TY - JOUR
AU - Yang, Guo-Yuan
AB - Abstract Rationale: Cerebral ischemia upregulates aquaporin-4 expression, increases blood-brain barrier (BBB) permeability, and induces brain edema. Mesenchymal stem cells (MSCs) can repress inflammatory cytokines and show great potential for ischemic stroke therapy. However, the effect of MSCs regarding the protection of ischemia-induced BBB break down is unknown. Objective: We test whether MSCs therapy protects BBB integrity and explore the molecular mechanisms of aquaporin-4 on BBB integrity. Methods and Results: Two hundred and twenty-eight adult CD1 male mice underwent 90 minutes transient middle cerebral artery occlusion and received 2 × 105 MSCs intracranial transplantation. The neurological severity score was improved and both ischemia-induced brain edema and BBB leakage were reduced in MSC-treated mice. MSCs therapy reduced astrocyte apoptosis and inhibited ischemia-induced aquaporin-4 upregulation. In addition, small-interfering RNA knockdown of aquaporin-4 after cerebral ischemia effectively reduced aquaporin-4 expression, brain edema, BBB leakage, and astrocyte apoptosis. Conditional medium from lipopolysaccharide (LPS)-activated microglia enhanced aquaporin-4 expression, p38 and JNK phosphorylation, and apoptosis of cultured astrocytes. MSC treatment reduced the expression of inflammatory cytokines in LPS-activated microglia, and subsequently reduced aquaporin-4 expression and apoptosis of astrocytes. Knockdown of aquaporin-4 in cultured astrocytes also reduced apoptosis. Treatment with p38 and JNK inhibitors showed that p38, but not the JNK signaling pathway, was responsible for the aquaporin-4 upregulation. Conclusion: MSCs protected BBB integrity by reducing the apoptosis of astrocytes after ischemic attack, which was due to the attenuation of inflammatory response and downregulation of aquaporin-4 expression via p38 signaling pathway. Stem Cells 2014;32:3150–3162 Aquaporin-4, Blood-brain barrier, Edema, Inflammation, Ischemia Introduction Brain edema exacerbates various brain injuries, including traumatic brain injury and ischemic or hemorrhagic stroke, among others [1-3]. Preventing brain edema with stem cell transplantation, drug administration, or exogenous target gene transduction could improve neurobehavioral outcomes [4-7]. Studies have demonstrated that inflammatory cytokines contribute significantly to the development of brain edema by disrupting blood-brain barrier (BBB) integrity [4, 8]. However, those studies mainly focused on the interaction between inflammatory cytokines and vascular endothelium; the effect of inflammatory cytokines on astrocytes after ischemic stroke was not explored. Aquaporin-4 (AQP4), a water channel protein expressed on the end-feet of astrocytes, has been widely studied as an inflammatory mediator [9]. When the brain is damaged, the changes of osmotic pressure alter transport systems and drive water into astrocytes, ultimately causing astrocytes to swell [10, 11]. This process is closely involved with AQP4 upregulation. Inhibiting AQP4 upregulation after brain injury has shown optimistic effects for the prognosis [12]. Moreover, AQP4 is upregulated in inflammation related diseases, which implies an innate relationship between the release of inflammatory cytokines and upregulation of AQP4 [13]. While there is evidence of AQP4 upregulation, the underlying molecular mechanism of AQP4 upregulation is unknown. As a treatment option for ischemic stroke, mesenchymal stem cell (MSC) therapy is under extensive investigation. Other than directly engrafting into the lesion site [14], the therapeutic effect of MSCs seems to be due to secretory factors like neurotrophins, chemokines, and cytokines [15, 16]. MSCs have also been documented to hamper experimental autoimmune encephalomyelitis [17, 18] and graft-versus-host disease [19]. Bone marrow-derived MSCs could also reduce the production of proinflammatory cytokines such as IL-1β, TNF-α, and IL-6 in astrocyte cultures [20]. These therapeutic benefits are the result of immunomodulatory properties of MSCs. However, very limited studies have focused on the effect of MSCs on microglia. The aim of this study is to explore (a) whether MSCs therapy attenuates BBB breakdown and reduces brain edema following transient ischemia; (b) the role of MSCs in maintaining BBB integrity; and (c) the molecular mechanism of AQP4 in MSCs therapy. Materials and Methods Experimental Group Adult male CD1 mice (n = 228) were randomly divided into 10 groups: in the MSCs transplantation study, (a) sham group, (b) phosphate buffered saline (PBS) group, (c) MSC-treated group, n = 42–51 per group. We added additional control (human umbilical vein endothelial cell [HUVEC]) group (d) to explore whether other cell lines are as efficient as MSCs (n = 12). In the mechanistic study, (e) sham group, (f) vehicle group, (g) P38 inhibitor (SB239063) treated group, n = 6–9 per group. In the AQP4 interference study, (h) PBS group, (i) scrambled small-interfering RNA (siRNA) group, and (j) siAQP4 group, n = 15 per group. MSCs Isolation and Identification Animal experimental protocols were approved by the Institutional Animal Care and Use Committee of Shanghai Jiao Tong University, Shanghai, China. All animals were housed in a SPF animal room with free access to water and food. Bone-derived MSCs (BMSCs) were isolated and harvested as previously described [21]. Briefly, adult male SD (Sprague Dawley) rats (Sippr-BK, Co., Shanghai, China, http://www.sippr-bk.cn) weighing 200–250 g were sacrificed. BMSCs were flushed from the femoral and tibial bones. The cells were suspended in Dulbecco’s modified Eagle’s medium (DMEM, Life Technologies, Carlsbad, CA, www.lifetechnologies.com) with 10% fetal bovine serum (FBS, Life Technologies, Carlsbad, CA) and then incubated at 37°C with 5% CO2. The medium containing nonadherent cells was discarded 24 hours later; and fresh medium was added. After isolation, the medium was changed every 3 days, and the primary cells were subcultured 1:2 when 80% confluence. Identification was performed by flow cytometry to examine the surface markers CD90 (1:100, Ebioscience, San Diego, CA, www.ebioscience.com), CD29 (1:100, Ebioscience, Sandiego, CA), CD31 (1:100, BD Biosciences, Mississauga, ON, www.bdbiosciences.com), and CD45 (1:100, Ebioscience, San Diego, CA) of cultured cells [22]. Cells (2 × 105) were incubated in a 200 µl buffer (PBS with 2 µl of CD29-APC, CD90-cy5.5, CD31-PE, FITC-CD45, isotope control antibodies, respectively) for 20 minutes on ice followed by three washes with PBS. Next, the cells were suspended in 200 µl of PBS and analyzed in an FACS instrument (BD Biosciences, mississauga, ON). For cellular tracking after transplantation, cells were labeled using Dil (Life Technologies, Carlsbad, CA) or adenovirus-GFP (AD-GFP) in some experiments. Transient Middle Cerebral Artery Occlusion Focal cerebral ischemia was induced by transient middle cerebral artery occlusion (tMCAO) according to previous studies [23]. Briefly, adult male CD1 mice (25–30 g) were anesthetized with ketamine/xylazine (100 mg/10 mg/kg, Sigma-Aldrich, San Louis, MO, www.sigmaaldrich.com). A heating pad (RWD Life Science, Shenzhen, China, www.rwdstco.com) was used to maintain the body temperature at 37°C ± 0.3°C. After the common carotid artery, the external carotid artery (ECA), and the internal carotid artery (ICA) were isolated; a suture (Covidien, Mansfield, MA, www.covidien.com) coated with silicon was inserted from the ECA stump into the ICA, and stopped at the opening of the middle cerebral artery (MCA). The distance from the furcation of the ECA/ICA to the opening of the MCA was 9 ± 0.5 mm. Reperfusion was performed 90 minutes after MCAO with suture withdrawal. To confirm successful occlusion and reperfusion, cerebral blood flow was measured in the left MCA territory using laser Doppler flowmetry (Moor Instruments, Devon, U.K., www.moor.co.uk). The sham-operated mice were subjected to the same procedure except for the suture insertion. MSCs Transplantation and siRNA Knockdown of AQP4 MSCs or siRNA were injected within 20 minutes after reperfusion. The animals were anesthetized with ketamine/xylazine intraperitoneally, and received stereotaxic transplantation. A small skull hole was made using a microsurgical drill followed by a 10 