Genetic Variation in CHI3L1 Contributes to Asthma Severity and Airway Expression of YKL-40

Abstract

Background

Single Nucleotide Polymorphisms (SNPs) in the CHI3L1 promoter, the gene encoding YKL-40, are associated with circulating YKL-40 levels and asthma prevalence. However, the effects of gene polymorphisms on asthma severity and airway expression of YKL-40 have not been examined.

Objective

Determine the effect of genetic variation in CHI3L1 on asthma severity and YKL-40 expression in individuals from the Yale Center for Asthma and Airways Disease (YCAAD) and the Severe Asthma Research Program (SARP).

Methods

SNPs spanning the CHI3L1 gene were genotyped in 261 YCAAD and 889 SARP individuals, respectively. Association and haplotype analyses were conducted to identify effects on airflow obstruction, YKL-40 levels and asthma severity.

Results

Fifteen SNPs in CHI3L1 were associated with FEV1 and/or serum YKL-40 levels. Rs12141494 (intron 6) was the only SNP in individuals of European ancestry in both cohorts that was associated with serum YKL-40 levels and post-bronchodilator FEV1. Conditional analysis demonstrated that the effect on lung function was independent of the promoter SNP rs4950928 and haplotype analysis demonstrated that G alleles at rs12141494 and rs4950928 are associated with lower YKL-40 levels and higher FEV1 % predicted. In individuals with asthma, the risk allele A at rs12141494 was associated with severe asthma and higher levels of YKL-40 in the airway (P ≤0.05).

Conclusion

In contrast to the promoter SNP, rs4950928, the intronic SNP, rs12141494 in CHI3L1 is associated with asthma severity, lung function and YKL-40 levels in the blood and airway. These data suggest that SNP rs12141494 modulates expression of YKL-40 in the airway and contributes to airway remodeling and asthma severity.

Introduction

YKL-40, a chitinase-like protein, belongs to the chitinase and chitinase-like family of proteins. These evolutionarily conserved molecules interact with chitin, the second most abundant polysaccharide on earth. Human and mechanistic studies have demonstrated that CHI3L1/YKL-40 plays a role in the pathobiology of asthma by modulating innate and adaptive immune and remodeling responses (1). In the first human asthma studies, we demonstrated that elevated serum levels of YKL-40 correlate with asthma severity, airway remodeling, and increased thickness of the sub-epithelial basement membrane in individuals with asthma (2). This work led to mechanistic studies demonstrating that bronchial epithelial cells exposed to YKL-40 generate higher levels of IL-8 and stimulate smooth muscle proliferation in vitro (3). An additional mechanism of YKL-40 increased bronchial smooth muscle proliferation involves the protease activated receptor-2 (4). Taken together, these studies established an important role for YKL-40 as a molecule uniquely juxtaposed between environmental exposure, inflammation, and the development of airway remodeling and severe asthma.

Genetic studies have also demonstrated that variation in CHI3L1 contributes to the pathogenesis of asthma. A Genome-wide Association Study (GWAS) of serum YKL-40 levels in a founder population of European descent by Ober et al. (5), identified an association between the rs4950928 promoter Single Nucleotide Polymorphism (SNP) and circulating YKL-40 levels. This SNP was also associated with the risk of asthma and FEV1 in three populations of European ancestry with mild asthma. Subsequent reports have shown an association with rs4950928 and asthma (6), although the risk allele was opposite of that reported by Ober (5). Additional publications have been conflicting with respect to the effect of genetic variation in the CHI3L1 gene and asthma (7, 8). To date, the effect of genetic variation in CHI3L1 on asthma severity has not been examined.

To determine the effects of genetic variation in CHI3L1 gene on YKL-40 expression in the airway, airway remodeling and asthma severity we examined two cohorts of individuals with asthma from the Yale Center for Asthma and Airways Disease (YCAAD) and the Severe Asthma Research Program (SARP). We hypothesized that genetic variation in CHI3L1 is associated with persistent airflow obstruction, serum YKL-40 levels, asthma severity, and airway expression of YKL-40. To examine this hypothesis, we characterized the effect of SNPs in the CHI3L1 gene on severe asthma traits, determined the interaction between SNPs by haplotype analysis, and correlated identified SNP with airway expression of YKL-40. Ultimately, we identified a novel polymorphism in CHI3L1 that is likely to contribute to airway remodeling and asthma severity through increased production of YKL-40 in the airway.

