Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2017 Jul 1.
Published in final edited form as: Allergy. 2016 Mar 24;71(7):1020–1030. doi: 10.1111/all.12869

Identification of STOML2 as a putative novel asthma risk gene associated with IL6R

Joana A Revez a, Melanie C Matheson b, Jennie Hui c,f, Svetlana Baltic g, AAGC collaborators h, Alan James i,j, John W Upham k, Shyamali Dharmage b, Philip J Thompson g, Nicholas G Martin a, John L Hopper b, Manuel AR Ferreira a
PMCID: PMC4907821  NIHMSID: NIHMS764444  PMID: 26932604

Abstract

Background

Functional variants in the interleukin-6 receptor gene (IL6R) are associated with asthma risk. We hypothesized that genes co-expressed with IL6R might also be regulated by genetic polymorphisms that are associated with asthma risk. The aim of this study was to identify such genes.

Methods

To identify genes whose expression was correlated with that of IL6R, we analyzed gene expression levels generated for 373 human lymphoblastoid cell lines by the Geuvadis consortium, and for 38 hematopoietic cell types by the Differentiation Map Portal (DMAP) project. Genes correlated with IL6R were then screened for nearby single nucleotide polymorphisms (SNPs) that were significantly associated with both variation in gene expression levels (eSNPs) and asthma risk.

Results

We identified 90 genes with expression levels correlated with those of IL6R and that also had a nearby eSNP associated with disease risk in a published asthma GWAS (N=20,776). For 16 (18%) genes, the association between the eSNP and asthma risk replicated with the same direction of effect in a further independent published asthma GWAS (N=27,378). Amongst the top replicated associations (FDR<0.05) were eSNPs for four known (IL18R1, IL18RAP, BCL6 and STAT6) and one putative novel asthma risk gene, Stomatin-like protein 2 (STOML2). The expression of STOML2 was negatively correlated with IL6R, while eSNPs that increased the expression of STOML2 were associated with increased asthma risk.

Conclusion

The expression of STOML2, a gene that plays a key role in mitochondrial function and T-cell activation, is associated with both IL-6 signaling and asthma risk.

Keywords: allergy, IL-6, IL-6R, mitochondria, SLP2

Introduction

The interleukin 6 receptor gene (IL6R) has been implicated in asthma pathophysiology both by human genetic association studies (1-3) and experimental animal models of asthma (4, 5). A common single nucleotide polymorphism (SNP) in IL6R was first identified as a risk factor for asthma in a meta-analysis of genome-wide association studies (GWAS) including 57,800 individuals (1). In that study, asthma risk was estimated to increase by 1.09-fold for each copy of the rs4129267:T allele, which has a frequency of 36% in Europeans. A second independent but less common SNP in IL6R (rs12083537, minor allele frequency of 19%) has also been reported to associate with asthma risk (OR=1.05 (2)). Notably, both rs4129267 and rs12083537 are associated with variation in serum protein levels of the soluble form of IL-6R (sIL-6R) (2, 6).

Variation in serum sIL-6R levels is thought to arise due to at least three mechanisms: (1) proteolytic shedding of the membrane bound form of IL-6R (mIL-6R) (7), which accounts for most variation in sIL-6R levels; (2) differential splicing of exon 9, which produces an isoform that directly encodes sIL-6R (8); and (3) release of microvesicle-associated IL-6R (9). The first two mechanisms are increased by the rs4129267:T asthma risk allele (6, 10). Increased shedding most likely arises because this allele is in phase (linkage disequilibrium [LD] r2 = 0.99 in Europeans) with the 358Ala amino-acid variant that is located within the main mIL-6R cleavage site (11) and that increases receptor shedding by ADAM proteases (10). As a result, the asthma rs4129267:T predisposing allele is associated with increased sIL-6R serum levels and decreased mIL-6R expression. This strongly suggests that increased asthma risk is associated with increased IL-6 trans-signaling, which is mediated by sIL-6R, and decreased IL6 classic signaling, mediated by mIL-6R (12).

The hypothesis that IL-6 trans-signaling contributes to disease pathophysiology has been experimentally demonstrated in mouse models of experimental asthma. Doganci et al. (4) showed that ovalbumin (OVA) -sensitized mice treated with sgp130-Fc, a fusion protein that inhibits trans- but not classic IL-6 signaling, had significantly lower eosinophilia and Th2 cytokine levels in bronchoalveolar lavage fluid (BALF) after OVA challenge, when compared to control mice. Similarly, Ullah et al. (5) found that both neutrophilia and eosinophilia induced by cockroach challenge were attenuated by sgp130-Fc, which was mirrored in significant decreases in IL-13, IL-17A and IL-17F levels in BALF. Furthermore, inhibition of IL-6 signaling had no protective effect in house dust mite challenged mice, which had lower levels of sIL-6R in the airways and so had reduced activation of IL-6 trans-signaling. These findings are consistent with the observation in the human genetic association studies that increased sIL-6R levels are associated with increased asthma risk.

Therefore, there is unambiguous evidence from both human and animal studies that dysregulation of IL6R expression contributes to the pathophysiology of a subset of individuals with asthma. On the other hand, there is some evidence that IL6R expression is associated with that of other immune and inflammatory genes (13, 14) but, to our knowledge, this possibility has not been studied systematically to date. In this study, we first analyzed two publicly available gene expression datasets to search for genes whose expression was correlated with that of IL6R (henceforth referred to as “IL6R-associated genes”); this information was then integrated with results from both asthma and transcriptome GWAS to identify the subset of IL6R-associated genes that are likely to be genetic risk factors for asthma. Identifying disease risk genes that are also associated with IL6R might provide new clues into the broader gene network that underlies the pathological effect of IL-6 signaling dysregulation in asthma.

Methods

Our analytical procedure is summarized in Figure 1 and described in detail below.

Figure 1. Flowchart of IL6R co-expression analysis.

Figure 1

Open in a new tab

Steps taken for the identification of genes with expression (i) associated with that of IL6R and (ii) regulated by asthma risk SNPs.

Analysis of gene expression levels in 373 Europeans studied by the Geuvadis consortium

To identify genes with expression levels correlated with those of IL6R, we first analyzed RNA-seq data generated by the Geuvadis consortium for human lymphoblastoid cell lines (LCLs) (15). LCLs are derived from peripheral blood B-cells and so represent a practical in vitro model to study gene expression patterns relevant to immune-related conditions.

We downloaded from the European Bioinformatics Institute (EBI) RPKM-normalized gene expression levels for 53,934 transcripts measured in LCLs of 462 unrelated individuals from the 1000 Genomes Project (accession E-GEUV-1, file GD660.GeneQuantRPKM.txt.gz). Transcripts expressed in >90% of individuals (N=18,702) were selected and distributions normalized using a rank-based inverse-normal transformation. We restricted our analysis to samples of European ancestry (N=373) and unique transcripts (unambiguous match between HGNC symbol and corresponding Ensembl ID) annotated in GENCODE (16) (N=15,440). The association between IL6R expression levels and the expression of each of the other 15,440 genes was then tested using linear regression, with covariates included in the model to adjust for the effects of sex, population (CEU, GBR, FIN, TSI), sequencing center and gene GC content (see below). Sixteen principal components (PCs) were also included as covariates to account for the effect of unmeasured confounders, as described below. These analyses were performed in R version 3.2.2.