µl Hamilton syringe (Hamilton, Bonaduz, Switzerland, www.hamiltoncompany.com) injection. The transplant coordinate was 2 mm lateral to the sagittal suture and 1 mm posterior to the coronal suture. The syringe was lowered into the brain 3 mm under the Dura. MSCs suspension with 2 × 105 cells in 5 µl PBS was injected at a rate of 0.4 µl/minute. Four siRNA duplexes against AQP4 (400 ng, siAQP4; Gene Pharma, Shanghai, China, www.genepharma.com) and nontargeting siRNA (400 ng, scrambled RNA; Gene Pharma, Shanghai, China) were mixed with INTERFERin (Polyplus Transfection, Illkirch, France, www.polyplus-transfection.com) at a final volume of 5 µl and incubated for 20 minutes before injection. A second siRNA injection was repeated 2 days later in all animals using the same injection protocol [24]. After completion of the injection, the needle was slowly withdrawn from the animals. The hole was sealed with bone wax and the wound was closed. PBS, HUVECs, or scrambled siRNA was injected separately as controls. The sham-operated mice were subjected to the same procedure except for injection. Neurological Severity Score Determination At 1 and 3 days after tMCAO, modified Neurological severity scores (mNSS) testing was performed to assess the neurological status of the animals, which included motor, sensory, balance, and reflex tests (normal score, 0; maximal deficit score, 14) [25]. For the motor test, the animal was raised by the tail and the flexion of its forelimb was observed (0–3), and its gait was observed after being placed on the floor (0–3). Beam balance tests were performed to look into the animals’ posture on a beam (0–6). The reflexes absent test (0–2) included the pinna reflex and the corneal reflex. Brain Water Content and BBB Permeability Assay Mice were sacrificed at 1 and 3 days after tMCAO with a high dose of chloral hydrate (10%) anesthesia. Brain samples were weighed before and after dehydration in an oven at 95°C for 24 hours. Brain water content was calculated using the formula: ([wet tissue weight-dry tissue weight]/wet tissue weight) × 100%. BBB permeability was assessed by measuring the extravasation of Evans Blue (EB, Sigma-Aldrich, St. Louis, MO, www.sigmaaldrich.com) and IgG. EB dye solution (2% in saline, 4 ml/kg) was injected through the left jugular vein at 1 and 3 days following tMCAO. After 2 hours of circulation, the mice were sacrificed for transcardial perfusion. The brain hemispheres were weighed and then EB was extracted by homogenizing the samples in 1 ml of 50% trichloroacetic acid solution followed by centrifuging at 12,000g for 20 minutes. The supernatant was diluted with 100% ethanol at a ratio of 1:3. The amount of EB was quantified at 610 nm by a spectrophotometer (Bio-Tek, Winooski, VT, www.biotek.com). IgG was examined as previously described [26]. Briefly, brain slices were incubated with biotinylated antibody for 30 minutes after fixation and blocked before incubating in ABC reagent (Vector Labs, Burlingame, CA, www.vectorlabs.com) for 30 minutes. The immunoreactivity was visualized using DAB (Vector Labs, Burlingame, CA) staining and slices were counterstained with hematoxylin. Six fields were randomly selected from the area of interest for each section and analyzed by Image Pro Plus 6.0 software (Media Cybernetics, Bethesda, MD, www.mediacy.com) for mean integrated optical density (IOD) analysis. Immunocytochemistry and Immunohistochemistry Cultured astrocytes or brain sections were fixed with 4% paraformaldehyde for 5 minutes and then incubated in PBS containing 0.1% Triton X-100 for 10 minutes at room temperature followed by blocking with 10% BSA for 1 hour at room temperature. Astrocytes were incubated with anti-aquaporin 4 (AQP4, 1:200, Santa Cruz Biotechnology, Santa Cruz, CA, www.scbt.com) and glial fibrillary acidic protein (GFAP) (1:1,000, Millipore, Bedford, MA, www.emdmillipore.com) antibodies. Brain sections were incubated with antibodies against occludin (1:200, Life Technologies, Carlsbad, CA), zonula occludens-1 (ZO-1, 1:200, Life Technologies, Carlsbad, CA), CD31 (1:200, R&D Systems, Minneapolis, MN, www.rndsystems.com), AQP4 (1:200, Santa Cruz Biotechnology, Santa Cruz, CA), and GFAP (1:1,000, Millipore, Bedford, MA) overnight at 4°C. After rinsing three times with PBS, the cells and sections were incubated with fluorescence conjugated secondary antibodies for 1 hour at room temperature. AQP4 staining for cultured astrocytes was computed as mean IOD. For ZO-1, or occludin staining, the gap length was presented as percentage (%) of whole tight junction staining. Similarly, six fields were randomly selected from each section and the data were analyzed by Image Pro Plus 6.0 software. For apoptosis analysis, apoptotic astrocytes were stained with GFAP and then counterstained with TUNEL using an in situ cell death detection kit (Roche, Penzberg, Germany, www.roche.com). TUNEL-positive cells merged with GFAP signal were counted as apoptotic astrocytes. Quantification of those cells was analyzed along the ischemic penumbra area. Real-Time PCR Analysis Total RNA from microglial cells and brain tissue samples was isolated using TRIzol Reagent (Life Technologies, Carlsbad, CA). The concentration of RNA was measured by a spectrophotometer (NanoDrop1000, Thermo Fisher, Wilmington, DE, www.thermofisher.com) followed by a reverse transcription process using PrimeScript RT reagent kit (TaKaRa, Dalian, China, www.takara.com.cn). The cDNA was used to perform Real-Time PCR by SYBR Premix Ex Tag Kit (TaKaRa, DaLian, China). A two stage RT-PCR amplification reaction was performed under the following conditions: 95°C for 30 seconds followed by 40 cycles of 95°C for 5 seconds and 60°C for 30 seconds. The primer sequences are listed in Table 1. Real-time PCR primers Gene . Forward primer (5′–3′) . Reverse primer (5′–3′) . Amplicon size (bp) . IL-6 TCTATACCACTTCACAAGTCGGA GAATTGCCATTGCACAACTCTTT 88 IL-1β GCAACTGTTCCTGAACTCAACT ATCTTTTGGGGTCCGTCAACT 89 TNF-α GGAACACGTCGTGGGATAATG GGCAGACTTTGGATGCTTCTT 213 AQP4 CTGGAGCCAGCATGAATCCAG TTCTTCTCTTCTCCACGGTCA 310 GAPDH AGGTCGGTGTGAACGGATTTG TGTAGACCATGTAGTTGAGGTCA 123 Gene . Forward primer (5′–3′) . Reverse primer (5′–3′) . Amplicon size (bp) . IL-6 TCTATACCACTTCACAAGTCGGA GAATTGCCATTGCACAACTCTTT 88 IL-1β GCAACTGTTCCTGAACTCAACT ATCTTTTGGGGTCCGTCAACT 89 TNF-α GGAACACGTCGTGGGATAATG GGCAGACTTTGGATGCTTCTT 213 AQP4 CTGGAGCCAGCATGAATCCAG TTCTTCTCTTCTCCACGGTCA 310 GAPDH AGGTCGGTGTGAACGGATTTG TGTAGACCATGTAGTTGAGGTCA 123 Abbreviations: AQP4, aquaporin-4; GAPDH, reduced glyceraldehyde-phosphate dehydrogenase; IL-6, interleukin-6; IL-1β, interleukin-1β; TNF-α, tumor necrosis factor-α. Open in new tab Real-time PCR primers Gene . Forward primer (5′–3′) . Reverse primer (5′–3′) . Amplicon size (bp) . IL-6 TCTATACCACTTCACAAGTCGGA GAATTGCCATTGCACAACTCTTT 88 IL-1β GCAACTGTTCCTGAACTCAACT ATCTTTTGGGGTCCGTCAACT 89 TNF-α GGAACACGTCGTGGGATAATG GGCAGACTTTGGATGCTTCTT 213 AQP4 CTGGAGCCAGCATGAATCCAG TTCTTCTCTTCTCCACGGTCA 310 GAPDH AGGTCGGTGTGAACGGATTTG TGTAGACCATGTAGTTGAGGTCA 123 Gene . Forward primer (5′–3′) . Reverse primer (5′–3′) . Amplicon size (bp) . IL-6 TCTATACCACTTCACAAGTCGGA GAATTGCCATTGCACAACTCTTT 88 IL-1β GCAACTGTTCCTGAACTCAACT ATCTTTTGGGGTCCGTCAACT 89 TNF-α GGAACACGTCGTGGGATAATG GGCAGACTTTGGATGCTTCTT 213 AQP4 CTGGAGCCAGCATGAATCCAG TTCTTCTCTTCTCCACGGTCA 310 GAPDH AGGTCGGTGTGAACGGATTTG TGTAGACCATGTAGTTGAGGTCA 123 Abbreviations: AQP4, aquaporin-4; GAPDH, reduced glyceraldehyde-phosphate dehydrogenase; IL-6, interleukin-6; IL-1β, interleukin-1β; TNF-α, tumor necrosis factor-α. Open in new tab Western Blot Analysis The brain tissue samples and astrocytes were collected and sonicated in homogenizing buffer (RIPA with protease cocktail inhibitor, phosphatase inhibitor, and Phenylmethanesulfonyl fluoride). The homogenate was centrifuged at 14,000g and the pellets were discarded. Protein concentrations were determined using a BCA kit (Thermo Scientific, Waltham, MA) and equal amounts of the samples were loaded onto 10% resolving gel for electrophoresis. Proteins were transferred to a nitrocellulose membrane (GE Healthcare Life Sciences, Pittsburgh, PA, www.gelifesciences.com) and blocked with 5% skim milk, then membranes were incubated with primary antibodies against AQP4 (1:500, Santa Cruz Biotechnology, Santa Crus, CA), phospho-p38 (p-p38, 1:1,000, Cell Signaling