AJP - Heart Calcium Transients and Cell-Sarcomere
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Am J Physiol Heart Circ Physiol 294: H891-H900, 2008. First published December 7, 2007; doi:10.1152/ajpheart.01248.2007
0363-6135/08 $8.00
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Supplemental Figures and Tables
Right arrow All Versions of this Article:
294/2/H891    most recent
01248.2007v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Fontijn, R. D.
Right arrow Articles by Horrevoets, A. J. G.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Fontijn, R. D.
Right arrow Articles by Horrevoets, A. J. G.

SOX-18 controls endothelial-specific claudin-5 gene expression and barrier function

Ruud D. Fontijn,1 Oscar L. Volger,1 Joost O. Fledderus,1 Arie Reijerkerk,2 Helga E. de Vries,2 and Anton J. G. Horrevoets1,2

1Department of Medical Biochemistry, Academic Medical Center, University of Amsterdam, Amsterdam, the Netherlands; and 2Department of Molecular Cell Biology and Immunology, VU University Medical Center, Amsterdam, The Netherlands

Submitted 29 October 2007 ; accepted in final form 5 December 2007


   ABSTRACT
TOP
 ABSTRACT
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 
Members of the claudin family constitute tight junction strands and are major determinants in specificity and selectivity of paracellular barriers. Transcriptional control of claudin gene expression is essential to establish individual claudin expression patterns and barrier properties. Using full genome expression profiling, we now identify sex-determining region Y-box (SOX)-18, a member of the SOX family of high-mobility group box transcription factors, as one of the most differentially induced genes during establishment of the endothelial barrier. We show that overexpression of SOX-18 and a dominant-negative mutant thereof, as well as SOX-18 silencing, greatly affect levels of claudin-5 (CLDN5). The relevance of an evolutionary conserved SOX-binding site in the CLDN5 promoter is shown using sequential promoter deletions, as well as point mutations. Furthermore, SOX-18 silencing abrogates endothelial barrier function, as measured by electric cell-substrate impedance sensing. Thus an obligatory role for SOX-18 in the regulation of CLDN5 gene expression in an endothelial-specific and cell density-dependent manner is established, as well as a crucial, nonredundant role for specifically SOX-18 in the formation of the endothelial barrier.

endothelium; claudin; promoter; sox

IN MULTICELLULAR organisms, epithelial and endothelial cell layers exert specific and selective control over the passage of water and solutes, thus allowing formation and maintenance of compartments that differ in fluid and solute composition. The paracellular aspect of this control is mainly attributed to tight junction strands, which require members of the claudin family of four-pass transmembrane proteins for their formation (9). Both in vivo and in vitro studies have provided evidence that claudins are major determinants in the selectivity of epithelial and endothelial tight junctions (11, 38, 41, 42). Several claudins were found to exhibit specific expression patterns in vivo (14, 30, 31, 38), whereas expression of other claudins is more ubiquitous (Ref. 30, review in Ref. 39). In addition, junction strands generally contain multiple claudin species that may interact in homo- and heterophilic ways, both within and between junction strands (10). Thus the view emerges that claudin expression patterns could account for the diverse functional characteristics of the resulting barriers. This paradigm implies an important role for mechanisms that establish and maintain correct spatial distribution and local concentration of claudins. Indeed, several reports provide evidence that functional mutations in claudin-encoding genes or knockout of individual claudin genes result in specific pathologies as a consequence of localized barrier dysfunction (12, 14, 34, 38). The contribution of transcriptional control to the regulation of claudin gene expression under normal and pathological conditions has been the subject of a, as yet, limited number of reports (17, 26, 29, 33, 36).

Claudin-5 was reported to be an endothelial-specific member of the claudin family (30, 31). Expression levels were found to vary among specific parts of the vasculature, being particularly high in lung and brain. Accordingly, the relative contribution of claudin-5 to the overall barrier of endothelium seems to vary (7, 34). Our laboratory reported that, in cultured endothelial cells, expression of claudin-5 occurs only after cells have reached confluency, which seemed relevant in relation to control of junction formation (7).

Here, we present a comparative analysis of transcriptomes from sparse and postconfluent endothelial cell cultures that was performed to search for genes relevant in endothelial cell barrier formation. Sex-determining region Y-box (SOX)-18, an endothelial-specific transcription factor, shows strong induction in postconfluent cells. In silico analysis of the claudin-5 (CLDN5) promoter reveals an evolutionary conserved SOX consensus binding site. Overexpression of SOX-18 and a dominant-negative mutant thereof, as well as SOX-18 silencing, affect levels of claudin-5. Using a newly developed lentiviral promoter-reporter system, we dissect the CLDN5 promoter and show the relevance of SOX-18 expression and a SOX consensus binding site in endothelial-specific regulation and timing of expression of the CLDN5 gene. Impedance measurements show that the barrier function of SOX-18-silenced endothelial cells is impaired. The data presented here point at a nonredundant function of SOX-18 in controlling endothelial barrier formation.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 
Cell culture. Human umbilical vein endothelial cells (HUVEC) were isolated and cultured as described (15, 20). The culture medium was composed of medium 199 (Invitrogen, Paisley, UK), supplemented with 20% (vol/vol) fetal bovine serum, 50 µg/ml heparin (Sigma, St. Louis, MO), 12.5 µg/ml endothelial cell growth supplement (Sigma), and 100 U/ml penicillin/streptomycin (Invitrogen). All culture surfaces were fibronectin coated.

Microarray probe synthesis and hybridization procedures. Total RNA was isolated using the TRIzol reagent (Invitrogen) and enriched for poly(A)+ RNA using the Oligotex mRNA Minikit (Qiagen, Hilden, Germany). Aminoallyl-labeled first-strand cDNA probes were synthesized from 0.5-µg poly(A)+ RNA using SuperScript II (Invitrogen), with a molecular ratio of aminoallyl-dUTP (Sigma) to dUTP of 4:1. Labeled cDNA was purified using the QIAquick PCR purification kit (Qiagen), and Cy3 or Cy5 monoreactive dyes (Amersham Biosciences, Piscataway, NJ) were coupled according to the manufacturer's instructions. Purified Cy3- and Cy5-labeled cDNAs were hybridized to the microarrays for 20 h at 40°C in microarray hybridization solution (Amersham) and 35% (vol/vol) formamide (Sigma). Microarrays were glass based containing 60-mer oligonucleotide sequences (Sigma/Compugen Library), which represents 18,650 human genes (Micro Array Department, Swammerdam Institute of Life Sciences, Amsterdam, the Netherlands). Microarray probe synthesis and hybridization procedures were performed in triplicate, using a dye-swap procedure. Images were acquired using the Agilent-II scanner (Agilent Technologies, Palo Alto, CA) and processed by ArrayVision 8.0 software (Imaging Research, St. Catharines, ON, Canada). Background-substracted intensities were Loess Normalized (Limma Package, Bioconductor Software, http://www.bioconductor.org) and imported into the Rosetta Resolver database and analysis software (Rosetta Biosoftware, Seattle, WA). Data are available at http://www.ncbi.nlm.nih.gov/projects/geo accession number GSE 9334.

