Interleukin (IL)-9 exerts a variety of functions in autoimmune diseases. However, its role in ischemic brain injury remains unknown. The present study explored the biological effects of IL-9 in ischemic stroke (IS). We recruited 42 patients newly diagnosed with IS and 22 age- and sex-matched healthy controls. The expression levels of IL-9 and percentages of IL-9-producing T cells, including CD3+CD4+IL-9+ and CD3+CD8+IL-9+ cells, were determined in peripheral blood mononuclear cells (PBMCs) obtained from patients and control individuals. We also investigated the effects of IL-9 on the blood–brain barrier (BBB) following oxygen–glucose deprivation (OGD) and the potential downstream signaling pathways. We found that patients with IS had higher IL-9 expression levels and increased percentages of IL-9-producing T cells in their PBMCs. The percentages of CD3+CD4+IL-9+ and CD3+CD8+IL-9+ T cells were positively correlated with the severity of illness. In in vitro experiments using bEnd.3 cells, exogenously administered IL-9 exacerbated the loss of tight junction proteins (TJPs) in cells subjected to OGD plus reoxygenation (RO). This effect was mediated via activation of IL-9 receptors, which increased the level of endothelial nitric oxide synthase (eNOS), as well as through up-regulated phosphorylation of signal transducer and activator of transcription 1 and 3 and down-regulated phosphorylated protein kinase B/phosphorylated phosphatidylinositol 3-kinase signaling. These results indicate that IL-9 has a destructive effect on the BBB following OGD, at least in part by inducing eNOS production, and raise the possibility of targetting IL-9 for therapeutic intervention in IS.

Introduction

It has become increasingly clear that inflammation plays an essential role in the neurological deterioration found in early stage of ischemic stroke (IS) [1]. Disruption of the blood–brain barrier (BBB) is a crucial event in IS, leading to an influx of circulating T cells and effective cytokines into the central nervous system, which contributes to neurological deterioration [2,3]. T cells, including T helper type 1 (Th1), γδT, regulatory T, and CD8+ T cells, have been reported to regulate post-ischemic inflammation in the brain [4]. Related cytokines, such as tumor necrosis factor-α, interferon-γ (IFN-γ), interleukin (IL) 17A (IL-17A) and IL-10, act as significant mediators in the delayed phase of ischemic brain injury [46] and inflammatory markers could be also used as prognostic factors for IS patient [7].

In addition to the above-mentioned cytokines, our recent study demonstrated that the expression of IL-9 is increased in an experimental stroke model in rats [8]. IL-9 has been shown to have pleiotropic functions in multiple diseases and conditions, including allergy and experimental autoimmune encephalomyelitis [9], but the effect of IL-9 in IS remains unknown. Therefore, in the current study, we investigated the expression of circulating IL-9 and the percentage of IL-9-positive (IL-9+) T cells in patients with IS of different severities. We also explored the role of IL-9 in vitro using bEnd.3 cells, a mouse brain microvascular endothelial cell line, to provide additional evidence regarding the biological function of IL-9 in IS.

Materials and methods

Ethics statement

The present study was approved by the ethics committee of the Third Affiliated Hospital of Sun Yat-sen University. Written informed consent was obtained from all the participants.

Patients and controls

A total of 42 patients with IS (22 females and 20 males; aged from 35 to 82 years), who were admitted to the Stroke Center of the Third Affiliated Hospital of Sun Yat-sen University between January 2015 and August 2015 were enrolled in this research. All patients fulfilled the following three criteria to be included in the present study: (a) onset age ≥18 years, (b) onset time ≤14 days of admission, and (c) lesions observed using diffusion-weighted imaging (DWI). The exclusion criteria were: patients (a) treated with systemic glucocorticoids or other immunosuppressive agents within 14 days of admission, or (b) those diagnosed with hemorrhagic cerebrovascular diseases, or (c) thromboembolism, collagen disease, disseminated intravascular coagulation, advanced liver disease, renal failure or malignant disease, or (d) asthma, systemic lupus erythematosus, tuberculosis, intestinal parasitic disease, or inflammatory bowel disease. Clinical and demographic characteristics of these patients were collected while they were hospitalized. Stroke severity was assessed by the National Institutes of Health Stroke Scale (NIHSS) score, and DWI in combination of the automated software program RAPID was used to estimate infarct volumes on the day of imaging. The subtypes of IS was classified according to the Trial of Org 10172 in Acute Stroke Treatment (TOAST) classification [10].

A total of 22 healthy controls (10 females and 12 males, aged from 21 to 57 years) were recruited from the physical examination center of our hospital.

Blood samples

In the patient group, blood samples were collected into vacutainers containing sodium heparin (Becton-Dickinson, U.S.A.) in the morning following admission. The blood of patients and healthy controls were subjected to centrifugation in a density gradient centrifuge to obtain peripheral blood mononuclear cells (PBMCs), which were then washed twice in Hank’s balanced salt solution. The cells were resuspended at a concentration of 1.5 × 106 cells/ml in complete RPMI 1640 medium (Gibco, U.S.A.) supplemented with 10% FBS (Sijiqing, China), 100 U/ml penicillin, 100 μg/ml streptomycin, 50 μM 2-mercaptoethanol, and 2 mM L-glutamine (all from Gibco, U.S.A.).

PBMC culture conditions

PBMCs were plated in 96-well round-bottom plates at a density of 1.5 × 106 cells/ml and stimulated for 48 h with or without PMA (20 ng/ml; Sigma–Aldrich) and ionomycin (1 μg/ml; Sigma–Aldrich). Cell-free supernatants were harvested and assessed for production of cytokines using ELISA. The cells were cultured with PMA (20 ng/ml) and ionomycin (1 μg/ml) in the presence of brefeldin A (10 μg/ml; Sigma–Aldrich) for 5 h and used for flow cytometry analysis. To investigate the effects of anti-IL-4 and anti-transforming growth factor-β1 (TGF-β1) on the IL-9 production of PBMCs obtained from patients, cells were pretreated with PMA (20 ng/ml) and ionomycin (1 μg/ml) in the presence of neutralizing monoclonal antibody (mAb) against IL-4 and/or TGF-β1 or with control IgG for 48 h. Cell-free supernatants were then analyzed for the production of IL-9. For quantitative real-time PCR (qRT-PCR), the cells were stimulated with PMA (20 ng/ml) and ionomycin (1 μg/ml) for 10 min. For Western blot analysis, PBMCs were stimulated for 2 days with immobilized anti-CD3 and anti-CD28 mAbs.

Cell surface and intracellular cytokine staining

Following stimulation, PBMCs were washed and stained with cell surface mAbs for 30 min at 4°C. The cells were then washed, fixed, and permeabilized. Cytokine-specific and isotype-matched control antibodies were added to the cells and incubated for 30 min at 4°C. After intracellular staining, cells were washed and resuspended in PBS. Data were acquired using a FACSCalibur cytometer (BD Biosciences) and were analyzed using the FlowJo Software (Treestar, San Carlos, CA, U.S.A.).

ELISA

The cell culture supernatants were harvested and assayed for IL-9, IFN-γ, and IL-4 using ELISA according to the manufacturers’ protocols (BD Pharmingen and eBioscience, Inc., U.S.A. and France).