Methods

Populations

Yale Center for Asthma and Airways Disease

Study participants in the YCAAD cohort based in New Haven, Connecticut, underwent an extensive phenotypic characterization after Institutional Review Board (IRB) approval. Inclusion, exclusion criteria, and study protocol have been described previously (2). Severe asthma was defined as outlined by the SARP clustering algorithm and was modified as follows: Severe asthma included subjects with a baseline FEV1 of less than 68% of predicted (SARP clusters 4 and 5, while non-severe asthma was defined by the presence of a baseline FEV1 equal or greater than 68% predicted (SARP clusters, 1, 2 and 3) (9). For specific details related to definitions and specific study measurements, see online supplement. A total of 259 individuals from this cohort were analyzed.

The Severe Asthma Research Program

Individuals in the SARP cohort completed study visits using established standard operating procedures as previously described (9). The definition for severe asthma outlined above was also used in this cohort. IRB approval at the SARP institutions was obtained for these studies. The characteristics of these subjects have been reported in previous publications (9, 10). A total of 919 subjects from this cohort were analyzed.

Sputum Induction

Individuals in the YCAAD cohort underwent sputum induction with inhaled hypertonic saline. Mucus plugs were removed using a dissecting microscope and washed to remove squamous cell contamination. The cellular and aqueous compartments were separated, cell counts, cell differentials, and viability were determined (Diff-Quik and trypan blue exclusion). Aliquots of supernatants were stored until processing. Samples with >20% squamous cells were considered contaminated and were not processed further.

Measurement of Serum and Sputum YKL-40 Levels

YKL-40 levels were measured in duplicate in serum and sputum specimens using a commercially available enzyme-linked immunosorbent assay (ELISA) kit for YKL-40 (Quidel). The mean value of the two duplicates was used in the statistical analyses. Duplicate samples with coefficients of variation (CV) greater than 20% were re-assayed. The mean CV for all samples was 10%. The limit of detection is 5.4 ng/mL. Values under the limit of detection were defaulted to 0 ng/mL.

SNP Genotyping

DNA was extracted from peripheral blood leukocytes using standard protocols. The Sequenom MassARRAY ® system (San Diego, California) genotyping platform was used. Genotyping was performed according to the manufacturer's iPLEX application guide at the Center for Human Genomics at Wake Forest University Health Sciences and the Yale Keck DNA sequencing facility for SARP and YCAAD samples, respectively (11). To identify rare and novel SNPs that could be associated with expression of YKL-40, resequencing of the CHI3L1 gene was conducted on a subset of 6 individuals of European Ancestry in the YCAAD cohort 3 with the CC genotype and 3 with GG genotype at the promoter SNP rs4950928, with high and low YKL-40 levels, respectively; amplicon distribution: 13 kilobases (kb) upstream-5′, 7.8 kb CHI3L1 gene, 1.2 kb downstream-3′) as previously described (12). The identified SNPs were compared to SNPs genotyped in previous publications (5, 6). These SNPs were analyzed with Haploview to identify tagging polymorphisms in the CHI3L1 gene using data from the HapMap project (version 3 release 27) (13). A total of 17 SNPs spanning the CHI3L1 gene, identified through sequencing, previous publications (5, 6) and haplotype tagging were genotyped in YCAAD (discovery). Simultaneously, 14 SNPs identified through haplotype tagging using data from the HapMap project (version 3 release 27) (13) were genotyped in SARP (validation). A total of 9 SNPs present in both datasets tagged all other SNPs reported in this study. Genotyping criteria for SNP quality included the removal of SNPs with low call rates (<80%), SNPs that violated Hardy-Weinberg (P-value <0.05), Minor Allele Frequency (MAF) <4%, samples with low call rates (<98%). Pair-wise marker linkage disequilibrium (LD) was estimated using Lewontin's D' statistic and r² using Haploview (13).

