| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
EXTRACELLULAR MATRIX, CELL INTERACTIONS
-induced alteration of the blood-cerebrospinal fluid barrier in vitro1Institut für Biochemie, Westfälische Wilhelms-Universität, Münster and 2Pädiatrische Infektiologie/Klinik für Allgemeine Pädiatrie, Heinrich Heine Universität, Düsseldorf, Germany
Submitted 31 August 2006 ; accepted in final form 10 May 2007
 |
ABSTRACT |
|---|
 TOP ABSTRACT MATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
(TNF-) on the blood-CSF barrier using an in vitro model based on porcine choroid plexus epithelial cells (PCPEC). TNF- evoked distinct inflammatory processes as shown by mRNA upregulation of intercellular adhesion molecule-1 and vascular cell adhesion molecule-1. The cytokine caused a drastic decrease in transepithelial electrical resistance within several hours representing an enhanced permeability of PCPEC monolayers. In addition, the distribution of tight junction proteins was altered. Moreover, MMP activity in PCPEC supernatants was significantly increased by TNF-, presumably due to a diminished expression of TIMP-3 that was similarly observed. MMP-2, -3, and -9 as well as TIMP-1 and -2 were also analyzed and found to be differentially regulated by the cytokine. The TNF--induced breakdown of the blood-CSF barrier could partially be blocked by the MMP inhibitor GM-6001. Our results show a contribution of MMPs to the inflammatory breakdown of the blood-CSF barrier in vitro. Thus TNF- may mediate the binding of leukocytes to cellular adhesion molecules and the transmigration across the blood-CSF barrier.
choroid plexus; matrix metalloproteases; tight junction; transepithelial electrical resistance; porcine choroid plexus epithelial cells; tumor necrosis factor-
CSF composition has been reported to be markedly changed during several neuroinflammatory diseases (22, 25, 28, 40). Production of proinflammatory cytokines such as tumor necrosis factor- (TNF-) and of major histocompatibility complex molecules by the CP further indicates its important contribution to immunological processes (2, 30, 49). A hallmark of inflammation is the infiltration of leukocytes into affected tissues. The cells need markers like cellular adhesion molecules (CAMs) to find and to attach to their desired sites of passage. The expression of CAM proteins like intercellular adhesion molecule-1 (ICAM-1) and vascular cell adhesion molecule-1 (VCAM-1), known to be generally enhanced in inflammatory processes, has been shown to be regulated by cytokines in particular (reviewed in Ref. 33). The production of CAMs by the CP has led to the assumption of a specific role of the blood-CSF barrier in the immunosurveillance of the CNS (17). In addition to the support by CAMs, immune cells evidently demand a certain impairment of the blood-CNS interfaces for transition. According to this, several authors have demonstrated barrier weakening in numerous inflammatory conditions, particularly regarding the blood-brain barrier (reviewed in Refs. 1, 13, 14, 16, 39).
Matrix metalloproteases (MMPs) and tissue inhibitors of matrix metalloproteases (TIMPs) belong to the CSF components that are altered due to inflammatory disorders (21, 29, 32, 45). MMPs have been known for quite a long time to be produced by a wide range of cell types. In particular, the enzymes and their inhibitors have been described to be expressed and secreted within the CNS (15, 27). Last, but not least, the CP has been shown to be a source of both MMPs and TIMPs (10, 38). Most recently, CP epithelial cells have been demonstrated to produce MMPs in vitro (48). The balance between MMPs and TIMPs, essential for physiological processes of any kind, was found in several studies to be delicately disturbed during neuroinflammatory conditions (reviewed in Refs. 42, 56). MMPs, generally known to degrade the components of the extracellular matrix as well as other proteins like tight junction components (24, 31, 47), have been reported to particularly impact the blood-CNS barriers under pathological conditions (6, 43).
In this study, we examined the effects of TNF- on the blood-CSF barrier in vitro using porcine CP epithelial cells (PCPEC; as characterized in Ref. 23). Occurring inflammation as a result of cytokine treatment was monitored by CAM mRNA expression. Changes in PCPEC permeability caused by TNF- were analyzed by measuring the transepithelial electrical resistance (TER) and by quantifying the passage of [14C]sucrose across the epithelial monolayers. In addition, inflammatory modifications of ZO-1, occludin, and the actin cytoskeleton were examined. Moreover, the production of MMPs and TIMPs by PCPEC was addressed concerning the regulatory impact of TNF-. Finally, alterations of the blood-CSF barrier induced by the cytokine were associated with changes in activity of MMPs, particularly of MMP-3.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Isolated PCPEC were seeded on laminin-coated permeable membranes (no. 3401, Costar, Cambridge, MA) with a diameter of 12 mm using a seeding density of 20 mg wet tissue/cm2 for the purpose of TER measurements, sucrose permeability studies, immunocytochemistry, and zymographic analysis of supernatants. Alternatively, cells were seeded on 25-cm2 culture flasks (Nunc, Wiesbaden, Germany) with a density of 10 mg wet tissue/cm2 for isolation of RNA and MMP activity measurements. The cell culture medium (DMEM-Ham's F-12, Biochrom, Berlin, Germany) was supplemented with 10% (vol/vol) fetal calf serum (PAA "Gold," PAA Laboratories, Pasching, Austria), 4 mM L-glutamine, 5 µg/ml insulin, as well as 100 U penicillin and 100 µg/ml streptomycin. Cytosine arabinoside (20 µM) was added to suppress the growth of contaminating cells. The purity of the cell culture was tested by staining of cell markers like glial fibrillary acidic protein and actin staining, and no contaminating cells of nonepithelial shape could be detected. After reaching confluence on in vitro days 7–9, growth medium was replaced by serum-free culture medium (SFM) to support cell differentiation and improve barrier function (23). All experiments were performed on cells on in vitro days 13–15 which had been cultured in SFM for at least 4 days in vitro and exhibited high TER values of at least 1,000 
·cm2, as controlled by impedance analysis.
