Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Investigating systemic immune responses in the pathophysiology of allergic rhinitis using peripheral… Kim, Young Woong 2018

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

Item Metadata

Download

Media
24-ubc_2018_may_kim_youngwoong.pdf [ 28.02MB ]
Metadata
JSON: 24-1.0364662.json
JSON-LD: 24-1.0364662-ld.json
RDF/XML (Pretty): 24-1.0364662-rdf.xml
RDF/JSON: 24-1.0364662-rdf.json
Turtle: 24-1.0364662-turtle.txt
N-Triples: 24-1.0364662-rdf-ntriples.txt
Original Record: 24-1.0364662-source.json
Full Text
24-1.0364662-fulltext.txt
Citation
24-1.0364662.ris

Full Text

INVESTIGATING SYSTEMIC IMMUNE RESPONSES IN THE PATHOPHYSIOLOGY OF ALLERGIC RHINITIS USING PERIPHERAL BLOOD  by  Young Woong Kim  B.Sc., Yonsei University, 2003 MMS., Yonsei University, 2004  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)  March 2018  © Young Woong Kim, 2018   ii Abstract  Allergic rhinitis (AR) is the most prevalent allergic disease worldwide, affecting up to 40% of the global population. AR is a symptomatic disorder of the nose induced by IgE-mediated allergic inflammation of the nasal mucosa. Many studies have focused on the local inflammatory site contributing to an understanding of the pathophysiology of AR. The systemic immune responses of AR, however, have not been well investigated, and approaches for such investigation are scant.  In order to identify systemic immune response signatures, we used peripheral blood collected from subjects with AR following nasal allergen challenge (NAC). The response signatures we identified consist of immune gene clusters associated with frequencies of corresponding immune cells that reflect dynamic immune responses after NAC. In particular, we found individual clusters associated with neutrophils, neutrophil/lymphocyte ratio (NLR), and lymphocytes, which demonstrated significantly different patterns between allergic and non-allergic subjects. The NLR-associated cluster was also moderately associated with clinical symptoms at 6 h post-NAC in cat allergic subjects. We tested how the identified signatures in cat allergy are expressed in birch and ragweed allergies, which are seasonal/intermittent allergies, using the NAC and Environmental Exposure Unit models. In the NAC model, while the birch allergic subjects did not demonstrate significantly (p > 0.05) different total nasal symptom score (TNSS) from the cat allergic subjects, there were fewer significantly (BH-FDR < 0.1) differentially expressed genes after allergen challenge in birch allergy (4 genes) than immune gene signatures (53 genes) identified in cat   iii allergy. The difference may be associated with the distinct clinical symptoms reported between seasonal/intermittent allergy and perennial/persistent allergy. We also used our systemic immune gene signature approach to help determine the possible mechanism of action of the novel immunotherapeutic Cat-PAD. The significant reduction of TNSS that allergic subjects experienced after the immunotherapy was strongly associated with significant increase of lymphocytes at 1 h post-NAC following treatment. As an exploratory data analysis, the approach identified five genes likely associated with the correlation.  Collectively, the systemic immune gene signature approach may be a useful and potentially objective method to diagnose AR and investigate the efficacy and mechanisms of AR treatments.    iv Lay Summary  Allergic rhinitis (AR, also known as hay fever) is the most prevalent allergic disease worldwide, affecting up to 40% of the global population. AR is troublesome in a patient’s daily life due to its symptoms, which include runny/stuffy nose, sneezing, itching, facial pain, and headaches. AR is induced by Immunoglobulin E– (a specific antibody, or molecule, associated with allergy) mediated allergic inflammation. This inflammation, or swelling, is a complex biological defense response that occurs when the body’s immune system misjudges a harmless allergen as a harmful pathogen (e.g., bacteria or parasite). We identified systemic immune response patterns, namely combinations of changes in cell counts and immune gene expression in blood, after allergen exposure. These immune response patterns may indicate how the body’s immune system communicates with cells and responds to allergen challenge. They may be useful to study AR and test AR therapeutics to improve treatment of these patients.       v Preface  I submit this dissertation for the degree of Doctor of Philosophy. The studies in this dissertation were conducted under the supervision of Professor Scott J. Tebbutt in the Department of Medicine (Respiratory Division) at the University of British Columbia, and in collaboration with Dr. Anne K. Ellis (Queen’s University), Dr. Helen Neighbour (McMaster University), Dr. Mark Larché (McMaster University), Dr. Elena Tonti (McMaster University), Dr. Daniel R. Gliddon (Circassia Ltd., Oxford, United Kingdom), and Dr. Pascal Hickey (Adiga Life Sciences, ON). My PhD progress was also supervised and supported by my supervisory committee, consisting of Dr. Don D. Sin and Dr. Pascal Bernatchez. I contributed to all of the studies described by designing and performing experiments, performing bioinformatics and statistical analyses, writing manuscripts, and presenting at conferences. The cat allergy study described in Chapters 3, 4, and 5 is registered on clinicaltrials.gov (Queen’s University, NCT01383590; McMaster University, NCT01383603), and was granted ethical clearance by the Queen’s University Health Sciences and Affiliated Teaching Hospitals Research Ethics Board at Queen’s University (DMED-1423-11) and the Hamilton Integrated Research Ethics Board at McMaster University (Research Project #11-3551). The birch allergy and ragweed allergy studies described in Chapter 4 were granted ethical clearance by the Queen’s University Health Sciences and Affiliated Teaching Hospitals Research Ethics Board at Queen’s University (DMED-1343-10 and DMED-1250-09, respectively). These studies were also cleared by the University of British Columbia Research Ethics Board of the Province Health Care Research Institute (H09-02114).   vi A version of Chapter 3 has been published in the Journal of Immunology: Kim YW, Singh A, Shannon CP, Thiele J, Steacy LM, Ellis AK, et al. “Investigating Immune Gene Signatures in Peripheral Blood from Subjects with Allergic Rhinitis Undergoing Nasal Allergen Challenge” J Immunol. 2017 Nov 15;199(10):3395–405. A poster version of Chapter 3 was presented at the Canadian Society of Allergy and Clinical Immunology (CSACI) 72nd Annual Scientific Meeting, Oct. 11-15th, 2017, where I won 2nd place for Best Poster in the Allergic Rhinitis/Asthma category, and at the 2017 Medicine Research Expo at UBC on Oct. 31st, 2017, where I won the Best Poster Award. A version of Chapter 4 will be submitted for peer-review publication. A version of Chapter 5 will also be submitted for peer-review publication. An abstract version of Chapter 5 has been published in the European Academy of Allergy and Clinical Immunology Online Library: Gliddon DR, Kim YW, Shannon CP, Singh A, Tebbutt SJ, Hickey PLC, Ellis AK, Neighbour H, Larché M. “Whole blood immune transcriptome profiling reveals systemic pathways associated with the mechanism of action of cat–synthetic peptide immune–regulatory” EAACI Online Library. Gliddon D. Jun 6, 2015; 104988. A second abstract version of Chapter 5 has been published in Allergy, Asthma & Clinical Immunology: Kim YW, Gliddon DR, Shannon CP, Singh A, Hickey PLC, Ellis AK, Neighbour H, Larché M, Tebbutt SJ. “Systemic immune pathways associated with the mechanism of CatSynthetic Peptide ImmunoRegulatory Epitopes, a novel immunotherapy, in whole blood of catallergic people” Allergy Asthma Clin Immunol. Aug 25, 2016, 12(Suppl 1): A53. A poster version of Chapter 5 has been presented at the CSACI 70th Annual Scientific Meeting/AllerGen Poster Competition, Oct. 21-24th, 2015, where I won 1st place for Best Poster in the PhD category. Publication and presentation during my PhD studies are listed in Appendix C and D.   vii Table of Contents  Abstract ..................................................................................................................................... ii Lay Summary .......................................................................................................................... iv Preface ....................................................................................................................................... v Table of Contents .................................................................................................................... vii List of Tables.......................................................................................................................... xiv List of Figures ......................................................................................................................... xv List of Abbreviations ........................................................................................................... xviii Acknowledgements ................................................................................................................. xx Dedication ............................................................................................................................. xxii Chapter 1: Introduction ........................................................................................................... 1 1.1 Thesis overview .......................................................................................................... 1 1.2 Allergic rhinitis (AR) .................................................................................................. 3 1.2.1 Rhinitis ................................................................................................................ 3 1.2.2 Allergic rhinitis ................................................................................................... 4 1.2.3 Intermittent allergic rhinitis (IAR) and persistent allergic rhinitis (PER) .............. 5 1.2.4 The pathophysiology of AR ................................................................................. 5 1.2.5 Systemic disease characteristic of allergic rhinitis ................................................ 8 1.3 Human model of AR ................................................................................................... 9 1.3.1 Nasal allergen challenge (NAC) model ................................................................ 9 1.3.2 Environmental exposure unit (EEU) model ........................................................ 10 1.4 Identifying patterns in gene expression data .............................................................. 10   viii 1.5 Hypotheses ................................................................................................................ 11 1.5.1 Hypothesis of Chapter 3 .................................................................................... 12 1.5.2 Hypothesis of Chapter 4 .................................................................................... 12 1.5.3 Hypothesis of Chapter 5 .................................................................................... 12 Chapter 2: Main experimental methods ................................................................................ 13 2.1 Human model of allergic rhinitis (AR) ...................................................................... 13 2.1.1 Nasal allergen challenge (NAC) model .............................................................. 13 2.1.2 Environmental exposure unit (EEU) model ........................................................ 14 2.2 Clinical symptom scores ............................................................................................ 16 2.2.1 Total nasal symptom score (TNSS) .................................................................... 17 2.2.2 Peak nasal inspiratory flow (PNIF) .................................................................... 17 2.3 Complete blood cells measurement with 4-part differential blood cells counting (CBC)    .................................................................................................................................. 18 2.4 Gene expression profiling .......................................................................................... 18 2.4.1 NanoString nCounter gene expression assay ...................................................... 18 2.4.1.1 Method .......................................................................................................... 18 2.4.1.2 Adjusting setting ............................................................................................ 19 2.4.1.2.1 Sample type ............................................................................................. 19 2.4.1.2.2 Reproducibility ........................................................................................ 22 2.4.1.2.3 Sample preparation sensitivity.................................................................. 22 2.4.1.3 Removing batch effects.................................................................................. 24 2.4.2 Microarray ......................................................................................................... 27 2.5 Clustering of profiled gene expression ....................................................................... 28   ix 2.5.1 Filtering immune genes for clustering ................................................................ 28 2.5.2 Fuzzy c-means (FCM) clustering ....................................................................... 28 2.5.3 Enrichment analyses .......................................................................................... 29 2.5.4 Statistical analysis ............................................................................................. 29 Chapter 3: Systemic immune response signature in peripheral blood from subjects with allergic rhinitis undergoing nasal allergen challenge ............................................................ 30 3.1 Sub-abstract .............................................................................................................. 30 3.2 Introduction ............................................................................................................... 30 3.3 Materials and methods ............................................................................................... 31 3.3.1 Study approval................................................................................................... 31 3.3.2 Subjects ............................................................................................................. 32 3.3.3 Nasal allergen challenge (NAC) ........................................................................ 34 3.3.4 Missing Data ..................................................................................................... 36 3.3.5 Canonical correlation ......................................................................................... 37 3.4 Results ...................................................................................................................... 37 3.4.1 Investigation of the clinical symptom scores during NAC .................................. 37 3.4.2 Investigation of immune cell frequencies and neutrophil/lymphocyte ratio during NAC  .......................................................................................................................... 38 3.4.3 Seven clusters of immune gene expression patterns during NAC ....................... 40 3.4.4 Relatedness of the identified clusters and immune cells ..................................... 43 3.4.5 Correlation of clinical symptoms and clusters or immune cell frequencies in 13 AR subjects. ...................................................................................................................... 44   x 3.4.6 Comparison of immune gene signature patterns between 13 AR subjects and 5 healthy non-allergic control subjects ................................................................................. 47 3.5 Discussion ................................................................................................................. 48 Chapter 4: Validation test of the systemic immune gene signature in seasonal allergy (birch and ragweed allergies) ............................................................................................................ 54 4.1 Sub-abstract .............................................................................................................. 54 4.2 Introduction ............................................................................................................... 54 4.3 Materials and methods ............................................................................................... 56 4.3.1 Study approval................................................................................................... 56 4.3.2 Subjects ............................................................................................................. 56 4.3.3 NAC .................................................................................................................. 59 4.3.4 EEU .................................................................................................................. 59 4.3.5 Clinical symptom scores collection .................................................................... 59 4.3.6 Blood collection ................................................................................................ 60 4.3.7 Microarray ......................................................................................................... 60 4.4 Results ...................................................................................................................... 61 4.4.1 Birch allergy ...................................................................................................... 61 4.4.1.1 Clinical symptom score in the NAC model .................................................... 61 4.4.1.2 Clinical symptom score in the EEU model ..................................................... 61 4.4.1.3 Immune cell frequencies and neutrophil/lymphocyte ratio in the NAC model 63 4.4.1.4 Immune cell frequencies and neutrophil/lymphocyte ratio in the EEU model . 64 4.4.1.5 Differentially expressed immune genes in birch allergy ................................. 64 4.4.2 Ragweed allergy ................................................................................................ 67   xi 4.4.2.1 Clinical symptom scores ................................................................................ 67 4.4.2.2 Immune cell frequencies and neutrophil/lymphocyte ratio.............................. 68 4.4.2.3 Comparison of differentially expressed immune genes ................................... 69 4.4.3 Comparison of pollen allergy to cat allergy ........................................................ 71 4.4.3.1 Clinical symptoms ......................................................................................... 71 4.4.3.1.1 PNIF in cat, birch, ragweed allergy studies............................................... 71 4.4.3.1.2 TNSS in cat, birch, ragweed allergy studies ............................................. 73 4.4.3.2 The ratios of immune cell frequencies ............................................................ 74 4.4.3.3 Immune gene signatures in pollen allergy ...................................................... 77 4.4.3.3.1 In the NAC model .................................................................................... 77 4.4.3.3.2 In the EEU model .................................................................................... 81 4.5 Discussion ................................................................................................................. 84 Chapter 5: Utility of the systemic immune gene signature approach ................................... 89 5.1 Sub-abstract .............................................................................................................. 89 5.2 Introduction ............................................................................................................... 89 5.3 Materials and methods ............................................................................................... 91 5.3.1 Study approval................................................................................................... 91 5.3.2 Subjects ............................................................................................................. 92 5.3.3 NAC .................................................................................................................. 93 5.3.4 Cat-PAD ............................................................................................................ 93 5.4 Results ...................................................................................................................... 95 5.4.1 Comparison of clinical symptom scores pre- and post-treatment ........................ 95 5.4.1.1 PNIF .............................................................................................................. 95   xii 5.4.1.2 TNSS ............................................................................................................. 95 5.4.2 Comparison of immune cell frequencies and neutrophil/lymphocyte ratio pre- and post-treatment ................................................................................................................... 98 5.4.3 Differentially expressed gene pre- and post-treatment ...................................... 100 5.4.4 Comparison of immune gene signatures pre- and post-treatment ...................... 102 5.4.5 Investigation of the mechanism of action of Cat-PAD using the systemic immune response signature approach ............................................................................................ 106 5.4.5.1 Relationship between clinical symptoms and immune cell frequencies......... 106 5.4.5.2 Investigation in cell type markers at 1h post-NAC ....................................... 109 5.4.5.3 Investigation of immune gene expression signatures at 1 h post-NAC .......... 109 5.4.5.3.1 Identification of five immune genes by a statistical test and a systemic immune gene signature approach ............................................................................. 111 5.4.5.3.2 Correlations between TNSS and the 5 immune gene expression difference (V3 - V1A) .............................................................................................................. 112 5.5 Discussion ............................................................................................................... 115 Chapter 6: Conclusions and future directions ..................................................................... 123 6.1 Overall summary and conclusions ........................................................................... 123 6.2 Strengths and Limitations ........................................................................................ 126 6.2.1 Strengths ......................................................................................................... 126 6.2.2 Limitations ...................................................................................................... 127 6.3 Future directions ..................................................................................................... 128 Bibliography ......................................................................................................................... 130 Appendices ............................................................................................................................ 144   xiii Appendix A Supplementary material for Chapter 3 ............................................................. 144 A.1 Immune gene list of the clusters identified (X: Not available). ............................. 144 A.2 Correlation between clinical symptoms and systemic immune response in 13 subjects ........................................................................................................................... 149 Appendix B Supplementary material for Chapter 5 ............................................................. 151 B.1 PNIF of pre-treatment (V1A) and post-treatment (V3) in the 10 cat allergic subjects .    ............................................................................................................................ 151 B.2 TNSS of pre-treatment (V1A) and post-treatment (V3) in the 10 cat allergic subjects    ............................................................................................................................ 152 B.3 Immune gene list of the clusters identified at V1A and V3 (X: Not available). ..... 153 B.4 Significantly higher expressed genes after Cat-PAD in 10 subjects (LIMMA paired test, V1A versus V3 at each time point, BH-FDR < 0.1).................................................. 157 B.5 Significantly lower expressed genes after Cat-PAD in 10 subjects (LIMMA paired test, V1A versus V3 at each time point, BH-FDR < 0.1).................................................. 163 Appendix C Publications arising during my PhD studies: .................................................... 177 Appendix D Presentations arising during my PhD studies: .................................................. 179    xiv List of Tables  Table 3-1. Characteristics of the Q and the M cohorts (Fisher's exact test). ............................... 33 Table 3-2. Characteristics of the 13 AR subjects and the 5 healthy non-allergic subjects (Fisher’s exact test). ................................................................................................................................ 34 Table 3-3 Similarities between clusters from both cohorts: ....................................................... 42 Table 4-1 Characteristics of subjects. ........................................................................................ 58 Table 5-1 Characteristics of the study populations: ................................................................... 94 Table 5-2 Differential expression patterns in 25 overlapped immune genes at V1A and V3. ... 104    xv List of Figures  Figure 1-1 Outline of dissertation chapters, including experiments performed. ............................ 2 Figure 1-2 The summary of the pathophysiology of AR. ............................................................. 6 Figure 2-1 NAC model. ............................................................................................................ 14 Figure 2-2 Layout of the Environmental Exposure Unit (EEU) (52). ......................................... 15 Figure 2-3 EEU model. ............................................................................................................. 16 Figure 2-4. Pearson correlations of technical replicates to compare lysate and RNA sample types from a PAXgene RNA blood tube of Sample 1. ........................................................................ 20 Figure 2-5 Bland-Altman plot of samples. ................................................................................ 21 Figure 2-6. Comparison between normal and high sensitivity protocols in nCounter prep station. ................................................................................................................................................. 23 Figure 2-7 Bland-Altman plot of comparison of normal and high sensitivity protocols.............. 24 Figure 2-8 Correlation comparison before and after batch effect correction. .............................. 26 Figure 2-9 PCA plot before and after batch effects correction. .................................................. 27 Figure 3-1 Diagram of an NAC model. ..................................................................................... 35 Figure 3-2 Clinical symptom scores after NAC. ........................................................................ 38 Figure 3-3 CBC after NAC. ...................................................................................................... 39 Figure 3-4 Seven clusters of differentially expressed immune genes in blood after NAC. ......... 41 Figure 3-5 The relationship between clusters and immune cell frequencies. .............................. 43 Figure 3-6 Correlations between clinical symptoms and systemic immune responses in 13 AR subjects. .................................................................................................................................... 45   xvi Figure 3-7 Comparison of immune gene signature patterns in 13 AR subjects to those of 5 healthy non-allergic subjects. .................................................................................................... 48 Figure 4-1 Clinical symptom scores after allergen challenge in birch allergy. ........................... 62 Figure 4-2 CBC pre- and post-allergen challenge. ..................................................................... 63 Figure 4-3 Volcano plot of gene expression data at comparison in NAC model. ........................ 65 Figure 4-4 Volcano plot of gene expression data at comparison in EEU model. ........................ 66 Figure 4-5 Clinical symptom scores. ......................................................................................... 68 Figure 4-6 CBC in ragweed allergic subjects pre- and post-allergen challenge. ......................... 69 Figure 4-7 Volcano plot of gene expression data at comparison in ragweed allergic subjects in the EEU model. ........................................................................................................................ 70 Figure 4-8 PNIF in all studies. .................................................................................................. 72 Figure 4-9 TNSS in all studies. ................................................................................................. 73 Figure 4-10 Ratios of immune cell frequencies. ........................................................................ 76 Figure 4-11 Immune gene clusters (using 4 significant genes) after a statistical cut-off in birch allergic subjects in the NAC model. .......................................................................................... 79 Figure 4-12 Immune gene clusters without a statistical cut-off in birch allergic subjects in the NAC model. ............................................................................................................................. 80 Figure 4-13 Immune gene clusters (using 11 significant genes) after a statistical cut-off in ragweed allergic subjects in the EEU model.............................................................................. 82 Figure 4-14 Immune gene clusters (using 43 genes) without a statistical cut-off in the EEU model. ...................................................................................................................................... 83 Figure 5-1 Diagram of an NAC model with Cat-PAD, an immunotherapy intervention. ............ 93 Figure 5-2 Comparison of PNIF pre- and post-treatment. .......................................................... 96   xvii Figure 5-3 Comparison of TNSS pre- and post-treatment. ......................................................... 97 Figure 5-4 Immune cell frequencies before and after immunotherapy intervention. ................... 99 Figure 5-5 Significantly differentially expressed genes after Cat-PAD. ................................... 101 Figure 5-6 Clustering in 10 AR subjects at V1A and V3. ........................................................ 103 Figure 5-7 Comparison of clusters pre- and post-treatment. .................................................... 105 Figure 5-8 Correlation between difference of TNSS and difference of immune cell frequencies. ............................................................................................................................................... 108 Figure 5-9 Cell type markers significantly more highly expressed at 1 h post NAC at V3 compared to V1A. .................................................................................................................. 110 Figure 5-10 The process to identify the five genes associated with the correlation between clinical symptom reduction and lymphocyte frequency change at 1 h post NAC in the 10 AR subjects. .................................................................................................................................. 112 Figure 5-11 Comparisons of expression of the five genes in the 10 AR subjects. ..................... 114 Figure 5-12 Potential mechanism of the action of Cat-PAD. ................................................... 118 Figure 5-13 Potential immune cells associated with the mechanism of the action of Cat-PAD. 121 Figure 6-1 Systemic immune response signatures, combinations of immune cell frequency and immune gene clusters found in Chapter 3. ............................................................................... 