Statistical Analysis

R software (R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna, Austria) was used for data analysis. R stats package version 3.0.1, the SNPassoc version 1.8-5, haplo.stats version 1.6.3 and car version 2.0-21 packages were implemented. Continuous variables were tested using non-parametric tests including Kruskal-Wallis test. Categorical variables were analyzed with chi-square tests. Results are reported as medians and Interquartile Ranges [IQRs] unless otherwise specified. Hardy-Weinberg equilibrium calculation was performed to assess the distribution of the SNPs in each group using Haploview (13). The SNPassoc package applies a generalized linear model to generate the association P-values. Using this statistical package dominant, codominant, recessive and additive models of association were tested for post-bronchodilator FEV1 percent predicted, used as a surrogate marker of airway remodeling, and serum YKL-40 levels on the discovery phase of the study. The additive association model was identified as the best-fit based on the Akaike information criterion (AIC) (14), all reported association values were generated using this model and were adjusted for age and gender, based on the effect these two features have on YKL-40 levels (15), pulmonary function (16), and on the best-fit based on the AIC criterion (14). In the discovery cohort (YCAAD) P values were adjusted for multiple testing with the Benjamini-Hochberg procedure (17)(False Discovery Rate (FDR) < 0.1) to balance power and the potential for false positives. While association values in the validation cohort (SARP) were adjusted for multiple testing with the Bonferroni method (P <0.05) to provide a more stringent and reliable measure of true association. We conducted a conditional analysis on individuals of European ancestry with the rs4950928 CC genotype to evaluate for independent associations on non-promoter SNPs, all P values on the conditional analyses were adjusted with the Bonferroni method. Haplotypes were inferred using the expectation maximization algorithm in Haploview (13). Haplotypes with a frequency greater than 5% were considered in the analysis. A generalized linear model using the Expectation Maximization (EM) algorithm described by Excoffier and Slatkin (18) was implemented in the haplotype analysis with the haplo.stats R package. Individuals with asthma and controls were considered in the discovery, validation, haplotype and conditional analyses. The Levene's test for homogeneity of variance on the two cohorts of individuals with asthma was implemented with the car package. The analysis of asthma severity considered asthma subjects only.

Results

Population characteristics and correlation of circulating YKL-40 levels with asthma severity and airway remodeling

To confirm that serum YKL-40 levels are associated with asthma severity and airway remodeling, two populations of individuals of European ancestry (EA) were evaluated separately: discovery was conducted in YCAAD (n=213; 174 individuals with asthma v. 39 healthy controls) and validation was completed in SARP (n=684; 562 individuals with asthma v. 122 healthy controls). Since the two cohorts are predominantly of EA, individuals of African Ancestry (AA) from both cohorts were pooled together (n=281; 259 individuals with asthma v. 22 healthy controls). Baseline characteristics of all three groups are presented in Table 1. Additional clinical characteristics of the YCAAD cohort are presented in Table E1 (online supplement). Individuals of EA with asthma in YCAAD and SARP had lower FEV1 % predicted compared to healthy controls. These individuals also had higher fractional exhaled nitric oxide (FeNO) and higher serum IgE levels. In individuals of AA, no differences were seen in age, Body Mass Index (BMI), or FeNO between subjects with asthma compared to healthy controls. YKL-40 levels were evaluated in 213 EA YCAAD, 143 EA SARP, and 92 AA and were significantly higher in individuals of EA with severe asthma compared to non-severe, and control subjects (Table 1). YKL-40 was not associated with severe asthma in AA, despite similar median values compared to EA in SARP. This is the result of a lack of statistical power due to the small number of AA individuals with available YKL-40 measurements. Additional characteristics of AA individuals with asthma are presented on Table E2 (online supplement). Post-bronchodilator FEV1 % predicted had a significant correlation with serum YKL-40 levels in individuals of EA in both YCAAD (Spearman rho=-0.19; P <0.01) and SARP (Spearman rho=-0.24; P < 0.01). Concordant with higher serum YKL-40 levels in individuals with severe asthma, the post-bronchodilator FEV1 % predicted was significantly lower in both YCAAD and SARP. These findings indicate that the two EA cohorts have similar characteristics and confirm that serum YKL-40 levels are associated with irreversible airflow obstruction and severe asthma in individuals of European ancestry.

Table 1.

Demographic characteristics.

