuclei to whole cell area ratio (Figure 4.3f,g).   114   Figure 4.2 Heart weight and heart to body weight ratio in AA and sham-grafted mice. The wet heart weight (a) and heart to body weight ratio (b) of AA (n=10) and healthy sham-grafted mice (n=11) revealed significantly heavier hearts in AA mice. Statistical significance was determined with Student’s t test where * denotes p<0.05.   115   Figure 4.3 AA mice had significantly fewer nuclei in atria tissue compared to sham-grafted mice. Sham-grafted control mouse atria (a) and ventricles (b) and AA mouse atria (c) and ventricles (d) were H&E stained. The number of nuclei in 100 µm2 was quantified (e). AA mice had significantly fewer nuclei in their atria but significantly more nuclei in their ventricles compared to the controls. The average size of nuclei in the atria was significantly larger than the control (f); the ratio of nuclei area to the whole cells in AA atria was significantly higher than the control as well (g). Statistical significance was determined with one-way ANOVA (e) and Student’s t-test (f, g) where * denotes p<0.05. Bar=20 μm.   116  4.3.3 AA mouse hearts exhibited increased collagen deposition Masson’s trichrome stain revealed collagen deposition distribution within the hearts. AA mouse hearts were identified with significantly higher total amount of collagen (by volume) compared to control hearts (Figure 4.4a). Collagen deposition was mostly localized to the periphery of blood vessels (Figure 4.4b). By measuring the average area, in pixels, encompassed by the collagen area, endothelial layer (blood vessel walls) and the total area (endothelial layer plus collagen), the average ratio of areas encompassed by collagen versus areas encompassed by blood vessel walls was shown to be significantly higher in AA mice (Figure 4.4c); an indication of increased collagen accumulation. In addition, average width of the endothelial layer in AA mice was found to be significantly lower than in controls (Figure 4.4d). From the comparison of histology results, AA mouse hearts presented with increased peri-vascular collagen and thinner blood vessel endothelial layers. Calcinosis within the hearts was minimal and not different between AA mice and controls (Figure 4.4e,f). 4.3.4 Concentrations of cTnI in heart tissue and plasma were higher in AA-affected mice ELISA analysis on both plasma and heart tissue protein homogenate cTnI levels revealed a trend for higher cTnI levels in AA affected mice compared to controls (Figure 4.5a,b). This is potentially an indication of overall deterioration of heart health. This result indicates that, despite the low gene expression (Figure 4.1b), there was still a relatively high level of cTnI protein in AA mouse hearts. 117   Figure 4.4 Evaluation of heart collagen deposition and calcinosis in AA and sham-grafted mice. Areas encompassed by collagen, lumen, blood vessels, and blood vessels plus surrounding collagens, were measured with ImageJ (a, c, d). There was a higher amount of collagen infiltration into the blood vessels within the hearts of AA mice (b, AA left; Control right). AA mice had significantly larger regions of collagen deposition around blood vessels but had significantly thinner blood vessel wall thickness compared to normal controls. Three random readings performed from each of 3 different non-consecutive slides per mouse for AA (n= 4) and control (n=4) mice. Heart sections from AA and sham-grafted control mice were stained with Alizarin Red S to detect calcium deposition (a, b). There were no noticeable difference in the level of cardiac calcinosis in the heart tissues of AA mice (e) compared to the sham-grafted controls (f). Statistical significance was determined with Student’s t-test where * denotes p<0.05. Bar=50 μm.    118   Figure 4.5 Significantly higher level of cTnI was detected in the heart tissues of AA mice.  ELISA revealed a higher amount of serum cTnI associated with AA (n=5 per group) compared to control mice, though not statistically significant (a). The concentration of cTnI in the heart tissue of AA mice was significantly higher than the control mice (b). Statistical significance was determined with Student’s t-test where * denotes p<0.05.   119  4.3.5 AA mouse skin and heart tissues showed significantly higher pro-inflammatory cytokine gene expression Further qPCR analysis on inflammatory genes associated with Il18 showed that Il18 receptor-1 (Il18r1) and Il18 binding protein (Il18bp) were significantly increased in the heart tissues of chronic (18 months) AA-affected mice, 4.5 and 5.2 fold respectively, compared to control mice. Similar trend was also observed in the AA skin where Il18r1 and Il18bp was significantly increased by 3.4 and 5.4 fold respectively (Figure 4.6a,b). Important marker of heart hypertrophy, atrial natriuretic factor (Nppa), was significantly increased 3.2 fold in AA mouse hearts compared to controls, but no significant change was observed for the mRNA expression of β-myosin heavy chain (Myh7) (Figure 4.6b).   120   Figure 4.6 qPCR analysis of selected genes in chronic stage AA and sham-grafted mice. In an initial screen for various heart disorder related gene markers, there was a significant increase of Il18, Il18r1 and Il18bp gene in both skin (a) and hearts (b) of AA mice (n=4) compared to sham-grafted control mice (n=5). The expression of Nppa was also significantly increased in the hearts of AA mice. qPCR analyses of relative fold change in gene expression were calculated using 2−ΔΔCt; average fold change is presented. Error bars represent the range factor difference (2−ΔΔCt±ΔCtSD). Statistical significance determined with Student’s t-test; * denotes p<0.05.   121  4.3.6 AA progression correlated to changes in the gene and protein expression of the Il18 family The dynamics of gene expression profiles for Il18, Il18r1, Il18bp, and Caspase-1 (Casp1) in hearts were investigated in response to AA skin grafts compared to control grafts (Figure 4.7a-d). AA hearts showed increased Il18 expression shortly after skin grafting prior to overt hair loss (4-8 weeks), but expression decreased when the mice first started to lose hair around 10 weeks after grafting. The expression of Il18bp showed significant increase in AA hearts across the first 12 weeks after receiving skin grafts. Similarly, Il18r1 and Casp1 showed overall significant increase in AA hearts compared to the controls except at just before the onset of AA (8 weeks). The overall gene expression of Il18bp, Il18r1 and Casp1 showed similar gene expression patterns as the mice with chronic AA. As for protein expression, IL18, IL18R1 and IL18BP showed decreasing trends as AA began to develop around eight weeks, similar to their respective gene expression at the similar time points. IL18BP expression in AA mouse hearts was significantly lower compared to the controls at 12 weeks. At 10 and 12 weeks, CASP1 expression in AA mouse hearts was significantly decreased compared to the controls (Figure 4.7e-h).   122   123  Figure 4.7 qPCR and ELISA analysis of Il18 and related genes and protein expression during the onset of AA compared to sham-grafted mice. Comparing AA mice to sham grafted mice (n=3 per time point), Il18 gene expression fold change in mouse hearts showed an increase in the expression pattern before the onset of AA, but decreased expression around the time of AA onset . Il18r1, Il18bp and Casp1 (a-d) showed mostly significant elevation both before and after onset of AA. At the protein level, the expression of IL18, IL18R1 and IL18BP displayed similar trends as with qPCR (e-g). However, Casp1 expression was significantly higher in sham grafted mice at 10 and 12 weeks (h). qPCR analyses for gene expression levels were calculated as fold change by using the 2−ΔΔCt, average fold change presented. Error bars represent the range factor difference (2−ΔΔCt±ΔCtSD). Statistical significance determined with Student’s t-test; * denotes p<0.05.   124  4.3.7 Localization of IL18 in the atria of AA mouse hearts AA mice displayed an atrial-specific localization of IL18 (Figure 4.8a) with immunohistochemistry (IHC) labelling. The control showed no specific labelling of IL18 in atrial tissues (Figure 4.8b). The pattern of IL18R1 protein expression was similar between AA and controls with a slightly higher localization in the apical areas of AA hearts (Figure 4.8c,d) while the IL18 antagonist, IL18BP, showed low and unspecific expression in both AA and controls (Figure 4.8e,f).    125   Figure 4.8 Immunohistochemistry of IL18 in AA and sham-grafted mice. IHC performed on AA and sham-grafted mouse heart sections revealed localized expression of IL18 (red) in the atria of AA mice (a). In healthy sham-grafted mice, there was no specific labeling in the atria of the heart (b). AA mouse hearts (c) had slightly more specific expression of IL18R1 near the apical part of the heart compared to healthy controls (d). No specific staining was observed with IL18BP in AA mouse hearts (e) compared to the control (f). Three random paraffin-embedded slides from each mouse, each containing 3 to 4 sections, were analyzed in parallel. Slides were counterstained with hematoxylin (blue). Bar=50 μm.   126  4.3.8 ACTH treatment in tissue culture resulted in differential gene expression in AA and control mouse atria ACTH exposure resulted in an increase of Il18, Il18r1, and Casp1 expression, compared to the no-treatment controls, in the atria of AA mice, but the differences were not statistically significant (Figure 4.9a-d). Conversely, ACTH exposure statistically significantly increased Il18bp, Il18r1, and Casp1 in control mouse atria tissues (Figure 4.9e-h). Both Il18r1 and Casp1 in control mouse atria showed a concentration dependent increasing trend that reached significance when exposed to 2 μM of ACTH.   127   128  Figure 4.9 qPCR analysis of atria treated with ACTH for 72 hours. For atria derived from AA mice, Il18 expression was highest at 1µM of ACTH (a); the expression of Casp1 showed similar trend (d). Il18r1 expression was dependent on ACTH concentration (b) while Il18bp was down-regulated (c). For atria derived from sham-grafted mice, Il18 expression was lowest at 1µM unlike AA (e). However, Il18bp was significantly increased at 1 µM (g). ACTH had a significant effect on the control atria with concentration dependent increase of Il18r1 (f) and Casp1 (h). Gene expression levels were calculated via qPCR as fold change compared to no-treatment control (0 µM ACTH). Error bars represent the range factor difference (2−ΔΔCt±ΔCtSD). Statistical significance determined with Student’s t-test; * denotes p<0.05.   129  4.3.9 ACTH exposure resulted in changes in collagen gene expression in the atria Collagen Ia1 (Col1a1), Collagen IIIa1 (Col3a1) and Collagen Va1 (Col5a1) in the atria of AA and normal mice showed different patterns of gene expression upon ACTH exposure. In general, expression of collagen genes increased in an ACTH concentration dependent manner. However, at both 1 μM and 2 μM of ACTH, the expression of Col5a1 was significantly increased in AA mouse atria (Figure 4.10a-c). 4.3.10 ACTH exposure resulted in increased release of cTnI by atria tissues ELISA analysis revealed that the release of cTnI by atrial tissues into the culture medium at the end of 72 hours period was ACTH concentration dependent for both AA and control mice. In comparison to no treatment controls, AA mouse atria released significantly higher cTnI at 2 μM ACTH while the difference in the control mouse atria was not significantly different (Figure 4.10d).    130   Figure 4.10 Gene expression of collagens and the release of cardiac troponin from AA and sham-grafted mouse atria in response to ACTH treatment after 72 hours. Both AA and control mice displayed ACTH concentration dependent increase of Col1a1 (a), Col3a1 (b), Col5a1 (c) and the release of cTnI from atria (d). However, significant increase was only observed in AA mouse atria for Col5a1 (at 1 μM and 2 μM ACTH) and release of cTnI (at 2 μM ACTH) compared to no-treatment control. qPCR analyses for gene expression levels were calculated as fold change by using the 2−ΔΔCt, average fold change presented. Error bars represent the range factor difference (2−ΔΔCt±ΔCtSD). For cTnI ELISA, a standard curve was generated with the standards provided by the manufacturer and the corresponding equation of the line was used to determine the cTnI concentration in the sample. Statistical significance determined with Student’s t-test (a-c) and one-way ANOVA (d); * denotes p<0.05.   131  4.4 Discussion Alopecia areata (AA) by itself does not lead to death in humans, however, the emotional stress associated with hair loss and other co-morbidities can have a devastating impact on affected individuals71. A relationship between androgenetic alopecia, cardiovascular disease and hypertension in humans has been demonstrated by various groups, though the exact biological mechanism remains elusive 298-301. In this study, we investigated the change in the physiology of the heart, as well as the ability of increased stress hormone ACTH, as observed in AA-affected C3H/HeJ mice65, to affect the health of hearts in the AA mouse model. Potentially, the skin grafting procedure itself could have adverse effects on the condition of the heart and this was not investigated in our study. However, the effect of surgery on mice was normalized by performing sham-grafting of their own skin (in a different orientation) in the controls. Furthermore, the size of the heart from mice without skin-grafts (both control and spontaneous AA) was similar to the respective grafted groups (data not shown). The potential for changes in heart physiology in AA patients has never been investigated and there are no prior reports of AA mice or affected human patients having an increased frequency of heart disorders. However, there is increasing evidence linking the development of cardiovascular disease, such as dilated cardiomyopathy (DCM), and depression with psoriasis patients175,176. As psoriasis is believed to share similar mechanisms of pathogenesis as AA, it is possible that AA is also associated with some form of inflammatory heart disease that results in abnormal hypertrophy in the heart.  The weight of the hearts was found to be significantly greater in AA mice compared to healthy, sham-grafted littermates in both wet weight and heart to body weight ratio (Figure 132  4.2); a symptom that resembles disease similar to DCM in humans302. However, the individual heart weights were quite variable when compared between individual mice. The variable extent of hair loss and its fluctuation over time may affect (and potentially reflect) the degree of heart damage, consistent with the high variability of results between each mouse. Further, the genetic resistance to atherosclerosis in C3H/HeJ mice303 may also affect the pathological presentation of any heart disease. Significant difference between the density of cardiomyocyte nuclei in the atria and ventricles of AA mice compared to the controls was discovered as one of the heart morphology changes in AA mice (Figure 4.3). AA hearts were presented with significantly fewer nuclei per unit area compared to the controls in the atria. Meanwhile, cardiomyocytes in AA hearts exhibited statistically significantly larger nuclei, and a greater nucleus to whole cell area ratio, compared to the controls. Conversely, the ventricles of AA mouse hearts showed significantly higher frequencies of nuclei than the controls (Figure 4.3e). These results share similarities to pathological features of atrial hypertrophy304. Excess deposition of peri-vascular collagen is a pathological event in many forms of cardiovascular disease305,306, excess peri-vascular collagen also likely stiffens and decreases the contractility of the blood vessels within the heart; a promotion of cardiovascular hypertrophy and fibrosis307. AA mouse hearts displayed significantly higher amounts of collagen compared to the control hearts, the deposition of collagen was also localized around the blood vessels (Figure 4.4a,b). There also appeared to be thinning of the blood vessels in AA hearts, possibly as a result of decreases in endothelial cell size, which is also a feature of hypertension308. The thinning of the blood vessel walls may also be a consequence of collagen reorganisation by cardiac fibroblasts309.  The accumulation of extracellular matrix 133  (ECM) in the heart can modulate cellular function and size; type I, III, IV collagen and fibronectin can decrease the migration, proliferation and the size of aortic endothelial cells310. Increased collagen deposition around and within blood vessel walls (Figure 4.4c) can also lead to hypertrophy and ultimately to heart failure311. Cardiac calcinosis has been reported to develop in C3H mice in a number of studies312-314, however calcium deposition in the hearts of AA and control mice showed no difference (Figure 4.4e,f). The gene expression screening in AA and control mice showed a significant increase of the Il18 gene in both the skin and heart tissues of AA mice (Figure 4.1). The pro-inflammatory nature of IL18 is associated with many forms of skin315,316 and heart diseases317,318 via the induction of interferon gamma (IFNγ) production from T-cells319-322. This led to the hypothesis that AA may be associated with heart inflammation and/or defects in the regulation of inflammation in multiple organs. In addition to Il18, the expression of Il18r1 (Il18 receptor), and Il18bp (Il18 binding protein; an antagonist) were also significantly increased in AA mouse skin and heart tissues (Figure 4.6). The significant increase of Il18r1 in both the skin and heart may have a synergistic effect with Il18 by increasing the sensitivity to the ligand323 while Il18bp gene expression alongside Il18 and Il18r1 may be a sign of an activated negative-feedback system to counter adverse effects inflicted by IL18324. There is also a moderate increase in Casp1 which may contribute to the increase in the amount of the activated form of IL18 since it is required to catalyze pro-IL18 to an active form. IL18 may also play a role in cardiac hypertrophy in AA mice by elevating Nppa (atrial natriuretic factor) gene expression in the heart (Figure 4.6b); Nppa is a vasodilator released by the atrial tissues in response to stretch and remodelling and its expression can be triggered by IL18325. 134  At the protein level, increased intensity and localization of IL18 was found only in the atria of AA mice; control mice showed no specific signals for IL18 in the atria or in the ventricles (Figure 4.8a,b). The presence of IL18 and heart remodelling in AA mice as observed in our study best resembles a condition called atrial fibrillation, in which the patients also have elevated IL18 and remodelling specifically in the atria317. Both IL18R1 (Figure 4.8c,d) and IL18BP (Figure 4.8e,f) showed very low intensity of expression and no obvious difference between AA and control mice in both atria and ventricles (IL18R1 showed a slightly higher intensity near the apical part of the heart near the atria). Potentially, the uniform expression of IL18R1 only exacerbates the effect of IL18 in the atria of AA hearts by increasing local sensitivity. Conversely, the uniform but low expression of IL18BP may not be able to elicit enough inhibitory effect against IL18 in the atria of AA mice. Injection of IL18 into mice has been shown to induce myocardial hypertrophy and heart remodeling326. Therefore, the elevation of Il18 expression in the heart as well as the specific protein localization in the atria, in conjunction with the abnormal hypertrophy in AA affected mice, may be a marker of inflammatory heart disease similar to observations with dilated cardiomyopathy319,327. As a pro-inflammatory cytokine, IL18 is able to induce the production of interferon γ (IFNγ) from lymphocytes and natural killer cells 319-322,328,329. IL18 plays a big role in the pathogenesis of various cardiovascular diseases (CVDs) such as ischemia-reperfusion injury, and atrial fibrillation317,327,330,331, as well as cell-mediated inflammation and myocardial fibrosis332-335. An alternative hypothesis is that dysfunctional caspase-1 activity can lead to increased secretion of IL18 and IL1β in a heart-specific “auto-inflammatory disease” in the absence of cell infiltration336,337 because there is no inflammatory cell infiltrate observed in 135  the AA hearts. The release of IL18 from the activated lymphocytes into circulation, as observed in other autoimmune diseases338, may also be an important contributor to cardiac hypertrophy and increased cell apoptosis without direct lymphocyte infiltration319. Elevation of IL18 has been found in AA patient plasma339; but the mechanism behind it and the exact link between IL18 and the changes in the hearts of AA mice remains unknown. The release of cardiac troponin I (cTnI) can precede the actual onset of more severe forms of heart disease and can serve as a hypertension marker340, it is also a myocardial regulatory protein that is elevated after cardiac injury340,341. The serum of AA mice exhibited higher levels of cTnI compared to the controls, though not statistically significant, but the trend is still consistent with heart tissue remodelling (Figure 4.5a). In addition, the significantly higher levels of cTnI in AA heart tissue compared to control mice (Figure 4.5b) suggest tissue remodelling and heart hypertrophy and thus increased demand for cTnI. Even though the gene expression of cTnI does not reflect protein expression, other regulators such as IGFBP could potentially be active and inhibit cardiac hypertrophy via limiting cTnI gene expression342. Abnormal regulation of stress hormones and receptors in AA affected mice have been reported65,343,344. Adrenocorticotrpic hormone (ACTH) is a stress hormone and a known inducer of IL18345,346, with aberrantly higher expression together with corticosterone in AA mouse plasma; potentially as a result of inflammatory cytokines released by activated lymphocytes involved in AA development65. With significant increase of Il18, Il18r1, and Casp1 gene expression in control mouse atria after ACTH exposure compared to AA mice, it is possible that, as AA mouse atria were already exposed to increased ACTH activity in vivo, there was limited opportunity for even greater modulation of gene expression by additional 136  ACTH exposure in the in vitro assay. The significant increase of Il18bp in control mouse atria may suggest that healthy, control mice may be somewhat better at coping with the effect of increased ACTH by elevating expression of antagonist in response to elevated IL18. Nevertheless, the net result of ACTH exposure suggests increased activity of IL18 in the heart. Targeting IL18 may also be a therapeutic strategy for patients with heart disease347, highlighting the importance of regulating IL18 within the heart. Previously, it was revealed that the physiological plasma level of ACTH in C3H/HeJ mice without AA is around 98 pg/mL (130 pg/mL in AA-affected mice)65. In our study a range of concentrations (0.1 µM, 1 µM and 2 µM) of ACTH was administered to the tissue culture; the concentration used was different than the physiological to ensure the effect of ACTH on the heart could be detected with qPCR and ELISA assays.  