Predicting asthma phenotypes: characterization of IL1RL1 in asthma
Dijk, Fokelina Nicole
IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite from it. Please check the document version below.
Document Version
Publisher's PDF, also known as Version of record
Publication date: 2018
Link to publication in University of Groningen/UMCG research database
Citation for published version (APA):
Dijk, F. N. (2018). Predicting asthma phenotypes: characterization of IL1RL1 in asthma. Rijksuniversiteit Groningen.
Copyright
Other than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).
Take-down policy
If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.
Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons the number of authors shown on this cover page is limited to 10 maximum.
Marie E. Ketelaar
-The research described in this thesis was performed at the Department of Pediatric Pul-monology and Pediatric Allergology, Beatrix Children’s Hospital, University Medical Centre Groningen, Groningen, The Netherlands.
The studies presented in this thesis were funded by the Lung Foundation of the Nether-lands (grant no. AF 95.05, AF 98.48, AF3.2.09.081JU and AF 3.4.05.033), the Stichting Ast-ma Bestrijding (SAB 2006/018), the MeDALL study (European Union, FP7 project, grant agreement No. 261357) and the Ubbo Emmius Fund.
Publication of this thesis was financially supported by the University Medical Centre Groningen, the University of Groningen, and the Graduate School GUIDE of the University of Groningen. The author gratefully acknowledge the financial support for printing of this thesis by Stichting Astma Bestrijding (SAB) and the Lung Foundation of the Netherlands. Cover design and lay out by Ellen Beck
Printed by Gildeprint
ISBN 978-94-034-1132-3 (printed version) ISBN 978-94-034-1131-6 (digital version) Copyright© F.N. Dijk, 2018
Prof. dr. D.S. Postma
Beoordelingscommissie
Prof. dr. H.M. Boezen Prof. dr. C.K. van der Ent Prof. dr. J.V. Fahy
naar op weg bent
-3.
4.
5.
Chapter 1General introduction (Page 8)
Chapter 2
Genetics of onset of asthma (Page 26)
Chapter 3
Association of IL33-IL-1 receptor-like 1 (IL1RL1) pathway polymorphisms with wheezing phenotypes and asthma in childhood (Page 40)
Chapter 4
Phenotypic and functional translation of IL1RL1 locus polymorphisms in lung tissue and airway epithelium in asthma
(Page 60) Chapter 5
Genetic regulation of IL1RL1 methylation and IL1RL1-a protein levels in asthma
8.
9.
7.
6.
Chapter 6
IL1RL1 gene variations are associated with asthma exacerbations in children and adolescents using inhaled corticosteroids (Page 134)
Chapter 7
TRPA1 gene polymorphisms and childhood asthma
(Page 156) Chapter 8
Predictive value of serum sST2 in preschool wheezers for development of asthma with high FeNO
(Page 180)
Chapter 9
Genetic risk scores do not improve asthma prediction in childhood (Page 194)
Chapter 10
Summary, general discussion and implication of results
(Page 214) Chapter 11
• Nederlandse samenvatting • Dankwoord
• About the author • List of publications (Page 234)
General introduction
_
General introduction
Asthma
Asthma is among the most frequent chronic diseases in children worldwide, with 100,0000 children be-ing affected in the Netherlands.1,2 There is an increase in asthma prevalence in the Western world, which
is accompanied by the observation that more people are concomitantly affected by one or more other allergic disorders such as eczema, rhinitis and food allergy.1 Asthma is a multifactorial disease caused by
a complex interaction between multiple genetic and environmental influences.1 The disease is
character-ized by ongoing airway inflammation associated with variable airflow obstruction, bronchial hyperres-ponsiveness (BHR) and airway remodeling.3 Asthma symptoms include episodes of wheeze, cough, and
shortness of breath and occur by exposure to various triggers, such as allergens, (viral) infections, specific weather conditions or exercise.
Compared to healthy peers, children with asthma experience a lower quality of life, and it is the leading cause of emergency room visits, hospitalizations, school absenteeism, and school underperformance.4,5
It also affects their psychological functioning, physical activity and their quality of life of that of their parents.6 Approximately 250,000 people die prematurely from asthma each year, with asthma mortality
rates in children ranging from 0.0-0.7 per 100,000.1,4 Moreover, the disease has a considerable economic
burden with high medical costs due to emergency room visits/hospitalizations and use of pharmaceuti-cal agents.1
Obtaining better insight in factors causing asthma will lead to better prevention strategies and improve-ment in therapeutic, personalized manageimprove-ment, which is greatly warranted.
Diagnosis of asthma
Thirty to 50% of preschool children experience symptoms suggestive of asthma, for instance they report recurrent wheeze or cough.7 However, only approximately 30% of these children with respiratory
symp-toms at preschool age will develop asthma at a later age. Due to the non-specific sympsymp-toms of asthma, it is hard to diagnose asthma or to determine which child will develop the disease and which child will not.8
Furthermore, there is no specific diagnostic test in childhood for asthma and lung function tests are often unreliable in preschool children.
Specific phenotypes of asthma
Asthma has a large phenotypic heterogeneity. Different asthma phenotypes can be distinguished based on the presence, timing, and severity of symptoms, such as age of onset and nocturnal symptoms, in com-bination with the presence of atopy, responsiveness to triggers, and type of airway inflammation. In the past, multiple approaches have been used to group different asthma phenotypes, in which cluster analyses of mostly adults has led to the identification of 5 specific groups based on age, gender, lung function, the presence of atopy, health care utilization and body mass index.9,10
Pediatric asthma differs from adult asthma in multiple ways11, therefore it is important to determine
Data driven approaches and longitudinal latent class analyses in children from two different birth co-horts have led to the identification of longitudinal wheezing phenotypes.12,13 A study performed in two
large birth cohorts identified 5 specific wheezing phenotypes, i.e. 1) never/infrequent wheeze, 2) tran-sient early wheeze, 3) intermediate onset wheeze, 4) late onset wheeze and 5) persistent wheeze.13 In
one birth cohort, transient wheeze was split into transient early wheeze and prolonged early wheeze. All those phenotypes were differentially associated with asthma, atopy, BHR and lung function at the age of 8 years, with intermediate onset, late onset and persistent wheeze being associated with an increased risk for doctor’s diagnosis of asthma and sensitization against common allergens, both at age 8 (Figure 1, Savenije et al.).13
Figure 1. The five discovered wheezing phenotypes in PIAMA, with their estimated prevalence of wheeze at each time point from birth to age 8 years. Reprinted from Savenije et al., with permission.13
In asthma, there has traditionally been thought to be a shift, or predilection, toward a Th2-cytokine profile resulting in eosinophilic inflammation characteristic of asthma. Aeroallergens or other environ-mental insults activate airway epithelial cells that release activators such as IL-25, IL-33 or TSLP, which in turn activate immune cells such as dendritic cells, Th2 cells and Innate type 2 lymphoid cells. The latter two are responsible for the increased production of type 2 cytokines such as IL-4, IL-5 and IL-13, leading to downstream cascade of inflammatory events (Figure 2, Fahy JV).14–16 Type 2 cytokines are thought to
be responsible for IgE overproduction by B-cells, chemo-attraction of effector cells (mast cells, eosino-phils and basoeosino-phils), development of BHR and remodeling of the airway epithelium, all characteristics of asthma.15,16 Type 2 associated asthma starts most frequently during childhood (childhood-onset
asth-ma) and is usually well responsive to corticosteroids.16,17 However, nowadays there is increasing evidence
that a large proportion of asthma patients has airway inflammation driven by other mechanisms. Cellu-lar research on gene expression in bronchial brushes has shown evidence for distinct Th2 high and Th2 low asthma phenotypes.17–19 In line with this, a classification based on eosinophilic and non-eosinophilic
asthma phenotypes has also been described.9,20,21 Those phenotypes have been found to have different
airway histopathology, airway structure, mechanisms of airway dysfunction, time of onset of symptoms and difference response to asthma treatment.
