Impact of Nfat5 on normoxia- and hypoxia-exposed mouse lung endothelial cells

Chronic hypoxia causes detrimental structural alterations in the lung, which are partially dependent on stress responses of the endothelium. In this context, we revealed that hypoxia-exposed murine lung endothelial cells (MLEC) activate nuclear factor of activated T-cells 5 (NFAT5/TonEBP) - a transcription factor that adjusts the cellular transcriptome to cope with multiple environmental stressors. Here, we studied the impact of NFAT5 on hypoxia-induced gene expression in MLEC by comparing the transcriptome of control and Nfat5-deficient MLEC, which were isolated from mouse lungs after exposure to normoxia and hypoxia for 7 days and processed for scRNA seq. Overall design: Nfat5fl/fl mice were crossed with Cdh5-CreERT2 mice and genetic ablation (deletion of loxP site-flanked exon 4 of Nfat5) of Nfat5 (Nfat5(EC)-/-) in the corresponding offspring was induced at 10-12 weeks of age by i.p. administration of 1 mg tamoxifen per day for 5 consecutive days or miglyol as solvent control (Nfat5fl/fl). Mice were used in experiments after a recovery period of 2-3 weeks. Mice were exposed to normobaric hypoxia for 7 days by placing them in a hypoxia chamber (A-Chamber/ProOx O2 controller, Biospherix, USA) with free access to drinking water and food (humidity: 50-65%, temperature: 21-23°C) with 10% oxygen and 90% nitrogen. Mouse lungs were digested by applying the Lung Dissociation Kit (Miltenyi Biotec, Bergisch Gladbach, Germany) following manufacturer's instruction. Single-cell suspensions were prepared by passing the digestion mixture through 18G and 19G cannula syringes and filtered through a 100 µm cell strainer. Erythrocytes were lysed after the addition of ACK buffer (154.4 mM ammonium chloride, 10 mM potassium bicarbonate, and 97.3 µM EDTA tetrasodium salt). CD31high cells were enriched by magnetic activated cell sorting (MACS). To exclude non-(vascular) endothelial cell populations, cell suspensions were treated with the following antibodies: anti-PDPN-AlexaFluor® 488 (eBioscience, Frankfurt am Main, Germany, 53–5381, RRID: AB_1106990), anti-LYVE1-AlexaFluor® 488 (eBioscience, 53–0443, RRID:AB_1633415), anti-CD45-FITC (BD Biosciences, Heidelberg, Germany, 553080, RRID:AB_394609), anti-LY76-FITC (BD Biosciences, 561032, RRID:AB_396936) and anti-EPCAM-FITC (eBioscience, 11-5791, RRID:AB_ 11151709) for 30 min at 4°C in PBS containing 5% FCS. Vascular endothelial cells were simultaneously labelled with anti-CD31-APC (BD Biosciences, 551262, RRID: AB_398497) and anti-CD146-PE-Cy7 (BioLegend, San Diego, CA, USA, 134713, RRID: AB_2563108). Dead cells were excluded by propidium iodide (PI, 1:2000) staining. CD45- LYVE1- LY76- PDPN- EPCAM- PI- CD31+ CD146+ cells were sorted using a BD FACSAria Fusion Flow Cytometer (BD Biosciences, Heidelberg, Germany). Immediately after sorting, cells were frozen in DMSO-supplemented MLEC medium and shipped to Single Cell Discoveries (Netherlands) for scRNAseq analysis. Upon arrival, cells were thawed, checked for viability (trypan blue), filtered and processed for scRNAseq analysis (~2-5,000 cells/sample) according to the indexing protocol CG000315. Sequencing was performed on Illumina Novaseq 6000 (167 GB, 2x50, S1, HMF, up to 50,000 reads/cell). Data analysis BCL files resulting from sequencing were transformed into FASTQ files using 10x Genomics Cell Ranger mkfastq. FASTQ files were mapped using Cell Ranger count. During sequencing, read 1 was assigned 28 base pairs, and was used to identify the Illumina library barcode, cell barcode and UMI. Read 2 was used to map to the mouse reference transcriptome (mm10). Filtering of empty barcodes was done by using Cell Ranger. Each sample represents the MLEC population isolated from one mouse lung.