Epigenetic Trajectories to Childhood Asthma

Item Type text; Electronic Dissertation

Authors DeVries, Avery

Publisher The University of Arizona.

Rights Copyright © is held by the author. Digital access to this material is made possible by the University Libraries, University of Arizona. Further transmission, reproduction, presentation (such as public display or performance) of protected items is prohibited except with permission of the author.

Download date 04/10/2021 09:24:24

Link to Item http://hdl.handle.net/10150/630233 1

EPIGENETIC TRAJECTORIES TO CHILDHOOD ASTHMA

Avery DeVries

A Dissertation Submitted to the Faculty of the

DEPARTMENT OF CELLULAR AND MOLECULAR MEDICINE

In Partial Fulfillment of the Requirements

For the Degree of

DOCTOR OF PHILOSOPHY

In the Graduate College

THE UNIVERSITY OF ARIZONA

2018

STATEMENT BY AUTHOR

This dissertation has been submitted in partial fulfillment of the requirements for an advanced degree at the University of Arizona and is deposited in the University Library to be made available to borrowers under rules of the Library.

Brief quotations from this dissertation are allowable without special permission, provided that an accurate acknowledgement of the source is made. Requests for permission for extended quotation from or reproduction of this manuscript in whole or in part may be granted by the head of the major department or the Dean of the Graduate College when in his or her judgment the proposed use of the material is in the interests of scholarship. In all other instances, however, permission must be obtained from the author.

SIGNED: Avery DeVries

ACKNOWLEDGEMENTS

I would like to thank my dissertation mentor, Donata Vercelli, for endless support throughout this journey and for always encouraging and challenging me to be a better scientist. And to my committee members, thank you for your support, critiques and comments, all of which have been invaluable. 5

DEDICATION

For my loving parents, my children, and my amazing husband, David. 6

TABLE OF CONTENTS

LIST OF FIGURES ……………………………………………………………. 8 LIST OF TABLES …………………………………………………………..... 9 LIST OF ABBREVIATIONS …………………………………………... 10 ABSTRACT ………………………………………………………………..…. 11 CHAPTER 1: INTRODUCTION …………………………………………... 13 1.1 – THE ASTHMA PUZZLE …………………………………... 13 1.1.1 – ASTHMA EPIDEMIOLOGY …………………………... 14 1.1.1.1. Asthma as a disease of childhood ……….….. 14 1.1.1.2. Risk factors for asthma …………………... 15 1.1.1.3. Molecular phenotypes (endotypes) of asthma ………..….………….…...……….…….………….……. 16 1.1.2 – ASTHMA AS A COMPLEX DISEASE …….…..… 17 1.1.2.1 – Asthma genetics …………………….…..… 17 1.1.2.2 – Asthma and the environment ……...…… 18 1.1.2.3 – Development and natural history ….…..…… 19 1.2 – EPIGENETICS …….…………..…….…………………..……… 21 1.2.1 – DNA METHYLATION ………………………..…. 21 1.2.2 – DNA METHYLATION PROFILING TECHNIQUES ………………………..…………………………………………. 23 1.3. – ASTHMA EPIGENETICS ………………………………..…. 26 1.3.1 – EPIGENOME-WIDE STUDIES …………..………. 26 1.3.1.1 – Asthma-associated epigenetic mechanisms in blood immune cells ………………………………..…. 28 1.3.1.2 – Asthma-associated epigenetic mechanisms in the airways ………………………………………..…. 32 1.3.1.3 – Conclusions ………………………………..…. 36 1.4 – DISSERTATION OUTLINE ………………………………..…. 39 CHAPTER 2: EPIGENOME-WIDE ANALYSIS LINKS SMAD3 METHYLATION AT BIRTH TO ASTHMA IN CHILDREN OF ASTHMATIC MOTHERS ……………………………………………………..……………. 42 2.1 – INTRODUCTION ………………………..…………………. 42 2.2 – METHODS …………..………………………………………. 45 2.3 – RESULTS ………………..…………………………………. 52 2.3.1 – Cord blood cells from IIS children harbor asthma- associated DMRs …………………………………………... 52 2.3.2 – Asthma-associated DMRs cluster in regulatory and pro- inflammatory gene networks ………….…….…………………. 55 2.3.3 – The SMAD3 promoter is significantly hypermethylated in asthmatic children of asthmatic mothers ………..…………. 57 2.3.4 – The association between SMAD3 promoter hypermethylation and childhood asthma replicates in two independent birth cohorts ………………..…………………. 60 2.3.5 – Maternal asthma modifies the relation between neonatal SMAD3 methylation and IL-1 producing capacity ………..…. 62 7

2.4 – DISCUSSION ……………………………..……………………. 65 CHAPTER 3: NEONATAL DNA METHYLATION PROFILES AND THE MATERNAL PRENATAL IMMUNE MILIEU …………………..………. 69 3.1 – INTRODUCTION …………………………..………………. 69 3.1.1 – Association between maternal immune profiles during pregnancy and risk for childhood asthma ………..…………. 70 3.2 – METHODS ………………………………..…………………. 72 3.3 – RESULTS ……………………………………..……………. 76 3.3.1 – Profiles of neonatal DNA methylation are associated with the maternal prenatal immune profile selectively in neonates born to non-asthmatic mothers …………..………………………. 76 3.3.2 – Neonatal CpG sites associated with the maternal IFN/IL- 13 ratio cluster in the TGFB1 pathway …………..………. 81 3.4 – DISCUSSION ……..……………………………..…...….....…... 86 CHAPTER 4: LEVERAGING TRANSCRIPTOMICS AND NETWORK APPROACHES TO EXPLORE THE ASTHMA-PROTECTIVE EFFECTS OF FARM EXPOSURE …………………………………………………………... 89 4.1 – INTRODUCTION ………………………………………..…. 89 4.1.1 – Intranasal treatment with allergen and Amish dust extract induces T cells in the lung ……………………….………….. 90 4.1.2 – T cells in the lungs of mice treated with Amish dust extract express V4 …………..………………………………. 91 4.1.3 – Network approaches to characterize the lung and 17 T cell transcriptomes in associated with asthma protection …... 92 4.2 – METHODS …………………………………………………... 94 4.3 – RESULTS …………………………………………………. 100 4.3.1 – Network analysis of transcriptome profiles in unfractionated lung cells …………………………………. 100 4.3.2 – Characterization of the transcriptome of isolated T cells .….….…………….……………………………………. 103 4.3.3 – Determining the relationships between modules associated with protection and the 17 T cell gene signature …………. 111 4.4 – DISCUSSION …………………………………………………. 118 CHAPTER 5: CONCLUSIONS AND FUTURE DIRECTIONS …………. 120 5.1 – CONCLUSIONS …………………………………………. 120 5.2 – FUTURE DIRECTIONS …………………………………. 125 APPENDIX: SUPPLEMENTARY MATERIALS FOR CHAPTER 2 …. 133 A.1 – METHODS …………………………………………………. 133 A.2 – SUPPLEMENTARY FIGURES …………………………. 137 A.3 – SUPPLEMENTARY TABLES …………………………. 145 REFERENCES …………………………………………………………. 239

LIST OF FIGURES

Figure 1.1. Literature on epigenetics and human asthma and allergic diseases, 2007-2017 …………………………………………………………………... 27 Figure 2.1. Overview of study design …………………………………... 47 Figure 2.2. Molecular interaction network of asthma-associated DMRs …... 54 Figure 2.3. Effects of maternal asthma on the association between neonatal SMAD3 methylation and childhood asthma in IIS, MAAS and COAST …... 59 Figure 2.4. Effects of maternal asthma on the relation between neonatal SMAD3 methylation and IL-1 protein production …………………………………... 63 Figure 3.1. Samples selected for pilot DNA methylation study …………... 74 Figure 3.2. Number of CpG sites associated with the maternal prenatal IFN/IL- 13 ratio at various levels of statistical significance ……………………….….. 78 Figure 3.3. Chromosomal location of neonatal CpG sites associated with the maternal prenatal IFN/IL-13 ratio in children of non-asthmatic mothers …... 79 Figure 3.4. Network of genes harboring differential neonatal methylation associated with the maternal prenatal IFN/IL-13 ratio in children of non- asthmatic mothers …………………………………………………………... 84 Figure 3.5. Relationships between neonatal DNA methylation in genes in the aryl hydrocarbon receptor and retinoic receptor pathways, the maternal prenatal IFN/IL-13 ratio, and childhood asthma in children of non-asthmatic mothers …………………………………………………………………………………... 85 Figure 4.1. Flow cytometry analysis of lung immune cells from a murine model of asthma …………………………………………………………….…..… 93 Figure 4.2. Experimental models and samples included in RNA-sequencing analysis ……………………………………………………………….….. 99 Figure 4.3. Module identification by WGCNA ………………...……….. 101 Figure 5.1. Working model: the relationship between TGF- signaling and potential trajectories to asthma …………………………………………. 124 Figure E1. Workflow of DNA methylation profiling …………………. 137 Figure E2. Correlation of bisulfite sequencing and DNA methylation microarray results …………………………………………………………………………. 138 Figure E3. Distribution of asthma-associated DMRs by chromosome and genomic location …………………………………………………………. 139 Figure E4. Regulatory potential of the SMAD3 DMR …………………. 140 Figure E5. Association between SMAD3 DNA methylation at birth and childhood asthma in the IIS cohort …………………………………………………. 141 Figure E6. SMAD3 CpG7 methylation in IIS and MAAS …………………. 142 Figure E7. Estimation of CBMC composition in the COAST cohort …. 143 Figure E8. Effects of maternal asthma on the association between neonatal SMAD3 methylation and childhood asthma in COAST after adjusting for CBMC composition …………………………………………………………………. 144 9

LIST OF TABLES

Table 1.1. Replicated asthma associations from GWAS ....………….….…. 20 Table 1.2. Epigenome-wide studies of asthma ………………..…………. 38 Table 2.1. Upstream regulators of genes containing asthma-associated DMRs ……....…………………………………………………………………………... 56 Table 2.2. LPS-stimulated CBMC-production of IL-1 in IIS children with or without a history of maternal asthma …………………………….…………….. 64 Table 3.1. Genomic distribution of neonatal CpG sites associated with the maternal prenatal IFN/IL-13 ratio in children of non-asthmatic mothers …... 80 Table 3.2. Top 5 upstream regulators of genes containing maternal IFN/IL-13 ratio-associated CpGs in children of non-asthmatic mothers ……….…….…… 83 Table 3.3. Top 5 canonical pathways enriched for genes containing maternal IFN/IL-13 ratio-associated CpGs in children of non-asthmatic mothers …... 83 Table 4.1. Correlations between WGCNA module eigengenes and phenotypes linked to asthma protection …………………………………………………. 102 Table 4.2. 17 T cell signature genes …………………………………. 105 Table 4.3. Modules enriched in 17 T cell signature genes …………………. 112 Table 4.4. Magenta module genes and their module membership …………. 112 Table 4.5. Enrichment of magenta module genes for canonical pathways …. 117 Table E1. Characteristics of IIS study subjects …………………………. 145 Table E2. Characteristics of non-asthmatic and asthmatic children in the IIS, MAAS and COAST study populations …………………………………. 146 Table E3. Complete list of annotated asthma-associated DMRs (n=589) in IIS children …………………………………………………………………. 147 Table E4. Primers used for bisulfite sequencing …………………………. 179 Table E5: Functional annotation of asthma-associated DMRs in IIS children …………………………………………………………………………………. 180 Table E6: Overlap between asthma-associated DMRs in IIS children and DNase I HSS in monocytes and naïve T cells …………………………………………. 231 Table E7. Asthma-associated DMRs in IIS children and DNase I HSS .… 237 Table E8: Association between child and parental risk factors at birth and asthma during childhood in IIS children …………………………………………. 238

LIST OF ABBREVIATIONS

450K: Illumina HumanMethylation450 BeadChip 5hmC: 5' hydroxymethylcytosine 5mC: 5' methylcytosine AEC: airway epithelial cells AHR: airway hyper-responsiveness BAL: Broncho-alveolar lavage CBMC: cord blood mononuclear cell CGI: CpG island CI: confidence interval COAST: Childhood Origins of ASThma Study CpG: CG dinucleotide DMC: differentially methylated CpG DMR: differentially methylated region eQTL: expression quantitative trait locus FC: fold change FDR: false discovery rate FEV1: forced expiratory volume in 1 sec GWAS: genome wide association study HSS: hypersensitive sites IIS: Infant Immune Study ILC2: innate lymphoid cell type 2 IPA: Ingenuity Pathway Analysis KD: knock down MAAS: Manchester Asthma and Allergy Study meQTL: methylation quantitative trait locus MIRA: Methylated CpG Island Recovery Assay MM: module membership OR: odds ratio PBMC: peripheral blood mononuclear cell RNA-seq: RNA sequencing RSV: respiratory syncytial virus RV: rhinovirus SNP: single nucleotide polymorphism TSS: transcription start site UTR: untranslated region WGCNA: Weighted Gene Co-expression Network Analysis 11

ABSTRACT

Asthma is the most common chronic disease of childhood and remains a public health concern in the United States and worldwide. Although it is characterized by recurrent, reversible bronchial obstruction, asthma is variable in its clinical expression and includes distinct cellular and molecular endotypes. In most cases, asthma manifestations (including subtle alterations to both adaptive and innate immunity) begin during the preschool years, even when chronic symptoms do not appear until adulthood. However, the lack of firm diagnostic criteria to distinguish children who will wheeze only transiently during early life viral illnesses from children who will wheeze persistently and then develop asthma prevents pinpointing the true inception of a child’s trajectory to the disease. Even though the evidence supporting the early origins of asthma is strong, the underlying mechanisms of asthma inception remain unknown. In this context, epigenetics is currently receiving much attention as a potential contributor to asthma pathogenesis in early life. To the extent that environmental and developmental factors are essential for asthma pathogenesis, epigenetic mechanisms are a plausible source of phenotypic variability because they mediate the responses to environmental stimuli and the timed unfolding of developmental programs. This work aims to better understand the roles of epigenetic and environmental mechanisms involved in asthma pathogenesis. The work presented herein shows that 1) an epigenetic trajectory to asthma is in place already at birth, at least in a subset of children, 2) distinct trajectories are evident in children born to asthmatic and non-asthmatic mothers, a finding that supports the importance of 12 pre-natal exposure, and 3) in mice, protection from asthma conferred by exposure to products from traditional farming environments likely involves 17 T cells.

Overall our data identify compelling candidates for future studies of the mechanisms underlying asthma development in humans and mouse models.

CHAPTER 1: INTRODUCTION

1.1 – THE ASTHMA PUZZLE

Asthma, a disease characterized by recurrent, reversible bronchial obstruction, is the most common chronic disease of childhood and remains a public health concern in the United States and worldwide [1]. The World Health

Organization has estimated in 2017 that over 235 million people are affected by asthma, a number which continues to increase [1]. It has also been estimated that the worldwide economic burden of asthma outweighs that of HIV/AIDS and tuberculosis combined [2].

Asthma is variable in its clinical expression, as evidenced by multiple, distinct cellular and molecular endotypes. In many cases, the disease can be controlled by medications, but for those who suffer from severe asthma, controlling symptoms proves difficult and remains an everyday struggle. Even among patients who adhere to daily therapy, airway remodeling and deficits in lung function beginning during early childhood create a significant risk for chronic lung disease, premature disability and early death [3].

It is well known that most cases of persistent asthma begin during the preschool years [4]. Several lines of evidence indicate that most people diagnosed with asthma in early adulthood had recurrent episodes of wheezing in early childhood, suggesting that the disease may have started years before the diagnosis

[4, 5]. However, the lack of firm diagnostic criteria to distinguish children who will wheeze only transiently during early life viral illnesses from children who 14 will wheeze persistently and then develop asthma prevents pinpointing the true inception of a child’s trajectory to the disease. While the epidemiological evidence supporting the early origins of asthma is strong [4, 6-8], the mechanisms underpinning asthma inception remain unknown. Genome-wide association studies have succeeded in identifying a number of asthma candidate genes of suggestive biological significance [9-12], but have failed to account for more than a modest proportion of phenotypic variance [13]. It is in this context that epigenetics is currently receiving much attention as a potential contributor to asthma pathogenesis in early life [14]. To the extent that environmental and developmental factors are essential for asthma pathogenesis [15], epigenetics, which sits squarely at the mechanistic intersection between these factors, is a plausible source of phenotypic variability.

1.1.1 – ASTHMA EPIDEMIOLOGY

1.1.1.1. Asthma as a disease of childhood

Even though asthma can be diagnosed at any age, several epidemiological studies support the notion that asthma typically, although not exclusively, begins in childhood [4, 15]. Birth cohorts and prospective studies have shown that even in individuals in whom chronic asthma symptoms do not occur until adulthood, reduced lung function [16] and higher frequency of episodic wheezing [4, 5] often appear during childhood. Although the lung is the organ most affected by asthma, epidemiological studies have shown that both adaptive and innate immunity are altered in asthmatics, often before a diagnosis is made [17, 18]. Together, these 15 findings emphasize that asthma pathogenesis unfolds along a trajectory that involves both respiratory and immune components.

1.1.1.2. Risk factors for asthma

Epidemiologists have identified many risk factors associated with asthma.

A child’s risk for asthma is greater if one or both parents are asthmatic, but risk is higher if the mother has asthma [19]. Indeed, maternal asthma is the strongest and most replicated risk factor for asthma in the child. In a meta-analysis of 33 studies, Lim and colleagues showed that a child’s risk for asthma was 3-fold higher if the mother had asthma and ~2.5-fold higher if the child’s father was asthmatic [20], and the contributions of maternal and paternal effects were significantly different (P=0.037) [20]. The mechanisms underlying these relationships are not well understood, but these findings suggest that the prenatal period is a developmental window critical in setting the trajectory to asthma during childhood. Other reports support this notion by showing associations between asthma in the child and prenatal factors such as maternal age, stress, and weight gain, smoking, infectious illness, and exposure to farming environments

[21-30]. Perinatal and early life factors (such as infections, mode of delivery, preterm birth, birthweight and exposures to microbes, air pollution, and tobacco smoke [22, 31, 32]) also modify risk for asthma, further emphasizing the lasting effects that early life exposures can have on the asthma trajectory. Some of these exposures will be further discussed in section 1.1.2.2.

1.1.1.3. Molecular phenotypes (endotypes) of asthma

It is becoming increasingly clear that asthma is a heterogeneous disease that consists of several cellular and molecular phenotypes, or endotypes. These distinct endotypes reflect the involvement of unique molecular pathways in different asthma trajectories. For instance, because it often co-occurs with allergy, asthma is often described as a type-2 disease dominated by increased levels of eosinophils, Th2 lymphocytes and type-2 innate lymphoid cells (ILC2), IgE and type-2 cytokine responses (IL-13, IL-4, IL-5) [33]. Indeed, IL-13, which is predominantly produced by Th2 and ILC2 cells, has been shown to induce airway remodeling and (in conjunction with IL-4) production of allergy-associated IgE antibodies by B cells [34]. IL-13 also promotes eosinophil migration to the lung and mucus production [34], both of which are key features of asthma.

Molecular pathways associated with asthma often lie upstream of type-2 inflammation and begin at the epithelium. For example, IL-33 is an alarmin produced by epithelial cells and other stromal cells in response to cellular damage or necrosis [35]. This cytokine can then signal to other cell types (including T and

B cells, eosinophils, mast cells, ILC2, and dendritic cells) through its receptor

ST2 (IL1RL1). Moreover, IL-33 is elevated in asthmatic patients and its levels correlate with disease severity [36, 37]. TSLP and IL-25 are also released by epithelial cells in response to damage, and induce an influx of inflammatory cells, primarily Th2 cells and ILC2, from the circulation [38].

Even though in many cases type-2 inflammation is critical for asthma pathogenesis, additional cellular and molecular mechanisms appear to be involved 17

[15, 39, 40]. For example, there is strong evidence that neutrophils and IL-17 are associated with asthma [39] and the TGF- signaling pathway, through its master regulator SMAD3, is thought to regulate both the asthma-protective T regulatory and the asthma-promoting Th17 developmental programs [15, 41].

Because distinct endotypes most likely represent distinct trajectories for developing asthma, understanding which pathways are involved in these endotypes remains an important but unanswered question in the field.

1.1.2 – ASTHMA AS A COMPLEX DISEASE

Asthma is widely considered as a prototypic complex disease, one in which genetic, environmental and developmental components contribute to susceptibility and severity [15].

1.1.2.1 – Asthma genetics

A role of genetics in asthma susceptibility has long been suggested by evidence that “asthma runs in families” [42-46]. More recently, genome-wide association studies (GWAS) have identified single nucleotide polymorphisms

(SNPs) that are associated with asthma risk [10, 11, 47-67] and replicate across studies (Table 1.1). These associations include both novel and previously known candidate genes. Novel associations, such as those with the chromosome 17q21 locus, illustrate the ability of GWAS to point to unexpected pathways of disease.

Of note, the association between 17q21 locus SNPs and childhood asthma is the most significant and most replicated reported so far (Table 1.1) [10, 11, 47, 52-56, 18

58, 59, 65-67]. Among the other most robust asthma-associated GWAS SNPs, several map to candidate genes (e.g. SMAD3, GATA3, IL33), suggesting variants in these genes play a role in disease pathogenesis.

However, in spite of these successes, even the most significant genome- wide associations have only modest effect sizes [odds ratios (OR) typically near

1.2], and only a small proportion of phenotypic variability can be explained by all

GWAS associations combined [13]. Whether this “missing heritability” results from failing to account for rare variants, heterogeneity within study populations, gene-gene and gene-environment interactions, or combinations of these factors is currently unclear.

1.1.2.2 – Asthma and the environment

The prevalence of asthma has increased drastically over the last several decades [68, 69]. Because genetics alone cannot explain such a steep increase in a relatively short time, the environment is likely to play an important role in asthma and its rise. The impact of the environment on asthma pathogenesis is likely complex. Not only can environmental exposures interact with certain genotypes to alter risk for asthma (as shown by the interaction between history of rhinovirus illness and 17q21 alleles [70]), but some environments have been shown to confer protection or risk depending on the timing of the exposure. For example, high endotoxin levels (a marker of bacterial load) during early life are inversely related to asthma risk, but similar exposures later in life are not protective and often have adverse effects for asthmatics [71, 72]. 19

Among the exposures known to modify asthma risk, growing up in a traditional farming environment [72] has the most replicated asthma-protective effects [73]. Within farm living, the exposures most associated with asthma protection include contact with cows and hay and consumption of raw milk [21,

72]. In contrast, early life infections of the lower airways with rhinovirus (RV) or respiratory syncytial virus (RSV) are often associated with recurrent wheezing

(section 1.1.1) and strongly increase the risk for asthma by age 5-7 years [31, 32].

Other early life exposures associated with increased asthma risk include tobacco smoke (prenatal as well as throughout childhood) and air pollution [23, 31, 32].

Importantly, viral infections, tobacco smoke and air pollution are also associated with asthma exacerbations in older children and adults [74].

1.1.2.3 – Development and natural history

As discussed in section 1.1.1, many epidemiological studies have shown that the first manifestations of asthma typically occur during childhood even though asthma may not be diagnosed until adulthood [15]. Indeed, decreased lung function [16] and subtle alterations of both innate and adaptive immunity in the first years of life [17, 18] accompany and often precede a diagnosis of asthma, suggesting that respiratory and immune alterations at critical developmental windows converge to alter the asthma trajectory. These considerations provide a strong rationale for studying the contribution of epigenetic mechanisms to asthma susceptibility, because these mechanisms control the timed unfolding of developmental programs and the responses to environmental stimuli. 20

Table 1.1. Replicated asthma associations from GWAS Chromosomal Reported Gene(s) Reference No. Region 1p13.3 VAV3 [51, 55] 1q21.3 IL6R [56-58] 1q24.2 CD247 [47, 58] 1q25.1 TNFSF18, TNFSF4 [47, 58] 2p25.1 LINC00299 [47, 58] 2q12.1 IL18R1, IL1RL2, IL1RL1 [47-49, 52, 53, 55, 56, 58] 2q37.3 D2HGDH [47, 58] 3q28 LPP [47, 58] 4p14 TLR1 [47, 55, 58] 5q22.1 SLC25A46, TSLP [47, 50, 52, 55, 56, 58, 59] 5q31.1 RAD50, IL13, SLC22A5 [10, 47, 54, 56, 58-60] 5q31.3 NDFIP1 [47, 53, 56, 58] [10, 47, 49, 50, 55, 56, 58, 6p21.32 HLA 60-62] 6p21.32 BTNL2 [49, 50, 56, 59] 7q22.3 CDHR3 [47, 54] 8q21.13 ZBTB10 [47, 55, 56, 58] 8q23.3 CSMD3 [59, 63] 9p24.1 RANBP6, IL33 [10, 47, 52, 54-56, 58] 10p14 GATA3 [47, 56, 58] 10p15.1 IL2RA [55, 58, 64] 11q13.5 LRRC32 [47, 56-58] 11q23.2 C11orf71 [52, 56] 12q13.2 IKZF4 [50, 58] 12q13.3 STAT6 [47, 56, 58] 14q24.1 RAD51B [47, 58] 15q22.2 RORA [10, 47, 56, 58] 15q22.33 SMAD3 [10, 47, 55, 56, 58] 16p13.13 CLEC16A [47, 55, 56, 58, 59] ZPBP2, GSDMB, [10, 11, 47, 52-56, 58, 59, 17q21.1 ORMDL3, GSDMA, IKZF3 65-67] 17q21.33 ZNF652 [56, 58] The table shows replicated asthma-associated genes from the GWAS catalog (https://www.ebi.ac.uk/gwas/). Included in the table are associations with the following traits: adult asthma, allergic disease (asthma, hay fever or eczema), asthma, asthma (bronchodilator response), asthma (childhood onset), asthma and hay fever, bronchial hyperresponsiveness in asthma, bronchodilator response in asthma (inhaled corticosteroid treatment interaction.

1.2 – EPIGENETICS

Epigenetic mechanisms (a term coined by C.H. Waddington in 1940) are defined as mitotically heritable changes in gene activity without changes in the underlying DNA sequence [75]. While epigenetic mechanisms include modifications of both DNA and chromatin structure—namely, DNA methylation and histone modifications—DNA methylation is the best characterized epigenetic mark in human populations because of technical and biological reasons.

Methylation profiles are stable when DNA is properly stored and studying these profiles does not require isolation of chromatin, a feature that makes it feasible to use previously isolated DNA. Additionally, the biological relationship between

DNA methylation and regulation of gene expression is robust and often decipherable.

1.2.1 – DNA METHYLATION

DNA methylation is an ancient adaptation used to distinguish an organism’s own DNA from that of invaders, such as viruses [76]. In eukaryotes,

DNA methylation has further evolved into an important epigenetic mechanism for controlling endogenous gene activity [77]. DNA methylation generally occurs at

CpG motifs—5’-CG dinucleotides methylated on cytosine, also known as 5- methylcytosine (5mC)—which are typically found in dense clusters known as

CpG islands (CGIs). CGIs are usually defined as any DNA sequence of at least

500 base pairs (excluding most repetitive Alu-elements) with GC content >50% and CpG observed/expected ratio >0.6 [78-80]. 22

In the demethylation process, 5mC is oxidized by the TET proteins to form 5-hydroxymethylcytosine (5hmC) [81]. The functional role of this epigenetic mark is incompletely understood. However, recent studies suggest that

5hmC is involved in resetting the epigenetic landscape to totipotency after an oocyte becomes fertilized by a sperm. Indeed, 5mC is converted to 5hmC in primordial germ cells, a process that is driven by high levels of TET1 and TET2.

Global conversion to 5hmC initiates asynchronously at embryonic day 9.5-10.5 and accounts for the process of epigenetic imprinting. However, rare regulatory elements escape systematic DNA demethylation. This process may provide a long-sought mechanism for trans-generational epigenetic inheritance [82].

Interestingly, in murine embryonic stem cells, 5hmC is enriched at transcription start sites, promoters and exons as well as at cis-acting enhancers [83, 84], suggesting it might have its own regulatory function.

While DNA methylation is known to regulate transcription of multiple genes, the functional outcome of this process appears to be locus- and location- specific [85]. Indeed, promoter methylation is typically associated with gene silencing, whereas an unmethylated promoter is necessary but not sufficient for gene expression. In contrast, methylation within gene bodies is often associated with gene expression [85, 86]. 23

1.2.2 – DNA METHYLATION PROFILING TECHNIQUES

DNA methylation profiling has risen to the forefront of epigenetic analytical techniques largely because it provides an effective tool for population- based studies. The same is not true of the methods required to study post- translational histone modifications, which involve large numbers of cells and cumbersome chromatin isolation and immunoprecipitation steps [87].

Current techniques to study DNA methylation rely on bisulfite conversion of DNA, a process in which unmethylated but not methylated cytosines are converted to uracil (and subsequently to thymine, after amplification), thereby allowing detection of methylated cytosine residues within the sequence. Bisulfite sequencing remains the golden standard for targeted studies of DNA methylation.

While the decreasing cost and higher quality of next-generation sequencing are improving research in many areas of genomics, whole-genome bisulfite sequencing remains problematic because of the cost and reduced complexity of the DNA sequence that results from converting the majority of cytosines to thymines. This makes it difficult to map 100-300 bp reads to unique regions of the genome. Therefore, other ways to study DNA methylation, such as targeted

(gene- and/or pathway-specific) bisulfite sequencing or genome-wide microarrays, are still extremely popular.

Until recently, the most widely used platform for genome-wide DNA methylation profiling was the Illumina HumanMethylation450 BeadChip array.

This platform interrogated approximately 485,000 individual CpG sites at single

CpG resolution, and its output could be intuitively understood as the percentage 24 of DNA methylation at any given CpG site on the array [80]. This array was designed to cover 99% of RefSeq genes (with a global average of 17.2 probes per gene region), 96% of CGIs, 92% of CpG shores (flanking sequences located 2kb upstream or downstream of a CGI), and 86% of CpG shelves (flanking sequences located 2kb upstream or downstream of a shore region) [80]. The Illumina

HumanMethylation450 BeadChip platform detected differences in DNA methylation by using the Infinium assay on bisulfite-converted DNA and incorporated both Infinium I and Infinium II probe types. Infinium I probes are designed in pairs—one against the methylated locus and the other against the unmethylated locus—whereas Infinium II probes are designed to bind both the methylated and the unmethylated locus, and methylation state is detected upon a single base extension. A distinct signal is given from labeled nucleotides [80].

One specific limitation of this approach is its inability to distinguish between 5mC and 5hmC. However, 5hmC occurs most commonly in the brain and in embryonic stem cells, and as such, its analysis is a concern primarily for studies focused on these tissues.

The microarray with the highest coverage to date, the 2.1M Human

Promoter Deluxe microarray (Roche-NimbleGen), profiled genome-wide DNA methylation using the Methylated CpG Island Recovery Assay (MIRA) technique.

This process utilized methyl binding protein-dependent capture [88] of methylated DNA (enriched fraction). For each sample, enriched and input DNA underwent one round of whole genome amplification and were then labeled with

Cy3 and Cy5 dye and co-hybridized to microarrays. The arrays included 25

2,137,192 experimental probes (from hg18 and covering ~10 kb of each annotated human gene surrounding the transcription start site, CpG islands and regulatory regions) as well as 6,726 positive, 4,257 negative, 963 non-CG and 38,763 random controls. In contrast to the Illumina HumanMethylation450 BeadChip that typically identifies individual differentially methylated CpGs (DMCs), this platform used a Probe Sliding Window-ANOVA [Roche-NimbleGen (13)] that identified differentially methylated regions of the genome (DMRs, ≥300 bp in length) that typically included multiple CpGs. This platform is no longer available.

Both the Roche-NimbleGen and the Illumina HumanMethylation450

BeadChip platforms were used in our experiments and will be further discussed in

Chapters 2 and 3.

1.3 – ASTHMA EPIGENETICS

The field of asthma and allergy epigenetics is still in its infancy. Figure

1.1 shows the number of papers published on these topics in the past decade.

Strikingly, in many years, reviews outnumbered primary publications, a paradox that illustrates the urgent interest of the scientific community in these types of studies.

1.3.1 – EPIGENOME-WIDE STUDIES

To date, most epigenome-wide studies in the asthma field have focused on epigenetic variation in the context of concurrent disease. While this design does not allow inferring whether asthma-related epigenetic signatures are a cause or an effect of the disease, linking epigenome variation to concurrent asthma in well- characterized populations may be informative. Most of these studies utilized peripheral blood cells to investigate the epigenetic contribution of the immune system to asthma pathogenesis, even though much work is now focusing on the airways. Other studies investigated specific asthma-related mechanisms using in vitro/ex vivo models. Finally, longitudinal epigenetic studies beginning in early life (still only a few) are seeking to delineate epigenetic trajectories to asthma during childhood.

Below I will comment on a few well-designed studies that emphasize the potential role of epigenetic mechanisms in asthma pathogenesis. Table 1.2 summarizes the design of each study.

Figure 1.1. Literature on the epigenetics of human asthma and allergic

diseases, 2007-2017. The y-axis shows the number of journal articles

published each year. Light grey bars indicate review articles, whereas dark

grey bars indicate primary research articles. The literature search was performed by entering into PubMed the following terms: (DNA methylation

OR epigenetics) AND (asthma OR allergy). Primary articles were excluded if

they did not include data from human tissues or cells. 28

1.3.1.1 – Asthma-associated epigenetic mechanisms in blood immune cells

Most epigenome-wide asthma studies have examined DNA methylation profiles in unfractionated blood cells. Because these studies can be influenced by differences in cellular composition, methods have been recently developed to estimate and adjust for these differences [89]. Overall, these studies provide evidence for associations between DNA methylation profiles and distinct asthma- related phenotypes.

Asthma-related DNA methylation in the inner city One of the first large-scale epigenome-wide studies of asthma examined a cross-sectional cohort of inner-city children to assess whether DNA methylation patterns in peripheral blood immune cells differed in 6-12 years old children with persistent atopic asthma and healthy controls (total n=194) [90]. Of the 81 asthma-associated DMRs (defined as 300 bp windows containing 2 probes) identified in this study, some mapped to immune genes (IL13, RUNX3, and TIGIT) and were hypomethylated in asthmatics. Fifty-five of these DMRs remained significant after adjusting for differences in cell proportions. Eleven DMRs were associated with serum IgE levels in asthmatics, and 16 were associated with percent predicted forced expiratory volume in 1 second (FEV1). Interestingly, asthma- and IgE-associated

DMRs were enriched for correlations with gene expression levels, but this was not the case for FEV1-associated DMRs. Differential methylation in RUNX3, IL4 and CAT was validated in an independent cohort of asthmatic children living in the inner city [90]. These findings indicate that epigenome-wide dysregulation in 29 immune cells is associated with asthma and might segregate with distinct endotypes associated with IgE or FEV1.

IgE-associated DNA methylation in Hispanic children While cohort studies mostly include Caucasian participants, other populations have higher risk for asthma [22, 31]. One recent study investigated whether genome-wide DNA methylation levels in white blood cells were associated with total IgE levels in two populations of Hispanic children (total n=879) [91]. By meta-analysis, 1,326 and 2,992 IgE-associated DMCs were identified [P< 2x10-7 and false discovery rate (FDR) < 0.01, respectively]. Interestingly, 189 of the top 200 IgE-associated

DMCs had been previously reported to be associated with IgE levels in

Caucasians (P<0.01) [92], suggesting that the epigenetic mechanisms involved in

IgE responses may be shared between these two populations. However, after adjusting for cell proportions, the number of significant IgE-associated DMCs was drastically reduced (n=4), and only the top 2 DMCs overlapped between the two analyses. Enrichment analysis for KEGG pathways performed on the genes closest to the top 1000 DMCs (adjusted for cell types) identified as a top hit asthma, a trait closely associated with total IgE. Next, the authors investigated which of the top 20 IgE-associated DMCs (adjusted for cell proportions) were associated with gene expression in at least one original cohort. Methylation at cg25087851 was found to be associated with expression of PTGDR2 (P=5.16x10-

17), a receptor known to mediate allergic inflammation [93]. Together, these results suggest that IgE-associated epigenetic mechanisms are largely enriched in 30 pathways associated with allergic asthma and immune responses and are similar between Hispanics and Caucasians.

Asthma- and wheeze- associated DNA methylation A more recent study examined asthma- and wheeze-associated DNA methylation [94]. In a cohort of nearly 1000 individuals, after adjusting for cell proportions (FDR < 0.05), the authors identified 302 asthma-associated DMCs and 445 wheeze-associated

DMCs at age 7.5 years. Two hundred forty-eight CpG sites overlapped between the two analyses. All associations were attenuated after adjusting for eosinophil counts, indicating that the differences in methylation were largely driven by differences in this variable. Not surprisingly, the genes annotated to the 302 asthma-associated DMCs were enriched for pathways related to movement of cellular/subcellular components, locomotion, IL-4 production and eosinophil migration. Interestingly, only 2 DMCs were associated with current asthma in children at 16.5 years of age. Thirty-two significant DMCs were associated with wheeze (ever) at 16.5 years of age after adjusting for cell proportions. In longitudinal models, the authors identified 4 CpG sites whose methylation at 7.5 years was associated with asthma at 16.5 years, and 28 CpG sites that were differentially methylated at 16.5 years and were associated with asthma at 7.5 years (FDR < 0.05); however, all associations were attenuated after adjusting for cell proportions. Using Mendelian randomization, the authors showed that there was a causal effect of asthma on DNA methylation at several DMCs at 7.5 years, but these associations did not persist after adjusting for multiple testing. In 31 contrast, no causal effect of asthma on DNA methylation at either of the 2 DMCs at 16.5 years was detected. The authors concluded that the majority of observed

DNA methylation differences were driven by differences in eosinophil counts in asthmatics at age 7.5 years, a time at which eosinophilic asthma is a dominating endotype. In contrast, eosinophilic asthma is more rare in adult-onset asthma

(16.5 years), and this may explain the lack of association. Again, these findings suggest that asthma-associated epigenetic mechanisms differ between endotypes.

Large scale meta-analysis for asthma-associated DNA methylation during childhood A recent multi-phase epigenome-wide study sought to highlight the acquisition of asthma-associated epigenetic trajectories over time. In the discovery phase, 35 asthma-associated DMCs were identified, 11 at age 4 years

(FDR < 0.10) and 26 at age 8 years (genome-wide significance P < 1.14x10-7)

[95]. A total of 14 sites were validated in a meta-analysis of 3,196 individuals. All

14 DMCs were replicated in a second population using whole blood DNA from

Canadian families (n=167) but were not replicated in cord blood cells (n=1,316).

Because adjusting for eosinophils attenuated the associations in the discovery phase, the authors investigated whether the 14 DMCs were replicated in previously published data from eosinophils [96]; indeed, 11 of the 14 asthma- associated DMCs were hypomethylated in eosinophils compared to other blood cell types (Fisher’s exact test, P=1.92x10-24). Again, in the Canadian cohort all 14

CpG sites in eosinophils were significantly associated with asthma and had on average 18% lower methylation in asthmatics. All 14 DMCs were significantly 32 associated with gene expression in whole blood, and three distinct expression patterns were identified by hierarchical clustering. The first cluster was inversely correlated with methylation levels at 7 DMCs and included genes important for eosinophils (SLC29A1, SIGLEC8, IL5RA, and ADORA3). The second cluster was strongly associated with 7 DMCs, included genes such as CCR7, SHMP7, and

LEF1, and was defined as a naïve CD4 and CD8 T cell gene signature. The third cluster contained genes such as TBX21, EOMES, CCL5, and GZMB, and was negatively correlated with 4 asthma-associated DMCs; this cluster was interpreted as an effector/memory CD8 T cell and natural killer cell gene signature. The lack of replication in cord blood suggests that certain epigenetic mechanisms are acquired over time, possibly as a consequence of epigenetic responses to early life viral infections.

1.3.1.2 – Asthma-associated epigenetic mechanisms in the airways

Epigenetic mechanisms are thought to be cell- and tissue-specific. Because asthma susceptibility has both airway and immune components, it is reasonable to surmise that studying both these components might provide distinct but complementary information about the epigenetic trajectory to asthma. An additional benefit of studying airway epithelial cells is their relative homogeneity in comparison to the high heterogeneity of peripheral blood.

Asthma-related DNA methylation in nasal epithelium of children from the inner In an inner-city cohort, the authors identified a total of 186 allergic 33 asthma-associated differentially methylated genes [119 DMRs representing 118 unique genes, and 118 single CpGs representing 107 unique genes] in nasal epithelial cells [97]. Several of these genes (e.g. ALOX15, CAPN14, HNMT, and

POSTN) have previously been found to be associated with asthma and allergies.

Interestingly, only 3 DMRs identified in peripheral blood [90] were replicated in nasal epithelium [97], supporting the notion that peripheral blood cells may better model immune events while nasal epithelia may better represent the airways component of the epigenetic trajectory to asthma. To further investigate the influences on nasal epithelial methylation profiles, the authors asked whether environmental exposures were associated with methylation at asthma-associated

DMRs, and showed that 48 of these DMRs were associated with environmental tobacco smoke. One of these (SFRP2) had been previously reported to be associated with tobacco smoking. In addition, 47 differentially methylated genes were also differentially expressed between asthmatics and controls, with inverse relationships between expression and methylation. When these asthma-associated genes were further investigated, an upstream regulator analysis revealed a significant enrichment of genes downstream of known regulators of transcription in asthmatics, including the cytokines IL-13, IL-4, IL-6, IFN, the growth factor

TGF- and the transcriptional regulator CIITA. Protein interaction network analysis revealed that the largest hub was RIPK2, which points to both adaptive and innate immune pathways important for attenuation of allergic airway inflammation [98]. Together, these findings suggest that genes whose expression is altered in the airways of asthmatics might be regulated by epigenetic processes. 34

Ex vivo studies of IL-13-mediated responses in airway epithelial cells In vitro and ex vivo studies of epigenetic responses to asthma-associated exposures have been performed to understand the mechanisms underlying disease pathogenesis

[13]. One recent paper illustrated this emerging trend by examining whether IL-

13-driven DNA methylation responses in airway epithelial cells are recapitulated in the airways of asthmatic patients [99]. Treatment of airway epithelial cells with

IL-13 yielded 6,522 DMCs, 1,590 of which were located near genes that were also differentially expressed in response to IL-13. Interestingly, IL-13-responsive

DMC-gene pairs were significantly enriched for asthma genes. These findings were confirmed in vivo by showing that 2,000 of the original 6,522 IL-13- responsive DMCs replicated in asthmatic patients. By network analysis, these

2,000 DMCs clustered into two co-methylation modules, one associated with asthma severity and one associated with blood eosinophilia. These modules included networks centered around ERK1/2 (module 1) and IFNγ and NF-B

(module 2), a result that highlights pathways which may coordinate IL-13- mediated DNA methylation changes promoting asthma pathophysiology. The results from this study yielded promising findings by showing that asthma- associated exposures may modify disease pathogenesis by coordinating epigenetic modifications.

DNA methylation in the airway epithelium and asthma endotypes In a separate study, the same authors asked how epigenetic variation in airway epithelial cells 35 impacts gene expression and asthma risk [100]. The authors identified over

40,000 DMCs between asthmatic and non-asthmatic airway epithelial cells (FDR

< 0.05). Among these, loci with larger difference in methylation were more strongly correlated with expression of the nearest gene. Asthma-associated DMCs were enriched for methylation quantitative trait loci (meQTLs); in contrast, expression (e)QTLs were not enriched among differentially expressed genes.

Moreover, SNPs that were meQTLs or eQTLs in airway epithelial cells were enriched for asthma SNPs from published GWAS from the EVE and GABRIEL consortia. The authors next investigated whether epigenetic variation was associated with distinct endotypes of asthma. To this end, they focused on loci with >5% methylation differences between asthmatics and non-asthmatics

(n=3,767 DMCs) and used network analysis to examine correlations among asthma-associated loci. DMCs were grouped into four co-methylation modules, and module eigengenes were tested for significant relationships with asthma- relevant phenotypes. Modules 1 and 4 were associated with asthma severity, module 2 was associated with airway eosinophilia (measured in bronchoalveolar lavage), and module 3 was associated with fractional exhaled nitric oxide. Module

3 was also enriched for IL-13-responsive DMCs from their previously published article [99]. Pathway analysis of DMC-gene pairs revealed that interaction networks associated with modules 1 and 4 were centered on the same hub genes involved in remodeling, cell growth and inflammatory pathways (e.g., ERK1/2,

NF-B, and Ras/Raf kinase), and yet only 9 genes overlapped between the modules, suggesting that module 1 and 4 represent different signaling pathways 36 that act through the same hub molecules. Moreover, module 1 was enriched for genes downstream of TNF (P=4.81×10-4) and module 4 was enriched for genes downstream of TGFB1 (P=7.74×10-9), further supporting this notion. The single interaction network created from module 2 DMC-gene pairs revolved around genes involved in eosinophil processes and migration (e.g. VEGF, SELE,

SMAD2/TGFB1), while the network from module 3 was centered on NOS2 and other components of the nitric oxide response. Together these findings supported the conclusion that asthma-associated methylation changes cluster in distinct molecular endotypes of asthma.

1.3.1.3 – Conclusions

Our understanding of the role of epigenetic mechanisms in asthma pathogenesis is only beginning [101-103]. The studies performed to date suggest that epigenome-wide variation is linked to disease pathogenesis. These studies, most of which focused on blood immune cells and to a more limited extent airway epithelial cells, indicate that distinct epigenetic trajectories to asthma likely exist and depend on the cell and/or tissue under investigation. Moreover, and perhaps most importantly, these studies emphasize that distinct epigenetic trajectories are likely associated with distinct asthma endotypes.

To further our understanding, we must overcome limitations in the assessment of epigenetic modifications and, as much as possible, the source of tissue. As experimental platforms continue to improve, and novel methods are proposed, some of these limitations may diminish. For instance, the Illumina 37

HumanMethylation450 arrays have recently been replaced by the

MethylationEPIC arrays, which drastically expand the coverage of their predecessors by incorporating CpG sites within enhancer regions [87]. This will in turn help explore the regulatory potential of differentially methylated regions.

Perhaps the most severe hurdle lies in the fact that most of the work so far has relied on pre-existing cross-sectional cohorts in which epigenetic relationships could only be explored in the context of concurrent asthma – a design that prevents determining whether epigenetic alterations are a cause or a consequence of the disease. Longitudinal studies in well characterized birth cohorts will be needed to illuminate the epigenetic trajectory to asthma and support plausible inferences about the role of epigenetic modifications in the inception of this disease.

One final note. This introduction is meant to provide a background for the experimental data on asthma epigenetics presented in Chapter 2 and 3. The work discussed in Chapter 4, on the other hand, decisively departs from this theme in both conceptual and technical terms. Therefore, for the sake of clarity and continuity, the introduction to Chapter 4 is provided at the beginning of the chapter itself.

Table 1.2. Epigenome-wide studies of asthma

Ref Method Sample Tissue Age Primary Outcome(s) Replication or Validation No. (platform) Size (n)

atopic asthma [90] 450K 194 PBMC 6-12 years YES; validation via gene expression (6-12 years) white 11.6 years YES; validation via meta-analysis; [91] 450K 879 blood total IgE (mean) replication in previously published data cells periphe 7.5 and 16.5 asthma or wheeze (7.5 or YES; replication in previously published [94] 450K 1018 ral years 16.5 years) data blood YES; validation via gene expression and whole [95] 450K 1548 4-5 and 8 years asthma (4-5 or 8 years) meta-analysis; replication in distinct cohorts blood and previously published data nasal atopic asthma [97] 450K 72 epitheli 10-12 years YES; validation via gene expression (10-12 years) um cultured 58 YES; validation via gene expression; [99] 450K primary 45 years (mean) in vitro IL-13 treatment replication in distinct cohort AEC primary 38.5 years asthma (mean age: 38.5 YES; validation via gene expression; [100] 450K 115 AEC (mean) years) replication in previously published data

AEC: airway epithelial cells, PBMC: peripheral blood mononuclear cells, 450K: Illumina Human Methylation450 39

1.4 – DISSERTATION OUTLINE

While recent work suggests that epigenetic mechanisms may play a role in asthma and its development, some fundamental questions remain unanswered. In particular, it is unknown when the trajectory to asthma begins, and whether epigenetic mechanisms are associated with disease inception. An attempt at answering these questions is the primary focus of most of this dissertation.

The work in Chapter 2 relies on the Infant Immune Study, an unselected birth cohort monitored for asthma and allergies throughout the first decade of life.

The study begins with an epigenome-wide search for DNA methylation signatures in neonatal (cord blood) immune cells that differentiate children who will and will not develop asthma by age 9 years. The initial identification of the promoter of

SMAD3 (the master regulator of the TGF- pathway) as an asthma-associated candidate region of differential methylation at birth is followed by targeted profiling of SMAD3 promoter methylation and replication of the results (stratified by maternal asthma) in two independent birth cohorts. This chapter consists entirely of a first-author paper entitled “Epigenome-wide Analysis Links SMAD3

Methylation at Birth to Asthma in Children of Asthmatic Mothers”, which was recently published in The Journal of Allergy and Clinical Immunology (J Allergy

Clin Immunol 140 (2): 534-542, 2017, PMID:28011059).

Chapter 3 builds on the findings of Chapter 2 and again relies on the Infant

Immune study to explore the possibility that the trajectory to asthma begins even before birth, that is, in utero. Because our group has recently shown that the maternal prenatal cytokine profile predicts asthma in children of non-asthmatic 40 mothers [104], this chapter presents a pilot study that investigated whether this relationship involves the neonatal epigenome and differs in children of asthmatic and non-asthmatic mothers. DNA methylation at over 500 CpG sites was significantly associated with prenatal maternal cytokine profiles in children of non-asthmatic mothers but not in children born to asthmatic mothers. Notably, the neonatal CpG sites associated with maternal cytokine profiles clustered predominantly in pathways regulated by TGFB1. An abridged version of these results was published in The Journal of Allergy and Clinical Immunology (J

Allergy Clin Immunol 141(6):1992-1996, 2018, PMID: 29709672) as a first- author Rostrum article entitled “Of Pleiotropy and Trajectories: Does the TGFB

Pathway Link Childhood Asthma and COPD?”.