ACTH has also been shown to increase bone mass by increasing the production of type I collagen in osteoblast cell lines348-351. Therefore, with the increase of ACTH production in AA mice, it may result in adverse effects by elevating collagen expression within the heart. The expression of important collagen genes, col1a1, col3a1, and col5a1 were all increased in the atria upon ACTH exposure, but they were less pronounced in control mouse atria (Figure 4.10a-c). Even though the gene expression of type I collagen did not significantly increase in the atria after ACTH exposure, the expression of col5a1 significantly increased in AA mouse atria in an ACTH concentration dependent manner (Figure 4.10c). Type V collagen is an important regulator for the assembly of type I collagen352, therefore significant increase of the Type V collagen gene (Col5a1) in AA mice may promote excess accumulation of type I collagen353. Such data may be an explanation as to how stress hormones can be associated with damage to the cardiovascular system.  137  In the development of hypertension, ACTH is elevated and is associated with the biosynthesis of aldosterone which may in turn be associated with hypertension, cardiac fibrosis and necrosis354. As such, injury in atria was confirmed by measuring the amount of cTnI released into culture medium after 72 hours (Figure 4.10d). Culture medium cTnI levels increased after ACTH exposure for both AA and control mouse atria indicating heart tissue changes in response to ACTH. However, AA mouse atria were more susceptible to ACTH activities and released significantly more cTnI with a concentration dependent trend.  We have provided evidence that AA development in mice is associated with abnormal heart hypertrophy, associated with elevation of Il18, Col5a1 and cardiac remodelling marker, cTnI. Our results presented here emphasize that the consequences of AA expression is not just restricted to the hair follicles. Stress hormones, such as ACTH, can accentuate the production of Il18 and may lead to damage in the heart and the release of cTnI. Therefore, the sequelae of AA development may have an impact on other tissues and organs beyond the skin. Results from our investigations suggest that AA onset can be a predisposing factor to abnormal heart remodelling and closer follow-up of patients with AA should be considered.     138  Chapter 5: The Effect of Alopecia Areata and Androgenetic Alopecia on the Expression of Heart Disease Markers in Humans and the Health of Cardiomyocytes  5.1 Introduction As discussed in Chapter 4. The development of AA can be associated with other inflammatory diseases such as thyroiditis267, vitiligo355,356, psoriasis71,268,357. A possible relationship between androgenetic alopecia (AGA) and hypertension has also been investigated by various groups, but the exact biological mechanisms linking the two conditions still remains elusive299,300,358. There are epidemiological studies suggesting that cardiovascular diseases (CVDs) and arterial stiffness are more prevalent in young males with AGA359,360. The association of alopecia areata (AA) with heart diseases has been briefly suggested in several epidemiological studies, but no active investigations have been conducted60,269,286.  Psoriasis is an inflammatory, autoimmune dermatological disease that shares some similarities with AA. The development of psoriasis is found to be associated with various forms of CVD including atherosclerosis, angina, dilated cardiomyopathy (DCM) and myocardial infarction175,176,361,362. It is believed that damage in the heart is induced by the upregulation of psoriasis-associated inflammatory cytokines274,275. As such, inflammatory cytokines or secretory factors produced during the development of AA could also potentially promote heart damage akin to observations in psoriasis. 139  Releasable molecules such as cardiac troponin I (cTnI) and C-Reactive Protein (CRP) have been widely used as a biomarkers for heart failure and cardiac infarction363,364. A major marker for cardiomyocyte injury is cTnI, the plasma level of cTnI is increased following myocardial infarction due to cardiac cell death and restructure363, but it can also be elevated in patients suffering from heart failure in advance of ischemia364. This property of cTnI offers early detection of abnormal cardiac remodelling, potentially in AA and AGA patients, even without the presence of active heart disease. In general, subjects with greater than 0.01 µg/mL of serum cTnI are believed to have subclinical forms of heart disease365. CRP is a marker for inflammatory immune response364. CRP is a strong predictor of new heart failure, myocardial infarction, stroke, cardiovascular cell death and new onset of diabetes364. Since CRP is both a mediator of inflammation as well as a marker for immune response366, it may also play a role in the progression of AA367, and the microinflammation associated with hair follicles in AGA368. In this study, we investigated the plasma level of cTnI and CRP in patients with AA and AGA compared to individuals with no hair loss (NHL). By identifying the elevation of CVD biomarkers in patients with either AA or AGA, hair loss could become an important clinical diagnostic feature for higher CVD risk in humans. 5.2 Materials and Methods 5.2.1 Study subject recruitment and blood plasma collection Study participants gave their consent for the study according to University of British Columbia Clinical Research Ethics Board approved protocols. AA affected subjects and 140  controls were recruited from Dr. Jerry Shapiro’s Hair Clinic in the Skin Care Centre of Vancouver General Hospital, and from Vancouver General Hospital - Gordon and Leslie Diamond Health Care Centre - Clinical Research Unit. AA subjects ranged from those with less than 25%, patchy hair loss to subjects with alopecia totalis (AT) and/or alopecia universalis (AU). Control subjects were screened to exclude those with inflammatory hair loss such as cicatricial alopecia and comprised individuals with either no hair loss (NHL) or androgenetic alopecia (AGA).  Investigators completed a questionnaire (Appendix B, C) for each individual regarding their hair loss and background information at the time of blood collection. Peripheral blood (20 mL) was collected by venous puncture into EDTA-coated vacutainers and processed. Peripheral blood mononuclear cells (PBMCs) were obtained by the Ficoll-paque density gradient centrifugation method. Whole blood was diluted 50% with complete medium R10 comprised of RPMI 1640 (catalog #11875-093) 10% FBS (catalog #16000-044), 2 mM Glutamax (catalog #35050-061), and 100 U/mL Streptomycin with 100 μg/mL Penicillin (catalog #15140-148); all from Invitrogen (Burlington, ON), and overlaid on equal volume of Ficoll-Paque Plus (GE Health Care, Burnaby, BC. #17-1440-03) and centrifuged at 600x g for 20 minutes with break set to off to prevent disruption of the buffy coat interphase. A portion of plasma was isolated and frozen in aliquots at -80 °C until use.  5.2.2 cTnI and CRP ELISA analysis Human Troponin I ELISA Kit (Calbiotech, Spring Valley, CA. Catalog #TI015C) and Human C-Reactive Protein ELISA Kit (eBioscience, San Diego, CA. Catalog #88-7502-28) were used to measure plasma levels of cTnI and CRP in the study subjects. Both kits utilized 141  the standard sandwich ELISA protocol. Assay standards of different concentrations were prepared using diluent supplied with the kits to cover a range of 0-75 ng/mL (cTnI) or 0-10 ng/mL (CRP). Equal volumes of sample (or standard) and enzyme conjugate were added to each well and incubated at room temperature for two hours. Subsequently, the wells were washed five times with wash buffer and blot dried. TMB reagent was added to each well and the development of blue color was observed as an indication of positive reaction (Around 10-20 minutes). The reaction was stopped with stop solution and color turned yellow. The plates were read with a spectrophotometer within 15 minutes at 450 nm. Calculation of the results were done by first creating a standard curve with GraphPad Prism (GraphPad Software, La Jolla, CA) which is capable of creating a four-parameter logistic fit. The results were deduced by GraphPad Prism from the average absorbance (OD) reading of the samples. Statistical significance was calculated with one-way ANOVA with multiple comparison using GraphPad Prism software and * denotes significance (p<0.05). 5.2.3 Human primary cardiomyocyte culture Human primary cardiomyocytes (HCMs) were purchased from Applied Biological Materials (ABM) Inc (Richmond, BC. Catalog #T4037). HCMs were cultured in T25 BioCoat Collagen I Coated Vented Flasks (BD Bioscience, Mississauga, ON. Catalog #354485) at 200,000 cells/mL in PriGrow I medium (ABM Inc, Catalog #TM001) supplemented with 10% FBS (Invitrogen, Burlington, ON. Catalog #16000-044) and 100 U/mL Streptomycin with 100 μg/mL Penicillin (Invitrogen, Catalog #15140-148). Half of the culture medium was removed every three days and replenished with fresh complete medium. Once the T25 was at 100% confluence, the HCMs were trypsinized with TrypLE Express 142  (Invitrogen, Catalog #12604-13) and subcultured into T75 BioCoat Collagen I Coated Vented Flasks (BD Bioscience, Catalog #354484). The cultures were maintained until passage 5 when enough HCMs were available for the apoptosis assay. HCMs at the end of passage 4 were subcultured into 24-well BioCoat Collagen I Coated Plates (BD Bioscience, Catalog #356408) at 60,000 cells/500 µL and rested for 2 days before switching to PriGrow medium supplemented with human plasma. 5.2.4 Human primary cardiomyocyte apoptosis assay After the HCMs attached to the collagen I coated plates, PriGrow medium supplemented with human plasma was used to culture the HCMs. Different aliquots of PriGrow medium were supplemented with plasma from either three different AA or AGA subjects with the highest levels of plasma cTnI, three AA or AGA subjects with lowest plasma cTnI, three NHL subjects (all NHL subjects except one had below detectable level of cTnI). Human plasma supplemented PriGrow media were added to each well of HCM to assess the level of apoptosis in HCM resulting from human plasma exposure either directly or indirectly associated with the level of cTnI. The cultures with human plasma were maintained for 10 days with fresh media replenished every 2 days until apoptosis could be observed from positive controls.  The positive controls for apoptosis were wells with either PriGrow media without any serum or RPMI 1640 (Invitrogen, Catalog #11875-093) without any serum. The negative control for apoptosis was PriGrow media with 10% FBS and 100 U/mL Streptomycin with 100 μg/mL Penicillin; the optimal culture condition for HCMs. At the end of 10 days, the HCMs were collected after trypsinization for flow cytometry. 143  5.2.5 Flow cytometry analysis The cell culture supernatant and trypsinized HCMs were combined together, centrifuged (600x g for 5 minutes) to collect live and apoptotic as well as necrotic cells. The cell pellets were washed with complete PriGrow (600x g for 5 minutes) and resuspended in 200 uL FACS staining buffer. The HCMs were washed again with staining buffer and once more with binding buffer as provided from an Annexin V apoptosis detection kit (eBioscience, 5 uL/100uL. catalog #88-8007-72). Following washing, the HCMs were labelled with Annexin V for 15 minutes in the dark on ice. After the staining with Annexin V was completed, the cells were washed once with staining buffer, and resusupend again in 400 µL staining buffer. Propidium Iodide (PI) was added into the cell suspension (5 µL/400 µL) 15 minutes before flow cytometry analysis. Flow cytometry analysis was performed on a BD FACSCanto II flow cytometer (BD Bioscience). Statistical significance was calculated with one-way ANOVA with multiple comparison followed by Fisher’s Least Significant Difference test using GraphPad Prism software and * denotes significance (p≤0.05). 5.3 Results 5.3.1 AA affected patients showed highest level of cTnI and CRP compared to AGA and NHL subjects In ELISA analysis of NHL subjects (n=34), AA (n=89) and AGA (n=72) patients, the plasma levels of cTnI in AA patients were significantly higher on average than both NHL and AGA patients (Figure 5.1a). AGA patients showed higher levels on average of cTnI compared to NHL subjects, but the difference was not statistically significant. The plasma 144  levels of CRP showed similar trends as cTnI, however statistical significance was not achieved (Figure 5.1b).  Figure 5.1 Plasma cTnI and CRP level was highest in AA patients and lowest in the NHL population. AA patients showed significantly higher levels of cTnI in the plasma compared to both AGA patients and the NHL population (a). CRP levels were higher in AA and AGA subject groups but did not reach statistical significance (b). Statistical significance was determined with one-way ANOVA where * denotes p<0.05.   145  5.3.2 AA males had highest levels of cTnI while female subjects with any form of hair loss had higher levels of CRP Males that were affected by AA or AGA had higher levels of cTnI compared to their female counterparts, conversely, males without any hair loss had lower cTnI on average compared to females (Figure 5.2a). Among the three different groups, AA males had significantly higher levels of cTnI compared to NHL males and AGA males. However, there were no significant differences between both genders within each respective group. A different trend was observed with the level of CRP in each group. Female subjects in any group showed higher levels of plasma CRP. Interestingly, AA and AGA males displayed similar levels of CRP; AA and AGA females also displayed similar levels. The comparisons showed a high degree of variability and significance was not achieved (Figure 5.2b).     146   Figure 5.2 AA males had highest levels of cTnI while females in general had highest levels of CRP. AA males had significantly higher levels of cTnI compared to males in other subject groups, but not with females within the same group (a). The level of CRP was higher in females from all subject groups, and was higher in AA and AGA groups, but was not able to reach statistical significance (b). Statistical significance was determined with one-way ANOVA where * denotes p<0.05.   147  5.3.3 Plasma levels of cTnI were higher in younger subjects with AA or AGA while CRP levels were higher in older subjects Younger patients suffering from AA (18-35) displayed higher levels of cTnI compared to age groups 35-50 and greater than 50, but results were not statistically significant. Similarly, AGA patients in the age group 18-35 displayed significantly higher levels of cTnI compared to 35-50 and 50+. While AA patients had overall higher levels of cTnI compared to AGA patients in age groups 35-50 and greater than 50, AGA patients age 18-35 had the highest levels of cTnI among all other groups (Figure 5.3a).  The trend of CRP levels was again discordant to that of cTnI. The results obtained suggested patients in older age groups had higher levels of CRP. Even though no statistically significant trend was obtained with AA patients, the result still points toward higher cTnI in subjects with older age (35-50). Meanwhile, it appeared that AGA patients released more CRP as age increased (Figure 5.3b).  5.3.4 Recent onset of AGA resulted in higher cTnI release while CRP levels were higher in patients with chronic AA AA patients did not show any significant difference between those with recent onset or chronic AA. However, AGA patients with recent onset of hair loss (less than 1 year) exhibited significantly higher levels of cTnI compared to those who had AGA for 1-5 years or over 5 years (Figure 5.4a). An increasing trend in CRP levels were observed in AA patients with more long-standing AA (1-5 years and over 5 years), but such trend was not observed in AGA patients (Figure 5.4b). 148   Figure 5.3 Younger onset of AA and AGA resulted in higher cTnI levels while CRP levels showed an increasing trend with age. In both AA and AGA groups, the level of cTnI was highest in those who were between the age of 18 and 35, but statistical significance was only observed in AGA patients (a). CRP levels did not show a consistent trend for AA patients, but increased linearly as age increased in AGA patients (b). Statistical significance was determined with one-way ANOVA where * denotes p<0.05.   Figure 5.4 Recent onset of AGA resulted in higher cTnI release. No significant difference in the release of cTnI was observed from AA patients with different durations of disease. In contrast, patients with recent onset (<1 year) of AGA had significantly higher levels of cTnI (a). AA patients displayed increasing trend of CRP expression as the duration of AA increased, this trend was not observed in AGA patients (b). Statistical significance was determined with one-way ANOVA where * denotes p<0.05.   149  5.3.5 Patients with AA currently receiving treatments had lower levels of cTnI Many patients suffering from AA are receiving a wide range of different treatments such as corticosteroids and minoxidil. The level of cTnI in those that were receiving treatments at the time of blood collection had significantly lower plasma levels of cTnI. AGA patients who were receiving treatments (finasteride and/or minoxidil) showed no difference in plasma cTnI levels compared to no treatment population (Figure 5.5a). Interestingly, an exact opposite trend was observed in AA patients in terms of plasma CRP level. AA patients who were receiving treatments had higher levels of CRP compared to no treatment, but the difference was not significant. AGA patients also showed no difference in CRP levels regardless of receiving treatments or not (Figure 5.5b). 5.3.6 AA patients with patchy hair loss had highest levels of cTnI but CRP levels were highest in those with 25-75% hair loss AA patients that had less than 25% (patchy) hair loss at the time of blood collection showed significantly higher levels of cTnI compared to those who had 25-75% hair loss. Interestingly, the level of cTnI in those with greater than 75% hair loss was higher than 25-75% hair loss albeit not as high as those with patchy AA (Figure 5.6a). The level of CRP in AA patients with different degree of hair loss exhibited an inversed trend compared to the level of cTnI. Patients with 25-75% hair loss had significantly higher levels of CRP compared to those with less than 25% or greater than 75% hair loss (Figure 5.6b).    150   Figure 5.5 AA patients not currently receiving treatments released higher levels of cTnI. AA patients who were not receiving any treatments for hair loss showed significantly higher levels of cTnI compared to those who were receiving treatments (a). However, the release of CRP was not significantly correlated to treatments (b). Statistical significance was determined with one-way ANOVA where * denotes p<0.05.   