If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Glycogen Synthase Kinase 3β Inhibition Prevents Monocyte Migration across Brain Endothelial Cells via Rac1-GTPase Suppression and Down-Regulation of Active Integrin Conformation
Address reprint requests to Yuri Persidsky, M.D., Ph.D., Department of Pathology and Laboratory Medicine, Temple University School of Medicine, 3401 N Broad St, Philadelphia, PA 19140
Glycogen synthase kinase (GSK) 3β has been identified as a regulator of immune responses. We demonstrated previously that GSK3β inhibition in human brain microvascular endothelial cells (BMVECs) reduced monocyte adhesion/migration across BMVEC monolayers. Herein, we tested the idea that GSK3β inhibition in monocytes can diminish their ability to engage the brain endothelium and migrate across the blood-brain barrier. Pretreatment of primary monocytes with GSK3β inhibitors resulted in a decrease in adhesion (60%) and migration (85%), with similar results in U937 monocytic cells. Monocyte-BMVEC interactions resulted in diminished barrier integrity that was reversed by GSK3β suppression in monocytic cells. Because integrins mediate monocyte rolling/adhesion, we detected the active conformational form of very late antigen 4 after stimulation with a peptide mimicking monocyte engagement by vascular cell adhesion molecule-1. Peptide stimulation resulted in a 14- to 20-fold up-regulation of the active form of integrin in monocytes that was suppressed by GSK3β inhibitors (40% to 60%). Because small GTPases, such as Rac1, control leukocyte movement, we measured active Rac1 after monocyte activation with relevant stimuli. Stimulation enhanced the level of active Rac1 that was diminished by GSK3β inhibitors. Monocytes treated with GSK3β inhibitors showed increased levels of inhibitory sites of the actin-binding protein, cofilin, and vasodilator-stimulated phosphoprotein-regulating conformational changes of integrins. These results indicate that GSK3β inhibition in monocytes affects active integrin expression, cytoskeleton rearrangement, and adhesion via suppression of Rac1-diminishing inflammatory leukocyte responses.
Leukocyte interactions with the endothelium are at the center of inflammatory responses. Such interactions are highly orchestrated, leading to the sequential steps of tethering, rolling, arrest, firm adhesion, and migration across the endothelium.
Integrins [lymphocyte function-associated antigen 1 or αLβ2 integrin, macrophage receptor 1, known as integrin αMβ2, and very late antigen 4 (VLA-4), also known as integrin α4β1] expressed on the surface of leukocytes are primarily responsible for arrest and firm adhesion, and they mediate leukocyte tethering and rolling.
The level of expression and, more recently, multifaceted conformational changes of integrins have emerged as key factors in the pro-inflammatory phenotype of leukocytes transitioning from the resting to the activated state, with enhanced adhesion potential.
Integrins can be quickly transformed to the active state by so-called inside-out signaling after stimulation of G-coupled receptors by chemokines or other inflammatory factors via intracellular signaling, leading to binding of talin to intracellular integrin domains.
Such alterations result in conformational unbending and increased affinity of integrins. The development of antibodies to detect expression of active integrin conformation has allowed a better understanding of the interactions of leukocytes with the endothelium.
One of these antibodies, HUTS21, detects the active conformational state of VLA-4 integrin that interacts with vascular cell adhesion molecule (VCAM)-1. Cross-linking with antibodies to integrin (eg, integrin α4, also known as CD49d) or stimulation with l-leucyl-l-aspartyl-l-valyl-l-prolyl-l-alanyl-l-alanyl-l-lysine (LDV) peptide mimics leukocyte endothelial interaction without actual endothelial cell engagement. LDV peptide is a peptide consensus sequence of VCAM-1 and fibronectin, shown by Chigaev et al
to activate integrin β1 conformational change. We used LDV peptides as a specific inducer for the active VLA-4 conformation and subsequently used the conformation-specific HUTS21 antibody to detect the change in VLA-4, to analyze the effects of glycogen synthase kinase (GSK) 3β inhibition on the processes regulating monocyte adhesion and migration.
GSK3β is a serine-threonine protein kinase that is constitutively active in cells, and numerous factors exert their effects by inhibiting GSK3β activity.
GSK3β activity is inhibited by phosphorylation of a specific serine residue (S9) located in the GSK3β N-terminal domain. Downstream targets of GSK3β include transcription factors (β-catenin, cAMP response element binding protein, and c-Jun), proteins bound to microtubules (τ), cell cycle mediators, and regulators of metabolism.
The anti-inflammatory effects of GSK3β inhibition have been recognized and include a decrease in cytokine-chemokine secretion and expression of adhesion molecules.
Yet, limited information exists about direct anti-inflammatory effects of GSK3β suppression in leukocytes.
Herein, we demonstrate, for the first time to our knowledge, that GSK3β inhibition in monocytes diminishes the expression of active VLA-4 and attenuates monocyte adhesion to and migration across brain microvascular endothelial cell (BMVEC) monolayers. We also show that GSK3β inhibition suppresses the activity of the small GTPase, Rac1; diminishes lamellipodia formation; and increases expression of phosphorylated forms of cofilin (an actin-binding protein that is involved in leukocyte migration) and vasodilator-stimulated phosphoprotein (VASP; controlling up-regulation of active VLA-4). These previously unrecognized effects of GSK3β in monocytes, coupled with our observations in BMVECs,
indicate that GSK3β inhibition could be an attractive target for treatment of inflammation and blood-brain barrier (BBB) injury.
