Inhibition of Hypoxia-Induced Increase of Blood-Brain Barrier Permeability by YC-1 through the Antagonism of HIF-1 Accumulation and VEGF Expression

Cerebral microvascular endothelial cells form the anatomical basis of the blood-brain barrier (BBB), and the tight junctions of the BBB are critical for maintaining brain homeostasis and low permeability. Ischemia/reperfusion is known to damage the tight junctions of BBB and lead to permeability changes. Here we investigated the protective role of 3-(5 -hydroxymethyl-2 furyl)-1-benzylindazole (YC-1), against chemical hypoxia and hypoxia/reoxygenation (H/R)-induced BBB hyperpermeability using adult rat brain endothelial cell culture (ARBEC). YC-1 significantly decreased CoCl2and H/R-induced hyperpermeability of fluorescein isothiocyanate (FITC)-dextran in cell culture inserts. It was found that the decrease and disorganization of tight junction protein zonular occludens-1 (ZO-1) in response to CoCl2, and H/R was antagonized by YC-1. The protection of YC-1 may result from the inhibition of HIF-1 accumulation and production of its downstream target vascular endothelial growth factor (VEGF). VEGF alone significantly increased FITCdextran permeability and down-regulated mRNA and protein levels of ZO-1 in ARBECs. We further used animal model to examine the effect of YC-1 on BBB permeability after cerebral ischemia/reperfusion. It was found that YC-1 significantly protected the BBB against ischemia/reperfusion-induced injury. Taken together, these results indicate that YC-1 may inhibit HIF-1 accumulation and VEGF production, which in turn protect BBB from injury caused by hypoxia. Pathological conditions such as tumors, inflammation, and ischemia are known to damage the blood-brain barrier (BBB) and to lead to the increase of permeability and development of vasogenic brain edema, and VEGF is likely to be a candidate to regulate the change of permeability (Schoch et al., 2002). Using in vitro model of the BBB consisting of brain microvascular endothelial cells, hypoxia-induced hyperpermeability is mediated by the VEGF/VEGF receptor system in an autocrine manner (Fischer et al., 1999). Increase of vascular permeability and subsequent inflammatory responses may contribute to pathogenetic cofactors responsible for the development of neurological damage. The blood-brain barrier constructed from brain microvessel endothelial cells forms a metabolic and physical barrier to protect the central nervous system from the compositional fluctuations that occur in the blood. The blood-brain barrier, which is different from peripheral microvascular endothelium, is the result of the presence of tight junctions (TJs) between neighboring endothelial cells. Tight junctions are complexes of transmembrane proteins that connect to the cytoarchitecture via membrane-associated accessory proteins. The claudin family and occludin are integral transmembrane proteins. Both of these proteins have multiple transmembrane domains and interact with adjacent cells to form homodimeric bridges (Feldman et al., 2005). Stabilization of TJs involves a network of claudins and occludin-linked to the actin cytoskeleton via the zonular occludens proteins (ZO-1, ZO-2, and ZO-3). ZO proteins are membrane-associated accessory proteins that mediate the linkage between This work was supported by research grants from the National Science Council of Taiwan. Article, publication date, and citation information can be found at http://molpharm.aspetjournals.org. doi:10.1124/mol.107.036418. ABBREVIATIONS: BBB, blood-brain barrier; VEGF, vascular endothelial growth factor; TJ, tight junction; ZO, zonular occludens; bHLH, basic helix-loop-helix; HIF, hypoxia inducible factor; HRE, hypoxia-response element; VEGF, vascular endothelial growth factor; YC-1, 3-(5 -hydroxymethyl-2 -furyl)-1-benzylindazole; sGC, soluble guanylate cyclase; ARBEC, adult rat brain endothelial cell; FITC, fluorescein isothiocyanate; PBS, phosphate-buffered saline; kb, kilobase pair(s); H/R, hypoxia/reoxygenation; PKG, protein kinase G; L-NAME, N -nitro-L-arginine methyl ester; 8-Br-cGMP, 8-bromo-cGMP; PCR, polymerase chain reaction; MTT, 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium; ODQ, 1H-[1,2,4]oxadiazolo[4,3-a]quinoxalin-1-one; KT5823, 9-methoxy-9-methoxycarbonyl-8-methyl-2,3,9,10-tetrahydro-8,11-epxoy-1H,8H,11H-2,7b-11a-triazadibenzo(a,g)cycloocta(cde)-trinden-1-one; TGF, transforming growth factor ; ELISA, enzyme-linked immunosorbent assay. 