Figure 5.6 The release of cTnI and CRP in AA patients with different extents of hair loss showed an inverse trend. AA patients with patchy hair loss (<25% hair loss) had significantly higher levels of cTnI compared to those with moderate amounts of hair loss (25%-75%), but not when compared to patients with extensive, chronic AA (a). Conversely, AA patients with moderate levels of hair loss released significantly higher levels of CRP than those with patchy and chronic hair loss (c). Statistical significance was determined with one-way ANOVA where * denotes p<0.05. 151  5.3.7 AA patient plasma samples with higher levels of cTnI could induce higher rates of apoptosis in HCM The cell culture media that were supplemented with human plasma with higher levels of cTnI instead of FBS were able to induce higher expression of early stage apoptosis marker (Annexin V) as well as late stage apoptosis marker (PI) in HCM, compared to media supplemented with low cTnI human plasma samples (Figure 5.7a,b). Statistical significance was achieved when AA with high cTnI plasma was used compared to low cTnI, as well as to AGA with high cTnI and low cTnI and NHL. There was a lower level of apoptosis observed in HCM cultured in AGA plasma supplemented culture media with either high or low cTnI. AGA plasma with high cTnI induced slightly higher apoptosis marker expression, but the increase was not significant. Media supplemented with plasma from NHL subjects resulted in close to baseline levels of apoptosis in HCM. AGA with high cTnI induced lower levels of necrosis compared to low AGA with low cTnI and NHL group (Figure 5.7c).   152   Figure 5.7 AA patient plasma samples with high levels of cTnI induced early stage apoptosis in primary human cardiomyocytes compared to samples with low levels of cTnI. AA plasma samples with higher levels of cTnI induced significantly higher levels of Annexin V expression in primary HCM compared to their low cTnI level counterpart (a). Significant difference was also observed between AA patients with high cTnI and NHL study group. AA plasma with high cTnI also induced significantly higher PI expression in HCM compared to low cTnI (b). AGA plasma samples with high cTnI actually had significantly lower apoptosis compared to AGA patients with low cTnI and NHL subjects(c). n=3 for each subject group, n=2 for no serum positive control, n=1 for RPMI positive control and n=1 for PriGrow + FBS negative control. Significance was determined using ANOVA followed by Fisher’s Least Significant Difference (LSD) test where * denotes p<0.05.   153  5.4 Discussion There is no paucity of epidemiological studies and meta-analyses that provide evidence of association for various forms of CVDs with the development of hair loss299,360,369-373. To our knowledge, direct investigations beyond clinical observations have rarely been performed, especially for the association of AA with CVDs. In animal studies, a change in the morphology of the heart related to the dysregulation of stress hormones in AA-affected mice was found as described in Chapter 5142. However, with so many variables, such as environmental input and genetic predispositions, associated with human populations, the data available now still cannot establish a firm link between the developments of different forms of hair loss to CVDs in humans.  We evaluated two important biomarkers and predictors for cardiac remodelling (cTnI) and response to inflammation (CRP) in the serum of populations affected with AA and AGA with no hair loss (NHL) subjects as controls. When evaluating the general populations, patients affected by AA showed significantly higher plasma cTnI levels on average compared to AGA and NHL populations. While plasma CRP levels showed similar trend, statistical significance was not achieved.  Clinical manifestation of AA with any active form of heart disease have not been reported beyond brief epidemiological observations60,269,286; routine tests for CVD markers are also not performed in dermatology clinics for hair loss patients. Based on the medical background of the study subjects investigated in this study, none of them were affected by any active forms of CVD. Therefore, the results indicate hair loss patients may be experiencing higher levels of cardiac remodelling even without overt heart diseases, 154  especially in AA patients. A long-term follow up study on a large population is required to confirm this and is beyond our capability.  Even though cTnI can be a predictor of heart failure in advance of myocardial ischemia364, the level of cTnI detected here for all three subject groups were below what normally classified as positive for heart diseases; patients with stable angina, chronic heart failure, left ventricular hypertrophy and subclinical heart diseases generally have at least 10 ng/mL of cTnI365; the average of the concentration of cTnI we obtained in NHL, AA and AGA was 0.26 ng/mL, 1.36 ng/mL and 0.65 ng/mL respectively. Strenuous, high intensity exercises can immediately, but transiently, increase the level of cTnI and normalizes within 24-48 hours; this is one of the main non-cardiac causes of cTnI elevation in healthy subjects365,374. Potentially, AA patients may be more susceptible to cardiac remodeling than AGA and NHL subjects as reflected by the higher plasma cTnI level but the evidence that establish linkage of AA to chronic heart diseases or subclinical forms of heart diseases is still lacking. Recent clinical observations revealed a 21.9% prevalence of hypertension in AA patients60 and early epidemiological studies showed association with subjects with pulmonary disease and open-heart surgery269. Direct comparison of the incidence of CVDs in AA patients with normal subjects has not been reported, it is possible that the increases in CVD biomarkers as identified here are indirectly associated with AA and other triggers are likely required for clinically significant CVDs. In contrast to AA, the development of AGA involves a different mechanism of pathogenesis56,  and its association with CVDs such as hypertension and non-fatal myocardial infarction, is more widely recognized300,360,369. Though notably, the level of both CVD markers analyzed here in AGA were lower than AA patients.  155   The rise of CRP levels in AA patients has been reported367 and it is expected because AA is an inflammatory disease much like psoriasis. It was hypothesized that the circulating inflammatory cytokines can promote heart tissue damage and development of atherosclerosis in psoriasis patients274 and CRP production is stimulated by the elevated inflammatory cytokines375.  Therefore, similar mechanisms may be involved in AA patients, resulting in the rise of CRP, but this alone cannot be an indicator of CVDs.  Our results showed males and females responded differently to the development of AA and AGA. Meta-analysis of AGA studies suggests a younger age of AGA onset in males has a higher association with coronary heart disease369. Similar levels of cTnI released by both male and female AGA patients was observed here, however, male AA patients released significantly higher amounts of cTnI than AGA males. The plasma level of cTnI is known to be higher with the male population even without the presence of CVDs376, but whether this is an indication of higher susceptibility to chronic CVDs in male overall population is not yet identified. Therefore, normal physiological difference in cTnI production by AA and AGA males and females may be what was reflected in our result. Since a significantly higher level of cTnI was observed in AA males compared to AGA males, we might expect a higher degree of cardiac remodeling is associated with AA development.  In an animal study, AA-affected mice showed higher levels of collagen deposition within the heart, around the blood vessels, as well as increased type V collagen gene expression142. Even though the mice in the study were all female, the data suggests a potential for human AA to be associated with cardiac fibrosis; diagnostic tools such as electrocardiogram (ECG) would be required to confirm fibrosis in human patients377.  156  CRP levels were elevated in females with both AA and AGA, but the differences were not significant. It has been reported that females with AGA have higher levels of CRP than without AGA and AGA males301. A similar elevation of CRP secretion in AA affected females over males was also observed in our results. It is still unknown why females with AA or AGA and even NHL would have higher CRP levels than the respective male groups, but one important factor that can affect CRP levels in post-menopausal women is from hormone replacement therapies (HRT)378. Post-menopausal hormones can affect various inflammatory markers including CRP379. Women receiving HRT to reduce discomfort from diminishing estrogen levels have elevated CRP, but we do not have the information on whether female study subjects were receiving HRT or not. Young age of AGA onset is known to be associated with increased risk of CVDs later in life358,360,369. The higher level of cTnI in AGA patients age 18-35 is consistent with these previous reports. The differences in AA patients followed the same trend, but cTnI levels were in general higher than the same age groups in AGA. This suggests that the risks of heart remodelling in AA may be less age-restricted compared to AGA.  The levels of CRP seemed to be dependent on the age of AGA patients, exactly opposite to the cTnI levels. However, this difference might be attributed to the increase of low-grade inflammation as age increases380,381. In AA patients, the level of CRP showed no obvious trend between different age groups. Therefore, as with cTnI, the secretion of CRP in AA patients is unlikely to be significantly affected by the age of the patients. The duration of hair loss in AA patients also did not have a significant correlation to the release of cTnI. While AA patients have overall significantly higher levels of cTnI 157  compared to AGA (except in those who developed AA in the past year), many patients have multiple episodes and different extents of AA hair loss. This could potentially affect the level of plasma cTnI if active AA has a direct and immediate effect on heart remodelling. Further, AA patients could experience hair regrowth on one site and lose hair in another, also changes in immune regulatory genes can occur before overt hair loss is observed125, this unpredictable nature of AA introduces difficulties in data interpretation.  Interestingly, patients with recent onset of AGA had significantly higher levels of cTnI compared to those who have had AGA for a longer time. However, it is still unclear why there was a significantly higher cTnI in AGA patients with more recent onset. There were only six subjects in AGA onset <1 year group which may introduce inaccurate representation of this cohort. No significant difference nor trend was found with the level of CRP, suggesting duration of AA and AGA has no significant effect on CRP production.  Different treatments for AA and AGA may have a significant impact of cardiovascular health. Our results showed that AA subjects not receiving any treatments at the time of blood collection had significantly higher cTnI levels than those who were receiving treatments. There are many different treatment regimens for AA patients based the severity and extent of AA46; 21% of the AA patients in our study were receiving minoxidil as part of their treatment program, a majority of them were also receiving other treatments as the same time. The exact mechanism of minoxidil in treating AA is still unknown, but it was originally developed as a therapy for hypertension because it is a vasodialator46,369; whether AA patients receiving minoxidil as part of their treatments have lower risk of hypertension is unknown. 158  Besides minoxidil, 29% of patients were receiving diphenylcyclopropenone (DPCP) as part of their treatment. DPCP is a contact sensitizer that works by inducing mild contact dermatitis and potentially redirecting the autoimmune inflammatory attack on the hair follicles46. Potentially, DPCP may redirect inflammation away from hair follicles but such inflammation can cause elevation of CRP compared to those without treatments. 15% of AGA patients also receive minoxidil as their treatment, there seems to be no difference on the release of cTnI and CRP in AGA patients regardless whether they were on treatments or not.  In AA patients, there was an inverse trend of cTnI versus CRP secretion in patients with different extents of hair loss. Patients with patchy AA (less than 25%) hair loss had significantly higher levels of cTnI compared to those with 25-75%, but levels were raised again in those with greater than 75% hair loss. It is uncertain whether the extent of hair loss has an effect on heart remodelling as each patient also has a different rate of AA development. Potentially, patchy AA could indicate presence of active inflammation due to presence of targets for inflammatory cells53 and thus higher impact on the health of the heart. In terms of CRP levels, the amount of inflammation in patients with 25-75% hair loss could be greater than those with less than 25% hair loss, however those with greater than 75% hair loss may be experiencing less inflammation due to decrease of terminal hairs, miniaturization of HFs and reduction of inflammatory cell numbers during chronic AA53.  Our investigation on plasma level of cTnI and CRP is limited by the paucity of study subjects in certain categories when the general populations is stratified into finer sub-categories such as in the comparison of different durations of AA and AGA onset. In many 159  cases, the release of cTnI may be heavily modified by the lifestyle of the subjects such as exercise365 while a high-sensitivity CRP ELISA kits (or similar protein assays) may be required to elucidate more subtle changes in CRP levels366.  In the comparison of hair loss durations, effects of treatments and different extents of AA, the results may also reflect more accurate and significant differences between each test variables if a complete medical records of hair loss and/or cardiovascular health status is available from each subject. The investigation on the production or cTnI and CRP would benefit from a well-designed clinical trial that take the factors highlighted above into account. For example, the effect of a specific or a family of AA/AGA drugs on any changes in heart tissues may require an assessment of large populations who is receiving one type of treatment but not another. A large population is also required to stratify test subjects into more well defined sub-populations with reasonable numbers of subjects in each group. For example, AA patients with multiple episodes of AA in a given duration time or patients with a single continuous episode for the same duration of time. The scale of this type of investigation is beyond our goals for this study.  The development of AGA is associated with heightened sensitivity to androgens due to the increase of androgen receptors on hair follicle dermal papilla cells in the balding scalp382 and the increase of free testosterone in the serum369. Testosterones can be metabolized into dihydrotestosterone (DHT) by 5α-reductase and induce miniaturization of hair follicles383. The association of androgen sensitivity to CVDs has been suggested as a result of the expression of 5α-reductase and DHT binding to DHT receptors in the blood vessels and the heart384. DHT can stimulate the proliferation of vascular smooth muscle and 160  directly involve in the progression of atherosclerosis384. Therefore, the elevation of androgens, and eventually DHT in AGA patients may be one of the potential mechanisms associating AGA with CVDs. However, as AA is a completely different form of hair loss, the same mechanism in AGA is unlikely to apply56. In the animal studies, dysregulated stress hormones and HPA axis65 was found and pro-inflammatory cytokine IL-18 was localized in the atria of AA C3H/HeJ mice142 (Chapter 4). The association of AA with CVDs is more likely to be a result of circulating inflammatory cytokines and other mediators due to autoimmune response (below). Even though detailed stratification of AA patients in the ELISA assays mostly revealed insignificant changes, the overall result does indicate positive association with heart tissue remodelling. Even though cTnI and CRP are both predictors for CVDs363,364, they are ultimately markers of different pathological features. The level of CRP can be modulated by many environmental inputs such as old age, gender, obesity and smoking366,378; it is an overall marker for inflammation that is not specific to inflammation in the heart. The level of cTnI has a more direct relationship with the heart since it is a specific type of muscle regulatory protein produced specifically by cardiac muscle cells363,364. This could explain the differences in the trend of secretion within AA or AGA patients. Since cTnI has a more direct relationship to heart damage and remodelling, we selected AA and AGA patients with high or low plasma cTnI to assess whether factors (as yet unknown) might be present in the plasma that could affect the health of human primary cardiomyocytes (HCMs). 161  The apoptosis assays from selected AA and AGA patients showed that AA patients with high levels of cTnI possess unknown factors in their plasma that is/are able to induce significantly higher levels of early and late stage apoptosis marker expression in HCMs, compared to similar AGA and NHL derived plasma samples. The trend indicates AA patients could be at risk of increased cardiac remodeling. Whether there is an association with CVDs such as hypertension60,71 via inflammatory cytokines as produced by the activated autoimmune cells like psoriasis274,275, is unknown. In this study, two important CVD biomarkers were found to be elevated in patients with AA and AGA. This suggests the development of AA hair loss may have a more systemic disease profile as opposed to current beliefs. From our results, individual risk groups cannot be identified. The results also were not corrected for many important environmental inputs (or lifestyles), which can have significant effect on the release of these biomarkers as well as the progression of hair loss. Despite this, we have provided some initial evidence that comorbidities such as heart tissue remodeling may occur at a subclinical level in AA patients and also young AGA patients. Investigations involving a large population with more comprehensive details on patient backgrounds will help elucidate the exact association of AA or AGA with CVD risk.   162  Chapter 6: Major Findings, Conclusions and Future Directions  6.1 Major Findings Although many hypotheses have been put forward to describe the pathogenesis of alopecia areata (AA) over the past decades, the current consensus describes AA as a cell-mediated, inflammatory hair loss disorder with autoimmune mechanisms203. The development of AA has also been described in association with significantly increased prevalence of a wide range of comorbidities over observations in the general population, ranging from atopy to other autoimmune diseases like psoriasis as well as psychiatric disorders71,161. To better understand AA disease mechanisms in vivo and the preclinical efficacy of various experimental treatments, animal models of AA play a crucial role and are often used to help achieve these goals3,189. Several animal models are available to study the biology of AA but each of them has their own strengths and limitations; none of the currently available models can completely reflect human AA3,190,193. The practical use of current animal models of AA are also often limited by the need to perform invasive procedures such as skin grafting96, requirement for human biopsies190, or they have a relatively low and unpredictable rate of spontaneous onset188,191. A method to rapidly and consistently generate a small rodent model that closely resembles human AA without invasive procedures is needed.  In Chapter 2, we have developed a new mouse model for AA, based on the widely used skin-grafted C3H/HeJ mouse model96. We have demonstrated that a cultured skin-163  draining lymph node cell (LNC) injection method of AA induction in C3H/HeJ mice provided a higher success rate of AA induction involving a relatively less skill-oriented procedure than the previous skin-grafting method. The cultured LNCs technique also yielded higher numbers of LNCs that will enable us to induce AA development in at least 50 mice compared to 10-20 from a single donor using the skin graft method.  