A novel IL-1RA-PEP fusion protein alleviates blood-brain barrier disruption after ischemia-reperfusion in male rats

Male Sprague-Dawley rats (250–290 g) were purchased from Vital River Co., Ltd. (Beijing, China). All rats were fed at the standard laboratory animal facility (25 °C, 12-h light/dark cycle) with ad libitum access to food and water for at least 2 weeks before the study.

Male Sprague-Dawley rats were subjected to transient MCAO, as reported previously [23]. Briefly, rats were anesthetized with 4% isoflurane (NO₂/O₂, 70%/30%). A nylon filament, 2 cm in length and diameter (φ) of 0.26 mm, was inserted into the right internal carotid artery (ICA) to occlude the origin of the right middle cerebral artery. After 2 h of MCAO, reperfusion was accomplished by withdrawing the filament. The body temperature of rats was maintained at 37.0 ± 0.5 °C throughout the procedure using a temperature control system. All rats were operated by the same surgeon in similar conditions to reduce variability.

The rats were randomly assigned to four groups. In the IL-1RA-PEP group, IL-1RA-PEP (a recombinant protein comprising fused IL-1RA and PEP-1, described in our previous study [24]) was diluted with saline and administered intravenously (50 mg/kg) 2 h after transient MCAO followed by reperfusion. In the vehicle group, saline was administered in the same manner. In the IL-1RA group, IL-1RA (50 mg/kg, a recombinant human protein prepared in our previous study [24]) was administered intravenously 2 h after transient MCAO followed by reperfusion. In the control group, rats underwent the same operation, without occlusion of the middle cerebral artery.

According to previously described methods [25], Evans blue (EB) dye (Sigma-Aldrich, St. Louis, MO, USA, 4% in saline, 3 mL/kg) was injected into the tail vein 22 h after transient MCAO. Two hours after injection of the EB dye, rats were anesthetized and transcardially perfused with 250–300 mL saline at room temperature. The brains of the rats were then removed, weighed, cut up, and soaked in formamide (1 mL/100 mg) at 55 °C for 24 h. Supernatants were obtained by centrifugation at 10,000×g for 10 min at 4 °C. The EB content in the tissue samples was measured as the level of fluorescence at 620 nm. It was then quantified using a linear regression standard curve derived from eight concentrations (0–10⁴ ng/mL) of the dye and expressed as μg/g of tissue.

Transmission electron microscopy (TEM) was used to determine BBB ultrastructure [26]. Twenty-four hours after ischemia-reperfusion injury, rats were anesthetized and perfused transcardially with 2% glutaraldehyde in 0.1 mol/L phosphate buffer. Approximately 1 mm³ of the ischemic penumbra of the cortex was taken and fixed in freshly prepared 3% glutaraldehyde overnight at 4 °C and post-fixed in 1% osmium tetroxide for 2 h. The specimens were then dehydrated through a graded series of ethanol and embedded in Epon 812. Ultrathin sections of the cortex were obtained and doubly stained with uranyl acetate and lead citrate. Sections were then examined using TEM (JEOL-1011, JEOL, Tokyo, Japan).

Leakage of endogenous IgG was also used to assess BBB disruption, according to previously described methods, with minor modifications [27]. Rats were anesthetized with isoflurane and then perfused transcardially with 0.9% saline, followed by 4% paraformaldehyde (PFA) in 0.1 mol/L phosphate buffer (PB, pH = 7.4). The brains of the rats were removed after 20 min perfusion fixation and then immersed in 4% paraformaldehyde overnight at 4 °C. The brains were then cryoprotected with 30% sucrose in 0.1 mol/L PB for 24 h and then sliced into 20-μm coronal sections in a cryostat. Coronal sections were stored at − 80 °C until further use.