The work presented in Chapter 4 represents a novel direction of my research and is somewhat distinct from the rest of the dissertation. This chapter builds on work performed by other members of our group [105, 106] aimed at characterizing the asthma-protective effects of farm exposure by comparing and contrasting two U.S. farming populations, the Amish and the Hutterites. Chapter 4 discusses work that is still ongoing and specifically utilizes lung gene expression data from mice exposed to Amish or Hutterite environmental dust to identify gene networks associated with asthma-protective phenotypes. I also investigate whether genes characteristic of T cells (which we have found to be recruited in large numbers to the lungs of mice exposed to Amish farm dust) are associated with asthma protection. 41

Finally, Chapter 5 seeks to tie together the studies presented in Chapters 2-

4 and proposes future research directions that may advance the distinct but complementary facets of these projects

CHAPTER 2: EPIGENOME-WIDE ANALYSIS LINKS SMAD3

METHYLATION AT BIRTH TO ASTHMA IN CHILDREN OF

ASTHMATIC MOTHERS

2.1 – INTRODUCTION

The timing and mechanisms of asthma inception remain imprecisely defined. Although epigenetic mechanisms likely contribute to asthma pathogenesis, little is known about their role in asthma inception. Therefore, in this study we assessed whether the trajectory to asthma beings already at birth and epigenetic mechanisms contribute to asthma inception. The data presented herein are published as a multi-author manuscript in The Journal of Allergy and Clinical

Immunology [J Allergy Clin Immunol 140(2): 534-542, 2017. PMID: 28011059].

All supplementary methods, figures, and tables for this manuscript can be found in the Appendix (sections A.1, A.2 and A.3, respectively).

Asthma is the most prevalent chronic disease of childhood [15].

Epidemiological evidence suggests that the disease often begins during the preschool years even when chronic symptoms appear much later in life [5]. However, firm criteria to pinpoint how early a child’s trajectory to asthma truly begins are currently lacking. The mechanisms underlying asthma inception also remain largely unknown. Subtle modifications of both innate and adaptive immune responses accompany and often precede the diagnosis of childhood asthma [17,

18], consistent with the notion that immune and respiratory alterations at an early 43 window of susceptibility converge to place the child on a path to the disease.

GWAS have identified multiple genetic variants that influence asthma susceptibility [107] but have accounted for only a modest proportion of the total phenotypic variance, providing a compelling rationale for seeking additional risk factors for asthma. In this context, epigenetic mechanisms are especially worth investigating because environmental and developmental influences are essential for asthma pathogenesis [15], and epigenetic processes ensure the timed unfolding of developmental programs and plastic responses to environmental cues, including those delivered in utero by the maternal milieu [108].

Little is known about the role of epigenetic mechanisms in childhood asthma [102]. A recent epigenome-wide study compared DNA methylation patterns in peripheral blood mononuclear cells from 6-12 year old inner-city children with persistent atopic asthma and healthy controls, and found that several immune genes involved in T cell maturation, Th2 immunity and oxidative stress were hypomethylated in asthmatic children [109]. While these results are novel, their significance remains unclear because it is difficult to determine whether epigenetic alterations concurrent with asthma are a cause or a consequence of the disease. Moreover, a cross-sectional study cannot provide insights into the timing and mechanisms of asthma inception. A recent candidate gene study pointed to an association between IL2 promoter methylation at birth and asthma exacerbations during childhood [110], but relevant pathways were not further interrogated.

In an attempt to define when the trajectory to asthma begins and which pathways are involved, we performed an epigenome-wide search for DNA 44 methylation signatures associated with childhood asthma in cord blood mononuclear cells (CBMCs) from children enrolled in the Infant Immune Study

(IIS). In this unselected birth cohort, the development of asthma and immune responses was monitored at multiple times from birth to age 9 [18, 111]. We reasoned that the detection at birth of DMRs associated with asthma during childhood would both support a perinatal origin of the disease and highlight epigenetic mechanisms potentially contributing to asthma inception. We show herein that DNA methylation signatures associated with asthma during childhood were indeed present in neonatal blood immune cells, and clustered in immunoregulatory and pro-inflammatory pathways. Moreover, hypermethylation of the SMAD3 promoter was selectively detected in asthmatic children of asthmatic mothers and was associated with risk of childhood asthma in the IIS population and in two comparable birth cohorts.

2.2 – METHODS

Study design and participants The IIS unselected birth cohort includes 482 children and was designed to assess patterns of immune maturation in early life and their impact on asthma risk [18, 111]. At enrollment, parents completed a respiratory health history questionnaire, a cord blood sample was obtained, and their child’s health was followed prospectively. Childhood asthma was defined as physician-diagnosed, with symptoms or medication use for asthma in the past year reported at least once on the age 2, 3, 5 or 9-year questionnaires.

Figure 2.1 shows an overview of our study design. We performed a nested case-control, epigenome-wide study of DNA methylation in CBMC (2-8x106 cells per sample) from a discovery population of 36 children from the IIS. This population was randomly selected among available samples but balanced for asthma status in the child (18 non-asthmatic, 18 asthmatic by age 9) and an effort was made to balance cases and controls by maternal asthma (Table E1).

Additional targeted analysis of DNA methylation was performed in the 31 samples available from the IIS discovery population and in samples from 29 additional IIS children of comparable characteristics (Table E2). The distribution of cases and controls is provided in Figure 2.1. Availability of adequate samples was the only inclusion criterion besides asthma status of the child and her/his mother. Overall, the study population did not differ from the rest of the IIS population except for a greater proportion of asthma and maternal asthma, as per study design (Table E1).

The results of targeted DNA methylation analysis performed in the IIS 46 were replicated in CBMC from 30 children with asthmatic mothers from the

Manchester Asthma and Allergy Study (MAAS) and 28 children from the

Childhood Origins of ASThma (COAST) study (Figure 2.1). The characteristics of the MAAS and COAST study populations are described in Table E2. The

MAAS unselected birth cohort includes 1085 children monitored for asthma and allergy from birth to age 11 years. An asthma diagnosis required at least one of the following criteria reported on age 5 or 8 questionnaires: 1) physician diagnosis of asthma, 2) the use of asthma medications during the previous 12 months.

Controls required a negative report on both of these criteria and no report of wheezing in the previous 12 months [112]. The COAST birth cohort enrolled 289 neonates at risk for asthma and allergy, i.e., having at least one parent with asthma and/or allergies. Asthma was diagnosed at age 6 years based on the documented presence of one or more of the following characteristics in the previous year: 1) physician diagnosis of asthma, 2) use of physician-prescribed albuterol for coughing or wheezing episodes, 3) use of a daily controller medication, 4) step-up plan including use of albuterol or short-term use of inhaled corticosteroids during illness, and 5) use of prednisone for asthma exacerbations [70]. All three studies were approved by the appropriate Institutional Review Boards. Informed consent was obtained from the parents of all research participants. 47

Figure 2.1. Overview of study design. IIS: Infant Immune Study, MAAS:

Manchester Asthma and Allergy Study, COAST: Child Origins of ASThma study.

NN: Non-asthmatic child with a Non-asthmatic mother, NA: Non-asthmatic child

with an Asthmatic mother, AN: Asthmatic child with a Non-asthmatic mother,

AA: Asthmatic child with an Asthmatic mother. 48

DNA methylation profiling and DMR identification DNA methylation was profiled by MIRA-chip (Roche-NimbleGen), as detailed in the Appendix, section

A.1 and Figure E1. DNA methylation microarray data from this publication were submitted to the NCBI Gene Expression Omnibus database and assigned the identifier GSE85228.

Regions that were differentially methylated between asthmatics and non- asthmatics (DMRs) were identified using a Probe Sliding Window-ANOVA

(Roche-NimbleGen [113]) that detects maximal inter-group differences in signal intensities relative to a user-defined threshold. Probability scores (P-values) are then assigned to each probe on the array using a repeated measure ANOVA model. In this study, DMRs were defined by region length ≥300 bp, magnitude

(between-group mean log2 ratio difference) ≥0.2, significance threshold (adjusted using the Benjamini-Hochberg FDR method to account for multiple testing) =

0.01, using a sliding window size = 750 bp (Table E3). A positive magnitude difference indicated hypermethylation in asthmatics relative to non-asthmatics.

DMRs containing ≤5 CpG sites within a 750 bp window centered on the DMR (a configuration likely to result in sub-optimal capture of methylated DNA [114]) were excluded from further analyses.

Results of this analysis were technically validated by bisulfite sequencing

(Appendix, section A.1 and Table E4). Microarray-derived estimates of DNA methylation intensity strongly correlated with DNA methylation levels measured by bisulfite sequencing over the entire DNA methylation range [Spearman correlation coefficient ()=0.48, P=1.2x10-13] and at intermediate DNA 49 methylation levels (8-92%, P=0.006; Figure E2). SMAD3 methylation estimates from bisulfite sequencing and microarrays were also highly correlated

(=0.46, P=0.009).

Functional DMR annotation is fully described in the Appendix, section A.1. In brief, DMRs were annotated to genes based on the closest transcription start site.

For pathway analysis, multiple genes were allowed to be associated with a single

DMR (RefSeq genes +/- 5 kb).

DNA methylation analysis in the COAST cohort was performed using the

Infinium HumanMethylation450 BeadChip array (Illumina) (Appendix, section

A.1).

Network analysis A molecular interaction network was constructed by uploading the list of DMR-containing genes into the Ingenuity Pathway Analysis

(IPA) software (Qiagen) and using all available interaction data in the Ingenuity

Systems Knowledge Base. Genes with no interactions were removed from the analysis to maximize the signal-to-noise ratio. The gene interaction network was then interrogated to detect enrichment for biological functions, canonical pathways and upstream regulators (defined as an enrichment for known targets of a given gene in a given gene list) using Ingenuity Pathway Analysis tools.

Cytokine measurements IL-1 concentrations were measured by ELISA

(Quantikine, R&D Systems) in the supernatants of LPS-stimulated IIS CBMC

(n=57) also tested for SMAD3 promoter methylation.

Statistical analyses DMRs were detected using a Probe Sliding Window-

ANOVA which uses a repeated measure ANOVA model for the probes in each sliding window [113]. Fisher’s Exact Test and one sample test of proportions were used to compare proportions for categorical variables, and Student’s t-test or

Wilcoxon two-sample test were used to compare mean levels for continuous variables. Variables with skewed distributions were log-transformed prior to assessment by t-test. Two-sided P-values less than 0.05 were considered significant. Spearman correlation was used to test for association between median

DNA methylation intensity (microarray) and percentage DNA methylation

(bisulfite sequencing). When analyzing the genomic locations of DMRs and the co-localization of DMRs and DNase I hypersensitive sites, permutations were performed by randomly selecting 589 independent probes (the number of DMRs) and calculating the2 statistic for each sampling. We recorded the number of times (out of 50,000) that the permuted2 statistic was larger than the observed2 statistic and divided by the number of permutations to get the empirical P-value.

Linear regression was used to test for an interaction between child asthma and maternal asthma on SMAD3 methylation. Pearson’s 2 test was used to identify significant associations between child and parental characteristics at birth and asthma during childhood. Meta-analysis of the association between SMAD3 CpG7 51 methylation and childhood asthma risk in neonates born to asthmatic mothers was performed using estimates from each study cohort to compute the combined estimate of risk.

2.3 – RESULTS

2.3.1 – Cord blood cells from IIS children harbor asthma-associated DMRs

Our multi-step study of the contribution of epigenetic mechanisms to the development of asthma in early life included: 1) a discovery phase in which an epigenome-wide approach was used to identify candidate regions that were differentially methylated in cord blood samples from 36 IIS children (18 non- asthmatic, 18 asthmatic by age 9); 2) a targeted analysis phase in which DNA methylation of the most compelling candidate region was measured in CBMC from 29 IIS neonates who did, and 31 IIS neonates who did not develop asthma by age 9; and 3) a final phase in which the results of the targeted analysis performed in IIS were replicated in two comparable birth cohorts, MAAS and

COAST (Figure 2.1).

Relying on the longitudinal design of the IIS birth cohort, we initially searched for neonatal epigenetic signatures of childhood asthma by profiling

DNA methylation in CBMC from a discovery cohort of 36 children (Figure 2.1,

Figure E1 and Table E1). Five hundred and eighty-nine independent regions were differentially methylated in asthmatic and non-asthmatic children (Table E3).

Among these, 199 were hypermethylated and 390 were hypomethylated at birth in children who became asthmatic.

Asthma-associated DMRs were distributed across all chromosomes but were non-randomly distributed with respect to genome location (2= 19.49, df =

7, P=0.007), with an overrepresentation at intergenic regions (standardized residual: 2.91) and an underrepresentation at transcription start sites (standardized 53 residual: -2.23, with absolute values > 1.96 significantly contributing to the overall 2 test statistics) (Figure E3). Of the DMRs with the most significant methylation differences, some mapped to biologically plausible genes such as

ATG9A (involved in autophagy and innate immune responses [115]; -log10 adjusted P-value 5.569), OR6K6 (an olfactory receptor expressed in sputum during asthma exacerbations [116]; -log10 adjusted P-value 4.408), and GALNT2

(an N-acetylgalactosaminyltransferase carrying polymorphisms associated with lung function [117]; -log10 adjusted P-value 4.895). Moreover, several asthma- associated DMRs mapped to loci (+/- 5kb from transcription start sites) identified in GWAS for asthma (RORA, SMAD3) and asthma-related traits such as atopic dermatitis (FLG), allergic rhinitis (TMEM232), lung function (ANK1, DLEU7,

SNRPN, RORA, CFDP1) and airflow obstruction (SPATA13)

(http://www.ebi.ac.uk/gwas). Functional annotation using Ingenuity Pathway

Analysis linked DMR-associated genes to biological processes including immune function and immune and lung development (Table E5). We also mapped the locations of neonatal asthma-associated DMRs to DNase I hypersensitive sites, regions of increased chromatin accessibility typically endowed with regulatory activity [118]. Overall, 72 asthma-associated DMRs mapped to DNase I hypersensitive sites (29 monocyte-specific, 15 T cell-specific, 28 in both cell types: Table E6), which represented a significant enrichment (Table E7, permutation P-value < 2x10-5 for both cell types). The co-localization of asthma- associated differential DNA methylation and enhanced chromatin accessibility supports the potential biological significance of asthma-associated DMRs. 54

Figure 2.2. Molecular interaction network of asthma-associated DMRs. The

network was constructed using all available interaction data in the Ingenuity

Systems Knowledge Base. The most highly connected gene (SMAD3, 17 connections) is highlighted in black. Targets of IL1B, the top upstream regulator

of network genes, are highlighted in gray. Types of interactive molecules are

defined in the Legend.

2.3.2 – Asthma-associated DMRs cluster in regulatory and pro-inflammatory gene networks

To elucidate the functional implications of our findings and gain a pathway-based view of asthma-associated methylome alterations at birth,

Ingenuity Pathway Analysis tools were used to construct a molecular interaction network of DMR-associated genes on the basis of prior knowledge of the physical and functional connections between the molecules encoded by those genes. The network included 146 genes organized around several nodes, the most connected of which was the transcription factor SMAD3 (Figure 2.2). When the network was interrogated to detect enrichment for known targets of upstream regulators, IL1B, an innate pro-inflammatory cytokine overexpressed in asthmatics [119-121], emerged as the top regulator of a gene subset that included RORA and RELB, transcription factors essential for innate and adaptive responses, UBD, an innate immunity gene regulated by the NF-B inhibitor A20/TNFAIP3 [122], and the asthma-associated neurotrophin BDNF [119] (overlap P=2.26x10-5, Table 2.1).

Table 2.1. Upstream regulators of genes containing asthma-associated DMRs TARGET UPSTREAM MOLECULE OVERLAP MOLECULES REGULATOR TYPE P* IN DATASET ACAN, BDNF, CFTR, COL10A1, CXCL10, CYP7A1, FGFR3, IL1B cytokine 2.26E-05 GNAS, LDHA, NCOA2, NEUROD1, PFKP, RELB, RORA, SCX, SNCA, UBD ACAN, COL10A1, HSPG2 enzyme 8.60E-05 FGFR3 BDNF, DUSP6, NRXN1, PRKCZ, PTBP1, BDNF growth factor 1.43E-04 RPL35A, SLC17A8, TMEM45A, TTC3 BDNF, CFL1, DNM1, SNCA enzyme 1.56E-04 SNCA, SYN3 CXCL10, CYP7A1, PRMT5 enzyme 3.51E-04 FASN * Enrichment for known targets of a given gene in a given gene list. The Table shows the top hits provided by this analysis.

2.3.3 – The SMAD3 promoter is significantly hypermethylated in asthmatic children of asthmatic mothers

SMAD3 is not only the most connected node in the network of asthma- associated DMRs (Figure 2.2), but is also a well-replicated asthma-associated gene from GWAS [10, 55, 123]. Moreover, SMAD3 acts as a master regulator of

TGF- signaling, thereby controlling the differentiation of Treg and Th17 cells that play critical and opposite roles in asthma [15, 41]. Finally, the SMAD3 DMR lies within the distal promoter of the gene, a location that provides ample opportunity for DNA methylation to influence SMAD3 gene expression (Figure

E4). Therefore, subsequent analyses of the nexus between neonatal DNA methylation and trajectory to childhood asthma specifically targeted SMAD3. We used bisulfite sequencing to precisely quantify methylation at the SMAD3 promoter DMR (321 bp, 8 consecutive CpG sites) in a total of 60 IIS neonates (31 non-asthmatics, 29 asthmatics: Figure 2.2). Mean SMAD3 methylation levels at birth were 41.3% [95% confidence interval (CI) 35.5-47.5] in non-asthmatics and

47.1% (95% CI 40.6-53.1) in asthmatics, a difference that did not reach statistical significance (P=0.2 by Wilcoxon two-sample test: Figure E5). Because these results may have reflected heterogeneity within the study population, we next examined the entire IIS population for associations between childhood asthma and potential risk factors measurable at birth in the child (sex, ethnicity, mode of delivery, total cord IgE, 17q21 rs8076131 genotype) and parents (maternal asthma, maternal allergy, paternal asthma, paternal allergy, maternal smoking during pregnancy). As shown in Table E8, maternal asthma exhibited a distinctive 58

association with childhood asthma in IIS (P=0.003 by  test). When the relation between SMAD3 methylation at birth and childhood asthma was examined separately in children with and without maternal asthma, SMAD3 methylation was found to be significantly increased in asthmatic compared to non-asthmatic children of asthmatic mothers (P=0.005 by Wilcoxon two-sample test). In contrast, asthmatic and non-asthmatic children of non-asthmatic mothers did not differ in their SMAD3 methylation levels [P for interaction (by linear regression)

= 0.001; Figure 2.3A].

Although associations between SMAD3 variants (rs17228058 [123], rs744910 [10], rs17294280 [55]) and asthma have been reported in GWAS, asthma-related SMAD3 methylation differences were unlikely to be influenced by

SMAD3 genotype. Indeed, sequencing identified no polymorphisms within the

SMAD3 DMR (data not shown). Moreover, asthma-associated SMAD3 variants are located at least 80 kb away from the SMAD3 DMR, whereas the relationship between DNA methylation and genetic variation appears to decay rapidly beyond

5 kbs [124].

Figure 2.3. Effects of maternal asthma on the association between neonatal

SMAD3 methylation and childhood asthma in IIS (panel A), MAAS (panel B) and

COAST (panel C). SMAD3 methylation In IIS and MAAS was assessed by bisulfite sequencing and expressed as mean percent DNA methylation across 8 consecutive CpG sites in the SMAD3 DMR. In COAST, percent SMAD3 methylation at cg02486855, the seventh CpG in the SMAD3 DMR, was assessed by the Infinium HumanMethylation450 BeadChip array. N: non-asthmatic, A: asthmatic. P-values by Wilcoxon two-sample test. The results presented in panel

A were unaffected after adjusting for ethnicity.

2.3.4 – The association between SMAD3 promoter hypermethylation and childhood asthma replicates in two independent birth cohorts

The association between neonatal SMAD3 hypermethylation and childhood asthma found in IIS neonates born to asthmatic mothers was assessed for replication in two birth cohorts, MAAS and COAST (Table E2). Bisulfite sequencing of the SMAD3 DMR in 30 MAAS children born to asthmatic mothers revealed significantly higher DNA methylation in the asthmatic group (P=0.049 by Wilcoxon two-sample test; Figure 2.3B). Comparable results were obtained by assessing SMAD3 methylation in 28 COAST neonates using the Illumina Infinium

HumanMethylation450 BeadChip array. Figure 2.3C shows that methylation at cg02486855, the seventh CpG in the SMAD3 DMR and the only SMAD3 DMR

CpG interrogated on the Illumina platform, did not significantly differ between asthmatic and non-asthmatic children of non-asthmatic mothers (P=0.4 by

Wilcoxon two-sample test). In contrast, significant cg02486855 hypermethylation was detected in neonates born to asthmatic mothers who developed asthma during childhood, compared to neonates who did not (P=0.04 by Wilcoxon two-sample test; Figure 2.3C). The data generated by bisulfite sequencing and on the Illumina platform were comparable because when percent methylation values for SMAD3

CpG7 were extracted from the bisulfite sequencing data for IIS and MAAS, again we found significant hypermethylation in asthmatic compared to non-asthmatic children of asthmatic mothers in IIS (P=0.004 by Wilcoxon two-sample test;

Figure E6), and a difference approaching significance in MAAS (P=0.09 by

Wilcoxon two-sample test; Figure E6). 61

Cord blood contains a mixture of cell types with potentially distinct DNA methylation profiles. The availability of COAST DNA methylation data generated on the Illumina450 BeadChip array allowed us to estimate cord blood cell proportions using an algorithm recently developed that integrates Illumina DNA methylation data with information from a cord blood reference panel [125].

Figure E7 shows that cell proportions thus estimated did not significantly differ between asthmatic and non-asthmatic children, regardless of maternal asthma history. Most importantly, the SMAD3 methylation differences detected between asthmatic and non-asthmatic children of asthmatic mothers remained significant

(P=0.04 by Wilcoxon two-sample test) even after adjusting for CBMC composition (Figure E8).

Finally, we asked whether the SMAD3 methylation levels measured at birth in children of asthmatic mothers are associated with risk for childhood asthma in our three cohorts. A meta-analysis revealed that for each 10% increase in SMAD3 CpG7 methylation there is nearly a two-fold increased risk of childhood asthma (meta-analysis OR=1.95, [95%CI: 1.23, 3.10], P=0.005; heterogeneity P=0.5).

2.3.5 – Maternal asthma modifies the relation between neonatal SMAD3 methylation and IL-1 producing capacity

SMAD3 knock-out mice exhibit increased expression of Il1b [126], which is also the top upstream regulator of genes containing asthma-associated DMRs in this study and a critical pro-inflammatory mediator in human asthma [119-121].

Therefore, we relied again on the IIS to explore the relationship between neonatal

LPS-induced IL-1production, SMAD3 promoter methylation and childhood asthma. The distributions of SMAD3 methylation and log IL-1 production in non-asthmatic and asthmatic children without and with a maternal history of asthma were compared by dividing mean percentage SMAD3 methylation and log

IL-1secretion at the median, thereby creating four groups (low/low, low/high, high/low and high/high). No distribution differences were observed among children of non-asthmatic mothers (P=0.79 by Fisher’s Exact Test; Figure 2.4A).

In contrast, the children of asthmatic mothers who developed asthma were almost entirely found among those with high SMAD3 promoter methylation and high IL-

1 secretion (P=0.009 by Fisher’s Exact Test; Figure 2.4B). Moreover, asthmatic children of asthmatic mothers had higher LPS-induced IL-1 than non-asthmatic children (P=0.03 by Student’s t test), whereas comparable IL-1 levels were measured in children without a maternal history of asthma (P=0.65 by Student’s t test; Table 2.2). These results suggest that a strong relationship exists between neonatal SMAD3 methylation, production of the innate cytokine IL-1, and 63 childhood asthma. Furthermore, our data suggest that this relationship is powerfully influenced by maternal asthma.

Figure 2.4. Effects of maternal asthma on the relation between neonatal SMAD3

methylation and IL-1 protein production. The distributions of SMAD3

methylation and log IL-1 producing capacity in non-asthmatic and asthmatic

children without and with a maternal history of asthma were compared by

dividing mean percentage SMAD3 methylation and log IL-1 secretion at the

median, thereby creating four groups (low/low, low/high, high/low and high/high). Distributions for asthmatics and non-asthmatics were compared across all four quadrants (panel A) or focusing the analysis on the high/high and low/low

quadrants (panel B) (P by Fisher’s Exact Test). N: non-asthmatic, A: asthmatic.

Table 2.2. LPS-stimulated CBMC-production of IL-1 in IIS children with or without a history of maternal asthma.

CHILDREN WITH CHILDREN WITH ASTHMATIC MOTHERS NON-ASTHMATIC MOTHERS Non- Non- Asthmatic* Asthmatic* asthmatic asthmatic (n=7) (n=19) (n=12) (n=16) GEOMETRIC MEAN 16.5 6.0 6.7 7.7 (ng/ml) 95% CI 6.9-39.4 4.8-7.6 4.1-10.9 4.7-12.5 (ng/ml) P† 0.03 0.65 * Physician-diagnosed with symptoms or medication use for asthma in the past year reported at least once on the age 2, 3, 5 or 9-year questionnaires † by Student's t-test after log10 transformation.

2.4 – DISCUSSION

Although asthma is the most common chronic complex disease of childhood [15], the timing and mechanisms of its inception remain largely unknown. This gap in knowledge severely hinders efforts aimed at preventing this disease, which no therapeutic regimen can currently cure. Our study, the first epigenome-wide search for asthma-associated methylome signatures at birth, sheds new light on these critical but still open questions. The finding that almost

600 genomic regions were differentially methylated at birth between children who did and did not develop asthma later in life strongly suggests that the trajectory to asthma begins at birth if not prenatally and involves epigenetic mechanisms. As importantly, data from three birth cohorts showed that SMAD3 promoter hypermethylation was associated with childhood asthma selectively in neonates with a maternal history of asthma - one of the strongest, and mechanistically one of the most elusive, risk factors for asthma in the child [20, 127]. These findings are especially noteworthy in the context of the current dearth of characteristics measurable at birth that effectively predict asthma during childhood. Of note, the link between epigenetic SMAD3 dysregulation at birth and childhood asthma appeared to be selective for neonates born to asthmatic mothers, a finding that suggests that heterogeneity is deeply embedded in the pathogenesis of, and the trajectory to, childhood asthma [15, 40]. More generally, to our knowledge this is the first time that a neonatal epigenetic characteristic linked to asthma during childhood is robust enough to replicate across three independent birth cohorts. 66

Our study highlighted SMAD3 and IL-1as main players in the trajectory to childhood asthma. SMAD3, the most connected node in the network of asthma- associated DMRs, is known to be critical for the regulation of both the asthma- protective Treg and the asthma-promoting Th17 cell differentiation programs.

Altered Treg and Th17 activities have been reported in childhood asthma [15, 41] and conversely, maternal exposure to asthma-protective environments such as farming has been shown to activate the Treg compartment [128] and influence the expression of Th17 markers [129]. On the other hand, IL-1, which emerged as the primary upstream regulator of genes harboring asthma-associated DMRs, is increasingly recognized as a key asthma mediator in both children [120] and adults, especially in neutrophilic asthma [119, 121]. Our data emphasize the functional connection between SMAD3 and IL-1. Indeed, in neonates who became asthmatic by age 9, SMAD3 promoter hypermethylation, an epigenetic configuration consistent with low SMAD3 expression, was strongly associated with high IL-1production. This convergence is likely to destabilize the Treg program, enhance inflammation, and promote Th17 differentiation [130], ultimately favoring the development of asthma. While it is unclear whether these mechanisms operate pre- and/or perinatally, detection of the relationship between

SMAD3 methylation and IL-1 production selectively among children of asthmatic mothers implies that the in utero environment is critical for directing the epigenetic trajectory towards childhood asthma.

Our results should be interpreted with caution because our discovery population was small, albeit longitudinally phenotyped for asthma in great detail, 67 and environmental exposures were not comprehensively assessed. Moreover,

DNA methylation did not distinguish between 5-methylcytosine and other cytosine modifications and was assessed in mixed rather than isolated cell populations. However, our data from the COAST population suggest that differential methylation by asthma status did not reflect asthma-associated differences in CBMC proportions. We also acknowledge that the DNA methylation differences we detected were not extreme, albeit more substantial than those recently reported in other studies [109, 131]. This is a recurring theme in the literature [103], and systematic studies are needed to comprehensively assess how relatively modest DNA methylation differences modify disease trajectories. In general, the functional impact of such differences will likely depend on the regulatory properties of the locus in which they reside, and the extent to which additional epigenetic processes, such as post-translational histone modifications, influence those properties.

Gene expression analyses are often used to complement epigenetic studies, but samples for such analyses were not collected at birth in our study populations.

On the other hand, samples for cytokine protein assessments were available and proved essential to integrate our epigenetic findings. Finally, our search for asthma-associated differential methylation returned almost 600 regions, only some of which mapped to genes involved in immunoregulation and inflammation.

DMRs that reside in functionally interesting genes but lack a link to immune regulation or inflammation may also contribute to asthma pathogenesis, albeit through different mechanisms. With these caveats, we propose that in a 68 proportion of children with childhood-onset asthma, a distinctive methylome is in place already at birth, particularly within innate immunoregulatory and pro- inflammatory pathways, and promotes a trajectory that may ultimately lead to clinical disease. Some of the epigenetic mechanisms that contribute to the inception of this trajectory are strongly influenced by the milieu associated with maternal asthma. A scenario in which epigenetic modifications at an early window of susceptibility promote a long-term developmental trajectory to asthma is consistent with the emerging paradigm that chronic non-communicable diseases have their origins in early life through an epigenetic calibration of set points for later responsiveness and function [132, 133].

CHAPTER 3: NEONATAL DNA METHYLATION PROFILES AND THE

MATERNAL PRENATAL IMMUNE MILIEU

3.1 – INTRODUCTION

As discussed in Chapter 1, recent evidence raises the possibility that epigenetic mechanisms may influence the development of asthma already at birth.

Specifically, our epigenome-wide DNA methylation survey in cord blood immune cells from the IIS birth cohort highlighted a link between neonatal DNA methylation profiles and development of asthma during the first decade of life

(Chapter 2 and Ref. [134]). In particular, we found that neonatal (cord blood)

SMAD3 promoter methylation was significantly increased in asthmatic children born to asthmatic mothers, while no such difference was found among children of non-asthmatic mothers.

The finding that an epigenetic trajectory to asthma appears to be in place already at birth led us to investigate whether this trajectory might begin even earlier, that is, in utero. This hypothesis is based on, and supported by, recent epidemiological work from our group that investigated the relationship between the mother’s immune profile during pregnancy and the child’s risk for asthma in the first decade of life.

3.1.1 – Association between maternal immune profiles during pregnancy and risk for childhood asthma

Little is known about whether the maternal immune status during pregnancy influences asthma development in the child. We measured cytokine production in supernatants from mitogen-stimulated peripheral blood immune cells collected during and after pregnancy from the mothers of children enrolled in the IIS birth cohort. Physician-diagnosed active asthma in children through age

9 and a history of asthma in their mothers were assessed through parental questionnaires. Maternal production of each of the cytokines IL-13, IL-4, IL-5,

IFN-γ, IL-10, and IL-17 was unrelated to childhood asthma, but upon creating cytokine ratios, we found that increases in maternal IFN-γ/IL-13 and IFN-γ/IL-4

(and only these two ratios) were associated with decreases in risk of childhood asthma (n=322; OR=0.33; 95%CI=0.17-0.66, P=0.002 and n=312; OR=0.36;

95%CI=0.18-0.71, P=0.003, respectively). The inverse relations of these ratios were strong in non-asthmatic mother-child pairs (OR=0.18; 95% CI=0.08-0.42,

P=0.00007 and OR=0.17; 95% CI=0.07-0.39, P=0.00003, respectively) but were not evident in asthmatic mother-child pairs (P for interaction by maternal asthma

=0.002). Paternal cytokine ratios were unrelated to childhood asthma. In marked contrast, maternal cytokine ratios in non-asthmatic mothers were unrelated to the child’s skin test reactivity, total IgE, physician-confirmed allergic rhinitis at age 5, or eczema in infancy [104]. To our knowledge this study provides the first evidence that cytokine profiles in pregnant non-asthmatic mothers relate to risk 71 for childhood asthma but not allergy, and suggests a process of asthma development that begins in utero and is independent of allergy.

The results of this work (currently in press in American Journal of

Respiratory Cell and Molecular Biology [104] ) supported the hypothesis that the trajectory to asthma can begin in utero, at least for a subgroup of children.

Therefore, I aimed to begin exploring whether this trajectory involves the neonatal epigenome. An additional goal of the work included in this chapter was to move to a different DNA methylation analysis platform (the Illumina

HumanMethylation450 BeadChip array) and delineate the analytical pipeline for a subsequent larger study now funded by R21AI133765 (awarded to my mentor,

Dr. Vercelli, and Dr. Susan V. Lynch, University of California San Francisco). An abridged version of this work has been published in The Journal of Allergy and

Clinical Immunology [J Allergy Clin Immunol 141(6): 1992-1996, 2018. PMID:

29709672].

3.2 – METHODS

Study design and participants The Tucson Infant Immune Study (IIS) has been described in detail in Chapter 2 (section 2.2 and Appendix section A.1).

Maternal prenatal immune profiles IFN and IL-13 were measured by ELISA

(R&D Systems and Diaclone, respectively) in the supernatants of maternal

PBMCs collected during the third trimester of pregnancy and stimulated with concanavalin A/phorbol 12-myristate 13-acetate as described previously [104].

Maternal prenatal IFN/IL-13 ratios were log-10 transformed to achieve a normal distribution.

Sample selection and DNA methylation analysis We selected 32 CBMC samples from neonates whose mothers’ prenatal IFN/IL-13 ratios were representative of the range of values found in the IIS mother population (Figure

3.1). An effort was also made to achieve a balanced representation of asthmatic and non-asthmatic children with and without maternal history of asthma.

Epigenome-wide DNA methylation was measured using the Illumina

HumanMethylation450 platform [135], in which ≈485,000 experimental probes target individual CpGs (section 1.2.2). Probes were removed from the analysis if they were located on sex chromosomes, overlapped the location of common SNPs or mapped to multiple regions of the genome after bisulfite-conversion [136], or if they provided signals not readily distinguishable from the background (detection

P-value > 0.01 in 75% of samples), leaving 327,278 probes for subsequent 73 analyses. Methylation data were processed using the minfi R package [137], and

Infinium type I and type II probe bias was corrected using the SWAN algorithm

[138]. Raw probe intensities were corrected for background and color imbalance by control-probe normalization. The methylation level at each CpG site was reported as a  value (i.e., the fraction of signal obtained from methylated beads over the sum of methylated and unmethylated bead signals), which was interpreted as percent methylation.

Proportions of CBMC cell types (CD4+ and CD8+ T cells, B cells, NK cells, and monocytes) were estimated using the minfi package as described by

Bakulski et. al [89]. Principal component analysis was performed to identify chip effects and potential confounding variables (cell proportions, chip, batches of

DNA extraction and bisulfite conversion, and child sex). Linear regression was used to remove variation due to differences in cell proportions for CD4+ and

CD8+ T cells and monocytes. The ComBat function in the sva R package [139] was used to adjust for chip effects. Surrogate variable analysis was used to identify additional confounding variables that were not associated with the maternal prenatal cytokine ratio or childhood asthma; these effects were removed by linear regression. Residual methylation  values were used for the remaining analysis. Linear regression was used to test for associations between the maternal prenatal IFN/IL-13 ratio (as a continuous variable) and neonatal DNA methylation at individual CpGs. Child sex was included as a covariate in this analysis.

Figure 3.1. Samples selected for pilot DNA methylation study. The figure shows

the cord blood mononuclear cell samples (n=32) that were selected for DNA

methylation analysis [samples from subjects with a non-asthmatic mother (red circles, n=19) or with an asthmatic mother (blue triangles, n=13)] and those that were not selected (small, dark grey dots). In order to illustrate that the mothers of these samples’ donors had prenatal cytokine profiles representative of the overall distribution (n=441), the y-axis shows each child’s mother’s prenatal IFN/IL-13 cytokine ratio plotted against the ranked IFN/IL-13 cytokine ratio value (x-axis). 75

Genomic annotation Maternal prenatal IFN/IL-13 ratio-associated CpGs were annotated using the Illumina manifest (human genome version hg19). When

CpGs were assigned more than one gene name and/or gene body annotation, only the first was kept for subsequent pathway analyses.

Pathway analyses Names of genes containing prenatal IFN/IL-13 ratio- associated CpGs were uploaded to IPA. Using the Ingenuity knowledge base, an upstream regulator analysis was performed to test whether there was an enrichment for prenatal IFN/IL-13 ratio-associated genes located downstream of a given gene regulator. The resulting overlap P-value measures the enrichment of downstream genes, without taking into account the direction of regulation [140].

Interaction networks were generated using the default settings. Network scores were calculated from a Fisher’s exact test and reported as the -log10(P-value), indicating the likelihood of finding the observed number of genes in a given network by chance [141].

Statistical analyses Standard tests were used for all analyses, including linear regression, 2 test, and Student’s t-test. The maternal prenatal IFN/IL-13 ratio was log10-transformed prior to testing for associations with neonatal DNA methylation. Due to the exploratory nature of this study and the small sample size, we chose a relaxed threshold of significance (FDR <0.15). All analyses were conducted in R (v. 3.3.0).

3.3 – RESULTS

3.3.1 – Profiles of neonatal DNA methylation are associated with the maternal prenatal immune profile selectively in neonates born to non- asthmatic mothers

This pilot study was designed to investigate the relationship between the maternal immune profile during pregnancy (a variable associated with childhood asthma risk [104]) and neonatal epigenome-wide DNA methylation, in order to begin assessing whether the epigenetic trajectory to asthma may begin in utero.

To address this question, I profiled epigenome-wide DNA methylation in CBMC samples from 32 children enrolled in the IIS cohort [104, 134, 142] using the

Illumina HumanMethylation450, and I tested for associations between neonatal

DNA methylation at individual CpGs and the maternal prenatal IFN/IL-13 ratio.

Neonatal DNA methylation was not significantly associated with the maternal prenatal IFN/IL-13 ratio in the entire study population (n=32). Because we previously showed that maternal asthma influences 1) the relationship between neonatal DNA methylation of immune cells and childhood asthma risk (section

2.3.3) and 2) the relationship between the maternal prenatal IFN/IL-13 ratio and childhood asthma (section 3.1.1 [104]), additional analyses were performed stratifying by maternal asthma. At an FDR<0.15, neonatal DNA methylation at

553 CpG sites was significantly associated with the maternal prenatal IFN/IL-13 ratio selectively in children with non-asthmatic mothers. In marked contrast, no

CpG sites were significantly associated with the IFN/IL-13 ratio in children of 77 asthmatic mothers even when the threshold of statistical significance was further relaxed (FDR < 0.25, Figure 3.2). These findings point to a relationship between a pregnant mother’s immune profile and her child’s developing epigenome, and this relationship is modified by her asthma status.

The CpG sites associated with the maternal prenatal IFN/IL-13 ratio in children of non-asthmatic mothers were distributes throughout the genome

(Figure 3.3) with no obvious clustering of significant hits with respect to genomic location (2 = 4.89, df = 6, P-value = 0.55, Table 3.1).

Figure 3.2. Number of CpG sites associated with the maternal prenatal IFN/IL-

13 ratio at various levels of statistical significance. Each circle represents the

number of CpG sites whose methylation levels at birth were significantly associated with the maternal IFN/IL-13 ratio at different thresholds of statistical significance [FDR <0.15 (yellow), 0.2 (green), and 0.25 (blue), respectively]. The

left group of circles represents the analysis limited to children of non-asthmatic mothers, while the right group of circles represents the analysis limited to children

of asthmatic mothers.

Figure 3.3. Chromosomal location of neonatal CpG sites associated with the maternal prenatal IFN/IL-13 ratio in children of non-asthmatic mothers. Each dot in the Manhattan plot represents a CpG plotted by chromosome (x-axis) and significance (y-axis) for the association between maternal prenatal IFN/IL-13 ratio and residual methylation -values at that CpG. The dashed line represents

the threshold for significance (FDR <0.15).

Table 3.1. Genomic distribution of neonatal CpG sites associated with the maternal prenatal IFN/IL-13 ratio in children of non-asthmatic mothers No. Maternal Prenatal % Genomic Expected No. in Genomic IFN/IL-13 ratio- Location Genomic Location Location associated CpGs (%) * Covered by Array† (from 2 test) Intergenic 126 (23) 23 127.2 1stExon 26 (4.7) 5 27.7 3'UTR 16 (2.9) 3.7 20.5 5'UTR 44 (8) 9.1 50.3 Gene Body 194 (35) 34 188 TSS1500‡ 94 (17) 15 83 TSS200§ 53 (9.6) 11 60.8 *Total n=553 CpGs; †Total N=327,278 CpGs; 2 = 4.89, df = 6, P-value = 0.55 ‡TSS1500: within 1500bp of TSS §TSS200: within 200bp of TSS UTR: untranslated region, TSS: transcription start site

3.3.2 – Neonatal CpG sites associated with the maternal IFN/IL-13 ratio cluster in the TGFB1 pathway

To begin deciphering the biology underpinning the relationship between maternal immune milieu and the child’s epigenetic profile at birth, the names of the genes containing the 553 CpG sites associated with the maternal IFN/IL-13 ratio were uploaded to IPA. I used the Ingenuity knowledge base to perform an upstream regulator analysis, which tested for an enrichment of differentially methylated genes that are associated with the maternal IFN/IL-13 ratio and are downstream of a given gene regulator. This analysis identified TGFB1 as a prominent upstream regulator (P<0.00009, Table 3.2) and showed that maternal

IFN/IL-13 ratio-associated CpGs primarily cluster in genes within the retinoic acid receptor and aryl hydrocarbon receptor pathways (Table 3.3). Interaction networks were created for the genes containing IFN/IL-13 ratio-associated CpGs using the Ingenuity knowledge base and scored based on the probability of finding the observed number of genes in a given network by chance. The network with the highest score again highlighted genes within the retinoic acid receptor and aryl hydrocarbon receptor pathways (RARA, NFE2L2, SMARCA4, and

NR2F2) (Figure 3.4). Together, these findings suggest that the maternal prenatal immune milieu may influence DNA methylation primarily in genes involved in the response to environmental regulatory signals [143-145].

As an illustration of the analyses I will conduct in future studies, I next explored whether the genes that harbor neonatal CpG sites associated with the 82

maternal IFN/IL-13 ratio in children of non-asthmatic mothers and map within the aryl hydrocarbon receptor and retinoic acid receptor pathways may also be involved in the pathogenesis of childhood asthma. To this end, I asked whether methylation levels in genes mapping to these pathways from the highest scoring interaction network (Figure 3.4, RARA, NFE2L2, SMARCA4, and NR2F2) differed between neonates who did and did not develop asthma during childhood.

Interestingly, in addition to methylation being significantly associated with the maternal prenatal immune profile, neonatal methylation levels at all four of these genes were also significantly different between children who did and did not become asthmatic (Figure 3.5), suggesting that genes involved in the response to environmental agents may also be important in setting the epigenetic trajectory to asthma in the child – a possibility that will be further explored in subsequent studies targeting the entire IIS population.

Table 3.2. Top 5 upstream regulators of genes containing maternal IFN/IL-13 ratio-associated CpGs in children of non-asthmatic mothers.

P-value Upstream Molecule of Target molecules in dataset Regulator Type overlap 1.87E- CXCR5, IL6, MAGI1, NCL, PTBP1, TERT, CDK9 kinase 05 VIM transcription 6.40E- CPT1A, FCER1A, GATA3, GHRL, NR5A1, USF2 regulator 05 TERT, THBS1 ACLY, CCNO, CCRL2, CDH4, CDHR1, CELF2, CHRNB4, CHST11, CNIH1, CXCR5, DNAH2, EPS8L3, FCER1A, FGF6, FLI1, FLT1, GABRD, GATA3, GJA1, GRHL1, HMGN2, growth 8.68E- IDI1, IL21, IL6, ITGAV, ITPR2, MAD1L1, TGFB1 factor 05 MMP13, NCR2, NEU3, PARP3, PILRA, PLAGL1, RARA, RASGRP3, RNH1, SGSH, SPAG4, SPARC, TERT, THBS1, TLL1, TMEM184B, VAT1L, VCAN, VIM, WISP1, WNT5B, YBX1 1.49E- ALDH1A2, CELF2, FLI1, GJA1, IL6, MMP13, Jnk group 04 NFE2L2, RARA, TBP, TERT, TNN, VCAN mature 1.65E- miR-138-5p ALDH1A2, TERT, VCAN microrna 04

Table 3.3. Top 5 canonical pathways enriched for genes containing maternal IFN/IL-13 ratio-associated CpGs in children of non-asthmatic mothers.

Canonical Pathways -log(P-value) Molecules RELA, PNRC1, RARA, ALDH1A2, Retinoic Acid Receptor 2.56 MAP3K1, NR2F2, PRKAR1B, Activation SMARCC1, SMARCD3, SMARCA4 Aryl Hydrocarbon RELA, NFIX, TFDP1, RARA, 2.36 Receptor Signaling ALDH1A2, IL6, NFE2L2, SMARCA4 RELA, CDH4, GNA12, MAP3K1, Gα12/13 Signaling 1.99 CDH17, ARHGEF1, CDH13

RELA, ITPR2, PLCG2, GNA12, TBP, Thrombin Signaling 1.89 ARHGEF1, GATA3, ARHGEF10, PLCD4

Role of BRCA1 in DNA E2F5, SMARCC1, BRE, SMARCD3, 1.85 Damage Response SMARCA4

Figure 3.4. Network of genes harboring differential neonatal methylation

associated with the maternal prenatal IFN/IL-13 ratio in children of non-

asthmatic mothers. The network score generated from Ingenuity Pathway

Analysis is 46, indicating an extremely low likelihood (P=1x10-46) of finding the

observed number of genes in the network by chance. Symbols shaded in grey

represent genes whose DNA methylation was significantly associated with the maternal prenatal cytokine ratio. Symbols further highlighted in yellow, blue and

green represent genes that are members of the retinoid acid receptor (RAR) or

aryl hydrocarbon receptor (AHR) canonical pathways, or both. 85

Figure 3.5. Relationships between neonatal DNA methylation in genes in the aryl

hydrocarbon receptor and retinoic receptor pathways, the maternal prenatal

IFN/IL-13 ratio, and childhood asthma in children of non-asthmatic mothers.

Scatterplots (left panels) show the relationship between the maternal prenatal

IFN/IL-13 ratio (x-axis) and residual methylation (adjusted for child sex) at

NFE2L2 cg13344867 (panel A), NR2F2 cg15379412 (panel B), SMARCA4

cg18685158 (panel C), and RARA cg26878655 (panel D) (y-axis); R2 and P-

values are reported from linear regression. Boxplots (right panels) show the

relationship between the residual methylation (adjusted for child sex) and

childhood asthma (P by Student’s t-test). Non-asthmatic and asthmatic children

are represented as grey and red dots, respectively. 86

3.4 – DISCUSSION

The study discussed in this chapter was designed to be exploratory and to set in place an analytical pipeline that will be leveraged in future, funded work.

Nevertheless, to our knowledge, this is the first study that investigates the relationship between the maternal prenatal immune status (here captured by the

IFN/IL-13 cytokine ratio during the third trimester of pregnancy) and neonatal

DNA methylation in cord blood immune cells. Although our results should be interpreted with caution because of the small sample size, they raise the possibility that the maternal prenatal immune milieu may influence DNA methylation profiles at birth, with effects selectively seen in children born to non- asthmatic mothers. Differential methylation primarily clustered within genes involved in the response to environmental regulatory signals. Notably, some of the genes that harbored neonatal differential methylation associated with the maternal cytokine ratio-associated (NFE2L2, SMARCA4, NR2F2 and RARA) were also differentially methylated between children who did and did not develop asthma by age 9, emphasizing a potential role for responses to environmental signals in disease pathogenesis. Collectively, our results - albeit preliminary - suggest that the epigenetic trajectory to childhood asthma may begin in utero.

When we interrogated maternal prenatal ratio-associated epigenetic signatures for biological significance, maternal IFN/IL-13 ratio-associated CpGs were enriched for genes downstream of TGFB1, an immunoregulatory gene responsible for balancing the Treg and Th17 developmental programs, which have been implicated in asthma pathogenesis with opposite roles [15, 41]. 87

Moreover TGF-, a pleiotropic growth factor, plays important roles in the development of several organs, including the lung [146]. Thus, the TGF- pathway may be relevant to both the immune and the respiratory components of the asthma trajectory. Interestingly, our previous study of the relation between

DNA methylation profiles at birth and risk of asthma in the first decade of life also highlighted a link to TGF- signaling through its master regulator, SMAD3.

Indeed, neonatal SMAD3 methylation was associated with asthma, but this relationship was limited to children of asthmatic mothers [134]. In contrast, TGF-

 appears to influence the relation between maternal immune status and neonatal methylome only in children of non-asthmatic mothers. Together, these findings appear to suggest that distinct epigenetic trajectories influenced by the prenatal environment and maternal history of asthma lead to childhood asthma and are differentially modulated by the TGF- pathway.

This study has limitations. Since our sample size was extremely small, our results should be considered purely exploratory. Moreover, we assessed DNA methylation in mixed cell populations, and therefore we cannot determine the cellular source of the signals we observed. We tried to minimize this problem by using state of the art computational methods to estimate the impact of differences in cell proportions on our data. For example, since we observed that the estimated proportion of T cells (both CD4+ and CD8+) and monocytes were associated with global methylation levels, we have adjusted for these cell types in our analyses to minimize bias associated with these differences. It would have been beneficial to validate our epigenetic data through a complementary approach (such as gene 88 expression), but adequate samples were not originally collected. Finally, we do not imply that the relative abundance of maternal IFN or IL-13 directly alters the developing epigenome of the fetus; it is also possible that the prenatal IFN/IL-13 ratio may reflect environmental exposures that prime both the immune profile of the pregnant mother and the epigenetic trajectory to asthma in the developing child. Further investigations required to better understand these relationships will be discussed in Chapter 5. 89

CHAPTER 4: LEVERAGING TRANSCRIPTOMICS AND NETWORK

APPROACHES TO EXPLORE THE ASTHMA-PROTECTIVE EFFECTS

OF FARM EXPOSURE

4.1 – INTRODUCTION

In the previous chapters I focused on epigenetic mechanisms associated with asthma inception, which are thought to be engaged by environmental stimuli.