We studied changes in cell surface markers in AA and control mouse LNCs after culture and found CD4+ and CD8+ T-cells expressing IFNγ were elevated in both AA and control mice (similar rate in AA compared to control). With this study, we also found a difference in the percentage of Tregs in the cultured LNCs where AA LNCs had significantly lower numbers compared to cultured control LNC populations. By labelling the LNCs before we injected them into naïve hosts, we were able to track the localization of LNCs after injection. We found that the injected LNCs did not directly participate in the inflammation of HFs in AA lesions; this suggests more complex interactions between the cultured LNCs with the host immune system may play a role in the onset of disease. We further confirmed the T-cell-mediated nature of AA pathogenesis with the cultured cell transfer technique. Even though the detailed analysis of injected LNC activities was limited by the low sample size, this model potentially enables other applications for studying the mechanism of AA in mice (discussed below).  The activity of immune cells play an important role in the development of AA as has been shown by various animal studies29,44,116,138 and implicated by the treatments used46,47 and histological analyses made in humans78,221. Specifically, autoreactive CD8+ cytotoxic T-cells (CTLs) appear to be the main mediator for the onset of AA89,116 and further exacerbated 164  by CD4+ helper T-cells45,98. However, even with the identification of various subsets of autoreactive T-cells, a key defining component that can confirm AA as an autoimmune disease is still lacking. One of the major questions outstanding is still regarding what is the key autoantigen epitope(s) that is/are involved in the activation of the autoreactive T-cells. Possible autoantigens and autoantigen epitopes have been suggested88,139,222. But, whether the previously identified autoantigens can elicit a CTL response against hair follicle (HF) cells has not been investigated in depth, especially in humans. It has been widely established by immunohistochemistry that AA lesions are associated with CD8+ T-cells (CTLs) invading into the outer and inner root sheaths of HFs with mostly perifollicular localization of CD4+ T-cells221. The presence of CTLs in close proximity of root sheaths persuaded us to investigate CD8 (MHC class I; HLA-A*0201) restricted autoantigen epitopes in AA because of the cells’ cytotoxic nature and potential secretion of pro-inflammatory cytokines upon HF-specific autoantigen epitope challenge116.  In Chapter 3, we compiled a panel of autoantigens that included some of the previously published antigens, and screened for epitope peptide(s) that could elicit higher frequencies of peripheral blood mononuclear cell (PBMC) activation, as well as inducing apoptosis of HF keratinocytes via secretory factors. The strength of this study is that all autoantigen epitope peptides were synthesized and analyzed using the same methods and we sequentially narrowed down the number of candidate peptides. We found that trichohyalin, tyrosinase, and tyrosinase related protein-2 epitope peptides were able to induce significantly higher frequencies of activation (via quantifying IFNγ production) in human AA PBMCs compared to healthy control PBMCs. Further confirmation with IFNγ intracellular cytokine 165  stain (ICS) was able to eliminate tyrosinase peptides for being relatively unspecific stimulators of PBMCs whether from control or AA patients. We stimulated PBMCs with these epitope peptides and treated HF keratinocytes with the conditioned media (CM) of PBMC cultures. We found trichohyalin peptides were able to induce AA PBMCs to secrete certain factors that can induce apoptosis in the keratinocytes. We were unable to detect significant differences in the level of many inflammatory cytokines between AA and control PBMC cultures with a cytokine array; possibly due to the low sensitivity of the array and/or targets detectable by the array we used.  Trichohyalin is a large granule produced by HF and nail keratinocytes and is involved in the mechanical strength of hair follicle242,385. Due to its size, trichohyalin could potentially produce many different epitope peptides that can stimulate AA-specific autoreactive CTLs. The higher frequency of activation from AA PBMCs against trichohyalin could also be the reason why some AA patients experience the “pitted nail” clinical presentation53.  This multistep validation of autoantigen epitopes in AA provides a solid foundation for potential future translational research. We utilized the same method to identify autoantigen epitope peptides in AA-affected C3H/HeJ mice with mouse homologs of the antigens. We found that instead of trichohyalin peptides, AA mouse LNCs were significantly activated by cytokeratin 16 peptides as well as MART1. It is likely that AA-affected C3H/HeJ mice have different primary antigen epitope target(s) than human AA. Despite that, we have shown that HF keratinocyte derived antigen epitopes were able to induce AA PBMC activation in both humans and mice with AA.  166  Whether direct cell-to-cell contact, or the presence of soluble factors, is the main mechanism of HF disruption by the CTLs is unexplored. Also the type of cytokines produced by the peptide-stimulated PBMCs/LNCs remains unknown and requires further investigation. Furthermore, as the study subjects had different states of AA, with possible epitope spreading involved over time256,257, the initiating autoantigen epitope(s) remain to be discovered. With the activation of autoimmune and auto-reactive T-cell responses in AA patients and mice, there is a concern raised as to whether the sequelae of AA can involve the onset of other diseases that could severely affect the quality of life60,71. However, much of the currently available data on disease associations with AA are derived from epidemiological or retrospective studies. There are very limited investigations on whether AA patients express disease makers for associated comorbid diseases, such as cardiovascular diseases (CVDs). There are also no studies utilizing animal models to address this question. In Chapter 4, we found that AA-affected C3H/HeJ mice exhibited enlarged hearts, higher deposition of collagen, as well as thinning of blood vessels in the heart, compared to control mice. We were able to relate the high IL18 expression with changes in heart morphology in AA mice by showing higher IL18 localization in the atria of AA mouse hearts. There was also a fluctuation in Il18 and related genes (Il18r1, Il18bp, Casp1) near the time of AA onset. Furthermore, we have shown that AA mice have elevated levels of a CVD marker, cardiac troponin I (cTnI), in the blood and heart tissues compared to controls.  The potential of stress hormone related changes in heart morphology was investigated as it has been shown that AA mice have dysregulated stress hormones65. We found that AA mouse atria tissues responded differently to increased levels of adrenocorticotrophic hormone 167  (ACTH) compared to control mice. The increase of ACTH increased the overall expression of Il18, Il18r1, and Casp1 while decreasing Il18bp in AA mouse atria.  This may suggest an increased level of the active form of IL18 (due to increase of Casp1) and increased sensitivity with decreased amounts of IL18 antagonist in AA mouse hearts; however, confirmatory experiments that evaluate the activity of caspase 1 are required to confirm this hypothesis. The increase of ACTH also significantly increased the expression of the type V collagen gene (promotes type I collagen assembly) and the release of cTnI. This result potentially suggests that the dysregulation of stress hormones such as ACTH in AA-affected mice could affect the health of the heart partially through an increase of IL18 expression in the heart. In Chapter 5 of this dissertation, we evaluated the levels of two different types of heart disease markers, cTnI and C-reactive protein (CRP), in the plasma of human patients with AA, androgenetic alopecia (AGA) and no hair loss (NHL). In this study, the level of the two CVD markers in the test subjects was stratified by gender, age groups, duration of hair loss, extent of hair loss and presence of treatments. While we found that both markers were generally elevated in patients with AA or AGA (in the overall population cTnI was significantly higher in AA patients compared to AGA and NHL), stratification of test subjects into small sub-populations yielded inconsistent trends.  However, we showed the plasma of AA patients with high levels of cTnI induced a significantly higher level of apoptosis in primary human cardiomyocytes. This could imply the higher level of cTnI in AA patients may be associated with cardiac remodeling. The remodeling of the heart is likely at a subclinical level because the concentration of cTnI in 168  AA patients was below the level indicated for CVDs. Nevertheless, these results provide an initial assessment of cardiovascular health in AA and AGA patients and how the development of hair loss, especially AA, may be associated with changes in CVD risk compared to a NHL population.  6.2 Conclusions 1. We have established a new mouse model for AA research by demonstrating AA is a cell-mediated disease where only the injection of cultured AA LNCs, but not control LNCs, can induce AA development. Further, we have found an interaction must have occurred between the LNCs we cultured and the host immune system as the injected LNCs did not localize in host AA lesions. 2. Among all the human autoantigen epitope peptides tested, epitope peptides derived from trichohyalin can induce significantly higher frequencies of AA PBMC activation and increase the production harmful soluble factors upon activation. These soluble factors were able to stimulate apoptosis in HF keratinocytes, which may be one of the mechanisms for hair loss in AA. A different group of keratinocyte autoantigens, cytokeratin 16 epitopes, were able to induce significantly higher frequencies of activation in AA mouse LNCs. Human and mouse data suggest that, even though HF keratinocyte autoantigens may not be the initial target, they are likely a primary target of autoreactive CTLs throughout the course of AA. 3. AA-affected mice showed significant changes in heart morphology compared to control mice. Human patients with AA also displayed significantly higher levels 169  of cTnI than AGA and NHL. These results suggest that even though the changes in the heart health may be at a subclinical level, the sequelae of AA may have an impact on other tissues and organs beyond the skin. In summary, the work performed in this dissertation provided an efficient method to generate an AA mouse model that closely resembles human AA. Through the studies on autoantigen epitopes in AA humans and mice, we provided novel data suggesting the important role of HF keratinocyte derived autoantigen epitopes in the development of AA. Finally, we provided evidence that AA is potentially a multi-organ disease that affects more than HFs; careful evaluation of cardiovascular health may be required for AA patients who also possess other CVD risk factors. 6.3 Future Directions In regards to the injection of cultured LNC AA mouse model, a more comprehensive cell phenotyping assay is required. This can be achieved with multi-color flow cytometry analysis on important T-cell surface markers. Better understanding of cell populations being expanded can not only enhance the efficiency of generating the model, but may also identify the important T-cell subset(s) that can induce AA, which may be an important first step to future targeted therapy. In addition, the activity of injected LNCs needs to be investigated, we have performed an initial assessment as to where the injected LNCs localize after the development of AA via fluorescence microscopy. However the study was limited to four weeks after the injection; a longer observation period is required. Further, the injected cells are subjected to normal cycling and apoptosis, resulting in dye dilution216, similar 170  histological analysis needs to be performed at earlier times in addition to flow cytometry performed in Chapter 2.  One of the important questions to ask is if LNCs obtained from AA mouse skin-draining lymph nodes can be re-stimulated with specific peptides identified in Chapter 3 instead of anti-CD3/anti-CD28 coated beads and expanded to enough number to induce AA in naïve mice. Alternatively, the same mouse IFNγ ELISpot assay can be performed on LNCs after culture to see whether there is an increase in certain epitope peptide-specific LNCs. Both of these approaches will provide insight on what are the key autoantigen epitopes in C3H/HeJ mice. In the present dissertation, some of the investigations were limited by small population size and heterogeneity in the human population. Recruitment of more subjects with HLA-A*0201 serotype in each disease state would be required to evaluate the ability of each single epitope peptides to elicit PBMC activation as well as to effectively correct for demographic and medical backgrounds. In order to translate the epitope peptide(s) identified in Chapter 3 to an effective clinical diagnostic tool, a larger population cohort is required to validate the consistency and accuracy of the results. This can potentially provide valuable information on identifying the primary autoantigen epitope(s) in patients with different AA disease manifestation or stage.  The use of MHC class I tetramers may be a valuable step in further validating the identified epitope peptides. MHC class I tetramers work by fusing multiple MHC class I complex designed with specificity to the epitope peptide of interest to form a probe that has high affinity to antigen specific T-cells386. Such a technique has been used to identify 171  autoantigen specific T-cell clones in Type 1 diabetes387. Potentially, this can translate the identified peptides in the laboratory into useful clinical biomarkers388. In Chapter 5, a larger sample size could enable more meaningful analysis of CVD biomarkers in specific sub-populations of patients with AA and AGA. If a large population size can enable the stratification of AA and AGA patients into individual CVD risk groups, this may potentially improve health care costs by helping physicians order specific tests or prescribe specific treatments for different populations instead of a large panel of tests. An example of this has been demonstrated recently and showed that tests for thyroid diseases are useful for AA patients over the age of 20, but not in children71. To confirm the association of CVDs with AA, additional protein assays for other commonly used CVD biomarkers may be used to establish a stronger link. Cardiac troponin T (cTnT) and creatin kinase (CK-MB) are some of the other biomarkers for myocardial injury and may be more relevant to changes in heart morphology, as observed in AA-affected mice, compared to using CRP (elevated during inflammation in general). In the study that involved mice, certain parameters for cardiovascular health can also be measured. Blood pressure of AA-affected mice can be compared with controls as hypertension can be correlated to the enlargement of the heart389. Magnetic resonance imaging (MRI)390 or echocardiography391 on the heart may also provide a non-invasive look on changes in heart morphologies.    172  Bibliography  1 Wang, E. & McElwee, K. J. Etiopathogenesis of alopecia areata: Why do our patients get it? Dermatologic therapy 24, 337-347, (2011). 2 Robinson, T. The etiology, pathology, and treatment of baldness & greyness. 2 edn,  (London, Henry Kimpton, 1882). 3 McElwee, K. J. et al. What can we learn from animal models of Alopecia areata? Dermatology 211, 47-53, (2005). 4 Sauvages. Nosologica Medica.  (Lyon, 1760). 5 Wasserman, D., Guzman-Sanchez, D. A., Scott, K. & McMichael, A. Alopecia areata. International journal of dermatology 46, 121-131, (2007). 6 Sabouraud, R. Sur les origines de la pelade. Annales de Dermatologie et de Syphiligraphie Series 3, 253-277, (1896). 7 Barber, H. W. & Zamora, A. M. Alopecia areata. With a note on the estimation of the pathogenicity of the tonsil. Brit J Dermatol Syph 33, 1-14, (1921). 8 Bowen, J. T. Two epidemics of alopecia areata in an asylum for girls. Journal of Cutaneous and Genito-Urinary Disease 17, 399-404, (1899). 9 Colcott Fox, T. On a small epidemic of alopecia areata. The British journal of dermatology 25, 51-56, (1913). 10 Davis, H. Epidemic alopecia areata. The British journal of dermatology 26, 207-210, (1914). 11 Ikeda, T. Produced alopecia areata based on the focal infection theory and mental motive theory. Dermatologica 134, 1-11, (1967). 12 Ormsby, O. S. & Montgomery, H. Diseases of the skin. 7 edn,  1365-1375 (London, Henry Kimpton, 1948). 13 Roxburgh, A. C. Common skin diseases. 9 edn,  406-412 (London, HK Lewis & Co. Ltd, 1950). 14 Savill, A. & Warner, E. C. Savill's system of clinical medicine. 11 edn,  799-801 (London, Edward Arnold & Co, 1939). 15 Skinner, R. B., Jr., Light, W. H., Bale, G. F., Rosenberg, E. W. & Leonardi, C. Alopecia areata and presence of cytomegalovirus DNA. JAMA : the journal of the American Medical Association 273, 1419-1420, (1995). 16 Skinner, R. B., Jr., Light, W. H., Leonardi, C., Bale, G. F. & Rosenberg, E. W. A molecular approach to alopecia areata. The Journal of investigative dermatology 104, 3S-4S, (1995). 17 McElwee, K. J. et al. Comparison of alopecia areata in human and nonhuman mammalian species. Pathobiology : journal of immunopathology, molecular and cellular biology 66, 90-107, (1998). 18 Garcia-Hernandez, M. J. et al. No evidence of cytomegalovirus DNA in alopecia areata. The Journal of investigative dermatology 110, 185, (1998). 19 Tosti, A. et al. No correlation between cytomegalovirus and alopecia areata. The Journal of investigative dermatology 107, 443, (1996). 173  20 Jackow, C. et al. Alopecia areata and cytomegalovirus infection in twins: genes versus environment? Journal of the American Academy of Dermatology 38, 418-425, (1998). 21 McElwee, K. J., Tobin, D. J., Bystryn, J. C., King, L. E., Jr. & Sundberg, J. P. Alopecia areata: an autoimmune disease? Exp Dermatol 8, 371-379, (1999). 22 Kingsbury, J. Three cases of alopecia areata associated with ametropia. J Cutaneous Disease 27, 211, (1909). 23 Freshwater, D. Alopecia areata. The Practitioner 95, 796-806, (1915). 24 Gohlke & Holtschmidt. Neurohistologische Studien bei Alopecia areata. Arch Dermatol Syphilol 191, 527, (1949). 25 Joseph, M. Experimentelle Untersuchungen uber die Atiologie der Alopecia areata. Monatsh Prakt Dermat 5, 483, (1886). 26 Wright, C. S. Etiology of alopecia areata - With special reference to the effect of experimental nerve injuries. Arch Dermatol Syph 19, 365-377, (1929). 27 Hordinsky, M. K., Kennedy, W., Wendelschafer-Crabb, G. & Lewis, S. Structure and function of cutaneous nerves in alopecia areata. The Journal of investigative dermatology 104, 28S-29S, (1995). 28 Hordinsky, M. K. & Ericson, M. E. Relationship between follicular nerve supply and alopecia. Dermatologic clinics 14, 651-660, (1996). 29 Gilhar, A., Paus, R. & Kalish, R. S. Lymphocytes, neuropeptides, and genes involved in alopecia areata. J Clin Invest 117, 2019-2027, (2007). 30 Jacquet, L. Pathogenic de la pelade. Bull Soc Fr Dermatol 13, 299-322, (1902). 31 Rousseau-Decelle, L. Bull et Mem Soc Med de Hop de Paris.  21, 72, (1909). 32 Lesclous, P. & Maman, L. An unusual case of alopecia areata of dental origin. Oral surgery, oral medicine, oral pathology, oral radiology, and endodontics 84, 290-292, (1997). 33 Bailly. L’origine gingivodentaire de la pelade.  (Lyon, Paris, 1910). 34 Kinnear, J. Gardiner's Handbook of Skin Diseases. 4 edn,  224-228 (Edinburgh, E&S Livingstone, 1939). 35 Sabouraud, R. Pelade et goitre exopthalmique. Annales de Dermatologie et de Syphiligraphie Series 5, 140-148, (1913). 36 Adamson, H. G. Goulstonian lectures on modern views upon the significance of skin eruptions.  (London, John Bale, Sons & Danielsson Ltd, 1912). 37 Dixon, W. E. Thallium. Proceedings of the Royal Society of Medicine 20, 1197-1200, (1927). 38 Dowling, G. B. Ringworm of the Scalp treated by Thallium Acetate Epilation. Proceedings of the Royal Society of Medicine 20, 1055-1056, (1927). 39 Giovannini, S. Recherches sur l'histologie pathologique de la pelade. Annales de Dermatologie et de Syphiligraphie Series 3, 923-957, (1891). 40 Van Scott, E. J. Morphologic changes in pilosebaceous units and anagen hairs in alopecia areata. The Journal of investigative dermatology 31, 35-43, (1958). 41 Van Scott, E. J. Evaluation of disturbed hair growth in alopecia areata and other alopecias. Ann N Y Acad Sci 83, 480-490, (1959). 174  42 Perret, C., Wiesner-Menzel, L. & Happle, R. Immunohistochemical analysis of T-cell subsets in the peribulbar and intrabulbar infiltrates of alopecia areata. Acta dermato-venereologica 64, 26-30, (1984). 43 Ranki, A., Kianto, U., Kanerva, L., Tolvanen, E. & Johansson, E. Immunohistochemical and electron microscopic characterization of the cellular infiltrate in alopecia (areata, totalis, and universalis). The Journal of investigative dermatology 83, 7-11, (1984). 