Technology, Danvers, MA, www.cellsignal.com), p38 (1:1,000, Cell Signaling Technology, Danvers, MA, www.cellsignal.com), phospho-stress-activated protein kinase/JNK (1:2,000, Cell Signaling Technology, Danvers, MA, www.cellsignal.com), stress-activated protein kinase/JNK(1:1,000, Cell Signaling Technology, Danvers, MA, www.cellsignal.com), phospho-ERK (1:2,000, Cell Signaling Technology, Danvers, MA, www.cellsignal.com), ERK (1:1,000, Cell Signaling Technology, Danvers, MA, www.cellsignal.com), and β-actin (1:1,000, Santa Cruz Biotechnology, Santa Cruz, CA) overnight at 4°C. After washing three times with TBST buffer, the membranes were incubated with HRP-conjugated secondary antibody and then reacted with an enhanced chemiluminescence substrate (Pierce, Rockford, IL, www.piercenet.com). The results of chemiluminescence were recorded with an imaging system (Bio-Rad, Hercules, CA, www.bio-rad.com). Preparation of MSCs Conditional Medium and Microglia Conditional Medium MSCs cultured in a 10 cm dish (Corning Incorporated, Corning, NY, www.corning.com) at a density of 1.5 × 106 cells were used for this experiment, fresh medium was added and followed by 24 hours of incubation at 37°C. The medium was collected and filtered for the following experiments. The BV-2 cells, immortalized mouse microglial cells, were plated in six-well plates and allowed to grow to 70% confluence (2 or 3 days), and then MSCs conditional medium (MCM) or fresh DMEM medium containing 10% FBS was added to each well and incubated for 24 hours. The medium was then renewed and supplemented with lipopolysaccharide (LPS) (500 ng/ml, Sigma-Aldrich, St. Louis, MO). After another 24 hours of incubation, the LPS-contained medium was replaced with a fresh medium. The medium was then collected and filtered for further experiments after 24 hours of incubation. Untreated cells were included as a control. These mediums were termed as microglia conditional medium (CM), LPS-activated microglia conditional medium (ACM), and MSC-treated LPS-activated microglia conditional medium (MACM). Microglia Total RNA Extraction BV-2 cells were plated in six-well plates and allowed to grow to 70% confluence (2 or 3 days). MCM or a fresh medium was added to each well and incubated for 24 hours. The medium was then renewed and supplemented with LPS (500 ng/ml). After another 24 hours of incubation, cells were washed with PBS and Trizol Reagent (Life Technologies, Carlsbad, CA) was added to each well for the extraction of total RNA. Astrocyte Isolation and Treatment Protocol Astrocytes were prepared from the C57BL/6 mice at the stage of P0 described previously [27]. Before experimental treatments, astrocyte cultures were passaged once in six-well plates. Cells were allowed to reach 90% confluence. CM, ACM, or MACM was added to each well and allowed to incubate for 24 hours. In mitogen-activated protein kinase inhibition studies, p38 and JNK inhibitors (SB239063 and SP600125, Gene Operation, Ann Arbor, MI, www.geneoperation.com) were added at the concentration of 10 µM with ACM. Untreated cells were included as a control. Astrocyte and bEnd.3 Cell Coculturing and Tight Junction Protein Expression in bEnd.3 Cells Astrocytes at a density of 3 × 105 cells per well were plated into six-well plates and then siRNA of AQP4 was added followed by 2 days of incubation. Then, ACM was added to each well with or without AQP4 knockdown. One day later, these treated astrocytes were cocultured with bEnd.3 (ATCC, Manassas, VA, www.atcc.org) cells plated on the trans-well membranes of an inner well at a density of 2 × 105/cm2. After 3 days of coculture, protein of bEnd.3 cells was harvested for Western blot examination. Statistical Analysis Data were expressed as mean ± SD. Statistical analysis was performed by SPSS for both parametric and nonparametric comparisons. Differences with p values < .05 were considered significant. Results MSCs Identification and Transplantation The cultured cells demonstrated a typical spindle-shaped morphology (Fig. 1A). Flow cytometry analysis showed that more than 99% of cells were CD29 and CD90 positive while less than 1% were CD31 and CD45 positive, which was highly consistent with previous studies [22]. For cell tracking, we labeled MSCs with AD-GFP and examined the green fluorescent protein (GFP) positive cells 3 days after transplantation. The results suggest that a considerable number of transplanted cells were detectable (Fig. 1B). We also labeled the cells with Dil and found the nuclei of the injected cells were clear and merged with Dil dye at day 3 after tMCAO (Supporting Information Fig. S1). Both results suggested that transplanted cells were still alive. Open in new tabDownload slide Mesenchymal stem cells (MSCs) isolation, identification, and injection. (A): MSCs isolation and identification. Cultured cells showed typically spindle-shaped morphology under phase-contrast microscopy (a). Flow cytometry analysis depicted that cultured cells were positive for CD29 (b) and CD90 (c), and negative for CD31 (e) and CD45 (f). (B): Survival of MSCs after injection. Green fluorescent protein (GFP+) MSCs were located in the ischemic hemisphere after 3 days of injection (a). GFP+ cells revealed the survival of transplanted MSCs (b). Scale bar = (A) 200 µm; (B) a, 500 µm, b, 25 µm. Open in new tabDownload slide Mesenchymal stem cells (MSCs) isolation, identification, and injection. (A): MSCs isolation and identification. Cultured cells showed typically spindle-shaped morphology under phase-contrast microscopy (a). Flow cytometry analysis depicted that cultured cells were positive for CD29 (b) and CD90 (c), and negative for CD31 (e) and CD45 (f). (B): Survival of MSCs after injection. Green fluorescent protein (GFP+) MSCs were located in the ischemic hemisphere after 3 days of injection (a). GFP+ cells revealed the survival of transplanted MSCs (b). Scale bar = (A) 200 µm; (B) a, 500 µm, b, 25 µm. MSCs Therapy Improved Neurological Outcomes, Attenuated Brain Edema, and BBB Leakage To explore the effects of MSCs on neurological outcomes, we examined the neurological deficits at 1 and 3 days after tMCAO using mNSS. We found that the neurological deficits in the MSCs group of mice were significantly attenuated compared to the PBS or HUVEC group at days 1 and 3 (Fig. 2A, p < .05). Water content was measured to determine the brain edema. We found that the water content of the ipsilateral hemisphere in MSC-treated mice was significantly lower compared to the control mice after 1 and 3 days of MCAO (p < .05). In contrast, no significant difference was observed between PBS and HUVEC group. To evaluate BBB permeability after ischemic brain injury, EB and IgG protein extravasation were measured. We demonstrated that MSCs remarkably reduced the EB leakage in the ipsilateral side, which indicated that BBB integrity was protected by MSCs treatment (Fig. 2B, p < .05). Further study confirmed that extravasated IgG was significantly increased after tMCAO while it was markedly decreased in the MSC-treated mice compared to the controls (Fig. 2C, p < .05). To investigate the mechanism of BBB disruption, we analyzed the localization of occludin and ZO-1 in cerebral vascular structures using CD31/occludin and CD31/ZO-1 double staining. Confocal microscopy analysis showed that the occludin and ZO-1 positive staining was continuously located on the endothelial cell margin of cerebral microvessels in sham mice; this continuity was disrupted by ischemic injury (Fig. 2D). It was noted that this process was reversed and gap formation was greatly reduced in the MSC-treated mice after tMCAO (p < .05). Open in new tabDownload slide MSCs improved neurological outcomes and attenuated edema formation in mice following transient middle cerebral artery occlusion (tMCAO) via maintaining the integrity of the blood-brain barrier. (A): MSCs improved neurological outcomes and reduced brain swelling. MSCs significantly ameliorated neurological outcomes (a) and reduced the brain water content (b) in the ischemic hemisphere at both days 1 and 3 when compared with the PBS or HUVEC group. (B): MSCs decreased the extravasation of Evans Blue (EB). Photographs represent the perfused brains after EB injection (a–f). The EB dye was rare in sham mice (a, d), the amount of EB increased after tMCAO in PBS-treated mice (b, e). However, the