In this chapter I will rely on transcriptomics and network approaches to directly explore how an exposure strongly associated with asthma protection in human populations affects airway and immune responses in experimental asthma models.

Because the work presented in this chapter builds off work performed from other members of our lab group, I will use first-person pronouns (I, my) and third- person pronouns (we, our) to distinguish the work that I have done from work that was performed by other members of our lab group.

Among the several exposures that have been studied for their impact on asthma risk [21-25, 28, 29, 31, 32], exposure to farming environments, especially in early life, has been found to be strongly associated with decreased risk for asthma and allergy across many studies performed in Europe, North America and

Australia [72, 73]. Recent work from members of our group characterized the relations between asthma prevalence and the environment in two U.S. farming populations, the Indiana Amish and the South Dakota Hutterites. Amish children are protected from asthma and have distinct innate immune profiles when 90 compared to Hutterite children [105]. Because significant differences were observed in the microbial load and composition of dust collected from Amish and

Hutterite homes, our laboratory developed mouse models to dissect the environmental impact of farming on asthma risk. By generating extracts from house dust collected from each community, we showed that intranasal treatment of mice with Amish dust extract was sufficient to protect from allergen-induced asthma-related phenotypes [e.g., airway hyper-responsiveness (AHR) and broncho-alveolar lavage (BAL) eosinophilia], whereas mice treated with allergen and Hutterite dust extract were not protected. Importantly, mice treated with allergen and Amish dust extract were no longer protected from experimental asthma if they lacked MyD88 and Trif and thus were deficient in innate immune signaling [105].

4.1.1– Intranasal treatment with allergen and Amish dust extract induces

T cells in the lung

The work mentioned above showed that protection from asthma involves the innate immune system, but the underlying mechanisms remained unknown.

Searching for potential cellular mechanisms of protection, our group used flow cytometry to characterize the cellular composition of the lung in mice treated with allergen and Amish dust extracts. Compared to mice that received allergen alone, recipients of allergen and Amish dust extract or Amish dust extract alone had a four-fold increase in the total number of T cells, nearly half of which were T 91

cells expressing the transcription factor Rort, the master regulator of IL-17- producing cells (Figure 4.1A). These cells were therefore 17 T cells. Induction of a 17 T cell response was specific to mice treated with Amish dust extracts because the number of 17 T cells was negligible in mice treated with allergen and Hutterite dust extracts or allergen alone. In more than 100 experiments performed to date, our lab has shown that induction of lung T cells almost invariably accompanies the induction of asthma protection by Amish dust extracts.

4.1.2– T cells in the lungs of mice treated with Amish dust extract express V4

Because it is known that mouse T cells have different functional properties depending on which T cell receptor V region gene they express [147-

152], we used flow cytometry to further characterize T cells from the lungs of mice treated with Amish dust extract with or without allergen, and we showed that these cells were >95% V4+ (Tonegawa’s nomenclature [153], Figure 4.1B).

V4+ T cells have been shown to preferentially produce IL-17 in response to

IL-1 and IL-23 [148]. Most importantly, these cells have been reported to decrease serum IgE and AHR, whereas V1+ T cells increase serum IgE levels

[147, 149-152]. Consistent with a suppressive effect of V4+ T cells on type

2 responses in the lung, using V4 mRNA expression as a signature of V4+17 92

T cell responses in the lungs of mice treated with Amish dust extracts, I found that levels of V4 mRNA were strongly and negatively correlated with both AHR

(P=4.54E-06) and BAL eosinophilia (P=0.007). These findings raise the possibility that V4+17 T cells play a mechanistic role in asthma protection in our model of Amish farm dust exposure.

4.1.3 –Network approaches to characterize the lung and 17 T cell transcriptomes in associated with asthma protection

I next took a transcriptome-based approach to begin investigating the potential role of T cells in protection from experimental allergic asthma.

First, I performed RNA-sequencing (RNA-seq) to characterize global lung gene expression profiles and identify protection-associated networks of co-regulated genes (modules). Next, I asked which protection-associated modules correlated with V4 mRNA levels, a specific 17 T cell signature. Finally, V4+17 T cells were isolated from the lungs of mice treated with Amish dust extracts in the presence or absence of allergen. Then, I characterized the gene expression profiles of these cells in our experimental model of asthma, and I asked whether these profiles were associated with protection from cardinal asthma-related phenotypes.

Figure 4.1. Flow cytometry analysis of lung immune cells from a murine model of

asthma. A) Immune cell subsets in the lungs of mice treated intra-nasally with allergen [house dust mite (HDM)] and Amish or Hutterite house dust extracts. DN

cells correspond to CD4-CD8- T cells, which implies these cells lack  T cell receptor and are instead classified as  T cells. B) Representative flow cytometry

plots showing lung cells isolated from allergen- and Amish dust extract-treated mice and stained with PE-pan-anti- T cell receptor monoclonal antibody (mAb)

(clone GL3, BD Pharmingen) or APC-anti-mouse TCR Vγ4 mAb (Tonegawa’s nomenclature [153], clone UC3-10A6, BioLegend), gating on live lymphocytes.

Percentages of T cells and V4+ T cells are shown in the upper-right

quadrants of the left and right plots, respectively. 94

4.2 – METHODS

Experimental models and treatment groups Figure 4.2 provides a detailed view of the design of our study. In order to maximize the number of samples under investigation and identify molecular mechanisms shared by different experimental conditions, we included three experimental models in our analysis. In the first model, beginning 10 days before the first allergen sensitization (day -10), 50 μl of

Amish or Hutterite house dust extract (generated as described previously [105,

106]) were instilled intranasally every 2-3 days (for a total of 14 times) into 8- week old BALB/c mice (Envigo). The mice received 100 μg of house dust mite

(Greer) intranasally on days 0, 7, 14 and 21. A subset of mice were treated with

Amish dust extract but were given PBS instead of house dust mite intranasally as described above. Mice were sacrificed on day 23 for terminal assessments.

In the second model, 50 μl of Amish house dust extract were instilled intranasally every 2 to 3 days (for a total of 14 times) into 7-week old BALB/c mice (Envigo), beginning on day 0. The mice were sensitized intraperitoneally with 20 μg of ovalbumin (grade V, Sigma) plus alum (Pierce) on days 0 and 14 and were challenged intranasally with 50 μg of ovalbumin on days 28 and 38. A subset of mice were treated with Amish dust extract but were given PBS instead of ovalbumin intraperitoneally and intranasally as described above. Mice were sacrificed for terminal assessments on day 38.

In the third model, beginning 5 days before the first allergen sensitization

(day -5), 50 μl of Amish dust extract were also instilled intranasally every 2 to 3 days (for a total of 14 times) into 7-week old C57/BL6 wild-type mice and into 95 mice deficient in both MyD88 and Trif [154] (Jackson Laboratories). These mice were sensitized intraperitoneally with 20 μg of ovalbumin plus alum on days 0 and 14 and challenged intranasally with 75 μg of ovalbumin on days 26, 27, and

28. Mice were sacrificed for terminal assessments on day 30.

Isolation of 17 T cells from lungs of mice treated with Amish dust extract

17 T cells were isolated from the lungs of mice treated with Amish dust extract in the presence or absence of allergen (Figure 4.2 model 2). Lung tissues were harvested, minced and digested with Liberase TM Research Grade (Roche) and

DNase I (Sigma) at 37°C for 1 hour. Single cell suspensions were prepared by passing digested lung tissue through the 70 m Cell Strainer (Falcon).  T cells were isolated from lung cell suspension using a magnetic TCR/+ T cell isolation kit (Miltenyi Biotec) according to the manufacturer protocol. Briefly, non-T cells were depleted using CD45R (B220) and CD11b magnetic beads followed by positive selection of T cells using biotinylated anti-TCR/ antibodies and anti-biotin magnetic beads. The purity of isolated T cells was between 77 and 89% as determined by flow cytometry.

RNA-sequencing (RNA-seq) Whole lungs (total n=91) and isolated T cells

(n=6) were collected from mice treated as described in Figure 4.2, and samples from each model were processed for RNA-seq as a separate batch. RNA was extracted using Qiagen RNeasy mini kits for all samples. cDNA libraries for 96 samples from experimental model 1 were constructed using Illumina TruSeq

(University of Chicago Genomics Facility: Knapp Center for Biomedical

Discovery), while cDNA libraries for samples from experimental models 2-3 were constructed (in separate batches) using KAPA Biosystems kits (in house); all libraries were run on the Illumina HiSeq 2000 platform at the University of

Chicago Genomics Facility, Knapp Center for Biomedical Discovery. Adapter sequences were trimmed from RNA-seq reads and only reads >25 bp long were included in the analysis. Reads were then mapped to the mouse genome (version mm10) via STAR [155]. HOMER (http://homer.ucsd.edu/homer/index.html) was used to select uniquely aligning reads and generate gene counts. The median number of uniquely mapped reads per mouse was 33,863,913 (range: 14,635,768-

82,116,549). Reads mapping to Tcrg [Tcrg-C1, Tcrg-V1, Trgv2, Tcrg-V3, Tcrg-

V4, Tcrg-V5, Tcrg-V6 and Tcrg-V7] and Tcrd [Trdc, Trdv1, Trdv2-1, Trdv2-2,

Trdv3, Trdv4, and Trdv5: GENCODE Version M16] genes were quantified using the annotatePeaks function in HOMER. Expression data were estimated for

24,538 genes and were filtered to retain genes that had ≥10 reads in ≥20% of samples, thereby reducing the total number of genes to 16,820. Gene counts were normalized by DESeq2 to account for library size and dispersion estimates [156].

Principal component analysis was used to identify confounding variables. Batch effects were removed using linear regression and residual expression data were used in subsequent analyses.

Differential expression analysis Lists of differentially expressed genes were created using the DESeq2 package in R. Unless otherwise stated, differentially expressed genes were defined as having |log2 fold change (log2FC)| > 1 and an

FDR-adjusted P-value < 0.05.

Weighted Gene Co-expression Network Analysis (WGCNA) Whole lung gene expression data for 16,820 genes were clustered into co-regulated modules using

WGCNA [157]. I determined the soft thresholding power to be 7 and used the

“signed” network which is a robust option for analyzing data from heterogeneous tissues. Modules were created using a deepSplit value of 2 and PAMstage was turned off. Modules were merged if their correlation to each other was greater than 0.9. All other settings were kept at the default values. Each gene was assigned to one module and one module only based on this clustering approach.

Expression data within each module were summarized using the module eigengene vector (i.e., the first principal component of the module), and their correlation to airway and immune phenotypes linked to protection was assessed using Pearson correlation. Module membership (MM) represents the correlation between each gene member and the module eigengene.

Pathway and enrichment analyses Module genes were uploaded to IPA.

Pathway analysis was performed to test for an enrichment of module genes that are members of individual molecular pathways. The resulting overlap P-value measures the enrichment in pathway member genes, without taking into account 98

the direction of regulation [140]. We tested for enrichment of 17 T cell signature genes within each module by hypergeometric test and resulting P-values were Bonferroni-corrected for multiple testing.

Figure 4.2. Experimental models and samples included in RNA-sequencing analysis. Abbreviations: I.N. – intra-nasal administration, I.P. – intra-peritoneal administration, OVA – ovalbumin, HDM – house dust mite, PBS – phosphate

buffered saline. 100

4.3 – RESULTS

4.3.1– Network analysis of transcriptome profiles in unfractionated lung cells

In order to identify potential mechanisms of asthma protection, I analyzed transcriptome profiles in unfractionated lungs from mice exposed to allergen in the presence or absence of Amish or Hutterite farm dust extracts (n=91, Figure

4.2). Initial differential expression analyses using conventional analytical approaches yielded inconclusive results, probably due to the high complexity of the tissue under examination. Therefore, I moved to a network approach based on

WGCNA, which seeks to identify groups (modules) of co-regulated genes.

Twelve modules emerged from this analysis (Figure 4.3). In order to assess which of these modules were associated with asthma protection, I summarized expression data within each module using the module eigengene vector (which is analogous to the first principal component), and I tested for associations between module eigengenes and the cardinal parameters of experimental allergic asthma:

AHR and BAL eosinophilia. Table 4.1 shows that the blue and magenta modules were associated with a full asthma-protective response (decreases in both AHR and BAL eosinophilia), while other modules were selectively associated with decreased AHR (purple and tan) or decreased BAL eosinophilia (brown).

Therefore, subsequent analyses focused on the blue and magenta modules.

Next, I asked whether the magenta and blue modules associated with asthma protection were also associated with a  T cell signature. To address 101

this question, I extracted from RNA-seq data expression levels for V4, the T cell receptor V region gene specifically expressed by T cells, and I examined the relation between this T cell signature and the blue and magenta modules.

This analysis showed that the magenta but not the blue module exhibited a highly significant (P=4.6E-11) association with V4 expression (Table 4.1), suggesting that T cell genes are represented in, or influence the activity of, this asthma- protective group of co-regulated genes.

Figure 4.3. Module identification by WGCNA. The figure depicts a gene

dendrogram where the distance between two genes indicates their correlation to one another. The color bar demonstrates which clusters of genes were assigned to

a given module. Note: the grey module includes genes that did not cluster into a

module (n=9,334 out of a total of 16,820 genes).

102

Table 4.1. Correlations between WGCNA module eigengenes and phenotypes linked to asthma protection. %Eosinophils Module No. Genes AHR* Tcrg-V4 (V4)† (BAL) 0.56 0.34 -0.22 TURQUOISE 2306 (8.7e-09) (0.001) (0.035) -0.56 -0.34 0.19 BLUE 1899 (7.6e-09) (0.0012) (0.071) -0.49 -0.24 -0.24 BROWN 1227 (6.5e-07) (0.02) (0.02) -0.17 -0.012 -0.2 YELLOW 386 (0.098) (0.91) (0.054) 0.34 0.51 GREEN 333 0.033 (8.6e-04) (0.76) (2.7e-07) 0.54 0.33 RED 290 0.13 (2.5e-08) (0.0014) (0.23) -0.24 -0.0078 -0.57 BLACK 253 (0.019) (0.94) (3.4e-09) -0.23 0.76 PINK 250 0.12 (0.25) (0.029) (2.8e-18) -0.39 -0.51 0.62 MAGENTA 198 (1.4e-04) (2.7e-07) (4.6e-11) -0.17 -0.64 0.84 PURPLE 149 (0.11) (1.2e-11) (9.5e-26) -0.16 -0.13 -0.27 GREENYELLOW 100 (0.13) (0.23) (0.0088) -0.25 -0.35 0.24 TAN 95 (0.016) (6.9e-04) (0.021) The table shows Pearson correlation coefficient (P-value) for each pairwise comparison. P-values ≤ 0.0014 significant (bolded) after adjusting for multiple testing using Bonferroni correction. *AHR was transformed by taking the slope of the curve for each mouse before assessing correlations with module eigengenes. †Expression data for Tcrg-V4 (V4) were extracted from the RNA-seq dataset.

103

4.3.2– Characterization of the transcriptome of isolated T cells

To dissect how 17 T cell genes contribute to the composition of asthma- protective lung gene modules (the magenta module, first and foremost), we next used magnetic cell sorting to isolate V4+17 T cells from lungs of mice treated with Amish dust extract in the presence or absence of allergen. RNA-seq analysis was then performed to characterize the transcriptomes of these cells (Figure 4.2, model 2).

Because this is the first time that the transcriptome of T cells elicited by Amish farm dust extracts has been profiled, I first compared 17 T cells isolated from mice treated with Amish dust extract in the presence or absence of allergen, and found that <1.25% of the genes (n=197) were differentially expressed in these conditions (|log2FC| > 1, FDR < 0.05). Therefore, data were combined in all subsequent analyses.

To identify an asthma-protective expression signature in whole lung that is induced by Amish dust extracts and is explained by the presence of  T cells, I relied on a three-step approach. In the first step, I compared the lungs of mice treated with Amish dust extract + allergen (which are protected from asthma and rich in 17 T cells) to the lungs of mice treated with allergen alone (which are not protected from asthma and lack 17 T cells) (log2FC > 1, FDR < 0.05). This comparison identified 2,004 genes induced by Amish dust extract treatment.

Because this treatment has complex effects on lung cellularity, this set of genes included both 17 T cell and non-17 T cell genes. 104

In the second step, I aimed to identify a set of T cell genes. To this end, I compared gene expression profiles from isolated T cells to the expression profiles of whole lungs from allergen-treated mice, which lack T cells but are populated by structural and inflammatory cells. This comparison

(performed using a high log2FC threshold: log2FC > 3) allowed me to exclude genes that are highly expressed by non-17 T cells in the lung and identified 617 differentially expressed  T cell genes (FDR < 0.05).

Finally, in the third step, I defined a T cell gene signature (n=127 genes, Table 4.2) as the overlap between the two comparisons outlined above.

Notably, among the top genes in the signature were Tcrg-V4 (V4) and Trdv5 (the

V5 region of the TCR  chain), a finding that supports my approach because it is consistent with the TCR expression profile of  T cells in our models. Il17a,

Rorc, and Il23r, genes characteristic of V4+17 T cells [148], were also among the top genes in my list. Moreover, this signature contained several other genes reported to be expressed in  T cells, including Icos [158], Pdcd1 [159], Cxcr6

[160], Tnf [161], Itgae [162], Blk [163], Il12rb1 [164], Rasgrp1 [165], Thy1

[166], Cd3g[167], and Cd3d [167].

105

Table 4.2. 17 T cell signature genes * † Gene Symbol Gene Name log2FC P-value Il17a interleukin 17A 9.275 1.99E-24 Il17f interleukin 17F 9.221 8.43E-30 RIKEN cDNA 5830411N06Rik 8.442 4.24E-25 5830411N06 gene neurofilament, light Nefl 8.166 1.13E-20 polypeptide T cell receptor delta, Trdv5 8.042 1.15E-17 variable 5 Il23r interleukin 23 receptor 7.960 6.49E-60 inducible T cell co- Icos 7.888 7.61E-38 stimulator programmed cell death Pdcd1 7.679 2.55E-27 1 T cell receptor gamma, Tcrg-V4 7.631 9.00E-16 variable 4 T cell receptor delta, Trdc 7.448 4.29E-23 constant region dual specificity Dusp4 7.061 8.13E-278 phosphatase 4 Krt83 keratin 83 7.017 7.36E-45 T cell receptor gamma, Tcrg-C1 6.855 2.30E-17 constant 1 lymphocyte antigen 6 Ly6g5b 6.829 3.54E-18 complex, locus G5B chemokine (C-X-C Cxcr6 6.798 2.19E-19 motif) receptor 6 Cd163l1 CD163 molecule-like 1 6.736 5.89E-58 RIKEN cDNA 1700012B07Rik 6.726 6.58E-24 1700012B07 gene leucine rich repeat and Lingo4 6.454 6.77E-14 Ig domain containing 4 CD3 antigen, gamma Cd3g 6.205 3.99E-22 polypeptide Tnf tumor necrosis factor 6.199 2.22E-59 integrin alpha E, Itgae 6.156 7.27E-39 epithelial-associated potassium voltage gated channel, Shaw- Kcnc1 6.152 8.24E-18 related subfamily, member 1 Blk B lymphoid kinase 6.103 3.78E-17 chemokine (C motif) Xcr1 6.050 6.44E-53 receptor 1 Cdh10 cadherin 10 5.969 1.70E-14 lysine (K)-specific Kdm2b 5.895 8.13E-278 demethylase 2B 106

RIKEN cDNA 9530052E02Rik 5.863 3.60E-137 9530052E02 gene Olfr60 olfactory receptor 60 5.860 1.57E-14 Gm15511 predicted gene 15511 5.817 2.49E-120 X-linked lymphocyte- Xlr4a 5.790 3.55E-22 regulated 4A pleckstrin and Sec7 Psd2 5.788 7.03E-14 domain containing 2 insulin receptor Irs2 5.765 1.40E-179 substrate 2 aryl hydrocarbon Arnt2 receptor nuclear 5.732 9.62E-24 translocator 2 UDP-Gal:betaGlcNAc beta 1,3- B3galt5 5.615 4.46E-30 galactosyltransferase, polypeptide 5 CD3 antigen, epsilon Cd3e 5.550 3.23E-31 polypeptide acyl-CoA synthetase Acsbg1 bubblegum family 5.476 3.71E-18 member 1 proprotein convertase Pcsk1 5.458 1.15E-80 subtilisin/kexin type 1 Cd96 CD96 antigen 5.452 3.52E-29 SH2 domain containing Sh2d2a 5.444 1.30E-29 2A linker for activation of Lat 5.440 1.92E-24 T cells Aqp3 aquaporin 3 5.356 2.58E-12 Apol7e apolipoprotein L 7e 5.328 4.54E-10 Rnf125 ring finger protein 125 5.307 2.51E-153 adhesion G protein- Adgrg5 5.271 2.73E-27 coupled receptor G5 trophoblast Tpbg 5.246 4.16E-154 glycoprotein acid phosphatase, Acpp 5.162 9.36E-29 prostate zeta-chain (TCR) Zap70 associated protein 5.042 8.09E-26 kinase thymus cell antigen 1, Thy1 4.997 3.54E-36 theta Msc musculin 4.958 9.08E-23 B cell Bcl11b leukemia/lymphoma 4.937 3.22E-35 11B G protein-coupled Gpr68 4.931 4.27E-32 receptor 68 107

proline-rich Prrt1 transmembrane protein 4.885 1.30E-25 1 tumor necrosis factor Tnfrsf25 receptor superfamily, 4.865 3.55E-46 member 25 zinc finger and BTB Zbtb1 4.817 8.13E-278 domain containing 1 beta-1,4-N-acetyl- B4galnt4 galactosaminyl 4.803 1.48E-13 transferase 4 Itgb7 integrin beta 7 4.784 7.26E-38 receptor (calcitonin) Ramp1 activity modifying 4.719 7.55E-28 protein 1 CD3 antigen, delta Cd3d 4.713 3.06E-32 polypeptide FYVE and coiled-coil Fyco1 4.669 3.95E-102 domain containing 1 endoplasmic reticulum Ern1 (ER) to nucleus 4.618 1.47E-178 signalling 1 reticuloendotheliosis Rel 4.596 1.88E-91 oncogene interleukin 27 receptor, Il27ra 4.548 7.90E-31 alpha Pde11a phosphodiesterase 11A 4.529 1.45E-20 a disintegrin and metallopeptidase Adam12 4.500 7.21E-28 domain 12 (meltrin alpha) nuclear receptor Nrip1 4.468 2.36E-92 interacting protein 1 carnosine N- Carnmt1 4.441 2.10E-95 methyltransferase 1 Ltb lymphotoxin B 4.370 4.44E-20 Hoxd4 homeobox D4 4.363 7.95E-23 interferon regulatory Irf4 4.243 3.52E-13 factor 4 Cdh20 cadherin 20 4.155 8.54E-09 collapsin response Crmp1 4.137 3.72E-11 mediator protein 1 DOT1-like, histone H3 Dot1l methyltransferase (S. 4.049 1.95E-166 cerevisiae) tudor domain Tdrd9 4.043 2.13E-33 containing 9 Ly75 lymphocyte antigen 75 4.011 3.02E-48 Gm10804 predicted gene 10804 3.994 3.07E-50 108

nanos homolog 1 Nanos1 3.977 2.71E-92 (Drosophila) RIKEN cDNA 0610040F04Rik 3.960 3.79E-63 0610040F04 gene sodium channel, Scn11a voltage-gated, type XI, 3.937 4.40E-12 alpha RIKEN cDNA 2900057B20Rik 3.871 1.32E-13 2900057B20 gene suppression inducing Sit1 transmembrane adaptor 3.837 4.06E-11 1 Scg5 secretogranin V 3.829 6.99E-35 ABI gene family, Abi3bp member 3 (NESH) 3.810 8.85E-20 binding protein Apol9a apolipoprotein L 9a 3.772 1.41E-09 contactin associated Cntnap2 3.711 1.30E-72 protein-like 2 capping protein Carmil2 regulator and myosin 1 3.690 4.34E-18 linker 2 family with sequence Fam129a similarity 129, member 3.662 2.51E-64 A NLR family, CARD Nlrc3 3.652 1.54E-23 domain containing 3 zinc finger, DHHC Zdhhc23 3.645 6.86E-67 domain containing 23 Podnl1 podocan-like 1 3.624 8.19E-07 interleukin 12 receptor, Il12rb1 3.620 6.76E-10 beta 1 aryl hydrocarbon Arntl receptor nuclear 3.603 6.35E-30 translocator-like RAR-related orphan Rorc 3.599 1.93E-15 receptor gamma Lta lymphotoxin A 3.548 6.71E-12 Rbl2 retinoblastoma-like 2 3.539 2.17E-68 RIKEN cDNA 4932438A13Rik 3.537 2.91E-53 4932438A13 gene protein tyrosine Ptprv phosphatase, receptor 3.524 1.28E-12 type, V mex3 RNA binding Mex3c 3.519 3.50E-105 family member C abhydrolase domain Abhd15 3.512 1.13E-22 containing 15 Apol7b apolipoprotein L 7b 3.452 2.15E-11 109

FBJ osteosarcoma Fos 3.439 1.19E-33 oncogene SLAIN motif family, Slain1os member 1, opposite 3.424 6.59E-38 strand junction-mediating and Jmy 3.401 9.04E-71 regulatory protein Gm13363 predicted gene 13363 3.373 2.24E-46 Galr2 galanin receptor 2 3.368 1.62E-17 potassium voltage- gated channel, shaker- Kcna3 3.351 7.74E-20 related subfamily, member 3 oxidative stress Oser1 3.343 8.49E-67 responsive serine rich 1 RIKEN cDNA A230020J21Rik 3.290 2.26E-51 A230020J21 gene yippee-like 4 Ypel4 3.267 1.66E-25 (Drosophila) Hlf hepatic leukemia factor 3.257 2.44E-26 immunoglobulin Igsf11 superfamily, member 3.251 1.30E-24 11 brain expressed, Bean1 associated with Nedd4, 3.234 1.96E-13 1 family with sequence Fam124b similarity 124, member 3.231 2.00E-31 B IKAROS family zinc Ikzf2 3.223 3.26E-22 finger 2 lymphocyte Lax1 transmembrane adaptor 3.214 8.36E-08 1 family with sequence Fam13b similarity 13, member 3.197 4.32E-102 B ankyrin repeat domain Ankrd53 3.174 3.17E-11 53 zinc finger homeobox Zfhx2os 3.159 3.22E-37 2, opposite strand RAS protein activator Rasal3 3.118 5.76E-14 like 3 Fcho1 FCH domain only 1 3.117 9.48E-11 ATP-binding cassette, Abca2 sub-family A (ABC1), 3.097 8.11E-36 member 2 Mir7077 microRNA 7077 3.085 0.037184 phosphoglucomutase 2- Pgm2l1 3.081 5.87E-45 like 1 110

nicotinamide riboside Nmrk2 3.074 1.71E-05 kinase 2 ubiquitin specific Usp44 3.071 3.95E-16 peptidase 44 RAS guanyl releasing Rasgrp1 3.070 2.40E-13 protein 1 Rimbp2 RIMS binding protein 2 3.056 4.70E-28 The table shows genes that were included in the 17 T cell signature. Genes that were among the top 2000 most highly expressed genes in isolated 17 T cells are bolded. *Log2FC is reported from the comparison between isolated 17 T cells to whole lungs from allergen-treated mice. †FDR-adjusted P-values.

111

4.3.3–Determining the relationships between modules associated with protection and the 17 T cell gene signature

Next, to assess whether T cells are associated with asthma-protective phenotypes in our model, I used a hypergeometric test to examine whether genes in the T cell signature were enriched within the gene modules associated with airway and immune phenotypes linked to protection (Table 4.1). Of the 12 modules of co-regulated genes, five were significantly enriched for T cell signature genes (Table 4.3). Notably, the magenta module (the one associated with decreased AHR, decreased BAL eosinophilia and increased expression of

V4: Table 4.1), was significantly enriched for the T cell gene signature, suggesting that this protection-associated module is either directly influenced by genes expressed by T cells or is influenced by other co-regulated networks that involve T cells. Further analyses therefore focused on the genes in the magenta module (Table 4.4). Interestingly, this module was enriched for genes within the TNFR1 and NF-κB signaling pathways (e.g., Irak3 and Map3k) (Table

4.5), a finding that further highlights the role of innate immunity in the response elicited by the asthma-protective Amish environment. In combination, these findings suggest that genes expressed by 17 T cells are members of innate immune gene networks associated with airway and immune phenotypes related to asthma protection.

112

Table 4.3. Modules enriched in 17 T cell signature genes No. Genes No. 17 T Module P-value* in Module genes in Module (%) TURQUOISE 2306 0 (0) 1 BLUE 1899 26 (1.4) 0.048 BROWN 1227 0 (0) 1 YELLOW 386 0 (0) 1 GREEN 333 10 (3.0) 0.008 RED 290 0 (0) 1 BLACK 253 0 (0) 1 PINK 250 37 (14.8) 4.63E-37 MAGENTA 198 10 (5.0) 8.61E-05 PURPLE 149 9 (6.0) 5.58E-05 GREENYELLOW 100 0 (0) 1 TAN 95 0 (0) 1 The table shows enrichment of 17 T cell signature genes in each module by hypergeometric test. P-values < 0.05 indicate enrichment for 17 T cell gene signature. *Bonferroni-corrected for multiple testing.

Table 4.4. Magenta module genes and their module membership Gene Symbol Gene Name MM Dock8 dedicator of cytokinesis 8 0.963 Irak3 interleukin-1 receptor-associated kinase 3 0.934 Wipf1 WAS/WASL interacting protein family, member 1 0.932 Pi4ka phosphatidylinositol 4-kinase, catalytic, alpha polypeptide 0.929 Abr active BCR-related gene 0.916 Rfx7 regulatory factor X, 7 0.912 Stk24 serine/threonine kinase 24 0.912 nuclear factor of activated T cells, cytoplasmic, Nfatc1 0.909 calcineurin dependent 1 Rel reticuloendotheliosis oncogene 0.906 nuclear factor of activated T cells, cytoplasmic, Nfatc3 0.902 calcineurin dependent 3 Arhgap17 Rho GTPase activating protein 17 0.900 Map3k1 mitogen-activated protein kinase kinase kinase 1 0.898 Plxnc1 plexin C1 0.891 Pik3ap1 phosphoinositide-3-kinase adaptor protein 1 0.891 Arhgap26 Rho GTPase activating protein 26 0.890 N-acetylglucosamine-1-phosphate transferase, alpha and Gnptab 0.888 beta subunits Clec16a C-type lectin domain family 16, member A 0.887 113

Ncoa7 nuclear receptor coactivator 7 0.883 Asxl1 additional sex combs like 1 0.882 Ccdc88c coiled-coil domain containing 88C 0.875 Ank progressive ankylosis 0.872 Mgat5 mannoside acetylglucosaminyltransferase 5 0.865 Ap3b1 adaptor-related protein complex 3, beta 1 subunit 0.864 Ly75 lymphocyte antigen 75 0.862 Papd4 PAP associated domain containing 4 0.862 Ldah lipid droplet associated hydrolase 0.860 Gramd1b GRAM domain containing 1B 0.859 Cnnm2 cyclin M2 0.859 Wdfy2 WD repeat and FYVE domain containing 2 0.857 Igf2r insulin-like growth factor 2 receptor 0.854 Mbp myelin basic protein 0.851 Pabpc1 poly(A) binding protein, cytoplasmic 1 0.851 Tyk2 tyrosine kinase 2 0.850 Fyco1 FYVE and coiled-coil domain containing 1 0.850 R3hdm1 R3H domain containing 1 0.849 Zcchc11 zinc finger, CCHC domain containing 11 0.846 Arsb arylsulfatase B 0.846 Rbl2 retinoblastoma-like 2 0.842 Clint1 clathrin interactor 1 0.841 Synj2 synaptojanin 2 0.840 Tram2 translocating chain-associating membrane protein 2 0.839 Cep192 centrosomal protein 192 0.835 Wdr7 WD repeat domain 7 0.835 Bach2os BTB and CNC homology 2, opposite strand 0.833 Whsc1l1 Wolf-Hirschhorn syndrome candidate 1-like 1 (human) 0.831 Lrrc8d leucine rich repeat containing 8D 0.830 Hs6st1 heparan sulfate 6-O-sulfotransferase 1 0.828 1700110K17Rik RIKEN cDNA 1700110K17 gene 0.828 Med15 mediator complex subunit 15 0.825 Kdm2b lysine (K)-specific demethylase 2B 0.825 Foxred2 FAD-dependent oxidoreductase domain containing 2 0.824 Cic capicua transcriptional repressor 0.824 Tank TRAF family member-associated Nf-kappa B activator 0.823 Tlr9 toll-like receptor 9 0.823 Atp10d ATPase, class V, type 10D 0.822 Lpar5 lysophosphatidic acid receptor 5 0.822 Wdr81 WD repeat domain 81 0.820 Mtss1 metastasis suppressor 1 0.820 Rap1gds1 RAP1, GTP-GDP dissociation stimulator 1 0.819 Zfp362 zinc finger protein 362 0.819 epidermal growth factor receptor pathway substrate 15- Eps15l1 0.818 like 1 114

Ambra1 autophagy/beclin 1 regulator 1 0.817 Cherp calcium homeostasis endoplasmic reticulum protein 0.816 Pde3b phosphodiesterase 3B, cGMP-inhibited 0.815 Bmp2k BMP2 inducible kinase 0.812 neutral sphingomyelinase (N-SMase) activation Nsmaf 0.812 associated factor Ttc39b tetratricopeptide repeat domain 39B 0.811 Sesn3 sestrin 3 0.811 Scaf4 SR-related CTD-associated factor 4 0.810 Man1c1 mannosidase, alpha, class 1C, member 1 0.810 Rab43 RAB43, member RAS oncogene family 0.809 Dgkd diacylglycerol kinase, delta 0.809 Myo9b myosin IXb 0.807 SWI/SNF related, matrix associated, actin dependent Smarca4 0.807 regulator of chromatin, subfamily a, member 4 Atad2b ATPase family, AAA domain containing 2B 0.805 Cabin1 calcineurin binding protein 1 0.805 Diaph1 diaphanous related formin 1 0.805 Naip6 NLR family, apoptosis inhibitory protein 6 0.803 Cd33 CD33 antigen 0.802 Smad3 SMAD family member 3 0.800 Dpy19l1 dpy-19-like 1 (C. elegans) 0.800 Slain1 SLAIN motif family, member 1 0.799 Arhgap39 Rho GTPase activating protein 39 0.798 UDP-N-acetyl-alpha-D-galactosamine:polypeptide N- Galnt10 0.797 acetylgalactosaminyltransferase 10 Arhgap15os Rho GTPase activating protein 15, opposite strand 0.797 Pip4k2b phosphatidylinositol-5-phosphate 4-kinase, type II, beta 0.797 Itpr3 inositol 1,4,5-triphosphate receptor 3 0.795 Slx4ip SLX4 interacting protein 0.794 Ulbp1 UL16 binding protein 1 0.794 Atg2a autophagy related 2A 0.794 Chd2 chromodomain helicase DNA binding protein 2 0.792 ribonuclease L (2', 5'-oligoisoadenylate synthetase- Rnasel 0.790 dependent) Prox1 prospero homeobox 1 0.789 Sfmbt1 Scm-like with four mbt domains 1 0.788 Clcn5 chloride channel, voltage-sensitive 5 0.784 pleckstrin homology domain containing, family M (with Plekhm1 0.783 RUN domain) member 1 Fut8 fucosyltransferase 8 0.782 Gm21188 predicted gene, 21188 0.781 Atp10a ATPase, class V, type 10A 0.781 Whrn whirlin 0.779 2310035C23Rik RIKEN cDNA 2310035C23 gene 0.779 Fnip2 folliculin interacting protein 2 0.778 115

Cnot6l CCR4-NOT transcription complex, subunit 6-like 0.777 Themis2 thymocyte selection associated family member 2 0.775 Mroh1 maestro heat-like repeat family member 1 0.775 Uggt1 UDP-glucose glycoprotein glucosyltransferase 1 0.773 Fbrsl1 fibrosin-like 1 0.771 Man2b2 mannosidase 2, alpha B2 0.770 Adcy3 adenylate cyclase 3 0.770 Plagl2 pleiomorphic adenoma gene-like 2 0.769 D030025P21Rik RIKEN cDNA D030025P21 gene 0.769 Zfp384 zinc finger protein 384 0.769 Abca1 ATP-binding cassette, sub-family A (ABC1), member 1 0.768 Dhx9 DEAH (Asp-Glu-Ala-His) box polypeptide 9 0.767 Gm10804 predicted gene 10804 0.766 6530402F18Rik RIKEN cDNA 6530402F18 gene 0.766 Tbc1d2 TBC1 domain family, member 2 0.765 Setd1a SET domain containing 1A 0.764 Zswim6 zinc finger SWIM-type containing 6 0.764 Helz helicase with zinc finger domain 0.763 Khsrp KH-type splicing regulatory protein 0.763 Abhd15 abhydrolase domain containing 15 0.762 Pom121 nuclear pore membrane protein 121 0.760 Ell elongation factor RNA polymerase II 0.758 Glcci1 glucocorticoid induced transcript 1 0.756 5031439G07Rik RIKEN cDNA 5031439G07 gene 0.755 Dvl3 dishevelled segment polarity protein 3 0.754 Zmynd8 zinc finger, MYND-type containing 8 0.752 Dcaf5 DDB1 and CUL4 associated factor 5 0.752 Srpk1 serine/arginine-rich protein specific kinase 1 0.751 Samd8 sterile alpha motif domain containing 8 0.751 Kdm4a lysine (K)-specific demethylase 4A 0.750 Fam102a family with sequence similarity 102, member A 0.749 Gramd4 GRAM domain containing 4 0.745 Tgfbr1 transforming growth factor, beta receptor I 0.745 Sufu suppressor of fused homolog (Drosophila) 0.743 Zbtb25 zinc finger and BTB domain containing 25 0.742 Slain1os SLAIN motif family, member 1, opposite strand 0.741 Ubap2 ubiquitin-associated protein 2 0.738 Chsy1 chondroitin sulfate synthase 1 0.737 Cdk5rap2 CDK5 regulatory subunit associated protein 2 0.737 Susd6 sushi domain containing 6 0.737 Pde7a phosphodiesterase 7A 0.736 Sirpb1a signal-regulatory protein beta 1A 0.736 Slc30a7 solute carrier family 30 (zinc transporter), member 7 0.735 Il16 interleukin 16 0.735 Nfam1 Nfat activating molecule with ITAM motif 1 0.734 116

Dhx37 DEAH (Asp-Glu-Ala-His) box polypeptide 37 0.734 Sh2b1 SH2B adaptor protein 1 0.733 Ncor1 nuclear receptor co-repressor 1 0.733 Bsn bassoon 0.730 Cyth1 cytohesin 1 0.728 Cnr2 cannabinoid receptor 2 (macrophage) 0.728 Pcbp2 poly(rC) binding protein 2 0.727 Siglec1 sialic acid binding Ig-like lectin 1, sialoadhesin 0.725 Htt huntingtin 0.722 Golm1 golgi membrane protein 1 0.721 Pak2 p21 protein (Cdc42/Rac)-activated kinase 2 0.719 Ubap2l ubiquitin-associated protein 2-like 0.718 Mfsd12 major facilitator superfamily domain containing 12 0.717 Sik3 SIK family kinase 3 0.714 Axl AXL receptor tyrosine kinase 0.711 Ttf2 transcription termination factor, RNA polymerase II 0.707 Bahd1 bromo adjacent homology domain containing 1 0.706 Rnf216 ring finger protein 216 0.705 Gm9961 predicted gene 9961 0.704 Coro7 coronin 7 0.703 Sulf2 sulfatase 2 0.703 AI504432 expressed sequence AI504432 0.702 Zbtb7a zinc finger and BTB domain containing 7a 0.702 Reln reelin 0.700 Gns glucosamine (N-acetyl)-6-sulfatase 0.698 Lclat1 lysocardiolipin acyltransferase 1 0.697 Zdhhc9 zinc finger, DHHC domain containing 9 0.697 Sgsh N-sulfoglucosamine sulfohydrolase (sulfamidase) 0.696 Pi4kb phosphatidylinositol 4-kinase, catalytic, beta polypeptide 0.695 Ffar1 free fatty acid receptor 1 0.691 Sirpb1b signal-regulatory protein beta 1B 0.687 Myo7a myosin VIIA 0.687 leucine-rich repeats and calponin homology (CH) domain Lrch4 0.687 containing 4 4930487H11Rik RIKEN cDNA 4930487H11 gene 0.687 Prrc2a proline-rich coiled-coil 2A 0.684 ATP-binding cassette, sub-family C (CFTR/MRP), Abcc1 0.683 member 1 Gm13807 predicted gene 13807 0.683 Tpbg trophoblast glycoprotein 0.682 Rhoq ras homolog family member Q 0.679 Skp2 S-phase kinase-associated protein 2 (p45) 0.676 Casp8 caspase 8 0.671 Acot11 acyl-CoA thioesterase 11 0.662 Ppp2r5c protein phosphatase 2, regulatory subunit B', gamma 0.662 117

Cbx4 chromobox 4 0.659 Ints1 integrator complex subunit 1 0.657 Nbeal2 neurobeachin-like 2 0.657 Lrp10 low-density lipoprotein receptor-related protein 10 0.656 Crlf3 cytokine receptor-like factor 3 0.653 solute carrier family 16 (monocarboxylic acid Slc16a10 0.652 transporters), member 10 Adam10 a disintegrin and metallopeptidase domain 10 0.644 Usp44 ubiquitin specific peptidase 44 0.636 Genes that are members of the 17 T cell signature are bolded.

Table 4.5. Enrichment of magenta module genes for canonical pathways

-LOG CANONICAL PATHWAYS MOLECULES (P-VALUE) TANK, NAIP1, PAK2, MAP3K1, TNFR1 SIGNALING 4.34 CASP8 ANTIPROLIFERATIVE ROLE IN 4.32 PABPC1, TGFBR1, SMAD3, SKP2 T CELL SIGNALING REGULATION OF IL-2 EXPRESSION IN ACTIVATED TGFBR1, NFATC3, SMAD3, 3.32 AND ANERGIC T MAP3K1, NFATC1 LYMPHOCYTES TANK, TGFBR1, MAP3K1, IRAK3, NF-ΚB SIGNALING 3.23 CASP8, TLR9, IGF2R SYNJ2, NFATC3, TYK2, TLR9, IL-4 SIGNALING 3.11 NFATC1 REGULATION OF ACTIN-BASED WIPF1, RHOQ, PAK2, PIP4K2B, 3.09 MOTILITY BY RHO PI4KA The table was created using Ingenuity Pathway Analysis.

118

4.4 – DISCUSSION

Our dataset on the lung transcriptome of two different strains of mice treated with different allergens and dust extracts provides a unique foundation to begin understanding the cellular and molecular events associated with asthma protection. While combining distinct experimental models may have introduced heterogeneity into our study, our aim was to identify a core of molecular mechanisms independent of model characteristics. Importantly, only one transcriptional module (magenta) was associated with inhibition of both AHR and lung eosinophilia. This module and the genes it includes is therefore a compelling candidate for future studies of the mechanisms underlying the protection from asthma conferred by Amish dust extracts. At the same time, the fact that other transcriptional modules were enriched in 17 T cell signature genes highlights the complex roles that these cells likely play in the response of the lung to environmental exposures.

Pathway analysis of the magenta module showed that genes expressed by

17 T cells are involved in networks important for innate immune signaling.

These data are consistent with our previous findings that peripheral blood leukocytes from Amish children had increased expression of innate immunity genes when compared to Hutterite children [147, 149, 168].

The work presented in this chapter is the first to explore the role of 17 T cell-associated genes in the protection from experimental asthma and was meant to be exploratory. As such, in its current version, it has some clear limitations.

While our study was designed to identify model-independent mechanisms of

118

119 asthma protection, mouse strains were not equally represented in our data set.

Therefore, we cannot exclude the possibility that inter-strain genetic differences influenced our findings. To address this issue, future analyses will be performed separately in a larger set of BALB/c mice (the strain from which T cells were originally isolated) and in C57BL/6 mice and their T cells. Moreover, I used an extremely high threshold of differential expression (log2FC > 3) to identify genes that are specifically expressed in 17 T cells compared to other lung cells (section 4.3.2). This approach was rigorous in its quest for specificity but likely excluded genes that are robustly, albeit less uniquely, expressed in 17

T cells and may play important functional roles in protective responses. In the future, I will focus on highly and moderately expressed genes from isolated

T cells defined by the upper and middle tertiles of expression and I will incorporate these genes into our definition of the T cell signature. The granularity of these comparisons will also be decisively enhanced by adding samples from lungs depleted of T cells after exposure to Amish dust extracts. This experimental condition will allow us to separate T and non-

T cell genes elicited by the same stimulus.

Finally, we acknowledge that at this stage our results are suggestive but remain descriptive. However, this work can provide the rationale for further hypothesis-driven studies in which mouse models will be used to assess the role of candidate T cell genes in asthma protection. This last step, while critical to confirm our findings, is beyond the scope of this dissertation.

119

120

CHAPTER 5: CONCLUSIONS AND FUTURE DIRECTIONS

5.1 – CONCLUSIONS

This dissertation presents the results of an integrated approach aimed at characterizing processes involved in the inception and pathogenesis of childhood asthma. We showed for the first time that epigenetic (DNA methylation) profiles in human immune cells at birth are associated with both risk for asthma during childhood (Chapter 2) and the maternal prenatal immune milieu (Chapter 3), and these relations are influenced by maternal asthma. On the other hand, our studies in mouse models of asthma highlighted a novel player in asthma pathogenesis by suggesting that 17 T cells may be involved in the protection from asthma conferred by inhalation of dust from traditional, microbe-rich farm environments

(Chapter 4).

Chapter 2 focused on asthma inception and showed that epigenetic signatures at birth are associated with asthma during the first decade of life. I also showed that neonatal SMAD3 promoter methylation is higher in children who will be diagnosed with asthma by age 9 compared to children who will not, and this relation occurs selectively in children of asthmatic mothers. These findings highlight an epigenetic mechanism that may potentially contribute to the increased risk of asthma found in some children of asthmatic mothers. Moreover, my work identified SMAD3 and IL-1 as key mediators of the trajectory to asthma during childhood. SMAD3, the master regulator of TGF-signaling, guides the development of both asthma-protective Treg cells and asthma-

120

121 promoting Th17 cells by activating the FOXP3 [169] and suppressing the RORC

[170] transcriptional programs (Figure 5.1). Importantly, Treg and Th17 alterations have been reported in childhood asthma [15, 41]. My data emphasize the functional connection between SMAD3 and IL-1a key asthma mediator in both children [120] and adults [119, 121]. Indeed, in neonates who became asthmatic by age 9, SMAD3 promoter hypermethylation (an epigenetic configuration consistent with low SMAD3 expression) was strongly associated with high IL-1production. The latter is expected to destabilize the Treg program, enhance inflammation, and promote Th17 differentiation [130], ultimately favoring the development of asthma (Figure 5.1A). On the other hand, the fact that neonatal SMAD3 methylation and IL-1 production are associated with asthma risk only among children of asthmatic mothers suggests that the in utero milieu critically influences the epigenetic trajectory towards childhood asthma.

The role of the in utero environment in the child’s trajectory to asthma is further supported by the data presented in Chapter 3, which again focused on epigenetic profiles at birth but looked back in time towards maternal influences.

In non-asthmatic mothers, the prenatal immune profile (i.e., the ratio between

IFN and IL-13 secreted by mitogen-stimulated PBMCs isolated during the third trimester of pregnancy) was strongly associated with both decreased risk for asthma in their children [104] and neonatal immune cell DNA methylation at

>500 CpG sites [171]. No such relation was detected in asthmatic mothers. These results not only reiterated the importance of the maternal asthma status in the

121

122 child’s trajectory to asthma, but also highlighted once again the potential role of the TGF-pathway. Indeed, neonatal differential methylation associated with the maternal prenatal immune profile clustered in pathways regulated by TGFB1

(Figure 5.1B). Because TGF- is critical not only for immunoregulation but also for lung development [146], this pathway may broadly impact both the immune and the respiratory components of the asthma trajectory.

Finally, in Chapter 4 my work merged with a unique project our laboratory has been pursuing for several years and took a genomic approach to explore how environmental exposures modify asthma risk. To this end I relied on our group’s demonstration that (1) exposure to traditional farming (such as the one practiced by the Indiana Amish) protects almost completely from asthma and allergic sensitization at the population level [105]; (2) this protection can be transferred to allergen-treated mice by exposing them to inhaled dust extracts from Amish farms [105, 106]; and (3) induction of asthma protection in mice treated with Amish farm dust extracts is almost invariably accompanied by the recruitment to the lung of a large population of 17 T cells (Pivniouk, in preparation). By analyzing the transcriptome of lung cells isolated from mice exposed to allergen with or without Amish dust extracts, I found that genes associated with asthma protection cluster in a few co-regulated networks, some of which are enriched for genes expressed by 17 T cells (Figure 5.1C).

Overall these findings suggest that distinct trajectories to asthma may share underlying pathways (e.g., the TGF- pathway), but the ways in which these pathways are altered may depend on key exposures during critical windows

122

123 of development (Figure 5.1). Indeed, the fact that many differentially methylated genes associated with the prenatal immune profile cluster in pathways regulated by TGFB1 (Chapter 3) is suggestive of alterations the modalities of which will be examined in further studies. At the same time, our results in Chapter 2 raise the possibility that the TGF- pathway be altered in asthmatics born to asthmatic mothers through hypermethylation of the SMAD3 promoter, which would indirectly favor the RORC transcriptional program. Activation of this program may have broad biological implications. At least three cell types express RORC and produce IL-17: adaptive Th17 cells and innate ILC type-3 and 17 T cells.

The involvement of TGF- signaling in promoting Th17 differentiation through the RORC transcriptional program is well characterized, but the role of TGF- in the differentiation of 17 T cells is unclear and will be worth investigating in light of the results presented in this dissertation.

