Annexin A1 protects against cerebral ischemia-reperfusion injury by modulating microglia/macrophage polarization via FPR2/ALX-dependent AMPK-mTOR pathway

Abstract

Background: Cerebral ischemia-reperfusion (I/R) injury is a major cause of early complications and unfavorable outcomes after endovascular thrombectomy (EVT) therapy in patients with acute ischemic stroke (AIS). Recent studies indicate that modulating microglia/macrophage polarization and subsequent inflammatory response may be a potential adjunct therapy to recanalization. Annexin A1 (ANXA1) exerts potent anti-inflammatory and pro-resolving properties in models of cerebral I/R injury. However, whether ANXA1 modulates post-I/R-induced microglia/macrophage polarization has not yet been fully elucidated.

Methods: We retrospectively collected blood samples from AIS patients who underwent successful recanalization by EVT and analyzed ANXA1 levels longitudinally before and after EVT and correlation between ANXA1 levels and 3-month clinical outcomes. We also established a C57BL/6J mouse model of transient middle cerebral artery occlusion/reperfusion (tMCAO/R) and an in vitro model of oxygen-glucose deprivation and reoxygenation (OGD/R) in BV2 microglia and HT22 neurons to explore the role of Ac2-26, a pharmacophore N-terminal peptide of ANXA1, in regulating the I/R-induced microglia/macrophage activation and polarization.

Results: The baseline levels of ANXA1 pre-EVT were significantly lower in 23 AIS patients, as compared with those of healthy controls. They were significantly increased to the levels found in controls 2-3 days post-EVT. The increased post-EVT levels of ANXA1 were positively correlated with 3-month clinical outcomes. In the mouse model, we then found that Ac2-26 administered at the start of reperfusion shifted microglia/macrophage polarization toward anti-inflammatory M2-phenotype in ischemic penumbra, thus alleviating blood-brain barrier leakage and neuronal apoptosis and improving outcomes at 3 days post-tMCAO/R. The protection was abrogated when mice received Ac2-26 together with WRW4, which is a specific antagonist of formyl peptide receptor type 2/lipoxin A4 receptor (FPR2/ALX). Furthermore, the interaction between Ac2-26 and FPR2/ALX receptor activated the 5' adenosine monophosphate-activated protein kinase (AMPK) and inhibited the downstream mammalian target of rapamycin (mTOR). These in vivo findings were validated through in vitro experiments.

Conclusions: Ac2-26 modulates microglial/macrophage polarization and alleviates subsequent cerebral inflammation by regulating the FPR2/ALX-dependent AMPK-mTOR pathway. It may be investigated as an adjunct strategy for clinical prevention and treatment of cerebral I/R injury after recanalization. Plasma ANXA1 may be a potential biomarker for outcomes of AIS patients receiving EVT.

Keywords: Annexin A1; Cerebral ischemia-reperfusion injury; Endovascular thrombectomy; Formyl peptide receptor 2; Microglial/macrophage polarization; Neuroinflammation.

Fig. 1

Expression profiles of endogenous ANXA1 in EVT-treated AIS patients and tMCAO/R-injured mice. a Plasma ANXA1 levels detected by ELISA in normal control subjects (n = 12) and AIS patients treated with EVT (n = 23). Data were presented as median and IQR, and were analyzed by Mann–Whitney U test. **p < 0.01, and ***p < 0.001. b Plasma ANXA1 levels or ∆ANXA1 levels in EVT-treated AIS patients with favorable (n = 11) and unfavorable (n = 12) clinical outcomes. ∆ANXA1: ANXA1 concentration on 2–3 days post-EVT minus pre-EVT ANXA1 concentration. Data were presented as median and IQR and were analyzed by Mann–Whitney U test. ns, not significant. **p < 0.01. c Spearman correlation coefficient analyses of correlation between plasma ANXA1 levels or ∆ANXA1 levels in AIS patients and 3-month mRS scores. d ELISA analyses of the expressions of ANXA1 in peripheral blood of tMCAO/R-injured mice. Data were presented as the mean ± SD (n = 12/group), and were analyzed by one-way ANOVA followed by Bonferroni’s multiple comparison test. *p < 0.05, **p < 0.01, and ***p < 0.001. e Representative western blotting bands and densitometric quantifications of ANXA1 in the peri-infarct cortex after tMCAO/R. Data were presented as the mean ± SD (n = 6/group) and were analyzed by one-way ANOVA followed by Bonferroni’s multiple comparison test. **p < 0.01, and ***p < 0.001

Fig. 2

Effect of Ac2-26 on neurological function, cerebral infarct volume, and cortical CBF at 3 days post-tMCAO/R. a Schematic diagram of the experimental design. b–e Neurological function was evaluated by mNSS test (b), corner test (c), foot fault test (d), and adhesive removal test (e). f Representative photographs of nissl staining and quantitative analyses of cerebral infarct volume. g Representative photographs of laser speckle contrast imaging and quantitative analyses of cortical CBF. PU, perfusion unit. Data were presented as the mean ± SD (b–e n = 12/group; f–g n = 6/group) and were analyzed by independent samples t test. *p < 0.05, **p < 0.01, and ***p < 0.001