Gene GC content can affect RNA-seq-derived expression levels differently in different individuals (17, 18). We tested this possibility and indeed found that in some individuals, increased gene GC content was associated with increased gene expression, while in others the reverse was true (Supplementary Figure 1). To correct for this potential confounder, we calculated a GC correction per individual as follows. From the 18,702 genes expressed in more than 90% of individuals, we (1) excluded a subset of 3,756 highly expressed genes (RPKM>21.48); (2) normalized the distribution of the remaining genes with a rank-based inverse-normal transformation; and (3) for each individual determined a correction value as the regression slope obtained from the regression of gene expression levels on gene GC content (for two examples, see Supplementary Figure 1). GC content for each gene was obtained from the Conditional Quantile Normalization R package for 15,277 genes (17). To correct for the effects of unmeasured confounders, PC analysis was applied to the matrix of quantile-normalized gene expression levels using the prcomp function of the R stats package (19), and the first 16 PCs were then extracted and included as covariates when testing the association between IL6R expression levels and the expression levels of other genes.

The top 10% of genes with expression most associated with that of IL6R were then selected for subsequent analyses. For a justification of how/why we selected a cut-off of 10% (instead of e.g. 5%), see section “Selection of cut-off to prioritize IL6R-associated genes” below.

Analysis of gene expression levels in 38 cell types from the DMAP project

We used a second independent approach to identify genes associated with IL6R; this approach differed from the analysis of the Geuvadis data in that we searched for genes correlated with IL6R within an individual but across immune cell types (e.g. cell types with increased expression of IL6R also have increased expression of gene X), instead of within a cell type but across individuals (e.g. individuals with increased expression of IL6R in LCLs also have increased expression of gene X).

A file (DMap_data.gct) containing normalized gene expression levels for 8,968 genes measured in 38 hematopoietic cell populations isolated from four to seven individuals was downloaded from the Differentiation Map Portal (DMAP) project (20) website. We then selected 7,372 unique genes annotated in GENCODE (16), and measured the association of their expression with IL6R expression using the cosine distance metric, as implemented in the GeneNeighbors module of GenePattern (21). As for the Geuvadis data, the top 10% of genes most associated with IL6R expression (based on the absolute of the distance metric) were selected for subsequent analyses.

Identification of SNPs associated with asthma risk and variation in gene expression

Not all genes associated with IL6R will contribute to asthma pathophysiology. To identify the subset of IL6R-associated genes whose expression might be causally-related to asthma risk, we combined information from published GWAS of asthma and published GWAS of gene expression levels. Specifically, for each gene X (e.g. BCL6) that is associated with IL6R, we (1) extracted summary association statistics (effect, risk allele, P-value) for SNPs within 1 Mb of the gene boundaries from the Ferreira et al. GWAS (22), which included 6,685 asthmatics and 14,091 controls; (2) selected the SNP with the most significant association with asthma (amongst those with P<0.05) in that 1 Mb interval; (3) identified all genes whose expression was previously reported to associate (typically, but not always, with FDR < 0.05 imposed by the original study) with this SNP (or a proxy, r2>0.8) in 11 published GWAS of gene expression levels conducted in cells/tissues relevant to asthma (15, 23-32); and (4) retained that SNP for further analysis if one of the genes known to be regulated by this SNP was also gene X (i.e. BCL6). In other words, we selected a subset of IL6R-associated genes that are regulated by a nearby SNP that associates with asthma risk. We considered an alternative more comprehensive analytical strategies (selection of all SNPs associated with asthma in step (2) above), but found that this would result in hundreds of SNPs being carried forward to validation, which would decrease power given the multiple testing burden (see below).

Selection of cut-off to prioritize IL6R-associated genes for downstream analyses

In the first step of our analytical procedure (Figure 1A), we used a 10% cut-off to select a group of genes whose expression was most strongly associated with that of IL6R. The choice of cut-off determined the number of genes selected for downstream analysis, and so also the number of SNPs that were moved to validation (see below). As such, an important consideration when deciding what cut-off to use was the power provided by the validation study to replicate a significant association after accounting for multiple SNP testing. Cut-offs of 5% and 10% resulted in 37 and 61 SNPs reaching the validation step, respectively. Based on the specific effect sizes and minor allele frequencies observed for these SNPs in the discovery GWAS, we estimated the power provided by the validation study to replicate the associations at a Bonferroni-corrected threshold of 0.0014 (0.05/37) and 0.0008 (0.05/61), respectively. The validation study was adequately powered to replicate 22 of the 37 (59%) SNP associations obtained with the 5% cut-off and 35 of the 61 (57%) associations obtained with the 10% cut-off. Therefore, the ability to replicate the associations with these two sets of SNPs was comparable between the two cut-offs (59% vs. 57%). The 10% cut-off maximised the opportunity to discover new risk SNPs, and so we selected that cut-off.

Validation of genetic associations with asthma risk

The previous analyses identified a set of SNPs found to regulate an IL6R-associated gene and that were also associated with asthma risk in the Ferreira et al. GWAS (22). We refer to these SNPs that are associated with variation in the expression of a nearby gene as “eSNPs”.

Some of these might represent novel risk variants for asthma. To test this possibility, we investigated whether these eSNPs were also associated with asthma risk in an independent study, the GABRIEL consortium GWAS (33). After excluding overlapping samples (N=1,207, Busselton cohort), results were based on the analysis of 12,077 asthmatics and 15,301 controls. A reproducible association was defined as a significant (P<0.05) and consistent (i.e. same direction of effect as in the Ferreira et al. GWAS (22) association in this analysis. The proportion of eSNPs for IL6R-associated genes with a reproducible association with asthma out of all eSNPs tested in the GABRIEL is denoted in the section below by fIL6R-genes. To obtain an overall measure of association between an eSNP and asthma risk, we used METAL (34) to meta-analyze results from the Ferreira et. al GWAS and the GABRIEL GWAS using a fixed-effects model.

Comparison with a random selection of genes

We applied the same analytical approach described above (Figure 1) to a random set of 1,000 genes, instead of IL6R. This allowed us to estimate the extent to which the enrichment of significant associations between asthma risk and eSNPs for IL6R-associated genes was because eSNPs in general – and not just those specific to IL6R-associated genes – are more likely to be disease-associated than randomly selected SNPs (35). To this end, we (1) selected a random gene tested in both the Geuvadis and DMAP datasets; (2) applied steps A through D (Figure 1) to identify genes co-expressed with that random gene (instead of IL6R); and (3) identified the proportion of eSNPs for genes co-expressed with that random gene that had a reproducible association asthma risk (and denote this fraction by frandom-genes). This procedure was repeated for 1,000 different random genes. We then calculated the mean frandom-genes across the 1,000 replicates, and the proportion of replicates where frandom-genes ≥ fIL6R-genes. Th is proportion represents the (empirically-derived) probability of observing reproducible eSNP associations with asthma risk more often than observed for IL6R-associated genes when selecting a random gene of interest.

Results

Identification of genes with expression levels associated with those of IL6R

To identify genes that were associated with IL6R, we used two complementary approaches (Figure 1 A). First, we tested the association between IL6R expression and that of 15,440 genes measured with RNA-seq in LCLs of 373 individuals of European descent by the Geuvadis consortium (15). In this analysis, the expression of 1,020 genes (6.6%) was associated with the expression of IL6R at a P<0.05. The most associated genes (N=1,544; top 10%) are listed in Supplementary Table 1 and include notable genes such as ATP8B2 and IL18R1, which are respectively a gene located near IL6R and a gene located in an established asthma risk locus (36).