Semiquantitative real-time RT-PCR. Total RNA was isolated using the Absolutely RNA RT-PCR miniprep kit (Stratagene, La Jolla, CA), according to the manufacturer's instructions. Total RNA (0.5 µg) was used for reversed transcription on (dT)12–18 primer using Superscript II (Invitrogen), according to the maufacturer's instructions. PCR reactions were performed in 15-µl reactions on an iCycler system (Biorad Laboratories, Veenendaal, The Netherlands). β-Actin levels were used to normalize levels of the assayed mRNAs. Primer sequences are as follows: CLDN5, (forward) 5'-CCC AGA GGC TCC TGC TGA C-3', (reverse) 5'-GCA GAT TCT TAG CCT TCC CAC TCC-3'; VE-cadherin (CDH5), (forward) 5'-CGT GAG CAT CCA GGC AGT GGT AGC-3', (reverse) 5'-GAG CCG CCG CCG CAG GAA G-3'; zonula occludens 1 (ZO-1), (forward) 5'-CCC GAA GGA GTT GAG CAG GAA ATC-3', (reverse) 5'-CCA CAG GCT TCA GGA ACT TGA GG-3'; β-actin (forward) GGG AAA TCG TGC GTG ACA TTA AG, (reverse) TGT GTT GGC GTA AG GTC TTT G; ELK-3, (forward) 5'-GAG CAG CCT TAG TCC AGT TG-3', (reverse) 5'-TGA GAG TTT GAA GAA AGC AGT AC-3'; and SOX-18, (forward) 5'-TTC GAC CAG TAC CTC AAC TGC-3', (reverse) 5'-GAC ATG GAA CCA AAC ATA CAC G-3'.

Lentiviral SOX-18 overexpression. The entire human SOX-18 open reading frame (ORF) cDNA was obtained by RT-PCR from full-length image clone 5589289 (Geneservice, Cambridge, UK) using primers (forward) 5'-AACTGCAGCCCAGCTGGAATGCAGAGATC-3' and (reverse) 5'-CGAATTCCTAGCCGGAGATGCACGCGC-3', incorporating PstI and EcoRI restriction sites (underlined). The resulting amplicon was subcloned in pGEM-T Easy vector (Promega, Madison, WI) and then transferred, using PstI-EcoRI sites, to the pRRL-cPPT-CMV-X2-PRE-SIN-IRES-eGFP vector (kindly provided by Dr. J. Seppen, Department of Experimental Hepatology, Academic Medical Center, Amsterdam, The Netherlands) and verified by sequencing. Similarly, a COOH-terminal truncated variant, similar to the ragged opossum (Raop) mutant of the mouse ortholog, was constructed by introducing an LEU266TER mutation, using 5'-CGAATTCCTAGGGAGCCCCGTAGCAACCGCC-3' as reverse primer in PCR. Lentiviruses were generated in human embryonic kidney (HEK)-293 T cells, as described (32, 43), and virus-containing supernatants were titrated on HUVEC to determine the titers needed to transduce >90% of the cells. The otherwise identical vector, but without SOX-18 cDNA, was used to generate mock viruses for control transductions. Freshly isolated or first-passage HUVEC cultures were transduced in normal growth medium at ~80% confluency during 24 h, after which they were used for further experimentation.

Immunofluorescence. Primary endothelial cells were grown on fibronectin-coated coverslips. Cells were washed with serum-free medium at 37°C and fixed for 15 min with methanol at room temperature. Phosphate-buffered saline was used as incubation and wash buffer and was supplemented with 1% (wt/vol) bovine serum albumin during antibody incubations. Antibodies were diluted according to the manufacturer's instructions (anti-VE-cadherin: BD Biosciences, Erembodegem, Belgium; anti-ZO-1 and anti-claudin-5: Invitrogen). Signals were detected using anti-rabbit Cy3 (Jackson Immuno Research, West Grove, PA). Coverslips were mounted in Mowiol (Calbiochem, La Jolla, CA), and images were recorded using a Zeiss axioplan2 microscope and Image-Pro software (CyberMedia, Silver Spring, MD).

Electrophoresis and immunoblotting. Cells were harvested by scraping in a buffer containing 50 mM Tris·Cl, pH 7.6, 1% (vol/vol) Triton X-100, 60 mM octyl β-D-glucopyranoside (Sigma), 150 mM NaCl, and supplemented with mammalian protease inhibitor cocktail (Sigma). Samples were denatured by boiling in sample buffer in the presence of 1% (wt/vol) SDS. Equal amounts of protein were used for SDS-PAGE followed by transfer to nitrocellulose for immunoblot analysis. Incubation and wash buffer contained 10 mM Tris·Cl, pH 7.4, 150 mM NaCl, 0.05% (vol/vol) Tween 20, and was supplemented with 5% (wt/vol) nonfat dry milk during blocking of the filter and incubation with antibodies. Primary antibodies were diluted according to the manufacturer's instructions. The following antibodies were used: anti-claudin-5 polyclonal antibody (Invitrogen) and anti-SOX-18 polyclonal antibody (Sigma). Signals were visualized using horseradish peroxidase-conjugated secondary antibodies (Bio-Rad, Hercules, CA), in combination with a chemiluminescent substrate (enhanced chemiluminescence, General Electric). Quantification of signals was performed using ImagQuant software (Amersham Biosciences, UK).

RNA silencing. Silencing of the SOX-18 mRNA was achieved by delivery of annealed silencer RNA (siRNA) 5'-GGGUUACAUUUUUGAAGCA-3' (Ambion, Silencer Pre-designed siRNA, no. 109098), using oligofectamine (Invitrogen), according to the manufacturer's instructions. 5'-CAGUCGCGUUUGCGACUGG-3' was used as nonspecific control.

Promoter-reporter constructs. CLDN5 promoter fragments were amplified from HUVEC chromosomal DNA using standard PCR procedures. Sequences of primers used for amplification, their position relative to the CLDN5 transcription start site, and their chromosomal localization can be found in Supplemental Table S2. (The online version of this article contains supplemental data.) The resulting amplicons were cloned in pGL3-basic (Promega) using KpnI-XhoI restriction sites and sequence verified.

The lentiviral vector pRRL-cPPT-CMV-X2-PRE-SIN (kindly provided by Dr. J. Seppen, Department of Experimental Hepatology, Academic Medical Center, Amsterdam, The Netherlands) was modified for use as carrier of promoter-reporter cassettes as follows: a unique NotI restriction site was eliminated by cleaving with NotI, filling of the overhangs, and religation. Next, a NheI-ClaI fragment containing the cytomegalovirus (CMV) promoter and the multiple cloning site were replaced by a new, synthetic multiple cloning site. The resulting lentiviral vector was then used to accommodate promoter-reporter cassettes, excised from existing plasmids.

First, a cassette containing Renilla luciferase under control of the constitutive herpes simplex virus thymidine kinase promoter was cloned. Second, CLDN5 promoter-reporter cassettes were transferred from pGL3-basic (Promega) as NotI-SalI fragments containing successively a synthetic pA/transcriptional pause site, a CLDN5 promoter fragment, the firefly luciferase reporter gene, and the simian virus 40 (late) polyadenylation signal. The resulting lentiviral CLDN5 promoter-firefly luciferase reporter vector was then packaged, in the presence of a small amount (5% wt/wt) of the herpes simplex virus thymidine kinase Renilla luciferase vector, using HEK-293 T cells as described (32, 43). Virus-containing supernatants were used for transductions, and luciferase activities were measured after at least 3 days, using the dual luciferase system (Promega). Renilla luciferase activities were used to normalize firefly luciferase activity.

Electric cell impedance sensing. Confluent HUVEC layers were monitored by electrical cell impedance sensing (ECIS) (13). HUVEC were cultured on gold electrodes (type 8W10E+, Applied Biophysics, Troy, NY) coated with fibronectin. One day after seeding, when the layers were near confluence, siRNA transfections were performed, and impedence measurements were initiated at 4 kHz.

