| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Endothelial and Epithelial Cell Biology Research Unit, Division of Cell Biology, Institute of Ophthalmology, University College London, London EC1V 9EL, UK;
* Institut Cochin, Département de Biologie Cellulaire, CNRS UMR8104, INSERM567, 75014 Paris, France;
Laboratoire d’Immunologie, Faculté de Médecine, Universite de la Méditerranée, Marseille, France; and
Division of Biochemistry and Molecular Biology, University of Glasgow, Glasgow, Lanark, Scotland G12 8QQ
1Correspondence: Division of Cell Biology, Institute of Ophthalmology, University College London, Bath Street, London EC1V 9EL. E-mail: J.Greenwood{at}ucl.ac.uk
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
subunits and inhibition (>80%) of lymphocyte migration without affecting lymphocyte adhesion. Aortic and high endothelial venule EC treated identically resulted in only partial inhibition of lymphocyte migration (<40%). Expression of ribosylation-resistant (PTX-insensitive) G-protein subunits in brain EC restored their ability to support lymphocyte migration after pretreatment with PTX. Treatment of brain EC with PTX did not inhibit ICAM-1-stimulated tyrosine phosphorylation of focal adhesion kinase, suggesting the effects of PTX in inhibiting EC facilitation of lymphocyte migration are distinct from activation of EC through ICAM-1. We conclude that a heterotrimeric G-protein-mediated signaling pathway in brain EC is essential for efficient transendothelial migration of T lymphocytes into the brain.—Adamson, P., Wilbourn, B., Etienne-Manneville, S., Calder, V., Beraud, E., Milligan, G., Couraud, P.-O., Greenwood, J. Lymphocyte trafficking through the blood–brain barrier is dependent on endothelial cell heterotrimeric G-protein signaling.
Key Words: endothelium • lymphocyte migration • EC
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The capture of circulating lymphocytes by vascular endothelium has been studied extensively and the basic principles governing this process delineated (1)
. These studies have shown that lymphocyte recruitment by vascular endothelia is regulated by the expression of surface adhesion molecules on both the lymphocyte and EC, which slow and secure passing lymphocytes. The extent to which this occurs is governed by the degree of adhesion molecule expression and by their state of activation. Adhesion molecule blockade studies have demonstrated that in non-CNS vessels, the process of migration, or diapedesis, is mediated largely by the pairing of LFA-1 on the lymphocyte and ICAM-1 on the endothelium (2)
. In the CNS, these molecules also appear to be central to lymphocyte adhesion (3
, 4)
and migration (5
6
7
8)
. Lymphocyte diapedesis through CNS endothelial monolayers in vitro is mediated by LFA-1/ICAM-1 interaction under basal conditions (primary cultures of EC constitutively express ICAM-1), with VLA-4/VCAM-1 interaction playing an additional minor role on cytokine-activated endothelia (6
, 8)
.
Although lymphocyte migration is largely governed by the activation state of the lymphocyte (9
, 10)
, it is now clear is that the EC plays an active part in controlling the process of lymphocyte diapedesis through the tight vascular barrier of the CNS. The ligation of ICAM-1 molecules on rat brain EC lines by cross-linking antibodies (used to mimic lymphocyte adhesion through the LFA-1 ligand) or T lymphocytes themselves results in the induction of intracellular events. These include reorganization of the endothelial actin cytoskeleton to form stress fibers (11)
, tyrosine phosphorylation of cortactin, focal adhesion kinase (FAK), paxillin, and p130Cas (Cas) (12
, 13)
, activation of c-Jun amino-terminal kinase (13)
, the small GTP binding protein Rho (11)
, protein kinase C, and phospholipase C
(14)
. Pretreatment of EC with inhibitors of Rho proteins (15
, 16)
has been shown to inhibit ICAM-1-mediated signaling (13)
and T lymphocyte migration both in vitro (11)
and in vivo (16)
. These results provide direct evidence of lymphocyte-to-endothelial signaling and that CNS endothelial cells are actively engaged in supporting T lymphocyte migration across the BBB by mechanisms that require an intact actin cytoskeleton and functional Rho proteins. Despite the apparent importance of ICAM-1 in supporting lymphocyte migration and in transducing signals to the endothelium, it is highly likely that other costimulatory signals are required to provide the specificity and differential migratory behavior of subsets of leukocytes that have been observed across different vascular beds.