The previous analysis aimed to identify genes associated with IL6R when comparing expression levels measured in a single cell type across different individuals. In a separate approach, we analyzed data from the DMAP project (20) to search for genes whose expression values were correlated with IL6R across 38 different hematopoietic cell populations. In this analysis, the expression of 3,643 genes (49%) was associated with that of IL6R at P<0.05. This excess of significant associations likely arises because cell types analyzed by the DMAP project derive from a common hematopoietic progenitor cell and many belong to the same lineage, and so their transcriptional profiles are highly correlated. (20) As for the analysis of the Geuvadis data, we then selected the top 10% of genes with expression patterns most associated with IL6R for subsequent analysis (N=737, Supplementary Table 1). Amongst the most IL6R-associated genes were for example LY96 and TLR2, two genes involved in the regulation of inflammation (37, 38).

After combining the lists of genes most associated with IL6R in these two independent approaches (top 10% of each), we obtained 2,203 genes (Figure 1 A), including those identified (1) in the Geuvadis dataset only (N=1,466); (2) in both the Geuvadis and DMAP datasets (N=78) and (3) in the DMAP dataset only (N=659).

Identification of asthma risk SNPs near IL6R-associated genes

The analysis above identified 2,203 genes with expression levels associated with IL6R, either across different individuals and/or cell types. We hypothesized that the expression of some of these genes might be causally related to asthma. If so, we would expect that genetic polymorphisms that regulate the expression of these genes would be associated with asthma risk. To test this possibility, for each gene, we first identified the nearby (+/− 1 Mb) single nucleotide polymorphism (SNP) with significant (P<0.05) and strongest association with asthma risk in a recently published asthma GWAS (22), that included 6,685 cases and 14,091 controls. For twelve genes there were no significant SNPs within 1 Mb, and so these were not considered further. The resulting list of 2,191 SNPs was further pruned by removing duplicate (i.e. same SNP selected for two or more genes; N=1,122) or correlated (linkage disequilibrium r2 > 0.1; N=79) SNPs, leaving for further analysis 990 independent SNPs that were associated with asthma risk and located near an IL6R-associated gene.

Regulation of gene expression by asthma risk SNPs near IL6R-associated genes

We then investigated if the association between each of these 990 independent SNPs and asthma risk could arise because these SNPs regulate the expression of the nearby IL6R-associated gene. Of the 990 SNPs, 358 (36%) were associated with the expression of at least one nearby gene in published GWAS of gene expression levels (which we refer to as eSNPs), including 84 (8.5%) that were eSNPs for the actual gene that was correlated with IL6R (Supplementary Table 2).

Replication of the association between asthma and eSNPs for IL6R-associated genes

Eighty four eSNPs were associated with both asthma risk in the Ferreira et al. GWAS (22) and expression levels of an IL6R-associated gene in published GWAS of gene expression. To validate the association observed between asthma risk and these 84 eSNPs, we analyzed publically available results from the GABRIEL asthma GWAS (33). After excluding samples that overlapped with the Ferreira et al. GWAS, results were available for 12,077 asthmatics and 15,301 controls. Of the 84 eSNPs, 61 were tested in the GABRIEL GWAS and for 14 of these (23%) there was a significant (P<0.05) and consistent (same direction of effect) association (Table 1 and Figure 1), when only about two reproducible associations were expected at this significance level by chance alone given multiple testing (61 [eSNPs tested] × 0.05 [P<0.05] × 0.5 [same direction] = 1.5).

Table 1.

Fourteen SNPs that are eSNPs for genes that are co-expressed with IL6R and also have a reproducible association with asthma risk.

Co-expression with IL6R
 Ferreira, 2014
 GABRIEL
No. eSNP Chr:bp Gene Cor Study OR, allele P-value Proxy SNP OR*, allele P-value
1 rs10197862 2:102350089 IL18R1 −0.202 G 1.24, A 4×10−11 rs13431828 1.2, C 2×10−9
rs10197862 2:102350089 IL18RAP −0.143 G 1.24, A 4×10−11 rs13431828 1.2, C 2×10−9
2 rs1464510 3:188394766 BCL6 0.324 D 0.91, A 0.0001 rs1559810 0.92, A 0.0001
3 rs7039317 9:34964538 STOML2 −0.346 D 1.11, T 0.0003 − 1.09, T 0.0010
4 rs324011 12:57108399 STAT6 0.415 D 1.09, T 0.0001 rs167769 1.07, T 0.0022
5 rs4833095 4:38798089 TLR1 0.448 D 1.2, T 5×10−12 - 1.07, T 0.0044
6 rs2357792 16:50734311 NOD2 0.648 D 1.08, A 0.0011 rs6500331 1.06, G 0.0050
7 rs13416555 2:8301605 ID2 0.324 D 1.12, C 1×10−5 rs10178845 1.06, G 0.0120
8 rs6511788 19:12369223 MAN2B1 0.425 D 1.07, T 0.0088 - 1.06, T 0.0160
9 rs11265424 1:160530060 SLAMF1 0.109 G 0.91, A 0.0005 rs1055880 0.95, C 0.0178
10 rs2435206 17:45980745 NSF 0.374 D 1.08, T 0.0014 rs2435211 1.05, T 0.0197
11 rs1253118 14:59497301 RTN1 0.556 D 1.07, T 0.0051 rs9323348 1.05, T 0.0210
12 rs13212921 6:27237643 BTN2A1 0.329 D 1.14, T 0.0002 rs13219354 1.08, C 0.0224
13 rs11000805 10:73903593 NDST2 0.349 D 1.07, C 0.0145 rs17741873 1.06, T 0.0316
14 rs9289837 3:151388204 P2RY13 0.452 D 0.88, T 1×10−5 rs7637803 0.95, T 0.0416
rs9289837 3:151388204 MED12L 0.129 G 0.88, T 1×10−5 rs7637803 0.95, T 0.0416
Open in a new tab

Highlighted in light grey are SNPs that had not previously been reported to associate with asthma risk. SNPs with significant association with asthma after correction for multiple testing (FDR<0.05) in the replication analysis are represented with bold font.

*

The OR is reported for the allele that is on the same haplotype as (i.e. in phase with) the allele reported for the Ferreira, 2014 GWAS SNP. Proxy SNPs were chosen based on an r2>0.8. Abbreviations: BP, base pair; Chr, chromosome; D, DMAP; G, Geuvadis; OR, Odds Ratio.

Of the 14 eSNPs (representing 16 genes; Figure 2) with a reproducible association with asthma risk in the replication analysis, four were significant after controlling for the 61 SNPs tested (FDR (39) < 0.05). Three of these are established risk variants for allergic disease, specifically those regulating the expression of IL18R1/IL18RAP (33), BCL6 and STAT6 (40). On the other hand, rs7039317 (combined discovery and replication analysis OR = 1.10 for the T allele, P=2× 10−6; Supplementary Table 3), an eSNP for STOML2, has not previously been reported to be associated with the risk of asthma or other allergic diseases. Therefore, STOML2 represents a putative novel risk gene for asthma.

Figure 2. Genes that are co-expressed with IL6R and have a reproducible association with asthma or allergies.

Figure 2

Open in a new tab

Underlined genes were significant at FDR<5% in the replication analysis. Distance of genes to IL6R, and of eSNPs to genes, represent strength of association (i.e. shorter distances represent stronger association).