RNA extraction and qRT-PCR analysis

Total RNA was extracted from PBMCs using TRIzol (Invitrogen, U.S.A.) according to the manufacturer’s instructions. The cDNA was synthesized from RNA using a Transcriptor First Strand cDNA Synthesis Kit (Takara Biotechnology). TaqMan primers and probes for human PU.1 and interferon regulatory factor 4 (IRF-4) were obtained from Takara Biotechnology, Japan. β-actin was used as an endogenous reference. Quantitative PCR using SYBR Green II (Takara Biotechnology) was performed using an ABI PRISM 7900 Sequence Detector system (Applied Biosystems). Target gene expression was normalized to the expression of β-actin, and the values were calculated relative to control values using the ΔΔCT method. The following primers were used: β-actin: F, 5′-TGGCACCCAGCACAATGAA-3′ and β-actin: R, 5′-CTAAGTCATAGTCCGCCTAGAAGCA-3′; PU.1: F, 5′-GAAGACCTGGTGCCCTATGA-3′ and PU.1: R, 5′-TGGAGCTCCGTGAAGTTGTT-3′; IRF-4: F, 5′-AGCTTGCCTTAGATGCTGTAAATTC-3′ and IRF-4: R, 5′-TGCTGATGTGTTCTGGTAAATCGTA-3′.

Western blotting for PBMCs

PBMCs were collected and lysed after stimulation. Lysates containing 20 μg of total protein were separated on a 10% polyacrylamide gel and then transferred on to PVDF membranes (Millipore, U.S.A.). The membranes were blocked with 5% fat-free milk and incubated with β-actin, PU.1, and IRF-4 primary antibodies (Cell Signaling Technology (CST), U.S.A) at 4°C overnight. The membranes were washed before treating with horseradish peroxidase–conjugated secondary antibodies (CST, U.S.A.) for 1 h at room temperature. The signals were visualized with a Tanon Imaging System (China), and the density of each band was quantitated using ImageJ (National Institutes of Health, U.S.A.).

bEnd.3 cell culture and the in vitro model of oxygen–glucose deprivation plus reoxygenation

The bEnd.3 cells were purchased from AllCells and grown to subconfluence before the exposure to oxygen–glucose deprivation (OGD: 94.95% N2, 0.05% O2, 5% CO2) or normoxic conditions (75% N2, 20% O2, 5% CO2) for 4 or 24 h. Cells were divided into the following four groups: normoxic (normal); recombinant murine IL-9-treated (IL-9); OGD-treatment for 4 h plus 20 h of reoxygenation (RO) (OGD + RO) and IL-9-treated followed by OGD and RO. The OGD experiments were performed in an anaerobic chamber (Stem Cell, Canada). After OGD treatment, the cells were incubated under normoxic conditions for another 20 h. Recombinant murine IL-9 (20 ng/ml, PeproTech, U.S.A.) was added 1 h before OGD treatment and maintained during the RO process. Cells were then collected for further study. All experiments were performed in triplicate.

Paracellular permeability assay

Briefly, bEnd.3 cells were seeded at a density of 2 × 104 cells/well in 100 μl of medium on to polycarbonate 24-well transwell chambers with a 0.4-μm mean pore size and a 0.33-cm2 surface area (Corning, U.S.A.). Cells were then exposed to normal, IL-9, OGD + RO, and OGD + RO + IL-9 conditions for 24 h. Cells were then incubated with fluorescein (NaF) (1 mg/ml) in medium for 30 min. Relative fluorescence passing through the chamber (in the lower chambers) was determined using an EnSpire Manager (PerkinElmer, U.S.A.) multimode plate reader at an excitation wavelength of 460 nm and an emission wavelength of 515 nm.

TEM analysis

The bEnd.3 cells were carefully scraped and fixed with 2.5% glutaraldehyde for 1 h at 4°C. After washing with cacodylate buffer, the cells were postfixed in 1% osmium tetroxide for 30 min, dehydrated in graded ethanal, and polymerized at 60°C for 48 h. The cells were cut with Reichert Jung Ultracut S (Leica, Vienna, Austria) and mounted on grids, stained with uranyl acetate and lead citrate, and then examined under a Zeiss EM 902 A electron microscope (Leo, Oberkochen, Germany). The changes of tight junction proteins (TJPs) were examined by TEM (JEM-1200EX; n143, each) at 60 kV.

The detection of TJPs and the potential mechanisms involved

For Western blotting, bEnd.3 cells were cultured in six-well plates and allowed to grow to confluence. Cultures were then exposed to normal, IL-9, OGD + RO, or OGD + RO + IL-9 conditions. Total protein was collected and the following procedure was used as described above (section on Western blotting for PBMCs). The following primary antibodies were used: claudin-5, zonula occludens 1 (ZO-1), occludin (Invitrogen, U.S.A.), IL-9 receptor (IL-9R), endothelial nitric oxide synthase (eNOS) (Abcam, U.K.), p-signal transducer and activator of transcription 1 (p-STAT1), p-STAT3, p-STAT5, total STAT1, total STAT3, total STAT5, p-protein kinase B (p-Akt), total Akt, p-phosphatidylinositol 3-kinase (p-PI3K), total PI3K, p-protein kinase Cα (p-PKCα), total PKCα, p-phospholipase Cγ (p-PLCγ), total PLCγ, p-extracellular signal regulated kinase (p-ERK), and total ERK antibody (all from CST, U.S.A.).

Immunofluorescence was used to verify the expression of the TJPs. The bEnd.3 cells grown to confluence on collagen-coated coverslips were subjected to the above-mentioned treatments and washed three times with PBS before 4% paraformaldehyde fixation. Fixed cells were permeabilized with 0.3% Triton X-100 for 5 min, and blocked for 1 h at room temperature with 10% goat serum in PBS. The cells were then incubated with anti-eNOS (Abcam, U.K.), anti-ZO-1, anti-occludin, or anti-claudin-5 primary antibodies (Invitrogen, U.S.A.) overnight at 4°C, followed by incubation with Alexa Fluor 555–conjugated anti-rabbit secondary antibodies (Invitrogen, U.S.A.) for 60 min. The coverslips were then mounted on to the glass slides with the antifade reagent Vectashield (Vector Laboratories). Images were acquired using an LSM 510 confocal laser-scanning microscope (Zeiss). ImageJ software (National Institute of Health, U.S.A.) was utilized for image processing and quantitation.

Statistical analysis

Data are presented as the mean ± S.D. or S.E.M. Comparisons between the two groups were performed by paired or unpaired Student’s t tests. Multiple comparisons were performed by one-way ANOVA. P-values <0.05 were considered statistically significant.

Results

Baseline characteristics

We recruited 42 patients newly diagnosed with IS and 22 healthy controls for the present study. There were no significant differences in age (63.5 ± 10.6 compared with 48.6 ± 12.4, P=0.123) or gender (P=0.561) between the two groups. Amongst the patients, 24 (60.0%) had hypertension, 14 (35.0%) had diabetes mellitus, 13 (32.5%) had dyslipidemia, 3 (7.5%) had coronary artery disease, 8 (20.0%) had previous IS (including transient ischemic attack), 2 (5%) had peripheral arterial disease, and 12 (30.0%) were active smokers. Patients were categorized into different subtypes according to TOAST classification: 38.1 % (16/42) were classified into large artery atherosclerosis, 4.76 % (2/42) into cardioembolism, 21.4 % (9/42) into small-vessel occlusion, 2.38% (1/42) into stroke of other determined etiologies, and 33.3 % (14/42) into stroke of undetermined etiology. The values for other parameters, including NIHSS score and infarct volume at sampling, are shown in Table 1.