Slides with the sections were completely air dried, treated with 0.01 mol/L sodium citrate buffer (pH = 6.0) for antigen retrieval, and washed three times with phosphate-buffered saline (PBS). After blocking with 10% normal goat serum (Vector Laboratories, Burlingame, CA, USA) at room temperature for 1 h, the sections were incubated overnight at 4 °C with Cy® 3 conjugated goat anti-rat IgG antibody (1:300, Molecular Probes, Eugene, OR, USA) under protection from light. After three rinses with PBS, 4′,6-diamidino-2-phenylindole (DAPI) (molecular probes) was applied to stain all the nuclei. The brain sections were mounted, coverslipped, and photographed under a Nikon A1 confocal on a Ti-E microscope (Nikon, Sola, Sweden).

For quantitation of BBB permeability, another set of rats subjected to ischemia-reperfusion injury was injected with fluorescein isothiocyanate-dextran (FITC-dextran) (40 kDa, 1 mg/kg body weight, Sigma) in the tail vein 1 h before sacrifice. In the same manner as described above, frozen serial coronal brain sections were prepared, stained with DAPI, and visualized directly under a fluorescent microscope. Images were captured, and distribution of the FITC-dextran tracer was calculated using ImageJ software.

Immunohistochemical staining was also used to detect penetration of IL-1RA and IL-1RA-PEP in the brain. Coronal frozen sections (20 μm) were produced in the same manner as described above. The sections were washed with PBS. They were then blocked with 10% normal rabbit serum for 30 min and incubated overnight at 4 °C with goat anti-human IL-1RA antibody (1:500, R&D Systems, Inc., Minneapolis, USA) and rabbit anti-NeuN antibody to label neuronal cells (1:100, Chemicon, Hampshire, UK). Brain sections were washed and then incubated with appropriate fluorochrome-conjugated secondary antibodies (Alexa Fluor 488 or Cy® 3, Invitrogen, Carlsbad, CA, USA) for 1 h at room temperature. The sections were counterstained with DAPI. Images were photographed with laser scanning confocal fluorescence microscopy (Zeiss LSM780, Carl Zeiss, Jena, Germany).

Brain coronal frozen sections (20 μm) were produced in the same manner as described above. The sections were blocked with 10% normal goat serum (Vector Laboratories, Burlingame, CA, USA) at room temperature for 1 h and then incubated with primary antibodies overnight at 4 °C. The primary antibodies used were as follows: rabbit anti-occludin (1:50, Proteintech Group Inc., Rosemont, USA), rabbit anti-ZO-1 (1:50, Proteintech Group Inc.), rabbit anti-claudin-5 (1:200, Affinity Biosciences, OH, USA), and mouse anti-vWF to label endothelial cells (1:50, Millipore, Temecula, CA, USA); or rabbit anti-MMP-2 (1:100, Abcam Inc., Cambridge, MA, USA), rabbit anti-MMP-9 (1:1000, Cell Signal Technology, Boston, USA), and mouse anti-GFAP to label astrocyte cells (1:100, Cell Signal Technology). Brain sections were rinsed with PBS and then incubated with Cy® 3 conjugated goat anti-rabbit IgG antibody (1:500, Invitrogen, Carlsbad, CA, USA) and Alexa Fluor 488 conjugated goat anti-mouse IgG antibody (1:300, Molecular Probes) for 1 h at room temperature. All nuclei were stained with DAPI (molecular probes). After being washed, the sections were observed under a laser scanning confocal microscope (FV10i, Olympus, Tokyo, Japan).

Rats were deeply anesthetized and decapitated 24 h after ischemia-reperfusion. Their brains were immediately resected, and the ischemic cerebral cortices were separated and stored at − 80 °C until further use.

Total RNA was extracted from frozen ischemic cerebral cortices. The RNA was reverse-transcribed and amplified using reverse transcriptase (RT)-PCR (7900 Real-time PCR System, Life Technologies, USA). The primers of vascular endothelial growth factor (VEGF), angiopoietin-1 (Ang-1), and glyceraldehyde 3-phosphate dehydrogenase (GAPDH) were designed as follows: VEGF, forward primer: 5′-CAC CAA AGC CAG CAC ATA GG-3′, reverse: 5′-TTT AAC TCA AGC TGC CTC GC-3′; Ang-1, forward: 5′-TGA TGG ACT GGG AAG GGA AC-3′, reverse: 5′-CAC AGG CAT CAA ACC ACC AA-3′; GAPDH, forward: 5′-AAG ATG GTG AAG GTC GGT GT-3′, reverse: 5′-TGA CTG TGC CGT TGA ACT TG-3′. The ΔΔCt values from each group were analyzed, and mRNA expression levels were normalized to 2-ΔΔCt. Expression of the specific genes of interest was compared across groups.

