J. Neurochem. (2010) 113, 447–453.
Normal neuronal functioning is dependent on the blood-brain barrier. This barrier is confined to specialized brain endothelial cells lining the inner vessel wall, and tightly controlling transport of nutrients, efflux of potentially harmful molecules and entry of immune cells into the brain. Loss of blood-brain barrier function is an early and significant event which contributes to inflammation in the brain and subsequent progression of neuronal deficits in a number of brain disorders and has been well-documented for the auto-immune disease multiple sclerosis. Extravasation of cells happens by paracellular transport across the endothelial junctions, transcellularly across the endothelial cells, or both, and requires the active participation of endothelial cells. We and others have shown that this process requires the activity of proteases, including tissue-type plasminogen activator. We here describe a novel role for NMDA receptor, a potential cellular target of tissue-type plasminogen activator, in human brain endothelial cells. Our results show that the NMDA receptor subunit 1 (NR1) is expressed in brain endothelial cells, regulates tissue-type plasminogen activator-induced signal transduction and controls the passage of monocytes through the brain endothelial cell barrier. Together, our results hold significant promise for the treatment of chronic inflammation in the brain.
aminoterminal domain of NR1
experimental allergic encephalomyelitis
extracellular signal related kinase
monocyte chemoattractant protein-1
sodium dodecyl sulfate
tissue-type plasminogen activator
Under normal conditions of immune surveillance in the brain traversal of immune cells from the bloodstream across neurovascular endothelium is strictly regulated and balanced. The brain vasculature, forming the so-called blood-brain barrier (BBB), is morphologically and functionally adapted to tightly regulate entry of molecules and cells thereby maintaining brain homeostasis and normal neuronal functioning.
Extensive extravasation of leukocytes, pre-dominantly monocytes, is an early event in the onset of inflammation playing an important role in the pathogenesis of neurodegenerative diseases including multiple sclerosis (MS), vascular dementia, and stroke (Dirnagl et al. 1999; Akiyama et al. 2000; Zipp and Aktas 2006; Zlokovic 2008). In MS, impaired BBB integrity and entry of circulating cells and molecules disturbs brain homeostasis and ultimately leads to the development of neuronal deficits. Infiltrated monocyte-derived macrophages form the major cell type in perivascular lesions and are key mediators of demyelination and axonal damage, two characteristic features of MS (Bruck et al. 1996).
To date, the precise molecular and cellular processes underlying BBB alterations and transendothelial passage of cells have been incompletely characterized. We and others previously showed that these detrimental processes involve cell adhesion molecules (vascular cell adhesion molecule-1, intercellular adhesion molecule-1, and activated leukocyte cell adhesion molecule), intracellular signaling events, reactive oxygen species and the extracellular proteases tissue-type plasminogen activator (tPA) and matrix metalloproteinases (Yednock et al. 1992; Walters et al. 2002; Gurney et al. 2006; Reijerkerk et al. 2006, 2008; Schreibelt et al. 2007; Cayrol et al. 2008).
A particular protease of interest is tPA. Recombinant tPA is the only Food and Drug Administration-approved drug for intravenous thrombolysis treatment and reopening of blood supply to the ischemic brain. We recently showed that tPA regulates monocyte trafficking across the BBB (Reijerkerk et al. 2008). It is currently unknown which brain endothelial receptors regulate the function of tPA during monocyte transmigration. A potential candidate is the ionotropic glutamate receptor and calcium channel NMDA receptor. NMDA receptors consist of different subunits: NR1 (whose presence is compulsory), NR2A–D, and the NR3A or NR3B subunits which in some cases attach to NR1/NR2 as a fifth subunit and reduce channel permeability for calcium ions. Upon activation, the NMDA receptor is able to trigger a pathway of MAPK that can activate extracellular signal related kinase (ERK) (Krapivinsky et al. 2003). Remarkably, tPA can induce intracellular signalling processes through the NMDA receptor in neurons (Nicole et al. 2001).
The aim of the current study was to elucidate the role of the NR1 subunit of the NMDA receptor in monocyte diapedesis and tPA-induced signal transduction in brain endothelial cells. Together, our results imply that NR1 plays a regulatory and specific role in the early neuro-inflammatory process of monocyte-diapedesis.