Table 1
Clinical characteristics of IS patients and healthy controls
 Patients (n=42) Controls (n=22) P-value 
Male gender (n, %) 25, 62.5% 7, 43.5% 0.561 
Age of onset (y, mean ± S.D.) 63.5 ± 10.6 48.6 ± 12.4 0.123 
Patients at acute stage (n, %) 32, 80.0% 
Hypertension (n, %) 24, 60.0% 
Diabetes mellitus (n, %) 14, 35.0% 
Dyslipidemia (n, %) 13, 32.5% 
Coronary artery disease (n, %) 3, 7.14% 
Previous stroke, (n, %) 8, 20.0% 
Current smoker (n, %) 12, 30.0% 
TOAST subtype    
LAA 16, 38.1%   
CE 2, 4.76%   
SVO 9, 21.4%   
SOE 1, 2.38%   
SUE 14, 33.3%   
NIHSS score at sampling, (mean, IQR) 6.45 (1, 24) 
Infarct volume at sampling, (median, IQR) 16.9 (0.21, 120) 
 Patients (n=42) Controls (n=22) P-value 
Male gender (n, %) 25, 62.5% 7, 43.5% 0.561 
Age of onset (y, mean ± S.D.) 63.5 ± 10.6 48.6 ± 12.4 0.123 
Patients at acute stage (n, %) 32, 80.0% 
Hypertension (n, %) 24, 60.0% 
Diabetes mellitus (n, %) 14, 35.0% 
Dyslipidemia (n, %) 13, 32.5% 
Coronary artery disease (n, %) 3, 7.14% 
Previous stroke, (n, %) 8, 20.0% 
Current smoker (n, %) 12, 30.0% 
TOAST subtype    
LAA 16, 38.1%   
CE 2, 4.76%   
SVO 9, 21.4%   
SOE 1, 2.38%   
SUE 14, 33.3%   
NIHSS score at sampling, (mean, IQR) 6.45 (1, 24) 
Infarct volume at sampling, (median, IQR) 16.9 (0.21, 120) 

Abbreviations: CE, cardioembolism; IQR, interquartile range; LAA, large artery atherosclerosis; SOE, stroke of other determined etiology; SUE, stroke of undetermined etiology, SVO, small-vessel occlusion.

Increased expression of IL-9 in PBMCs of patients with IS

To determine the expression of inflammatory cytokines in PBMCs derived from patients with IS and controls, we measured the concentrations of IL-9, IFN-γ, and IL-4 in cell-free supernatants using ELISA. The concentrations of IL-9 (143.96 ± 37.17 compared with 78.67 ± 23.56 pg/ml, P<0.01, Figure 1A) and IL-4 (453.22 ± 190.11 compared with 276.83 ± 164.41 pg/ml, P<0.05, Figure 1C) were significantly higher in patients than those in controls, suggesting that IS in humans is associated with higher expression of these cytokines. No significant difference was found in the expression of IFN-γ between patients and controls (61.94 ± 13.11 compared with 59.36 ± 10.14 ng/ml, P=0.341, Figure 1B).

Increased expression of IL-9 in patients with IS.

Figure 1
Increased expression of IL-9 in patients with IS.

The concertrations of IL-9 (A), IFN-γ (B) and IL-4 (C) in the supernatant after 48 h culture of PBMCs of patients (n=42) and healthy controls (n=22) were detected by ELISA. Data are expressed as mean ± S.E.M. *P<0.05, **P<0.01.

Figure 1
Increased expression of IL-9 in patients with IS.

The concertrations of IL-9 (A), IFN-γ (B) and IL-4 (C) in the supernatant after 48 h culture of PBMCs of patients (n=42) and healthy controls (n=22) were detected by ELISA. Data are expressed as mean ± S.E.M. *P<0.05, **P<0.01.

Increased frequency of circulating CD3+IL-9+ T cells in patients with IS

We explored the characteristics of the IL-9-producing cells in PBMCs and found a significantly increased percentage of CD3+IL-9+ cells in the PBMCs from patients compared with that from controls (0.89 ± 0.21% compared with 0.39 ± 0.24%, P<0.01), whereas no significant difference was observed between these two groups in the percentage of CD3IL-9+ cells (0.38 ± 0.24% compared with 0.51 ± 0.33%, P=0.211). We also determined the percentages of CD3+CD4+IL-9+ (T helper type 9 (Th9)) cells and CD3+CD8+IL-9+ (T cytotoxic type 9 (Tc9)) cells. As shown in Figure 2, compared with healthy controls, patients with IS showed a significant increase in the percentages of both Th9 cells (0.75 ± 0.39% compared with 0.29 ± 0.16%, P<0.01) and Tc9 cells (0.83 ± 0.48% compared with 0.28 ± 0.20%, P<0.01) in PBMCs. These results indicated that IL-9-producing T cells may contribute to the peripheral immune response in patients with IS.

The cellular source of IL-9 in PBMCs of IS patients.

Figure 2
The cellular source of IL-9 in PBMCs of IS patients.

(A) Representative dot plots illustrating the proportion of CD3+IL-9+ and CD3IL-9+ T cells in the total T cells in PBMCs. (B) Quantitative analyses of the percentage of CD3+IL-9+ and CD3IL-9+ T cells in the total T-cell populations between the controls (n=22) and patients (n=35). (C) Representative dot plots illustrating the proportion of CD4+IL-9+/CD8+IL-9+ T cells in the total CD4+/CD8+ T-cell populations. (D) Quantitative analyses of the percentage of CD4+IL-9+/CD8+IL-9+ T cells in the total CD4+/CD8+ T-cell populations between the groups. Data are expressed as mean ± S.E.M. **P<0.01.

Figure 2
The cellular source of IL-9 in PBMCs of IS patients.

(A) Representative dot plots illustrating the proportion of CD3+IL-9+ and CD3IL-9+ T cells in the total T cells in PBMCs. (B) Quantitative analyses of the percentage of CD3+IL-9+ and CD3IL-9+ T cells in the total T-cell populations between the controls (n=22) and patients (n=35). (C) Representative dot plots illustrating the proportion of CD4+IL-9+/CD8+IL-9+ T cells in the total CD4+/CD8+ T-cell populations. (D) Quantitative analyses of the percentage of CD4+IL-9+/CD8+IL-9+ T cells in the total CD4+/CD8+ T-cell populations between the groups. Data are expressed as mean ± S.E.M. **P<0.01.

Elevated PU.1 and IRF-4 gene and protein levels in PBMCs from patients with IS

Next we examined whether the expression levels of PU.1 and IRF-4 were altered in patients with IS. We used qRT-PCR and Western blot analyses to determine the mRNA and protein levels, respectively, in PBMCs obtained from patients with IS (n=8) and from healthy controls (n=8). The results showed that the levels of both PU.1 and IRF-4 mRNA (P<0.05, Figure 3A) and protein (n=6, P<0.01, Figure 3B) were significantly higher in PBMCs from patients with IS than those from controls.

Increased IL-9 mRNA and protein levels in PBMCs of patients with IS.

Figure 3
Increased IL-9 mRNA and protein levels in PBMCs of patients with IS.

IL-9 mRNA (A) and protein (B) levels in normal controls and patients with IS. β-actin was used as the control. Data are expressed as mean ± S.E.M. (n=6). *P<0.05, **P<0.01.

Figure 3
Increased IL-9 mRNA and protein levels in PBMCs of patients with IS.

IL-9 mRNA (A) and protein (B) levels in normal controls and patients with IS. β-actin was used as the control. Data are expressed as mean ± S.E.M. (n=6). *P<0.05, **P<0.01.

Neutralizing IL-4 and/or TGF-β1 decrease the expression of IL-9 in PBMCs from patients with IS

To determine the upstream factors that may contribute to the increased expression of IL-9+ T cells, we pretreated PBMCs from patients with PMA (20 ng/ml) and ionomycin (1 μg/ml) in the presence of anti-IL-4 and/or anti-TGF-β1 antibodies. Nonspecific rabbit IgG was used as isotype control. We observed that inhibition of the biologic activity of IL-4, TGF-β1, or both with neutralizing antibodies down-regulated the production of IL-9 (Figure 4). The inhibitory effect was the strongest when antibodies to both IL-4 and TGF-β1 were used.