Materials and Methods
Reagents and Cells
Recombinant human tumor necrosis factor (TNF)-α and monocyte chemoattractant protein 1/chemokine ligand (CCL)2 were obtained from R&D Systems (Minneapolis, MN). GSK3β inhibitors were obtained as follows: lithium chloride (LiCl; Sigma/Aldrich Co, Ltd, St Louis, MO), 5-iodo-indirubin-3′-monoxime (I3′M; EMD Chemicals, San Diego, CA), N-(4-methoxybenzyl)-N′-(5-nitro-1,3-thiazol-2-yl) urea [AR-A014418 (AR); Sigma/Aldrich Co, Ltd], and 3-(2,4-dichlorophenyl)-4-(1-methyl-1H-indol-3-yl)-1H-pyrrole-2,5-dione [SB 216763 (SB2); Cayman Chemical Co; Ann Arbor, MI]. Rac1 GTPase-specific inhibitor, NSC23766 (NSC), was obtained from EMD Chemicals. Concentrations of inhibitors were based on previously published work
and our initial dose-response experiments with different exposure times. Lysophosphatidic acid (LPA) and lipopolysaccharide (LPS) were obtained from Sigma/Aldrich Co, Ltd, and Cayman Chemical Co, respectively.
Primary BMVECs were supplied by Dr. Michael Bernas and Dr. Marlys Witte (University of Arizona, Tucson). BMVECs were isolated from the vessels from the brain resection path of patients (showing no abnormalities) undergoing surgery for the treatment of intractable epilepsy, as previously described.
BMVEC cultures were used until passage 5 and were expanded in Dulbecco's modified Eagle's medium/F-12 media supplemented with 10% heat-inactivated fetal bovine serum (FBS; Invitrogen/Life Technologies, Carlsbad, CA), endothelial cell growth supplement (BD Bioscience, Franklin Lakes, NJ), heparin (1 mg/mL; Sigma/Aldrich Co, Ltd), amphotericin B (2.5 μg/mL; Invitrogen/Life Technologies), penicillin (100 U/mL; Invitrogen/Life Technologies), and streptomycin (100 μg/mL; Invitrogen/Life Technologies). Before experimentation, BMVEC monolayers were placed in media containing the previously described supplements but lacking endothelial cell growth supplement and heparin. Under these conditions, the BMVEC cultures were routinely evaluated for expression of endothelial cell markers (von Willebrand factor and CD31/platelet endothelial cell adhesion molecule-1) and tight junction proteins (zonula occludens protein 1, occludin, and claudin-5) and the lack of macrophage (CD68 and CD163), astrocyte (glial fibrillary acidic protein), and pericyte (α-myosin) markers. BMVEC purity was usually approximately 97%.
Activation of peroxisome proliferator-activated receptor gamma (PPARgamma) suppresses Rho GTPases in human brain microvascular endothelial cells and inhibits adhesion and transendothelial migration of HIV-1 infected monocytes.
Primary human monocytes were supplied by the Human Immunology Core of the University of Pennsylvania, Philadelphia. Flow cytometry analysis was used to confirm that the cells were >90% CD14 positive.
The cells were isolated by countercurrent centrifugal elutriation and maintained in suspension culture in Dulbecco's modified Eagle's medium containing heat-inactivated 10% FBS, penicillin (100 U/mL), streptomycin (100 μg/mL), and l-glutamine (2 mmol/L)
The human myeloid leukemia cell line, U937 (obtained from ATCC, Manassas, VA), was grown in RPMI 1640 medium containing 10% heat-inactivated FBS supplemented with 100 U/mL penicillin, 100 μg/mL streptomycin, and 2 mmol/L l-glutamine.
Primary monocytes and U937 cells were maintained in 5-mL polystyrene round-bottom tubes in suspension (Becton Dickinson, Franklin Lakes, NJ) before use and detached by shaking the tube. Both primary monocytes and U937 cells were pretreated with inhibitors overnight before experiments. Treatment with inhibitors did not cause any toxic effects on viability of the cells, as determined using the Live/Dead Cell Viability Assay (Invitrogen/Life Technologies) (see Supplemental Figure S1B at http://ajp.amjpathol.org).
Adhesion and Migration Assays
Quantitative adhesion assays were performed as previously described.
Activation of peroxisome proliferator-activated receptor gamma (PPARgamma) suppresses Rho GTPases in human brain microvascular endothelial cells and inhibits adhesion and transendothelial migration of HIV-1 infected monocytes.
In brief, BMVECs were seeded on collagen type I–coated 96-well plates at a density of 2.5 × 104 cells per well. Confluent monolayers were then exposed to TNF-α (20 ng/mL) for 24 hours to activate the BMVECs. U937 cells or freshly isolated human monocytes at 5 × 106 cells/mL were pretreated with GSK3β inhibitors and then labeled with 5 μmol/L fluorescent tracer, calcein-AM (Invitrogen/Life Technologies). Pretreatment by GSK3β inhibitors did not have any effect on calcein-AM staining, as assessed by fluorescence-activated cell sorting (FACS; see Supplemental Figure S1A at http://ajp.amjpathol.org). All treatments were removed from the endothelial cells, and from monocytes, before adding monocytes at 2.5 × 105 cells per well. Monocytes were added to BMVECs, incubated for 15 minutes, and then rinsed three times with one times PBS to eliminate nonadherent monocytes. The fluorescence from adherent or migrated monocytes was measured using a Spectramax M5 fluorescence plate reader (Molecular Devices, Sunnyvale, CA). The number of migrated monocytes was determined from external standards of known numbers of labeled monocytes. The results for adhesion assays are presented as the mean ± SEM fold adhesion (number of adherent monocytes for each experimental condition/basal adhesion of the untreated control).
The transendothelial migration assay was performed as previously described.