124    xviii List of Abbreviations  AR, allergic rhinitis BAU, bioequivalent allergen unit  BH-FDR, Benjamini–Hochberg false discovery rate  BMT, bone marrow transplantation  CBC, complete blood count ComBat, combatting batch effects DCs, dendritic cells EB, empirical Bayes EEU, environmental exposure unit EPR, early phase response FCM clustering, Fuzzy c-means clustering ILCs, innate lymphoid cells ILC2, type 2 innate lymphoid cells LME, linear mixed effect model LPR, late phase response M cohort, McMaster University study subjects in the cat allergy study NAC, nasal allergen challenge NGS, next-generation sequencing NLR, neutrophil/lymphocyte ratio PCA, principal component analysis PCR, polymerase chain reaction   xix PNIF, peak nasal inspiratory flow PNU, protein nitrogen unit Q cohort, Queen’s University study subjects in the cat allergy study SD, standard deviation Tcm, central memory T cells Th1, type 1 T helper cells Th17, T helper 17 cells Th2, type 2 T helper cells TNSS, total nasal symptom score Treg, regulatory T cells V1A, Visit 1A (pre-treatment visit) V3, Visit 3 (post-treatment visit)    xx Acknowledgements “The fruit of the righteous is a tree of life, and the one who is wise saves lives.” <Proverbs 11:30>  In the world, I am a little one, but Jesus Christ makes me happy with my beloved people. Without their love and support, my Ph.D. study would have been impossible. I would like to deeply thank my supervisor and mentor Dr. Scott Tebbutt for his tremendous love and support. He always shows me open-minded consideration, warm encouragement, and wonderful insight. He considers not only his students’ research qualifications but also their personal needs, such as in my case, a dependent family. I would also like to thank my Ph.D. supervisory committee Dr. Don Sin and Dr. Pascal Bernatchez who give me immense support and love that have guided me to keep walking on the way to a successful Ph.D. study. I also want to thank Dr. R. Robert Schellenberg, Dr. Fiona Brinkman, and Dr. Andrew Sandford who were members of my Ph.D. comprehensive exam. And many thanks to my Ph.D. oral examining committee (Chair Dr. Karen H. Bartlett, External Examiner Dr. Andrew Craig, and University Examiner Dr. Kelly McNagny and Dr. Sara Mostafavi). I would like to thank my collaborators: McMaster University (Dr. Mark Larché, Dr. Helen Neighbour, and Dr. Elena Tonti), Queen’s University (Dr. Anne K Ellis, Ms. Jenny Thiele, Ms. Lisa M. Steacy, and Mr. Mena Soliman), Adiga Life Sciences (Dr. Pascal L. C. Hickey), and Circassia Ltd. (Dr. Daniel R. Gliddon). I want to acknowledge the financial support from Adiga Life Sciences, the Allergy, Genes and the Environment Network for Centres of Excellence (AllerGen NCE), Circassia Ltd., Mitacs Accelerate, and the Prevention of Organ Failure (PROOF) Centre of Excellence. I also thank Canada and South Korea.   xxi Acknowledgement also goes to my colleagues, Dr. Amrit Singh is my beloved brother, best friend, and best colleague who have taught me bioinformatics and statistics with tremendous patience and consideration. I always appreciate his love and friendship that have shown to my family and me. He makes me feel that Canada is like my home country South Korea. Ms. Yolanda (Chen Xi) Yang and Ms. Amreen Toor are my good friends and colleagues. I remember the time we study together. Special thanks to Mr. Casey P. Shannon, Mr. Nick Fishbane, and Mr. Yunlong Nie for the wonderful introduction of R programming language. I also want to thank my lab members – Mr. Daniel He, Mr. Luka Culbrik, Ms. Karen Tam, and Ms. Carys A. Croft: Special thanks to my friends at St. Paul’s hospital – Dr. Loubna Akhabir, Ms. Beth Whalen, Ms. Basak Sahin, Ms. Sheena Tam, Dr. Sima Allahverdian, Dr. Aida Eslami, Dr. Anthony Tam, and Mr. David Ngan. I am thankful to my beloved mentors who are my buttresses – Dr. Yunhee Kim Kwon, Dr. Min Goo Lee, Dr. Woong Sun, and Dr. Seung Won Ra (South Korea); Dr. Jean-François Côté, Dr. Yves Berthiaume, Dr. André Dagenais and the late Michelle Harkness (Canada). I would also like to deeply thank my beloved friends, Ms. Mengxi Yang, Ms. Yan Wang, Ms. Leah Graystone, and Mr. Oh Hyoung Kwon. Finally, I deeply and sincerely thank and love my family. I love my late father Won Hyoung Kim, I believe you see me in heaven, mother Young Soon Choi, my wife’s father Tae Won Hwang, mother Tae Ok Choi, my wife So Jung Hwang, my kids Hyemin Kim and Hyunwoong Kim, my brother’s family (Sung Woong Kim, Mi Jung Park, and nephew Ji Woong Kim), my sister’s family (Jina Kim, Sungho Noh, and niece Mimi Noh), my wife’s sister Su Min Hwang, and my wife’s brother Junsik Hwang. And many thanks to my beloved congregation of the Church of the LORD’s Word and my precious and beloved people who I couldn’t mention in given limit pages.   xxii Dedication To God and my beloved people     1 Chapter 1: Introduction  1.1 Thesis overview             The goal of this thesis is to identify systemic immune response signatures associated with the pathophysiology of allergic rhinitis (AR) in peripheral blood samples. These systemic immune response signatures may provide a cross-sectional and alternative view of the pathophysiology of AR, in contrast to intensive studies on local inflammatory responses that are focused on the nose. We are interested in systemic immune responses to AR specifically as measured in peripheral blood, which includes information from blood vessels, the direct passage between the nose and the immune system. AR is associated with not only local inflammation in the nose, but also with the immune system (e.g., lymphoid organs and peripheral blood). To investigate the systemic immune response triggered by allergen challenge, we used two established human models of AR: the nasal allergen challenge (NAC, also known as nasal allergen provocation) model and the environment exposure unit (EEU) model. Chapter 1 introduces the background of these studies. Chapter 2 describes the main experimental methods used in our studies. Chapter 3 describes how we identified the systemic immune response signatures associated with the pathophysiology of allergic rhinitis in cat allergy. Chapter 4 describes validation studies to test the trends of systemic immune gene signatures identified in Chapter 3, using seasonal/intermittent allergic rhinitis (birch allergy and ragweed allergy) samples. Herein, we have used the term ‘validation’ as any empirically determined results that support our findings in an independent dataset. This term could also be construed as ‘replication.’ In particular, Chapter 4 describes the results of white birch pollen challenge in both human NAC and EEU allergic rhinitis models.    2   Figure 1-1 Outline of dissertation chapters, including experiments performed. We investigated systemic immune response signatures associated with the pathophysiology of AR using peripheral  blood collected after allergen challenge in this study.   3  Chapter 5 discusses the utility of the systemic immune gene signature approach by describing how the approach was used to investigate the mechanism of an allergen-specific immunotherapy. Chapter 6 summarizes all results of this study and describes its strengths and weaknesses as well as future directions for investigation. Figure 1-1 pictorially depicts the outline of this dissertation and the experiments involved.  1.2 Allergic rhinitis (AR) 1.2.1 Rhinitis             Rhinitis is characterized by one or more nasal symptoms: nasal congestion, nasal rhinorrhea, sneezing, and nasal itching. Rhinitis can be caused by allergic, infectious, hormonal, occupational, and other factors (1,2). The main categories of rhinitis are AR and non-allergic rhinitis (NAR). The ratio of AR to NAR in rhinitis is 3:1 (AR/NAR) (3–5). NAR can be infectious or noninfectious (6).             Vasomotor rhinitis (VMR, also known as idiopathic rhinitis) is the most prevalent cause of NAR. Common triggers for VMR include changes in the environment (temperature, barometric pressure, humidity, odors, tobacco smoke, drugs and ingestion of alcohol) and menstrual-related hormones, which result in a female subject predominance of VMR (ratio, female:male; 2:1). About 60% of patients with AR will develop nasal symptoms in response to non-allergic environmental triggers (2,5,7). 44-87% of patients with rhinitis may have mixed rhinitis, a combination of AR and NAR (3). 50% of the patients with persistent rhinitis are affected by AR based on clinical manifestations and a positive skin prick test (SPT) and/or serum specific IgE (sIgE) to airborne allergens (8). Although rhinitis is common in older adults due to   4 increasing NAR with aging, AR in older adults contributes to increased risk of stroke, asthma and hospital admissions for other pathologies (9). Rhinitis is a complex disease, which makes it difficult to understand the various aspects of its pathophysiology. It is an important causative factor of asthma, sinusitis, and otitis media with effusion, occurring when rhinitis causes a chronic state of nasal inflammation (2,10).  1.2.2 Allergic rhinitis             Allergic rhinitis (AR) is the most prevalent allergic disease worldwide, affecting up to 40% of the global population. AR affects between 10 and 30% of all adults and as many as 10-40% of children (2,3). Prevalence has increased progressively over the last three decades in industrialized societies, with 30% of the population in Europe and USA affected (7,11–14). AR affects approximately 20-25% of Canadians (15). Poorly/not controlled AR is one of the most common reasons for visits to the family doctor. The most often reported symptom of Canadian patients with AR is nasal congestion (69% of patients) followed by nasal rhinorrhea (52%) and sneezing (47%) (15,16).             AR is a symptomatic disorder of the nose induced after allergen exposure by an IgE-mediated inflammation of the nasal mucosa driven by Th2 cells (7,12). The most common allergic triggers for AR include pollens, fungi, dust mites, furry animals, and insect emanations (3).             AR is a relatively modern disease, being rarely described in antiquity (12–14). In 1929, AR was defined by three cardinal symptoms – sneezing, nasal obstruction, and mucous discharge (17). AR is characterized by early phase response (EPR) and, in some patients, a subsequent late   5 phase response (LPR). Although it also includes sneezing, congestion, and rhinorrhea symptoms similar to EPR, LPR is predominantly characterized by nasal congestion (3).  1.2.3 Intermittent allergic rhinitis (IAR) and persistent allergic rhinitis (PER)             Traditionally, AR was subdivided into perennial AR, seasonal AR, and occupational AR based on the type of allergen and occurrence of symptoms during the year. This classification is not entirely satisfactory, however, since: 1) some pollens and molds are perennial allergens in certain areas; 2) symptoms of some patients allergic to perennial allergens are seasonal; 3) the majority of patients are polysensitized; and 4) nonspecific irritants such as air pollution may affect the severity of AR (12,18,19). The Allergic Rhinitis and its Impact on Asthma (ARIA) world health initiative on allergic rhinitis therefore revised the classification. The new classification of ‘intermittent’ and ‘persistent’ AR is based on the duration of symptoms, not the type of allergen. Currently, both the traditional and new classification systems are in use (12,18,20).   1.2.4 The pathophysiology of AR             The phenotypic and functional heterogeneity of airway epithelial cells, gene-environment interactions, and various components of allergens may be associated with the different responses of nasal mucosa, such as development, severity, and limit resolution of allergic inflammation (21). The summary of the pathophysiology of AR are described in Figure 1-2.             The nasal cavity is protected from the outside environment by mucus and an epidermal barrier of nasal mucosa. Decreased antiprotease activity in nasal mucus of AR makes the barrier of the epidermis less protective against allergen protease attack (22). Epidermal barrier    6   Figure 1-2 The summary of the pathophysiology of AR.  dysfunction is also associated with gene mutations such as mutations in filament aggregating protein (FLG) gene, which encodes filaggrin, in some patients with allergic dermatitis (23).  Proteolytically active allergen directly and indirectly breaks the intercellular junctions of the epithelium, comprised of tight junctions, adherens junctions, and desmosomes (24). Increased epithelial permeability allows allergens to pass through the barrier and cause allergic responses. At the lumen of the epidermis, allergens can also be taken up by protrusions of dendrites on dendritic cells (DCs) such as Langerhans cells, inducing antigen-specific Th2 responses (24,25). A favourable microenvironment for DCs that supports maturation and activation of DCs is also an important factor for allergic sensitization by DCs. Activation of Toll-like receptor (TLR) -2, TLR-4, C-type lectin domain family 7 member A (CLEC7A, Dectin-1), and protease activated receptor 2 (PAR-2) receptors is important for the process (26).   7             The allergic responses of persons already sensitized to allergen are triggered when the allergen passes through the epidermal barrier. The responses are characterized by the EPR and LPR. The EPR occurs within 5-30 min after allergen challenge and is caused by an immediate IgE-mediated mast cell response. Subsequent LPR follows EPR in 3-75% of patients with AR, occurring after several hours and peaking 6-9 h after allergen exposure (20,21,27). Allergic responses are associated with polarized type 2 immune responses derived by Th2 cells, enhancing barrier defense at mucosal surfaces and inducing defense against parasites. The damaged epidermal barrier produces interleukin (IL) -25, IL-33, and thymic stromal lymphopoietin (TSLP), which induce activation of type 2 innate lymphoid cells (ILC2) cells and DCs. Secreted IL-5 and IL-9 by ILC2 cells recruit and activate mast cells and eosinophils. Degranulation of mast cells and basophils in the epidermal barrier is also an important allergic response occurring when allergens are recognized by IgE on the cells. The accumulated activated mast cells in the nasal mucosa and submucosa secret histamine, tryptase, leukotrienes, prostaglandins, and cytokines such as IL-4, IL-5, and IL-6, which are causing the allergic responses (28). Mast cells and basophils are the most relevant source of histamine release (28,29). TSLP also promotes the degranulation, differentiation, and proliferation of basophils (30–32). These processes lead to the recruitment of inflammatory cells as positive feedback. LPR is characterized by recruitment into the nasal mucosa of eosinophils, basophils, and lymphocytes such as B cells and Th2 cells (18,21). Nasal itching, sneezing,  and rhinorrhea are considered as irritative phenomena by neuroactive and vasoactive substances such as histamine, prostaglandin D2, and cysteinyl leukotrienes, but nasal congestion is predominantly related to mucosal inflammation (7,18). Th2 cytokines such as IL-4, IL-5 and IL-13 enhance the action of cysteinyl leukotrienes through up-regulation of the synthesis of cysteinyl leukotrienes and CysLT1   8 receptors on inflammatory cells such as mast cells, eosinophils, monocytes, and macrophages (7,33). The interaction of activated immune cells associated with the type 2 immune response plays a major role in causing the allergic inflammation that characterizes the pathophysiology of AR. The hypersensitive reactions are abnormal immune responses with a bias towards Th2 rather than Th1 (21).   1.2.5 Systemic disease characteristic of allergic rhinitis             Many studies on AR have focused on the local inflammatory site. However, investigations into the pathophysiological systemic immune responses of AR have been scant and the response has not been well described (34). It is important to study the systemic immune response of the pathophysiology of AR to not only better understand and diagnose AR, but also to examine the efficacy of AR treatments and investigate their mechanism of action. The potential utility of the systemic immune response approach is supported by the established utility of the SPT and/or measurement of serum sIgE response to airborne allergens in diagnosis of AR. AR is associated with allergic dermatitis (also known as eczema), and allergic asthma (21,35). Nasal symptoms are experienced by 78% of patients with asthma and asthma is experienced by 38% of patients with AR (36). Both AR and NAR are a strong risk factor for new-onset asthma (2). Associated allergic diseases reflect the systemic disease characteristic of AR regardless of whether the allergies are sequentially developed such as in allergic march or whether multiple allergies coexist (37). In other words, allergic diseases are not only an epithelial organ-specific response of the skin, lung, and nose, but also a systemic response. For example, complications of bone marrow transplantation (BMT), which mainly consist of graft-versus-host disease (GVHD) and immunodeficiency, demonstrate the role of the immune system in allergic conditions.   9 Following BMT from an allergic donor, non-allergic recipients have been reported to develop allergies. In contrast, a recipient’s allergic dermatitis was resolved after a BMT from a non-allergic donor (35,38,39).             On the other hand, local AR (LAR; also known as entopy) is an interesting phenotype of AR, characterized by a nasal Th2 allergic inflammatory response with local production of specific IgE (sIgE) and a positive response to the nasal allergen challenge (NAC; also known as nasal allergen provocation) test without evidence of systemic atopy assessed by conventional diagnostic tests such as SPT or measurement of sIgE in serum (40,41). Although it has a negative response to the SPT, LAR is associated with recruitment of immune cells such as mature B cells. LAR may be associated with the development of AR or allergic asthma with a positive response to the SPT or sIgE tests. Investigation of the pathophysiological systemic immune responses of AR may help understand the association between LAR and AR.   1.3 Human model of AR             To investigate systemic immune responses following allergen challenge in patients with AR, we used two established human models of AR: the nasal allergen challenge (NAC) and environmental exposure unit (EEU) models. These models are in many ways similar, but differ in terms of administered allergen dose and allergen exposure method (42–44).   1.3.1 Nasal allergen challenge (NAC) model             The nasal allergen challenge (NAC, also known as nasal allergen provocation) model is an established human model of AR used to measure the efficacy of AR treatments and investigate the pathophysiology of AR (45–49). The NAC model delivers a standardized allergen   10 directly and locally to a participant’s nasal mucosa through both nostrils. The administered allergen dose can be adjusted to the individual allowing all participants to achieve the required level of AR symptoms (49). The individual titration of allergen dose and easy allergen challenge method makes the NAC model more suitable (than the EEU model) for collection of clinical symptom scores as well as biospecimens (49–51). The NAC model is advantageous in pilot studies with a small sample size because it allows for collection of biospecimens, such as peripheral blood, after the allergen challenge.  1.3.2 Environmental exposure unit (EEU) model             The environmental exposure unit (EEU, also known as environmental challenge chamber) model was developed at Kingston General Hospital and Queen’s University in Kingston, Ontario, Canada in 1981 by Dr. James H. Day (42,52). The EEU was designed for highly and effectively replicating the study conditions of an outdoor study. The EEU is a room (21 x 15 x 3 m, or 924 m3) with a feeder system that continuously delivers controlled levels of allergen, such as pollen, into the seating area by fans. EEU allows for allergen specificity, control of antigen exposure levels, temperature, and air quality, and an accurate way of tracking self-reported and physician-observed symptoms for determining response to medication (42,53). Ragweed pollen (Ambrosia artemisiiflia), birch pollen (Betula pendula) and grass pollen (Dactylis glomerata) are typical allergens used for the EEU.  1.4 Identifying patterns in gene expression data             Transcriptional changes in genes are not always highly correlated with protein levels and specific cell frequencies because they are controlled by many complex steps. Nevertheless,   11 clustering of gene expression data correlated with corresponding cell frequency may allow for identification of traceable pattern signatures (54) which may provide more objective biomarkers for diagnosis, prognosis, and pathophysiological or pharmacological understanding. For example, although phenotyping using cell surface markers is difficult for classification of acute myeloid leukemia, the gene expression signature (based on consistent difference of signaling patterns between primitive and mature leukemia subpopulations, the mean expression of which is correlated with corresponding subpopulation frequency), is useful for prognosis of overall patient survival (55).              Models using time series gene expression data are appropriate to test and understand response and developmental processes in dynamic biological processes (54,56,57). Oscillations in gene expression can be detected by a statistical approach using a clustering method, the K-medoids algorithm. This method allowed for the identification and characterization of oscillating gene groups to infer a gene’s phase in a cell cycle study analyzing the profiles of single undifferentiated human embryonic stem cells (57). The other approach using time series gene expression data is a method that integrates static and time series data from multiple individuals to reconstruct condition-specific response networks in an unsupervised way (56). Identifying patterns in time series gene expression data is challenging, but necessary in order to understand hidden dynamic mechanisms.  1.5 Hypotheses             The main hypothesis of this dissertation is that if we investigate immune gene expression and immune cell frequencies in peripheral blood collected following allergen challenge in   12 subjects with allergic rhinitis (AR), we will identify systemic immune responses associated with the pathophysiology of AR.  1.5.1 Hypothesis of Chapter 3             Systemic immune response signatures are associated with the pathophysiology of AR, and can be determined from immune gene expression and immune cell frequencies measured in peripheral blood collected following allergen challenge in subjects with cat allergy.  1.5.2 Hypothesis of Chapter 4             The identified systemic immune response signatures in cat allergy will be validated in birch and ragweed allergies using peripheral blood collected after allergen challenge in NAC or EEU models.  1.5.3 Hypothesis of Chapter 5             Systemic immune response signatures following allergen challenge in a patient with allergic rhinitis receives immunotherapy will be differentially expressed in the peripheral blood samples collected pre- and post-immunotherapy. And the systemic immune response signature changes will provide a cross-sectional view to further investigate the mechanism of action of the immunotherapy.   13 Chapter 2: Main experimental methods  2.1 Human model of allergic rhinitis (AR) 2.1.1 Nasal allergen challenge (NAC) model             We used the nasal allergen challenge (NAC) method, which is a standardized protocol to study the pathophysiology of AR as well as the mechanism of action and therapeutic effects of AR treatments by the Allergic Rhinitis – Clinical Investigator Collaborative (AR-CIC), formed with funding from the AllerGen Network of Centres of Excellence (AllerGen NCE) (49). The NAC was implemented during two visits – the screening visit and a NAC visit. The NAC visit followed the screening visit by at least six days to allow a sufficient wash-out period (Figure 2-1). Participants with nonspecific nasal hyper-responsiveness were identified and excluded using an initial nasal wash with 5 ml of 0.9% saline before allergen challenge at the screening visit. After the nasal wash, 100 µl of serially diluted allergen (from 2048-fold to 2-fold dilution, using a 4-fold dilution factor between dilutions, at a time interval of 15 min) was sprayed into each nostril using the Aptar Bi-dose device (Aptar Pharma, New York, USA) starting with the lowest concentration. The qualifying concentration was the concentration that first met the criteria (which differed depending on the study site): a TNSS ≥ 8/12 and/or a PNIF reduction of ≥ 50% from baseline.             At the NAC visit, the allergen challenge was conducted once in the morning. The allergen dose administered was determined differently depending on the study site, and evaluated either as the cumulative dose (from the lowest to the qualifying concentration dose) or the qualifying concentration dose (based on 100 µl of the qualifying concentration).    14  Figure 2-1 NAC model.  A, screening visit and NAC visit. In NAC model, each subject undergoes NAC visit to collect clinical symptom scores and biospecimens such as peripheral blood after determining a successful NAC dose at the screening visit.  B, time points of recording clinical symptom scores (PNIF and TNSS) at NAC visit.              Healthy non-allergic subjects underwent a similar NAC protocol except the allergen dose used was different: two times diluted stock allergen (the maximum allergen dose used in the NAC model).  2.1.2 Environmental exposure unit (EEU) model             The EEU experimental setting at Queen’s University is described in the literature (52,58). The EEU is a room (21 x 15 x 3 m, 924 m3) with a feeder system that continuously delivers controlled levels of allergen, such as pollen, into the seating area, distributed by fans (Figure 2-2). Briefly, at the allergen (herein, pollen from white birch or short ragweed) exposure visit, all subjects were continuously exposed to 3500 ± 500 grains/m3 concentration of pollen (Greer     15  Figure 2-2 Layout of the Environmental Exposure Unit (EEU) (52).  Laboratories, Lenoir, North Carolina, USA) for 3 or 4 hours. To keep the concentration of the pollen consistent, 7 Rotorod samplers (Sampling Technologies Inc, Minnetonka, Minnesota, USA) positioned throughout the subject seating area were used to monitor pollen levels every 30 minutes   16  Figure 2-3 EEU model.  A, allergen exposure visit. B, time points of collecting clinical symptom scores (PNIF and TNSS) at allergen exposure visit.  and to adjust the pollen levels as required. Subjects with AR and healthy non-allergic subjects underwent the pollen exposure in the same EEU on the same allergen exposure visit day (Figure 2-3).  2.2 Clinical symptom scores             Clinical symptom scores following allergen challenge were measured using total nasal symptom score (TNSS) and peak nasal inspiratory flow (PNIF). In the NAC model, clinical symptom scores from all subjects were collected prior to NAC (baseline, 0 h), and additionally at      17 15 min, 30 min, and every hour from 1 h to 12 h post-NAC (Figure 2-1). In the EEU model, clinical symptom scores from all subjects were collected prior to allergen exposure (baseline, 0 h), every 30 min during allergen exposure, and every hour from the end of allergen exposure to 12 h after the start of allergen exposure.  2.2.1 Total nasal symptom score (TNSS)             TNSS is a subjective assessment of clinical symptoms of allergic rhinitis. Participants were asked to record their total nasal symptom score, which consists of four categorized nasal symptoms recorded on diary cards (49): runny nose, nasal congestion, sneezing and nasal itching. Each symptom was scored from 0 to 3, as follows: 0: absence; 1: mild; 2: moderate and bothersome; 3: severe and intolerable. The symptom diary card was designed for automatic scanning and reading using Optical Mark Recognition (OMR), which is the process of capturing human-marked data from the document form, to allow for automated data entry into the system.  2.2.2 Peak nasal inspiratory flow (PNIF)             Participants were asked to record their baseline PNIF using the ‘In-Check’ inspiratory flow measurement device (Clement Clarke International Ltd., Essex, UK), which allows for measure of nasal inspiratory flow between 30-370 L/min. The ‘In-Check’ device works as follows: when a participant inhales through the nose, causing air to be drawn through the device, a cursor on the meter moves along a scale to indicate the speed of inhalation. The position of the cursor against the calibrated scale of the meter indicates the flow rate. The maximum of three measurements at each time point represents the PNIF value at the time point. PNIF provides an objective assessment of nasal airway patency (59).    18  2.3 Complete blood cells measurement with 4-part differential blood cells counting (CBC)             CBC measurement with differential at NAC visit was generated using automated hematology analyzers (Queen’s University: Sysmex XE-2100TM; McMaster University: Sysmex XN-3000TM, Sysmex, Kobe, Japan).  2.4 Gene expression profiling             We used the NanoString nCounter gene expression assay and Affymetrix microarray to profile transcripts of peripheral blood.  2.4.1 NanoString nCounter gene expression assay             The NanoString nCounter gene expression assay is a sensitive and highly reproducible technique (60–62). The NanoString nCounter PanCancer Immune Profiling Panel (NanoString, Seattle, WA, USA) was used to profile 730 canonical immune genes from peripheral blood. We tested the sample type (RNA and lysate) and prep sensitivity options (Normal and High) of the assay. We also demonstrated how batch effects, which can result when different lots of CodeSets and master kit are used, can be corrected using a bioinformatics tool.  2.4.1.1 Method             PAXgene Blood RNA tubes (PreAnalytiX, Hombrechtikon, Switzerland) collected at the NAC visit in the NAC model or the allergen exposure visit in the EEU model were stored at -80°C after preprocessing (incubation at room temperature over 2 h) as per the manufacturer’s protocol.   19 The lysates and RNA were extracted from 5 ml of each PAXgene Blood RNA sample tube using the PAXgene Blood miRNA kit (PreanalytiX).              The cellular lysates and RNA were profiled using NanoString methodology based on the manufacturer’s protocol. In brief, 4 ml of each lysate or 100 ng of RNA were applied to the NanoString nCounter PanCancer Immune Profiling Panel to measure differential mRNA expression of 770 genes (730 immune genes and 40 housekeeping genes). CodeSets comprise 50 nt-sized Reporter probes and Capture probes of the target genes. The lysate or RNA was mixed with the Reporter probes and Capture probes for hybridization at 65°C for 18 h. The hybridized samples were processed using the nCounter Prep Station using the Normal Sensitivity or the High Sensitivity protocols and followed by quantitative detection by nCounter Digital Analyzer using the maximum resolution (MAX FOV). The raw data of the NanoString assay was normalized using nSolver Analysis software (v2.6 or v3.0) using positive controls and housekeeping genes. Normalization was processed using the geometric mean method: positive control (excluding the lowest concentration, POS_F), and housekeeping genes (excluding lower (count < 32) expressed genes: CC2D1B, GPATCH3, and ZKSCAN5). After normalization, log2 transformed data were used for further analysis.  2.4.1.2 Adjusting setting 2.4.1.2.1 Sample type             While gene expression profiling assays such as real-time polymerase chain reaction (PCR), microarray, and next-generation sequencing (NGS) usually use RNA only, the NanoString nCounter gene expression assay allows the use of not only RNA but also cellular    20  Figure 2-4. Pearson correlations of technical replicates to compare lysate and RNA sample types from a PAXgene RNA blood tube of Sample 1.  The x-axis reference line of scattering plot is log2 transformed counts of gene expression of the up-side sample. The y-axis reference line of scattering plot is log2 transformed counts of gene expression of the right-side sample.    21  Figure 2-5 Bland-Altman plot of samples.  A, comparison of Lysate and RNA samples in a PAXgene RNA blood sample. B, comparison of technical replicates of a lysate sample.   22 lysates. Using the Pearson correlation, we tested the linear relationship between lysate and RNA of the same sample. Lysates and RNA in technical replicates had strong correlations (r ³ 0.987) (Figure 2-4). The difference of gene expression between lysate and RNA in a PAXgene RNA blood sample is shown in Figure 2-5A.             A technical replication test in a lysate sample demonstrated a strong correlation (r = 0.990), and the difference of gene expression is shown in Figure 2-5 B. We decided to use lysates instead of RNA because lysates can be extracted more easily and quickly. An efficient method with less complexity and that is less time consuming may have higher utility for diagnosis and investigation.  