Novel phenotypic strategies have attempted to cluster asthma patients based on 1) association with environmental factors (air pollution, cigarette smoke), 2) specific symptoms or clinical characteristics (age of onset, obesity) and 3) the presence of certain biomarkers (eosinophilic, neutrophilic). One ex-ample is the definitions proposed by Hekking et al. (Figure 3)18, who suggested 13 distinct clinical
Identification of specific (biological) phenotypes is helpful in better understanding of disease pathogen-esis and susceptibility as well as starting personalized treatment. Moreover, the identification of asthma phenotypes aids the interpretation of asthma associated genes and can lead to more targeted genetic and perhaps subsequent intervention studies.
Figure 2. Type 2 immune responses characteristic for asthma. Airway epithelial derived cytokines, such as interleu-kin-33 (IL-33) and thymic stromal lymphopoeitin (TSLP), induces the expression of OX40 ligand (OX40L) on dendrit-ic cells (DCs). This leads to their mobilization to nearby lymph nodes where they activate naive CD4+ T cells to an IL-4-competent state. These IL-4-competent T cells migrate to B cell zones and differentiate into T follicular helper (TFH) cells and move into the circulation to mature to Th2 cells. IL-4-secreting TFH cells in parafollicular B cell zones mediate IgE class-switching in B cells, whereas TH2 cells that migrate to the airway epithelium and to the subepithelial mucosa secrete IL-5 and IL-13 to mediate inflammatory and remodelling changes in the airway mucosa that predis-pose an individual to asthma and to asthma exacerbations. Reprinted from Fahy JV, with permission.16
Figure 3. Grouping of asthma phenotypes in adulthood, divided by environmental factors, symptoms, clinical character-istics and eosinophilic versus non- eosinophilic airway inflammation. Reprinted from Hekking et al., with permission.18 The (epi)genome
The genome is the complete set of a person’s nucleic acid sequences, encoded as DNA within 22 chro-mosome pairs and one pair of sex chrochro-mosomes. DNA is located in the nucleus of each cell and in small DNA molecules within mitochondria. Human DNA is composed of base pairs that form a double helix structure when joined together. There are four bases; the base adenosine (A) always binds with thymine (T), and the base cytosine (C) always binds with guanine (G). Somatic cells of humans consist of six billion base pairs. Amino acids are formed from translation of three base pairs and they are the cornerstones to build proteins. DNA consists of coding and non-coding DNA, in which only coding DNA, around 1.5% of the entire DNA, is responsible for the production of proteins. Proteins are formed by transcription from DNA to mRNA followed by translation from mRNA to protein. Transcription and translation of a single gene results in one protein or different protein isoforms. The human genome contains approximately 25,000 genes, but several biological processes (such as alternative mRNA splicing) can lead to the forma-tion of many more unique proteins than the number of genes.22,23
The DNA of individuals differs only on 0.1% of this genetic variation, leading to variation in about 12 mil-lion base pairs. A variation in one base pair between persons that is prevalent in the population is called a single nucleotide polymorphism (SNP) (Figure 4), whereas rare variants (allele frequency <1%) are called mutations. The frequency at which the second most common allele is present in a population is called the minor allele frequency (MAF).22,24
SNPs in close physical proximity are frequently inherited together which results in a correlation between specific variants. This correlation is called linkage disequilibrium (LD, expressed as r2 or D’) when the
fre-quency of association of different SNPs is higher than what would be expected if the SNPs were inde-pendently inherited from each other.25
Figure 4. Overview of DNA located in a chromosome in the nucleus of a cell. DNA is composed of a double helix struc-ture with base pairs adenosine (A) and thymine (T), and cytosine and guanine (G). A single nucleotide polymorphism (SNP) is variation in one base pair between individuals.
Besides genetics, the study of genes, genetic variation, and heredity, there has become more and more attention to epigenetics, which is the study of heritable changes in gene function that do not involve changes in the DNA sequence.23,24 This can involve for example changes that affect gene expression and
function. Mechanisms responsible for such changes are DNA methylation, histone modification and the presence of small or long non-coding mRNAs. DNA methylation is the process in which a methyl group (CH3) is added to carbon 5 in a cytosine base to create 5-methylcytosine, which affects the regulation of gene expression. Methylation usually occurs at the sequence CG (5′-C-phosphate-G-3′) (CpG) and regions with a high frequency of CpG sites (CpG islands) are highly regulatory units.
Genetic association studies are performed to identify SNPs that are associated with diseases or specific traits (Figure 5). Different variants of a SNP called alleles, are being compared between patients (cases) and non-affected individuals (controls) (Figure 6). Allelic variation can regulate levels of mRNA expres-sion (called expresexpres-sion Quantitative Trait Locus, eQTL), DNA methylation at CpG sites (methylation Quan-titative Trait Locus, meQTLs), protein levels (protein QuanQuan-titative Trait Locus, pQTLs) or protein function.24
Figure 5. Flow chart of genetic association studies that can be performed to identify single nucleotide polymorphisms (SNPs) that are associated with methylation (such as with CpG sites, meQTL), expression (such as with mRNA, eQTL), or protein (such as with receptors, pQTL). eQTM, expression quantitative trait methylations; pQTM, protein quantitative trait methylations.
SNPs in genes and CpG sites on genes associated with the disease can be discovered by performing genome wide association studies (GWAS) or epigenome wide association studies (EWAS).
Thus, by combining the results of these association studies, more knowledge can be gained on how all processes influ-ence each other (e.g. expression and protein) and the biological pathways underlining specific disease associations. In addition, these studies allow us to assess effects on asthma treatment and asthma phenotypes.
Figure 6. The concept of genetic association testing in which the differences in allele frequency between cases and controls is being tested. Significant findings are identified as risk/protective alleles for the trait being tested.
(Epi)genetics of asthma
Genetic family based and twin studies have shown that the heritability of asthma is around 60%.26
Iden-tification of the genetic factors that are responsible for asthma is difficult, since the disease is highly poly-genic; disparate genes are leading to the same disease in different subjects. Publication of the HapMAP project27 and the 1000 genomes project28 let to a detailed catalogue of human genetic variation, with the
presence of sequence variants of more than 2000 participants from a number of different ethnicities. With this data it has become more accessible with increase in power to perform large hypothesis free genome wide association studies (GWAS). Over more than 90 GWAS have been performed in asthma, with the identification of more than 850 associated variants in multiple susceptibility genes.29 Recently, two
consor-tia performed large meta-analyses of published GWAS on asthma30 and allergic diseases (either asthma,
hay fever or eczema)31, which led to doubling of the number of genetic risk factors that are associated with
asthma. The TAGC consortium meta-analysis performed in 142,000 subjects from different ethnicities has identified 18 asthma associated loci. Moreover, 136 independent genetic asthma and allergy variants were discovered in the meta-analyses of the SHARE consortium performed in 360,000 subjects.
Besides the identification of genes that are associated with asthma, epigenome wide studies (EWAS) have additionally been performed, showing that epigenetic regulation, such as methylation, has a strong association with asthma development as well.32,33 Epigenetics changes with exposures and with age,
moreover it can specifically occur in one or a few cell types, therefore it is important to consider methyla-tion stages of different cells at different stages in life.
The discovery of (epi)genetic associations with asthma is promising, but the step from association to the mechanisms leading to the development of the disease is complex. To make it more complex, some genes will be more specifically related to the time of onset of the disease34, while others are more associated with
severity or response to treatment.35 In addition, gene-gene36,37, gene-environment36,38 and more recently
gene-epigenetic39 interactions have been described.