Effect of blocking IL-4 and TGF-β 1 on the IL-9 production of patients' PBMCs.

Figure 4
Effect of blocking IL-4 and TGF-β 1 on the IL-9 production of patients' PBMCs.

The concentration of IL-9 in supernatants was examined by ELISA. Data are expressed as mean ± S.E.M. (n=12). *P<0.05; **P<0.01.

Figure 4
Effect of blocking IL-4 and TGF-β 1 on the IL-9 production of patients' PBMCs.

The concentration of IL-9 in supernatants was examined by ELISA. Data are expressed as mean ± S.E.M. (n=12). *P<0.05; **P<0.01.

Association of the percentage of CD3+IL-9+ T cells with NIHSS score and infarct volume in patients with IS

To analyze the relationship between the expression of IL-9 and the disease state in patients with IS, we correlated the percentages of CD4+IL-9+ T cells and CD8+IL-9+ T cells with a number of clinical and laboratory parameters. We found a positive correlation between the percentage of IL-9+ T cells and NIHSS score (P=0.03 for Th9; P=0.049 for Tc9) and between the percentage of IL-9+ T cells and infarct volume as determined using DWI (P=0.018 for Th9, P=0.024 for Tc9; Table 2). Our findings suggest that CD3+IL-9+ T cells may play a critical role in the inflammatory pathophysiologic processes of cerebral ischemic infarction.

Table 2
Correlation amongst IL-9, CD3+CD4+IL-9+ cells (Th9 cells), CD3+CD8+IL-9+ cells (Tc9 cells) of IS patients with NIHSS and infarct volume on DWI (n=35)
 NIHSS Infarct volume 
 Correlation coefficient (Spearman) P-value Correlation coefficient (Spearman) P-value 
IL-9 0.006 0.960 0.186 0.250 
Th9 cells 0.367 0.030* 0.397 0.018* 
Tc9 cells 0.325 0.049* 0.380 0.024* 
 NIHSS Infarct volume 
 Correlation coefficient (Spearman) P-value Correlation coefficient (Spearman) P-value 
IL-9 0.006 0.960 0.186 0.250 
Th9 cells 0.367 0.030* 0.397 0.018* 
Tc9 cells 0.325 0.049* 0.380 0.024* 

*P<0.05.

Detection of IL-9R on bEnd.3 cells

The finding that the percentage of IL-9+ T cells correlated with infarct volume suggested the presence of serious BBB damage in patients with IS. Thus, we used bEnd.3 cells to explore the effects of IL-9 in a model of the BBB. Previous studies have shown that the IL-9R is expressed in mouse aortic endothelial cells and in human carotid plaques [11,12]. However, no study has reported the existence of IL-9R in brain microvascular endothelial cells. Here, we found the expression of IL-9Rs in bEnd.3 cells, and the expression was significantly up-regulated after OGD (Figure 5A,B).

Effect of IL-9 on morphology and permeability of endothelial cells following OGD condition.

Figure 5
Effect of IL-9 on morphology and permeability of endothelial cells following OGD condition.

(A) Representative Western blots for IL-9R in bEnd.3 cells with the loading control, β-actin; (B) quantitative analyses of IL-9R. Data were expressed as mean ± S.D. (n=3); (C) Phase-contrast image showing the morphology of bEnd.3 cells under normoxic condition (normal), recombinant murine IL-9 (IL-9), OGD + RO, or OGD + RO + IL-9 conditions. (D) Quantitative analyses of cell numbers of bEnd.3 under different conditions. The results were shown as mean ± S.E.M. (n=4); (E) Sodium fluorescein (NaF) across bEnd.3 monolayers under different conditons. Data are expressed as mean ± S.E.M. (n=4). Scale bar =200 μm. #P<0.05 compared with normal; *P<0.05 compared with OGD treatment.

Figure 5
Effect of IL-9 on morphology and permeability of endothelial cells following OGD condition.

(A) Representative Western blots for IL-9R in bEnd.3 cells with the loading control, β-actin; (B) quantitative analyses of IL-9R. Data were expressed as mean ± S.D. (n=3); (C) Phase-contrast image showing the morphology of bEnd.3 cells under normoxic condition (normal), recombinant murine IL-9 (IL-9), OGD + RO, or OGD + RO + IL-9 conditions. (D) Quantitative analyses of cell numbers of bEnd.3 under different conditions. The results were shown as mean ± S.E.M. (n=4); (E) Sodium fluorescein (NaF) across bEnd.3 monolayers under different conditons. Data are expressed as mean ± S.E.M. (n=4). Scale bar =200 μm. #P<0.05 compared with normal; *P<0.05 compared with OGD treatment.

Effects of IL-9 on the morphology and permeability of endothelial cells following OGD

We tested the effects of exogenously administered IL-9 on endothelial cell morphology as well as on the BBB integrity with or without the induction of OGD. There was no significant difference in the morphological changes and cell numbers of bEnd.3 amongst the four groups (Figure 5C,D). In contrast, exposure of cells to 4 h of hypoxic conditions followed by 20 h of RO significantly increased the level of NaF that leaked through the cell monolayers compared with that under normal (normoxic) conditions. Additionally, IL-9 treatment in the presence of OGD resulted in a four-fold increase in the passage of NaF relative to that in normoxic and hypoxic cells (P<0.05, Figure 5E). Notably, IL-9 alone did not change the endothelial paracellular permeability compared with that under normal conditions.

Changes in the ultrastructure of TJPs after IL-9 treatment

In order to clarify the mechanism for the change of BBB permeability, we next detected the ultrastructure of TJPs by TEM. The bEnd.3 cells were incubated with IL-9 for 20 h with or without 4-h OGD. As shown in Figure 6, results of TEM revealed that the TJPs were dense and appeared as an electron-dense zone on cell–cell contact in bEnd.3 cells in normal and IL-9 treatment groups. However, the TJPs were opened and the intercellular clefts were widened following OGD treatment, and the damage aggravated in groups with IL-9 treatment after OGD induction.

Effect of IL-9 on ultrastructure of TJPs in bEnd.3 cells shown by transmission electon micrographs.

Figure 6
Effect of IL-9 on ultrastructure of TJPs in bEnd.3 cells shown by transmission electon micrographs.

Representative images of TJPs were presented in normal, IL-9, OGD + RO, and OGD + RO + IL-9 groups. Arrows show TJPs of bEnd.3 cells and micrographs were taken at ×12500. Scale bar =2 μm.

Figure 6
Effect of IL-9 on ultrastructure of TJPs in bEnd.3 cells shown by transmission electon micrographs.

Representative images of TJPs were presented in normal, IL-9, OGD + RO, and OGD + RO + IL-9 groups. Arrows show TJPs of bEnd.3 cells and micrographs were taken at ×12500. Scale bar =2 μm.

Effect of IL-9 on TJP expression with OGD plus RO

We then examined whether the increased endothelial cell monolayer permeability by IL-9 treatment was due to the loss of TJPs (occludin, claudin-5, or ZO-1) induced by OGD. After exposure to IL-9 for 24 h under normoxia or OGD conditions, the levels of claudin-5, occludin, and ZO-1 in a monolayer of bEnd.3 cells were measured using Western blotting. Statistical analyses indicated that exposure to IL-9 alone did not change claudin-5 levels in the endothelial cells under normoxic conditions. However, exposure to IL-9 following OGD markedly exacerbated the loss of claudin-5 (P<0.05, Figure 7A,B). Similar results were observed in the levels of occludin and ZO-1, that is, IL-9 augmented the OGD-induced loss of these TJPs. The results were confirmed using single-label immunofluorescence methods (Figure 7C–F). These findings indicate that the loss of TJPs is likely to be the mechanism underlying the observed endothelial monolayer disruption induced by OGD and IL-9 stimulation.