European Ancestry YCAAD	Controls n=39	Non-Severe n=137	Severe n=37	P value
Age (years)	45 [31-57]	46 [36-56]	59 [50-68]	<0.01
Female sex	27(69%)	98 (72%)	23 (62%)	0.52
Age of Onset	-	22 [5-40]	21 [4-45]	0.46
BMI (Kg/m2)	25.5 [23.5-27.8]	26.3 [22.7-31.2]	29.3 [25.0-31.5]	0.08
FEV1 Pre-BD PP	99 [91-105]	89 [79-98]	53 [39-59]	<0.01
FEV1 Post-BD PP	104 [96-110]	95 [84-106]	60 [45-70]	<0.01
FeNO (ppb)	22 [13-29]	26 [19-60]	40 [24-54]	0.02
Total IgE (IU/mL)	20 [11-72]	95 [29-225]	129 [26-289]	0.61
Serum YKL-40 (ng/mL)*	43 [31-67]	65 [37-105]	83 [37-110]	0.01
European Ancestry SARP	Controls n=122	Non-Severe n=234	Severe n=328	P value
Age (years)	27 [23-40]	31 [22-41]	41 [31-51]	<0.01
Female sex	68 (66%)	168 (72%)	190 (58%)	0.29
Age of Onset	-	10 [4-22]	12 [3-30]	0.06
BMI (Kg/m2)	24.4 [21.7-28.4]	25.1 [22.1-29.3]	28.7 [24.4-34.0]	<0.01
FEV1 Pre-BD PP	98 [93-105]	89 [82-99]	60 [43-75]	<0.01
FEV1 Post-BD PP	105 [97-110]	97 [89-107]	71 [36-86]	<0.01
FeNO (ppb)	14 [11-20]	26 [17-52]	29 [17-52]	0.03
Total IgE (IU/mL)	35 [15-83]	117 [49-232]	115 [29-305]	0.05
Serum YKL-40 (ng/mL)*	43 [33-53]	44 [32-60]	60 [41-101]	<0.01
African Ancestry SARP and YCAAD	Controls n=22	Non-Severe n=132	Severe n=127	P value
Age (years)	35 [29-40]	29 [22-40]	36 [22-45]	0.79
Female sex	16 (73%)	88 (66%)	82 (65%)	0.48
Age of Onset	-	8 [2-18]	4 [1-17]	0.66
BMI (Kg/m2)	28.7 [25.8-34.7]	29.0 [24.7 -36.6]	32.3 [23.5-37.5]	0.50
FEV1 Pre-BD PP	103 [96-110]	87 [80-96]	62 [52-77]	<0.01
FEV1 Post-BD PP	108 [101-117]	98 [89-107]	80 [66-93]	<0.01
FeNO (ppb)	18 [16-28]	28 [15-55]	30 [18-44]	0.63
Total IgE (IU/mL)	66 [39-162]	150 [47-376]	201 [69-569]	0.02
Serum YKL-40 (ng/mL)*	41 [29-47]	39 [23-65]	60 [32-79]	0.33

Values expressed as n and (percentage) or median and [interquartile range].

^*Number (Percentage) with available measurements: European Ancestry YCAAD=213 (100%); European Ancestry SARP=143 (22%); African Ancestry SARP and YCAAD=92 (33%)

Definition of Abbreviations: YCAAD = Yale Center for Asthma and Airways Disease; SARP = Severe Asthma Research Program; BMI = Body Mass Index; FEV1 = Forced Expiratory Volume in One Second; Pre-BD = Pre-bronchodilator; PP= Percent of predicted; Post-BD=Post-bronchodilator; FeNO = Fractional exhaled nitric oxide; ppb = parts per billion; IgE = Immunoglobulin E.

CHI3L1 SNPs and Linkage Disequilibrium Structure

To visualize the LD coefficients in the EA and AA individuals, we generated haplotype maps using Haploview (13). Nine common SNPs in YCAAD and SARP were used to generate a haplotype map for European ancestry, Figure 1 and for African ancestry, Supplemental figure E1A. This analysis revealed two haplotypic blocks, one in the promoter region that includes SNPs in the 5′ region of the gene and the promoter SNP rs4950928, and one including the Intronic/Exonic (I/E) regions of the CHI3L1 gene. These haplotypic blocks show common SNPs and a similar structure to those identified in the HapMap project in the Utah residents with Northern and Western European ancestry from the CEPH collection (CEU group-Figure E1B), and African ancestry in Southwest USA (ASW group- Figure E1C), respectively. This indicates that the SNP distribution and LD patterns seen in our study populations were consistent with HapMap populations of similar ancestry.

Figure 1. Haplotype map for individuals of European ancestry in the CHI3L1 gene in YCAAD and SARP.