IL1RL1, asthma associated gene
In 2008, Reijmerink et al. from our group were the first to describe the association of SNPs in the interleu-kin-1 receptor–like 1 (IL1RL1), and IL18R1 gene with asthma and associated phenotypes.40
In 2009, a strong conformation of IL1RL1 as a new asthma gene, was provided by a GWAS on peripheral blood eosinophilia in the Icelandic population, and subsequently association with asthma was found in ten populations.41 A year later this gene was reported to be associated with (childhood onset) asthma in
one of the largest consortium based GWAS on asthma.34 This finding has been reproduced by multiple
studies42, with (SNPs) in IL1RL1 being associated with asthma and other atopic traits.40,43–50 Because of a
complex LD structure in the IL1RL1 region it has been difficult to identify the true causal variants related to asthma.42
IL1RL1 (also called ST2) is part of the Toll-like/IL-1 receptor superfamily with expression on inflammato-ry cells present in the lung.51 The gene is located on 2q12, consisting of a distal and proximal promotor
and 13 exons. Alternative splicing leads to four different isoforms: IL1RL1-a (soluble ST2, sST2), which can be measured in serum; a transmembrane receptor, IL1RL1-b (S2TL); and two less well-character-ized isoforms, isoform 3 and IL1RL1-c (ST2V)52,53 (Figure 7, Dijk et al. adjusted).54
Figure 7. The interleukin-1 receptor-like 1 (IL1RL1) gene (GRCh37/hg19; chr2:102,927,962–102,968,497) with transcript annotation of IL1RL1-a, the soluble variant (sST2) (ENST00000311734.6), IL1RL1-b, the transmembrane variant (ST2L) (ENST00000233954.5), and two less well known variants, IL1RL1-c (ST2V) (ENST00000427077.1) and isoform 3 (ENST00000404917.6). Adjusted from Dijk et al.54
IL1RL1 is the receptor for Interleukin-33 (IL-33), a cytokine that is important in the regulation of several allergic disorders.55 Binding of IL-33 to the receptor complex of IL1RL1-b and IL-1 receptor accessory
pro-tein (IL-1RAcP) present on Th2 cells, basophils and mast cells and type 2 innate lymphoid cells leads to an inflammatory signaling cascade with the release of pro-inflammatory Th2 cytokines that facilitate inflammation in the lung.56 In contrast IL1RL1-a is thought to serve as a decoy receptor, and sequestering
IL-33 leading to blockade of its function (Figure 8, Grotenboer et al.).57
Figure 8. The IL33-IL1RL1 signaling pathway assocaited with airway inflammation. Airway epithelial damage caused by allergic or enviromental stimuli will lead to IL-33 release. IL-33 binds to a receptor complex of IL1RL1-b and IL1RAcP which results in activation of multiple adaptors and signaling proteins, such as Mal or MyD88 (see lower insertion). This signaling cascade will initiates the release of pro-inflammatory Th2 cytokines promoting airway inflammation. IL1RL1-b can in addition inhibit receptor signaling through TIR domain–dependent sequestration of the adaptor pro-teins Mal and MYD88 (see upper insertion). The neutralizing effect of IL-33 by IL1RL1-a is also shown. Reprinted from Grotenboer et al., with permission.42
In this thesis, we hypothesize that SNPs and methylation of IL1RL1 are associated with a disbalance in IL1RL1 gene expression, resulting in increased IL1RL1-b transcription and upregulation of pro-inflamma-tory Th2 responses in asthma. In addition, downregulation of IL1RL-a levels may possibly attenuate Th2 inflammation. Finally, in patients with asthma induced by IL1RL1 dysfunction particularly, the disease might be characterized by specific Th2 inflammatory (bio)markers such as eosinophilic inflammation and an increase in fractional exhaled nitric oxide (FeNO).
Pharmacogenetics
Asthma treatment is based on a stepwise protocol, yet the different phenotypes as mentioned above are currently treated very similar. Most asthma medications (ICS, short- and long-acting b2-agonists, and leukotriene receptor antagonists) are effective in most patients. Notwithstanding this, there is still a considerable proportion of asthma patients that responds less to these treatments or exhibits side- effects.58 It is thought that this heterogeneous response to medication might be genetically determined.
Pharmacogenetics is the study of how genetic variations influence drug response, with recent candidate and GWAS studies discovering multiple genes that exert an effect on asthma treatment response.59
Recently, molecular insight into asthma development has led to the identification of new therapeutic regimens focused on for example cytokine inhibition, such as anti-IgE, anti-IL-5 or anti-IL-13, all with the aim to downregulate the inflammatory response.60
Eosinophilic, Th2 associated airway inflammation has been associated with responsiveness to ICS in asth-ma.17,61 Since IL1RL1 is thought to be important in type 2 inflammatory responses, it might be that patients
with specific IL1RL1 risk variants represent responders to ICS therapy. Identification of such genetic variants could lead to better, personalized treatment.
Prediction of asthma
Accurately predicting which child will develop asthma is important for primary prevention and for improve-ments in asthma diagnosis. Yet it is extremely difficult at childhood to diagnose asthma, this may lead to over-treatment with steroids in children that do effectively not have asthma, but also to a late diagnosis and possibly under-treatment, which is associated with reduced lung growth, a high personal burden due to recur-rent symptoms, school absence and consequently a low quality of life.62,63
Several prediction models based on personal and environmental factors have been developed to improve the early diagnosis of asthma64–66 All these predictions are based on preschool children with respiratory
symptoms, a subjective measure. Next to symptoms, asthma is characterized by variable airflow obstruc-tion, an objective measure. We know that the impairment of lung function at one month after birth is a risk factor for asthma at age 8.67 Further, approximately 75% of all children with mild-to-moderate se-vere asthma have abnormal patterns of lung growth and early decline of lung function.68 Up to now there
is no risk model that can predict the development of asthma directly after birth when no symptoms have yet occurred. Attempts have been made to develop genetic risks scores for asthma.69,70 However,
predic-tion models made thus far have no sufficiently sensitive or specific clinical value. Since the multifactorial origin of asthma, with personal- and environmental factors as well as genetic factors being involved, it might be that better prediction can be achieved by combining all these factors into one predictive score.
Study groups
To identify (epi)genetic associations and link those to phenotypic traits or search for interactions with environmental factors large cohorts and studies are needed. For the purpose of this study we were able to use data from different studies, like birth cohorts, family studies and case-control studies. These studies originated from all over the world and multiple ethnicities were included. Information present in the cohorts included complete phenotypic, perinatal, familial, (epi) genetic and protein data.
For the investigation in heterogeneous diseases such as asthma there is the need to use large study groups. By combining data from multiple cohorts, when for example performing meta-analyses or repli-cation studies, we were able to increase power in our research. Harmonization of definitions, serum mea-surements or genetic data is important in achieving a well powered and reliable outcome. The included cohorts in this thesis are summarized in Table 1.
In almost all the studies performed for this thesis we used data from the Prevention and Incidence of Asthma and Mite Allergy (PIAMA) study.71 This is a multicenter Dutch prospective birth cohort, which was
initiated in 1996. Children were recruited from the north (Groningen and surrounding), middle (Utrecht and surrounding) and west (Rotterdam and surrounding) in the Netherlands. 7862 pregnant women (2779 with allergy and 5083 without allergy) were invited to participate in the study. At the end 3963 live-born children participated in the study (1327 with a mother with allergy were defined as high-risk, and 2726 children with a mother without allergy were defined as low-risk). Questionnaires for parental com-pletion, partly based on the International Study of Asthma and Allergies in Childhood core question-naires, were sent to the parents during pregnancy, when the infant was 3 months old, yearly from 1 until 8 years, and at ages 11, 14 and 17 years. Subgroups of high-risk children and low-risk children were selected for an extensive medical examination at age 4, 8, 12 and 16 years. Blood or a buccal brush was used to obtain DNA data.
The PIAMA cohort is a great example of how birth cohorts can help in the investigation of risk factors related to disease such as house dust mites for allergy72 or sugar-containing beverages for asthma.73 The
fact that the cohort has participants from rural and more urban areas provides the ability for example to investigate the effect of air pollution differences on asthma development throughout childhood and adolescence.74 The identification of the aforementioned longitudinal phenotypes13, which was possible
because of the intensive follow-up present, is furthermore an example of the power of this study. The effectiveness of collaboration between cohorts is proven by the studies we performed in cohorts as part of the Mechanisms of the Development of Allergy (MeDALL) project75 and the Pharmacogenomics in
Childhood Asthma (PiCA) consortium.76 The MeDALL project consists of 14 European birth cohorts (with
PIAMA being one of them), including 44,010 participants, which were followed up between pregnancy and age 20 years. With usage of a standardized MeDALL Core Questionnaire and a strictly defined defi-nition for asthma, there is a high improvement in linking epidemiologic, clinical, and functional research among studies, needed to understand the complex mechanisms behind multifactorial diseases. For the pharmacogenetic study described in this thesis we used data from 5 cohorts of the PiCA consor-tium76, an international collaboration cohort, integrating a total of 21 studies worldwide, with the
inclu-sion of 14,227 children/adolescences from 12 different countries. The consortium contains 619 users of ICS who have data on asthma symptoms, exacerbations and treatment response. This research project provides a unique change to discover (pharmaco)genetic differences between multiple ethnicities which could lead to novel precision based asthma medicine.