We have investigated whether heterotrimeric G-protein signaling pathways in brain microvascular EC are involved in the support of lymphocyte diapedesis. Brain EC monolayers were pretreated with pertussis toxin (PTX) leading to ribosylation and inactivation of G-protein subunits, which resulted in the inhibition of lymphocyte transmonolayer migration. In this paper, we describe for the first time that heterotrimeric G-protein-mediated signaling in CNS endothelial cells is essential for efficient transendothelial migration of T lymphocytes into the CNS.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, Gi2, and Gi3 and the common carboxyl-terminal of Gi1 and Gi2 subunits were obtained from G. Milligan (University of Glasgow). Polyclonal antibodies against Go/Gi3, Gi1, Gi1/Gi2, and Gi3 were obtained from Calbiochem (Notts, UK). Pertussis toxin was from Sigma (Poole, UK) and catalytically inactive pertussis toxin (PTXL129E) was obtained from Dr. R. Rappuoli (Sienna, Italy). Na51Cr2O7 was purchased from Amersham International (Amersham, UK) and [32P-adenylate] NAD+ was obtained from New England Nuclear (Herts, UK). Unless stated otherwise, all other reagents were purchased from Sigma. RSV puro expression vector was created after ligation of the BglII-HindIII fragment of pRC/RSV containing the RSV promoter sequence into BglII-HindIII digested pcDNA3/CMV. The puromycin resistance gene was excised from GB pac using SmaI and ClaI and ligated into SmaI/ClaI digested pcDNA3/RSV.
Endothelial cells
Primary microvascular brain endothelial cell cultures
Primary cultures of rat brain EC were grown from Lewis rats as described (17
, 18)
. Cells grown according to this protocol have been described extensively elsewhere (3
4
5
6
7
, 17
18
19)
. Rat cerebral cortex was dispersed by enzymatic digestion and microvessel fragments were separated from other material and single cells by density-dependent centrifugation. The microvessel fragments were washed and plated onto collagen-coated plastic culture plates. Cultures formed contact inhibited monolayers of fusiform cells that were maintained in Ham’s F-10 medium supplemented with 17.5% plasma-derived serum (First Link Ltd., West Midlands, UK), 75 µg/ml EC growth supplement (First Link Ltd., West Midlands, UK), 80 µg/ml heparin, 2 mM glutamine, 5 µg/ml vitamin C, 100 U/ml penicillin, and 100 µg/ml streptomycin at 37°C in 5% CO2; medium was replaced every 3 days until confluence was reached.
Immortalized brain endothelial cell lines
Two well-established Lewis rat brain EC lines were used. The SV40 large T immortalized rat brain EC line GP8/3.9 (19)
was maintained in Ham’s F-10 medium supplemented with 10% fetal calf serum (FCS), 2 mM glutamine, 100 U/ml penicillin, and 100 µg/ml streptomycin. The E1A immortalized rat brain EC line RBE4 (20)
was maintained in Ham’s F-10/MEM (1:1), 10% FCS, 300 µg/ml G418, 2 mM glutamine, 100 U/ml penicillin, and 100 µg/ml streptomycin. GP8/3.9 and RBE4 cultures were maintained at 37°C in 5% CO2 and media was replaced every 3 days until the cells reached confluence.
Aortic and high endothelial vein endothelial cells
To compare CNS-derived with non-CNS-derived endothelia, rat aortic and high endothelial vein (HEV) EC were cultured using similar protocols. Rat aortic endothelium were isolated by the method described by McGuire and Orkin (21)
. Rat aorta was removed by dissection, cut into small pieces (2–5 mm), placed luminal side down onto collagen-coated 24-well plates, and cultured in RPMI supplemented with 20% FCS, 7.5 µg/ml EC growth supplement, 80 µg/ml heparin, 2 mM glutamine, 100 U/ml penicillin, and 100 µg/ml streptomycin. After 3 days, the explants were removed and outgrowing cells of contact-inhibited cobblestone morphology were expanded and passaged by trypsinization and reseeding. Cells were used between passage 3 and 12. Rat high endothelial venule EC (22)
were generously supplied by A. Ager (NIMR, London, UK). HEV cells were grown in the same media as described above for aortic endothelia and used between passage 6 and 12.
Lymphocytes
Interleukin 2 (IL-2) -dependent myelin basic protein (MBP)-specific CD4+ T cell lines were used in the migration assays. These cells are extremely migratory in vitro and model activated circulating memory T cells. The MBP T cell lines were established from guinea pig MBP-primed Lewis rat lymph nodes (23)
. T cells were maintained in culture by alternating antigen stimulation (2 days) and propagation phases (4 days) in the presence of IL-2. These cells have been characterized as MHC-class II restricted CD4+ T cells that are encephalitogenic in vivo (24
, 25)
. T cells were used postantigen activation and after 2 to 4 days stimulation with IL-2.