123

124

Figure 5.1. Working model: the relationship between TGF- signaling and

potential trajectories to asthma. A) SMAD3 promoter hypermethylation in

asthmatic neonates born to asthmatic mothers may destabilize the Treg program

(FOXP3), thereby promoting the Th17 differentiation (RORC). This scenario would lead to increased inflammation (including increased production of IL-1)

and asthma development. B) In children of non-asthmatic mothers, the maternal

prenatal immune profile (marked by the IFN/IL-13 ratio) is associated with

neonatal DNA methylation in genes regulated by TGFB1. How these genes

impact on downstream asthma-related events remains to be determined. C)

Exposure to asthma-protective farm environments activates a population of

RORC-expressing17 T cells that might be TGF--dependent. Red arrows

represent mechanisms associated with increased asthma risk; blue arrows

represent mechanisms associated with decreased asthma risk.

124

125

5.2 – FUTURE DIRECTIONS

In Chapter 2, I showed that SMAD3 promoter hypermethylation is associated with increased LPS-induced production of IL-1 selectively in children of asthmatic mothers. These findings suggest a possible mechanism underlying this path to asthma. Future studies will focus on two main areas: an extensive follow-up in children, particularly those born to asthmatic mothers, and the implications of altered SMAD3 methylation.

Essential information will be obtained from a large birth cohort carefully phenotyped for asthma and related traits at least through age 5 (the earliest time at which a robust asthma diagnosis can be made). For these studies we will compare asthmatic and non-asthmatic children with asthmatic or non-asthmatic mothers, and we will further characterize the association between SMAD3 promoter methylation and risk for asthma in these different groups of subjects. The most urgent questions to address, and the methods to address them, include:

1. What cell types carry asthma-associated SMAD3 promoter

hypermethylation?

Method: We will assess the DNA methylation profile of the SMAD3

DMR in CBMCs as well as isolated T cell subsets and monocytes. To this

end we will use bisulfite sequencing and/or extract relevant SMAD3

promoter CpG DNA methylation data from the Illumina450K or EPIC

arrays.

2. Does SMAD3 promoter methylation correlate with gene expression?

125

126

Method: SMAD3 expression will be measured using qRT-PCR in CBMCs

and isolated T cell subsets and monocytes and correlated with the SMAD3

methylation profiles characterized as described above.

3. Does SMAD3 promoter hypermethylation detected at birth persist? How

long? Does the relationship between SMAD3 methylation and gene

expression change over time?

Method: SMAD3 DNA methylation profiling and SMAD3 expression will

be assessed longitudinally from birth to age 60 months as described above.

These profiles will be measured in isolated T cell subsets and monocytes

from CBMCs and PBMCs.

4. Does genetic variation at nearby SNPs influence the methylation at the

SMAD3 promoter?

Method: SMAD3 SNPs (prioritized to asthma-associated loci from

GWAS) will be genotyped in individuals from our birth cohort and will be

tested for associations with DNA methylation levels at individual CpGs.

We will use mediation analysis to ask whether genetic risk for asthma is

mediated by methylation at the SMAD3 promoter. Again, these studies

will be performed in isolated T cell subsets and monocytes from CBMCs

and PBMCs.

Once we determine which cell types express SMAD3 and carry asthma- associated SMAD3 promoter hypermethylation, we will ask what effects decreased SMAD3 expression has on its downstream transcriptional targets. This work will rely on siRNA-mediated knock down (KD) of SMAD3 expression in

126

127 unstimulated and LPS-stimulated CBMC subsets to characterize differences in

DNA methylation profiles, transcription programs, and SMAD3 transcription factor occupancy compared to mock-treated cells. These experiments will also better characterize the functional relationship between SMAD3 and IL-1. The key questions to address are:

1. When SMAD3 is present in limiting amounts, where in the genome

does SMAD3 occupancy change? What genes are differentially

expressed? How do methylation profiles change?

Method: We will use ChIP-seq, RNA-seq, and IlluminaEPIC arrays to

profile SMAD3 transcription factor occupancy, epigenome-wide DNA

methylation, and transcriptome profiles, respectively, in the relevant

cell types that carry asthma-associated SMAD3 promoter

hypermethylation. We will compare profiles of occupancy,

methylation, and transcriptome between SMAD3 KD and mock-treated

control cells.

2. Does LPS-stimulated IL-1 production differ in cells with decreased

SMAD3 expression compared to normal cells?

Method: To address this question we will compare LPS-stimulated

SMAD3 KD to LPS-stimulated mock-treated cells. Because IL-1 may

or may not be produced by the same cells carrying asthma-associated

SMAD3 hypermethylation, siRNA and LPS treatments will be

conducted in both isolated cells and CBMC samples, thereby allowing

127

128

interactions between distinct cell types. IL-1 will be measured in cell

supernatants using ELISA.

In combination, the proposed studies will help clarify, on a mechanistic level, the relationships between SMAD3 promoter hypermethylation, LPS-induced

IL-1 production, and risk for asthma in the child in relation to maternal asthma status. This work will provide a deeper, longitudinal characterization of the relationship between SMAD3 promoter hypermethylation and its impact on the expression of SMAD3 and its downstream targets. The fact that SMAD3 is associated with asthma in several GWAS also makes it a good candidate to further investigate the relationship between genetic variation and DNA methylation profiles [126].

The study presented in Chapter 3 was designed, in part, to provide preliminary data and an analytical blueprint for a grant application that has now been funded (R21AI133765). With this additional support, we are extending our analysis of the neonatal methylome to the entire IIS population (n=234 samples) and we will further explore the questions raised in Chapter 3. Specifically, we will examine the relationships among maternal immune IFN/IL-13 profiles during pregnancy, neonatal epigenome, and early life microbiome (in both environmental and nasal samples). The latter work will per performed in collaboration with Dr.

Susan Lynch’s laboratory at the University of California San Francisco.

Specifically, we will explore these relationships to better understand how they act individually and in combination to alter the trajectory to childhood asthma and the

128

129 asthma-related phenotypes that have been carefully documented in the IIS population. For example, this larger dataset will allow us to assess to what extent the relationship between the maternal IFN/IL13 ratio and childhood asthma is mediated by neonatal DNA methylation. If we find that both the maternal prenatal immune profile and childhood asthma are associated with differential methylation at a subset of CpG sites, we will use mediation analysis to explore causal relationships among these variables.

In addition, this larger dataset will be analyzed on MethylationEPIC arrays, which were designed to analyze 850,000 CpGs, including more than 90% of the CpGs from the 450K arrays and an additional 413,745 CpGs [172]. This array drastically expands the coverage provided by its predecessor by incorporating more than 300,000 CpGs within regulatory elements located in enhancer regions identified from the ENCODE and FANTOM5 databases.

Importantly, while the 450K arrays were designed largely by for use in cancer and aging studies, the expanded scope of the EPIC arrays allows them to provide more insights into regulatory processes [172]. At the same time, recent work demonstrated a high correlation among replicates measured on both the 450K and

EPIC arrays, suggesting that the data collected on these platforms are robust.

Because our preliminary data suggest that TGFB1 regulates genes containing differential methylation associated with the maternal prenatal immune profile, we will focus some of our analyses on pathways regulated by TGFB1 in order to minimize the multiple testing issue.

129

130

In Chapter 4, my work showed for the first time that the protection from asthma conferred by Amish farm dust exposure likely involves 17 T cells carrying a V4 TCR. Because  T cell function appears to depend on the expressed TCR, and transcriptome profiles of V4+ T have not been extensively described, it is worth comparing V4+ to V4-  T cells, which represent 2-5% of the  T cells found in the lungs of mice treated with Amish dust extract. After isolating  T cells from lungs of mice treated with Amish dust extract by magnetic sorting with a pan-anti-mAb, I will use the same technology to further separate V4+ and V4-  T cells and characterize their transcriptome.

This will allow me to identify genes and networks thereof that are truly unique to

17 T cells compared to other  T cell subset. While some preliminary work has characterized the transcriptome of immature V4+ cells from the fetal thymus using microarrays, adult V4+ cells were not examined ([173], https://www.immgen.org/). Moreover, these data are only available for C57BL/6 mice. Therefore, our work (which will be performed primarily in BALB/c mice) will fill a current gap in knowledge because V4+ cells from this strain have not been characterized in the context of asthma-protective responses. We will also compare the transcriptome of unfractionated lungs treated with Amish dust extracts to lung cells depleted of 17 T cells from the same mice. This additional comparison will provide the granularity required to define 17 T cell and non-

17 T cell genes without using extreme thresholds for differential expression.

130

131

We will further investigate the pathways in protection-associated gene modules to generate hypotheses about causal mechanisms, as in the examples given in Chapter 4. A specific area of focus will be to confirm that 17 T cells are critical for protection. In this respect, it has been shown that Blk plays an important, unexpected role in the development and function of V4+ T cells

[163]. Blk-deficient mice offer a powerful tool to investigate the role of 17 T cells in asthma protection because Blk function is dispensable for B cells and these mice have no overt defects in immunity [174]. For this experiment, Blk- deficient mice will be compared to both Tcrd-deficient and wild-type C57BL/6 mice, with the expectation that Blk-deficient mice treated with Amish dust extract will be no longer or less protected from AHR or lung eosinophilia. Because we are interested in characterizing the impact of selective 17 T cells depletion on our asthma protection model and Tcrd-deficient mice will lack all  T cell subsets regardless of their opposing roles in experimental models of asthma [147,

149-152], Tcrd-deficient mice will serve as an additional control rather than our primary interest group.

A final area of research will focus on assessing whether 17 T cells in mice treated with Amish dust extract are TGF-dependent, since this pathway appears to be a target of epigenetic mechanisms that influence asthma risk

(Chapters 2 and 3). While TGF-has been suggested to be important for the development of 17 T cells [175], it is thought that 17 T cells can either develop and gain their effector functions in the thymus as natural TGF-- dependent 17 T cells, or they can remain in a naïve state in the periphery and

131

132 can gain effector function upon antigen stimulation [175]. These studies will involve specific knock-out models (including Smad3 knock-out mice [176]) and a detailed time course characterization of how the 17 T cell response to Amish dust extract treatment develops over time.

Overall these experiments may provide novel mechanistic insights about the involvement of the TGF- pathway and 17 T cells in the trajectory to childhood asthma and the protection from this disease elicited by exposure to environments rich in microbes.

132

133

APPENDIX: SUPPLEMENTARY MATERIALS FOR CHAPTER 2

Epigenome-wide Analysis Links SMAD3 Methylation at Birth to Asthma in Children of Asthmatic Mothers

A.1 - METHODS

DNA methylation profiling by Methylated CpG Island Recovery Assay (MIRA)- chip High molecular weight genomic DNA was isolated from CBMC using the DNeasy Blood and Tissue Kit (Qiagen) and fragmented with a Bioruptor sonicator (Diagenode) to produce fragments of 200-700 bp (Figure E1, Appendix p. 127). For each sample, a DNA aliquot was set aside to serve as the non- enriched, input DNA fraction. The MethylCollector Ultra Kit (Active Motif) was then used for methyl binding protein-dependent capture (MBDCap) [88] of methylated DNA (enriched fraction). The protocol is based on the MIRA technique [88] and has the advantage of requiring limited amounts of DNA. For each sample, enriched and input DNA underwent one round of whole genome amplification (WGA) using the GenomePlex Whole Genome Amplification Kit (Sigma). The entire output of the MethylCollector capture (for enriched DNA) or 20 ng of input DNA were used as starting material. WGA reactions were run on 1% agarose gels (1x TBE) to assess fragment size distribution and purified with the QiaQuick PCR Purification Kit (Qiagen). The efficiency of methylated DNA enrichment was evaluated by quantitative PCR analysis of three control regions that are typically methylated (XIST and NBR2) and unmethylated (APC) across tissues. Primers are provided with the MethylCollector Kit. WGA-input and WGA-enriched DNA (5 ng/l, 5 l) were amplified using PerfeCTa SYBR Green FastMix ROX (Quanta BioSciences) and optimal cycling conditions. Samples were run in triplicate and their Ct values averaged. Fold-enrichment was calculated as 2(Ct WGA input – Ct WGA

133

134 enriched). All samples met the required quality control criteria (≥3-fold enrichment in XIST and NBR2, methylation depletion in APC). WGA-input and WGA-enriched samples were labeled with Cy3 and Cy5 respectively, and co-hybridized to 2.1M Human Promoter Deluxe microarrays (Roche-NimbleGen). The arrays include 2,137,192 experimental probes [from hg18 and covering ~10 kb of each annotated human gene surrounding the transcription start site, CpG islands and regulatory regions] as well as 6726 positive, 4257 negative, 963 non-CG and 38763 random controls. Hybridization intensities were extracted separately for the Cy3 and Cy5 channels and used for subsequent processing and analysis. Labeling, hybridization, scanning and quality control were performed at Roche-NimbleGen. For each sample, the hybridization intensities of the input and enriched channels were quantile-normalized separately to remove technical noise. Probes mapping to sex chromosomes were excluded from the analysis. Log2 ratios of enriched/input probe intensities were then calculated from the normalized data. DNA methylation microarray data from this publication were submitted to the NCBI Gene Expression Omnibus (GEO) database and assigned the identifier GSE85228.

Technical validation of results from DNA methylation microarrays Results of genome-wide DNA methylation analysis on the NimbleGen platform were validated by bisulfite sequencing of 16 DMRs (300-600 bp) containing a total of 264 CpG sites and representative of a range of CpG densities and DNA methylation levels (Figure E2, Appendix p. 128). After bisulfite conversion (EZ DNA Methylation-Gold Kit, Zymo Research), target regions were amplified with Platinum Taq DNA Polymerase High Fidelity (Invitrogen) using primers designed to amplify exclusively converted DNA (BiSearch, http://bisearch.enzim.hu) (Table E4, Appendix p. 161). PCR products were run on 1% agarose gels and cloned into the pCR4-Topo vector (Invitrogen). After transformation into E.coli One Shot TOP10 competent cells, individual colonies were picked and analyzed. For each sample, 10-25 clones were sequenced with forward or reverse M13 primers. Sequences were edited and manually curated in CodonCode Aligner.

134

135

Percent methylation per CpG site per clone was determined using BISMA (http://biochem.jacobs-university.de/BDPC/BISMA). Bisulfite sequencing data were summarized by the median percentage methylation across the CpG sites in each region for a total of 216 DNA methylation estimates. The corresponding estimates of microarray DNA methylation intensity were derived from the median probe signal intensities (normalized log2 ratios of enriched/input sample) across the same region [177].

Functional DMR annotation DMRs were annotated using data from the hg19 version of the genome in Homer (Hypergeometric Optimization of Motif EnRichment, http://biowhat.ucsd.edu/homer/), which determines the distance to the closest transcription start site, assigns the DMR to that gene, and then determines the genomic annotation of the region occupied by the center of the DMR. DMR coordinates were also used to search for overlaps with DNase I hypersensitive sites identified by the ENCODE project (http://www.genome.ucsc.edu/ENCODE/) in peripheral blood CD14+ monocytes (RO01746) and CD4+ Naive T cells (Wb11970640). For all subsequent gene annotation analyses, DMR coordinates were overlapped with coordinates of RefSeq genes +/- 5 kb, allowing multiple genes to be associated with a single DMR.

DNA methylation analysis in the COAST cohort was performed using the Infinium HumanMethylation450 BeadChip array (Illumina) [135]. Probes located on the sex chromosomes and those that could not be distinguished from the background (detection P-value > 0·01 in 75% of the samples) were removed. Probes were also removed if they mapped to multiple regions of the genome after bisulfite conversion or if they overlapped with the location of known SNPs [136]. This reduced the number of probes from 485,512 to 327,214. Methylation data were processed using the minfi R package [137] and Infinium type I and type II probe bias was corrected using the SWAN algorithm [138]. The raw probe values were then corrected for color imbalance and background by controls

135

136

normalization. The methylation level at each CpG site was reported as a  value (i.e., the fraction of signal obtained from methylated beads over the sum of methylated and unmethylated bead signals), which is interpreted as percent methylation. Principal component analysis was performed to identify chip effects and confounding variables, and the ComBat function in the sva R package was used to adjust for these variables.

136

137

A.2 – SUPPLEMENTARY FIGURES

Figure E1. Workflow of DNA methylation profiling. Genomic DNA was fragmented by sonication and divided into two aliquots: one underwent methyl- binding protein methylation capture (MBDCap) and was enriched for methylated DNA, while the other was left untreated. The two fractions were then whole genome-amplified, fluorescently labeled [captured (enriched)=red, untreated (input)=green], mixed and hybridized to Human DNA Methylation 2.1M Deluxe Promoter Arrays (Roche NimbleGen). After scanning the arrays, the hybridization intensities for each probe were extracted separately for the red and green channels, and were quantile-normalized to remove technical noise. The difference between normalized intensities [log2 enriched – log2 input = log2 (enriched/input), referred to as the log2 ratio] was then calculated and used as a measure of methylation levels. Measurements for 16 genomic regions were validated by bisulfite sequencing. Regions differentially methylated in the asthmatic and non- asthmatic groups were detected using Roche-Nimblegen scripts that implement a probe sliding-window ANOVA. DMRs were functionally annotated using Homer and Ingenuity Pathway Analysis.

137

138

Figure E2. Correlation of bisulfite sequencing and DNA methylation microarray results. Each dot corresponds to the median normalized log2 ratios of enriched/input sample across probes (for microarray data) or the median percentage DNA methylation across CpG sites (for bisulfite sequencing data) for each region in each subject. The correlation between median normalized log2 ratios from microarray analysis and median percent DNA methylation from bisulfite sequencing was assessed by measuring the Spearman correlation coefficient (ρ).

138

139

Figure E3. Distribution of asthma-associated DMRs by chromosome and genomic location. A) Manhattan plot of asthma-associated DMRs. The x axis depicts the chromosomal location of each DMR. The y-axis shows the Benjamini Hochberg (BH)-adjusted P-values from the probe sliding window ANOVA. The minimum value of the y-axis was set at 2 because a significance threshold of 0.01 was imposed on the adjusted P-values. Genes associated with top hits are labeled. B) Genomic location of asthma-associated DMRs. The annotation is based on the center of the DMR, as defined by the Homer package. The genomic distribution of DMRs differed significantly from the distribution of probes across the entire array (2= 19.49, df = 7, P-value = 0.007). Fifty-thousand permutations were performed to assess whether this distribution differed from random expectations (permutation P-value = 0.008). Standardized residuals were computed for each category in order to assess which genomic locations contributed to the overall statistical significance. Absolute values > 1.96 significantly contribute to the overall 2 test statistics. UTR: untranslated region, TSS: transcription start site (- 1kb to +100 bp), TTS: transcription termination site (-100 bp to +1kb).

139

140

Figure E4. Regulatory potential of the SMAD3 DMR. The SMAD3 DMR was mapped to the human genome (hg19/GRCh37) using University of California Santa Cruz (UCSC) Genome Browser tools. Tracks are specified on the left. UCSC genes are shown in blue. Nearby CpG islands (with identifier) are shown in green. DNase I hypersensitive sites (HSS) identified in CD14+ monocytes and CD4+ naïve T cells are shown in red and blue, respectively. Post-translational histone modifications (H3K27Ac, H3K4Me1, and H3K4Me3) identified in CD14+ monocytes, CD4+ naïve T cells, and peripheral blood mononuclear cells (PBMC) (Roadmap Epigenomics Project) mark poised/active regulatory regions [178]. The bottom track shows common SNPs (MAF > 10%) from the HapMap CEU population, none of which are associated with asthma. The genomic region that surrounds the SMAD3 DMR (+/- 100 bp) is highlighted in red. The figure covers 5 kb upstream and downstream of the SMAD3 DMR.

140

141

Figure E5. Association between SMAD3 DNA methylation at birth and childhood asthma in the IIS cohort. SMAD3 methylation was assessed by bisulfite sequencing and expressed as mean percentage DNA methylation across 8 consecutive CpG sites in the SMAD3 DMR. N: non-asthmatic, A: asthmatic. P- value by Wilcoxon two-sample test.

141

142

Figure E6. SMAD3 CpG7 methylation in IIS and MAAS. SMAD3 methylation was assessed by bisulfite sequencing and expressed as percent methylation at SMAD3 CpG7 in neonates from the IIS and MAAS cohorts. N: non-asthmatic, A: asthmatic. P-values by Wilcoxon two-sample test.

efs for tables … take

142

143

Figure E7. Estimation of CBMC composition in the COAST cohort. Proportions of CBMC cell types were estimated using DNA methylation data generated on the Illumina HumanMethylation450 BeadChip and a cord blood reference panel [89]. N: non-asthmatic, A: asthmatic. P-values by Wilcoxon two-sample test.

143

144

Figure E8. Effects of maternal asthma on the association between neonatal SMAD3 methylation and childhood asthma in COAST after adjusting for CBMC composition. Residual methylation values for SMAD3 (cg02486855, Illumina HumanMethylation450 BeadChip) were calculated by adjusting for CBMC composition using linear regression. N: non-asthmatic, A: asthmatic. P-values by Wilcoxon two-sample test.

144

145

A.3 – SUPPLEMENTARY TABLES

Table E1. Characteristics of IIS study subjects.

% (n/N)

Asthma, Maternal Paternal Maternal Paternal Female Household Caucasian ages 2 to 9* asthma† asthma† atopy‡ atopy‡ child smoking at birth child

Discovery Cohort, 50.0 38.9 17.1 80.0 82.4 58.3 14.7 75.0 N=36 (18/36) (14/36) (6/35) (28/35) (28/34) (21/36) (5/34) (27/36)

Additional Subjects included in the Targeted 51.7 31.0 8.0 64.3 75.9 58.6 17.9 51.7 Analysis, (15/29) (9/29) (2/25) (18/28) (20/26) (17/29) (5/28) (15/29) N=29

Remaining IIS subjects, 12.1 16.7 16.5 66.2 77.6 50.4 24.5 (101/412) 56.8 (237/417) N=417 (44/365) (67/402) (60/364) (251/379) (264/340) (210/417)

P value§ <0.001 0.002 0.623 0.238 0.874 0.492 0.363 0.082 * Physician-diagnosed with symptoms or medication use for asthma in the past year reported at least once on the age 2, 3, 5 or 9-year questionnaires † Physician-diagnosed asthma ever, as assessed by questionnaire. ‡ Positive skin test response to any of 17 local aeroallergens shortly after the child's birth. § by Fisher’s Exact test.

145

146

Table E2. Characteristics of non-asthmatic and asthmatic children in the IIS, MAAS and COAST study populations.

% (n/N) Household Maternal Paternal Maternal Paternal Female smoking asthma‖ asthma‖ atopy** atopy** child at birth DISCOVERY COHORT (IIS) NO ASTHMA 44.4 11.1 76.5 94.1 66.7 11.1 (N=18) (8/18) (2/18) (13/17) (16/17) (12/18) (2/18) ASTHMA* 33.3 23.5 83.3 70.6 50.0 18.8 (N=18) (6/18) (4/17) (15/18) (12/17) (9/18) (3/16) P value† N/A †† 0.4 0.69 0.18 0.5 0.65 TARGETED ANALYSIS (IIS)‡ NO ASTHMA 41.9 6.5 70.0 80.0 64.5 12.9 (N=31) (13/31) (2/31) (21/30) (24/30) (20/31) (4/31) ASTHMA* 27.6 16.7 75.0 80.0 44.8 19.2 (N=29) (8/29) (4/24) (21/28) (20/25) (13/29) (5/26) P value† N/A †† 0.39 0.77 >0.99 0.19 0.72 REPLICATION COHORT 1 (MAAS) NO ASTHMA 100 17.6 100 100 64.7 17.6 (N=17) (17/17) (3/17) (17/17) (17/17) (11/17) (3/17) ASTHMA§ 100 15.4 100 92.3 46.2 7.7 (N=13) (13/13) (2/13) (13/13) (12/13) (6/13) (1/13) P value† - >0.99 - 0.43 0.46 0.61 REPLICATION COHORT 2 (COAST) NO ASTHMA 37.5 25.0 100 84.6 50.0 31.3 (N=16) (6/16) (4/16) (16/16) (11/13) (8/16) (5/16) ASTHMA¶ 58.3 25.0 90.9 81.8 41.7 41.7 (N=12) (7/12) (3/12) (10/11) (9/11) (5/12) (5/12) P value† 0.45 >0.99 0.41 >0.99 0.72 0.7 * Physician-diagnosed with symptoms or medication use for asthma in the past year reported at least once on the age 2, 3, 5 or 9-year questionnaires. † by Fisher’s Exact test comparing characteristics of asthmatic and non-asthmatic children separately in each cohort. ‡ Includes 31 individuals from the IIS Discovery Cohort (17 non-asthmatics, 14 asthmatics). § At least one of the following criteria reported on age 5 or 8 questionnaires: (1) physician diagnosis of asthma, (2) the use of asthma medications during the previous 12 months. ¶ At least one of the following criteria reported at age 6 years: (1) physician diagnosis of asthma, (2) use of albuterol for coughing or wheezing episodes (prescribed by physician), (3) use of a daily controller medication, (4) step-up plan including use of albuterol or short-term use of inhaled corticosteroids during illness, and (5) use of prednisone for asthma exacerbation. ‖ Physician-diagnosed asthma ever, as assessed by questionnaire. ** Positive skin test response and/or allergen-specific IgE response to local aeroallergens as described previously [179-181].

†† Not applicable - maternal asthma was included as selection criteria.

146

147

Table E3. Complete list of annotated asthma-associated DMRs (n=589) in IIS children* * Identified by a 750 bp within-probe sliding window ANOVA (size ≥300 bp, magnitude difference≥0.2, BH-adjusted p value<0.01) † Adjusted for multiple testing (-log10 BH-adjusted P-value). ‡ Within a 750 bp window centered on the DMR OVERLAPPING DMR IDENTIFICATION CLOSEST TSS GENES (± 5 KB) Magnitu Chromosomal Size de No. Genomic Promoter Entrez Gene ID P-value† Distance Refseq Genes Location (bp) differen CpG‡ Location ID ID Name ce D1 chr1:28029-28475 446 -0.242 2.518 7 intron 1118 NR_024540 653635 WASH7P WASH7P chr1:619416- OR4F29, D2 875 -0.262 3.171 5 intergenic 2181 NM_001005277 81399 OR4F16 620291 OR4F16, OR4F3 chr1:2078616- D3 400 -0.303 4.435 9 intron 42661 NM_001033582 5590 PRKCZ PRKCZ 2079016 chr1:3195396- 1004229 D4 375 0.392 2.944 42 intron 151044 NR_036215 MIR4251 PRDM16 3195771 68 chr1:3446116- D5 525 0.235 2.985 26 intron 30976 NR_030277 693135 MIR551A MEGF6 3446641 chr1:7916430- D6 352 0.212 2.184 24 intergenic -3055 NM_021995 10911 UTS2 UTS2 7916782 chr1:11790134- D7 375 0.235 2.192 18 intergenic -5821 NM_001040196 57085 AGTRAP intergenic 11790509 chr1:13008095- PRAMEF D8 325 -0.25 2.28 6 promoter-TSS -851 NM_001010889 440561 PRAMEF6 13008420 6 chr1:13636362- PRAMEF D9 394 -0.222 3.797 5 intergenic -5414 NM_001010890 343070 intergenic 13636756 9 chr1:13640855- PRAMEF PRAMEF9, D10 320 -0.24 2.247 8 promoter-TSS -958 NM_001010890 343070 13641175 9 PRAMEF15 chr1:16924065- D11 344 -0.295 2.516 9 intron 15745 NM_017940 55672 NBPF1 NBPF1 16924409 chr1:20985704- D12 415 -0.237 3.06 10 intron 2126 NM_005216 1650 DDOST KIF17, DDOST 20986119 chr1:24826914- RCAN3, D13 375 0.201 2.798 17 intron -1740 NM_001251984 11123 RCAN3 24827289 RCAN3AS

147

148

chr1:26579075- D14 501 -0.236 2.374 11 intron 18632 NM_022778 64793 CEP85 CEP85 26579576 chr1:26947080- 1003021 D15 495 0.252 2.84 79 intergenic 66294 NR_031740 MIR1976 intergenic 26947575 90 chr1:28259405- SMPDL3 D16 333 0.368 2.468 10 intergenic -1933 NM_014474 27293 SMPDL3B 28259738 B chr1:43857017- D17 380 -0.33 2.88 17 intron 1651 NM_015284 23334 SZT2 MED8, SZT2 43857397 chr1:45082683- D18 595 0.227 3.334 63 intron 57119 NM_024587 79639 TMEM53 RNF220 45083278 chr1:45251683- D19 425 0.29 2.246 45 exon 1531 NM_153274 266675 BEST4 BEST4 45252108 chr1:49511297- D20 320 -0.215 2.486 7 exon -268910 NM_024603 79656 BEND5 AGBL4 49511617 chr1:67521434- D21 490 -0.241 3.141 7 intergenic -1599 NM_015139 23169 SLC35D1 SLC35D1 67521924 chr1:84607359- D22 395 -0.211 3.061 5 intron -2396 NM_182948 5567 PRKACB PRKACB 84607754 chr1:86171241- D23 528 -0.24 3.097 5 intron 2611 NM_001170670 54680 ZNHIT6 ZNHIT6 86171769 chr1:92947121- D24 400 0.219 3.137 33 intron 2035 NM_001127215 2672 GFI1 GFI1 92947521 chr1:118468419- D25 572 -0.238 3.024 5 intron 3597 NM_017686 54834 GDAP2 GDAP2, WDR3 118468991 chr1:144929866- D26 572 -0.235 4.134 6 intron 1880 NM_001002811 9659 PDE4DIP PDE4DIP 144930438 chr1:145045538- D27 403 -0.225 2.309 6 intron -5747 NM_001198832 9659 PDE4DIP PDE4DIP 145045941 chr1:146549619- 185 LOC7289 D28 0.35 4.28 51 intergenic -35947 NR_024442 728989 intergenic 146551474 5 89 chr1:147958520- PPIAL4A, D29 513 -0.24 2.486 6 intergenic -3357 NM_178230 164022 PPIAL4A 147959033 PPIAL4B chr1:147960243- PPIAL4A, D30 355 -0.204 2.94 6 intergenic -5001 NM_178230 164022 PPIAL4A 147960598 PPIAL4B chr1:148579506- NBPF15, D31 320 -0.228 2.553 5 exon 2309 NM_001102663 728936 NBPF16 148579826 NBPF16 chr1:151740167- OAZ3, MRPL9, D32 495 0.261 2.244 7 intron 1283 NM_016178 51686 OAZ3 151740662 TDRKH

148

149

chr1:152298811- D33 690 -0.247 3.425 7 intergenic -1477 NM_002016 2312 FLG FLG 152299501 chr1:152390374- D34 540 0.294 3.65 17 intergenic -3894 NM_016190 49860 CRNN CRNN 152390914 chr1:153091540- D35 376 -0.244 2.689 7 intergenic -5739 NM_001014450 6705 SPRR2F intergenic 153091916 chr1:154195857- C1orf43, D36 318 -0.239 3.163 6 intron 2691 NM_014847 9898 UBAP2L 154196175 UBAP2L chr1:158547123- D37 368 -0.242 3.735 5 intergenic 2382 NM_001004477 128367 OR10X1 OR10X1 158547491 chr1:158688854- D38 525 -0.246 2.698 8 intergenic -1211 NM_001005327 391114 OR6K3 OR6K3 158689379 chr1:158717724- D39 315 -0.219 2.694 12 intergenic -6725 NM_001005184 128371 OR6K6 intergenic 158718039 chr1:158720911- D40 305 -0.239 4.408 7 intergenic -3543 NM_001005184 128371 OR6K6 OR6K6 158721216 chr1:176155727- D41 487 -0.269 2.627 9 intron 20400 NM_001001740 64326 RFWD2 RFWD2 176156214 chr1:197870276- D42 600 -0.24 3.868 5 intergenic -1106 NM_001024594 388722 C1orf53 C1orf53 197870876 chr1:200375870- D43 310 -0.255 2.681 10 3' UTR 3141 NM_012482 23528 ZNF281 ZNF281 200376180 chr1:202859572- D44 370 -0.227 3.315 8 TTS -1372 NM_002871 5877 RABIF KLHL12, RABIF 202859942 chr1:203061380- D45 887 -0.258 3.342 6 intergenic -6657 NM_002479 4656 MYOG intergenic 203062267 chr1:210128503- D46 385 -0.226 2.617 5 intron 17176 NR_027459 255928 SYT14 SYT14 210128888 chr1:211313605- D47 333 -0.207 2.267 6 intergenic -6314 NM_172362 3756 KCNH1 intergenic 211313938 chr1:215176873- D48 370 -0.248 2.335 10 intergenic -1827 NM_001017424 3776 KCNK2 KCNK2 215177243 chr1:220954769- MARC2, D49 314 -0.232 2.752 5 intron -5113 NM_022746 64757 MARC1 220955083 MARC1 chr1:221921025- D50 310 -0.223 2.848 8 intergenic -5664 NM_144729 11221 DUSP10 intergenic 221921335 chr1:230197990- D51 675 -0.256 4.895 8 intergenic -4629 NM_004481 2590 GALNT2 GALNT2 230198665

149

150

chr1:241686365- D52 590 -0.278 3.059 5 intergenic -3575 NM_000143 2271 FH FH 241686955 chr1:241697860- D53 424 -0.239 2.567 6 intron 2638 NM_003679 8564 KMO KMO 241698284 chr1:243791053- D54 484 -0.281 3.665 8 intron 215289 NM_181690 10000 AKT3 AKT3 243791537 chr1:247667137- D55 500 -0.252 3.435 8 intergenic 13017 NM_001004698 441932 OR2W5 intergenic 247667637 chr1:247879633- D56 600 -0.246 4.246 5 intergenic -3876 NM_001005286 343169 OR6F1 OR6F1 247880233 chr10:3160852- 1005070 LOC1005 D57 305 0.214 2.316 34 exon -22789 NR_038284 PFKP 3161157 34 07034 chr10:4864339- D58 490 -0.239 3.19 6 intergenic -3785 NR_073127 83592 AKR1E2 AKR1E2 4864829 chr10:5133284- D59 385 -0.249 3.383 6 intron -3092 NM_001253909 8644 AKR1C3 AKR1C3 5133669 chr10:5486603- D60 495 -0.248 3.182 8 intron -1664 NM_005863 10276 NET1 NET1 5487098 chr10:14051467- D61 490 0.219 3.582 24 intron 321154 NM_018027 55691 FRMD4A FRMD4A 14051957 chr10:23215392- D62 985 -0.234 3.33 9 intergenic -1070 NM_173081 219681 ARMC3 ARMC3 23216377 chr10:42672545- LOC4416 D63 470 0.215 2.812 87 intergenic 190713 NR_024380 441666 intergenic 42673015 66 chr10:55581713- D64 535 -0.214 2.725 9 exon 979071 NM_001142765 65217 PCDH15 PCDH15 55582248 chr10:55888543- D65 405 -0.209 2.901 6 intron 672306 NM_001142765 65217 PCDH15 PCDH15 55888948 chr10:55965520- D66 527 -0.22 3.146 7 intron 595268 NM_001142765 65217 PCDH15 PCDH15 55966047 chr10:64016217- D67 425 -0.236 2.975 5 intron 12037 NM_145307 219790 RTKN2 RTKN2 64016642 chr10:71809023- D68 380 -0.222 2.581 6 intergenic -3144 NM_018649 55506 H2AFY2 H2AFY2 71809403 chr10:71890257- D69 510 -0.241 2.58 7 intron 2178 NM_001198696 84883 AIFM2 AIFM2 71890767 chr10:85892270- D70 390 -0.361 2.955 5 intergenic -6720 NM_014394 27069 GHITM intergenic 85892660

150

151

chr10:92454928- D71 312 -0.233 2.279 5 intergenic 162587 NM_019860 3363 HTR7 intergenic 92455240 chr10:105344175- D72 310 0.217 4.101 102 exon 90595 NM_004210 9148 NEURL NEURL 105344485 chr10:115422789- D73 471 -0.231 2.22 8 intron 805 NM_001261463 4892 NRAP NRAP 115423260 chr10:118931854- 1005008 D74 400 -0.249 2.776 11 intergenic -4769 NR_037436 MIR3663 MIR3663 118932254 93 chr10:134067135- D75 302 0.225 2.708 28 intron 54191 NM_173575 282974 STK32C STK32C 134067437 chr10:134149160- D76 405 0.22 2.878 12 intron -1249 NR_026559 80313 LRRC27 LRRC27 134149565 chr10:135012691- D77 370 0.223 4.265 39 intron -30902 NM_003577 8433 UTF1 KNDC1 135013061 chr10:135254566- LOC6192 D78 485 0.217 2.17 32 intergenic -12624 NR_002934 619207 intergenic 135255051 07 chr11:1763722- D79 325 0.347 3.015 7 intron 7940 NM_001170820 402778 IFITM10 MOB2, IFITM10 1764047 chr11:2011138- 1001335 MRPL23- H19, MRPL23- D80 411 0.229 2.662 27 promoter-TSS -193 NR_024471 2011549 45 AS1 AS1 chr11:2068424- 142 D81 -0.255 2.68 7 intergenic -50069 NR_002196 283120 H19 intergenic 2069844 0 chr11:2076419- D82 900 -0.248 2.312 5 intergenic -57804 NR_002196 283120 H19 intergenic 2077319 chr11:2085620- D83 930 -0.281 2.757 6 intergenic -67020 NR_002196 283120 H19 intergenic 2086550 chr11:4393525- D84 378 0.367 2.369 8 intergenic -4098 NM_001005161 143496 OR52B4 OR52B4 4393903 chr11:4903522- D85 500 -0.247 2.942 11 exon 723 NM_001004759 401665 OR51T1 OR51T1 4904022 chr11:5019521- D86 302 -0.246 3.34 7 promoter-TSS -541 NM_001004755 119682 OR51L1 OR51L1 5019823 chr11:5111850- D87 405 -0.244 3.076 6 intergenic -31195 NM_001005164 119678 OR52E2 intergenic 5112255 chr11:5443751- OR51B5, D88 400 -0.22 3.878 15 exon 610 NM_001004757 390061 OR51Q1 5444151 OR51Q1 chr11:5743837- D89 309 -0.218 3.131 5 intergenic -13687 NM_001005180 387748 OR56B1 intergenic 5744146

151

152

chr11:5880266- D90 375 -0.232 3.537 9 intergenic -1521 NM_001005168 390079 OR52E8 OR52E8 5880641 chr11:6050754- D91 380 -0.213 2.367 5 intergenic -1973 NM_001001917 120796 OR56A1 OR56A1 6051134 chr11:6052435- D92 446 -0.227 3.186 9 intergenic -3687 NM_001001917 120796 OR56A1 OR56A1 6052881 chr11:6912947- D93 510 -0.248 3.407 8 exon 529 NM_003700 120776 OR2D2 OR2D2 6913457 chr11:10773833- D94 470 -0.247 2.895 9 intron 1257 NM_014633 9646 CTR9 CTR9 10774303 chr11:11857771- D95 530 -0.246 3.331 7 intergenic -4934 NM_017944 55031 USP47 USP47 11858301 chr11:16760509- D96 370 0.225 2.505 31 intron 546 NM_014267 10944 C11orf58 C11orf58 16760879 chr11:18410042- D97 400 0.23 2.733 13 intergenic -5694 NM_001165416 3939 LDHA intergenic 18410442 chr11:18417431- D98 585 0.296 2.305 47 promoter-TSS -90 NM_001165414 3939 LDHA LDHA 18418016 chr11:18955777- MRGPRX D99 470 0.249 2.763 25 exon 537 NM_147199 259249 MRGPRX1 18956247 1 chr11:27743674- D100 306 0.236 3.027 52 promoter-TSS -222 NM_170731 627 BDNF BDNF 27743980 chr11:32850713- D101 500 0.463 3.122 23 promoter-TSS -518 NM_024081 79056 PRRG4 PRRG4 32851213 chr11:55563441- D102 490 -0.257 3.218 7 exon 654 NM_001004735 219436 OR5D14 OR5D14 55563931 chr11:57945518- D103 310 -0.227 3.384 9 intron -12233 NM_001005283 219957 OR9Q2 OR9Q1 57945828 chr11:58346012- ZFP91- ZFP91, ZFP91- D104 616 0.336 2.988 67 promoter-TSS -267 NR_024091 386607 58346628 CNTF CNTF, LPXN chr11:62341248- D105 470 0.301 2.707 45 promoter-TSS -23 NM_001404 1937 EEF1G EEF1G, TUT1 62341718 chr11:64408326- 120 D106 -0.28 2.904 7 intron 1859 NM_138734 9379 NRXN2 NRXN2 64409531 5 chr11:65629666- EFEMP2, CFL1, D107 370 0.324 2.405 11 intron 1979 NM_025128 80198 MUS81 65630036 MUS81 chr11:70668933- 1005008 D108 490 0.254 2.601 30 intron 49295 NR_037437 MIR3664 SHANK2 70669423 44

152

153

chr11:73883801- D109 307 -0.239 2.243 5 intron 1586 NM_016147 51400 PPME1 C2CD3, PPME1 73884108 chr11:75139113- D110 610 0.224 3.565 40 intron 2256 NM_001039548 283212 KLHL35 KLHL35 75139723 chr11:94707446- D111 478 -0.257 2.956 14 promoter-TSS 840 NM_018039 55693 KDM4D CWC15, KDM4D 94707924 chr11:98892394- D112 380 -0.241 2.909 7 intron 878 NM_014361 53942 CNTN5 CNTN5 98892774 chr11:101789553- ANGPTL KIAA1377, D113 510 -0.252 3.742 6 intron -2555 NM_178127 253935 101790063 5 ANGPTL5 chr11:103108580- DYNC2H D114 400 -0.204 3.245 5 intron 128620 NM_001080463 79659 DYNC2H1 103108980 1 chr11:104915738- D115 515 0.342 2.374 6 promoter-TSS 56 NM_001017534 114769 CARD16 CARD16 104916253 chr11:114312263- D116 493 -0.213 3.152 5 intron 2401 NM_015523 25996 REXO2 REXO2 114312756 chr11:114550477- D117 611 -0.227 4.063 5 intron 1582 NM_182495 120406 NXPE2 FAM55B 114551088 chr11:116790538- D118 595 -0.232 2.626 10 intron -82497 NM_000039 335 APOA1 SIK3 116791133 chr11:116874432- D119 305 -0.221 2.133 7 intron 94409 NM_025164 23387 SIK3 SIK3 116874737 chr11:116881940- D120 370 -0.208 2.336 9 intron 86868 NM_025164 23387 SIK3 SIK3 116882310 chr11:131214392- D121 625 -0.224 2.649 11 intergenic -25667 NM_001048209 50863 NTM intergenic 131215017 chr12:1748591- D122 602 0.264 2.657 62 exon 10480 NM_032642 81029 WNT5B WNT5B 1749193 chr12:4641451- D123 375 -0.23 2.636 8 intron 5999 NM_020374 57102 C12orf4 C12orf4 4641826 chr12:4914276- D124 370 -0.276 2.692 8 intergenic -3881 NM_002235 3742 KCNA6 KCNA6 4914646 chr12:7902903- D125 385 -0.27 3.127 9 intergenic -1026 NM_203503 170482 CLEC4C CLEC4C 7903288 chr12:9824116- D126 393 0.293 2.329 17 intron 2008 NR_036693 29121 CLEC2D CLEC2D 9824509 chr12:11341172- D127 580 -0.246 2.621 5 intergenic -1919 NM_181429 353164 TAS2R42 TAS2R42 11341752

153

154

chr12:12814205- D128 305 -0.213 3.753 7 exon 34764 NM_006143 2842 GPR19 GPR19 12814510 chr12:13198821- KIAA146 D129 495 0.258 3.359 10 intron 1753 NM_020853 57613 KIAA1467 13199316 7 chr12:21589244- PYROXD D130 400 -0.218 3.064 5 intergenic -1094 NM_024854 79912 PYROXD1 21589644 1 chr12:21995017- D131 480 -0.212 2.906 7 exon -67502 NM_004982 3764 KCNJ8 ABCC9 21995497 chr12:23099370- D132 390 -0.244 3.824 6 intergenic 321489 NM_001039481 55500 ETNK1 intergenic 23099760 chr12:48723063- D133 330 0.223 3.306 47 exon 465 NM_181788 341567 H1FNT H1FNT 48723393 chr12:49524725- D134 411 0.279 2.704 84 intron 374 NM_006082 10376 TUBA1B TUBA1B 49525136 chr12:54719022- D135 400 0.228 2.446 22 intron 311 NM_016057 22818 COPZ1 COPZ1 54719422 chr12:54720215- D136 415 0.216 2.736 15 intron 1511 NM_016057 22818 COPZ1 COPZ1 54720630 chr12:66537890- D137 595 -0.262 2.83 6 intron -13654 NM_032338 84298 LLPH TMBIM4 66538485 chr12:68653983- D138 382 -0.248 2.909 7 intergenic -6893 NM_020525 50616 IL22 intergenic 68654365 chr12:72333232- D139 495 -0.232 5.089 5 intron 853 NM_173353 121278 TPH2 TPH2 72333727 chr12:75699193- D140 480 0.229 2.395 35 intron 24403 NM_032606 84698 CAPS2 CAPS2 75699673 chr12:89743694- D141 525 0.333 2.263 11 intron 2340 NM_001946 1848 DUSP6 DUSP6 89744219 chr12:100748413- D142 385 -0.262 2.749 5 intergenic -2252 NM_139319 246213 SLC17A8 SLC17A8 100748798 chr12:120687466- D143 610 0.318 3.322 9 5' UTR 193 NM_025157 5829 PXN PXN 120688076 chr12:120906313- DYNLL1, D144 491 0.296 2.919 52 intron 1000 NM_003769 8683 SRSF9 120906804 SRSF9, GATC chr12:132394036- D145 502 0.212 3.456 56 intron 15008 NM_003565 8408 ULK1 ULK1 132394538 chr12:132434342- D146 375 0.248 2.253 120 promoter-TSS 64 NM_015409 57634 EP400 EP400 132434717

154

155

chr12:133019759- D147 325 0.211 2.465 41 intergenic -47236 NM_001142641 57666 FBRSL1 intergenic 133020084 chr13:21714409- D148 305 0.244 2.483 62 promoter-TSS -92 NM_005870 10284 SAP18 SAP18 21714714 chr13:23752625- D149 400 -0.224 2.8 10 intergenic -2235 NM_000231 6445 SGCG SGCG 23753025 chr13:24816697- 1008742 SPATA13 D150 391 -0.268 2.67 6 intron 11685 NR_046531 SPATA13 24817088 31 -AS1 chr13:25419251- D151 570 -0.24 3.533 7 intron 77491 NR_047595 55835 CENPJ RNF17 25419821 chr13:27185922- D152 575 -0.259 3.987 7 intron 54369 NM_006646 10810 WASF3 WASF3 27186497 chr13:30816333- KATNAL D153 490 -0.22 3.092 9 intron 64613 NM_001014380 84056 KATNAL1 30816823 1 chr13:36163256- NBEA, D154 320 -0.246 4.563 5 intron 112530 NM_001204197 26960 NBEA 36163576 MIR548F5 chr13:43570879- D155 304 -0.223 4.039 6 intergenic -4624 NM_033255 94240 EPSTI1 EPSTI1 43571183 chr13:44200823- D156 520 -0.218 2.651 5 intron 2530 NM_001127615 55068 ENOX1 ENOX1 44201343 chr13:48613616- D157 886 0.239 2.94 24 intron 2356 NM_018283 55270 NUDT15 NUDT15 48614502 chr13:51416034- D158 500 -0.226 3.197 10 intron 1601 NM_198989 220107 DLEU7 DLEU7 51416534 chr13:52376962- D159 385 0.235 2.758 8 intron 1144 NM_001031719 79758 DHRS12 DHRS12 52377347 chr13:52435278- D160 470 -0.237 3.653 6 promoter-TSS -604 NM_031290 83446 CCDC70 CCDC70 52435748 chr13:84453945- D161 310 0.233 3.549 32 exon 2428 NM_052910 114798 SLITRK1 SLITRK1 84454255 chr13:88329305- D162 420 0.232 3.138 52 exon 4645 NM_015567 26050 SLITRK5 SLITRK5 88329725 chr13:99665699- D163 675 -0.226 3.541 13 intron -35698 NM_001130050 23348 DOCK9 DOCK9 99666374 chr13:109241531- D164 400 -0.323 3.266 7 intergenic -6769 NM_015011 23026 MYO16 intergenic 109241931 chr13:114108569- ADPRHL ADPRHL1, D165 365 0.232 2.697 26 promoter-TSS -912 NM_138430 113622 114108934 1 DCUN1D2

155

156

chr14:19922332- D166 559 -0.397 3.725 8 intergenic 97661 NM_001145442 641455 POTEM intergenic 19922891 chr14:50154049- D167 400 0.22 3.201 27 intron 849 NM_002692 5427 POLE2 POLE2 50154449 chr14:62539883- LINC006 D168 610 -0.252 3.099 5 intron -43887 NR_015358 646113 SYT16 62540493 43 chr14:63786679- 100 D169 0.29 2.727 29 intergenic -1590 NM_145171 122876 GPHB5 GPHB5 63787687 8 chr14:89054472- D170 305 -0.219 2.69 6 intron -6106 NM_207662 79882 ZC3H14 ZC3H14 89054777 chr14:101188142- D171 490 0.236 2.441 15 intergenic -4815 NM_003836 8788 DLK1 DLK1 101188632 chr14:102834884- D172 416 -0.222 2.456 9 intron 5792 NM_014844 9895 TECPR2 TECPR2 102835300 chr15:20717355- D173 400 -0.238 3.581 10 intergenic -6122 NR_036432 283755 HERC2P3 intergenic 20717755 chr15:22733726- GOLGA6 D174 870 -0.218 3.106 8 intergenic -2085 NM_001001413 283767 GOLGA6L1 22734596 L1 chr15:23428157- GOLGA8 D175 572 -0.281 3.636 9 intergenic -6627 NR_033350 390535 intergenic 23428729 EP chr15:25200266- D176 685 0.243 3.189 49 intron 473 NM_003097 6638 SNRPN SNRPN, SNURF 25200951 chr15:26108457- D177 327 0.261 2.288 86 promoter-TSS -271 NM_024490 57194 ATP10A ATP10A 26108784 chr15:27138906- D178 305 0.207 2.328 44 intron 26785 NM_001165037 2558 GABRA5 GABRA5 27139211 chr15:28573237- D179 405 -0.22 3.055 8 intergenic -6141 NM_004667 8924 HERC2 intergenic 28573642 chr15:28617077- 1001325 GOLGA8 D180 390 -0.234 3.021 10 intergenic -6512 NR_033351 intergenic 28617467 65 F chr15:28620752- 1001325 GOLGA8 GOLGA8G, D181 575 -0.327 3.143 5 intergenic -2745 NR_033351 28621327 65 F GOLGA8F chr15:28780531- GOLGA8 GOLGA8G, D182 540 -0.277 2.763 5 intergenic -2658 NR_033353 283768 28781071 G GOLGA8F chr15:28784414- GOLGA8 D183 400 -0.236 2.978 11 intergenic -6471 NR_033353 283768 intergenic 28784814 G chr15:28871033- D184 365 -0.231 2.484 6 intergenic -28373 NR_036443 440248 HERC2P9 intergenic 28871398