Materials and methods
The human brain endothelial cell line hCMEC/D3 (Weksler et al. 2005) was provided by dr. P-O. Couraud (Institut Cochin, Université Paris Descartes, Paris, France.) and grown in Endothelial Cell Basal Medium-2 supplemented with human epidermal growth factor, hydrocortisone, GA-1000, vascular endothelial growth factor, fibroblast growth factor, R3-insulin-like growth factor-1, ascorbic acid and 2.5% fetal calf serum (Lonza, Basel, Switzerland). Monocyte chemoattractant protein-1 (MCP-1) was from Peprotech Inc. (Rocky Hill, NJ, USA). Pefabloc® tPA was purchased from Pentapharm LTD (Basel, Switzerland). MK-801 and collagen type I (calf skin) were from Sigma-Aldrich (St. Louis, MO, USA). Phalloidin-rhodamine was from Molecular Probes (Leiden, The Netherlands). Antibodies against ERK1/2 and phospho(Thr202/Tyr204)-ERK1/2 (Cell Signaling Technology Inc., Danvers, MA, USA) were used for immunoblot analyses. Anti-NMDAζ1 which recognizes the amino acid sequence of the C-terminus of the NMDA type I (NR1) receptor, was purchased from Santa Cruz Biotechnology (Heidelberg, Germany).
The pQE100-Double Tag vector (Qiagen, Valencia, CA, USA), which encodes for His6 at the amino terminus of the NR1 cDNA (Fernandez-Monreal et al. 2004) was transformed into Escherichia coli (M15 strain) and recombinant aminoterminal domain of NR1 (ATD-NR1) was purified from isopropyl 1-thio-β-d-galactopyranoside-induced cultures (on a nickel affinity matrix as described by the manufacturer Qiagen).
Immunofluorescence analysis of chronic experimental autoimmune encephalomyelitis tissue
Chronic experimental autoimmune encephalomyelitis (EAE) was induced in 8- to 12-week-old female (FVB) mice via subcutaneous immunization with 200 μg recombinant myelin oligodendrocyte glycoprotein (1–125) in an emulsion mixed (volume ratio 1 : 1) with Complete Freund’s Adjuvant (Difco Laboratories, Detroit, MI, USA) containing 500 μg of heat-killed Mycobacterium tuberculosis H37Ra (Difco). Animals were treated intraperitoneally (i.p.) with 200 ng pertussis toxin derived from Bordetella pertussis (Sigma, Zwijndrecht, The Netherlands) in 200 μL saline at the time of, and after 24 h following immunization. Recombinant myelin oligodendrocyte glycoprotein (1–125) was synthesized as described (Abdul-Majid et al. 2000). Clinical symptoms in mice were examined daily and scored as described (Breij et al. 2005). All experimental procedures were reviewed and approved by the Ethical Committee for Animal Experiments of the VU University Medical Center (Amsterdam, The Netherlands). Animals were kept under standard laboratory conditions and received food and water ad libitum. Leukocyte infiltration and endothelial cells were analysed in animals afflicted with complete hindlimb paralysis using antibodies against CD45 and PECAM-1 (CD31). The tight junction protein zona-occludens-1 was stained with rabbit anti-zona occludens-1 (ZO-1) from Zymed (San Francisco, CA, USA). Nuclei were stained with 4′,6-diamidino-2-phenylindole (DAPI).
Quantitative real-time RT-PCR
One μg of total RNA was reverse-transcribed using the Promega RT system (Promega, Charbonnieres, France; RT: 42°C for 1 h). Two primers were designed for each gene using the Beacon Designer software (Bio-Rad, Marnes-la-Coquette, France). Primer alignments were performed with the BLAST database in order to ensure the specificity of primers.
NR1: GTGCCTCAGTGTTGCTACGG and CCTTCTTGTTGCTGTTGTTCACC; NR2a: ACGTCTACCTTCTTCCAGTTTGG and GCCAGTCATAATCCTGCATGATCT; NR2b: CCCAGTCTGGTGGCAGGG and GCAGCAGTGGTGATTATGGCA; NR2d: CGCTATGGTCGCTTCCTGCA and ACGATGACAAACGGCCTTTCC; NR2c: CCTCAAGATCAGTACCCACCTTTC and GTGTGCATGTCACGGTAGTTACT; NR3a: CTGAGTTTGTGGTGGCTGCTG and GATGCGCTTGAGGATCTGGC; NR3b: GAGGGCTATGGGATCGGACTG and AAGGCACCATCTTGTACCACTTG; Cyclophilin: CAGGGTGGTGACTTTACACGC and TGTTTGGTCCAGCATTTGCCA.