FluoroBlok (BD Bioscience, Bedford, MA) cell culture inserts, designed to block the transmission of fluorescent light between 490 and 700 nm, were used to allow for continuous detection of fluorescently labeled monocytes migrating across endothelial monolayers. BMVECs were seeded on collagen type I–coated pore (3 μm thick) 24-well tissue culture FluoroBlok inserts at a density of 2.5 × 104 cells per insert. Confluent monolayers were then exposed to TNF-α (20 ng/mL) for 24 hours to activate the BMVECs. After activation, BMVECs were rinsed with fresh medium, and the medium was replaced. U937 or primary monocytes were pretreated with or without GSK3β inhibitors overnight and rinsed with fresh medium, and the medium was replaced, before calcein-AM labeling. For migration assays, calcein-AM–labeled monocytes were added to the upper chamber of the tissue culture insert system, whereas the chemoattractant, CCL2/monocyte chemoattractant protein 1 (50 ng/mL), was added to the lower chamber to stimulate migration. The cells were then incubated for 2 hours at 37°C in a tissue culture incubator. The fluorescence from migrated monocytes was measured using a Spectramax M5 fluorescence plate reader (Molecular Devices). The number of migrated monocytes was determined from external standards of known numbers of labeled monocytes. The results for migration assays are shown as the average fold migration ± SEM; the fold migration is derived from the number of migrated monocytes for each experimental condition/number of migrated monocytes in the untreated, no chemoattractant control.
TEER Data
To determine the integrity of brain endothelial monolayers after engagement with monocytes, transendothelial electrical resistance (TEER) was measured using the 1600R Electric Cell-substrate Impedance Sensing (ECIS) system (Applied Biophysics, Troy, NY). By using the free ions in the culture media, the instrument generates an a.c. flow between the electrode and counterelectrode located in specialized tissue culture arrays and measures the change in impedance. The ECIS system provides real-time complex impedance, providing readouts for impedance, resistance, and capacitance. In brief, BMVECs were plated on collagen type I–coated electrode arrays (96W10E; Applied Biophysics). The cells were then allowed to form monolayers, reaching stable TEER values. After achieving basal TEER readings because of monolayer and tight junction formation, the BMVECs were incubated overnight in the absence or presence of TNF-α (20 ng/mL). Medium was replaced on the BMVECs, and monocytes (untreated or pretreated with GSK3β inhibitors) were added at 5 × 104 cells per well (10:1 ratio of monocytes/BMVECs). All treatments were removed from the BMVECs, and monocytes were washed before adding them to the wells. Measurements were taken every 30 minutes at 4000 Hz. The average baseline TEER readings varied between 800 and 1500 Ω/cm2. The results are presented as the average ± SEM percentage change of baseline TEER from at least three independent experiments consisting of at least two replicates.
VLA-4 Ligation and Cross-Linking
For integrin engagement studies, U937 cells were suspended in RPMI 1640 medium/1% FBS/l-glutamine at 5 × 106 cells/mL. Surface CD49d antigen was ligated with monoclonal antibody (HP2/1; Beckman Coulter Inc., Brea, CA) at 4°C for 25 minutes at 15 μg/mL, washed twice in cold RPMI 1640 medium, and then cross-linked with 5 μg/mL goat anti-mouse F(ab')2 (Thermo Scientific, Rockford, IL) for 25 minutes at 4°C. This concentration of primary antibody was saturating by flow cytometry.
VLA-4 integrin cross-linking on human monocytic THP-1 cells induces tissue factor expression by a mechanism involving mitogen-activated protein kinase.
Cells were washed twice in cold RPMI 1640 medium, resuspended in RPMI 1640 medium/1% fetal calf serum/l-glutamine, and incubated at 37°C and 5% CO2 for times ranging from 0 minutes to 1 hour. Reactions were stopped by placing the cells on ice. In inhibition studies, U937 cells were pre-incubated in the presence of GSK3β inhibitors overnight.
Rac1 GTPase activity was measured in cell lysates prepared from unstimulated (basal) or stimulated primary monocytes or U937 cells (with or without pretreatment with GSK3β inhibitors) by cross-linking integrin CD49d with antibody or alternatively stimulated with LPS (100 ng/mL) or LPA (100 ng/mL), known activators of GTPases, which served as positive controls.
To measure Rac1 GTPase activity, a commercially available G-LISA Rac1 Activation Assay kit (Cytoskeleton Inc., Denver, CO) was used, according to manufacturer instructions.
Western Blot Analysis
After integrin cross-linking, U937 cells were lysed in ice-cold cell lysis buffer containing 1% Triton X-100, 150 mmol/L NaCl, 10 mmol/L Tris-HCl (pH 7.4), 2 mmol/L sodium orthovanadate, 10 μg/mL leupeptin, 50 mmol/L NaF, 5 mmol/L EDTA, 1 mmol/L EGTA, and 1 mmol/L phenylmethylsulfonyl fluoride. Supernatants were collected after centrifugation at 10,000 × g for 5 minutes at 4°C. The protein content in the cell lysates was determined using the Bradford protein assay (BioRad, Hercules, CA). Protein fractions containing 20 μg of protein were mixed with two times Laemmli (sample loading) buffer containing β-mercaptoethanol and then boiled for 5 minutes. The proteins were then resolved by SDS-PAGE (4% to 20% precast gels; Thermo Scientific), followed by electrophoretic transfer to nitrocellulose membranes. The following primary antibodies, diluted in one times PBS/0.1% Tween 20, were used to detect target proteins: anti-phosphorylated-cofilin and anti-phosphorylated-VASP (Cell Signaling Technology, Beverly, MS), anti-cofilin, anti-VASP, anti-Grb-2 protein, and loading control (Santa Cruz Biotechnology, Santa Cruz, CA). All antibodies were incubated with the membranes overnight at 4°C with gentle shaking. Bound primary antibodies were exposed to the corresponding species-specific fluorescent IRDye680 or IRDye800 infrared secondary antibodies (diluted 1:20,000; LI-COR, Lincoln, NE) for 30 minutes at room temperature, and detected using the LI-COR Odyssey imaging system.
Conformational Changes of VLA-4
VLA-4–activated conformation, detected by HUTS21 binding, was studied by FACS, as previously described.
