| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Neuroimmunology Research Group, Molecular Cell Biology and Immunology, VU University Medical Center, Amsterdam, The Netherlands
1Correspondence: Neuroimmunology Research Group, Molecular Cell Biology and Immunology, VU University Medical Center, P.O. Box 7057, 1007 MB Amsterdam, The Netherlands. E-mail: a.reijerkerk{at}vumc.nl
ABSTRACT
The blood-brain barrier (BBB), a selective barrier formed by endothelial cells and dependent on the presence of tight junctions, is compromised during neuroinflammation. A detailed study of tight junction dynamics during transendothelial migration of leukocytes has been lacking. Therefore, we retrovirally expressed green fluorescent protein (GFP) fused to the N-terminus of the tight junction protein occludin in the rat brain endothelial cell line GP8/3.9. Confocal microscopy analyses revealed that GFP-occludin colocalized with the intracellular tight junction protein, ZO-1, localized at intercellular connections, and was absent at cell borders lacking apposing cells. Using live cell imaging we found that monocytes scroll over the brain endothelial cell surface toward cell-cell contacts, induce gap formation, which is associated with local disappearance of GFP-occludin, and subsequently traverse the endothelium paracellularly. Immunoblot analyses indicated that loss of occludin was due to protein degradation. The broad spectrum matrix metalloproteinase (MMP) inhibitor BB-3103 significantly inhibited endothelial gap formation, occludin loss, and the ability of monocytes to pass the endothelium. Our results provide a novel insight into the mechanism by which leukocytes traverse the BBB and illustrate that therapeutics aimed at the stabilization of the tight junction may be beneficial to resist a neuroinflammatory attack.—Reijerkerk, A., Kooij, G., van der Pol, S. M. A., Khazen, S., Dijkstra, C. D., de Vries, H. E. Diapedesis of monocytes is associated with MMP-mediated occludin disappearance in brain endothelial cells.
Key Words: tight junction • occludin • blood-brain barrier • matrix metalloproteinase • diapedesis
THE BLOOD-BRAIN BARRIER (BBB) protects the brain from entry of circulating molecules and cells that may disturb the neuroparenchymal microenvironment. This barrier consists of highly specialized capillary endothelial cells lining the inner vessel wall that are connected by intercellular tight junctions. Morphologically, tight junctions consist of a continuous network of parallel, intramembrane fibril structures spanning the apical region of the intercellular cleft of brain endothelium (1
, 2)
. Tight junctions are composed of a combination of transmembrane and cytoplasmic proteins linked to the cytoskeleton, which allows formation of an impermeant seal. To a lesser extent, barrier properties can be attributed to cell-cell adherence junctions found in close vicinity to tight junction complexes in brain endothelial cells (3)
.
Studies have demonstrated that tight junction alterations in brain endothelial cells are a common phenomenon of neurological diseases where inflammatory cell extravasation is a major event, including HIV-associated dementia (4)
and encephalitis (5)
, multiple sclerosis (6
, 7)
, bacterial meningitis (8)
, stroke, and brain trauma (9)
.
Occludin is a 65 kDa protein localized exclusively at tight junctions of epithelial and endothelial cells (10)
. Sequence analyses predict the presence of four transmembrane regions and intracellular localization of the amino and the carboxyl terminus, suggesting the formation of two extracellular loops. Although occludin is not essential for functional tight junction formation (11
, 12)
, its expression is associated with increased junction tightness and decreased monolayer permeability (13
, 14)
. The cytoplasmic carboxy-terminal domain of occludin interacts with several cytosolic accessory proteins, including zona occludens (ZO) family members, ZO-1 (15
, 16)
, ZO-2 (15)
, and ZO-3 (17)
. These proteins are necessary for the formation and maintenance of tight junctions, and they support structures for signaling and actin cytoskeleton protein binding to establish a tight seal still capable of rapid regulation (18
, 19)
.