PCR solutions were prepared with RNase-free water containing primers and IQ SYBR Green Supermix (Bio-Rad). For PCR amplification, 20 μL of mix were added to 5 μL of reverse transcription reaction previously diluted (1 : 20). Two negative controls were performed during each experiment. In the first control, we used samples without reverse transcription as a template in order to control contamination of RNA with genomic DNA. In the second control, we used RNase-free water instead of cDNA to prove that qPCR mixes were not contaminated with DNA. Assays were run in triplicate on the iCycler iQ real-time PCR detection system (Bio-Rad). The amplification conditions were as follows: Hot Goldstar enzyme activation, 95°C for 3 min; 50 cycles of PCR at 95°C, 15 s and 60°C, 1 min. The levels of expression of interest gene were computed as follows: relative mRNA expression = 2(Ct of Cyclophilin − Ct of gene of interest) where Ct is the threshold cycle value.
Transendothelial migration of primary monocytes
Human blood monocytes were isolated from buffy coats of healthy donors by Ficoll gradient and CD14-positive beads (Elkord et al. 2005). Primary monocyte transmigration through a monolayer of hCMEC/D3 was assayed using time-lapse microscopy as described previously (Reijerkerk et al. 2008). For MCP-1 stimulated transmigration studies hCMEC/D3 cells were cultured onto collagen-coated transwell filters (pore size 4 μm; Corning Incorporated, Corning, NY, USA). Monocytes were suspended in Endothelial Cell Growth Medium-2 (Lonza, Basel, Switzerland) containing 2.5% fetal calf serum at a concentration of 1 × 106/mL and 100 μL was added to the upper compartment of the transwell. As a chemoattractant recombinant human MCP-1 was diluted to 10 ng/mL in 600 μL endothelial cell growth medium and added to the lower chamber. The wells were incubated for 8 h at 37°C, 5% CO2. The suspension of transmigrated monocytes was harvested and quantified using beads (Flow-count fluorospheres, Beckman Coulter, Inc., Brea, CA, USA) and subsequent FACScan flow cytometer (Becton-Dickinson, San Jose, CA, USA) analyses. To test the role of NR1 in monocyte diapedesis, inhibitors including 100 μM MK-801 or 0.1 μM of a recombinant aminoterminal fragment of NR1 comprising its tPA binding site, were added simultaneously with the monocytes to the abluminal side of endothelial cell monolayers.
hCMEC/D3 cells were washed with ice-cold phosphate-buffered saline and lysed in Cell Lysis Buffer (Cell Signaling Technology Inc, Boston, MA, USA) containing complete protease inhibitor cocktail (Roche, Almere, The Netherlands) at 4°C for 30 min. Cell nuclei were pelleted by centrifugation and discarded. Protein G-sepharose beads (Pharmacia Biotech, Uppsala, Sweden) were absorbed with anti-NMDAζ1 antibodies which recognize the amino acid sequence of the C-terminus of the NMDA type I (NR1) receptor or antibodies against the tight junction protein ZO-1 (San Francisco, CA, USA), washed and subsequently incubated with cleared cell lysates (0.1–0.5 mg of protein) at 4°C overnight with end-over-end rotation. The washed immune complexes were eluted with sodium dodecyl sulfate (SDS) sample buffer (100 mM Tris–HCl pH 6.8), 4% SDS, 20% glycerol and 5%β-mercaptoethanol) and heated to 95°C for 5 min, and proteins were resolved on SDS–polyacrylamide gel electrophoresis.
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 less than 0.05.