2.4.1.2.2 Reproducibility             The NanoString nCounter gene expression assay had high reproducibility. The correlations between technical replicates of lysates or RNA were high (r > 0.9) (Figure 2-4 and Figure 2-8).  2.4.1.2.3 Sample preparation sensitivity             The NanoString nCounter gene expression assay provides two options to process translocated hybridized samples to a cartridge with different incubation time. The normal sensitivity protocol, which gives less 30 min incubation time in the cartridge than the high sensitivity protocol, had a strong correlation (r > 0.9) with the high sensitivity protocol in a technical replication test (Figure 2-6). The difference of gene expression between the protocols is shown in Figure 2-7. We decided to use the high sensitivity protocol having a longer incubation time.   23    Figure 2-6. Comparison between normal and high sensitivity protocols in nCounter prep station.  The number and the plot at crossed position between the protocols are their Pearson correlation and the scatter plot of the expression data.    24  Figure 2-7 Bland-Altman plot of comparison of normal and high sensitivity protocols.  2.4.1.3 Removing batch effects             The NanoString assay has batch effects or non-biological experimental variation across different lots of CodeSets and master kit. Batch effects make samples in different batches not directly comparable (63).              The cat allergy subjects who underwent Cat-PAD treatment (as described in Chapters 3 and 5) were involved in two projects. The first project compared samples using a 6 h post-NAC   25 interval between pre-treatment (V1A) and post-treatment (V3). The second project included the other time points (baseline, 1 and 2 h post-NAC). Both projects used the PanCancer Immune Profiling Assay, but the batches of the assay were different between projects. The batch effects between projects were demonstrated using a principal component analysis (PCA) plot (Figure 2-9). The batch effects were corrected by an established batch effect correction method, combatting batch effects (ComBat) (Figure 2-8 and Figure 2-9). The difference between before and after the correction is shown in technical replicates (Figure 2-8). ComBat, based on parametric and non-parametric empirical Bayes (EB) frameworks for adjusting data to remove batch effects, is robust to outliers in small sample sizes and performs comparably to existing methods, such as methods for location and scale adjustments for large samples (63). The assumption of the EB method is that phenomena resulting in batch effects affect many genes in a similar way; for example, many genes may demonstrate increased expression and higher variability (63,64). Eliminating confounding variables by batch effects correction using ComBat is important to reduce bias when investigating gene expression data (65).     26  Figure 2-8 Correlation comparison before and after batch effect correction.  A, replicates of one sample collected at V1A to compare the difference before and after batch effect correction intra-project and inter-project. B, replicates of one sample collected at V3 to compare the difference before and after batch effect correction intra-project and inter-project.     27  Figure 2-9 PCA plot before and after batch effects correction.  2.4.2 Microarray             RNA from PAXgene blood RNA tubes were profiled using Affymetrix Human Gene 1.0 ST (Affymetrix, Santa Clara, CA, USA), which provides genome-wide expression profiling through measurement of protein coding and long intergenic non-coding RNA transcripts. RNA labeling and array hybridization were implemented by the Centre for Translational and Applied Genomics at the BC Cancer Agency (Vancouver, BC, Canada). Raw data of the microarray results were normalized using Robust Multi-array Average (RMA), a typical normalization method (66). After RMA normalization, log2 transformed data were used for further analysis.     28 2.5 Clustering of profiled gene expression             Using statistical and bioinformatics tools, we identified reproducible systemic immune gene signatures associated with immune cell frequencies and clinical symptom scores.  2.5.1 Filtering immune genes for clustering             Fuzzy c-mean (FCM) clustering, a widely used method to discover significant patterns in a given data set, was used to partition the gene expression data into clusters. We chose FCM over other clustering methods, such as K-means and hierarchical clustering, because FCM allows a more exploratory approach to find natural boundaries in data (67,68). To solve the challenge of FCM – sensitivity to noise and outliers of a given data set (67) – we applied two filtering procedures to our data before clustering: i) a statistical test using a linear mixed effect model (LME) and ii) a fold-change threshold. The first filter procedure is to select genes significantly (BH-FDR < 0.1) differentially expressed at a time point post-allergen challenge compared to baseline, considering missing data and subject differences as a nested random effect. The second filter procedure is to select genes that have greater than ± 1.2 fold change between minimum and maximum of expression value over time points measured.  2.5.2 Fuzzy c-means (FCM) clustering              FCM clustering was applied to immune genes retained after two filtering procedures using the R statistical computing program and packages nlme [ver. 3.1-128], (Pinheiro J, Bates D, DebRoy S, Sarkar D and R Core Team, 2016, nlme: Linear and Nonlinear Mixed Effects Models. R package version 3.1-128, http://CRAN.R-project.org/package=nlme) and Mfuzz [ver. 2.30.0] (69). The condition of FCM clustering (mfuzz function of Mfuzz package) was 1.8   29 fuzzification parameter (m = 1.8) and 7 cluster centroids (c=7). The function mfuzz used the FCM algorithm based on minimization of a weighted square error function.  2.5.3 Enrichment analyses             To test the association between clusters and immune cell types, cell enrichment analyses were performed using Enrichr, a web-based, mobile software application (70,71), and Simple Enrichment Analysis in R (SEAR, https://www.github.com/cashoes/sear), which has detailed cell subsets. Pathway enrichment analysis using identified significant genes was also performed using Enrichr.  2.5.4 Statistical analysis             Statistical analyses were performed using the R statistical computing program (Packages: nlme [ver. 3.1-128], GeneOverlap [ver.1.6.0], compareGroups [3.2.4] and mixOmics [ver.6.0.0]). Linear mixed effects models were used for the comparison between baseline and each time point after NAC in clinical symptoms, CBC and gene expression data. The similarities of clusters between cohorts were tested using the Fisher’s exact test (‘testGeneOverlap’ function of GeneOverlap package[ver.1.6.0], https://www.github.com/shenlab-sinai/geneoverlap). A p-value of less than 0.05 was considered statistically significant. The Benjamini-Hochberg false discovery rate (BH-FDR) method was also used.       30 Chapter 3: Systemic immune response signature in peripheral blood from subjects with allergic rhinitis undergoing nasal allergen challenge  3.1 Sub-abstract             Our hypothesis was “Systemic immune response signatures are associated with the pathophysiology of Allergic rhinitis (AR), and can be determined from immune gene expression and immune cell frequencies measured in peripheral blood collected following allergen challenge in subjects with cat allergy.” We identified significantly differentially expressed 7 clusters of canonical immune genes in 9 subjects with cat allergy (Queen’s University study subjects, or Q cohort) using Fuzzy c-means (FCM) clustering method. The clusters were reproduced in the other 9 subjects with cat allergy (McMaster University study subjects, or M cohort). Cluster 2, 3, and 4 were respectively associated with neutrophils, neutrophil/lymphocyte ratio (NLR), and lymphocytes in cell enrichment analyses of each gene set and canonical correlation analysis between cluster expression patterns and the frequency patterns of the corresponding immune cells. The systemic immune response signatures, namely immune gene expression patterns associated with their corresponding immune cell frequencies, after the allergen challenge in allergic subjects had modest correlations with clinical symptoms. These systemic immune response signatures were not reproducible in healthy non-allergic subjects.  3.2  Introduction             Allergic rhinitis (AR) is a local nasal inflammatory disease, but systemic immune responses also play an important role in its pathophysiology. Understanding the mechanisms of   31 the systemic immune system in AR responses may provide insight into not only the pathophysiology of AR but also the mechanism of action of AR treatments. To examine the systemic immune responses of AR, we designed an approach using peripheral blood collected in a human AR model. Peripheral blood was collected from AR patients undergoing a NAC in the Allergic Rhinitis – Clinical Investigator Collaborative (AR-CIC) project, part of the Allergy, Genes and the Environment Networks for Centres of Excellence (AllerGen NCE), which has developed a standard NAC protocol (49). The collected time series data in the NAC model is useful for investigating dynamic immune responses after allergen challenge. Here, we examined the hypothesis that immune gene sets clustered by gene expression pattern in peripheral blood from subjects with AR undergoing NAC are associated with immune cell frequencies as a systemic immune signature of the pathophysiology of AR.  3.3 Materials and methods 3.3.1 Study approval             The Queen’s University study (NCT01383590), or Q cohort, and McMaster University study (NCT01383603), or M cohort, were registered on clinicaltrials.gov. The Q cohort study (Study Code: DMED-1423-11) and healthy control study (DMED-1343-10) were granted ethical clearance by the Queen’s University Health Sciences and Affiliated Teaching Hospitals Research Ethics Board at Queen’s University; the M cohort study (Research Project #11-3551) was cleared by the Hamilton Integrated Research Ethics Board at McMaster University.    32 3.3.2 Subjects             This study was performed in 18 AR subjects and 5 healthy non-allergic subjects who underwent NAC. Samples of 9 AR subjects (Queen’s University study subjects, or Q cohort) and 5 healthy non-allergic subjects were from Queen’s University. Samples of another 9 AR subjects (McMaster University study subjects, or M cohort) from McMaster University were used for validation of our systemic immune response signature approach (Table 3-1 and Table 3-2). The correlation analyses between clinical symptom scores and identified immune signature were performed in 13 of 18 AR subjects who had no missing data (Table 3-2).             In AR subjects, inclusion criteria included a minimum 1-year documented history of AR on exposure to cats and positive skin prick test to cat allergen with a wheal diameter at least 3 mm larger than that produced by the negative control.              In healthy non-allergic subjects, inclusion criteria included a history of non-allergy to any aeroallergens and a negative skin prick test to all of the common aeroallergens tested.   33 Table 3-1. Characteristics of the Q and the M cohorts (Fisher's exact test). Q cohort M cohort p value Sex     < 0.05      Men 0 5       Women 9 4   Age (year, mean ± SD) 35 ± 8 40 ± 12 > 0.05 BMI (kg/m2, mean ± SD) 26.8 ± 4.8 26.6 ± 3.4 > 0.05 Race   > 0.05      Caucasian 8 7        Asian 1 2  Study site (NAC model experiment) Queen’s University McMaster University  The criteria of qualifying concentration for successful NAC at screening visit  i) a PNIF reduction of ≥ 50% from baseline      ii) a TNSS ≥ 8/12 And Or  Allergen dose administrated at NAC visit cumulative dose qualifying concentration dose  Allergen dose administered at NAC visit (BAU, mean ± SD) 408.5 ± 185.7 67.7 ± 55.1 < 0.05   34             Participants were excluded if they had a diagnosis of asthma, a history of anaphylaxis to cat allergen, an FEV1 < 80% of predicted, an FEV1/ FVC ratio < 0.7, vital signs (blood pressure,  pulse rate, respiratory rate, body temperature) that were outside normal limits, significant history of alcohol or drug abuse, any history of vasovagal reaction in response to needles or blood donation, a history of any significant disease or disorder, or if the subject was a smoker or quit smoking less than 3 months prior to the screening date.  Table 3-2. Characteristics of the 13 AR subjects and the 5 healthy non-allergic subjects (Fisher’s exact test).   13 AR subjects 5 healthy non-allergic subjects p value Sex     > 0.05      Men 2 3       Women 11 2   Age (year, mean ± SD) 36 ± 9 31 ± 8 > 0.05 BMI (kg/m2, mean ± SD) 26.5 ± 4.1 29.2 ± 6.1 > 0.05 Race   > 0.05      Caucasian 11 4        Asian 2 1    3.3.3 Nasal allergen challenge (NAC)             The Q and the M cohorts underwent a similar NAC protocol except for the allergen dose administered at the visit (Table 3-1). The NAC method is described in our previous paper (49). NAC was implemented at two visits – the screening visit and the NAC visit (Figure 3-1).     35  Figure 3-1 Diagram of an NAC model.  A, individual NAC model test procedure from screening visit to NAC visit. B, a schedule of collecting peripheral blood samples and clinical symptoms before and during NAC onset in both Q and M cohorts; in healthy non-allergic subjects, blood samples were collected at two time points (before NAC and 1 h post-NAC).              The allergen administered at both study sites was a standardized cat allergen extract (10,000 bioequivalent allergen unit [BAU]/ml, ALK-Abello Pharmaceuticals Inc., Hørsholm, Denmark) with the same DIN (02235299) and lot number (ID0142). After the nasal wash, 100 µl of serially diluted allergen (from 4.9 to 5000 BAU/ml, using a 4-fold dilution factor between dilutions, at a time interval of 15 min) was sprayed into each nostril using the Aptar Bi-dose device (Aptar Pharma, New York, USA) starting with the lowest concentration. The qualifying concentration was the concentration that first met the criteria (Table 3-1).              At the NAC visit, the allergen challenge was given once in the morning (before 10 a.m.). The allergen dose administered was determined differently depending on the study site, and evaluated as the cumulative dose (from the lowest to the qualifying concentration dose) in the Q   36 cohort or the qualifying concentration dose (based on 100 µl of the qualifying concentration) in the M cohort.	            Healthy non-allergic subjects underwent a similar NAC protocol except for the specific allergen: two times diluted stock birch allergen (39,000 protein nitrogen unit [PNU]/ml, ALK-Abello Pharmaceuticals Inc., Hørsholm, Denmark), which was the maximum allergen dose used in the NAC model.  3.3.4 Missing Data             Clinical symptom scores and blood were collected during NAC. Clinical symptom scores from all subjects were collected prior to NAC (baseline, 0 h), and at 15 min, 30 min, and every hour from 1 h to 12 h post-NAC. Whole peripheral blood from AR subjects was collected in EDTA blood tubes and PAXgene blood RNA tubes at 4 time points: prior to NAC (baseline, 0 h), and 1, 2, and 6 h post-NAC (Figure 3-1). Equivalent blood specimens were collected from healthy non-allergic subjects at two-time points (prior to NAC, and 1 h post-NAC); we were unable to obtain the additional blood samples at 2 h and 6 h post-NAC due to these healthy subjects being part of a different study. However, the 1 h post-NAC blood sample represented a convenient comparator for some of our gene signature analyses.              Clinical symptom and blood data from AR subjects was partially missing due either to failed collection or no procedure for collecting. These missing data included: in clinical symptom scores, the M cohort (1 subject 0-12 h post-NAC in PNIF, 1 subject 0-6 h post-NAC in PNIF, 1 subject 7-12 h post-NAC in PNIF, 2 subjects 7-12 h post-NAC in TNSS); in CBC data, the Q cohort (1 subject 2 and 6 h post-NAC) and the M cohort (1 subject 0, 1, 2 and 6 h post-NAC); in   37 gene expression data, the Q cohort (1 subject 2 and 6 h post-NAC) and the M cohort (1 subject 2 h post-NAC).  3.3.5 Canonical correlation             The canonical correlation of immune cell frequencies in CBC and the cluster centres in gene expression data was calculated by partial least squares regression (PLS) using the mixOmics package [ver.6.0.0] (72). The cluster centres are arithmetic means of the standardized expression of genes of each cluster at each subject’s repeated measurements.  3.4 Results 3.4.1 Investigation of the clinical symptom scores during NAC             Both Q and M cohorts had moderate negative correlations (Pearson correlation, r < -0.5) between PNIF and TNSS, and experienced peak TNSS or minimum PNIF score at 15 min post-NAC (Figure 3-2, A and B). A reduction of PNIF means difficulty of breathing through the nostrils. An increase of TNSS indicates the occurrence of AR symptoms (runny nose, nasal congestion, sneezing and nasal itching). PNIF of the Q cohort decreased significantly from baseline at 15 min to 5 h, and 9 to 10 h post-NAC. This cohort experienced a significant increase of TNSS at 15 min to 4 h, and 10 h post-NAC. PNIF of the M cohort significantly decreased at 15 and 30 min post-NAC, and participants experienced a significant increase of TNSS at 15 min to 1 h post-NAC.             Both cohorts used a different decision process to determine the cat allergen doses administered at the NAC visit. The fewer significant changes in the M cohort may primarily be related to the allergen doses used, which were significantly (p < 0.05) (six times) lower than in    38  Figure 3-2 Clinical symptom scores after NAC.  A, clinical Symptom scores, and Pearson correlation (r) of PNIF and TNSS in the Q cohort (n = 9). B, clinical Symptom scores, and Pearson correlation of PNIF and TNSS in the M cohort (n = 9, partial missing data). Error bars: mean ± SEM, *p < 0.05, Linear mixed effect model (LME).  the Q cohort (Table 3-1). Previous studies have demonstrated a relationship between the incidence of the stronger AR response (such as LPR) and higher allergen dose (73,74).   3.4.2 Investigation of immune cell frequencies and neutrophil/lymphocyte ratio during NAC             In CBC data of the Q cohort, leukocytes and monocytes significantly (p < 0.05) increased at 2 h and 6 h post-NAC, while platelets significantly increased 6 h post-NAC (Figure 3-3 A). In the M cohort, leukocytes significantly increased at 1 h and 2 h post-NAC, while platelets significantly increased at 1 h and 6 h post-NAC. Monocytes showed no significant change at any time point post-NAC compared to baseline (Figure 3-3 B). Both cohorts displayed similar    39  Figure 3-3 CBC after NAC.  A, CBC in the Q cohort (n = 9, partial missing data). B, CBC in the M cohort (n = 8). C, NLR in the Q (n = 9, partial missing data) and M (n = 8) cohorts. Error bars: mean ± SEM. *p < 0.05, LME.  patterns in three leukocyte subtypes and NLR: neutrophils increased at 1 h and 2 h post-NAC; eosinophils decreased at 1 h and 2 h post-NAC; lymphocytes increased at 6 h post-NAC; and, NLR increased at 1 h and 2 h post-NAC (Figure 3-3).    40 3.4.3 Seven clusters of immune gene expression patterns during NAC             We sought signatures of immune gene transcripts of blood associated with immune cell frequencies in order to identify distinct pathophysiological systemic immune responses of AR.          To investigate the signatures using a clustering method, we profiled 730 canonical immune genes using a NanoString nCounter assay on blood samples that were sequentially collected. We used two filtering procedures, a linear mixed effect model (LME) and a range threshold, to select significantly expressed immune genes after NAC. Using LME and considering missing data and subject differences as a nested random effect, we identified 224 genes as significantly (BH-FDR < 0.1) different when compared between baseline and other time points post-NAC. One hundred and twenty of these 224 genes were retained after applying a cut-off filter (> ± 1.2 fold-change) over time points in mean gene expression of all the subjects in the Q cohort before clustering. Fuzzy c-means (FCM) clustering identified seven clusters based on the standardized (mean: 0, SD: 1) mean gene expression pattern over the examined time points (Figure 3-4 A).              To validate the FCM clustering, we independently applied the analysis to the M cohort. The seven clusters were reproducible in the M cohort (Figure 3-4 B). A total of 207 genes from the profiled 730 immune genes were filtered and grouped into the seven clusters in the M cohort. For both cohorts, Clusters 1, 2, and 3 displayed an increasing expression at 1 h and 2 h post-NAC; Clusters 4 and 5 showed a maximum change at 6 h post-NAC; and Clusters 6 and 7 showed a decreasing expression at 1 h and 2 h post-NAC (Figure 3-4, A and B, and Appendix A.1). The seven clusters in each cohort were tested for similarities using the Fisher’s exact test on 85 shared genes between 120 genes of the Q cohort and 207 genes of the M cohort (Table II). Five clusters had significant (BH-FDR < 0.01) similarities: Q2/M2, Q3/M3, Q4/M4, Q6/M6, and Q7/M7. This indicates the five clusters had strong similarities between Q an M cohorts.    41  Figure 3-4 Seven clusters of differentially expressed immune genes in blood after NAC.  A, seven clusters in the Q cohort (n=9, partial missing data). B, seven clusters in the M cohort (n=9, partial missing data). C, cell enrichment analyses of the seven clusters.  42 Table 3-3 Similarities between clusters from both cohorts:  The Q and the M cohorts. Numbers in parentheses indicate total number of shared genes (Fisher’s exact test). BH-FDR (Overlapped genes) Shared 85 genes of 120 genes (Q cohort) 3 genes 19 genes 23 genes 8 genes 16 genes 6 genes 10 genes Cluster Q1 Cluster Q2 Cluster Q3 Cluster Q4 Cluster Q5 Cluster Q6  Cluster Q7 Shared 85 genes of 207 genes (M cohort)  0 gene Cluster M1 1 1 1 1 1 1 1 19 genes Cluster M2 0.68 (2) 1.23x10-6 (14) 1 (3) 1 1 1 1 36 genes Cluster M3 1 1 (4) 9.26x10-5 (19) 1 0.03 (12) 1 1 (1) 7 genes Cluster M4 1 1 1 1.07x10-5 (6) 1 1 1 9 genes Cluster M5 1 1 (1) 1 (1) 1 0.42 (4) 1 0.42 (3) 9 genes Cluster M6 1 (1) 1 1 0.98 (2) 1 2.16x10-4 (5) 1 (1) 5 genes Cluster M7 1 1 1 1 1 1 9.42x10-5 (5)     43  Figure 3-5 The relationship between clusters and immune cell frequencies.  A, canonical correlation between cell frequencies and clusters in the Q cohort (n = 9, partial missing data). B, canonical correlation between cell frequencies and clusters in the M cohort (n = 8). Partial least squares regression was used.  3.4.4 Relatedness of the identified clusters and immune cells             To test the association between clusters and immune cell types, cell enrichment analyses were performed using Enrichr, a web-based, mobile software application (70,71), and Simple Enrichment Analysis in R (SEAR, https://www.github.com/cashoes/sear), which has detailed cell subsets. Both analyses demonstrated similar results (Figure 3-4 C). Cluster Q2/M2 was primarily associated with myeloid cells, such as neutrophils. Cluster Q3/M3 was also associated with myeloid cells, such as neutrophils and monocytes. Cluster Q4/M4 was associated with B cells and T cells. Cluster Q6/M6 was associated with T cells.             We asked whether the clusters were correlated with the frequencies of the corresponding immune cells, which were demonstrated on the cell enrichment analyses, after NAC. To identify the correlations, we calculated the canonical correlations of immune cell frequencies with the cluster centres using partial least squares regression (Figure 3-5, A and B). Cluster Q2/M2 and Q3/M3 were strongly positively correlated to neutrophil count and NLR (canonical correlation > 0.75, Q2 and NLR = 0.69) (Figure 3-5, A and B). Cluster Q4/M4 was positively correlated   44 (canonical correlation, Q4 = 0.86, M4 = 0.54) with lymphocyte count (Figure 3-5, A and B). Interestingly, the M cohort, whose Cluster M4 had weaker correlation with lymphocyte frequency than Cluster Q4, had no significant LPR in the clinical symptom scores analysis (Figure 3-2, A and B).  3.4.5 Correlation of clinical symptoms and clusters or immune cell frequencies in 13 AR subjects.             After removing subjects who had missing data, we used 13 AR subjects (Q cohort: 8, M cohort: 5; Sex: 2 Male, 11 Female, Table 3-2) to investigate the association of clinical symptoms and the identified systemic immune response patterns (frequencies of neutrophils and lymphocytes, NLR and clusters 2, 3 and 4): cluster 2 was associated with the TNF-related apoptosis-inducing ligand (TRAIL) and Platelet-derived growth factor receptor-β (PDGFR-β) signaling pathway, and clusters 3 and 4, with IL-4– mediated signaling events in pathway enrichment analysis (BH-FDR < 0.1, Enrichr). When the Pearson correlations (r) were calculated, the original values and the ratios (relative numbers at each time point post-NAC compared to the baseline value) of the variables were used (Figure 3-6 and Appendix A.2). A sum of clinical symptom scores in EPR (baseline to 6 h post-NAC), LPR (7 to 12 h post-NAC) or over all examined time points was used because of different time intervals between clinical symptoms and systemic immune response patterns: we separated EPR and LPR periods based on the change of the mean values of clinical symptoms.    45  Figure 3-6 Correlations between clinical symptoms and systemic immune responses in 13 AR subjects.  The correlations (r > +0.5 or r < -0.5) between sum of clinical symptom scores (TNSS, PNIF, and PNIF.Ratio) and immune cell frequencies (lymphocytes, neutrophils, and NLR; lymphocytes.Ratio, neutrophils.Ratio, and NLR.Ratio) or immune gene clusters (clusters 2, 3, and 4) were demonstrated (n = 13). Sum.ALL: a sum of scores over all measured time points (baseline to 12 h post-NAC in clinical symptoms; baseline to 6 h post-NAC in immune cell frequencies); Sum.EPR: a sum of scores over time points in EPR (baseline to 6 h post-NAC); Sum.LPR: a sum of scores over time points in LPR (7–12 h post-NAC). Error bars: mean ± SEM.   46             The Pearson correlations between the sum of TNSS and the sum of PNIF or the sum of the ratio of PNIF were negligible (r < 0.3 or r > -0.3, A.2). This may reflect the fact that PNIF measures just one of the four symptom categories of TNSS; although PNIF is an objective measurement, it is unlikely to represent overall symptoms after NAC.             The sum of TNSS in EPR was moderately associated with the lymphocyte ratio (values at 2 h post-NAC: r = 0.51, the sum of all-over time points: r = 0.55) and the NLR ratio at 6 h post-NAC (r = -0.61). The sum of TNSS over all time points was moderately associated with the NLR ratio at 6 h post-NAC (r = -0.51). The sum of PNIF in LPR was moderately associated with neutrophil at 2 h post-NAC (r = -0.51). The sum of the ratio of PNIF in LPR was moderately associated with NLR at 6 h post-NAC (r = 0.67, Figure 3-6).              The immune genes shared in the same clusters of both cohorts were used for the correlation calculation. The geometric mean of the ratio of cluster 3 (19 genes in cluster Q3/M3, Table 3-3) at 6 h post-NAC was moderately (r = -0.55) associated with the sum of TNSS in EPR (Figure 3-6).             While neutrophil and lymphocyte frequencies were positively associated with the intensities of EPR or LPR, NLR and its ratio at 6 h post-NAC were negatively associated with the severity of clinical symptoms (Figure 3-6). In other words, considering NLR, higher lymphocytes at 2 h and 6 h post-NAC may be associated with the severity of AR; neutrophil count at 2 h post-NAC was also positively related with the severity of AR, but inversely related at 6 h post-NAC. NLR at baseline has been suggested to be an indicator of inflammation and severity of AR (75,76).   47 3.4.6 Comparison of immune gene signature patterns between 13 AR subjects and 5 healthy non-allergic control subjects             Five healthy non-allergic subjects whose blood was collected at two time points (baseline and 1 h post-NAC) were used for comparison of immune gene signature patterns with AR subjects to address whether the systemic immune responses directly resulted from diurnal variation. The healthy non-allergic subjects had no significant change in TNSS after NAC (baseline: 0). There was no significant difference (p > 0.05, Fisher’s exact test, Table 3-2) in demographics (age, sex, race, and BMI) between the 13 AR subjects and 5 healthy non-allergic subjects.             While immune cell frequencies (neutrophils, lymphocytes, and NLR) and their ratios in 13 AR subjects had significant changes after NAC, those in the 5 healthy non-allergic subjects had no significant change (Figure 3-7), at least within the first hour post-NAC. AR subjects and the healthy controls were significantly different in all variables except neutrophils (Figure 3-7 A).  The ratios of clusters 2, 3 and 4 at 1 h post-NAC were significantly different between the AR subjects and the healthy subjects. In comparison to baseline, the ratios of clusters 2 and 3 at 1 h post-NAC significantly increased in only the AR subjects, but the ratio of cluster 4 at 1 h post-NAC significantly increased in only the healthy subjects (Figure 3-7 B).    48  Figure 3-7 Comparison of immune gene signature patterns in 13 AR subjects to those of 5 healthy non-allergic subjects.  A, frequencies and their ratios of neutrophils, lymphocytes, and NLR. B, ratios of immune gene clusters (clusters 2, 3 and 4). Error bars: Mean ± SEM; *p < 0.05, Wilcoxon Rank Sum test: 13 AR subjects vs. 5 healthy subjects; #p < 0.05, Wilcoxon Signed Rank test/paired t test: baseline vs. 1 h post-NAC in the AR subjects; +p < 0.05, Wilcoxon Signed Rank test/paired t test: baseline vs. 1 h post-NAC in the healthy non-allergic subjects (allergic subjects, n = 13; non-allergic subjects, n = 5).  3.5 Discussion             Our hypothesis was that clustered immune gene sets based on patterns of time series gene expression in peripheral blood from subjects with AR undergoing NAC would be associated with immune cell frequencies and clinical symptoms of AR. Herein, we demonstrated a systemic   49 immune response signature, or in other words, a clustered immune gene signature associated with corresponding immune cell frequencies in whole peripheral blood collected following allergen challenge. The identified systemic immune response signatures were associated with the pathophysiological systemic immune responses of AR triggered by allergen challenge in the human model of AR. The immune gene signatures, which were reproducible in cat allergy cohorts, associated with significantly changed immune cell frequencies: cluster 3 was moderately associated with clinical symptoms at 6 h post-NAC in the AR subjects. clusters 2, 3 and 4 were significantly different between the AR subjects and the healthy non-allergic subjects, at least within the first hour post-NAC, which was the only comparison we could perform given the more limited blood sampling available for the healthy non-allergic subjects.             We determined associations between seven immune gene clusters and immune cell frequencies (leukocytes, platelets, neutrophils, lymphocytes, monocytes, eosinophils, and NLR) using canonical correlation analysis, which is a multivariate form of the general linear model, adjusting for inter-subject variability and time points (72,77). Finally, we investigated the Pearson correlation between clinical symptom scores and the identified systemic immune response pattern. Although TNSS is an ordinal variable, we used the Pearson correlation because TNSS has been considered as an interval variable (49,78) and we were interested in linear relationships between the intensity of clinical symptoms and other variables for a potentially objective measurement.             The Q and the M cohorts experienced peak TNSS or minimum PNIF score at 15 min post-NAC (Figure 3-2, A and B), and significantly increased clinical symptoms were shown by several hours, even at 9 or 10 h post-NAC in the Q cohort. This is consistent with previous studies, which demonstrated peak of clinical symptoms at a very early time point (2-20 min)   50 after allergen challenge in AR and allergic conjunctivitis (27,49,79,80). Subsequent LPR follows EPR after several hours and peaks 6-9 h after allergen exposure (20,21,27).             We investigated immune cell frequencies in CBC data, which is a straightforward, rapid, and relatively inexpensive method that provides reliable counts of subtypes of leukocytes as a standard diagnostic tool, e.g., neutrophils, lymphocytes, monocytes, and eosinophils (81–83). Basophil counts have been shown to be unreliable between hematology analyzers (81,83). Considering the minimum detectable cell count (0.1 x 109 cells/ L) of the analyzers, the basophil count was negligible in our CBC data (range 0.0 - 0.1 x 109 cells/L; median, 0.0). Thus, we excluded basophil count in this study, although basophils are an important effector cell type that is functionally and developmentally similar with mast cells associated with type 2 immune responses by type 2 T helper cells (84).             