Table 1. Description of the cohorts and data used for this thesis.
*The PIAMA, BAMSE and INMA cohort are part of the Mechanisms of the Development of Allergy (MeDALL) project. #The PACMAN, GALAII, SAGE, SLOVENIA and ESTATe cohort are all part of the Pharmacogenomics in Childhood Asthma (PiCA) consortium.
Aims of this thesis
The first aim of this thesis is to improve our understanding of the role of (epi)genetics in specific asthma phe-notypes, with a special focus on the IL1RL1 gene. The second aim of this study is to gain more insight on the role of IL1RL1 in the effect of ICS treatment in asthma. And finally, the third aim is to investigate the added value of genetics in asthma prediction.
Scope of the thesis
Chapter 2 gives an overview of the progress in defining early wheezing phenotypes and describes genetic
factors associated with the age of onset of asthma. Furthermore a comparison between asthma and ato-py associated genes is made.
In Chapter 3 we show the results of a candidate gene study performed in two birth cohorts in whom we
investigated the association of the IL33-IL1RL1 pathway with asthma and longitudinal wheezing pheno-types. The phenotypes used have been identified and studied previously by our group as described above, with the important finding that specific phenotypes show different associations with lung function, sensitization and asthma. In this chapter we also searched for the association of gene-gene interactions within the IL33-IL1RL1 pathway with asthma and the asthma associated wheezing phenotypes.
With the use of three independent asthma cohorts and re-sequencing data we investigated in Chapter 4 the
association of independent IL1RL1 asthma related genetic variants to specific features of asthma as defined by clinical and immunological measures. Moreover, with the use resequencing data and in vitro studies we tried to identify SNP driven mechanisms that may contribute to the identified genetic association signals in lung and airway structural cells.
Since the (epi)genetic regulation of IL1RL1 protein expression has not been established we focus in Chap-ter 5 on the association between IL1RL1 genetic variants, IL1RL1 blood DNA methylation and serum
IL-1RL1-a protein levels. With this study, performed in one adult cohort and three birth cohorts, we aimed to identify causal pathways in asthma relating to genetic variation in IL1RL1, white blood cell IL1RL1 methyl-ation and IL1RL1-a protein expression.
The translation of genetic and functional data related to a disease back to the patients is of great impor-tance. In Chapter 6 we therefore investigated the role of IL1RL1 genetic variants on exacerbations, asthma
control, FeNO levels and FEV1% predicted in multi-ancestry asthma patients during ICS treatment. Fur-thermore we aimed to identify whether there is a pharmacogenetic effect of IL1RL1 SNPs on ICS treat-ment response in asthma patients.
In Chapter 7 we tried to replicate experimental data in animals in human studies. In this chapter we focus
on the transient receptor potential ankyrin-1 (TRPA1) gene, which has been thought to play a key role in pro-moting airway inflammation in asthma and may mediate effects of paracetamol on asthma. We inves-tigated in two large birth cohorts the association between TRPA1 gene variants with childhood asthma and total IgE concentration. Especially, we searched for interactions between TRPA1 and prenatal parac-etamol exposure on these outcomes.
Asthma prediction might be improved with the identification of novel biomarkers, more specific defi-nitions of asthma subphenotypes and better understanding of the genetic and biological pathways in-volved in asthma. In Chapter 8 we investigated in preschool wheezers whether serum IL1RL1-a levels can
be used to predict asthma, with a special focus on asthma with elevated FeNO, as a marker of eosinophilic asthma. We compared this to the commonly used Asthma Prediction Index (API).
In Chapter 9 we aimed to generate a prediction model for asthma in the first 8 years of life based on
the combination of family, perinatal, environmental and genetic risk factors. Several prediction models based on personal and environmental factors have been developed to improve early diagnosis of asthma. With this study set out to investigate the added value of genetics at predicting asthma at birth.
References
1. Masoli M, Fabian D, Holt S, Beasley R, Program GI for A (GINA). The global burden of asthma: executive summary of the GINA Dissemination Committee report. Allergy. 2004;59:469–78.
2. Volksgezondheidenzorg.info (2018): https://www. volksgezondheidenzorg.info/onderwerp/astma, RIVM: Bilthoven, juli 2018.
3. Global Initiative for Asthma. Global Strategy for Asthma Management and Prevention, 2018. Available from: www.ginasthma.org.
4. Asher I, Pearce N. Global burden of asthma among children. Int J Tuberc Lung Dis. 2014;18:1269–78. 5. Stridsman C, Dahlberg E, Zandrén K, Hedman L.
Asthma in adolescence affects daily life and school attendance - Two cross-sectional population-based studies 10 years apart. Nurs Open. 2017;4:143–8. 6. Cerdan NS, Alpert PT, Moonie S, Cyrkiel D, Rue S.
Asthma severity in children and the quality of life of their parents. Appl Nurs Res. 2012;25:131–7. 7. Sears MR, Greene JM, Willan AR, Wiecek EM, Taylor
DR, Flannery EM, et al. A longitudinal, population-based, cohort study of childhood asthma followed to adulthood. N Engl J Med. 2003;349:1414–22. 8. Savenije OE, Kerkhof M, Koppelman GH, Postma
DS. Predicting who will have asthma at school age among preschool children. J Allergy Clin Immunol. 2012;130:325–31.
9. Haldar P, Pavord ID, Shaw DE, Berry MA, Thomas M, Brightling CE, et al. Cluster analysis and clinical asthma phenotypes. Am J Respir Crit Care Med. 2008;178:218–24.
10. Moore WC, Meyers DA, Wenzel SE, Teague WG, Li H, Li X, et al. Identification of asthma phenotypes using cluster analysis in the severe asthma research program. Am J Respir Crit Care Med. 2010;181:315–23.
11. Gelfand EW. Pediatric Asthma: A Different Disease. Proc Am Thorac Soc. 2009;6:278–82.
12. Martinez FD, Wright AL, Taussig LM, Holberg CJ, Halonen M, Morgan WJ. Asthma and wheezing in the first six years of life. The Group Health Medical Associates. N Engl J Med. 1995;332:133–8.
13. Savenije OE, Granell R, Caudri D, Koppelman GH, Smit HA, Wijga A, et al. Comparison of childhood wheezing phenotypes in 2 birth cohorts: ALSPAC and PIAMA. J Allergy Clin Immunol. 2011;127:1505–12.e14. 14. Barnes PJ. Review series The cytokine network in
asthma and chronic obstructive pulmonary disease. J Clin Invest. 2008;118:3546–56.
15. Schatz M, Rosenwasser L. The Allergic Asthma Phenotype. J Allergy Clin Immunol Pract. 2014;2:645–8. 16. Fahy J V. Type 2 inflammation in asthma — present
in most, absent in many. Nat Rev Immunol. 2014;15:57–65.
17. Woodruff PG, Modrek B, Choy DF, Jia G, Abbas AR, Ellwanger A, et al. T-helper Type 2 – driven Inflammation Defines Major Subphenotypes of Asthma. Am J Respir Crit Care Med. 2009;180:388-95. 18. Hekking PPW, Bel EH. Developing and emerging
clinical asthma phenotypes. J Allergy Clin Immunol Pract. 2014;2:671–80.
19. Zhang XY, Simpson JL, Powell H, Yang IA, Upham JW, Reynolds PN, et al. Full blood count parameters for the detection of asthma inflammatory phenotypes. Clin Exp Allergy. 2014;44:1137–45.
20. Haldar P, Pavord ID. Noneosinophilic asthma: A distinct clinical and pathologic phenotype. J Allergy Clin Immunol. 2007;119:1043–52.
