FTY720 Protects Against Ischemia-Reperfusion Injury by Preventing the Redistribution of Tight Junction Proteins and Decreases Inflammation in the Subacute Phase in an Experimental Stroke Model

Abstract

Injury due to brain ischemia followed by reperfusion (I/R) may be an important therapeutic target in the era of thrombectomy. FTY720, a widely known sphingosine-1-phosphate receptor agonist, exerts various neuroprotective effects. The aim of this study was to examine the protective effect of FTY720 with respect to I/R injury, especially focusing on blood-brain barrier (BBB) protection and anti-inflammatory effects. Male rats were subjected to transient ischemia and administered vehicle or 0.5 or 1.5 mg/kg of FTY720 immediately before reperfusion. Positron emission tomography (PET) with [¹⁸F]DPA-714 was performed 2 and 9 days after the insult to serially monitor neuroinflammation. Bovine and rat brain microvascular endothelial cells (MVECs) were also subjected to oxygen-glucose deprivation (OGD) and reperfusion, and administered FTY720, phosphorylated-FTY720 (FTY720-P), or their inhibitor. FTY720 dose-dependently reduced cell death, the infarct size, cell death including apoptosis, and inflammation. It also ameliorated BBB disruption and neurological deficits compared to in the vehicle group. PET indicated that FTY720 significantly inhibited the worsening of inflammation in later stages. FTY720-P significantly prevented the intracellular redistribution of tight junction proteins but did not increase their mRNA expression. These results suggest that FTY720 can ameliorate I/R injury by protecting the BBB and regulating neuroinflammation.

Keywords: Apoptosis; Blood–brain barrier; Brain ischemia–reperfusion; FTY720; Inflammation; Sphingosine-1-phosphate receptor.

Fig. 1

Effect of FTY720 on infarct size, mortality ratio, and neurological recovery in a rat model of transient middle cerebral artery occlusion. a Animals (vehicle, n = 20; 0.5 mg/kg, n = 6; 1.5 mg/kg, n = 6) were sacrificed at day 7 after transient middle cerebral artery occlusion and 2,3,5-triphenyltetrazolium chloride staining of coronal brain sections was performed to analyze the efficacy of FTY720 for reducing the infarct size. High-dose FTY720 (1.5 mg/kg) significantly reduced the infarct size compared to that in the vehicle group (††P < 0.01 versus 1.5 mg/kg). b FTY720 showed a dose-dependent improvement in mortality ratio (vehicle, n = 36; 0.5 mg/kg, n = 19; 1.5 mg/kg, n = 12). c, d Neurological function was assessed by the modified Neurological Severity Score (c) and modified Bederson score (d) in the vehicle, low-dose FTY720, and high-dose FTY720 groups (vehicle, n = 24; 0.5 mg/kg, n = 12; 1.5 mg/kg, n = 12). Vehicle group, blue; 0.5 mg/kg group, red; 1.5 mg/kg, green. Significant differences between the vehicle and treatment groups. Values are the mean; *P < 0.05 versus 0.5 mg/kg, †P < 0.05 versus 1.5 mg/kg, **P < 0.01 versus 0.5 mg/kg, ††P < 0.01 versus 1.5 mg/kg

Fig. 2

Immunohistochemistry staining (day 7) and TUNEL staining (day 3) of transient middle cerebral artery occlusion rat model showing the extent of inflammation and cell death including apoptosis in the peri-infarct areas. Iba-1-positive activated microglia (a) and macrophages (b) were dose-dependently decreased in the FTY720 groups (n = 6 per group); *P < 0.05 versus 0.5 mg/kg, ††P < 0.01 versus 1.5 mg/kg. FTY720 significantly reduced the number of cell death including apoptotic cells in the FTY720-treated groups (n = 6 per group); **P < 0.01 versus 0.5 mg/kg, ††P < 0.01 versus 1.5 mg/kg. Scale bar represents 25 μm. Higher magnification figure on the right upper corner (× 5)