Recombinant porcine TNF- was purchased from Pierce Biotechnology (Rockford, IL) and applied at 20 ng/ml in all experiments performed. Using PCPEC on culture filters, the cytokine was added from both the apical and the basolateral fluid compartment. The MMP inhibitor GM-6001 was purchased from Sigma-Aldrich (Deisenhofen, Germany). The cell-permeable superoxide dismutase mimetic and peroxynitrite scavenger Mn(III)tetrakis(4-benzoic acid)Porphyrin chloride (MnTBAP; Calbiochem, San Diego, CA) was used at 50 µM and the radical scavenger N-tert-butyl--phenylnitrone (PBN; Calbiochem) was used at 100 µM and 1 mM, respectively.
Quantitative real-time PCR.
For quantifying MMPs and TIMPs by real-time PCR, RNA from PCPEC grown on culture flasks was isolated using the RNeasy Mini Kit (Qiagen, Hilden, Germany). cDNA was obtained by means of the Reverse Transcription Core Kit (Eurogentec, Seraing, Belgium) using 200 ng RNA per sample. Quantitative real-time PCR (qRT-PCR) was performed using the qPCR Core Kit SYBRgreen I (Eurogentec, Seraing, Belgium) and the GeneAmp 5700 Sequence Detection System (Applied Biosystems, Foster City, CA). cDNA quantities were measured as critical threshold (CT) values, which were then normalized using simultaneously measured 
-actin levels (
CT). Final CT values were obtained by comparing TNF--treated with nonstimulated cells using PCPEC from three different preparations (n = 3; ± SD). Primers used are shown in Table 1.
|
Briefly, the cell-covered Transwell filters were placed in the setup resembling a 12-well plate. The bottom of the basolateral (lower) compartments is covered with a common stainless steel electrode. Individual stamplike electrodes, also made from stainless steel, reach in the apical (upper) filter compartment. Thus, in each well, the cell layer is sandwiched between two electrodes that are used for current injection and voltage reading. Impedance spectra were recorded between 1 and 105·s–1 as described in Ref. 51. TER was calculated by fitting the parameters of an appropriate equivalent circuit to the experimental data using a nonlinear least squares fit.
Permeability measurements. Apical-to-basolateral sucrose permeation was measured across PCPEC monolayers. The experiments were performed according to Ref. 20. Fifty microliters of the apical culture medium were replaced by 50 µl of SFM containing 1 µCi [14C]sucrose. At the same time, the basolateral medium was renewed. After 10, 20, 30, 40, 60, and 80 min, 50-µl samples were taken out of the basolateral compartment and replaced by SFM. The samples were mixed with a tenfold volume of scintillation fluid and the radioactivity was detected by using a scintillation counter (model LS 6500, Beckman). In parallel, the flux of [14C]sucrose across the pure laminin-coated filter was determined. Permeability (in cm/s) was calculated according to Ref. 20.
Immunocytochemistry. PCPEC grown on filters were washed with phosphate-buffered saline (PBS). For phalloidin staining of the actin cytoskeleton as well as tight junction staining of ZO-1 and occludin, PCPEC were fixed with 4% (wt/vol) paraformaldehyde-PBS at room temperature for 20 min and permeabilized by applying PBS containing 0.2% (vol/vol) Triton-X 100 and 1% (wt/vol) bovine serum albumin. Afterward, the cells were washed, and the filter membranes were excised and placed onto glass slides. Immunofluorescence staining was performed using primary polyclonal antibodies rabbit anti-ZO-1 or anti-occludin at 1:250 dilution (overnight at room temperature) obtained from Zymed (San Francisco, CA) and fluorophore-labeled secondary polyclonal antibody chicken anti-rabbit-IgG Alexafluor 594 at 1:250 dilution (60 min at 4°C) obtained from Molecular Probes (Eugene, OR). Actin was stained by incubating the cells with phalloidin Alexa Fluor 488 (Molecular Probes) for 60 min at 4°C. Images were acquired with Zeiss Apotome and Axiovision software (Carl Zeiss, Jena, Germany) using a x63/1.4-objective lens. This system provides an optical slice view reconstructed from fluorescent samples.
MMP activity assay.
Supernatants obtained from PCPEC grown on culture flasks were concentrated 25-fold using Microcon YM-3 centrifugal filter devices (Millipore, Bedford, MA). Supernatants had been conditioned over 48 h without changing the culture medium. TNF- was present during the last 24 h before applying the samples to the assay. MMP activity was detected according to Ref. 31 by means of fluorogenic peptide cleavage. The increase in fluorescence was recorded over 15 min and fitted linearly. The slope of the regression lines is given in relative fluorescence units per minute, mirroring MMP activity. Detection of MMP-3 activity was performed similarly by using the substrate NFF-3 (Calbiochem, Bad Soden, Germany) according to Ref. 36.
Zymography.
Gelatin zymography was conducted to detect MMPs in cultured PCPEC supernatants. In initial experiments, purchased MMP-containing fibroblast supernatants (MMP control-1, Sigma-Aldrich, Taufkirchen, Germany) were used to scale the obtained zymograms together with a prestained molecular mass standard (Sigma-Aldrich, Steinheim, Germany). PCPEC grown on Transwell filters were treated with TNF- for 48 h, and samples from the apical and basolateral compartment were then collected and mixed with equal volumes of nonreducing 2 x SDS sample buffer. Samples were loaded on 12% SDS polyacrylamide gels containing 0.1% (wt/vol) gelatin obtained from porcine skin (Sigma-Aldrich, Deisenhofen, Germany). After electrophoresis, SDS was removed by rinsing the gels 2 x 30 min in 2.7% (wt/vol) Triton-X 100 solution. Triton was removed by washing the gels four to five times in deionized water. Gels were then incubated in developing buffer [50 mM Tris, 200 mM NaCl, 5 mM CaCl2, 0.02% (vol/vol) Brij-35, pH 7.5] overnight at 37°C. The gelatine digestion by MMPs was stopped by application of fixation buffer [50% (vol/vol) methanol, 10% (vol/vol) acetic acid] for 30 min. The gels were incubated for 30 min in staining solution [25% (vol/vol) isopropanol, 10% (vol/vol) acetic acid, 0.25% (wt/vol) bromophenol blue]. After destaining with buffer I [50% (vol/vol) methanol, 10% (vol/vol) acetic acid] and buffer II [10% (vol/vol) methanol, 10% (vol/vol) acetic acid], respectively, the gels were scanned and then dried using drying solution [20% (vol/vol) methanol, 2% (vol/vol) glycerol].
|
RESULTS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
on ICAM-1 and VCAM-1 mRNA expression.