increase was reduced in MSC treated mice (c, f). Extravasated EB was analyzed by a spectrophotometer at 610 nm (g). (C): IgG staining displayed that IgG protein leaked into brain tissue in sham (a), PBS treated mice (b), and MSC-treated mice (c) at day 3. The images in square frames of graphs (a, b, c) were amplified as images (d, e, h), respectively. The insets were higher amplification, scale bar = 200 µm. Quantitative analysis of leaked IgG protein, less IgG protein leaked into brain tissue in MSC-treated mice compared to PBS treated mice (g). Data are mean ± SD, n = 6 per group. (D): MSCs reversed gap formation of occludin and ZO-1. Sections from ischemic penumbra were stained for occludin and ZO-1 (green), and then costained with endothelial marker CD31 (red). Nuclei were stained with 4,6-diamidino-2-phenylindole. Three-dimension reconstruction of confocal microscopy images showed continuous and linear labeling of occludin and ZO-1 along the vessels in the sham group (a, d). In contrast, discontinuous labeling and gap formation were observed in ipsilateral brains following 3 days of tMCAO (b, e, white arrows). MSCs significantly reduced gap formation and maintained the integrity of occludin and ZO-1 (c, f). Data are mean ± SD, *, p < .05; **, p < .01. Abbreviations: HUVEC, human umbilical vein endothelial cell; PBS, phosphate buffered saline; MSCs, mesenchymal stem cells; ZO-1, zonula occludens-1. Open in new tabDownload slide MSCs improved neurological outcomes and attenuated edema formation in mice following transient middle cerebral artery occlusion (tMCAO) via maintaining the integrity of the blood-brain barrier. (A): MSCs improved neurological outcomes and reduced brain swelling. MSCs significantly ameliorated neurological outcomes (a) and reduced the brain water content (b) in the ischemic hemisphere at both days 1 and 3 when compared with the PBS or HUVEC group. (B): MSCs decreased the extravasation of Evans Blue (EB). Photographs represent the perfused brains after EB injection (a–f). The EB dye was rare in sham mice (a, d), the amount of EB increased after tMCAO in PBS-treated mice (b, e). However, the increase was reduced in MSC treated mice (c, f). Extravasated EB was analyzed by a spectrophotometer at 610 nm (g). (C): IgG staining displayed that IgG protein leaked into brain tissue in sham (a), PBS treated mice (b), and MSC-treated mice (c) at day 3. The images in square frames of graphs (a, b, c) were amplified as images (d, e, h), respectively. The insets were higher amplification, scale bar = 200 µm. Quantitative analysis of leaked IgG protein, less IgG protein leaked into brain tissue in MSC-treated mice compared to PBS treated mice (g). Data are mean ± SD, n = 6 per group. (D): MSCs reversed gap formation of occludin and ZO-1. Sections from ischemic penumbra were stained for occludin and ZO-1 (green), and then costained with endothelial marker CD31 (red). Nuclei were stained with 4,6-diamidino-2-phenylindole. Three-dimension reconstruction of confocal microscopy images showed continuous and linear labeling of occludin and ZO-1 along the vessels in the sham group (a, d). In contrast, discontinuous labeling and gap formation were observed in ipsilateral brains following 3 days of tMCAO (b, e, white arrows). MSCs significantly reduced gap formation and maintained the integrity of occludin and ZO-1 (c, f). Data are mean ± SD, *, p < .05; **, p < .01. Abbreviations: HUVEC, human umbilical vein endothelial cell; PBS, phosphate buffered saline; MSCs, mesenchymal stem cells; ZO-1, zonula occludens-1. MSCs Attenuated Astrocyte Apoptosis and Suppressed AQP4 Upregulation in the Ischemic Penumbra Region Astrocytes are very important for maintaining the integrity of the BBB [28]. To examine whether MSCs therapy after tMCAO had an optimistic effect on astrocyte survival, we performed TUNEL/GFAP double staining. No apoptosis cells were found in control mice while ischemia induced a significant increase in apoptotic astrocytes (Fig. 3A). Interestingly, the increase in apoptotic astrocytes following tMCAO was significantly reduced in MSC treated mice (p < .01). Brain injury-induced AQP4 upregulation contributes to astrocyte swelling, which leads to an apoptosis of astrocytes and to the breakdown of the BBB [12]. To explore the potential relationship between BBB disruption and AQP4 expression, we examined the AQP4 at mRNA and protein levels. Results showed that AQP4 expression was upregulated following 1 and 3 days of tMCAO. MSCs therapy significantly reduced the AQP4 upregulation at day 3 (Fig. 3B). We further found AQP4 upregulation was mainly located in the ischemic peri-focal region. Brain injury-induced AQP4 upregulation was significantly reduced in the MSC-treated mice (Fig. 3C). Open in new tabDownload slide MSCs reduced apoptotic astrocytes and AQP4 upregulation in vivo. (A): MSCs reduced apoptotic astrocytes. The analyzed area was displayed in 3C, square box a. Few TUNEL-positive cells were observed in the sham group (a), fewer TUNEL-positive astrocytes were seen in the MSCs treatment group (c) than the PBS group (b, white arrows). Quantification of TUNEL-positive astrocytes showed the decrease of apoptotic astrocytes in MSC-treated mice. Scale bar = 50 µm. (B): MSCs significantly downregulated AQP4 expression at day 3 in mRNA and protein levels. Bar graph showed changes of the AQP4 mRNA expression at days 1 and 3. MSCs significantly reduced AQP4 gene expression at day 3 after transient middle cerebral artery occlusion (tMCAO) (a). Western blot revealed that upregulation of AQP4 was inhibited by MSCs after tMCAO; the difference was significant at day 3 (b). (C): Schematic diagram illustrated the area where AQP4 immunofluorescence and TUNEL assay were analyzed. The dotted line indicates infarction border. The square box was chosen as ischemic periphery for AQP4 expression and TUNEL-positive astrocyte analysis (a). AQP4 expression was upregulated in ischemic penumbra following tMCAO (c, white arrows) compared to sham mice (b) and MSCs restrained its upregulation (d). Scale bar = 50 µm. Data are mean ± SD, n = 6 per group, **, p < .01. Abbreviations: AQP4, aquaporin-4; PBS, phosphate buffered saline; MSCs, mesenchymal stem cells. Open in new tabDownload slide MSCs reduced apoptotic astrocytes and AQP4 upregulation in vivo. (A): MSCs reduced apoptotic astrocytes. The analyzed area was displayed in 3C, square box a. Few TUNEL-positive cells were observed in the sham group (a), fewer TUNEL-positive astrocytes were seen in the MSCs treatment group (c) than the PBS group (b, white arrows). Quantification of TUNEL-positive astrocytes showed the decrease of apoptotic astrocytes in MSC-treated mice. Scale bar = 50 µm. (B): MSCs significantly downregulated AQP4 expression at day 3 in mRNA and protein levels. Bar graph showed changes of the AQP4 mRNA expression at days 1 and 3. MSCs significantly reduced AQP4 gene expression at day 3 after transient middle cerebral artery occlusion (tMCAO) (a). Western blot revealed that upregulation of AQP4 was inhibited by MSCs after tMCAO; the difference was significant at day 3 (b). (C): Schematic diagram illustrated the area where AQP4 immunofluorescence and TUNEL assay were analyzed. The dotted line indicates infarction border. The square box was chosen as ischemic periphery for AQP4 expression and TUNEL-positive astrocyte analysis (a). AQP4 expression was upregulated in ischemic penumbra following tMCAO (c, white arrows) compared to sham mice (b) and MSCs restrained its upregulation (d). Scale bar = 50 µm. Data are mean ± SD, n = 6 per group, **, p < .01. Abbreviations: AQP4, aquaporin-4; PBS, phosphate buffered saline; MSCs, mesenchymal stem cells. MSCs Decreased Inflammatory Response and Astrocyte Apoptosis To determine whether the effect of MSCs on AQP4 expression after tMCAO is involved in the immunomodulatory influence of MSCs, we examined IL-1β, IL-6, and TNF-α mRNA expression both in vivo and in vitro. We demonstrated that IL-1β, IL-6, and TNF-α mRNA were increased at days 1 and 3 following tMCAO. The three cytokines significantly decreased in MSC-treated mice at day 3 compared to the controls (Fig. 4A, p < .05). These findings were confirmed by in vitro gene expression data using the BV-2 cell line (Fig. 4B, p < .05). We further used the conditional medium of BV-2 to