Cerebral invasion of immune cells through the paracellular route requires mechanisms that regulate tight junction opening. In neuroinflammatory diseases, matrix metalloproteinases (MMPs) occupy a central role in BBB opening and subsequent cerebral influx of immune cells (20
21
22)
. MMPs are a family of zinc endopeptidases comprising 25 members that degrade many extracellular matrix components. The majority of MMPs are secreted as soluble enzymes. Six membrane-bound MMPs have been identified that are implicated in pericellular proteolysis. MMPs are secreted as proenzymes and are activated by autocatalytic cleavage or other proteases, including plasmin. Although MMPs can be produced by a variety of cells, including endothelial cells and leukocytes, it is important to note that activation is highly regulated by cell-cell interaction.
Because tight junctions may play a pivotal role in regulating cerebral influx of inflammatory cells, components of these complexes are of major interest. We focused on the effects of monocytes, immune cells that are substantially involved in inflammatory responses that occur in the central nervous system (CNS), on the expression and localization of occludin during diapedesis. In the present study we describe a brain endothelial cell line expressing occludin N-terminally tagged to green fluorescent protein. Using live cell imaging, we show that transendothelial migration of monocytes coincides with the local disappearance of occludin and subsequent endothelial gap formation. Simultaneous administration of BB-3103, a broad-spectrum inhibitor of MMPs, prevented occludin disappearance, gap formation, and diapedesis. Our results further strengthen the idea that extracellular proteases, including MMPs, play an important role in BBB opening and subsequent immune cell diapedesis.
MATERIALS AND METHODS
Chemicals and antibodies
Ham’s F12 medium, penicillin, streptomycin, RPMI 1640 medium, L-glutamine, fetal calf serum (FCS), and trypsin/EDTA were obtained from Gibco BRL (Breda, The Netherlands). Collagen type I (calf skin) and collagenase type I were from Sigma (St. Louis, MO, USA). Mouse anti-occludin and rabbit anti-ZO-1 antibodies were from Zymed (San Francisco, CA, USA). 3H-Indolium, 5-[[4-(chloromethyl)benzoyl]amino]methyl]-2-[3-(1,3-dihydro-3,3-dimethyl-1-octadecyl-2H-indol-2-ylidene)-1-propenyl]-3,3-dimethyl-1-octadecyl-, chloride (CM-DiI), 2',7'-bis-(2-carboxyethyl)-5-(and-6)-carboxyfluorescein, acetoxymethyl ester (AM) (BCECF-AM), 7-aminoactinomycin-D (7AAD), rabbit anti-GFP antibodies and Alexa 546-conjugated goat anti-rabbit and goat anti-mouse antibodies were from Molecular Probes (Leiden, The Netherlands). Mouse anti-rat intercellular adhesion molecule-1 (ICAM-1; clone 1A29) was a gift from Dr. T. Tamatani. Mouse anti-rat vascular cell adhesion molecule-1 (VCAM-1; clone SF10) was a gift from Dr. R. Lobb, Biogen, Inc. (Cambridge, MA, USA). Monoclonal antibodies against rat major histocompatibility complex class II (MHC class II; OX-6) were from Serotec (Oxford, UK). Rabbit anti-human von Willebrand factor (vWF) and horseradish peroxidase-conjugated antibodies were from DakoCytomation B.V. (Heverlee, Belgium). BB-3103 was from British Biotech Pharmaceuticals Ltd. (Oxford, UK).
Cell culture
The Lewis rat brain endothelial cell line (GP8/3.9) (23)
was routinely cultured in collagen type I-coated flasks in Ham’s F12 medium supplemented with 10% FCS (heat inactivated), 2 mM L-glutamine, 100 µg/ml streptomycin, and 100 U/ml penicillin.
The rat alveolar macrophage cell line NR8383 was obtained from the American Type Culture Collection (ATCC) (Manassas, VA, USA). These nonadherent, not fully differentiated cells with monocytic characteristics (24)
were cultured in RPMI 1640 medium supplemented with 10% FCS (heat inactivated), 2 mM L-glutamine, 100 U/ml penicillin, and 100 µg/ml streptomycin. Every 3 days cells were harvested and subcultured in fresh medium at 7.0 x 105 cells/ml.