The isolated ischemic cerebral cortices were homogenized in ice-cold radioimmunoprecipitation assay (RIPA) lysis buffer (CWBiotech, Beijing, China), containing a protease inhibitor cocktail (Roche, Indianapolis, IN, USA) for the extraction of protein samples. Protein concentration was determined using the bicinchoninic acid (BCA) method. Equal amounts of proteins (100 μg) were separated on SDS-polyacrylamide gels at 100 V for 80–120 min and transferred to polyvinylidene difluoride (PVDF) membranes (Millipore Corporation, Billerica, MA, USA) at 250 mA for 100 min. The membranes were blocked with 5% nonfat dried milk in tris-buffered saline Tween-20 (TBST) for 1 h at room temperature and then incubated overnight at 4 °C with primary antibodies against occludin, ZO-1, MMP-2, MMP-9 (1:1000, Proteintech Group Inc.), claudin-5 (1:500, Affinity Biosciences), VEGF, Ang-1 (1:1000, Abcam Inc.), and GAPDH (1:1000, CWBiotech). The membranes were washed and then incubated with the respective horseradish peroxidase-conjugated secondary antibodies (1:2500, ZSGB-BIO, Beijing, China) for 1 h at 37 °C. Immunoreactive bands were visualized using a chemiluminescent detection system kit (Millipore Corporation, Billerica, MA, USA). Band intensities were measured with the ImageJ software (National Institute of Health [NIH], Bethesda, MD, USA).

After the NF-κB inhibitor, JSH-23 (50 mg/mL) (Selleck Chemicals, USA), was added, ZO-1, MMP-9, Ang-1, and VEGF were again detected by the methods described above.

All data were analyzed with the statistical software, GraphPad Prism 5 (La Jolla, CA, USA) and presented as mean ± SEM. The parametric data of two groups were analyzed using the Student’s t test, and Welch’s correction was applied for data with unequal variances. All differences in other data between groups were assessed by one-way ANOVA analysis, followed by the Bonferroni test for multiple comparisons. Statistically significant, very significant and highly significant differences were determined by P < 0.05, P < 0.01, and P < 0.001, respectively.

The experimental protocol is presented in Fig. 1a. Extravasation of Evans blue, the BBB ultrastructure, leakage of endogenous IgG, and permeability of FITC-dextran tracer in the brain were all used to evaluate BBB breakdown 24 h after ischemia-reperfusion.

Evans blue extravasation in the vehicle group was notably exacerbated, in comparison to the control group, whereas extravasation of Evans blue in IL-1RA-PEP (50 mg/kg)-treated rats was significantly alleviated, in comparison to the vehicle group (6.79 ± 0.88 versus 15.12 ± 1.39 μg/g, P < 0.001). Extravasation was also significantly reduced in the positive IL-1RA (50 mg/kg) group (11.00 ± 0.41 μg/g), in comparison to the vehicle group. Moreover, extravasation in the IL-1RA-PEP group was significantly lower than that in the IL-1RA group (P < 0.05) (Fig. 1b, c).

Ultrastructural changes of the BBB were examined in the different groups (Fig. 1d). The complete basement membrane and tight junctions were observed in the control group. After the ischemia-reperfusion injury, the basement membrane was disrupted and tight junctions were damaged in the vehicle group. However, in the IL-1RA-treated and IL-1RA-PEP-treated groups, the basement membrane was preserved and the tight junctions were slowly restored.