Statistical analysis. Data are reported as means ± SD. Differences in mean values were analyzed using Student's t-test. Differences were considered significant at the P < 0.05 level.


   RESULTS
TOP
 ABSTRACT
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 GRANTS
 REFERENCES
 
Genes induced in postconfluent cultures of HUVECs. In an attempt to select genes that, analogous to CLDN5, are specifically expressed in postconfluent endothelial monolayers and encode gene products that, based on further in silico analysis may be involved in junction formation, we compared microarray mRNA expression profiles of sparse- and postconfluent monolayers of primary human endothelial cells. Previously, we have shown that, under the applied culture conditions, claudin-5 was not detectable in sparse cells and showed induction only after cells had reached confluency. VE-cadherin, a transmembrane component of endothelial adherens junctions (22), and ZO-1, a cytoplasmic protein connecting multiple transmembrane junction proteins to the actin cytoskeleton (8, 19), both showed basal expression under sparse conditions and a moderate increase, compared with claudin-5, when cells became confluent (7). Microarray data from three different isolates of HUVEC were analyzed by unpaired Cyber-T statistical analysis and corrected for false discovery rate by the Benjamini-Hochberg method (5). Comparison between sparse cultures and postconfluent cultures revealed a large set (N = 511) of reproducibly differential genes at statistical significance P < 0.05. In addition to established endothelial cell-specific markers as endoglin, endothelin 1, von Willebrand factor, and plasminogen inactivator 1 (SERPIN1), this set contains several genes encoding proteins involved in junction formation, platelet endothelial cell adhesion molecule, tight junction protein 1/ZO-1, and VE-cadherin (CDH5), thus showing consistency with our laboratory's earlier findings (7). The corresponding list of genes is given in Supplemental Table S1. We validated the analysis by performing a semiquantitative RT-PCR assay in which PCR reaction products were gel analyzed at intervals of five amplification cycles. In addition to the three well-described genes ZO-1, CDH5, and CLDN5, we tested 10 additional genes, with microarray P values ranging from 1.14E-2 to 1.19E-3. From these genes, 9 out of 10 showed specific cell density-dependent induction, thus independently confirming the result of our microarray analysis (Supplemental Fig. S1).

Interestingly, in our set of differentially expressed genes, signals of SOX-18 and ELK-3 show the highest confluent-to-sparse signal ratios. PCR analysis suggests an on/off type of switch, rather than gradual differences in expression levels, as observed for all other genes tested. This was further confirmed by semiquantitative real-time RT-PCR (Fig. 1). SOX-18 is a member of the SOX gene family that shows endothelial expression in nascent blood vessels and hair follicles. Mutations in SOX-18 are associated with ragged phenotypes in mice and hypotrichosis-lymphedema-telangiectasia (HLTS) in humans. In both cases, defects in vascular (lymphatic) development are accompanied by diffuse (lymph) edema (18, 35). ELK-3 is an ETS ternary complex transcription factor that, like many other members of the ETS family, is expressed in the endothelium of the developing vasculature (3, 4). An interspecies comparative analysis of the CLDN5 promoter revealed the presence of a single evolutionary conserved SOX consensus transcription factor binding site between position –183 and –167 relative to the CLDN5 transcription initiation site. Like many other endothelial-specific genes, the CLDN5 promoter contains multiple conserved ETS consensus sites, i.e., four sites between position –165 and –74 relative to the CLDN5 transcription initiation site (Supplemental Table S2). The importance of ETS transcription factors in endothelial biology is well established and comprises junction formation via regulation of CDH5 (24). Here, we tested the hypothesis that SOX-18 is involved in timing and endothelial specificity of the CLDN5 promoter.


Figure 1
View larger version (9K):
[in this window]
[in a new window]

  		Fig. 1. Expression of claudin-5 (CLDN5), sex-determining region Y-box (SOX)-18, and ELK-3 as function of human umbilical vein endothelial cell (HUVEC) density. Sparse, HUVEC were seeded at 1 x 104/cm2 and grown for 24 h; confluent, cells were seeded at 3 x 104/cm2 and grown to confluency, after which culture was continued for 48 h before RNA was harvested. Specific mRNAs were measured using semiquantitative RT-PCR. Results were normalized for β-actin and expressed as percentage of the maximum value measured for the specific mRNA. CLDN5, SOX-18, and ELK-3 were significantly upregulated in confluent cells compared with sparse cells. Means ± SD from four HUVEC isolates are shown. *P < 0.05, **P < 0.01, confluent vs. sparse.

 
Endothelial specificity of the CLDN5 promoter and SOX-18 expression. Endothelial-specific expression of claudin-5 has been described in tissue of mouse- and human origin (25, 30, 31, 34). We examined whether transcriptional regulation is involved in endothelial specificity and whether this regulation can be studied under in vitro conditions. We developed a lentivirus-based system that allowed highly efficient transduction of CLDN5-promoter-reporter constructs in virtually any cell type, thus making long-term stable expression from a chromosomal environment in both homologous and heterologous cell types possible. Based on in silico analysis, an evolutionary conserved hypothetical core promoter comprising 300 bp upstream of the CLDN5 transcription start site and including the conserved SOX binding site was defined (Supplemental Table S2). This fragment was cloned upstream of the firefly luciferase gene in a viral vector, and virus particles were packaged for transduction of several types of primary cells and cell lines, as described in MATERIALS AND METHODS. In HUVEC, the hypothetical CLDN5 core promoter increased luciferase activity 42-fold, relative to a promoter-less luciferase control virus. As shown in Fig. 2A, the CLDN5 core promoter induced transcriptional activity at a considerably higher rate in HUVEC compared with primary arterial smooth muscle cells, as well as CaCo, HeLa, and HEK-293 cell lines.


Figure 2
View larger version (12K):
[in this window]
[in a new window]

  		Fig. 2. Endothelial specificity of claudin-5 promoter activity and SOX-18 expression. A: an evolutionary conserved fragment of 300 bp upstream of the CLDN5 transcription start site was cloned upstream of the firefly luciferase reporter gene in a lentiviral vector, which was then packaged into lentiviral particles. During packaging, a small amount of a similar vector carrying the constitutive herpes simplex virus thymidine kinase (HSV-TK) promoter in front of the Renilla luciferase reporter gene was spiked for normalization. Both types of luciferase were measured after at least 48 h. Ratios of firefly luciferase over Renilla luciferase are plotted relative to ratios in HUVEC. B: approximately 48 h after the cells had reached confluency, RNA was harvested, and SOX-18, ELK-3, CLDN5, VE-cadherin (CDH5), and zonula occludens-1 (ZO-1, TJP1) mRNA were measured by semiquantitative RT-PCR. Values were normalized for β-actin mRNA and are expressed relative to HUVEC. SMC, smooth muscle cells; TJP1, tight junction protein 1.

 
Next, we examined cell specificity of SOX-18 gene expression by measuring its mRNA levels in different cell types, using real-time semiquantitative RT-PCR, and compared its actin-normalized relative mRNA expression to that of CDH5 and CLDN5, two established endothelial-specific junction components, ELK-3, of unknown cellular specificity, and ZO-1, a ubiquitously expressed component of the junctional plaque. As shown in Fig. 2B, SOX-18 is exclusively expressed in endothelial cells. ELK-3 expression is preferentially in endothelial cells, and ZO-1 shows preferred expression in endothelial/epithelial cell types. Together, these data point at cell-specific transcriptional control of CLDN5 expression under culture conditions and may suggest a role for SOX-18 in this control.