confirm whether inflammatory cytokines contributed to the AQP4 upregulation in cultured astrocytes. We demonstrated that LPS-activated microglia conditional medium (ACM) induced AQP4 expression sharply compared to controls. The induction of AQP4 in the MACM was much less than that in the ACM group (Fig. 4C, p < .05). To determine the effect of AQP4 on astrocyte apoptosis, we used AQP4 siRNA to treat cultured astrocytes stimulated by ACM. We demonstrated that AQP4 knockdown reduced ACM induced astrocyte apoptosis (p < .01). Our results also showed that MACM induced less apoptosis of astrocytes when compared with ACM (Fig. 4D, p < .01). The AQP4 knocking down efficiency was examined by RT-PCR and immunofluorescence (Fig. 4E, p < .01). Coculture of astrocytes and bEnd.3 cells increased the occludin expression in bEnd.3 cells. ACM-treated astrocytes induced weaker occludin expression in bEnd.3 cells while AQP4 knockdown and ACM-treated astrocytes maintained occludin expression (Fig. 4F, p < .01). Open in new tabDownload slide MSCs diminished the upregulation of AQP4 by decreasing the production of inflammatory cytokines. (A): Modulation of inflammatory cytokine gene expression in vivo. The relative mRNA expression of IL-1β (a), IL-6 (b), and TNF-α (c) normalized to GAPDH was detected at days 1 and 3 following transient middle cerebral artery occlusion. The expression of IL-1β, IL-6, and TNF-α mRNA was significantly decreased in the MSC-treated group at day 3 compared to the PBS group, however, at day 1, only IL-1β was significantly reduced in the MSC-treated group compared to the PBS group. n = 6 per group. (B): Influence of MSCs conditional medium on proinflammatory cytokine expression by microglia. MSCs conditional medium attenuated LPS-induced IL-1β (a), IL-6 (b), and TNF-α (c). n = 3 per group. (C): MSC-treated microglia conditional medium induced more moderate upregulation of AQP4 in cultured astrocytes. Immunofluorescence staining showed that more than 95% of cultured cells were positive for glial fibrillary acidic protein (a), scale bar = 20 µm. The AQP4 expression was examined in control (b), CM (c), ACM (d), and MACM (e) treated astrocytes, scale bar = 100 µm. Quantifying the expression of AQP4 showed that ACM induced the fiercest upregulation of AQP4, which was significant when compared with CM and MACM and suggested that it was the inflammatory cytokines that were responsible for the induction of AQP4. (D): AQP4 knockdown reduced apoptosis in cultured astrocytes. Few apoptotic astrocytes (white arrows) were found in control (a) and CM-treated astrocytes (b). AQP4 knockdown (e) reduced ACM (c) induced astrocyte apoptosis. MACM (d) was also found to induce less apoptosis of astrocytes when compared with ACM (c), scale bar = 50 µm. (E): RT-PCR (a) and immunofluorescence (b, c) revealed that the AQP4 knockdown was efficient. (F): AQP4 knockdown maintained the expression of tight junction protein occludin in an inflammed environment. Coculture of astrocytes and bEnd.3 cells increased the expression of occludin in bEnd.3 cells (Coc) comparing to bEnd.3 cell alone (Con). ACM-treated astrocytes induced weaker expression of occludin in bEnd.3 cells (CA) while AQP4 knockdown as well as ACM treated astrocytes maintained occludin expression (KCA). n = 3 per group. Data are mean ± SD, *, p < .05; **, p < .01. Abbreviations: ACM, LPS-activated microglia conditional medium; AQP4, aquaporin-4; CM, microglia conditional medium; LPS, lipopolysaccharide; MACM, MSC-treated LPS-activated microglia conditional medium; MSCs, mesenchymal stem cells; PBS, phosphate buffered saline; IOD, integrated optical density. Open in new tabDownload slide MSCs diminished the upregulation of AQP4 by decreasing the production of inflammatory cytokines. (A): Modulation of inflammatory cytokine gene expression in vivo. The relative mRNA expression of IL-1β (a), IL-6 (b), and TNF-α (c) normalized to GAPDH was detected at days 1 and 3 following transient middle cerebral artery occlusion. The expression of IL-1β, IL-6, and TNF-α mRNA was significantly decreased in the MSC-treated group at day 3 compared to the PBS group, however, at day 1, only IL-1β was significantly reduced in the MSC-treated group compared to the PBS group. n = 6 per group. (B): Influence of MSCs conditional medium on proinflammatory cytokine expression by microglia. MSCs conditional medium attenuated LPS-induced IL-1β (a), IL-6 (b), and TNF-α (c). n = 3 per group. (C): MSC-treated microglia conditional medium induced more moderate upregulation of AQP4 in cultured astrocytes. Immunofluorescence staining showed that more than 95% of cultured cells were positive for glial fibrillary acidic protein (a), scale bar = 20 µm. The AQP4 expression was examined in control (b), CM (c), ACM (d), and MACM (e) treated astrocytes, scale bar = 100 µm. Quantifying the expression of AQP4 showed that ACM induced the fiercest upregulation of AQP4, which was significant when compared with CM and MACM and suggested that it was the inflammatory cytokines that were responsible for the induction of AQP4. (D): AQP4 knockdown reduced apoptosis in cultured astrocytes. Few apoptotic astrocytes (white arrows) were found in control (a) and CM-treated astrocytes (b). AQP4 knockdown (e) reduced ACM (c) induced astrocyte apoptosis. MACM (d) was also found to induce less apoptosis of astrocytes when compared with ACM (c), scale bar = 50 µm. (E): RT-PCR (a) and immunofluorescence (b, c) revealed that the AQP4 knockdown was efficient. (F): AQP4 knockdown maintained the expression of tight junction protein occludin in an inflammed environment. Coculture of astrocytes and bEnd.3 cells increased the expression of occludin in bEnd.3 cells (Coc) comparing to bEnd.3 cell alone (Con). ACM-treated astrocytes induced weaker expression of occludin in bEnd.3 cells (CA) while AQP4 knockdown as well as ACM treated astrocytes maintained occludin expression (KCA). n = 3 per group. Data are mean ± SD, *, p < .05; **, p < .01. Abbreviations: ACM, LPS-activated microglia conditional medium; AQP4, aquaporin-4; CM, microglia conditional medium; LPS, lipopolysaccharide; MACM, MSC-treated LPS-activated microglia conditional medium; MSCs, mesenchymal stem cells; PBS, phosphate buffered saline; IOD, integrated optical density. P38 Was Involved in AQP4 Upregulation in Astrocytes After Microglia Conditional Medium Stimulation To study the potential mechanism of AQP4 upregulation in astrocytes stimulated by CM, ACM, and MACM; p38, JNK, and ERK1/2 were examined 24 hours later by Western blot analysis. We found that p38 and JNK phosphorylation increased after CM, ACM, and MACM treatment. The increase was milder in the MACM group compared to the ACM group (p < .05). At the same time, the increase was also significant in ACM-treated group compared to the CM-treated group (Fig. 5A, p < .05). To further elucidate whether JNK, p38, or both pathways were involved in AQP4 upregulation, we compared the effect of p38 and JNK inhibitors (SB239063 and SP600125) on AQP4 expression. As shown in Figure 5B, 10 µM of SB203580, but not 10 µM of SP600125 significantly reduced AQP4 expression (p < .05). The validity of these two inhibitors was confirmed by Western blot analysis (Fig. 5C). We then detected p38 activation and AQP4 expression in vivo after tMCAO. The results revealed that SB203580 effectively dampened AQP4 upregulation after tMCAO. The effectiveness of the p38 inhibitor was confirmed (Fig. 5D, p < .05). Open in new tabDownload slide Inflammatory cytokines upregulated AQP4 via p38 but not the JNK and ERK signaling pathways. (A): Inflammatory cytokines activated p38 and JNK mitogen-activated protein kinase (MAPK) pathways. Western blot demonstrated that ACM could efficiently phosphorylate p38 and JNK (a, b) compared with CM, while the phosphorylation decreased in MCAM-treated astrocytes when compared with ACM. However, there were no detectable changes of ERK1/2 between groups. n = 3 per group. (B): p38 MAPK signaling pathway was responsible for the upregulation of AQP4. AQP4 was highly upregulated after ACM stimulation (b) compared to control (Con) (a). P38 inhibitor (SB239063) significantly diminished this tendency (d). In contrast, JNK inhibitor (SP600125) exhibited no significant effects on AQP4 expression (c). n = 3 per group, scale bar = 100 µm. (C): The validity