Construction of a brain endothelial cell line overexpressing GFP-occludin
The full-length DNA sequence encoding rat occludin was amplified from pBOS/oc.FLAG (a generous gift from Norimasa Sawada and Yasuo Kokai, Sapporo Medical University School of Medicine, Japan) and inserted into the pEGFP vector (Clontech, Palo Alto, CA, USA). Subsequently, GFP-occludin was subcloned into the modified retroviral vector LZRS-IRES-zeocin (25)
(provided by Dr. P. L. Hordijk, Sanquin Research, Amsterdam, The Netherlands). The resulting construct, LZRS-GFP-occludin-IRES-zeocin, was transfected using calcium phosphate into amphotropic Phoenix retrovirus producer cells (25)
for the generation of helper free amphotropic retroviruses. Virus-containing supernatant was used to transduce GP8/3.9 brain endothelial cells that were pretreated with 1 mg/ml diethylaminoethyl (DEAE) -dextran. After 20 h, cells were allowed to recover in fresh medium. Transduced GP8/3.9 cells were selected with 25 µg/ml Zeocin (Invitrogen). Expression and localization of GFP-occludin were assessed by confocal laser scanning microscopy (CLSM; Leica TCS SP2 AOBS microscope, HCX PL APO 63x/1.30 lens; Leica Microsystems B.V., Rijswijk, The Netherlands). GP8/3.9 cells expressing GFP-occludin were grown on collagen-coated glass coverslips and fixed in 3.7% paraformaldehyde in PBS for 20 min at RT. GFP-occludin was labeled with monoclonal mouse anti-occludin IgG, followed by Alexa 594-conjugated goat anti-mouse IgG. Mounted coverslips were analyzed by sequential excitation at 488 and 594 nm using CLSM.
Flow cytometric analysis
Flow cytometric analysis of ICAM-1, VCAM-1, MHC class II, and vWF expression was performed as described (26)
. Brain endothelial cells were detached from 24-well culture plates by collagenase type I treatment (1 mg/ml). Washed cells were incubated with antibodies against ICAM-1, VCAM-1, MHC class II, or vWF (1 µg/ml) for 30 min at 4°C. Binding was detected using biotinylated secondary antibodies followed by streptavidin-conjugated peridinin chlorophyll-a protein-Cy5.5 (BD Biosciences PharMingen, NJ, USA). Omission of the primary antibodies served as negative control. Fluorescence intensity was determined using a FACScan flow cytometer (Becton & Dickinson, San Jose, CA, USA). FACS analysis was performed on 10,000 viable cells, selected by 7AAD exclusion.
Monocyte adhesion
Monocyte adhesion to brain endothelium was analyzed as described by de Vries et al. (24)
In short, NR8383 monocytes were fluorescently labeled with BCECF-AM for 15 min at 37°C in growth medium, washed, and resuspended in RPMI containing 1% BSA at 1, 0.75, 0.5, 0.25, and 0.125 x 106 cells/ml. Brain endothelial monolayers were washed with RPMI and different amounts of labeled monocytes were allowed to adhere for 1 h at 37°C and 5% CO2. After incubation, nonadherent cells were removed by gentle washing with RPMI containing 0.5% BSA. Adhered cells were lysed with 0.1 M NaOH and fluorescence intensity was measured (Fluostar 32, BMG; excitation 485 nm, emission 535 nm).
Endothelial permeability measurement
GP8/3.9 cells expressing GFP-occludin were seeded at confluency onto collagen-coated Costar Transwell filter (pore-size 0.4 µm; Corning Incorporated, Corning, NY, USA) in growth medium containing 2.5% FCS and grown for 5 days. Paracellular permeability to FITC-dextran (150 kDa, 500 µg/ml in culture medium Sigma) in the apical to basolateral direction and the influence of 2 h preincubation with 1.3 x 105 macrophages/cm2 endothelial monolayer were assayed in the presence or absence of MMP inhibitor. At various time points after addition of FITC-dextran, samples were collected from the acceptor chambers for measurement of fluorescence intensity using a FLUOstar Galaxy microplate reader (BMG Labtechnologies, Offenburg, Germany), excitation 485 nm, emission 520 nm.
