Fig 2: CXCL1 is controlled by MyD88 in lung epithelial cells and prolongs survival in MyD88(−/−) mice following A. fumigatus challenge.(A) BALF (B) lung neutrophil recruitment in WT → WT (black bars), MyD88(−/−) → WT (dark grey bars), WT → MyD88(−/−) (light grey bars), and MyD88(−/−) → MyD88(−/−) (white bars) BM chimeric mice 10 h p.i. with 3 × 107 conidia. Data are expressed as the fold change when compared to the WT → WT group and were pooled from 3 experiments with 12–15 mice per group. (C) BALF and (D) lung neutrophil recruitment in IL1R(−/−) → WT (black circles), and WT → IL1R(−/−) (white circles) BM chimeric mice 10 h p.i. with 3 × 107 conidia. Data are expressed as mean (±SEM) and are from an experiment with 9 mice per group. (E-H) Mean (+SEM) BALF (E) neutrophil recruitment, (F) CXCL1, (G) CXCL2 and (H) CXCL5 levels, in MyD88(−/−) CC10-MyD88 (CC10-MyD88+; black bars) and in MyD88(−/−) transgene-negative littermate controls (CC10-MyD88−; grey bars) 10 h p.i. with 3 × 107 conidia. Data were pooled from 2 experiments and include 7–9 mice per genotype. (I) Kaplan-Meier survival plot of MyD88(−/−) mice challenged with 6–7 × 107 conidia and treated 4 h p.i. with 50 ng rCXCL1 (white circles, n = 11), or PBS vehicle (grey circles, n = 12). Data were pooled from 2 experiments (p = 0.026, Gehan-Breslow-Wilcoxon test).

Fig 3: The clinical impact of CXCL5-correlated genes.A–E CXCL5 expression is correlated with many genes associated with hypoxia (A), glycolysis-related pathways (e.g., glycolysis, PPP, serine/glycine biosynthesis, one-carbon metabolism, and glutathione metabolism) (B), monosaccharide metabolism (C), nitrogen metabolism (D), and sterol biosynthesis (E). F–K The overall survival of PAAD patients with different expression levels of the genes mentioned in A–E. The corresponding metabolic pathways are above the plots.

Fig 4: Cxcl5−/− cells are susceptible to hypoxia-induced cell death due to defective upregulation of Hif1α.A Heatmap analysis of hypoxia-related genes. B, C Hif1α expression in 2D- or 3D-cultured WT and Cxcl5−/− Panc02 cells assessed by qRT-PCR (B) and western blotting (C) (mean ± SEM, n = 3; 9–12 spheroids per replicate). D Immunofluorescence micrographs (left) showing hypoxic regions of WT and Cxcl5−/− Panc02 spheroids detected with Cyanine 5-conjugated anti-EF5 antibody (red). Nuclei (blue) were visualized with DAPI. E Immunofluorescence micrographs (left) showing cell death as detected by TUNEL staining (green). Bar graph (right) showing the percentage of TUNEL-positive cells (mean ± SEM, n = 3). F Flow cytometry plot (left) showing annexin V+ cells (WT and Cxcl5−/−) cultured in 2D normoxic or hypoxic conditions. Bar graph (right) showing the percentages of annexin V+ cells (mean ± SEM, n = 3). G Bar graph showing the percentage of WT, Cxcl5−/−, and Hif1αOECxcl5−/− cells cultured in 2D normoxic or hypoxic conditions (mean ± SEM, n = 3). B–G Data are representative of three independent experiments. Significance was assessed via two-way (B) or one-way ANOVA (F, G) with Tukey’s post-hoc tests or via unpaired two-tailed Student’s t-tests (E). D, E Magnification, 20X. Scale bar = 200 μm.

Fig 5: Cxcl5 deficiency impairs one-carbon metabolism, increasing oxidative stress responses.A Volcano plot showing metabolites observed at significantly different levels in the comparison of 3D-cultured Cxcl5−/− cells to 3D-cultured WT cells. B Joint pathway analysis combining DEGs and differential metabolites using MetaboAnalyst 6.0. The metabolic pathways with p < 0.1 and an impact score > 0.5 appear in red. C Schematic of one-carbon metabolic pathways showing up- (red) or down-regulated (blue) genes and metabolites in the comparison between 3D-cultured Cxcl5−/− cells and 3D-cultured WT cells. The trans-sulfuration process appears in yellow. THF Tetrahydrofolate, me-THF 5,10-methyleneTHF, mTHF 5-methylTHF, F-THF 10-formylTHF, MET Methionine, SAM S-adenosylmethionine, SAH S-adenosylhomocysteine, hCYS homocysteine. D Flow cytometry plot showing mitochondrial ROS stained with MitoSOXTM Red (left) and the MFI of MitoSOXTM Red (right, mean ± SEM, n = 3). E MMP polarization was analyzed by flow cytometry and measured as the ratio of red (aggregates) and green (monomers) JC-1 dye fluorescence (mean ± SEM, n = 3). F Flow cytometry plots (left) showing lipid peroxidation detected as oxidized BODIPY-C11 signals. The percentages of oxidized BODIPY-C11+ cells are also shown (right, mean ± SEM, n = 3). For 3D-cultured samples of D–F, 12–18 spheroids were pooled per replicate. Data are representative of three independent experiments. Significance was assessed via two-way ANOVA with Tukey’s post-hoc tests.