of inhibitors was confirmed by Western blot 24 hours after stimuli, which showed obvious inhibition of the activation of p38 (a) and JNK (b), n = 3 per group. (D): Preventing the activation of p38 by SB239063 reduced the upregulation of AQP4 in vivo. The p38 inhibitor was administered immediately after reperfusion. Western blot revealed that SB239063 sufficiently dampened the phosphorylation of p38 compared to the vehicle (a). The expression of AQP4 was also significantly inhibited in SB239063 treated mice when compared with vehicle treated mice (b). n = 6 per group. Data are mean ± SD, *, p < .05. Abbreviations: ACM, lipopolysaccharide-activated microglia conditional medium; AQP4, aquaporin-4; MACM, MSC-treated lipopolysaccharide-activated microglia conditional medium; SB, SB239063; SP, SP600125; IOD, integrated optical density. Open in new tabDownload slide Inflammatory cytokines upregulated AQP4 via p38 but not the JNK and ERK signaling pathways. (A): Inflammatory cytokines activated p38 and JNK mitogen-activated protein kinase (MAPK) pathways. Western blot demonstrated that ACM could efficiently phosphorylate p38 and JNK (a, b) compared with CM, while the phosphorylation decreased in MCAM-treated astrocytes when compared with ACM. However, there were no detectable changes of ERK1/2 between groups. n = 3 per group. (B): p38 MAPK signaling pathway was responsible for the upregulation of AQP4. AQP4 was highly upregulated after ACM stimulation (b) compared to control (Con) (a). P38 inhibitor (SB239063) significantly diminished this tendency (d). In contrast, JNK inhibitor (SP600125) exhibited no significant effects on AQP4 expression (c). n = 3 per group, scale bar = 100 µm. (C): The validity of inhibitors was confirmed by Western blot 24 hours after stimuli, which showed obvious inhibition of the activation of p38 (a) and JNK (b), n = 3 per group. (D): Preventing the activation of p38 by SB239063 reduced the upregulation of AQP4 in vivo. The p38 inhibitor was administered immediately after reperfusion. Western blot revealed that SB239063 sufficiently dampened the phosphorylation of p38 compared to the vehicle (a). The expression of AQP4 was also significantly inhibited in SB239063 treated mice when compared with vehicle treated mice (b). n = 6 per group. Data are mean ± SD, *, p < .05. Abbreviations: ACM, lipopolysaccharide-activated microglia conditional medium; AQP4, aquaporin-4; MACM, MSC-treated lipopolysaccharide-activated microglia conditional medium; SB, SB239063; SP, SP600125; IOD, integrated optical density. Knockdown of AQP4 Reduced Brain Edema, BBB Disruption, and Astrocyte Apoptosis To further investigate the effects of AQP4 on edema formation, BBB disruption, and astrocyte apoptosis, siRNAs against AQP4 were injected after tMCAO. The knockdown efficiency was confirmed by immunofluorescence and Western blot analysis (Fig. 6A, p < .01). Our results showed that the knockdown of AQP4 reduced brain edema (Water content: PBS, 84.6% ± 0.6%, Scrambled, 84.4% ± 0.4%, siAQP4, 81.4% ± 0.4%; p < .05, PBS vs. siAQP4, Scrambled vs. siAQP4), IgG leakage, and astrocyte apoptosis (Fig. 6B, 6C, p < .01) when compared with PBS or the scrambled siRNA control group. Open in new tabDownload slide The knockdown of AQP4 reduced ischemia-induced blood-brain barrier breakdown and astrocyte apoptosis. (A): The validity of AQP4 knockdown efficiency. Immunofluorescence and Western blot analysis demonstrated that siAQP4 significantly reduced the expression of AQP4 (a, b, c, d), scale bar = 100 µm. (B): IgG staining displayed that IgG protein leaked into brain tissue in PBS-treated mice (a), scrambled siRNA-treated mice (b), and siAQP4-treated mice (c) at day 3. The insets were higher amplification, scale bar = 200 µm. Quantitative analysis of leaked IgG protein, less IgG protein leaked into brain tissue in siAQP4-treated mice compared to scrambled siRNA treated mice or PBS group. (C): The knockdown of AQP4 reduced apoptotic astrocytes. Fewer TUNEL-positive astrocytes (white arrows) were seen in the siAQP4 treatment group (c) than the PBS (a) or scrambled siRNA group (b). Quantification of TUNEL-positive astrocytes showed the decrease of apoptotic astrocytes in siAQP4 treated mice. Scale bar = 50 µm. n = 6 per group. Scrambled: scrambled siRNA. Data are mean ± SD, **, p < .01. Abbreviations: AQP4, aquaporin-4; IOD, integrated optical density. Open in new tabDownload slide The knockdown of AQP4 reduced ischemia-induced blood-brain barrier breakdown and astrocyte apoptosis. (A): The validity of AQP4 knockdown efficiency. Immunofluorescence and Western blot analysis demonstrated that siAQP4 significantly reduced the expression of AQP4 (a, b, c, d), scale bar = 100 µm. (B): IgG staining displayed that IgG protein leaked into brain tissue in PBS-treated mice (a), scrambled siRNA-treated mice (b), and siAQP4-treated mice (c) at day 3. The insets were higher amplification, scale bar = 200 µm. Quantitative analysis of leaked IgG protein, less IgG protein leaked into brain tissue in siAQP4-treated mice compared to scrambled siRNA treated mice or PBS group. (C): The knockdown of AQP4 reduced apoptotic astrocytes. Fewer TUNEL-positive astrocytes (white arrows) were seen in the siAQP4 treatment group (c) than the PBS (a) or scrambled siRNA group (b). Quantification of TUNEL-positive astrocytes showed the decrease of apoptotic astrocytes in siAQP4 treated mice. Scale bar = 50 µm. n = 6 per group. Scrambled: scrambled siRNA. Data are mean ± SD, **, p < .01. Abbreviations: AQP4, aquaporin-4; IOD, integrated optical density. Discussion We focused on the protective effects of MSCs on BBB permeability since maintaining BBB integrity is pivotal for reducing secondary brain injury following cerebral ischemia. The disturbance of BBB function is related to several pathological changes in the CNS, such as stroke attack, head trauma, Parkinson’s disease, and Alzheimer’s disease [28]. BBB disruption directly resulting from vascular endothelial damage potentially aggravates vasogenic edema and worsens prognosis [29]. Conditional medium from MSCs preserved vascular endothelial integrity in pulmonary endothelial cells by preserving adherens junction [30]. This finding was also confirmed in human umbilical vascular endothelial cell studies [31]. Another report showed that TIMP3 released by MSCs could stabilize BBB integrity following traumatic brain injury [32]. In this study, we demonstrated that MSCs therapy is beneficial to maintaining BBB integrity following ischemic stroke. Our results support that this was due to an attenuated inflammatory response and reduced apoptosis of astrocytes. Proinflammatory cytokines were capable of disrupting the epithelial barrier by decreasing tight junction protein expression [33, 34]. Therefore, we believe that the effect of MSCs on BBB integrity was from the suppression of proinflammatory cytokine production, which directly resulted in less gap formation of tight junction protein. In addition, astrocyte-endothelial crosstalk is important to maintain the function of the BBB. Astrocytes contributed to BBB formation by inducing the BBB phenotypic characteristics of endothelial cells [28, 35]. Furthermore, astrocytes also showed a capability to promote tight junction formation in the brain capillary endothelium [36, 37]. It is noted that different type of cells such as neurons from fetal brain, endothelial progenitor cells (EPCs) from umbilical cord blood, neural stem cells (NSCs), neural progenitor cells, and embryonic stem cells have been reported to be effective in the treatment of experimental stroke animals. The therapeutic benefits of these cells are probably due to neuroprotection and regeneration by promoting neurogenesis and angiogenesis [38, 37, 39, 40]. However, each cell type has its own properties and molecular mechanism in the repair of ischemic stroke. For example, EPCs could increase vascular endothelial growth factor level and promote angiogenesis [23]. NSCs were capable of differentiating into neurons and contributed to neurogenesis, thus inducing regeneration of damaged brain tissue. Moreover, NSCs transplantation also showed anti-inflammatory properties in hemorrhagic stroke and decreased apoptosis and brain edema [41]. We chose rat bone marrow-derived MSCs because we want to explore therapeutic effects of exogenous