Changes in immune cell frequency of CBC are the net result of loss and gain of leukocytes in the blood. Many leukocyte subtypes are known to migrate to local inflammatory sites after NAC (13,20,85–87). Leukocyte frequency was mainly affected by neutrophils and lymphocytes that constitute large proportions of leukocytes. Additionally, the Q cohort had higher leukocyte count at 6 h post-NAC with higher monocytes in contrast to the M cohort. In a previous study, monocytes were significantly recruited at the site of allergen challenge in AR patients at 12 h post-NAC (88).             The Q and the M cohorts displayed similar changes after NAC in NLR and the frequencies of three leukocyte subtypes (neutrophils, eosinophils, and lymphocytes). However, the Q cohort, which had worse clinical symptoms at 2 h post-NAC compared to the M cohort, showed significantly higher cell counts for neutrophils and lymphocytes at this time point compared to 1 h post-NAC (p < 0.05, paired t test).   51             We observed lower eosinophil counts in the blood at 1 and 2 h post-NAC (Figure 3-3, A and B). This may be associated with eosinophil influx into nasal lavage fluids, and is consistent with a study that used intranasal heparin to reduce symptom scores at 1 and 6 h post-NAC (85), while demonstrating lower influx of eosinophils. Another study reported eosinophils to be significantly increased in nasal and bronchial biopsies and blood from AR subjects at 24 h post-NAC compared to baseline (89), though earlier time point samples were not collected.              In the correlation analysis using the 13 subjects who had no missing data, counts of neutrophils and lymphocytes at 2 and 6 h post-NAC or over all time points may be associated with change of immune response phase: innate and adaptive immune response; cell-mediated and humoral immune response; and EPR and LPR. These may be directly associated with the severity of AR, with neutrophils representing the innate immune system (90) and lymphocytes representing the adaptive immune system. High neutrophil frequency and NLR at early time points after NAC may reflect the highly activated innate immune response against the allergen, but their values at 6 h post-NAC are likely related to the interaction between the innate and adaptive immune systems.             We cannot determine the subtype frequencies of lymphocytes using CBC data, but the identified immune gene cluster, Q4/M4, corresponding to lymphocyte frequency, was comprised of genes related to lymphocyte function: Fc fragment of IgE receptor II (FCER2, low affinity receptor for IgE) and CD180 for B cells; transcription factor 7 (TCF7), lymphocyte antigen 9 (LY9), and Fms related tyrosine kinase 3 ligand (FLT3LG) for T cells. Although peripheral whole blood has not been previously well studied in the NAC model, recent papers reported significant (p < 0.05) changes in the proportion of CD4+ T cell subsets and type 2 innate lymphoid cells (ILC2s) in purified peripheral blood mononuclear cells following NAC (34,91).   52 While CD4+CCR4+ T cells decreased, CD4+CD25lo T cells, CD4+CD152+ T cells, and CD4+CRTH2+ T cells increased at 6 h post-NAC compared to pre-NAC or control (diluent) (34). ILC2s increased at 4 h post-NAC compared to baseline (91). These results are consistent with our own and may be associated with LPR and amplification of the IgE response. Levels of the Th2 cytokines IL-4, IL-5, IL-9, and IL-13 significantly increased in nasal fluid at 8 h post-NAC (48). The microenvironment of the nasal mucosa after NAC is thought to be modified by recruitment of leukocytes causing tissue remodeling and providing germinal center-like reactions, which facilitate the isotype switching to IgE+ B cells and their proliferation and maturation to IgE-producing plasmablasts/plasma (13,21,86,92–94). These responses may also be related to nonspecific nasal hyper-responsiveness (7,87).             Furthermore, the seven clusters of immune genes had deconvoluted information that demonstrated significantly differentially expressed immune genes associated with specific leukocyte subtypes and immune responses (Figure 3-4). For example, cluster 3 (Q3/M3) corresponding to neutrophils and NLR, was significantly (BH-FDR < 0.01) related to Toll-like receptor (TLR)–, IL-1–, and IL-4– mediated signaling pathways in an enrichment analysis (Enrichr). The geometric mean of the ratio at 6 h post-NAC was negatively associated with EPR (Figure 3-6, sum of TNSS in EPR); to minimize effect of extreme values, we used the geometric mean instead of the arithmetic mean because the ratio of immune gene expression was not standardized.              AR is a complex disease involving interactions between the local inflammatory site and the systemic immune system (e.g., lymphoid organs and peripheral blood). Systemic characteristics of AR have previously been demonstrated as alterations of the nervous system caused by immune responses of AR such as eosinophil recruitment to nasal nerves after NAC   53 (95,96). Systemic immune responses of AR triggered by allergen challenge may also be influenced by many factors such as stress, diurnal variation, and dehydration. The ratio of cluster 4 in the 5 healthy non-allergic subjects demonstrated that they may have a pattern in the normal diurnal condition, which was significantly different from the 13 AR subjects (Figure 3-7 B). Future studies comparing allergic with healthy non-allergic subjects at later time points will be needed to establish, or definitively rule out, any elements of diurnal variability.             In conclusion, we successfully tested our hypothesis in the cat allergy studies. The identified immune gene signatures associated with the frequency change patterns of corresponding immune cells after the allergen challenge had modest correlations with clinical symptoms.      54 Chapter 4: Validation test of the systemic immune gene signature in seasonal allergy (birch and ragweed allergies)  4.1 Sub-abstract             Our hypothesis was “The identified systemic immune response signatures in cat allergy will be validated in birch and ragweed allergies using peripheral blood collected after allergen challenge in NAC or EEU models.” We tested our hypothesis using birch allergy and ragweed allergy studies. In the given limited conditions such as small sample size and suboptimal time points of blood sample collection, we found less or no significant genes in the comparison between baseline and post-allergen challenge. This may reflect that intermittent allergy such as pollen allergy has less clinical symptoms than persistent allergy such as animal allergy.  4.2 Introduction             In Chapter 3 we identified systemic immune response signatures associated with the pathophysiology of allergic rhinitis (AR) using samples of cat allergic subjects. Cat allergy is a typical example of animal allergy, which is a perennial/persistent allergy. Animal allergy has stronger symptoms than pollen allergy, which is a seasonal/intermittent allergy (18,97). Thus, we investigated how the identified systemic immune response signatures were expressed in pollen allergy. Herein, we have used the term ‘validation’ as any empirically determined results that support our findings in an independent dataset. This term could also be construed as ‘replication.’   55             We were interested in birch allergy and ragweed allergy because they are typical examples of pollen allergy in North America (98,99). The pollens of the Betulaceae family, mainly the genera birch and alder, are a frequent cause of AR in Europe. The major birch (Betula pendula) pollen allergen is Bet v 1, which shares molecular homology with many plants of the Betulaceae family. Birch allergy may therefore be associated with a series of allergies to different Betulaceae pollen allergens (100–102).             Birch allergy is also associated with a form of food allergy know as oral allergy syndrome or pollen food syndrome, which is an allergic reaction to tree nuts via secondary cross-reactivity mechanisms to birch pollen (100,103).              The major allergen of short ragweed (Ambrosia artemisiifolia) is Amb a 1 (also known as antigen E), and 95% of ragweed-sensitive individuals react to Amb a 1 in a skin prick test (104). In USA, 10% of the population aged 6 to 74 years demonstrates a positive skin prick test to ragweed (99). The increased length of the ragweed pollen season and concentration of pollen due to global environmental change, such as the warming global surface temperature, may be the cause of the increased prevalence of seasonal or intermittent allergic rhinitis (105–108).             Our hypothesis in this chapter was that the identified systemic immune response signatures in cat allergy could be validated in birch allergy and ragweed allergy using peripheral blood collected after allergen challenge in NAC or EEU models.             To test the hypothesis, we studied birch allergy and ragweed allergy using human models of AR.  Seven birch allergic subjects and 5 non-allergic subjects underwent allergen challenge and their immune genes were profiled using a NanoString nCounter gene expression assay. In particular, all birch allergic subjects and 3 of 5 non-allergic subjects underwent both the NAC   56 and EEU allergen challenge models. Fourteen ragweed allergic subjects underwent the EEU model and their RNA transcripts in peripheral blood were profiled using a microarray analysis. After analyzing the results of clinical symptom score, immune cell frequency, and gene expression data in each study, we compared these results with those of the 13 cat allergic subjects in Chapter 3.  4.3 Materials and methods 4.3.1 Study approval             The birch allergy study (Study Code: DMED-1343-10) and ragweed allergy study (DMED-1250-09) were granted ethical clearance by the Queen’s University Health Sciences and Affiliated Teaching Hospitals Research Ethics Board at Queen’s University. The studies (H09-02114) were also cleared by the University of British Columbia Research Ethics Board of the Providence Health Care Research Institute.  4.3.2 Subjects             In the birch allergy study, 7 birch allergic subjects and 5 non-allergic subjects underwent the allergen challenge and their immune genes were profiled using a NanoString nCounter gene expression assay (Table 4-1). Seven birch allergic subjects and 3 non-allergic subjects took part in both models: the NAC model experiment, and, after at least 289 days of break, the EEU model experiment. Two additional non-allergic subjects underwent the NAC or the EEU models to increase the sample size of non-allergic subjects in each study (Table 4-1). White birch pollen exposure was implemented for 4 hours in EEU.   57             In the ragweed allergy study, 14 ragweed allergic subjects underwent short ragweed pollen exposure for 3 hours in the EEU model and their RNA transcripts in peripheral blood were profiled using a microarray analysis (Table 4-1).             In AR subjects, inclusion criteria included that the subject be a healthy, ambulatory male or female volunteer with a clinical history of developing nasal symptoms consistent with allergic rhinitis on exposure to birch pollen and a positive skin prick test (SPT) to birch pollen with a wheal diameter at least 3 mm larger than that produced by the negative control (diluent only). SPTs conducted at the research site within the last 12 months were also acceptable.             In healthy non-allergic subjects, inclusion criteria included a history of non-allergy to any aeroallergens and a negative skin prick test to all of the common aeroallergens tested. Participants were excluded if they had a diagnosis of asthma; a history of anaphylaxis to cat allergen; an FEV1 < 80% of predicted; an FEV1/ FVC ratio < 0.7; vital signs (blood pressure,  pulse rate, respiratory rate, body temperature) that were outside normal limits; significant history of alcohol or drug abuse; any history of vasovagal reaction in response to needles or blood donation; a history of any significant disease or disorder; the subject was a smoker or quit smoking less than 3 months prior to the screening date; a female who is pregnant, lactating or actively trying to become pregnant; an upper or lower respiratory infection within the 2 weeks prior to the screening or challenge visit; any structural nasal abnormalities or nasal polyps on examination, a history of frequent nasal bleeding, or nasal surgery within the previous 3 months; signs/symptoms of active perennial rhinitis or seasonal allergic rhinitis (Total Nasal Symptom Score (TNSS) of greater than 2 at screening or immediately prior to allergen challenge); a history  58  Table 4-1 Characteristics of subjects.  No significant difference (p > 0.05, Fisher’s exact test) with 13 cat allergic subjects, #p < 0.05, allergic versus non-allergic subjects in the NAC model, +p < 0.05, allergic versus non-allergic subjects in the EEU model.  NAC model EEU model Birch allergy Cat allergy Birch allergy Ragweed allergy Allergy status Allergic Non-allergic Allergic  Allergic Non-allergic Allergic Sex                  Men 3 3 2 3 2 7      Women 4 2 11 4 3 7 Age (year, mean ± SD) 42 ± 7# 31 ± 8# 36 ± 9 42 ± 7+ 31 ± 5+ 34 ± 10 BMI (kg/m2, mean ± SD) 31.7 ± 8.5 29.2 ± 6.1 26.5 ± 4.1 32.1 ± 7.1 26.4 ± 5.0 27.8 ± 6.2 Race                  Caucasian 6 4 11 6 4 14      Asian 1 1 2 1 1 0 Allergen dose administered at NAC. Cat: BAU/ml; Birch: PNU/ml, Median (Range) 1500.0# (1500.0 - 19500.0) 19500.0# 1666.7  (102.0-5000.0)       Allergen dose administered in EEU (pollen grains/m3)       3500 3500 3500   59 of a positive test to HIV, TB (not due to vaccination), hepatitis B (not due to vaccination) or hepatitis C; any other clinical trial and received an investigational product within the previous 30 days; a participant unable and/or unlikely to comprehend and/or follow the protocol over the duration of the study.   4.3.3 NAC             Seven birch allergic subjects and 5 non-allergic subjects underwent NAC using white birch pollen (39,000 PNU/ml, ALK-Abello Pharmaceuticals Inc., Hørsholm, Denmark, DIN 00299987). The NAC protocol was described in Chapter 2 and 3. Non-allergic subjects underwent NAC as in Chapter 3.  4.3.4 EEU             The EEU experimental setting at Queen’s University was described in the previous literature (52,58). Briefly, at the pollen exposure visit, all subjects were continuously exposed to 3500 grains/m3 concentration of short ragweed pollen (Greer Laboratories, Lenoir, North Carolina, USA) for 3 hours or white birch pollen (ALK-Abello Pharmaceuticals Inc., Hørsholm, Denmark) for 4 hours. To keep the concentration of the pollen constant, 7 Rotorod samplers (Sampling Technologies Inc, Minnetonka, Minnesota, USA) positioned throughout the subject seating area were used to monitor pollen levels every 30 minutes and to adjust the pollen levels as required.  4.3.5 Clinical symptom scores collection             In the birch allergy study in the NAC model, clinical symptom scores from all subjects were collected prior to NAC (baseline, 0 h), and at 15 min, 30 min, and every hour from 1 h to   60 12 h post-NAC, using symptom diary cards. In the EEU model, clinical symptom scores from all subjects were collected prior to challenge, every 30 minutes during pollen exposure (allergen challenge) for 4 hours, and every 1 hour between 4 h and 12 h following the allergen challenge protocol.             In the ragweed allergy study using the EEU model, clinical symptom scores from all subjects were collected prior to challenge, every 30 minutes during pollen exposure for 3 hours, and every 1 hour between 6 h and 12 h following the allergen challenge protocol.  4.3.6 Blood collection             Whole peripheral blood from all subjects was collected in EDTA blood tubes and PAXgene blood RNA tubes at two time points: pre-allergen challenge (baseline, 0 h) and post-allergen challenge (birch allergy in the NAC model, 1 h post-NAC; birch allergy in the EEU model, 4 h post-start of allergen challenge; and ragweed allergy in the EEU model, 3 h post-start of allergen challenge). Complete blood count (CBC) with differential was generated using an automated hematology analyzer.  4.3.7 Microarray             RNA from PAXgene blood RNA tubes was profiled using Affymetrix Human Gene 1.0 ST (Affymetrix, Santa Clara, CA, USA), which provides genome-wide expression profiling through measurement of protein coding and long intergenic non-coding RNA transcripts. RNA labeling and array hybridization were implemented by the Centre for Translational and Applied Genomics at the BC Cancer Agency (Vancouver, BC, Canada).    61 4.4 Results 4.4.1 Birch allergy 4.4.1.1 Clinical symptom score in the NAC model             In contrast to non-allergic subjects having no significant change after allergen challenge, allergic subjects had a weak correlation (Pearson correlation, r = -0.36) between PNIF and TNSS, and experienced peak TNSS (mean ± SD, 7.42 ± 1.90) at 15 min and minimum PNIF score (mean ± SD, 101.43 ± 39.34 L/min) at 30 min post-NAC (Figure 4-1 A). PNIF of allergic subjects decreased significantly (p < 0.05, linear mixed effect model (LME)) from 15 min to 12 h post-NAC (over all time points post-NAC) from baseline. These allergic subjects experienced a significant increase of TNSS between 15 min to 10 h post-NAC.   4.4.1.2 Clinical symptom score in the EEU model             In the EEU model, non-allergic subjects had no significant change during four hours of allergen challenge (by 4 h post-start of allergen challenge) and after the allergen exposure. Allergic subjects having a weak correlation (Pearson correlation, r = -0.33) between PNIF and TNSS, however, experienced peak TNSS (mean ± SD, 7.43 ± 0.98) at 3.5 h and minimum PNIF score (mean ± SD, 94.29 ± 41.98 L/min) at 4 h post-start of allergen challenge (Figure 4-1 B). PNIF of allergic subjects decreased significantly (p < 0.05, LME) from 1 to 12 h post-start of allergen challenge compared to baseline. These allergic subjects experienced a significant increase of TNSS at 30 min to 11 h post-start of allergen challenge.      62  Figure 4-1 Clinical symptom scores after allergen challenge in birch allergy.  A, TNSS and PNIF in the NAC model. B, TNSS and PNIF in the EEU model. Error bars: mean ± SEM. (birch allergic subjects, n = 7; non-allergic subjects, n = 5).   63  Figure 4-2 CBC pre- and post-allergen challenge.  A, birch allergic and non-allergic subjects in the NAC model. B, birch allergic and non-allergic subjects in the NAC model. Error bars: mean ± SEM. #p < 0.05, paired t test: baseline versus post-allergen challenge in the AR subjects; +p < 0.05, Paired t test: baseline versus allergen challenge in the healthy non-allergic subjects. (birch allergic subjects, n = 7; non-allergic subjects, n = 5).  4.4.1.3 Immune cell frequencies and neutrophil/lymphocyte ratio in the NAC model             In allergic subjects, leukocytes and neutrophils significantly (p < 0.05) increased post-allergen challenge (1 h post-NAC) compared to baseline. In non-allergic subjects, leukocytes significantly increased post-allergen challenge, but not neutrophils. There was no significant   64 difference between allergic and non-allergic subjects at both pre- and post-allergen challenge time points (Figure 4-2 A).  4.4.1.4 Immune cell frequencies and neutrophil/lymphocyte ratio in the EEU model             While non-allergic subjects had no significant change after allergen challenge, allergic subjects had significantly increased counts in platelets, leukocytes, neutrophils, and lymphocytes after allergen challenge for 4 hours in EEU. There was no significant difference between allergic and non-allergic subjects at both pre- and post-allergen challenge time points (Figure 4-2 B).  4.4.1.5 Differentially expressed immune genes in birch allergy             The canonical 730 immune genes of peripheral blood collected from birch-allergic and non-allergic subjects in the NAC and EEU models were profiled using the NanoString nCounter PanCancer Immune Profiling Panel. We investigated how immune genes were differentially expressed after allergen challenge through two comparisons.              The first comparison was between differentially expressed genes pre- versus post-allergen challenge in allergic subjects and non-allergic subjects. Some immune genes were identified by a nominal p value cut-off of 0.05, but there were no genes significantly differentially expressed after allergen challenge at a BH-FDR cut-off of 10% in both NAC and EEU models (Figure 4-3 A and B, and Figure 4-4 A and B).             The second comparison was between allergic subjects and non-allergic subjects pre- or post-allergen challenge. Like the first comparison, the result had no significant genes in both NAC and EEU models at a BH-FDR cut-off of 10% (Figure 4-3 C and D, and Figure 4-4 C and D).   65  Figure 4-3 Volcano plot of gene expression data at comparison in NAC model.  Pink and blue dot, p < 0.05, Paired LIMMA test (pre- versus post-allergen challenge), LIMMA test (allergic versus non-allergic subjects).   66  Figure 4-4 Volcano plot of gene expression data at comparison in EEU model.  Pink and blue dot, p < 0.05, Paired LIMMA test (pre- versus post-allergen challenge), LIMMA test (allergic versus non-allergic subjects).    67 4.4.2 Ragweed allergy 4.4.2.1 Clinical symptom scores             Ragweed allergic subjects experienced a minimum PNIF score at 1.5 h and peak TNSS at 3 h post-start of allergen challenge (Figure 4-5, A and B). Subjects had a negligible correlation (Pearson correlation, r = -0.28) between TNSS and PNIF (Figure 4-5 C). PNIF significantly decreased during allergen challenge (3 hours duration) and at 7 h post-start of allergen challenge. TNSS demonstrated monotonically significantly increased scores during the 3 hours of allergen challenge: TNSS increased sharply by 1.5 h post-start of allergen challenge, then it closed to stabilization by the end of allergen exposure (3 h post-start of allergen challenge). Although TNSS decreased after pollen exposure time, its level stayed significantly higher between 6 h to 12 h post-start of allergen challenge compared to baseline (Figure 4-5 B).     68  Figure 4-5 Clinical symptom scores.  A, PNIF. B, TNSS. C, The Pearson correlation between TNSS and PNIF. Error bars: mean ± SEM. *p < 0.05, LME. (n=14)  4.4.2.2 Immune cell frequencies and neutrophil/lymphocyte ratio             Leukocytes, neutrophils, and monocytes significantly (p < 0.05, paired t test) increased post-allergen challenge (3 h post-start of allergen challenge) compared to baseline (Figure 4-6).    69  Figure 4-6 CBC in ragweed allergic subjects pre- and post-allergen challenge.  Error bars: mean ± SEM. *p < 0.05, Wilcoxon signed rank test/paired t test (pre- versus post-allergen challenge). (n = 13, one subject with missing data).  4.4.2.3 Comparison of differentially expressed immune genes             RNA transcripts of peripheral blood collected pre- and post-allergen challenge (3 hours of ragweed pollen exposure in EEU) in ragweed allergic subjects were profiled using the Affymetrix Human Gene 1.0 ST. After filtering with a signal threshold (defined as a gene that in    70  Figure 4-7 Volcano plot of gene expression data at comparison in ragweed allergic subjects in the EEU model.  Red and blue dots indicate a BH-FDR < 0.1, paired LIMMA test (pre- versus post-allergen challenge).  over 20% of samples had greater than log2 32 in the value of gene expression), 14,137 genes were retained for comparison between pre- and post-allergen challenge. Seventy genes were identified at a BH-FDR cut-off of 10% (Figure 4-7). In a pathway enrichment analysis, the genes were associated with IL-2–mediated signaling events (BH-FDR < 0.001).   71 4.4.3 Comparison of pollen allergy to cat allergy             Our hypothesis in this chapter is that the identified systemic immune response signatures in cat allergy will be validated in birch and ragweed allergies using peripheral blood collected after allergen challenge in NAC or EEU models. An additional hypothesis is that the signatures will be different between allergic and non-allergic subjects.             The identified systemic immune response signatures of immune cell frequencies and immune gene signatures, which comprise 53 genes clustered according to the same clusters as developed for the test and validation cohorts in Chapter 3, were tested in birch and ragweed allergies and compared to those of the 13 cat allergy subjects (Table 3-2).  4.4.3.1 Clinical symptoms 4.4.3.1.1 PNIF in cat, birch, ragweed allergy studies             Non-allergic subjects had no significant reduction in PNIF scores after allergen challenge in the EEU and NAC models (Figure 4-8). There were no significant differences (p > 0.05, t test) at baseline between studies.             While cat allergic subjects experienced minimum PNIF (mean ± SD, 64.62 ± 25.61 L/min) at 15 min after allergen challenge in the NAC model, birch allergic subjects experienced minimum PNIF (mean ± SD, 101.43 ± 39.34 L/min) at 30 min after allergen challenge in the NAC model. Birch allergic subjects in the EEU model experienced minimum PNIF (mean ± SD, 94.29 ± 41.98 L/min) at 4 h post-start of allergen challenge. Ragweed allergic subjects in the EEU model experienced minimum PNIF (mean ± SD, 81.07 ± 33.64 L/min) at 1.5 h post-start of allergen challenge. Allergic subjects in the EEU model experienced the maximun reduction in    72  Figure 4-8 PNIF in all studies.  Error bars: mean ± SEM. (cat allergic subjects, n = 13; birch allergic subjects, n = 7; non-allergic subjects, n = 5).  PNIF levels during allergen exposure, but at a later time point than allergic subjects in the NAC model.   73  Figure 4-9 TNSS in all studies.  Error bars: mean ± SEM. (cat allergic subjects, n = 13; birch allergic subjects, n = 7; ragweed allergic subjects, n = 14; non-allergic subjects, n = 5).  4.4.3.1.2 TNSS in cat, birch, ragweed allergy studies             Non-allergic subjects had no significant change in TNSS after allergen challenge in EEU and NAC models (Figure 4-9). In the NAC model, both cat allergic subject and birch allergic subject cohorts experienced peak TNSS (respectively, mean ± SD, 8.08 ± 2.81, 7.43 ± 1.90) at 15 min after allergen challenge. Birch allergic subjects in the EEU model experienced peak TNSS (mean ± SD, 7.43 ± 0.98) at 3.5 h post-start of allergen challenge. Ragweed allergic   74 subjects in the EEU model experienced peak TNSS (mean ± SD, 9.14 ± 1.70) at 3 h. Both allergic subject cohorts had peak TNSS during allergen exposure.              All of the allergic subject cohorts had significantly (p < 0.05, Wilcoxon rank sum test) different values for sum of TNSS (time points: 0, 0.5, 1, 2, 3 h and every hour from 6 to 12 h; common time points in all studies) from the non-allergic subject cohorts in both the NAC and EEU models. The birch allergic subject cohort (mean ± SD, 42.86 ± 17.68) was not significantly different from the cat allergic subject cohort (mean ± SD, 39.15 ± 18.52) in the sum of TNSS values (time points: 0, 0.25, 0.5 h, and every hour from 1 to 12 h).             The birch allergic subject cohort underwent both EEU and NAC models. Their sum of TNSS (time points: 0, 0.5 h, and every hour from 1 to 12 h) in the EEU model (mean ± SD, 57.29 ± 16.67) was significantly higher (p < 0.05, paired t test) than that in the NAC model (mean ± SD, 35.43 ± 16.41).  4.4.3.2 The ratios of immune cell frequencies             To investigate trends of immune cell frequencies in pollen allergic subjects and non-allergic subjects compared to cat allergic subjects, we used the ratio of immune cell frequency at a given time point to the frequency at baseline, because the ratios enabled comparison of the cell count changes on an equivalent scale as compared to baseline. Birch allergic subjects and non-allergic subjects in the NAC model had 1 h post-NAC data similar to that of cat allergic subjects, but birch allergic subjects, ragweed allergic subjects, and non-allergic subjects in the EEU model had no time point similar to that of cat allergic subjects (Figure 4-10).             We therefore used different statistical tests for the NAC model and EEU model samples. First, the t test or Wilcoxon rank sum test were used to test the difference between samples from   75 the NAC model and cat allergic subjects. Second, the Mann-Kendall trend test (M-K test) was used to demonstrate trends between cat allergic subjects and birch allergic subjects, ragweed allergic subjects or non-allergic subjects from the EEU model. While the Wilcoxon rank sum test/t test demonstrated whether the birch allergic subjects or non-allergic subjects were significantly different from cat allergic subjects in the same NAC protocol setting, the M-K test demonstrated whether the birch allergic subjects, ragweed allergic subjects, or non-allergic subjects behaved according to the same trend as the cat allergic subjects in the different human model settings.             In the NAC model, there was no significant difference between birch allergic subjects and cat allergic subjects in terms of the ratios of neutrophil or eosinophil frequencies after allergen challenge (1 h post-NAC) to the frequencies at baseline. The same ratios were significantly (p < 0.05, Wilcoxon rank sum test/t test) different between cat allergic and non-allergic subjects.             Lymphocyte ratios at 1 h post-NAC in both birch allergic and non-allergic subjects were significantly different from those in cat allergic subjects. There was significant difference in the ratio of NLR at 1 h post-NAC to baseline in cat allergic subjects compared to birch allergic subjects and non-allergic subjects: while the ratio significantly (p < 0.05, Wilcoxon signed rank test/paired t test) increased after allergen challenge in cat allergic subjects, birch allergic subjects and non-allergic subjects had no significant difference in the ratio of NLR at 1 h post-NAC to baseline. Both birch allergic and non-allergic subjects had no significant difference in the ratios of leukocytes and monocytes compared to cat allergic subjects.             In the EEU samples, the ratio of lymphocyte frequency post-allergen challenge to baseline in ragweed allergic subjects and non-allergic subjects trended significantly (p < 0.05,    76  Figure 4-10 Ratios of immune cell frequencies.  A, allergic subject cohorts. B, non-allergic subject cohorts. Error bars: mean ± SEM. Comparison with cat allergic subjects: Wilcoxon rank sum test/t test, *p < 0.05, versus birch allergic subjects in the NAC model (NAC.Birch.allergic), #p < 0.05, versus non-allergic subjects in the NAC mode (NAC.Birch.non-allergic); one-  77 tailed M-K test, +p < 0.05, with non-allergic subjects in the EEU model (EEU.Birch.non-allergic), $p < 0.05, with ragweed allergic subjects in the EEU model (EEU.ragweed.allergic). (cat allergic subjects, n = 13; birch allergic subjects, n = 7; ragweed allergic subjects, n = 13, one subject with missing data ; non-allergic subjects, n = 5).  one-tailed M-K test; birch allergic subjects had a p value of 0.052) with that of cat allergic subjects. Interestingly, the ratio of NLR in ragweed allergic subjects trended significantly (p < 0.05, one-tailed M-K test) with that of cat allergic subjects. The ratios of leukocytes, neutrophils, monocytes, and eosinophils were not significant in the trend test (Figure 4-10).  4.4.3.3 Immune gene signatures in pollen allergy             We tested the systemic immune signature of birch allergy and ragweed allergy using the immune gene clusters identified in Chapter 3. These immune gene clusters comprise 53 genes clustered based on the test and validation cohorts of cat allergy.  4.4.3.3.1 In the NAC model             In birch allergy using the NAC model, 730 canonical immune genes and 40 housekeeping genes were profiled from peripheral blood samples collected pre- and post-allergen challenge. After filtering with a signal threshold (defined as a gene that in over 20% samples had greater than log2 32 in the value of gene expression) to remove genes with low signal intensity, 480 genes were retained for further study.             Comparing pre- and post-allergen challenge samples, no gene of the 480 genes was identified to have significance at a BH-FDR cutoff of 10% in both birch allergic and non-allergic subjects. We, however, then focused on the 53 genes identified in the cat allergy study. Forty-five of the 53 genes overlapped with the 480 genes retained after a signal threshold was applied   78 in birch allergy. We found that 4 of the overlapped 45 genes had significantly (BH-FDR < 0.1, paired LIMMA test) differential expression after allergen challenge in birch allergic subjects. Using the selected 45 of 53 genes in the birch allergic subjects may allow for a less stringent BH-FDR than using the 480 genes originally identified, based on the assumption that the differential expressions of the genes are not random events. Three genes – C-C motif chemokine receptor 7 (CCR7), interleukin 7 receptor (IL7R), and IL2 inducible T-cell kinase (ITK) – occur in cluster 6 and one gene – REL proto-oncogene, NF-KB subunit (REL) – in cluster 3. The geometric mean of expression of genes in each cluster, or sole value of the gene, when the cluster had a single gene, were used as the representative value of the cluster (Figure 4-11). There was no significant difference between cat allergic subjects and birch allergic or non-allergic subjects (p > 0.05, Wilcoxon rank sum test/t test) when comparing the 4 significant immune genes.              