Association between STOML2 expression, IL6R expression and asthma risk

To further characterize the association observed between the expression of STOML2 and that of IL6R in the DMAP project, we examined the expression of both genes across the 38 hematopoietic cell types studied. There was a negative correlation between IL6R and STOML2 expression (Figure 3); for example, central memory CD4+ T Cells and neutrophilic metamyelocytes had high IL6R expression and low STOML2 expression, whereas megakaryocytes and erythroids had low IL6R expression and high STOML2 expression.

Figure 3. Expression of IL6R and STOML2 across 38 hematopoietic cell types.

Figure 3

Open in a new tab

High gene expression levels are portrayed in red and low gene expression levels in white.

We also combined results from the asthma and transcriptome GWAS to predict the direction of effect of STOML2 expression on asthma risk. The rs7039317:T allele that was associated with increased asthma risk was associated with increased expression of STOML2 in whole-blood (29) (Table 2), suggesting that increased STOML2 expression has a predisposing effect on asthma.

Table 2.

Effect of asthma-associated SNPs on gene expression.

Asthma association
(Ferreira 2014 GWAS)
 Gene expression association
eSNP in LD with
asthma SNP
SNP rs ID Effect
allele		 OR rs ID r2 Gene P-value Study Tissue Effect on
expression*
rs10197862 A 1.24 rs10197862 1.00 IL18RAP 2.8×10−137 (29) Whole blood Decreased
rs10197862 A 1.24 rs950881 0.95 IL18R1 0.00025 (29) Whole blood Decreased
rs11000805 C 1.07 rs17741873 0.86 NDST2 2.7×10−10 (29) Whole blood Increased
rs11265424 A 0.91 rs6670721 0.81 SLAMF 1 3.0×10−15 (29) Whole blood Decreased
rs1253118 T 1.07 rs956901 0.87 RTN1 0.00160 (29) Whole blood Decreased
rs13212921 T 1.14 rs13217285 0.92 BTN2A1 1.7×10−15 (15) LCLs N/A
rs13416555 C 1.12 rs891058 0.99 ID2 6.1×10−08 (27) PBMCs Increased
rs1464510 A 0.91 rs9864529 0.81 BCL6 5.2×10−10 (27) PBMCs Increased
rs2357792 A 1.08 rs7342715 0.89 NOD2 3.5×10−11 (27) PBMCs Decreased
rs2435206 T 1.08 rs2435211 0.96 NSF 9.4×10−05 (24) LCLs N/A
rs324011 T 1.09 rs12368672 0.89 STAT6 9.8×10−198 (29) Whole blood Decreased
rs4833095 T 1.20 rs12233670 0.97 TLR1 2.8×10−57 (31) Whole blood Increased
rs6511788 T 1.07 rs10411986 0.98 MAN2B1 5.2×10−06 (15) LCLs N/A
rs7039317 T 1.11 rs10972275 0.93 STOML2 9.6×10−05 (29) Whole blood Increased
rs9289837 T 0.88 rs13327359 1.00 MED12L 2.4×10−37 (29) Whole blood Decreased
rs9289837 T 0.88 rs7637803 0.85 P2RY13 2.0×10−14 (29) Whole blood Decreased
Open in a new tab
*

Corresponding to the eSNP allele that is on the same haplotype as the asthma effect allele. Abbreviations: LD, linkage disequilibrium; N/A, not available; OR, odds ratio; SNP, single nucleotide polimorphism.

Other genes near STOML2 regulated by rs7039317

Many eSNPs are known to be associated with the expression of multiple genes, as we observed for rs10197862, an eSNP for both IL18R1 and IL18RAP. Although rs7039317 was only associated with the expression of a single gene that was correlated with IL6R, it could nonetheless be associated with the expression of other nearby genes unrelated to the IL-6 signaling pathway. When we queried results from 11 published GWAS of gene expression levels, there were an additional three genes whose expression was correlated with this SNP, namely VCP (P=2×10−17; (27, 29)), PIGO (P=3×10−6; (29)) and DNAJB5 (P=6×10−4; (29)). Therefore, in addition to STOML2, these three genes represent putative target genes for rs7039317 and so might also be related to asthma pathophysiology.

Comparison with a random selection of genes

The aim of this study was to identify asthma risk genes amongst IL6R-associated genes. This was assessed by identifying eSNPs with a reproducible association with asthma risk amongst eSNPs for IL6R-associated genes. We found 14 such eSNPs amongst 61 eSNPs tested (23%), a significant enrichment over the 2.5% expectation (0.05 [P<0.05] × 0.5 [same direction] = 0.025). It is possible that this enrichment of significant asthma associations amongst the eSNPs tested arose mostly because eSNPs in general are more likely to be disease-associated than randomly selected SNPs (35). To test this, we applied steps A through D (Figure 1) to a random gene of interest in place of IL6R, and repeated this analysis 1,000 times (see Methods for details). On average, we found that 17.4% (SD = 4.7%) of eSNPs for genes co-expressed with a given random gene had a reproducible association with disease risk, a 1.3-fold decrease when compared to 23% for the list of IL6R-associated genes, but this difference was not statistically significant (empirical P=0.120; see Methods for details). The observation that the enrichment of reproducible associations between eSNPs and asthma risk is comparable whether we considered IL6R or a random selection of genes does not imply that the associations are false-positives. Instead, it is consistent with previous studies that demonstrate that eSNPs are more likely to be associated with human traits than frequency-matched SNPs that are not related to gene expression (35) . This observation is important to interpret our results but does not detract from the identification of eSNPs that are related to asthma risk as well as the expression of IL6R-associated genes.

Discussion

In this study, we investigated the possibility that genes whose expression is associated with that of IL6R might also be causally related to asthma pathophysiology.

To identify genes with expression levels associated with that of IL6R, we used two complementary approaches, each with its own strengths and weaknesses. The RNA-seq dataset generated by the Geuvadis consortium (15) for 373 unrelated individuals provided a unique opportunity to identify genes that share transcriptional regulatory mechanisms with IL6R constitutively in LCLs; for example, these could be shared regulatory elements (e.g. enhancers) or transcription factors. The top three genes most correlated with IL6R expression in this analysis were: FYN (negative correlation), which regulates mast cell function (41), B-cell development (42) and is phosphorylated upon IL-6 binding to IL-6R (43); CD180 (positive correlation), which belongs to the Toll-like receptor family and is involved in innate immune response to mycobacteria (44); and ATP8B2 (positive correlation), which is located in close proximity (53 kb) to IL6R, suggesting that both genes might share a nearby regulatory element. Our results demonstrate that the expression of these genes is to some extent coordinated with that of IL6R in LCLs, and is consistent with the previously suggested role for IL6R in Treg development (4), innate immunity (45) and inflammation (46). The major caveat of this analysis was the use of an immortalized cell line that, although derived from a relevant cell type (B-cells), provides a very limited representation of gene expression patterns that are relevant for asthma.

To partly address the limitation of using LCLs to uncover IL6R-associated genes that are relevant for asthma, we also analyzed publicly available gene expression patterns measured in 38 different hematopoietic cell populations by the DMAP project (20). The strengths of this approach included the use of primary cells collected from volunteers (instead of cell lines) and the opportunity to identify genes associated with IL6R across cell types within individuals (instead of within a cell type across individuals). As such, the two approaches used (analyses of Geuvadis and DMAP data) shared the same aim but were conceptually distinct. An association between IL6R and another gene across cell types within an individual could arise, for example, if differentiation into specific cell lineages from a common progenitor requires temporal coordination of the expression of both genes (e.g. simultaneous expression; expression of one gene but not the other).