156

157

chr15:30692464- CHRFAM D185 354 -0.265 2.302 7 intergenic -6777 NM_139320 89832 intergenic 30692818 7A chr15:36927825- D186 514 -0.213 3.282 6 intron -4724 NM_032499 84529 C15orf41 C15orf41 36928339 chr15:41786297- D187 609 0.244 2.751 87 exon 545 NM_002220 3706 ITPKA ITPKA 41786906 chr15:49446102- COPS2, GALK2, D188 305 -0.224 2.614 7 intron 1600 NM_001143887 9318 COPS2 49446407 LOC100306975 chr15:61522646- D189 602 -0.232 3.852 11 intergenic -1445 NM_134261 6095 RORA RORA 61523248 chr15:67356706- D190 321 0.219 2.356 24 intergenic -1329 NM_005902 4088 SMAD3 SMAD3 67357027 chr15:74377911- GOLGA6 D191 331 -0.24 2.374 6 intergenic -3185 NM_001038640 342096 GOLGA6A 74378242 A chr15:81510763- D192 691 -0.223 2.857 6 intron -6532 NM_172217 3603 IL16 IL16 81511454 chr15:82720077- GOLGA6 D193 428 -0.241 3.696 8 intergenic -1894 NM_198181 440295 GOLGA6L9 82720505 L9 chr15:83025933- 1001348 UBE2Q2 D194 346 -0.258 2.322 11 intron 2333 NR_004847 UBE2Q2P3 83026279 69 P2 chr15:83113821- GOLGA6 D195 515 0.239 2.912 51 intergenic 15368 NM_198181 440295 intergenic 83114336 L9 chr15:85775471- LOC6424 D196 316 -0.297 3.13 8 intergenic -27111 NR_049748 642423 intergenic 85775787 23 chr15:89348238- D197 405 0.245 2.988 9 intron 1766 NM_013227 176 ACAN ACAN 89348643 chr15:91833384- D198 430 -0.232 2.37 8 intron 190060 NM_014848 9899 SV2B SV2B 91833814 chr15:93614715- D199 380 0.236 3.848 26 intron 1484 NM_001166286 56963 RGMA RGMA 93615095 chr15:101332621- ALDH1A D200 304 -0.22 3.219 11 intergenic -87236 NM_000693 220 intergenic 101332925 3 chr15:102466837- 119 D201 -0.237 3.156 10 intergenic -4172 NM_001004195 26682 OR4F4 OR4F4 102468032 5 chr16:1559764- D202 570 0.232 3.653 37 TTS 16697 NM_016111 9894 TELO2 IFT140, TELO2 1560334 chr16:2828985- D203 680 0.294 2.225 22 intergenic -2028 NM_207013 6923 TCEB2 TCEB2, PRSS33 2829665

157

158

chr16:6422898- 100 D204 -0.236 3.033 8 intron 354266 NM_018723 54715 RBFOX1 RBFOX1 6423898 0 chr16:6748783- D205 320 -0.217 2.857 10 intron -74867 NM_001142334 54715 RBFOX1 RBFOX1 6749103 chr16:12009069- D206 398 0.215 2.414 105 exon 557 NM_001130006 2935 GSPT1 GSPT1 12009467 chr16:18443504- 1004229 MIR3180- D207 485 0.212 3.412 44 intergenic 52378 NR_036142 intergenic 18443989 56 2 chr16:20777313- D208 305 -0.212 2.679 6 intron 2153 NM_005622 6296 ACSM3 ACSM3 20777618 chr16:21531140- SLC7A5P D209 330 0.333 3.69 75 non-coding 460 NR_002594 387254 SLC7A5P2 21531470 2 chr16:22305349- D210 395 -0.278 2.796 8 intergenic -3150 NM_018119 55718 POLR3E POLR3E 22305744 chr16:22522466- 1001322 LOC1001 D211 566 -0.264 2.67 8 intergenic -2095 NM_001135865 LOC100132247 22523032 47 32247 chr16:25931053- D212 690 -0.282 4.105 6 intron 228051 NM_006040 9951 HS3ST4 HS3ST4 25931743 chr16:25958955- D213 475 -0.228 2.747 10 intron 255845 NM_006040 9951 HS3ST4 HS3ST4 25959430 chr16:26204939- D214 905 -0.305 3.822 5 intergenic 502044 NM_006040 9951 HS3ST4 intergenic 26205844 chr16:26343060- D215 780 -0.296 3.419 8 intergenic 640103 NM_006040 9951 HS3ST4 intergenic 26343840 chr16:31232871- D216 378 -0.253 2.752 10 intron -4665 NM_152901 260434 PYDC1 PYDC1, TRIM72 31233249 chr16:33070434- D217 411 0.226 2.589 82 intergenic -134946 NM_016212 24150 TP53TG3 intergenic 33070845 chr16:62284966- D218 485 -0.229 3.256 5 intergenic -214469 NM_001796 1006 CDH8 intergenic 62285451 chr16:62468784- D219 600 -0.228 2.882 5 intergenic -398345 NM_001796 1006 CDH8 intergenic 62469384 chr16:70208313- D220 540 0.253 4.984 28 intron 655 NM_173619 283971 CLEC18C CLEC18C 70208853 chr16:71494556- D221 325 -0.233 3.78 6 intron 1399 NM_145911 7571 ZNF23 ZNF23 71494881 chr16:75468283- D222 330 0.21 3.124 24 intergenic -1061 NM_006324 10428 CFDP1 CFDP1 75468613

158

159

chr16:80569566- DYNLRB D223 610 -0.284 3.925 6 intergenic -4983 NM_130897 83657 DYNLRB2 80570176 2 chr16:82208779- MPHOSP D224 498 -0.228 2.861 6 intergenic -5199 NM_005792 10200 MPHOSPH6 82209277 H6 chr16:87524869- ZCCHC1 D225 530 0.225 3.51 53 intron 326 NM_015144 23174 ZCCHC14 87525399 4 chr16:90148322- D226 575 0.306 3.78 45 intergenic -6271 NM_001098173 11105 PRDM7 intergenic 90148897 chr17:2997224- D227 501 -0.231 2.695 6 intergenic -1184 NM_002548 4991 OR1D2 OR1D2 2997725 chr17:6938284- SLC16A1 D228 401 0.231 2.899 7 promoter-TSS -910 NM_201566 201232 SLC16A13 6938685 3 chr17:8114593- D229 370 -0.218 2.58 10 promoter-TSS -834 NM_001256834 9212 AURKB AURKB 8114963 chr17:16555471- D230 320 -0.217 2.296 6 intron 1536 NM_020787 57547 ZNF624 ZNF624 16555791 chr17:19545683- ALDH3A D231 385 -0.257 3.207 6 intergenic -6189 NM_000382 224 intergenic 19546068 2 chr17:31255661- D232 390 -0.216 2.557 18 intron 928 NM_001033504 26022 TMEM98 TMEM98 31256051 chr17:38277847- D233 589 0.304 2.343 81 promoter-TSS -649 NM_001012241 339287 MSL1 MSL1 38278436 chr17:45062246- D234 319 -0.297 3.934 10 intergenic -5791 NM_203400 388394 RPRML intergenic 45062565 chr17:48792207- D235 395 -0.232 2.563 10 intergenic -4522 NM_016424 51747 LUC7L3 LUC7L3 48792602 chr17:48828262- LINC004 D236 498 -0.268 3.494 10 3' UTR 16365 NR_073199 55018 LUC7L3 48828760 83 chr17:55167220- D237 375 -0.209 2.492 8 intron 4272 NM_001242902 8165 AKAP1 AKAP1 55167595 chr17:56249191- D238 375 -0.264 2.578 7 intergenic 2361 NM_001004707 124538 OR4D2 OR4D2 56249566 chr17:61430940- D239 407 -0.243 2.897 7 intron 87064 NM_001017917 1534 CYB561 TANC2 61431347 chr17:74349827- D240 685 0.308 2.567 62 promoter-TSS 61 NM_002766 5635 PRPSAP1 PRPSAP1 74350512 chr17:78948420- D241 400 -0.249 3.256 9 intergenic -17021 NM_024591 79643 CHMP6 intergenic 78948820

159

160

chr17:79361084- D242 334 0.211 4.662 83 intergenic -12289 NM_001080519 57597 BAHCC1 intergenic 79361418 chr17:79917839- D243 470 0.235 3.566 107 intron 983 NM_178493 147111 NOTUM NOTUM 79918309 chr17:80036750- D244 675 0.235 2.969 54 exon -13390 NM_022156 64118 DUS1L FASN 80037425 chr18:22036495- D245 331 -0.227 2.063 9 intergenic -3933 NM_001143828 59340 HRH4 IMPACT, HRH4 22036826 chr18:25903262- D246 475 -0.242 2.775 8 intergenic -146054 NM_001792 1000 CDH2 intergenic 25903737 chr18:31800204- D247 305 -0.229 2.791 7 intron 2078 NM_001198547 8715 NOL4 NOL4 31800509 RPL17, RPL17- C18ORF32, chr18:47019290- D248 498 -0.475 3.42 12 promoter-TSS -604 NM_001199345 6139 RPL17 SNORD58A, 47019788 SNORD5B, SNORD58C chr18:56982879- D249 505 -0.254 2.933 9 intron 2750 NM_181654 339302 CPLX4 CPLX4 56983384 chr18:61225101- SERPINB D250 495 -0.269 3.027 5 intron 1955 NM_080474 89777 SERPINB12 61225596 12 chr18:61432732- SERPINB D251 320 -0.214 3.714 6 intron -9717 NM_001261831 8710 SERPINB7 61433052 7 chr18:61438138- SERPINB D252 410 -0.242 3.807 5 intron -4266 NM_001261831 8710 SERPINB7 61438548 7 chr18:66507007- CCDC102 D253 410 -0.213 2.863 5 intron 41895 NM_024781 79839 CCDC102B 66507417 B chr18:77196024- D254 500 0.219 2.609 47 intron 35948 NM_172389 4772 NFATC1 NFATC1 77196524 chr18:77791162- D255 520 -0.224 3.233 8 intergenic -2924 NM_001171967 79863 RBFA RBFA 77791682 D256 chr19:69351-70083 732 -0.279 3.136 7 intron 1249 NR_033266 375690 WASH5P WASH5P chr19:585719- D257 505 0.286 2.474 13 intergenic -3922 NM_001194 610 HCN2 BSG, HCN2 586224 chr19:735961- D258 472 0.219 2.62 45 intron -14949 NM_173481 126353 C19orf21 PALM 736433

160

161

chr19:805752- 1006164 D259 700 0.339 3.751 78 intron 1162 NR_039900 MIR4745 PTBP1, MIR4745 806452 59 chr19:1021495- D260 330 0.241 3.03 40 promoter-TSS -519 NM_033420 91304 C19orf6 C19orf6, CNN2 1021825 chr19:1228365- D261 580 0.219 4.613 101 TTS 9335 NM_152769 255057 C19orf26 C19orf26, STK11 1228945 chr19:1876344- FAM108 D262 590 0.27 2.843 46 TTS 8879 NM_001130111 81926 FAM108A1 1876934 A1 chr19:4557006- D263 325 0.223 3.104 37 intron 2603 NM_032108 10501 SEMA6B SEMA6B 4557331 chr19:5135093- D264 305 0.261 2.284 31 intron 166121 NM_015015 23030 KDM4B KDM4B 5135398 chr19:10907627- 1006164 D265 576 0.226 3.466 28 intron 16985 NR_039903 MIR4748 DNM2 10908203 25 chr19:10934228- MIR199A D266 605 0.247 2.605 21 exon -6358 NR_029586 406976 DNM2 10934833 1 chr19:16051903- D267 324 -0.241 4.468 6 intergenic -6389 NM_021187 57834 CYP4F11 intergenic 16052227 chr19:17445165- D268 310 0.207 3.571 84 promoter-TSS 318 NM_020959 57719 ANO8 GTPBP3, ANO8 17445475 chr19:17449965- D269 525 0.248 3.767 37 intron 1901 NM_133644 84705 GTPBP3 GTPBP3, ANO8 17450490 chr19:18041207- D270 405 -0.23 2.612 16 intergenic -2415 NM_001136203 115098 CCDC124 CCDC124 18041612 chr19:21860585- LOC6413 D271 370 0.241 2.901 37 intergenic -72777 NR_024523 641367 intergenic 21860955 67 chr19:37020091- D272 385 0.237 2.749 21 intergenic -1035 NM_001166036 339324 ZNF260 ZNF260 37020476 chr19:37263404- D273 590 0.25 3.935 53 promoter-TSS 17 NM_001267779 342892 ZNF850 ZNF850 37263994 chr19:38084769- D274 379 0.215 2.643 26 promoter-TSS 715 NM_016536 51276 ZNF571 ZNF540, ZNF571 38085148 chr19:44570828- D275 470 -0.229 3.281 11 exon -5234 NM_001037813 342909 ZNF284 ZNF284, ZNF223 44571298 chr19:45515333- D276 310 0.208 3.033 56 exon 10781 NM_006509 5971 RELB RELB 45515643 chr19:47102132- 1005060 LOC100506012, D277 375 -0.213 2.422 7 intron 2138 NM_001205281 PPP5D1 47102507 12 CALM3

161

162

chr19:47356165- D278 330 0.429 2.054 5 intergenic -2127 NM_004069 1175 AP2S1 AP2S1 47356495 chr19:47634525- D279 325 0.258 3.13 36 intron 572 NR_027280 10055 SAE1 SAE1 47634850 chr19:49842935- D280 310 0.219 3.706 49 TTS 4413 NM_001774 951 CD37 CD37, TEAD2 49843245 chr19:51572668- D281 324 -0.218 2.753 11 intergenic -4463 NM_015596 26085 KLK13 KLK13 51572992 chr19:51586807- D282 330 0.244 2.771 18 intron 530 NM_022046 43847 KLK14 KLK14 51587137 chr19:51774231- SIGLECL D283 515 0.292 2.608 34 intergenic 13524 NM_173635 284369 C19orf75 51774746 1 chr19:54972295- LENG9, LENG8, D284 410 0.217 2.484 39 TTS 2394 NM_198988 94059 LENG9 54972705 CDC42EP5 chr19:56163345- CCDC106, D285 510 0.233 3.287 44 intron -1816 NM_007279 11338 U2AF2 56163855 U2AF2 chr19:58566138- ZSCAN1, D286 403 0.246 2.633 27 TTS -4268 NM_003436 7694 ZNF135 58566541 ZNF135 chr19:58874038- D287 330 0.282 2.105 82 promoter-TSS 11 NM_001207009 162968 ZNF497 ZNF497 58874368 chr2:557919- D288 525 0.246 3.382 43 intergenic 119258 NM_152834 129787 TMEM18 intergenic 558444 chr2:10151509- D289 586 0.248 4.112 56 intergenic -31880 NM_003597 8462 KLF11 intergenic 10152095 chr2:16153858- MYCNO D290 375 0.213 2.431 32 intergenic -72200 NR_026766 10408 intergenic 16154233 S chr2:20103745- D291 331 -0.214 3.239 6 intergenic -2166 NM_001008237 130502 TTC32 TTC32 20104076 chr2:33657519- RASGRP D292 430 -0.25 2.822 11 intergenic -3682 NM_170672 25780 RASGRP3 33657949 3 chr2:36924606- D293 415 -0.223 3.41 8 intron 980 NM_001177969 5212 VIT VIT 36925021 chr2:37895804- CDC42EP D294 370 -0.231 4.415 6 intron 3353 NM_006449 10602 CDC42EP3 37896174 3 chr2:42556470- COX7A2 D295 325 -0.206 3.494 10 intron 31724 NM_004718 9167 EML4 42556795 L chr2:43860636- PLEKHH D296 310 -0.246 2.388 6 intergenic -3648 NM_172069 130271 PLEKHH2 43860946 2

162

163

chr2:45231533- D297 410 0.246 2.557 47 TTS 4804 NM_016932 10736 SIX2 SIX2 45231943 chr2:50577893- D298 424 -0.239 5.112 5 intron -3211 NM_138735 9378 NRXN1 NRXN1 50578317 chr2:51704426- D299 380 -0.235 3.385 5 intergenic -444942 NM_001135659 9378 NRXN1 intergenic 51704806 chr2:51719209- D300 505 -0.227 2.656 7 intergenic -459787 NM_001135659 9378 NRXN1 intergenic 51719714 chr2:51950325- D301 600 -0.24 3.499 6 intergenic -690951 NM_001135659 9378 NRXN1 intergenic 51950925 chr2:54557572- D302 325 0.242 2.761 52 promoter-TSS -337 NM_001100396 129852 C2orf73 C2orf73 54557897 chr2:58470499- D303 604 -0.223 3.084 7 intergenic -2286 NM_018062 55120 FANCL FANCL 58471103 chr2:88463530- D304 520 -0.241 3.166 11 intergenic -6024 NM_001244676 55258 THNSL2 intergenic 88464050 chr2:92261990- ACTR3B D305 517 0.223 3.959 48 intergenic 133089 NR_027714 440888 intergenic 92262507 P2 chr2:97910656- 1005061 LOC1005 D306 725 -0.256 3.637 7 exon -37211 NR_040097 ANKRD36 97911381 23 06123 chr2:98128084- 1005061 LOC1005 D307 528 -0.216 2.576 9 exon -37299 NR_040097 ANKRD36B 98128612 23 06123 chr2:100760155- D308 585 -0.229 3.542 10 intergenic -1410 NM_002285 3899 AFF3 AFF3 100760740 chr2:105853033- D309 375 0.219 3.383 28 intergenic -4980 NM_007227 11250 GPR45 GPR45 105853408 chr2:107073701- D310 871 -0.216 2.898 5 intron 10665 NM_001144013 653489 RGPD3 RGPD3 107074572 chr2:116065044- D311 780 -0.265 3.61 6 intron 145750 NM_001004360 57628 DPP10 DPP10 116065824 chr2:118757608- D312 370 -0.24 2.562 7 intron 13946 NM_019044 54520 CCDC93 CCDC93 118757978 chr2:166935866- D313 308 -0.24 2.936 5 intron -5871 NM_006920 6323 SCN1A SCN1A 166936174 chr2:174212569- D314 305 -0.222 2.674 10 intergenic -6840 NM_031942 83879 CDCA7 intergenic 174212874 chr2:182547413- NEUROD D315 615 0.313 2.648 43 intergenic -2328 NM_002500 4760 NEUROD1 182548028 1

163

164

chr2:193061133- D316 375 -0.247 2.548 16 intergenic -1676 NM_016192 23671 TMEFF2 TMEFF2 193061508 chr2:197676027- D317 875 -0.28 4.502 6 intergenic -1464 NM_213608 401027 C2orf66 C2orf66 197676902 chr2:200774538- D318 425 -0.212 3.305 5 intergenic -1229 NM_153689 205327 C2orf69 C2orf69 200774963 chr2:202564922- D319 308 -0.216 2.53 6 TTS -1659 NM_033066 58538 MPP4 MPP4, ALS2 202565230 chr2:219576276- D320 770 0.376 3.481 6 intron 1093 NM_014640 9654 TTLL4 TTLL4 219577046 chr2:220083145- D321 385 0.226 5.569 68 TTS 375 NM_005689 10058 ABCB6 ABCB6, ATG9A 220083530 chr2:234987728- D322 430 -0.238 3.205 8 intergenic 28597 NM_006944 6694 SPP2 SPP2 234988158 chr2:242448454- D323 390 0.305 2.648 57 promoter-TSS -615 NM_006374 10494 STK25 STK25 242448844 chr20:3714870- D324 385 -0.225 2.307 13 intron 1745 NM_001197327 116835 HSPA12B HSPA12B 3715255 chr20:8005813- D325 627 -0.232 2.504 8 intergenic -5733 NM_021156 56255 TMX4 intergenic 8006440 chr20:18123530- PET117, D326 519 0.387 3.142 9 TTS 921 NM_020536 57325 CSRP2BP 18124049 CSRP2BP chr20:23413449- D327 475 -0.245 2.58 6 intergenic -6636 NM_138283 128817 CSTL1 intergenic 23413924 chr20:36304618- 1002877 LOC1002 D328 325 -0.218 2.701 8 promoter-TSS -532 NR_040021 LOC100287792 36304943 92 87792 chr20:46300740- D329 492 0.274 2.918 23 exon 113822 NM_018837 55959 SULF2 SULF2 46301232 chr20:57202945- D330 385 0.214 4.187 10 intergenic -23172 NR_037943 8675 STX16 intergenic 57203330 chr20:57416607- 108 GNAS-AS1, D331 0.221 2.865 47 non-coding 2356 NM_016592 2778 GNAS 57417696 9 GNAS chr20:57426818- GNAS-AS1, D332 791 0.246 3.747 39 promoter-TSS -823 NM_080425 2778 GNAS 57427609 GNAS chr20:60873143- OSBPL2, D333 585 0.308 2.481 20 intergenic -4592 NM_175573 11047 ADRM1 60873728 ADRM1 chr20:60888727- LAMA5, D334 500 0.233 3.176 53 intron 10950 NM_007002 11047 ADRM1 60889227 ADRM1

164

165

chr20:62580073- UCKL1, UCKL1- D335 490 0.272 4.022 30 intron 2209 NM_001193379 54963 UCKL1 62580563 AS1 chr20:62901069- D336 605 0.256 3.883 50 intron 14323 NM_001104925 55251 PCMTD2 PCMTD2 62901674 chr21:9437391- 100 1005008 D337 0.22 2.906 112 intergenic -387941 NR_037421 MIR3648 intergenic 9438392 1 62 chr21:9438899- 1005008 D338 301 0.232 2.728 85 intergenic -386783 NR_037421 MIR3648 intergenic 9439200 62 chr21:15436148- ANKRD2 D339 575 0.238 2.246 65 intergenic -83670 NR_027270 391267 intergenic 15436723 0A11P chr21:30674834- D340 390 -0.231 2.512 8 intron -2531 NR_027655 571 BACH1 BACH1 30675224 chr21:31315685- D341 502 -0.241 2.766 6 intergenic -3654 NM_175611 2897 GRIK1 GRIK1 31316187 chr21:33961591- D342 385 -0.212 3.328 7 intergenic -3938 NM_144659 140290 TCP10L TCP10L 33961976 chr21:34673773- D343 395 0.336 2.781 5 intergenic -23244 NM_000629 3454 IFNAR1 IL10RB 34674168 chr21:34756882- D344 515 0.22 3.368 15 intergenic -18063 NM_005534 3460 IFNGR2 intergenic 34757397 chr21:36262421- D345 593 0.342 2.891 75 intron -1730 NM_001001890 861 RUNX1 RUNX1 36263014 chr21:38476940- D346 515 0.262 2.595 8 intron 21950 NM_003316 7267 TTC3 TTC3 38477455 chr21:43236051- D347 575 0.338 4.421 43 intron -49089 NM_020639 54101 RIPK4 PRDM15 43236626 chr21:47742584- D348 515 0.22 2.457 47 5' UTR 944 NM_058180 54058 C21orf58 PCNT, C21orf58 47743099 chr22:16190145- D349 356 -0.256 2.229 7 intergenic 97614 NM_001136213 23784 POTEH intergenic 16190501 chr22:19712690- GP1BB, SEPT5- D350 730 0.236 3.456 61 TTS 1989 NM_000407 2812 GP1BB 19713420 GP1BB, SEPT5 chr22:21530255- D351 698 -0.273 3.52 5 intergenic 73299 NR_037566 400892 BCRP2 intergenic 21530953 chr22:21531173- D352 330 -0.254 3.05 9 intergenic 74033 NR_037566 400892 BCRP2 intergenic 21531503 chr22:21533789- D353 315 -0.248 2.804 6 intergenic 76641 NR_037566 400892 BCRP2 intergenic 21534104

165

166

chr22:32101660- D354 325 -0.252 2.228 15 intron 44298 NM_173566 253143 PRR14L PRR14L 32101985 chr22:33281882- D355 475 0.309 2.866 12 intron 85317 NM_000362 7078 TIMP3 SYN3 33282357 chr22:40676734- D356 400 -0.237 2.783 7 intron -65570 NM_001123378 158 ADSL TNRC6B 40677134 chr22:44211099- D357 328 -0.223 2.674 8 intergenic -3046 NM_022785 64800 EFCAB6 EFCAB6 44211427 chr22:48730086- 1004229 D358 320 0.301 2.295 34 intergenic 60070 NR_036172 MIR3201 intergenic 48730406 16 chr3:13972732- 1001325 LOC1001 D359 370 -0.231 2.337 9 intergenic -1636 NR_036481 LOC100132526 13973102 26 32526 chr3:30898339- D360 624 -0.234 2.772 9 intron 37502 NM_207359 339896 GADL1 GADL1 30898963 chr3:33150809- D361 369 -0.237 2.339 6 intergenic -4457 NM_006371 10491 CRTAP CRTAP 33151178 chr3:46599590- D362 630 0.244 4.231 21 intron 8135 NM_024512 79442 LRRC2 LRRC2 46600220 chr3:52088932- D363 700 0.256 2.63 28 intron 1179 NM_001947 1849 DUSP7 DUSP7 52089632 chr3:54998632- D364 403 -0.253 2.184 8 intron -36761 NM_020678 57408 LRTM1 CACNA2D3 54999035 chr3:89157491- D365 315 -0.217 4.13 7 intron 974 NM_182644 2042 EPHA3 EPHA3 89157806 chr3:93776193- ARL13B, D366 367 -0.252 3.084 5 TTS 5455 NM_176815 200895 DHFRL1 93776560 DHFRL1 chr3:100210474- TMEM45 D367 316 -0.279 2.215 12 promoter-TSS -831 NM_018004 55076 TMEM45A 100210790 A chr3:110783860- 1005065 PVRL3- D368 369 -0.222 2.696 6 intron 4762 NR_045114 PVRL3-AS1 110784229 55 AS1 chr3:112712434- D369 365 -0.225 2.678 11 intron 2816 NM_138485 29083 GTPBP8 GTPBP8 112712799 chr3:124229077- D370 500 -0.294 3.709 5 intron -74179 NM_007064 8997 KALRN KALRN 124229577 chr3:125685462- D371 385 -0.227 2.448 6 intergenic -2374 NM_001012337 152015 ROPN1B ROPN1B 125685847 chr3:140662821- SLC25A3 D372 370 -0.208 3.367 5 intron 2344 NM_001104647 55186 SLC25A36 140663191 6

166

167

chr3:142294654- D373 600 -0.222 2.934 7 intron 2714 NM_001184 545 ATR ATR 142295254 chr3:155833456- D374 700 -0.247 3.117 10 intergenic -4531 NM_172160 7881 KCNAB1 KCNAB1 155834156 chr3:167046807- D375 685 -0.258 3.062 6 intron 50922 NM_001199202 79740 ZBBX ZBBX 167047492 chr3:170818668- D376 521 -0.225 2.651 10 intron 5620 NR_030295 693154 MIR569 TNIK 170819189 chr3:171001952- D377 386 -0.25 2.998 7 intron 176052 NM_001161564 23043 TNIK TNIK 171002338 chr3:176746328- TBL1XR D378 503 -0.225 3.055 5 intron 168469 NM_024665 79718 TBL1XR1 176746831 1 chr3:178793387- D379 323 -0.231 2.964 8 intergenic -3892 NM_022470 64393 ZMAT3 ZMAT3 178793710 chr3:182699611- DCUN1D D380 505 -0.26 3.195 12 intergenic -1537 NM_020640 54165 DCUN1D1 182700116 1 chr3:183750534- D381 305 0.423 2.914 7 promoter-TSS 67 NM_001163646 200909 HTR3D HTR3D 183750839 chr3:188882790- D382 705 -0.419 4.192 6 intergenic -6621 NM_198485 285386 TPRG1 intergenic 188883495 chr3:192130716- D383 428 -0.222 2.93 5 intron -4092 NM_021032 2257 FGF12 FGF12 192131144 chr3:194786013- D384 400 0.244 2.936 42 intergenic 205682 NM_152531 152002 XXYLT1 XXYLT1 194786413 chr3:195426815- D385 470 -0.251 3.372 9 TTS 778 NR_030296 693155 MIR570 MIR570 195427285 chr3:197676983- D386 500 0.225 2.457 27 promoter-TSS 181 NM_000996 6165 RPL35A IQCG, RPL35A 197677483 chr4:124613- D387 515 0.211 2.951 44 intron 71593 NM_001039127 255403 ZNF718 ZNF718 125128 chr4:698291- D388 305 0.207 2.686 47 intergenic -1130 NM_006315 10336 PCGF3 PCGF3 698596 chr4:1794245- D389 580 0.311 3.045 61 promoter-TSS -504 NM_000142 2261 FGFR3 FGFR3 1794825 chr4:2847892- D390 505 0.387 2.789 6 intron 2560 NM_014189 118 ADD1 ADD1 2848397 chr4:3042964- 1007503 D391 375 0.222 2.834 46 TTS 33090 NR_045414 HTT-AS1 GRK4 3043339 26

167

168

chr4:3252885- MSANTD D392 415 0.247 2.512 18 intron 2325 NM_001042690 345222 MSANTD1 3253300 1 chr4:10028678- D393 487 -0.224 3.092 12 intron -5807 NM_020041 56606 SLC2A9 SLC2A9 10029165 chr4:13605282- D394 414 -0.234 2.774 7 exon 23839 NM_148894 259282 BOD1L1 BOD1L1 13605696 chr4:41534117- D395 305 -0.234 3.438 11 intron -80650 NM_001112719 22998 LIMCH1 LIMCH1 41534422 chr4:47469443- COMMD D396 401 -0.242 3.064 10 intergenic -3967 NM_017845 54951 COMMD8 47469844 8 chr4:74489443- D397 812 -0.248 3.603 8 intergenic -3501 NM_001270392 166824 RASSF6 RASSF6 74490255 chr4:74715813- D398 315 -0.27 2.239 7 intergenic -3043 NM_002620 5197 PF4V1 PF4V1 74716128 chr4:76943285- D399 365 -0.251 3.305 6 intron 1222 NM_001565 3627 CXCL10 ART3, CXCL10 76943650 chr4:76993566- D400 600 -0.293 2.692 5 intron -1982 NM_001179 419 ART3 ART3 76994166 chr4:81947252- D401 575 -0.257 4.205 6 intergenic -4580 NM_001201 651 BMP3 BMP3 81947827 chr4:89051945- D402 380 -0.243 3.563 9 intron 27876 NM_004827 9429 ABCG2 ABCG2 89052325 chr4:90234239- D403 486 -0.221 2.666 10 intergenic -5321 NM_198281 285513 GPRIN3 intergenic 90234725 chr4:90753936- LOC6442 SNCA, D404 620 -0.264 3.74 5 intron -3306 NR_045481 644248 90754556 48 LOC644248 chr4:94747689- D405 490 -0.237 2.377 8 intergenic -2144 NM_005172 474 ATOH1 ATOH1 94748179 chr4:99063393- D406 330 -0.245 2.845 6 intron 833 NM_174952 285555 STPG2 C4orf37 99063723 chr4:100240587- D407 575 -0.247 3.318 5 intron 1698 NM_000668 125 ADH1B ADH1B 100241162 chr4:118009944- TRAM1L D408 402 -0.215 2.691 5 intergenic -3409 NM_152402 133022 TRAM1L1 118010346 1 chr4:118500576- D409 797 -0.289 4.044 6 intergenic -454526 NM_004784 9348 NDST3 intergenic 118501373 chr4:118528065- D410 591 -0.255 2.966 6 intergenic -427140 NM_004784 9348 NDST3 intergenic 118528656

168

169

chr4:118711658- D411 420 -0.219 3.132 6 intergenic -243632 NM_004784 9348 NDST3 intergenic 118712078 chr4:125591600- ANKRD5 D412 305 -0.204 2.606 10 exon 42135 NM_001167882 57182 ANKRD50 125591905 0 chr4:141296446- SCOC, D413 375 -0.244 2.371 8 intron 1969 NM_001153663 60592 SCOC 141296821 LOC100129858 chr4:155418051- D414 494 -0.266 3.062 5 intergenic -5368 NM_001142553 54798 DCHS2 intergenic 155418545 PLEKHG D415 chr5:84782-85252 470 0.235 3.18 58 intergenic -55356 NM_052909 153478 intergenic 4B chr5:2739664- D416 698 0.278 2.224 69 intergenic 11756 NM_033267 153572 IRX2 intergenic 2740362 chr5:36001428- D417 505 -0.237 2.318 6 promoter-TSS -550 NM_001171873 133688 UGT3A1 UGT3A1 36001933 chr5:56238259- D418 395 -0.302 3.1 8 intron 9498 NM_152622 166968 MIER3 MIER3 56238654 chr5:56285774- D419 790 -0.297 3.442 6 intergenic -38215 NM_152622 166968 MIER3 intergenic 56286564 chr5:65437589- D420 590 0.297 2.621 5 intergenic -2162 NM_001077199 140890 SREK1 SREK1 65438179 chr5:68974058- D421 965 -0.328 2.898 5 intron 31732 NR_027386 653188 GUSBP3 GUSBP3 68975023 chr5:68976563- D422 750 -0.268 3.297 6 intron 29334 NR_027386 653188 GUSBP3 GUSBP3 68977313 chr5:69555415- 154 D423 -0.25 3.226 8 intron 29817 NR_029426 11039 SMA4 SMA4 69556960 5 chr5:69704768- GTF2H2 D424 312 -0.228 3.259 5 intergenic -6273 NR_033417 653238 intergenic 69705080 B chr5:69849579- 136 1000490 D425 -0.295 3.113 5 intron 891 NR_033968 GUSBP9 SMA5, GUSBP9 69850939 0 76 chr5:69851754- 154 1000490 D426 -0.261 3.269 8 intron -1375 NR_033968 GUSBP9 SMA5, GUSBP9 69853296 2 76 chr5:70555908- 121 1000490 D427 -0.236 3.098 9 intergenic -1393 NR_033968 GUSBP9 GUSBP9 70557123 5 76 chr5:75469940- D428 415 0.214 2.427 29 intron 90842 NM_014979 22987 SV2C SV2C 75470355 chr5:75696649- D429 700 -0.253 3.761 7 intergenic -2150 NM_006633 10788 IQGAP2 IQGAP2 75697349

169

170

chr5:82368664- TMEM16 TMEM167A, D430 420 -0.224 2.402 10 intron 4398 NM_174909 153339 82369084 7A XRCC4 TMEM161B- chr5:87567189- 1005058 TMEM16 D431 375 -0.232 2.912 7 intron 2535 NR_039994 AS1, 87567564 94 1B-AS1 TMEM161B chr5:94885207- D432 909 -0.246 3.413 8 intron 5048 NM_014639 9652 TTC37 TTC37, ARSK 94886116 chr5:95065474- RHOBTB D433 590 -0.235 2.509 6 intergenic -1081 NM_014899 22836 RHOBTB3 95066064 3 chr5:109962533- TMEM23 D434 595 -0.222 3.386 5 intron 99620 NM_001039763 642987 TMEM232 109963128 2 chr5:115292255- D435 630 -0.235 3.162 6 intergenic -5581 NM_173800 206338 AQPEP intergenic 115292885 chr5:126114144- D436 310 0.22 2.879 50 intron 1454 NM_001198557 4001 LMNB1 LMNB1 126114454 chr5:131485248- D437 570 -0.252 2.746 7 intergenic 76048 NM_000758 1437 CSF2 intergenic 131485818 chr5:132228740- D438 380 -0.239 3.48 9 intron 19572 NM_052971 116842 LEAP2 AFF4 132229120 PCDHB15, chr5:140620404- PCDHB1 D439 490 0.237 2.935 31 non-coding 960 NR_001282 84054 PCDHB18, 140620894 9P PCDHB19P PCDHGA2, chr5:140719011- PCDHGA D440 615 0.222 2.382 33 exon 964 NM_018915 56113 PCDHGA1, 140719626 2 PCDHGA3 chr5:142174675- ARHGAP D441 310 -0.203 2.758 5 intron 24538 NM_001135608 23092 ARHGAP26 142174985 26 chr5:147260474- SCGB3A D442 305 -0.21 3.073 6 intron 2352 NM_054023 117156 SCGB3A2 147260779 2 chr5:156487454- D443 365 -0.24 3.115 6 intergenic -1666 NM_001173393 26762 HAVCR1 HAVCR1 156487819 chr5:156570589- D444 305 0.262 3.601 27 promoter-TSS -820 NM_004270 9443 MED7 MED7 156570894 chr5:161270539- D445 480 -0.235 2.684 8 intergenic -3418 NM_000806 2554 GABRA1 GABRA1 161271019 chr5:175393973- D446 545 -0.246 3.534 12 exon 1300 NM_032361 84321 THOC3 THOC3 175394518

170

171

chr5:175488222- FAM153 D447 405 0.213 3.694 64 intergenic -2288 NM_001265615 202134 intergenic 175488627 B chr5:177303153- LOC7285 D448 625 -0.259 3.5 17 intron 1203 NR_003615 728554 LOC728554 177303778 54 chr5:180648315- 1006163 TRIM41, D449 705 0.228 3.595 33 TTS 966 NR_039781 MIR4638 180649020 42 MIR4638 chr5:180794543- OR4F16, OR4F3, D450 645 -0.243 2.814 8 exon 577 NM_001005224 26683 OR4F3 180795188 OR4F29 chr5:180795933- OR4F16, OR4F3, D451 916 -0.27 3.436 5 intergenic 2103 NM_001005224 26683 OR4F3 180796849 OR4F29 chr6:2902234- SERPINB D452 807 0.234 3.144 27 intron 909 NM_004155 5272 SERPINB9 2903041 9 chr6:4020994- D453 495 0.241 2.985 35 promoter-TSS -328 NM_003913 8899 PRPF4B PRPF4B 4021489 chr6:10319513- 1001302 LOC1001 D454 534 0.261 4.52 42 intergenic -92771 NR_033910 intergenic 10320047 75 30275 chr6:10407836- 1001302 LOC1001 TFAP2A, D455 375 -0.217 3.782 7 intron -4528 NR_033910 10408211 75 30275 LOC100130275 chr6:25650732- D456 330 -0.228 2.942 5 intergenic -1532 NM_006998 10590 SCGN SCGN 25651062 chr6:25878141- D457 530 -0.228 2.861 5 intergenic -3935 NM_006632 10786 SLC17A3 SLC17A3 25878671 chr6:29530427- D458 477 -0.243 3.013 10 intergenic -2963 NM_006398 10537 UBD UBD 29530904 chr6:33776419- D459 498 -0.254 3.503 13 intergenic -4875 NM_002418 4295 MLN MLN 33776917 chr6:34499206- D460 525 0.214 2.285 33 exon 16819 NM_001199583 29993 PACSIN1 PACSIN1 34499731 chr6:56817935- D461 395 -0.232 2.421 12 intergenic -1641 NM_152731 221336 BEND6 BEND6 56818330 chr6:56821525- D462 620 -0.244 3.064 6 intron 2062 NM_152731 221336 BEND6 BEND6 56822145 chr6:70505077- D463 623 -0.242 3.275 5 intron 1661 NM_018368 55788 LMBRD1 LMBRD1 70505700 chr6:72589922- D464 490 -0.221 2.539 8 intergenic -6239 NM_014989 22999 RIMS1 intergenic 72590412 chr6:73733245- 1004230 D465 310 -0.208 2.377 5 intron -55924 NR_036244 MIR4282 KCNQ5 73733555 05

171

172

chr6:99803202- D466 404 -0.275 3.094 6 intergenic -5873 NM_032511 84553 FAXC intergenic 99803606 chr6:109767871- MICAL1, D467 410 0.243 2.679 16 intron -5702 NM_001111298 285755 PPIL6 109768281 SMPD2 chr6:116450645- COL10A1, D468 302 -0.228 2.24 8 intron -3500 NM_000493 1300 COL10A1 116450947 NT5DC1 chr6:132335452- D469 699 -0.241 4.041 6 intergenic -63283 NM_001901 1490 CTGF intergenic 132336151 chr6:132534976- 1005072 LOC1005 D470 695 -0.236 2.755 7 intergenic 80205 NR_038981 inter 132535671 54 07254 chr6:135308670- 1005008 D471 310 -0.237 2.404 6 exon -8255 NR_037435 MIR3662 HBS1L 135308980 80 chr6:149198979- 1001281 LOC1001 D472 515 0.209 4.052 48 intron 86584 NR_038408 UST 149199494 76 28176 chr6:151711985- D473 725 0.327 3.029 99 intron 330 NM_020861 57621 ZBTB2 ZBTB2 151712710 chr6:152963082- D474 480 -0.256 3.287 11 intergenic -4788 NM_182961 23345 SYNE1 SYNE1 152963562 chr6:153322336- D475 505 -0.258 3.58 10 intron 1337 NM_019041 54516 MTRF1L MTRF1L 153322841 chr6:156952081- D476 475 0.286 2.811 39 intergenic -146746 NM_017519 57492 ARID1B intergenic 156952556 chr6:165993899- D477 578 -0.309 3.988 6 intron 81400 NM_001130690 10846 PDE10A PDE10A 165994477 chr6:167410746- 1005008 FGFR1OP, D478 375 -0.282 2.362 8 TTS 467 NR_037504 MIR3939 167411121 57 MIR3939 chr6:167835314- D479 323 0.228 2.937 62 intergenic -37477 NM_004610 6953 TCP10 intergenic 167835636 chr6:170147361- D480 500 -0.233 3.02 6 intron 4027 NM_174910 6991 TCTE3 TCTE3, C6orf70 170147861 chr6:170474407- LOC1544 D481 500 0.223 2.92 46 intergenic 97000 NR_002787 154449 intergenic 170474907 49 chr7:6691952- D482 830 0.232 2.675 99 intergenic 36840 NM_017560 54753 ZNF853 intergenic 6692782 chr7:12412350- D483 969 -0.237 4.086 8 exon 31018 NM_001135924 221806 VWDE VWDE 12413319 chr7:25025686- D484 328 -0.211 2.216 5 intergenic -6090 NM_145321 26031 OSBPL3 intergenic 25026014

172

173

chr7:26431539- LOC4412 D485 400 -0.23 3.431 9 intergenic -11369 NR_015364 441204 intergenic 26431939 04 chr7:27327949- D486 390 -0.256 2.34 6 intergenic 45980 NM_001989 2128 EVX1 intergenic 27328339 chr7:27361924- D487 398 -0.228 2.484 6 intergenic 79959 NM_001989 2128 EVX1 intergenic 27362322 chr7:27422529- D488 325 -0.24 3.047 9 intergenic 140527 NM_001989 2128 EVX1 intergenic 27422854 chr7:49813357- D489 630 0.258 2.969 87 5' UTR 415 NM_198570 375567 VWC2 VWC2 49813987 chr7:64595524- D490 639 -0.286 4.106 6 intergenic 97111 NR_033416 643180 CCT6P3 intergenic 64596163 chr7:65106699- D491 638 -0.282 3.608 5 intergenic -5759 NR_027392 644619 INTS4L2 LOC441242 65107337 chr7:89883719- D492 470 -0.242 2.49 10 intron 9466 NM_001160138 79846 C7orf63 C7orf63 89884189 chr7:98610088- 1005008 D493 633 0.257 3.113 26 3' UTR 131131 NR_037403 MIR3609 TRRAP 98610721 19 chr7:99479857- D494 690 -0.218 3.402 6 intergenic -5546 NM_001005276 81392 OR2AE1 intergenic 99480547 chr7:100763435- SERPINE D495 370 0.347 2.117 13 intergenic -6750 NM_000602 5054 intergenic 100763805 1 chr7:107381088- D496 400 -0.237 2.906 5 intergenic -2991 NM_024814 79872 CBLL1 CBLL1 107381488 chr7:114203667- 1005008 D497 370 -0.222 3.026 5 intron -89548 NR_037439 MIR3666 FOXP2 114204037 96 chr7:114220756- 1005008 D498 400 -0.325 2.511 6 intron -72444 NR_037439 MIR3666 FOXP2 114221156 96 chr7:114507033- D499 315 -0.21 2.774 12 intergenic -55019 NM_001166346 29969 MDFIC intergenic 114507348 chr7:114581249- D500 395 -0.246 2.838 6 intron 19237 NM_001166345 29969 MDFIC MDFIC 114581644 chr7:114688062- D501 601 -0.251 3.329 6 intergenic 126153 NM_001166345 29969 MDFIC intergenic 114688663 chr7:114876936- D502 411 -0.25 3.001 5 intergenic 314932 NM_001166345 29969 MDFIC intergenic 114877347 chr7:116089354- D503 698 -0.238 3.621 6 intergenic -49952 NM_198212 858 CAV2 intergenic 116090052

173

174

chr7:116243038- D504 314 -0.23 2.215 9 intergenic -69264 NM_000245 4233 MET intergenic 116243352 chr7:117009622- D505 625 -0.234 2.344 8 intron -46591 NR_024047 7472 WNT2 ASZ1 117010247 chr7:117141954- D506 368 -0.226 3.16 7 intron 22121 NM_000492 1080 CFTR CFTR 117142322 chr7:117442255- CTTNBP D507 369 -0.239 2.793 5 intron 71122 NM_033427 83992 CTTNBP2 117442624 2 chr7:117627456- CTTNBP D508 700 -0.264 3.482 7 intergenic -114245 NM_033427 83992 intergenic 117628156 2 chr7:124535340- D509 605 -0.245 2.79 6 intron 34395 NM_001042594 25913 POT1 POT1 124535945 chr7:126078689- D510 680 -0.247 3.246 8 non-coding 619209 NR_030323 693177 MIR592 GRM8 126079369 chr7:126187791- D511 365 -0.22 2.478 8 intron 510265 NR_030323 693177 MIR592 GRM8 126188156 chr7:126427756- D512 330 -0.212 2.997 5 intron 270317 NR_030323 693177 MIR592 GRM8 126428086 chr7:127175557- D513 315 -0.219 2.893 5 intergenic 49940 NM_024523 79571 GCC1 intergenic 127175872 chr7:128765686- LOC4078 D514 490 0.342 2.656 7 promoter-TSS -394 NR_002144 407835 LOC407835 128766176 35 chr7:129244625- D515 475 -0.349 4.137 8 intergenic -6693 NM_005011 4899 NRF1 intergenic 129245100 chr7:129981463- D516 425 -0.237 3.403 11 intergenic -2955 NM_001127442 93979 CPA5 CPA5 129981888 MEST, chr7:130132572- D517 600 0.24 3.01 46 intron 973 NM_001253900 4232 MEST MESTIT1, 130133172 MIR335 chr7:143699592- D518 330 -0.255 3.49 6 intergenic -1333 NM_001005281 135946 OR6B1 OR6B1 143699922 chr7:148844069- D519 430 0.258 2.509 61 promoter-TSS -276 NM_170686 57541 ZNF398 ZNF398 148844499 chr7:152590887- D520 405 0.213 2.727 30 intergenic 134255 NR_073000 57180 ACTR3B intergenic 152591292 chr7:154865062- 1001282 LOC1001 HTR5A, D521 390 0.205 2.345 10 intron -1990 NR_038945 154865452 64 28264 LOC100128264

174

175

chr8:1048469- LOC2860 D522 505 0.239 2.642 58 intergenic 202106 NR_033895 286083 intergenic 1048974 83 chr8:1321117- LOC2860 D523 485 0.245 2.765 54 intergenic -70532 NR_033895 286083 intergenic 1321602 83 chr8:1818044- ARHGEF D524 305 0.205 2.931 31 intron 46047 NM_014629 9639 ARHGEF10 1818349 10 chr8:6291306- D525 365 -0.223 2.155 9 intron 27375 NM_024596 79648 MCPH1 MCPH1 6291671 chr8:6795035- D526 400 -0.233 3.225 12 intron 551 NM_001925 1669 DEFA4 DEFA4 6795435 chr8:9950169- D527 530 -0.243 3.742 6 intron -2632 NM_001199729 4482 MSRA MSRA 9950699 chr8:11246567- D528 365 -0.233 2.128 17 intron 20838 NR_026814 83656 C8orf12 C8orf12 11246932 chr8:12597658- 1005008 MIR3926- D529 410 -0.23 2.841 7 intron -13050 NR_037492 LONRF1 12598068 70 1 chr8:14722653- D530 325 -0.226 2.639 8 intron -11796 NR_029875 494332 MIR383 SGCZ 14722978 chr8:16051096- D531 365 -0.238 4.191 13 promoter-TSS -978 NM_138715 4481 MSR1 MSR1 16051461 chr8:17015551- D532 387 -0.21 3.02 9 intron 1908 NM_016353 51201 ZDHHC2 ZDHHC2 17015938 chr8:39437313- ADAM18, D533 400 -0.221 3.013 8 non-coding -4574 NM_001190956 8749 ADAM18 39437713 LOC100130964 chr8:41653347- D534 325 -0.215 2.922 9 intron 1631 NM_020475 286 ANK1 ANK1 41653672 chr8:59410332- D535 490 -0.254 3.237 6 intron 2143 NM_000780 1581 CYP7A1 CYP7A1 59410822 chr8:71317824- D536 421 -0.223 2.573 10 intergenic -2014 NM_006540 10499 NCOA2 NCOA2 71318245 chr8:72266841- D537 380 -0.217 3.531 7 exon 1948 NM_172058 2138 EYA1 EYA1 72267221 chr8:79421345- D538 410 -0.269 3.874 5 intergenic -6786 NM_181839 5569 PKIA intergenic 79421755 chr8:87876176- D539 400 -0.221 3.206 7 intergenic -2300 NM_173538 168975 CNBD1 CNBD1 87876576 chr8:92076812- D540 398 -0.221 3.39 8 intergenic -5413 NM_016023 51633 OTUD6B intergenic 92077210