44 McElwee, K. J., Spiers, E. M. & Oliver, R. F. In vivo depletion of CD8+ T cells restores hair growth in the DEBR model for alopecia areata. The British journal of dermatology 135, 211-217, (1996). 45 McElwee, K. J., Spiers, E. M. & Oliver, R. F. Partial restoration of hair growth in the DEBR model for Alopecia areata after in vivo depletion of CD4+ T cells. The British journal of dermatology 140, 432-437, (1999). 46 Alkhalifah, A., Alsantali, A., Wang, E., McElwee, K. J. & Shapiro, J. Alopecia areata update: part II. Treatment. Journal of the American Academy of Dermatology 62, 191-202, quiz 203-194, (2010). 47 Hordinsky, M. & Donati, A. Alopecia areata: an evidence-based treatment update. American journal of clinical dermatology 15, 231-246, (2014). 48 Lundin, M. et al. Gender differences in alopecia areata. Journal of drugs in dermatology : JDD 13, 409-413, (2014). 49 Kavak, A. et al. Alopecia areata in Turkey: demographic and clinical features. Journal of the European Academy of Dermatology and Venereology : JEADV 22, 977-981, (2008). 50 Safavi, K. H., Muller, S. A., Suman, V. J., Moshell, A. N. & Melton, L. J., 3rd. Incidence of alopecia areata in Olmsted County, Minnesota, 1975 through 1989. Mayo Clin Proc 70, 628-633, (1995). 51 Mirzoyev, S. A., Schrum, A. G., Davis, M. D. & Torgerson, R. R. Lifetime incidence risk of alopecia areata estimated at 2.1% by Rochester Epidemiology Project, 1990-2009. The Journal of investigative dermatology 134, 1141-1142, (2014). 52 Kos, L. & Conlon, J. An update on alopecia areata. Curr Opin Pediatr 21, 475-480, (2009). 53 Alkhalifah, A., Alsantali, A., Wang, E., McElwee, K. J. & Shapiro, J. Alopecia areata update: part I. Clinical picture, histopathology, and pathogenesis. Journal of the American Academy of Dermatology 62, 177-188, quiz 189-190, (2010). 54 Price, V. H. Therapy of alopecia areata: on the cusp and in the future. The journal of investigative dermatology. Symposium proceedings / the Society for Investigative Dermatology, Inc. [and] European Society for Dermatological Research 8, 207-211, (2003). 55 Thomas, E. A. & Kadyan, R. S. Alopecia areata and autoimmunity: a clinical study. Indian journal of dermatology 53, 70-74, (2008). 56 Breitkopf, T., Leung, G., Yu, M., Wang, E. & McElwee, K. J. The basic science of hair biology: what are the causal mechanisms for the disordered hair follicle? Dermatologic clinics 31, 1-19, (2013). 175  57 Aghaei, S., Saki, N., Daneshmand, E. & Kardeh, B. Prevalence of psychological disorders in patients with alopecia areata in comparison with normal subjects. ISRN dermatology 2014, 304370, (2014). 58 Inui, S., Inoue, T. & Itami, S. Psychosocial impact of wigs or hairpieces on perceived quality of life level in female patients with alopecia areata. J Dermatol 40, 225-226, (2013). 59 Sinclair, R. D. Alopecia areata and suicide of children. The Medical journal of Australia 200, 145, (2014). 60 Huang, K. P., Mullangi, S., Guo, Y. & Qureshi, A. A. Autoimmune, atopic, and mental health comorbid conditions associated with alopecia areata in the United States. JAMA dermatology 149, 789-794, (2013). 61 Colon, E. A., Popkin, M. K., Callies, A. L., Dessert, N. J. & Hordinsky, M. K. Lifetime prevalence of psychiatric disorders in patients with alopecia areata. Comprehensive psychiatry 32, 245-251, (1991). 62 Bashir, K., Dar, N. R. & Rao, S. U. Depression in adult dermatology outpatients. Journal of the College of Physicians and Surgeons--Pakistan : JCPSP 20, 811-813, (2010). 63 Ghanizadeh, A. Comorbidity of psychiatric disorders in children and adolescents with alopecia areata in a child and adolescent psychiatry clinical sample. International journal of dermatology 47, 1118-1120, (2008). 64 Kim, H. S. et al. Immunoreactivity of corticotropin-releasing hormone, adrenocorticotropic hormone and alpha-melanocyte-stimulating hormone in alopecia areata. Exp Dermatol 15, 515-522, (2006). 65 Zhang, X. et al. Development of alopecia areata is associated with higher central and peripheral hypothalamic-pituitary-adrenal tone in the skin graft induced C3H/HeJ mouse model. The Journal of investigative dermatology 129, 1527-1538, (2009). 66 Hsieh, F. I. & Chiou, H. Y. Stroke: morbidity, risk factors, and care in taiwan. Journal of stroke 16, 59-64, (2014). 67 Martinac, M. et al. Metabolic syndrome, activity of the hypothalamic-pituitary-adrenal axis and inflammatory mediators in depressive disorder. Acta clinica Croatica 53, 55-71, (2014). 68 Ebner, N. & von Haehling, S. Iron deficiency in heart failure: a practical guide. Nutrients 5, 3730-3739, (2013). 69 Wang, S. J. et al. Increased risk for type I (insulin-dependent) diabetes in relatives of patients with alopecia areata (AA). American journal of medical genetics 51, 234-239, (1994). 70 Petukhova, L. et al. Genome-wide association study in alopecia areata implicates both innate and adaptive immunity. Nature 466, 113-117, (2010). 71 Chu, S. Y. et al. Comorbidity profiles among patients with alopecia areata: The importance of onset age, a nationwide population-based study. Journal of the American Academy of Dermatology, (2011). 72 Cho, H. H. et al. Clinical characteristics and prognostic factors in early-onset alopecia totalis and alopecia universalis. Journal of Korean medical science 27, 799-802, (2012). 176  73 Lu, W. et al. Alopecia areata: pathogenesis and potential for therapy. Expert reviews in molecular medicine 8, 1-19, (2006). 74 Gilhar, A., Etzioni, A. & Paus, R. Alopecia areata. The New England journal of medicine 366, 1515-1525, (2012). 75 Zhang, J. G. & Oliver, R. F. Immunohistological study of the development of the cellular infiltrate in the pelage follicles of the DEBR model for alopecia areata. The British journal of dermatology 130, 405-414, (1994). 76 Gilhar, A. & Kalish, R. S. Alopecia areata: a tissue specific autoimmune disease of the hair follicle. Autoimmun Rev 5, 64-69, (2006). 77 Todes-Taylor, N., Turner, R., Wood, G. S., Stratte, P. T. & Morhenn, V. B. T cell subpopulations in alopecia areata. Journal of the American Academy of Dermatology 11, 216-223, (1984). 78 Whiting, D. A. Histopathologic features of alopecia areata: a new look. Arch Dermatol 139, 1555-1559, (2003). 79 El Darouti, M., Marzouk, S. A. & Sharawi, E. Eosinophils in fibrous tracts and near hair bulbs: A helpful diagnostic feature of alopecia areata. Journal of the American Academy of Dermatology 42, 305-307, (2000). 80 Elston, D. M., McCollough, M. L., Bergfeld, W. F., Liranzo, M. O. & Heibel, M. Eosinophils in fibrous tracts and near hair bulbs: a helpful diagnostic feature of alopecia areata. Journal of the American Academy of Dermatology 37, 101-106, (1997). 81 Spath, U. & Steigleder, G. K. [Number of mast cells (MC) in Alopecia areata]. Zeitschrift fur Haut- und Geschlechtskrankheiten 45, 435-436, (1970). 82 Freyschmidt-Paul, P. & Hoffmann, R. in Hair Growth and Disorders   (eds D. Whiting, A. Tosti, & R. Trueb)  311-320 (Springer, 2008). 83 Kutner, A. & Friedman, A. Hair loss in the dermatology office: an update on alopecia areata. Journal of drugs in dermatology : JDD 12, 588-593, (2013). 84 Messenger, A. G. & Bleehen, S. S. Alopecia areata: light and electron microscopic pathology of the regrowing white hair. The British journal of dermatology 110, 155-162, (1984). 85 Messenger, A. G., Bleehen, S. S., Slater, D. N. & Rooney, N. Expression of HLA-DR in hair follicles in alopecia areata. Lancet 2, 287-288, (1984). 86 Brocker, E. B., Echternacht-Happle, K., Hamm, H. & Happle, R. Abnormal expression of class I and class II major histocompatibility antigens in alopecia areata: modulation by topical immunotherapy. The Journal of investigative dermatology 88, 564-568, (1987). 87 Paus, R., Slominski, A. & Czarnetzki, B. M. Is alopecia areata an autoimmune-response against melanogenesis-related proteins, exposed by abnormal MHC class I expression in the anagen hair bulb? The Yale journal of biology and medicine 66, 541-554, (1993). 88 Leung, M. C., Sutton, C. W., Fenton, D. A. & Tobin, D. J. Trichohyalin is a potential major autoantigen in human alopecia areata. Journal of proteome research 9, 5153-5163, (2010). 89 Xing, L. et al. Alopecia areata is driven by cytotoxic T lymphocytes and is reversed by JAK inhibition. Nature medicine, (2014). 177  90 Mehling, A. & Beissert, S. Dendritic cells under investigation in autoimmune disease. Critical reviews in biochemistry and molecular biology 38, 1-21, (2003). 91 Ueno, H. et al. Dendritic cell subsets in health and disease. Immunological reviews 219, 118-142, (2007). 92 Freyschmidt-Paul, P. et al. Treatment with an anti-CD44v10-specific antibody inhibits the onset of alopecia areata in C3H/HeJ mice. The Journal of investigative dermatology 115, 653-657, (2000). 93 Zoller, M., Gupta, P., Marhaba, R., Vitacolonna, M. & Freyschmidt-Paul, P. Anti-CD44-mediated blockade of leukocyte migration in skin-associated immune diseases. Journal of leukocyte biology 82, 57-71, (2007). 94 Zoller, M. et al. Chronic delayed-type hypersensitivity reaction as a means to treat alopecia areata. Clinical and experimental immunology 135, 398-408, (2004). 95 Gupta, P. et al. A chronic contact eczema impedes migration of antigen-presenting cells in alopecia areata. The Journal of investigative dermatology 126, 1559-1573, (2006). 96 McElwee, K. J., Boggess, D., King, L. E., Jr. & Sundberg, J. P. Experimental induction of alopecia areata-like hair loss in C3H/HeJ mice using full-thickness skin grafts. The Journal of investigative dermatology 111, 797-803, (1998). 97 Gilhar, A. et al. Mediation of alopecia areata by cooperation between CD4+ and CD8+ T lymphocytes: transfer to human scalp explants on Prkdc(scid) mice. Arch Dermatol 138, 916-922, (2002). 98 McElwee, K. J. et al. Transfer of CD8(+) cells induces localized hair loss whereas CD4(+)/CD25(-) cells promote systemic alopecia areata and CD4(+)/CD25(+) cells blockade disease onset in the C3H/HeJ mouse model. The Journal of investigative dermatology 124, 947-957, (2005). 99 Carroll, J. M., McElwee, K. J., L, E. K., Byrne, M. C. & Sundberg, J. P. Gene array profiling and immunomodulation studies define a cell-mediated immune response underlying the pathogenesis of alopecia areata in a mouse model and humans. The Journal of investigative dermatology 119, 392-402, (2002). 100 Zauli, D. et al. Autoantibodies in alopecia areata. The British journal of dermatology 111, 247, (1984). 101 Betterle, C., Peserico, A., Del Prete, G. & Trisotto, A. Letter: Autoantibodies in alopecia areata. Arch Dermatol 111, 927-928, (1975). 102 Cochran, R. E., Thomson, J. & MacSween, R. N. An auto-antibody profile in alopecia totalis and diffuse alopecia. The British journal of dermatology 95, 61-65, (1976). 103 Galbraith, G. M., Thiers, B. H., Vasily, D. B. & Fudenberg, H. H. Immunological profiles in alopecia areata. The British journal of dermatology 110, 163-170, (1984). 104 Friedmann, P. S. Alopecia areata and auto-immunity. The British journal of dermatology 105, 153-157, (1981). 105 Muller, H. K., Rook, A. J. & Kubba, R. Immunohistology and autoantibody studies in alopecia areata. The British journal of dermatology 102, 609-610, (1980). 106 Kemp, E. H., Sandhu, H. K., Weetman, A. P. & McDonagh, A. J. Demonstration of autoantibodies against tyrosine hydroxylase in patients with alopecia areata. The British journal of dermatology 165, 1236-1243, (2011). 178  107 Tobin, D. J., Orentreich, N., Fenton, D. A. & Bystryn, J. C. Antibodies to hair follicles in alopecia areata. The Journal of investigative dermatology 102, 721-724, (1994). 108 Tobin, D. J., Orentreich, N. & Bystryn, J. C. Autoantibodies to hair follicles in normal individuals. Arch Dermatol 130, 395-396, (1994). 109 Tobin, D. J. & Bystryn, J. C. Immunity to hair follicles in alopecia areata. The Journal of investigative dermatology 104, 13S-14S, (1995). 110 Bystryn, J. C., Orentreich, N. & Stengel, F. Direct immunofluorescence studies in alopecia areata and male pattern alopecia. The Journal of investigative dermatology 73, 317-320, (1979). 111 Tobin, D. J., Hann, S. K., Song, M. S. & Bystryn, J. C. Hair follicle structures targeted by antibodies in patients with alopecia areata. Arch Dermatol 133, 57-61, (1997). 112 Tobin, D. J. Characterization of hair follicle antigens targeted by the anti-hair follicle immune response. The journal of investigative dermatology. Symposium proceedings / the Society for Investigative Dermatology, Inc. [and] European Society for Dermatological Research 8, 176-181, (2003). 113 McElwee, K. J., Pickett, P. & Oliver, R. F. The DEBR rat, alopecia areata and autoantibodies to the hair follicle. The British journal of dermatology 134, 55-63, (1996). 114 Gilhar, A., Pillar, T., Assay, B. & David, M. Failure of passive transfer of serum from patients with alopecia areata and alopecia universalis to inhibit hair growth in transplants of human scalp skin grafted on to nude mice. The British journal of dermatology 126, 166-171, (1992). 115 McElwee, K. J. et al. Resistance to alopecia areata in C3H/HeJ mice is associated with increased expression of regulatory cytokines and a failure to recruit CD4+ and CD8+ cells. The Journal of investigative dermatology 119, 1426-1433, (2002). 116 Bodemer, C. et al. Role of cytotoxic T cells in chronic alopecia areata. The Journal of investigative dermatology 114, 112-116, (2000). 117 Freyschmidt-Paul, P. et al. Fas-deficient C3.MRL-Tnfrsf6(lpr) mice and Fas ligand-deficient C3H/HeJ-Tnfsf6(gld) mice are relatively resistant to the induction of alopecia areata by grafting of alopecia areata-affected skin from C3H/HeJ mice. The journal of investigative dermatology. Symposium proceedings / the Society for Investigative Dermatology, Inc. [and] European Society for Dermatological Research 8, 104-108, (2003). 118 Ito, T. & Tokura, Y. The role of cytokines and chemokines in the T cell-mediated autoimmune process in alopecia areata. Exp Dermatol, (2014). 119 Ghoreishi, M., Martinka, M. & Dutz, J. P. Type 1 interferon signature in the scalp lesions of alopecia areata. The British journal of dermatology 163, 57-62, (2010). 120 Kasumagic-Halilovic, E., Prohic, A. & Karamehic, J. Serum concentrations of interferon-gamma (IFN-g) in patients with alopecia areata: correlation with clinical type and duration of the disease. Medicinski arhiv 64, 212-214, (2010). 121 Tobin, D. J., Fenton, D. A. & Kendall, M. D. Cell degeneration in alopecia areata. An ultrastructural study. The American Journal of dermatopathology 13, 248-256, (1991). 179  122 Gregoriou, S. et al. Cytokines and other mediators in alopecia areata. Mediators of inflammation 2010, 928030, (2010). 123 Ito, T., Meyer, K. C., Ito, N. & Paus, R. Immune privilege and the skin. Current directions in autoimmunity 10, 27-52, (2008). 124 Paus, R., Ito, N., Takigawa, M. & Ito, T. The hair follicle and immune privilege. The journal of investigative dermatology. Symposium proceedings / the Society for Investigative Dermatology, Inc. [and] European Society for Dermatological Research 8, 188-194, (2003). 125 Kang, H. et al. Hair follicles from alopecia areata patients exhibit alterations in immune privilege-associated gene expression in advance of hair loss. The Journal of investigative dermatology 130, 2677-2680, (2010). 126 Paus, R., Christoph, T. & Muller-Rover, S. Immunology of the hair follicle: a short journey into terra incognita. The journal of investigative dermatology. Symposium proceedings / the Society for Investigative Dermatology, Inc. [and] European Society for Dermatological Research 4, 226-234, (1999). 127 Ito, T. et al. Collapse and restoration of MHC class-I-dependent immune privilege: exploiting the human hair follicle as a model. Am J Pathol 164, 623-634, (2004). 128 Ghersetich, I., Campanile, G. & Lotti, T. Alopecia areata: immunohistochemistry and ultrastructure of infiltrate and identification of adhesion molecule receptors. International journal of dermatology 35, 28-33, (1996). 129 Gilhar, A. Collapse of immune privilege in alopecia areata: coincidental or substantial? The Journal of investigative dermatology 130, 2535-2537, (2010). 130 Botchkareva, N. V., Ahluwalia, G. & Shander, D. Apoptosis in the hair follicle. The Journal of investigative dermatology 126, 258-264, (2006). 131 Weedon, D. & Strutton, G. Apoptosis as the mechanism of the involution of hair follicles in catagen transformation. Acta dermato-venereologica 61, 335-339, (1981). 132 Lindner, G. et al. Analysis of apoptosis during hair follicle regression (catagen). Am J Pathol 151, 1601-1617, (1997). 133 Eichmuller, S. et al. Clusters of perifollicular macrophages in normal murine skin: physiological degeneration of selected hair follicles by programmed organ deletion. The journal of histochemistry and cytochemistry : official journal of the Histochemistry Society 46, 361-370, (1998). 134 Chow, S., Rizzo, C., Ravitskiy, L. & Sinha, A. A. The role of T cells in cutaneous autoimmune disease. Autoimmunity 38, 303-317, (2005). 135 McElwee, K., Freyschmidt-Paul, P., Ziegler, A., Happle, R. & Hoffmann, R. Genetic susceptibility and severity of alopecia areata in human and animal models. Eur J Dermatol 11, 11-16, (2001). 136 Tobin, D. J., Sundberg, J. P., King, L. E., Jr., Boggess, D. & Bystryn, J. C. Autoantibodies to hair follicles in C3H/HeJ mice with alopecia areata-like hair loss. The Journal of investigative dermatology 109, 329-333, (1997). 137 Albert, M. L., Sauter, B. & Bhardwaj, N. Dendritic cells acquire antigen from apoptotic cells and induce class I-restricted CTLs. Nature 392, 86-89, (1998). 138 McElwee, K. J. et al. Alopecia areata in C3H/HeJ mice involves leukocyte-mediated root sheath disruption in advance of overt hair loss. Veterinary pathology 40, 643-650, (2003). 180  139 Gilhar, A. et al. Melanocyte-associated T cell epitopes can function as autoantigens for transfer of alopecia areata to human scalp explants on Prkdc(scid) mice. The Journal of investigative dermatology 117, 1357-1362, (2001). 140 Barahmani, N., Schabath, M. B., Duvic, M. & National Alopecia Areata, R. History of atopy or autoimmunity increases risk of alopecia areata. Journal of the American Academy of Dermatology 61, 581-591, (2009). 141 Somers, E. C., Thomas, S. L., Smeeth, L. & Hall, A. J. Are individuals with an autoimmune disease at higher risk of a second autoimmune disorder? American journal of epidemiology 169, 749-755, (2009). 142 Wang, E. et al. Development of autoimmune hair loss disease alopecia areata is associated with cardiac dysfunction in C3H/HeJ mice. PloS one 8, e62935, (2013). 143 McElwee, K. J., Silva, K., Beamer, W. G., King, L. E., Jr. & Sundberg, J. P. Melanocyte and gonad activity as potential severity modifying factors in C3H/HeJ mouse alopecia areata. Exp Dermatol 10, 420-429, (2001). 144 McElwee, K. J. et al. Dietary soy oil content and soy-derived phytoestrogen genistein increase resistance to alopecia areata onset in C3H/HeJ mice. Exp Dermatol 12, 30-36, (2003). 145 Rodriguez, T. A., Duvic, M. & National Alopecia Areata, R. Onset of alopecia areata after Epstein-Barr virus infectious mononucleosis. Journal of the American Academy of Dermatology 59, 137-139, (2008). 146 Wise, R. P., Kiminyo, K. P. & Salive, M. E. Hair loss after routine immunizations. JAMA : the journal of the American Medical Association 278, 1176-1178, (1997). 147 Sundberg, J. P. et al. Recombinant human hepatitis B vaccine initiating alopecia areata: testing the hypothesis using the C3H/HeJ mouse model. Veterinary dermatology 20, 99-104, (2009). 148 McDonagh, A. J. & Tazi-Ahnini, R. Epidemiology and genetics of alopecia areata. Clinical and experimental dermatology 27, 405-409, (2002). 149 Rodriguez, T. A., Fernandes, K. E., Dresser, K. L., Duvic, M. & National Alopecia Areata, R. Concordance rate of alopecia areata in identical twins supports both genetic and environmental factors. Journal of the American Academy of Dermatology 62, 525-527, (2010). 150 Sundberg, J. P., Silva, K. A., Li, R., Cox, G. A. & King, L. E. Adult-onset Alopecia areata is a complex polygenic trait in the C3H/HeJ mouse model. The Journal of investigative dermatology 123, 294-297, (2004). 151 Jagielska, D. et al. Follow-up study of the first genome-wide association scan in alopecia areata: IL13 and KIAA0350 as susceptibility loci supported with genome-wide significance. The Journal of investigative dermatology 132, 2192-2197, (2012). 152 Gupta, M. A., Gupta, A. K. & Watteel, G. N. Stress and alopecia areata: a psychodermatologic study. Acta dermato-venereologica 77, 296-298, (1997). 153 Perini, G. I. et al. Life events and alopecia areata. Psychotherapy and psychosomatics 41, 48-52, (1984). 154 Garcia-Hernandez, M. J., Ruiz-Doblado, S., Rodriguez-Pichardo, A. & Camacho, F. Alopecia areata, stress and psychiatric disorders: a review. J Dermatol 26, 625-632, (1999). 181  155 Chrousos, G. P. The hypothalamic-pituitary-adrenal axis and immune-mediated inflammation. The New England journal of medicine 332, 1351-1362, (1995). 