21. Brooks CR, Van Dalen CJ, Zacharasiewicz A, Simpson JL, Harper JL, Le Gros G, et al. Absence of airway inflammation in a large proportion of adolescents with asthma. Respirology. 2016;21:460–6.
22. E. S. Lander et al. Initial sequencing and analysis of the human genome. 2001;409.
23. Dunham I, Kundaje A, Aldred SF, Collins PJ, Davis CA, Doyle F, et al. An integrated encyclopedia of DNA elements in the human genome. Nature. 2012;489:57–74.
24. McCarthy MI, Abecasis GR, Cardon LR, Goldstein DB, Little J, Ioannidis JP, et al. Genome-wide association studies for complex traits: consensus, uncertainty and challenges. Nat Rev. 2008;9:356–69.
25. Slatkin M. Linkage disequilibrium - Understanding the evolutionary past and mapping the medical future. Nat Rev Genet. 2008;9:477–85.
26. Koppelman GH, Los H, Postma DS. Genetic and environment in asthma: the answer of twin studies. Eur Respir J. 1999;13:2–4.
27. Altshuler DM, Gibbs RA, Peltonen L, Altshuler DM, Gibbs RA, Peltonen L, et al. Integrating common and rare genetic variation in diverse human populations. Nature. 2010;467:52–8.
28. Auton A, Abecasis GR, Altshuler DM, Durbin RM, Abecasis GR, Bentley DR, et al. A global reference for human genetic variation. Nature. 2015;526:68–74. 29. MacArthur J, Bowler E, Cerezo M, Gil L, Hall P, Hastings
E, et al. The new NHGRI-EBI Catalog of published genome-wide association studies (GWAS Catalog). Nucleic Acids Res. 2017;45:896–901.
30. Demenais F, Margaritte-Jeannin P, Barnes KC, Cookson WOC, Altmüller J, Ang W, et al. Multiancestry association study identifies new asthma risk loci that colocalize with immune-cell enhancer marks. Nat Genet. 2018;50:42–50.
31. Ferreira MA, Vonk JM, Baurecht H, Marenholz I, Tian C, Hoffman JD, et al. Shared genetic origin of asthma, hay fever and eczema elucidates allergic disease biology. Nat Genet. 2017;49:1752–7.
32. Davidson EJ, Yang I V. Role of epigenetics in the development of childhood asthma. Curr Opin Allergy Clin Immunol. 2018;18:132-38.
33. Xu CJ, Söderhäll C, Bustamante M, Baïz N, Gruzieva O, Gehring U, et al. DNA methylation in childhood asthma: an epigenome-wide meta-analysis. Lancet Respir Med. 2018;6:379–88.
34. Moffatt MF, Gut IG, Demenais F, Strachan DP, Bouzigon E, Heath S, et al. A large-scale, consortium-based genomewide association study of asthma. N Engl J Med. 2010;363:1211–21.
35. Perin P, Potočnik U. Polymorphisms in recent GWA identified asthma genes CA10, SGK493, and CTNNA3 are associated with disease severity and treatment response in childhood asthma. Immunogenetics. 2014;66:143–51.
36. Su MW, Tung KY, Liang PH, Tsai CH, Kuo NW, Lee YL. Gene-gene and gene-environmental interactions of childhood asthma: A multifactor dimension reduction approach. PLoS One. 2012;7:1–9.
37. Murk W, DeWan AT. Genome-wide search identifies a gene-gene interaction between 20p13 and 2q14 in asthma. BMC Genet. 2016;17:1–9.
38. Lee J-U, Kim JD, Park C-S. Gene-Environment Interactions in Asthma: Genetic and Epigenetic Effects. Yonsei Med J. 2015;56:877–86.
39. Berlivet S, Moussette S, Ouimet M, Verlaan DJ, Koka V, Al Tuwaijri A, et al. Interaction between genetic and epigenetic variation defines gene expression patterns at the asthma-associated locus 17q12-q21 in lymphoblastoid cell lines. Hum Genet. 2012;131:1161–71.
40. Reijmerink NE, Postma DS, Bruinenberg M, Nolte IM, Meyers DA, Bleecker ER, et al. Association of IL1RL1, IL18R1, and IL18RAP gene cluster polymorphisms with asthma and atopy. J Allergy Clin Immunol. 2008;122:651–4.e8.
41. Gudbjartsson DF, Bjornsdottir US, Halapi E, Helgadottir A, Sulem P, Jonsdottir GM, et al. Sequence variants affecting eosinophil numbers associate with asthma and myocardial infarction. Nat Genet. 2009;41:342–7. 42. Grotenboer NS, Ketelaar ME, Koppelman GH, Nawijn MC. Decoding asthma: translating genetic variation in IL33 and IL1RL1 into disease pathophysiology. J Allergy Clin Immunol. 2013;131:856–65.
43. Ferreira MAR, McRae AF, Medland SE, Nyholt DR, Gordon SD, Wright MJ, et al. Association between ORMDL3, IL1RL1 and a deletion on chromosome 17q21 with asthma risk in Australia. Eur J Hum Genet. 2011;19:458–64.
44. Torgerson DG, Ampleford EJ, Chiu GY, Gauderman WJ, Gignoux CR, Graves PE, et al. Meta-analysis of genome-wide association studies of asthma in ethnically diverse North American populations. Nat Genet. 2011;43:887–92.
45. Castano R, Bosse Y, Endam LM, Desrosiers M. Evidence of association of interleukin-1 receptor-like 1 gene polymorphisms with chronic rhinosinusitis. Am J Rhinol Allergy. 2009;23:377–84.
46. Shimizu M, Matsuda A, Yanagisawa K, Hirota T, Akahoshi M, Inomata N, et al. Functional SNPs in the distal promoter of the ST2 gene are associated with atopic dermatitis. Hum Mol Genet. 2005;14:2919–27. 47. Bonnelykke K, Matheson MC, Pers TH, Granell R,
Strachan DP, Alves AC, et al. Meta-analysis of genome-wide association studies identifies ten loci influencing allergic sensitization. Nat Genet. 2013;45:902–6. 48. Ferreira MAR, Matheson MC, Tang CS, Granell R, Ang
W, Hui J, et al. Genome-wide association analysis identifies 11 risk variants associated with the asthma with hay fever phenotype. J Allergy Clin Immunol. 2014;133:1564–71.
49. Gordon ED, Palandra J, Wesolowska-Andersen A, Ringel L, Rios CL, Lachowicz-Scroggins ME, et al. IL1RL1 asthma risk variants regulate airway type 2 inflammation. JCI Insight. 2016;1:e87871.
50. Sarnowski C, Sugier PE, Granell R, et al. Identification of a new locus at 16q12 associated with time to asthma onset. J Allergy Clin Immunol. 2016; 138: 1071–80. 51. Coyle AJ, Lloyd C, Tian J, Nguyen T, Erikkson C, Wang
L, et al. Crucial role of the interleukin 1 receptor family member T1/ST2 in T helper cell type 2-mediated lung mucosal immune responses. J Exp Med. 1999;190:895– 902.
52. Tominaga S, Kuroiwa K, Tago K, Iwahana H, Yanagisawa K, Komatsu N. Presence and expression of a novel variant form of ST2 gene product in human leukemic cell line UT-7/GM. Biochem Biophys Res Commun. 1999;264:14–8.
53. Tago K, Noda T, Hayakawa M, Iwahana H, Yanagisawa K, Yashiro T, et al. Tissue distribution and subcellular localization of a variant form of the human ST2 gene product, ST2V. Biochem Biophys Res Commun. 2001;285:1377–83.
54. Dijk FN, Xu C, Melén E, Carsin A-E, Kumar A, Nolte IM, et al. Genetic regulation of IL1RL1 methylation and IL1RL1-a protein levels in asthma. Eur Respir J. 2018;51:1701377.
55. Saluja R, Ketelaar ME, Hawro T, Church MK, Maurer M, Nawijn MC. The role of the IL-33/IL-1RL1 axis in mast cell and basophil activation in allergic disorders. Mol Immunol. 2015;63:80–5.