Real time analysis of the affinity regulation of alpha 4-integrin: the physiologically activated receptor is intermediate in affinity between resting and Mn(2+) or antibody activation.
Conformational regulation of alpha 4 beta 1-integrin affinity by reducing agents: “inside-out” signaling is independent of and additive to reduction-regulated integrin activation.
indicated that binding of the VLA-4–specific ligand, LDV, to integrin induces a series of conformational changes resulting in the exposure of the HUTS21 epitope. In our experiments, we used LDV as a specific stimulator of active VLA-4 conformation. LDV was synthesized by Invitrogen/Life Technologies. FACS analysis was performed using mouse anti-human integrin β1 (CD29) clone HUTS21 (phosphatidylethanolamine), mouse anti-human CD29 clone MAR4 (activated protein C), phosphatidylethanolamine, and activated protein C isotype control (mouse IgG2a κ) clone G155 to G178. Cells (106 cells/mL) removed from ice were warmed in HEPES buffer containing 0.1% bovine serum albumin for 10 minutes at 37°C. First, samples were analyzed for 30 to 120 seconds to establish a baseline for unstained cells. Next, the tube was removed, HUTS21 antibody (Ab) was added (20 μL of Ab to 1 mL of cell suspension), and acquisition was re-established. To induce VLA-4 conformational change, 12 nmol/L LDV (saturating concentration
) was added and mixed. Acquisition was re-established, and data were acquired, collecting 10,000 events. Analysis was performed using an FACS BD Calibur flow cytometer (BD Biosciences, San Jose, CA) using Cell Quest software (BD Biosciences) or FlowJo software (Tree Star, Inc., Ashland, OR). Quantitation of integrin β1 conformational activation was performed, in which the mean fluorescence intensity (MFI) of activated nontreated cells was assigned a value of 100, and a value of 0 was assigned to the MFI of cell autofluorescence. All other calculations were performed using the regression curve calculation tool of Prism version 5 software (GraphPad Software Inc., La Jolla, CA).
Visualization of Lamellipodia
For immunofluorescence studies, cells were fixed with 4% formaldehyde in PBS for 15 minutes on ice. Cells were permeabilized with 0.1% Triton X-100 in PBS for 10 minutes. For localization of F-actin filaments, cells were incubated with 14 μmol/L Acti-stain 488 fluorescent phalloidin (Cytoskeleton Inc.) for 45 minutes at room temperature and DAPI (300 nmol/L) for 20 minutes at room temperature. Images of 20 cells per treatment were taken at ×63 (1024 × 1024 pixel area) using a Leica Sp5 confocal laser-scanning microscope (Leica Microsystems GmbH, Mannheim, Germany) using Leica LAS AF software, and were processed with Adobe Photoshop CS3 software (Adobe Systems Inc., San Jose, CA). The area (in pixels) of each lamellipodium per cell was measured using ImageJ software (NIH, Bethesda, MD), and the average lamellipodium area per cell in each captured image was used for statistical analysis.
Human Brain Tissue
To investigate expression of total and active forms of phosphorylated GSK3β in neuroinflammation, we used seven cases (frontal cortex) of HIV-1 encephalitis (HIVE) and four seronegative age-matched controls provided by the National NeuroAIDS Consortium (Washington, DC), previously described.
The Institutional Review Board approved these studies. Serial paraffin sections (5 μm thick) were cut and double immunostained (by indirect immunofluorescence) for CD68 (monoclonal Abs; dilution, 1:200; DakoCytomation, Carpentaria, CA), and the GSK3β active form was phosphorylated on T216 (polyclonal Abs; dilution, 1:50; Abcam, Cambridge, MA). The infiltration of activated macrophages and the level of infection were assessed using monoclonal Abs to HLA-DR (dilution, 1:200; DakoCytomation) and HIV-1 p24 (dilution, 1:10; DakoCytomation) on paraffin serial sections with the avidin-biotin immunoperoxidase Vectastain Elite ABC kit (Vector Laboratories, Burlingame, CA).
The values shown in all figures and those mentioned in the text represent the average ± SEM of experiments that were conducted multiple times (as indicated). Multiple group comparisons were performed by one-way analysis of variance with Dunnet's post hoc tests (adhesion assay, migration assay, FACS, and G-LISA). Statistical analyses were performed using Prism version 5 software (GraphPad Software Inc.). Differences were considered significant at P < 0.05.
Results
Active Form of GSK3β Is Expressed in Infiltrating Monocytes and Macrophages in HIVE
Our previous work suggested that GSK3β was activated in brain endothelium in HIVE.
Herein, we investigated whether activated GSK3β was up-regulated in mononuclear phagocytes within human brain tissues during neuroinflammation (eg, HIVE). Although control brain tissues showed minimal staining for HLA-DR (a macrophage activation marker) and no staining for HIV-1 p24 (Figure 1, A and B), HIVE cases demonstrated numerous HLA-DR– and p24-positive perivascular macrophages (Figure 1, D and E). Next, we used phospho-specific antibodies to T216 (on GSK3β), which is located on the activation loop of GSK3β, indicating an increase in GSK3β catalytic activity
detecting the active form of GSK3β (p-GSK3β, phosphorylated at T216) and CD68 (a monocyte/macrophage marker). Control brains demonstrated minimal to no staining with antibodies to p-GSK3β in perivascular macrophages (Figure 1C). A high level of p-GSK3β was detected in perivascular CD68-positive mononuclear cells (Figure 1F), macrophages, and microglia within microglial nodules in brain tissues of patients with HIVE (data not shown). Total GSK3β showed similar expression in control and HIVE brain tissues (data not shown). These results indicated that GSK3β was activated in mononuclear phagocytes, in addition to brain endothelium, during neuroinflammation (eg, HIVE).
