Fig 2: GILZ expression in human and murine blood samples from young and aged individuals. (a) GILZ expression in human blood samples. Expression data were retrieved from the GTEx database and are shown as TPM (transcripts per million) (20–29 years: 27 ?, 57 ?; 70–79 years: 9 ?, 18 ?). (b) Representative histograms showing GILZ expression in myeloid subsets in young mice. Colored: GILZ signal; gray: isotype control. (c) Representative histograms showing GILZ signals in cells from aged and young mice. Gray line: young; colored line: aged. (d) Background-subtracted GILZ median fluorescence intensity (MFI) in myeloid blood cells from young mice (n = 14). **p < .01, ***p < .001 compared with MFI values for total CD45+ cells by ANOVA with Bonferroni's post hoc test. (e) GILZ expression in myeloid subsets of aged mice (n = 13). The GILZ MFI of young cells was set as 100% for each subset.Box plots show the 25-75th percentile (box), mean (circle), median (line), and SD (whiskers). *p < .05, **p < .01 compared with the same subset in young animals by Mann–Whitney U test. Mono: monocytes

Fig 3: GILZ expression in myeloid cells from young and aged mice. (a, b) GILZ expression in PMs from young and aged mice was determined by flow cytometry. (a) Representative histogram. (b) Background-subtracted GILZ median fluorescence intensity (MFI) (young: n = 8, aged: n = 10). (c) GILZ MFI in myeloid cells in the liver and lymphoid tissues of young mice (n = 10). ***p < .001 compared with MFI values for total CD45+ (liver) or total cells (spleen and lymph node), ## p < .01, ### p < .001 compared with all other subsets (ANOVA with Bonferroni's post hoc test. (d) GILZ expression in myeloid subsets of aged mice. The GILZ MFI of young cells was set as 100% for each subset (n = 8). *p < .05, **p < .01, ***p < .001 compared with the same subset in young animals by Mann–Whitney U test. Box plots show the 25–75th percentiles (box), mean (circle), median (line), and SD (whiskers). Mono: monocytes; Mo/Mac: monocytes and macrophages; cDC: classical dendritic cells; KC: Kupffer cells

Fig 4: GILZ knockout mimics inflammaging. (a) Serum TNF-a levels in young GILZ knockout (KO) mice were determined by ELISA and compared with serum concentrations in young and aged wild-type (WT) mice (see Figure 1). The mean of the values for young untreated WT mice was set as 1 (blue line). The red line indicates the mean TNF-a level in LPS-treated young WT mice, expressed as x-fold of untreated young WT mice. Untreated young KO: n = 12, LPS-treated young KO (5 mg/kg BW, 4 hr): n = 6. *p < .05, **p < .01 compared with equally treated young WT animals by ANOVA with Bonferroni's post hoc test. (b) PMs obtained from young and aged WT, as well as young KO mice, were stimulated with LPS (100 ng/ml, 4 hr), and TNF production was assessed by bioassay. Values for young WT mice were set as 1 (n = 4, triplicates). (c) H2O2 production was measured in PM supernatants by homovanillic acid assay (young WT and KO: n = 5, aged: n = 6). (d, e) ERK phosphorylation after LPS treatment (100 ng/ml, 15') was analyzed by Western blot. (d) One representative blot out of six is shown. Total ERK (tERK) and tubulin served as loading controls. (e) Densitometric analysis of LPS-induced ERK phosphorylation. pERK1/2 signal intensities were normalized to tERK1/2 and expressed as x-fold of young WT. Box plots show the 25–75th percentiles (box), mean (circle), median (line), and SD (whiskers). Bars show the mean ± SEM, and circles within bars indicate the mean of each experiment. *p < .05, **p < .01, ***p < .001 by ANOVA with Bonferroni's post hoc test