Fig. 3

Serial PET imaging of inflammation in transient middle cerebral artery occlusion rat model. a Left figures show representative PET images (yellow: ischemic area; and brown: healthy area) (n = 4 per group), 3,5-triphenyltetrazolium chloride (TTC) staining (red represents healthy area and white represents ischemic area), and autoradiography (ARG; black represents ischemia and gray represents healthy area) of rat brains. Color bars indicate the scale of standardized uptake values (SUVs) in PET images (right) and qualitative scale in ARG (left). b Graphs show the max SUV (left) and the mean SUV (right); †P < 0.05 versus 1.5 mg/kg. Vehicle-treated animals showed a significant increase in inflammation on day 9 compared to that on day 2, whereas FTY720 suppressed this effect. Particularly, in the high-dose group (1.5 mg/kg), nearly no exacerbation of inflammation was observed

Fig. 4

Effect of FTY720 on blood–brain barrier (BBB) permeability in the brain of ischemia–reperfusion (I/R rats). After middle cerebral artery occlusion (MCAO)/reperfusion, Evans blue, 10-kD fluorescent dextran (FITC dextran), or 3-kD FITC dextran was infused by i.v. injection at day 1. The brains were isolated, and the supernatant was measured at 620 nm for absorbance using a spectrophotometer. Representative pictures and the spectrophotometry optical density of Evans blue was compared between groups (a) (n = 6 per group). Immunohistochemistry staining of brain microvessels (red) and leaked FITC (green) were compared (b; 10-kD FITC dextran, c; 3-kD FITC dextran). Scale bar represents 50 μm. **P < 0.01 versus 0.5 mg/kg, ††P < 0.01 versus 1.5 mg/kg

Fig. 5

Effect of FTY720/FTY720-P on BBB permeability in in vitro ischemia model. Relative expression levels were examined in a Evans blue, b 10-kD FITC dextran, and c 3-kD FITC dextran. OGD condition (vehicle) disrupted BBB permeability compared to control. PTX appears to abolish the effect of FTY720-P. However, FTY720 or FTY720-P showed no protective role compared to OGD. FTY720 rather showed increased permeability in Evans blue and 10-kD FITC dextran compared to OGD. *P < 0.05 vehicle versus FTY720, †P < 0.05 vehicle versus FTY720-P + PTX

Fig. 6

mRNA expression of tight and adherens junctional proteins in transient middle cerebral artery occlusion rat model. mRNA expression for ZO-1 (a), Occludin (b), Claudin-5 (CLDN5) (c), and VE-cadherin (d) were evaluated. Ischemia–reperfusion (I/R) injury resulted in decreased mRNA expression; however, FTY720 and FTY720P did not restore the expression of mRNA

Fig. 7

Translocation of tight and adherens junction proteins in transient middle cerebral artery occlusion rat model (a ZO-1, b VE-cadherin, c occludin, d Claudin-5). Whereas ZO-1 and VE-cadherin were distributed at the lamellipodia under physiological conditions (a, b; control, arrow), disruption and retraction of these proteins into the cytoplasm was observed after ischemia–reperfusion (I/R) injury (a, b; vehicle, arrowhead). While FYT720 groups (a, b; FTY720, arrowhead) showed the same results as the vehicle group, specifically that junctional proteins were not preserved at the lamellipodia, FTY720-P successfully maintained the junctional proteins similar to that observed in the control group (a, b; FTY720-P; arrow). This effect was completely abolished by adding the S1P₁ receptor antagonist, pertussis toxin (PTX) (a, b; FTY720-P+PTX, arrowhead). These protective roles of FTY720P was not observed in the occludin (c) and claudin-5 (d). Scale bar = 10 μm. *P < 0.05 versus vehicle group, **P < 0.01 versus control group, ††P < 0.01 versus PTX+FTY720-P group

Fig. 8

Western blot analysis was performed to elucidate the downstream signaling of FTY720-P; ERK1/2 signaling was significantly increased in the FTY720-P group and completely abolished by PTX treatment (n = 3 per group); *P < 0.05 versus vehicle group, **P < 0.01 versus control group, ††P < 0.01 versus PTX+FTY720-P group