SOX-18 specifically increases claudin-5 expression at the mRNA and protein level. We examined possible effects of SOX-18 on claudin-5 expression levels by lentiviral transduction of human SOX-18 encoding cDNA into a native, human vascular cell type (7). In addition, we tested a COOH-terminally truncated mutant of SOX-18 ({Delta}CT), which was analogous to the dominant-negative Raop mutant of the mouse ortholog (21, 35). SOX-18 {Delta}CT was constructed by introducing a LEU266TER mutation, resulting in a truncated SOX-18 protein. In Raop, a frame shift at this position leads to missense coding and premature truncation. SOX-18 and SOX-18 {Delta}CT cDNA was cloned into a lentiviral vector under control of the CMV promoter, and infectious virus particles were produced and titrated, as described in MATERIALS AND METHODS. At the RNA level, SOX-18 expression was typically between 15- and 25-fold increased (data not shown). Immunostaining on Western blot detected SOX-18 and SOX-18 {Delta}CT at 46 kDa and between 34 and 40 kDa, respectively (Fig. 3A). Comparison to predicted sizes of 41 kDa (Swiss prot. no. p35713) and 30 kDa suggests posttranslational modification of, as yet, unknown nature. Using immunofluorescence microscopy, we examined expression and localization of wild-type SOX-18 and SOX-18 {Delta}CT, as well as their possible effects on claudin-5, VE-cadherin, and ZO-1 expression in HUVEC. Figure 3B shows that both overexpressed wild type and COOH-terminally truncated SOX-18 localize to the nucleus, as does endogenous SOX-18. SOX-18 {Delta}CT signals are slightly higher and show less variation in signal intensity compared with overexpressed wild-type SOX-18. In both cases, signals are considerably higher than that observed for endogenous SOX-18, and nuclear localization seems more pronounced. Relative to vector and SOX-18 {Delta}CT transduced cells, signals of claudin-5 are increased in SOX-18 transduced cells. As expected, claudin-5 fluorescence concentrates at the areas of cell-cell contact. Neither VE-cadherin nor ZO-1 expression seems affected by SOX-18 or SOX-18 {Delta}CT. Next, we examined effects of SOX-18 overexpression on mRNA levels of CLDN5, ZO-1, and CDH5 by real-time semiquantitative RT-PCR in both postconfluent cells and sparse cells, i.e., in the absence and presence of endogenous SOX-18, respectively. As shown in Fig. 3C, in postconfluent cells, CLDN5 mRNA expression is significantly increased compared with mock-transduced cells, whereas ZO-1 and CDH5 mRNA levels remain unaltered, thus confirming, at the RNA level, the observations made by immunofluorescence microscopy. Under sparse conditions, SOX-18 overexpression does not have an effect on any of the mRNA levels measured. The effects of SOX-18 {Delta}CT on CLDN5 mRNA expression were limited; a small decrease (n = 6, mean value relative to control 0.75, P value 0.03) was observed, likely due to a dominant-negative effect on endogenous transcriptional activity from SOX F family members. Finally, we quantified claudin-5 protein levels of SOX-18, SOX-18 {Delta}CT, and mock transduced cells on Western blot (Fig. 4). We could detect significantly increased claudin-5 levels in SOX-18 overexpressing cells, whereas claudin-5 levels were significantly decreased in SOX-18 {Delta}CT cells, thus substantiating our results obtained by immunofluorescence and semiquantitative PCR. Together, the finding that SOX-18 overexpression enhances CLDN5 gene expression and the observed endothelial specificity of SOX-18 expression support the view that SOX-18 is an important determinant in endothelial-specific expression of CLDN5.


Figure 3
View larger version (26K):
[in this window]
[in a new window]

  		Fig. 3. Lentiviral overexpression of SOX-18 and SOX-18 dCt in HUVEC. A: SOX-18 overexpression detected by Western blotting. Four days after lentiviral transduction of HUVEC, lysates were analyzed by immunoblotting with SOX-18 antibodies. Mock, transduced with virus containing vector without SOX-18 insert; SOX-18, transduction with virus containing the SOX-18 open reading frame (ORF) under control of the cytomegalovirus (CMV) promoter; SOX-18 dCt, transduction with virus encoding a (Leu266Ter) SOX-18 mutant under control of the CMV promoter. Equal loading was confirmed with anti-{alpha} tubulin antibodies. B: effect of SOX-18 and SOX-18 dCt overexpression on fluorescence patterns of SOX-18, claudin-5, VE-cadherin, and ZO-1. After transduction with either mock, SOX-18 dCt, or SOX-18 virus (rows), cells were seeded on fibronectin-coated glass coverslips and grown for 4 days. Cells were fixed and stained with antibodies against SOX-18, claudin-5, VE-cadherin, and ZO-1 (columns) and a Cy-3-conjugated secondary antibody. Recordings were made using a Zeiss axioplan2 microscope, with settings optimized for SOX-18 transduced cells. Subsequently, mock and SOX-18 dCt transduced cells were recorded with identical settings. SOX-18 transduced cells show increased fluorescence of claudin-5 compared with mock and SOX-18 dCt-transduced cells. VE-cadherin and ZO-1 fluorescence are not affected by SOX-18. Bar: 60 µm. C: effect of overexpression of SOX-18 on mRNA levels of ZO-1 (TJP1), CDH5, and CLDN5 under sparse and confluent conditions. Cells were transduced with virus containing the SOX-18 ORF under control of the CMV promoter. As control, transductions with virus containing vector without insert were performed (mock). Three days after transduction, either growth was continued (confluent), or cells were reseeded at a fivefold lower density (sparse). Cells were then grown another 24 h before harvest. mRNA levels were determined using real-time semiquantitative RT-PCR. Values were normalized for β-actin and plotted as value relative to mock. SOX-18 transduced cells show significantly increased levels of CLDN5 mRNA under confluent conditions but not under sparse conditions. Means ± SD from 3 (ZO-1 and CDH5) or 6 (CLDN5) HUVEC isolates are shown. *Significant differences with mock transductions are indicated, P < 0.05.

 

Figure 4
View larger version (17K):
[in this window]
[in a new window]

  		Fig. 4. Effect of lentiviral overexpression of SOX-18 and SOX-18 dCt on claudin-5 expression in HUVEC. A: claudin-5 protein detected by Western blotting. Four days after lentiviral transduction of HUVEC, lysates were analyzed by immunoblotting with claudin-5 antibodies. SOX-18, transduction with virus containing the SOX-18 ORF under control of the CMV promoter; SOX-18 dCt, transduction with virus encoding a (Leu266Ter) SOX-18 mutant under control of the CMV promoter; mock, transduced with virus containing vector without insert. Equal loading was confirmed with anti-{alpha}-tubulin antibodies. B: quantitative representation of the experiment shown under A. SOX-18 values are normalized on {alpha}-tubulin. SOX-18 induces an increase in claudin-5 protein levels, whereas SOX-18 dCt has the opposite effect. Means ± SD from 6 HUVEC isolates are shown. Significant differences with mock transductions are indicated: *P < 0.05, **P < 0.01.