In this second approach, the proportion of genes with expression associated with that of IL6R (49%) at P<0.05 far exceeded the 5% nominal expectation. However, this is perhaps not too unexpected given that the cell types analyzed by the DMAP project not only derive from a common hematopoietic progenitor cell (hematopoietic stem cells CD133+ CD34dim), but also include cell types that belong to the same lineage: for example, CD8 T cells and CD4 T cells, early B cells and pro B cells. As highlighted in the original DMAP publication (20), the global transcription profiles from cell types of related lineages are highly correlated.

The top three genes whose expression was most closely associated with that of IL6R in the DMAP project were (all with a positive correlation): CD4, a cell surface antigen that is expressed on subsets of T cells, as well as on monocytes and macrophages, and plays an important role in T-helper cell development and activation (47); VIPR1, a high-affinity G protein-coupled receptor for vasoactive intestinal peptide, a neuropeptide that controls both innate and adaptive immunity (48, 49) and that has long been linked to asthma (50); and RGL1, a downstream effector protein of the Ras pathway of IL-6 signal transduction.

The clear literature links between IL-6 signaling and the function of some of the top IL6R-associated genes in the Geuvadis and DMAP analyses, suggests that both approaches were indeed able to identify specific components of the IL-6 signaling transduction pathway, its upstream regulators or downstream target genes. As for IL6R, we hypothesized that genetic dysregulation of the expression of these genes could affect asthma risk, for example by interfering with cellular differentiation or function in response to IL-6 stimulation. To address this hypothesis, we identified the subset of IL6R-associated genes whose expression was regulated by a SNP that was also a risk factor for asthma in a published GWAS (22). This list (90 genes) included genes directly relevant for both IL-6 signaling and asthma, for example TLR1 (51), IL17RA (52, 53) and NDFIP1 (54), amongst others.

Due to multiple testing, the SNP associations with asthma for some of these 90 genes were likely to represent false-positive findings. To identify those with a reproducible association, we extracted results for these SNPs from an independent asthma GWAS (33) and found that for 16 genes the SNP association with asthma was both significant and consistent. As such, this represents the group of genes which our analyses most convincingly link to both asthma etiology and IL-6 signaling. Five of these genes were identified at a significance threshold that controls for multiple SNP testing in the replication study (FDR < 0.05): IL18R1, IL18RAP, BCL6, STAT6 and STOML2. The first four, but not STOML2, are known target genes of SNPs that have an established (i.e. genome-wide significant) association with allergic disease (33, 40).

IL18R1 and IL18RAP encode the alpha and beta chains of the IL-18 receptor; IL-18 signaling through this receptor can activate IL-6 production (55), and IL-18 serum levels have been found to be positively correlated with sIL-6R levels (56). Therefore, together with these studies, our results indicate that the IL-18 and IL-6 signaling pathways are associated, and that this could in part be achieved by the coordinated transcription of IL6R, IL18R1 and IL18RAP.

BCL6 is a transcription factor with a critical role in the activation and differentiation of germinal center (GC) B cells (57) and CD4 T cell differentiation into T follicular helper (Tfh) cells (58). IL-6 stimulation is required for BCL6 expression, while in turn BCL6 induces IL6R expression (59). Thus, the expression of both genes is associated during specific B and T cell differentiation programs. Our finding that BCL6 expression is positively correlated with IL6R expression across hematopoietic cell types further supports the importance of coordinated transcription of these genes during normal cell lineage commitment. The association between SNPs that regulate the expression of these two genes and asthma risk suggests that their effect on asthma pathophysiology might be related to dysregulation of GC B cell and/or Tfh cell differentiation.

STAT6 is a transcription factor that plays a key role in adaptive immunity, by mediating IL-4 and IL-13 signaling through their cognate receptors (60), but also in innate immunity (61). Few studies have reported an association between IL-6 and STAT6, suggesting that these two signaling pathways might be mostly independent. However, in macrophages, IL-6 can induce the expression of the IL-4 receptor and augment IL-4-induced STAT6 signaling (62). As such, IL6R and STAT6 co-expression might be important for innate immune responses. Consistent with this possibility, in the DMAP project, the cell types with highest expression of both IL6R and STAT6 were monocytes (not shown), the precursors for macrophages.

Lastly, results from our analyses indicate that STOML2 is a putative novel asthma risk gene whose expression is negatively associated with that of IL6R. The overall statistical evidence we found for an association between the STOML2 eSNP and asthma risk (P=2×10−6) would be considered suggestive and not genome-wide significant in the context of a GWAS. As such, validation of this association in a well powered independent study is required to unambiguously confirm this eSNP as a novel risk factor for asthma.

STOML2 encodes for a protein that is mostly expressed in mitochondria, is required for the correct development of cell respiratory chain complexes (63), and promotes T cell activation (64). In our study, we found that the eSNP associated with increased STOML2 expression was associated with an increased disease risk, suggesting that STOML2 may have a pro-inflammatory effect in asthma. This is consistent with the observation that T cells from STOML2-deficient mice have decreased IL-2 production in response to cell activation, and this translated into reduced CD4+ T cell responses (65). In the DMAP project, CD4+ T cells had high expression of IL6R and low expression of STOML2. We speculate that in individuals with the rs7039317:T allele that increases STOML2 expression (29), mitochondria biogenesis and function in CD4+ T cells is increased, which results in stronger T cell responses to allergens, thereby increasing asthma risk. Functional studies that formally test this hypothesis are warranted.

Interestingly, IL-6R blockade has been shown to attenuate the development of cachexia by promoting mitochondrial biogenesis and dynamics (66), which are induced by STOML2 (65). These observations are consistent with our finding of a negative correlation between IL6R and STOML2 expression and, collectively, suggest that STOML2 is part of a gene network that underlies the effect of IL-6 on muscle loss. Of note, skeletal muscle dysfunction is common in chronic obstructive pulmonary disease (67) and is induced by chronic intake of corticosteroids (68).

In conclusion, we identified 16 genes whose expression was associated with IL6R and that were regulated by common polymorphisms that had a reproducible association with asthma risk. This list included 5 known and 11 putative new asthma risk genes, of which STOML2 had the strongest SNP association with asthma. These genes provide new clues into broader gene networks that are associated with IL-6 signaling and that contribute to asthma pathophysiology.

Supplementary Material

Supp info

Acknowledgments

We thank all study participants, including customers of 23andMe who answered surveys, as well as the employees of 23andMe, who together made this research possible. This work was supported in part by the National Human Genome Research Institute of the National Institutes of Health (R44HG006981) and the National Health and Medical Research Council of Australia (613627 and APP1036550). J.A.R. is supported by a doctoral grant (SFRH/BD/92907/2013) from Fundação para a Ciência e Tecnologia, Portugal.