Effect of IL-9 on the expression of TJPs in bEnd.3 cells following OGD conditions.

Figure 7
Effect of IL-9 on the expression of TJPs in bEnd.3 cells following OGD conditions.

(A) Representative Western blots showing expression levels of the TJPs (occludin, claudin-5, and ZO-1) in bEnd.3 cells following exposure to normal, IL-9, OGD + RO, or OGD + RO + IL-9 conditions. Intensity was normalized to β-actin. (B) Quantitative results are given as mean ± S.E.M. (n=4). (CE) The representative photomicrograph of immunofluorescence for the claudin-5, occludin, ZO-1 (red), and nuclear staining with DAPI (blue) in bEnd.3 monolayers under different conditons (n=3). Scale bar =50 μm. (F) Quantitative analyses of immunofluorescence results of TJPs under different conditions. The results were shown as mean florescence intensities (MFI) ± SEM (n=3). #P<0.05 compared with normal; *P<0.05 compared with OGD conditions.

Figure 7
Effect of IL-9 on the expression of TJPs in bEnd.3 cells following OGD conditions.

(A) Representative Western blots showing expression levels of the TJPs (occludin, claudin-5, and ZO-1) in bEnd.3 cells following exposure to normal, IL-9, OGD + RO, or OGD + RO + IL-9 conditions. Intensity was normalized to β-actin. (B) Quantitative results are given as mean ± S.E.M. (n=4). (CE) The representative photomicrograph of immunofluorescence for the claudin-5, occludin, ZO-1 (red), and nuclear staining with DAPI (blue) in bEnd.3 monolayers under different conditons (n=3). Scale bar =50 μm. (F) Quantitative analyses of immunofluorescence results of TJPs under different conditions. The results were shown as mean florescence intensities (MFI) ± SEM (n=3). #P<0.05 compared with normal; *P<0.05 compared with OGD conditions.

The involvement of eNOS via STAT1/3 and PI3K/Akt signaling pathways in mediating the destructive role of IL-9 on OGD-induced BBB breakdown in vitro

We next investigated the potential mechanisms by which IL-9 accumulation disrupts the BBB after ischemia using endothelial cell monolayers. eNOS and related signaling pathways have been reported to mediate BBB disruption in IS. Therefore, we hypothesized that these related elements participated in IL-9-induced BBB disruption after OGD. Western blot analysis and immunostaining were used to verify this hypothesis. Our results showed that OGD indeed up-regulated the expression of eNOS, and IL-9 administration exacerbated this up-regulation (P<0.05, Figure 8). OGD increased the level of p-STAT1 and p-STAT3, with no change in p-STAT5 level (Figure 8). The levels of p-Akt and p-PI3K were decreased by OGD compared with those in the controls. Co-treatment of IL-9 and OGD significantly exacerbated the changes induced by OGD alone in p-STAT1, p-STAT3, p-Akt, and p-PI3K levels (Figures 9 and 10A,B). However, no change in the level of phosphorylation of these proteins was observed following IL-9 administration in the absence of OGD. These data suggest that the STAT1, STAT3, and PI3K/Akt pathways are involved in the harmful effects of IL-9 in hypoxia, which is partially dependent on the overproduction of eNOS. No changes were observed for PKC-α/PLC-γ/ERK signaling with the addition of IL-9 after OGD (Figure 10C,D).

The expression of eNOS in bEnd.3 cells following exposure to normal, IL-9, OGD+RO or OGD+RO+IL-9 conditions.

Figure 8
The expression of eNOS in bEnd.3 cells following exposure to normal, IL-9, OGD+RO or OGD+RO+IL-9 conditions.

(A) Representative Western blots showing levels of eNOS under different conditions, β-actin was used as the loading control. (B) Quantitative analyses of eNOS by Western blots. Results are given as mean ± S.E.M. (n=3). (C) Immunofluorescence staining of confluent bEnd.3 monolayers for of eNOS (red), nuclei were labeled by DAPI (blue) (n=4). Scale bar =100 μm. (D) Quantitative analyses of immunofluorescence results of eNOS under different conditions. The results were shown as mean florescence intensities (MFI) ± S.E.M. (n=3). #P<0.05 compared with normal, *P<0.05 compared with 4-h OGD.

Figure 8
The expression of eNOS in bEnd.3 cells following exposure to normal, IL-9, OGD+RO or OGD+RO+IL-9 conditions.

(A) Representative Western blots showing levels of eNOS under different conditions, β-actin was used as the loading control. (B) Quantitative analyses of eNOS by Western blots. Results are given as mean ± S.E.M. (n=3). (C) Immunofluorescence staining of confluent bEnd.3 monolayers for of eNOS (red), nuclei were labeled by DAPI (blue) (n=4). Scale bar =100 μm. (D) Quantitative analyses of immunofluorescence results of eNOS under different conditions. The results were shown as mean florescence intensities (MFI) ± S.E.M. (n=3). #P<0.05 compared with normal, *P<0.05 compared with 4-h OGD.

Effect of IL-9 on the expression of STAT signal pathways.

Figure 9
Effect of IL-9 on the expression of STAT signal pathways.

(A) Representative Western blots showing the expression of phosphorylation of STAT (p-STAT)1/3/5 after OGD-induced BBB destruction or treatment with IL-9. Total STAT1/3/5 were used as the loading controls. (BD) Quantitative analyses of p-STAT1, p-STAT3, and p-STAT5. Results are given as mean ± S.E.M. (n=4). #P<0.05 compared with normal, *P<0.05 compared with 4-h OGD.

Figure 9
Effect of IL-9 on the expression of STAT signal pathways.

(A) Representative Western blots showing the expression of phosphorylation of STAT (p-STAT)1/3/5 after OGD-induced BBB destruction or treatment with IL-9. Total STAT1/3/5 were used as the loading controls. (BD) Quantitative analyses of p-STAT1, p-STAT3, and p-STAT5. Results are given as mean ± S.E.M. (n=4). #P<0.05 compared with normal, *P<0.05 compared with 4-h OGD.

Effect of IL-9 on the expression of AKT/PI3K and PKCα/PLCγ/ERK signal pathways.

Figure 10
Effect of IL-9 on the expression of AKT/PI3K and PKCα/PLCγ/ERK signal pathways.

(A) Representative Western blots showing the expression of p-AKT and p-PI3K under normal, IL-9, OGD + RO, or OGD + RO + IL-9 conditions. Total AKT and total PI3K were used as the loading controls. (B) Quantitative analyses of p-AKT and p-PI3K. (C) Representative Western blots showing the expression of p-PKCα, p-PLCγ, and p-ERK under these different conditions. Total PKCα, total PLCγ, and total ERK were used as the loading controls. (D) Quantitative analyses of p-PKCα, p-PLCγ, and p-ERK. Data are presented as mean ± S.E.M. (n=3). #P<0.05 compared with normal, *P<0.05 compared with 4-h OGD.

Figure 10
Effect of IL-9 on the expression of AKT/PI3K and PKCα/PLCγ/ERK signal pathways.

(A) Representative Western blots showing the expression of p-AKT and p-PI3K under normal, IL-9, OGD + RO, or OGD + RO + IL-9 conditions. Total AKT and total PI3K were used as the loading controls. (B) Quantitative analyses of p-AKT and p-PI3K. (C) Representative Western blots showing the expression of p-PKCα, p-PLCγ, and p-ERK under these different conditions. Total PKCα, total PLCγ, and total ERK were used as the loading controls. (D) Quantitative analyses of p-PKCα, p-PLCγ, and p-ERK. Data are presented as mean ± S.E.M. (n=3). #P<0.05 compared with normal, *P<0.05 compared with 4-h OGD.