175

176

chr8:92257633- D541 415 -0.231 2.183 9 intergenic -3676 NM_134266 115111 SLC26A7 SLC26A7 92258048 chr8:94713994- FAM92A LINC00535, D542 390 -0.215 2.555 6 intron 1416 NM_145269 137392 94714384 1 FAM92A1 chr8:98785186- LAPTM4 D543 330 0.239 2.584 20 intergenic -2458 NM_018407 55353 LAPTM4B 98785516 B chr8:104507492- D544 305 -0.218 2.648 8 intergenic -5332 NM_001100117 9699 RIMS2 intergenic 104507797 chr8:118836458- D545 315 -0.211 2.757 8 intron 287443 NM_000127 2131 EXT1 EXT1 118836773 chr8:119266746- D546 325 -0.219 2.199 10 intron -142850 NM_000127 2131 EXT1 SAMD12 119267071 chr8:124214717- 1001317 LOC1001 FAM83A, D547 666 -0.242 3.702 9 promoter-TSS -67 NR_024479 124215383 26 31726 LOC100131726 chr8:133872708- D548 530 -0.203 2.259 9 intergenic -6232 NM_003235 7038 TG intergenic 133873238 chr8:143625560- D549 605 0.278 2.995 61 3' UTR 69971 NM_015193 23237 ARC BAI1 143626165 chr8:144631520- D550 315 0.307 2.663 23 intergenic -3880 NM_001166237 79792 GSDMD GSDMD 144631835 chr8:145317415- HEATR7A, D551 471 0.225 3.656 34 TTS -3867 NM_001080514 642658 SCXB 145317886 SCXB, SCXA chr8:145460790- FAM203 D552 366 0.219 2.555 29 intergenic 23093 NM_016458 51236 intergenic 145461156 A chr9:4840010- D553 410 -0.215 2.469 6 intron -10082 NR_029836 406894 MIR101-2 RCL1 4840420 chr9:15508453- D554 320 -0.217 2.892 8 intron 1674 NM_021144 11168 PSIP1 PSIP1 15508773 chr9:27102462- D555 410 -0.288 3.339 7 intergenic -6480 NM_000459 7010 TEK intergenic 27102872 chr9:27521612- D556 305 0.231 2.682 13 intron -2548 NM_020124 56832 IFNK IFNK, MOB3B 27521917 chr9:41958910- MGC218 KGFLP2, D557 600 -0.259 2.416 5 non-coding -4134 NR_015363 389741 41959510 81 MGC21881 chr9:42023436- D558 310 -0.211 2.638 6 intergenic -4007 NR_003670 654466 KGFLP2 KGFLP2 42023746 chr9:42024036- D559 515 -0.219 2.986 11 intergenic -4709 NR_003670 654466 KGFLP2 KGFLP2 42024551

176

177

chr9:42473899- 1001330 FAM95B D560 330 -0.235 3.386 11 non-coding 5475 NR_026759 FAM95B1 42474229 36 1 chr9:42699546- 1000365 FOXD4L D561 858 -0.268 3.182 5 intergenic -17259 NM_001099279 intergenic 42700404 19 2 chr9:43089861- ANKRD2 ANKRD20A2, D562 355 -0.226 2.494 5 TTS 43506 NM_001012419 441425 43090216 0A3 ANKRD20A3 chr9:44983637- 1001329 D563 999 -0.29 3.595 6 intergenic -6100 NR_027421 FAM27C intergenic 44984636 48 chr9:46682564- D564 860 -0.224 2.727 12 intergenic -4563 NR_003674 387628 KGFLP1 KGFLP1 46683424 chr9:67799758- 1001331 D565 869 -0.287 3.555 5 intergenic -6003 NR_027422 FAM27B intergenic 67800627 21 chr9:69217378- FOXD4L D566 345 -0.205 3.019 5 intron -15346 NM_001085476 653404 CBWD6 69217723 6 chr9:69219107- FOXD4L D567 718 -0.263 3.001 5 intron -17262 NM_001085476 653404 CBWD6 69219825 6 chr9:69424001- ANKRD2 D568 355 -0.234 3.472 5 3' UTR 42197 NM_001098805 728747 ANKRD20A4 69424356 0A4 chr9:69787085- 1001339 LOC1001 D569 320 0.215 2.39 94 intergenic 135884 NR_024443 intergenic 69787405 20 33920 chr9:70195445- 110 FOXD4L D570 -0.249 3.006 5 intergenic -17180 NM_001126334 653427 intergenic 70196545 0 5 chr9:70446362- 110 1000365 FOXD4L CBWD3, D571 -0.272 3.701 5 intron -17182 NM_001099279 70447464 2 19 2 CBWD5 chr9:70883054- D572 671 -0.338 4.185 5 intron 26550 NM_201453 445571 CBWD3 CBWD3 70883725 chr9:70899486- FOXD4L D573 484 -0.237 2.684 7 intron -18055 NM_199135 286380 CBWD3 70899970 3 chr9:71392971- FAM122 FAM122A, D574 415 -0.232 2.864 6 intron -1786 NM_138333 116224 71393386 A PIP5K1B chr9:82180820- D575 375 -0.244 2.339 7 intergenic -5871 NM_007005 7091 TLE4 intergenic 82181195 chr9:85953365- D576 494 -0.289 3.342 6 intron -6736 NM_001244961 257019 FRMD3 FRMD3 85953859 chr9:94710496- D577 375 0.221 4.192 23 intron 1761 NM_004560 4920 ROR2 ROR2 94710871 chr9:94716003- D578 616 -0.263 3.475 10 intergenic -3867 NM_004560 4920 ROR2 ROR2 94716619

177

178

chr9:107378101- D579 405 -0.21 3.152 5 intergenic 2182 NM_001001956 286362 OR13C9 OR13C9 107378506 chr9:113317535- D580 500 -0.22 2.925 7 intron 24375 NM_153366 79987 SVEP1 SVEP1 113318035 chr9:114373142- LRRC37 D581 325 -0.228 2.003 9 intron 2529 NR_034087 652972 LRRC37A5P 114373467 A5P DNAJC25- chr9:114416816- D582 505 -0.224 3.141 11 TTS -6783 NM_001198664 2790 GNG10 GNG10, 114417321 DNAJC25 chr9:115824896- D583 936 0.247 2.805 40 intergenic -6368 NM_003408 7539 ZFP37 intergenic 115825832 chr9:124988166- D584 395 0.217 2.634 66 intron 1502 NM_001242334 26468 LHX6 LHX6 124988561 chr9:130980777- D585 515 0.254 3.833 44 intron -14372 NM_001131015 25792 CIZ1 DNM1 130981292 chr9:131901819- D586 470 0.248 3.064 18 intron 28087 NM_001193397 5524 PPP2R4 PPP2R4 131902289 chr9:132138020- D587 400 -0.257 3.575 7 intergenic 54925 NM_001012715 414318 C9orf106 intergenic 132138420 chr9:139690341- KIAA198 KIAA1984, D588 570 0.299 2.646 11 promoter-TSS -164 NM_001039374 84960 139690911 4 TMEM141 chr9:140375158- D589 370 0.226 2.263 30 intron -21557 NM_001130969 26012 NELF PNPLA7 140375528

178

179

Table E4. Primers used for bisulfite sequencing. Amplicon Extension Length No. Closest Primers * coordinates Temp. (bp) CpGs gene 5’-3’ (hg19) (C) † GAGTTTTTTTTTTAATTTTTAATGG chr2:50,201,165-50,201,630 466 16 NRNX1 ATTTCCCCTCTATATTTAAC 50 AAATTGATTTTGGTAAAGTAG chr10:131,696,855-31,697,384 530 30 EBF3 AACACTCTACTAATAAACAAC 50 TTTAGTAGTTGAATTTGAGTGGT chr19:805,877-806,280 404 45 PTBP1 CTACCCTAAATACCTATAAAA 50 GTGAGTTAAGGTTATGTTATTGTT chr16:90,148,245-90,148,740 496 29 PRDM7 AAAACTTTTTCCTCATCTTC 53 GATGTTTTTGTGTAGTTTTATG chr21:43,236,140-43,236,400 261 16 PRDM15 ATATTACAATCAATCCCTCC 50 TAGGAGTAGTATAGTTATTAGGGTT chr1:205,819,150-205,819,562 413 16 PM20D1 ATTTTACTTTTCCTCCTTTA 50 TAGGAGAAGGAAATAAGATA chr7:114,220,317-114,220,913 597 10 FOXP2 ACTCAATACTAACTACCTAC 50 GAAGATAAAGAAGATAGTAAG chr6:165,994,226-165,994,582 357 5 PDE10A TCTCCATAATATCACCATAA 50 GTTTTGGGAGGTGTTTAGTATTTTATT chr4:100,360,216-100,360,593 378 4 ADH7 ATTTCTCTCATATTTCTACTTCC 53 GGTTTTATGTTTGTAGTGATTAAG chr22:22,342,517-22,342,893 377 5 TOP3B ATTACCTACTATTACCCTAT 50 GGGGAAATATTTTTGGTTTATGT chr17:5,019,708-5,020,144 437 22 USP6 CACTTACCATCAATTTTCTCTT 57 TTGGTTTTTAGTTAAGATTGTGT chrX:139,591,526-139,591,985 460 14 SOX3 ACTTTTTTAAAAATACCCCC 53 GTTGGTGGAAATGAAGTTATAATT chrX:70,752,148-70,752,507 360 15 OGT CAATCAATTCAAAAAATTACCTCTC 53 GGGTTTTGGTTTTAGAGGTTTATTTA chrX:153,363,724-153,363,953 230 6 MECP2 TCTTCCAAACAAAACTAACATTACC 59 TTTTTTGGGTATTAGTAATAGG chrX:43,514,821-43,515,213 393 12 MAOA RCCTACCTTAACACTAAAAA 50 GAAGATAAGAAAGGGGTATTATT chrX:75,392,703-75,393,068 366 19 CXORF26 AAAAAAAAAAAAAACTCCACC 53 GTTAAAAATAAATAAAGGGTTTAG chr15:67,356,796-67,357,298 503 15 SMAD3 ACTAAAAACACCAATAACAACCT 54 *Forward (top) and Reverse (bottom) primers used

179

180

Table E5: Functional annotation of asthma-associated DMRs in IIS children. * Overrepresentation of biological functions was evaluated using Fisher's Exact Test as implemented in Ingenuity Pathway Analysis.

CATEGORY FUNCTION P-VALUE* MOLECULES # AKAP1, AKT3, ATR, BACH1, BDNF, CDC42EP3, CFTR, CYP7A1, DNM1, DNM2, DUSP6, DYNLL1, EP400, EYA1, FASN, FGFR3, GFI1, CELL DEATH AND apoptosis 1.54E-07 GNAS, IFNK, LAMA5, LDHA, MSR1, NCOA2, NEUROD1, NFATC1, 36 SURVIVAL POT1, PRDM16, PRKCZ, RELB, RUNX1, SMAD3, SNCA, STK11, TMBIM4, UBD, XRCC4 innervation of outer CELL MORPHOLOGY 1.80E-07 BDNF, FGFR3, GABRA5, GFI1 4 hair cells CELLULAR innervation of outer 1.80E-07 BDNF, FGFR3, GABRA5, GFI1 4 MOVEMENT hair cells NERVOUS SYSTEM innervation of outer DEVELOPMENT AND 1.80E-07 BDNF, FGFR3, GABRA5, GFI1 4 hair cells FUNCTION AKAP1, AKT3, ATR, BACH1, BDNF, CDC42EP3, CYP7A1, DNM1, DNM2, DUSP6, DYNLL1, EP400, FASN, FGFR3, GABRA5, GFI1, CELL DEATH AND necrosis 1.75E-06 GNAS, LAMA5, LDHA, MSR1, NEUROD1, NFATC1, POT1, PRKCZ, 33 SURVIVAL RELB, RUNX1, SMAD3, SNCA, STK11, STK25, TMBIM4, UBD, XRCC4 AKAP1, AKT3, ATR, BACH1, BDNF, CALM1 (includes others), CDC42EP3, CFTR, CYP7A1, DNM1, DNM2, DUSP6, DYNLL1, EP400, CELL DEATH AND cell death 1.79E-06 EYA1, FASN, FGFR3, GABRA5, GFI1, GNAS, IFNK, LAMA5, LDHA, 39 SURVIVAL MSR1, NCOA2, NEUROD1, NFATC1, POT1, PRDM16, PRKCZ, RELB, RUNX1, SMAD3, SNCA, STK11, STK25, TMBIM4, UBD, XRCC4 ABCC9, AKT3, ATR, BDNF, CFL1, CFTR, COL10A1, CXCL10, FASN, ORGANISMAL organismal death 3.55E-06 FGFR3, GABRA1, GFI1, GNAS, KCNAB1, MSR1, NFATC1, RELB, 25 SURVIVAL RUNX1, SMAD3, SNCA, STK11, SULF2, TRRAP, UBD, XRCC4

180

181

congenital anomaly of DEVELOPMENTAL ABCC9, ACAN, COL10A1, DNM2, EYA1, FGFR3, GABRA1, GABRA5, musculoskeletal 4.25E-06 13 DISORDER GNAS, PCNT, SMAD3, STK25, SYNE1 system

SKELETAL AND congenital anomaly of ABCC9, ACAN, COL10A1, DNM2, EYA1, FGFR3, GABRA1, GABRA5, MUSCULAR musculoskeletal 4.25E-06 13 GNAS, PCNT, SMAD3, STK25, SYNE1 DISORDERS system

DEVELOPMENTAL congenital anomaly of ABCC9, ACAN, COL10A1, EYA1, FGFR3, GABRA1, GABRA5, GNAS, 7.18E-06 10 DISORDER skeletal bone PCNT, SMAD3

SKELETAL AND congenital anomaly of ABCC9, ACAN, COL10A1, EYA1, FGFR3, GABRA1, GABRA5, GNAS, MUSCULAR 7.18E-06 10 skeletal bone PCNT, SMAD3 DISORDERS

CONNECTIVE TISSUE congenital anomaly of ABCC9, ACAN, COL10A1, EYA1, FGFR3, GABRA1, GABRA5, GNAS, 7.18E-06 10 DISORDERS skeletal bone PCNT, SMAD3

CELL-TO-CELL BDNF, CFL1, CXCL10, DDOST, FGFR3, GABRA1, GFI1, GNAS, IFNK, SIGNALING AND activation of cells 7.27E-06 MSR1, PRKCZ, RELB, RORA, RUNX1, SMAD3, SNCA, STK11, 18 INTERACTION XRCC4 ABCC9, AKT3, ATR, BDNF, CALM1 (includes others), CFDP1, CFL1, CXCL10, DNM2, DUSP6, EP400, EYA1, FASN, FGFR3, GFI1, GNAS, CELLULAR GROWTH proliferation of cells 1.03E-05 IFNK, LAMA5, LDHA, MSR1, NCOA2, NEUROD1, NFATC1, PALM, 39 AND PROLIFERATION PFKP, POT1, PRDM16, PRKCZ, RELB, RORA, RTKN2, RUNX1, SMAD3, SNCA, STK11, SULF2, TNIK, TRRAP, XRCC4 CELLULAR BDNF, CFL1, DNM1, DNM2, DYNLL1, EYA1, GNAS, GRK4, LAMA5, organization of ASSEMBLY AND 1.12E-05 NEUROD1, PACSIN1, PALM, PCNT, PRKCZ, RELB, SNCA, STK11, 20 cytoskeleton ORGANIZATION SULF2, SYNE1, TNIK BDNF, CFL1, DNM1, DNM2, DYNLL1, EYA1, GNAS, GRK4, LAMA5, CELLULAR FUNCTION organization of 1.12E-05 NEUROD1, PACSIN1, PALM, PCNT, PRKCZ, RELB, SNCA, STK11, 20 AND MAINTENANCE cytoskeleton SULF2, SYNE1, TNIK

181

182

DEVELOPMENTAL skeletal dysplasia 1.19E-05 ABCC9, ACAN, COL10A1, FGFR3, GNAS, PCNT 6 DISORDER SKELETAL AND MUSCULAR skeletal dysplasia 1.19E-05 ABCC9, ACAN, COL10A1, FGFR3, GNAS, PCNT 6 DISORDERS CONNECTIVE TISSUE skeletal dysplasia 1.19E-05 ABCC9, ACAN, COL10A1, FGFR3, GNAS, PCNT 6 DISORDERS synthesis of nitric BDNF, CALM1 (includes others), CFTR, DNM2, DYNLL1, FASN, IFNK, CELL SIGNALING 1.26E-05 9 oxide RORA, SNCA SMALL MOLECULE synthesis of nitric BDNF, CALM1 (includes others), CFTR, DNM2, DYNLL1, FASN, IFNK, 1.26E-05 9 BIOCHEMISTRY oxide RORA, SNCA DEVELOPMENTAL hypoplasia of thymus 1.33E-05 COL10A1, GFI1, RELB, RUNX1, SMAD3, XRCC4 6 DISORDER gland IMMUNOLOGICAL hypoplasia of thymus 1.33E-05 COL10A1, GFI1, RELB, RUNX1, SMAD3, XRCC4 6 DISEASE gland CELLULAR BDNF, CFL1, DNM1, DNM2, DYNLL1, EYA1, GNAS, GRK4, LAMA5, organization of ASSEMBLY AND 1.43E-05 NEUROD1, PACSIN1, PALM, PCNT, PRKCZ, RELB, SNCA, STK11, 21 cytoplasm ORGANIZATION STK25, SULF2, SYNE1, TNIK BDNF, CFL1, DNM1, DNM2, DYNLL1, EYA1, GNAS, GRK4, LAMA5, CELLULAR FUNCTION organization of 1.43E-05 NEUROD1, PACSIN1, PALM, PCNT, PRKCZ, RELB, SNCA, STK11, 21 AND MAINTENANCE cytoplasm STK25, SULF2, SYNE1, TNIK CELLULAR BDNF, CFL1, DNM1, DNM2, EYA1, GNAS, GRK4, LAMA5, ASSEMBLY AND microtubule dynamics 1.82E-05 NEUROD1, PACSIN1, PALM, PCNT, PRKCZ, RELB, SNCA, STK11, 18 ORGANIZATION SULF2, SYNE1 BDNF, CFL1, DNM1, DNM2, EYA1, GNAS, GRK4, LAMA5, CELLULAR FUNCTION microtubule dynamics 1.82E-05 NEUROD1, PACSIN1, PALM, PCNT, PRKCZ, RELB, SNCA, STK11, 18 AND MAINTENANCE SULF2, SYNE1 EMBRYONIC size of embryonic 2.31E-05 EYA1, FGFR3, LAMA5, NFATC1, STK11 5 DEVELOPMENT tissue size of embryonic TISSUE MORPHOLOGY 2.31E-05 EYA1, FGFR3, LAMA5, NFATC1, STK11 5 tissue

182

183

DEVELOPMENTAL Seckel syndrome 4 2.43E-05 ATR, PCNT 2 DISORDER HEREDITARY Seckel syndrome 4 2.43E-05 ATR, PCNT 2 DISORDER NERVOUS SYSTEM quantity of sensory DEVELOPMENT AND 2.48E-05 BDNF, EYA1, FGFR3, GFI1, NEUROD1 5 neurons FUNCTION quantity of sensory TISSUE MORPHOLOGY 2.48E-05 BDNF, EYA1, FGFR3, GFI1, NEUROD1 5 neurons AKT3, CFTR, CXCL10, GABRA1, GABRA5, GFI1, GNAS, IFNK, CELLULAR FUNCTION cellular homeostasis 3.07E-05 MSR1, NEUROD1, NFATC1, PRKCZ, RELB, RORA, RUNX1, SIK3, 21 AND MAINTENANCE SMAD3, SNCA, STK11, UBD, XRCC4 BDNF, CFL1, CXCL10, EYA1, FGFR3, GFI1, GNAS, IFNK, NEUROD1, CELLULAR differentiation of cells 3.15E-05 NFATC1, PACSIN1, PRDM16, PRKCZ, RELB, RORA, RTKN2, RUNX1, 26 DEVELOPMENT SCGB3A2, SIX2, SMAD3, SNCA, STK11, SYNE1, TTC3, UBD, XRCC4 ADH1B, AKAP1, AKT3, CALM1 (includes others), CFTR, CXCL10, DUSP6, DYNLL1, EP400, EYA1, FASN, FGFR3, FH, FRMD3, GNAS, CANCER carcinoma 3.15E-05 GSPT1, LAMA5, LDHA, MSR1, NFATC1, PFKP, POLE2, PRKCZ, 33 RELB, RUNX1, SHANK2, SMAD3, STK11, SYNE1, TMBIM4, TNIK, TRRAP, TTC3 neurodegeneration of TISSUE MORPHOLOGY 3.47E-05 BDNF, EYA1, GABRA5, GFI1, SNCA 5 nervous tissue NEUROLOGICAL neurodegeneration of 3.47E-05 BDNF, EYA1, GABRA5, GFI1, SNCA 5 DISEASE nervous tissue ADH1B, AKAP1, AKT3, BDNF, CALM1 (includes others), CFTR, CXCL10, DUSP6, DYNLL1, EP400, EYA1, FASN, FGFR3, FH, FRMD3, CANCER epithelial neoplasia 3.72E-05 GNAS, GSPT1, LAMA5, LDHA, MSR1, NFATC1, PFKP, POLE2, 34 PRKCZ, RELB, RUNX1, SHANK2, SMAD3, STK11, SYNE1, TMBIM4, TNIK, TRRAP, TTC3

183

184

CELLULAR GROWTH generation of cells 3.86E-05 BDNF, CXCL10, DYNLL1, EP400, NEUROD1, RORA, SMAD3, STK11 8 AND PROLIFERATION TISSUE generation of cells 3.86E-05 BDNF, CXCL10, DYNLL1, EP400, NEUROD1, RORA, SMAD3, STK11 8 DEVELOPMENT ATR, BDNF, CXCL10, DNM1, DNM2, DUSP6, FGFR3, FH, GRK4, CELL DEATH AND cell viability 4.27E-05 LAMA5, MSR1, NEUROD1, PRKCZ, RELB, RUNX1, SMAD3, SNCA, 19 SURVIVAL STK11, XRCC4 DEVELOPMENTAL kyphoscoliosis 5.12E-05 FGFR3, SMAD3, SYNE1 3 DISORDER SKELETAL AND MUSCULAR kyphoscoliosis 5.12E-05 FGFR3, SMAD3, SYNE1 3 DISORDERS NERVOUS SYSTEM differentiation of BDNF, CFL1, EYA1, FGFR3, GFI1, NEUROD1, PACSIN1, RUNX1, DEVELOPMENT AND 5.41E-05 10 neurons SNCA, TTC3 FUNCTION CELLULAR differentiation of BDNF, CFL1, EYA1, FGFR3, GFI1, NEUROD1, PACSIN1, RUNX1, 5.41E-05 10 DEVELOPMENT neurons SNCA, TTC3 formation of cellular BDNF, CFL1, DNM1, DNM2, LAMA5, NEUROD1, PACSIN1, PALM, CELL MORPHOLOGY 6.21E-05 14 protrusions PCNT, RELB, SNCA, STK11, SULF2, SYNE1 CELLULAR formation of cellular BDNF, CFL1, DNM1, DNM2, LAMA5, NEUROD1, PACSIN1, PALM, ASSEMBLY AND 6.21E-05 14 protrusions PCNT, RELB, SNCA, STK11, SULF2, SYNE1 ORGANIZATION CELLULAR FUNCTION formation of cellular BDNF, CFL1, DNM1, DNM2, LAMA5, NEUROD1, PACSIN1, PALM, 6.21E-05 14 AND MAINTENANCE protrusions PCNT, RELB, SNCA, STK11, SULF2, SYNE1 BDNF, CFTR, COL10A1, CXCL10, DNM1, DNM2, EP400, FASN, CELL MORPHOLOGY morphology of cells 6.45E-05 FGFR3, GFI1, MSR1, NCOA2, NEUROD1, PRKCZ, RELB, RORA, 21 SIK3, SMAD3, SNCA, STK11, SULF2 POST- phosphorylation of AKT3, ATR, CALM1 (includes others), CFL1, DUSP6, DYNLL1, FGFR3, TRANSLATIONAL 7.05E-05 14 protein GRK4, PRKCZ, SIK3, SNCA, STK11, STK25, TNIK MODIFICATION

184

185

abnormal morphology CELL MORPHOLOGY 7.26E-05 BDNF, GFI1 2 of vestibular hair cells

NERVOUS SYSTEM abnormal morphology DEVELOPMENT AND 7.26E-05 BDNF, GFI1 2 of vestibular hair cells FUNCTION

abnormal morphology TISSUE MORPHOLOGY 7.26E-05 BDNF, GFI1 2 of vestibular hair cells

NERVOUS SYSTEM activation of DEVELOPMENT AND 7.26E-05 BDNF, NEUROD1 2 hippocampus FUNCTION CELLULAR infiltration by 7.26E-05 CXCL10, SMAD3 2 MOVEMENT myofibroblasts EMBRYONIC development of ear 7.33E-05 BDNF, EYA1, FGFR3, GFI1, NEUROD1, SIX2 6 DEVELOPMENT TISSUE development of ear 7.33E-05 BDNF, EYA1, FGFR3, GFI1, NEUROD1, SIX2 6 DEVELOPMENT AUDITORY AND VESTIBULAR SYSTEM development of ear 7.33E-05 BDNF, EYA1, FGFR3, GFI1, NEUROD1, SIX2 6 DEVELOPMENT AND FUNCTION ORGAN development of ear 7.33E-05 BDNF, EYA1, FGFR3, GFI1, NEUROD1, SIX2 6 DEVELOPMENT

185

186

ORGANISMAL development of ear 7.33E-05 BDNF, EYA1, FGFR3, GFI1, NEUROD1, SIX2 6 DEVELOPMENT DEVELOPMENTAL COL10A1, EYA1, GFI1, NCOA2, RELB, RUNX1, SIX2, SMAD3, hypoplasia of organ 7.68E-05 9 DISORDER XRCC4 AKAP1, AKT3, BDNF, DNM1, DNM2, DYNLL1, FASN, FGFR3, GFI1, CELL DEATH AND apoptosis of tumor 8.66E-05 GNAS, LAMA5, MSR1, POT1, PRKCZ, RUNX1, SMAD3, SNCA, 18 SURVIVAL cell lines TMBIM4 NEUROLOGICAL hearing loss 8.88E-05 BDNF, EYA1, FGFR3, GABRA5, GFI1, NEUROD1 6 DISEASE AUDITORY DISEASE hearing loss 8.88E-05 BDNF, EYA1, FGFR3, GABRA5, GFI1, NEUROD1 6 ACAN, ADH1B, AKAP1, AKT3, BDNF, CALM1 (includes others), CFTR, CXCL10, DUSP6, DYNLL1, EP400, EYA1, FASN, FGFR3, FH, FRMD3, GFI1, GNAS, GSPT1, LAMA5, LDHA, MSR1, NEUROD1, CANCER Cancer 1.01E-04 41 NFATC1, PFKP, POLE2, POT1, PRDM16, PRKCZ, RELB, RUNX1, SHANK2, SMAD3, STK11, STK25, SYNE1, TMBIM4, TNIK, TRRAP, TTC3, XRCC4 CELLULAR FUNCTION GFI1, IFNK, MSR1, NFATC1, PRKCZ, RELB, RORA, RUNX1, SMAD3, T cell development 1.07E-04 11 AND MAINTENANCE STK11, XRCC4 CELLULAR GFI1, IFNK, MSR1, NFATC1, PRKCZ, RELB, RORA, RUNX1, SMAD3, T cell development 1.07E-04 11 DEVELOPMENT STK11, XRCC4 CELL-MEDIATED GFI1, IFNK, MSR1, NFATC1, PRKCZ, RELB, RORA, RUNX1, SMAD3, T cell development 1.07E-04 11 IMMUNE RESPONSE STK11, XRCC4 HEMATOLOGICAL SYSTEM GFI1, IFNK, MSR1, NFATC1, PRKCZ, RELB, RORA, RUNX1, SMAD3, T cell development 1.07E-04 11 DEVELOPMENT AND STK11, XRCC4 FUNCTION GFI1, IFNK, MSR1, NFATC1, PRKCZ, RELB, RORA, RUNX1, SMAD3, HEMATOPOIESIS T cell development 1.07E-04 11 STK11, XRCC4 LYMPHOID TISSUE GFI1, IFNK, MSR1, NFATC1, PRKCZ, RELB, RORA, RUNX1, SMAD3, STRUCTURE AND T cell development 1.07E-04 11 STK11, XRCC4 DEVELOPMENT

186

187

AKAP1, AKT3, ATR, BDNF, DNM1, DNM2, DYNLL1, FASN, FGFR3, CELL DEATH AND cell death of tumor 1.13E-04 GFI1, GNAS, LAMA5, MSR1, POT1, PRKCZ, RELB, RUNX1, SMAD3, 20 SURVIVAL cell lines SNCA, TMBIM4 SMALL MOLECULE uptake of lipid 1.15E-04 BDNF, CFTR, CYP7A1, MSR1, NCOA2, PRKCZ 6 BIOCHEMISTRY LIPID METABOLISM uptake of lipid 1.15E-04 BDNF, CFTR, CYP7A1, MSR1, NCOA2, PRKCZ 6 MOLECULAR uptake of lipid 1.15E-04 BDNF, CFTR, CYP7A1, MSR1, NCOA2, PRKCZ 6 TRANSPORT DIGESTIVE SYSTEM development of BDNF, COL10A1, EYA1, FGFR3, LAMA5, NEUROD1, RELB, RUNX1, DEVELOPMENT AND 1.20E-04 9 digestive system SMAD3 FUNCTION quantity of TISSUE MORPHOLOGY 1.26E-04 FGFR3, SIK3, SMAD3 3 chondrocytes CONNECTIVE TISSUE quantity of DEVELOPMENT AND 1.26E-04 FGFR3, SIK3, SMAD3 3 chondrocytes FUNCTION DEVELOPMENTAL chondrodysplasia 1.26E-04 ABCC9, ACAN, COL10A1, PCNT 4 DISORDER SKELETAL AND MUSCULAR chondrodysplasia 1.26E-04 ABCC9, ACAN, COL10A1, PCNT 4 DISORDERS CONNECTIVE TISSUE chondrodysplasia 1.26E-04 ABCC9, ACAN, COL10A1, PCNT 4 DISORDERS HEREDITARY chondrodysplasia 1.26E-04 ABCC9, ACAN, COL10A1, PCNT 4 DISORDER ATR, CALM1 (includes others), CFL1, DNM1, MPHOSPH6, STK11, CELL CYCLE M phase 1.39E-04 7 TRRAP DERMATOLOGICAL DISEASES AND acanthosis nigricans 1.45E-04 BDNF, FGFR3 2 CONDITIONS

187

188

CELLULAR FUNCTION endocytosis by 1.45E-04 DNM1, DNM2 2 AND MAINTENANCE fibroblasts CELLULAR FUNCTION differentiation of T 1.48E-04 GFI1, IFNK, NFATC1, RELB, RORA, RUNX1, SMAD3, STK11, XRCC4 9 AND MAINTENANCE lymphocytes CELLULAR differentiation of T 1.48E-04 GFI1, IFNK, NFATC1, RELB, RORA, RUNX1, SMAD3, STK11, XRCC4 9 DEVELOPMENT lymphocytes CELL-MEDIATED differentiation of T 1.48E-04 GFI1, IFNK, NFATC1, RELB, RORA, RUNX1, SMAD3, STK11, XRCC4 9 IMMUNE RESPONSE lymphocytes HEMATOLOGICAL SYSTEM differentiation of T 1.48E-04 GFI1, IFNK, NFATC1, RELB, RORA, RUNX1, SMAD3, STK11, XRCC4 9 DEVELOPMENT AND lymphocytes FUNCTION differentiation of T HEMATOPOIESIS 1.48E-04 GFI1, IFNK, NFATC1, RELB, RORA, RUNX1, SMAD3, STK11, XRCC4 9 lymphocytes LYMPHOID TISSUE differentiation of T STRUCTURE AND 1.48E-04 GFI1, IFNK, NFATC1, RELB, RORA, RUNX1, SMAD3, STK11, XRCC4 9 lymphocytes DEVELOPMENT EMBRYONIC development of inner 1.65E-04 BDNF, EYA1, FGFR3, GFI1, NEUROD1 5 DEVELOPMENT ear TISSUE development of inner 1.65E-04 BDNF, EYA1, FGFR3, GFI1, NEUROD1 5 DEVELOPMENT ear AUDITORY AND VESTIBULAR SYSTEM development of inner 1.65E-04 BDNF, EYA1, FGFR3, GFI1, NEUROD1 5 DEVELOPMENT AND ear FUNCTION ORGAN development of inner 1.65E-04 BDNF, EYA1, FGFR3, GFI1, NEUROD1 5 DEVELOPMENT ear ORGANISMAL development of inner 1.65E-04 BDNF, EYA1, FGFR3, GFI1, NEUROD1 5 DEVELOPMENT ear

188

189

BDNF, DNM1, GFI1, NCOA2, NEUROD1, NFATC1, PRKCZ, RELB, GENE EXPRESSION transactivation 1.78E-04 12 RORA, RUNX1, SMAD3, TRRAP SMALL MOLECULE AKT3, BDNF, CFTR, CYP7A1, DNM2, FASN, GNAS, KCNA6, MSR1, concentration of lipid 1.79E-04 14 BIOCHEMISTRY RORA, RUNX1, SIK3, SMAD3, SNCA AKT3, BDNF, CFTR, CYP7A1, DNM2, FASN, GNAS, KCNA6, MSR1, LIPID METABOLISM concentration of lipid 1.79E-04 14 RORA, RUNX1, SIK3, SMAD3, SNCA MOLECULAR AKT3, BDNF, CFTR, CYP7A1, DNM2, FASN, GNAS, KCNA6, MSR1, concentration of lipid 1.79E-04 14 TRANSPORT RORA, RUNX1, SIK3, SMAD3, SNCA ORGANISMAL AKT3, BDNF, CFTR, COL10A1, FGFR3, GABRA1, GFI1, NCOA2, size of body 1.80E-04 15 DEVELOPMENT NEUROD1, RUNX1, SIK3, SMAD3, SNCA, SULF2, SYNE1 CANCER testicular cancer 2.36E-04 FGFR3, GNAS, LDHA, STK11 4 ENDOCRINE SYSTEM testicular cancer 2.36E-04 FGFR3, GNAS, LDHA, STK11 4 DISORDERS REPRODUCTIVE testicular cancer 2.36E-04 FGFR3, GNAS, LDHA, STK11 4 SYSTEM DISEASE NEUROLOGICAL tonic-clonic seizure 2.36E-04 AKT3, GABRA1, GABRA5, NEUROD1 4 DISEASE CELL-TO-CELL SIGNALING AND delamination of cells 2.41E-04 CFL1, NEUROD1 2 INTERACTION

CELLULAR FUNCTION differentiation of 2.45E-04 IFNK, NFATC1, RELB, RORA, RUNX1, SMAD3 6 AND MAINTENANCE helper T lymphocytes

CELLULAR differentiation of 2.45E-04 IFNK, NFATC1, RELB, RORA, RUNX1, SMAD3 6 DEVELOPMENT helper T lymphocytes

CELL-MEDIATED differentiation of 2.45E-04 IFNK, NFATC1, RELB, RORA, RUNX1, SMAD3 6 IMMUNE RESPONSE helper T lymphocytes

189

190

HEMATOLOGICAL SYSTEM differentiation of 2.45E-04 IFNK, NFATC1, RELB, RORA, RUNX1, SMAD3 6 DEVELOPMENT AND helper T lymphocytes FUNCTION

differentiation of HEMATOPOIESIS 2.45E-04 IFNK, NFATC1, RELB, RORA, RUNX1, SMAD3 6 helper T lymphocytes

LYMPHOID TISSUE differentiation of STRUCTURE AND 2.45E-04 IFNK, NFATC1, RELB, RORA, RUNX1, SMAD3 6 helper T lymphocytes DEVELOPMENT

ORGANISMAL abnormal morphology 2.50E-04 CFTR, RELB, STK11 3 DEVELOPMENT of distended abdomen

BDNF, CFTR, CYP7A1, DNM1, GABRA1, GABRA5, GFI1, KCNAB1, BEHAVIOR behavior 2.50E-04 15 NCOA2, NEUROD1, PRKCZ, SHANK2, SIK3, SNCA, SYNE1 ABCC9, AKT3, BDNF, CFL1, CFTR, CYP7A1, FH, GABRA1, GABRA5, MOLECULAR transport of molecule 2.51E-04 KCNA6, KCNAB1, MSR1, NEUROD1, NFATC1, PRKCZ, SIX2, 18 TRANSPORT SMAD3, STK11 DEVELOPMENTAL dwarfism 2.54E-04 ABCC9, ACAN, COL10A1, FGFR3, PCNT 5 DISORDER HEREDITARY dwarfism 2.54E-04 ABCC9, ACAN, COL10A1, FGFR3, PCNT 5 DISORDER NERVOUS SYSTEM development of DEVELOPMENT AND 2.76E-04 BDNF, FGFR3, GABRA5, GFI1, NEUROD1, RUNX1 6 neurons FUNCTION CELLULAR development of 2.76E-04 BDNF, FGFR3, GABRA5, GFI1, NEUROD1, RUNX1 6 DEVELOPMENT neurons TISSUE development of 2.76E-04 BDNF, FGFR3, GABRA5, GFI1, NEUROD1, RUNX1 6 DEVELOPMENT neurons

190

191

CELLULAR GROWTH proliferation of EYA1, FGFR3, LAMA5, NCOA2, NEUROD1, NFATC1, RELB, RUNX1, 2.78E-04 9 AND PROLIFERATION epithelial cells SMAD3

abnormal morphology CELL MORPHOLOGY 2.81E-04 BDNF, FGFR3, GFI1 3 of hair cells

NERVOUS SYSTEM abnormal morphology DEVELOPMENT AND 2.81E-04 BDNF, FGFR3, GFI1 3 of hair cells FUNCTION

abnormal morphology TISSUE MORPHOLOGY 2.81E-04 BDNF, FGFR3, GFI1 3 of hair cells

NERVOUS SYSTEM synaptic transmission DEVELOPMENT AND 2.81E-04 BDNF, PRKCZ, SNCA 3 of hippocampal cells FUNCTION CELL-TO-CELL synaptic transmission SIGNALING AND 2.81E-04 BDNF, PRKCZ, SNCA 3 of hippocampal cells INTERACTION differentiation of CELLULAR central nervous 2.98E-04 BDNF, FGFR3, NEUROD1, PACSIN1, RUNX1 5 DEVELOPMENT system cells DEVELOPMENTAL COL10A1, EYA1, FGFR3, GFI1, NCOA2, RELB, RUNX1, SIX2, Hypoplasia 3.04E-04 10 DISORDER SMAD3, XRCC4 CARDIOVASCULAR valvular regurgitation 3.15E-04 GABRA1, GABRA5, NFATC1 3 DISEASE LYMPHOID TISSUE morphology of STRUCTURE AND 3.24E-04 AKT3, COL10A1, EYA1, GFI1, MSR1, RELB, RUNX1, SMAD3, STK11 9 lymphoid organ DEVELOPMENT

191

192

ORGAN morphology of 3.24E-04 AKT3, COL10A1, EYA1, GFI1, MSR1, RELB, RUNX1, SMAD3, STK11 9 MORPHOLOGY lymphoid organ LYMPHOID TISSUE morphology of AKT3, COL10A1, EP400, EYA1, GFI1, MSR1, RELB, RUNX1, SMAD3, STRUCTURE AND lymphatic system 3.37E-04 10 STK11 DEVELOPMENT component CELLULAR development of blood GFI1, IFNK, MSR1, NFATC1, PRKCZ, RELB, RORA, RTKN2, RUNX1, 3.41E-04 12 DEVELOPMENT cells SMAD3, STK11, XRCC4 HEMATOLOGICAL SYSTEM development of blood GFI1, IFNK, MSR1, NFATC1, PRKCZ, RELB, RORA, RTKN2, RUNX1, 3.41E-04 12 DEVELOPMENT AND cells SMAD3, STK11, XRCC4 FUNCTION development of blood GFI1, IFNK, MSR1, NFATC1, PRKCZ, RELB, RORA, RTKN2, RUNX1, HEMATOPOIESIS 3.41E-04 12 cells SMAD3, STK11, XRCC4 ADH1B, AKAP1, CFTR, CXCL10, DUSP6, DYNLL1, FASN, FGFR3, CANCER digestive organ tumor 3.50E-04 FRMD3, GNAS, LDHA, POLE2, PRDM16, PRKCZ, RELB, RUNX1, 22 SMAD3, STK11, SYNE1, TMBIM4, TNIK, TTC3 ADH1B, AKAP1, CFTR, CXCL10, DUSP6, DYNLL1, FASN, FGFR3, GASTROINTESTINAL digestive organ tumor 3.50E-04 FRMD3, GNAS, LDHA, POLE2, PRDM16, PRKCZ, RELB, RUNX1, 22 DISEASE SMAD3, STK11, SYNE1, TMBIM4, TNIK, TTC3 degeneration of spiral TISSUE MORPHOLOGY 3.52E-04 EYA1, GABRA5, GFI1 3 ganglion NEUROLOGICAL degeneration of spiral 3.52E-04 EYA1, GABRA5, GFI1 3 DISEASE ganglion shape change of CELL MORPHOLOGY 3.52E-04 CFTR, MSR1, TNIK 3 epithelial cell lines

abnormal morphology TISSUE MORPHOLOGY 3.58E-04 COL10A1, EYA1, SIK3, SMAD3 4 of cartilage tissue

192

193

CONNECTIVE TISSUE abnormal morphology DEVELOPMENT AND 3.58E-04 COL10A1, EYA1, SIK3, SMAD3 4 of cartilage tissue FUNCTION SKELETAL AND MUSCULAR SYSTEM abnormal morphology 3.58E-04 COL10A1, EYA1, SIK3, SMAD3 4 DEVELOPMENT AND of cartilage tissue FUNCTION

CELL DEATH AND degeneration of 3.60E-04 BDNF, SNCA 2 SURVIVAL cholinergic neurons

CELLULAR degeneration of 3.60E-04 BDNF, SNCA 2 COMPROMISE cholinergic neurons

mitogenesis of kidney CELL CYCLE 3.60E-04 CXCL10, GRK4 2 cells AKT3, ATR, BDNF, FASN, FGFR3, GFI1, GNAS, LAMA5, LDHA, CELLULAR GROWTH proliferation of tumor 3.76E-04 PFKP, PRDM16, PRKCZ, RELB, RORA, RUNX1, SMAD3, STK11, 19 AND PROLIFERATION cell lines TNIK, TRRAP AKT3, ATR, BDNF, FASN, FGFR3, GFI1, GNAS, LAMA5, LDHA, CELLULAR proliferation of tumor 3.76E-04 PFKP, PRDM16, PRKCZ, RELB, RORA, RUNX1, SMAD3, STK11, 19 DEVELOPMENT cell lines TNIK, TRRAP

abnormal morphology CELL MORPHOLOGY 3.79E-04 BDNF, DNM1, FGFR3, GFI1, NEUROD1, SNCA, SULF2 7 of neurons

NERVOUS SYSTEM abnormal morphology DEVELOPMENT AND 3.79E-04 BDNF, DNM1, FGFR3, GFI1, NEUROD1, SNCA, SULF2 7 of neurons FUNCTION

abnormal morphology TISSUE MORPHOLOGY 3.79E-04 BDNF, DNM1, FGFR3, GFI1, NEUROD1, SNCA, SULF2 7 of neurons

193

194

AUDITORY AND VESTIBULAR SYSTEM abnormal morphology 3.91E-04 EYA1, FGFR3, GABRA5 3 DEVELOPMENT AND of cochlea FUNCTION

ORGAN abnormal morphology 3.91E-04 EYA1, FGFR3, GABRA5 3 MORPHOLOGY of cochlea

CFTR, CXCL10, GABRA1, GABRA5, MSR1, RELB, SNCA, SYNE1, INFECTIOUS DISEASE Bacterial Infection 4.27E-04 9 UBD DEVELOPMENTAL osteochondrodysplasia 4.33E-04 ABCC9, ACAN, PCNT 3 DISORDER SKELETAL AND MUSCULAR osteochondrodysplasia 4.33E-04 ABCC9, ACAN, PCNT 3 DISORDERS CONNECTIVE TISSUE osteochondrodysplasia 4.33E-04 ABCC9, ACAN, PCNT 3 DISORDERS HEREDITARY osteochondrodysplasia 4.33E-04 ABCC9, ACAN, PCNT 3 DISORDER AKT3, DYNLL1, FGFR3, LDHA, NFATC1, PRKCZ, RELB, RUNX1, CANCER cell transformation 4.42E-04 10 STK11, TRRAP TISSUE cartilage development 4.49E-04 ACAN, COL10A1, FGFR3, SIX2, SMAD3 5 DEVELOPMENT SKELETAL AND MUSCULAR SYSTEM cartilage development 4.49E-04 ACAN, COL10A1, FGFR3, SIX2, SMAD3 5 DEVELOPMENT AND FUNCTION LYMPHOID TISSUE morphology of STRUCTURE AND 4.65E-04 AKT3, COL10A1, EYA1, GFI1, RUNX1 5 thymus gland DEVELOPMENT

194

195

ORGAN morphology of 4.65E-04 AKT3, COL10A1, EYA1, GFI1, RUNX1 5 MORPHOLOGY thymus gland

arrest in G1 phase of CELL CYCLE 4.77E-04 FASN, GSPT1, STK11 3 colon cancer cell lines

NERVOUS SYSTEM DEVELOPMENT AND quantity of hair cells 4.77E-04 EYA1, FGFR3, GFI1 3 FUNCTION

TISSUE MORPHOLOGY quantity of hair cells 4.77E-04 EYA1, FGFR3, GFI1 3

CARBOHYDRATE binding of 4.95E-04 CXCL10, LAMA5, PACSIN1, SULF2 4 METABOLISM carbohydrate AUDITORY AND VESTIBULAR SYSTEM morphology of ear 4.99E-04 BDNF, EYA1, FGFR3, GABRA5, GFI1 5 DEVELOPMENT AND FUNCTION ORGAN morphology of ear 4.99E-04 BDNF, EYA1, FGFR3, GABRA5, GFI1 5 MORPHOLOGY

SMALL MOLECULE concentration of 5.02E-04 CFTR, SNCA 2 BIOCHEMISTRY docosahexaenoic acid

concentration of LIPID METABOLISM 5.02E-04 CFTR, SNCA 2 docosahexaenoic acid

MOLECULAR concentration of 5.02E-04 CFTR, SNCA 2 TRANSPORT docosahexaenoic acid

195

196

NERVOUS SYSTEM quantity of vestibular DEVELOPMENT AND 5.02E-04 EYA1, GFI1 2 hair cells FUNCTION

quantity of vestibular TISSUE MORPHOLOGY 5.02E-04 EYA1, GFI1 2 hair cells

AUDITORY AND VESTIBULAR SYSTEM quantity of vestibular 5.02E-04 EYA1, GFI1 2 DEVELOPMENT AND hair cells FUNCTION EMBRYONIC kidney development 5.03E-04 BDNF, EYA1, LAMA5, SIK3, SIX2, SMAD3, SULF2 7 DEVELOPMENT TISSUE kidney development 5.03E-04 BDNF, EYA1, LAMA5, SIK3, SIX2, SMAD3, SULF2 7 DEVELOPMENT ORGAN kidney development 5.03E-04 BDNF, EYA1, LAMA5, SIK3, SIX2, SMAD3, SULF2 7 DEVELOPMENT ORGANISMAL kidney development 5.03E-04 BDNF, EYA1, LAMA5, SIK3, SIX2, SMAD3, SULF2 7 DEVELOPMENT RENAL AND UROLOGICAL SYSTEM kidney development 5.03E-04 BDNF, EYA1, LAMA5, SIK3, SIX2, SMAD3, SULF2 7 DEVELOPMENT AND FUNCTION CELLULAR quantity of ASSEMBLY AND 5.25E-04 AKT3, DNM1, RELB 3 mitochondria ORGANIZATION EMBRYONIC development of 5.76E-04 BDNF, EYA1, LAMA5, SIX2 4 DEVELOPMENT metanephric bud TISSUE development of 5.76E-04 BDNF, EYA1, LAMA5, SIX2 4 DEVELOPMENT metanephric bud

196

197

ORGAN development of 5.76E-04 BDNF, EYA1, LAMA5, SIX2 4 DEVELOPMENT metanephric bud ORGANISMAL development of 5.76E-04 BDNF, EYA1, LAMA5, SIX2 4 DEVELOPMENT metanephric bud RENAL AND UROLOGICAL SYSTEM development of 5.76E-04 BDNF, EYA1, LAMA5, SIX2 4 DEVELOPMENT AND metanephric bud FUNCTION SKELETAL AND ACAN, BDNF, COL10A1, CXCL10, DNM1, DYNLL1, EEF1G, FASN, MUSCULAR arthritis 5.78E-04 15 GABRA1, GABRA5, NCOA2, PRKCZ, SMAD3, SNCA, SYNE1 DISORDERS CONNECTIVE TISSUE ACAN, BDNF, COL10A1, CXCL10, DNM1, DYNLL1, EEF1G, FASN, arthritis 5.78E-04 15 DISORDERS GABRA1, GABRA5, NCOA2, PRKCZ, SMAD3, SNCA, SYNE1 INFLAMMATORY ACAN, BDNF, COL10A1, CXCL10, DNM1, DYNLL1, EEF1G, FASN, arthritis 5.78E-04 15 DISEASE GABRA1, GABRA5, NCOA2, PRKCZ, SMAD3, SNCA, SYNE1 CELLULAR quantity of cellular ASSEMBLY AND 5.91E-04 BDNF, EYA1, PACSIN1, PALM, SNCA 5 protrusions ORGANIZATION CELLULAR FUNCTION quantity of cellular 5.91E-04 BDNF, EYA1, PACSIN1, PALM, SNCA 5 AND MAINTENANCE protrusions CELL CYCLE arrest in G1 phase 6.06E-04 ATR, EP400, FASN, GFI1, GSPT1, RUNX1, STK11 7