Figure 1Active forms of GSK3β are detected in mononuclear phagocytes in HIVE. Paraffin sections (5 μm thick) of control human brains (A–C) or HIVE brains (D–F) are cut. The sections are immunostained with anti-HLA-DR (A and D) or p24 (B and E) or double immunostained with p-GSK3β Ab (green; C and F) and CD68/DAPI (red/blue; C and F). There is no staining for HLA or p24 in control brain tissues (A and B). HIVE cases demonstrate HLA-DR– (D) and p24-positive (E) perivascular monocytes/macrophages (arrowheads). Similarly, minimal staining for p-GSK3β is found in CD68-positive macrophages (C). In HIVE, infiltrating monocytes/macrophages (arrowheads) show a high level of p-GSK3β (F). L, lumen; P.S., perivascular space. Original magnification: ×200 (A, B, D, and E); ×400 (C and F).
GSK3β Inhibition in Monocytes Attenuate Their Adhesion to Brain Endothelium and Transendothelial Migration
Previous studies from our laboratory showed protection of BBB integrity by GSK3β inhibition in BMVECs associated with diminished monocyte adhesion to endothelial cells and migration across BBB models.
We also observed substantial diminution in leukocyte adhesion to brain endothelium in vivo using intravital microscopy. Because both endothelium and leukocytes could be the subject of GSK3β inhibition in vivo, in the present study, we investigated whether suppression of GSKβ in monocytes would also result in decreased inflammatory responses. We pretreated monocytic cells, U937, or primary human monocytes with GSK3β inhibitors to study their effects on monocyte adhesion to brain endothelium. Adhesion assays revealed that exposure of BMVEC to TNF-α induced an increase in adhesion of U937 monocytic cells (Figure 2A) and primary monocytes (Figure 2C), fourfold and sevenfold, respectively. Pretreatment of U937 monocytic cells with GSK3β inhibitors (Figure 2A) led up to a 60% dose-dependent reduction (P < 0.01) in adhesion (with the highest doses of LiCl, I3′M, and SB2) compared with adhesion of untreated cells to TNF-α–activated BMVECs. Pretreatment of U937 cells with the Rac1-specific inhibitor, NSC,
had a similar effect on attachment of U937 cells to the endothelial monolayer (reduction of approximately 40%). A decline in adhesion could also result in a decrease in the number of monocytes crossing the BBB. By using migration assays in an in vitro BBB model, we tested whether GSK3β inhibition in monocytes could prevent monocyte passage across BMVEC monolayers. We used CCL2 as a relevant cytokine that is up-regulated in the brain under neuroinflammatory conditions. The addition of CCL2 to the lower chamber of the BBB model resulted in a 2.5-fold enhanced migration of U937 cells (Figure 2B) and a fivefold increase in migration of primary human monocytes (Figure 2D) compared with control constructs without chemokine addition. Treatment of monocytes with the GSK3β inhibitors, LiCl, I3′M, or SB2, led to attenuation in the migration of U937 monocytic cells by 85% ± 8%. Exposure of primary monocytes to GSK3β inhibitors prevented their ability to cross the BBB by 93% ± 3%. These results suggested that GSK3β inhibition in monocytes reduced both adhesion and migration across activated endothelial monolayers. Pretreatment of primary monocytes with GSK3β inhibitors (Figure 2C) resulted in approximately 26% to 53% reduction of monocyte attachment compared with adhesion of untreated monocytes to TNF-α–treated BMVECs.
Figure 2GSK3β inhibitors reduce monocyte adhesion to and migration across BMVEC monolayers. A and C: BMVECs are stimulated with TNF-α (20 ng/mL, overnight) or left untreated. Calcein-labeled U937 cells (pretreated with different concentrations of LiCl, I3′M, SB2, or NSC for 24 hours) (A) or primary human monocytes (pretreated with 20 mmol/L LiCl or 1 μmol/L I3′M) (C) are washed with PBS, added to BMVECs, and allowed to adhere for 15 minutes. Unattached cells are rinsed with PBS, and the fluorescence is measured. The number of adherent cells is calculated based on a standard curve derived from the fluorescence intensity of known amounts of labeled cells. B and D: Transendothelial migration of U937 cells (B) or monocytes (D) is performed across TNF-α–stimulated BMVEC monolayers using CCL2 (50 ng/mL) as the chemotactic stimulus. Calcein-labeled U937 cells (B) or monocytes (D) pretreated with GSK3β inhibitors (20 mmol/L LiCl, 1 μmol/L I3′M, or 5 μmol/L SB2) are added to BMVEC monolayers, and the monocytes are allowed to migrate for 2 hours. Fluorescence is measured as described in Materials and Methods. Data represent the mean ± SEM values of fold difference from three different experiments performed in triplicate; primary monocytes are from two different donors. *P < 0.01, †P < 0.05 indicates significant inhibition.
GSK3β Inhibition in Monocytes Diminishes BBB Injury
One of the early stages of neuroinflammation was the adhesion and engagement of leukocytes with microvascular endothelial cells. Engagement by leukocytes and subsequent migration across the BBB led to enhanced barrier permeability. We hypothesized that, because GSK3β inhibition in monocytes reduced their ability to attach and migrate through BMVEC monolayers (Figure 2), it would also prevent monocyte-mediated disruption of the BBB. To evaluate this idea, we treated U937 cells with GSK3β inhibitors before their addition to BMVEC monolayers plated on ECIS electrode arrays, with or without TNF-α exposure. The integrity of the barrier was evaluated by measuring TEER (Figure 3). The addition of untreated monocytes to BMVECs resulted in a 9.5% ± 0.5% decrease of TEER during the first 9 hours, whereas the addition of GSK3β inhibitor-treated monocytes produced a 4% decrease in TEER. Stimulation of BMVECs by TNF-α, without addition of monocytes, did not produce a significant effect on transendothelial electrical resistance. Figure 3 shows that, in wells in which GSK3β inhibitor-treated monocytes were added to BMVECs, tightness of the barrier recovered to initial levels after 21 hours, whereas tightness of BMVEC monolayers with untreated monocytes was reduced by approximately 12%, even at 21 hours. The addition of untreated or treated U937 cells to BMVECs had similar effects on TEER (data not shown). Thus, GSK3β inhibition in monocytes protected against the leakiness of the BBB because of diminished immune cell engagement with the brain endothelium.
