Asthma is a heterogenous condition that remains poorly understood. Current guidelines do not sufficiently capture this heterogeneity, leading to suboptimal management and treatment strategies. A more comprehensive classification of asthma into biologically meaningful subgroups is needed.

We performed plasma metabolomic profiling of 1155 asthmatic children across four platforms covering a broad range of the metabolome; C8-positive, HILIC-positive, C18-negative and targeted Amide-Negative. We generated patient similarity networks for each platform that connected asthmatics via edges representing the similarity in their metabolomic profiles (controlling for age, sex and body mass index). We then fused the four networks using Similarity Network Fusion and performed spectral clustering on the resulting fused network.

We identified four clusters, or metabo-endotypes, and determined there was a significant difference across the metabo-endotypes in asthma relevant phenotypes including lung function (FEV1/FVC ratio, p-ANOVA=1.7x10-6); eosinophil count (p-ANOVA=0.04) and prevalent hay fever (p-ANOVA=0.01) a common asthma co-morbidity. We then recapitulated the four metabo-endotypes using plasma metabolomic profiles from an independent population of childhood asthmatics (n=955) and demonstrated these recapitulated metabo-endotypes shared the same clinical features as the discovery endotypes, with significant differences (p<0.05) in the same asthma phenotypes. We further observed that the metabo-endotype defined by the mildest asthma, was associated with higher levels of anti-inflammatory metabolites, while the opposite was seen in the metabo-endotype with the most severe asthmatics.

These findings demonstrate that clinically meaningful endotypes can be derived and validated using metabolomic data, and that interrogating the drivers of these metabo-endotypes can help understand their pathophysiology and generate therapeutic targets.

BACKGROUND: Total serum IgE (tIgE) is elevated in individuals suffering from allergic diseases such as asthma, rhinitis and eczema, and elevated tIgE has also been associated with coronary artery disease. Of the genome-wide association study (GWAS) loci identified for tIgE, the human leukocyte antigen (HLA) region has been the most consistently associated. In this study, we used next generation sequence data from the NHLBI Trans-Omics for Precision Medicine (TOPMed) program and graph-based alignment to perform an HLA association study of tIgE levels.

METHODS: HLA-LA was used to conduct HLA typing on the NHLBI BioData Catalyst cloud-based platform. Subjects for whom tIgE levels and TOPMed freeze5b CRAM files were available were included in the analysis (subjects from the Barbados Asthma Genetics Study, the Genetic Epidemiology of Asthma in Costa Rica, COPDGene, Genetics of Cardiometabolic Health in the Amish and the Framingham Heart Study). Analysis strata were defined based on study, race/ethnicity and asthma status (n=5,580 total; n=818 African, n=4,298 European, n=465 Hispanic/Latino ancestry; and n=798 asthmatics). Linear mixed effect models were used to test for association between presence/absence of a specific HLA allele and tIgE, separately for each stratum. Analyses were limited to 159 2-digit HLA alleles with a frequency of at least 1% in one or more strata. Age, sex and the first five principal components of genetic ancestry were included as fixed effect covariates in the models, and a kinship matrix was included as random effect to account for relatedness between individuals. Results were combined using inverse-variance meta-analysis.

RESULTS: HLA allele calls were of high quality, with a mean proportion of 0.008 discordant calls between duplicated samples and a mean Mendelian error rate of 0.016 in parent-offspring trios. HLA DRB1*13:03 was associated with increased levels of tIgE (Bonferroni-corrected p=0.008) in the European ancestry meta-analysis. HLA-DRB1*13 has previously been reported to have a higher frequency in Alternaria-sensitive moderate-severe asthmatics compared to Alternaria-sensitive mild asthmatics; Alternaria spores are the most common airborne mold in the USA.

Chronic lower respiratory disease (CLRD) is known to have both genetic and environmental (e.g., smoking) risk factors. Large genome-wide association studies (GWAS) have been performed for chronic obstructive pulmonary disease, asthma, and emphysema, pinpointing to hundreds of loci associated with CLRD and CLRD-related phenotypes. However, the majority of the SNPs associated with CLRD traits are in noncoding regions with an unknown regulatory mechanism, causal gene, and etiologic basis. Therefore, catalogs of molecular quantitative trait loci (molQTLs) incorporating gene-environment interaction effects are essential to gaining additional biological insights.
Here, we present the results of methylation quantitative trait loci (mQTL) mapping in ~900 participants of diverse ancestry as part of the Trans-Omics for Precision Medicine (TOPMed) Multi-Ethnic Study of Atherosclerosis (MESA) pilot project. We identified mQTLs for 40% of the tested methylation probes in whole blood with false discovery rate < 0.05. Given the cellular heterogeneity of the tissue under study, we estimated the proportions of six cell types in whole blood and mapped cell type interacting mQTLs (imQTLs) to study cell type-specific effects. Colocalization analysis using mQTLs and imQTLs revealed the role of variation in methylation levels as a likely molecular mechanism for some of the GWAS loci previously implicated in lung function in a multiethnic population of 90,715 individuals. Also, we identified additional colocalized loci that were driven by secondary signals or masked in bulk tissue by relaxing the assumption of one causal variant per trait or using cell type imQTLs, respectively. Furthermore, we decomposed the effects of methylation and gene expression as exposures to lung function.
Given the notable role of environmental factors in CLRD, we also mapped gene-by-environment (GxE) imQTLs, e.g., smoking imQTLs and age imQTLs. We observed that the GxE imQTLs were also enriched for being cell type imQTLs. Moreover, simulation and moderated mediation analyses suggested that these higher order GxE imQTLs may manifest their effect through mediation via changes in cell type composition. As an example, current smoking decreases NK cell proportions, and proportions of B cells decrease with aging. In turn, changes in these cell type proportions moderate the genotype effect on DNA methylation.
Taken together, mQTLs enhance our knowledge of regulatory genetic variants underlying risk in CLRD. Our results highlight the importance of cell type composition effects in colocalization with GWAS loci and as likely drivers of higher order gene-environment interactions.