Discussion

Recent emerging evidence has indicated that IL-9 plays detrimental roles in the pathogenesis of cancer immunity and transplant tolerance [1316]. Moreover, IL-9 has been found to increase in patients with rheumatoid arthritis and coronary atherosclerosis [11,17]. However, little is known about the role of IL-9 in the development of IS.

To the best of our knowledge, the present study is the first to demonstrate that IL-9 is differentially expressed in PBMCs derived from patients with IS and those from healthy controls. We determined the cellular source of IL-9 in these PBMCs. A previous study has suggested that the main cellular sources of IL-9 are mast cells, natural killer T cells, and T-cell subsets [18]. In our study, we identified CD3+ T cells as the main source of IL-9 in PBMCs derived from patients with IS. Within the CD3+ T-cell population, the percentages of Th9 and Tc9 cells in the PBMCs were both significantly higher than those in healthy controls, indicating that hypoxia exposure was beneficial for the differentiation of IL-9+ T cells. This suggests that local ischemic brain lesions may contribute to PBMC differentiation toward the Th9 and Tc9 subsets during the early stage of IS.

A previous study reported an increased level of IFN-γ in both the peripheral blood and ischemic brain [4]. However, our study found no difference in the IFN-γ level but a higher IL-4 expression in patients with IS. The concurrent changes in IL-9 and IL-4 levels we observed in IS patients are consistent with the notion that IL-4 promotes the secretion of IL-9 from T cells, which is mediated by STAT6 and IRF4 [19,20]. The development of Th9 and Tc9 cells is also driven by the transcription factor PU.1, which increases in response to TGF-β1 [21,22]. Not only did we found that PMBCs derived from patients with IS exhibited increased levels of PU.1 and IRF-4, but we also demonstrated that neutralizing IL-4 and TGF-β1 led to a reduction in IL-9 in PBMCs. Taken together, these results provide strong evidence that IS skewed PBMC differentiation toward Th9 and Tc9 cells.

We also observed strong positive correlations between the infarct volume in patients with IS and the percentages of Th9 and Tc9 cells, suggesting potential immunopathological roles of IL-9+ T cells in IS. This result is consistent with previously published data regarding tumor immunity and allergic airway hyperreactivity [2326]. Although both the expression level of IL-9 and the number of circulating IL-9+ T cells were elevated, IL-9+ T cells but not IL-9 was significantly associated with IS severity in our patients. A plausible explanation for this observation is that polarized Th9 and Tc9 cells may have infiltrated the ischemic brain and played an effective role in local injury, whereas increased IL-9 levels may have existed only under a specific stimulation or may merely have been an accompanying change in peripheral blood. Tuttolomondo et al. [27] proved that peripheral frequency of CD4+CD28 cells may be useful to differentiate TOAST diagnostic subtypes, whereas circulating IL-9+ T cells did not differ significantly between subgroups in our study (results not shown).

The breakdown of the BBB is one of the main contributors interfering with brain recovery from ischemic damage [28]. We found a positive association between IL-9+ T cells and infarct volume as assessed by DWI, which has been shown to correlate with BBB damage in IS [29]. Previous studies have revealed that Th9- and Tc9-mediated immunological responses critically dependent on IL-9 production in vivo [9]. We speculated that IL-9+ T cells may affect the BBB by secreting IL-9 locally. Therefore, we used bEnd.3 cells, a mouse brain microvascular endothelial cell line, to conduct the in vitro experiments examining the BBB. Prior to the present study, there was no report on the existence of IL-9Rs on brain microvascular endothelial cells. Our data not only revealed that IL-9Rs are expressed in bEnd.3 cells, but also that these receptors were more activated following OGD than under normoxic conditions. IL-9 treatment disrupted the BBB integrity following OGD plus RO (Figure 5) but did not harm the BBB under normal conditions, and the damaged BBB was associated with the down-regulation of three types of TJPs. These TJPs, claudin-5, occludin, and ZO-1, are important BBB structural components that seal the gaps between adjacent endothelial cells and thus restrict paracellular permeability.The down-regulation of these TJPs reflects a compromise in the integrity of BBB [30]. Hence, our results indicated that IL-9 disrupted the BBB through the IL-9R/TJPs pathway in bEnd.3 cells.

eNOS confers stroke protection by increasing cerebral blood flow through the vasodilation properties of endothelial cells [31]. It has also been shown to increase the permeability of the BBB under pathological conditions [32,33]. Cavtratin, an eNOS inhibitor, abrogates vascular endothelial cell growth factor-A induced BBB disruption and protects against neurologic deficit in experimental autoimmune encephalomyelitis [33]. Our results provide direct evidence that eNOS participates in OGD-induced BBB disruption and that IL-9 amplifies the production of eNOS, indicating that eNOS may be a downstream effector of IL-9.

The IL-9–IL-9R interaction is known to activate classic STAT signaling [34], which has been shown to play a role in the inflammatory signaling cascades triggered by lipopolysaccharides and some cytokines [35,36]. Xuan et al. [37] has reported that ischemic preconditioning of the myocardium activates Ser727 phosphorylation of STAT1 and STAT3 and then modulates infract size by up-regulating STAT-dependent genes, such as COX-2. Our data showed that IL-9 induced phosphorylation of STAT1 and STAT3 in bEnd.3 cells subjected to OGD, indicating that the activation of STAT1/STAT3 by IL-9 may account for the up-regulation of eNOS in endothelium. PI3K/Akt signaling has been shown to have neuroprotective effects, as represented by its mediation of neuronal survival or cellular angiogenesis under a wide variety of circumstances [38,39]. Several studies have also shown that activation of the PI3K/Akt signaling pathway is beneficial in stroke [4042]. In the present study, we demonstrated in vitro that the role of IL-9 in OGD-induced BBB breakdown was related to the deactivation of the PI3K/Akt pathway, illustrating a positive role of the PI3K/Akt pathway in a model of OGD. Whereas as a signaling cascade promoting the production of eNOS [33], PLCγ–PKC stabilization remained constant in our experiments. Thus, taken together, we found that IL-9 increased eNOS production, which down-regulated TJPs in OGD-treated bEnd.3 cells via IL-9R through activated STATs but blocked AKT/PI3K signaling pathways. This is distinctly different from the pathogenesis of experimental autoimmune encephalomyelitis, in which IL-9 was shown to induce the expression of chemokine CC ligand 20 in astrocytes, thereby attracting Th17 cells into the central nervous system [43].

Our study has some limitations that should be acknowledged. First, we administered exogenous IL-9 but not IL-9 transgenic cells. Second, the mechanisms concerning other pro-inflammatory effects of IL-9 on brain microvascular endothelial cells were not addressed in the present study. Finally, we did not examine the direct function of IL-9 in vivo. Nevertheless, through the combined results of patient samples and in vitro experiments, the present study presented evidence suggesting a vital role of IL-9 in IS. Further studies are needed to validate and extend the current results.

In conclusion, we found that expression levels of IL-9 as well as the percentages of Th9 and Tc9 cells were significantly higher amongst PBMCs derived from patients with IS than amongst those from healthy controls. To the best of our knowledge, this is the first study demonstrating that IL-9 treatment that compromised the BBB integrity by regulating the IL-9R/STAT1,3 and PI3K/Akt pathways under ischemic conditions. We further identified that the downstream effector eNOS mediated the levels of TJPs. On the basis of our results, we believe that targetting the IL-9 pathway may provide a new therapeutic strategy for cerebrovascular diseases.