Fig. 3

Effect of Ac2-26 on microglial/macrophage polarization in the peri-infarct cortex at 3 days post-tMCAO/R. a–b Representative photographs of double immunostaining and quantitative analyses of microglial/macrophage polarization. M1-phenotype: CD16/32⁺ (red) and Iba-1⁺ (green); M2-phenotype: CD206⁺ (red) and Iba-1⁺ (green). Scale bar = 200 μm. c–d Representative western blotting bands and densitometric quantifications of activated microglial/macrophage marker Iba-1, M1-phenotype markers (CD16 and iNOS), and M2-phenotype markers (CD206 and Arg-1). e qRT-PCR analyses of mRNA expressions of M1-phenotype markers (CD16, CD32, and iNOS) and M2-phenotype markers (CD206, Arg-1, and YM1). f ELISA analyses of the expressions of a pro-inflammatory cytokine IL-1β (M1-phenotype) and an anti-inflammatory cytokine IL-10 (M2-phenotype). Data were presented as the mean ± SD (n = 6/group) and were analyzed by independent samples t test. *p < 0.05, **p < 0.01, and ***p < 0.001

Fig. 4

Effect of Ac2-26 on BBB disruption and neuronal apoptosis in the peri-infarct cortex at 3 days post-tMCAO/R. a Representative photographs of immunostaining and quantitative analyses of EB dye extravasation. White arrows indicated exudative EB dye. Scale bar = 100 μm. b Representative western blotting bands and densitometric quantifications of albumin extravasation and TJ (occludin and claudin-5) expressions. c–d Representative photographs of immunostaining and quantitative analyses of TJ (occludin and claudin-5; red) expression and neuronal apoptosis (NeuN, red; TUNEL, green). Scale bar = 200 μm. Data were presented as the mean ± SD (n = 6/group), and were analyzed by independent samples t test. *p < 0.05, **p < 0.01, and ***p < 0.001

Fig. 5

Effect of FPR2/ALX-dependent AMPK-mTOR pathway on Ac2-26-mediated microglial/macrophage polarization in the peri-infarct cortex at 3 days post-tMCAO/R. a Neurological function was evaluated by mNSS test. b Representative nissl staining and quantitative analyses of cerebral infarct volume. c Quantitative analyses of EB dye extravasation. d Representative western blotting bands and densitometric quantifications of activated microglial/macrophage marker Iba-1, M1-phenotype marker CD16, and M2-phenotype marker CD206. e qRT-PCR analyses of mRNA expressions of M1-phenotype markers (CD16, CD32, and iNOS) and M2-phenotype markers (CD206, Arg-1, and YM1). f ELISA analyses of the expressions of a pro-inflammatory cytokine IL-1β (M1-phenotype) and an anti-inflammatory cytokine IL-10 (M2-phenotype). g Representative western blotting bands and densitometric quantifications of phosphorylation of AMPKα and mTOR. Data were presented as the mean ± SD (A n = 12/group; b–g n = 6/group) and were analyzed by one-way ANOVA followed by Bonferroni’s multiple comparison test. *p < 0.05, **p < 0.01, and ***p < 0.001, ^#p < 0.05, ^##p < 0.01, and ^###p < 0.001

Fig. 6

Effect of FPR2/ALX-dependent AMPK-mTOR pathway on Ac2-26-mediated in vitro BV2 microglial polarization. a Schematic diagram of the experimental design. b Representative photographs of double immunostaining and quantitative analyses of BV2 microglial polarization. M1-phenotype: iNOS⁺ (green) and Iba-1⁺ (red); M2-phenotype: Arg-1⁺ (green) and Iba-1⁺ (red). Scale bar = 100 μm. c qRT-PCR analyses of mRNA expressions of M1-phenotype markers (CD32 and iNOS) and M2-phenotype markers (CD206 and Arg-1). d ELISA analyses of the expressions of a pro-inflammatory cytokine IL-1β (M1-phenotype) and an anti-inflammatory cytokine IL-10 (M2-phenotype). e Representative western blotting bands and densitometric quantifications of phosphorylation of AMPKα and mTOR. Data were presented as the mean ± SD (n = 6/group) and were analyzed by one-way ANOVA followed by Bonferroni’s multiple comparison test. *p < 0.05, **p < 0.01, and ***p < 0.001, ^#p < 0.05, ^##p < 0.01, and ^###p < 0.001

Fig. 7

Effect of Ac2-26-mediated microglia polarization on post-OGD/R HT22 neuronal apoptosis. a Schematic diagram of the experimental design. b Quantitative analyses of treated HT22 neuronal death using LDH release assay. c Representative photographs of immunostaining and quantitative analyses of HT22 neuronal apoptosis (TUNEL, green). Scale bar = 100 μm. d Representative western blotting bands and densitometric quantifications of cleaved-caspase-3. CL: cleaved. Data were presented as the mean ± SD (n = 6/group) and were analyzed by one-way ANOVA followed by Bonferroni's multiple comparison test. **p < 0.01, and ***p < 0.001, ^###p < 0.001