On the other hand, the results of testing using all of the 45 immune genes without a statistical cut-off are shown in Figure 4-12. In cluster 2, both birch allergic and non-allergic subjects were significantly (p < 0.05, Wilcoxon rank sum test/t test) different post-NAC compared to cat allergic subjects. In clusters 3, 4, and 6, while non-allergic subjects were significantly different post-NAC compared to cat allergic subjects, birch allergic subject had no significant difference. In clusters 5 and 7, there was no significant difference between cat allergic subjects and birch allergic or non-allergic subjects.        79  Figure 4-11 Immune gene clusters (using 4 significant genes) after a statistical cut-off in birch allergic subjects in the NAC model.  Error bars: mean ± SEM. There was no significant (p > 0.05, Wilcoxon rank sum test/t test) difference between studies at 1 h post-NAC. (cat allergic subjects, n = 13; birch allergic subjects, n = 7; non-allergic subjects, n = 5).     80  Figure 4-12 Immune gene clusters without a statistical cut-off in birch allergic subjects in the NAC model.  Error bars: mean ± SEM. *p < 0.05, Wilcoxon rank sum test/t test (cat allergic subjects vs. birch allergic subjects), #p < 0.05, Wilcoxon rank sum test/t test (cat allergic subjects vs. birch non-allergic subjects). (cat allergic subjects, n = 13; birch allergic subjects, n = 7; non-allergic subjects, n = 5).   81 4.4.3.3.2 In the EEU model             In birch allergic subjects using the EEU model, when comparing pre- and post-allergen challenge, no gene was found to be significant at a BH-FDR cutoff of 10%, regardless of whether 480 genes were used, or whether only genes shared with the 53 immune genes of cat allergy were used (i.e., the 46 genes that overlapped between the 480 birch allergy genes in the EEU model and 53 genes of cat allergy.)             In ragweed-allergic subjects, however, 70 genes were significantly (BH-FDR < 0.1, paired LIMMA test) differentially expressed after allergen challenge. The profiled 14,137 genes of ragweed allergy after signal threshold and the 53 immune genes of cat allergy shared 48 immune genes. Of these 48 immune genes, 11 genes were significantly (BH-FDR < 0.1, paired LIMMA test) differentially expressed following allergen challenge. Using the selected 11 genes, we applied the immune gene signatures identified in Chapter 3. In clusters 3, 5, and 6, ragweed-allergic subjects at post-allergen challenge (3 h post-start of allergen challenge) had significantly (p < 0.05, one-tailed M-K test) the same trend as cat-allergic subjects at 2 h and 6 h post-NAC (Figure 4-13).             Without using a statistical cut-off, we investigated ragweed-allergic subjects, birch allergic subjects, and non-allergic subjects with the systemic immune gene signature and compared these subjects with cat-allergic subjects using the 43 immune genes shared in the studies (Figure 4-14). Ragweed-allergic subjects (3 h post-start of allergen challenge) and birch allergic subjects (4 h post-start of allergen challenge) had significantly (p < 0.05, one-tailed M-K test) the same trend as cat-allergic subjects at 2 h and 6 h post-NAC in clusters 3, 4, 5, and 6. Non-allergic subjects (4 h post-start of allergen challenge) had significantly the same trends as cat-allergic subjects at 2 h and 6 h post-NAC in cluster 5 (Figure 4-14).   82  Figure 4-13 Immune gene clusters (using 11 significant genes) after a statistical cut-off in ragweed allergic subjects in the EEU model.  Error bars: mean ± SEM. One-tailed M-K test (with cat allergic subjects), *p < 0.05, ragweed allergic subjects. (cat allergic subjects, n = 13; ragweed allergic subjects, n = 14).    83  Figure 4-14 Immune gene clusters (using 43 genes) without a statistical cut-off in the EEU model.  Error bars: mean ± SEM. One-tailed M-K test (with cat allergic subjects), *p < 0.05, ragweed allergic subjects, #p < 0.05, birch allergic subjects, +p < 0.05, birch non-allergic subjects. (cat allergic subjects, n = 13; birch allergic subjects, n = 7; ragweed allergic subjects, n = 14; non-allergic subjects, n = 5).   84 4.5 Discussion             Herein, we investigated seasonal/intermittent allergic rhinitis (AR) using the NAC and EEU models. While the NAC model allows for measurement of administered allergen dose and challenge subjects individually, the EEU model provides a controlled allergen exposure environment to all subjects simultaneously (109). In both models, non-allergic subjects had no significant symptoms due to allergen challenge. Birch allergic subjects and ragweed allergic subjects, however, demonstrated significant AR responses triggered by allergen challenge, although symptom patterns were different between the NAC and the EEU models.              Birch allergic subjects in the NAC model had a similar symptom pattern to cat allergic subjects in TNSS and PNIF: their peak of symptoms occurred early. Birch and ragweed-allergic subjects in the EEU model had monotonically increasing TNSS values, but PNIF of ragweed-allergic subjects started to return to baseline 1.5 h post-start of allergen challenge.             Birch-allergic and ragweed-allergic subjects had a weak or a negligible negative correlation between TNSS and PNIF, in contrast to cat-allergic subjects having a moderate negative correlation (r < -0.5). The peak TNSS and minimum PNIF of cat allergic subjects were recorded at the same time (15 min post-NAC). The peak TNSS of birch-allergic subjects occurred 15 min post-NAC, followed by the minimum PNIF at 30 min post-NAC. The result may be associated with symptom characteristics of seasonal and perennial allergic rhinitis since PNIF reflects nasal congestion. While typical symptoms of seasonal allergic rhinitis to pollen include watery rhinorrhea, itching, and sneezing, along with frequent ocular allergy symptoms but not severe nasal congestion, most of the patients with perennial allergic rhinitis have predominantly nasal congestion and mucous production symptoms caused by a swollen nasal mucosa or sneezing paroxysms (110).    85             Although the EEU model differed in allergen exposure time and allergen dose to the NAC model, we tested whether a trend analysis may demonstrate an aspect of systemic responses of immune cell frequency changes associated with the pathophysiology of AR that differs between the studies. The M-K test assessed whether EEU samples (ragweed allergic subjects at 3 h post-start of allergen challenge; birch allergic and non-allergic subjects at 4 h) trended between 2 h and 6 h post-NAC in cat allergic subject samples.              The changes of immune cell frequency and immune gene expression in birch-allergic subjects were less than those of the cat-allergic subjects in the NAC model, although birch-allergic subjects had no significant difference from cat-allergic subjects in TNSS intensity.  When, without statistical filtering, we tested immune gene signatures in pollen allergy using the 53 immune genes identified in the cat allergy study in Chapter 3, birch allergic subjects and ragweed allergic subjects had more similar trends to cat allergic subjects as compared to non-allergic subjects. Our results, however, must be validated in a large-scale study with more time points to collect blood samples given the low statistical power of our small sample size (in particular, our small sample size of only 7 birch allergic subjects and 5 non-allergic subjects). In the EEU model, there was no significant immune gene in the birch allergic subjects, and only 11 immune genes of the systemic immune gene signature of cat allergy were significantly (BH-FDR < 0.1) differentially expressed in ragweed-allergic subjects; in the NAC model, we discovered 4 immune genes in birch allergic subjects.             The less significant results described may correlate with the limitations of the studies. First, blood was collected at only two time points. AR symptoms were enhanced by longer allergen exposure time in the EEU model, but post-allergen challenge blood was collected only at later time points, 3 h or 4 h post-start of allergen challenge, and not at the time point of rapid   86 change of symptoms. Many systemic immune gene signatures of cat allergy demonstrated a return to baseline at 6 h post-NAC. Second, the sample size was smaller in the birch allergy study than the cat allergy study: there were only 7 birch allergic subjects. In contrast, the cat allergy study had 13 subjects. Third, the significant (p < 0.05, Fisher’s exact test) difference in age between allergic subjects and non-allergic subjects was a potential confounding factor and a limitation of the statistical tests used in the birch allergy study.              On the other hand, the less significant results we report may also be related to the allergen components we compared. Pollen allergy is generally less severe than animal allergy such as cat allergy (12,18,79,97). There may be several explanations for the difference in severity. First, the allergen exposure frequency of pollen allergy is limited and temporary because of seasonal pollen production, which also varies considerably depending on the time (day, season) and geographic area (50), so pollen allergy results in less chronic inflammation (which is associated with priming of AR in the nasal mucosa) than perennial/persistent allergy such as cat allergy. The other explanation is related to TLR activation via a lipid transfer mechanism in animal allergy. Cat allergen predominantly includes Fel d 1 (uteroglobin), Fel d 2 (serum albumin) and Fel d 4 (lipocalin). In particular Fel d 1 enhances AR symptoms by helping TLR activation via a lipid transfer mechanism (97). It may therefore trigger stronger systemic immune responses in animal allergy than pollen allergy.             Furthermore, different protease activity in birch pollen and ragweed pollen may be associated with the significant (p < 0.05, Wilcoxon rank sum test) difference in the peak TNSS intensities between the birch allergic subjects and the ragweed allergic subjects. Although birch allergic subjects experienced one hour longer pollen exposure than the ragweed allergic subjects, the birch allergic subjects (mean ± SD, 7.43 ± 0.98) had significantly less intense TNSS than the   87 ragweed allergic subjects (mean ± SD, 9.14 ± 1.70). The protease activity of allergen is important for sensitizing and exacerbating AR. Proteolytically-active allergen directly and indirectly breaks intercellular junctions, comprised of tight junctions, adherens junctions, and desmosomes, of the epithelium (24). Increased epithelial permeability allows the allergen to invade the barrier and pass through to cause allergic responses. At the lumen of the epidermis, allergen can be also taken up by protrusions of dendrites of dendritic cells (DCs) such as the Langerhans cells, where it induces antigen-specific Th2 responses (24,25). The reason that ragweed allergic subjects had more significant results than birch allergic subjects in our study may be due to the stronger protease activity of ragweed. Ragweed has stronger (3-9 times depending on the substrate) protease activity compared to birch allergy, and its activity is less variable in the pH 5.5-9.0 range. The protein concentration of ragweed pollen extract in Tris buffer (pH 9.0) was 410 µg/m, three times higher than that of birch pollen extract (130 µg/ml) (111).              Non-allergic subjects in the EEU model had a similar trend in lymphocyte ratio as cat allergic subjects (Figure 4-10). The trend may be representative of factors, such as diurnal variation, that affect both allergic or non-allergic subjects, and also the fact that pollen allergens may induce pollen-specific CD4+ T cells such as Th1 and Treg in non-allergic subjects (112–114).             In NAC model, the successful allergen dose was decided by the criteria of TNSS (³ 8) and/or PNIF (reduction ³ 50 % of baseline) at the screening visit. TNSS, therefore, was relatively scored by subjects in both cat allergy and birch allergy studies as the subjective score measurement may already lose its role to compare the studies. TNSS cannot be used to objectively compare the severity of clinical symptoms between seasonal/intermittent and   88 perennial/persistent AR. An objective method such as the systemic immune signature may better reflect the intensity of pathophysiological systemic immune responses of AR.     89 Chapter 5: Utility of the systemic immune gene signature approach  5.1 Sub-abstract             Our hypothesis was "Systemic immune response signatures following allergen challenge in a patient with allergic rhinitis receives immunotherapy will be differentially expressed in the peripheral blood samples collected pre- and post-immunotherapy. And the systemic immune response signature changes will provide a cross-sectional view to further investigate the mechanism of action of the immunotherapy." We tested the hypothesis using samples from patients with allergic rhinitis who had received peptide immunotherapy. The comparison between pre- and post-treatment demonstrated significant changes in the systemic immune response signatures. Interestingly, the significant reduction of clinical symptoms after the immunotherapy was strongly associated with the increase of lymphocytes at 1 h post-NAC at post-treatment. Statistical tests and the systemic immune gene signature approach identified five immune genes associated with the mechanism of the action of the immunotherapy: integrin subunit alpha E (ITGAE, CD103), CD180 (LY64), neural cell adhesion molecule 1 (NCAM1, CD56), C-C motif chemokine receptor 7 (CCR7, CD197), and leucine rich repeat neuronal 3 (LRRN3, NLRR3).  5.2 Introduction             Many treatment types are currently in use or in development for patients who suffer from allergic rhinitis (AR), the most prevalent allergic disease worldwide. When optimal allergen avoidance is impossible in practice, intranasal corticosteroids, oral antihistamines and leukotriene receptor antagonists have been applied for various severities of AR, but these   90 treatments are not effective for all patients. Allergen specific immunotherapy (AIT) is recommended for patients with AR who have inadequate response with pharmacologic therapy (115,116). Although AIT poses a risk of anaphylactic reactions, when it is prescribed by physicians who are adequately trained in the treatment of allergies, it has many benefits including prevention of the development of asthma and reduced sensitization to other allergens (115,117,118). Two AITs that are currently used in clinical practice are subcutaneous immunotherapy (SCIT) and sublingual immunotherapy (SLIT). SCIT can be performed for both single and multiple allergens, while SLIT is limited to single allergens, due to its increased cost at higher doses and the inconvenience of taking multiple tablets for multiple allergens (119). Intralymphatic immunotherapy (ILIT), involving injection of allergen directly into a lymph node, and intradermal immunotherapy (IDIT), involving intradermal injection, are new injection approaches to administer allergen. Besides allergen injection forms of immunotherapy, non-injection forms such as epicutaneous, nasal, and oral immunotherapies are in development (119–122).             While the mechanism of action of immunotherapy has not been completely elucidated, the most plausible explanation is that immunotherapy modifies the immune response, resulting in less production of specific immunoglobulin E (sIgE) and more of the other isotype antibodies, such as IgG; in a rush immunotherapy, however, any association with change in sIgE or specific IgG was not reported (123). Immunotherapy also changes the response of T cells by activation of Treg cells, modifying their systemic (peripheral) and local (nasal mucosal) responses to allergen. These are the result of a redirection of allergen-specific Th2 response or immune deviation, which have been reported by many studies (19,123–127).    91             Various approaches are needed to understand AR and investigate the mechanism of action of immunotherapy. As described in Chapter 3, we recently identified a systemic immune response pattern in the whole peripheral blood of AR patients during nasal allergen challenge (NAC) (128). The immune gene expression clusters of the blood were associated with frequency change of the corresponding immune cells. Herein we test the utility of the systemic immune response approach to investigate the mechanism of a novel peptide immunotherapy.             The 18 cat allergic subjects in Chapter 3 participated in a pilot mechanistic research trial of a novel form of immunotherapy. The immunotherapy is Cat peptide antigen desensitization (Cat-PAD, Circassia Ltd, Oxford, UK), that is an equimolar mixture of seven short synthetic peptides (13-17 amino acids) being used against cat allergy in a clinical trial. The peptide sequences are derived from the primary sequence of Fel d 1, the major cat allergen. Ten of the 18 AR subject completed the NAC protocol pre- and post-treatment without any missing data. Using the clinical symptoms and immune gene expression data of the 10 AR subjects we tested our hypothesis that the identified systemic immune response signatures will be differentially expressed pre- and post-treatment and will provide a cross-sectional view to further investigate the mechanism of action of the immunotherapy. Our approach can provide an understanding of the immune response changes occurring due to the Cat-PAD immunotherapy.  5.3 Materials and methods 5.3.1 Study approval             This study is a follow-up study to the studies described in Chapter 3. The approval number is same as that provided in the previous chapter.     92 5.3.2 Subjects             This study was performed on 10 AR subjects and 5 healthy non-allergic subjects who also took part in the Chapter 3 study (Table 5-1). The 10 AR subjects received Cat-PAD treatment between NAC Visit 1A (V1A) and Visit 3 (V3) (Figure 5-1). The collection of clinical symptom scores and blood during NAC was completed without missing data. There were no significant demographic differences (p > 0.05, Fisher’s exact test: age, sex, race, and BMI) between the 10 AR subjects and 5 healthy non-allergic subjects.             In AR subjects, inclusion criteria included a minimum 1-year documented history of AR on exposure to cats and a positive skin prick test to cat allergen with a wheal diameter at least 3 mm larger than that produced by the negative control.              In healthy non-allergic subjects, inclusion criteria included a history of non-allergy to any aeroallergens and a negative skin prick test to all of the common aeroallergens tested. Subjects were excluded if they had a diagnosis of asthma, a history of anaphylaxis to cat allergen, an FEV1 < 80% of predicted, an FEV1/ FVC ratio < 0.7, vital signs (blood pressure, pulse rate, respiratory rate, body temperature) outside normal limits, a significant history of alcohol or drug abuse, any history of vasovagal reaction in response to needles or blood donation, a history of any significant disease or disorder, or if the subject was a smoker or quit smoking less than 3 months prior to the screening date.     93  Figure 5-1 Diagram of an NAC model with Cat-PAD, an immunotherapy intervention.  A, individual NAC model test procedure from screening visit (V1) to post-treatment visit (V3). B, a schedule of collecting peripheral blood samples and clinical symptoms before and during NAC onset in 10 AR subjects; in healthy non-allergic subjects, blood samples were collected at two time points (before NAC and 1 h post-NAC).  5.3.3 NAC             Ten AR subjects underwent NAC using cat allergen at visits V1A and V3 (Figure 5-1). The allergen administered at the NAC visits was a standardized cat allergen extract (10000 BAU/ml, ALK-Abello Pharmaceuticals Inc., Hørsholm, Denmark) with the same DIN (02235299) and lot number (ID0142). The NAC protocol was described in Chapter 2 and 3. Non-allergic subjects underwent NAC as described in Chapter 3.  5.3.4 Cat-PAD             Cat-PAD was supplied as a lyophilisate. To reconstitute Cat-PAD, water was added before injection so that a final dosing volume of 60 µl contained 6 nmol of each of the peptides. Cat-PAD was administered by a series of four intradermal injections with 4-week intervals between injections.    94  Table 5-1 Characteristics of the study populations:  There is no significant difference (p > 0.05, Fisher’s exact test: age, sex, race, and BMI) between the 10 allergic rhinitis subjects and 5 healthy non-allergic subjects. F: female; M: male; CAU: Caucasian; ALL.TNSS.Sum: sum of TNSS over 0 -12 h post-NAC   Subject Age Sex Race BMI Used allergen for NAC Allergen Dose  for NAC (BAU/ml) Days  (V3 - V1A) Difference of ALL.TNSS.Sum (V3-V1A)  10 allergic rhinitis subjects Subj 1 40 F CAU 30.4 Cat 5000.0 139 -33 Subj 2 29 F CAU 24.3 Cat 5000.0 153 -27 Subj 3 42 F CAU 28.4 Cat 5000.0 116 -16 Subj 4 33 F CAU 33.2 Cat 5000.0 119 3 Subj 5 30 F CAU 32.2 Cat 102.0 119 -18 Subj 6 24 F CAU 20.2 Cat 1666.7 119 -16 Subj 7 37 F CAU 23.6 Cat 5000.0 119 -55 Subj 8 30 F CAU 25.2 Cat 1250.0 141 -27 Subj 9 50 M CAU 27.7 Cat 1250.0 139 -26 Subj 10 41 F Asian 26.1 Cat 1250.0 133 0 Mean ± SD 35.6 ±7.8   27.1 ± 4.1     -21.5 ± 16.6 Median (Range)           3333.4  (102.0 - 5000.0) 126  (116 - 153)                         Subject Age Sex Race BMI Used allergen for NAC Allergen Dose  for NAC (PNU/ml)  ALL.TNSS.Sum 5 healthy non-allergic subjects Subj 11 36 M CAU 34.3 Birch 19500.0 1 Subj 12 25 M Asian 28.5 Birch 19500.0 0 Subj 13 29 F CAU 22.0 Birch 19500.0 0 Subj 14 42 M CAU 36.4 Birch 19500.0 0 Subj 15 23 F CAU 24.7 Birch 19500.0 1 Mean ± SD 31 ± 7.9     29.2 ± 6.1       Median (Range)           19500.0 0  (0-1)   95 5.4 Results 5.4.1 Comparison of clinical symptom scores pre- and post-treatment 5.4.1.1 PNIF             Subjects who received Cat-PAD treatment had a demonstrated improvement of their clinical symptom score after allergen challenge. Although minimum PNIF at 15 min post-NAC was similar between V1A and V3, the reduction of PNIF was significantly (p < 0.05, Wilcoxon signed rank test/paired t test) lower at 2, 4 and 12 h post-NAC at V3 compared to V1A (Figure 5-2 and Appendix B.1).  5.4.1.2 TNSS             Peak TNSS at 15 min post-NAC at V3 was significantly lower than at V1A. TNSS at V3 significantly decreased at 15 min, 1, 2 and 4 h post-NAC compared to V1A (Figure 5-3). The comparison of the sum of TNSS 0-12 h post-NAC between V1A and V3 in the 10 cat allergic subjects showed there to be a significant (p = 0.0027, mean of difference = -21.5, paired t test) reduction in clinical symptoms post-treatment, though two of the ten AR subjects experienced no treatment benefit (Table 5-1 and Appendix B.2).    96  Figure 5-2 Comparison of PNIF pre- and post-treatment.  A, PNIF at V1A and V3 in the 10 AR subjects. B, comparison of PNIF at V1A and V3. PNIF reduction at 2, 4 and 12 h post-NAC significantly decreased at V3 than V1A. Error bars: Mean ± SEM; * p < 0.05, Wilcoxon Signed Rank test/paired t test (V1A versus V3). (n = 10)   97  Figure 5-3 Comparison of TNSS pre- and post-treatment.  A, TNSS at V1A and V3 in the 10 AR subjects. B, comparison of TNSS at V1A and V3. TNSS at 15 min, 1, 2 and 4 h post-NAC significantly decreased at V3 than V1A. Error bars: Mean ± SEM; * p < 0.05, Wilcoxon Signed Rank test/paired t test (V1A versus V3). (n = 10).   98 5.4.2 Comparison of immune cell frequencies and neutrophil/lymphocyte ratio pre- and post-treatment             We investigated immune cell frequencies and their ratio (a relative number at each time point compared with the baseline value of V1A). Immune cell frequencies and neutrophil/lymphocyte ratio (NLR) dynamically changed after Cat-PAD treatment, except for eosinophils (Figure 5-4). Leukocytes and their ratio at V3 significantly (p < 0.05, paired t test) decreased at 1 h post-NAC compared to V1A. The frequency of neutrophils at V3 was significantly lower at baseline and 1 h post-NAC than V1A; and their ratio at V3 was significantly lower at not only baseline and 1 h but also 2 h post-NAC compared to V1A. Lymphocytes and their ratio at V3 significantly increased at 1 and 2 h post-NAC compared to V1A. Monocytes and their ratio at V3 significantly increased at 2 h post-NAC. Both NLR and its ratio at baseline, 1, and 2 h post-NAC at V3 were significantly lower than at V1A. But eosinophils and eosinophil ratio had no significant difference between V1A and V3.          99  Figure 5-4 Immune cell frequencies before and after immunotherapy intervention.  A, comparison of immune cell frequencies at V1A and V3. B, ratios of immune cell frequencies at V1A and V3: Error bars: Mean ± SEM; *p < 0.05, paired t test (V1A versus V3). (n = 10).   100 5.4.3 Differentially expressed gene pre- and post-treatment             Using the NanoString nCounter PanCancer Immune Profiling Panel, we profiled 730 canonical immune genes from peripheral blood samples collected post-NAC at four time points at both pre-treatment (V1A) and post-treatment (V3) visits (Figure 5-1). The LIMMA paired test identified significantly (BH-FDR < 0.1) higher or lower expressed genes at V3 compared to V1A at baseline and 1, 2 and 6 h post-NAC (Figure 5-5 and Appendix B.4 and B.5). These significantly expressed genes were applied to a pathway enrichment analysis (Enrichr) in order to find the immune response associated with the genes at each time point. The associated top 5 pathways (BH-FDR < 0.0001) are shown in Figure 5-5. The genes associated with IL-2–, IL-4–, and IL-6–mediated signaling events had lower expression at V3. T-cell receptor (TCR)– and IL-12–mediated signaling events were associated with more highly expressed genes at V3 (Figure 5-5).         101  Figure 5-5 Significantly differentially expressed genes after Cat-PAD.  The red and blue dots at the volcano plot at each time point demonstrate significantly (respectively, higher and lower) expressed genes at V3 than V1A. Top 5 results of pathway enrichment analysis associated with the significantly expressed genes at each time point were described. BH-FDR < 0.1, LIMMA paired test (V1A versus V3 at each time point). Pathway enrichment analysis (Enrichr, BH-FDR < 0.0001) (Appendix B.4 and B.5).    102 5.4.4 Comparison of immune gene signatures pre- and post-treatment             In Chapter 3, we used Fuzzy c-means clustering to identify systemic immune gene signatures in peripheral blood collected following allergen challenge. Herein we used the systemic immune gene signature approach method to investigate how immune gene expression changed after Cat-PAD treatment in 10 cat allergic subjects. As described in Chapter 3, two filtering procedures –comparison of gene expression at all time points post-NAC with baseline using a linear mixed effect model (LME) and a signal range cutoff filter (> ± 1.2 fold change) – identified significantly differentially expressed immune gene after allergen challenge before clustering.  As a result of two filtering procedures, 126 immune genes were identified as they significantly differentially expressed after allergen challenge at V1A in the 10 AR subjects. In contrast, 53 immune genes were identified at V3 in the 10 AR subjects and 25 of these 53 immune genes overlapped with the 126 immune genes identified at V1A (Figure 5-6 A and Appendix B.3). The significantly expressed immune genes at V1A and V3 were clustered into 7 groups based on their standardized (mean: 0, SD: 1) mean expression patterns (Figure 5-6 B and C). One hundred one of 126 immune genes at V1A had not significant expression at V3. Moreover, the all of overlapped 25 immune genes, which were significantly expressed pre- and post-treatment, had not consistent patterns at both V1A and V3 (Table 5-2). These indicate the changes of systemic immune responses in the AR subjects after Cat-PAD treatment.    103  Figure 5-6 Clustering in 10 AR subjects at V1A and V3.  A, significantly differentially expressed genes at V1A and V3. B, clusters at V1A using significant 126 immune genes after filtering. C, clusters at V3 using significant 53 immune genes after filtering.        104 Table 5-2 Differential expression patterns in 25 overlapped immune genes at V1A and V3. Overlapped  25 genes V3.Cluster 1 V3.Cluster 2 V3.Cluster 3 V3.Cluster 4 V3.Cluster 5 V3.Cluster 6 V3.Cluster 7 V1A.Cluster 1 0 0 0 2 1 0 0 V1A.Cluster 2 0 1 2 1 0 0 0 V1A.Cluster 3 2 0 2 0 0 1 0 V1A.Cluster 4 0 0 0 0 1 0 0 V1A.Cluster 5 0 3 0 0 0 4 1 V1A.Cluster 6 0 0 0 0 3 0 0 V1A.Cluster 7 0 0 0 0 0 0 1              We investigated how the clusters of immune genes that were identified in Chapter 3 expressed differentially between V1A and V3 in the 10 AR subjects. To address this question, 50 genes (50 of the 126 significant genes at V1A) that overlapped with the 53 significant immune genes used in chapters 3 and 4 were compared between V1A and V3 using geometric mean of expression ratios (a relative number at each time point compared with the baseline value of V1A) of genes at each cluster. Similarly to immune cell frequencies that dynamically and significantly differentially changed (Figure 5-4) between V1A and V3, clusters that identified in Chapter 3 also significantly changed after Cat-PAD treatment (Figure 5-7). Cluster 2 of Chapter 3 were significantly (p < 0.05, Wilcoxon signed rank test/paired t test) differentially expressed at baseline and 1, 2, and 6 h post-NAC between V1A and V3; clusters 3 and 4 at baseline, 1 and 2 h post-NAC; cluster 5 at 6 h post-NAC; cluster 6 at 1, 2, and 6 h post-NAC (Figure 5-7).    105  Figure 5-7 Comparison of clusters pre- and post-treatment.  Error bars: Mean ± SEM; *p < 0.05, Wilcoxon signed rank test/paired t test (V1A versus V3 in 10 AR subjects). (n = 10).    106 5.4.5 Investigation of the mechanism of action of Cat-PAD using the systemic immune response signature approach 5.4.5.1 Relationship between clinical symptoms and immune cell frequencies             We were interested in the relationship between changes in clinical symptoms and systemic immune responses in peripheral blood after onset of AR response using the NAC model. We determined how TNSS change, which is a typical clinical symptom score measure that includes the four main symptoms of AR, is related to immune cell frequency changes by calculating the difference in value between V1A and V3 (Figure 5-8 A). As described in Chapter 3, although TNSS is an ordinal variable that represents a participant’s subjective evaluation of symptoms, by considering TNSS as an interval variable we could use the difference in value and the Pearson correlation (r), instead of the Spearman correlation, to evaluate the linearity between variables. Here we used the sum of TNSS scores, calculated as the total TNSS measured at given time points: ALL.TNSS.Sum, the sum of TNSS over 0–12 h post-NAC (Table 5-1); EPR.TNSS.Sum, the sum of TNSS over 0–6 h post-NAC; and LPR.TNSS.Sum, the sum of TNSS over 7–12 h post-NAC. We defined early phase response (EPR) as clinical symptoms 0–6 h post-NAC and late phase response (LPR) as clinical symptoms 7–12 h post-NAC based on the change in mean value of the participants’ clinical symptoms. The mean value of TNSS returned to near baseline at 6 h post-NAC and increased 7 h post-NAC (Figure 5-2 and Figure 5-3). Using the sum of TNSS, we could also put a weight on each measure. In other words, EPR could be weighted more than LPR because the interval of TNSS recording time points in the 1h post-NAC (15, 30 min and 1 h post-NAC) was shorter than the 1 hour intervals that followed. The weighted TNSS in the early time point may be able to reflect the strong AR symptoms in the EPR (Figure 5-2 and Figure 5-3). In immune cell frequency data, we used cell counts at each time point and   107 the sum of cell counts 0–6 h post-NAC. Basophil counts in CBC have been shown to be unreliable between hematology analyzers (81,83) and the basophil counts were negligible in our data. Thus we excluded basophil counts in this study as we had done in Chapter 3.              The difference of lymphocyte level at 1 h post-NAC had strong correlation with the difference of ALL.TNSS.Sum (r = - 0.7927) and the difference of EPR.TNSS.Sum (r = -0.8819) (Figure 5-8, A and B). The difference (V3-V1A) in lymphocytes increased from -0.1 to 0.5 x 109 cells/L with significant difference between V1A and V3 (p < 0.05, paired t test) (Figure 5-4). The difference of ratio of lymphocytes at 1 h post-NAC also had strong correlation with the difference of ALL.TNSS.Sum (r = -0.7117) and the difference of EPR.TNSS.Sum (r = - 0.8138).  In other words, the higher lymphocyte frequency at the first-time point following allergen challenge at V3 was positively associated with the reduction of clinical symptoms after the treatment.     