Fig 2: Inhibition of hPGDS prevents PGD2 production by human monocyte-derived macrophages and CD68+ phagocytes strongly express hPGDS in human lung tissue. (A) Human monocyte-derived macrophages were incubated with hPGDS inhibitors or vehicle overnight and stimulated with LPS/IFN-? in the presence of inhibitors for 6 h. (B) Monocyte-derived macrophages at the time of collection (representative of n = 4 donors; scale bar 100 µm). (C) hPGDS inhibitors Cay16256 (10 µM) and HQL-79 (20 µM) significantly reduced PGD2 production by macrophages. (D) Serial cut precision-cut lung slices (PCLS) from human lung tissue resections were used. (E) Co-staining of CD68 and hPGDS showed that macrophage-like cells already express hPGDS in unstimulated PCLS and to a stronger extent in LPS/IFN-?-treated PCLS. (F) hPGDS+ tryptase+ mast cells were present in donor PCLS, but hPGDS levels were less pronounced compared to macrophages and unaffected by stimulation with LPS/IFN-?. Data are shown as box-and-whisker plot, n = 5. One-way ANOVA for repeated measurements (mixed model) with Sidak’s post hoc test, *** p < 0.001.

Fig 3: Macrophages and monocytes sorted from the healthy murine lung express hPGDS, but not LPGDS, and release PGD2 upon stimulation with LPS/IFN-?. (A) Schematic experimental approach. Total lung was extracted and digested with Dispase II. Monocytes, macrophages and mast cells were sorted from whole lung single-cell suspension according to displayed gating strategy. Representative images of populations obtained after sorting were stained with Diff/Quik (pooled from 2 mice, n = 4). (B) Mean percentage of monocytes, macrophages and mast cells of total viable cells obtained from healthy lungs. (C) Flow cytometry showed that lung-associated monocytes express high levels of hPGDS. (D) hPGDS transcriptional expression was comparable in all three sorted populations, while (E) LPGDS mRNA was very low. (F) Stimulation with LPS/IFN-? for 18 h significantly increased PGD2 levels in monocyte- and macrophage-conditioned media. Data are shown as box-and-whisker plots with one data point representing one mouse, n = 4–8, (C–E) one-way ANOVA for repeated measurements with Tukey’s post hoc test (log-transformed) or (F) two-way ANOVA for repeated measurements with Fisher’s LSD post hoc test, * p < 0.05, ** p < 0.01, *** p < 0.0001.

Fig 4: Evaluation of hematopoietic PGD synthase expression in human peripheral blood leukocytes highlights monocytes as a potential PGD2 source. (A) CD14+ monocytes, CD4+ T-cells, NK/T and NK cells as well as plasma cells stained highly positive for hPGDS. (B) Gating strategy for peripheral blood leukocytes, which were stained in the PMNL or PBMC fraction of peripheral blood from human donors. (C) For comparison, hPGDS expression in circulating mast cells and monocytes was determined in whole blood. Data are shown as box-and-whisker, n = 10, Student’s t-test (*** p < 0.001).

Fig 5: LPS/IFN-?, but not IL-4/IL-13, induce hPGDS-dependent PGD2 release and cytokine secretion by human donor precision-cut lung slices. (A) Serial precision-cut lung slices (PCLS) from human donors were pre-treated and stimulated in presence of hPGDS inhibitors HQL-79 (20 µM) or Cay16256 (10 µM) for indicated durations and activated with LPS/IFN-? (100 ng/mL and 20 ng/mL, respectively) or IL-4/IL-13 (20 ng/mL). (B) LPS/IFN-?, but not IL-4/IL-13, initiated a significant release of PGD2. (C) Inhibition of hPGDS significantly reduced PGD2 release as well as (D) TNF-a, (E) IL-6 and (F) IL-10 secretion by PCLS after LPS/IFN-? stimulation for 6 h. After 24 h stimulation with LPS/IFN-?, HQL-79 still significantly reduced (G) PGD2 and (H) TNF-a release. There were no significant changes in (I) IL-6 and (J) IL-10 levels. Data are shown as box-and-whisker or before-after plots, n = 5–9. (B) Two-way ANOVA for repeated measurements (mixed model), (C–E,G,H) One-way ANOVA for repeated measurements (mixed model) with Sidak’s post hoc test (log-transformed), (F,J) Kruskal–Wallis test, * p < 0.05, ** p < 0.01, *** p < 0.001.