56. Traister RS, Uvalle CE, Hawkins GA, Meyers DA, Bleecker ER, Wenzel SE. Phenotypic and genotypic association of epithelial IL1RL1 to human TH2-like asthma. J Allergy Clin Immunol. 2015;135:92–9. 57. Lloyd CM. IL-33 family members and asthma - bridging
innate and adaptive immune responses. Curr Opin Immunol. 2010;22:800–6.
58. Drazen JM, Silverman EK, Lee TH. Heterogeneity of therapeutic responses in asthma. Br Med Bull. 2000;56:1054–70.
59. Kersten ETG, Koppelman GH. Pharmacogenetics of asthma : toward precision medicine. Curr Opin Pulm Med. 2017;23:12–20.
60. Chung KF. Asthma phenotyping: A necessity for improved therapeutic precision and new targeted therapies. J Intern Med. 2016;279:192–204. 61. Wadsworth SJ, Sandford AJ. Personalised medicine
and asthma diagnostics/management. Current Allergy and Asthma Reports. 2013;13:118–29.
62. Caudri D, Wijga AH, Smit HA, Koppelman GH, Kerkhof M, Hoekstra MO, et al. Asthma symptoms and medication in the PIAMA birth cohort: Evidence for under and overtreatment. Pediatr Allergy Immunol. 2011;22:652–9.
63. Halterman JS, Aligne A, Auinger P, McBride JT, Szilagyi PG. Inadequate therapy for asthma among children in the United States. Psychol Heal. 2000;105:25–50. 64. Caudri D, Wijga A, A. Schipper CM, Hoekstra M,
Postma DS, Koppelman GH, et al. Predicting the long-term prognosis of children with symptoms suggestive of asthma at preschool age. J Allergy Clin Immunol. 2009;124:903–910.e7.
65. Hafkamp-De Groen E, Lingsma HF, Caudri D, Levie D, Wijga A, Koppelman GH, et al. Predicting asthma in preschool children with asthma-like symptoms: Validating and updating the PIAMA risk score. J Allergy Clin Immunol. 2013;132:1303-10.
66. Castro-Rodriguez JA. Another Predictive Score for Childhood Asthma: The Search Remains. J Allergy Clin Immunol Pract. 2014;2:716–8.
67. Bisgaard H, Jensen SM, Bonnelykke K. Interaction between asthma and lung function growth in early life. Am J Respir Crit Care Med. 2012;185:1183–9.
68. Wise RA, Szefler SJ, Sharma S, Kho AT, Cho MH, Chonka DCC. Patterns of Growth and Decline in Lung Function in Persistent Childhood Asthma. 2016;19:1842-1852
69. Spycher BD, Henderson J, Granell R, Evans DM, Smith GD, Timpson NJ, et al. Genome-wide prediction of childhood asthma and related phenotypes in a longitudinal birth cohort. J Allergy Clin Immunol. 2012;130:503–9.e7.
70. Belsky DW, Sears MR, Hancox RJ, Harrington H, Houts R, Moffitt TE, et al. Polygenic risk and the development and course of asthma: An analysis of data from a four-decade longitudinal study. Lancet Respir Med. 2013;1:453–61.
71. Wijga AH, Kerkhof M, Gehring U, de Jongste JC, Postma DS, Aalberse RC, et al. Cohort profile: the prevention and incidence of asthma and mite allergy (PIAMA) birth cohort. Int J Epidemiol. 2014;43:527–35. 72. Antens CJM, Oldenwening M, Wolse A, Gehring U,
Smit HA, Aalberse RC, et al. Repeated measurements of mite and pet allergen levels in house dust over a time period of 8 years. Clin Exp Allergy. 2006;36:1525–31.
73. Berentzen NE, Stokkom VL Van, Gehring U, Koppelman GH, Schaap LA, Smit HA, et al. Associations of sugar-containing beverages with asthma prevalence in 11-year-old children : the PIAMA birth cohort. Eur J Clin Nutr. 2015;69:303-8.
74. Gehring U, Wijga AH, Hoek G, Bellander T, Berdel D, Brüske I, et al. Exposure to air pollution and development of asthma and rhinoconjunctivitis throughout childhood and adolescence: A population-based birth cohort study. Lancet Respir Med. 2015;3:933–42.
75. Bousquet J, Anto J, Auffray C, Akdis M, Cambon-Thomsen A, Keil T, et al. MeDALL (Mechanisms of the Development of ALLergy): an integrated approach from phenotypes to systems medicine. Allergy. 2011;66:596–604.
76. Farzan N, Vijverberg SJ, Andiappan AK, Arianto L, Berce V, Blanca-López N, et al. Rationale and design of the multiethnic Pharmacogenomics in Childhood Asthma consortium. Pharmacogenomics. 2017;18:931–43.
F. Nicole Dijk, Johan C. de Jongste, Dirkje S. Postma, and Gerard H. Koppelman
Genetics of
onset of asthma
_
Chapter 2
Abstract
Purpose of review
Most asthma starts early in life. Defining phenotypes of asthma at this age is difficult as many preschool children have asthma-like respiratory symptoms. This review discusses progress in defining early wheez-ing phenotypes and describes genetic factors associated with the age of onset of asthma.
Recent findings
Latent class analyses confirmed transient and persistent wheezing phenotypes, and identified a novel in-termediate-onset wheezing phenotype that was strongly associated with atopy and asthma at age 8 years. However, no single cross-sectional or longitudinal definition of respiratory symptoms in childhood strongly predicts asthma later in life. Genome-wide association (GWA) studies have identified a locus on chromo-some 17q12–21 (encoding ORMDL3 and GSDMB) as a risk factor for predominantly childhood-onset asth-ma, but not for atopy, and overall not for adult-onset asthma. Other loci found by GWA studies appear to increase asthma risk both in children and adults. Atopy genes do not explain early-onset asthma.
Summary
Although most asthma starts early in life, no valid test is able to identify asthma at that age period. GWA studies have provided more insight into the unique and common genetic origins of adult-onset and childhood-onset asthma. The 17q12–21 locus is predominantly associated with childhood-onset asthma.
Abbreviations
ALSPAC - Avon Longitudinal Study of Parents and Children AHR - Airway hyperresponsiveness
AUC - Area under the curve
ETS - Environmental tobacco smoke ERS - European Respiratory Society GWA - Genome-wide association LD - Linkage disequilibrium
LLCA - Longitudinal latent class analysis
Introduction
Asthma is a common chronic lower respiratory disease that results from the interactions between genes and environmental factors. It affects more than 300 million people worldwide and its prevalence is still increasing globally.1 Asthma is characterized by chronic airway inflammation associated with variable
airflow obstruction and airway hyperresponsiveness (AHR) leading to recurrent episodes of wheeze, cough, and shortness of breath.2 Different asthma phenotypes can be distinguished based on the
pres-ence, timing, and severity of symptoms (e.g., age of onset, nocturnal symptoms), atopy, responsiveness to triggers, and type of airway inflammation eosinophilic or neutrophilic). Recent findings indicate that ge-netic factors of childhood-onset asthma differ from those of adult-onset asthma. Therefore, this review will focus on recent progress made in defining phenotypes of early-onset asthma, and describe genetic factors associated with the age of onset of asthma.
Defining early-onset asthma
It is problematic to define when asthma starts, because there is no conclusive diagnostic test available at preschool age. Respiratory symptoms suggestive of asthma, such as wheeze, cough, and shortness of breath are highly prevalent in early childhood, yet not specific for asthma, and commonly occur during viral respiratory infections. In clinical practice, a history of recurrent wheezing episodes is frequently con-sidered as the onset of asthma in children, often defined retrospectively and, hence, inaccurately. Longitudinal studies revealed that wheezing in the first 6 years of life comprise distinct phenotypes, with dissimilar causes and outcomes. A 1995 seminal paper from the Tucson Children’s Respiratory Study introduced a longitudinal wheezing classification.3 The ‘early transient wheezing’ phenotype included
children with onset of wheezing during the first 3 years of life with no symptoms at age 6, the ‘persistent wheezing’ phenotype included children with wheezing during the first 3 years of life that continued until age 6, and the ‘late-onset wheezing’ phenotype included children with development of wheezing be-tween 3 and 6 years of age.