108  Figure 5-8 Correlation between difference of TNSS and difference of immune cell frequencies.  A, a heatmap of Pearson correlation between the differences (V3-V1A) in TNSS and Immune cell counts. B, Pearson correlation of difference in lymphocyte count at 1 h post-NAC with difference in ALL.TNSS.Sum or EPR.TNSS.Sum. ALL.TNSS.Sum: a sum of TNSS over all measured time points (baseline to 12 h post-NAC); EPR.TNSS.Sum: a sum of TNSS over time points in EPR (baseline to 6 h post-NAC).   109 5.4.5.2 Investigation in cell type markers at 1h post-NAC             We sought to understand why the increase of lymphocytes at 1 h post-NAC at V3 is associated with clinical symptom change. To investigate the reason, we first considered the cell type markers provided by the NanoString nCounter gene expression assay, which we used for profiling immune genes in the peripheral blood. We assumed that more highly expressed genes at 1 h post-NAC may be associated with the increased lymphocytes. Therefore, we focused on 50 genes more highly expressed 1 h post-NAC at V3 compared to V1A (Figure 5-5). Among the genes, 10 overlapped with the 108 cell type marker genes in the NanoString nCounter gene expression assay. The genes were associated with B cells (CD19 gene) and T cells (CD3E, CD3G, CD28, ICOS, CTLA4, STAT4, FLT3LG, GZMM, and DOCK9) (Figure 5-9). Particularly, Th1 cells, CD8+ T cells, and central memory T (Tcm) cells were associated. Although our analysis was limited from identifying more specific cell types, the immune cells associated with the cell type marker are likely to be involved in the immune response driven by Th1. Our results are consistent with the mechanisms involving redirection of allergen-specific Th2 response or immune deviation that have been reported by many studies of AITs (124–127).  5.4.5.3 Investigation of immune gene expression signatures at 1 h post-NAC             We next applied a different approach, namely the systemic immune response signature approach, to understand why the higher lymphocyte count at 1 h post-NAC at V3 compared to V1A is associated with clinical symptom change.      110  Figure 5-9 Cell type markers significantly more highly expressed at 1 h post NAC at V3 compared to V1A.  The NanoString nCounter PanCancer Immune Profiling Panel provides 108 of 730 immune genes as cell type markers. Among 50 significantly higher expressed immune genes at 1h post-NAC at V3, 10 genes were overlapped with the cell type markers. Tcm: central memory T cells.     111 5.4.5.3.1 Identification of five immune genes by a statistical test and a systemic immune gene signature approach             Our focus was on the increase of lymphocytes at 1 h post-NAC at V3. In order to investigate possible characteristics of the lymphocytes at 1h post-NAC of V3, we used the following analyses of gene expression data. First, we identified seven clusters of profiled immune genes at V3 (Figure 5-6 C). Fifty-three of 730 immune genes were retained for clustering after two filtering procedures: i) a statistical test using a LME, and ii) a fold-change threshold. FCM clustering, a widely used method to discover significant patterns in a given dataset, was used to partition the 53 genes of V3 into the clusters.             Next, we focused on the 50 genes significantly (BH-FDR < 0.1, paired LIMMA test) more highly expressed at 1 h post-NAC at V3 (Figure 5-5), based on our assumption that increased lymphocyte level at V3 may be correlated with higher expression of specific immune genes than at V1A. Five of the 53 immune genes of the clusters overlapped with the 50 immune genes more highly expressed at 1 h post-NAC at V3 compared to V1A: integrin subunit alpha E (ITGAE, CD103), CD180 (LY64), neural cell adhesion molecule 1 (NCAM1, CD56), C-C motif chemokine receptor 7 (CCR7, CD197), and leucine rich repeat neuronal 3 (LRRN3, NLRR3). ITGAE and NCAM1 were members of Cluster 1, with peak expression at 1 h post-NAC at V3. CD180 was a member of Cluster 4 that had a continuously increasing relative abundance after NAC. CCR7 and LRRN3 were members of Cluster 5 that had peaks at 6 h post-NAC (Figure 5-10).   112  Figure 5-10 The process to identify the five genes associated with the correlation between clinical symptom reduction and lymphocyte frequency change at 1 h post NAC in the 10 AR subjects.  Error bars: Mean ± SEM; * BH-FDR < 0.1, LIMMA paired test (V1A versus V3 at each time point) (n = 10).  5.4.5.3.2 Correlations between TNSS and the 5 immune gene expression difference (V3 - V1A)             Correlations were determined between the difference (V3-V1A) in the sum of TNSS and the difference in the gene expression ratio of the 5 identified immune genes. The gene expression ratio enabled comparison of the expression change of the genes on an equivalent scale compared to baseline at V1A. In the correlation tests, ITGAE and LRRN3 had moderate negative   113 correlations (r = -0.6357 and r = -0.6103, respectively), NCAM1 had a weak negative correlation (r = -0.4614), CCR7 had a weak negative correlation (r = -0.3421), and CD180 had a negligible correlation (r = -0.1011) (Figure 5-11 A).             We also tested correlations between differences (V3-V1A) in gene expression ratios and differences in lymphocyte count or ratios. CCR7 had weak positive correlations with lymphocyte count (r = 0.3526) and lymphocyte ratio (r = 0.4938), LRRN3 had weak positive correlations (r = 0.3912 and r = 0.3828, respectively), NCAM1 (r = 0.3834 and r = 0.2275), ITGAE had negligible correlations (r = -0.2893 and r = -0.1241), and CD180 also had negligible correlations (r = -0.0023 and r = 0.0607) (Figure 5-11 A).              The results were compared with those from 5 healthy non-allergic subjects (Table 5-1). We tested how the gene expression ratios at 1 h post-NAC of the AR subjects (V1A or V3) were different from the healthy non-allergic subjects who also underwent NAC. AR subjects at V3 and the healthy non-allergic subjects had no significant difference in the comparisons of gene expression ratios of the 5 immune genes (ratio: a relative number at each time point (at V1A and V3) compared with the baseline value of V1A in the AR subjects; a relative number at each time point compared with the baseline in the healthy non-allergic subjects). But, the gene expression ratios of LRRN3 and CD180 at V1A of the AR subjects were significantly (p < 0.05, t test/Wilcoxon rank sum test) differently expressed from the healthy non-allergic subjects (Figure 5-11 B). In the AR subjects, the identified genes, other than LRRN3 (p = 0.05997, paired t test), had significantly (p < 0.05, paired t test) higher expression at V3 than V1A in this comparison (Figure 5-11 B).    114  Figure 5-11 Comparisons of expression of the five genes in the 10 AR subjects. A, Pearson correlation of the difference (V3-V1A) in the sum of TNSS with the difference in the five gene expressions at 1 h post-NAC in the AR subjects. B, comparison of the gene expression ratios of the five genes at 1 h post-NAC at V3 with V1A in the AR subjects or at the time point of 5 healthy non-allergic subjects. C, the expression of IFN-γ over time points. Error bars: mean ± SEM. #p < 0.05, Wilcoxon rank sum test/t test (the AR   115 subjects versus the 5 healthy subjects at 1h post-NAC), *p < 0.05, Wilcoxon signed rank test/paired t test (V1A versus V3 at each time point in the AR subjects). (AR subjects, n = 10; non-allergic subjects, n = 5).  5.5 Discussion             Herein, we used a new approach, namely analysis of systemic immune gene signature in peripheral blood collected following a nasal allergen challenge (NAC), to further elucidate mechanism of action hypotheses of the novel immunotherapeutic Cat-PAD. This study was an open label clinical trial using a small population: 10 patients with allergic rhinitis (AR) who underwent NAC to test the efficacy of Cat-PAD without missing data. Cat-PAD is a peptide immunotherapy designed to modulate T cells through interaction of the small peptide epitopes and T-cell receptor (TCR) without the cross-linking of IgE on mast cells and basophils (122,129), and induced immune tolerance over a relatively short, standardised treatment course in the trial. Eight patients experienced reduction of total nasal symptom scores (TNSS) after allergen challenge at all measured time points following V3 NAC compared to V1A NAC. The V3 NAC was one month after the last dose of Cat-PAD administrated by intradermal injection, i.e., from 116 to 153 days (median, 126 days) following the V1A NAC. The benefit of the intervention was clear, given that subjects experienced reduction of TNSS in a relatively short time period, compared to traditional immunotherapies, which need to be administered for a couple of years before benefit is seen.             Interestingly, the reductions of the sum of TNSS over 0–6 h post-NAC (EPR.TNSS.Sum) or the sum of TNSS over 0–12 h post-NAC (ALL.TNSS.Sum) at V3 were strongly associated with both absolute and relative lymphocytes count differences at 1 h post-NAC (Figure 5-8). It was an interesting finding that the change in lymphocyte level at the early time point post-NAC could reflect the clinical symptom reduction over several hours and be associated with the   116 mechanism of the treatment. The lymphocyte counts were calculated as net of the input and output in the blood by migration, recirculation, as well as no cell response after allergen challenge. The relation between lymphocyte count and TNSS reduction may be reflective of a causality, reverse causality, or simultaneity. Although the relationship was seen in a small sample size (that may have biased information), it may be a snapshot of the mechanism explaining why the AR patients experienced the benefit of Cat-PAD following only a single short course treatment. We assumed that a subset of lymphocytes may be involved in the difference between V1A and V3. We used profiled canonical immune genes of peripheral blood collected following NAC to seek out candidate cell types associated with the higher lymphocytes counts at 1 h post-NAC at V3 in order to understand the mechanism of Cat-PAD. We approached our question using three ways of analyzing the immune gene expression data: i) pathway enrichment analysis; ii) cell type markers; and iii) systemic immune gene signatures. Pathway enrichment analysis at each time point demonstrated that immune genes with lower expression at V3 were associated with various inflammatory immune responses including IL-6, granulocyte-macrophage colony-stimulating factor (GM-CSF), B cell receptor (BCR) signaling and the interferon gamma (IFN-γ) pathways. The IL-6, GM-CSF and BCR signaling pathways are associated with the AR response driven by Th2 cells. While peripheral tolerance to allergens by multiple suppressive factors (e.g., IL-10, transforming growth factor (TGF)-β, cytotoxic T lymphocyte–associated antigen 4 and programmed death-1) has been reported, breaking tolerance was related to activity of myeloid dendritic cells, Toll-like receptor (TLR) 4 or TLR8, and the pro-inflammatory cytokines IL-1β, IL-4 or IL-6 (119). Furthermore, IL-6 is related with other types of differentiation of CD4+ T cells: IL-6 promotes T helper 17 (Th17) cells with TGF-β, and follicular helper T (Tfh) cells with IL-21 (130,131). The predominantly identified IL-6   117 signaling pathway may be important in the pathophysiology of AR and the development of allergic asthma (AA) given its relationship with Th17, an important effector cell in chronic inflammation (132).             Higher Th17 counts in peripheral blood of patients with AR or AA compared to healthy controls have previously been reported (133,134). The counts in AA were higher than AR, and the higher Th17/Treg ratio in peripheral blood was associated with severity of LPR in AA. Continuously triggered allergic immune responses may make AR symptoms more severe because of higher frequency of Th17 mediated by IL-6 (135), and may be associated with higher IL-17 serum levels in patients with severe AR (136,137). Perennial allergy such as cat allergy has a higher frequency of allergen exposure than seasonal allergy such as pollen allergy. It may be related to the previously reported result that the risk of developing AA in perennial AR is two times higher than in seasonal AR (138). The relationship may be reflected in the weak systemic immune responses of pollen allergy compared to cat allergy that we demonstrated in Chapter 4.  In contrast, more highly expressed immune genes at V3 were associated with changes of systemic immune response toward CD4+ and CD8+ T cell activation associated with IL-12 after Cat-PAD treatment. IL-12, mainly secreted from antigen presenting cells in response to pathogens, is also a pro-inflammatory heterodimeric cytokine like IL-6, but promotes CD4+ cell differentiation into Th1 cells through induction of transcription factor T-bet (139–141). A positive relationship between clinical symptom reduction, inhibition of the late cutaneous response, and an increase in the number of IL-12 expressing cells – predominantly macrophages (CD68 +) – after grass pollen immunotherapy was reported at allergen challenge sites in AR patients 24 h following allergen challenge. The IL-12 expressed cells had a positive correlation with IFN-γ expressing cells and a negative correlation with IL-4 expressing cells (124,142,143).   118   Figure 5-12 Potential mechanism of the action of Cat-PAD.  Comparison of immune gene expression between V1A and V3 NAC visits suggested that the systemic immune responses were changed from type 1 immune response to type 2 immune response after the immunotherapy. The changes were likely associated with immune tolerance such as central and peripheral tolerance.              The change of immune responses in the peripheral blood between pre- and post-treatment had a trend toward type 1 immune response associated with Th1 activation from type 2 immune   119 response associated with Th2 activation after the immune therapy (Figure 5-12). The significantly higher expression of IFN-γ at 6 h post-NAC at V3 (Figure 5-11 C) supports the trend. This process is considered as a process to achieve the immune tolerance by immunotherapy (144). The proportion of T cell subtypes in peripheral blood of allergic subjects were significantly different from healthy non-allergic subjects (113,145). The changes of immune cell subtype proportions and immune responses were derived by central and peripheral tolerance. Regulatory T cells (Treg), regulatory B cells (Breg), and dendritic cells (DC) involve in the immune tolerance (146–149). Even not only activation but also apoptosis of DC controls the immune tolerance progress (150). Plasmacytoid dendritic cells (pDC) and conventional dendritic cells (cDC) after uptake antigens migrate into thymus from the blood and can promote central tolerance such as clonal deletion and Treg induction (151).             Cell type marker and systemic immune response analyses allowed us to study more possible candidate cell types of the lymphocytes associated with clinical symptom reduction. Cell type marker analysis suggested T helper cells, especially Th1, CD8+ T cells, Tcm and B cells.              The systemic immune gene signature approach identified five significant immune genes: ITGAE (CD103), CD180 (LY64), NCAM1 (CD56), CCR7 (CD197), and LRRN3 (NLRR3). The five genes identified may represent altered immune responses by one or more subsets of immune cells associated with the mechanism of Cat-PAD (i.e., proportions of Th1, Th2, Treg, ILCs, and DC cells) (Figure 5-13). The higher expression of CCR7 at 1 h post-NAC at V3 was associated with clinical symptom reduction and lymphocyte frequency. It may be associated with allergen-specific CD27+ CCR7+ CD4+ memory T cells (152). The immune cell subset found in non-allergic subjects secretes IFN-γ and represents allergen-specific Th1 and Treg1 cells.   120 Significantly (p < 0.05, paired t test) highly expressed IFN-γ at 6 h post-NAC at V3 compared to V1A may therefore be associated with the mechanism of the treatment (Figure 5-11 C). LRRN3 in plasma is associated with activation of MAPK activity and endocytosis and control of cholesterol level (153). Although it had no correlation with clinical symptom reduction, CD180 expression was higher on IgM+, IgG+, and IgA+ memory B cells than naïve mature B cells except for IgE+ memory B cells (154). ITGAE and NCAM1 are markers of DC cells and NK cells, respectively. Both are associated with CD8+ Treg cells, NKp44+ ITGAE + ILC1s, NKp44+ NK cells, and plasmacytoid dendritic cell (pDC), which play roles in the immune responses by cell-mediated immunity and immune tolerance (155–158).              ITGAE and cytotoxic T-lymphocyte associated protein 4 (CTLA4) expression are associated with the phenotype and function of Treg cells (146,147). Some subset of conventional DC (e.g., Batf3-dependent cDC subsets) and pDC express ITGAE for migration and they are involved in CD8+ T cell activation (157,159). Highly expressed ITGAE and NCAM1 were important for activation of pDC for migration and antigen presentation with cytolytic activity (157,160). Production of type I IFNs, IL-12 and IL-18 by pDCs supports the accumulation and effector functions of CD8+ T cells and NK cells, as well as the differentiation of CD4+ T cells into Treg cells and Th1 cells (157,161).     121  Figure 5-13 Potential immune cells associated with the mechanism of the action of Cat-PAD.  The subtypes of T cells, innate lymphoid cells, and dendritic cells were likely associated with the identified five immune genes and higher expressed IFN-γ at V3 (post-treatment). Error bars: Mean ± SEM; * BH-FDR < 0.1, LIMMA paired test (V1A versus V3 at each time point) (n = 10).              NK cells, which are defined as NCAM1+ CD3- in humans, represent 10% of peripheral blood mononuclear cells (PBMC) comprising the third largest population of lymphocytes following B and T cells. Human NK cells have two subsets based on NCAM1 expression: NCAM1dim, which are 90% of NK cells; and NCAM1bright, which are 10% of NK cells. NCAM1bright NK cells may have an important role in early immune response, involved in   122 producing cytokines such as IFN-γ, GM-CSF and IL-10. ITGAE + NK cells produce IL-22 stimulated epithelial cells to secrete IL-10 to reduce inflammation. The role of NK cells is found to be either allergy-controlling or -enhancing (158,162,163). Group 1 innate lymphoid cells (ILCs) comprise NK cells and ILC1s that reside within peripheral tissues. ILC1s have an important role in the innate immune response and also secrets IFN-γ (164).              As a small scale pilot study, the limitations of our study include its small sample size, lack of placebo group, and limited information – such as no detailed lymphocyte subset frequencies. The association between the five genes and lymphocyte subset frequencies needs to be further investigated. However, we could identify the utility of the systemic immune response signature approach to explore the various aspects of the pathophysiology of AR and mechanisms of AIT. This approach helped us build hypotheses to explore the mechanisms of Cat-PAD immunotherapy by providing candidate cells associated with clinical symptom reduction given the limited information available on lymphocyte subset frequencies. Our result suggests that the change of lymphocyte subsets associated with the differently expressed genes may be an important process for the induction of tolerance by the immunotherapy. Future large-scale studies and follow-up mechanistic studies may validate this work by identifying the subsets of leukocytes associated with immunotherapies and providing biomarkers to better phenotype patients for precision medicine.    123 Chapter 6: Conclusions and future directions  6.1 Overall summary and conclusions             The defense mechanisms of the immune system distinguishing self and a non-self as an invader is important to protect human body and keep homeostasis. Even computer scientists are fascinated by the immune system, developing biologically-inspired intrusion detection algorithms modeled on the immune response in information technology (165). In vertebrates, there are two types of immune system: the innate and adaptive immune systems. Immune responses are also categorized into type 1 and type 2 immune responses (30,166). The immune system has plasticity of immune response depending on the pathogen type such as bacteria, viruses, helminths, and cancer. T cell plasticity is crucial to promote more effective immunity (167). This plasticity, however, is also associated with immune-based diseases such as allergies, autoimmunity, and cancer.             A bias towards Th2 immune responses inducing abnormal hypersensitive reactions to allergens results in allergic inflammation, causing allergic disorders (7,21). Allergic rhinitis (AR) is developed by allergic inflammation of the nasal mucosa. While many studies have focused on the local site of inflammation to investigate the pathophysiology of allergic rhinitis, the pathophysiological aspect of the systemic immune responses of AR is not well elucidated. This dissertation suggests a novel approach, the systemic immune gene signature approach, to investigate systemic immune responses associated with the pathophysiology and mechanism of the action of AR treatment.    124  Figure 6-1 Systemic immune response signatures, combinations of immune cell frequency and immune gene clusters found in Chapter 3.  Ratio: a relative number at each time point post-NAC compared to the baseline value in the subjects.    125             In Chapter 3, we tested the hypothesis “Systemic immune response signatures are associated with the pathophysiology of AR, and can be determined from immune gene expression and immune cell frequencies measured in peripheral blood collected following allergen challenge in subjects with cat allergy.” Clustered seven immune gene sets based on patterns of time series gene expression data in peripheral blood from subjects with AR undergoing NAC were tested whether the clusters were associated with immune cell frequencies and clinical symptoms. We identified systemic immune gene signatures from the clusters, traceable signatures of immune gene expression clusters corresponding to immune cell frequencies (Figure 6-1).              In Chapter 4, we tested the hypothesis “The identified systemic immune response signatures in cat allergy will be validated in birch and ragweed allergies using peripheral blood collected after allergen challenge in NAC or EEU models.” For this,  systemic immune response signatures identified in cat allergy were compared to the results in birch and ragweed allergies using peripheral blood collected after allergen challenge in NAC and/or EEU models. Our pollen allergy results demonstrated less significant changes after allergen challenge as previous studies had reported that seasonal allergy has less clinical symptoms than persistent allergy such as cat allergy. Ragweed allergy in EEU model demonstrated more significant results than birch allergy based on cluster trend analysis in the given limited condition. It may be associated with the stronger protease activity of ragweed than birch.              In Chapter 5, we tested the hypothesis that identified systemic immune response signatures would be differentially expressed pre- and post-immunotherapy, and would provide a cross-sectional view to investigate the mechanism of action of the immunotherapy. We found dynamic changes in immune cell frequencies and immune gene signatures of peripheral blood   126 following immune therapy. The systemic immune response signature approach identified five immune genes that may be associated with lymphocyte difference pre- and post-immunotherapy at 1h post-NAC. These can be further studied to help elucidate the mechanism of action of the immune therapy. Through our research we addressed the main hypothesis that clustering of profiled immune gene expression data from peripheral blood collected following allergen challenge would identify systemic immune response signatures associated with the pathophysiology of AR.             We concluded that peripheral blood collected following allergen challenge reflected the systemic immune response, whose signature, a combination of patterns of immune gene expression and immune cell frequencies, may reflect various aspects of the pathophysiology of AR depending on the allergen and the changes of systemic immune response induced by an immunotherapy.  6.2 Strengths and Limitations 6.2.1 Strengths             Clustering methods are an unsupervised learning method for exploratory data analysis that are used to discover unknown subgroups in data. Although unsupervised learning methods, such as principal components analysis, are of growing importance in exploratory data analysis, they are difficult to interpret in order to predict responses (Y) using features (X); in contrast, supervised learning methods, such as linear regression models, can be more easily used for prediction (168). The immune gene signatures identified in Chapter 3 were based on immune gene clusters grouped by Fuzzy c-mean clustering, an unsupervised learning method. We, however, incorporated immune cell frequency information into the clusters of immune gene   127 expression using a cell enrichment analysis and canonical correlation analysis. The combination made the identified systemic immune response signatures accessible indicators to investigate the pathophysiological systemic immune responses of AR.              Although our sample size was small, in Chapter 3 we validated the identified systemic immune response signatures using the test cohort of Queen’s University with an independent procedure using a validation cohort from McMaster University. Investigation of birch and ragweed allergy in nasal allergen challenge (NAC) or with an environmental exposure unit (EEU) model provided an understanding about different intensities of systemic immune responses of pollen allergy compared to cat allergy, a perennial allergy, in Chapter 4. In Chapter 5 we assessed the utility of applying the systemic immune response approach in further study using samples collected pre- and post-immunotherapy (i.e., repeated measurements in a longitudinal study).   6.2.2 Limitations             In our studies, we used immune cell frequencies in CBC data that provided a reliable count of different subtypes of leukocytes as a standard diagnostic tool, e.g., neutrophils, lymphocytes, monocytes, and eosinophils (81–83). But CBC data is unable to investigate the frequencies of leukocyte subsets such as Th1, Th2 and ILC2s. Detailed frequency information of immune cell subsets, collected by observation of the local inflammatory sites, blood, and reservoirs of immune cells, is needed to fill in the gaps in the systemic immune response approach. We identified dynamic trajectories of systemic immune response signatures through time series blood collection at baseline, 1, 2 and 6 h post-NAC in Chapters 3 and 5, but using just two time points pre- and post-allergen challenge, it was difficult to investigate the dynamic   128 change of systemic immune responses (as seen in Chapter 4). The limited time points used for collecting blood samples and clinical symptom scores were a weakness of our study. The absence of placebo or healthy non-allergic subject groups for cat allergy and ragweed allergy were a limitation, and did not permit us to consider the influence of factors such as stress, diurnal variation, and dehydration. The studies in this dissertation were pilot studies using small sample sizes comprising predominantly Caucasians. They may therefore be more sensitive to outliers than a large-scale study with randomized sampling.  6.3 Future directions             The necessity of improving diagnostic methods and subcategorizing AR phenotypes is increasing due to the limitations of current diagnosis methods and the variety of different phenotypes of AR that could benefit from directed, personalized treatment strategies. Disagreement between the sIgE test and skin prick test (SPT), established methods to diagnose AR, has been reported in allergic subjects in previous studies (169,170). 44 -87% of patients with rhinitis may have mixed rhinitis, a combination of AR and non-allergic rhinitis (3). Allergic responses of AR demonstrate various clinical features depending on the allergen type (perennial and seasonal AR), symptom duration (persistent and intermittent AR), SPT results (AR and local AR), and sensitization number (monosensitization and polysensitization) (8,12,171).  The systemic immune response signature approach identified in this study may provide a way to improve diagnostic measurements, evaluate AR and test the efficacy of AR treatment based on a better understanding of the pathophysiology of AR and the mechanisms of AR treatment. The benefits of such an approach are clear, but large-scale studies with various AR samples and randomized samples are necessary to identify critical systemic immune response signatures and   129 consider confounding factors and causalities. Moreover, gene expression data of isolated leukocyte subsets and detailed frequency information of leukocyte subsets by flow cytometry and/or mass cytometry may elaborate the systemic immune response signature approach that we suggest through this dissertation. This approach may also be used to investigate the pathophysiology and the mechanisms of treatments in other diseases associated with systemic immune responses, such as allergic asthma.   130 Bibliography  1.  Dykewicz MS, Hamilos DL. Rhinitis and sinusitis. J Allergy Clin Immunol. 2010 Feb 1;125(2):S103–15.  2.  Scadding GK, Kariyawasam HH, Scadding G, Mirakian R, Buckley RJ, Dixon T, et al. BSACI guideline for the diagnosis and management of allergic and non-allergic rhinitis (Revised Edition 2017; First edition 2007). Clin Exp Allergy. 2017 Jul 1;47(7):856–89.  3.  Wallace DV, Dykewicz MS, Bernstein DI, Blessing-Moore J, Cox L, Khan DA, et al. The diagnosis and management of rhinitis: An updated practice parameter. J Allergy Clin Immunol. 2008 Aug;122(2, Supplement):S1–84.  4.  Pattanaik D, Lieberman P. Vasomotor Rhinitis. Curr Allergy Asthma Rep. 2010 Mar 1;10(2):84–91.  5.  Halderman A, Sindwani R. Surgical management of vasomotor rhinitis: a systematic review. Am J Rhinol Allergy. 2015 Apr 3;29(2):128–34.  6.  Sin B, Togias A. Pathophysiology of Allergic and Nonallergic Rhinitis. Proc Am Thorac Soc. 2011 Mar 1;8(1):106–14.  7.  Wheatley LM, Togias A. Allergic Rhinitis. N Engl J Med. 2015 Jan 29;372(5):456–63.  8.  Rondón C, Romero JJ, López S, Antúnez C, Martín-Casañez E, Torres MJ, et al. Local IgE production and positive nasal provocation test in patients with persistent nonallergic rhinitis. J Allergy Clin Immunol. 2007 Apr 1;119(4):899–905.  9.  De Martinis M, Sirufo MM, Ginaldi L. Allergy and Aging: An Old/New Emerging Health Issue. Aging Dis. 2017 Apr;8(2):162–75.  10.  Skoner DP. Complications of allergic rhinitis. J Allergy Clin Immunol. 2000 Jun 1;105(6, Part 2):S605–9.  11.  Sly RM. Changing prevalence of allergic rhinitis and asthma. Ann Allergy Asthma Immunol. 1999 Mar 1;82(3):233–52.  12.  Bousquet J, Khaltaev N, Cruz AA, Denburg J, Fokkens WJ, Togias A, et al. Allergic Rhinitis and its Impact on Asthma (ARIA) 2008*. Allergy. 2008 Apr 1;63:8–160.  13.  Pawankar R, Mori S, Ozu C, Kimura S. Overview on the pathomechanisms of allergic rhinitis. Asia Pac Allergy. 2011 Oct;1(3):157–67.    131 14.  Katelaris CH, Lee BW, Potter PC, Maspero JF, Cingi C, Lopatin A, et al. Prevalence and diversity of allergic rhinitis in regions of the world beyond Europe and North America. Clin Exp Allergy. 2012 Feb 1;42(2):186–207.  15.  Keith PK, Desrosiers M, Laister T, Schellenberg RR, Waserman S. The burden of allergic rhinitis (AR) in Canada: perspectives of physicians and patients. Allergy Asthma Clin Immunol Off J Can Soc Allergy Clin Immunol. 2012 Jun 1;8(1):7.  16.  Dranitsaris G, Ellis AK. Sublingual or subcutaneous immunotherapy for seasonal allergic rhinitis: an indirect analysis of efficacy, safety and cost. J Eval Clin Pract. 2014 Jun 1;20(3):225–38.  17.  Hansel FK. Clinical and histopathologic studies of the nose and sinuses in allergy. J Allergy. 1929 Nov 1;1(1):43–70.  18.  Ciprandi G, Cirillo I, Vizzaccaro A, Tosca M, Passalacqua G, Pallestrini E, et al. Seasonal and perennial allergic rhinitis: is this classification adherent to real life? Allergy. 2005 Jul 1;60(7):882–7.  19.  Calderon MA, Alves B, Jacobson M, Hurwitz B, Sheikh A, Durham S. Cochrane review: Allergen injection immunotherapy for seasonal allergic rhinitis. Evid-Based Child Health Cochrane Rev J. 2010 Sep 1;5(3):1279–379.  20.  Broide DH. Allergic rhinitis: Pathophysiology. Allergy Asthma Proc. 2010 Oct 9;31(5):370–4.  21.  Galli SJ, Tsai M, Piliponsky AM. The development of allergic inflammation. Nature. 2008 Jul 24;454(7203):445–54.  22.  Tomazic PV, Birner-Gruenberger R, Leitner A, Obrist B, Spoerk S, Lang-Loidolt D. Nasal mucus proteomic changes reflect altered immune responses and epithelial permeability in patients with allergic rhinitis. J Allergy Clin Immunol. 2014 Mar 1;133(3):741–50.  23.  