Figure 3GSK3β inhibition in monocytes improves barrier function. TEER is measured in BMVECs with TNF-α pretreatment during engagement by monocytes. Results from a representative experiment with the GSK3β inhibitor, SB2 (5 μmol/L), are shown. The percentage change in TEER of BMVECs at 3, 9, and 21 hours after addition of monocytes, with or without pretreatment with GSK3β inhibitor, is shown. Results are presented as an average from at least two independent experiments consisting of at least three replicates. *P < 0.01, †P < 0.05 versus untreated control.
Because GSK3β inhibition in monocytes affected their ability to attach to and migrate across endothelial monolayers, we hypothesized that the activation status of integrin (ensuring rolling and firm adhesion of leukocytes) would be affected in monocytes after GSK3β inhibition. To study active integrins, we used conformationally sensitive monoclonal Abs binding to epitopes usually hidden in resting cells and exposed only in the active conformation. Chigaev et al
showed rapid conformational changes in VLA-4 using the conformationally sensitive monoclonal Ab, HUTS21, after stimulation with a VLA-4 ligand (LDV peptide), an inducer of conformational change. By using the same approach, we analyzed HUTS21 binding in U937 cells (Figure 4). The addition of HUTS21 antibody to U937 cells resulted in 4% to 5% binding in non–LDV-stimulated samples (Figure 4A). Exposure to LDV peptide shifted integrin β1 conformation toward the activated, open form (Figure 4A). To test if GSK3β inhibition of monocytes could affect activation of VLA-4, we treated U937 cells with GSK3β inhibitors (LiCl, I3′M, or AR) (Figure 4A). HUTS21 binding to activated integrin β was quantified by FACS before and after LDV stimulation in inhibitor-treated cells. LDV exposure increased the change to the active VLA-4 configuration by 20-fold, and GSK3β inhibition decreased activation of integrin β1 by 60% ± 6% (Figure 4B). Because Rac1 GTPase was involved in activation of VLA-4 integrins,
we treated U937 cells with the Rac1-specific inhibitor, NSC. Pretreatment of U937 cells with NSC showed a similar diminution in active integrin expression, as in GSK3β inhibitor-treated cells.
Figure 4GSK3β inhibition prevents conformational activation of VLA-4 in U937 cells. A: Stimulation of U937 cells with LDV (a peptide consensus sequence of VCAM-1 and fibronectin
) leads to changes in integrin β1 conformation from a closed (not active, gray) to an open (active, black) form. U937 cells treated with LiCl, I3′M, AR, or NSC active form of VLA-4 (green). The expression of total VLA-4 is not affected by inhibitors (A, right panel). B: Quantitation of integrin β1 conformational activation in which the MFI of activated nontreated cells is assigned a value of 100. The results are shown as the mean ± SEM from three independent experiments. *P < 0.05 versus untreated control. Max, maximum.
Next, we measured activation of integrin β1 in primary monocytes using the same approach (Figure 5). In primary monocytes, LDV enhanced expression of active VLA by 14-fold. The Rac1 inhibitor, NSC, attenuated integrin activation by 40% ± 2%. LiCl, I3′M, and AR treatment resulted in 40%, 45%, and 30% diminution in active VLA-4 conformation, respectively. Total levels of integrin β1 were not affected by any treatment in either U937 cells (Figure 4) or primary monocytes (Figure 5). In summary, GSK3β or Rac1 inhibition suppressed the conformational activation of VLA-4.
Figure 5GSK3β inhibition prevents conformational activation of VLA-4 (integrin β1) in primary human monocytes. Stimulation with LDV results in expression of active VLA-4 (integrin β1) detected by HUTS21 monoclonal Ab. A: Pretreatment of primary monocytes with LiCl, I3′M, AR, or NSC significantly down-regulates the active form of VLA-4 (green). GSK3β inhibitors did not affect total expression of VLA-4 (integrin β1) (right column). B: Quantitation of integrin β1 conformational activation in human monocytes in which the MFI of activated nontreated cells is assigned a value of 100. The results are shown as the mean ± SEM from three independent experiments. *P < 0.05 versus untreated control. Max, maximum.
Rac1 Activation Is Reduced in Monocytes by GSK3β Inhibition
To detect the active, GTP-bound form of Rac1, we used a commercially available G-LISA kit (Cytoskeleton Inc.). Activation of integrin signaling can be achieved by cross-linking with antibodies mimicking engagement with endothelium.
VLA-4 integrin cross-linking on human monocytic THP-1 cells induces tissue factor expression by a mechanism involving mitogen-activated protein kinase.
Cross-linking of the α4 subunit of VLA-4 induced Rac1 activation 2- and 2.5-fold in U937 cells and primary monocytes, respectively (Figure 6, A and C). The use of nonimmune IgG (negative control) had no effect on Rac1 activation. Treatment with the GSK3β inhibitors, LiCl and SB2, resulted in 95% and 80% reduction in Rac1 activation, respectively (P < 0.01), in U937 cells, and in 41% and 46% reduction in primary monocytes, respectively. The addition of I3′M had less of an effect on Rac1 activation in U937 cells (approximately 50% reduction) (P < 0.05) and in primary monocytes (approximately 33% reduction).