An increased expression of CAMs has previously been shown to be the result of TNF- treatment (reviewed in Ref. 33). In particular, an inflammatory upregulation of CAMs was reported at CP epithelium of mice (46). We now analyzed mRNA expression of ICAM-1 and VCAM-1 to quantify the degree of PCPEC inflammation induced by TNF- (Fig. 1). A drastic upregulation of both CAMs was observed by means of qRT-PCR technique, which was more intense after 4 h than after 24 h [–CT values (± SD): 8.2 ± 1.1 (4 h), 6.1 ± 0.2 (24 h) for VCAM-1; 3.6 ± 1.2 (4 h), 2.3 ± 0.7 (24 h) for ICAM-1].
|
on TER, [14C]sucrose permeability, and tight junction formation.
Since previous investigations have revealed a permeability increasing effect of TNF- on cellular barriers, particularly on the endothelial blood-brain barrier (12, 44, 55), we tested the influence of the cytokine on the tightness of PCPEC monolayers.
As the exemplar in Fig. 2 shows, TNF- caused a marked decrease in TER in a significant biphasic manner, while control cells (no TNF- treatment) showed stable TER values of 
80–90% of initial values during the whole experiment. The TER of PCPEC treated with TNF- fell to 60% of initial values within the first 4 h, then remained stable for another 4 h, and finally started again to drop with 10% per hour. Final TER (measured after 16–20 h of TNF- treatment in different experiments) was in the range of 30–40% (500–400 ·cm2) of initial values. In general, cells from different preparations showed only slight variations concerning the impact of TNF- on the TER. A drop in TER of 10–15% within the first 2–4 h was seen for all cells in all experiments, apparently due to the setup itself.
|
for 18 h evoked a distinct increase in sucrose permeability. Using PCPEC taken from different preparations, factors of inflammatory permeability augmentation were in the range of 2.9–5.1. As can be seen in Fig. 3B, sucrose permeability increases with time of TNF- treatment. Whereas after 2–4 h all cells showed similar permeability values in the range of 3–4 cm–7/s, after 24 h control cells (no TNF- treatment) exhibited permeability values of 2.0 ± 0,3 cm–7/s, whereas addition of the cytokine evoked a rise in permeability to 11.3 ± 1.7 cm–7/s.
|
-induced changes in PCPEC permeability, it seemed reasonable to also analyze alterations of tight junctions after cytokine stimulation. In the present work, we checked the localization of ZO-1, occludin, and the actin cytoskeleton in PCPEC after 24 h of TNF- treatment by immunofluorescence studies (Fig. 4). The Apotome images show an xy en face view of a cell culture monolayer shown in a maximum-intensity projection through the z-axis as well as xz- and yz-stack. TNF--treated cells showed a fuzzy and blurred distribution pattern, whereas nonstimulated cells displayed a clear and smooth protein expression at the cell borders. TNF- also induced dramatic polarized F-actin restructuring with stress fiber appearance in cabled bundles in the basolateral cell compartment. In some areas, we found reduced colocalization of ZO-1 and occludin with the perijunctional actin belt, whereas a subset of F-actin fibers remained colocalized. The protein expression levels of ZO-1 and occludin, however, were not altered after application of TNF- (not shown).
|
on MMP activity.
Because many cytokines, including TNF-, have been shown to be responsible for an increased MMP production (reviewed in Ref. 41), we measured MMP activity as relative fluorescence units (RFU) per minute in concentrated PCPEC supernatants by detecting the degradation of an appropriate fluorophore substrate with time. Whereas untreated cells showed 2.0 ± 0.2 RFU/min (n = 6; ± SD of mean), this value was doubled to 4.0 ± 0.4 RFU/min (n = 6; ± SD of mean) after 24 h of TNF- treatment (Fig. 5A).
|
on MMP-3 activity by utilizing a different peptide (Fig. 6). The cytokine evoked a drastic rise in MMP-3 activity. Coincubation with the broad spectrum MMP inhibitor GM-6001 (50 µM) weakened the increase of activity 1.9-fold. Similarly, a strong upregulation of the MMP-3 gene was observed (Fig. 7), which was slightly more intense after 4 h than after 24 h of cytokine treatment [–CT values (± SD): 6.3 ± 0.3 (4 h), 5.7 ± 0.4 (24 h)].
|
|
on the one hand evokes an increased MMP activity in cell supernatants and on the other hand decreases the TER. It seemed therefore reasonable to hypothesize a direct contribution of the proteases to the decrease of PCPEC tightness. Indeed, as shown in Fig. 5B, coincubation with 50 µM GM-6001, beginning at 60 min before applying TNF- and starting the measurement, diminished the drop in TER caused by the cytokine. TER was at 40.9 ± 1.9% of initial values after 16 h of PCPEC treatment with TNF- only (n = 12; ± SD of mean) and at 54.2 ± 2.2% when GM-6001 was also present (n = 12; ± SD of mean). At the same time, the inhibitor alone caused a marginal not significant decrease in TER compared with nontreated PCPEC (not shown). Preincubation of PCPEC with the antioxidants MnTBAP (50 µM) or PBN (100 µM and 1 mM) could not diminish the drop in TER induced by TNF- (not shown).
Effect of TNF- on TIMP-3 mRNA expression.
Because TIMP-3 has been described as the most prominent MMP inhibitor at CP epithelium (38), we checked whether its expression by PCPEC was downregulated by TNF- to possibly provide an explanation of the increase in MMP activity as a result of cytokine treatment. Indeed, TIMP-3 mRNA expression by PCPEC was significantly decreased after TNF- treatment (Fig. 8). The effect was more pronounced after 24 h than after only 4 h of cytokine exposure [–CT values (± SD): –1.5 ± 0.5 (24 h), –0.7 ± 0.3 (4 h)].
|
on other TIMPs and MMPs.