0026-895X/07/7202-440–449$20.00 MOLECULAR PHARMACOLOGY Vol. 72, No. 2 Copyright © 2007 The American Society for Pharmacology and Experimental Therapeutics 36418/3235287 Mol Pharmacol 72:440–449, 2007 Printed in U.S.A. 440 at A PE T Jornals on N ovem er 7, 2017 m oharm .aspeurnals.org D ow nladed from actin and the cytoplasmic tail of claudins and occludin (Gloor et al., 2001). Moreover, ZO proteins are members of membrane-associated guanylate kinase family conserved guanylate kinase domain, an SH3 domain and multiple PDZ domains, suggesting that ZO proteins may participate in signal transduction cascade (González-Mariscal et al., 2000). HIF-1 is a heterodimer composed of two subunits—HIF-1 (120kDa) and HIF-1 (91–94kDa)—that are basic helix-loophelix protein of the PAS [PER/ARNT/SIM (periodicity/aryl hydrocarbon receptor nuclear translocator/simple-minded)] family (Wang et al., 1995). HIF-1 is ubiquitinated by von Hippel-Lindau protein and subjected to proteasomal degradation in nonhypoxic cells, whereas HIF-1 is expressed constitutively in all cells (Salceda and Caro, 1997). Exposure of cells to hypoxia or transition metals like cobalt and iron chelators induces HIF-1 expression and inhibits HIF-1 ubiquitination by dissociating von Hippel-Lindau protein from HIF-1 . HIF-1 comprises a bHLH domain near the amino (N) terminal, which is essential for DNA binding to hypoxia-response elements (HREs) in the HIF target genes such as vascular endothelial growth factor (VEGF), endothelin-1, and erythropoietin (Sharp and Bernaudin, 2004). VEGF, also named vascular permeability factor, is one of the most well known HIF target genex involved in vascular biology (Forsythe et al., 1996). Effects of VEGF on the endothelial cells are evidently mediated by the high-affinity cell surface receptors VEGFR-1 and VEGFR-2. Thus, endothelial cells have a unique and specific spectrum of responses to VEGF (Connolly, 1991). Although many recent reports focused on the angiogenic effect of VEGF, VEGF was initially discovered as a tumor-secreted factor that increases vascular permeability (Senger et al., 1983). During vascular sprouting, cell junctions are partially disorganized, which allows endothelial cells to migrate and proliferate and also increases vascular permeability (Dvorak et al., 1995; Dejana, 2004). YC-, was first introduced to increase NO-soluble guanylate cyclase (sGC) activity and cGMP level on platelets (Ko et al., 1994). However, YC-1-mediated responses via cGMP-independent pathway have also been reported (Hsu et al., 2003; Chien et al., 2005). In addition, growing evidence suggests that YC-1 exerts a novel inhibitory effect on the accumulation of HIF-1 , which in turn blocks the expression of VEGF and antiangiogenesis (Yeo et al., 2003, 2004). Here, we examined whether YC-1 inhibits the destruction and hyperpermeability of BBB induced by cobalt, hypoxia/reoxygenation, or ischemia/reperfusion. We demonstrated that YC-1 significantly inhibits HIF-1 accumulation and VEGF production caused by hypoxia treatment in ARBECs. The in vivo study also reveals that YC-1 is able to protect BBB from ischemia/ reperfusion-induced injury. Materials and Methods Materials. YC-1 was provided by Yung-Shin Pharmaceutical Industry Co. Ltd. (Taichung, Taiwan). Dimethyl sulfoxide served as vehicle and was purchased from Sigma (St. Louis, MO). CoCl2, a hypoxia mimetic agent, was purchased from Wako Pure Chemicals (Tokyo, Japan). Recombinant human VEGF, goat anti-rat VEGF antibody, and goat control IgG antibody were purchased from R&D Systems (Minneapolis, MN). Cell Cultures. Immortalized adult rat brain endothelial cells (ARBECs) were a generous gift from National Research Council of Canada (Garberg et al., 2005). ARBECs were seeded onto 75-cm flasks coated with type I rat tail collagen (50 g/ml; Sigma, St. Louis, MO) and maintained in M199 (Invitrogen, Carlsbad, CA) containing 1% D-glucose solution, 1% Eagle’s basal medium amino acid solution, 1% Eagle’s basal medium vitamin solution (Sigma), 100 U/ml penicillin, 100 mg/ml streptomycin, 10% heat-inactivated fetal bovine serum (Hyclone, Logan, UT) at 37°C in a humidified incubator under 5% CO2 and 95% air. Confluent cultures were passaged by trypsinization. Cells from passage 35 45 were used and starved overnight before experiments. Hypoxia-Reoxygenation of Cultured ARBECs. To induce hypoxia, confluent monolayers of ARBEC cultures were placed into special chamber (Anaerobic System PROOX model 110; BioSpherix, Redfield, NY), which was closed and placed inside an incubator at 37°C and gassing the special chamber with a gas mixture consisting of 95% N2/5% CO2. After 24 h, cell cultures were moved