Identification of CHI3L1 polymorphisms associated with serum YKL-40 and irreversible airflow obstruction

To identify polymorphisms in CHI3L1 that are associated with serum YKL-40 or post-bronchodilator FEV1 % predicted, individuals of EA in the YCAAD cohort (discovery) were analyzed using an additive model and adjusted for multiple comparisons (FDR < 0.10) as described in methods (Table 2 summarizes the significant associations in the EA and AA groups). Consistent with previous reports (5), multiple SNPs were associated with serum YKL-40 in the discovery cohort including rs4950928. In individuals of EA with the rs4950928 CC genotype serum YKL-40 levels were higher compared to those with the GG genotype (72 vs. 34 ng/mL, P <0.01) (Figure E2A). In addition, four CHI3L1 SNPs (rs12141494, rs880633, rs884209 and rs6698204) were associated with post-bronchodilator FEV1 % predicted (Table 2).

Table 2.

Significant associations of genetic variation in CHI3L1 with lung function and YKL-40 protein levels.

Cohort	SNP	Position relative to TSS	Genomic coordinates	Location	Minor Allele	MAF	HW P-value	Risk Allele	FEV1 post-BD PP P-value	Serum YKL-40 P-value
Discovery EA YCAAD
	rs884209	8462	201413912	3′ UTR	A	0.44	0.57	A	0.085	0.160
	rs12141494	4326	201418048	Intron 6	A	0.45	0.44	A	0.085	0.170
	rs880633	2950	201419424	Exon5	C	0.45	0.53	C	0.085	0.164
	rs1538372	1219	201421155	Intron 2	A	0.37	0.08	G	0.213	0.031
	rs4950928	-131	201422505	Promoter	G	0.22	0.85	C	0.213	8.1 × 10-5
	rs10399931	-329	201422703	5′ UTR	C	0.26	0.52	C	0.136	1.1 × 10-4
	rs10920579	-3221	201425595	5′ UTR	A	0.21	0.96	G	0.221	6.7 × 10-5
	rs6698204	-3432	201425806	5′ UTR	G	0.44	0.58	A	0.085	0.039
	rs4950929	-4375	201426749	5′ UTR	G	0.21	0.96	T	0.221	6.7 × 10-5
	rs946263	-9630	201432004	5′ UTR	G	0.19	0.95	A	0.198	6.7 × 10-5
	rs2486064	-13013	201435387	5′ UTR	A	0.41	0.38	G	0.660	0.030
Validation EA SARP
	rs2364572	9165	201413209	3′ UTR	T	0.04	0.42	T	0.006	NA
	rs2275353	5678	201416696	Intron 7	A	0.18	1.00	G	0.017	1
	rs12141494	4326	201418048	Intron 6	G	0.46	0.75	A	0.005	0.001
	rs1538372	1219	201421155	Intron 2	A	0.31	1	G	1	0.006
	rs4950928	-131	201422505	Promoter	G	0.20	0.82	C	1	0.011
	rs10399805	-247	201422621	5′ UTR	A	0.14	0.65	G	0.035	1
	rs946261	-2122	201424496	5′ UTR	G	0.38	0.27	A	0.011	0.053
African Ancestry
	rs6691378	-1371	201423745	5′ UTR	A	0.27	0.72	A	1	0.006

For all the genotyped SNPs and phenotypic associations, see online supplement, Table E2.

^*Genomic coordinates in Chromosome 1 using the Human Genome NCBI Build 36.1 (March, 2006)

All P values are adjusted for age and gender. Followed by correction for multiple comparison. FDR threshold for the discovery-EA YCAAD cohort < 0.1. Bonferroni threshold for the Validation EA SARP and AA cohorts < 0.05

Definition of Abbreviations: SNP = Single Nucleotide Polymorphism; TSS: Transcription start site; MAF = Minor Allele Frequency; HW = Hardy-Weinberg; FEV1 = Forced Expiratory Volume in One Second; PP= Percent of predicted; Post-BD=Post-bronchodilator; YCAAD = Yale Center for Asthma and Airways Disease; SARP = Severe Asthma Research Program; UTR = Untranslated Region.