 
A SOX consensus binding site in the CLDN5 promoter is required for SOX-18 enhanced transcriptional activity. We tested whether the SOX-18-induced increase in claudin-5 expression was due to increased transcriptional activity from the CLDN5 promoter. To this purpose, the 300-bp hypothetical core promoter that was used to establish endothelial specificity of CLDN5 promoter activity (Fig. 2) was now tested in either the presence or absence of SOX-18 or SOX18 {Delta}CT overexpression. In addition, a larger fragment comprising 1,000 bp upstream of the CLDN5 transcription start was tested. These fragments were cloned in a lentiviral reporter vector and transduced into HUVEC, as described in MATERIALS AND METHODS. Twenty-four hours after this transduction, cells were transduced with SOX-18, SOX-18 {Delta}CT, or mock virus. Three days after the second transduction, lysates were prepared for measurement of reporter activity. As shown in Fig. 5, the effects of SOX-18 and SOX-18 {Delta}CT overexpression on reporter activity are essentially the same as observed earlier for claudin-5 mRNA and protein, indicating that SOX-18 enhances claudin-5 expression via transcriptional activation of the CLDN5 promoter. In addition, the viral promoter-reporter system reliably reflects the behavior of the endogenous CLDN5 promoter in response to SOX-18 or SOX-18 {Delta}CT.


Figure 5
View larger version (12K):
[in this window]
[in a new window]

  		Fig. 5. Overexpression of SOX-18 enhances transcription directed by the CLDN5 promoter. 1,000-bp and 300-bp fragments upstream of the CLDN5 transcription start site were cloned upstream of a firefly luciferase reporter gene on a lentiviral vector, which was then packaged into lentiviral particles. During packaging, a small amount of a similar vector carrying the constitutive HSV-TK promoter in front of the Renilla luciferase reporter gene was spiked for normalization. Resulting virus was used to transduce HUVEC. In a subsequent second round of transduction, SOX-18 or SOX-18 dCt under control of the CMV promoter were introduced. As a control, virus containing the viral expression vector without insert was transduced (mock). After 3 days, cells were lysed, and luciferase activities were measured. Normalized promoter activity was plotted relative to mock transduced cells. SOX-18 enhances transcriptional activity of the CLDN5 promoter. Means ± SD from 4 HUVEC isolates are shown. *Significant differences with mock transductions are indicated: P < 0.05.

 
We then further fragmented the CLDN5 promoter along lines of evolutionary conservation (Fig. 6A and Supplemental Table S2). Immediately upstream of the transcription start site, the well-conserved 300-bp hypothetical core promoter was identified, whereas overall sequence conservation shows a sharp decline further upstream, as well as downstream of the transcription start. Within the conserved 300-bp core promoter fragment (designated –300), we identified three clusters of conserved transcription factor binding sites (Fig. 6B and Supplemental Table S2). A promoter-deletion analysis was performed in which these clusters were covered by fragments –300, –230, –148, and control (empty reporter vector). Two additional short sequences that each contain two overlapping transcription factor binding sites are located in the –490 and –362 fragments. Reporter constructs were transduced into HUVEC, which was followed after 24 h by transduction with either SOX-18 or mock virus. Reporter activity was measured after 3 days. As can be inferred from Fig. 6B, both basal (endogenous) and SOX-18-enhanced transcriptional activity are maintained in the consecutive promoter fragments –490, –362, –300, and –230. However, a significant decrease in both basal and SOX-18-enhanced reporter activity is observed upon deletion of the –230 to –149 fragment, which contains, among others, the conserved SOX consensus binding site. The importance of this site was confirmed by testing the effect of local point mutations. In the 300-bp reporter construct, we converted the SOX core sequence CAAT to either TGAT or TGTG and measured the effect on reporter activity after transduction of either SOX-18 or mock virus. Figure 6C shows that, both under conditions of SOX-18 overexpression and under control (mock) conditions, point mutations in the core SOX binding site cause a significant decrease in reporter activity. These findings imply that the SOX site is involved not only in the enhanced transcriptional activity from the CLDN5 promoter upon overexpression of SOX-18, but also in basal, endogenously directed transcription of the CLDN5 gene.


Figure 6
View larger version (15K):
[in this window]
[in a new window]

  		Fig. 6. Stimulation of the CLDN5 promoter by SOX-18 is critically dependent on an evolutionary conserved SOX-binding site. A: evolutionary conservation of the CLDN5 locus. Prealigned genome sequences for CLDN5 from human, mouse, rat, and dog were analyzed for evolutionary conservation of regulatory sequences using the VISTA browser (http://genome.llbl.gov/vista). The boxed plot represents organization and evolutionary conservation (human vs. mouse) of the CLDN5 locus using a 50-bp sliding window. Next, the Genomatix suite (http://www.genomatix.com) was used to identify conserved potential transcription factor binding sites within the conserved regions. Potential SOX, ETS, and LEF binding sites are indicated. For a complete overview of conserved transcription factor binding sites, see Supplemental Table S2. B: progressive 5' deletion of the CLDN5 promoter. Deletion fragments of the CLDN5 promoter were generated by PCR (Supplemental Table S2). The resulting fragments were cloned upstream of the firefly luciferase reporter gene on a lentiviral vector, which was then packaged into lentiviral particles. During packaging, a small amount of a similar vector carrying the constitutive HSV-TK promoter in front of the Renilla luciferase reporter gene was spiked for normalization. Resulting virus was used to transduce HUVEC. In a second round of transduction, SOX-18 under control of the CMV promoter was introduced. As a control, virus containing the viral expression vector without insert was transduced (mock). After 3 days, cells were lysed, and luciferase activities were measured. Normalized values of firefly luciferase activity (Fluc) over Renilla luciferase activity (Rluc) are plotted. The region –230 to –147 is critical in both basal and SOX-18-induced CLDN5 transcription. Means ± SD from 3 HUVEC isolates are shown. Significant differences between SOX-18 and mock transductions are indicated: *P < 0.05, **P < 0.01. C: in the –300 CLDN5 upstream fragment, mutations in the core CAAT sequence of an evolutionary conserved SOX-binding site were introduced using PCR. The resulting 300-bp TGAT and 300-bp TGTG fragments were cloned and tested for their promoter activity in presence and absence of overexpressed SOX-18, described under B. Normalized values of firefly luciferase activity (Fluc) over Renilla luciferase activity (Rluc) are plotted. Means ± SD from 3 HUVEC isolates are shown. Significant differences with the 300-bp CAAT fragment are indicated: *P < 0.05.

 
RNA silencing of SOX-18 expression specifically reduces basal, endogenous transcription from the CLDN5 promoter. Our findings from the previous section are consistent with the hypothesized role of SOX-18 in regulation of the CLDN5 promoter. However, given the fact that members of the SOX family tend to bind to a common consensus site (23, 28), the question as to which particular SOX protein is involved in basal CLDN5 expression still has to be answered conclusively. In particular, members of the SOX-F subfamily (SOX-18, -17, and -7) that are thought to display functional redundancy in endothelial cells (27, 37) are candidates. Inspection of our microarray data revealed that, in addition to the SOX-18 gene, SOX-17, -2, -3, and -6 genes are expressed. In contrast to SOX-18, expression of none of the other SOX genes mentioned showed significant dependence on cell density.

We specifically targeted SOX-18 expression by transfecting double-stranded SOX-18-siRNAs into HUVEC and monitored their effect by measuring mRNA levels for SOX-18 and its potential target, CLDN5, by real-time semiquantitative RT-PCR. As a negative control, we examined expression of ZO-1, which we had found not to be affected by overexpression of SOX-18. Compared with control siRNA, we observed that SOX-18-siRNA induced efficient silencing of SOX-18, accompanied by a profound decrease of CLDN5 mRNA levels, whereas ZO-1 mRNA levels remained unchanged (Fig. 7A). These data support the view that SOX-18 is the specific member of the SOX family involved in transcriptional control of CLDN5. Western blotting of cell lysates from non-sense and SOX-18 siRNA-treated cells (Fig. 7B) further substantiated these results. Strikingly, the decrease of CLDN5 at the protein level is more pronounced than at the mRNA level. Our CLDN5 real-time RT-PCR has been proven to accurately measure CLDN5 mRNA expression in a wide dynamic range (Fig. 1 and Ref. 7). We speculate that stability of claudin-5 protein is decreased upon SOX-18 silencing. A decrease of claudin-5 concentration in the cell membrane may result in a lower degree of claudin strand formation and, hence, in lower stability. Thus specific silencing of SOX-18 expression reveals a nonredundant and obligatory role for SOX-18 in the control of CLDN5 expression.