stem cell on mice after tMCAO. Previous studies revealed that human derived stem cells had positive effects on ischemic mice or rat [42, 43]. In addition, bone marrow-derived MSCs had low immunogenicity and had immunosuppressive effect on graft-versus-host disease [44, 45]. We found that MSC treatment reduced apoptotic astrocytes in ischemia, which we believe significantly contributed to the BBB integrity. In addition, our in vitro data showed that astrocytes were capable of inducing occludin over-expression in bEnd.3 cells, indicating that astrocytes do regulate the function of endothelial cells. Downregulated AQP4 expression could explain the beneficial effects of MSCs on astrocytes in the ischemic mouse brain. Brain water content and swelling of astrocyte foot processes in AQP4-deficient mice were significantly reduced in a model of brain edema [12]. Further study showed that reduced astrocyte edema at an early time point subsequently led to fewer apoptotic astrocytes in an oxidative stress model [46]. We demonstrated that MSCs effectively downregulated AQP4 expression and reduced astrocyte apoptosis after ischemic injury, suggesting that AQP4 was a key target in the induction of astrocyte apoptosis. Our in vitro data also showed that AQP4 knockdown reduced apoptotic astrocytes. The mechanism of AQP4 downregulation in astrocytes could be due to the immunomodulatory functions of MSCs. Here, we showed MSCs significantly reduced inflammatory cytokine production both in vivo and in vitro. We found that the time course of inflammatory cytokine release and AQP4 expression were associated, implying an intrinsic relationship between them. We further explored the relationship between inflammatory cytokines, AQP4, and p38. The ACM could increase AQP4 expression via activating p38 and JNK signaling pathways while the increase of AQP4 in CM-treated astrocytes was much less, suggesting that it is the inflammatory cytokines (IL-1β, IL6, and TNF-α) released by the microglia that induced the upregulation of AQP4. AQP4 upregulation was more moderate in MCAM-treated astrocytes further confirming that inflammatory cytokines were responsible for AQP4 upregulation. It was noted that p38 and JNK activation occurred in activated astrocytes. Nevertheless, our study displayed that only the p38 pathway was involved in AQP4 upregulation, suggesting that p38 played a crucial role in the mediation of AQP4 over-expression. However, the receptor that mediated the activation of p38 in response to inflammatory cytokines was not under investigation. IL-1 receptor type1 may be involved in this process [47]. Although a number of studies have shown that the p38 signaling pathway plays an important role in regulation of AQP4 expression [48], our study was the first to report that inflammatory cytokines took the responsibility for the upregulation of AQP4 by activating the p38 signaling pathway. Conclusions Our study shows for the first time that MSCs exert a potent regulating function on AQP4 expression in astrocytes, which consequently attenuates astrocyte apoptosis, ischemia-induced BBB disruption, and brain edema. The molecular mechanism of AQP4 upregulation was through p38 but not ERK1/2 or JNK signaling pathways in response to inflammatory cytokines. Based on these findings, we believe that the effect of MSCs on the ischemic brain is associated with the integrity of the BBB and edema formation, which represents a novel mechanism of MSC therapy (Supporting Information Fig. S2). Acknowledgments The study was supported by the National Key Basic Research Program of China, 973 Program, 2011CB504405 (G.Y.Y., Y.W.), the National Natural Science Foundation of China, U1232205 (G.Y.Y.) and 81371305 (Y.W.), Science and Technology Commission of Shanghai Municipality, 13140903500 (G.Y.Y.) and 13ZR1422600 (Z.J.Z.), Shanghai Jiao Tong University Foundation for technological innovation of major projects 2X190030021 (G.Y.Y.), and KC Wong Foundation (G.Y.Y.). We thank Meijie Qu for editorial assistance. Author Contributions G.T.: conception and design, provision of study material, collection and/or assembly of data, data analysis and interpretation, manuscript writing, and final approval of manuscript; Y.Liu: provision of study material, collection and assembly of data, data analysis and interpretation, revised manuscript writing, and final approval of manuscript; Z.Z.: conception and design; Y.Lu: provision of study material; Yang W., J.H., Y.Li, X.C., and X.G.: collection and/or assembly of data and administrative support; Yongting W. and G.Y.: conception and design, financial support, administrative support, provision of study material, data analysis and interpretation, manuscript writing, and final approval of manuscript. G.T. and Y.L. contributed equally to this article. Disclosure of Potential Conflicts of Interest The authors indicate no potential conflicts of interest. References 1 Song EC , Chu K, Jeong SW et al. Hyperglycemia exacerbates brain edema and perihematomal cell death after intracerebral hemorrhage . Stroke 2003 ; 34 : 2215 – 2220 . Google Scholar Crossref Search ADS PubMed WorldCat 2 Lin TN , He YY, Wu G et al. Effect of brain edema on infarct volume in a focal cerebral ischemia model in rats . Stroke 1993 ; 24 : 117 – 121 . Google Scholar Crossref Search ADS PubMed WorldCat 3 Beaumont A , Marmarou A, Hayasaki K et al. The permissive nature of blood brain barrier (BBB) opening in edema formation following traumatic brain injury . Acta Neurochir Suppl 2000 ; 76 : 125 – 129 . Google Scholar PubMed OpenURL Placeholder Text WorldCat 4 Huang J , Li Y, Tang Y et al. CXCR4 antagonist AMD3100 protects blood-brain barrier integrity and reduces inflammatory response after focal ischemia in mice . Stroke 2013 ; 44 : 190 – 197 . Google Scholar Crossref Search ADS PubMed WorldCat 5 Onda T , Honmou O, Harada K et al. Therapeutic benefits by human mesenchymal stem cells (hMSCs) and Ang-1 gene-modified hMSCs after cerebral ischemia . J Cereb Blood Flow Metab 2008 ; 28 : 329 – 340 . Google Scholar Crossref Search ADS PubMed WorldCat 6 Chan PH , Yang GY, Chen SF et al. Cold-induced brain edema and infarction are reduced in transgenic mice overexpressing CuZn-superoxide dismutase . Ann Neurol 1991 ; 29 : 482 – 486 . Google Scholar Crossref Search ADS PubMed WorldCat 7 Masada T , Hua Y, Xi G et al. Attenuation of intracerebral hemorrhage and thrombin-induced brain edema by overexpression of interleukin-1 receptor antagonist . J Neurosurg 2001 ; 95 : 680 – 686 . Google Scholar Crossref Search ADS PubMed WorldCat 8 Han X , Fink MP, Delude RL. Proinflammatory cytokines cause NO*-dependent and -independent changes in expression and localization of tight junction proteins in intestinal epithelial cells . Shock 2003 ; 19 : 229 – 237 . Google Scholar Crossref Search ADS PubMed WorldCat 9 Nielsen S , Nagelhus EA, Amiry-Moghaddam M et al. Specialized membrane domains for water transport in glial cells: High-resolution immunogold cytochemistry of aquaporin-4 in rat brain . J Neurosci 1997 ; 17 : 171 – 180 . Google Scholar Crossref Search ADS PubMed WorldCat 10 Hoffmann EK , Dunham PB. Membrane mechanisms and intracellular signalling in cell volume regulation . Int Rev Cytol 1995 ; 161 : 173 – 262 . Google Scholar Crossref Search ADS PubMed WorldCat 11 Fu X , Li Q, Feng Z et al. The roles of aquaporin-4 in brain edema following neonatal hypoxia ischemia and reoxygenation in a cultured rat astrocyte model . Glia 2007 ; 55 : 935 – 941 . Google Scholar Crossref Search ADS PubMed WorldCat 12 Manley GT , Fujimura M, Ma T et al. Aquaporin-4 deletion in mice reduces brain edema after acute water intoxication and ischemic stroke . Nat Med 2000 ; 6 : 159 – 163 . Google Scholar Crossref Search ADS PubMed WorldCat 13 Aoki-Yoshino K , Uchihara T, Duyckaerts C et al. Enhanced expression of aquaporin 4 in human brain with inflammatory diseases . Acta Neuropathol 2005 ; 110 : 281 – 288 . Google Scholar Crossref Search ADS PubMed WorldCat 14 Ortiz LA , Gambelli F, McBride C et al. Mesenchymal stem cell engraftment in lung is enhanced in response to bleomycin exposure and ameliorates its fibrotic effects . Proc Natl Acad Sci USA 2003 ; 100 : 8407 – 8411 . Google Scholar Crossref Search ADS PubMed WorldCat 15 Takai K , Hara J, Matsumoto K et al. Hepatocyte growth factor is constitutively produced