Adhesion assays were performed using cells harvested from Lewis rat peripheral lymph nodes (PLN). PLN-derived lymphocytes were isolated from macrophages by panning on plastic tissue culture Petri dishes for 1 h and mitogen-activated by treating with type V concanavalin A (5 µg/ml) for 18–24 h before use. The isolated lymphocytes are adhesive when activated with mitogen but exhibit negligible migration over the duration of the adhesion assay (4
, 10)
.
Treatment of endothelial cell monolayers
EC monolayers were treated with either PTX (100 ng/ml) or catalytically inactive pertussis toxin (PTXL129E, 100 ng/ml) for 2 h at 37°C in 5% C02. The cells were washed thoroughly in Hanks buffered salt solution (HBSS; six rinses); fresh media was added and the cells were incubated for another 2 h. The monolayers were then washed thoroughly again (six washes) before the addition of lymphocytes for the adhesion and migration assays.
To establish whether residual EC-bound PTX was back-contaminating and affecting T cell function during the migration assay, mutant PTXL129E (1 µg/ml) was added to the 2 h incubation that followed the 2 h PTX treatment so as to compete off any residual surface-bound PTX. The adhesion and migration of lymphocytes was then assessed as described below. The ability of mutant PTXL129E to competitively inhibit the effect of PTX was evaluated by adding increasing concentrations of PTXL129E (1–100 µg/ml) to the 2 h PTX (100 ng/ml) incubation. Cells were washed and reincubated in fresh medium as described above before adhesion and migration assays.
Lymphocyte adhesion assay
Adhesion assays were carried out as described previously (3
, 26)
. Mitogen-activated PLN cells were labeled with 3 µCi 51Cr (sodium chromate) per 106 cells in HBSS for 90 min at 37°C. After washing, cells were resuspended in RPMI 1640 medium containing 10% FCS. EC monolayers were grown on 96-well plates; 200 µl of 51Cr-labeled PLN cells (1x106/ml) was added to each well, including blank wells, and incubated at 37°C for 90 min. After incubation, nonadherent cells were removed by washing with 37°C HBSS as described (4)
. Adherent PLN cells were lysed with 2% SDS, the lysate was removed, and emissions were quantitated by spectrometry. For each assay, emissions were determined from a minimum of six replicate blank wells to provide a value for the total amount of radioactivity added per well and to allow calculation of the specific activity of the cells. Results were obtained from a minimum of 6–12 separate wells in a minimum of three independent assays. At the end of the 90 min adhesion assay, PLN cell adhesion to the EC monolayer was 18.6 ± 1.0% for primary cultured brain EC, 17.4 ± 1.1% for the GP8/3.9 brain EC line, 18.4 ± 2.4% for the RBE4 brain EC line, 17.6 ± 0.7% for the HEV EC, and 29.7 ± 1.8% for the aortic EC. The results are expressed as a percentage of control values (means±SE).
Lymphocyte migration assay
Quantitative migration assays were carried out by phase-contrast videomicroscopy as previously reported (6
, 10
, 27)
. Endothelial monolayers derived from rat microvascular CNS endothelia, aortic or HEV tissue were grown to confluence in 96-well tissue culture plates. Antigen-activated lymphocytes (2x104 cells/well) were added and cultured with untreated or pretreated EC for 4 h. At the 4 h time point, cocultures were recorded for 10 min using time-lapse videomicroscopy and replayed at 160x real time to evaluate the number of migrated lymphocytes. These could be readily distinguished from surface lymphocytes by their attenuated and phase dark appearance (10)
(see http://www.ucl.ac.uk/ioo/video/sequence3.avi). At the end of the 4 h migration assay, T cell migration through the EC monolayer was 38.5 ± 1.5% for primary cultured brain EC, 37.2 ± 0.9% for the GP8/3.9 brain EC line, 33.0 ± 0.9% for the RBE4 brain EC line, 40.8 ± 2.8% for the HEV EC, and 23.9 ± 1.1% for the aortic EC. The number of migrated cells after different treatments was expressed as a percentage of control migration through untreated monolayers.