Live cell analysis of monocyte migration through GP8-GFP-occludin monolayer
For live analyses, GP8/3.9 cells expressing GFP-occludin were grown on collagen-coated glass coverslips until confluence. After washing with NR8383 growth medium, coverslips were mounted in a confocal laser scanning microscope (CLSM). NR8383 monocytes (5x106/ml) were labeled with 5 µg/ml CM-DiI for 5 min at 37°C, washed and suspended in growth medium. Labeling did not affect cellular functions including migration, adhesion, and adhesion molecule expression in our assays (24
, 27)
. Endothelial transmigration was analyzed after addition of 1 ml containing 1,2 x 106 NR8383 monocytes to confluent endothelial monolayers (8.5x104 NR8383 monocytes/cm2 endothelial cell monolayer, NR8383:EC ratio 1.4:1). Four to eight time-lapse images were collected at different time points for 10 frames at 37°C using CLSM and accumulated. Migration was quantified by counting the number of monocytes that had induced gap formation within 50 min upon association with an intact occludin junction (generally within 10 min after addition of NR8383 monocytes). Data are expressed relative to the number of occludin-associated monocytes at 10 min.
Immunoblotting
For immunoblotting analyses, GP8-GFP-occludin cells were grown to confluence in collagen-coated 24-well plates. For coculture experiments, NR8383 monocytes were harvested, pelleted, and resuspended in culture medium. After washing endothelial cells with NR8383 culture medium, different concentrations of NR8383 monocytes were added for 2 h at 37°C. Cell homogenates were prepared by replacing the culture medium with sodium dodecyl sulfate (SDS) sample buffer containing 5% ß-mercaptoethanol and subsequent heating at 95°C for 5 min to prevent nonspecific proteolysis of occludin (28)
. Samples were separated by SDS-PAGE and blotted onto nitrocellulose membranes. Membranes were blocked with PBS containing 5% normal rabbit serum and incubated with mouse anti-occludin antibody or blocked with 1% BSA, followed by rabbit anti-GFP incubation. Immunoreactive proteins were detected with horseradish peroxidase-conjugated secondary antibodies and visualized using an enhanced chemoluminescence detection kit. Occludin band intensity was quantified using ImageQuant TL software (Amersham Biosciences, NJ, USA) and presented as percentage of control.
Statistical analysis
Statistical analysis was performed with the Student’s t test (Prism 4.0; GraphPad Software, San Diego, CA, USA), and results were considered significant if P was <0.05.
RESULTS
Occludin is localized at cell junctions when expressed in rat brain endothelial cells
To gain insight into tight junction dynamics involved in monocyte diapedesis, we generated a brain endothelial cell line that stably expressed occludin N-terminally tagged to enhanced green fluorescence protein (enhanced GFP; Fig. 1
A). We used the brain endothelial cell line GP8/3.9, which under normal conditions expresses low levels of occludin. Immunoblotting analysis using anti-occludin and anti-GFP antibodies showed that GFP-occludin migrates as a single molecular mass with the expected size of 
90 kDa (Fig. 1B
). CLSM analyses revealed that GFP-occludin was mainly localized at the plasma membrane and was recognized by anti-occludin monoclonal antibodies (Fig. 1C-E
). Moreover, expression was associated with intercellular junctions and was absent at cell borders lacking apposing cells (Fig. 1C
). As shown in Fig. 1F-H
, immunostaining with anti-ZO-1 polyclonal antibodies demonstrates that GFP-occludin colocalized with the intracellular tight junction protein ZO-1 at cell-cell contacts. To investigate the effect of GFP-occludin expression on barrier function, we determined the paracellular permeability of endothelial monolayers to FITC-dextran, with an average molecular mass of 150 kDa. The GFP-occludin expressing cells showed significantly decreased permeability compared with control cells (Fig. 1I
). Mutant cells retained endothelial properties including cell morphology, proliferation rate, and adhesive capacity. Flow cytometric analysis further indicated normal expression of proteins that are involved in cell-cell interaction and/or transendothelial migration. Expression of intercellular adhesion molecule-1 (ICAM-1), vascular cell adhesion molecule-1 (VCAM-1), the cell activation marker major histocompatibility complex class II (MHC class II), and the endothelial marker von Willebrand factor (vWF) were unaffected by GFP-occludin expression compared with GP8/3.9 brain endothelial cells (Fig. 1J
). In addition, firm adhesion kinetics of NR8383 monocytes to rat brain endothelial cells expressing GFP-occludin was similar to nontransduced control GP8/3.9 cells (Fig. 1K
).