INFECTIOUS DISEASE infection of mammalia 6.09E-04 CFTR, CXCL10, MSR1, PRKCZ, RELB, SMAD3, SNCA, UBD 8

abnormal morphology BDNF, CFTR, COL10A1, DNM1, EP400, FGFR3, GFI1, NCOA2, CELL MORPHOLOGY 6.15E-04 15 of cells NEUROD1, RELB, SIK3, SMAD3, SNCA, STK11, SULF2

DERMATOLOGICAL DISEASES AND keratosis 6.34E-04 BDNF, CYP7A1, FGFR3, RELB 4 CONDITIONS

197

198

CELLULAR FUNCTION receptor-mediated 6.34E-04 DNM1, DNM2, MSR1, SYNE1 4 AND MAINTENANCE endocytosis SKELETAL AND ACAN, BDNF, COL10A1, CXCL10, DNM1, DYNLL1, EEF1G, FASN, MUSCULAR Rheumatic Disease 6.39E-04 16 GABRA1, GABRA5, GFI1, NCOA2, PRKCZ, SMAD3, SNCA, SYNE1 DISORDERS CONNECTIVE TISSUE ACAN, BDNF, COL10A1, CXCL10, DNM1, DYNLL1, EEF1G, FASN, Rheumatic Disease 6.39E-04 16 DISORDERS GABRA1, GABRA5, GFI1, NCOA2, PRKCZ, SMAD3, SNCA, SYNE1 INFLAMMATORY ACAN, BDNF, COL10A1, CXCL10, DNM1, DYNLL1, EEF1G, FASN, Rheumatic Disease 6.39E-04 16 DISEASE GABRA1, GABRA5, GFI1, NCOA2, PRKCZ, SMAD3, SNCA, SYNE1 TISSUE development of 6.43E-04 ACAN, COL10A1, FGFR3, NFATC1, RUNX1, SIX2, SMAD3 7 DEVELOPMENT connective tissue CELLULAR accumulation of ASSEMBLY AND 6.67E-04 CFTR, NEUROD1 2 zymogen granules ORGANIZATION CELL-TO-CELL activation of cerebral SIGNALING AND 6.67E-04 BDNF, GABRA1 2 cortex cells INTERACTION

CELLULAR chemotaxis of hepatic 6.67E-04 CXCL10, SMAD3 2 MOVEMENT stellate cells

CONNECTIVE TISSUE chemotaxis of hepatic DEVELOPMENT AND 6.67E-04 CXCL10, SMAD3 2 stellate cells FUNCTION HEPATIC SYSTEM chemotaxis of hepatic DEVELOPMENT AND 6.67E-04 CXCL10, SMAD3 2 stellate cells FUNCTION CELLULAR dysfunction of 6.67E-04 BDNF, SNCA 2 COMPROMISE neurons

198

199

GASTROINTESTINAL obstruction of 6.67E-04 CFTR, STK11 2 DISEASE intestine DEVELOPMENTAL BDNF, CFTR, CYP7A1, FGFR3, GFI1, NEUROD1, SMAD3, STK11, Growth Failure 7.30E-04 11 DISORDER SULF2, SYNE1, XRCC4 NERVOUS SYSTEM quantity of synaptic DEVELOPMENT AND 7.45E-04 BDNF, DNM1, SNCA 3 vesicles FUNCTION CELLULAR quantity of synaptic ASSEMBLY AND 7.45E-04 BDNF, DNM1, SNCA 3 vesicles ORGANIZATION

ORGAN abnormal morphology 8.55E-04 EYA1, SMAD3 2 MORPHOLOGY of small thyroid gland

ENDOCRINE SYSTEM abnormal morphology DEVELOPMENT AND 8.55E-04 EYA1, SMAD3 2 of small thyroid gland FUNCTION NEUROLOGICAL incoordination 8.55E-04 BDNF, GABRA1 2 DISEASE DNA REPLICATION, double-stranded DNA RECOMBINATION, 8.73E-04 EYA1, SMAD3, TRRAP, XRCC4 4 break repair AND REPAIR AUDITORY AND VESTIBULAR SYSTEM morphology of inner 8.73E-04 BDNF, EYA1, FGFR3, GABRA5 4 DEVELOPMENT AND ear FUNCTION ORGAN morphology of inner 8.73E-04 BDNF, EYA1, FGFR3, GABRA5 4 MORPHOLOGY ear CELLULAR GROWTH production of 8.74E-04 CXCL10, RUNX1, STK11 3 AND PROLIFERATION lymphocytes

199

200

CELLULAR FUNCTION production of 8.74E-04 CXCL10, RUNX1, STK11 3 AND MAINTENANCE lymphocytes HEMATOLOGICAL SYSTEM production of 8.74E-04 CXCL10, RUNX1, STK11 3 DEVELOPMENT AND lymphocytes FUNCTION SMALL MOLECULE transport of 8.74E-04 CFTR, CYP7A1, MSR1 3 BIOCHEMISTRY cholesterol transport of LIPID METABOLISM 8.74E-04 CFTR, CYP7A1, MSR1 3 cholesterol MOLECULAR transport of 8.74E-04 CFTR, CYP7A1, MSR1 3 TRANSPORT cholesterol CELL-TO-CELL activation of blood BDNF, CXCL10, DDOST, GFI1, IFNK, MSR1, RELB, RORA, RUNX1, SIGNALING AND 8.85E-04 12 cells SMAD3, SNCA, STK11 INTERACTION HEMATOLOGICAL SYSTEM activation of blood BDNF, CXCL10, DDOST, GFI1, IFNK, MSR1, RELB, RORA, RUNX1, 8.85E-04 12 DEVELOPMENT AND cells SMAD3, SNCA, STK11 FUNCTION CELLULAR cytokinesis 8.90E-04 CALM1 (includes others), CFL1, DNM1, STK11, TRRAP 5 MOVEMENT CELL CYCLE cytokinesis 8.90E-04 CALM1 (includes others), CFL1, DNM1, STK11, TRRAP 5

CELL MORPHOLOGY branching of neurites 8.93E-04 BDNF, NEUROD1, PACSIN1, PALM, RELB, SULF2 6

NERVOUS SYSTEM DEVELOPMENT AND branching of neurites 8.93E-04 BDNF, NEUROD1, PACSIN1, PALM, RELB, SULF2 6 FUNCTION CELLULAR ASSEMBLY AND branching of neurites 8.93E-04 BDNF, NEUROD1, PACSIN1, PALM, RELB, SULF2 6 ORGANIZATION

200

201

CELLULAR FUNCTION branching of neurites 8.93E-04 BDNF, NEUROD1, PACSIN1, PALM, RELB, SULF2 6 AND MAINTENANCE EMBRYONIC branching of neurites 8.93E-04 BDNF, NEUROD1, PACSIN1, PALM, RELB, SULF2 6 DEVELOPMENT CELLULAR branching of neurites 8.93E-04 BDNF, NEUROD1, PACSIN1, PALM, RELB, SULF2 6 DEVELOPMENT TISSUE branching of neurites 8.93E-04 BDNF, NEUROD1, PACSIN1, PALM, RELB, SULF2 6 DEVELOPMENT BDNF, CFTR, COL10A1, CXCL10, DYNLL1, EYA1, FGFR3, GABRA1, TISSUE MORPHOLOGY quantity of cells 8.96E-04 GFI1, MSR1, NCOA2, NEUROD1, NFATC1, PRDM16, RELB, RORA, 21 RUNX1, SIK3, SMAD3, SNCA, STK11 ORGAN size of brain 9.12E-04 AKT3, BDNF, NEUROD1, PRDM16 4 MORPHOLOGY SKELETAL AND ACAN, BDNF, COL10A1, CXCL10, DNM1, DYNLL1, EEF1G, MUSCULAR rheumatoid arthritis 9.13E-04 12 GABRA1, GABRA5, NCOA2, SNCA, SYNE1 DISORDERS CONNECTIVE TISSUE ACAN, BDNF, COL10A1, CXCL10, DNM1, DYNLL1, EEF1G, rheumatoid arthritis 9.13E-04 12 DISORDERS GABRA1, GABRA5, NCOA2, SNCA, SYNE1 IMMUNOLOGICAL ACAN, BDNF, COL10A1, CXCL10, DNM1, DYNLL1, EEF1G, rheumatoid arthritis 9.13E-04 12 DISEASE GABRA1, GABRA5, NCOA2, SNCA, SYNE1 INFLAMMATORY ACAN, BDNF, COL10A1, CXCL10, DNM1, DYNLL1, EEF1G, rheumatoid arthritis 9.13E-04 12 DISEASE GABRA1, GABRA5, NCOA2, SNCA, SYNE1 BEHAVIOR circling behavior 9.43E-04 BDNF, GFI1, NEUROD1 3 NERVOUS SYSTEM DEVELOPMENT AND quantity of neuroglia 9.52E-04 BDNF, FGFR3, NEUROD1, SNCA 4 FUNCTION

TISSUE MORPHOLOGY quantity of neuroglia 9.52E-04 BDNF, FGFR3, NEUROD1, SNCA 4 BDNF, NEUROD1, PACSIN1, PALM, RELB, SNCA, STK11, SULF2, CELL MORPHOLOGY neuritogenesis 9.87E-04 9 SYNE1

201

202

NERVOUS SYSTEM BDNF, NEUROD1, PACSIN1, PALM, RELB, SNCA, STK11, SULF2, DEVELOPMENT AND neuritogenesis 9.87E-04 9 SYNE1 FUNCTION CELLULAR BDNF, NEUROD1, PACSIN1, PALM, RELB, SNCA, STK11, SULF2, ASSEMBLY AND neuritogenesis 9.87E-04 9 SYNE1 ORGANIZATION CELLULAR FUNCTION BDNF, NEUROD1, PACSIN1, PALM, RELB, SNCA, STK11, SULF2, neuritogenesis 9.87E-04 9 AND MAINTENANCE SYNE1 CELLULAR BDNF, NEUROD1, PACSIN1, PALM, RELB, SNCA, STK11, SULF2, neuritogenesis 9.87E-04 9 DEVELOPMENT SYNE1 TISSUE BDNF, NEUROD1, PACSIN1, PALM, RELB, SNCA, STK11, SULF2, neuritogenesis 9.87E-04 9 DEVELOPMENT SYNE1 CELL MORPHOLOGY polarization of cells 1.00E-03 CFL1, CXCL10, LAMA5, PRKCZ, STK11 5

NERVOUS SYSTEM GABA-mediated DEVELOPMENT AND 1.07E-03 GABRA1, GABRA5 2 receptor currents FUNCTION CELL-TO-CELL GABA-mediated SIGNALING AND 1.07E-03 GABRA1, GABRA5 2 receptor currents INTERACTION DIGESTIVE SYSTEM abnormal morphology DEVELOPMENT AND 1.07E-03 CFTR, SIK3 2 of gall bladder FUNCTION

ORGAN abnormal morphology 1.07E-03 CFTR, SIK3 2 MORPHOLOGY of gall bladder

HEPATIC SYSTEM abnormal morphology DEVELOPMENT AND 1.07E-03 CFTR, SIK3 2 of gall bladder FUNCTION

202

203

MOLECULAR internalization of 1.08E-03 CFL1, DNM1, DNM2, SIX2 4 TRANSPORT protein PROTEIN internalization of 1.08E-03 CFL1, DNM1, DNM2, SIX2 4 TRAFFICKING protein

CELLULAR GROWTH proliferation of AKT3, ATR, EP400, FGFR3, LDHA, NFATC1, RELB, RUNX1, SMAD3, 1.11E-03 10 AND PROLIFERATION connective tissue cells XRCC4

CELL-TO-CELL activation of CXCL10, DDOST, GFI1, IFNK, MSR1, RELB, RORA, RUNX1, SMAD3, SIGNALING AND 1.12E-03 11 leukocytes SNCA, STK11 INTERACTION HEMATOLOGICAL SYSTEM activation of CXCL10, DDOST, GFI1, IFNK, MSR1, RELB, RORA, RUNX1, SMAD3, 1.12E-03 11 DEVELOPMENT AND leukocytes SNCA, STK11 FUNCTION IMMUNE CELL activation of CXCL10, DDOST, GFI1, IFNK, MSR1, RELB, RORA, RUNX1, SMAD3, 1.12E-03 11 TRAFFICKING leukocytes SNCA, STK11 INFLAMMATORY activation of CXCL10, DDOST, GFI1, IFNK, MSR1, RELB, RORA, RUNX1, SMAD3, 1.12E-03 11 RESPONSE leukocytes SNCA, STK11 ORGANISMAL INJURY Lesion Formation 1.12E-03 ABCC9, BDNF, CFTR, CXCL10, MSR1, RUNX1, SMAD3, SNCA 8 AND ABNORMALITIES SKELETAL AND MUSCULAR Parkinson's disease 1.14E-03 BDNF, GABRA1, GABRA5, GNAS, LDHA, SNCA 6 DISORDERS NEUROLOGICAL Parkinson's disease 1.14E-03 BDNF, GABRA1, GABRA5, GNAS, LDHA, SNCA 6 DISEASE CELL DEATH AND antiapoptosis 1.14E-03 BDNF, CFDP1, CFL1, PRKCZ, SNCA, TMBIM4 6 SURVIVAL CELL-TO-CELL activation of SIGNALING AND 1.16E-03 CXCL10, IFNK, MSR1, RORA, SNCA 5 macrophages INTERACTION

203

204

HEMATOLOGICAL SYSTEM activation of 1.16E-03 CXCL10, IFNK, MSR1, RORA, SNCA 5 DEVELOPMENT AND macrophages FUNCTION IMMUNE CELL activation of 1.16E-03 CXCL10, IFNK, MSR1, RORA, SNCA 5 TRAFFICKING macrophages INFLAMMATORY activation of 1.16E-03 CXCL10, IFNK, MSR1, RORA, SNCA 5 RESPONSE macrophages CELLULAR differentiation of 1.17E-03 GFI1, IFNK, NFATC1, RELB, RUNX1, UBD 6 DEVELOPMENT phagocytes HEMATOLOGICAL SYSTEM differentiation of 1.17E-03 GFI1, IFNK, NFATC1, RELB, RUNX1, UBD 6 DEVELOPMENT AND phagocytes FUNCTION differentiation of HEMATOPOIESIS 1.17E-03 GFI1, IFNK, NFATC1, RELB, RUNX1, UBD 6 phagocytes CELLULAR senescence of 1.17E-03 ATR, FASN, STK11 3 DEVELOPMENT fibroblasts CONNECTIVE TISSUE senescence of DEVELOPMENT AND 1.17E-03 ATR, FASN, STK11 3 fibroblasts FUNCTION senescence of CELL CYCLE 1.17E-03 ATR, FASN, STK11 3 fibroblasts CANCER growth of tumor 1.19E-03 AKT3, DNM2, FGFR3, NEUROD1, RELB, RORA, RUNX1, TNIK 8 CELLULAR differentiation of GFI1, IFNK, NFATC1, PRDM16, RELB, RORA, RUNX1, SMAD3, 1.20E-03 11 DEVELOPMENT leukocytes STK11, UBD, XRCC4 HEMATOLOGICAL SYSTEM differentiation of GFI1, IFNK, NFATC1, PRDM16, RELB, RORA, RUNX1, SMAD3, 1.20E-03 11 DEVELOPMENT AND leukocytes STK11, UBD, XRCC4 FUNCTION

204

205

differentiation of GFI1, IFNK, NFATC1, PRDM16, RELB, RORA, RUNX1, SMAD3, HEMATOPOIESIS 1.20E-03 11 leukocytes STK11, UBD, XRCC4 CELL-TO-CELL activation of antigen SIGNALING AND 1.22E-03 CXCL10, IFNK, MSR1, RELB, RORA, SNCA 6 presenting cells INTERACTION HEMATOLOGICAL SYSTEM activation of antigen 1.22E-03 CXCL10, IFNK, MSR1, RELB, RORA, SNCA 6 DEVELOPMENT AND presenting cells FUNCTION

IMMUNE CELL activation of antigen 1.22E-03 CXCL10, IFNK, MSR1, RELB, RORA, SNCA 6 TRAFFICKING presenting cells

INFLAMMATORY activation of antigen 1.22E-03 CXCL10, IFNK, MSR1, RELB, RORA, SNCA 6 RESPONSE presenting cells

NEUROLOGICAL BDNF, CFTR, FGFR3, GABRA1, GABRA5, GFI1, GNAS, KCNAB1, Movement Disorders 1.29E-03 15 DISEASE LDHA, NEUROD1, PDE4DIP, RELB, RORA, SAP18, SNCA

cell spreading of CELL MORPHOLOGY 1.30E-03 MSR1, TNIK 2 embryonic cell lines

EMBRYONIC cell spreading of 1.30E-03 MSR1, TNIK 2 DEVELOPMENT embryonic cell lines

ORGAN function of brain 1.30E-03 BDNF, RORA 2 DEVELOPMENT NEUROLOGICAL hypophagia 1.30E-03 FASN, KCNA6 2 DISEASE NUTRITIONAL hypophagia 1.30E-03 FASN, KCNA6 2 DISEASE

205

206

PSYCHOLOGICAL hypophagia 1.30E-03 FASN, KCNA6 2 DISORDERS CELL DEATH AND loss of neuroglia 1.30E-03 FGFR3, SNCA 2 SURVIVAL arrest in G1 phase of CELL CYCLE 1.33E-03 ATR, FASN, GFI1, GSPT1, STK11 5 tumor cell lines CELLULAR differentiation of GFI1, IFNK, NFATC1, PRDM16, RELB, RORA, RTKN2, RUNX1, 1.42E-03 12 DEVELOPMENT blood cells SMAD3, STK11, UBD, XRCC4 HEMATOLOGICAL SYSTEM differentiation of GFI1, IFNK, NFATC1, PRDM16, RELB, RORA, RTKN2, RUNX1, 1.42E-03 12 DEVELOPMENT AND blood cells SMAD3, STK11, UBD, XRCC4 FUNCTION INFLAMMATORY inflammation of 1.43E-03 CFTR, RELB, SMAD3 3 RESPONSE secretory structure morphogenesis of CELL MORPHOLOGY 1.52E-03 BDNF, NEUROD1, PACSIN1, PALM, RELB, SULF2, SYNE1 7 neurites NERVOUS SYSTEM morphogenesis of DEVELOPMENT AND 1.52E-03 BDNF, NEUROD1, PACSIN1, PALM, RELB, SULF2, SYNE1 7 neurites FUNCTION CELLULAR morphogenesis of ASSEMBLY AND 1.52E-03 BDNF, NEUROD1, PACSIN1, PALM, RELB, SULF2, SYNE1 7 neurites ORGANIZATION CELLULAR FUNCTION morphogenesis of 1.52E-03 BDNF, NEUROD1, PACSIN1, PALM, RELB, SULF2, SYNE1 7 AND MAINTENANCE neurites CELLULAR morphogenesis of 1.52E-03 BDNF, NEUROD1, PACSIN1, PALM, RELB, SULF2, SYNE1 7 DEVELOPMENT neurites TISSUE morphogenesis of 1.52E-03 BDNF, NEUROD1, PACSIN1, PALM, RELB, SULF2, SYNE1 7 DEVELOPMENT neurites

206

207

DEVELOPMENTAL biliary atresia 1.55E-03 CFTR, CYP7A1 2 DISORDER GASTROINTESTINAL biliary atresia 1.55E-03 CFTR, CYP7A1 2 DISEASE HEPATIC SYSTEM biliary atresia 1.55E-03 CFTR, CYP7A1 2 DISEASE NERVOUS SYSTEM myelination of spinal DEVELOPMENT AND 1.55E-03 BDNF, CXCL10 2 cord FUNCTION TISSUE myelination of spinal 1.55E-03 BDNF, CXCL10 2 DEVELOPMENT cord CELL DEATH AND cell viability of tumor ATR, BDNF, DNM1, DNM2, DUSP6, FGFR3, GRK4, PRKCZ, RELB, 1.66E-03 11 SURVIVAL cell lines SNCA, XRCC4 NERVOUS SYSTEM BDNF, CALM1 (includes others), DNM1, GABRA1, GABRA5, KCNAB1, DEVELOPMENT AND neurotransmission 1.67E-03 8 PRKCZ, SNCA FUNCTION CELL-TO-CELL BDNF, CALM1 (includes others), DNM1, GABRA1, GABRA5, KCNAB1, SIGNALING AND neurotransmission 1.67E-03 8 PRKCZ, SNCA INTERACTION LYMPHOID TISSUE abnormal morphology STRUCTURE AND 1.77E-03 COL10A1, EYA1, GFI1, RUNX1 4 of thymus gland DEVELOPMENT

ORGAN abnormal morphology 1.77E-03 COL10A1, EYA1, GFI1, RUNX1 4 MORPHOLOGY of thymus gland

CELLULAR stabilization of ASSEMBLY AND 1.77E-03 CFL1, GNAS, GRK4, STK11 4 filaments ORGANIZATION ORGANISMAL viability 1.77E-03 ACAN, GFI1, RUNX1, SYNE1 4 SURVIVAL

207

208

lymphohematopoietic AKT3, CXCL10, FASN, FGFR3, GFI1, GNAS, LDHA, MSR1, PRKCZ, CANCER 1.80E-03 13 cancer RUNX1, STK11, TNIK, XRCC4

abnormal morphology CELL MORPHOLOGY 1.83E-03 FGFR3, GFI1 2 of outer hair cells

NERVOUS SYSTEM abnormal morphology DEVELOPMENT AND 1.83E-03 FGFR3, GFI1 2 of outer hair cells FUNCTION

abnormal morphology TISSUE MORPHOLOGY 1.83E-03 FGFR3, GFI1 2 of outer hair cells

CELL DEATH AND apoptosis of epithelial 1.83E-03 EYA1, MSR1 2 SURVIVAL tissue SMALL MOLECULE binding of heparin 1.83E-03 LAMA5, SULF2 2 BIOCHEMISTRY CARBOHYDRATE binding of heparin 1.83E-03 LAMA5, SULF2 2 METABOLISM DRUG METABOLISM binding of heparin 1.83E-03 LAMA5, SULF2 2 NEUROLOGICAL complex partial 1.83E-03 GABRA1, GABRA5 2 DISEASE seizure interphase of stem CELL CYCLE 1.83E-03 PRDM16, RUNX1 2 cells

NERVOUS SYSTEM long-term potentiation DEVELOPMENT AND of hippocampal 1.83E-03 BDNF, SNCA 2 FUNCTION neurons

208

209

CELL-TO-CELL long-term potentiation SIGNALING AND of hippocampal 1.83E-03 BDNF, SNCA 2 INTERACTION neurons

NERVOUS SYSTEM DEVELOPMENT AND maturation of synapse 1.83E-03 BDNF, PALM 2 FUNCTION CELL-TO-CELL SIGNALING AND maturation of synapse 1.83E-03 BDNF, PALM 2 INTERACTION CELLULAR ASSEMBLY AND maturation of synapse 1.83E-03 BDNF, PALM 2 ORGANIZATION CELLULAR FUNCTION maturation of synapse 1.83E-03 BDNF, PALM 2 AND MAINTENANCE TISSUE maturation of synapse 1.83E-03 BDNF, PALM 2 DEVELOPMENT mitosis of colon CELL CYCLE 1.83E-03 ATR, UBD 2 cancer cell lines CELL DEATH AND degeneration of 1.84E-03 BDNF, GABRA5, GFI1 3 SURVIVAL sensory neurons CELLULAR degeneration of 1.84E-03 BDNF, GABRA5, GFI1 3 COMPROMISE sensory neurons CELL CYCLE G1 phase 1.84E-03 ATR, EP400, FASN, GFI1, GSPT1, NFATC1, RUNX1, STK11 8 CELLULAR migration of 1.90E-03 ACAN, BDNF, CFL1, LAMA5 4 MOVEMENT embryonic cells EMBRYONIC migration of 1.90E-03 ACAN, BDNF, CFL1, LAMA5 4 DEVELOPMENT embryonic cells

209

210

CELLULAR FUNCTION engulfment of cells 1.91E-03 CXCL10, DNM1, DNM2, MSR1, PRKCZ, RELB, SNCA 7 AND MAINTENANCE

CELLULAR chemotaxis of 1.95E-03 CXCL10, RELB, SMAD3 3 MOVEMENT connective tissue cells

CELLULAR FUNCTION movement of 1.95E-03 DNM2, STK11, SYNE1 3 AND MAINTENANCE organelle NEUROLOGICAL addiction 2.03E-03 ADH1B, GABRA1, GABRA5, SNCA 4 DISEASE PSYCHOLOGICAL addiction 2.03E-03 ADH1B, GABRA1, GABRA5, SNCA 4 DISORDERS differentiation of CELLULAR antigen presenting 2.10E-03 GFI1, IFNK, NFATC1, RELB, UBD 5 DEVELOPMENT cells HEMATOLOGICAL differentiation of SYSTEM antigen presenting 2.10E-03 GFI1, IFNK, NFATC1, RELB, UBD 5 DEVELOPMENT AND cells FUNCTION differentiation of HEMATOPOIESIS antigen presenting 2.10E-03 GFI1, IFNK, NFATC1, RELB, UBD 5 cells CELL-TO-CELL excitation of cerebral SIGNALING AND 2.13E-03 BDNF, SNCA 2 cortex cells INTERACTION

CELLULAR GROWTH excitation of cerebral 2.13E-03 BDNF, SNCA 2 AND PROLIFERATION cortex cells

CELLULAR GROWTH generation of 2.13E-03 BDNF, EP400 2 AND PROLIFERATION fibroblasts

210

211

TISSUE generation of 2.13E-03 BDNF, EP400 2 DEVELOPMENT fibroblasts DEVELOPMENTAL stuttering 2.13E-03 GABRA1, GABRA5 2 DISORDER NEUROLOGICAL stuttering 2.13E-03 GABRA1, GABRA5 2 DISEASE PSYCHOLOGICAL stuttering 2.13E-03 GABRA1, GABRA5 2 DISORDERS

NERVOUS SYSTEM synaptic transmission DEVELOPMENT AND of hippocampal 2.13E-03 BDNF, SNCA 2 FUNCTION neurons

CELL-TO-CELL synaptic transmission SIGNALING AND of hippocampal 2.13E-03 BDNF, SNCA 2 INTERACTION neurons

binding of NFkB GENE EXPRESSION 2.17E-03 CFL1, GFI1, GNAS, RELB 4 binding site NERVOUS SYSTEM DEVELOPMENT AND generation of neurons 2.18E-03 BDNF, NEUROD1, SMAD3 3 FUNCTION CELLULAR GROWTH generation of neurons 2.18E-03 BDNF, NEUROD1, SMAD3 3 AND PROLIFERATION TISSUE generation of neurons 2.18E-03 BDNF, NEUROD1, SMAD3 3 DEVELOPMENT TISSUE involution 2.18E-03 CFTR, DYNLL1, SMAD3 3 DEVELOPMENT ORGAN involution 2.18E-03 CFTR, DYNLL1, SMAD3 3 MORPHOLOGY

211

212

CARBOHYDRATE binding of 2.30E-03 CXCL10, LAMA5, SULF2 3 METABOLISM polysaccharide CELLULAR migration of neural 2.30E-03 ACAN, CFL1, LAMA5 3 MOVEMENT crest cells EMBRYONIC migration of neural 2.30E-03 ACAN, CFL1, LAMA5 3 DEVELOPMENT crest cells quantity of myeloid TISSUE MORPHOLOGY 2.41E-03 BDNF, GFI1, RELB, RORA, RUNX1, SMAD3, SNCA 7 cells HEMATOLOGICAL SYSTEM quantity of myeloid 2.41E-03 BDNF, GFI1, RELB, RORA, RUNX1, SMAD3, SNCA 7 DEVELOPMENT AND cells FUNCTION CELL CYCLE arrest in interphase 2.43E-03 ATR, EP400, FASN, GFI1, GSPT1, PRDM16, RUNX1, STK11 8 cell spreading of CELL MORPHOLOGY 2.45E-03 MSR1, TNIK 2 epithelial cell lines HAIR AND SKIN cell spreading of DEVELOPMENT AND 2.45E-03 MSR1, TNIK 2 epithelial cell lines FUNCTION CELLULAR chemotaxis of 2.45E-03 CXCL10, SMAD3 2 MOVEMENT microglia NERVOUS SYSTEM chemotaxis of DEVELOPMENT AND 2.45E-03 CXCL10, SMAD3 2 microglia FUNCTION HEMATOLOGICAL SYSTEM chemotaxis of 2.45E-03 CXCL10, SMAD3 2 DEVELOPMENT AND microglia FUNCTION IMMUNE CELL chemotaxis of 2.45E-03 CXCL10, SMAD3 2 TRAFFICKING microglia

212

213

INFLAMMATORY chemotaxis of 2.45E-03 CXCL10, SMAD3 2 RESPONSE microglia CELLULAR formation of ASSEMBLY AND 2.45E-03 CFTR, UBD 2 aggresome ORGANIZATION CELLULAR FUNCTION formation of 2.45E-03 CFTR, UBD 2 AND MAINTENANCE aggresome PROTEIN formation of 2.45E-03 CFTR, UBD 2 DEGRADATION aggresome formation of PROTEIN SYNTHESIS 2.45E-03 CFTR, UBD 2 aggresome ORGAN function of left 2.45E-03 ABCC9, GNAS 2 DEVELOPMENT ventricle CARDIOVASCULAR SYSTEM function of left 2.45E-03 ABCC9, GNAS 2 DEVELOPMENT AND ventricle FUNCTION SMALL MOLECULE oxidation of NADPH 2.45E-03 CALM1 (includes others), FASN 2 BIOCHEMISTRY DNA REPLICATION, RECOMBINATION, oxidation of NADPH 2.45E-03 CALM1 (includes others), FASN 2 AND REPAIR

ENERGY PRODUCTION oxidation of NADPH 2.45E-03 CALM1 (includes others), FASN 2

NUCLEIC ACID oxidation of NADPH 2.45E-03 CALM1 (includes others), FASN 2 METABOLISM DEVELOPMENTAL pectus excavatum 2.45E-03 GABRA1, GABRA5 2 DISORDER

213

214

SKELETAL AND MUSCULAR pectus excavatum 2.45E-03 GABRA1, GABRA5 2 DISORDERS CONNECTIVE TISSUE pectus excavatum 2.45E-03 GABRA1, GABRA5 2 DISORDERS NERVOUS SYSTEM DEVELOPMENT AND quantity of dendrites 2.45E-03 BDNF, PACSIN1 2 FUNCTION CELLULAR ASSEMBLY AND quantity of dendrites 2.45E-03 BDNF, PACSIN1 2 ORGANIZATION CELLULAR FUNCTION quantity of dendrites 2.45E-03 BDNF, PACSIN1 2 AND MAINTENANCE NERVOUS SYSTEM quantity of ganglion DEVELOPMENT AND 2.45E-03 BDNF, NEUROD1 2 cells FUNCTION quantity of ganglion TISSUE MORPHOLOGY 2.45E-03 BDNF, NEUROD1 2 cells

CELLULAR development of 2.55E-03 ACAN, FGFR3, NFATC1, SMAD3 4 DEVELOPMENT connective tissue cells

TISSUE development of 2.55E-03 ACAN, FGFR3, NFATC1, SMAD3 4 DEVELOPMENT connective tissue cells

CONNECTIVE TISSUE development of DEVELOPMENT AND 2.55E-03 ACAN, FGFR3, NFATC1, SMAD3 4 connective tissue cells FUNCTION

CELL MORPHOLOGY cell polarity formation 2.69E-03 CFL1, PRKCZ, STK11 3

214

215

CELL-TO-CELL quantity of actin stress SIGNALING AND 2.69E-03 CFL1, DNM2, PRKCZ 3 fibers INTERACTION CELLULAR quantity of actin stress ASSEMBLY AND 2.69E-03 CFL1, DNM2, PRKCZ 3 fibers ORGANIZATION TISSUE quantity of actin stress 2.69E-03 CFL1, DNM2, PRKCZ 3 DEVELOPMENT fibers REPRODUCTIVE SYSTEM litter size 2.76E-03 AKAP1, CFTR, GABRA1, NCOA2, SULF2 5 DEVELOPMENT AND FUNCTION NERVOUS SYSTEM abnormal morphology DEVELOPMENT AND 2.79E-03 BDNF, EYA1 2 of geniculate ganglion FUNCTION

ORGAN abnormal morphology 2.79E-03 BDNF, EYA1 2 MORPHOLOGY of geniculate ganglion

SMALL MOLECULE concentration of bile 2.79E-03 CYP7A1, SIK3 2 BIOCHEMISTRY salt concentration of bile LIPID METABOLISM 2.79E-03 CYP7A1, SIK3 2 salt MOLECULAR concentration of bile 2.79E-03 CYP7A1, SIK3 2 TRANSPORT salt CELLULAR FUNCTION flux of chloride 2.79E-03 CFTR, GABRA5 2 AND MAINTENANCE MOLECULAR flux of chloride 2.79E-03 CFTR, GABRA5 2 TRANSPORT EMBRYONIC formation of neural 2.79E-03 CFL1, STK11 2 DEVELOPMENT fold

215

216

TISSUE formation of neural 2.79E-03 CFL1, STK11 2 DEVELOPMENT fold ORGANISMAL formation of neural 2.79E-03 CFL1, STK11 2 DEVELOPMENT fold BEHAVIOR panic-like anxiety 2.79E-03 GABRA1, GABRA5 2 NERVOUS SYSTEM DEVELOPMENT AND quantity of microglia 2.79E-03 BDNF, SNCA 2 FUNCTION

TISSUE MORPHOLOGY quantity of microglia 2.79E-03 BDNF, SNCA 2 HEMATOLOGICAL SYSTEM quantity of microglia 2.79E-03 BDNF, SNCA 2 DEVELOPMENT AND FUNCTION SKELETAL AND skelatal muscle MUSCULAR 2.79E-03 GABRA1, GABRA5 2 spasticity DISORDERS SMALL MOLECULE synthesis of 2.83E-03 BDNF, CFTR, CYP7A1 3 BIOCHEMISTRY cholesterol synthesis of LIPID METABOLISM 2.83E-03 BDNF, CFTR, CYP7A1 3 cholesterol VITAMIN AND synthesis of MINERAL 2.83E-03 BDNF, CFTR, CYP7A1 3 cholesterol METABOLISM EMBRYONIC morphology of lung 2.89E-03 CFTR, EYA1, FGFR3, NEUROD1, SYNE1 5 DEVELOPMENT TISSUE morphology of lung 2.89E-03 CFTR, EYA1, FGFR3, NEUROD1, SYNE1 5 DEVELOPMENT ORGAN morphology of lung 2.89E-03 CFTR, EYA1, FGFR3, NEUROD1, SYNE1 5 DEVELOPMENT

216

217

ORGANISMAL morphology of lung 2.89E-03 CFTR, EYA1, FGFR3, NEUROD1, SYNE1 5 DEVELOPMENT ORGAN morphology of lung 2.89E-03 CFTR, EYA1, FGFR3, NEUROD1, SYNE1 5 MORPHOLOGY RESPIRATORY SYSTEM morphology of lung 2.89E-03 CFTR, EYA1, FGFR3, NEUROD1, SYNE1 5 DEVELOPMENT AND FUNCTION CELLULAR BDNF, CXCL10, DNM1, DUSP6, GNAS, PRKCZ, RELB, RUNX1, homing of cells 2.91E-03 9 MOVEMENT SMAD3 development of CELLULAR antigen presenting 2.98E-03 GFI1, NFATC1, RELB 3 DEVELOPMENT cells HEMATOLOGICAL development of SYSTEM antigen presenting 2.98E-03 GFI1, NFATC1, RELB 3 DEVELOPMENT AND cells FUNCTION development of HEMATOPOIESIS antigen presenting 2.98E-03 GFI1, NFATC1, RELB 3 cells LYMPHOID TISSUE development of STRUCTURE AND antigen presenting 2.98E-03 GFI1, NFATC1, RELB 3 DEVELOPMENT cells LYMPHOID TISSUE abnormal morphology STRUCTURE AND 3.00E-03 COL10A1, EYA1, GFI1, RELB, RUNX1, SMAD3, STK11 7 of lymphoid organ DEVELOPMENT

ORGAN abnormal morphology 3.00E-03 COL10A1, EYA1, GFI1, RELB, RUNX1, SMAD3, STK11 7 MORPHOLOGY of lymphoid organ

217

218

abnormal morphology CELL MORPHOLOGY 3.15E-03 COL10A1, RELB 2 of proerythroblasts

abnormal morphology TISSUE MORPHOLOGY 3.15E-03 COL10A1, RELB 2 of proerythroblasts

HEMATOLOGICAL SYSTEM abnormal morphology 3.15E-03 COL10A1, RELB 2 DEVELOPMENT AND of proerythroblasts FUNCTION

abnormal morphology HEMATOPOIESIS 3.15E-03 COL10A1, RELB 2 of proerythroblasts

CARDIOVASCULAR SYSTEM abnormal morphology 3.15E-03 COL10A1, RELB 2 DEVELOPMENT AND of proerythroblasts FUNCTION arrest in cell cycle progression of CELL CYCLE 3.15E-03 ATR, POT1 2 cervical cancer cell lines NERVOUS SYSTEM differentiation of DEVELOPMENT AND 3.15E-03 BDNF, NEUROD1 2 granule cells FUNCTION CELLULAR differentiation of 3.15E-03 BDNF, NEUROD1 2 DEVELOPMENT granule cells differentiation of CELLULAR myeloid dendritic 3.15E-03 RELB, UBD 2 DEVELOPMENT cells

218

219

HEMATOLOGICAL differentiation of SYSTEM myeloid dendritic 3.15E-03 RELB, UBD 2 DEVELOPMENT AND cells FUNCTION differentiation of HEMATOPOIESIS myeloid dendritic 3.15E-03 RELB, UBD 2 cells TISSUE involution of thymus 3.15E-03 CFTR, SMAD3 2 DEVELOPMENT gland ORGAN involution of thymus 3.15E-03 CFTR, SMAD3 2 MORPHOLOGY gland NERVOUS SYSTEM quantity of nerve DEVELOPMENT AND 3.15E-03 BDNF, SNCA 2 ending FUNCTION CELLULAR quantity of nerve ASSEMBLY AND 3.15E-03 BDNF, SNCA 2 ending ORGANIZATION NERVOUS SYSTEM recycling of synaptic DEVELOPMENT AND 3.15E-03 DNM1, PACSIN1 2 vesicles FUNCTION CELLULAR recycling of synaptic ASSEMBLY AND 3.15E-03 DNM1, PACSIN1 2 vesicles ORGANIZATION SKELETAL AND MUSCULAR torticollis 3.15E-03 BDNF, GFI1 2 DISORDERS SKELETAL AND MUSCULAR SYSTEM morphology of axial 3.15E-03 COL10A1, FGFR3, SIK3, SMAD3, SULF2 5 DEVELOPMENT AND skeleton FUNCTION

219

220

AKT3, CALM1 (includes others), EYA1, FASN, GNAS, GSPT1, LAMA5, CANCER mammary tumor 3.15E-03 14 MSR1, NFATC1, POLE2, RELB, RUNX1, SMAD3, SYNE1 CELLULAR GROWTH production of cells 3.25E-03 CXCL10, FGFR3, RUNX1, STK11 4 AND PROLIFERATION CELLULAR FUNCTION production of cells 3.25E-03 CXCL10, FGFR3, RUNX1, STK11 4 AND MAINTENANCE CELL-TO-CELL activation of central SIGNALING AND 3.28E-03 BDNF, GABRA1, SNCA 3 nervous system cells INTERACTION CELL DEATH AND cell death of motor 3.28E-03 BDNF, RUNX1, SNCA 3 SURVIVAL neurons CELLULAR differentiation of stem 3.50E-03 BDNF, GFI1, NEUROD1, RUNX1, SMAD3 5 DEVELOPMENT cells ORGANISMAL morphology of body BDNF, CFTR, COL10A1, EYA1, FGFR3, RELB, SIK3, SMAD3, STK11, 3.51E-03 10 DEVELOPMENT region SULF2 NERVOUS SYSTEM quantity of outer hair DEVELOPMENT AND 3.53E-03 FGFR3, GFI1 2 cells FUNCTION quantity of outer hair TISSUE MORPHOLOGY 3.53E-03 FGFR3, GFI1 2 cells AUDITORY AND VESTIBULAR SYSTEM quantity of outer hair 3.53E-03 FGFR3, GFI1 2 DEVELOPMENT AND cells FUNCTION CELLULAR organization of ASSEMBLY AND 3.53E-03 ACAN, CFL1, DNM1, PCNT, POT1, SNCA, STK25, SYNE1 8 organelle ORGANIZATION transformation of CANCER 3.55E-03 AKT3, DYNLL1, FGFR3, LDHA, NFATC1, RUNX1 6 fibroblast cell lines

220

221

EMBRYONIC morphogenesis of 3.60E-03 EYA1, LAMA5, SIX2 3 DEVELOPMENT metanephric bud TISSUE morphogenesis of 3.60E-03 EYA1, LAMA5, SIX2 3 DEVELOPMENT metanephric bud ORGAN morphogenesis of 3.60E-03 EYA1, LAMA5, SIX2 3 DEVELOPMENT metanephric bud ORGANISMAL morphogenesis of 3.60E-03 EYA1, LAMA5, SIX2 3 DEVELOPMENT metanephric bud RENAL AND UROLOGICAL SYSTEM morphogenesis of 3.60E-03 EYA1, LAMA5, SIX2 3 DEVELOPMENT AND metanephric bud FUNCTION NERVOUS SYSTEM DEVELOPMENT AND quantity of neurons 3.60E-03 BDNF, EYA1, FGFR3, GFI1, NEUROD1, SNCA 6 FUNCTION

TISSUE MORPHOLOGY quantity of neurons 3.60E-03 BDNF, EYA1, FGFR3, GFI1, NEUROD1, SNCA 6

EMBRYONIC morphology of limb 3.65E-03 COL10A1, FGFR3, SMAD3, SULF2 4 DEVELOPMENT bone TISSUE morphology of limb 3.65E-03 COL10A1, FGFR3, SMAD3, SULF2 4 DEVELOPMENT bone ORGAN morphology of limb 3.65E-03 COL10A1, FGFR3, SMAD3, SULF2 4 DEVELOPMENT bone ORGANISMAL morphology of limb 3.65E-03 COL10A1, FGFR3, SMAD3, SULF2 4 DEVELOPMENT bone CONNECTIVE TISSUE morphology of limb DEVELOPMENT AND 3.65E-03 COL10A1, FGFR3, SMAD3, SULF2 4 bone FUNCTION

221

222

ORGAN morphology of limb 3.65E-03 COL10A1, FGFR3, SMAD3, SULF2 4 MORPHOLOGY bone SKELETAL AND MUSCULAR SYSTEM morphology of limb 3.65E-03 COL10A1, FGFR3, SMAD3, SULF2 4 DEVELOPMENT AND bone FUNCTION CELLULAR ASSEMBLY AND quantity of filaments 3.65E-03 CFL1, DNM2, PRKCZ, SMAD3 4 ORGANIZATION ORGAN size of bone 3.72E-03 COL10A1, MSR1, NCOA2, SIK3, SMAD3 5 MORPHOLOGY SKELETAL AND MUSCULAR SYSTEM size of bone 3.72E-03 COL10A1, MSR1, NCOA2, SIK3, SMAD3 5 DEVELOPMENT AND FUNCTION NERVOUS SYSTEM action potential of DEVELOPMENT AND 3.75E-03 BDNF, CALM1 (includes others), GABRA1, KCNAB1 4 cells FUNCTION CELL-TO-CELL action potential of SIGNALING AND 3.75E-03 BDNF, CALM1 (includes others), GABRA1, KCNAB1 4 cells INTERACTION

abnormal morphology EMBRYONIC of epiphyseal growth 3.76E-03 COL10A1, FGFR3, SMAD3 3 DEVELOPMENT plate

abnormal morphology TISSUE MORPHOLOGY of epiphyseal growth 3.76E-03 COL10A1, FGFR3, SMAD3 3 plate

222

223

abnormal morphology TISSUE of epiphyseal growth 3.76E-03 COL10A1, FGFR3, SMAD3 3 DEVELOPMENT plate

abnormal morphology ORGAN of epiphyseal growth 3.76E-03 COL10A1, FGFR3, SMAD3 3 DEVELOPMENT plate

abnormal morphology ORGANISMAL of epiphyseal growth 3.76E-03 COL10A1, FGFR3, SMAD3 3 DEVELOPMENT plate

CONNECTIVE TISSUE abnormal morphology DEVELOPMENT AND of epiphyseal growth 3.76E-03 COL10A1, FGFR3, SMAD3 3 FUNCTION plate

abnormal morphology ORGAN of epiphyseal growth 3.76E-03 COL10A1, FGFR3, SMAD3 3 MORPHOLOGY plate

SKELETAL AND abnormal morphology MUSCULAR SYSTEM of epiphyseal growth 3.76E-03 COL10A1, FGFR3, SMAD3 3 DEVELOPMENT AND plate FUNCTION NEUROLOGICAL dyssomnia 3.76E-03 BDNF, GABRA1, GABRA5 3 DISEASE PSYCHOLOGICAL dyssomnia 3.76E-03 BDNF, GABRA1, GABRA5 3 DISORDERS

223

224

hematological AKT3, CXCL10, FASN, FGFR3, GFI1, LDHA, MSR1, PRKCZ, RUNX1, CANCER 3.85E-03 12 neoplasia STK11, TNIK, XRCC4 HEMATOLOGICAL hematological AKT3, CXCL10, FASN, FGFR3, GFI1, LDHA, MSR1, PRKCZ, RUNX1, 3.85E-03 12 DISEASE neoplasia STK11, TNIK, XRCC4 GENE EXPRESSION repression of RNA 3.88E-03 NCOA2, PRDM16, PRKCZ, RUNX1, SAP18 5 ACAN, AKT3, BDNF, CFL1, CFTR, CXCL10, DNM1, DNM2, DUSP6, CELLULAR cell movement 3.93E-03 FH, GNAS, LAMA5, MSR1, NEUROD1, NFATC1, PRKCZ, RELB, 21 MOVEMENT RUNX1, SMAD3, STK11, STK25 DIGESTIVE SYSTEM abnormal morphology DEVELOPMENT AND 3.93E-03 CFTR, STK11 2 of duodenum FUNCTION

ORGAN abnormal morphology 3.93E-03 CFTR, STK11 2 MORPHOLOGY of duodenum

CELL DEATH AND cell death of spinal 3.93E-03 BDNF, RUNX1 2 SURVIVAL neuron DERMATOLOGICAL DISEASES AND formation of abscess 3.93E-03 CFTR, SMAD3 2 CONDITIONS ORGANISMAL INJURY formation of abscess 3.93E-03 CFTR, SMAD3 2 AND ABNORMALITIES DEVELOPMENTAL scoliosis 3.93E-03 FGFR3, SIK3 2 DISORDER SKELETAL AND MUSCULAR scoliosis 3.93E-03 FGFR3, SIK3 2 DISORDERS NEUROLOGICAL status epilepticus 3.93E-03 GABRA1, GABRA5 2 DISEASE

224

225

SMALL MOLECULE synthesis of acyl- 3.93E-03 FASN, SNCA 2 BIOCHEMISTRY coenzyme A synthesis of acyl- LIPID METABOLISM 3.93E-03 FASN, SNCA 2 coenzyme A NUCLEIC ACID synthesis of acyl- 3.93E-03 FASN, SNCA 2 METABOLISM coenzyme A

CARDIOVASCULAR valvular regurgitation 3.93E-03 GABRA1, GABRA5 2 DISEASE of mitral valve

EMBRYONIC abnormal morphology 3.93E-03 SIK3, SMAD3, SULF2 3 DEVELOPMENT of sternum

TISSUE abnormal morphology 3.93E-03 SIK3, SMAD3, SULF2 3 DEVELOPMENT of sternum

ORGAN abnormal morphology 3.93E-03 SIK3, SMAD3, SULF2 3 DEVELOPMENT of sternum

ORGANISMAL abnormal morphology 3.93E-03 SIK3, SMAD3, SULF2 3 DEVELOPMENT of sternum

CONNECTIVE TISSUE abnormal morphology DEVELOPMENT AND 3.93E-03 SIK3, SMAD3, SULF2 3 of sternum FUNCTION

ORGAN abnormal morphology 3.93E-03 SIK3, SMAD3, SULF2 3 MORPHOLOGY of sternum

225

226

SKELETAL AND MUSCULAR SYSTEM abnormal morphology 3.93E-03 SIK3, SMAD3, SULF2 3 DEVELOPMENT AND of sternum FUNCTION GASTROINTESTINAL diarrhea 3.93E-03 BDNF, CFTR, CYP7A1 3 DISEASE EP400, FASN, FGFR3, GNAS, LAMA5, LDHA, MSR1, NEUROD1, CANCER genital tumor 3.97E-03 13 POLE2, PRKCZ, RELB, STK11, TRRAP REPRODUCTIVE EP400, FASN, FGFR3, GNAS, LAMA5, LDHA, MSR1, NEUROD1, genital tumor 3.97E-03 13 SYSTEM DISEASE POLE2, PRKCZ, RELB, STK11, TRRAP CELLULAR GROWTH formation of cells 4.07E-03 BDNF, DNM2, NCOA2, NEUROD1, NFATC1, PRKCZ, SMAD3 7 AND PROLIFERATION cell viability of CELL DEATH AND cervical cancer cell 4.12E-03 ATR, DNM1, DUSP6, GRK4, XRCC4 5 SURVIVAL lines

CELL DEATH AND degeneration of 4.35E-03 BDNF, SNCA 2 SURVIVAL dopaminergic neurons

CELLULAR degeneration of 4.35E-03 BDNF, SNCA 2 COMPROMISE dopaminergic neurons

CELLULAR development of 4.35E-03 ACAN, FGFR3 2 DEVELOPMENT chondrocytes TISSUE development of 4.35E-03 ACAN, FGFR3 2 DEVELOPMENT chondrocytes CONNECTIVE TISSUE development of DEVELOPMENT AND 4.35E-03 ACAN, FGFR3 2 chondrocytes FUNCTION