Generation of ribosylation-resistant GP8/3.9 endothelial cell lines
cDNA constructs encoding ribosylation-resistant G subunits, in which the acceptor cysteine residue for ADP-ribosylation by PTX was mutated to isoleucine, were constructed according to the method described by Bahia et al. (28)
. The cDNA constructs Go (C351 224 I), Gi2 (C352 224 I), and Gi3 (C351 224 I) were subcloned into the pcDNA3 expression vector. GP8/3.9 cells, already neomycin resistant, were cotransfected with the above cDNA constructs encoding ribosylation-resistant G-protein subunits and the rsvPURO vector at a molar ratio of 10:1 using the FuGENE6TM transfection reagent (Boehringer Mannheim, Mannheim, Germany) according to the manufacturer’s instructions. Control cells were transfected with the rsvPURO vector alone. Positive clones were selected by growing cells in normal GP8/3.9 growth medium supplemented with 20 µg/ml puromycin in 100 mm Petri dishes. Emerging colonies were detached using cloning rings and replated for expansion.
Analysis of ICAM expression on GP8/3.9 endothelial cell lines
To evaluate whether manipulation of the EC lines resulted in alteration of ICAM expression, surface expression of ICAM-1 and ICAM-2 was evaluated using flow cytometry and enzyme-linked immunosorbent assay (ELISA). Endothelial cells were seeded at confluent density onto either 6-well plates (for flow cytometry) or 96-well plates (for ELISA) and cultured for 3 days.
For flow cytometry, cells were detached with nonenzymatic cell dissociation solution (Sigma), washed, and either anti-ICAM-1 mAb (10 µg/ml) or anti-ICAM-2 pAb (10 µg/ml) was added to the cell pellet in 100 µl total volume over ice for 30 min. Cells were washed and fluorophore-conjugated secondary antibodies were added for another 30 min. After washing, cells were resuspended in 200–500 µl of phosphate-buffered saline (PBS) for flow cytometric analysis on a Becton Dickinson FACScan (Oxford, UK). The data was analyzed with CellQuestTM software.
For ELISA, cells were washed four times in ice-cold HBSS and fixed with 0.1% glutaraldehyde in PBS for 10 min at room temperature. Aldehydes were subsequently quenched with 50 mM Tris-HCl, pH 7.5 for 20 min at room temperature. Primary antibodies directed against ICAM-1 and ICAM-2 were diluted in 100 µl HBSS containing 100 µg/ml normal rabbit IgG and 4 mg/ml bovine serum albumin (BSA) and incubated with cells for 45 min at 37°C. Cells were washed four times with PBS containing 0.2% Tween-20 and incubated with biotinylated-anti-mouse-IgG (1:700; Amersham International) for 45 min at 37°C. Cells were again washed four times with PBS containing 0.2% Tween-20 and incubated with streptavidin-HRP (1:700; Amersham International) for 45 min at 37°C. Cells were washed four times in PBSA containing 0.2% Tween-20 before incubation with 100 µl tetramethylbenzidine (0.1 mg/ml) and 0.03% H2O2 in citrate-phosphate buffer (35 mM citric acid, 66 mM Na2HPO4, pH5) for 10 min. Reactions were stopped by the addition of 50 µl 1M sulfuric acid and product quantitated by optical density at 450 nm.
Analysis of heterotrimeric G-protein expression in brain endothelial cells
Control and transfected GP8/3.9 EC were washed three times with cold PBS and 5 ml of PBS containing 1 mg/ml phenylmethylsulfonyl fluoride (PMSF) and detached using cell scrapers. Cells were centrifuged for 5 min at 800 g. After resuspending in 5 ml of cold PBS containing PMSF, cells were disrupted by 30 strokes in a Dounce homogenizer and the crude cell homogenates were centrifuged at 1000 g for 30 min at 4°C. Supernatants were collected and centrifuged at 75,000 g at 4°C for 90 min in an ultracentrifuge to pellet a crude membrane fraction. Crude membranes were resuspended in 50 µl of Laemmli sample buffer containing 20 mM dithiothreitol and heated to 95°C for 5 min. Membrane samples containing equivalent amounts of protein (50 µg) were separated on 10% SDS-polyacrylamide (PAGE) gels and electrotransferred to nitrocellulose (Schleicher and Schuell, Dassel, Germany) at 10V for 30 min. Membranes were blocked with 5% casein in PBS for 1 h, followed by 0.2% Nonidet P-40 in PBS for 15 min. Blots were then washed extensively with PBS before overnight incubation with rabbit anti-sera raised against Go, Gi2 and Gi3 and the common carboxyl-terminal of Gi1, Gi2 subunits at a titer of 1:2000 in PBS containing 1% casein. Nitrocellulose filters were washed with PBS and treated with 0.2% Nonidet P-40 in PBS for 15 min. Blots were incubated with goat antisera specific for rabbit immunoglobulins for 2 h at a titer of 1:15,000 in PBS containing 1.0% casein. Blots were washed extensively with PBS before exposure to ECL chemiluminescence reagent (Pharmacia Amersham Biotech, UK) and autoradiographic film.