Figure 7
View larger version (17K):
[in this window]
[in a new window]

  		Fig. 7. Effects of SOX-18 silencing (Si) in HUVEC. A: effects of SOX-18 Si on SOX-18, CLDN5, and ZO-1 (TJP1) mRNA. HUVEC cultures were treated with complexes of oligofectamine and either SOX-18 Si or control (C) double-stranded RNA oligonucleotides. Two days after transfection, RNA was isolated, and SOX-18, CLDN5, and ZO-1 mRNA levels were determined using real-time semiquantitative RT-PCR. Values were normalized for β-actin and are plotted relative to control double-stranded RNA. SOX-18 and CLDN5 mRNAs are significantly reduced, whereas ZO-1 mRNA is not affected. Means ± SD from 3 HUVEC isolates are shown. **Significant differences with control double-stranded RNA are indicated: P < 0.01. B: claudin-5 protein detected by Western blotting. HUVEC cultures were treated with complexes of oligofectamine and either SOX-18 Si or C double-stranded RNA oligonucleotides. Two days after transfection, lysates were analyzed by immunoblotting with claudin-5 antibodies. Equal loading was confirmed with anti-{alpha}-tubulin antibodies. SOX-18 Si transfected cells show decreased levels of claudin-5 protein. C: phase-contrast microscopy of HUVEC 48 h after SOX-18 Si RNA or C transfections. D: electrical cell impedance sensing (ECIS) measurement of SOX-18 Si HUVEC. Cultures were treated as under B, and development of barrier function was monitored using ECIS. Resistance measured at 4 kHz at the times indicated is plotted as percentage of the mean resistance, with 100% corresponding to 1,196 {Omega}. *Significant differences with control double-stranded RNA are indicated: *P < 0.05.

 
RNA silencing of SOX-18 expression results in impairment of endothelial barrier function as measured by ECIS. To investigate whether SOX-18 regulates integrity of HUVEC barriers, transendothelial electrical resistance of confluent endothelial monolayers was measured in time using ECIS (13). Two days after transfection with SOX-18 siRNA, cell shape became elongated, but monolayers remained intact, as observed by phase-contrast microscopy (Fig. 7C). Sixty hours after transfection, resistance was significantly reduced in SOX-18 siRNA cells compared with control-treated cells and had dropped to levels observed in nonmatured monolayers (Fig. 7D). The complete ECIS time course graph is available as Supplemental Fig. S2.


   DISCUSSION
TOP
 ABSTRACT
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
GRANTS
 REFERENCES
 
Previously, our laboratory has reported the transcriptional upregulation of several genes involved in endothelial barrier formation during maturation of endothelial monolayers (7). In the present study, SOX-18 emerged as one of the most strongly induced genes in an analysis of transcriptomes from sparse vs. postconfluent endothelial cells. A role for SOX-18 in control of endothelial barriers was suggested by reports on SOX-18 mutations in vascular anomalies in ragged (Ra) mice (21, 35) and in the human HLTS syndrome (18). The associated pathologies are characterized by coat defects, chylous ascites, edema, and cardiovascular dysfunction. All four described mouse mutations, and one out of three mutations described for HLTS have in common that they encode truncated versions of the SOX-18 protein with impaired trans-activation activity. These truncated SOX-18 proteins are thought to interfere, in a dominant-negative fashion, with other members of the SOX F subfamily (21, 27, 35, 37).

Here, we observed that, of several cell types tested, only HUVEC expressed SOX-18. Overexpression of SOX-18 and a dominant-negative derivative thereof, SOX-18 {Delta}CT, induce an increase and decrease of claudin-5 expression, respectively. We observed that the human CLDN5 promoter displayed cell-specific activity and identified a potential SOX binding site. Controlled deletion analysis of the CLDN5 promoter, as well as mutational analysis of the potential SOX binding site, showed that the latter is required for both basal and SOX-18 overexpression driven CLDN5 expression. Finally, silencing of SOX-18 showed that endogenous SOX-18 is essential for basal CLDN5 transcription. These findings identify the CLDN5 gene as an important SOX-18 target.

SOX-18 is a member of the SOX gene family that shares a conserved high-mobility group domain. Individual members are responsible for regulation of cell fate and differentiation processes (review in Ref. 23) and show characteristic spatiotemporal expression patterns during development and adulthood. SOX-18 was found in endothelial cells of pre- and postnatal developing vasculature in mice (16, 27, 35). These observations are consistent with our view that SOX-18 regulation of the CLDN5 promoter contributes to endothelial cell specificity of CLDN5 expression.

Here, one may speculate that CLDN5, being a target of SOX-18, may be involved in pathologies associated with HLTS in humans and ragged mutations in mice, as these result from dominant-negative SOX-18 mutants. The phenotype of CLDN5–/– mice, however, shows no overlap with the ragged phenotype: CLDN5–/– mice have a normally developed vasculature, do not show edema or vascular leakage, but only display partial failure of the blood-brain barrier (34). Indeed, the importance of claudin-5 in non-blood-brain barrier endothelium seems limited (7, 34). Thus, despite the fact that we identify CLDN5 as SOX-18 target gene, it seems unlikely that CLDN5 is a major effector gene in pathologies associated with HLTS or ragged phenotypes. However, ECIS measurements applied to SOX-18-silenced HUVEC monolayers show impaired barrier function, which is in marked contrast to the lack of a barrier-related phenotype in SOX-18–/– mice (35). Apparently, masking of effects resulting from the loss of SOX-18 function by redundant activity of other members of the SOX F subfamily, as observed in mouse knockout models (27, 35, 37), does not occur under our culture conditions. Thus we show that loss of SOX-18 function results in a severely compromised endothelial barrier function. A future inventory of SOX-18-downstream genes might identify potential effector genes.

At the subcellular level, claudins are functionally associated with tight junction complexes via interaction with ZO proteins (19). In epithelial cells, this interaction plays a role in spatiotemporal control of tight junction formation (40). Initially, ZO-1 and ZO-2 accumulate on primordial adherence junctions between adjacent cells. In a later stage, claudins are recruited to these structures where their polymerization into functional strands is facilitated (1, 2, 40). Therefore, it may be speculated that claudin genes are subject to regulatory mechanisms that exert temporal control during junction formation. Consistent with earlier findings (7), our data reveal tight control of SOX-18 over CLDN5 transcription, which is dependent on cell density and thus coincides with junction formation. Under sparse conditions, when endogenously expressed SOX-18 is absent, we observed no significant effect of SOX-18 overexpression on CLDN5 mRNA levels, suggesting that additional, obligatory transcription factors are lacking. Alternatively, the CLDN5 promoter may be actively repressed under sparse conditions. Under confluent conditions, the response of the CLDN5 gene to SOX-18 overexpression is limited, likely due to the background of endogenous SOX-18. Silencing of endogenous SOX-18 in postconfluent cells revealed an essential and nonredundant role for SOX-18 in CLDN5 gene expression. This observation, together with the observed induction of SOX-18 in postconfluent cells, reveals a potent mechanism for temporal control of CLDN5 during junction formation. These findings are consistent with the paradigm that SOX family members are key regulators in spatiotemporal control of gene expression (23).