Abbreviations

DDENDA2

Myeloid Dendritic Cell

TCELLA6

Naïve CD4+ T Cells

TCELLA7

Effective Memory CD4+ T Cell

GMP

Granulocyte-Monocyte Progenitor

TCELLA2

Naïve CD8+ T Cells

GRAN3

Neutrophil

MONO2

Monocyte

MONO1

Colony Forming Unit-Monocyte

GRAN2

Neutrophilic Metamyelocyte

TCELLA8

Central Memory CD4+ T Cell

ERY4

Erythroid 4

ERY3

Erythroid 3

HSC1

Hematopoietic Stem Cell 1

MEGA1

Colony Forming Unit-Megakaryocytic

HSC3

Hematopoietic Stem Cell 3

ERY5

Erythroid 5

ERY1

Erythroid 1

MEP

Megakaryocyte/Erythroid Progenitor

MEGA2

Megakaryocyte

BCELLA4

Mature B Cell Class Switched

PRE_BCELL3

Pro B Cells

NKA4

NKT

DENDA1

Plasmacytoid Dendritic cell

ERY2

Erythroid 2

PRE_BCELL2

Early B Cells

CMP

Common Myeloid Progenitor

EOS2

Eosinophil

GRAN1

Colony Forming Unit-Granulocyte

BASO1

Basophil

TCELLA4

Central Memory CD8+ T Cell

NKA1

Mature NK Cell 1

TCELLA1

Effective Memory RA CD8+ T Cell

TCELLA3

Effective Memory CD8+ T Cell

BCELLA3

Mature B Cell

BCELLA1

Naïve B Cells

NKA2

Mature NK Cell 2

NKA3

Mature NK Cell 3

BCELLA2

Mature B Cell Class Able To Switch

Footnotes

Author contributions

J.A.R. and M.A.R.F. conducted the analyses and wrote the manuscript. M.C. M., J.H., S.B., A.J., P.J.T., J.W.U., S.D., N.G.M and J.L.H. were involved in data collection and analysis for the asthma GWAS. All authors read and approved the manuscript for publication.