Clinical perspectives

  • IL-9 has been shown to have pleiotropic functions in multiple diseases, including allergy and experimental autoimmune encephalomyelitis, but the effects of IL-9 in IS remains unknown.

  • Patients with IS had a significantly higher expression of IL-9 as well as the percentages of Th9 and Tc9 cells than those from healthy controls; and we first demonstrated that IL-9 treatment compromised the BBB integrity by regulating the IL-9R/STAT1, 3 and PI3K/Akt pathways under ischemic conditions. We further identified that the downstream effector eNOS mediated the levels of TJPs.

  • Based on our results, we believe that targetting the IL-9 pathway may provide a new therapeutic strategy for cerebrovascular diseases.

Competing interests

The authors declare that there are no competing interests associated with the manuscript.

Funding

This work was supported by the National Natural Science Foundation of China [grant numbers 81271328, 81671178]; and the Natural Science Foundation of Guangdong Province, China [grant number 2014A030312001].

Author contribution

S.T. and Y.S. designed the research. S.T., Y.W., Y.L., and S.L. performed the research. Y.S., Z.D., L.Z., W.C., Q.Z., L.Z., B.Z., X.M., and H.L. analyzed the data. C.W. and X.H. participated in design and co-ordination of the study. S.T., Y.S., L.P., and Z.L. wrote the manuscript. All the authors read and approved the final manuscript.