Recent studies have shown that the regulation of claudin gene expression seems to have diverged greatly, in marked contrast to the highly conserved sequence and structural homology of the claudin proteins, thus allowing a specific regulation of local barrier functions. Promoters of mouse claudin-3, -4, and -7 genes were repressed by Snail during epithelial-to-mesenchyme transition (17). T-cell factor (TCF)-4 binding elements in the 5' flanking region of the claudin-1 gene were found to be responsible for regulation by the β-catenin/TCF complex in colorectal cancers (29). Overlapping GATA/nuclear factor Y binding sites were found to be responsible for dual-transcriptional control of CLDN11 in the blood-testis barrier (26). Tissue-specific regulation was described for CLDN18 in lung by T/EBP/NKX2.1 homeodomain transcription factor (33). Sakaguchi et al. (36) reported on the role of two intestine-specific caudal-type homeobox proteins, CDX1 and CDX2, and a member of the POU homeodomain family, TCF-1 (hepatocyte nuclear factor-1{alpha}), in regulation of the CLDN2 gene in intestine and liver. In the present study, we now identify SOX-18 as an essential regulator of endothelial claudin-5 expression, as well as barrier function, supporting the view that individual claudin genes are controlled by unique transcriptional regulatory mechanisms to achieve correct tissue-specific and temporal expression.


   GRANTS
TOP
 ABSTRACT
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 GRANTS
REFERENCES
 
This study was supported by Molecular Cardiology Grant M93.007 from the Netherlands Heart Foundation, The Hague, The Netherlands and NWO-Genomics Grant 050–110-1014 from the Netherlands Organization for Scientific Research, The Hague, The Netherlands.


   FOOTNOTES
 

Address for reprint requests and other correspondence: A. J. G. Horrevoets, Dept. of Medical Biochemistry, Academic Medical Center K1-114, Meibergdreef 15, 1105 AZ Amsterdam, The Netherlands (e-mail: a.j.horrevoets{at}amc.uva.nl)

The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.