Figure 6GSK3β inhibitors decrease Rac1 activation in primary monocytes. Rac1 GTPase activity is measured in U937 cells (A and B) and primary monocytes (C and D). A and C: Rac1 is activated in monocytes by cross-linking of the VLA-4 receptor with α-CD49d antibody (mimicking leukocyte engagement with endothelial cells). Nonimmune IgG serves as the negative control. Cells are starved (1% FBS/RPMI 1640 media supplemented with penicillin-streptomycin for 24 hours) and then pretreated with inhibitors (20 mmol/L LiCl, 1 μmol/L I3′M, or 5 μmol/L SB2) overnight before cross-linking. The level of Rac1 activity in non–cross-linked cells is assigned a value of 1 to calculate the relative ratio of activation. B and D: Rac1 is activated in monocytes by LPA (100 ng/mL, black bars) or LPS (100 ng/mL, hatched bars) in the presence or absence of GSK3β inhibitors (20 mmol/L LiCl, 1 μmol/L I3′M, or 5 μmol/L SB2). The level of Rac1 activity in nonstimulated and nontreated cells is assigned a value of 1 to calculate the relative ratio of activation. The results are shown as the mean ± SEM value from three independent experiments. *P < 0.01, †P < 0.05 versus untreated control. neg, negative; NT, non-treated.
We next examined the ability of GSK3β inhibitors to decrease Rac1 activation in primary monocytes stimulated with LPA or LPS. A significant increase in Rac1 activation was observed in monocytes stimulated with LPA or LPS (Figure 6, B and D), and a substantial inhibition (P < 0.01) was observed in the presence of GSK3β inhibitors (LiCl, I3′M, and SB2), albeit at different levels. Inhibitors did not have any effect on Rac1 baseline activity in monocytes. In summary, for the first time to our knowledge, we showed that GSK3β inhibition altered Rac1 activation in monocytes.
GSK3β Inhibition Prevents Dephosphorylation of Inhibitory Sites of VASP and Cofilin
To control and coordinate actin filament-based lamellipodia formation and cell movement, efficient disassembly of pre-existing actin filaments was necessary. A search for proteins that regulate filamentous and globular actin levels revealed an increase in the levels of an inactive phosphorylated form of the F-actin–severing protein, cofilin, in cells unable to generate actin-based lamellipodia. Rac1 has been implicated in the dynamic regulation of cofilin activity in several cell types.
After detection of cofilin phosphorylation in U937 cells without stimulation, we examined the dynamics of cofilin phosphorylation in cells undergoing changes in actin cytoskeletal reorganization. As a stimulus for actin cytoskeletal rearrangements, we used cross-linking of integrins by antibodies, similar to the cross-linking used in the Rac1 activation assays. Cross-linking did not substantially affect cofilin; however, treatment of U937 cells with the GSK3β inhibitors, LiCl and I3′M, resulted in a 50% ± 6% increase of phosphorylation levels of the inhibitory site, pS3, of cofilin.
SB2 treatment produced a 95% ± 7% increase in pS3-cofilin (Figure 7). The effect of the Rac1-specific inhibitor, NSC, on the phosphorylation level of pS3-cofilin was similar to that of LiCl and I3′M.
Figure 7GSK3β inhibition increases levels of pS3 cofilin and pS157 VASP. U937 cells are treated with 20 mmol/L LiCl, 1 μmol/L I3′M, or 5 μmol/L SB2 before cross-linking experiments. Cells are lysed, and levels of total/phosphorylated cofilin and VASP are analyzed by using Western blot analysis. The level of phosphorylation in non–cross-linked cells is assigned a value of 1 to calculate the relative ratio. The results are shown as the mean ± SEM value from three independent experiments. *P < 0.01, †P < 0.05 versus untreated control. neg, negative.
Because GSK3β inhibitors affect Rac1 activity, they could also implicate phosphorylation of VASP protein in leukocytes. Because S157 and S239 sites are conserved among mammals,
we tested the levels of phosphorylation of these sites by using Western blot analysis. Treatment of U937 cells with LiCl or SB2 resulted in a significant (P < 0.05) increase (30% ± 7%) in phosphorylation of S157-VASP (Figure 7). Application of the GSK3β inhibitor, I3′M, or the Rac1 inhibitor, NSC, had little effect on pS157-VASP levels. Treatment of U937 cells with the inhibitors previously described did not have any effect on phosphorylation of pS239 (data not shown). Together, these results strongly suggested that GSK3β inhibition affected actin-binding proteins, cofilin and VASP, via Rac1 inhibition.
Inhibition of GSK3β Reduces Lamellipodia Formation
Formation of lamellipodia was necessary for leukocyte migration associated with recruitment of active Rac1 into the leading edge of the cell.
To assess the possibility that inhibition of GSK3β affected lamellipodia formation, the lamellipodia area of both U937 and primary monocytes cells was measured. Maximum lamellipodia protrusion/formation was observed 30 minutes after addition of LDV peptide or CN04, the G-switch direct activator of small G protein (Cytoskeleton Inc.) (Figure 8A). The area occupied by lamellipodia increased by 1.8- to 2-fold after treatment with LDV or CN04. When primary monocytes or U937 cells were treated with GSK3β inhibitors (as described for the previous experiments), the area of the lamellipodia was reduced by 1.8- and 2.2-fold (P < 0.01; Figure 8, B and C). Inhibition of Rac1 by NSC also led to a 1.7- and 2.4-fold decrease of lamellipodia formation in primary monocytes and U937 cells, respectively, when compared with nontreated LDV-stimulated control cells. These data indicated that inhibition of GSK3β activity reduced lamellipodia protrusion in monocytic cells via Rac1 inhibition.