References

  • 1.Ferreira MA, Matheson MC, Duffy DL, Marks GB, Hui J, Le Souef P, et al. Identification of IL6R and chromosome 11q13.5 as risk loci for asthma. Lancet. 2011;378(9795):1006–1014. doi: 10.1016/S0140-6736(11)60874-X. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Revez JA, Bain L, Chapman B, Powell JE, Jansen R, Duffy DL, et al. A new regulatory variant in the interleukin-6 receptor gene associates with asthma risk. Genes Immun. 2013;14(7):441–446. doi: 10.1038/gene.2013.38. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Hawkins GA, Robinson MB, Hastie AT, Li X, Li H, Moore WC, et al. The IL6R variation Asp(358)Ala is a potential modifier of lung function in subjects with asthma. J Allergy Clin Immunol. 2012;130(2):510–515. doi: 10.1016/j.jaci.2012.03.018. e511. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Doganci A, Eigenbrod T, Krug N, De Sanctis GT, Hausding M, Erpenbeck VJ, et al. The IL-6R alpha chain controls lung CD4+CD25+ Treg development and function during allergic airway inflammation in vivo. J Clin Invest. 2005;115(2):313–325. doi: 10.1172/JCI22433. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Ullah MA, Revez JA, Loh Z, Simpson J, Zhang V, Bain L, et al. Allergen-induced IL-6 trans-signaling activates gammadelta T cells to promote type 2 and type 17 airway inflammation. J Allergy Clin Immunol. 2015;136(4):1065–1073. doi: 10.1016/j.jaci.2015.02.032. [DOI] [PubMed] [Google Scholar]
  • 6.Ferreira RC, Freitag DF, Cutler AJ, Howson JM, Rainbow DB, Smyth DJ, et al. Functional IL6R 358Ala allele impairs classical IL-6 receptor signaling and influences risk of diverse inflammatory diseases. PLoS Genet. 2013;9(4):e1003444. doi: 10.1371/journal.pgen.1003444. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Mullberg J, Dittrich E, Graeve L, Gerhartz C, Yasukawa K, Taga T, et al. Differential shedding of the two subunits of the interleukin-6 receptor. FEBS Lett. 1993;332(1-2):174–178. doi: 10.1016/0014-5793(93)80507-q. [DOI] [PubMed] [Google Scholar]
  • 8.Lust JA, Donovan KA, Kline MP, Greipp PR, Kyle RA, Maihle NJ. Isolation of an mRNA encoding a soluble form of the human interleukin-6 receptor. Cytokine. 1992;4(2):96–100. doi: 10.1016/1043-4666(92)90043-q. [DOI] [PubMed] [Google Scholar]
  • 9.Schumacher N, Meyer D, Mauermann A, von der Heyde J, Wolf J, Schwarz J, et al. Shedding of Endogenous Interleukin-6 Receptor (IL-6R) Is Governed by A Disintegrin and Metalloproteinase (ADAM) Proteases while a Full-length IL-6R Isoform Localizes to Circulating Microvesicles. J Biol Chem. 2015;290(43):26059–26071. doi: 10.1074/jbc.M115.649509. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Garbers C, Monhasery N, Aparicio-Siegmund S, Lokau J, Baran P, Nowell MA, et al. The interleukin-6 receptor Asp358Ala single nucleotide polymorphism rs2228145 confers increased proteolytic conversion rates by ADAM proteases. Biochim Biophys Acta. 2014;1842(9):1485–1494. doi: 10.1016/j.bbadis.2014.05.018. [DOI] [PubMed] [Google Scholar]
  • 11.Mullberg J, Oberthur W, Lottspeich F, Mehl E, Dittrich E, Graeve L, et al. The soluble human IL-6 receptor. Mutational characterization of the proteolytic cleavage site. J Immunol. 1994;152(10):4958–4968. [PubMed] [Google Scholar]
  • 12.Rose-John S, Scheller J, Elson G, Jones SA. Interleukin-6 biology is coordinated by membrane-bound and soluble receptors: role in inflammation and cancer. J Leukoc Biol. 2006;80(2):227–236. doi: 10.1189/jlb.1105674. [DOI] [PubMed] [Google Scholar]
  • 13.Jones GW, McLoughlin RM, Hammond VJ, Parker CR, Williams JD, Malhotra R, et al. Loss of CD4+ T cell IL-6R expression during inflammation underlines a role for IL-6 trans signaling in the local maintenance of Th17 cells. J Immunol. 2010;184(4):2130–2139. doi: 10.4049/jimmunol.0901528. [DOI] [PubMed] [Google Scholar]
  • 14.Sindhu S, Thomas R, Shihab P, Sriraman D, Behbehani K, Ahmad R. Obesity Is a Positive Modulator of IL-6R and IL-6 Expression in the Subcutaneous Adipose Tissue: Significance for Metabolic Inflammation. PLoS One. 2015;10(7):e0133494. doi: 10.1371/journal.pone.0133494. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Lappalainen T, Sammeth M, Friedlander MR, t Hoen PA, Monlong J, Rivas MA, et al. Transcriptome and genome sequencing uncovers functional variation in humans. Nature. 2013;501(7468):506–511. doi: 10.1038/nature12531. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Harrow J, Frankish A, Gonzalez JM, Tapanari E, Diekhans M, Kokocinski F, et al. GENCODE: the reference human genome annotation for The ENCODE Project. Genome Res. 2012;22(9):1760–1774. doi: 10.1101/gr.135350.111. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Hansen KD, Irizarry RA, Wu Z. Removing technical variability in RNA-seq data using conditional quantile normalization. Biostatistics. 2012;13(2):204–216. doi: 10.1093/biostatistics/kxr054. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Benjamini Y, Speed TP. Summarizing and correcting the GC content bias in high-throughput sequencing. Nucleic Acids Res. 2012;40(10):e72. doi: 10.1093/nar/gks001. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Team RC R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing. 2014 [Google Scholar]
  • 20.Novershtern N, Subramanian A, Lawton LN, Mak RH, Haining WN, McConkey ME, et al. Densely interconnected transcriptional circuits control cell states in human hematopoiesis. Cell. 2011;144(2):296–309. doi: 10.1016/j.cell.2011.01.004. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Reich M, Liefeld T, Gould J, Lerner J, Tamayo P, Mesirov JP. GenePattern 2.0. Nat Genet. 2006;38(5):500–501. doi: 10.1038/ng0506-500. [DOI] [PubMed] [Google Scholar]
  • 22.Ferreira MA, 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(6):1564–1571. doi: 10.1016/j.jaci.2013.10.030. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Fairfax BP, Makino S, Radhakrishnan J, Plant K, Leslie S, Dilthey A, et al. Genetics of gene expression in primary immune cells identifies cell type-specific master regulators and roles of HLA alleles. Nat Genet. 2012;44(5):502–510. doi: 10.1038/ng.2205. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Dimas AS, Deutsch S, Stranger BE, Montgomery SB, Borel C, Attar-Cohen H, et al. Common regulatory variation impacts gene expression in a cell type-dependent manner. Science. 2009;325(5945):1246–1250. doi: 10.1126/science.1174148. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Ding J, Gudjonsson JE, Liang L, Stuart PE, Li Y, Chen W, et al. Gene expression in skin and lymphoblastoid cells: Refined statistical method reveals extensive overlap in cis-eQTL signals. Am J Hum Genet. 2010;87(6):779–789. doi: 10.1016/j.ajhg.2010.10.024. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Dixon AL, Liang L, Moffatt MF, Chen W, Heath S, Wong KC, et al. A genome-wide association study of global gene expression. Nat Genet. 2007;39(10):1202–1207. doi: 10.1038/ng2109. [DOI] [PubMed] [Google Scholar]
  • 27.Zeller T, Wild P, Szymczak S, Rotival M, Schillert A, Castagne R, et al. Genetics and beyond--the transcriptome of human monocytes and disease susceptibility. PLoS One. 2010;5(5):e10693. doi: 10.1371/journal.pone.0010693. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Hao K, Bosse Y, Nickle DC, Pare PD, Postma DS, Laviolette M, et al. Lung eQTLs to help reveal the molecular underpinnings of asthma. PLoS Genet. 2012;8(11):e1003029. doi: 10.1371/journal.pgen.1003029. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Westra HJ, Peters MJ, Esko T, Yaghootkar H, Schurmann C, Kettunen J, et al. Systematic identification of trans eQTLs as putative drivers of known disease associations. Nat Genet. 2013;45(10):1238–1243. doi: 10.1038/ng.2756. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Consortium GT Human genomics. The Genotype-Tissue Expression (GTEx) pilot analysis: multitissue gene regulation in humans. Science. 2015;348(6235):648–660. doi: 10.1126/science.1262110. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Battle A, Mostafavi S, Zhu X, Potash JB, Weissman MM, McCormick C, et al. Characterizing the genetic basis of transcriptome diversity through RNA-sequencing of 922 individuals. Genome Res. 2014;24(1):14–24. doi: 10.1101/gr.155192.113. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Luo W, Obeidat M, Di Narzo AF, Chen R, Sin DD, Pare PD, et al. Airway Epithelial Expression Quantitative Trait Loci Reveal Genes Underlying Asthma and Other Airway Diseases. Am J Respir Cell Mol Biol. 2016;54(2):177–187. doi: 10.1165/rcmb.2014-0381OC. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.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(13):1211–1221. doi: 10.1056/NEJMoa0906312. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Willer CJ, Li Y, Abecasis GR. METAL: fast and efficient meta-analysis of genomewide association scans. Bioinformatics. 2010;26(17):2190–2191. doi: 10.1093/bioinformatics/btq340. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Nicolae DL, Gamazon E, Zhang W, Duan S, Dolan ME, Cox NJ. Trait-associated SNPs are more likely to be eQTLs: annotation to enhance discovery from GWAS. PLoS Genet. 2010;6(4):e1000888. doi: 10.1371/journal.pgen.1000888. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.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(3):342–347. doi: 10.1038/ng.323. [DOI] [PubMed] [Google Scholar]