To elucidate the possible contributions of other TIMPs and MMPs to the observed effects of TNF- on PCPEC, further qRT-PCR studies were performed. Whereas MMP-2 and TIMP-2 mRNAs were found to be not significantly regulated by the cytokine (not shown), MMP-9 and TIMP-1, unexpectedly, showed opposed effects after TNF- treatment relating to the measured increase in MMP activity (Figs. 7 and 8). While both genes were not markedly regulated after only 4 h of cytokine exposure, TNF- caused a downregulation of MMP-9 expression and an upregulation of TIMP-1 expression by PCPEC after 24 h [–CT values (± SD): –1.4 ± 0.5 for MMP-9 and 1.5 ± 0.3 for TIMP-1]. Zymographic detection of the gelatinases in apical PCPEC supernatants (Fig. 9) similarly revealed a decrease in pro-MMP-9 secretion in response to TNF- (48 h), especially regarding the MMP-9 dimer. Production of pro-MMP-2 seems to be slightly diminished by the cytokine. Gelatinase secretion into the basolateral PCPEC fluid compartment was less pronounced but qualitatively similar to the apical zymogram shown.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
considering that MMPs are involved in the weakening of the barrier. Primary cultured PCEPC served as a well characterized in vitro model to analyze inflammatory changes in TER and tight junction expression. MMPs were regarded concerning their expression, secretion, and regulation by TIMPs, respectively. The study clearly shows that MMPs contribute to changes in the tightness of the blood-CSF barrier in vitro during inflammatory conditions.
Inflammation of CP epithelium in vitro caused by TNF-.
As can be seen in Fig. 1, the application of TNF- to PCPEC leads to drastic mRNA upregulation of ICAM-1 and VCAM-1, respectively, indicating intense inflammatory processes taking place after cytokine exposure. A decline of about one-fourth concerning the expression of both CAMs could be detected comparing the mRNA levels after 4 and 24 h of cytokine stimulation. According to this, inflammation of PCPEC appears to be rapidly inducible by TNF- and seems to be rather long lasting. In preliminary experiments, we showed that 2.5 ng/ml TNF- also evokes a clear induction of CAM expression (not shown). In an earlier study on CP of mice, the expression of ICAM-1 and VCAM-1, being restricted to CP epithelial cells, turned out to be upregulated during experimental autoimmune encephalomyelitis (46). In the same study, an augmented CAM expression was also observed as the result of stimulation with cytokines such as TNF- using a primary culture model of murine CP epithelial cells. No differences in expression levels were described between different cytokine incubation times of 4 and 16 h, respectively. Because the CAM-mediated binding of lymphocytes to CP epithelium was additionally shown, the authors assumed an important role of the blood-CSF barrier in the immunosurveillance of the CNS. Additionally, Wolburg et al. (54) found CAMs to be exclusively located at the microvilli on the apical surface of CP epithelium and suggested the involvement of the CP in the transmigration of leukocytes leaving the brain ventricles. The fact that the CP is capable to produce both pro-inflammatory cytokines (18, 49) and major histocompatibility complex molecules (17, 37) further supports the assumption of a fundamental significance of the blood-CSF barrier to the immunological communication between CNS and periphery. The results of our present study corroborate the hypothesis of an increased importance of such a communication during inflammatory conditions.
Effects of TNF- on PCPEC barrier function.
Several authors have pointed out a barrier-weakening effect of TNF- regarding diverse cellular systems (7, 12, 34, 44, 55, 57). For PCPEC, a decrease in TER has been shown before to be caused by serum and growth factors (23) or phorbol esters (5). In the present study, we examined the impact of the cytokine on the blood-CSF barrier in vitro built up by high-resistance PCECP. It could be demonstrated that TNF- intensely decreases the TER of PCPEC within several hours (Fig. 2).
In addition to the drop in TER, TNF- evokes a significant rise in PCPEC sucrose permeability (Figs. 3A and B). The detection of a small decrease in permeability with time found for noninflamed PCPEC (Fig. 3B) is due to the fact that the cells pump liquid from the basolateral into the apical filter compartment. Apparently, the cytokine markedly impairs the cultured epithelial barrier system. Such an influence is, for instance, already well known for the endothelial blood-brain barrier (reviewed in Ref. 14).
Because TNF- leads to a clear increase in PCPEC permeability, we additionally addressed the tight junction expression of the cultured barrier. ZO-1 and occludin as well as actin distribution was found to be significantly altered after cytokine exposure (Fig. 4), whereas former experiments had revealed an unchanged expression level of both tight junction proteins. In earlier studies, our laboratory has observed similar modifications of tight junctions (ZO-1, occludin, and claudin-1, respectively) when serum was added to the cells (4). TNF-, in combination with other cytokines, has already been shown to cause changes in tight junction and actin expression on other systems (7, 8, 57). As a result of the inflammatory stimulation, cellular permeability was increased in addition. Furthermore, the authors report on modifications of F-actin, which we report to be also existent in the case of TNF--treated PCPEC. In our experiments, we observed the massive inflammatory appearance of stress fibers and a decreased colocalization of ZO-1 and occludin with actin. This "unhitching" of the tight junction link between the transmembrane proteins and the actin cytoskeleton by decreased protein-protein interactions may lead to actin strands that are no longer apically located or "tethered" (19). We thus speculate that disruption of interactions between ZO-1 and occludin in PCPEC caused by TNF- evokes the inability to maintain transmembrane tight junction proteins at or to recruit them to the apical strand complex, resulting in the formation of ectopic actin strands in the lateral membrane and cytoplasm. In a recent study on interferon-
-pretreated Caco-2 cells, Wang et al. (50) demonstrated a drastic TNF--induced reorganization of tight junction proteins ZO-1, occludin, and claudin-1, with no detectable changes in total cellular amounts. At the same time, TER was reduced. Applying TNF- (together with interleukin-1) to cultured CP epithelial cells of rats, Strazielle et al. (48) could not detect apparent differences concerning the expression of occludin and claudin, respectively. Instead, the authors report a decrease in organic anion transport by the epithelium after cytokine exposure. At the same time, sucrose permeability was not significantly altered (48). It has to be mentioned that for the rat model of the blood-CSF barrier values of sucrose permeability were measured to be much higher than for the porcine epithelium discussed here.