out of the hypoxia chamber and reoxygenated in a regular normoxic incubator (95% air, 5% CO2) for another 4 h. For comparison, control cultures were incubated under normoxic conditions for the same duration. Cell Viability Assay. Cell viability was assessed by MTT assay. ARBECs cultured onto 24-well plates coated with type I rat tail collagen were induced chemical hypoxia using CoCl2, or suffered an H/R insult in the presence or absence of YC-1. Culture medium was aspirated 24 h after treatment and MTT (0.5 mg/ml; Sigma) was added in each well. MTT was then removed 30 min later and cells were lysed by DMSO. The absorbance was measured at 550 nm by microplate reader (Bio-Tek, Winooski, VT). Assay of Paracellular Permeability. The permeability assay using cell culture insert system was performed as described previously (Andriopoulou et al., 1999) with minor modifications. ARBECs 4 10 were seeded onto cell culture inserts (diameter, 10 mm; pore size, 0.4 m; polycarbonate membrane, Nalge Nunc International, Rochester, NY) coated with type I rat tail collagen (50 g/ml, 100 l). When cultured cells were confluent, FITC-dextran (1 mg/ml, 500 l; average molecular mass, 43,200; Sigma) was then added into the upper compartment followed by addition of CoCl2 or putting the culture inserts into hypoxia chamber in the absence or presence of YC-1. At the indicated time points, 50 l of sample medium was taken from the lower compartment. After a dilution of the sample medium to 500 l with PBS, fluorescence intensity of FITC-dextran was measured at excitation and emission wavelengths of 492 nm and 520 nm, respectively, by a fluorescent reader (Spectra MAX Gemini XS). Western Blot Analysis. ARBECs were seeded onto 6-cm dishes coated with type I rat tail collagen. Cells were exposed to drugs for 6 or 24 h for detection of HIF-1 or TJ proteins, respectively. After washing with ice-cold PBS, cells were lysed with radioimmunoprecipitation assay buffer (200 l/dish) on ice for 30 min. After centrifugation at 18,000g for 20 min, the supernatant was used for Western blotting. Protein concentration was measured by BCA assay kit (Pierce, Rockford, IL) with BSA as a standard. Equal proteins (30 g for TJ proteins and 80 g for HIF-1 ) were separated on 8% SDS polyacrylamide gels and transferred to polyvinylidene difluoride (PVDF) membranes (Millipore, Billerica, MA). The membranes were incubated for 1 h with 4% dry skim milk in PBS buffer to block nonspecific binding and then incubated with rabbit antibodies against occludin (1:1000; Santa Cruz Biotechnology, Santa Cruz, CA), claudin-1, ZO-1 (1:1000; Zymed, South San Francisco, CA) or mouse antibodies against HIF-1 (1:1000; Novus Biologicals, Littleton, CO), -tubulin (1:1000; Santa Cruz Biotechnology) for 1 h. After washing with PBS-Tween 20, the membranes were incubated with goat anti-rabbit or anti-mouse peroxidase-conjugated secondary antibody (1:1000; Santa Cruz Biotechnology) for 1 h. The blots were visualized by enhanced chemiluminescence (ECL; Santa Cruz Biotechnology) using Kodak X-OMAT LS film (Eastman Kodak, Rochester, NY). Immunocytofluorescent Staining. ARBECs were seeded onto glass coverslips coated with type I rat tail collagen. After exposure to drugs for 24 h, cells were washed with PBS and fixed with PBS Inhibition of BBB Permeability by YC-1 441 at A PE T Jornals on N ovem er 7, 2017 m oharm .aspeurnals.org D ow nladed from containing 4% paraformaldehyde for 15 min and then permeabilized with 1% Triton X-100 for 20 min. After blocking with 4% dry skim milk in PBS buffer, cells were incubated with rabbit antibodies against ZO-1 (1:100) overnight at 4°C. After a brief wash, cells were then incubated with goat anti-rabbit FITC-conjugated secondary antibody (1:200; Leinco Technologies, Inc., St. Louis, MO) for 1 h. Finally, cells were washed again, mounted, and visualized by fluorescence microscopy (Carl Zeiss Inc., Thornwood, NY). Transfection and Reporter Gene Assay. ARBECs were seeded onto 12-well plates coated with type I rat tail collagen. Cells were cotransfected with 0.4 g of lac-Z vector and 0.8 g of HRE-luciferase reporter gene, 1.5 kb of VEGF-luciferase reporter gene, or 1.2 kb of HRE-deleted VEGF-luciferase reporter gene, respectively (gifts from M.L. Kuo, National Taiwan University, Taipei, Taiwan). Plasmid DNA and Lipofectamine 2000 (10 g/ml; Invitrogen) were premixed with Opti-MEM I (Invitrogen) separately for 5 min and then mix with each other for 25 min and then applied to the cells (500 l/well). After 24-h transfection, the medium was replaced with fresh serumfree