In 2004, these wheezing phenotypes were further classified incorporating markers of atopy. ‘Transient wheezing’ was not associated with allergic sensitization and was associated with a slightly lower lung function, which was already present before any lower respiratory tract illness had occurred, probably caused by relatively narrow airways.3 ‘Nonatopic wheezing’ was associated with viral respiratory
infec-tions, without allergic sensitization.4 These children had normal lung function in infancy but slightly
lower lung function in later childhood, combined with increased AHR. ‘Atopic wheezing’ was recognized as the classical asthma phenotype, with normal lung function early in life but impairment later in life. Earlier onset of symptoms was associated with more severe asthma.3-5
Recently, two longitudinal birth cohort studies in the UK and the Netherlands analyzed annual reports of wheezing until age 8 and identified similar wheezing phenotype patterns using the unbiased, lon-gitudinal latent class analyses (LLCAs). A novel wheezing phenotype ‘intermediate-onset wheeze’ was identified, characterized by a low prevalence of wheeze until the age of 1.5 years with a rapidly increas-ing prevalence of symptoms thereafter.6 Persistent, late-onset, and intermediate-onset wheezing
phe-notypes were strongly associated with doctor-diagnosed asthma at age 8 years compared with never, transient early, and prolonged early wheezing phenotypes.6,7 Risk factors for persistence of wheeze were
As longitudinal phenotypes can only be categorized retrospectively, an European Respiratory Society task force proposed a cross-sectional classification of wheezing illness based on triggers: episodic viral wheeze (EVW) and multitrigger wheeze (MTW).8 It is often thought that MTW may precede asthma, yet
literature does not fully support this. Inconsistent clinical and pathological differences between these phenotypes have been reported9-11 and significant phenotypic switching between these wheezing
phe-notypes was observed.12 Therefore, this classification is of limited value when studying onset of asthma.13
When does asthma start?
In most patients, asthma has its origins in early childhood.14 A recent Danish study of high-risk children
showed that children with asthma at the age of 7 years already had increased AHR and lower lung func-tion as neonates. These lower lung funcfunc-tion levels progressively declined into childhood.15 Moreover,
characteristic pathological features of asthma in airway wall biopsies, such as eosinophilic inflammation and basement membrane thickening were observed in children with MTW at age 3, preceding symptoms of wheeze in toddlers.16,17 Finally, asthma at age 22 years was predicted by late-onset or persistent
wheez-ing at age 6, a decreased lung function, and AHR.18 There is still much uncertainty about the
relation-ship, similarities, and differences between childhood and adult asthma. Comparing the characteristics of childhood and adult asthma reveals similarities between risk factors, such as environmental tobacco smoke (ETS) exposure19,20, AHR, a family history of asthma6,7,18,21, the role of viral infections22, allergic
rhi-nitis23, atopic status, and eczema3,6,7,24,25,, but also differences (Table 1).26-30 Table 1. Clinical differences between childhood-onset and adulthood-onset asthma.
Susceptibility to childhood-asthma: answers from
genetic studies
Genetic factors were estimated to explain 34% of the variation in age of onset of asthma.31 Several
candidate gene studies have provided evidence for an age-specific genetic effect.32 Recently, novel
ge-nome-wide association (GWA) studies have allowed to disentangle childhood-onset and adult-onset asthma susceptibility in a hypothesis-free manner (Table 2 lists genetic terminology). Thus far, 14 GWA studies have been published revealing 27 loci with genome-wide significant risk variants for the develop-ment of asthma. Half of these studies were performed in childhood-onset asthma (Table 3).33-46
Table 2. Glossary table: genome-wide association study.
The role of the 17q12–21 locus in childhood onset
Asthma In the first GWA study on asthma, the 17q12–21 IKZF3-ZPBP2-GSDMB-ORMDL3 region was identi-fied as an asthma susceptibility locus in childhood-onset asthma in a white population. This observation was extensively replicated34,40-42,44,47-52 and thus far only not confirmed in African–Americans.53 A remarkable
finding is that the 17q12–21 SNPs have a particularly strong association with (early) childhood-onset asth-ma. This was first shown in a French population wherein 17q12–21 risk alleles were strongly associated with asthma when retrospectively defined as starting before the age of 12, but not with asthma that started at a later age. Subsequently, a prospective Danish study confirmed that this ORMDL3/GSDMB locus conferred an increased risk of recurrent wheeze and asthma, but only in case of early-onset disease, that is, before the age of 3 years, but not in later-onset asthma. Furthermore, an association was observed with severe exac-erbations until age 6 years and AHR until age 4, but not with atopy.48 Other studies also indicated that the
17q12–21 locus is not an atopy-susceptibility locus.40,47,49,54,55 In addition, a recent study performed in the Avon
Longitudinal Study of Parents and Children (ALSPAC) cohort revealed a significant association between 17q12–21 SNPs and the persistent-onset and intermediate-onset wheezing phenotypes.55
A large-scale GWA meta-analysis of the GABRIEL (A Multidisciplinary Study to Identify the Genetic and En-vironmental Causes of Asthma in the European Community) consortium of European ancestry individuals classified asthmatic patients in two age-of-onset groups: before or after 16 years of age. Significant associ-ations of ORMDL3/GSDMB SNPs were restricted to childhood-onset asthma. The SNP with the strongest association had an odds ratio (OR) of 0.64 in the childhood-onset group, compared with an OR of 1.03 in individuals with a later onset (mean difference P value: 1.34x10-9).40 Two other studies reported a
signifi-cant association of the 17q12–21 locus with childhood-onset asthma, but not adult-onset asthma54 and
with severe childhood asthma.56 Furthermore, the effect of 17q12–21 SNPs on early-onset asthma seems to
stronger association between respiratory infections and asthma onset less than 4 years of age if they had early ETS exposure. Moreover, the latter effect was also present in childhood asthma that remits in adult-hood.58 In contrast, a complete restriction of the 17q12–21 variants to early-onset asthma was refuted in a study
of predominantly adult-onset severe asthma (mean age of onset 21 years). Yet a genome-wide significant as-sociation with this locus was reported.44 Furthermore, a study59 performed in US asthma patients reported a
similar OR for 17q12–21 risk variants in their early-onset (<4 years) and late onset (≥4 years) asthma group. The functionality of the genes at the 17q12–21 locus is not well understood. Asthma risk alleles at this lo-cus are strongly correlated over multiple genes and, therefore, it is difficult to discover the causal asthma variant(s). The asthma-associated 17q12–21 alleles were correlated with the transcript levels of predom-inantly two genes, ORM1-like 3 (Saccharomyces cerevisiae) (ORMDL3) and GSDMB, which indicates that both genes are coregulated.33,54,55,60,61 A functional SNP in this region affects the binding of the nuclear
CCCTC-binding factor and, thus, alters nucleosome occupancy leading to altered regulation of transcrip-tion.60 ORMDL3 encodes for a transmembrane protein anchored in the endoplasmic reticulum. Its
func-tion might be related to endoplasmatic reticulum-mediated calcium signaling with modulafunc-tion of an unfolded protein response.62,63 ORMDL3 may modify important pathways in the process of T-lymphocyte
activation.64 A recent study65 in mice demonstrated that ORMDL3 is an airway epithelial gene that is
acti-vated by Th2 cytokines and allergens. Furthermore, its function has been linked to sphingolipid homeo-stasis.66 GSDMB may play a role in apoptosis and tumorgenesis and is expressed in epithelial cells.67
In conclusion, although different definitions of early-onset and late-onset asthma have been used, it seems that the asthma-associated 17q12–21 polymorphisms are related to a nonatopic, more severe ear-ly-onset asthma phenotype, and associations with adult-onset asthma are less prominent.
Table 3. Published GWA studies and meta-analysis of GWA studies for childhood and adult asthma which obtained genetic associations at a genome-wide significant level, either in the original cohort, the replication cohort or in a combined analysis.
*Reported as mean (±SD) or range, mn=month, y=year, † Bold genes were genome wide significant (P≤5x10-8) in the original cohort, underlined genes were genome wide significant in the replication/combined analyses, normal printed genes showed nominal significance in original, replication or combined analyses, ‡ No association in African Ameri-can population, § Associated with total IgE levels, not with asthma, ∫ Only associated in AfriAmeri-can AmeriAmeri-cans.
