Lipidomics Imaging of Human Postnatal Lung in Health and Bronchopulmonary Dysplasia

This dataset provides spatial lipidomics maps of human postnatal lung tissue from a premature infant with bronchopulmonary dysplasia (BPD) and an age‑matched non‑BPD control, generated to support the study “Multiscale Mapping of Altered Metabolism in Bronchopulmonary Dysplasia via Co‑Registered Mass Spectrometry and Non‑linear Optical Imaging.” Lung tissue was procured through the LungMAP Human Tissue Core (BRINDL) from the left upper lobe of two donors (one BPD, one control), each approximately 3–4 months corrected age. From each donor, four agarose‑inflated, CMC‑embedded fresh‑frozen tissue blocks (n = 4 healthy; n = 4 BPD) were selected for imaging.

Serial 20 µm cryosections were mounted on glass slides and analyzed by matrix‑assisted laser desorption/ionization Fourier transform ion cyclotron resonance mass spectrometry imaging (MALDI‑FTICR‑MSI) in positive and negative ion modes. Lipid imaging was performed at 35 µm lateral resolution within regions of interest that contained at least one bronchi or bronchiole, a muscularized vessel, and adjacent alveolar parenchyma. Annotations were generated through METASPACE using the SwissLipids, LipidMaps, and an internal LC‑MS/MS lung lipid database derived from matched 200 µm sections processed by MPLEx. In total, 280 lipids were putatively annotated (FDR ≤ 20%), with 108 species present in ≥70% of regions.

Histology (H&E) on the same sections was used by a pathologist to delineate functional tissue units (FTUs): alveolar parenchyma, bronchi, bronchioles, and large vessels. These FTUs were then used to aggregate MALDI signals into FTU‑resolved lipid profiles. These maps enabled comparison of lipid classes and specific species (e.g., phosphatidylcholines, phosphatidylglycerols, sphingolipids) between healthy and BPD tissues, revealing BPD‑associated reductions in key surfactant‑related PCs and PGs in alveoli, enrichment of longer‑chain PCs in airways, and distinct sphingolipid signatures in vessels.

This lipidomics imaging dataset is tightly integrated with the companion “U‑FLIP Multimodal Metabolic Imaging of Human Postnatal Lung in Health and Bronchopulmonary Dysplasia” dataset. The same sections were subsequently imaged by the U‑FLIP platform to obtain high‑resolution maps of optical redox state, lipid saturation, and collagen organization. Cross‑modal co‑registration via the HiMReg network permitted the direct correlation between FTU‑resolved lipid compositions and label‑free optical biomarkers of metabolism and extracellular matrix structure across the same tissue architecture.