|
Real-time analyses of monocyte transmigration
In this study we used the GFP-occludin expressing endothelial cell line for live observation of the dynamics of occludin localization during transendothelial monocyte migration. Endothelial cells were grown on glass coverslips until confluence, incubated with fluorescently labeled NR8383 monocytes, and migration was followed by CLSM. A comparison of collected images at different time points revealed that adhered monocytes first migrate toward a continuous junction of GFP-occludin and within 30 min induce the formation of a gap between two adjacent endothelial cells subsequently facilitating diapedesis (Fig. 2
A). We further observed that gap formation is associated with local disappearance of occludin from the plasma membrane of apposing endothelial cells (Fig. 2A, B
). Our results do not provide evidence for migration of monocytes through endothelial cells, since monocytes exclusively transmigrated the endothelial cell monolayer through cell-cell junctions, as clearly indicated by 3-dimensional analysis of x,y sections (Fig. 2B
). To further explore this notion, we analyzed the impact of NR8383 monocytes on occludin protein level in more detail by immunoblotting. Results showed that addition of monocytes led to a decrease in GFP-occludin levels that correlated highly with the number of monocytes added (Fig. 2C
).
|
MMPs are involved in monocyte migration through occludin junctions
The observed loss of occludin band intensity was indicative of occludin degradation. Occludin contains a putative extracellular MMP cleavage site (29)
, and GP8/3.9 endothelial cells have been shown to generate high levels of MMP-2 and MMP-9 (30)
. As a first step to understand the regulatory mechanism underlying occludin disappearance, we examined the role of MMPs in gap formation and monocyte diapedesis. Therefore, we performed live cell analyses of NR8383 monocyte migration through occludin junctions in endothelial monolayers in the presence or absence of 10 µM BB-3103, a broad-spectrum inhibitor of MMPs. Our results showed that BB-3103 did not affect monocyte adhesion and the subsequent approach of junctional occludin (not shown). Strikingly, addition of BB-3103 potently suppressed monocyte-induced loss of occludin, endothelial gap formation, and diapedesis through junctional occludin (Fig. 3
). Moreover, analyses of monocyte-induced gap formation by means of permeability studies pointed to an important role of MMPs in this process. Inhibition of MMP activity abolished monocyte-induced paracellular permeability to FITC-dextran without affecting basal permeability levels (Fig. 3D
).
|
However, BB-3103 did not block monocyte-induced occludin degradation (not shown), suggesting that endothelial gap formation, which is associated with local disappearance of occludin (regulated by MMPs), and occludin degradation (not regulated by MMPs) may be independent processes.
DISCUSSION
An important feature of the BBB is the presence of specialized endothelial cells connected by tight junctions. Tight junctions form a primary barrier to prevent cerebral entry of blood-borne macromolecules and immune cells via the paracellular route, and tight junction alterations have been demonstrated in several neuroinflammatory diseases. Occludin is a regulatory component of tight junctions. Here, we studied the behavior of occludin during transendothelial migration of monocytes. For this purpose, we generated a rat brain endothelial cell line expressing a GFP-occludin fusion protein, which localized at intercellular junctions along with endogenous ZO-1, another tight junction protein member. Real-time observation of GFP-occludin during monocyte diapedesis revealed that this process is associated with endothelial gap formation and local disappearance of occludin. Moreover, inhibitor studies showed that gap formation and loss of occludin depend on the activity of matrix metalloproteinases.