  • 37.Shimazu R, Akashi S, Ogata H, Nagai Y, Fukudome K, Miyake K, et al. MD-2, a molecule that confers lipopolysaccharide responsiveness on Toll-like receptor 4. J Exp Med. 1999;189(11):1777–1782. doi: 10.1084/jem.189.11.1777. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Darville T, O'Neill JM, Andrews CW, Jr., Nagarajan UM, Stahl L, Ojcius DM. Toll-like receptor-2, but not Toll-like receptor-4, is essential for development of oviduct pathology in chlamydial genital tract infection. J Immunol. 2003;171(11):6187–6197. doi: 10.4049/jimmunol.171.11.6187. [DOI] [PubMed] [Google Scholar]
  • 39.Benjamini Y, Hochberg Y. Controlling the False Discovery Rate: A Practical and Powerful Approach to Multiple Testing. Journal of the Royal Statistical Society. Series B (Methodological) 1995;57(1):289–300. [Google Scholar]
  • 40.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(8):902–906. doi: 10.1038/ng.2694. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Parravicini V, Gadina M, Kovarova M, Odom S, Gonzalez-Espinosa C, Furumoto Y, et al. Fyn kinase initiates complementary signals required for IgE-dependent mast cell degranulation. Nat Immunol. 2002;3(8):741–748. doi: 10.1038/ni817. [DOI] [PubMed] [Google Scholar]
  • 42.Saijo K, Schmedt C, Su IH, Karasuyama H, Lowell CA, Reth M, et al. Essential role of Src-family protein tyrosine kinases in NF-kappaB activation during B cell development. Nat Immunol. 2003;4(3):274–279. doi: 10.1038/ni893. [DOI] [PubMed] [Google Scholar]
  • 43.Hallek M, Neumann C, Schaffer M, Danhauser-Riedl S, von Bubnoff N, de Vos G, et al. Signal transduction of interleukin-6 involves tyrosine phosphorylation of multiple cytosolic proteins and activation of Src-family kinases Fyn, Hck, and Lyn in multiple myeloma cell lines. Exp Hematol. 1997;25(13):1367–1377. [PubMed] [Google Scholar]
  • 44.Yu CH, Micaroni M, Puyskens A, Schultz TE, Yeo JC, Stanley AC, et al. RP105 Engages Phosphatidylinositol 3-Kinase p110delta To Facilitate the Trafficking and Secretion of Cytokines in Macrophages during Mycobacterial Infection. J Immunol. 2015;195(8):3890–3900. doi: 10.4049/jimmunol.1500017. [DOI] [PubMed] [Google Scholar]
  • 45.Hoge J, Yan I, Janner N, Schumacher V, Chalaris A, Steinmetz OM, et al. IL-6 controls the innate immune response against Listeria monocytogenes via classical IL-6 signaling. J Immunol. 2013;190(2):703–711. doi: 10.4049/jimmunol.1201044. [DOI] [PubMed] [Google Scholar]
  • 46.Rabe B, Chalaris A, May U, Waetzig GH, Seegert D, Williams AS, et al. Transgenic blockade of interleukin 6 transsignaling abrogates inflammation. Blood. 2008;111(3):1021–1028. doi: 10.1182/blood-2007-07-102137. [DOI] [PubMed] [Google Scholar]
  • 47.Yamane H, Paul WE. Early signaling events that underlie fate decisions of naive CD4(+) T cells toward distinct T-helper cell subsets. Immunol Rev. 2013;252(1):12–23. doi: 10.1111/imr.12032. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 48.Smalley SG, Barrow PA, Foster N. Immunomodulation of innate immune responses by vasoactive intestinal peptide (VIP): its therapeutic potential in inflammatory disease. Clin Exp Immunol. 2009;157(2):225–234. doi: 10.1111/j.1365-2249.2009.03956.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 49.Ganea D, Gonzalez-Rey E, Delgado M. A novel mechanism for immunosuppression: from neuropeptides to regulatory T cells. J Neuroimmune Pharmacol. 2006;1(4):400–409. doi: 10.1007/s11481-006-9044-0. [DOI] [PubMed] [Google Scholar]
  • 50.Barnes PJ. Neuropeptides in human airways: function and clinical implications. Am Rev Respir Dis. 1987;136(6 Pt 2):S77–83. doi: 10.1164/ajrccm/136.6_Pt_2.S77. [DOI] [PubMed] [Google Scholar]
  • 51.Mikacenic C, Schneider A, Radella F, Buckner JH, Wurfel MM. Cutting edge: Genetic variation in TLR1 is associated with Pam3CSK4-induced effector T cell resistance to regulatory T cell suppression. J Immunol. 2014;193(12):5786–5790. doi: 10.4049/jimmunol.1401185. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 52.Willis CR, Siegel L, Leith A, Mohn D, Escobar S, Wannberg S, et al. IL-17RA Signaling in Airway Inflammation and Bronchial Hyperreactivity in Allergic Asthma. Am J Respir Cell Mol Biol. 2015;53(6):810–821. doi: 10.1165/rcmb.2015-0038OC. [DOI] [PubMed] [Google Scholar]
  • 53.Zrioual S, Toh ML, Tournadre A, Zhou Y, Cazalis MA, Pachot A, et al. IL-17RA and IL-17RC receptors are essential for IL-17A-induced ELR+ CXC chemokine expression in synoviocytes and are overexpressed in rheumatoid blood. J Immunol. 2008;180(1):655–663. doi: 10.4049/jimmunol.180.1.655. [DOI] [PubMed] [Google Scholar]
  • 54.Ramon HE, Beal AM, Liu Y, Worthen GS, Oliver PM. The E3 ubiquitin ligase adaptor Ndfip1 regulates Th17 differentiation by limiting the production of proinflammatory cytokines. J Immunol. 2012;188(8):4023–4031. doi: 10.4049/jimmunol.1102779. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 55.Lee JK, Kim SH, Lewis EC, Azam T, Reznikov LL, Dinarello CA. Differences in signaling pathways by IL-1beta and IL-18. Proc Natl Acad Sci U S A. 2004;101(23):8815–8820. doi: 10.1073/pnas.0402800101. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 56.Tsirakis G, Pappa CA, Kaparou M, Boula A, Katsomitrou V, Xekalou A, et al. The relationship between soluble receptor of interleukin-6 with angiogenic cytokines and proliferation markers in multiple myeloma. Tumour Biol. 2013;34(2):859–864. doi: 10.1007/s13277-012-0618-6. [DOI] [PubMed] [Google Scholar]
  • 57.Basso K, Dalla-Favera R. Roles of BCL6 in normal and transformed germinal center B cells. Immunol Rev. 2012;247(1):172–183. doi: 10.1111/j.1600-065X.2012.01112.x. [DOI] [PubMed] [Google Scholar]
  • 58.Kroenke MA, Eto D, Locci M, Cho M, Davidson T, Haddad EK, et al. Bcl6 and Maf cooperate to instruct human follicular helper CD4 T cell differentiation. J Immunol. 2012;188(8):3734–3744. doi: 10.4049/jimmunol.1103246. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 59.Nurieva RI, Chung Y, Martinez GJ, Yang XO, Tanaka S, Matskevitch TD, et al. Bcl6 mediates the development of T follicular helper cells. Science. 2009;325(5943):1001–1005. doi: 10.1126/science.1176676. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 60.Hebenstreit D, Wirnsberger G, Horejs-Hoeck J, Duschl A. Signaling mechanisms, interaction partners, and target genes of STAT6. Cytokine Growth Factor Rev. 2006;17(3):173–188. doi: 10.1016/j.cytogfr.2006.01.004. [DOI] [PubMed] [Google Scholar]
  • 61.Chen H, Sun H, You F, Sun W, Zhou X, Chen L, et al. Activation of STAT6 by STING is critical for antiviral innate immunity. Cell. 2011;147(2):436–446. doi: 10.1016/j.cell.2011.09.022. [DOI] [PubMed] [Google Scholar]
  • 62.Mauer J, Chaurasia B, Goldau J, Vogt MC, Ruud J, Nguyen KD, et al. Signaling by IL-6 promotes alternative activation of macrophages to limit endotoxemia and obesity-associated resistance to insulin. Nat Immunol. 2014;15(5):423–430. doi: 10.1038/ni.2865. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 63.Mitsopoulos P, Chang YH, Wai T, Konig T, Dunn SD, Langer T, et al. Stomatin-like protein 2 is required for in vivo mitochondrial respiratory chain supercomplex formation and optimal cell function. Mol Cell Biol. 2015;35(10):1838–1847. doi: 10.1128/MCB.00047-15. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 64.Kirchhof MG, Chau LA, Lemke CD, Vardhana S, Darlington PJ, Marquez ME, et al. Modulation of T cell activation by stomatin-like protein 2. J Immunol. 2008;181(3):1927–1936. doi: 10.4049/jimmunol.181.3.1927. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 65.Christie DA, Mitsopoulos P, Blagih J, Dunn SD, St-Pierre J, Jones RG, et al. Stomatin-like protein 2 deficiency in T cells is associated with altered mitochondrial respiration and defective CD4+ T cell responses. J Immunol. 2011;189(9):4349–4360. doi: 10.4049/jimmunol.1103829. [DOI] [PubMed] [Google Scholar]
  • 66.White JP, Puppa MJ, Sato S, Gao S, Price RL, Baynes JW, et al. IL-6 regulation on skeletal muscle mitochondrial remodeling during cancer cachexia in the ApcMin/+ mouse. Skelet Muscle. 2012;2:14. doi: 10.1186/2044-5040-2-14. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 67.Schols AM, Soeters PB, Dingemans AM, Mostert R, Frantzen PJ, Wouters EF. Prevalence and characteristics of nutritional depletion in patients with stable COPD eligible for pulmonary rehabilitation. Am Rev Respir Dis. 1993;147(5):1151–1156. doi: 10.1164/ajrccm/147.5.1151. [DOI] [PubMed] [Google Scholar]
  • 68.Bodine SC, Furlow JD. Glucocorticoids and Skeletal Muscle. Adv Exp Med Biol. 2015;872:145–176. doi: 10.1007/978-1-4939-2895-8_7. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supp info
NIHMS764444-supplement-Supp_info.docx (367.8KB, docx)

RESOURCES