Inhibition of the immunoproteasome LMP2 ameliorates ischemia/hypoxia-induced blood-brain barrier injury through the Wnt/β-catenin signalling pathway

Abstract

Background: Disruption of the blood-brain barrier (BBB) after a stroke can lead to brain injury and neurological impairment. Previous work confirmed the involvement of the immunoproteasome subunit of low molecular mass peptide 2 (LMP2) in the pathophysiology of ischemia stroke. However, the relationship between the immunoproteasome LMP2 and the BBB remains unclear.

Methods: Adult male Sprague-Dawley rats were subjected to transient middle cerebral artery occlusion/reperfusion (MCAO/R). Three days before MCAO, the rats were treated with lentivirus-mediated LMP2 shRNA preparations by stereotactical injection into the ipsilateral hemispheric region. The rat brain microvascular endothelial cell (RBMVEC) line was exposed to oxygen-glucose deprivation/reperfusion (OGD/R) to mimic ischemic conditions in vitro. The RNA interference-mediated knockdown of LMP2 or β-catenin was analysed in vivo and in vitro. Analysis of the quantity of extravasated Evans blue (EB) and cerebral fluorescent angiography were performed to evaluate the integrity of the BBB. Immunofluorescence and Western blotting were employed to detect the expression of target proteins. Cell migration was evaluated using a scratch migration assay. The results of immunofluorescence, Western blotting and cell migration were quantified using the software ImageJ (Version 1.53m). Parametric data from different groups were compared using one-way ANOVA followed by the least significant difference (LSD) test.

Results: Cerebral ischemia led to lower levels of structural components of the BBB such as tight junction proteins (occludin, claudin-1 and ZO-1) in the MCAO/R group compared with the sham group (P < 0.001). However, inhibition of the immunoproteasome LMP2 restored the expression of these proteins, resulting in higher levels of occludin, claudin-1 and ZO-1 in the LMP2-shRNA group compared with the control-shRNA group (P < 0.001). In addition, inhibition of the immunoproteasome LMP2 contributed to higher microvascular density and decreased BBB permeability [e.g., the quantity of extravasated EB: LMP2-shRNA group (58.54 ± 7.37) µg/g vs. control-shRNA group (103.74 ± 4.32) µg/g, P < 0.001], and promoted the upregulation of Wnt-3a and β-catenin proteins in rats following MCAO/R. In vitro experiments, OGD/R induced marked upregulation of LMP2, proapoptotic protein Bax and cleaved caspase-3, and downregulation of occludin, claudin-1, ZO-1 and Bcl-2, as well as inhibition of the Wnt/β-catenin pathway Wnt-3a and β-catenin proteins in RBMVECs, compared with the control group under normal culture conditions (P < 0.001). However, silencing of LMP2 gene expression reversed these protein changes and promoted proliferation and migration of RBMVECs following OGD/R. Silencing of β-catenin by transfection of RBMVECs with β-catenin-siRNA aggravated the downregulation of tight junction proteins, and reduced the proliferation and migration of RBMVECs following OGD/R, compared with the control-siRNA group (P < 0.001). LMP2-siRNA and β-catenin-siRNA co-transfection partly counteracted the beneficial effects of silencing LMP2-siRNA on the levels of tight junction proteins in RBMVECs exposed to OGD/R.

Conclusion: This study suggests that inhibition of the immunoproteasome LMP2 ameliorates ischemia/hypoxia-induced BBB injury, and that the molecular mechanism involves the immunoproteasome-regulated activation of the Wnt/β-catenin signalling pathway under ischemic conditions.

Keywords: Blood–brain barrier; Cerebral ischemia; Immunoproteasome; Oxygen–glucose deprivation/reperfusion; Wnt/β-catenin pathway.

Fig. 1

LMP2 inhibition increased the levels of tight junction proteins, improved BBB integrity and upregulated Wnt/β-catenin signalling after MCAO. a Western blotting showed the levels of tight junction proteins occludin, claudin-1, ZO-1 proteins in each group rats. b Immunofluorescence staining and quantitative analysis of CD31 in brain cortex tissue from each group rats. c FITC-dextran angiographic micrographs and quantitative analysis of FITC in brain cortex tissue from each group rats. d Representative Western blotting images of Wnt-3a and β-catenin proteins in brain cortex tissue from each group rats. Data are presented as the mean ± standard deviation (SD) from three independent experiments (each group rats n = 4). *P < 0.001, compared with the sham group; ^#P < 0.001, compared with the control-shRNA group. All scale bars = 50 µm