Taken together, our present study, in contradiction to results reported in Ref. 48, provides evidence for an inflammatory breakdown of the blood-CSF barrier in vitro.
Contributions of MMPs to TNF--induced changes in PCPEC permeability.
We have previously demonstrated that the cytokine TNF- inflames the cultured CP epithelium derived from pigs and, in addition, weakens the blood-CSF barrier in vitro. It was a further aim of this study to verify whether MMPs are involved in the impairment of the used model system, because these enzymes have been associated before with several neuroinflammatory disorders (reviewed in Ref. 42) and, in particular, cellular barrier breakdown (6, 31, 35, 43).
We measured an increased MMP activity in concentrated PCPEC supernatants after stimulation of the cells with TNF- compared with controls (Fig. 5A). In nonconcentrated supernatants, no MMP activity could be detected using the chosen system. This is in remarkable contrast to studies performed on cultured porcine brain capillary endothelial cells (31). The authors measured MMP activity in nontreated supernatants using exactly the same setup as we did in the present study. Apparently, the endothelial blood-brain barrier and the epithelial blood-CSF barrier show fundamental differences concerning the secretion of (active) MMPs and/or TIMPs. Because the utilized MMP substrate does not account for stromelysins, we additionally checked the impact of TNF- on MMP-3 and found its activity as well as its mRNA expression to be enhanced as well (Figs. 6 and 7). An upregulation of MMP-3 expression after cytokine exposure was, for example, also described for murine brain astrocytes (53). Because TIMP-3 had been found before to be the outstanding MMP inhibitor at the CP (38), we assumed changes in its expression to be at least in part responsible for the observed increase in MMP activity after cytokine stimulation. Indeed, TIMP-3 mRNA expression by PCPEC was shown to be significantly downregulated by TNF- (Fig. 8). Bugno et al. (9) observed a drastic decrease in TIMP-3 expression by brain capillary endothelial cells stimulated with TNF- and interleukin-1. The authors found TIMP-1, in contrast, to be intensely upregulated, which could also be demonstrated for PCPEC (see below).
In the present study, the decrease in TIMP-3 expression may specifically contribute to the biphasic attenuation of TER after treatment of PCPEC with TNF-, because TIMP-3 is known to inhibit the TNF- converting enzyme (3). A decreased concentration of the MMP inhibitor might therefore cause an increased concentration of TNF- at PCPEC and thereby result in a downstream drop of TER. It is worth noting in this context that TNF- is known to be produced by CP epithelium (49).
In the present study, we could furthermore show that the broad spectrum MMP inhibitor GM-6001 weakens the TNF--induced decrease in TER of PCPEC (Fig. 5B), proving a direct impact of MMPs on inflammatory blood-CSF barrier breakdown. Because the MMP inhibitor is able to prevent the barrier breakdown to a minor extend only, it is obvious that TNF- action on PCPEC takes also place via other pathways than MMP regulation. This seems quite reasonable, because the CP is supposed to be very accurately balanced concerning its barrier function. An occurring inflammation, for instance, should not be exclusively dependent on one single mechanism to keep the running processes well managed. In a study by de Vries et al. (12), for instance, the TNF--induced increase in permeability of the endothelial blood-brain barrier could be reversed by cyclooxygenase inhibition. Furthermore, TNF- can lead to the release of reactive oxygen species, which can also contribute to barrier disruption and cell damage (11). In our experiments, however, the two antioxidants MnTBAP and PBN had no effect on the TNF--induced weakening of barrier function (not shown).
Unlike the drop in TER, the increase of sucrose permeability was not significantly influenced by the MMP inhibitor GM-6001. Although both parameters reflect a paracellular flow across the epithelium, there are striking differences between the two analytic methods. First, sensitivity is much higher for impedance analysis than for detecting sucrose permeability, as can be directly concluded from the degree of statistical spread. Moreover, concerning sucrose permeability transcytotic effects have to be taken into account. Finally, the TER, depending on the cellular tightness, partly reflects the resistance of the epithelial membranes and not only the flux through the intercellular clefts. Therefore, the disagreement between TER and sucrose permeability concerning the measurable impact of MMPs is striking but not completely surprising.
In the present work, TNF--induced modifications of tight junction distribution at PCPEC were not significantly affected by the utilized MMP inhibitor (not shown). It is worth noting that GM-6001 concentrations higher than 50 µM distinctly affected the cultured barrier system and therefore could not be applied to PCPEC. Harkness et al. (24) could block MMP-9-induced alterations of ZO-1 expression by brain capillary endothelium using an MMP inhibitor. It has to be mentioned, however, that the described modifications of the tight junction protein were only subtle and, furthermore, of another kind compared with the ones observed for PCPEC.
Evidence of a direct influence of MMPs on barrier weakening is rarely found. Rosenberg et al. (43) could partially block blood-brain barrier breakdown by applying a synthetic MMP inhibitor. In vitro, Lohmann et al. (31) associated an impairment of brain capillary endothelium with increased MMP activity and could prevent the occurring degradation of occludin by several MMP inhibitors. On cultured CP epithelium of rats, however, increased MMP production was not accompanied by an impairment of cellular integrity (48).