Our in vitro findings indicate that cellular infiltration may directly contribute to loss of tight junction proteins in neuroinflammation. Abnormal expression of occludin and ZO-1 was observed in inflamed tissue in brains of patients with multiple sclerosis (6
, 7)
and HIV-1-mediated encephalitis (5)
. In addition, 
-carrageenan-induced inflammation led to decreased brain endothelial expression of occludin and enhanced BBB permeability (31
, 32)
. In rats, interleukin (IL) -1ß-induced, neutrophil-dependent BBB breakdown coincided with vascular loss of occludin and ZO-1 (33)
. During experimental autoimmune encephalomyelitis, an animal model of multiple sclerosis, claudin-3 is selectively lost in inflamed vessels (34)
. Similar results are reported from studies of tight junction function in the blood-retinal barrier. During experimental autoimmune uveoretinitis, disrupted claudin-1/3, occludin, and ZO-1 in retinal venules correlates with sites of leukocyte extravasation (35)
. Development of cerebral inflammation on stroke-induced loss of cerebral vessel integrity (36)
could also suggest that reduced endothelial tight junction expression predisposes toward cell extravasation. Our previous findings indicating that reduced cerebrovascular integrity precedes infiltration of inflammatory cells in experimental acute encephalomyelitis in rats point to such a mechanism (37)
.
Although the process of transendothelial migration has been investigated intensely for decades, the pathway by which inflammatory cells cross endothelium is still a subject of debate. Recently, in vitro (38)
and in vivo (39
, 40)
experiments have challenged the idea that leukocytes only extravasate through endothelial cell junctions (the paracellular route) and have revealed that leukocytes can also traverse by a route through endothelial cells (the transcellular route). Our studies using brain endothelial cells in an in vitro system did not provide evidence of transcellular migration. We show that monocytes selectively migrated toward endothelial cell-cell contacts and exclusively transmigrated an endothelial cell monolayer in a paracellular fashion, i.e., through cell-cell junctions. Real-time imaging studies of the adherence junction protein vascular endothelial-cadherin by Luscinskas’ group also reported paracellular trafficking of leukocytes across peripheral endothelium (41
42
43)
. It could well be that stimulation of endothelial cells or monocytes increases the incidence of transendothelial traversal via a transcellular route. For live analyses of occludin behavior during monocyte diapedesis, it was not necessary to activate brain endothelial cells or monocytes with stimulators such as cytokines or chemokines. Transcellular diapedesis through human umbilical vein endothelial cells by monocytes, neutrophils, and lymphocytes was described only through tumor necrosis factor 
-stimulated endothelial cells and treatment with monocyte chemoattractant protein-1, platelet activating factor, and stromal cell-derived factor-1, respectively (38)
. We are the first to report on live analyses of monocyte diapedesis through occludin junctions in cultured brain endothelium. Different vascular beds may have distinct preferences for either paracellular or transcellular passage of leukocytes.
In the brain, transendothelial migration through the paracellular route must require mechanisms that regulate opening of tight junctions. Several studies point to a crucial role of MMPs in tight junction dynamics. For instance, tight junction proteins of the claudin family colocalize with membrane type-1 matrix metalloproteinase (MT1-MMP) and promote MT1-MMP-mediated MMP-2 activation (44)
. Moreover, occludin is absent in invasive cells expressing high levels of MT1-MMP (45)
. It is generally thought that cell-cell interactions involve the extracellular loops of occludin and are mediated by the formation of adhesive homopolymers (46)
. Occludin contains a putative MMP cleavage site in its first extracellular loop, and peptides comprising this loop specifically block occludin-dependent fibroblast aggregation (47)
. Others have shown that monocyte migration through a monolayer of human or rat brain endothelial cells could be inhibited by treatment with a physiological MMP inhibitor, the tissue inhibitor of metalloproteinase (TIMP-1) (48)
. However, no direct role of MMPs in monocyte-associated occludin disappearance and endothelial gap formation has been reported before. Our results show for the first time that diapedesis of monocytes is associated with local disappearance of occludin, which could be blocked by a MMP inhibitor, suggesting active participation of MMPs in these processes.
