Eosinophils are key regulators of perivascular adipose tissue and vascular functionality

Figures

• 

  Figure 1. Obese mice have impaired vascular reactivity. Age-matched WT and HFD WT mice were analyzed for vascular reactivity and adipose tissue alterations. (a) The anti-contractile effect of PVAT observed in WT mice is lost in HFD mice (mean ± SEM from one experiment; n = 8 (WT) and 5 (HFD); P = NS, two-way ANOVA) (no PVAT (-■-) and PVAT (-□-)). (b) Representative H&E staining of mesenteric adipose tissue from control and HFD mice, and (c) analysis of adipocyte size (HFD: n = 5, and WT: n = 8; ***P < 0.0001, Student’s t-test). Scale bar shows 100 μm. (d) Mesenteric adipose tissue of HFD and control mice was analyzed for number of eosinophils (mean ± SEM; HFD: n = 5, and WT: n = 8; *P = 0.0113, Student’s t-test). (e) Representative flow cytometric plots of mesenteric adipose tissue eosinophils from control and HFD mice.

• 

  Figure 2. Eosinophil-deficient mice have impaired vascular reactivity which is restored by in vivo reconstitution with eosinophils but has no effect on other immune cell populations. (a) Flow cytometric analysis of mesenteric adipose tissue eosinophils from WT, ∆dblGATA-1 and eosinophil-reconstituted (AdBac) mice (n = 11, 6 and 5, respectively). (b) Eosinophil-deficient mice have impaired vascular reactivity. NE-induced constriction of arteries from WT and ΔdblGATA-1 mice (n = 15, data pooled from 3 experiments; ***P < 0.0001, two-way ANOVA) (ΔdblGATA-1 no PVAT (-•-), ΔdblGATA-1 PVAT (-ο-), WT no PVAT (-■-) and WT PVAT (-□-). (c) The PVAT-mediated anti-contractile response of eosinophil-deficient ΔdblGATA-1 mice (no PVAT (-•-) and PVAT (-ο-)) to NE-induced constriction is restored upon eosinophil-reconstitution (AdBac PVAT (-◊-) vs. AdBac no PVAT (-♦-) mean ± SEM; n = 10; P = 0.0023, two-way ANOVA). (d) Flow cytometry quantification of SMA+ contractile pericytes (mean ± SEM frequency pooled from 4 individual experiments; n = 6, P = NS, student’s t-test) showed no significant differences between WT (■) and ∆dblGATA-1 mice (-•-) in mesenteric adipose tissue, and (e) mesenteric adipose weight…

• 

  Figure 3. Exogenous application of purified eosinophils induces vessel relaxation. Isolated splenic eosinophils were added to mesenteric arteries preconstricted with norepinephrine (10−5 M). Relaxation was measured by wire myography. (a) Exogenous application of eosinophils induced dose-dependent relaxation of ΔdblGATA-1 arteries in the presence (-ο-) but not absence (-•-) of PVAT (n = 8, pooled from 8 individual experiments; ***P < 0.0001, two-way ANOVA). (b) Trace representative of a pre-constricted ΔdblGATA-1 artery ± PVAT responding to 10,000 eosinophils. Note the rapid drop in constriction following addition of eosinophils (dotted line). (c) Solution transfer experiments to assess generation of a transferable factor that mediates relaxation (n = 10, pooled from 2 experiments; *P = 0.049, Student’s t-test. (d) Vessel relaxation in response to application of NE-stimulated eosinophils (−Fil) or filtered supernatant from NE-stimulated eosinophils (+Fil), by passing through a 0.2 μm filter to retain eosinophils, to pre-constricted WT mesenteric arteries in the presence of PVAT. (e) Addition of NE-stimulated (10−5 M) eosinophil supernatant or unstimulated eosinophil supernatant supplemented…

• 

  Figure 4. Eosinophils induce adiponectin- and NO-dependent PVAT-induced vessel relaxation. The functional role of adiponectin and NO were investigated by wire myograph analysis of mesenteric vessels of ΔdblGATA-1 mice using pharmacological tools and in eosinophil-reconstituted ΔdblGATA-1 mice. (a,b) Vessel relaxation in response to exogenous application of 10,000 NE-stimulated eosinophils (eos) to preconstricted ΔdblGATA-1 mesenteric arteries + PVAT following incubation of (a) PVAT (L-NMMA, anti-IL-4: n = 9, ABP: n = 8; *P < 0.05 and **P < 0.01, one-way ANOVA with post hoc Dunnett’s), or (b) PVAT and eosinophils (L-NMMA, ABP, anti-IL-4: n = 9; **P < 0.01, ***P < 0.001, one-way ANOVA with post hoc Dunnett’s). (c,d) The PVAT anti-contractile effect of NE constricted arteries from ΔdblGATA-1 mice reconstituted with (c) adiponectin−/− (--), (c,d) WT (-□-), or (d) iNOS−/− (-×-) eosinophils (adipo−/− vs WT: n = 4, **P = 0.0085, two-way ANOVA; iNOS−/− vs WT: n = 5; P = NS, two-way ANOVA).

• 

  Figure 5. Eosinophils stimulate β3 adrenoreceptors through the release of soluble factors. (a) Samples of PVAT from WT mice were incubated with or without antagonists of β1-β3 (propranolol 1 μM) and, α1 (phentolamine 0.1 μM,) or β3 (SR-592,30 A 0.3 μM) adrenoreceptors and the anti-contractile response of NE-constricted arteries assessed by wire myography (β3 adrenoreceptor antagonist vs. time control: ***P < 0.001, one-way ANOVA; β3 adrenoreceptor antagonist vs. vessel control, n = 6; *P < 0.05, one-way ANOVA). (b) The β3 adrenoreceptor agonist (CL-316,243) was added to ΔdblGATA-1 arteries with PVAT and contractile responses compared with WT (P < 0.001, n = 5, one-way ANOVA with post hoc Dunnett’s). (c) Immunocytochemical analysis for tyrosine hydroxylase in purified and cytospun splenic eosinophils and IgG control staining. (d) Eosinophils were incubated with or without a tyrosine hydroxylase inhibitor (AMPT) and the vascular reactivity of the vessels in response to addition of 10,000 eosinophils tested (*p = 0.0472, n = 5, t-test). (e) Levels of epinephrine, norepinephrine and dopamine were measured from unstimulated and stimulated eosinophils (n = 4) by ELISA.