Validation of SNPs with an effect on serum YKL-40 levels and post-bronchodilator FEV1 % predicted that were identified in the discovery cohort was conducted in individuals of EA in the SARP cohort. These association values were adjusted for multiple comparisons (Bonferroni < 0.05) as described in methods (Table 2). This analysis identified multiple SNPs associated with serum YKL-40 or post-bronchodilator FEV1. Consistent with the findings in the discovery cohort, individuals of EA with the rs4950928 CC genotype have higher serum YKL-40 levels compared to those with the GG genotype (53 vs. 16 ng/mL, P <0.01) (Figure E2B). Importantly, of the SNPs identified in the discovery cohort, only rs12141494 was significantly associated with both post-bronchodilator FEV1 % predicted (P < 0.01) and serum YKL-40 levels (P < 0.01) in the SARP cohort. Specifically the rs12141494 A allele was associated with decreased post-bronchodilator FEV1 percent predicted (Effect size) (95% Confidence Interval (CI)): -3.6 (-5.6/-1.6)). No associations were seen between the promoter SNP rs4950928 and pulmonary function in any of the cohorts (Table 2). In individuals of African ancestry only one SNP was associated with the quantitative traits evaluated, therefore, we restricted further analyses to individuals of European ancestry. Taken together, these analyses indicate that the non-promoter SNP, rs12141494 in intron 6, is strongly associated with persistent airflow obstruction and serum YKL-40 levels in individuals of European ancestry.

Effect of CHI3L1 Haplotypes on serum YKL-40 and lung function

To define potential interactions of CHI3L1 SNPs validated above on lung function and YKL-40 levels, a haplotype model was tested on all individuals of EA with asthma and healthy controls from the YCAAD and SARP cohorts. Of the nine SNPs genotyped in both cohorts, five SNPs were selected for this analysis based on their significant association with FEV1 % predicted or serum YKL-40 levels in both the discovery and validation steps (Table 2). Haplotype A present in 18% of individuals of EA, involving the SNP interaction between the minor allele in the promoter SNP rs4950928 (G) and the (G) allele in rs12141494 was associated with improved post-bronchodilator FEV1 and lower YKL-40 (Table 3). While haplotype B in 14 % of the population was associated with improved post-bronchodilator FEV1, involving the (G) allele in rs12141494. This analysis demonstrates that G alleles at both the promoter and the rs12141494 intronic SNP have a protective effect characterized by lower YKL-40 levels and higher post-bronchodilator FEV1 % predicted, demonstrating an interaction between promoter and non-promoter SNPs in the CHI3L1 gene.

Table 3. Haplotype analysis in individuals of European ancestry in YCAAD and SARP.

Haplotype	Frequency	Phenotype	Effect size and direction	P Value	rs2275353	rs12141494	rs1538372	rs495928	rs1399805
A	18%	FEV1 PP Post-BD	4	<0.01	G	G	A	G	G
A	20%	Serum YKL-40	-24	<0.01	G	G	A	G	G
B	14%	FEV1 PP Post-BD	6	<0.01	A	G	G	C	A
B	15%	Serum YKL-40	-0.6	0.93	A	G	G	C	A
C	12%	FEV1 PP Post-BD	2	0.19	G	G	A	C	G
C	13%	Serum YKL-40	-0.1	0.99	G	G	A	C	G
Base	51%				G	A	G	C	G

Identification of CHI3L1 SNPs associated with asthma severity

To determine the effect CHI3L1 SNPs on asthma severity, individuals of EA with asthma from YCAAD and SARP were combined (n=736). The Levene's test for homogeneity of variance was used before combining individuals from the two cohorts. Asthma severity and serum YKL-40 levels were included as independent variables. No differences in the variability of the two groups were seen. A linear regression analysis using an additive model of association was conducted in the combined group to identify their relationship with asthma severity (Table 4). The Hardy-Weinberg values for these SNPs were evaluated before implementing the association test to maintain the fundamental assumptions of equilibrium seen in the discovery and validation groups (Table 4). Of the SNPs evaluated, only rs12141494 was associated with asthma severity in individuals of EA (Bonferroni-adjusted P=0.04). The G allele had a protective effect on severity (OR (95% CI): 0.7 (0.6/0.9)), and was associated with lower serum YKL-40 levels (Bonferroni-adjusted P 0.04; Effect size (95% CI): -14.3 (-24.1/-4.4) ng/mL) and higher post-bronchodilator FEV1 % predicted (Bonferroni-adjusted P <0.01; Effect size (95% CI): 3.8 (1.9/5.7) % predicted). This observation indicates that rs12141494 is strongly associated with the severe asthma phenotype, including SARP defined severity, serum YKL-40 levels and persistent airflow obstruction.

Table 4.

Significant SNP associations in individuals of European ancestry with asthma.