156 Brajac, I., Tkalcic, M., Dragojevic, D. M. & Gruber, F. Roles of stress, stress perception and trait-anxiety in the onset and course of alopecia areata. J Dermatol 30, 871-878, (2003). 157 Picardi, A. et al. Psychosomatic factors in first-onset alopecia areata. Psychosomatics 44, 374-381, (2003). 158 van der Steen, P., Boezeman, J., Duller, P. & Happle, R. Can alopecia areata be triggered by emotional stress? An uncontrolled evaluation of 178 patients with extensive hair loss. Acta dermato-venereologica 72, 279-280, (1992). 159 Hall, J. M. et al. Psychological Stress and the Cutaneous Immune Response: Roles of the HPA Axis and the Sympathetic Nervous System in Atopic Dermatitis and Psoriasis. Dermatology research and practice 2012, 403908, (2012). 160 Lin, T. K. et al. Paradoxical Benefits of Psychological Stress in Inflammatory Dermatoses Models Are Glucocorticoid Mediated. The Journal of investigative dermatology, (2014). 161 Chu, S. Y. et al. Psychiatric comorbidities in patients with alopecia areata in Taiwan: a case-control study. The British journal of dermatology 166, 525-531, (2012). 162 Willemsen, R., Vanderlinden, J., Roseeuw, D. & Haentjens, P. Increased history of childhood and lifetime traumatic events among adults with alopecia areata. Journal of the American Academy of Dermatology 60, 388-393, (2009). 163 Paus, R. & Arck, P. Neuroendocrine perspectives in alopecia areata: does stress play a role? The Journal of investigative dermatology 129, 1324-1326, (2009). 164 Matzer, F., Egger, J. W. & Kopera, D. Psychosocial stress and coping in alopecia areata: a questionnaire survey and qualitative study among 45 patients. Acta dermato-venereologica 91, 318-327, (2011). 165 Willemsen, R., Haentjens, P., Roseeuw, D. & Vanderlinden, J. Hypnosis and alopecia areata: Long-term beneficial effects on psychological well-being. Acta dermato-venereologica 91, 35-39, (2011). 166 Willemsen, R. & Vanderlinden, J. Hypnotic approaches for alopecia areata. The International journal of clinical and experimental hypnosis 56, 318-333, (2008). 167 Katsarou-Katsari, A., Singh, L. K. & Theoharides, T. C. Alopecia areata and affected skin CRH receptor upregulation induced by acute emotional stress. Dermatology 203, 157-161, (2001). 168 Isgor, C., Cecchi, M., Kabbaj, M., Akil, H. & Watson, S. J. Estrogen receptor beta in the paraventricular nucleus of hypothalamus regulates the neuroendocrine response to stress and is regulated by corticosterone. Neuroscience 121, 837-845, (2003). 169 Ito, N. et al. Corticotropin-releasing hormone stimulates the in situ generation of mast cells from precursors in the human hair follicle mesenchyme. The Journal of investigative dermatology 130, 995-1004, (2010). 170 Ito, T. Hair follicle is a target of stress hormone and autoimmune reactions. Journal of dermatological science 60, 67-73, (2010). 171 Toyoda, M., Makino, T., Kagoura, M. & Morohashi, M. Expression of neuropeptide-degrading enzymes in alopecia areata: an immunohistochemical study. The British journal of dermatology 144, 46-54, (2001). 182  172 Siebenhaar, F. et al. Substance P as an immunomodulatory neuropeptide in a mouse model for autoimmune hair loss (alopecia areata). The Journal of investigative dermatology 127, 1489-1497, (2007). 173 Cetin, E. D., Savk, E., Uslu, M., Eskin, M. & Karul, A. Investigation of the inflammatory mechanisms in alopecia areata. The American Journal of dermatopathology 31, 53-60, (2009). 174 Evers, A. W. et al. How stress gets under the skin: cortisol and stress reactivity in psoriasis. The British journal of dermatology 163, 986-991, (2010). 175 Vena, G. A., Vestita, M. & Cassano, N. Psoriasis and cardiovascular disease. Dermatologic therapy 23, 144-151, (2010). 176 Eliakim-Raz, N., Shuvy, M., Lotan, C. & Planer, D. Psoriasis and dilated cardiomyopathy: coincidence or associated diseases? Cardiology 111, 202-206, (2008). 177 Trost, L. B., Bergfeld, W. F. & Calogeras, E. The diagnosis and treatment of iron deficiency and its potential relationship to hair loss. Journal of the American Academy of Dermatology 54, 824-844, (2006). 178 Boffa, M. J., Wood, P. & Griffiths, C. E. Iron status of patients with alopecia areata. The British journal of dermatology 132, 662-664, (1995). 179 White, M. I., Currie, J. & Williams, M. P. A study of the tissue iron status of patients with alopecia areata. The British journal of dermatology 130, 261-263, (1994). 180 Gonul, M., Cakmak, S. K., Soylu, S., Kilic, A. & Gul, U. Serum vitamin B12, folate, ferritin, and iron levels in Turkish patients with alopecia areata. Indian journal of dermatology, venereology and leprology 75, 552, (2009). 181 Arnold, H. L., Jr. Alopecia areata; prevalence in Japanese and prognosis after reassurance. A.M.A. archives of dermatology and syphilology 66, 191-196, (1952). 182 Wei, H., Bowen, R., Cai, Q., Barnes, S. & Wang, Y. Antioxidant and antipromotional effects of the soybean isoflavone genistein. Proceedings of the Society for Experimental Biology and Medicine. Society for Experimental Biology and Medicine 208, 124-130, (1995). 183 McElwee, K. J. et al. Murine cytomegalovirus is not associated with alopecia areata in C3H/HeJ mice. The Journal of investigative dermatology 110, 986-987, (1998). 184 Offidani, A., Amerio, P., Bernardini, M. L., Feliciani, C. & Bossi, G. Role of cytomegalovirus replication in alopecia areata pathogenesis. Journal of cutaneous medicine and surgery 4, 63-65, (2000). 185 Delamere, F. M., Sladden, M. M., Dobbins, H. M. & Leonardi-Bee, J. Interventions for alopecia areata. The Cochrane database of systematic reviews, CD004413, (2008). 186 Dombrowski, N. C. & Bergfeld, W. F. Alopecia areata: what to expect from current treatments. Cleveland Clinic journal of medicine 72, 758, 760-751, 765-756 passim, (2005). 187 Shapiro, J. Clinical practice. Hair loss in women. The New England journal of medicine 357, 1620-1630, (2007). 188 Michie, H. J., Jahoda, C. A., Oliver, R. F. & Johnson, B. E. The DEBR rat: an animal model of human alopecia areata. The British journal of dermatology 125, 94-100, (1991). 183  189 Sun, J., Silva, K. A., McElwee, K. J., King, L. E., Jr. & Sundberg, J. P. The C3H/HeJ mouse and DEBR rat models for alopecia areata: review of preclinical drug screening approaches and results. Exp Dermatol 17, 793-805, (2008). 190 Gilhar, A. et al. Autoimmune disease induction in a healthy human organ: a humanized mouse model of alopecia areata. The Journal of investigative dermatology 133, 844-847, (2013). 191 Sundberg, J. P., Cordy, W. R. & King, L. E., Jr. Alopecia areata in aging C3H/HeJ mice. The Journal of investigative dermatology 102, 847-856, (1994). 192 McElwee, K. J., Freyschmidt-Paul, P., Zoller, M. & Hoffmann, R. Alopecia areata susceptibility in rodent models. The journal of investigative dermatology. Symposium proceedings / the Society for Investigative Dermatology, Inc. [and] European Society for Dermatological Research 8, 182-187, (2003). 193 McElwee, K. J., Freyschmidt-Paul, P., Sundberg, J. P. & Hoffmann, R. The pathogenesis of alopecia areata in rodent models. The journal of investigative dermatology. Symposium proceedings / the Society for Investigative Dermatology, Inc. [and] European Society for Dermatological Research 8, 6-11, (2003). 194 Sundberg, J. P. et al. Alopecia areata: updates from the mouse perspective. The journal of investigative dermatology. Symposium proceedings / the Society for Investigative Dermatology, Inc. [and] European Society for Dermatological Research 16, S23-24, (2013). 195 Tobin, D. J. et al. A natural canine homologue of alopecia areata in humans. The British journal of dermatology 149, 938-950, (2003). 196 Hoolahan, D. E., White, S. D., Outerbridge, C. A., Shearer, P. L. & Affolter, V. K. Equine alopecia areata: a retrospective clinical descriptive study at the University of California, Davis (1980-2011). Veterinary dermatology 24, 282-e264, (2013). 197 Tobin, D. J., Alhaidari, Z. & Olivry, T. Equine alopecia areata autoantibodies target multiple hair follicle antigens and may alter hair growth. A preliminary study. Exp Dermatol 7, 289-297, (1998). 198 De Villez, R. L. & Buchanan, J. M. The graying phenomenon: an unusual manifestation of alopecia areata. International journal of dermatology 21, 344-346, (1982). 199 Guin, J. D., Kumar, V. & Petersen, B. H. Immunofluorescence findings in rapid whitening of scalp hair. Arch Dermatol 117, 576-578, (1981). 200 McBride, A. K. & Bergfeld, W. F. Mosaic hair color changes in alopecia areata. Cleveland Clinic journal of medicine 57, 354-356, (1990). 201 Oliver, R. F. & Lowe, J. G. Oral cyclosporin A restores hair growth in the DEBR rat model for alopecia areata. Clinical and experimental dermatology 20, 127-131, (1995). 202 Gilhar, A., Keren, A. & Paus, R. A new humanized mouse model for alopecia areata. The journal of investigative dermatology. Symposium proceedings / the Society for Investigative Dermatology, Inc. [and] European Society for Dermatological Research 16, S37-38, (2013). 203 McElwee, K. J. et al. What causes alopecia areata? Exp Dermatol 22, 609-626, (2013). 184  204 Tang, L., Cao, L., Sundberg, J. P., Lui, H. & Shapiro, J. Restoration of hair growth in mice with an alopecia areata-like disease using topical anthralin. Exp Dermatol 13, 5-10, (2004). 205 Harnaha, J. et al. Interleukin-7 is a survival factor for CD4+ CD25+ T-cells and is expressed by diabetes-suppressive dendritic cells. Diabetes 55, 158-170, (2006). 206 Nanjappa, S. G., Walent, J. H., Morre, M. & Suresh, M. Effects of IL-7 on memory CD8 T cell homeostasis are influenced by the timing of therapy in mice. J Clin Invest 118, 1027-1039, (2008). 207 Weng, N. P., Liu, K., Catalfamo, M., Li, Y. & Henkart, P. A. IL-15 is a growth factor and an activator of CD8 memory T cells. Ann N Y Acad Sci 975, 46-56, (2002). 208 Picker, L. J. et al. IL-15 induces CD4 effector memory T cell production and tissue emigration in nonhuman primates. J Clin Invest 116, 1514-1524, (2006). 209 Bere, A., Denny, L., Hanekom, W., Burgers, W. A. & Passmore, J. A. Comparison of polyclonal expansion methods to improve the recovery of cervical cytobrush-derived T cells from the female genital tract of HIV-infected women. Journal of immunological methods 354, 68-79, (2010). 210 Sallusto, F., Lenig, D., Forster, R., Lipp, M. & Lanzavecchia, A. Two subsets of memory T lymphocytes with distinct homing potentials and effector functions. Nature 401, 708-712, (1999). 211 Forster, R., Davalos-Misslitz, A. C. & Rot, A. CCR7 and its ligands: balancing immunity and tolerance. Nature reviews. Immunology 8, 362-371, (2008). 212 Thompson, C. B. et al. CD28 activation pathway regulates the production of multiple T-cell-derived lymphokines/cytokines. Proceedings of the National Academy of Sciences of the United States of America 86, 1333-1337, (1989). 213 Tartar, D. M. et al. FoxP3+RORgammat+ T helper intermediates display suppressive function against autoimmune diabetes. Journal of immunology 184, 3377-3385, (2010). 214 Veerapathran, A., Pidala, J., Beato, F., Yu, X. Z. & Anasetti, C. Ex vivo expansion of human Tregs specific for alloantigens presented directly or indirectly. Blood 118, 5671-5680, (2011). 215 Nagahama, K., Nishimura, E. & Sakaguchi, S. Induction of tolerance by adoptive transfer of Treg cells. Methods in molecular biology 380, 431-442, (2007). 216 Progatzky, F., Dallman, M. J. & Lo Celso, C. From seeing to believing: labelling strategies for in vivo cell-tracking experiments. Interface focus 3, 20130001, (2013). 217 Harris, J. E. et al. A mouse model of vitiligo with focused epidermal depigmentation requires IFN-gamma for autoreactive CD8(+) T-cell accumulation in the skin. The Journal of investigative dermatology 132, 1869-1876, (2012). 218 Asquith, D. L., Miller, A. M., McInnes, I. B. & Liew, F. Y. Animal models of rheumatoid arthritis. European journal of immunology 39, 2040-2044, (2009). 219 Ransohoff, R. M. Animal models of multiple sclerosis: the good, the bad and the bottom line. Nature neuroscience 15, 1074-1077, (2012). 220 Constantinescu, C. S., Farooqi, N., O'Brien, K. & Gran, B. Experimental autoimmune encephalomyelitis (EAE) as a model for multiple sclerosis (MS). British journal of pharmacology 164, 1079-1106, (2011). 185  221 Zhang, B. et al. Early stage alopecia areata is associated with inflammation in the upper dermis and damage to the hair follicle infundibulum. The Australasian journal of dermatology 54, 184-191, (2013). 222 Erb, U., Freyschmidt-Paul, P. & Zoller, M. Tolerance induction by hair-specific keratins in murine alopecia areata. Journal of leukocyte biology 94, 845-857, (2013). 223 Messenger, A. G. & Bleehen, S. S. Expression of HLA-DR by anagen hair follicles in alopecia areata. The Journal of investigative dermatology 85, 569-572, (1985). 224 Finner, A. M. Alopecia areata: Clinical presentation, diagnosis, and unusual cases. Dermatologic therapy 24, 348-354, (2011). 225 Ellis, J. M. et al. Frequencies of HLA-A2 alleles in five U.S. population groups. Predominance Of A*02011 and identification of HLA-A*0231. Human immunology 61, 334-340, (2000). 226 Rammensee, H., Bachmann, J., Emmerich, N. P., Bachor, O. A. & Stevanovic, S. SYFPEITHI: database for MHC ligands and peptide motifs. Immunogenetics 50, 213-219, (1999). 227 Parker, K. C., Bednarek, M. A. & Coligan, J. E. Scheme for ranking potential HLA-A2 binding peptides based on independent binding of individual peptide side-chains. Journal of immunology 152, 163-175, (1994). 228 Valmori, D. et al. Enhanced generation of specific tumor-reactive CTL in vitro by selected Melan-A/MART-1 immunodominant peptide analogues. Journal of immunology 160, 1750-1758, (1998). 229 Wolfel, T. et al. Two tyrosinase nonapeptides recognized on HLA-A2 melanomas by autologous cytolytic T lymphocytes. European journal of immunology 24, 759-764, (1994). 230 Kawakami, Y. et al. T cell immune responses against melanoma and melanocytes in cancer and autoimmunity. Pigment cell research / sponsored by the European Society for Pigment Cell Research and the International Pigment Cell Society 13 Suppl 8, 163-169, (2000). 231 Parkhurst, M. R. et al. Improved induction of melanoma-reactive CTL with peptides from the melanoma antigen gp100 modified at HLA-A*0201-binding residues. Journal of immunology 157, 2539-2548, (1996). 232 Cellerai, C. et al. Proliferation capacity and cytotoxic activity are mediated by functionally and phenotypically distinct virus-specific CD8 T cells defined by interleukin-7R{alpha} (CD127) and perforin expression. J Virol 84, 3868-3878, (2010). 233 Ferrari, G. et al. Absence of immunodominant anti-Gag p17 (SL9) responses among Gag CTL-positive, HIV-uninfected vaccine recipients expressing the HLA-A*0201 allele. Journal of immunology 173, 2126-2133, (2004). 234 Spangenberg, H. C. et al. Intrahepatic CD8+ T-cell failure during chronic hepatitis C virus infection. Hepatology 42, 828-837, (2005). 235 Kamath, A. B. et al. Cytolytic CD8+ T cells recognizing CFP10 are recruited to the lung after Mycobacterium tuberculosis infection. The Journal of experimental medicine 200, 1479-1489, (2004). 236 Fassnacht, U., Ackermann, A., Staeheli, P. & Hausmann, J. Immunization with dendritic cells can break immunological ignorance toward a persisting virus in the 186  central nervous system and induce partial protection against intracerebral viral challenge. The Journal of general virology 85, 2379-2387, (2004). 237 Malik, A. et al. Identification of a nonameric H-2Kk-restricted CD8+ cytotoxic T lymphocyte epitope on the Plasmodium falciparum circumsporozoite protein. Infection and immunity 63, 1955-1959, (1995). 238 Leggatt, G. R. et al. The importance of pairwise interactions between peptide residues in the delineation of TCR specificity. Journal of immunology 161, 4728-4735, (1998). 239 Kellenberger, C., Roussel, A. & Malissen, B. The H-2Kk MHC peptide-binding groove anchors the backbone of an octameric antigenic peptide in an unprecedented mode. Journal of immunology 175, 3819-3825, (2005). 240 Majlessi, L., Rojas, M. J., Brodin, P. & Leclerc, C. CD8+-T-cell responses of Mycobacterium-infected mice to a newly identified major histocompatibility complex class I-restricted epitope shared by proteins of the ESAT-6 family. Infection and immunity 71, 7173-7177, (2003). 241 Pechhold, K. et al. Low dose streptozotocin-induced diabetes in rat insulin promoter-mCD80-transgenic mice is T cell autoantigen-specific and CD28 dependent. Journal of immunology 166, 2531-2539, (2001). 242 O'Keefe, E. J., Hamilton, E. H., Lee, S. C. & Steinert, P. Trichohyalin: a structural protein of hair, tongue, nail, and epidermis. The Journal of investigative dermatology 101, 65S-71S, (1993). 243 Bernot, K. M., Coulombe, P. A. & McGowan, K. M. Keratin 16 expression defines a subset of epithelial cells during skin morphogenesis and the hair cycle. The Journal of investigative dermatology 119, 1137-1149, (2002). 244 Martinez, N. R. et al. Disabling an integral CTL epitope allows suppression of autoimmune diabetes by intranasal proinsulin peptide. J Clin Invest 111, 1365-1371, (2003). 245 Johnston, A., Gudjonsson, J. E., Sigmundsdottir, H., Love, T. J. & Valdimarsson, H. Peripheral blood T cell responses to keratin peptides that share sequences with streptococcal M proteins are largely restricted to skin-homing CD8(+) T cells. Clinical and experimental immunology 138, 83-93, (2004). 246 Bredenbeck, A. et al. Identification of noncanonical melanoma-associated T cell epitopes for cancer immunotherapy. Journal of immunology 174, 6716-6724, (2005). 247 Hansson, L., Rabbani, H., Fagerberg, J., Osterborg, A. & Mellstedt, H. T-cell epitopes within the complementarity-determining and framework regions of the tumor-derived immunoglobulin heavy chain in multiple myeloma. Blood 101, 4930-4936, (2003). 248 Liu, G. Y. et al. Low avidity recognition of self-antigen by T cells permits escape from central tolerance. Immunity 3, 407-415, (1995). 249 Muraro, P. A. et al. Immunodominance of a low-affinity major histocompatibility complex-binding myelin basic protein epitope (residues 111-129) in HLA-DR4 (B1*0401) subjects is associated with a restricted T cell receptor repertoire. J Clin Invest 100, 339-349, (1997). 250 Chang, K. Y. & Unanue, E. R. Prediction of HLA-DQ8beta cell peptidome using a computational program and its relationship to autoreactive T cells. International immunology 21, 705-713, (2009). 187  251 Gilhar, A., Kam, Y., Assy, B. & Kalish, R. S. Alopecia areata induced in C3H/HeJ mice by interferon-gamma: evidence for loss of immune privilege. The Journal of investigative dermatology 124, 288-289, (2005). 252 Sundberg, J. P. S., Kathleen A; Edwards, Kathryn; Black, Steven; Bennett Jenson, A; King, Lloyd E. Failure to induce alopecia areata in C3H/HeJ mice with exogenous interferon gamma. Journal of Experimental Animal Science 43, (2007). 253 Wong, P. & Pamer, E. G. CD8 T cell responses to infectious pathogens. Annual review of immunology 21, 29-70, (2003). 254 Lee, S. C. et al. The structure of human trichohyalin. Potential multiple roles as a functional EF-hand-like calcium-binding protein, a cornified cell envelope precursor, and an intermediate filament-associated (cross-linking) protein. The Journal of biological chemistry 268, 12164-12176, (1993). 255 Panagiotopoulos, C., Qin, H., Tan, R. & Verchere, C. B. Identification of a beta-cell-specific HLA class I restricted epitope in type 1 diabetes. Diabetes 52, 2647-2651, (2003). 256 Standifer, N. E. et al. Identification of Novel HLA-A*0201-restricted epitopes in recent-onset type 1 diabetic subjects and antibody-positive relatives. Diabetes 55, 3061-3067, (2006). 257 You, S. & Chatenoud, L. Proinsulin: a unique autoantigen triggering autoimmune diabetes. J Clin Invest 116, 3108-3110, (2006). 258 von Herrath, M., Sanda, S. & Herold, K. Type 1 diabetes as a relapsing-remitting disease? Nature reviews. Immunology 7, 988-994, (2007). 259 Elmore, S. Apoptosis: a review of programmed cell death. Toxicologic pathology 35, 495-516, (2007). 260 Tobin, D. J., Fenton, D. A. & Kendall, M. D. Ultrastructural observations on the hair bulb melanocytes and melanosomes in acute alopecia areata. The Journal of investigative dermatology 94, 803-807, (1990). 261 Shellow, W. V., Edwards, J. E. & Koo, J. Y. Profile of alopecia areata: a questionnaire analysis of patient and family. International journal of dermatology 31, 186-189, (1992). 262 Tobin, D. J. & Bystryn, J. C. Different populations of melanocytes are present in hair follicles and epidermis. Pigment cell research / sponsored by the European Society for Pigment Cell Research and the International Pigment Cell Society 9, 304-310, (1996). 263 Lo, B. K. et al. CXCR3 ligands promote expression of functional indoleamine 2,3-dioxygenase in basal cell carcinoma keratinocytes. The British journal of dermatology 165, 1030-1036, (2011). 264 Zhang, L., Nakayama, M. & Eisenbarth, G. S. Insulin as an autoantigen in NOD/human diabetes. Current opinion in immunology 20, 111-118, (2008). 265 Babad, J., Geliebter, A. & DiLorenzo, T. P. T-cell autoantigens in the non-obese diabetic mouse model of autoimmune diabetes. Immunology 131, 459-465, (2010). 