Abbreviations

     
  • BBB

    blood–brain barrier

  •  
  • CST

    Cell Signaling Technology

  •  
  • DWI

    diffusion-weighted imaging

  •  
  • eNOS

    endothelial nitric oxide synthase

  •  
  • IFN-γ

    interferon-γ

  •  
  • IL

    interleukin

  •  
  • IL-9R

    interleukin-9 receptor

  •  
  • IRF-4

    interferon regulatory factor 4

  •  
  • IS

    ischemic stroke

  •  
  • mAb

    monoclonal antibody

  •  
  • NIHSS

    National Institutes of Health Stroke Scale

  •  
  • OGD

    oxygen–glucose deprivation

  •  
  • PBMC

    peripheral blood mononuclear cell

  •  
  • p-AKT

    p-protein kinase B

  •  
  • p-ERK

    p-extracellular signal regulated kinase

  •  
  • p-PI3K

    p-phosphatidylinositol 3-kinase

  •  
  • p-PKCα

    p-protein kinase Cα

  •  
  • p-PLCγ

    p-phospholipase Cγ

  •  
  • p-STAT1

    p-signal transducer and activator of transcription 1

  •  
  • qRT-PCR

    quantitative real-time PCR

  •  
  • RO

    reoxygenation

  •  
  • Tc9

    T cytotoxic type 9

  •  
  • Th1

    T helper type 1

  •  
  • Th9

    T helper type 9

  •  
  • TGF-β1

    transforming growth factor-β1

  •  
  • TJP

    tight junction protein

  •  
  • TOAST

    Trial of Org 10172 in Acute Stroke Treatment

  •  
  • ZO-1

    zonula occludens 1

References

References
1
McColl
B.W.
,
Allan
S.M.
and
Rothwell
N.J.
(
2009
)
Systemic infection, inflammation and acute ischemic stroke
.
Neuroscience
158
,
1049
1061
2
Sandoval
K.E.
and
Witt
K.A.
(
2008
)
Blood–brain barrier tight junction permeability and ischemic stroke
.
Neurobiol. Dis.
32
,
200
219
3
Seifert
H.A.
and
Pennypacker
K.R.
(
2014
)
Molecular and cellular immune responses to ischemic brain injury
.
Transl. Stroke Res.
5
,
543
553
4
Yilmaz
G.
,
Arumugam
T.V.
,
Stokes
K.Y.
and
Granger
D.N.
(
2006
)
Role of T lymphocytes and interferon-gamma in ischemic stroke
.
Circulation
113
,
2105
2112
5
Macrez
R.
,
Ali
C.
,
Toutirais
O.
,
Le Mauff
B.
,
Defer
G.
,
Dirnagl
U.
et al. 
(
2011
)
Stroke and the immune system: from pathophysiology to new therapeutic strategies
.
Lancet Neurol.
10
,
471
480
6
Shichita
T.
,
Sakaguchi
R.
,
Suzuki
M.
and
Yoshimura
A.
(
2012
)
Post-ischemic inflammation in the brain
.
Front. Immunol.
3
,
132
7
Tuttolomondo
A.
,
Pedone
C.
,
Pinto
A.
,
Di Raimondo
D.
,
Fernandez
P.
,
Di Sciacca
R.
et al. 
(
2008
)
Predictors of outcome in acute ischemic cerebrovascular syndromes: the GIFA study
.
Int. J. Cardiol.
125
,
391
396
8
Lin
Y.
,
Zhang
L.
,
Dai
Y.
,
Li
H.
,
Wang
Y.
,
Zhang
B.
et al. 
(
2015
)
Expression of interleukin-9 and its upstream stimulating factors in rats with ischemic stroke
.
Neurol. Sci.
36
,
913
920
9
Stassen
M.
,
Schmitt
E.
and
Bopp
T.
(
2012
)
From interleukin-9 to T helper 9 cells
.
Ann. N. Y. Acad. Sci.
1247
,
56
68
10
Adams
H.P.
Jr
,
Bendixen
B.H.
,
Kappelle
L.J.
,
Biller
J.
,
Love
B.B.
,
Gordon
D.L.
et al. 
(
1993
)
Classification of subtype of acute ischemic stroke. Definitions for use in a multicenter clinical trial. TOAST. Trial of Org 10172 in acute stroke treatment
.
Stroke
24
,
35
41
11
Gregersen
I.
,
Skjelland
M.
,
Holm
S.
,
Holven
K.B.
,
Krogh-Sorensen
K.
,
Russell
D.
et al. 
(
2013
)
Increased systemic and local interleukin 9 levels in patients with carotid and coronary atherosclerosis
.
PLoS ONE
8
,
e72769
12
Zhang
W.
,
Tang
T.
,
Nie
D.
,
Wen
S.
,
Jia
C.
,
Zhu
Z.
et al. 
(
2015
)
IL-9 aggravates the development of atherosclerosis in ApoE-/- mice
.
Cardiovasc. Res.
106
,
453
464
13
Vegran
F.
,
Apetoh
L.
and
Ghiringhelli
F.
(
2015
)
Th9 cells: a novel CD4 T-cell subset in the immune war against cancer
.
Cancer Res.
75
,
475
479
14
Kim
I.K.
,
Kim
B.S.
,
Koh
C.H.
,
Seok
J.W.
,
Park
J.S.
,
Shin
K.S.
et al. 
(
2015
)
Glucocorticoid-induced tumor necrosis factor receptor-related protein co-stimulation facilitates tumor regression by inducing IL-9-producing helper T cells
.
Nat. Med.
21
,
1010
1017
15
Fabrega
E.
,
Lopez-Hoyos
M.
,
San Segundo
D.
,
Casafont
F.
,
Moraleja
I.
,
Sampedro
B.
et al. 
(
2012
)
Interleukin-9 in stable liver transplant recipients
.
Transplant. Proc.
44
,
1536
1538
16
Noelle
R.J.
and
Nowak
E.C.
(
2010
)
Cellular sources and immune functions of interleukin-9
.
Nat. Rev. Immunol.
10
,
683
687
17
Rojas-Zuleta
W.G.
and
Vasquez
G.
(
2016
)
Th9 lymphocytes: a recent history from IL-9 to its potential role in rheumatic diseases
.
Autoimmun. Rev.
15
,
649
655
18
Goswami
R.
and
Kaplan
M.H.
(
2011
)
A brief history of IL-9
.
J. Immunol.
186
,
3283
3288
19
Staudt
V.
,
Bothur
E.
,
Klein
M.
,
Lingnau
K.
,
Reuter
S.
,
Grebe
N.
et al. 
(
2010
)
Interferon-regulatory factor 4 is essential for the developmental program of T helper 9 cells
.
Immunity
33
,
192
202
20
Goswami
R.
,
Jabeen
R.
,
Yagi
R.
,
Pham
D.
,
Zhu
J.
,
Goenka
S.
et al. 
(
2012
)
STAT6-dependent regulation of Th9 development
.
J. Immunol.
188
,
968
975
21
Veldhoen
M.
,
Uyttenhove
C.
,
van Snick
J.
,
Helmby
H.
,
Westendorf
A.
,
Buer
J.
et al. 
(
2008
)
Transforming growth factor-beta ‘reprograms’ the differentiation of T helper 2 cells and promotes an interleukin 9-producing subset
.
Nat. Immunol.
9
,
1341
1346
22
Chang
H.C.
,
Sehra
S.
,
Goswami
R.
,
Yao
W.
,
Yu
Q.
,
Stritesky
G.L.
et al. 
(
2010
)
The transcription factor PU.1 is required for the development of IL-9-producing T cells and allergic inflammation
.
Nat. Immunol.
11
,
527
534
23
Purwar
R.
,
Schlapbach
C.
,
Xiao
S.
,
Kang
H.S.
,
Elyaman
W.
,
Jiang
X.
et al. 
(
2012
)
Robust tumor immunity to melanoma mediated by interleukin-9-producing T cells
.
Nat. Med.
18
,
1248
1253
24
Kerzerho
J.
,
Maazi
H.
,
Speak
A.O.
,
Szely
N.
,
Lombardi
V.
,
Khoo
B.
et al. 
(
2013
)
Programmed cell death ligand 2 regulates TH9 differentiation and induction of chronic airway hyperreactivity
.
J. Allergy Clin. Immunol.
131
,
1048
1057
25
Visekruna
A.
,
Ritter
J.
,
Scholz
T.
,
Campos
L.
,
Guralnik
A.
,
Poncette
L.
et al. 
(
2013
)
Tc9 cells, a new subset of CD8(+) T cells, support Th2-mediated airway inflammation
.
Eur. J. Immunol.
43
,
606
618
26
Lu
Y.
,
Hong
B.
,
Li
H.
,
Zheng
Y.
,
Zhang
M.
,
Wang
S.
et al. 
(
2014
)
Tumor-specific IL-9-producing CD8+ Tc9 cells are superior effector than type-I cytotoxic Tc1 cells for adoptive immunotherapy of cancers
.
Proc. Natl. Acad. Sci. U.S.A.
111
,
2265
2270
27
Tuttolomondo
A.
,
Pecoraro
R.
,
Casuccio
A.
,
Di Raimondo
D.
,
Buttà
C.
,
Clemente
G.
et al. 
(
2015
)
Peripheral frequency of CD4+ CD28- cells in acute ischemic stroke: relationship with stroke subtype and severity markers
.
Medicine
94
,
e81
28
Persidsky
Y.
,
Ramirez
S.H.
,
Haorah
J.
and
Kanmogne
G.D.
(
2006
)
Blood–brain barrier: structural components and function under physiologic and pathologic conditions
.
J. Neuroimmune Pharmacol.
1
,
223
236
29
Kahles
T.
,
Luedike
P.
,
Endres
M.
,
Galla
H.J.
,
Steinmetz
H.
,
Busse
R.
et al. 
(
2007
)
NADPH oxidase plays a central role in blood-brain barrier damage in experimental stroke
.
Stroke
38
,
3000
3006
30
Wolburg
H.
and
Lippoldt
A.
(
2002
)
Tight junctions of the blood–brain barrier: development, composition and regulation
.
Vasc. Pharmacol.
38
,
323
337
31
Zhu
J.
,
Song
W.
,
Li
L.
and
Fan
X.
(
2016
)
Endothelial nitric oxide synthase: a potential therapeutic target for cerebrovascular diseases
.
Mol. Brain
9
,
30
32
Yamagata
K.
,
Tagami
M.
,
Takenaga
F.
,
Yamori
Y.
and
Itoh
S.
(
2004
)
Hypoxia-induced changes in tight junction permeability of brain capillary endothelial cells are associated with IL-1beta and nitric oxide
.
Neurobiol. Dis.
17
,
491
499
33
Argaw
A.T.
,
Asp
L.
,
Zhang
J.
,
Navrazhina
K.
,
Pham
T.
,
Mariani
J.N.
et al. 
(
2012
)
Astrocyte-derived VEGF-A drives blood-brain barrier disruption in CNS inflammatory disease
.
J. Clin. Invest.
122
,
2454
2468
34
Knoops
L.
and
Renauld
J.C.
(
2004
)
IL-9 and its receptor: from signal transduction to tumorigenesis
.
Growth Factors
22
,
207
215
35
Levy
D.E.
and
Darnell
J.E.
Jr
(
2002
)
Stats: transcriptional control and biological impact
.
Nat. Rev. Mol. Cell Biol.
3
,
651
662
36
Murray
P.J.
(
2007
)
The JAK-STAT signaling pathway: input and output integration
.
J. Immunol.
178
,
2623
2629
37
Xuan
Y.T.
,
Guo
Y.
,
Zhu
Y.
,
Wang
O.L.
,
Rokosh
G.
and
Bolli
R.
(
2007
)
Endothelial nitric oxide synthase plays an obligatory role in the late phase of ischemic preconditioning by activating the protein kinase C epsilon p44/42 mitogen-activated protein kinase pSer-signal transducers and activators of transcription1/3 pathway
.
Circulation
116
,
535
544
38
Qi
D.
,
Ouyang
C.
,
Wang
Y.
,
Zhang
S.
,
Ma
X.
,
Song
Y.
et al. 
(
2014
)
HO-1 attenuates hippocampal neurons injury via the activation of BDNF-TrkB-PI3K/Akt signaling pathway in stroke
.
Brain Res.
1577
,
69
76
39
Lu
Y.
,
Lei
S.
,
Wang
N.
,
Lu
P.
,
Li
W.
,
Zheng
J.
et al. 
(
2016
)
Protective effect of minocycline against ketamine-induced injury in neural stem cell: involvement of PI3K/Akt and Gsk-3 Beta pathway
.
Front. Mol. Neurosci.
9
,
135
40
Davis
S.M.
and
Pennypacker
K.R.
(
2016
)
Targeting antioxidant enzyme expression as a therapeutic strategy for ischemic stroke
.
Neurochem. Int.
107
,
23
32
41
Wang
J.
,
He
C.
,
Zhou
T.
,
Huang
Z.
,
Zhou
L.
and
Liu
X.
(
2016
)
NGF increases VEGF expression and promotes cell proliferation via ERK1/2 and AKT signaling in Muller cells
.
Mol. Vis.
22
,
254
63
42
Zhu
H.
,
Zhang
Y.
,
Shi
Z.
,
Lu
D.
,
Li
T.
,
Ding
Y.
et al. 
(
2016
)
The neuroprotection of liraglutide against ischaemia-induced apoptosis through the activation of the PI3K/AKT and MAPK pathways
.
Sci. Rep.
6
,
26859
43
Zhou
Y.
,
Sonobe
Y.
,
Akahori
T.
,
Jin
S.
,
Kawanokuchi
J.
,
Noda
M.
et al. 
(
2011
)
IL-9 promotes Th17 cell migration into the central nervous system via CC chemokine ligand-20 produced by astrocytes
.
J. Immunol.
186
,
4415
4421

Author notes

*

These authors contributed equally to this work.