   REFERENCES
TOP
 ABSTRACT
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 

  1. Adams CL, Nelson WJ, Smith S. Quantitative analysis of cadherin-catenin-actin reorganization during development of cell-cell adhesion. J Cell Biol 135: 1899–1911, 1996.[Abstract/Free Full Text]
  2. Ando-Akatsuka Y, Yonemura S, Itoh M, Furuse M, Tsukita S. Differential behaviour of E-cadherin and occludin in their colocalization with ZO-1 during the establishment of epithelial cell polarity. J Cell Physiol 179: 115–125, 1999.[CrossRef][Web of Science][Medline]
  3. Ayadi A, Suelves M, Dollé P, Wasylyk B. Net, an Ets ternary complex transcription factor, is expressed in sites of vasculogenesis, angiogenesis, and chondrogenesis during mouse development. Mech Dev 102: 205–208, 2001.[CrossRef][Web of Science][Medline]
  4. Ayadi A, Zheng H, Sobieszczuk P, Buchwalter G, Moerman P, Alitalo K, Wasylyk B. Net-targeted mutant mice develop a vascular penotype and upregualte egr-1. EMBO J 20: 5139–5152, 2001.[CrossRef][Web of Science][Medline]
  5. Baldi P, Long AD. A Bayesian framework for the analysis of microarray expression data: regularized t-test and statistical inferences of gene changes. Bioinformatics 17: 509–519, 2001.[Abstract/Free Full Text]
  6. Dekker RJ, Boon RA, Rondaij MG, Kragt A, Volger OL, Elderkamp YW, Meijers JCM, Voorberg J, Pannekoek H, Horrevoets AJG. KLF2 provokes a gene expression pattern that establishes functional quiescent differentiation of the endothelium. Blood 107: 4354–4363, 2006.[Abstract/Free Full Text]
  7. Fontijn RD, Rohlena J, van Marle J, Pannekoek H, Horrevoets AJG. Limited contribution of claudin-5 dependent tight junction strands to endothelial barrier function. Eur J Cell Biol 85: 1131–1144, 2006.[CrossRef][Web of Science][Medline]
  8. Furuse M, Itoh M, Hirase T, Nagasuchi A, Yonemura S, Tsukita S. Direct association of occludin with ZO-1 and its possible involvement in the localization of occludin at the tight junction. J Cell Biol 127: 1617–1626, 1994.[Abstract/Free Full Text]
  9. Furuse M, Sasaki H, Fujimoto K, Tsukita S. A single gene product, claudin-1 or -2, reconstitutes tight junction strands and recruits occludin in fibroblasts. J Cell Biol 143: 391–401, 1998.[Abstract/Free Full Text]
  10. Furuse M, Sasaki H, Tsukita S. Manner of interaction of heterogenous claudin species within and between tight junction strands. J Cell Biol 147: 891–903, 1999.[Abstract/Free Full Text]
  11. Furuse M, Furuse K, Sasaki H, Tsukita S. Conversion of zonulea occludentes from tight to leaky stand type by introducing claudin-2 into Madin-Darby canine kidney I cells. J Cell Biol 153: 263–272, 2001.[Abstract/Free Full Text]
  12. Furuse M, Hata M, Furuse K, Yoshida Y, Haratake A, Sugitani Y, Noda T, Kubo A, Tsukita S. Claudin-based tight junctions are crucial for the mammalian epidermal barrier: a lesson from claudin-1-deficient mice. J Cell Biol 156: 1099–10111, 2002.[Abstract/Free Full Text]
  13. Giaever I, Keese CR. Micromotion of mammalian cells measured electrically. Proc Natl Acad Sci USA 88: 7896–7900, 1991.[Abstract/Free Full Text]
  14. Gow A, Southwood CM, Li JS, Pariali M, Riordan GP, Brodie SE, Danias J, Bronstein JM, Kachar B, Lazzarani RA. CNS myelin and sertoli cell tight junction strands are absent in Osp/claudin-11 null mice. Cell 99: 649–659, 1999.[CrossRef][Web of Science][Medline]
  15. Horrevoets AJG, Fontijn RD, van Zonneveld AJ, de Vries CJM, ten Cate JW, Pannekoek H. Vascular endothelial genes that are responsive to tumor necrosis factor-{alpha} in vitro are expressed in atherosclerotic lesions, including inhibitor of apoptosis protein-1, stannin, and two novel genes. Blood 93: 3418–3431, 1999.[Abstract/Free Full Text]
  16. Hosking BM, Wyeth JR, Pennisi DJ, Wang SC, Koopman P, Muscat GEO. Cloning and functional analysis of the Sry-related HMG box gene, Sox18. Gene 262: 239–247, 2001.[CrossRef][Web of Science][Medline]
  17. Ikenouchi J, Matsuda M, Furuse M, Tsukita S. Regulation of tight junctions during epithelium-mesenchyme transition: direct repression of claudins/occludin by snail. J Cell Sci 116: 1959–1967, 2003.[Abstract/Free Full Text]
  18. Irrthum A, Devriendt K, Chitayat D, Matthijs G, Glade C, Steijlen PM, Fryns JP, VanSteensel MA, Vikkula M. Mutations in the transcription factor gene SOX18 underlie recessive and dominant forms of hypotrichosis-lymphedema-telangiectasia. Am J Hum Genet 2001: 34–38, 2003.
  19. Itoh M, Furuse M, Morita K, Kubota K, Saitou M, Tsukita S. Direct binding of three tight junction-associated MAGUKs, ZO-1, ZO-2 and ZO-3, with the COOH termini of claudins. J Cell Biol 147: 1351–1363, 1999.[Abstract/Free Full Text]
  20. Jaffe EA, Nachman RL, Becker CG, Minick CR. Culture of human endothelial cells derived from umbilical veins. Identification by morphologic and immunologic criteria. J Clin Invest 52: 2745–2756, 1973.[Web of Science][Medline]
  21. James K, Hosking B, Gardner J, Muscat GEO, Koopman P. Sox18 mutations in the ragged mouse alleles ragged-like and opossum. Genesis 36: 1–3, 2003.[CrossRef][Web of Science][Medline]
  22. Lampugnani MG, Resnati M, Raiteri M, Pigott R, Pisacane A, Houen G, Ruco LP, Dejana E. A novel endothelial-specific membrane protein is a marker of cell-cell contacts. J Cell Biol 118: 1511–1522, 1992.[Abstract/Free Full Text]
  23. Lefebvre V, Dumitriu B, Penzo-Méndez A, Han Y, Pallavi B. Control of cell fate and differentiation by Sry-related high-mobility-group box (Sox) transcription factors. Int J Biochem Cell Biol 39: 2195–2214, 2007.[CrossRef][Web of Science][Medline]
  24. Lelièvre E, Mattot V, Huber P, Vandenbunder B, Soncin F. Ets lowers capillary endothelial cell density at confluence and induces the expression of VE-cadherin. Oncogene 19: 2438–2446, 2000.[CrossRef][Web of Science][Medline]
  25. Liebner S, Fischmann A, Rascher G, Duffner F, Grote EH, Kalbacher H, Wolburg H. Claudin-1 and claudin-5 expression and tight junction morphology in blood vessels of human glioblastoma multiforme. Acta Neuropathol (Berl) 100: 323–331, 2000.[CrossRef][Medline]
  26. Lui WY, Wong EWP, Guan Y, Lee WM. Dual transcriptional control of claudin-11 via an overlapping GATA/NF-Y motif: positive regulation through the interaction of GATA, NF-YA, and CREB and negative regulation through the interaction of Smad, HDAC1 and mSin3A. J Cell Physiol 211: 638–648, 2007.[CrossRef][Web of Science][Medline]
  27. Matsui T, Kanai-Azuma M, Hara K, Matoba S, Hiramatsu R, Kawakami H, Kurohmaru M, Koopman P, Kanai Y. Redundant roles of Sox17 and Sox18 in postnatal angiogensis in mice. J Cell Sci 119: 3513–3526, 2006.[Abstract/Free Full Text]
  28. Mertin S, McDowall SG, Harley VR. The DNA-binding specificity of SOX9 and other SOX proteins. Nucleic Acids Res 27: 1359–1364, 1990.[CrossRef]
  29. Miwa N, Furuse M, Tsukita S, Niikawa N, Nakamura Y, Furukawa Y. Involvement of claudin-1 in the β-catenin/Tcf signalling pathway and its frequent upregulation in human colorectal cancers. Oncol Res 12: 469–476, 2000.
  30. Morita K, Furuse M, Fujimoto K, Tsukita S. Claudin multigene family encoding four-transmembrane domain protein components of tight junction strands. Proc Natl Acad Sci USA 96: 511–516, 1999.[Abstract/Free Full Text]
  31. Morita K, Sasaki H, Furuse M, Tsukita S. Endothelial claudin: claudin-5/TMVCF constitutes tight junction strands in endothelial cells. J Cell Biol 147: 185–194, 1999.[Abstract/Free Full Text]
  32. Naldini L, Blomer U, Gallay P, Ory D, Mulligan R, Gage FH, Verma IM, Trono D. In vivo gene delivery and stable transduction of nondividing cells by a lentiviral vector. Science 272: 263–267, 1996.[Abstract]
  33. Niimi T, Nagashima K, Ward JM, Minoo P, Zimonjic DB, Popescu NC, Kimura S. Claudin-18, a novel downstream target gene for the T/EBP/NKX2.1 homeodomain transcription factor, encodes lung- and stomach-specific isoforms through alternative splicing. Mol Cell Biol 21: 7380–7390, 2001.[Abstract/Free Full Text]
  34. Nitta T, Hata M, Gotoh S, Seo Y, Sasaki H, Hashimoto N, Furuse M, Tsukita S. Size-selective loosening of the blood-brain barrier in claudin-5 deficient mice. J Cell Biol 161: 653–660, 2003.[Abstract/Free Full Text]
  35. Pennisi D, Gardner J, Chambers D, Hosking B, Peters J, Muscat G, Abbot C, Koopman P. Mutations in Sox18 underlie cardiovascular and hair follicle defects in ragged mice. Nat Genet 28: 1473–1480, 2000.
  36. Sakaguchi T, Gu X, Golden HM, Suh E, Rhoads DB, Reinecker HC. Cloning of the human claudin-2 5' flanking region revealed a TATA-less promoter with conserved binding sites in mouse and human for caudal-related homeodomain proteins and hepatocyte nuclear factor-1{alpha}. J Biol Chem 277: 21361–21370, 2002.[Abstract/Free Full Text]
  37. Sakamoto Y, Hara K, Kanai-Azuma M, Matsui T, Miura Y, Tsunekawa N, Kurohmaru M, Saijoh Y, Koopman P, Kanai Y. Redundant roles of Sox17 and Sox18 in early cardiovascular development of mouse embryos. Biochem Biophys Res Commun 360: 539–544, 2007.[CrossRef][Web of Science][Medline]
  38. Simon DB, Lu Y, Choate KA, Velazques H, Al-Sabban E, Praga M, Casari G, Bettinelli A, Colussi G, Rodriguez-Soriano J, McCredie D, Milford D, Sanjad S, Lifton RP. Paracellin-1, a renal tight junction protein required for paracellular Mg2+ resorption. Science 285: 103–106, 1999.[Abstract/Free Full Text]
  39. Turksen K, Troy TC. Bariers built on claudins. J Cell Sci 117: 2435–2447, 2000.[CrossRef]
  40. Umeda K, Ikenouchi J, Katahira-Tayama S, Furuse K, Sasaki H, Nakayama M, MatsuiT, Tsukita S, Furuse M, Tsukita S. ZO-1 and ZO-2 independently determine where claudins are polymerized in tight-junction strand formation. Cell 126: 741–754, 2006.[CrossRef][Web of Science][Medline]
  41. Van Itallie C, Rahner C, Anderson JM. Regulated expression of claudin-4 decreases paracellular conductance through a selective decrease in sodium permeability. J Clin Invest 107: 1319–1327, 2001.[Web of Science][Medline]
  42. Yu ASL, Enck AH, Wayne IL, Schneeberger E. Claudin-8 expression in Madin-Darby canine kidney cells augments the paracellular barrier to cation permeation. J Biol Chem 278: 17350–17359, 2003.[Abstract/Free Full Text]
  43. Zufferey R, Dull T, Mandel RJ, Bukovsky A, Quiroz D, Naldini L, Trono D. Self-inactivating lentivirus vector for safe and efficient in vivo gene delivery. J Virol 72: 9873–9880, 1998.[Abstract/Free Full Text]




This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Supplemental Figures and Tables
Right arrow All Versions of this Article:
294/2/H891    most recent
01248.2007v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Fontijn, R. D.
Right arrow Articles by Horrevoets, A. J. G.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Fontijn, R. D.
Right arrow Articles by Horrevoets, A. J. G.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Visit Other APS Journals Online
Copyright © 2008 by the American Physiological Society.