266 Chaparro, R. J. & Dilorenzo, T. P. An update on the use of NOD mice to study autoimmune (Type 1) diabetes. Expert review of clinical immunology 6, 939-955, (2010). 188  267 Kasumagic-Halilovic, E. Thyroid autoimmunity in patients with alopecia areata. Acta dermatovenerologica Croatica : ADC 16, 123-125, (2008). 268 Tsai, T. F. et al. Epidemiology and comorbidities of psoriasis patients in a national database in Taiwan. Journal of dermatological science 63, 40-46, (2011). 269 Ro, B. I. Alopecia areata in Korea (1982-1994). J Dermatol 22, 858-864, (1995). 270 Kurd, S. K., Troxel, A. B., Crits-Christoph, P. & Gelfand, J. M. The risk of depression, anxiety, and suicidality in patients with psoriasis: a population-based cohort study. Arch Dermatol 146, 891-895, (2010). 271 Palijan, T. Z., Kovacevic, D., Koic, E., Ruzic, K. & Dervinja, F. The impact of psoriasis on the quality of life and psychological characteristics of persons suffering from psoriasis. Coll Antropol 35 Suppl 2, 81-85, (2011). 272 Weigl, B. A. The significance of stress hormones (glucocorticoids, catecholamines) for eruptions and spontaneous remission phases in psoriasis. International journal of dermatology 39, 678-688, (2000). 273 Kim, J. E. et al. Expression of the corticotropin-releasing hormone-proopiomelanocortin axis in the various clinical types of psoriasis. Exp Dermatol 16, 104-109, (2007). 274 Ghazizadeh, R., Shimizu, H., Tosa, M. & Ghazizadeh, M. Pathogenic mechanisms shared between psoriasis and cardiovascular disease. International journal of medical sciences 7, 284-289, (2010). 275 Kremers, H. M., McEvoy, M. T., Dann, F. J. & Gabriel, S. E. Heart disease in psoriasis. Journal of the American Academy of Dermatology 57, 347-354, (2007). 276 Towbin, J. A. & Bowles, N. E. The failing heart. Nature 415, 227-233, (2002). 277 Martino, T. A., Liu, P. & Sole, M. J. Viral infection and the pathogenesis of dilated cardiomyopathy. Circ Res 74, 182-188, (1994). 278 Wexler, R. K., Elton, T., Pleister, A. & Feldman, D. Cardiomyopathy: an overview. Am Fam Physician 79, 778-784, (2009). 279 Towbin, J. A. Hypertrophic cardiomyopathy. Pacing Clin Electrophysiol 32 Suppl 2, S23-31, (2009). 280 Fu, M. Autoimmunity and idiopathic dilated cardiomyopathy: where we stand? Autoimmunity 41, 415-418, (2008). 281 Jahns, R., Boivin, V., Schwarzbach, V., Ertl, G. & Lohse, M. J. Pathological autoantibodies in cardiomyopathy. Autoimmunity 41, 454-461, (2008). 282 San Martin, M. A., Garcia, A., Rodriguez, F. J. & Terol, I. [Dilated cardiomyopathy and autoimmunity: an overview of current knowledge and perspectives]. Rev Esp Cardiol 55, 514-524, (2002). 283 Bernstein, D., Fajardo, G. & Zhao, M. THE ROLE OF beta-ADRENERGIC RECEPTORS IN HEART FAILURE: DIFFERENTIAL REGULATION OF CARDIOTOXICITY AND CARDIOPROTECTION. Progress in pediatric cardiology 31, 35-38, (2011). 284 Fajardo, G., Zhao, M., Powers, J. & Bernstein, D. Differential cardiotoxic/cardioprotective effects of beta-adrenergic receptor subtypes in myocytes and fibroblasts in doxorubicin cardiomyopathy. Journal of molecular and cellular cardiology 40, 375-383, (2006). 189  285 Zheng, M., Han, Q. D. & Xiao, R. P. Distinct beta-adrenergic receptor subtype signaling in the heart and their pathophysiological relevance. Sheng li xue bao : [Acta physiologica Sinica] 56, 1-15, (2004). 286 Walsh, D. Circulatory Disorders In Relation To Baldness. The British Medical Journal 2, 1007-1011, (1913). 287 Yu, M. et al. Superficial, nodular, and morpheiform basal-cell carcinomas exhibit distinct gene expression profiles. The Journal of investigative dermatology 128, 1797-1805, (2008). 288 Livak, K. J. & Schmittgen, T. D. Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)) Method. Methods 25, 402-408, (2001). 289 Leite, S. N., Jordao Junior, A. A., Andrade, T. A., Masson Ddos, S. & Frade, M. A. Experimental models of malnutrition and its effect on skin trophism. An Bras Dermatol 86, 681-688, (2011). 290 Ito, N. et al. Human hair follicles display a functional equivalent of the hypothalamic-pituitary-adrenal axis and synthesize cortisol. The FASEB journal : official publication of the Federation of American Societies for Experimental Biology 19, 1332-1334, (2005). 291 O'Shaughnessy, P. J. et al. Adrenocorticotropic hormone directly stimulates testosterone production by the fetal and neonatal mouse testis. Endocrinology 144, 3279-3284, (2003). 292 Wen, X. L., Villee, D. B., Ellison, P., Todd, R. & Loring, J. Effects of adrenocorticotropic hormone, human chorionic gonadotropin, and insulin on steroid production by human adrenocortical carcinoma cells in culture. Cancer research 45, 3974-3978, (1985). 293 Brade, T. et al. Retinoic acid stimulates myocardial expansion by induction of hepatic erythropoietin which activates epicardial Igf2. Development 138, 139-148, (2011). 294 Chang, A. N., Parvatiyar, M. S. & Potter, J. D. Troponin and cardiomyopathy. Biochem Biophys Res Commun 369, 74-81, (2008). 295 Nellessen, U. et al. Serial analysis of troponin I levels in patients with ischemic and nonischemic dilated cardiomyopathy. Clin Cardiol 29, 219-224, (2006). 296 Miller, W. L. et al. Serial biomarker measurements in ambulatory patients with chronic heart failure: the importance of change over time. Circulation 116, 249-257, (2007). 297 Lim, W., Cook, D. J., Griffith, L. E., Crowther, M. A. & Devereaux, P. J. Elevated cardiac troponin levels in critically ill patients: prevalence, incidence, and outcomes. Am J Crit Care 15, 280-288; quiz 289, (2006). 298 Arias-Santiago, S., Gutierrez-Salmeron, M. T., Buendia-Eisman, A., Giron-Prieto, M. S. & Naranjo-Sintes, R. Hypertension and aldosterone levels in women with early-onset androgenetic alopecia. The British journal of dermatology 162, 786-789, (2010). 299 Gutersohn, T. & Scheidegger, E. P. Is baldness bad for the heart? Dermatology 211, 72-74, (2005). 300 Ahouansou, S., Le Toumelin, P., Crickx, B. & Descamps, V. Association of androgenetic alopecia and hypertension. Eur J Dermatol 17, 220-222, (2007). 190  301 Arias-Santiago, S., Gutierrez-Salmeron, M. T., Castellote-Caballero, L., Buendia-Eisman, A. & Naranjo-Sintes, R. Androgenetic alopecia and cardiovascular risk factors in men and women: a comparative study. Journal of the American Academy of Dermatology 63, 420-429, (2010). 302 Jefferies, J. L. & Towbin, J. A. Dilated cardiomyopathy. Lancet 375, 752-762, (2010). 303 Potier, M., Karl, M., Elliot, S. J., Striker, G. E. & Striker, L. J. Response to sex hormones differs in atherosclerosis-susceptible and -resistant mice. Am J Physiol Endocrinol Metab 285, E1237-1245, (2003). 304 Porrello, E. R., Widdop, R. E. & Delbridge, L. M. Early origins of cardiac hypertrophy: does cardiomyocyte attrition programme for pathological 'catch-up' growth of the heart? Clinical and experimental pharmacology & physiology 35, 1358-1364, (2008). 305 Van Kerckhoven, R., Kalkman, E. A., Saxena, P. R. & Schoemaker, R. G. Altered cardiac collagen and associated changes in diastolic function of infarcted rat hearts. Cardiovasc Res 46, 316-323, (2000). 306 Biernacka, A. & Frangogiannis, N. G. Aging and Cardiac Fibrosis. Aging Dis 2, 158-173, (2011). 307 Park, J. B. & Schiffrin, E. L. Cardiac and vascular fibrosis and hypertrophy in aldosterone-infused rats: role of endothelin-1. Am J Hypertens 15, 164-169, (2002). 308 Rummery, N. M., McKenzie, K. U., Whitworth, J. A. & Hill, C. E. Decreased endothelial size and connexin expression in rat caudal arteries during hypertension. J Hypertens 20, 247-253, (2002). 309 Souders, C. A., Bowers, S. L. & Baudino, T. A. Cardiac fibroblast: the renaissance cell. Circ Res 105, 1164-1176, (2009). 310 Madri, J. A., Pratt, B. M. & Yannariello-Brown, J. Matrix-driven cell size change modulates aortic endothelial cell proliferation and sheet migration. Am J Pathol 132, 18-27, (1988). 311 Mustonen, E. et al. Thrombospondin-4 expression is rapidly upregulated by cardiac overload. Biochem Biophys Res Commun 373, 186-191, (2008). 312 Ivandic, B. T. et al. A locus on chromosome 7 determines myocardial cell necrosis and calcification (dystrophic cardiac calcinosis) in mice. Proceedings of the National Academy of Sciences of the United States of America 93, 5483-5488, (1996). 313 Aherrahrou, Z. et al. A locus on chromosome 7 determines dramatic up-regulation of osteopontin in dystrophic cardiac calcification in mice. Am J Pathol 164, 1379-1387, (2004). 314 Meng, H. et al. Identification of Abcc6 as the major causal gene for dystrophic cardiac calcification in mice through integrative genomics. Proceedings of the National Academy of Sciences of the United States of America 104, 4530-4535, (2007). 315 Konishi, H. et al. IL-18 contributes to the spontaneous development of atopic dermatitis-like inflammatory skin lesion independently of IgE/stat6 under specific pathogen-free conditions. Proceedings of the National Academy of Sciences of the United States of America 99, 11340-11345, (2002). 191  316 Wittmann, M., Macdonald, A. & Renne, J. IL-18 and skin inflammation. Autoimmun Rev 9, 45-48, (2009). 317 Luan, Y. et al. Interleukin-18 among atrial fibrillation patients in the absence of structural heart disease. Europace 12, 1713-1718, (2010). 318 Jefferis, B. J. et al. Interleukin 18 and coronary heart disease: prospective study and systematic review. Atherosclerosis 217, 227-233. 319 Wang, M., Markel, T. A. & Meldrum, D. R. Interleukin 18 in the heart. Shock 30, 3-10, (2008). 320 Wang, S. F., Xu, J. C. & Liang, Q. X. [Levels of plasma interleukin-18 and interferon-gamma in patients with coronary heart disease]. Zhongguo Wei Zhong Bing Ji Jiu Yi Xue 18, 237-239, (2006). 321 Baran, W., Szepietowski, J. C., Mazur, G. & Baran, E. IFN-gamma promoter gene polymorphism in psoriasis vulgaris. Biomarkers 13, 52-58, (2008). 322 Dinarello, C. A. Interleukin-18 and the pathogenesis of inflammatory diseases. Semin Nephrol 27, 98-114, (2007). 323 Gerdes, N. et al. Expression of interleukin (IL)-18 and functional IL-18 receptor on human vascular endothelial cells, smooth muscle cells, and macrophages: implications for atherogenesis. The Journal of experimental medicine 195, 245-257, (2002). 324 Dinarello, C. A. Novel targets for interleukin 18 binding protein. Annals of the rheumatic diseases 60 Suppl 3, iii18-24, (2001). 325 Chandrasekar, B., Mummidi, S., Claycomb, W. C., Mestril, R. & Nemer, M. Interleukin-18 is a pro-hypertrophic cytokine that acts through a phosphatidylinositol 3-kinase-phosphoinositide-dependent kinase-1-Akt-GATA4 signaling pathway in cardiomyocytes. The Journal of biological chemistry 280, 4553-4567, (2005). 326 Woldbaek, P. R. et al. Daily administration of interleukin-18 causes myocardial dysfunction in healthy mice. Am J Physiol Heart Circ Physiol 289, H708-714, (2005). 327 Mallat, Z. et al. Evidence for altered interleukin 18 (IL)-18 pathway in human heart failure. The FASEB journal : official publication of the Federation of American Societies for Experimental Biology 18, 1752-1754, (2004). 328 Dinarello, C. A. IL-18: A TH1-inducing, proinflammatory cytokine and new member of the IL-1 family. J Allergy Clin Immunol 103, 11-24, (1999). 329 Xiang, Y. & Moss, B. IL-18 binding and inhibition of interferon gamma induction by human poxvirus-encoded proteins. Proceedings of the National Academy of Sciences of the United States of America 96, 11537-11542, (1999). 330 Jefferis, B. J. et al. Interleukin 18 and coronary heart disease: prospective study and systematic review. Atherosclerosis 217, 227-233, (2011). 331 Naito, Y. et al. Increased circulating interleukin-18 in patients with congestive heart failure. Heart 88, 296-297, (2002). 332 Holzinger, C. et al. Phenotypic patterns of mononuclear cells in dilated cardiomyopathy. Circulation 92, 2876-2885, (1995). 333 Liao, C. H. et al. Cardiac mast cells cause atrial fibrillation through PDGF-A-mediated fibrosis in pressure-overloaded mouse hearts. J Clin Invest 120, 242-253, (2010). 192  334 Platis, A. et al. The effect of daily administration of IL-18 on cardiac structure and function. Perfusion 23, 237-242, (2008). 335 Fix, C., Bingham, K. & Carver, W. Effects of interleukin-18 on cardiac fibroblast function and gene expression. Cytokine 53, 19-28, (2011). 336 Johansen, C., Moeller, K., Kragballe, K. & Iversen, L. The activity of caspase-1 is increased in lesional psoriatic epidermis. The Journal of investigative dermatology 127, 2857-2864, (2007). 337 Dinarello, C. A. Blocking interleukin-1beta in acute and chronic autoinflammatory diseases. J Intern Med 269, 16-28, (2011). 338 Dos Santos, J. C. et al. Increased circulating pro-inflammatory cytokines and imbalanced regulatory T-cell cytokines production in chronic idiopathic urticaria. Int Immunopharmacol 8, 1433-1440, (2008). 339 Lee, D. et al. Serum levels of IL-18 and sIL-2R in patients with alopecia areata receiving combined therapy with oral cyclosporine and steroids. Exp Dermatol 19, 145-147, (2010). 340 Babuin, L. & Jaffe, A. S. Troponin: the biomarker of choice for the detection of cardiac injury. Cmaj 173, 1191-1202, (2005). 341 Adams, J. E., 3rd et al. Cardiac troponin I. A marker with high specificity for cardiac injury. Circulation 88, 101-106, (1993). 342 Ito, H. et al. Insulin-like growth factor-I induces hypertrophy with enhanced expression of muscle specific genes in cultured rat cardiomyocytes. Circulation 87, 1715-1721, (1993). 343 Guo, H. W. et al. An implication for post-transcriptional control: reciprocal changes of melanocortin receptor type 2 mRNA and protein expression in alopecia areata. Med Hypotheses 76, 122-124, (2011). 344 Guo, H. W. et al. Melanocortin receptor type 2 (MC2R, ACTH receptor) expression in patients with alopecia areata. Exp Dermatol 19, 1020-1022, (2011). 345 Sekiyama, A. et al. A stress-induced, superoxide-mediated caspase-1 activation pathway causes plasma IL-18 upregulation. Immunity 22, 669-677, (2005). 346 Park, H. J. et al. Adrenocorticotropin hormone stimulates interleukin-18 expression in human HaCaT keratinocytes. The Journal of investigative dermatology 127, 1210-1216, (2007). 347 O'Brien L, C. et al. Interleukin-18 as a therapeutic target in acute myocardial infarction and heart failure. Molecular medicine 20, 221-229, (2014). 348 Otis, M., Campbell, S., Payet, M. D. & Gallo-Payet, N. Expression of extracellular matrix proteins and integrins in rat adrenal gland: importance for ACTH-associated functions. J Endocrinol 193, 331-347, (2007). 349 Kahai, S., Vary, C. P., Gao, Y. & Seth, A. Collagen, type V, alpha1 (COL5A1) is regulated by TGF-beta in osteoblasts. Matrix Biol 23, 445-455, (2004). 350 Evans, J. F., Fernando, A. & Ragolia, L. Functional melanocortin-2 receptors are expressed by mouse aorta-derived mesenchymal progenitor cells. Mol Cell Endocrinol 355, 60-70, (2012). 351 Isales, C. M., Zaidi, M. & Blair, H. C. ACTH is a novel regulator of bone mass. Ann N Y Acad Sci 1192, 110-116, (2010). 193  352 Wenstrup, R. J. et al. Type V collagen controls the initiation of collagen fibril assembly. The Journal of biological chemistry 279, 53331-53337, (2004). 353 Caulfield, J. B. & Janicki, J. S. Structure and function of myocardial fibrillar collagen. Technol Health Care 5, 95-113, (1997). 354 Mizuno, Y. et al. Adrenocorticotropic hormone is produced in the ventricle of patients with essential hypertension. J Hypertens 23, 411-416, (2005). 355 Alikhan, A., Felsten, L. M., Daly, M. & Petronic-Rosic, V. Vitiligo: a comprehensive overview Part I. Introduction, epidemiology, quality of life, diagnosis, differential diagnosis, associations, histopathology, etiology, and work-up. Journal of the American Academy of Dermatology 65, 473-491, (2011). 356 Oiso, N., Suzuki, T., Fukai, K., Katayama, I. & Kawada, A. Nonsegmental vitiligo and autoimmune mechanism. Dermatology research and practice 2011, 518090, (2011). 357 Sharma, V. K., Dawn, G. & Kumar, B. Profile of alopecia areata in Northern India. International journal of dermatology 35, 22-27, (1996). 358 Arias-Santiago, S., Gutierrez-Salmeron, M. T., Buendia-Eisman, A., Giron-Prieto, M. S. & Naranjo-Sintes, R. Hypertension and aldosterone levels in women with early-onset androgenetic alopecia. The British journal of dermatology 162, 786-789. 359 Agac, M. T. et al. Androgenetic alopecia is associated with increased arterial stiffness in asymptomatic young adults. Journal of the European Academy of Dermatology and Venereology : JEADV, (2014). 360 Sharma, K. H. & Jindal, A. Association between androgenetic alopecia and coronary artery disease in young male patients. International journal of trichology 6, 5-7, (2014). 361 Khalid, U. et al. Psoriasis and risk of heart failure: a nationwide cohort study. European journal of heart failure 16, 743-748, (2014). 362 Khraishi, M., Aslanov, R., Rampakakis, E., Pollock, C. & Sampalis, J. S. Prevalence of cardiovascular risk factors in patients with psoriatic arthritis. Clinical rheumatology, (2014). 363 Choudhary, R. et al. Heart failure biomarkers. Journal of cardiovascular translational research 6, 471-484, (2013). 364 Liquori, M. E., Christenson, R. H., Collinson, P. O. & Defilippi, C. R. Cardiac biomarkers in heart failure. Clinical biochemistry 47, 327-337, (2014). 365 Agewall, S., Giannitsis, E., Jernberg, T. & Katus, H. Troponin elevation in coronary vs. non-coronary disease. European heart journal 32, 404-411, (2011). 366 Ridker, P. M. Clinical application of C-reactive protein for cardiovascular disease detection and prevention. Circulation 107, 363-369, (2003). 367 Mahamid, M., Abu-Elhija, O., Samamra, M., Mahamid, A. & Nseir, W. Association between vitamin D levels and alopecia areata. The Israel Medical Association journal : IMAJ 16, 367-370, (2014). 368 Mahe, Y. F. et al. Androgenetic alopecia and microinflammation. International journal of dermatology 39, 576-584, (2000). 369 Yamada, T., Hara, K., Umematsu, H. & Kadowaki, T. Male pattern baldness and its association with coronary heart disease: a meta-analysis. BMJ open 3, (2013). 194  370 Mansouri, P., Mortazavi, M., Eslami, M. & Mazinani, M. Androgenetic alopecia and coronary artery disease in women. Dermatology online journal 11, 2, (2005). 371 Schnohr, P., Lange, P., Nyboe, J., Appleyard, M. & Jensen, G. Gray hair, baldness, and wrinkles in relation to myocardial infarction: the Copenhagen City Heart Study. American heart journal 130, 1003-1010, (1995). 372 Ford, E. S., Freedman, D. S. & Byers, T. Baldness and ischemic heart disease in a national sample of men. American journal of epidemiology 143, 651-657, (1996). 373 Cotton, S. G., Nixon, J. M., Carpenter, R. G. & Evans, D. W. Factors discriminating men with coronary heart disease from healthy controls. British heart journal 34, 458-464, (1972). 374 Mahajan, V. S. & Jarolim, P. How to interpret elevated cardiac troponin levels. Circulation 124, 2350-2354, (2011). 375 Calabro, P., Willerson, J. T. & Yeh, E. T. Inflammatory cytokines stimulated C-reactive protein production by human coronary artery smooth muscle cells. Circulation 108, 1930-1932, (2003). 376 Eggers, K. M., Jaffe, A. S., Lind, L., Venge, P. & Lindahl, B. Value of cardiac troponin I cutoff concentrations below the 99th percentile for clinical decision-making. Clinical chemistry 55, 85-92, (2009). 377 Tharakan, J. A. Electrocardiogram in endomyocardial fibrosis. Indian pacing and electrophysiology journal 11, 129-133, (2011). 378 Barinas-Mitchell, E., Cushman, M., Meilahn, E. N., Tracy, R. P. & Kuller, L. H. Serum levels of C-reactive protein are associated with obesity, weight gain, and hormone replacement therapy in healthy postmenopausal women. American journal of epidemiology 153, 1094-1101, (2001). 379 Cushman, M. et al. Effect of postmenopausal hormones on inflammation-sensitive proteins: the Postmenopausal Estrogen/Progestin Interventions (PEPI) Study. Circulation 100, 717-722, (1999). 380 Kravitz, B. A., Corrada, M. M. & Kawas, C. H. Elevated C-reactive protein levels are associated with prevalent dementia in the oldest-old. Alzheimer's & dementia : the journal of the Alzheimer's Association 5, 318-323, (2009). 381 Woods, J. A., Wilund, K. R., Martin, S. A. & Kistler, B. M. Exercise, inflammation and aging. Aging Dis 3, 130-140, (2012). 382 Hibberts, N. A., Howell, A. E. & Randall, V. A. Balding hair follicle dermal papilla cells contain higher levels of androgen receptors than those from non-balding scalp. J Endocrinol 156, 59-65, (1998). 383 Trueb, R. M. Molecular mechanisms of androgenetic alopecia. Experimental gerontology 37, 981-990, (2002). 384 Fujimoto, R. et al. Androgen receptors, 5 alpha-reductase activity and androgen-dependent proliferation of vascular smooth muscle cells. The Journal of steroid biochemistry and molecular biology 50, 169-174, (1994). 385 Steinert, P. M., Parry, D. A. & Marekov, L. N. Trichohyalin mechanically strengthens the hair follicle: multiple cross-bridging roles in the inner root shealth. The Journal of biological chemistry 278, 41409-41419, (2003). 195  386 McHeyzer-Williams, M. G., Altman, J. D. & Davis, M. M. Enumeration and characterization of memory cells in the TH compartment. Immunological reviews 150, 5-21, (1996). 