GATA6 controls lipid accumulation in resident peritoneal macrophages.

Mean fluorescence intensity (MFI) flow-cytometric analysis of BODIPY staining performed on freshly isolated (A) and 12 hour cultured (B) Gata6-WT and Gata6-KOmye pMФ from male mice. Each symbol represents an individual mouse. Statistical significance was determined using unpaired t-test, **p<0.01, ***p<0.001. (C) Confocal microscopy analysis of BODIPY staining localisation in Gata6-WT and Gata6-KOmye pMФ. (D) DIC images (left) from Gata6-WT and Gata6-KOmye pMФ and corresponding false colour images generated from the overlay of FSC3 concentration maps of components C3 (blue, lipid component), C4 (green, protein), C5 (red, lipid component). Both images are on the same RGB scale, from 0 to a colour maximum scale M (vol/vol) with Mred = 0.167, Mgreen = 0.082. Scale bar: 10 μm. Analysis was performed on fields of view highlighted in black boxes in Figure S1B. (E) Susceptibility spectra from the FSC3 analysis. c5 has been multiplied by 0.2 to allow all components to be seen clearly on a single graph.

Dysregulation of sphingolipids in Gata6-deficient resident pMФ.

(A) Log2FC (Gata6-KOmye / -WT pMФ) of sphingolipids by lipid class and sub-class by chain-length. Log2FC is indicated by spot size. (B) Bar chart representation of sphingolipids by lipid class and sub-class by chain-length significantly changed in Gata6-KOmye pMФ compared to Gata6-WT. Statistical significance determined using t-test. Each lipid was analysed individually, without assuming a consistent SD, the graphs depict the mean±SEM, each symbol represents an individual mouse. Colours correspond to specific lipid category, with Gata6-KOmye in paler shading. (C) qPCR analysis of Smpd1 and Gba2 expression in RAW264.7 cells following infection with lentivirus containing shRNA specifically directed against the target genes or non-silencing control shRNA (Control). Analyses were performed on cells after selection by flow-cytometric sorting to high purity (>95%) on reporter gene expression (eGFP+), 7 days after infection. Data expressed as mean±SEM from three independent experiments. (D) Heatmaps of significantly changed sphingolipids in RAW264.7 cells (log2 FC) following infection with lentivirus containing shRNA against Smpd1 or Gba2 and changes to that lipid in Gata6-KOmye pMФ. The heatmaps show the difference (fold-change) between the indicated group and their respective controls: nd - levels of lipid not exceeding background, ns – not significant from its respective control. *p<0.05, **p<0.01, ***p<0.001.

Differential eicosanoid production in the absence of myeloid GATA6.

(A) Bar graphs showing expression of changed eicosanoids inside freshly isolated Gata6-WT and - KOmye pMФ and (B) in cell free peritoneum, measured using LC-MS/MS. Statistical significance was determined using Mann-Whitney test. (C) Relative expression of Ltc4s in Gata6-WT and Gata6-KOmye pMФ, measured by previously published microarray. (D) ImageStream analysis of LTC4S in Gata6-WT and -KOmye pMФ from 2 independent experiments indicated by the symbol colour and (E) a representative gallery from one experiment. (F) Immgen gene expression of enzymes involved in LTC4 to LTE4 metabolism in different cell types in the peritoneal cavity. P-value, *<0.05, **<0.01, ***<0.001. Data in bar graphs represents mean±SD, and each symbol represents an individual mouse, apart from (A) where multiple mice (10-18) of the same genotype were pooled for each point.

Dysregulated peritoneal eosinophil homeostasis in the absence of GATA6.

(A) Eosinophil proportion and absolute count in the peritoneal cavity of Gata6-WT and Gata6-KOmye female (black) and male (blue) mice. (B) Marker surface expression on peritoneal eosinophils (pre-gated by their characteristic SSChigh, Siglec-F+ phenotype) in Gata6-WT and -KOmye mice with corresponding representative flow cytometry density plots. (C) Annexin and PI staining of freshly isolated peritoneal eosinophils from Gata6-WT and Gata6-KOmye mice and quantification of Annexin V and PI double positive eosinophils. Indicated statistical analysis was performed. *p<0.05, ***p<0.001, ****p<0.0001. Each symbol represents an individual mouse.

Role of IL5 in peritoneal eosinophil regulation.

(A) Levels of IL5 (measured by ELISA) in cell free peritoneal lavage obtained from steady-state Gata6-WT and Gata6-KOmye mice. (B-D) Gata6-WT and Gata6-KOmye mice were treated with 20 µg of α-IL5 neutralising antibody (TRGK5) or isotype matched control antibody via intraperitoneal injection. After 72 hours, the proportion and absolute count of eosinophils in the (B) peritoneal cavity, (C) bone marrow and (D) peripheral blood was measured. (E) Proportion of Annexin V and propidium iodine (PI) double positive eosinophils in the peritoneal cavity was investigated in the same samples. Data are expressed as mean ± SD and analysis were performed using two-way ANOVA analysis, apart from (A) where t-test was used. Each symbol represents an individual mouse.

5-lipoxygenase inhibition counters increased eosinophil numbers seen in Gata6-KOmye mice.

(A) A schematic overview of the experiment performed in B and C. Gata6-WT and Gata6-KOmye mice were gavaged with 1mg/kg of indomethacin or 0.05% Ethanol control, twice 24 hours apart. (B) Analysis of eicosanoids from a single experiment is shown n≥2. (C) Peritoneal lavage samples were collected from these mice 24 hours after the last treatment and were investigated for the proportion, absolute count and percentage apoptosis (Annexin V, propidium iodine) of eosinophils. n≥10 individual mice per group, from 4 independent experiments. (D) Correlation plots between the amount of detectable LTE4 in cell free peritoneal lavage from Gata6-WT and Gata6-KOmye mice and the frequency of eosinophils in these samples. Non-parametric Spearman R correlation and simple linear regression. n≥11 individual mice from 2 independent experiments. (E) A schematic overview of the experiment performed in F and G. Gata6-KOmye mice were fed a complete nutrition Nutra-Gel containing 70mg/kg of Zileuton or control Nutra-Gel without any addition for 7 days with fresh supplemented Nutra-Gel provided daily. Peritoneal lavages were performed after 3 (Supplementary Figure 7E, F) or 7 days and (F) the levels of LTE4 in cell free lavage and (G) the proportion, absolute count and percent apoptosis (Annexin V, Propidium Iodine (PI)) of eosinophils were measured. n≥11 individual mice per group, from 2 independent experiments indicated by the symbol colour. Data are expressed as mean±SD and analysis were performed using two-way ANOVA analysis. *p<0.05, **p<0.01, ***p<0.001, ****p<0.0001.