Cytokines and metabolites are the main descriptors of severe asthma in children when performing multi-omics analysis of bronchoaleveolar lavages

Abstract

Abstract Background. Severe asthma (SA) is a heterogeneous condition with multiple phenotypes. There is still an unmet need to characterize and understand underlying mechanisms taking place in the lungs in order to propose the most suitable therapeutic strategies for SA. For this purpose, we aimed to identify a local signature of severe asthma by conducting comprehensive multi-omics analysis of bronchoalveolar lavages fluids (BALs) from children with SA versus non-asthmatic (NA) controls. Method. BALs were collected from twenty children with SA and from ten age-matched NA. We previously analyzed soluble and cellular immune components in those samples, and now propose to perform comprehensive analysis of their microbiota and their metabolome. Briefly, DNA from BALs was extracted and 16S rRNA gene (V3-V4 region) was amplified by PCR and sequenced. In parallel, untargeted metabolomics was performed using liquid chromatography coupled to high resolution mass spectrometry (LC-HRMS) following an established workflow for sample preparation, data acquisition and treatment. Each microbiome and metabolome dataset was first analysed independently by unsupervised multivariate analyses (Principal component analyses, PCA). Differences between groups for microbiota diversity indices, the relative distribution of each phyla and genera were then analysed. Metabolite set enrichment analysis (MSEA) and univariate supervised analysis were also performed. To identify a local signature of severe asthma, microbiota and metabolome data were further integrated, together with immune and with clinical data, using unsupervised Multi-Omics Factor Analysis (MOFA). Results. Microbiota diversity was higher in children with SA versus NA, with higher relative abundances of Streptococcus, Corynebacterium, Tropheryma whipplei, Dolosigranulum pigrum and Moraxella nonliquefaciens. We identified 88 metabolites in BALs, but unsupervised PCA of corresponding data did not differentiate children with SA from NA. However, MSEA evidenced that biotin and carnitine synthesis, lysine degradation, methionine metabolism and spermidine and spermine biosynthesis pathways were significantly enriched in children with SA. Finally, multiblocks data integration identified a signature of SA, mainly described by metabolites and cytokines. Conclusion. By integrating metabolome, microbiome and cytokines data obtained on BALs from children with severe asthma versus NA, our study uniquely described a local signature of SA.