Fig. 2: Lyve1⁺Tim4⁺ ATMs are resident and persist during obesity.

a, b Gating strategy used to define F4/80^highTim4⁺ (blue), F4/80^highTim4⁻ (cyan), and F4/80^low (green) macrophage populations in AT (a) and histogram of the fluorescence intensity of Lyve1, MHCII, CD206, RELMα, and CSF1R with fluorescence minus one (FMO) in black (b). c Hosts were partially irradiated (limbs) and reconstituted with Ccr2^+/+ (Wild Type (WT)) or Ccr2^−/− mice. Non-host chimerism (%) among Ly6C^high monocytes, eosinophils, and F4/80⁺ macrophages in the epididymal AT, 8 weeks post reconstitution with WT (circle) or Ccr2^−/− (triangle) BM. d Non-host chimerism among ATM subsets, normalized to Ly6C^high blood monocyte non-host chimerism. e–j Hosts were partially irradiated (head and forelimbs) and reconstituted with WT BM. After recovery, animals were put on CD (solid bar) or HFD (stripped bar) for 8 weeks (e). Non-host chimerism was normalized to Ly6C^high blood monocyte chimerism. Number and non-host chimerism of the whole F4/80⁺ ATM populations (f) and proportions (%) of ATMs in epididymal AT (g). Non-host chimerism among ATM subsets. i Total number of ATMs (solid) and BM-derived ATMs (squared pattern). Statistical analyses were performed to compare total ATMs in CD vs HFD (*) and total ATMs vs BM-derived ATMs in HFD (^#) (h). ATM proliferation measured by percentage of Ki-67⁺ cells (j). Data pooled from n = 10 mice per groups from 2 to 3 independent experiments. Error bars show SEM. Kruskal–Wallis test with Dunn’s multiple comparisons test or ANOVA with Sidak’s multiple comparisons test were applied after assessing normality using D’Agostino and Pearson Normality test. Significant differences are indicated by *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001, ^##P < 0.01, ^####P < 0.0001, ns = non-significant.