by human bone marrow stromal cells and indirectly promotes hematopoiesis . Blood 1997 ; 89 : 1560 – 1565 . Google Scholar Crossref Search ADS PubMed WorldCat 16 Labouyrie E , Dubus P, Groppi A et al. Expression of neurotrophins and their receptors in human bone marrow . Am J Pathol 1999 ; 154 : 405 – 415 . Google Scholar Crossref Search ADS PubMed WorldCat 17 Constantin G , Marconi S, Rossi B et al. Adipose-derived mesenchymal stem cells ameliorate chronic experimental autoimmune encephalomyelitis . Stem Cells 2009 ; 27 : 2624 – 2635 . Google Scholar Crossref Search ADS PubMed WorldCat 18 Zappia E , Casazza S, Pedemonte E et al. Mesenchymal stem cells ameliorate experimental autoimmune encephalomyelitis inducing T-cell anergy . Blood 2005 ; 106 : 1755 – 1761 . Google Scholar Crossref Search ADS PubMed WorldCat 19 Keating A . How do mesenchymal stromal cells suppress T cells? Cell Stem Cell 2008 ; 2 : 106 – 108 . Google Scholar Crossref Search ADS PubMed WorldCat 20 Schafer S , Calas AG, Vergouts M et al. Immunomodulatory influence of bone marrow-derived mesenchymal stem cells on neuroinflammation in astrocyte cultures . J Neuroimmunol 2012 ; 249 : 40 – 48 . Google Scholar Crossref Search ADS PubMed WorldCat 21 Horwitz EM , Gordon PL, Koo WK et al. Isolated allogeneic bone marrow-derived mesenchymal cells engraft and stimulate growth in children with osteogenesis imperfecta: Implications for cell therapy of bone . Proc Natl Acad Sci USA 2002 ; 99 : 8932 – 8937 . Google Scholar Crossref Search ADS PubMed WorldCat 22 Fernandez Vallone VB , Romaniuk MA, Choi H et al. Mesenchymal stem cells and their use in therapy: What has been achieved? Differentiation 2013 ; 85 : 1 – 10 . Google Scholar Crossref Search ADS PubMed WorldCat 23 Li Q , Tang G, Xue S et al. Silica-coated superparamagnetic iron oxide nanoparticles targeting of EPCs in ischemic brain injury . Biomaterials 2013 ; 34 : 4982 – 4992 . Google Scholar Crossref Search ADS PubMed WorldCat 24 Badaut J , Ashwal S, Adami A et al. Brain water mobility decreases after astrocytic aquaporin-4 inhibition using RNA interference . J Cereb Blood Flow Metab 2011 ; 31 : 819 – 831 . Google Scholar Crossref Search ADS PubMed WorldCat 25 Li Y , Chopp M, Chen J et al. Intrastriatal transplantation of bone marrow nonhematopoietic cells improves functional recovery after stroke in adult mice . J Cereb Blood Flow Metab 2000 ; 20 : 1311 – 1319 . Google Scholar Crossref Search ADS PubMed WorldCat 26 Tanno H , Nockels RP, Pitts LH et al. Breakdown of the blood-brain barrier after fluid percussive brain injury in the rat. Part 1: Distribution and time course of protein extravasation . J Neurotrauma 1992 ; 9 : 21 – 32 . Google Scholar Crossref Search ADS PubMed WorldCat 27 Shao W , Zhang SZ, Tang M et al. Suppression of neuroinflammation by astrocytic dopamine D2 receptors via alphaB-crystallin . Nature 2013 ; 494 : 90 – 94 . Google Scholar Crossref Search ADS PubMed WorldCat 28 Abbott NJ , Ronnback L, Hansson E. Astrocyte-endothelial interactions at the blood-brain barrier . Nat Rev Neurosci 2006 ; 7 : 41 – 53 . Google Scholar Crossref Search ADS PubMed WorldCat 29 Unterberg AW , Stover J, Kress B et al. Edema and brain trauma . Neuroscience 2004 ; 129 : 1021 – 1029 . Google Scholar Crossref Search ADS PubMed WorldCat 30 Pati S , Gerber MH, Menge TD et al. Bone marrow derived mesenchymal stem cells inhibit inflammation and preserve vascular endothelial integrity in the lungs after hemorrhagic shock . PLoS One 2011 ; 6 : e25171 . Google Scholar Crossref Search ADS PubMed WorldCat 31 Pati S , Khakoo AY, Zhao J et al. Human mesenchymal stem cells inhibit vascular permeability by modulating vascular endothelial cadherin/beta-catenin signaling . Stem Cells Dev 2011 ; 20 : 89 – 101 . Google Scholar Crossref Search ADS PubMed WorldCat 32 Menge T , Zhao Y, Zhao J et al. Mesenchymal stem cells regulate blood-brain barrier integrity through TIMP3 release after traumatic brain injury . Sci Transl Med 2012 ; 4 : 161ra150 . Google Scholar Crossref Search ADS PubMed WorldCat 33 Al-Sadi RM , Ma TY. IL-1beta causes an increase in intestinal epithelial tight junction permeability . J Immunol 2007 ; 178 : 4641 – 4649 . Google Scholar Crossref Search ADS PubMed WorldCat 34 Yang R , Han X, Uchiyama T et al. IL-6 is essential for development of gut barrier dysfunction after hemorrhagic shock and resuscitation in mice . Am J Physiol Gastrointest Liver Physiol 2003 ; 285 : G621 – 629 . Google Scholar Crossref Search ADS PubMed WorldCat 35 Dehouck MP , Meresse S, Delorme P et al. An easier, reproducible, and mass-production method to study the blood-brain barrier in vitro . J Neurochem 1990 ; 54 : 1798 – 1801 . Google Scholar Crossref Search ADS PubMed WorldCat 36 Arthur FE , Shivers RR, Bowman PD. Astrocyte-mediated induction of tight junctions in brain capillary endothelium: An efficient in vitro model . Brain Res 1987 ; 433 : 155 – 159 . Google Scholar PubMed OpenURL Placeholder Text WorldCat 37 Lee SW , Kim WJ, Choi YK et al. SSeCKS regulates angiogenesis and tight junction formation in blood-brain barrier . Nat Med 2003 ; 9 : 900 – 906 . Google Scholar Crossref Search ADS PubMed WorldCat 38 Song M , Kim Y, Kim Y et al. MRI tracking of intravenously transplanted human neural stem cells in rat focal ischemia model . Neurosci Res 2009 ; 64 : 235 – 239 . Google Scholar Crossref Search ADS PubMed WorldCat 39 Jin K , Sun Y, Xie L et al. Comparison of ischemia-directed migration of neural precursor cells after intrastriatal, intraventricular, or intravenous transplantation in the rat . Neurobiol Dis 2005 ; 18 : 366 – 374 . Google Scholar Crossref Search ADS PubMed WorldCat 40 Chen J , Sanberg PR, Li Y et al. Intravenous administration of human umbilical cord blood reduces behavioral deficits after stroke in rats . Stroke 2001 ; 32 : 2682 – 2688 . Google Scholar Crossref Search ADS PubMed WorldCat 41 Lee ST , Chu K, Jung KH et al. Anti-inflammatory mechanism of intravascular neural stem cell transplantation in haemorrhagic stroke . Brain 2008 ; 131 : 616 – 629 . Google Scholar Crossref Search ADS PubMed WorldCat 42 Fan Y , Shen F, Frenzel T et al. Endothelial progenitor cell transplantation improves long-term stroke outcome in mice . Ann Neurol 2010 ; 67 : 488 – 497 . Google Scholar Crossref Search ADS PubMed WorldCat 43 Kelly S , Bliss TM, Shah AK et al. Transplanted human fetal neural stem cells survive, migrate, and differentiate in ischemic rat cerebral cortex . Proc Natl Acad Sci USA 2004 ; 101 : 11839 – 11844 . Google Scholar Crossref Search ADS PubMed WorldCat 44 Oh JY , Lee RH, Yu JM et al. Intravenous mesenchymal stem cells prevented rejection of allogeneic corneal transplants by aborting the early inflammatory response . Mol Ther 2012 ; 20 : 2143 – 2152 . Google Scholar Crossref Search ADS PubMed WorldCat 45 Nevruz O , Avcu F, Ural AU et al. Immunosuppressive effects of multipotent mesenchymal stromal cells on graft-versus-host disease in rats following allogeneic bone marrow transplantation . Turk J Haematol 2013 ; 30 : 256 – 262 . Google Scholar Crossref Search ADS PubMed WorldCat 46 Jou MJ , Peng TI, Reiter RJ et al. Visualization of the antioxidative effects of melatonin at the mitochondrial level during oxidative stress-induced apoptosis of rat brain astrocytes . J Pineal Res 2004 ; 37 : 55 – 70 . Google Scholar Crossref Search ADS PubMed WorldCat 47 Ito H , Yamamoto N, Arima H et al. Interleukin-1beta induces the expression of aquaporin-4 through a nuclear factor-kappaB pathway in rat astrocytes . J Neurochem 2006 ; 99 : 107 – 118 . Google Scholar Crossref Search ADS PubMed WorldCat 48 Nito C , Kamada H, Endo H et al. Involvement of mitogen-activated protein kinase pathways in expression of the water channel protein aquaporin-4 after ischemia in rat cortical astrocytes . J Neurotrauma 2012 ; 29 : 2404 – 2412 . Google Scholar Crossref Search ADS PubMed WorldCat © 2014 AlphaMed Press This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)
TI - Mesenchymal Stem Cells Maintain Blood-Brain Barrier Integrity by Inhibiting Aquaporin-4 Upregulation After Cerebral Ischemia
JO - Stem Cells
DO - 10.1002/stem.1808
DA - 2014-12-01
UR - https://www.deepdyve.com/lp/oxford-university-press/mesenchymal-stem-cells-maintain-blood-brain-barrier-integrity-by-xcX5S2f4sp
SP - 3150
EP - 3162
VL - 32
IS - 12
DP - DeepDyve
ER -