culture medium and exposed to drugs for 24 h for the detection of HRE luciferase and VEGF luciferase activity. To prepare lysates, 100 l of reporter lysis buffer (Promega, Madison, WI) was added to each well, and cells were scraped from plates. The supernatant was collected after centrifugation at 15,000g for 3 min. Aliquots of cell lysates (20 l) containing equal amounts of proteins were placed into wells of an opaque white 96-well microplate. The luciferase activity was determined using a dual-luciferase reporter assay system (Promega) and activity value was normalized to transfection efficiency monitored by the cotransfected lacZ vector. Reverse Transcriptase-PCR and Quantitative Real TimePCR. ARBECs were seeded onto six-well plates coated with type I rat tail collagen. After exposure to drugs for 6 or 24 h, total RNA were extracted using a TRIzol kit (MDBio, Inc., Taipei, Taiwan). Two micrograms of RNA were used for reverse transcription by using a commercial kit (Invitrogen, Carlsbad, CA). PCR was performed using an initial step of denaturation (5 min at 95°C), 30 cycles of amplification (95°C for 30 s, 60°C for 1 min, and 72°C for 30 s), and an extension (72°C for 2 min). PCR products were analyzed on 2% agarose gels. Quantitative real time-PCR was proceeded using SYBR Green I Master Mix and analyzed with a model 7900 Sequence Detector System (Applied Biosystems, Foster City, CA). After preincubation at 50°C for 2 min and 95°C for 10 min, the PCR was performed as 40 cycles of 95°C for 10 s and 60°C for 1 min. The threshold was set above the nontemplate control background and within the linear phase of target gene amplification to calculate the cycle number at which the transcript was detected (denoted as CT). The oligonucleotide primers were: ZO-1: forward, 5 -GCGAGGCATCGTTCCTAATAAG-3 ; reverse: 5 -TCGCCACCTGCTGTCTTTG-3 ; VEGF: forward, 5 -ACGAAAGCGCAAGAAATCCC-3 ; reverse, 5 -TTAACTCAAGCTGCCTCGCC-3 ; and -actin: forward, 5 -AGGCTCTTTTCCAGCCTTCCT-3 ; reverse, 5 -GTCTTTACGGATGTCAACGTCACA-3 . Enzyme-Linked Immunosorbent Assay. ARBECs were seeded onto 24-well plates coated with type I rat tail collagen. After exposure to drugs for 24 h, 100 l of culture medium was collected and frozen at 80°C until measurement by Quantikine Rat VEGF Immunoassay ELISA kit (R&D Systems, Minneapolis, MN). After adding 50 l of assay diluent into each microplate well, 50 l of sample medium was then added and incubated for 2 h at room temperature on the shaker. After a brief wash, 100 l of conjugate buffer was added and incubated for 1 h. Finally, 100 l of substrate solution (hydrogen peroxide and chromogen tetramethylbenzidine) was added and incubated in the dark, and 100 l of stop solution (diluted HCl) was added 30 min after. The absorbance was measured at 450 nm by an ELISA reader (Bio-Tek, Winooski, VT). Increase of Blood-Brain Barrier Permeability by IschemiaReperfusion in Rat. Male Sprague-Dawley rats were obtained from the National Laboratory Animal Center of Taiwan and kept on a 12-h light/dark cycle with ad libitum access to food and water. Rats were acclimated to the environment for 7 days before the experiments. The rats (250 300 g) were then anesthetized with trichloroacetaldehyde (400 mg/kg), and two common carotid arteries were exposed and occluded by artery clips through a neck skin incision. Under an operating microscope, a piece of skull was removed through a head skin incision, and the middle cerebral artery was tied with a polyamide monofilament nonabsorbable suture (diameter, 150 m; Johnson and Johnson) to produce ischemia, and rectal temperature was maintained at 37°C by external warming. Ninety minutes after occlusion, YC-1 (1 mg/kg) was injected into femoral vein just before untying of suture. Evans blue dye (100 mg/kg; Sigma) was injected into femoral vein 2 h after reperfusion. Rats were perfused with saline 4 h later through the left ventricle until colorless perfusion fluid was obtained from the right atrium. After decapitation, the brain was removed from the skull, and each hemicortex was dissected. Samples were weighed and soaked in 1 ml of 50% trichloroacetic acid solution. After homogenization and centrifugation, the extracted Evans blue dye was diluted with ethanol (1:3), and fluorescence intensity was measured at excitation and emission wavelengths of 620 nm and 680 nm, respectively, using a fluorescence reader (Belayev et al., 1996). The tissue content of Evans blue dye was quantified from a linear standard curve derived from known amounts of the dye and was expressed as micrograms per microgram