SNP	Location	Minor Allele	MAF	HW P-value	Risk Allele	FEV1 post-BD PP	Serum YKL-40	Asthma Severity
rs12141494	Intron 6	G	0.47	0.11	A	9.2 × 10-4	0.043	0.040
rs1538372	Intron 2	A	0.31	0.07	G	0.348	0.012	1
rs4950928	Promoter	G	0.20	0.16	C	0.633	1.2 × 10-5	1

Effect of rs12141494 on the expression of YKL-40 in the airway

To determine the contribution of rs12141494 to the airway expression of YKL-40, we quantified the levels of YKL-40 in the sputum supernatants of individuals with asthma in the YCAAD cohort that underwent sputum induction as described in methods. Individuals of EA homozygous for the risk allele A had higher YKL-40 sputum levels compared to individuals homozygous for the G allele (23 [0-59] vs. 6 [0-33], ng/mL, P =0.043) (Figure 3). The promoter SNP rs4950928 was not associated with sputum levels (data not shown). This observation, shows that rs12141494 has an effect on the expression of YKL-40 in the airway of individuals with asthma and is consistent with the findings that the rs12141494 G allele is associated with lower levels of YKL-40, milder asthma and better lung function compared to individuals carrying the A allele.

Figure 3.

The rs12141494 AA genotype in individuals of European ancestry with asthma is associated with higher YKL-40 sputum levels.

In silico analysis of functional genomic impact of intronic CHI3L1 SNPs

An in silico analysis for the functional impact of rs12141494 was conducted using the Human Splicing Finder Version 2.4.1 (http://www.umd.be/HSF/HSF.html). The exon 6 CHI3L1 sequence was analyzed by selecting 1000 bp on the 5′ and 3′ flanks of exon 6 to include rs12141494. Mutation analysis of the position 1344 from C to T (complementary to G and A), identified enhancement in the splice site strength according to the Max Entropy algorithm (19), suggesting a potential effect of rs12141494 on pre-messenger RNA processing. Analysis of CHI3L1 mRNA expression in the circulation of a subset of individuals in YCAAD demonstrates a significant genotype effect concordant with that seen at the protein level in the airway (data not shown).

Discussion

These studies are the first to examine the effect of genetic variation in CHI3L1 on asthma severity and demonstrate that the intronic SNP, rs12141494, is associated with disease severity, airflow obstruction, and YKL-40 levels in the serum. This effect is independent of the promoter SNP rs4950928 and rs12141494 was the only SNP found to be associated with airway expression of the CHI3L1 gene product, YKL-40. Taken together, these findings suggest that rs12141494 contributes to asthma severity by altering airway expression of YKL-40 and contributes to airway remodeling, persistent airflow obstruction, and asthma severity.

Previous population studies of genetic variation in the CHI3L1 gene and asthma have shown an association with the promoter SNP rs4950928, asthma prevalence, and YKL-40 expression, however, the risk allele reported in the studies has differed. A European population study (6), identified the G allele in rs4950928 as the risk allele for atopic asthma in never smokers, while the study by Ober et al (5) reported that the C allele was associated with an increased risk of asthma and higher YKL-40 levels in the circulation. An additional publication has suggested that CHI3L1 polymorphisms are not associated with asthma (7), and in Korean children, other promoter polymorphisms are associated with atopy but not with asthma (8). In the current study, we confirmed that the C allele at rs4950928 is associated with higher serum levels of YKL-40 in two cohorts of individuals of European ancestry (Figure E2). Our study was underpowered to detect associations in individuals of African ancestry, therefore additional studies are needed to evaluate these associations in these populations. Interestingly, we did not find that the rs4950928 polymorphism was associated with airflow obstruction or airway levels of YKL-40 in either of the two populations of asthmatics that were examined which suggests that this polymorphism does not directly contribute to asthma severity.

Since serum YKL-40 levels were associated with severe asthma and lung function impairment, we suspected that other promoter or non-promoter polymorphisms could contribute to severe asthma. This led us to discover that a novel intronic SNP, rs12141494, is associated with airway expression of YKL-40 and persistent airflow obstruction in severe asthma. This SNP was not examined in previous genetic studies of CHI3L1 in asthma and was identified by overlapping direct DNA sequencing data of the CHI3L1 gene and HapMap data. Analysis of the CHI3L1 LD structure shows that rs12141494 resides in intron 6 within one of two haplotypic blocks separate from the promoter SNP rs4950928 (Figure 1). This SNP was associated with post-bronchodilator FEV1 % predicted and serum YKL-40 levels, and unlike rs4950928, was associated with the levels of YKL-40 in the sputum. The observation that the additive model had the best fit in conjunction with the quantitative differences seen in these phenotypes, suggests that there is a gene-dosage effect of the risk allele in these associations. The effect of this SNP on lung function persisted after controlling for the risk genotype (CC) at rs4950928 (Table E4) and evaluation of haplotypes, with the A allele associated with higher serum YKL-40 levels, lower lung function, and more severe asthma (Table 3). Individuals with asthma that are homozygous for the A allele of rs12141494 have significantly higher YKL-40 sputum levels, compared to homozygotes for the G allele, further supporting our finding that the rs12141494 genotype has a functional effect on the airway expression of YKL-40 (Figure 3). This suggests that rs12141494 is linked to severe asthma traits, whereas rs4950928 is not. Rs12141494 may contribute to circulating YKL-40 levels through production of YKL-40 in the airway and this effect may be enhanced by the presence of the rs4950938 C allele.