Natsuga K. Epidermal Barriers. Cold Spring Harb Perspect Med [Internet]. 2014 Apr;4(4). Available from: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3968788/ 24.  Nawijn MC, Hackett TL, Postma DS, van Oosterhout AJM, Heijink IH. E-cadherin: gatekeeper of airway mucosa and allergic sensitization. Trends Immunol. 2011 Jun 1;32(6):248–55.  25.  Kubo A, Nagao K, Amagai M. Epidermal barrier dysfunction and cutaneous sensitization in atopic diseases. J Clin Invest. 2012 Feb 1;122(2):440–7.  26.  Gandhi VD, Vliagoftis H. Airway Epithelium Interactions with Aeroallergens: Role of Secreted Cytokines and Chemokines in Innate Immunity. Front Immunol [Internet]. 2015 Apr 2;6. Available from: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4382984/   132 27.  De Bruin-Weller, Weller, De Monchy. Repeated allergen challenge as a new research model for studying allergic reactions. Clin Exp Allergy. 1999 Feb 1;29(2):159–65.  28.  Howarth PH. Allergic rhinitis: not purely a histamine-related disease. Allergy. 2000 Dec 1;55:7–16.  29.  Borriello F, Iannone R, Marone G. Histamine Release from Mast Cells and Basophils. In: Histamine and Histamine Receptors in Health and Disease [Internet]. Springer, Cham; 2017 [cited 2018 Mar 7]. p. 121–39. (Handbook of Experimental Pharmacology). Available from: https://link-springer-com.ezproxy.library.ubc.ca/chapter/10.1007/164_2017_18 30.  Licona-Limón P, Kim LK, Palm NW, Flavell RA. TH2, allergy and group 2 innate lymphoid cells. Nat Immunol. 2013 Jun;14(6):536–42.  31.  Hammad H, Lambrecht BN. Barrier Epithelial Cells and the Control of Type 2 Immunity. Immunity. 2015 Jul 21;43(1):29–40.  32.  Cosmi L, Liotta F, Maggi L, Annunziato F. Role of Type 2 Innate Lymphoid Cells in Allergic Diseases. Curr Allergy Asthma Rep. 2017 Sep 11;17(10):66.  33.  Peters-Golden M, Gleason M, Togias A. Cysteinyl leukotrienes: multi-functional mediators in allergic rhinitis. Clin Exp Allergy. 2006 Jun 1;36(6):689–703.  34.  Shamji MH, Bellido V, Scadding GW, Layhadi JA, Cheung DKM, Calderon MA, et al. Effector cell signature in peripheral blood following nasal allergen challenge in grass pollen allergic individuals. Allergy. 2015 Feb 1;70(2):171–9.  35.  Pucci S, Incorvaia C. Allergy as an organ and a systemic disease. Clin Exp Immunol. 2008 Sep 1;153:1–2.  36.  Corren J. Allergic rhinitis and asthma: How important is the link? J Allergy Clin Immunol. 1997 Feb 1;99(2):S781–6.  37.  Goksör E, Loid P, Alm B, Åberg N, Wennergren G. The allergic march comprises the coexistence of related patterns of allergic disease not just the progressive development of one disease. Acta Paediatr. 2016 Dec 1;105(12):1472–9.  38.  Dholaria B, Bhasin A, Krishna M, Finn L. Adoptive transfer of food allergy via unrelated allogeneic bone marrow transplant. Ann Allergy Asthma Immunol. 2016 Jul;117(1):96–7.  39.  Garzorz N, Thomas J, Eberlein B, Haferlach C, Ring J, Biedermann T, et al. Newly acquired kiwi fruit allergy after bone marrow transplantation from a kiwi-allergic donor. J Eur Acad Dermatol Venereol. 2016 Jul 1;30(7):1136–9.  40.  Powe DG, Bonnin AJ, Jones NS. ‘Entopy’: local allergy paradigm. Clin Exp Allergy. 2010 Jul 1;40(7):987–97.    133 41.  Campo P, Rondón C, Gould HJ, Barrionuevo E, Gevaert P, Blanca M. Local IgE in non-allergic rhinitis. Clin Exp Allergy. 2015 May 1;45(5):872–81.  42.  Ellis AK, North ML, Walker T, Steacy LM. Environmental exposure unit: a sensitive, specific, and reproducible methodology for allergen challenge. Ann Allergy Asthma Immunol. 2013 Nov;111(5):323–8.  43.  North ML, Soliman M, Walker T, Steacy LM, Ellis AK. Controlled Allergen Challenge Facilities and Their Unique Contributions to Allergic Rhinitis Research. Curr Allergy Asthma Rep. 2015 Apr 1;15(4):11.  44.  Hohlfeld JM, Holland-Letz T, Larbig M, Lavae-Mokhtari M, Wierenga E, Kapsenberg M, et al. Diagnostic value of outcome measures following allergen exposure in an environmental challenge chamber compared with natural conditions. Clin Exp Allergy. 2010 Jul;40(7):998–1006.  45.  Mygind N, Johnsen NJ, Thomsen J. Intranasal allergen challenge during corticosteroid treatment. Clin Exp Allergy. 1977 Jan 1;7(1):69–74.  46.  Holmberg K, Bake B, Pipkorn U. Nasal mucosal blood flow after intranasal allergen challenge. J Allergy Clin Immunol. 1988 Mar 1;81(3):541–7.  47.  Litvyakova LI, Baraniuk JN. Nasal provocation testing: a review. Ann Allergy Asthma Immunol. 2001 Apr 1;86(4):355–65.  48.  Scadding GW, Eifan A, Penagos M, Dumitru A, Switzer A, McMahon O, et al. Local and systemic effects of cat allergen nasal provocation. Clin Exp Allergy. 2015 Mar 1;45(3):613–23.  49.  Ellis AK, Soliman M, Steacy L, Boulay M-È, Boulet L-P, Keith PK, et al. The Allergic Rhinitis – Clinical Investigator Collaborative (AR-CIC): nasal allergen challenge protocol optimization for studying AR pathophysiology and evaluating novel therapies. Allergy Asthma Clin Immunol. 2015 Apr 24;11(1):16.  50.  Akerlund A, Andersson M, Leflein J, Lildholdt T, Mygind N. Clinical trial design, nasal allergen challenge models, and considerations of relevance to pediatrics, nasal polyposis, and different classes of medication. J Allergy Clin Immunol. 2005 Mar;115(3, Supplement):S460–82.  51.  Day JH, Ellis AK, Rafeiro E, Ratz JD, Briscoe MP. Experimental models for the evaluation of treatment of allergic rhinitis. Ann Allergy Asthma Immunol. 2006 Feb 1;96(2):263–78.  52.  Ellis AK, Steacy LM, Hobsbawn B, Conway CE, Walker TJ. Clinical validation of controlled grass pollen challenge in the Environmental Exposure Unit (EEU). Allergy Asthma Clin Immunol. 2015 Jan 27;11:5.    134 53.  Heffer MJ, Ratz JD, Miller JD, Day JH. Comparison of the Rotorod to other air samplers for the determination of Ambrosia artemisiifolia pollen concentrations conducted in the Environmental Exposure Unit. Aerobiologia. 2005 Sep 1;21(3–4):233–9.  54.  Bar-Joseph Z, Gitter A, Simon I. Studying and modelling dynamic biological processes using time-series gene expression data. Nat Rev Genet. 2012 Aug;13(8):552–64.  55.  Levine JH, Simonds EF, Bendall SC, Davis KL, Amir ED, Tadmor MD, et al. Data-Driven Phenotypic Dissection of AML Reveals Progenitor-like Cells that Correlate with Prognosis. Cell. 2015 Jul 2;162(1):184–97.  56.  Wise A, Bar-Joseph Z. SMARTS: reconstructing disease response networks from multiple individuals using time series gene expression data. Bioinformatics. 2015 Apr 15;31(8):1250–7.  57.  Leng N, Chu L-F, Barry C, Li Y, Choi J, Li X, et al. Oscope identifies oscillatory genes in unsynchronized single-cell RNA-seq experiments. Nat Methods. 2015 Oct;12(10):947–50.  58.  Ellis AK, Ratz JD, Day AG, Day JH. Factors that affect the allergic rhinitis response to ragweed allergen exposure. Ann Allergy Asthma Immunol. 2010 Apr 1;104(4):293–8.  59.  Starling-Schwanz R, Peake HL, Salome CM, Toelle BG, Ng KW, Marks GB, et al. Repeatability of peak nasal inspiratory flow measurements and utility for assessing the severity of rhinitis. Allergy. 2005 Jun;60(6):795–800.  60.  Malkov VA, Serikawa KA, Balantac N, Watters J, Geiss G, Mashadi-Hossein A, et al. Multiplexed measurements of gene signatures in different analytes using the Nanostring nCounterTM Assay System. BMC Res Notes. 2009 May 9;2(1):80.  61.  Reis PP, Waldron L, Goswami RS, Xu W, Xuan Y, Perez-Ordonez B, et al. mRNA transcript quantification in archival samples using multiplexed, color-coded probes. BMC Biotechnol. 2011;11:46.  62.  Veldman-Jones MH, Brant R, Rooney C, Geh C, Emery H, Harbron CG, et al. Evaluating Robustness and Sensitivity of the NanoString Technologies nCounter Platform to Enable Multiplexed Gene Expression Analysis of Clinical Samples. Cancer Res. 2015 Jul 1;75(13):2587–93.  63.  Johnson WE, Li C, Rabinovic A. Adjusting batch effects in microarray expression data using empirical Bayes methods. Biostatistics. 2007 Jan 1;8(1):118–27.  64.  Leek JT, Johnson WE, Parker HS, Jaffe AE, Storey JD. The sva package for removing batch effects and other unwanted variation in high-throughput experiments. Bioinformatics. 2012 Mar 15;28(6):882–3.    135 65.  Soneson C, Gerster S, Delorenzi M. Batch Effect Confounding Leads to Strong Bias in Performance Estimates Obtained by Cross-Validation. PLoS ONE. 2014 Jun 26;9(6):e100335.  66.  Irizarry RA, Hobbs B, Collin F, Beazer-Barclay YD, Antonellis KJ, Scherf U, et al. Exploration, normalization, and summaries of high density oligonucleotide array probe level data. Biostatistics. 2003 Apr 1;4(2):249–64.  67.  Leski JM. Fuzzy c-ordered-means clustering. Fuzzy Sets Syst. 2016 Mar 1;286:114–33.  68.  Zhou K, Yang S. Exploring the uniform effect of FCM clustering: A data distribution perspective. Knowl-Based Syst. 2016 Mar 15;96:76–83.  69.  Kumar L, E. Futschik M. Mfuzz: A software package for soft clustering of microarray data. Bioinformation. 2007 May 20;2(1):5–7.  70.  Chen EY, Tan CM, Kou Y, Duan Q, Wang Z, Meirelles G, et al. Enrichr: interactive and collaborative HTML5 gene list enrichment analysis tool. BMC Bioinformatics. 2013;14(1):128.  71.  Kuleshov MV, Jones MR, Rouillard AD, Fernandez NF, Duan Q, Wang Z, et al. Enrichr: a comprehensive gene set enrichment analysis web server 2016 update. Nucleic Acids Res. 2016 Jul 8;44(W1):W90-97.  72.  González I, Cao K-AL, Davis MJ, Déjean S. Visualising associations between paired ‘omics’ data sets. BioData Min. 2012;5:19.  73.  Frew AJ, Kay AB. The relationship between infiltrating CD4+ lymphocytes, activated eosinophils, and the magnitude of the allergen-induced late phase cutaneous reaction in man. J Immunol. 1988 Dec 15;141(12):4158–64.  74.  Bonini S, Bonini S, Bucci MG, Berruto A, Adriani E, Balsano F, et al. Allergen dose response and late symptoms in a human model of ocular allergy. J Allergy Clin Immunol. 1990 Dec 1;86(6):869–76.  75.  Dogru M, Evcimik MF, Cirik AA. Is neutrophil–lymphocyte ratio associated with the severity of allergic rhinitis in children? Eur Arch Otorhinolaryngol. 2015 Nov 2;1–4.  76.  Templeton AJ, McNamara MG, Šeruga B, Vera-Badillo FE, Aneja P, Ocaña A, et al. Prognostic Role of Neutrophil-to-Lymphocyte Ratio in Solid Tumors: A Systematic Review and Meta-Analysis. J Natl Cancer Inst. 2014 Jun 1;106(6):dju124.  77.  Thompson B. Canonical Correlation Analysis. In: Encyclopedia of Statistics in Behavioral Science [Internet]. John Wiley & Sons, Ltd; 2005 [cited 2017 Apr 20]. Available from: http://onlinelibrary.wiley.com/doi/10.1002/0470013192.bsa068/abstract   136 78.  Scadding GW, Calderon MA, Bellido V, Koed GK, Nielsen N-C, Lund K, et al. Optimisation of grass pollen nasal allergen challenge for assessment of clinical and immunological outcomes. J Immunol Methods. 2012 Oct 31;384(1–2):25–32.  79.  Naclerio RM. Pathophysiology of perennial allergic rhinitis. Allergy. 1997;52(36 Suppl):7–13.  80.  Bacon AS, Ahluwalia P, Irani A-M, Schwartz LB, Holgate ST, Church MK, et al. Tear and conjunctival changes during the allergen-induced early- and late-phase responses. J Allergy Clin Immunol. 2000 Nov;106(5):948–54.  81.  Meintker L, Ringwald J, Rauh M, Krause SW. Comparison of Automated Differential Blood Cell Counts From Abbott Sapphire, Siemens Advia 120, Beckman Coulter DxH 800, and Sysmex XE-2100 in Normal and Pathologic Samples. Am J Clin Pathol. 2013 May 1;139(5):641–50.  82.  Danise P, Maconi M, Rovetti A, Avino D, Di Palma A, Gerardo Pirofalo M, et al. Cell counting of body fluids: comparison between three automated haematology analysers and the manual microscope method. Int J Lab Hematol. 2013 Dec 1;35(6):608–13.  83.  Seo JY, Lee S-T, Kim S-H. Performance evaluation of the new hematology analyzer Sysmex XN-series. Int J Lab Hematol. 2015 Apr 1;37(2):155–64.  84.  Voehringer D. Protective and pathological roles of mast cells and basophils. Nat Rev Immunol. 2013 May;13(5):362–75.  85.  Vancheri C, Mastruzzo C, Armato F, Tomaselli V, Magrì S, Pistorio MP, et al. Intranasal heparin reduces eosinophil recruitment after nasal allergen challenge in patients with allergic rhinitis. J Allergy Clin Immunol. 2001 Nov;108(5):703–8.  86.  Galli SJ, Tsai M. IgE and mast cells in allergic disease. Nat Med. 2012 May;18(5):693–704.  87.  Eifan AO, Durham SR. Pathogenesis of rhinitis. Clin Exp Allergy. 2016 Sep 1;46(9):1139–51.  88.  Eguíluz-Gracia I, Bosco A, Dollner R, Melum GR, Lexberg MH, Jones AC, et al. Rapid recruitment of CD14+ monocytes in experimentally induced allergic rhinitis in human subjects. J Allergy Clin Immunol. 2016 Jun;137(6):1872–1881.e12.  89.  Braunstahl G-J, Overbeek SE, KleinJan A, Prins J-B, Hoogsteden HC, Fokkens WJ. Nasal allergen provocation induces adhesion molecule expression and tissue eosinophilia in upper and lower airways. J Allergy Clin Immunol. 2001 Mar;107(3):469–76.  90.  Kolaczkowska E, Kubes P. Neutrophil recruitment and function in health and inflammation. Nat Rev Immunol. 2013 Mar;13(3):159–75.    137 91.  Doherty TA, Scott D, Walford HH, Khorram N, Lund S, Baum R, et al. Allergen challenge in allergic rhinitis rapidly induces increased peripheral blood type 2 innate lymphoid cells that express CD84. J Allergy Clin Immunol. 2014 Apr;133(4):1203–1205.e7.  92.  Davies JM, Platts-Mills TA, Aalberse RC. The enigma of IgE+ B-cell memory in human subjects. J Allergy Clin Immunol. 2013 Apr;131(4):972–6.  93.  Henriques A, Nunes R, Loureiro G, Martinho A, Pais M, Segorbe-Luís A, et al. Alterations on peripheral blood B cell subsets induced by allergic rhinitis. Inflamm Res. 2015 Feb 21;64(3–4):145–9.  94.  Looney TJ, Lee J-Y, Roskin KM, Hoh RA, King J, Glanville J, et al. Human B-cell isotype switching origins of IgE. J Allergy Clin Immunol. 2016 Feb;137(2):579–586.e7.  95.  Thornton MA, Akasheh N, Walsh M-T, Moloney M, Sheahan PO, Smyth CM, et al. Eosinophil recruitment to nasal nerves after allergen challenge in allergic rhinitis. Clin Immunol. 2013 Apr;147(1):50–7.  96.  Undem BJ, Taylor-Clark T. Mechanisms underlying the neuronal-based symptoms of allergy. J Allergy Clin Immunol. 2014 Jun;133(6):1521–34.  97.  Herre J, Grönlund H, Brooks H, Hopkins L, Waggoner L, Murton B, et al. Allergens as Immuno-Modulatory Proteins: the cat dander protein Fel d 1 enhances Toll-like receptor activation by lipid ligands. J Immunol Baltim Md 1950 [Internet]. 2013 Aug 15;191(4). Available from: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3836235/ 98.  Asam C, Hofer H, Wolf M, Aglas L, Wallner M. Tree pollen allergens—an update from a molecular perspective. Allergy. 2015 Oct 1;70(10):1201–11.  99.  Gergen PJ, Turkeltaub PC, Kovar MG. The prevalence of allergic skin test reactivity to eight common aeroallergens in the U.S. population: Results from the second National Health and Nutrition Examination Survey. J Allergy Clin Immunol. 1987 Nov 1;80(5):669–79.  100.  Ciprandi G, Comite P, Ferrero F, Bignardi D, Minale P, Voltolini S, et al. Birch allergy and oral allergy syndrome: The practical relevance of serum immunoglobulin E to Bet v 1. Allergy Asthma Proc. 2016 Feb 1;37(1):43–9.  101.  Jensen-Jarolim E. Happy 25th birthday, Bet v 1! World Allergy Organ J. 2014 Jun 3;7:14.  102.  Kleine-Tebbe J, Ballmer-Weber BK, Breiteneder H, Vieths S. Bet v 1 and its Homologs: Triggers of Tree-Pollen Allergy and Birch Pollen-Associated Cross-Reactions. In: Molecular Allergy Diagnostics [Internet]. Springer, Cham; 2017 [cited 2017 Nov 5]. p. 21–42. Available from: https://link-springer-com.ezproxy.library.ubc.ca/chapter/10.1007/978-3-319-42499-6_2   138 103.  McWilliam V, Koplin J, Lodge C, Tang M, Dharmage S, Allen K. The Prevalence of Tree Nut Allergy: A Systematic Review. Curr Allergy Asthma Rep. 2015 Sep 1;15(9):54.  104.  Rafnar T, Griffith IJ, Kuo MC, Bond JF, Rogers BL, Klapper DG. Cloning of Amb a I (antigen E), the major allergen family of short ragweed pollen. J Biol Chem. 1991 Jan 15;266(2):1229–36.  105.  Yli-Panula E, Fekedulegn DB, Green BJ, Ranta H. Analysis of Airborne Betula Pollen in Finland; a 31-Year Perspective. Int J Environ Res Public Health Basel. 2009 Jun;6(6):1706–23.  106.  Ziska L, Knowlton K, Rogers C, Dalan D, Tierney N, Elder MA, et al. Recent warming by latitude associated with increased length of ragweed pollen season in central North America. Proc Natl Acad Sci U S A. 2011;108(10):4248–51.  107.  García-Mozo H, Yaezel L, Oteros J, Galán C. Statistical approach to the analysis of olive long-term pollen season trends in southern Spain. Sci Total Environ. 2014 Mar 1;473–474(Supplement C):103–9.  108.  Zhang Y, Bielory L, Mi Z, Cai T, Robock A, Georgopoulos P. Allergenic pollen season variations in the past two decades under changing climate in the United States. Glob Change Biol. 2015 Apr 1;21(4):1581–9.  109.  Ellis AK, Tenn MW. Advances in rhinitis—models and mechanisms. Ann Allergy Asthma Immunol [Internet]. Available from: https://www.sciencedirect.com/science/article/pii/S1081120617311705 110.  Skoner DP. Allergic rhinitis: Definition, epidemiology, pathophysiology, detection, and diagnosis. J Allergy Clin Immunol. 2001 Jul;108(1, Supplement):S2–8.  111.  Gunawan H, Takai T, Ikeda S, Okumura K, Ogawa H. Protease Activity of Allergenic Pollen of Cedar, Cypress, Juniper, Birch and Ragweed. Allergol Int. 2008;57(1):83–91.  112.  Ebner C, Schenk S, Najafian N, Siemann U, Steiner R, Fischer GW, et al. Nonallergic individuals recognize the same T cell epitopes of Bet v 1, the major birch pollen allergen, as atopic patients. J Immunol. 1995 Feb 15;154(4):1932–40.  113.  Akdis M, Verhagen J, Taylor A, Karamloo F, Karagiannidis C, Crameri R, et al. Immune Responses in Healthy and Allergic Individuals Are Characterized by a Fine Balance between Allergen-specific T Regulatory 1 and T Helper 2 Cells. J Exp Med. 2004 Jun 7;199(11):1567–75.  114.  Allakhverdi Z, Bouguermouh S, Rubio M, Delespesse G. Adjuvant activity of pollen grains. Allergy. 2005 Sep 1;60(9):1157–64.  115.  Small P, Kim H. Allergic rhinitis. Allergy Asthma Clin Immunol. 2011;7(1):S3.    139 116.  Seidman MD, Gurgel RK, Lin SY, Schwartz SR, Baroody FM, Bonner JR, et al. Clinical Practice Guideline: Allergic Rhinitis. Otolaryngol Neck Surg. 2015 Feb 1;152(1_suppl):S1–43.  117.  Moldaver D, Larché M. Immunotherapy with peptides. Allergy. 2011 Jun 1;66(6):784–91.  118.  Akdis M, Akdis CA. Mechanisms of allergen-specific immunotherapy: Multiple suppressor factors at work in immune tolerance to allergens. J Allergy Clin Immunol. 2014 Mar;133(3):621–31.  119.  Burks AW, Calderon MA, Casale T, Cox L, Demoly P, Jutel M, et al. Update on allergy immunotherapy: American Academy of Allergy, Asthma & Immunology/European Academy of Allergy and Clinical Immunology/PRACTALL consensus report. J Allergy Clin Immunol. 2013 May;131(5):1288–1296.e3.  120.  Senti G, Crameri R, Kuster D, Johansen P, Martinez-Gomez JM, Graf N, et al. Intralymphatic immunotherapy for cat allergy induces tolerance after only 3 injections. J Allergy Clin Immunol. 2012 May;129(5):1290–6.  121.  Hylander T, Latif L, Petersson-Westin U, Cardell LO. Intralymphatic allergen-specific immunotherapy: An effective and safe alternative treatment route for pollen-induced allergic rhinitis. J Allergy Clin Immunol. 2013 Feb;131(2):412–20.  122.  Casale TB, Stokes JR. Immunotherapy: What lies beyond. J Allergy Clin Immunol. 2014 Mar;133(3):612–9.  123.  Frew AJ. Injection immunotherapy. British Society for Allergy and Clinical Immunology Working Party. BMJ. 1993 Oct 9;307(6909):919–23.  124.  Durham SR, Till SJ. Immunologic changes associated with allergen immunotherapy. J Allergy Clin Immunol. 1998 Aug 1;102(2):157–64.  125.  Romagnani S. The role of lymphocytes in allergic disease. J Allergy Clin Immunol. 2000 Mar 1;105(3):399–408.  126.  Larché M, Akdis CA, Valenta R. Immunological mechanisms of allergen-specific immunotherapy. Nat Rev Immunol. 2006 Oct;6(10):761–71.  127.  Ellis AK, Frankish CW, O’Hehir RE, Armstrong K, Steacy L, Larché M, et al. Treatment with grass allergen peptides improves symptoms of grass pollen–induced allergic rhinoconjunctivitis. J Allergy Clin Immunol [Internet]. [cited 2017 Apr 9]; Available from: http://www.sciencedirect.com/science/article/pii/S0091674917300374 128.  Kim YW, Singh A, Shannon CP, Thiele J, Steacy LM, Ellis AK, et al. Investigating Immune Gene Signatures in Peripheral Blood from Subjects with Allergic Rhinitis Undergoing Nasal Allergen Challenge. J Immunol. 2017 Oct 18;ji1700378.    140 129.  Worm M, Lee HH, Kleine-Tebbe J, Hafner RP, Laidler P, Healey D, et al. Development and preliminary clinical evaluation of a peptide immunotherapy vaccine for cat allergy. J Allergy Clin Immunol. 2011;127:89–97, 97 e1-14.  130.  Dienz O, Rincon M. The effects of IL-6 on CD4 T cell responses. Clin Immunol Orlando Fla. 2009 Jan;130(1):27–33.  131.  Hunter CA, Jones SA. IL-6 as a keystone cytokine in health and disease. Nat Immunol. 2015 May;16(5):448–57.  132.  Miossec P, Kolls JK. Targeting IL-17 and TH17 cells in chronic inflammation. Nat Rev Drug Discov. 2012 Oct;11(10):763–76.  133.  Ciprandi G, Filaci G, Battaglia F, Fenoglio D. Peripheral Th-17 cells in allergic rhinitis: New evidence. Int Immunopharmacol. 2010 Feb 1;10(2):226–9.  134.  Singh A, Yamamoto M, Ruan J, Choi JY, Gauvreau GM, Olek S, et al. Th17/Treg ratio derived using DNA methylation analysis is associated with the late phase asthmatic response. Allergy Asthma Clin Immunol. 2014 Jun 24;10:32.  135.  Bajoriuniene I, Malakauskas K, Lavinskiene S, Jeroch J, Gasiuniene E, Vitkauskiene A, et al. Response of Peripheral Blood Th17 Cells to Inhaled Dermatophagoides pteronyssinus in Patients with Allergic Rhinitis and Asthma. Lung. 2012 Oct 1;190(5):487–95.  136.  Ciprandi G, Fenoglio D, De Amici M, Quaglini S, Negrini S, Filaci G. Serum IL-17 levels in patients with allergic rhinitis. J Allergy Clin Immunol. 2008 Sep 1;122(3):650–651.e2.  137.  Ciprandi G, De Amici M, Negrini S, Marseglia G, Tosca MA. TGF-β and IL-17 serum levels and specific immunotherapy. Int Immunopharmacol. 2009 Sep 1;9(10):1247–9.  138.  Linneberg A, Henrik Nielsen N, Frølund L, Madsen F, Dirksen A, Jørgensen T. The link between allergic rhinitis and allergic asthma: A prospective population-based study. The Copenhagen Allergy Study. Allergy. 2002 Nov 1;57(11):1048–52.  139.  Croxford AL, Kulig P, Becher B. IL-12-and IL-23 in health and disease. Cytokine Growth Factor Rev. 2014 Aug;25(4):415–21.  140.  Floros T, Tarhini AA. Anticancer Cytokines: Biology and Clinical Effects of Interferon-α2, Interleukin (IL)-2, IL-15, IL-21, and IL-12. Semin Oncol. 2015 Aug 1;42(4):539–48.  141.  Lazarevic V, Glimcher LH, Lord GM. T-bet: a bridge between innate and adaptive immunity. Nat Rev Immunol. 2013 Nov;13(11):777–89.  142.  Hamid QA, Schotman E, Jacobson MR, Walker SM, Durham SR. Increases in IL-12 messenger RNA+ cells accompany inhibition of allergen-induced late skin responses after successful grass pollen immunotherapy. J Allergy Clin Immunol. 1997 Feb;99(2):254–60.    141 143.  Moingeon P, Batard T, Fadel R, Frati F, Sieber J, Van Overtvelt L. Immune mechanisms of allergen-specific sublingual immunotherapy. Allergy. 2006 Feb;61(2):151–65.  144.  Teixeira LK, Fonseca BP, Barboza BA, Viola JP. The role of interferon-gamma on immune and allergic responses. Mem Inst Oswaldo Cruz. 2005 Mar;100:137–44.  145.  Bartemes KR, Kephart GM, Fox SJ, Kita H. Enhanced innate type 2 immune response in peripheral blood from patients with asthma. J Allergy Clin Immunol. 2014 Sep 1;134(3):671–678.e4.  146.  Sakaguchi S, Wing K, Onishi Y, Prieto-Martin P, Yamaguchi T. Regulatory T cells: how do they suppress immune responses? Int Immunol. 2009 Oct 1;21(10):1105–11.  147.  Dasgupta A, Saxena R. Regulatory T cells: A review. HttpnmjiinarchivesVolume-25Issue-6Review-Artic [Internet]. 2012 Nov [cited 2017 May 22]; Available from: http://imsear.hellis.org/handle/123456789/156310 148.  Kim AS, Doherty TA, Karta MR, Das S, Baum R, Rosenthal P, et al. Regulatory B cells and T follicular helper cells are reduced in allergic rhinitis. J Allergy Clin Immunol. 2016 Oct 1;138(4):1192–1195.e5.  149.  Rosser EC, Mauri C. Regulatory B Cells: Origin, Phenotype, and Function. Immunity. 2015 Apr 21;42(4):607–12.  150.  Chen M, Wang Y-H, Wang Y, Huang L, Sandoval H, Liu Y-J, et al. Dendritic Cell Apoptosis in the Maintenance of Immune Tolerance. Science. 2006 Feb 24;311(5764):1160–4.  151.  Hadeiba H, Lahl K, Edalati A, Oderup C, Habtezion A, Pachynski R, et al. Plasmacytoid Dendritic Cells Transport Peripheral Antigens to the Thymus to Promote Central Tolerance. Immunity. 2012 Mar 23;36(3):438–50.  152.  Wambre E, DeLong JH, James EA, LaFond RE, Robinson D, Kwok WW. Differentiation stage determines pathologic and protective allergen-specific CD4+ T-cell outcomes during specific immunotherapy. J Allergy Clin Immunol. 2012 Feb;129(2):544–551.e7.  153.  Won H-H, Kim SR, Bang OY, Lee S-C, Huh W, Ko J-W, et al. Differentially expressed genes in human peripheral blood as potential markers for statin response. J Mol Med. 2012 Feb 1;90(2):201–11.  154.  Berkowska MA, Heeringa JJ, Hajdarbegovic E, van der Burg M, Thio HB, van Hagen PM, et al. Human IgE+ B cells are derived from T cell–dependent and T cell–independent pathways. J Allergy Clin Immunol. 2014 Sep 1;134(3):688–697.e6.  155.  Wu M, Lou J, Zhang S, Chen X, Huang L, Sun R, et al. Gene expression profiling of CD8+ T cells induced by ovarian cancer cells suggests a possible mechanism for CD8+ Treg cell production. Cell Prolif. 2016 Dec 1;49(6):669–77.    142 156.  Fuchs A, Vermi W, Lee JS, Lonardi S, Gilfillan S, Newberry RD, et al. Intraepithelial Type 1 Innate Lymphoid Cells Are a Unique Subset of IL-12- and IL-15-Responsive IFN-γ-Producing Cells. Immunity. 2013 Apr 18;38(4):769–81.  157.  Swiecki M, Colonna M. The multifaceted biology of plasmacytoid dendritic cells. Nat Rev Immunol. 2015 Aug;15(8):471–85.  158.  Cella M, Fuchs A, Vermi W, Facchetti F, Otero K, Lennerz JKM, et al. A human natural killer cell subset provides an innate source of IL-22 for mucosal immunity. Nat Lond. 2009 Feb 5;457(7230):722–5.  159.  Merad M, Sathe P, Helft J, Miller J, Mortha A. The Dendritic Cell Lineage: Ontogeny and Function of Dendritic Cells and Their Subsets in the Steady State and the Inflamed Setting. Annu Rev Immunol [Internet]. 2013 [cited 2017 May 17];31. Available from: http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3853342/ 160.  Tel J, Smits EL, Anguille S, Joshi RN, Figdor CG, Vries IJM de. Human plasmacytoid dendritic cells are equipped with antigen-presenting and tumoricidal capacities. Blood. 2012 Nov 8;120(19):3936–44.  161.  Reizis B, Bunin A, Ghosh HS, Lewis KL, Sisirak V. Plasmacytoid dendritic cells: recent progress and open questions. Annu Rev Immunol. 2011;29:163–83.  162.  Poli A, Michel T, Thérésine M, Andrès E, Hentges F, Zimmer J. CD56bright natural killer (NK) cells: an important NK cell subset. Immunology. 2009 Apr;126(4):458–65.  163.  Mandal A, Viswanathan C. Natural killer cells: In health and disease. Hematol Oncol Stem Cell Ther. 2015 Jun;8(2):47–55.  164.  Jiao Y, Huntington ND, Belz GT, Seillet C. Type 1 Innate Lymphoid Cell Biology: Lessons Learnt from Natural Killer Cells. Front Immunol [Internet]. 2016 Oct 12 [cited 2018 Mar 3];7. Available from: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5059362/ 165.  Kim J, Bentley PJ, Aickelin U, Greensmith J, Tedesco G, Twycross J. Immune system approaches to intrusion detection – a review. Nat Comput. 2007 Dec 1;6(4):413–66.  166.  Iwasaki A, Medzhitov R. Regulation of Adaptive Immunity by the Innate Immune System. Science. 2010 Jan 15;327(5963):291–5.  167.  DuPage M, Bluestone JA. Harnessing the plasticity of CD4+ T cells to treat immune-mediated disease. Nat Rev Immunol. 2016 Mar;16(3):149–63.  168.  James G, Witten D, Hastie T, Tibshirani R. An Introduction to Statistical Learning [Internet]. New York, NY: Springer New York; 2013. (Springer Texts in Statistics; vol. 103). Available from: http://link.springer.com/10.1007/978-1-4614-7138-7   143 169.  Vos G de. Skin Testing Versus Serum-Specific IgE Testing: Which Is Better for Diagnosing Aeroallergen Sensitization and Predicting Clinical Allergy? Curr Allergy Asthma Rep. 2014 May 1;14(5):430.  170.  Schoos A-MM, Chawes BLK, Følsgaard NV, Samandari N, Bønnelykke K, Bisgaard H. Disagreement between skin prick test and specific IgE in young children. Allergy. 2015 Jan 1;70(1):41–8.  171.  Ciprandi G, Cirillo I. Monosensitization and polysensitization in allergic rhinitis. Eur J Intern Med. 2011 Dec 1;22(6):e75–9.    144  Appendices Appendix A  Supplementary material for Chapter 3 A.1 Immune gene list of the clusters identified (X: Not available).No Gene Clusters Q cohort M cohort (120genes) (207genes) 1 ADORA2A Q1 M2 2 IL1RN Q1 M2 3 CD7 Q1 M6 4 CD74 Q1 X 5 CX3CR1 Q1 X 6 MAP3K1 Q1 X 7 MX1 Q1 X 8 OAS3 Q1 X 9 PSEN2 Q1 X 10 RELA Q1 X 11 BCL6 Q2 M2 12 CFP Q2 M2 13 CR1 Q2 M2 14 FUT7 Q2 M2 15 ICAM1 Q2 M2 16 MAPK3 Q2 M2 17 MEFV Q2 M2 18 NOTCH1 Q2 M2 19 PIK3CG Q2 M2 20 SH2B2 Q2 M2 21 STAT3 Q2 M2 22 TNFRSF10B Q2 M2 23 TNFRSF10C Q2 M2 24 TNFRSF8 Q2 M2 25 HLA-C Q2 M3 26 IGF2R Q2 M3 No Gene Clusters Q cohort M cohort (120genes) (207genes) 27 LILRA5 Q2 M3 28 MME Q2 M3 29 PDGFRB Q2 M5 30 CREB5 Q2 X 31 IFITM2 Q2 X 32 SBNO2 Q2 X 33 TBX21 Q2 X 34 TNFRSF1B Q2 X 35 FADD Q3 M2 36 LCN2 Q3 M2 37 TNFRSF9 Q3 M2 38 ARG1 Q3 M3 39 BST1 Q3 M3 40 CASP8 Q3 M3 41 CEBPB Q3 M3 42 CSF3R Q3 M3 43 ENTPD1 Q3 M3 44 FCER1G Q3 M3 45 IL1R1 Q3 M3 46 ITGA1 Q3 M3 47 ITGAX Q3 M3 48 LILRA1 Q3 M3 49 LILRB2 Q3 M3 50 LILRB3 Q3 M3 51 LTF Q3 M3 52 NCF4 Q3 M3   145 No Gene Clusters Q cohort M cohort (120genes) (207genes) 53 REL Q3 M3 54 SLC11A1 Q3 M3 55 TLR4 Q3 M3 56 TREM1 Q3 M3 57 MERTK Q3 M5 58 CLEC7A Q3 X 59 CXCR1 Q3 X 60 NOD2 Q3 X 61 CD180 Q4 M4 62 CD5 Q4 M4 63 FCER2 Q4 M4 64 FLT3LG Q4 M4 65 LY9 Q4 M4 66 TCF7 Q4 M4 67 CD274 Q4 M6 68 CD4 Q4 M6 69 TLR7 Q4 X 70 CAMP Q5 M3 71 CD58 Q5 M3 72 CKLF Q5 M3 73 CREBBP Q5 M3 74 FOS Q5 M3 75 IFIT2 Q5 M3 76 IL18R1 Q5 M3 77 LGALS3 Q5 M3 78 LY96 Q5 M3 79 S100A12 Q5 M3 80 SPP1 Q5 M3 81 USP9Y Q5 M3 82 CLEC4A Q5 M5 83 CLEC5A Q5 M5 No Gene Clusters Q cohort M cohort (120genes) (207genes) 84 GZMB Q5 M5 85 IL1RL1 Q5 M5 86 C5 Q5 X 87 CD1C Q5 X 88 CEACAM8 Q5 X 89 IL13RA1 Q5 X 90 IL1R2 Q5 X 91 IL1RAP Q5 X 92 IRAK2 Q5 X 93 S100A8 Q5 X 94 XCL2 Q5 X 95 CD28 Q6 M4 96 BCL2 Q6 M6 97 CCR7 Q6 M6 98 IL6ST Q6 M6 99 IL7R Q6 M6 100 ITK Q6 M6 101 CCR3 Q7 M3 102 CCL4 Q7 M5 103 CCND3 Q7 M5 104 KLRC1 Q7 M5 105 SOCS1 Q7 M6 106 ANXA1 Q7 M7 107 HMGB1 Q7 M7 108 IDO1 Q7 M7 109 IL5RA Q7 M7 110 CXCL8 Q7 M7 111 CASP3 Q7 X 112 CD24 Q7 X 113 CD3D Q7 X 114 GZMK Q7 X   146 No Gene Clusters Q cohort M cohort (120genes) (207genes) 115 IFNGR1 Q7 X 116 IL32 Q7 X 117 PRKCE Q7 X 118 RPS6 Q7 X 119 TIGIT Q7 X 120 TNFSF8 Q7 X 121 CCR4 X M1 122 IL4R X M1 123 IRGM X M1 124 PLA2G6 X M1 125 TLR10 X M1 126 ALCAM X M2 127 BID X M2 128 CD46 X M2 129 CDH1 X M2 130 CEACAM1 X M2 131 CSF1R X M2 132 CXCL1 X M2 133 F13A1 X M2 134 F2RL1 X M2 135 FAS X M2 136 FCGR2A X M2 137 ICAM3 X M2 138 ICAM4 X M2 139 IFI35 X M2 140 IFNAR2 X M2 141 IL6R X M2 142 ITGB3 X M2 143 JAM3 X M2 144 LAMP2 X M2 145 MAP2K4 X M2 No Gene Clusters Q cohort M cohort (120genes) (207genes) 146 MAP3K5 X M2 147 MAPK1 X M2 148 MAPKAPK2 X M2 149 PTGS2 X M2 150 PTPRC X M2 151 SELL X M2 152 STAT6 X M2 153 TAL1 X M2 154 TFE3 X M2 155 THBD X M2 156 THBS1 X M2 157 TLR1 X M2 158 TLR6 X M2 159 TLR8 X M2 160 TNFRSF1A X M2 161 TNFSF13 X M2 162 TNFSF14 X M2 163 TOLLIP X M2 164 AMICA1 X M3 165 ATG7 X M3 166 CASP10 X M3 167 CD1D X M3 168 CD53 X M3 169 CD9 X M3 170 CHUK X M3 171 CTSS X M3 172 FCGR2B X M3 173 IGF1R X M3 174 IKBKG X M3 175 IL18 X M3 176 INPP5D X M3   147 No Gene Clusters Q cohort M cohort (120genes) (207genes) 177 IRAK1 X M3 178 ITGA5 X M3 179 ITGAM X M3 180 JAK2 X M3 181 MAPK14 X M3 182 MICA X M3 183 MSR1 X M3 184 PECAM1 X M3 185 PLAUR X M3 186 PPBP X M3 187 PSEN1 X M3 188 PYCARD X M3 189 TNFSF10 X M3 190 TNFSF4 X M3 191 TRAF6 X M3 192 TYK2 X M3 193 UBC X M3 194 VEGFA X M3 195 CD19 X M4 196 CD22 X M4 197 CD79A X M4 198 CXCR5 X M4 199 DOCK9 X M4 200 DPP4 X M4 201 FCGR1A X M4 202 HLA-DOB X M4 203 ICOS X M4 204 LRRN3 X M4 205 LTA X M4 206 LTB X M4 207 SMAD3 X M4 No Gene Clusters Q cohort M cohort (120genes) (207genes) 208 TNFRSF13C X M4 209 TP53 X M4 210 CCL3 X M5 211 CD63 X M5 212 CMKLR1 X M5 213 GZMH X M5 214 HAVCR2 X M5 215 IL18RAP X M5 216 KIR2DS1 X M5 217 LY86 X M5 218 NCR1 X M5 219 PIN1 X M5 220 PVR X M5 221 SH2D1B X M5 222 CD247 X M6 223 CD27 X M6 224 CD3E X M6 225 CD3G X M6 226 CD8B X M6 227 CD96 X M6 228 DDX58 X M6 229 ETS1 X M6 230 IL21R X M6 231 ILF3 X M6 232 ITGA6 X M6 233 LCK X M6 234 NFKB1 X M6 235 SPN X M6 236 ST6GAL1 X M6 237 TXK X M6 238 CLEC4C X M7   148 No Gene Clusters Q cohort M cohort (120genes) (207genes) 239 GNLY X M7 240 PTGDR2 X M7 241 SMPD3 X M7 242 TNFRSF17 X M7    149 A.2 Correlation between clinical symptoms and systemic immune response in 13 subjects  Figure A.2 Correlation between clinical symptoms and systemic immune response in 13 subjects who had no missing data.  The Pearson correlation between sum of clinical symptom scores (TNSS, PNIF and PNIF.Ratio) and immune cell frequencies (lymphocytes, neutrophils and NLR; lymphocytes.Ratio, neutrophils.Ratio and NLR.Ratio) or immune gene clusters (geometric mean  of the ratio of Cluster 2, 3 and 4 using shared genes in both Q and M cohrots). n = 13, Sum.ALL: a sum of score over all measured time points (baseline to 12h post-NAC in clinical symptoms;   150 baseline to 6h post-NAC in immune cell frequencies), Sum.EPR: a sum of score over time points in EPR (baseline to 6h post-NAC), Sum.LPR: a sum of score over time points in LPR (7 to 12h post-NAC).  