In terms of endogenous IgG leakage, 24 h after MCAO, leakage in the vehicle group was more obvious than that in the control group. However, the leakage of endogenous IgG in the right cerebral hemisphere of IL-1RA-PEP-treated rats was very significantly brought down, in comparison to that in the vehicle rats (7.58% ± 1.62 versus 34.21% ± 2.08%, P < 0.001). Leakage (19.65% ± 3.16%) was also significantly reduced in the IL-1RA-treated group in comparison to the vehicle group. Furthermore, leakage in IL-1RA-PEP-treated rats was significantly lower than that in IL-1RA-treated rats (P < 0.05) (Fig. 2a, b).

The FITC-dextran permeability assay was performed to examine and quantify the changes in BBB integrity 24 h after ischemia-reperfusion injury. As shown in Fig. 2a, c, in comparison to control rats, cerebrovascular permeability to FITC-dextran in vehicle rats was significantly aggravated. The permeability of FITC-dextran was attenuated by post-ischemic treatment with IL-1RA-PEP (50 mg/kg), in comparison to the vehicle group (28.27% ± 2.84 versus 52.40% ± 3.57%, P < 0.001). Furthermore, in comparison to the vehicle group, permeability was significantly ameliorated by treatment with IL-1RA (50 mg/kg) (31.87% ± 2.45%) (P < 0.01).

The permeation effects of IL-1RA and IL-1RA-PEP on the brain of rats were analyzed by immunohistochemical technology. In rats subjected to MCAO that had been intravenously administered IL-1RA (50 mg/kg) or IL-1RA-PEP (50 mg/kg), a low concentration of IL-1RA (Fig. 3a) and a high concentration of IL-1RA-PEP were detected in the brain (Fig. 3b), respectively. Moreover, the magnification of double immunofluorescence staining indicated that IL-1RA only accumulated on the cell surface of neurons (Fig. 3c), but IL-1RA-PEP was able to both bind to IL-1R and penetrate neurons (Fig. 3d). In comparisons of the positive staining regions between the cerebral cortex and striatum, the staining intensity of IL-1RA-PEP that reached the brain was significantly higher than that of IL-1RA (in the striatum of MCAO rats, 13.05% ± 2.11% versus 6.82% ± 1.38%, P < 0.05) (Fig. 3e).

The permeation of IL-1RA and IL-1RA-PEP into brain tissues was analyzed by ELISA, 4 h after intravenous administration in rats subjected to MCAO. In comparisons of IL-1RA concentration per microgram of protein in the ischemic brain hemisphere, the concentration of IL-1RA-PEP that reached brain tissues (the ipsilateral cerebral cortex and striatum area) was significantly higher than that of IL-1RA (in the striatum of rats subjected to MCAO, 14.50 ± 3.20 versus 3.09 ± 0.60 ng mL⁻¹, P < 0.05) (Fig. 3f).

These results show that IL-1RA-PEP possessed an enhanced capacity to penetrate brain tissue in comparison to IL-1RA, under conditions of ischemia-reperfusion injury.

The TJ proteins, ZO-1, occludin, and claudin-5 are crucial proteins that are involved in the process of BBB breakdown following transient MCAO in rats [28]. Given this finding, the expression and localization of the TJ proteins ZO-1, occludin, and claudin-5 in brain tissue were investigated. In comparison to the control group, 24 h after ischemia-reperfusion, levels of ZO-1, occludin, and claudin-5 were significantly decreased in the vehicle group. However, in comparison to the vehicle group, the IL-1RA-PEP group showed a marked increase in these protein levels (ZO-1, 0.52 ± 0.04 versus 0.34 ± 0.03, P < 0.05; occludin, 0.77 ± 0.03 versus 0.55 ± 0.04, P < 0.05; claudin-5, 1.92 ± 0.13 versus 1.07 ± 0.12, P < 0.05) (Fig. 4a–e). Moreover, IL-1RA-PEP significantly enhanced the reconstitution of ZO-1, occludin, and claudin-5 proteins in cerebral tissues, following transient MCAO (ZO-1, 79.13% ± 12.98% versus 34.26% ± 6.19%, P < 0.05; occludin, 80.24% ± 7.72% versus 14.72% ± 4.44%, P < 0.01; claudin-5, 45.59% ± 5.41% versus 12.20% ± 6.33%, P < 0.05) (Fig. 4f–i). In addition, rats administered IL-1RA showed little improvement in the expression levels and distribution of these proteins, in comparison to rats in the vehicle group. Nevertheless, the expression of claudin-5 (1.18 ± 0.16) and redistribution of occludin (40.02% ± 11.51%) in rats administered IL-1RA were notably lower than those in rats administered IL-1RA-PEP (P < 0.05). The data indicate that MCAO-induced changes in characteristics of the TJ proteins were significantly restored by IL-1RA-PEP.