Immunoblotting analysis suggested that monocyte-induced disappearance of occludin was due to protein degradation. Surprisingly, analyses using anti-occludin or anti-GFP antibodies did not reveal the presence of occludin degradation products. In addition, occludin degradation was not inhibited by MMP blockage. These results suggest that occludin breakdown is independent of, though related to, endothelial gap formation and local disappearance of occludin. Therefore, the exact nature of occludin degradation warrants further study. It has been shown that tight junction stability is regulated by the ubiquitin-proteasome pathway. Occludin interacts directly with the E3 ubiquitin-protein ligase Itch. Moreover, occludin is ubiquitinated in vivo and its turnover is retarded after treatment with a proteasome inhibitor (49)
. Whether such a pathway is involved in occludin degradation remains to be determined. Data published by Harkness and co-workers (30)
show that dexamethasone-induced GP8/3.9 endothelial barrier function is due to decreased MMP-9 activity and mediated by elevated junctional ZO-1 expression in endothelial cells. These results could implicate that occludin disappearance is caused by ZO-1 redistribution, which is dependent on MMP activity. It is also possible that intracellular signaling processes may have a role in occludin degradation. Evidence suggests that signaling activities may affect tight junctions directly by inducing phosphorylation. Occludin binds to extracellular signal related kinase (ERK) (50)
and phosphatidylinositol-3-kinase (51)
, can be hyperphosphorylated on Ser/Thr residues (52)
and is dephosphorylated during tight junction disruption (53)
. Since others have reported that MMPs regulate disruption of cell-cell contacts and proteolysis of occludin in porcine brain capillary endothelial cells treated with phenylarsine oxide, a tyrosine phosphatase inhibitor (54
, 55)
and apoptotic epithelial cells (29)
, it will be interesting to investigate whether MMPs play a general role in occludin redistribution. Studies in other tissues where tight junctions actively contribute to barrier function further highlight such behavior of MMPs. For example, ischemia injury induced alteration of endothelial cell properties in the kidney is associated with occludin degradation and coincides with increased MMP-9 protein levels (56)
. Moreover, in vitro experiments using retinal endothelial cells or corneal epithelial cells suggest that the increase in vascular and epithelial permeability observed in early diabetic retinopathy (57
, 58)
and a model of experimental dry eye (59)
might be facilitated by MMP-dependent activity.
We show that monocyte diapedesis through brain endothelium is associated with loss of endothelial cell-cell contacts and local disappearance of the tight junction protein occludin. Furthermore, our results indicate that MMPs play a crucial role in the initial stages of neuroinflammation by mediating cellular migration through endothelial junctions. The described data will help to unravel the function and dynamics of tight junctions in neuroinflammation, and illustrate that compounds that strengthen the tight junction or protect tight junction proteins from enzymatic cleavage are attractive therapeutical candidates and may be beneficial in neurological disorders to resist an inflammatory attack.
ACKNOWLEDGMENTS
We thank Drs. Norimasa Sawada and Yasuo Kokai, Sapporo Medical University School of Medicine, Japan, for providing rat occludin cDNA, Esther Hendrikx for technical assistance on the confocal laser scanning microscope, and Drs. Jack van Horssen, Sipke Dijkstra, and Gerty Schreibelt for critical reading of the manuscript. This work was supported by grants from the Netherlands Organization of Scientific Research (androgen receptor, G.K., H.E.deV., S.K., grant 016.046.314), MS Research Foundation (H.E.deV., grant 00–427), and Senter Novum (S.v.deP.).
Received for publication March 29, 2006. Accepted for publication July 17, 2006.
REFERENCES
This article has been cited by other articles:
|
|
 |
|
  A. Reijerkerk, G. Kooij, S. M. A. van der Pol, T. Leyen, B. van het Hof, P.-O. Couraud, D. Vivien, C. D. Dijkstra, and H. E. de Vries Tissue-Type Plasminogen Activator Is a Regulator of Monocyte Diapedesis through the Brain Endothelial Barrier J. Immunol., September 1, 2008; 181(5): 3567 - 3574. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. J. Grab, E. Nyarko, N. C. Barat, O. V. Nikolskaia, and J. S. Dumler Anaplasma phagocytophilum-Borrelia burgdorferi Coinfection Enhances Chemokine, Cytokine, and Matrix Metalloprotease Expression by Human Brain Microvascular Endothelial Cells Clin. Vaccine Immunol., November 1, 2007; 14(11): 1420 - 1424. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Schreibelt, G. Kooij, A. Reijerkerk, R. van Doorn, S. I. Gringhuis, S. van der Pol, B. B. Weksler, I. A. Romero, P.-O. Couraud, J. Piontek, et al. Reactive oxygen species alter brain endothelial tight junction dynamics via RhoA, PI3 kinase, and PKB signaling FASEB J, November 1, 2007; 21(13): 3666 - 3676. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