226

227

SKELETAL AND MUSCULAR SYSTEM development of 4.35E-03 ACAN, FGFR3 2 DEVELOPMENT AND chondrocytes FUNCTION NEUROLOGICAL myoclonic seizure 4.35E-03 GABRA1, GABRA5 2 DISEASE CELL-TO-CELL SIGNALING AND uptake of dopamine 4.35E-03 BDNF, SNCA 2 INTERACTION SMALL MOLECULE uptake of dopamine 4.35E-03 BDNF, SNCA 2 BIOCHEMISTRY MOLECULAR uptake of dopamine 4.35E-03 BDNF, SNCA 2 TRANSPORT

DRUG METABOLISM uptake of dopamine 4.35E-03 BDNF, SNCA 2

CELL CYCLE senescence of cells 4.37E-03 ATR, FASN, POT1, RUNX1, STK11 5 NUTRITIONAL weight gain 4.44E-03 ABCC9, BDNF, GABRA1, GABRA5, RUNX1, SMAD3 6 DISEASE NEUROLOGICAL cognitive impairment 4.48E-03 BDNF, CXCL10, SNCA 3 DISEASE NERVOUS SYSTEM DEVELOPMENT AND quantity of neurites 4.48E-03 BDNF, PACSIN1, SNCA 3 FUNCTION CELLULAR ASSEMBLY AND quantity of neurites 4.48E-03 BDNF, PACSIN1, SNCA 3 ORGANIZATION CELLULAR FUNCTION quantity of neurites 4.48E-03 BDNF, PACSIN1, SNCA 3 AND MAINTENANCE CELL MORPHOLOGY size of cells 4.65E-03 AKT3, BDNF, CFL1, FASN, GNAS, NCOA2, SIK3 7

227

228

NERVOUS SYSTEM abnormal morphology DEVELOPMENT AND 4.67E-03 BDNF, NEUROD1, SULF2 3 of nervous system FUNCTION EMBRYONIC abnormal morphology 4.67E-03 FGFR3, SIK3, SMAD3 3 DEVELOPMENT of rib TISSUE abnormal morphology 4.67E-03 FGFR3, SIK3, SMAD3 3 DEVELOPMENT of rib ORGAN abnormal morphology 4.67E-03 FGFR3, SIK3, SMAD3 3 DEVELOPMENT of rib ORGANISMAL abnormal morphology 4.67E-03 FGFR3, SIK3, SMAD3 3 DEVELOPMENT of rib CONNECTIVE TISSUE abnormal morphology DEVELOPMENT AND 4.67E-03 FGFR3, SIK3, SMAD3 3 of rib FUNCTION ORGAN abnormal morphology 4.67E-03 FGFR3, SIK3, SMAD3 3 MORPHOLOGY of rib SKELETAL AND MUSCULAR SYSTEM abnormal morphology 4.67E-03 FGFR3, SIK3, SMAD3 3 DEVELOPMENT AND of rib FUNCTION CELLULAR invasion of lung 4.67E-03 AKT3, BDNF, STK11 3 MOVEMENT cancer cell lines EMBRYONIC lung development 4.72E-03 CFTR, EYA1, FGFR3, LAMA5, NEUROD1, SYNE1 6 DEVELOPMENT TISSUE lung development 4.72E-03 CFTR, EYA1, FGFR3, LAMA5, NEUROD1, SYNE1 6 DEVELOPMENT ORGAN lung development 4.72E-03 CFTR, EYA1, FGFR3, LAMA5, NEUROD1, SYNE1 6 DEVELOPMENT ORGANISMAL lung development 4.72E-03 CFTR, EYA1, FGFR3, LAMA5, NEUROD1, SYNE1 6 DEVELOPMENT

228

229

RESPIRATORY SYSTEM lung development 4.72E-03 CFTR, EYA1, FGFR3, LAMA5, NEUROD1, SYNE1 6 DEVELOPMENT AND FUNCTION

EMBRYONIC abnormal morphology 4.80E-03 EYA1, SULF2 2 DEVELOPMENT of basisphenoid bone

TISSUE abnormal morphology 4.80E-03 EYA1, SULF2 2 DEVELOPMENT of basisphenoid bone

ORGAN abnormal morphology 4.80E-03 EYA1, SULF2 2 DEVELOPMENT of basisphenoid bone

ORGANISMAL abnormal morphology 4.80E-03 EYA1, SULF2 2 DEVELOPMENT of basisphenoid bone

CONNECTIVE TISSUE abnormal morphology DEVELOPMENT AND 4.80E-03 EYA1, SULF2 2 of basisphenoid bone FUNCTION

ORGAN abnormal morphology 4.80E-03 EYA1, SULF2 2 MORPHOLOGY of basisphenoid bone

SKELETAL AND MUSCULAR SYSTEM abnormal morphology 4.80E-03 EYA1, SULF2 2 DEVELOPMENT AND of basisphenoid bone FUNCTION SKELETAL AND MUSCULAR SYSTEM density of bone 4.80E-03 FGFR3, MSR1 2 DEVELOPMENT AND FUNCTION

229

230

DERMATOLOGICAL DISEASES AND desquamation 4.80E-03 CYP7A1, NCOA2 2 CONDITIONS CELLULAR endocytosis of 4.80E-03 DNM1, SNCA 2 MOVEMENT synaptic vesicles NERVOUS SYSTEM endocytosis of DEVELOPMENT AND 4.80E-03 DNM1, SNCA 2 synaptic vesicles FUNCTION CELLULAR endocytosis of ASSEMBLY AND 4.80E-03 DNM1, SNCA 2 synaptic vesicles ORGANIZATION CELLULAR FUNCTION endocytosis of 4.80E-03 DNM1, SNCA 2 AND MAINTENANCE synaptic vesicles CELLULAR FUNCTION function of 4.80E-03 RELB, SMAD3 2 AND MAINTENANCE keratinocytes INFLAMMATORY inflammation of 4.80E-03 CFTR, RELB 2 RESPONSE salivary gland INFLAMMATORY inflammation of 4.80E-03 RELB, SMAD3 2 RESPONSE stomach AKT3, CALM1 (includes others), EYA1, FASN, GNAS, GSPT1, LAMA5, CANCER breast cancer 4.87E-03 13 NFATC1, POLE2, RELB, RUNX1, SMAD3, SYNE1 EMBRYONIC morphogenesis of 4.90E-03 EYA1, LAMA5, SIX2, SMAD3 4 DEVELOPMENT embryonic tissue TISSUE morphogenesis of 4.90E-03 EYA1, LAMA5, SIX2, SMAD3 4 DEVELOPMENT embryonic tissue ORGANISMAL morphogenesis of 4.90E-03 EYA1, LAMA5, SIX2, SMAD3 4 DEVELOPMENT embryonic tissue

230

231

Table E6: Overlap between asthma-associated DMRs in IIS children and DNase I HSS in monocytes and naïve T cells. * Magnitude difference

† -log10(P-value) ‡ Number of CpGs within 750bp window, centered on the DMR DNase I HSS DNase I HSS DMRs monocytes naïve T-cells Location Size Mag* Sig† CpGs‡ Start - End Start - End Genes chr1:7916430- 7916580- 352 0.212 2.184 24 . UTS2 7916782 7916730 chr1:11790134- 11790500- 375 0.235 2.192 18 . intergenic (DRAXIN-AGTRAP) 11790509 11790650 chr1:11790134- 11790200- 375 0.235 2.192 18 . intergenic (DRAXIN-AGTRAP) 11790509 11790350 chr1:26947080- 26947340- 26947220- 495 0.252 2.84 79 intergenic (RPS6KA1-ARID1A) 26947575 26947490 26947370 chr1:26947080- 26947060- 495 0.252 2.84 79 . intergenic (RPS6KA1-ARID1A) 26947575 26947210 chr1:45251683- 45251880- 45251860- 425 0.29 2.246 45 BEST4 45252108 45252030 45252010 chr1:118468419- 118468740- 572 -0.238 3.024 5 . WDR3/GDAP2 118468991 118468890 chr2:54557572- 54557880- 54557880- 325 0.242 2.761 52 C2orf73 54557897 54558030 54558030 chr2:219576276- 219576240- 770 0.376 3.481 6 . TTLL4 219577046 219576390 chr2:220083145- 220083200- 385 0.226 5.569 68 . ABCB6/ATG9A 220083530 220083350 chr3:46599590- 46599940- 46599900- 630 0.244 4.231 21 LRRC2 46600220 46600090 46600050 chr3:197676983- 197677200- 500 0.225 2.457 27 . RPL35A/IQCG 197677483 197677350

231

232

124520- 125020- chr4:124613-125128 515 0.211 2.951 44 ZNF718 124670 125170 124760- 124740- chr4:124613-125128 515 0.211 2.951 44 ZNF718 124910 124890 chr4:3042964- 3043200- 375 0.222 2.834 46 . GRK4 3043339 3043350 chr4:89051945- 89051840- 380 -0.243 3.563 9 . ABCG2/AFF4 89052325 89051990 chr5:82368664- 82368860- 420 -0.224 2.402 10 . XRCC4/TMEM167A 82369084 82369010 chr5:126114144- 126114020- 126114040- 310 0.22 2.879 50 LMNB1 126114454 126114170 126114190 chr5:142174675- 142174960- 310 -0.203 2.758 5 . ARHGAP26 142174985 142175110 chr5:156570589- 156570600- 156570600- 305 0.262 3.601 27 MED7 156570894 156570750 156570750 chr5:180648315- 180648420- 705 0.228 3.595 33 . TRIM41/MIR4638 180649020 180648570 chr6:2902234- 2903000- 807 0.234 3.144 27 . SERPINB9 2903041 2903150 chr6:4020994- 4021460- 4021040- 495 0.241 2.985 35 PRPF4B 4021489 4021610 4021190 chr6:4020994- 4021400- 495 0.241 2.985 35 . PRPF4B 4021489 4021550 chr6:34499206- 34499700- 525 0.214 2.285 33 . PACSIN1 34499731 34499850 chr6:151711985- 151712140- 151712680- 725 0.327 3.029 99 ZBTB2 151712710 151712290 151712830 chr6:167410746- 167411080- 375 -0.282 2.362 8 . FGFR1OP/MIR3939 167411121 167411230 chr7:148844069- 148844320- 148844180- 430 0.258 2.509 61 ZNF398 148844499 148844470 148844330 chr8:119266746- 119266760- 325 -0.219 2.199 10 . SAMD12 119267071 119266910

232

233

chr8:124214717- 124215120- 666 -0.242 3.702 9 . LOC100131726/FAM83A 124215383 124215270 chr8:144631520- 144631620- 144631560- 315 0.307 2.663 23 GSDMD 144631835 144631770 144631710 chr9:130980777- 130981100- 130981100- 515 0.254 3.833 44 DNM1 130981292 130981250 130981250 chr10:14051467- 14051580- 490 0.219 3.582 24 . FRMD4A 14051957 14051730 chr11:1763722- 1763680- 325 0.347 3.015 7 . MOB2/IFITM10 1764047 1763830 chr11:16760509- 16760800- 16760580- 370 0.225 2.505 31 C11orf58 16760879 16760950 16760730 chr11:16760509- 16760860- 370 0.225 2.505 31 . C11orf58 16760879 16761010 chr11:18417431- 18417660- 18417780- 585 0.296 2.305 47 LDHA 18418016 18417810 18417930 chr11:18417431- 18417460- 585 0.296 2.305 47 . LDHA 18418016 18417610 chr11:32850713- 32851200- 500 0.463 3.122 23 . PRRG4 32851213 32851350 chr11:58346012- 58346380- 58346380- 616 0.336 2.988 67 LPXN/ZFP91/ZFP91-CNTF 58346628 58346530 58346530 chr11:62341248- 62341400- 62341400- 470 0.301 2.707 45 TUT1/EEF1G 62341718 62341550 62341550 chr11:75139113- 75139360- 75139340- 610 0.224 3.565 40 KLHL35 75139723 75139510 75139490 chr11:104915738- 104916000- 104916040- 515 0.342 2.374 6 CARD16 104916253 104916150 104916190 chr11:104915738- 104915740- 515 0.342 2.374 6 . CARD16 104916253 104915890 chr11:116881940- 116882020- 370 -0.208 2.336 9 . SIK3 116882310 116882170 chr12:9824116- 9824120- 393 0.293 2.329 17 . CLEC2D 9824509 9824270

233

234

chr12:49524725- 49524600- 49525100- 411 0.279 2.704 84 TUBA1B 49525136 49524750 49525250 chr12:49524725- 49525120- 411 0.279 2.704 84 . TUBA1B 49525136 49525270 chr12:54719022- 54719100- 400 0.228 2.446 22 . COPZ1 54719422 54719350 chr12:89743694- 89743740- 525 0.333 2.263 11 . DUSP6 89744219 89743890 chr12:132434342- 132434260- 375 0.248 2.253 120 . EP400 132434717 132434410 chr13:21714409- 21714540- 21714460- 305 0.244 2.483 62 SAP18 21714714 21714690 21714610 chr13:43570879- 43570780- 304 -0.223 4.039 6 . EPSTI1 43571183 43570930 chr14:50154049- 50154160- 400 0.22 3.201 27 . POLE2 50154449 50154310 chr15:25200266- 25200240- 685 0.243 3.189 49 . SNRPN/SNURF 25200951 25200390 chr15:25200266- 25200520- 685 0.243 3.189 49 . SNRPN/SNURF 25200951 25200670 chr15:26108457- 26108560- 26108560- 327 0.261 2.288 86 ATP10A 26108784 26108710 26108710 chr15:41786297- 41786420- 609 0.244 2.751 87 . ITPKA 41786906 41786570 chr16:12009069- 12009140- 398 0.215 2.414 105 . GSPT1 12009467 12009290 chr16:21531140- 21531460- 21531460- 330 0.333 3.69 75 SLC7A5P2 21531470 21531610 21531610 chr16:21531140- 21531040- 330 0.333 3.69 75 . SLC7A5P2 21531470 21531190 chr16:31232871- 31233240- 378 -0.253 2.752 10 . PYDC1/TRIM72 31233249 31233390 chr16:90148322- 90148460- 90148460- 575 0.306 3.78 45 intergenic (PRDM7) 90148897 90148610 90148610

234

235

chr17:6938284- 6938580- 401 0.231 2.899 7 . SLC16A13 6938685 6938730 chr17:38277847- 38278120- 589 0.304 2.343 81 . MSL1 38278436 38278270 chr17:74349827- 74350200- 74349980- 685 0.308 2.567 62 PRPSAP1 74350512 74350350 74350130 chr17:74349827- 74349960- 74350240- 685 0.308 2.567 62 PRPSAP1 74350512 74350110 74350390 chr17:79361084- 79361100- intergenic (LOC100130370- 334 0.211 4.662 83 . 79361418 79361250 BAHCC1) RPL17/RPL17- chr18:47019290- 47019720- 498 -0.475 3.42 12 . C18ORF32/SNORD58A/SNORD58 47019788 47019870 B/SNORD58C chr18:77196024- 77196040- 500 0.219 2.609 47 . NFATC1 77196524 77196190 chr19:1021495- 1021620- 1021620- 330 0.241 3.03 40 CNN2/C19orf6 1021825 1021770 1021770 chr19:1876344- 1876440- 590 0.27 2.843 46 . FAM108A1 1876934 1876590 chr19:10907627- 10907800- 576 0.226 3.466 28 . DNM2 10908203 10907950 chr19:37263404- 37263940- 37263940- 590 0.25 3.935 53 ZNF850 37263994 37264090 37264090 chr19:37263404- 37263680- 37263660- 590 0.25 3.935 53 ZNF850 37263994 37263830 37263810 chr19:38084769- 38085100- 379 0.215 2.643 26 . ZNF540/ZNF571 38085148 38085250 chr19:47634525- 47634420- 325 0.258 3.13 36 . SAE1 47634850 47634570 chr19:51774231- 51774480- 515 0.292 2.608 34 . C19orf75 51774746 51774630 chr19:58874038- 58874200- 58874200- 330 0.282 2.105 82 ZNF497 58874368 58874350 58874350

235

236

chr20:62580073- 62580520- 62580460- 490 0.272 4.022 30 UCKL1/UCKL1-AS1 62580563 62580670 62580610 chr20:62580073- 62580260- 62579960- 490 0.272 4.022 30 UCKL1/UCKL1-AS1 62580563 62580410 62580110 chr21:30674834- 30674740- 390 -0.231 2.512 8 . BACH1 30675224 30674890 chr21:34756882- 34757220- 515 0.22 3.368 15 . intergenic (IFNAR1-IFNGR2) 34757397 34757370 chr21:36262421- 36262400- 593 0.342 2.891 75 . RUNX1 36263014 36262550 chr21:47742584- 47742440- 515 0.22 2.457 47 . PCNT/C21orf58 47743099 47742590

236

237

Table E7. Asthma-associated DMRs in IIS children and DNase I HSS

N. OF DMRS % TOTAL % HSS COVERED P† IN HSS DMRS* BY ARRAY

DNASE I HSS IN CD14+ 2.05E 57 9.68 4.54 MONOCYTES -09

DNASE I HSS IN CD4+ 1.79E 43 7.3 3.4 NAÏVE T CELLS -07

* Total number of DMRs = 589 † by one-sample test of proportions

237

238

Table E8: Association between child and parental risk factors at birth and asthma during childhood in IIS children.

A) CHILD % WITH N P-VALUE† CHARACTERISTICS ASTHMA* Male 208 20.7 -- Child sex Female 222 15.3 0.148 Caucasian 265 16.6 -- Child ethnicity All others 165 20.0 0.372 Vaginal delivery 334 18.9 -- Mode of delivery Cesarean section 95 13.7 0.243 Undetectable 204 20.6 -- Total cord ige Detectable 147 17.0 0.400 AA 80 21.3 -- 17q21 rs8076131‡ AG/GG 178 15.2 0.230

B) PARENTAL % WITH N P-VALUE† CHARACTERISTICS ASTHMA* No 343 15.2 -- Maternal asthma Yes 83 28.9 0.003 No 133 17.3 -- Maternal allergy Yes 284 17.6 0.938 No 330 16.4 -- Paternal asthma Yes 63 20.6 0.409 No 82 15.9 -- Paternal allergy Yes 299 17.1 0.796 Maternal smoking No 411 18.3 -- (pregnancy) Yes 19 10.5 0.391 *Physician-diagnosed with symptoms or medication use for asthma in the past year reported at least once on the age 2, 3, 5 or 9-year questionnaires †by Pearson's 2 test ‡Analysis limited to Caucasian children.

238

239

REFERENCES

1. W.H.O., Asthma - Fact sheet. 2017. 2. W.H.O. Bronchial asthma - Fact sheet n. 206. [Internet] 2015; Available from: http://www.who.int/mediacentre/factsheets/fs206/en/. 3. Burrows, B., et al., The course and prognosis of different forms of chronic airways obstruction in a sample from the general population. N Engl J Med, 1987. 317(21): p. 1309-1314. 4. Martinez, F.D., What have we learned from the Tucson Children's Respiratory Study? Paediatr. Respir. Rev., 2002. 3: p. 193-197. 5. Stern, D.A., et al., Wheezing and bronchial hyper-responsiveness in early childhood as predictors of newly diagnosed asthma in early adulthood: a longitudinal birth-cohort study. Lancet., 2008. 372(9643): p. 1058-64. 6. Martinez, F.D., Maturation of immune responses at the beginning of asthma. J. Allergy Clin. Immunol., 1999. 103: p. 355-361. 7. Martinez, F.D., Toward asthma prevention--does all that really matters happen before we learn to read? N. Engl. J. Med., 2003. 349: p. 1473-1475. 8. Martinez, F.D., The origins of asthma and chronic obstructive pulmonary disease in early life. Proc Am Thorac Soc, 2009. 6(3): p. 272-7. 9. Li, X., et al., Genome-wide association study of asthma identifies RAD50-IL13 and HLA-DR/DQ regions. J. Allergy Clin. Immunol., 2010. 125: p. 328-335. 10. Moffatt, M.F., et al., A large-scale, consortium-based genomewide association study of asthma. N. Engl. J. Med., 2010. 363(13): p. 1211-21. 11. Moffatt, M.F., et al., Genetic variants regulating ORMDL3 expression contribute to the risk of childhood asthma. Nature, 2007. 448(7152): p. 470-473. 12. Torgerson, D.G., et al., Meta-analysis of genome-wide association studies of asthma in ethnically diverse North American populations. Nat. Genet., 2011. 43(9): p. 887-892. 13. Ober, C., Asthma Genetics in the Post-GWAS Era. Ann Am Thorac Soc., 2016. 13 Suppl 1: p. S85-90. 14. Lovinsky-Desir, S. and R.L. Miller, Epigenetics, Asthma, and Allergic Diseases: A Review of the Latest Advancements. Curr Allergy Asthma Rep, 2012. 15. Martinez, F.D. and D. Vercelli, Asthma. Lancet., 2013. 382(9901): p. 1360-72. 16. Berry, C.E., et al., A Distinct Low Lung Function Trajectory from Childhood to the Fourth Decade of Life. Am J Respir Crit Care Med, 2016. 194(5): p. 607-12. 17. Halonen, M., et al., Perinatal TNF-alpha production influenced by maternal pregnancy weight gain, predicts childhood asthma. Am. J. Respir. Crit. Care Med., 2013. 188(1): p. 35-41. 18. Rothers, J., et al., Adaptive cytokine production in early life differentially predicts total IgE levels and asthma through age 5 years. J. Allergy Clin. Immunol., 2011. 128(2): p. 397-402. 19. Litonjua, A.A., et al., Parental history and the risk for childhood asthma. Does mother confer more risk than father? Am J Respir Crit Care Med, 1998. 158(1): p. 176-81. 20. Lim, R.H., L. Kobzik, and M. Dahl, Risk for asthma in offspring of asthmatic mothers versus fathers: a meta-analysis. PLoS One., 2010. 5(4): p. e10134. 21. Douwes, J., et al., Farm exposure in utero may protect against asthma, hay fever and eczema. Eur Respir J., 2008. 32(3): p. 603-11. 22. Schwartz, J., et al., Predictors of asthma and persistent wheeze in a national sample of children in the United States. Association with social class, perinatal events, and race. Am Rev Respir Dis, 1990. 142(3): p. 555-62.

239

240

23. Neuman, A., et al., Maternal smoking in pregnancy and asthma in preschool children: a pooled analysis of eight birth cohorts. Am J Respir Crit Care Med, 2012. 186(10): p. 1037-43. 24. Calvani, M., et al., Infectious and uterus related complications during pregnancy and development of atopic and nonatopic asthma in children. Allergy, 2004. 59(1): p. 99-106. 25. Xu, B., et al., Maternal infections in pregnancy and the development of asthma among offspring. Int J Epidemiol, 1999. 28(4): p. 723-7. 26. Halonen, M., et al., Perinatal tumor necrosis factor-alpha production, influenced by maternal pregnancy weight gain, predicts childhood asthma. Am J Respir Crit Care Med., 2013. 188(1): p. 35-41. 27. Harpsoe, M.C., et al., Maternal obesity, gestational weight gain, and risk of asthma and atopic disease in offspring: a study within the Danish National Birth Cohort. J Allergy Clin Immunol., 2013. 131(4): p. 1033-40. 28. Stensballe, L.G., et al., Use of antibiotics during pregnancy increases the risk of asthma in early childhood. J Pediatr., 2013. 162(4): p. 832-838 e3. 29. Khan, T.K., D.J. Palmer, and S.L. Prescott, In-utero exposures and the evolving epidemiology of paediatric allergy. Curr Opin Allergy Clin Immunol., 2015. 15(5): p. 402-8. 30. Khashan, A.S., et al., Prenatal stress and risk of asthma hospitalization in the offspring: a Swedish population-based study. Psychosom Med., 2012. 74(6): p. 635-41. 31. Toskala, E. and D.W. Kennedy, Asthma risk factors. Int Forum Allergy Rhinol, 2015. 5 Suppl 1: p. S11-6. 32. Beasley, R., A. Semprini, and E.A. Mitchell, Risk factors for asthma: is prevention possible? Lancet, 2015. 386(9998): p. 1075-1085. 33. Tattersfield, A.E., et al., Asthma. Lancet, 2002. 360(9342): p. 1313-22. 34. Corren, J., Role of Interleukin-13 in Asthma. Curr Allergy Asthma Rep, 2013. 13(5): p. 415-420. 35. Makrinioti, H., et al., Role of interleukin 33 in respiratory allergy and asthma. Lancet Respir Med, 2014. 2(3): p. 226-37. 36. Bahrami Mahneh, S., et al., Serum IL-33 Is Elevated in Children with Asthma and Is Associated with Disease Severity. Int Arch Allergy Immunol, 2015. 168(3): p. 193-6. 37. Guo, Z., et al., IL-33 promotes airway remodeling and is a marker of asthma disease severity. J Asthma, 2014. 51(8): p. 863-9. 38. Kumar, R.K., P.S. Foster, and H.F. Rosenberg, Respiratory viral infection, epithelial cytokines, and innate lymphoid cells in asthma exacerbations. J Leukoc Biol, 2014. 96(3): p. 391-396. 39. Moore, W.C., et al., Identification of asthma phenotypes using cluster analysis in the Severe Asthma Research Program. Am J Respir Crit Care Med, 2010. 181(4): p. 315-23. 40. Moore, W.C. and E.R. Bleecker, Asthma heterogeneity and severity-why is comprehensive phenotyping important? Lancet Respir Med, 2014. 2(1): p. 10-1. 41. Wenzel, S.E., Asthma phenotypes: the evolution from clinical to molecular approaches. Nat Med, 2012. 18(5): p. 716-25. 42. Thomsen, S.F., et al., Estimates of asthma heritability in a large twin sample. Clin Exp Allergy, 2010. 40(7): p. 1054-61. 43. Hemminki, K., et al., Familial risks for asthma among twins and other siblings based on hospitalizations in Sweden. Clin Exp Allergy, 2007. 37(9): p. 1320-5.

240

241

44. Willemsen, G., et al., Heritability of self-reported asthma and allergy: a study in adult Dutch twins, siblings and parents. Twin Res Hum Genet, 2008. 11(2): p. 132-42. 45. Crestani, E., et al., Parental asthma as a risk factor for the development of early skin test sensitization in children. J Allergy Clin Immunol, 2004. 113(2): p. 284- 90. 46. Henderson, A.J. and J.O. Warner, Fetal origins of asthma. Semin Fetal Neonatal Med, 2012. 17(2): p. 82-91. 47. Pickrell, J.K., et al., Detection and interpretation of shared genetic influences on 42 human traits. Nat Genet, 2016. 48(7): p. 709-17. 48. Barreto-Luis, A., et al., Genome-wide association study in Spanish identifies ADAM metallopeptidase with thrombospondin type 1 motif, 9 (ADAMTS9), as a novel asthma susceptibility gene. J Allergy Clin Immunol, 2016. 137(3): p. 964- 6. 49. Ramasamy, A., et al., Genome-wide association studies of asthma in population- based cohorts confirm known and suggested loci and identify an additional association near HLA. PLoS One, 2012. 7(9): p. e44008. 50. Hirota, T., et al., Genome-wide association study identifies three new susceptibility loci for adult asthma in the Japanese population. Nat Genet, 2011. 43(9): p. 893-6. 51. Forno, E., et al., Genome-wide interaction study of dust mite allergen on lung function in children with asthma. J Allergy Clin Immunol, 2017. 140(4): p. 996- 1003.e7. 52. Torgerson, D.G., et al., Meta-analysis of genome-wide association studies of asthma in ethnically diverse North American populations. Nat Genet, 2011. 43(9): p. 887-92. 53. Wan, Y.I., et al., Genome-wide association study to identify genetic determinants of severe asthma. Thorax, 2012. 67(9): p. 762-8. 54. Bonnelykke, K., et al., A genome-wide association study identifies CDHR3 as a susceptibility locus for early childhood asthma with severe exacerbations. Nat Genet, 2014. 46(1): p. 51-5. 55. Ferreira, M.A., et al., Genome-wide association analysis identifies 11 risk variants associated with the asthma with hay fever phenotype. J Allergy Clin Immunol, 2014. 133(6): p. 1564-71. 56. Demenais, F., et al., Multiancestry association study identifies new asthma risk loci that colocalize with immune-cell enhancer marks. Nat Genet, 2018. 50(1): p. 42-53. 57. Ferreira, M.A., et al., Identification of IL6R and chromosome 11q13.5 as risk loci for asthma. Lancet, 2011. 378(9795): p. 1006-14. 58. Ferreira, M.A., et al., Shared genetic origin of asthma, hay fever and eczema elucidates allergic disease biology. Nat Genet, 2017. 49(12): p. 1752-1757. 59. Almoguera, B., et al., Identification of Four Novel Loci in Asthma in European American and African American Populations. Am J Respir Crit Care Med, 2017. 195(4): p. 456-463. 60. Li, X., et al., Genome-wide association study of asthma identifies RAD50-IL13 and HLA-DR/DQ regions. J Allergy Clin Immunol, 2010. 125(2): p. 328-335 e11. 61. Lasky-Su, J., et al., HLA-DQ strikes again: genome-wide association study further confirms HLA-DQ in the diagnosis of asthma among adults. Clin Exp Allergy, 2012. 42(12): p. 1724-33.

241

242

62. Noguchi, E., et al., Genome-wide association study identifies HLA-DP as a susceptibility gene for pediatric asthma in Asian populations. PLoS Genet, 2011. 7(7): p. e1002170. 63. Wu, A.C., et al., Inhaled corticosteroid treatment modulates ZNF432 gene variant's effect on bronchodilator response in asthmatics. J Allergy Clin Immunol, 2014. 133(3): p. 723-8.e3. 64. Duan, Q.L., et al., A genome-wide association study of bronchodilator response in asthmatics. Pharmacogenomics J, 2014. 14(1): p. 41-7. 65. Yan, Q., et al., A meta-analysis of genome-wide association studies of asthma in Puerto Ricans. Eur Respir J, 2017. 49(5). 66. Ferreira, M.A., et al., Association between ORMDL3, IL1RL1 and a deletion on chromosome 17q21 with asthma risk in Australia. Eur J Hum Genet, 2011. 19(4): p. 458-64. 67. Nieuwenhuis, M.A., et al., Combining genomewide association study and lung eQTL analysis provides evidence for novel genes associated with asthma. Allergy, 2016. 71(12): p. 1712-1720. 68. Bach, J.F., The effect of infections on susceptibility to autoimmune and allergic diseases. N Engl J Med, 2002. 347(12): p. 911-20. 69. Akinbami, L.J., et al., Trends in asthma prevalence, health care use, and mortality in the United States, 2001-2010. NCHS Data Brief, 2012(94): p. 1-8. 70. Calışkan, M., et al., Rhinovirus wheezing illness and genetic risk of childhood- onset asthma. N Engl J Med, 2013. 368(15): p. 1398-407. 71. Ege, M.J., et al., Exposure to environmental microorganisms and childhood asthma. N Engl J Med, 2011. 364(8): p. 701-9. 72. Riedler, J., et al., Exposure to farming in early life and development of asthma and allergy: a cross-sectional survey. Lancet, 2001. 358(9288): p. 1129-33. 73. von Mutius, E. and D. Vercelli, Farm living: effects on childhood asthma and allergy. Nat Rev Immunol, 2010. 10(12): p. 861-8. 74. Jackson, D.J., et al., Asthma exacerbations: origin, effect, and prevention. J Allergy Clin Immunol, 2011. 128(6): p. 1165-74. 75. Jirtle, R.L. and M.K. Skinner, Environmental epigenomics and disease susceptibility. Nat Rev Genet, 2007. 8(4): p. 253-62. 76. Ansel, K.M., et al., Regulation of Th2 differentiation and Il4 locus accessibility. Annu. Rev. Immunol., 2006. 24: p. 607-56. 77. Wilson, C.B., et al., DNA methylation and the expanding epigenetics of T cell lineage commitment. Semin. Immunol., 2005. 17: p. 105-119. 78. Takai, D. and P.A. Jones, Comprehensive analysis of CpG islands in human chromosomes 21 and 22. Proc Natl Acad Sci U S A, 2002. 99(6): p. 3740-5. 79. Gardiner-Garden, M. and M. Frommer, CpG islands in vertebrate genomes. J Mol Biol, 1987. 196(2): p. 261-82. 80. Bibikova, M., et al., High density DNA methylation array with single CpG site resolution. Genomics, 2011. 98(4): p. 288-95. 81. Pfeifer, G.P., S. Kadam, and S.G. Jin, 5-hydroxymethylcytosine and its potential roles in development and cancer. Epigenetics Chromatin, 2013. 6(1): p. 10. 82. Hackett, J.A., et al., Germline DNA demethylation dynamics and imprint erasure through 5-hydroxymethylcytosine. Science, 2013. 339(6118): p. 448-52. 83. Shen, L. and Y. Zhang, 5-Hydroxymethylcytosine: generation, fate, and genomic distribution. Curr Opin Cell Biol, 2013. 25(3): p. 289-296. 84. Pastor, W.A., et al., Genome-wide mapping of 5-hydroxymethylcytosine in embryonic stem cells. Nature, 2011. 473(7347): p. 394-7.

242

243

85. Jones, P.A., Functions of DNA methylation: islands, start sites, gene bodies and beyond. Nat Rev Genet, 2012. 13(7): p. 484-92. 86. Tirado-Magallanes, R., et al., Whole genome DNA methylation: beyond genes silencing. Oncotarget, 2017. 8(3): p. 5629-5637. 87. Furey, T.S., ChIP-seq and beyond: new and improved methodologies to detect and characterize protein-DNA interactions. Nat Rev Genet, 2012. 13(12): p. 840- 52. 88. Rauch, T.A. and G.P. Pfeifer, DNA methylation profiling using the methylated- CpG island recovery assay (MIRA). Methods, 2010. 52(3): p. 213-7. 89. Bakulski, K.M., et al., DNA methylation of cord blood cell types: Applications for mixed cell birth studies. Epigenetics, 2016: p. 1-9. 90. Yang, I.V., et al., DNA methylation and childhood asthma in the inner city. J Allergy Clin Immunol., 2015. 136(1): p. 69-80. 91. Chen, W., et al., An epigenome-wide association study of total serum IgE in Hispanic children. J Allergy Clin Immunol, 2017. 140(2): p. 571-577. 92. Liang, L., et al., An epigenome-wide association study of total serum immunoglobulin E concentration. Nature, 2015. 93. Uller, L., et al., Antagonism of the prostaglandin D2 receptor CRTH2 attenuates asthma pathology in mouse eosinophilic airway inflammation. Respir Res, 2007. 8: p. 16. 94. Arathimos, R., et al., Epigenome-wide association study of asthma and wheeze in childhood and adolescence. Clin Epigenetics, 2017. 9: p. 112. 95. Xu, C.J., et al., DNA methylation in childhood asthma: an epigenome-wide meta- analysis. Lancet Respir Med, 2018. 96. Reinius, L.E., et al., Differential DNA methylation in purified human blood cells: implications for cell lineage and studies on disease susceptibility. PLoS One, 2012. 7(7): p. e41361. 97. Yang, I.V., et al., The nasal methylome and childhood atopic asthma. J Allergy Clin Immunol, 2017. 139(5): p. 1478-1488. 98. Goh, F.Y., et al., Receptor-interacting protein 2 gene silencing attenuates allergic airway inflammation. J Immunol, 2013. 191(5): p. 2691-9. 99. Nicodemus-Johnson, J., et al., Genome-Wide Methylation Study Identifies an IL- 13-induced Epigenetic Signature in Asthmatic Airways. Am J Respir Crit Care Med, 2016. 193(4): p. 376-85. 100. Nicodemus-Johnson, J., et al., DNA methylation in lung cells is associated with asthma endotypes and genetic risk. JCI Insight, 2016. 1(20): p. e90151. 101. Yang, I.V. and D.A. Schwartz, Epigenetic mechanisms and the development of asthma. J. Allergy Clin. Immunol., 2012. 130(6): p. 1243-55. 102. DeVries, A. and D. Vercelli, Early predictors of asthma and allergy in children: The role of epigenetics. Curr. Opin. Allergy Clin. Immunol., 2015. 15(5): p. 435- 439. 103. DeVries, A. and D. Vercelli, Epigenetics in allergic diseases. Curr Opin Pediatr, 2015. 27(6): p. 719-23. 104. Rothers, J., et al., Maternal prenatal cytokine profiles predict asthma in children of nonasthmatic mothers. Am J Respir Cell Mol Biol, 2018. 105. Stein, M.M., et al., Innate Immunity and Asthma Risk in Amish and Hutterite Farm Children. N Engl J Med, 2016. 375(5): p. 411-421. 106. Gozdz, J., C. Ober, and D. Vercelli, Innate Immunity and Asthma Risk. N Engl J Med, 2016. 375(19): p. 1898-1899. 107. Meyers, D.A., et al., Asthma genetics and personalised medicine. Lancet Respir Med, 2014. 2(5): p. 405-15.

243

244

108. Barua, S. and M.A. Junaid, Lifestyle, pregnancy and epigenetic effects. Epigenomics., 2015. 7(1): p. 85-102. 109. Yang, I.V., et al., DNA methylation and childhood asthma in the inner city. J Allergy Clin Immunol, 2015. 136(1): p. 69-80. 110. Curtin, J.A., et al., Methylation of IL-2 promoter at birth alters the risk of asthma exacerbations during childhood. Clin Exp Allergy, 2013. 43(3): p. 304-11. 111. Oddy, W.H., et al., TGF- in human milk is associated with wheeze in infancy. J. Allergy Clin. Immunol., 2003. 112(4): p. 723-728. 112. Deliu, M., et al., Impact of rhinitis on asthma severity in school-age children. Allergy, 2014. 69(11): p. 1515-21. 113. van Breda, S.G., et al., Epigenetic mechanisms underlying arsenic-associated lung carcinogenesis. Arch Toxicol, 2015. 89(11): p. 1959-69. 114. Yegnasubramanian, S., et al., Combination of methylated-DNA precipitation and methylation-sensitive restriction enzymes (COMPARE-MS) for the rapid, sensitive and quantitative detection of DNA methylation. Nucleic Acids Res., 2006. 34(3): p. e19. 115. Saitoh, T., et al., Atg9a controls dsDNA-driven dynamic translocation of STING and the innate immune response. Proc. Natl. Acad. Sci. U S A, 2009. 106(49): p. 20842-20846. 116. Bosco, A., et al., Decreased activation of inflammatory networks during acute asthma exacerbations is associated with chronic airflow obstruction. Mucosal Immunol., 2010. 3(4): p. 399-409. 117. Pendergrass, S.A., et al., Phenome-wide association study (PheWAS) for detection of pleiotropy within the Population Architecture using Genomics and Epidemiology (PAGE) Network. PLoS Genet., 2013. 9(1): p. e1003087. 118. Maurano, M.T., et al., Systematic localization of common disease-associated variation in regulatory DNA. Science, 2012. 337(6099): p. 1190-5. 119. Hastie, A.T., et al., Analyses of asthma severity phenotypes and inflammatory proteins in subjects stratified by sputum granulocytes. J Allergy Clin Immunol., 2010. 125(5): p. 1028-1036 e13. 120. Raedler, D., et al., Identification of novel immune phenotypes for allergic and nonallergic childhood asthma. J Allergy Clin Immunol, 2014. 121. Simpson, J.L., et al., Elevated expression of the NLRP3 inflammasome in neutrophilic asthma. Eur Respir J, 2014. 43(4): p. 1067-76. 122. Damrauer, S.M., et al., A20 modulates lipid metabolism and energy production to promote liver regeneration. PLoS One, 2011. 6(3): p. e17715. 123. Hinds, D.A., et al., A genome-wide association meta-analysis of self-reported allergy identifies shared and allergy-specific susceptibility loci. Nat Genet, 2013. 45(8): p. 907-11. 124. Bell, J.T., et al., DNA methylation patterns associate with genetic and gene expression variation in HapMap cell lines. Genome Biol, 2011. 12(1): p. R10. 125. Bakulski, K.M., et al., DNA methylation of cord blood cell types: Applications for mixed cell birth studies. Epigenetics, 2016. 11(5): p. 354-62. 126. Anthoni, M., et al., Transforming growth factor-beta/Smad3 signalling regulates inflammatory responses in a murine model of contact hypersensitivity. Br J Dermatol, 2008. 159(3): p. 546-54. 127. Paaso, E.M., et al., Allergic diseases and asthma in the family predict the persistence and onset-age of asthma: a prospective cohort study. Respir Res, 2014. 15: p. 152.

244

245

128. Schaub, B., et al., Maternal farm exposure modulates neonatal immune mechanisms through regulatory T cells. J. Allergy Clin. Immunol., 2009. 123(4): p. 774-82. 129. Lluis, A., et al., Regulation of TH17 markers early in life through maternal farm exposure. J Allergy Clin Immunol, 2014. 133(3): p. 864-71. 130. Basu, R., et al., IL-1 signaling modulates activation of STAT transcription factors to antagonize retinoic acid signaling and control the TH17 cell-iTreg cell balance. Nat Immunol, 2015. 16(3): p. 286-95. 131. Nicodemus-Johnson, J., et al., Genome-wide Methylation Study Identifies an IL- 13 Induced Epigenetic Signature in Asthmatic Airways. Am J Respir Crit Care Med, 2015. in press. 132. Wang, G., et al., Epigenetics and early life origins of chronic noncommunicable diseases. J Adolesc Health, 2013. 52(2 Suppl 2): p. S14-21. 133. McDade, T.W., Early environments and the ecology of inflammation. Proc Natl Acad Sci U S A, 2012. 109 Suppl 2: p. 17281-8. 134. DeVries, A., et al., Epigenome-wide analysis links SMAD3 methylation at birth to asthma in children of asthmatic mothers. J Allergy Clin Immunol 2017. 140(2): p. 534-542. 135. Dedeurwaerder, S., et al., Evaluation of the Infinium Methylation 450K technology. Epigenomics, 2011. 3(6): p. 771-84. 136. Banovich, N.E., et al., Methylation QTLs Are Associated with Coordinated Changes in Transcription Factor Binding, Histone Modifications, and Gene Expression Levels. PLoS Genet, 2014. 10(9): p. e1004663. 137. Aryee, M.J., et al., Minfi: a flexible and comprehensive Bioconductor package for the analysis of Infinium DNA methylation microarrays. Bioinformatics, 2014. 30(10): p. 1363-9. 138. Maksimovic, J., L. Gordon, and A. Oshlack, SWAN: Subset-quantile within array normalization for illumina infinium HumanMethylation450 BeadChips. Genome Biol, 2012. 13(6): p. R44. 139. Johnson, W.E., C. Li, and A. Rabinovic, Adjusting batch effects in microarray expression data using empirical Bayes methods. Biostatistics, 2007. 8(1): p. 118- 27. 140. Kramer, A., et al., Causal analysis approaches in Ingenuity Pathway Analysis. Bioinformatics, 2014. 30(4): p. 523-30. 141. Calvano, S.E., et al., A network-based analysis of systemic inflammation in humans. Nature, 2005. 437(7061): p. 1032-7. 142. Rothers, J., et al., Adaptive cytokine production in early life differentially predicts total IgE levels and asthma through age 5 years. J Allergy Clin Immunol., 2011. 128(2): p. 397-402 e2. 143. Quintana, F.J., The aryl hydrocarbon receptor: a molecular pathway for the environmental control of the immune response. Immunology, 2013. 138(3): p. 183-189. 144. Park, B.V. and F. Pan, The role of nuclear receptors in regulation of Th17/Treg biology and its implications for diseases. Cell Mol Immunol, 2015. 12(5): p. 533- 542. 145. Hall, Jason A., et al., The Role of Retinoic Acid in Tolerance and Immunity. Immunity, 2011. 35(1): p. 13-22. 146. Bartram, U. and C.P. Speer, The role of transforming growth factor beta in lung development and disease. Chest, 2004. 125(2): p. 754-65.

245

246

147. Huang, Y., et al., The influence of IgE-enhancing and IgE-suppressive gammadelta T cells changes with exposure to inhaled ovalbumin. J Immunol, 2009. 183(2): p. 849-55. 148. Pang, D.J., et al., Understanding the complexity of gammadelta T-cell subsets in mouse and human. Immunology, 2012. 136(3): p. 283-90. 149. Huang, Y., et al., gammadelta T Cells Shape Preimmune Peripheral B Cell Populations. J Immunol, 2016. 196(1): p. 217-31. 150. Murdoch, J.R. and C.M. Lloyd, Resolution of allergic airway inflammation and airway hyperreactivity is mediated by IL-17-producing {gamma}{delta}T cells. Am J Respir Crit Care Med, 2010. 182(4): p. 464-76. 151. Hahn, Y.S., et al., Different potentials of gamma delta T cell subsets in regulating airway responsiveness: V gamma 1+ cells, but not V gamma 4+ cells, promote airway hyperreactivity, Th2 cytokines, and airway inflammation. J Immunol, 2004. 172(5): p. 2894-902. 152. Hahn, Y.S., et al., V gamma 4+ gamma delta T cells regulate airway hyperreactivity to methacholine in ovalbumin-sensitized and challenged mice. J Immunol, 2003. 171(6): p. 3170-8. 153. Heilig, J.S. and S. Tonegawa, Diversity of murine gamma genes and expression in fetal and adult T lymphocytes. Nature, 1986. 322(6082): p. 836-40. 154. Yamamoto, M., et al., Role of adaptor TRIF in the MyD88-independent toll-like receptor signaling pathway. Science, 2003. 301(5633): p. 640-3. 155. Dobin, A., et al., STAR: ultrafast universal RNA-seq aligner. Bioinformatics, 2013. 29(1): p. 15-21. 156. Love, M.I., W. Huber, and S. Anders, Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biol, 2014. 15(12): p. 550. 157. Langfelder, P. and S. Horvath, WGCNA: an R package for weighted correlation network analysis. BMC Bioinformatics, 2008. 9: p. 559. 158. Brandes, M., et al., Flexible migration program regulates gamma delta T-cell involvement in humoral immunity. Blood, 2003. 102(10): p. 3693-701. 159. Nishimura, H., et al., Developmentally regulated expression of the PD-1 protein on the surface of double-negative (CD4-CD8-) thymocytes. Int Immunol, 1996. 8(5): p. 773-80. 160. Hunter, S., et al., Human liver infiltrating gammadelta T cells are composed of clonally expanded circulating and tissue-resident populations. J Hepatol, 2018. 161. Fleming, C., et al., γδ T Cells: Unexpected Regulators of Cancer Development and Progression. Trends Cancer, 2017. 3(8): p. 561-570. 162. Do, J.S., et al., gammadelta T Cells Coexpressing Gut Homing alpha4beta7 and alphaE Integrins Define a Novel Subset Promoting Intestinal Inflammation. J Immunol, 2017. 198(2): p. 908-915. 163. Laird, R.M., K. Laky, and S.M. Hayes, Unexpected role for the B cell-specific Src family kinase B lymphoid kinase in the development of IL-17-producing gammadelta T cells. J Immunol, 2010. 185(11): p. 6518-27. 164. Zeng, X., et al., γδ T Cells Recognize a Microbial Encoded B Cell Antigen to Initiate a Rapid Antigen-Specific Interleukin-17 Response. Immunity, 2012. 37(3): p. 524-534. 165. Chen, Y., et al., Differential requirement of RasGRP1 for gammadelta T cell development and activation. J Immunol, 2012. 189(1): p. 61-71. 166. Zorbas, M. and R. Scollay, Development of gamma delta T cells in the adult murine thymus. Eur J Immunol, 1993. 23(7): p. 1655-60.

246

247

167. Munoz-Ruiz, M., et al., TCR signal strength controls thymic differentiation of discrete proinflammatory gammadelta T cell subsets. Nat Immunol, 2016. 17(6): p. 721-727. 168. Huang, Y., et al., gammadelta T cells affect IL-4 production and B-cell tolerance. Proc Natl Acad Sci U S A, 2015. 112(1): p. E39-48. 169. Lu, L., et al., Role of SMAD and Non-SMAD Signals in the Development of Th17 and Regulatory T Cells. J Immunol, 2010. 184(8): p. 4295-4306. 170. Yoon, J.-H., et al., Phosphorylation status determines the opposing functions of Smad2/Smad3 as STAT3 cofactors in TH17 differentiation. Nat Commun, 2015. 6: p. 7600. 171. DeVries, A. and D. Vercelli, Of pleiotropy and trajectories: Does the TGF-beta pathway link childhood asthma and chronic obstructive pulmonary disease? J Allergy Clin Immunol, 2018. 141(6): p. 1992-1996. 172. Moran, S., C. Arribas, and M. Esteller, Validation of a DNA methylation microarray for 850,000 CpG sites of the human genome enriched in enhancer sequences. Epigenomics, 2016. 8(3): p. 389-99. 173. Narayan, K., et al., Intrathymic programming of effector fates in three molecularly distinct gammadelta T cell subtypes. Nat Immunol, 2012. 13(5): p. 511-8. 174. Texido, G., et al., The B-cell-specific Src-family kinase Blk is dispensable for B- cell development and activation. Mol Cell Biol, 2000. 20(4): p. 1227-33. 175. Chien, Y.-h., X. Zeng, and I. Prinz, The natural and the inducible: interleukin (IL)-17-producing γδ T cells. Trends Immunol, 2013. 34(4): p. 151-154. 176. Do, J.-s., et al., Cutting Edge: Spontaneous Development of IL-17–Producing γδ T Cells in the Thymus Occurs via a TGF-β1–Dependent Mechanism. J Immunol, 2010. 184(4): p. 1675-1679. 177. Aryee, M.J., et al., Accurate genome-scale percentage DNA methylation estimates from microarray data. Biostatistics, 2011. 12(2): p. 197-210. 178. An integrated encyclopedia of DNA elements in the human genome. Nature, 2012. 489(7414): p. 57-74. 179. Rothers, J., et al., Cord blood 25-hydroxyvitamin D levels are associated with aeroallergen sensitization in children from Tucson, Arizona. J Allergy Clin Immunol, 2011. 128(5): p. 1093-9.e1-5. 180. Marinho, S., et al., Quantification of atopy and the probability of rhinitis in preschool children: a population-based birth cohort study. Allergy, 2007. 62(12): p. 1379-86. 181. Jackson, D.J., et al., Wheezing rhinovirus illnesses in early life predict asthma development in high-risk children. Am J Respir Crit Care Med, 2008. 178(7): p. 667-72.

247