Studies have shown that the expression and localization of MMPs strongly influence the integrity of the BBB [29]. In view of this, the effects of IL-1RA and IL-1RA-PEP on the expression and localization of MMPs in the brain, following transient MCAO were investigated. The levels of MMP-9 and MMP-2 in cerebral tissues of the vehicle group were markedly increased in comparison to the control group. Changes in these protein levels were attenuated by post-ischemic treatment with IL-1RA-PEP (MMP-9, 0.34 ± 0.09 versus 1.41 ± 0.22, P < 0.01; MMP-2, 0.41 ± 0.11 versus 1.26 ± 0.13, P < 0.05) (Fig. 5a–c). Additionally, transient MCAO-induced redistribution of MMP-9 and MMP-2 proteins in cerebral tissues of the IL-1RA-PEP-treated group was significantly suppressed in comparison to the vehicle group (MMP-9, 175.0% ± 24.7% versus 441.7% ± 59.4%, P < 0.05; MMP-2, 146.7% ± 20.33% versus 436.0% ± 31.7%, P < 0.01) (Fig. 5d–f). These results show that MCAO-induced transient changes in MMP characteristics were alleviated by IL-1RA-PEP, which might contribute to the BBB protection of IL-1RA-PEP.

Blood vessels comprise an essential component of the structure of the BBB. After MCAO induces breakdown of the BBB, angiogenesis occurs [30]. Angiopoietin-1 and VEGF are important factors associated with the process of angiogenesis. Western blot and real-time quantitative PCR analyses were applied to detect the expression of angiopoietin-1 and VEGF in cerebral tissues after transient MCAO. In comparison to the control group, occlusion of the middle cerebral artery remarkably decreased mRNA and protein expression of angiopoietin-1 and notably increased mRNA and protein expression of VEGF. Conversely, in the IL-1RA-PEP treatment group, a marked elevation in angiopoietin-1 expression and reduction in VEGF expression was observed, in comparison to the vehicle group (angiopoietin-1 mRNA, 5.09 ± 0.27 versus 1.03 ± 0.18, P < 0.001; angiopoietin-1 protein, 0.41 ± 0.06 versus 0.20 ± 0.04, P < 0.05; VEGF mRNA, 0.76 ± 0.02 versus 1.00 ± 0.04, P < 0.05; VEGF protein, 0.19 ± 0.04 versus 0.56 ± 0.04, P < 0.001). In addition, IL-1RA treatment had a profound influence on the expression of angiopoietin-1 and VEGF (Fig. 6).

The JSH-23 inhibitor of p65 phosphorylation was used to block the p65/NF-κB pathway. In comparison to the vehicle group, the expression of ZO-1, MMP-9, and angiopoietin-1 in the vehicle + JSH-23 group was significantly altered 24 h after ischemia-reperfusion injury (Fig. 7). The data indicate that the NF-κB pathway significantly regulated the function of the TJ proteins and angiogenic factors.

Moreover, in comparison to the IL-1RA-PEP group, 24 h after transient MCAO, the expression of ZO-1, MMP-9, angiopoietin-1, and VEGF in the IL-1RA-PEP + JSH-23 group was significantly changed (Fig. 7). The data show that after administration of the p65/NF-κB inhibitor, the effects of IL-1RA-PEP on ZO-1, MMP-9, angiopoietin-1, and VEGF were altered.

Therefore, our results indicate that the p65/NF-κB signaling pathway was involved in IL-1RA-PEP-mediated protection against BBB disruption, induced by ischemia-reperfusion injury.