In contradiction to the rise in overall MMP activity measured in PCPEC supernatants, we found TIMP-1 mRNA to be up-regulated and MMP-9 mRNA to be downregulated by TNF- on mRNA level (Figs. 8 and 7). Moreover, mRNA expression levels of MMP-2 and TIMP-2, respectively, were shown to be not significantly altered by the cytokine (not shown). Interestingly, changes in TIMP-1 and MMP-9 mRNA expression by TNF- are delayed compared with the regulation of MMP-3 and TIMP-3, respectively. This observation supports the assumption that in the case of TIMP-1 and MMP-9 the cultured epithelium generates self-protecting mechanisms against the inflammatory upregulation of proteolytic activity. Zymographic analysis of gelatinase secretion by PCPEC, which our laboratory had already carried out before (5), equally revealed downregulation of (latent) MMP-9 as a result of TNF- exposure (Fig. 9) and thus provides further support of the unexpected finding. Most strikingly, secretion of the dimeric form of MMP-9 (mentioned before, for instance, in Ref. 26) is decreased after cytokine exposure. It is worth noting that the production of MMP-9 is known to be most prominently enhanced in various inflammatory conditions. Our results concerning this gelatinase, on the one hand, disagree with a wide range of studies but, on the other hand, indicate a complex reaction of the blood-CSF barrier in vitro to an inflammatory stimulation. In addition to the inflammatory drop in MMP-9 production, a slight downregulation of pro-MMP-2 secretion caused by TNF- could be observed in the majority of the analyzed PCPEC preparations. In general, supernatants for zymographic analysis had to be collected over at least 48 h to obtain visible bands, especially accounting for monomeric MMP-9.
Not only in terms of the impact of TNF- but also concerning the gelatinolytic MMP secretion profile itself, PCPEC show important differences compared with the cultured CP epithelium of rats as described by Strazielle et al. (48), who were the first to demonstrate MMP secretion by the blood-CSF barrier in vitro. For example, active gelatinase species could only be detected as weak bands using concentrated PCPEC supernatants by zymographic analysis (not shown). This finding is in accordance with two other facts. On the one hand, as mentioned above, no overall MMP activity could be measured in nonconcentrated conditioned media. On the other hand, we could show by direct zymographic comparison using commercially available active MMP-2 that the intense gelatinolytic band found at 70 kDa in PCPEC supernatants is due to the latent (but not the active) form of the enzyme (not shown).
One can only speculate about the physiological relevance of the impact of MMPs on inflammatory breakdown of the blood-CSF barrier. Leukocytes penetrating into or leaving the CNS need gates of passage and are thought to find them at the cerebral endothelium and at the site of CSF production. Cellular adhesion molecules expressed by the CP epithelium have been shown to help to closely attach inflammatory cells, allowing subsequent transmigration. MMPs, principally capable of degrading extracellular matrix components as well as tight junction proteins, potentially mediate both the transcellular and the paracellular way of diapedesis.
In summary, the present study clearly shows a contribution of MMPs to the TNF--induced breakdown of the blood-CSF barrier in vitro. The cultured epithelium generates both favoring and antagonizing mechanisms to answer the inflammatory changes induced by the cytokine.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
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 |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
2. Alldinger S, Wunschmann A, Baumgartner W, Voss C, Kremmer E. Up-regulation of major histocompatibility complex class II antigen expression in the central nervous system of dogs with spontaneous canine distemper virus encephalitis. Acta Neuropathol (Berl) 92: 273–280, 1996.[CrossRef][Medline]
3. Amour A, Slocombe PM, Webster A, Butler M, Knight CG, Smith BJ, Stephens PE, Shelley C, Hutton M, Knauper V, Docherty AJ, Murphy G. TNF-alpha converting enzyme (TACE) is inhibited by TIMP-3. FEBS Lett 435: 39–44, 1998.[CrossRef][Web of Science][Medline]
4. Angelow S, Zeni P, Galla HJ. Usefulness and limitation of primary cultured porcine choroid plexus epithelial cells as an in vitro model to study drug transport at the blood-CSF barrier. Adv Drug Deliv Rev 56: 1859–1873, 2004.[CrossRef][Web of Science][Medline]
5. Angelow S, Zeni P, Hohn B, Galla HJ. Phorbol ester induced short- and long-term permeabilization of the blood-CSF barrier in vitro. Brain Res 1063: 168–179, 2005.[CrossRef][Web of Science][Medline]
6. Anthony DC, Miller KM, Fearn S, Townsend MJ, Opdenakker G, Wells GM, Clements JM, Chandler S, Gearing AJ, Perry VH. Matrix metalloproteinase expression in an experimentally-induced DTH model of multiple sclerosis in the rat CNS. J Neuroimmunol 87: 62–72, 1998.[CrossRef][Web of Science][Medline]
7. Blum MS, Toninelli E, Anderson JM, Balda MS, Zhou J, O'Donnell L, Pardi R, Bender JR. Cytoskeletal rearrangement mediates human microvascular endothelial tight junction modulation by cytokines. Am J Physiol Heart Circ Physiol 273: H286–H294, 1997.
8. Bruewer M, Luegering A, Kucharzik T, Parkos CA, Madara JL, Hopkins AM, Nusrat A. Proinflammatory cytokines disrupt epithelial barrier function by apoptosis-independent mechanisms. J Immunol 171: 6164–6172, 2003.
9. Bugno M, Witek B, Bereta J, Bereta M, Edwards DR, Kordula T. Reprogramming of TIMP-1 and TIMP-3 expression profiles in brain microvascular endothelial cells and astrocytes in response to proinflammatory cytokines. FEBS Lett 448: 9–14, 1999.[CrossRef][Web of Science][Medline]
10. Canete Soler R, Gui YH, Linask KK, Muschel RJ. MMP-9 (gelatinase B) mRNA is expressed during mouse neurogenesis and may be associated with vascularization. Brain Res Dev Brain Res 88: 37–52, 1995.[CrossRef][Medline]
11. Corda S, Laplace C, Vicaut E, Duranteau J. Rapid reactive oxygen species production by mitochondria in endothelial cells exposed to tumor necrosis factor-alpha is mediated by ceramide. Am J Respir Cell Mol Biol 24: 762–768, 2001.
12. de Vries HE, Blom-Roosemalen MC, van Oosten M, de Boer AG, van Berkel TJ, Breimer DD, Kuiper J. The influence of cytokines on the integrity of the blood-brain barrier in vitro. J Neuroimmunol 64: 37–43, 1996.[CrossRef][Web of Science][Medline]
13. de Vries HE, Kuiper J, de Boer AG, Van Berkel TJ, Breimer DD. The blood-brain barrier in neuroinflammatory diseases. Pharmacol Rev 49: 143–155, 1997.