Human and mouse brain endothelial cells express the NR1 subunit of the NMDA receptor
To investigate the role of NR1 in inflammatory processes associated with the blood-brain barrier, we first established the presence and localization of NR1 in the brain vasculature of EAE and control mice (Fig. 1a–c). NR1 expression, leukocyte infiltration (CD45), and endothelial cells (CD31) were analysed in animals at the peak of the disease, suffering from complete hindlimb paralysis. CD45 stainings indicated the presence of several perivascular infiltrates of leukocytes, including T-cells and monocytes, at this stage of the disease. In addition to widespread expression of NR1 in brains derived from EAE and control mice, the results clearly demonstrated a scattered expression pattern in brain endothelial cells and at endothelial cell-cell junctions. The localization of NR1 at cell-cell junctions in control and EAE brains was confirmed by co-staining with antibodies against the tight junction protein ZO-1 (Fig. 1b and c). Control stainings with secondary antibodies were negative (data not shown).
NR1 is expressed in cultured human brain endothelial cells and mediates tPA-induced activation of ERK
We previously reported that diapedesis of monocytes through brain endothelial cell barriers is dependent on tPA and tPA-induced signal transduction processes, including the activation of the ERK1/2 pathway (Reijerkerk et al. 2008). As NR1 is a potential cellular receptor for tPA possessing cell signalling properties towards ERK1/2 in neurons (Nicole et al. 2001) we next investigated its function in tPA-mediated ERK1/2 activation in brain endothelial cells. Immunofluorescence staining, western blot analyses of immunoprecipitated NR1 and quantitative PCR analyses confirmed the abundant expression of NR1 in cultured human brain endothelial cells (Fig. 2a–c). Importantly, analysis of the immunoprecipitated NR1 corroborated our in vivo findings and showed the association with ZO-1 (Fig. 2b). Next, phosphorylation of ERK1/2 at Thr202/Tyr204 was assayed in human brain endothelial cells treated with tPA for 5 min, which resulted in maximal activation of ERK1/2, in the presence or absence of the NR1 inhibitor MK-801 (100 μM). The results clearly indicate that tPA-induced ERK1/2 activation involves NR1 (Fig. 2d, upper). tPA treatment did not affect NR1 levels as determined by western blot (Fig. 2d, lower).
Monocyte passage of a brain endothelial barrier involves NR1
Next, we tested whether NR1 has a functional role during monocyte diapedesis, one of the early processes of EAE lesion development, using an in vitro model for the brain endothelial barrier. The involvement of NR1 in transendothelial migration of primary monocytes was shown by using specific inhibitors of NR1 (Fig. 3a and b). First, passage by monocytes of a confluent monolayer of endothelial cells cultured on collagen-coated culture plates was reduced by the NMDA receptor blocker MK-801 (69.2 ± 4.6%; n = 4; p < 0.0027 vs. control) or Pefabloc tPA (58.2 ± 2.0%; n = 4; p < 0.0001 vs. control).
Second, MCP-1 induced transendothelial passage of monocytes in a transwell system was almost completely abolished upon MK-801 or Pefabloc tPA treatment (control: 197.7 ± 6.7%; MK-801: 112.8 ± 8.5%; p = 0.0002 vs. control; Pefabloc tPA: 129.8 ± 18.7%; p = 0.0142 vs. control).
Finally, to allow transient and specific prevention of the interaction of tPA with the NR1 subunit we used purified recombinant N-terminal domain of NR1 (ATD-NR1) and tested the influence on monocyte transmigration. The results show that the exogenous treatment with ATD-NR1 (0.2 μM) slightly reduced monocyte transmigration under normal conditions (Fig. 3c; 80.5 ± 5.5%; n = 8; p = 0.0233 vs. control) but almost completely abolished MCP-1 stimulated passage of monocytes (Fig. 3d; control: 193.0 ± 19.5%; ATD-NR1: 99.7 ± 5.5%; p = 0.01 vs. control; n = 3).
Our current observations show that the NR1 subunit of the calcium channel NMDA receptor regulates monocyte transmigration of the brain endothelial barrier.