151 Appendix B  Supplementary material for Chapter 5 B.1 PNIF of pre-treatment (V1A) and post-treatment (V3) in the 10 cat allergic subjects  Figure B.1 PNIF of the 10 cat allergic subjects.  V1A, pre-treatment; V3, post-treatment.    152 B.2 TNSS of pre-treatment (V1A) and post-treatment (V3) in the 10 cat allergic subjects  Figure B.2 TNSS of the 10 cat allergic subjects.  V1A, pre-treatment; V3, post-treatment.   153 B.3 Immune gene list of the clusters identified at V1A and V3 (X: Not available).  No Gene Clusters at V1A (126 genes) Clusters at V3 (53 genes) 1 ARG1 V1A.C3 V3.C1 2 CAMP V1A.C3 V3.C1 3 CCL4 V1A.C5 V3.C2 4 CCND3 V1A.C5 V3.C7 5 CCR3 V1A.C5 V3.C6 6 CCR7 V1A.C6 V3.C5 7 CD180 V1A.C1 V3.C4 8 CX3CR1 V1A.C1 V3.C4 9 FCER2 V1A.C4 V3.C5 10 GZMB V1A.C5 V3.C2 11 IL18RAP V1A.C3 V3.C3 12 IL1R2 V1A.C5 V3.C6 13 IL1RAP V1A.C5 V3.C6 14 IL1RL1 V1A.C5 V3.C6 15 IL5RA V1A.C7 V3.C7 16 ITGA6 V1A.C6 V3.C5 17 KLRC1 V1A.C5 V3.C2 18 LILRA5 V1A.C2 V3.C3 19 LRRN3 V1A.C6 V3.C5 No Gene Clusters at V1A (126 genes) Clusters at V3 (53 genes) 20 LTF V1A.C3 V3.C6 21 PTGS2 V1A.C2 V3.C3 22 SH2B2 V1A.C2 V3.C4 23 TBX21 V1A.C2 V3.C2 24 TLR4 V1A.C3 V3.C3 25 TLR7 V1A.C1 V3.C5 26 AMICA1 V1A.C3 X 27 ANXA1 V1A.C7 X 28 BCL2 V1A.C6 X 29 BCL6 V1A.C2 X 30 BST1 V1A.C3 X 31 C5 V1A.C5 X 32 CASP3 V1A.C7 X 33 CASP8 V1A.C3 X 34 CCRL2 V1A.C7 X 35 CD1C V1A.C5 X 36 CD24 V1A.C5 X 37 CD27 V1A.C4 X 38 CD4 V1A.C4 X   154 No Gene Clusters at V1A (126 genes) Clusters at V3 (53 genes) 39 CD5 V1A.C4 X 40 CD58 V1A.C3 X 41 CD7 V1A.C1 X 42 CD74 V1A.C1 X 43 CD79A V1A.C1 X 44 CEACAM1 V1A.C3 X 45 CEACAM8 V1A.C5 X 46 CKLF V1A.C3 X 47 CLEC4A V1A.C5 X 48 CLEC4C V1A.C7 X 49 CLEC5A V1A.C5 X 50 CLEC6A V1A.C5 X 51 CLEC7A V1A.C3 X 52 CR1 V1A.C2 X 53 CREBBP V1A.C5 X 54 CSF3R V1A.C3 X 55 CXCL1 V1A.C2 X 56 DOCK9 V1A.C4 X 57 ENTPD1 V1A.C3 X 58 ETS1 V1A.C6 X 59 FADD V1A.C3 X No Gene Clusters at V1A (126 genes) Clusters at V3 (53 genes) 60 FCER1G V1A.C3 X 61 FLT3LG V1A.C4 X 62 FOXJ1 V1A.C5 X 63 FPR2 V1A.C3 X 64 FUT7 V1A.C2 X 65 HMGB1 V1A.C7 X 66 ICAM1 V1A.C2 X 67 IDO1 V1A.C7 X 68 IFIT2 V1A.C3 X 69 IFITM2 V1A.C2 X 70 IFNGR1 V1A.C5 X 71 IGF1R V1A.C3 X 72 IGF2R V1A.C3 X 73 IL11RA V1A.C5 X 74 IL12RB1 V1A.C3 X 75 IL13RA1 V1A.C5 X 76 IL18R1 V1A.C5 X 77 IL1R1 V1A.C3 X 78 IL1RN V1A.C2 X 79 IL32 V1A.C5 X 80 IL6ST V1A.C6 X   155 No Gene Clusters at V1A (126 genes) Clusters at V3 (53 genes) 81 IL7R V1A.C6 X 82 CXCL8 V1A.C5 X 83 ITGA1 V1A.C3 X 84 ITGAX V1A.C3 X 85 ITK V1A.C6 X 86 KIR2DS1 V1A.C3 X 87 LCN2 V1A.C3 X 88 LGALS3 V1A.C5 X 89 LILRA1 V1A.C5 X 90 LILRB2 V1A.C3 X 91 LILRB3 V1A.C3 X 92 LY9 V1A.C4 X 93 MAP3K5 V1A.C2 X 94 MAPK3 V1A.C2 X 95 MAPKAPK2 V1A.C3 X 96 MEFV V1A.C2 X 97 MME V1A.C2 X 98 NCF4 V1A.C3 X 99 NOD2 V1A.C3 X 100 NOTCH1 V1A.C2 X 101 PDGFRB V1A.C3 X No Gene Clusters at V1A (126 genes) Clusters at V3 (53 genes) 102 PIK3CG V1A.C2 X 103 PLAUR V1A.C3 X 104 PRKCE V1A.C5 X 105 PYCARD V1A.C3 X 106 REL V1A.C3 X 107 RELA V1A.C1 X 108 RPS6 V1A.C7 X 109 S100A12 V1A.C5 X 110 S100A8 V1A.C5 X 111 SBNO2 V1A.C2 X 112 SLC11A1 V1A.C3 X 113 SOCS1 V1A.C7 X 114 ST6GAL1 V1A.C4 X 115 TCF7 V1A.C4 X 116 TIRAP V1A.C3 X 117 TLR10 V1A.C1 X 118 TNFRSF10B V1A.C2 X 119 TNFRSF10C V1A.C2 X 120 TNFRSF8 V1A.C2 X 121 TNFRSF9 V1A.C2 X 122 TNFSF14 V1A.C2 X   156 No Gene Clusters at V1A (126 genes) Clusters at V3 (53 genes) 123 TNFSF8 V1A.C7 X 124 TREM1 V1A.C3 X 125 USP9Y V1A.C5 X 126 XCL2 V1A.C5 X 127 BID X V3.C3 128 BLNK X V3.C4 129 C4B X V3.C6 130 CCL3L1 X V3.C3 131 CCR6 X V3.C5 132 CD36 X V3.C5 133 CD40 X V3.C5 134 CHUK X V3.C4 135 CR2 X V3.C5 136 DDX58 X V3.C5 137 GZMH X V3.C2 138 IFI16 X V3.C5 139 IL2RA X V3.C7 140 ITGAE X V3.C1 141 KLRC2 X V3.C2 142 KLRD1 X V3.C2 143 LTA X V3.C5 No Gene Clusters at V1A (126 genes) Clusters at V3 (53 genes) 144 LY96 X V3.C3 145 MARCO X V3.C7 146 MR1 X V3.C4 147 MSR1 X V3.C4 148 NCAM1 X V3.C1 149 PVR X V3.C7 150 SH2D1B X V3.C2 151 TLR8 X V3.C4 152 TNFSF13B X V3.C4 153 TXK X V3.C7 154 VEGFA X V3.C6            157 B.4 Significantly higher expressed genes after Cat-PAD in 10 subjects (LIMMA paired test, V1A versus V3 at each time point, BH-FDR < 0.1).  Baseline (0 h) Num Gene Fold change (log base 2) Mean of expression (log base 2) p value BH-FDR 1 BATF 0.32143 5.50105 0.00619 0.08049 2 SIGIRR 0.19572 7.95071 0.00630 0.08049 3 CD7 0.21174 7.39755 0.00788 0.08159 4 PVR 0.28226 5.56774 0.00811 0.08159 5 CARD11 0.20607 7.26802 0.00845 0.08230 6 LCK 0.17369 9.54186 0.01040 0.08858 7 CD8B 0.24155 6.36165 0.01112 0.09110 8 IRF3 0.25257 5.94833 0.01148 0.09196 9 ST6GAL1 0.15338 9.13952 0.01258 0.09333             1 h post-NAC Num Gene Fold change (log base 2) Mean of expression (log base 2) p value BH-FDR 1 CD28 0.35827 6.93537 0.00026 0.00930 2 CD96 0.27162 8.72891 0.00098 0.01177 3 STAT4 0.25728 7.94462 0.00106 0.01209 4 RORA 0.26066 7.75057 0.00132 0.01394 5 LY9 0.27888 8.56457 0.00185 0.01564 6 BMI1 0.30410 6.84477 0.00190 0.01564 7 ITGAE 0.35580 6.63418 0.00191 0.01564 8 IL7R 0.24620 11.17913 0.00230 0.01733 9 ICOS 0.35961 6.85905 0.00234 0.01733 10 IL11RA 0.39518 5.20333 0.00242 0.01752   158 Num Gene Fold change (log base 2) Mean of expression (log base 2) p value BH-FDR    11 TCF7 0.28772 9.90263 0.00285 0.01977 12 SH2D1A 0.44982 6.63821 0.00327 0.02177 13 BTLA 0.34730 8.27634 0.00329 0.02177 14 CD79B 0.24132 8.58242 0.00461 0.02875 15 CD19 0.28596 6.67448 0.00466 0.02875 16 CD3G 0.26683 8.08728 0.00512 0.02991 17 LCK 0.25425 9.46578 0.00532 0.03044 18 ZAP70 0.22023 8.57082 0.00598 0.03351 19 CD3D 0.43941 9.04856 0.00612 0.03395 20 CTLA4 0.26637 6.28108 0.00667 0.03557 21 CXCR5 0.33081 5.05149 0.00745 0.03814 22 CD84 0.22778 8.31408 0.00748 0.03814 23 CD27 0.24435 8.32787 0.00771 0.03853 24 MIF 0.24671 7.47731 0.01033 0.04718 25 GZMM 0.24072 6.06913 0.01036 0.04718 26 RPS6 0.42903 12.00797 0.01040 0.04718 27 CD3E 0.19453 8.98308 0.01075 0.04797 28 EIF2B4 0.24464 6.97331 0.01152 0.04981 29 CD247 0.19816 9.34241 0.01204 0.05138 30 ITK 0.19752 9.07259 0.01207 0.05138 31 DOCK9 0.21374 7.42101 0.01266 0.05305 32 CD40LG 0.24389 7.38683 0.01369 0.05526 33 CD180 0.22401 6.77866 0.01385 0.05550 34 NCAM1 0.20612 6.67776 0.01458 0.05758 35 CD48 0.29292 10.46164 0.01564 0.06088 36 NT5E 0.21330 5.52717 0.01775 0.06767 37 ELK1 0.25544 5.32603 0.01854 0.06915 38 CCR7 0.20333 7.42430 0.01904 0.06944 39 CD7 0.19991 7.45329 0.01940 0.07006   159 Num Gene Fold change (log base 2) Mean of expression (log base 2) p value BH-FDR 40 GZMK 0.33548 7.74303 0.02059 0.07291 41 HLA-DRA 0.22797 12.01868 0.02078 0.07309 42 KLRK1 0.23170 8.95749 0.02090 0.07309 43 SLAMF1 0.22115 6.88457 0.02159 0.07454 44 IKBKE 0.17998 7.17768 0.02190 0.07513 45 DDX50 0.21121 7.50190 0.02447 0.08190 46 FLT3LG 0.20913 8.14381 0.02580 0.08380 47 ICOSLG 0.29535 5.51319 0.02603 0.08406 48 HLA-DMB 0.16269 9.45119 0.02783 0.08882 49 FCER1A 0.43538 7.45131 0.02872 0.09114 50 LRRN3 0.29431 6.64382 0.03141 0.09796             2 h post-NAC Num Gene Fold change (log base 2) Mean of expression (log base 2) p value BH-FDR 1 S100B 0.47062 4.75520 0.00028 0.00959 2 IL7R 0.30146 11.16400 0.00098 0.01384 3 ZAP70 0.27873 8.62727 0.00116 0.01520 4 ITK 0.30029 9.03949 0.00124 0.01553 5 KLRC2 0.44304 6.20865 0.00132 0.01577 6 GZMA 0.49086 8.41334 0.00142 0.01589 7 HLA-DQB1 0.30013 6.96063 0.00158 0.01653 8 CD28 0.36128 6.92217 0.00173 0.01655 9 KLRB1 0.41879 8.08426 0.00268 0.02302 10 GZMK 0.35647 7.74626 0.00281 0.02355 11 AKT3 0.29694 7.38633 0.00301 0.02396 12 LRRN3 0.35744 6.59440 0.00314 0.02396 13 STAT4 0.31806 7.95023 0.00377 0.02617   160 Num Gene Fold change (log base 2) Mean of expression (log base 2) p value BH-FDR 14 KLRC1 0.37668 6.83225 0.00395 0.02619 15 CD3D 0.34623 9.02461 0.00397 0.02619 16 TIGIT 0.31189 6.56840 0.00409 0.02642 17 LCK 0.23481 9.46193 0.00441 0.02689 18 NUP107 0.28229 7.61362 0.00513 0.02964 19 RORC 0.29470 4.98882 0.00513 0.02964 20 CD48 0.28839 10.42022 0.00549 0.03080 21 HLA-DRA 0.27612 12.05249 0.00699 0.03691 22 CD84 0.21336 8.37455 0.00731 0.03787 23 CD96 0.23367 8.75468 0.00763 0.03913 24 HLA-DQA1 0.25388 8.02561 0.00788 0.03936 25 RPS6 0.40280 12.00968 0.00789 0.03936 26 BTLA 0.22214 8.24017 0.00833 0.04023 27 CD79B 0.22597 8.61917 0.00855 0.04084 28 SIGIRR 0.22337 7.94045 0.00905 0.04226 29 IL12RB1 0.27648 5.77363 0.01104 0.04930 30 MS4A1 0.20240 8.39154 0.01137 0.05034 31 ICOS 0.28775 6.76143 0.01299 0.05535 32 KLRK1 0.27641 9.01560 0.01301 0.05535 33 CCL4 0.28586 6.36485 0.01421 0.05865 34 SH2D1A 0.35000 6.64956 0.01435 0.05879 35 FLT3LG 0.25824 8.10341 0.01461 0.05941 36 RORA 0.26267 7.76483 0.01790 0.06919 37 FCER1A 0.47847 7.41398 0.01954 0.07349 38 ETS1 0.17557 10.98918 0.02138 0.07876 39 CD160 0.36478 5.21877 0.02212 0.08017 40 CCR4 0.21299 5.57213 0.02220 0.08017 41 BATF 0.22259 5.50610 0.02347 0.08385 42 MR1 0.16128 6.31088 0.02731 0.09491   161 Num Gene Fold change (log base 2) Mean of expression (log base 2) p value BH-FDR 43 TCF7 0.21428 9.90792 0.02789 0.09630             6 h post-NAC Num Gene Fold change (log base 2) Mean of expression (log base 2) p value BH-FDR 1 CD1C 0.54296 5.69992 0.00097 0.08165 2 MSR1 0.65925 4.73136 0.00147 0.08165 3 GZMH 0.51888 7.08861 0.00306 0.08165 4 PSMB7 0.26076 8.45201 0.00447 0.08165 5 CCR6 0.30276 6.65202 0.00455 0.08165 6 CD40 0.21761 7.03146 0.00534 0.08234 7 KIR2DS1 0.57539 4.04533 0.00633 0.08424 8 CD3G 0.24360 8.31784 0.00797 0.08550 9 CD2 0.33899 6.39806 0.00848 0.08617 10 KLRK1 0.25991 8.90112 0.00897 0.08955 11 KLRD1 0.25166 7.49172 0.00965 0.09215 12 KLRC1 0.38608 6.51164 0.01038 0.09350 13 ITGA6 0.16434 8.11269 0.01174 0.09831 14 CD3D 0.47605 9.06872 0.01242 0.09933 15 ANXA1 0.60851 9.56851 0.01259 0.09933 16 FCER1A 0.77387 7.70470 0.01320 0.09933 17 GZMM 0.29893 6.16122 0.01393 0.09933 18 GZMB 0.39447 8.26164 0.01434 0.09933 19 GZMA 0.57536 8.26301 0.01441 0.09933 20 IFNG 0.71667 4.45295 0.01543 0.09933 21 HMGB1 0.44563 8.99768 0.01602 0.09933 22 CD8A 0.16397 7.88750 0.01672 0.09933 23 CD84 0.18309 8.39579 0.01675 0.09933   162 Num Gene Fold change (log base 2) Mean of expression (log base 2) p value BH-FDR 24 AKT3 0.23042 7.43004 0.01678 0.09933 25 XCL2 0.38149 5.97729 0.01691 0.09933 26 REPS1 0.16299 8.23985 0.01697 0.09933 27 LRRN3 0.39201 7.30075 0.01801 0.09991 28 ATG5 0.30196 7.41852 0.01802 0.09991    163 B.5 Significantly lower expressed genes after Cat-PAD in 10 subjects (LIMMA paired test, V1A versus V3 at each time point, BH-FDR < 0.1).  Baseline (0 h) Num Gene Fold change (log base 2) Mean of expression (log base 2) p value BH-FDR 1 CEBPB -0.36897 10.45097 0.00002 0.01341 2 CXCL16 -0.34919 7.07523 0.00013 0.03668 3 MME -0.53743 9.38393 0.00027 0.04168 4 SLC11A1 -0.51501 10.15784 0.00032 0.04168 5 NCF4 -0.34438 9.82608 0.00038 0.04168 6 IFNGR1 -0.31427 10.16595 0.00058 0.04562 7 ICAM1 -0.31629 6.87578 0.00059 0.04562 8 BCL6 -0.48381 10.34537 0.00066 0.04562 9 LILRB3 -0.43407 8.91633 0.00108 0.05184 10 CR1 -0.41438 10.58086 0.00112 0.05184 11 CSF3R -0.42324 12.42183 0.00114 0.05184 12 FCGR2A -0.44954 12.47014 0.00118 0.05184 13 IL18RAP -0.36784 8.10449 0.00132 0.05184 14 TLR4 -0.35048 9.41160 0.00140 0.05184 15 IGF2R -0.39445 9.50742 0.00147 0.05184 16 PLAUR -0.33489 8.17271 0.00162 0.05184 17 CD46 -0.28910 11.30961 0.00164 0.05184 18 TNFRSF10C -0.41524 11.93074 0.00189 0.05184 19 SH2B2 -0.29027 5.81908 0.00190 0.05184 20 IL1R2 -0.45285 8.78508 0.00191 0.05184 21 TLR8 -0.35273 9.12268 0.00203 0.05184 22 FOS -0.46200 9.15421 0.00221 0.05184 23 CSF2RB -0.34337 10.60753 0.00225 0.05184 24 HCK -0.29887 11.02399 0.00234 0.05184   164 Num Gene Fold change (log base 2) Mean of expression (log base 2) p value BH-FDR 25 TLR5 -0.35250 6.04782 0.00244 0.05184 26 AMICA1 -0.29412 12.32307 0.00246 0.05184 27 CXCR2 -0.45919 12.23328 0.00259 0.05266 28 FPR2 -0.47748 10.27811 0.00289 0.05673 29 TREM1 -0.42236 10.13164 0.00338 0.06400 30 STAT5B -0.21817 10.10015 0.00374 0.06671 31 ICAM3 -0.33561 11.54792 0.00385 0.06671 32 LYN -0.36669 9.60762 0.00395 0.06671 33 FCGR3A -0.39145 12.69118 0.00402 0.06671 34 TXNIP -0.22609 13.56357 0.00424 0.06671 35 ITGAX -0.35585 11.23029 0.00425 0.06671 36 PECAM1 -0.26943 10.92281 0.00448 0.06705 37 CXCR1 -0.36235 10.42928 0.00461 0.06705 38 PRKCD -0.26864 9.59845 0.00464 0.06705 39 CD97 -0.28833 11.48576 0.00485 0.06829 40 IFITM2 -0.38540 13.40967 0.00507 0.06956 41 ENTPD1 -0.31557 8.06209 0.00615 0.08049 42 PTPRC -0.33603 11.74232 0.00696 0.08159 43 CREB5 -0.39163 9.52519 0.00712 0.08159 44 IL6R -0.28964 9.20442 0.00716 0.08159 45 TNFRSF1A -0.26093 10.41229 0.00718 0.08159 46 LCP1 -0.21401 12.63548 0.00730 0.08159 47 ALCAM -0.23349 5.87018 0.00749 0.08159 48 IL13RA1 -0.33636 9.54697 0.00776 0.08159 49 TLR6 -0.32355 8.92358 0.00810 0.08159 50 STAT3 -0.24585 10.33458 0.00812 0.08159 51 SYK -0.23803 9.28111 0.00817 0.08159 52 CLEC7A -0.27824 10.13885 0.00854 0.08230 53 HLA-G -0.30913 8.99462 0.00914 0.08617   165 Num Gene Fold change (log base 2) Mean of expression (log base 2) p value BH-FDR 54 IL17RA -0.23210 9.21665 0.00927 0.08617 55 IL1B -0.43779 6.79563 0.00942 0.08617 56 THBD -0.27142 7.19875 0.00978 0.08803 57 CXCL1 -0.22969 7.02101 0.01012 0.08857 58 TNFRSF9 -0.37474 5.80484 0.01016 0.08857 59 MAP2K4 -0.24741 7.04344 0.01049 0.08858 60 IL1RAP -0.34095 8.95254 0.01088 0.09048 61 TLR1 -0.26287 10.98474 0.01156 0.09196 62 TRIM39 -0.20414 5.99055 0.01181 0.09263 63 CTSS -0.20312 12.90980 0.01218 0.09310 64 CXCR4 -0.20158 11.23692 0.01221 0.09310 65 ITGAM -0.28917 9.12664 0.01246 0.09333 66 NFKBIA -0.28549 10.02395 0.01275 0.09333 67 CD59 -0.21097 8.17375 0.01297 0.09371 68 MAPKAPK2 -0.25423 7.37751 0.01373 0.09790             1 h post-NAC Num Gene Fold change (log base 2) Mean of expression (log base 2) p value BH-FDR 1 SH2B2 -0.42925 6.02222 0.00004 0.00919 2 CXCL16 -0.51237 7.11425 0.00007 0.00919 3 IFITM2 -0.57478 13.63717 0.00009 0.00919 4 MME -0.67545 9.73726 0.00010 0.00919 5 ICAM3 -0.46128 11.69698 0.00011 0.00919 6 TNFRSF10C -0.50106 12.17761 0.00012 0.00919 7 SELPLG -0.40222 9.86928 0.00016 0.00919 8 IL6R -0.49408 9.32626 0.00016 0.00919 9 SLC11A1 -0.61748 10.44784 0.00018 0.00919   166 Num Gene Fold change (log base 2) Mean of expression (log base 2) p value BH-FDR 10 PECAM1 -0.41122 11.01328 0.00018 0.00919 11 FOS -0.44052 9.28078 0.00020 0.00919 12 TREM1 -0.49283 10.31882 0.00020 0.00919 13 LILRB2 -0.43913 8.83497 0.00027 0.00930 14 FPR2 -0.63732 10.44498 0.00029 0.00930 15 CD46 -0.37799 11.46722 0.00029 0.00930 16 CXCR2 -0.56360 12.38269 0.00030 0.00930 17 BCL6 -0.61347 10.62100 0.00034 0.00930 18 CXCR1 -0.48038 10.60998 0.00034 0.00930 19 CEBPB -0.38693 10.64674 0.00037 0.00930 20 LAMP2 -0.42434 10.05176 0.00038 0.00930 21 LILRB3 -0.54450 9.09908 0.00038 0.00930 22 FCGR2A -0.48476 12.66895 0.00039 0.00930 23 MAPK3 -0.40446 7.89827 0.00041 0.00930 24 TLR8 -0.44267 9.27357 0.00050 0.00980 25 CSF3R -0.47589 12.65277 0.00053 0.00980 26 IGF2R -0.61143 9.70412 0.00054 0.00980 27 PLAUR -0.54069 8.36615 0.00054 0.00980 28 LYN -0.53249 9.70071 0.00057 0.00980 29 TNFSF14 -0.29478 7.28391 0.00057 0.00980 30 TLR4 -0.42691 9.66203 0.00059 0.00980 31 SBNO2 -0.37809 7.93750 0.00059 0.00980 32 LCP1 -0.36187 12.72480 0.00060 0.00980 33 HCK -0.41343 11.11635 0.00062 0.00980 34 ITGAM -0.44442 9.21407 0.00062 0.00980 35 STAT5B -0.34373 10.15374 0.00068 0.01024 36 SPP1 -0.49849 5.04972 0.00069 0.01024 37 TNFRSF1A -0.36733 10.57325 0.00077 0.01069 38 CD97 -0.39962 11.43345 0.00077 0.01069   167 Num Gene Fold change (log base 2) Mean of expression (log base 2) p value BH-FDR 39 LILRA1 -0.34728 7.64404 0.00078 0.01069 40 PYCARD -0.36158 7.55069 0.00080 0.01069 41 IL1B -0.49747 6.94773 0.00089 0.01157 42 G6PD -0.37810 9.07178 0.00093 0.01177 43 ITGB2 -0.36557 10.21652 0.00096 0.01177 44 CFP -0.32398 8.99847 0.00099 0.01177 45 SELL -0.32724 11.78041 0.00102 0.01189 46 FCGR3A -0.42532 12.86052 0.00116 0.01300 47 LILRA5 -0.37735 8.39723 0.00119 0.01311 48 ITGA5 -0.34919 9.11538 0.00127 0.01372 49 CSF2RB -0.43689 10.74932 0.00142 0.01471 50 MAPK1 -0.33648 9.90171 0.00145 0.01475 51 NOTCH1 -0.38309 8.45817 0.00155 0.01551 52 IGF1R -0.45268 9.01475 0.00159 0.01556 53 PRKCD -0.40453 9.66496 0.00167 0.01564 54 STAT3 -0.41472 10.44553 0.00174 0.01564 55 MEFV -0.42093 8.73817 0.00176 0.01564 56 PSEN1 -0.27758 9.39337 0.00178 0.01564 57 CR1 -0.56501 10.77591 0.00180 0.01564 58 TNFRSF1B -0.34952 9.67764 0.00184 0.01564 59 TFEB -0.35507 6.95673 0.00185 0.01564 60 MYD88 -0.32247 9.94446 0.00186 0.01564 61 HLA-C -0.28264 12.82023 0.00198 0.01600 62 F2RL1 -0.37059 6.85214 0.00203 0.01616 63 CREB5 -0.50566 9.68736 0.00213 0.01672 64 ICAM1 -0.43053 7.02257 0.00217 0.01674 65 STAT6 -0.31061 10.78655 0.00232 0.01733 66 IRF1 -0.33136 9.98113 0.00241 0.01752 67 PIK3CD -0.28766 10.09258 0.00250 0.01782   168 Num Gene Fold change (log base 2) Mean of expression (log base 2) p value BH-FDR 68 SYK -0.40222 9.29223 0.00281 0.01977 69 TYK2 -0.24736 8.86966 0.00306 0.02100 70 PIK3CG -0.28518 8.91012 0.00325 0.02177 71 IFNAR1 -0.26987 7.18106 0.00334 0.02186 72 NCF4 -0.42746 10.08186 0.00387 0.02497 73 MAPKAPK2 -0.28172 7.58962 0.00435 0.02779 74 CD59 -0.25768 8.23113 0.00477 0.02875 75 IL18RAP -0.45348 8.40744 0.00479 0.02875 76 IL1RAP -0.44851 8.96738 0.00482 0.02875 77 IL13RA1 -0.41704 9.59964 0.00483 0.02875 78 AMICA1 -0.35871 12.44931 0.00487 0.02875 79 ITGAX -0.47159 11.38282 0.00518 0.02994 80 HLA-B -0.25088 13.76096 0.00569 0.03221 81 KIR2DS1 -0.44659 4.31737 0.00630 0.03461 82 LAMP1 -0.24547 9.25516 0.00651 0.03524 83 LTBR -0.23104 7.95589 0.00655 0.03524 84 IL1R2 -0.58828 8.95366 0.00699 0.03687 85 CD53 -0.19541 12.16929 0.00716 0.03745 86 TLR6 -0.41146 9.09645 0.00750 0.03814 87 TGFB1 -0.26505 10.08417 0.00772 0.03853 88 PTPRC -0.32583 11.90161 0.00793 0.03921 89 ENTPD1 -0.36552 8.39080 0.00828 0.04060 90 LCN2 -0.31924 7.16996 0.00843 0.04096 91 MAP3K5 -0.27852 9.68357 0.00854 0.04114 92 UBC -0.17917 13.81696 0.00872 0.04161 93 HLA-E -0.24404 12.54664 0.00917 0.04338 94 IL4R -0.23683 8.77086 0.00991 0.04650 95 TLR5 -0.36719 6.23876 0.01007 0.04686 96 IL17RA -0.34454 9.23696 0.01072 0.04797   169 Num Gene Fold change (log base 2) Mean of expression (log base 2) p value BH-FDR 97 MAPK14 -0.33669 10.18840 0.01090 0.04827 98 FUT7 -0.30326 7.57428 0.01111 0.04881 99 CD63 -0.22607 9.47920 0.01132 0.04934 100 BST1 -0.34862 9.78121 0.01263 0.05305 101 NFKBIA -0.26841 9.87495 0.01286 0.05314 102 HLA-A -0.21417 13.11221 0.01287 0.05314 103 TMUB2 -0.21936 8.20372 0.01297 0.05315 104 EP300 -0.22366 9.61516 0.01328 0.05400 105 THBD -0.37110 7.29845 0.01407 0.05596 106 JAK1 -0.22423 9.42365 0.01526 0.05983 107 NOD2 -0.29074 8.00214 0.01597 0.06175 108 BCL10 -0.19581 8.86594 0.01642 0.06305 109 INPP5D -0.21213 9.75152 0.01820 0.06885 110 TXNIP -0.25282 13.46559 0.01831 0.06885 111 TNFRSF9 -0.38813 5.96035 0.01864 0.06915 112 TAPBP -0.22339 9.64090 0.01883 0.06940 113 TLR2 -0.40233 10.12561 0.01910 0.06944 114 MAP3K1 -0.18328 9.61892 0.01996 0.07164 115 PSMB9 -0.20825 10.54096 0.02016 0.07187 116 POLR2A -0.24348 8.40076 0.02122 0.07374 117 CTSS -0.20403 13.06263 0.02243 0.07586 118 CD14 -0.25061 9.93360 0.02244 0.07586 119 IRF2 -0.22656 9.39948 0.02252 0.07586 120 MAP2K4 -0.22928 7.20740 0.02522 0.08380 121 RELA -0.25518 5.82976 0.02552 0.08380 122 TLR1 -0.29911 11.20574 0.02578 0.08380 123 CCND3 -0.23709 9.39215 0.02579 0.08380 124 PVR -0.19642 5.56870 0.02658 0.08534 125 CLEC6A -0.31330 4.80231 0.03023 0.09538   170 Num Gene Fold change (log base 2) Mean of expression (log base 2) p value BH-FDR 126 IL1R1 -0.39943 6.28416 0.03133 0.09796             2 h post-NAC Num Gene Fold change (log base 2) Mean of expression (log base 2) p value BH-FDR 1 PLAUR -0.51240 8.36709 0.00003 0.00959 2 PECAM1 -0.37635 11.03564 0.00004 0.00959 3 ICAM1 -0.55325 7.02482 0.00006 0.00959 4 LYN -0.45517 9.74705 0.00007 0.00959 5 IL6R -0.50806 9.31245 0.00012 0.00959 6 TNFRSF10C -0.54796 12.18445 0.00015 0.00959 7 IFITM2 -0.50461 13.67101 0.00017 0.00959 8 TNFRSF1B -0.31460 9.72041 0.00018 0.00959 9 STAT5B -0.40791 10.15548 0.00019 0.00959 10 CR1 -0.56050 10.81906 0.00021 0.00959 11 CXCR1 -0.50010 10.56324 0.00023 0.00959 12 SLC11A1 -0.61373 10.45845 0.00024 0.00959 13 FCGR2A -0.50239 12.65967 0.00026 0.00959 14 MME -0.74876 9.64544 0.00028 0.00959 15 CXCR2 -0.49724 12.39902 0.00028 0.00959 16 BCL6 -0.56151 10.69963 0.00030 0.00959 17 SELPLG -0.38085 9.76854 0.00034 0.00959 18 LILRB3 -0.50753 9.11462 0.00035 0.00959 19 CSF3R -0.47960 12.67138 0.00036 0.00959 20 G6PD -0.40975 9.13063 0.00038 0.00959 21 MAPKAPK2 -0.34594 7.58361 0.00038 0.00959 22 IGF2R -0.58828 9.75678 0.00041 0.00982 23 STAT3 -0.38150 10.50393 0.00045 0.01023   171 Num Gene Fold change (log base 2) Mean of expression (log base 2) p value BH-FDR 24 SYK -0.35933 9.31122 0.00052 0.01096 25 ITGB2 -0.33136 10.22623 0.00052 0.01096 26 SELL -0.37794 11.75628 0.00057 0.01125 27 TREM1 -0.48919 10.35975 0.00059 0.01125 28 LILRB2 -0.36728 8.83897 0.00062 0.01125 29 CD46 -0.32723 11.45702 0.00063 0.01125 30 IL17RA -0.44205 9.23355 0.00064 0.01125 31 PSEN1 -0.27221 9.42105 0.00066 0.01133 32 PRKCD -0.36726 9.69118 0.00077 0.01259 33 MEFV -0.43634 8.74304 0.00079 0.01259 34 FCGR3A -0.41216 12.87720 0.00081 0.01259 35 FPR2 -0.53158 10.43375 0.00083 0.01259 36 LAMP2 -0.41256 10.06701 0.00087 0.01265 37 ICAM3 -0.42918 11.73023 0.00088 0.01265 38 TNFRSF1A -0.36970 10.58467 0.00115 0.01520 39 HLA-B -0.24860 13.74151 0.00115 0.01520 40 PTGS2 -0.44882 7.29358 0.00123 0.01553 41 NOTCH1 -0.48163 8.52429 0.00133 0.01577 42 MAP3K5 -0.29831 9.73135 0.00135 0.01577 43 TFE3 -0.29296 6.56471 0.00138 0.01577 44 NCF4 -0.43205 10.05580 0.00154 0.01653 45 PIK3CD -0.32459 10.14508 0.00160 0.01653 46 MAP2K4 -0.32177 7.13756 0.00161 0.01653 47 ITGAM -0.40621 9.27163 0.00163 0.01653 48 TGFB1 -0.26008 10.12622 0.00169 0.01653 49 MAPK3 -0.35591 7.96989 0.00169 0.01653 50 CSF2RB -0.41381 10.77604 0.00175 0.01655 51 POU2AF1 -0.36498 5.14908 0.00187 0.01743 52 CD97 -0.38846 11.46020 0.00193 0.01770   172 Num Gene Fold change (log base 2) Mean of expression (log base 2) p value BH-FDR 53 ITGAX -0.46866 11.40345 0.00224 0.02018 54 SH2B2 -0.40315 6.02263 0.00238 0.02102 55 HLA-E -0.24263 12.55186 0.00241 0.02102 56 PIK3CG -0.24030 8.96590 0.00289 0.02355 57 LCP1 -0.26840 12.75551 0.00291 0.02355 58 CREB5 -0.43935 9.68336 0.00292 0.02355 59 TLR8 -0.39182 9.23350 0.00312 0.02396 60 BCL10 -0.26026 8.85860 0.00314 0.02396 61 POLR2A -0.28401 8.48227 0.00327 0.02457 62 HCK -0.34262 11.13835 0.00336 0.02495 63 ALCAM -0.27672 5.90080 0.00352 0.02577 64 LAMP1 -0.22429 9.33342 0.00366 0.02617 65 IL1R1 -0.41692 6.31430 0.00370 0.02617 66 HLA-C -0.26343 12.83725 0.00375 0.02617 67 ITGA5 -0.31684 9.16518 0.00386 0.02619 68 IRF2 -0.23592 9.42625 0.00394 0.02619 69 INPP5D -0.25107 9.80007 0.00401 0.02619 70 ENTPD1 -0.41597 8.34658 0.00422 0.02663 71 IL13RA1 -0.40665 9.52235 0.00427 0.02663 72 IRF1 -0.26501 10.01429 0.00430 0.02663 73 CEBPB -0.33663 10.62898 0.00432 0.02663 74 TAPBP -0.29800 9.66703 0.00458 0.02765 75 CFP -0.24133 8.98738 0.00470 0.02806 76 IGF1R -0.43936 9.01075 0.00500 0.02954 77 LILRA1 -0.24965 7.60738 0.00518 0.02964 78 SBNO2 -0.36061 7.94965 0.00550 0.03080 79 C5 -0.30073 4.90110 0.00567 0.03142 80 STAT6 -0.28028 10.82927 0.00596 0.03272 81 IFNAR1 -0.26983 7.17498 0.00606 0.03295   173 Num Gene Fold change (log base 2) Mean of expression (log base 2) p value BH-FDR 82 AMICA1 -0.34028 12.45552 0.00671 0.03611 83 IL1RAP -0.37635 8.88676 0.00695 0.03691 84 TAP1 -0.21760 8.33540 0.00722 0.03773 85 IL1R2 -0.57008 8.87703 0.00780 0.03936 86 IFNAR2 -0.20690 10.12686 0.00803 0.03970 87 TOLLIP -0.30889 7.60416 0.00830 0.04023 88 OSM -0.39570 4.43704 0.00835 0.04023 89 EP300 -0.24648 9.68384 0.00901 0.04226 90 MAPK1 -0.26848 9.95290 0.00908 0.04226 91 TMUB2 -0.20838 8.22090 0.01029 0.04744 92 ATG7 -0.21612 8.14732 0.01043 0.04744 93 TLR4 -0.32998 9.67077 0.01045 0.04744 94 FOS -0.34730 9.22377 0.01063 0.04785 95 BCL2L1 -0.59454 11.80576 0.01176 0.05164 96 THBD -0.34440 7.29652 0.01215 0.05293 97 CD59 -0.22503 8.24192 0.01270 0.05489 98 ARG1 -0.64546 4.88123 0.01314 0.05550 99 TLR6 -0.31358 9.10480 0.01357 0.05685 100 TNFRSF9 -0.52100 6.03715 0.01401 0.05828 101 CD53 -0.27752 12.08973 0.01486 0.06000 102 IL1B -0.41996 6.90036 0.01553 0.06224 103 TFEB -0.25226 7.00648 0.01663 0.06615 104 FUT7 -0.30676 7.53266 0.01675 0.06617 105 F2RL1 -0.27544 6.85213 0.01694 0.06643 106 ITGA1 -0.29020 5.16582 0.01721 0.06699 107 JAK1 -0.22808 9.47475 0.01821 0.06992 108 AMMECR1L -0.18473 7.13276 0.01849 0.07051 109 GUSB -0.20450 6.07115 0.01935 0.07327 110 TLR1 -0.28521 11.19203 0.02027 0.07570   174 Num Gene Fold change (log base 2) Mean of expression (log base 2) p value BH-FDR 111 CLEC7A -0.27937 10.28716 0.02100 0.07788 112 CXCL1 -0.35449 7.27461 0.02208 0.08017 113 MYD88 -0.19914 9.97411 0.02352 0.08385 114 FCGR1A -0.35619 6.63647 0.02473 0.08761 115 SAP130 -0.22998 6.98024 0.02531 0.08907 116 HLA-A -0.15112 13.10322 0.02582 0.09030 117 CD14 -0.20406 9.95385 0.02826 0.09696             6 h post-NAC Num Gene Fold change (log base 2) Mean of expression (log base 2) p value BH-FDR 1 HCK -0.39285 11.04826 0.00134 0.08165 2 SBNO2 -0.42319 7.86391 0.00137 0.08165 3 MEFV -0.37942 8.54294 0.00185 0.08165 4 CSF2RB -0.48445 10.68831 0.00197 0.08165 5 IFITM2 -0.46748 13.52347 0.00201 0.08165 6 TNFRSF1B -0.30845 9.58090 0.00221 0.08165 7 TNFRSF1A -0.35071 10.47364 0.00238 0.08165 8 TNFRSF10C -0.50087 12.04725 0.00251 0.08165 9 CD97 -0.40704 11.40961 0.00255 0.08165 10 IL17RA -0.40142 9.11382 0.00285 0.08165 11 IL6R -0.36129 9.29517 0.00286 0.08165 12 LYN -0.44656 9.68098 0.00295 0.08165 13 ITGB2 -0.35045 10.15336 0.00300 0.08165 14 FPR2 -0.41510 10.29383 0.00317 0.08165 15 TAPBP -0.30555 9.66464 0.00337 0.08165 16 IGF2R -0.47109 9.46373 0.00351 0.08165 17 LAMP2 -0.30864 9.91907 0.00369 0.08165   175 Num Gene Fold change (log base 2) Mean of expression (log base 2) p value BH-FDR 18 TNFSF14 -0.36091 7.20889 0.00373 0.08165 19 MTMR14 -0.21584 8.43101 0.00389 0.08165 20 BCL6 -0.57448 10.55257 0.00391 0.08165 21 THBD -0.49402 7.14347 0.00391 0.08165 22 LILRB3 -0.43090 8.80632 0.00409 0.08165 23 CXCR2 -0.48975 12.25472 0.00450 0.08165 24 FCGR2A -0.40149 12.55261 0.00466 0.08165 25 IRF2 -0.20569 9.36328 0.00470 0.08165 26 ICAM3 -0.36654 11.65531 0.00473 0.08165 27 RELA -0.25801 5.98722 0.00477 0.08165 28 JAK3 -0.29538 9.15445 0.00491 0.08165 29 IKBKG -0.24127 7.51329 0.00538 0.08234 30 IRF1 -0.28507 9.83129 0.00541 0.08234 31 NFKB2 -0.24894 7.63490 0.00578 0.08234 32 MAVS -0.25558 7.42761 0.00580 0.08234 33 STAT5B -0.37717 10.12680 0.00596 0.08234 34 CXCR1 -0.43010 10.42416 0.00600 0.08234 35 CSF3R -0.35041 12.41145 0.00669 0.08424 36 STAT6 -0.31979 10.58640 0.00687 0.08424 37 MAPK3 -0.34442 7.87145 0.00719 0.08424 38 CEBPB -0.32468 10.42017 0.00727 0.08424 39 CD46 -0.22243 11.39774 0.00734 0.08424 40 PECAM1 -0.23804 10.93394 0.00745 0.08424 41 CREB5 -0.39337 9.72132 0.00751 0.08424 42 PIK3CD -0.40343 10.04419 0.00752 0.08424 43 PSEN1 -0.22709 9.28644 0.00789 0.08550 44 ITGAX -0.43224 11.01325 0.00817 0.08550 45 PLAUR -0.24897 8.21881 0.00825 0.08550 46 IFNAR2 -0.21353 10.01553 0.00957 0.09215   176 Num Gene Fold change (log base 2) Mean of expression (log base 2) p value BH-FDR 47 SYK -0.31195 9.24604 0.00974 0.09215 48 NFKB1 -0.29228 7.44270 0.01015 0.09350 49 IL4R -0.32796 8.84652 0.01039 0.09350 50 STAT3 -0.36791 10.32957 0.01095 0.09694 51 CR1 -0.45070 10.63237 0.01131 0.09831 52 PRKCD -0.29821 9.59968 0.01178 0.09831 53 TYK2 -0.19877 8.65190 0.01182 0.09831 54 IL1B -0.34481 7.05582 0.01275 0.09933 55 LAMP1 -0.22042 9.38729 0.01287 0.09933 56 SLC11A1 -0.39947 10.09550 0.01338 0.09933 57 TMUB2 -0.21722 8.05263 0.01379 0.09933 58 LTBR -0.26908 7.84020 0.01381 0.09933 59 SH2B2 -0.16936 6.06142 0.01461 0.09933 60 GTF3C1 -0.22987 6.56645 0.01469 0.09933 61 TLR8 -0.22117 9.32320 0.01562 0.09933 62 PIK3CG -0.19763 8.88822 0.01565 0.09933 63 JAK1 -0.23822 9.49253 0.01583 0.09933 64 ITGAM -0.29310 9.14088 0.01619 0.09933 65 ITGA5 -0.26183 9.00595 0.01625 0.09933 66 ITGAL -0.19363 9.10291 0.01633 0.09933 67 MYD88 -0.17660 9.93749 0.01681 0.09933 68 ENTPD1 -0.34479 7.90124 0.01701 0.09933 69 AMICA1 -0.23777 12.22409 0.01729 0.09972 70 TFEB -0.34537 6.80774 0.01747 0.09972 71 TXNIP -0.16276 13.36454 0.01762 0.09972   177 Appendix C  Publications arising during my PhD studies:  (1) Journals 1)  Kim YW, Singh A, Shannon CP, Thiele J, Steacy LM, Ellis AK, et al. Investigating Immune Gene Signatures in Peripheral Blood from Subjects with Allergic Rhinitis Undergoing Nasal Allergen Challenge. J Immunol. 2017 Nov. 15;199(10):3395–405. 2)  Singh A, Shannon CP, Kim YW, Yang CX, Balshaw R, Cohen Freue GV, et al. Novel Blood-based Transcriptional Biomarker Panels Predict the Late Phase Asthmatic Response. Am J Respir Crit Care Med. 2018 Feb 15;197(4):450-462.  3) Yang CX, Singh A, Kim YW, Conway EM, Carlsten C, Tebbutt SJ. Diagnosis of Western Red Cedar Asthma Using a Blood-based Gene Expression Biomarker Panel. Am J Respir Crit Care Med. 2017 May 2; 196(12):1615-1617. 4) Pesenacker AM, Wang AY, Singh A, Gillies J, Kim YW, Piccirillo CA, et al. A Regulatory T-Cell Gene Signature Is a Specific and Sensitive Biomarker to Identify Children with New-Onset Type 1 Diabetes. Diabetes. 2016 Apr.;65(4):1031–9.  (2) Conferences 1) Kim YW, Gliddon DR, Shannon CP, Singh A, Hickey PLC, Ellis AK, Neighbour H, Larché M, Tebbutt SJ. “Systemic immune pathways associated with the mechanism of CatSynthetic Peptide ImmunoRegulatory Epitopes, a novel immunotherapy, in whole blood of catallergic people” Allergy Asthma Clin Immunol. Aug. 25, 2016, 12(Suppl 1): A53 2) Gliddon DR, Kim YW, Shannon CP, Singh A, Tebbutt SJ, Hickey PLC, Ellis AK,   178 Neighbour H, Larché M. “Whole blood immune transcriptome profiling reveals systemic pathways associated with the mechanism of action of cat–synthetic peptide immune–regulatory” EAACI Online Library. Gliddon D. Jun. 6, 2015; 104988 3) Kim YW, Singh A, Shannon CP, Gauvreau GM, Tebbutt SJ. “Transcriptional networks in whole blood of asthmatics” Allergy Asthma Clin Immunol 2014 Dec. 18;10(Suppl 2): A58    179 Appendix D  Presentations arising during my PhD studies:  1) Kim YW, Singh A, Shannon CP, Ellis AK, Neighbour H, Larché M, Tebbutt SJ. “Immune gene signatures in blood of patients with allergic rhinitis following nasal allergen challenge” 2017 Medicine Research Expo at UBC, Vancouver, BC (poster presentation, the Best Poster Award), Oct. 31st, 2017. 2) Kim YW, Singh A, Shannon CP, Ellis AK, Neighbour H, Larché M, Tebbutt SJ. “Immune gene signatures in blood of patients with allergic rhinitis following nasal allergen challenge” 2017 (72nd) CSACI Annual Scientific Meeting, Toronto, ON (two-minute oral presentation and poster presentation, 2nd place of Best Poster Award – Allergic Rhinitis/Asthma), Oct. 11-15th, 2017. 3) Kim YW, “Investigation of systemic immune response of allergic rhinitis using peripheral blood during nasal allergen challenge” The 11th AllerGen Trainee Symposium, Burlington, ON (one-minute oral presentation), May. 3-5th, 2017. 4)  Kim YW, Shannon CP, Singh A, Ellis AK, Neighbour H, Larché M, Tebbutt SJ. “Investigating systemic immune responses in the pathophysiology of allergic rhinitis under the nasal allergen challenge model” 2016 (71st) CSACI Annual Scientific Meeting & AllerGen Poster Competition, Montréal, QC (two-minute oral presentation and poster presentation), Sept. 29-Oct. 2nd, 2016. 5) Kim YW, Shannon CP, Singh A, Ellis AK, Neighbour H, Larché M, Tebbutt SJ. “Investigating systemic immune responses in peripheral blood of cat allergic people under the nasal allergen challenge model” 2016 AllerGen’s 8th Research Conference, Vancouver, BC (one-minute oral presentation and poster presentation), May. 29-Jun. 1st, 2016   180 6) Kim YW, Gliddon DR, Shannon CP, Singh A, Hickey PLC, Ellis AK, Neighbour H, Larché M, Tebbutt SJ. “Systemic immune pathways associated with the mechanism of Cat-Synthetic Peptide Immuno-Regulatory Epitopes, a novel immunotherapy, in whole blood of cat-allergic people” 2015 (70th) CSACI Annual Scientific Meeting & AllerGen Poster Competition, Vancouver, BC (two-minute oral presentation and poster presentation, 1st place of Best Poster Award – Ph.D.), Oct. 21-24th, 2015. 7) Kim YW, “Systemic immune pathways associated with the mechanism of Cat-Synthetic Peptide Immuno-Regulatory Epitopes, a novel immunotherapy, in whole blood of cat-allergic people” The 10th Annual AllerGen Trainee Symposium, Toronto, ON (one-minute oral presentation), Apr. 29-May. 1st, 2015. 8) Kim YW, Singh A, Shannon CP, Yang CX, Ellis AK, Neighbour H, Tebbutt SJ. “Blood transcriptional changes in subjects undergoing nasal allergen challenge” The 4th Annual Norman Bethune Symposium, Vancouver, BC (one-minute oral presentation and poster presentation), Apr. 16th, 2015. 9) Kim YW, Singh A, Shannon CP, Yang CX, Ellis AK, Neighbour H, Tebbutt SJ. “Blood transcriptional changes in subjects undergoing nasal allergen challenge” 2015 Heart + Lung Scientific Symposium, Vancouver, BC (poster presentation), Mar. 26-27th, 2015. 10) Kim YW, Singh A, Shannon CP, Gauvreau GM, Tebbutt SJ. “Transcriptional networks in whole blood of asthmatics” 2014 (69th) CSACI Annual Scientific Meeting & AllerGen Poster Competition, Ottawa, ON (two-minute oral presentation and poster presentation), Oct. 23-26th, 2014.   181 11) Kim YW, Singh A, Shannon CP, Gauvreau GM, Tebbutt SJ. “Transcriptional networks in whole blood of asthmatics” Summer Student Research Day 2014, Vancouver, BC (15-minute oral presentation), Aug. 14th, 2014.   

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:
https://iiif.library.ubc.ca/presentation/dsp.24.1-0364662/manifest

Comment

Related Items