Fig. 2

Expression of LMP2 protein in RBMVECs exposed to OGD/R. a Immunofluorescence staining and quantitative analysis of LMP2 in RBMVECs exposed to OGD/R. b Western blotting showed that the expression of LMP2 protein in RBMVECs in each group. Data are presented as the mean ± standard deviation (SD) from three independent experiments. *P < 0.001, compared with the normal group. All scale bars = 50 µm. OGD/R oxygen–glucose deprivation and reoxygenation. RBMVECs rat brain microvascular endothelial cells

Fig. 3

Changes to occludin, claudin-1 and ZO-1 proteins in RBMVECs after OGD/R. a Immunofluorescence staining and quantitative analysis of occludin, claudin-1 and ZO-1 proteins in RBMVECs in each group. b Representative Western blotting images of occludin, claudin-1 and ZO-1 in RBMVECs in each group. Data are presented as the mean ± standard deviation (SD) from three independent experiments. *P < 0.001, compared with the normal group. All scale bars = 50 µm. OGD/R oxygen–glucose deprivation and reoxygenation. RBMVECs rat brain microvascular endothelial cells

Fig. 4

Silencing LMP2 reversed the downregulated expression of the occludin, claudin-1 and ZO-1 proteins in RBMVECs following OGD/R. a Transfection efficiency of LMP2-siRNA was verified by RT-PCR and Western blotting. *P < 0.001, compared with the Si1 group, ^#P < 0.001, compared with the Si2 group. ^&P < 0.001, compared with the NC group. b Western blotting showed that the levels of occludin, claudin-1 and ZO-1 proteins in RBMVECs in each group. Data are presented as the mean ± standard deviation (SD) from three independent experiments. *P < 0.001, compared with the normal group; ^#P < 0.001, compared with the control-siRNA group. OGD/R oxygen–glucose deprivation and reoxygenation; NC group negative control group; RBMVECs rat brain microvascular endothelial cells

Fig. 5

Silencing LMP2 rescued the downregulation of the Wnt-3a and β-catenin proteins in RBMVECs following OGD/R. a Immunofluorescence and quantitative analysis of β-catenin protein in RBMVECs under normal condition culture and OGD/R conditions. Scale bars = 50 µm. b and c Western blotting indicated that the expression of β-catenin and Wnt-3a proteins in RBMVECs in each group. Data are presented as the mean ± standard deviation (SD) from three independent experiments. *P < 0.001, compared with the normal group; ^#P < 0.001, compared with the control-siRNA group. OGD/R oxygen–glucose deprivation and reoxygenation; RBMVECs rat brain microvascular endothelial cells

Fig. 6

Silencing LMP2 decreased the expression of apoptosis-related proteins and promoted the proliferation and migration of RBMVECs following OGD/R. a Western blotting and quantitative analysis the levels of Bcl-2, Bax and cleaved caspase3-p17 proteins in RBMVECs in each group. b Cell scratch test showed that proliferation and migration repair ability of RBMVECs in each group. Scale bars = 250 µm. Data are presented as the mean ± standard deviation (SD) from three independent experiments. *P < 0.001, compared with the normal group; ^#P < 0.001, compared with the control-siRNA group. OGD/R oxygen–glucose deprivation and reoxygenation; RBMVECs rat brain microvascular endothelial cells

Fig. 7

Silencing of β-catenin aggravated downregulation of the occludin, claudin-1 and ZO-1 proteins, and reduced the proliferation and migration of RBMVECs following OGD/R. a Western blotting showed that Si1 was the most effective siRNA for silencing endogenous β-catenin gene expression. ^&P < 0.001, compared with the NC group; *P < 0.001, compared with the Si3 group; ^#P < 0.001, compared with the Si2 group. b Representative Western blotting images of occludin, claudin-1 and ZO-1 proteins in RBMVECs in each group. c Cell scratch test showed that proliferation and migration repair ability of RBMVECs in each group. All scale bars = 250 µm. Data are presented as the mean ± standard deviation (SD) from three independent experiments. *P < 0.001, compared with the normal group; ^#P < 0.001, compared with the control-siRNA group. NC group negative control group; OGD/R oxygen–glucose deprivation and reoxygenation; RBMVECs rat brain microvascular endothelial cells

Fig. 8

Effects of combined silencing of the β-catenin and LMP2 genes on the expression of occludin, claudin-1 and ZO-1 proteins in RBMVECs following OGD/R. Western blotting and quantitative analysis of the expressions of occludin, claudin-1and ZO-1proteins in RBMVECs in each group. Data are presented as the mean ± standard deviation (SD) from three independent experiments. *P < 0.001, compared with the normal group; ^#P < 0.001, compared with the LMP2-siRNA group. OGD/R oxygen–glucose deprivation and reoxygenation. RBMVECs rat brain microvascular endothelial cells