ADP-ribosylation of G-protein subunits
Crude membrane preparations were resuspended in 10 mM HEPES, pH8, 1 mM EDTA, 10 mM thymidine, 0.025% SDS, 0.2 mg/ml BSA, 5 mM DTT, 5 µCi [32P-adenylate]NAD+, and 1 µg activated PTX at 30°C for 1 h. Proteins were subsequently precipitated in 30% TCA and proteins were pelleted at 10,000 g for 10 min. Pellets were washed in 10% TCA recentrifuged at 10,000 g for 10 min and pellets were washed in ethanol. Pellets were air dried, resuspended in sample buffer, and proteins were resolved on 10% SDS-PAGE. Gels were dried and exposed to autoradiographic film. Activated PTX was prepared after reduction of PTX (whole molecule) for 30 min in 50 mM HEPES, pH8, 1 mg/ml BSA, and 20 mM DTT at 30°C.
Immunoprecipitation of cortactin and FAK and Western blot analysis of phosphotyrosine
GP8/3.9 brain EC were seeded at a density of 104 cells/cm2 and, after 3–4 days in culture, were incubated with serum- and bFGF-free culture medium containing 100 U/ml IFN- (to up-regulate ICAM-1 expression) for 48 h. Cells were washed in PBS before cross-linking of ICAM-1 with a specific monoclonal antibody (1A29) for 10 min and subsequent addition of RAM antibody for 30 min (13)
. Cells were then washed with ice-cold PBS containing 1 mM orthovanadate and lysed for 30 min at 4°C in Nonidet P-40 buffer (10 mM Tris-HCl, pH 7.5, 140 mM NaCl, 1 mM orthovanadate, 1% Nonidet P-40, 2 mM PMSF, 5 mM EDTA, 10 µg/ml leupeptin, 1 µg/ml pepstatin, and 10 µg/ml aprotinin). Nuclei were discarded after centrifugation at 10,000 g for 10 min. Lysates were incubated overnight at 4°C with specific antibodies against cortactin and FAK and immune complexes were harvested after 2 h incubation with protein A-agarose. Immunoprecipitates were collected by centrifugation and washed extensively in Nonidet P-40 buffer. Immunoprecipitated proteins were eluted with SDS sample buffer and resolved by 10% SDS-PAGE, followed by detection by immunoblotting with the antiphosphotyrosine antibody 4G10 (12
, 13)
.
Statistical analysis
Statistical analysis was carried out using a Student’s t test. Mean differences between groups were considered significant at P < 0.05.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
B). In contrast, PTX pretreatment of brain EC resulted in a highly significant reduction of lymphocyte migration through the EC monolayer (17.3±2.4% of control migration, mean±SE; P<0.0001). Contrary to these findings, treatment of EC in an identical manner with a mutant of PTX (PTXL129E), in which the catalytic activity had been destroyed through a leucine to glutamine substitution at codon 129, had no effect on lymphocyte migration (Fig. 1A
). To compete off residual EC cell surface-bound PTX that may cause subsequent G subunit ribosylation in adherent lymphocytes, endothelial monolayers were treated during the 2 h wash period with 1 µg/ml PTXL129E before T cell adhesion and migration assays. Lymphocyte migration in the presence of excess PTXL129E during the 2 h wash was 18.4 ± 5.3% (mean±SE) of control values and not significantly different from the inhibition of lymphocyte migration induced by PTX treatment alone (Fig. 1A
). To determine whether PTXL129E was able to reverse the inhibitory effect of PTX on lymphocyte migration, increasing concentrations of PTXL129E were added to the 100 ng/ml PTX during the initial 2 h incubation. No significant antagonistic effect of PTXL129E was observed at 1 or 10 µg/ml, but at 100 µg/ml there was a significant reversal in the effects of PTX (data not shown). Complete reversal could not be achieved, as higher concentrations of PTXL129E were found to be cytotoxic to EC. Treatment of the two brain EC lines (GP8/3.9 and RBE4) with PTX led to identical inhibition of lymphocyte migration through GP8/3.9 and RBE4 cell monolayers (14.4±1.2% and 15.5±1.0% of control migration, respectively, P<0.0001), with no effect on lymphocyte adhesion to either cell line (Fig. 1C, D
).