14. Deli MA, Abraham CS, Kataoka Y, Niwa M. Permeability studies on in vitro blood-brain barrier models: physiology, pathology, and pharmacology. Cell Mol Neurobiol 25: 59–127, 2005.[CrossRef][Web of Science][Medline]
15. Dzwonek J, Rylski M, Kaczmarek L. Matrix metalloproteinases and their endogenous inhibitors in neuronal physiology of the adult brain. FEBS Lett 567: 129–135, 2004.[CrossRef][Web of Science][Medline]
16. Edens HA, Parkos CA. Modulation of epithelial and endothelial paracellular permeability by leukocytes. Adv Drug Deliv Rev 41: 315–328, 2000.[CrossRef][Web of Science][Medline]
17. Engelhardt B, Wolburg-Buchholz K, Wolburg H. Involvement of the choroid plexus in central nervous system inflammation. Microsc Res Tech 52: 112–129, 2001.[CrossRef][Web of Science][Medline]
18. Eriksson C, Nobel S, Winblad B, Schultzberg M. Expression of interleukin 1 alpha and beta, and interleukin 1 receptor antagonist mRNA in the rat central nervous system after peripheral administration of lipopolysaccharides. Cytokine 12: 423–431, 2000.[CrossRef][Web of Science][Medline]
19. Fanning AS, Jameson BJ, Jesaitis LA, Anderson JM. The tight junction protein ZO-1 establishes a link between the transmembrane protein occludin and the actin cytoskeleton. J Biol Chem 273: 29745–29753, 1998.
20. Franke H, Galla H, Beuckmann CT. Primary cultures of brain microvessel endothelial cells: a valid and flexible model to study drug transport through the blood-brain barrier in vitro. Brain Res Brain Res Protoc 5: 248–256, 2000.[CrossRef][Medline]
21. Gijbels K, Masure S, Carton H, Opdenakker G. Gelatinase in the cerebrospinal fluid of patients with multiple sclerosis and other inflammatory neurological disorders. J Neuroimmunol 41: 29–34, 1992.[CrossRef][Web of Science][Medline]
22. Gray LD, Fedorko DP. Laboratory diagnosis of bacterial meningitis. Clin Microbiol Rev 5: 130–145, 1992.
23. Hakvoort A, Haselbach M, Wegener J, Hoheisel D, Galla HJ. The polarity of choroid plexus epithelial cells in vitro is improved in serum-free medium. J Neurochem 71: 1141–1150, 1998.[Web of Science][Medline]
24. Harkness KA, Adamson P, Sussman JD, Davies-Jones GA, Greenwood J, Woodroofe MN. Dexamethasone regulation of matrix metalloproteinase expression in CNS vascular endothelium. Brain 123: 698–709, 2000.
25. Hauser SL, Doolittle TH, Lincoln R, Brown RH, Dinarello CA. Cytokine accumulations in CSF of multiple sclerosis patients: frequent detection of interleukin-1 and tumor necrosis factor but not interleukin-6. Neurology 40: 1735–1739, 1990.
26. Justicia C, Panes J, Sole S, Cervera A, Deulofeu R, Chamorro A, Planas AM. Neutrophil infiltration increases matrix metalloproteinase-9 in the ischemic brain after occlusion/reperfusion of the middle cerebral artery in rats. J Cereb Blood Flow Metab 23: 1430–1440, 2003.[CrossRef][Web of Science][Medline]
27. Kaczmarek L, Lapinska-Dzwonek J, Szymczak S. Matrix metalloproteinases in the adult brain physiology: a link between c-Fos, AP-1 and remodeling of neuronal connections? Embo J 21: 6643–6648, 2002.[CrossRef][Web of Science][Medline]
28. Khuth ST, Strazielle N, Giraudon P, Belin MF, Ghersi-Egea JF. Impairment of blood-cerebrospinal fluid barrier properties by retrovirus-activated T lymphocytes: reduction in cerebrospinal fluid-to-blood efflux of prostaglandin E2. J Neurochem 94: 1580–1593, 2005.[CrossRef][Web of Science][Medline]
29. Kolb SA, Lahrtz F, Paul R, Leppert D, Nadal D, Pfister HW, Fontana A. Matrix metalloproteinases and tissue inhibitors of metalloproteinases in viral meningitis: upregulation of MMP-9 and TIMP-1 in cerebrospinal fluid. J Neuroimmunol 84: 143–150, 1998.[CrossRef][Web of Science][Medline]
30. Lindsley MD, Patick AK, Prayoonwiwat N, Rodriguez M. Coexpression of class I major histocompatibility antigen and viral RNA in central nervous system of mice infected with Theiler's virus: a model for multiple sclerosis. Mayo Clin Proc 67: 829–838, 1992.[Web of Science][Medline]
31. Lohmann C, Krischke M, Wegener J, Galla HJ. Tyrosine phosphatase inhibition induces loss of blood-brain barrier integrity by matrix metalloproteinase-dependent and -independent pathways. Brain Res 995: 184–196, 2004.[CrossRef][Web of Science][Medline]
32. Lorenzl S, Albers DS, LeWitt PA, Chirichigno JW, Hilgenberg SL, Cudkowicz ME, Beal MF. Tissue inhibitors of matrix metalloproteinases are elevated in cerebrospinal fluid of neurodegenerative diseases. J Neurol Sci 207: 71–76, 2003.[CrossRef][Web of Science][Medline]
33. Meager A. Cytokine regulation of cellular adhesion molecule expression in inflammation. Cytokine Growth Factor Rev 10: 27–39, 1999.[CrossRef][Web of Science][Medline]
34. Mullin JM, Laughlin KV, Marano CW, Russo LM, Soler AP. Modulation of tumor necrosis factor-induced increase in renal (LLC-PK1) transepithelial permeability. Am J Physiol Renal Fluid Electrolyte Physiol 263: F915–F924, 1992.
35. Mun-Bryce S, Rosenberg GA. Gelatinase B modulates selective opening of the blood-brain barrier during inflammation. Am J Physiol Regul Integr Comp Physiol 274: R1203–R1211, 1998.