387 Giuliani, L. et al. Detection of GAD65 autoreactive T-cells by HLA class I tetramers in type 1 diabetic patients. Journal of biomedicine & biotechnology 2009, 576219, (2009). 388 Nepom, G. T. MHC class II tetramers. Journal of immunology 188, 2477-2482, (2012). 389 Gerdts, E. et al. Correlates of left atrial size in hypertensive patients with left ventricular hypertrophy: the Losartan Intervention For Endpoint Reduction in Hypertension (LIFE) Study. Hypertension 39, 739-743, (2002). 390 Abbasi, S. A. et al. Impact of cardiovascular magnetic resonance on management and clinical decision-making in heart failure patients. Journal of cardiovascular magnetic resonance : official journal of the Society for Cardiovascular Magnetic Resonance 15, 89, (2013). 391 Wheeler, R. & Masani, N. D. The role of echocardiography in the management of atrial fibrillation. European journal of echocardiography : the journal of the Working Group on Echocardiography of the European Society of Cardiology 12, i33-38, (2011).    196  Appendices  Appendix A   Induction of Alopecia Areata in C3H/HeJ Mice via Cultured Cell Transfer A.1 Introduction Alopecia areata (AA) is believed to be an inflammatory, non-scarring, cell-mediated autoimmune hair loss disease. There is a 1.7% life-time risk for AA in humans, it can affect both genders in all age groups1. Patients affected with AA typically experience patchy hair loss on the scalp, but it can progress to complete loss of scalp or even body hair. AA has an unpredictable nature and there are currently no effective treatments, because of this, AA can have a severe psychological burden, especially in women and children53. There are several different species of animals that spontaneously develop AA-like symptoms similar to humans, however, the rate of spontaneous AA development in most of these species is extremely low making them difficult to use as practical models to study AA3. Therefore, an easily accessible, small animal model that can be consistently induced to develop AA, while sharing a high biological resemblance to human AA, is required. A.2 General Experimental Design Animal studies and protocols were approved by the University of British Columbia Animal Care Committee. Normal haired, female C3H/HeJ mice can be obtained from The Jackson Laboratory (Bar Harbor, ME). A C3H/HeJ mouse that already has AA is needed to perform this technique; this can be obtained by aging a small colony of mice or purchase of a 197  C3H/HeJ mouse that already has AA induced by skin grafting from The Jackson Laboratory. From our experience, the mice must be on low-fat diet to obtain the best success rate144. The naïve cell injection recipient also must be at least 10-weeks old before they receive cell injection as it was previously identified that younger C3H/HeJ mice were relatively immune to AA transfer96,115; we have injected cultured cells into mice that were around one year old and still achieved the same efficacy. We have only used mice with over 70% hair loss as cell donors and have used donors that developed AA spontaneously or via skin-grafting and achieved the same results.  When collecting skin-draining lymph nodes, we collected cervical lymph nodes, auxiliary lymph nodes and inguinal lymph nodes. Mice with active AA will have noticeably enlarged lymph nodes compared to the same lymph nodes from healthy mice. The lymph nodes were processed immediately. We produced single cell suspensions via grinding the lymph nodes gently against a 70 μm cell strainer in complete medium, keeping the lymph nodes and LNCs moist during the procedure. We could achieve around 8-10 fold expansion of cell numbers from a 6-day culture. For cell injection, 10 million LNCs were used to induce AA in each mouse, therefore the number of LNCs that will be used to initiate the culture must be determined beforehand depending on the anticipated number of recipients. Depending on the application of this technique, the investigator may wish to include a comparative control where a separate mouse group receives similarly cultured LNCs derived from a healthy donor mouse without AA.    198  A.3 Materials and Reagent Setup REAGENTS • Naïve female C3H/HeJ mice as cell recipients (at least 10 weeks old, no AA phenotype; The Jackson Laboratory, Bar Harbor, ME).  CAUTION: All experiments involving live rodents must be approved by local regulatory institutions.  CRITICAL: Mice younger than 10 weeks old are relatively immune to AA induction. • AA affected female C3H/HeJ mice as cell donors (at least 70% hair loss. The Jackson Laboratory).  CAUTION: All experiments involving live rodents must be approved by local regulatory institutions. • Dynabeads Mouse T-Activator CD3/CD28 (Gibco, Burlington, ON. Catalog #114.52D) • Human Recombinant IL-2 (Roche Life Science, Laval, QC. Catalog #11011456001) • Mouse Recombinant IL-7 (R&D Systems, Minneapolis, MN. Catalog #407-ML-005) • Mouse Recombinant IL-15 (R&D Systems, Minneapolis, MN. Catalog #447-ML-010) • Sterile PBS, pH=7.4 (Mediatech Inc, Manassas, VA. Catalog #46-013-CM) • Sterile DPBS (Sigma, Oakville, ON. Catalog #D1408) 199  • EDTA (0.5 M), pH 8.0 (Ambion, Burlington, ON. Catalog #AM9260G) • Advanced RPMI 1640 (Gibco, Burlington, ON. Catalog #12633-012) • GlutaMAX (Gibco, Burlington, ON. Catalog #35050-061) • Fetal Bovine Serum (FBS; Gibco, Burlington, ON. Catalog #16000044) • Penicillin-Streptomycin (Gibco, Burlington, ON. Catalog #15140-148) • Trypan blue (Invitrogen, Burlington, ON. Catalog #15250-061)  EQUIPMENT • Centrifuge tubes • Petri dishes, 35x10 mm (Fisher Scientific, Ottawa, ON. Catalog #08-757-100A) • Centrifuge capable of up to 600x g • Hemocytometer • Inverted microscope • 37 ºC Incubator with 5% CO2 • Surgical scissors • Forceps • Electric hair shaver 200  • Non-treated Tissue culture plates – 24 well (BD Bioscience, Mississauga, ON. Catalog #351147) • Suspension cell TC flasks T25/T75 vented (Sarstedt, Montreal, QC. Catalog #83-1810-502 and 83-1813-502) • Sterile round bottom falcon tubes, 12x75 mm (VWR, Mississauga, ON. Catalog #734-0445) • 70uM cell strainer (Fisher Scientific, Ottawa, ON. Catalog #08-771-2) • 10 mL Syringes (BD Bioscience, Mississauga, ON. Catalog #309604) • EasySep™ Magnet (StemCell Technologies, Vancouver, BC. Catalog #18000) • 1cc Insulin syringes (BD, Mississauga, ON. Catalog #329420) • Heat pad or heat lamp • Anaesthesia and euthanasia apparatus  REAGENT SETUP Dynabead wash buffer: Prepare ahead of time, Sterile PBS with 0.1% FBS and 2 mM EDTA, pH 7.4. 1x PBS and DPBS: Prepare ahead of time by diluting with sterile distilled H2O. Complete medium AR10: Prepare ahead of time, Advanced RPMI 1640 with 10% FBS, 2mM GlutaMAX and 100 U/mL Penicillin-Streptomycin. 201  Mouse Recombinant IL-7: Reconstitute to 25 ng/µL in sterile PBS with 0.1% FBS. Aliquot and store in -20 ºC. Mouse Recombinant IL-15: Reconstitute to 50 ng/µL in sterile PBS with 0.1% FBS. Aliquot and store in -20 ºC.  Complete medium AR10 supplemented with cytokines: Prepare fresh, AR10 supplemented with 30 U/mL human recombinant IL-2, 25 ng/mL mouse recombinant IL-7 and 50 ng/mL mouse recombinant IL-15. Dynabead mixture: Prepare fresh, follow the manufacturer’s protocol to resuspend the dynabeads via vortex for 30 seconds. 25 µL of dynabeads is required for each 1 million LNCs. Pipette dynabeads into a sterile round bottom centrifuge tube and add 1 mL of dynabead wash buffer, pipette gently 30 times without generating air bubbles. Place the centrifuge tube into an EasySep magnet for 1 minute. Decant/Pipette the wash buffer into a waste container while keeping the tube inside the magnet; the dynabeads will adhere to the wall of the tube. Remove the tube from the magnet and add AR10 supplemented with cytokines at 500 µL/25 µL dynabeads. Make sure all of the dynabeads on the wall of the tube are reuspended into the medium. A.4 Detailed Procedures Isolation of skin-draining lymph nodes  1 Euthanize the donor mice using institution approved procedures. Clean mouse skin with 70% ethanol. 2 Use surgical scissors and forceps to dissect the mice from the ventral side. 202  3 Excise the inguinal, auxiliary and cervical skin-draining lymph nodes with forceps or scissors and trim away the extra fat and connective tissues. 4 Place the trimmed skin-draining lymph nodes into complete AR10 medium and keep on ice. CRITICAL STEP: Process the lymph nodes as soon as possible, prevent drying of tissues. Separation of lymph node cells (LNCs) into single cell suspension 5 Work in a biosafety cabinet, with aseptic techniques, fill a small petri dish with 1 mL of 1x DPBS. 6 Remove the lymph nodes from AR10 and transfer into DPBS to wash by brief rinsing. 7 In another clean petri dish, add 1 mL of fresh AR10 and place a 70 µM cell strainer in the middle of the dish. 8 Pick up the lymph nodes with a pair of forceps and transfer into the middle of the cell strainer. 9 Use the plunger end of a syringe to gently grind the lymph nodes against the cell strainer for about 5 minutes. 10 The culture medium will turn cloudy as the lymph nodes disintegrate, leaving a small amount of white connective tissue in the strainer. 203  CRITICAL STEP: While grinding the lymph nodes, change the angle and direction often to ensure complete breakdown of the tissues to recover the maximum amount of LNCs. 11 Pipette 1 mL of fresh AR10 and rinse the bottom of the cell strainer to wash the extra LNCs into the petri dish. There should be 2 mL of LNC suspension at this step. 12 Remove the cell strainer and transfer all of the LNC suspension into a clean 15 mL centrifuge tube. 13 Wash the petri dish with another 1 mL of fresh AR10 and transfer into the centrifuge tube. 14 Top up centrifuge tube to 10 mL with another 7 mL of fresh AR10, pipette thoroughly but gently to wash the LNC suspension. 15 Take out 10 µL and count the cell number with a hemocytometer and trypan blue, or use another method to count the cells. CRITICAL STEP: Once the total number of cells is calculated, determine how many cells will be used for expansion. 10 million cultured cells are required to inject into a single recipient; the number of cells will expand around 8-10 fold during the 6-day culture. Therefore, we initiate the culture with 2 million cells for each mouse we expect to inject to ensure we will have enough cells for the target number of cell recipients. 16 Prepare complete medium AR10 with cytokines. Make enough AR10 with cytokines to resuspend the cell pellet into 2 million cells/mL.  204  17 Centrifuge the LNC suspension for 5 minutes at 350x g at room temperature.  18 Remove the supernatant carefully leaving the cell pellet. 19 Resuspend the cell pellet to 2 million cells/mL with AR10 supplemented with cytokines. 20 Pipette 1 mL of LNC suspension into a 24-well, non-tissue culture treated plate until desired number of wells are filled. Each single well should now have 2 million LNCs and each is expected to expand 8-10 fold after 6 days. CRITICAL STEP: Keeping the LNCs at a high density throughout the culturing process allows better cell-to-cell contact as well as interaction with the antibody coated magnetic beads. In pilot experiments, initiating the cell culture in a T25 flask with a larger volume of media did not yield as high cell number compared to initiating cultures in a 24-well culture plate. 21 Place the 24-well plates in the 37 ºC incubator with 5% CO2. 22 Prepare dynabead mixture at this point, resuspend well. 23 Take out the 24-well plates from the incubator and add 500 µL of dynabead mixture into each well of the plate. Each well should have 2 million cells in 1.5 mL of medium supplemented with cytokine and dynabeads. 24 Transfer the plate back into 37 ºC incubator with 5% CO2. Activation and expansion of LNCs 25 Check the culture daily but do not disturb the LNC suspension. 205  CRITICAL STEP: Clumping of LNCs with the dynabeads can be observed after 24 hours (Figure 2a). Do not attempt to break up the clumps as they will disappear after 48 hours.  26 The LNCs will expand and cover the entire bottom of the wells after 48 hours, the color of culture medium will turn light yellow.  27 Split each well to two by gently resuspending the culture and pipetting 750 μL to an adjacent well. Add 750 μL of AR10 supplemented with cytokines so each well contains 1.5 mL. 28 After another 24 hours, combine 2x wells into 1x T25 flask and add 3 mL of AR10 supplemented with cytokines. 29 After another 24 hours, combine 2x T25 into 1x T75 flask and add 8 mL of AR10 supplemented with cytokines. CRITICAL STEP: The passaging protocol here serves as a general guideline only. The purpose is to keep the LNCs at very high density, around 1.5 to 2 million cells/mL. Preparation of expanded LNCs for injection 30 Combine the LNC suspension in T75 flasks into 50 mL centrifuge tube(s). 31 Place a clean round bottom centrifuge tube into an EasySep magnet or similar. 32 Transfer LNCs suspension into the centrifuge tube to a level at the same height as the magnet and let it set for 1 minute. 206  33 Gently pipette the LNC suspension from the centrifuge tube to a clean 50 mL centrifuge tube. 34 Repeat steps 32-33 until removal of all dynabeads from all LNC suspension has been achieved. CRITICAL STEP: Remember to keep the centrifuge tube inside the magnet while transferring LNC suspension to a new 50 mL falcon tube. If pipetting, do not scratch the wall of the tube as the dynabeads will come off back into the LNC suspension. 35 Take 10 μL and count the number of cells with a hemocytometer. 36 Centrifuge the LNC suspension for 5 minutes at 350x g at room temperature.  37 Remove the supernatant from the LNC pellet and resuspend into 10 million cells/100 μL with sterile PBS. 38 Draw the resuspended LNC suspension into insulin syringe(s) or similar and keep on ice. CRITICAL STEP: It is preferable to avoid using normal syringes as there will be a small volume inside the syringe that is very difficult to eject, insulin syringes bypass this problem because syringe and needle are sealed as one unit. Inject the LNCs as soon as possible. Adoptive transfer of LNCs into naïve C3H/HeJ mice 39 Anaesthetize the recipient mice using isoflurane or similar following the standardized protocols as provided by the local institution. 207  CRITICAL STEP: Recipient mice must be at least 10 weeks old or they are unlikely to be susceptible to AA induction. 40 Shave a small area on the lower back of the mice to expose an area of skin for intra-dermal injection. 41 Gently pinch up a section of the skin with a pair of blunt forceps and insert the syringe needle almost parallel to the plane of the skin pinched up by the forceps, into the dermis layer. CRITICAL STEP: Be careful not to go too deep or the LNCs will be injected into subcutaneous layer. Preliminary investigations revealed significantly lower success rates with subcutaneous injection. 42 Inject 100 μL of LNC suspension slowly into the skin. A small bulge will appear at the site of injection. 43 Retract the syringe, but use the forceps to hold the skin for a few seconds to allow the injected LNCs to dissipate a little bit as well as to allow the injection site opening to close. 44 Transfer the injected mice back to their cage but take care not to press on the injection site. 45 Repeat steps 40-43 until all mice are injected. 46 The mice should recover from anaesthesia within 10 minutes.  208  Development of alopecia areata 47 Continue to feed the mice with a low-fat diet. High fat/oil diets reduce success rates of AA transfer. 48 Hair at the site of injection will start to grow back after about one week due to injury induced hair growth. 49 Monitor the ventral side of the mice every few days as hair loss can start to develop as early as two weeks post injection and typically initially occurs away from the injection site. Timing Steps 1-4, Isolation of skin-draining lymph nodes: 30 minutes Steps 5-24, Separation of lymph node cells (LNCs) into single cell suspension: 1 hour Steps 25-29, Activation and expansion of LNCs: 6 days Steps 30-38, Preparation of expanded LNCs for injection: 1.5 hours Steps 39-46, Adoptive transfer of LNCs into naïve C3H/HeJ mice: 5-10 min per mouse Steps 47-49, Initial development of alopecia areata: Up to 18 weeks A.5 Anticipated Results Single cell suspension and LNC expansion The number of fresh LNCs from each donor AA mouse is variable, we can typically obtain around 50 to 90 million cells from one donor. The rate of cell expansion can also be variable, usually at 48 hours changes in cell morphology can be observed. By the 72nd hour, the number of cells should cover over 90% of the bottom of the 24-well plate. The rate of cell 209  expansion will start to slow down after 6 days, therefore, it is not advisable to culture cells beyond that point.  The progression of alopecia areata in cell injected mice We have observed the mice start to lose hair as early as two weeks and as late as 18 weeks after injection with AA cells; between 7 to 10 weeks seems to be the peak where most mice will first develop AA hair loss. The hair loss does not begin at the site of injection, rather, it usually progresses from patches first observed on the ventral side to the dorsal skin; a pattern similar to that observed with skin grafted mice. It should be anticipated that not all AA mice receiving cultured AA LNCs will develop AA within 20 weeks, some mice are apparently resistant to AA induction while any AA development after 20 weeks is unlikely to be a result of cell injection. Conversely, it is also possible for a small numbers of control mice that received cultured control LNCs to develop AA simply because this strain is known to develop spontaneous AA at a low rate. General troubleshooting suggestions are provided in Table A.1.  210  Table A.1 Troubleshooting suggestions. Step Problem Solution 28, 35 Low cell recovery. Even though LNCs are suspension cells, some of them still tend to adhere to the bottom of plates or flasks. Wash the surface with additional AR10 supplemented with cytokines and combine with the suspension. 37 Red color precipitates within the cell pellets. Dynabeads were not completely removed. Resuspend the cell pellet with 1-2 mL of PBS in the round bottom centrifuge tubes and place into the magnet for another 2 minutes; repeat the wash step. 49 Cell injected mice don’t develop hair loss within 20 weeks. Ensure female C3H/HeJ mice that were 10 weeks or older was used. Check if they are on the right type of low fat diet. Inject the LNCs as soon as they are removed from culture, preferably within one hour. During injection, make sure to inject slowly into the dermis, not into the subcutaneous layer. Do not inject male derived LNCs into female recipients, the injected cells may be rejected (female derived LNCs injected to male recipients should be accepted).    211  Appendix B  Questionnaire for Alopecia Areata Patients  212    213     214  Appendix C  Questionnaire for Control Subjects  215   216    217  Appendix D  Primer Sequences Table E.1 Primer sequences used in chapter 5 Gene Symbol Forward Reverse Cardiac Troponin I Cti TCTGCCAACTACCGAGCCTAT CTCTTCTGCCTCTCGTTCCAT Cardiac Troponin T Ctt CAGAGGAGGCCAACGTAGAAG TCGATCAGAGTCTGTAGCTCATT Fas Fas TGCAAGTGCAAACCAGACTTC GTCAACAACCATAGGCGATTTCT Fas Ligand Fasl ACCCCCACTCAAGGTCCAT CGAAGTACAACCCAGTTTCGT St2 (IL1 Receptor-like 1) St2 ATGGGAGAGACCTGTTACCTG CCTGCTCGTAGGCAAATTCCT Myosin light polypeptide kinase Mlik TGGGGGGACGTGAAACTGTTTG GGGGCAGAATGAAAGCTGG Interleukin 6 Il6 TCCAGTTGCCTTCTTGGGAC GTGTAATTAAGCCTCCGACTTG TNFα Tnfa TTCTGTCTACTGAACTTCGGGGTGATCGGTCC GTATGAGATAGCAAATCGGCTGACGGTGTGGG Adrenomedullin Ad GGAATAAGTGGGCGCTAAGTC CAAGAGTCTGGGTAGGAACTGT Natriuretic peptide type B Nppb CCCAAAAGAGTCCTTCGGTC CGGTCTATCTTGTGCCCAAAG Granzyme B Gzmb TGGACCCTACATGGCCTTAC TGGGGAATGCATTTTACCAT  Interleukin 18 Il18 GTGAACCCCAGACCAGACTG CCTGGAACACGTTTCTGAAAGA Matrix metallopeptidase 2 Mmp2 TTTGCTCGGGCCTTAAAAGTAT   CCATCAAACGGGTATCCATCTC  Matrix metallopeptidase 9 Mmp9 TGCCCATTTCGACGACGAC   GTGCAGGCCGAATAGGAGC  IL18 Receptor-1 Il18r1 TAATCATCGTTCTCAGCCAGAGT   GGACTGTCAGCCCTCCATTTT  IL18 Binding Protein Il18bp CCTACTTCAGCATCCTCTACTGG   AGGGTTTCTTGAGAAGGGGAC  218   Gene Symbol Forward Reverse IL18 Binding Protein Il18bp CCTACTTCAGCATCCTCTACTGG   AGGGTTTCTTGAGAAGGGGAC  B cell lymphoma 2 Bcl2 GCTACCGTCGTGACTTCGC   CAACCAGACATGCACCTACCC  BCL2-like 1 Bclxl GGGATGGAGTAAACTGGGGTC   TGTTCCCGTAGAGATCCACAAA  Tissue inhibitor of MMP2 Timp2 TCAGAGCCAAAGCAGTGAGC   GCCGTGTAGATAAACTCGATGTC  Caspase 1 Casp1 ACAAGGCACGGGACCTATG   TCCCAGTCAGTCCTGGAAATG  Collagen Ia1 Col1a1 AGCTTTGTGGACCTCCGGCT ACACAGCCGTGCCATTGTGG Collagen IIIa1 Col3a1 TGCCCACAGCCTTCTACACCT CCAGCTGGGCCTTTGATACCT Collagen Va1 Col5a1 TGAATTCAAGCGTGGGAAACT CCGCAGGAAGGTCATTTGTAC Atrial Natriuretic Factor Nppa GTACAGTGCGGTGTCCAACA TCTCCTCCAGGTGGTCTAGCA β Myosin Heavy Chain Myh7 GCATTCTCCTGCTGTTTCCTT TGGATTCTCAAACGTGTCTAGTGA  219  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.24.1-0167061/manifest

Comment

Related Items