|
Pertussis toxin treatment of aortic and HEV endothelial cells partially inhibits lymphocyte transendothelial monolayer migration
Using an identical PTX treatment protocol as described above for brain EC, it was found that the level of migration through monolayers of aortic and HEV EC was significantly reduced to 65.0 ± 7.7% (P<0.005) and 67.1 ± 2.7% (P<0.0001) of control values, respectively (mean±SE) (Fig. 2
A). The degree of inhibition of lymphocyte migration through aortic and HEV EC monolayers, however, did not approach that achieved after PTX treatment of brain EC, where migration was significantly lower (17.3±2.4%: P<0.0001 vs. PTX-treated aortic and HEV EC). PTX treatment had no effect on the capacity of lymphocytes to adhere to the endothelial monolayers (Fig. 2B
).
|
GP8/3.9 brain EC express G-protein subunits
Proteins from crude membrane preparations obtained from GP8/3.9 brain capillary EC were resolved on 10% SDS-PAGE and analyzed by Western blotting analysis. Immunoreactive proteins were identified using polyclonal antibodies (Fig. 3
A). Immunoblotting with anti-Gi1 showed that this protein was not expressed in GP8/3.9 EC, consistent with a previous report describing the lack of this subunit in rat brain EC (29)
. Two bands were observed when immunoblotting with the specific antiserum for Go. This observation suggests that the GP8/3.9 cell line may express two of the four isoforms of this subunit reported to be present in brain tissue (30
31
32)
. Likewise, two bands were frequently observed after staining with antiserum against the carboxyl-terminal of Gi1/Gi2, suggesting two isoforms of Gi2.
|
Pertussis toxin treatment of GP8/3.9 brain EC results in ADP-ribosylation of G-protein subunits
Treatment of GP8/3.9 cell lysates with activated PTX and [32P-adenylate]NAD+ showed [32P]ADP-ribosylation of G-protein subunits. However, when GP8/3.9 EC were pretreated in culture with 100 ng/ml PTX for 2 h and lysates subsequently exposed to activated PTX and [32P-adenylate]NAD+, there was no [32P]ADP-ribosyl labeling. These data demonstrate that treatment of cells in culture with 100 ng/ml PTX is capable of inhibiting all G-protein subunit substrates within EC (Fig. 3C
) as no further ADP-ribosylation by PTX is possible.
ICAM-1-mediated tyrosine phosphorylation of cortactin and FAK are insensitive to PTX
As described previously (13)
, ICAM-1 cross-linking resulted in enhanced tyrosine phosphorylation of cortactin and FAK (Fig. 4
A). Treatment of cells with AlF4-, which activates heterotrimeric G-proteins, did not mimic the effects of ICAM-1 cross-linking in inducing cortactin phosphorylation but was able to result in tyrosine phosphorylation of FAK (Fig. 4B
), demonstrating that FAK phosphorylation can be induced through stimulation of heterotrimeric G-proteins. The effects of AlF4- in enhancing the tyrosine phosphorylation of FAK were partially inhibited after prior exposure of cells to 100 ng/ml PTX (data not shown). However, prior exposure of cells to PTX was unable to inhibit the tyrosine phosphorylation of either cortactin or FAK after ICAM-1 cross-linking (Fig. 4B
), suggesting that PTX treatment does not alter signaling pathways initiated through ICAM-1.
|
Expression of ribosylation-resistant G subunits by GP8/3.9 brain EC
GP8/3.9 cells were cotransfected with rsvPURO and cDNA constructs encoding ribosylation-resistant (PTX-insensitive) G-protein subunits. After puromycin selection, cells lines expressing PTX-insensitive forms of Go, Gi2, and Gi3 were expanded and designated as GoC351I, Gi2C352I, and Gi3C351I, respectively. All cell lines grew as contact inhibited monolayers of cobblestone morphology similar to the parent line. The overexpression of ribosylation-resistant G-protein subunits in each cell line was determined by preparing crude membrane fractions as described above for wild-type cells. In this case, however, 10 µg of crude membranes was resolved by SDS-PAGE and transferred to nitrocellulose filters, as this concentration fell below the detection limit of the available antisera for identifying endogenous G-protein subunits. After Western blotting with subunit-specific antibodies, all three cell lines were found to be positive for the expression of specific ribosylation-resistant proteins (Fig. 3B
).