36. Nagase H, Fields CG, Fields GB. Design and characterization of a fluorogenic substrate selectively hydrolyzed by stromelysin 1 (matrix metalloproteinase-3). J Biol Chem 269: 20952–20957, 1994.
37. Nathanson JA, Chun LL. Immunological function of the blood-cerebrospinal fluid barrier. Proc Natl Acad Sci USA 86: 1684–1688, 1989.
38. Pagenstecher A, Stalder AK, Kincaid CL, Shapiro SD, Campbell IL. Differential expression of matrix metalloproteinase and tissue inhibitor of matrix metalloproteinase genes in the mouse central nervous system in normal and inflammatory states. Am J Pathol 152: 729–741, 1998.[Abstract]
39. Petty MA, Lo EH. Junctional complexes of the blood-brain barrier: permeability changes in neuroinflammation. Prog Neurobiol 68: 311–323, 2002.[CrossRef][Web of Science][Medline]
40. Rieckmann P, Nunke K, Burchhardt M, Albrecht M, Wiltfang J, Ulrich M, Felgenhauer K. Soluble intercellular adhesion molecule-1 in cerebrospinal fluid: an indicator for the inflammatory impairment of the blood-cerebrospinal fluid barrier. J Neuroimmunol 47: 133–140, 1993.[CrossRef][Web of Science][Medline]
41. Ries C, Petrides PE. Cytokine regulation of matrix metalloproteinase activity and its regulatory dysfunction in disease. Biol Chem Hoppe Seyler 376: 345–355, 1995.[Web of Science][Medline]
42. Rosenberg GA. Matrix metalloproteinases in neuroinflammation. Glia 39: 279–291, 2002.[CrossRef][Web of Science][Medline]
43. Rosenberg GA, Estrada EY, Dencoff JE. Matrix metalloproteinases and TIMPs are associated with blood-brain barrier opening after reperfusion in rat brain. Stroke 29: 2189–2195, 1998.
44. Rosenberg GA, Estrada EY, Dencoff JE, Stetler-Stevenson WG. Tumor necrosis factor-alpha-induced gelatinase B causes delayed opening of the blood-brain barrier: an expanded therapeutic window. Brain Res 703: 151–155, 1995.[CrossRef][Web of Science][Medline]
45. Sporer B, Paul R, Koedel U, Grimm R, Wick M, Goebel FD, Pfister HW. Presence of matrix metalloproteinase-9 activity in the cerebrospinal fluid of human immunodeficiency virus-infected patients. J Infect Dis 178: 854–857, 1998.[Web of Science][Medline]
46. Steffen BJ, Breier G, Butcher EC, Schulz M, Engelhardt B. ICAM-1, VCAM-1, and MAdCAM-1 are expressed on choroid plexus epithelium but not endothelium and mediate binding of lymphocytes in vitro. Am J Pathol 148: 1819–1838, 1996.[Abstract]
47. Sternlicht MD, Werb Z. How matrix metalloproteinases regulate cell behavior. Annu Rev Cell Dev Biol 17: 463–516, 2001.[CrossRef][Web of Science][Medline]
48. Strazielle N, Khuth ST, Murat A, Chalon A, Giraudon P, Belin MF, Ghersi-Egea JF. Pro-inflammatory cytokines modulate matrix metalloproteinase secretion and organic anion transport at the blood-cerebrospinal fluid barrier. J Neuropathol Exp Neurol 62: 1254–1264, 2003.[Web of Science][Medline]
49. Tarlow MJ, Jenkins R, Comis SD, Osborne MP, Stephens S, Stanley P, Crocker J. Ependymal cells of the choroid plexus express tumour necrosis factor-alpha. Neuropathol Appl Neurobiol 19: 324–328, 1993.[Web of Science][Medline]
50. Wang F, Graham WV, Wang Y, Witkowski ED, Schwarz BT, Turner JR. Interferon-gamma and tumor necrosis factor-alpha synergize to induce intestinal epithelial barrier dysfunction by up-regulating myosin light chain kinase expression. Am J Pathol 166: 409–419, 2005.
51. Wegener J, Abrams D, Willenbrink W, Galla HJ, Janshoff A. Automated multi-well device to measure transepithelial electrical resistances under physiological conditions. Biotechniques 37: 590, 592–594, 596–597, 2004.
52. Wegener J, Hakvoort A, Galla HJ. Barrier function of porcine choroid plexus epithelial cells is modulated by cAMP-dependent pathways in vitro. Brain Res 853: 115–124, 2000.[CrossRef][Web of Science][Medline]
53. Witek-Zawada B, Koj A. Regulation of expression of stromyelysin-1 by proinflammatory cytokines in mouse brain astrocytes. J Physiol Pharmacol 54: 489–496, 2003.[Web of Science][Medline]
54. Wolburg K, Gerhardt H, Schulz M, Wolburg H, Engelhardt B. Ultrastructural localization of adhesion molecules in the healthy and inflamed choroid plexus of the mouse. Cell Tissue Res 296: 259–269, 1999.[CrossRef][Web of Science][Medline]
55. Wong D, Dorovini-Zis K, Vincent SR. Cytokines, nitric oxide, and cGMP modulate the permeability of an in vitro model of the human blood-brain barrier. Exp Neurol 190: 446–455, 2004.[CrossRef][Web of Science][Medline]
56. Yong VW, Power C, Forsyth P, Edwards DR. Metalloproteinases in biology and pathology of the nervous system. Nat Rev Neurosci 2: 502–511, 2001.[CrossRef][Web of Science][Medline]
57. Zech JC, Pouvreau I, Cotinet A, Goureau O, Le Varlet B, de Kozak Y. Effect of cytokines and nitric oxide on tight junctions in cultured rat retinal pigment epithelium. Invest Ophthalmol Vis Sci 39: 1600–1608, 1998.
This article has been cited by other articles:
|
|
 |
|
  P.-L. Yao, Y.-C. Lin, and J. H. Richburg TNF Alpha-Mediated Disruption of Spermatogenesis in Response to Sertoli Cell Injury in Rodents Is Partially Regulated by MMP2 Biol Reprod, March 1, 2009; 80(3): 581 - 589. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
