Metabolome and microbiome multi-omics integration from a murine lung inflammation model of bronchopulmonary dysplasia

Abstract

Background: Respiratory tract microbial dysbiosis can exacerbate inflammation and conversely inflammation may cause dysbiosis. Dysbiotic microbiome metabolites may lead to bronchopulmonary dysplasia (BPD). Hyperoxia and lipopolysaccharide (LPS) interaction alters lung microbiome and metabolome, mediating BPD lung injury sequence.

Methods: C57BL6/J mice were exposed to 21% (normoxia) or 70% (hyperoxia) oxygen during postnatal days (PND) 1-14. Pups were injected with LPS (6 mg/kg) or equal PBS volume, intraperitoneally on PND 3, 5, and 7. At PND14, the lungs were collected for microbiome and metabolomic analyses (n = 5/group).

Results: Microbiome alpha and beta diversity were similar between groups. Metabolic changes included hyperoxia 31 up/18 down, LPS 7 up/4 down, exposure interaction 8. Hyperoxia increased Intestinimonas abundance, whereas LPS decreased Clostridiales, Dorea, and Intestinimonas; exposure interaction affected Blautia. Differential co-expression analysis on multi-omics data identified exposure-altered modules. Hyperoxia metabolomics response was integrated with a published matching transcriptome, identifying four induced genes (ALDOA, GAA, NEU1, RENBP), which positively correlated with BPD severity in a published human newborn cohort.

Conclusions: We report hyperoxia and LPS lung microbiome and metabolome signatures in a clinically relevant BPD model. We identified four genes correlating with BPD status in preterm infants that are promising targets for therapy and prevention.

Impact: Using multi-omics, we identified and correlated key biomarkers of hyperoxia and LPS on murine lung micro-landscape and examined their potential clinical implication, which shows strong clinical relevance for future research. Using a double-hit model of clinical relevance to bronchopulmonary dysplasia, we are the first to report integrated metabolomic/microbiome landscape changes and identify novel disease biomarker candidates.