Ribosylation-resistant mutant brain EC lines are insensitive to PTX-induced inhibition of lymphocyte migration
PTX treatment of the rsvPuro control EC significantly inhibited lymphocyte migration to 21.1 ± 3.0% of control values compared with untreated cells (P<0.0001) and was similar to the level of inhibition achieved with untransfected GP8/3.9 cells (14.4±1.2% of controls, P<0.0001) (Fig. 5
A). With the GoC351I, Gi2C352I, and Gi3C351I cell lines, the degree of lymphocyte migration was significantly greater than rsvPuro-transfected controls being 144.0 ± 7.1% (P<0.0001), 122.7 ± 8.4% (P<0.05), and 126.5 ± 4.0% (P<0.0001) of control values, respectively (Fig. 5A
). When treated with PTX, there was a small decrease in migration to 79.0% (GoC351I, P<0.005), 89.8% (Gi2C352I, not significant), and 74.1% (Gi3C351I, P<0.005) of migration through untreated monolayers of corresponding mutant cell lines. The degree of inhibition did not approach the level of inhibition achieved after PTX treatment of untransfected or rsvPuro-transfected GP8/3.9 cells, demonstrating a significant level of rescue with GoC351I (P<0.001), Gi2C352I (P<0.001), and Gi3C351I (P<0.001) cell lines (Fig. 5A
).
|
In the cell lines expressing Go, Gi2, and Gi3 mutants, there was a significant increase in the level of lymphocyte adhesion of 157.4 ± 4.4% (P<0.0001), 189.1 ± 5.6% (P<0.0001), and 146.6 ± 7.5% (P<0.0001) of wild-type control adhesion, respectively (Fig. 5B
). There was no inhibitory effect of PTX treatment of the mutant cell lines on lymphocyte adhesion except with the Gi2 mutant, in which there was a small increase (P<0.05) (Fig. 5B
).
Ribosylation-resistant mutant brain EC lines have increased expression of ICAM that is not affected by PTX
GP8/3.9 brain EC constitutively expressed ICAM-1 (100%) and to a lesser extent ICAM-2 (12%) as assessed by flow cytometry (Fig. 6
A, B, respectively). Treatment of these cells with PTX for 2 h, followed by 2 h in normal medium did not alter the percentage of ICAM-1 or ICAM-2 expression. Using ELISA, the relative levels of ICAM-1 and ICAM-2 expressed on GP8/3.9 EC and the ribosylation-resistant G-protein EC lines were evaluated in the absence and presence of PTX. The overexpression of ribosylation-resistant G-proteins led to a significant increase in ICAM-1 expression in the rsvPuro control line (P<0.0001), GoC351I (P<0.05), and Gi2C352I (P<0.001) cell lines but not the Gi3C351I cell line compared with control GP8/3.9 cells. Similarly, ICAM-2 expression was up-regulated in rsvPuro and ribosylation-resistant G-proteins cell lines (P<0.0001 vs. GP8/3.9 cells) (Fig. 6C, D
). In all cases, no decrease in ICAM-1 or ICAM-2 expression was observed after PTX treatment.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
subunits in GP8/3.9 cell monolayers, as we demonstrated that further PTX treatment of cell lysates from treated monolayers was unable to further [32P]-label G-protein subunits. This strongly suggests that endothelial G-protein subunits are directly involved in the EC-mediated facilitation of lymphocyte migration through the BBB.
The most compelling evidence that EC heterotrimeric G-proteins are central to the process of leukocyte migration through the BBB and that the effect is not due to back contamination of added T cells is provided by the data obtained with the ribosylation-resistant cell lines. The ability of GoC351I, Gi2C352I, and Gi3C351I to reverse the effect of PTX on T lymphocyte migration through CNS EC suggests that each of these subunits is capable of interacting with specific effector molecules. Alternatively, it may be that these proteins possess sufficient homology to interact with specific effectors when overexpressed. The unexpected finding that the mutant EC lines supported a significantly greater level of T lymphocyte migration is likely to be accounted for by overexpression of the G-protein subunits. The observation that lymphocyte adhesion is significantly increased in cells expressing the GoC351I and Gi2C352I mutants, however, is more difficult to interpret, although this may be explained in part by the increased expression of both ICAM-1 and ICAM-2 in these cell lines. Although the mutant cell lines were able to significantly rescue the effect of PTX, there was still a significant reduction in migration after PTX treatment, which is likely to be due to inhibition of the endogenous heterotrimeric G-proteins.