We have previously shown that tPA is secreted into the extracellular compartment upon interaction of monocytes with brain endothelial cells (Reijerkerk et al. 2008). The current results show that human brain endothelial cells express the NR1 subunit of NMDA receptor which is functionally involved in cellular migration. Activation of this receptor can lead to a loss of the in vitro brain endothelial barrier (Koenig et al. 1992; Krizbai et al. 1998; Sharp et al. 2003; Scott et al. 2007). Our results point to a mechanism of tPA-induced signal transduction in brain endothelial cells and monocyte diapedesis. First, activation of ERK1/2 in brain endothelial cells induced by tPA was inhibited by NMDA receptor blockade and second, monocyte diapedesis was reduced by a recombinant fragment of NR1, comprising its binding site for tPA. Our results are in accordance with an earlier study showing that ERK activation in brain endothelial cells significantly contributes to monocyte diapedesis (Reijerkerk et al. 2008).
The underlying mechanism of tPA-mediated NMDA receptor activation remains controversial, but there is no doubt that tPA can modulate NMDA receptor function. In vitro and in vivo analyses have demonstrated that tPA can bind and cleave NR1 and thereby exacerbates NMDA receptor calcium influx in neurons and neuronal injury (Nicole et al. 2001; Fernandez-Monreal et al. 2004; Benchenane et al. 2007). As tPA treatment of brain endothelial cells did not affect NR1 levels, our results point to a different role for NR1 in brain endothelium which is distinct from its function in neurons. Others have reported a proteolysis-independent effect of tPA on NMDA receptor-mediated intracellular effects through direct interaction of tPA with the NMDA receptor (Medina et al. 2005). Interestingly, recent data provide evidence that tPA-mediated enhancement of NMDA receptor function requires a co-receptor for tPA, low density lipoprotein receptor-related protein 1 (Martin et al. 2008; Samson et al. 2008), a cell-membrane protein also expressed in brain endothelial cells (Wang et al. 2003).
It has been shown that inflammatory polymorphonuclear leukocytes (Collard et al. 2002) and mononuclear leukocytes, either HIV-infected (Jiang et al. 2001) or after inflammatory stimulation (Klegeris et al. 1997) can release glutamate. The fact that in our experiments glutamate levels were not affected by co-culturing monocytes and brain endothelial cells (data not shown) indicates the involvement of another factor, like tPA, which triggers the NMDA receptor pathway.
Our data suggest that the NMDA receptor could provide a potential target for the treatment of MS, an option which is supported by previous findings. The NMDA receptor antagonists memantine and MK-801 (Wallstrom et al. 1996; Bolton and Paul 1997; Paul and Bolton 2002) and also riluzole (Gilgun-Sherki et al. 2003) treatment, all reduce neurological deficits in experimental autoimmune encephalomyelitis (EAE), the animal model of MS. It has been suggested that NMDA receptor involvement coincides with EAE symptom onset and BBB breakdown (Paul and Bolton 2002), processes associated with monocyte diapedesis. Indeed, it has recently been shown that NMDA receptor antagonism can reduce the infiltration of immune cells into the central nervous system of EAE mice (Basso et al. 2008). Investigations in non-immune models have also identified the NMDA receptor as an important mediator of cerebral vessel disruption. Activation of the NMDA receptor increases intracellular Ca2+ and subsequent production of damaging intracellular reactive oxygen species (Sharp et al. 2005) and peroxynitrite (Scott et al. 2007) in cerebral endothelial cells, which underlies the subsequent loss of the tight junction protein occludin and barrier function (Sharp et al. 2003; Andras et al. 2007). Others have shown that the NMDA receptor antagonist MK-801 maintains brain capillary endothelial cell barrier function during hypoxia (Giese et al. 1995) or following cryogenic injury (Koenig et al. 1992) and reverses tissue plasminogen activator-induced ischemic injury (Kilic et al. 2005). Finally, recent studies have indicated a role of NMDA receptor in hypoxia-induced blood-brain-barrier disruption and leukocyte adhesion (Kuhlmann et al. 2008).
Together, our results point out that the NR1 subunit of the NMDA receptor is expressed by human brain endothelial cells and controls monocyte transmigration. Moreover, our findings suggest that NR1 mediates tPA function at the blood-brain barrier, as we have described previously (Reijerkerk et al. 2008). These studies could pave the way for the development of novel means for anti-inflammatory treatment of neurological disease.
This work was supported by grants from the Netherlands Organization of Scientific Research (AR, GK, HEdV), MS Research Foundation (HEdeV), Senter Novem (SMAvdP) and Hersenstichting Nederland (AR). The authors have no conflicting financial interests.