#Studies: MAGICS: German Multicenter Asthma Genetics in Childhood Study, ISAAC Phase II: International Study of Asthma and Allergies in Childhood cohort, phase II, CAMP: Childhood Asthma Management Program, GACRS: Genet-ics of Asthma in Costa Rica Study, BAMSE: Swedish birth cohort study Barn Allergi Milieu Stockholm Epidemiologi, PACT: Pediatric Asthma Controller Trial.
Other asthma genes found by genome-wide
association studies: age-studies
Eleven asthma genes were identified in seven GWA studies performed in populations with childhood-on-set asthma. The GABRIEL GWA study found, next to the childhood-onchildhood-on-set asthma ORMDL3/GSDMB lo-cus, also more prominent effects in childhood-onset asthma for the IL1RL1/IL18R1 and IL33 genes.40 In
contrast, HLA-DQ variants were slightly more strongly associated among the later-onset individuals, but none of these differences was significant. In childhood-onset asthma, an association with cAMP-specific 3’,5’-cyclic phosphodiesterase 4D (PDE4D)34 and DENN/MADD Domain containing 1B (DENND1B)35 gene
poly-morphisms was identified in US white populations. PDE4D was reported to be important in signaling pathways and airway smooth muscle contraction68, whereas DENND1B encodes for a protein that may
interact with tumor necrosis factor-a. It is expressed by natural killer and dendritic cells, which have a critical role in the inflammatory pathogenesis of asthma.35,63 A subsequent analysis of DENND1B
vari-ants with age of onset of asthma was performed in children of European and African ancestry, all with asthma diagnosed before age 6. This yielded contradictory results, with opposite alleles being associated in children with asthma symptoms at older age, compared with younger age, and also different alleles associated in populations of African ancestry compared with European ancestry. The latter might be due to differences in underlying genetic architecture between populations.35
In Japanese childhood-onset (<15 years) asthma patients, the genes Major histocompatibility complex, class II, DP (HLA-DP) and Solute Carrier Family 30 member 8 (SLC30A8) emerged as genome-wide significant vari-ants.36 The HLA-DP gene product is a cell surface receptor for foreign or self-antigens, with a function
in immune-related diseases. The gene seems to have a protective role in Th1-derived diseases such as diabetes type 1 and rheumatoid arthritis, which suggests that it regulates Th1/Th2 balance.69,70 SLC30A8
is associated with type 2 diabetes, and its link to asthma may lie in its localization in epithelial cells and epithelial integrity.71
In the Isle of Wight study, SNPs and haplotypes at chromosome 1p33-p32.31 were associated with asthma
and asthma severity in children with asthma at age 10 using a pooled GWA study approach.37 This region
contained the ATP Synthase Mitochondrial F1 Complex Assembly Factor 1 (ATPAF1), C1ORF223, and KIAA0494 genes. In replication cohorts, wherein asthma was diagnosed during childhood, significant trends for as-sociation were identified in this linkage disequilibrium block in several ethnic groups, except in Hispanic populations. In an Australian study43 of adults and children, SNPs at the IL-6 receptor gene IL6R1 and on
chromosome 11q13.5 were identified as asthma susceptibly variants in individuals of European ancestry of whom 53.9% had childhood-onset asthma, defined as onset 16 years of age or less (26.1% >16 years, 20% unknown). No significant difference in strength of association of these risk alleles with age of onset was found.
Other asthma GWA studies have identified important asthma susceptibility genes, yet included both adults and children with often an unknown age of asthma onset (Table 3), thus not allowing conclusions as to differential effects according to age of onset of asthma.
In conclusion, it is difficult to put forward a solid opinion about distinct genetic origins of childhood-on-set and adult-onchildhood-on-set asthma. Most studies made no formal comparison between the two age groups. Fur-thermore, age of onset is subject to recall bias because it was often defined in a retrospective manner.72
Until now, only the childhood-onset associated ORMDL3/GSDMB region has been replicated when age of onset was taken into account and this has not been observed so strongly for other GWA loci.
Genome-wide association studies of age of onset asthma
and wheezing phenotypes
Until now, only one GWA study reported on the association between asthma and age of onset in childhood as an outcome measurement. This study included 573 US non-Hispanic white children with a mean±SD age of asthma onset of 3.1±2.5 years38 and replicated in three independent cohorts with European,
Amer-ican, and Latin American children (age of onset 2.5±2.3, 3.1±2.4, and 3.3±2.7 years, respectively). Two SNPs on chromosome 3 and 11 were associated with earlier onset of asthma using a survival analysis. The com-bined effect of the associated SNPs led to a mean age of onset of 3.4 in children having 0 risk alleles, compared with a mean age of onset of 2.5 years (P<0.0001) in children having at least one risk allele. Both SNPs were not located in any known gene, but one of them was close to the IL-5 receptor a gene (IL5RA). This receptor is selectively expressed in the bronchial muscle and shows an eosinophil-independent role in AHR.38 Recently, a GWA study39 performed in the ALSPAC cohort investigated whether the combined
signal of asthma risk SNPs was predictive of childhood-onset asthma (<8 years). Asthma-associated SNPs from the GABRIEL GWA study (excluding ALSPAC) were used to construct a genetic prediction score. Children were classified into six LLCA-derived wheezing phenotypes6: persistent, late-onset,
inter-mediate-onset, prolonged early, transient early, and never/infrequent wheezing. Despite weak discrimi-nation (area under the receiver operating characteristic curve scores≤0.60), an association was examined
between the top 46 ranked SNPs (including mostly 17q12–21 SNPs) and two symptom-based phenotypes, early-onset persistent wheeze and intermediate-onset wheeze. Interestingly, doctor-diagnosed asthma was associated with the lower ranked SNPs, which could indicate that early-onset persistent wheeze and intermediate-onset wheeze are better predicted by the 17q12–21 locus than doctor-diagnosed asthma. However, the clinical relevance of prediction scores in the determination of disease risk is poor.39,40,43
Genome-wide association studies of asthma and atopy:
a comparison
Atopy is a strong risk factor for childhood asthma. Therefore, it is of interest to compare atopy genesthat regulate total or specific IgE production, and blood eosinophils to asthma susceptibility SNPs (Table 4). 73-80,81,82 Remarkably, published data suggest that early childhood asthma is not explained by atopic genetic
susceptibility, as atopy genes thus far discovered through GWA studies are not detected as asthma sus-ceptibility genes. This observation is paralleled by studies that showed no association between allergen exposure and asthma83,84, and reports that avoidance of allergens did not lead to a reduction in asthma
development.85 The causal model in which allergic sensitization directly leads to an increased risk of
asth-ma should, therefore, be re-considered.86
Conclusion
Although most asthma starts early in life, there is no valid test to diagnose asthma in the first years of life. GWA studies have provided more insight into the specific and common genetic origins of adult-onset and childhood-onset asthma. The 17q12–21 asthma locus is important for childhood-onset, nonatopic, more severe asthma. Other asthma risk loci do not have a strong relation with age of onset. Finally, atopy risk alleles do not explain genetic susceptibility to asthma. Future genetic studies are needed to systematical-ly compare age of onset of asthma in large groups of patients including interactions with environmental factors, especially in critical periods, and additionally implementing the role of epigenetic mechanism. Studies that assess the effects of asthma risk SNPs on gene expression (expression quantitative trait loci) can guide further functional studies that will provide more insight into asthma pathogenesis.87 Finally, in
this way, early detection of asthma using genetic markers interacting with personal and environmental factors might become feasible.
Key points
• Novel longitudinal symptom based wheezing phenotypes have been discovered by the use of latent class analyses.
• GWA studies have revealed that 17q12-21 is predominantly a childhood-onset asthma locus, with modified effects through ETS exposure.
• Two loci have been associated with age of onset in childhood asthma, one locus being close to the IL5RA gene.
• Genetic susceptibility to asthma cannot be explained by atopic risk alleles.
• Genetic profiles of childhood and adult asthma, with the combination of environmental risk factors and associated biomarkers may improve the early detection of asthma.