An alternative explanation for the observed inhibition of migration is that the carbohydrate binding subunits of PTX, which have homology to selectin molecules, may interfere with lymphocyte binding (33)
and hence migration. This has been discounted because EC pretreated with a mutant form of PTX (L129E) containing a Leu to Glu substitution at codon 129 that results in a protein able to bind to and enter EC but has no catalytic activity (34)
had no discernible effect on lymphocyte adhesion and migration through these cell monolayers.
We are confident that this effect is not due to PTX from the treated EC back-contaminating the cocultured lymphocytes and inhibiting their ability to migrate. This is clearly not the case as the inhibitory effects of PTX on lymphocyte migration through aortic and HEV endothelial monolayers were significantly less than that observed with brain endothelium. This suggests that the mechanism is functionally more important in CNS endothelial cells and may be an important factor in enabling efficient leukocyte migration into the CNS. Indeed, this is consistent with a recent report showing that PTX treatment of HEV EC does not inhibit lymphocyte migration (35)
.
The Gi protein-coupled receptor through which the signaling is putatively induced has not been identified, but intriguing possibilities include chemokine receptors known to be expressed on brain EC (36)
. Such receptors may be necessary to trigger EC costimulatory pathways that are independent of those induced by ICAM-1. The signaling role of EC G-protein subunits in facilitating leukocyte migration across the BBB is currently unclear although recent evidence suggests that PTX-sensitive G-protein subunits may be important in the regulation and biogenesis of tight junctions. It has emerged from studies in which MDCK cells treated with AlF4- or overexpressing a constitutively activated form of Gi2 displayed increases in transepithelial resistance and an increase in the rate of tight junction development (37)
. The Go protein has been implicated in tight junction biogenesis and the Gi2 and Go proteins are both found to localize to tight junction areas with Go showing significant colocalization and coimmunoprecipitation with ZO-1 (38)
.
This study has demonstrated for the first time that the presence of an EC PTX-sensitive pathway is central to the regulation of lymphocyte migration through specialized vascular endothelium of the BBB. Future identification of this G-protein-coupled receptor could lead to the development of new pharmacological strategies for inhibiting the abhorrent lymphocyte migration into the central nervous system that occurs in diseases such as multiple sclerosis.
|
ACKNOWLEDGMENTS |
|---|
Received for publication February 12, 2002.
Revision received April 11, 2002.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, tumour necrosis factor and interleukin-1. Immunology 64,677-681[Medline]
determines the extent of activation by the 2A adrenoreceptor. Biochemistry 37,11555-11562[CrossRef][Medline]
in the maintenance and biogenesis of epithelial cell tight junctions. J. Biol. Chem. 273,21629-21633 subunit in tight junction biogenesis. J. Biol. Chem. 271,25750-25753This article has been cited by other articles:
|
|
 |
|
  R. Martinelli, M. Gegg, R. Longbottom, P. Adamson, P. Turowski, and J. Greenwood ICAM-1-mediated Endothelial Nitric Oxide Synthase Activation via Calcium and AMP-activated Protein Kinase Is Required for Transendothelial Lymphocyte Migration Mol. Biol. Cell, February 1, 2009; 20(3): 995 - 1005. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. H. Ramirez, D. Heilman, B. Morsey, R. Potula, J. Haorah, and Y. Persidsky Activation of Peroxisome Proliferator-Activated Receptor {gamma} (PPAR{gamma}) Suppresses Rho GTPases in Human Brain Microvascular Endothelial Cells and Inhibits Adhesion and Transendothelial Migration of HIV-1 Infected Monocytes J. Immunol., February 1, 2008; 180(3): 1854 - 1865. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. Amend, H. Doster, C. Lange, E. Dubois, H. Kalbacher, A. Melms, and F. Bischof Induction of Autoimmunity by Expansion of Autoreactive CD4+CD62Llow Cells In Vivo J. Immunol., October 1, 2006; 177(7): 4384 - 4390. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. T. Hawkins and T. P. Davis The Blood-Brain Barrier/Neurovascular Unit in Health and Disease Pharmacol. Rev., June 1, 2005; 57(2): 173 - 185. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Bonig, G. V. Priestley, L. M. Nilsson, Y. Jiang, and T. Papayannopoulou PTX-sensitive signals in bone marrow homing of fetal and adult hematopoietic progenitor cells Blood, October 15, 2004; 104(8): 2299 - 2306. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