What are the possible functional effects of rs12141494 on the architectural changes characteristic of airway remodeling? The presence of higher YKL-40 levels in the airway of individuals with asthma and the rs12141494 A allele suggests that increased expression of YKL-40 is the primary effect of this polymorphism. Bronchial epithelial cells exposed to YKL-40 have been shown to increase expression of IL-8, which, in turn, leads to proliferation and migration of bronchial smooth muscle cells in vitro (3). In addition, epithelial expression of YKL-40 correlates with bronchial smooth muscle mass and activation of the protease activated receptor-2, AKT, ERK and p38 pathways (4). Therefore, higher local production of YKL-40 in the bronchial epithelium in individuals with the AA rs12141494 genotype may contribute to changes in airway structure, persistent airflow obstruction, and higher levels of YKL-40 in the circulation. This is supported by the results of the haplotype analysis that reveal that the G alleles at both SNPs are associated with lower levels of YKL-40 (Table 3). In addition, in silico analysis suggests that the effect of the rs12141494 A allele on pre-messenger RNA processing enhances the strength of the intron 6 splice site, resulting in increased CHI3L1 messenger RNA and YKL-40 protein levels. While analysis of mRNA expression in the circulation of a subset of individuals in YCAAD demonstrates a genotype effect concordant with that seen at the protein level in the airway.

In summary, these studies identify a novel SNP in CHI3L1, rs12141494, that is associated with persistent airflow obstruction, YKL-40 levels in the airway and blood, and severe asthma. The functional effect of the risk allele A may increase YKL-40 in the airway and circulation and may have synergistic effects in conjunction with the CHI3L1 SNP rs4950928. We used a rigorous definition of asthma, two cohorts of individuals with asthma that are representative of centers across the United States, and a single genotyping methodology to reduce bias to conduct these studies. While future studies will determine the effect of rs12141494 on other ethnic groups and the cellular control of YKL-40 synthesis and expression, these observations add to the body of evidence demonstrating that genetic variation in CHI3L1 and YKL-40 play important roles in the development of airway remodeling and severe asthma.

Figure 2.

The A allele of rs12141494 in individuals of European ancestry with asthma is associated with worse airflow obstruction (A), higher serum YKL-40 levels (B) and increased asthma severity (C).

Key Messages.

• Intronic variation, rs12141494, in the CHI3L1 gene encoding YKL-40 is associated with persistent airflow obstruction, airway remodeling and severe asthma.

• Individuals with the AA risk genotype at rs12141494 have higher sputum levels of YKL-40, suggesting that its effect may be mediated through increased protein levels in the airway.

Abbreviations

SNP

Single Nucleotide Polymorphism

CHI3L1

Chitinase 3-like 1

YCAAD

Yale Center for Asthma and Airways Disease

SARP

Severe Asthma Research Program

FEV1

Forced Expiratory Volume 1 second

GWAS

Genome-wide Association Study

IRB

Institutional Review Board

ELISA

Enzyme-linked immunosorbent assay

CV

Coefficient of variation

kb

kilobase

MAF

Minor Allele Frequency

LD

Linkage Disequilibrium

AIC

Akaike Information Criterion

FDR

False Discovery Rate

IQR

Interquartile Range

EM

Expectation Maximization

EA

European Ancestry

AA

African Ancestry

FeNO

Fractional exhaled nitric oxide

IgE

Immunoglobulin E

BMI

Body Mass Index

CEU

Utah residents with Northern and Western European Ancestry from the CEPH collection

ASW

African ancestry in Southwest USA

DNA

Deoxyribonucleic acid

RNA

Ribonucleic acid

OR

Odds Ratio

CI

Confidence Interval