Figure 8Effects of GSK3β inhibition on lamellipodia formation. Lamellipodia areas are measured at the time of maximum protrusion after 30 minutes of incubation of primary monocytes with LDV peptide. A: Images (taken at ×63 objective) of primary monocytes stained with Acti-stain 488 fluorescent phalloidin. Protruding lamellipodia are indicated (arrowheads). Primary monocytes treated with or without GSK3β inhibitors (20 mmol/L LiCl, 1 μmol/L I3′M, or 5 μmol/L SB2), Rac1 inhibitor (75 mmol/L NSC), or GTPase activator (1 μg/mL CN04, positive control) are shown. The average lamellipodium area (in pixels) per cell (20 cells per treatment) is calculated for primary monocytes (B) and U937 cells (C). The results are shown as the mean ± SEM value from two independent experiments. *P < 0.01 compared with untreated control.
Leukocyte-endothelial interactions are at the center of any inflammatory response and ensuing subsequent tissue injury. Monocyte infiltration across the BBB plays an important role in neuroinflammatory conditions, such as multiple sclerosis and HIVE.
Diminution of monocyte invasion is associated with lessened brain damage. In the present study, we tested whether GSK3β inhibition could attenuate monocyte adhesion and migration across monolayers of primary human brain endothelial cells. Enhanced adhesion of monocytes to TNF-α–stimulated BMVECs and their migration across BMVEC monolayers (driven by CCL2 chemotaxis) were decreased by 60% and 85%, respectively, after treatment of primary monocytes or U937 cells with GSK3β inhibitors. Limited data are available regarding the functional role of GSK3β and the Wnt pathway in leukocytes during inflammatory processes. Inhibition of GSK3β by LiCl or activation of the Wnt–β-catenin pathway results in decreased monocyte migration across endothelial monolayers.
Glycogen synthase kinase 3 activation is important for anthrax edema toxin-induced dendritic cell maturation and anthrax toxin receptor 2 expression in macrophages.
The Rho GTPases are regulatory molecules that link surface receptors to organize the actin cytoskeleton, transendothelial leukocyte migration, oxidative stress, and inflammation.
In leukocytes, Rac1 and Cdc42 regulate cell polarity, lamellipodia formation, and direct migration, whereas RhoA controls leukocyte tail retraction during transmigration.
RhoA signaling is essential for integrin-dependent adhesion to the endothelium via leukocyte function-associated antigen (or αLβ2) for intercellular adhesion molecule-1 and very late antigen (VLA-4; or α4β1) for VCAM-1.
We sought to determine whether Rac1 inhibition would mediate a decreased adhesion/migration of monocytes across BMVEC monolayers. We measured the activation of Rac1 by G-LISA after cross-linking with antibodies to CD49d or by stimulation by LPA or LPS in both primary monocytes and U937 monocytic cells. The significant increase in active Rac1 after cross-linking or stimulation was suppressed in both cell types by GSK3β inhibitors. Little is known about the connection between GSK and Rac1 in inflammatory cells. Inactivation of GSK3β in monocytes from diabetic patients was associated with Rac1 inhibition; however, it paradoxically led to increased monocyte migration.
Differences in cell culture conditions and stimulus (platelet-activating factor) may explain the observed differences. More important, several reports have indicated suppression of immune cell adhesion to
the BBB in animal models of systemic inflammation and experimental autoimmune encephalomyelitis that could be, in part, mediated by GSK3β inhibition in leukocytes.
Up-regulation of conformationally active VLA-4 was linked to active Cdc42 and Rac1.
Decreased migration of monocytic cells correlated with diminished expression of active VLA-4 (detected by HUTS21) on the cell surface and a fourfold decrease in active Rac1 and Cdc42.
Because we observed diminished adhesion of monocytes treated with GSK3β inhibitors, we explored the idea that suppression of GSK3β may result in conformational changes of active integrin and/or total integrin expression. We used LDV consensus peptide stimulation to reproduce activation of monocytes by VCAM-1 and fibronectin during leukocyte adhesion.
LDV stimulation increased expression of conformationally active VLA–4 (14– to 20–fold) within minutes; GSK3β inhibitors or a specific inhibitor of Rac1 decreased expression of active VLA–4 by 40% to 60%. Expression of total VLA–4 was not changed after LDV treatment, indicating that the conformational status of VLA–4 is primarily responsible for the adhesion observed in the presence of GSK3β inhibitors. Previously, memory T cells that constitutively express activation/ligand–induced epitopes on β1 integrins recognized by HUTS21 exhibited significantly higher rates of attachment and accumulation on VCAM–1–expressing cells compared with other T–cell subsets without active epitope expression.
have also established that the rate of HUTS21 binding is also related to the VLA-4 activation state, even at a saturating ligand concentration. Conformational unbending of the integrin molecule could facilitate both tethering and rolling.
More important, the specific inhibitor of Rac1, NSC, decreased expression of conformationally active VLA-4 to the same extent as did the GSK3β inhibitors (Figure 4, Figure 5).
have demonstrated that Rac1 recruits high-affinity integrins localized at the leading edge of spreading and migrating cells. VLA-4 activation localizes to lamellipodia
that are required for leukocyte migration. Therefore, we evaluated the effects of LDV stimulation on lamellipodia formation and cytoskeleton alterations. Treatment with LDV resulted in a significant increase in lamellipodia and redistribution of actin (Figure 8). GSK3β inhibitors diminished lamellipodia formation to control levels in primary human monocytes and U937 cells (Figure 8, A–C). Stimulation of Rho GTPases resulted in a similar lamellipodia increase, which was inhibited by a specific Rac1 inhibitor, indicating the involvement of Rac1 in this process.
Because GSK3β inhibition in monocytes decreased monocyte adhesion to and migration across BMVEC monolayers, we investigated whether it could also protect barrier function. Previously, we showed
that BMVEC engagement by monocytes resulted in diminished barrier tightness (indicated by a TEER decrease) that was reversed by GSK3β suppression in BMVECs. The current study demonstrated that suppression of GSK3β in U937 cells and in primary human monocytes attenuated barrier dysfunction (Figure 3), paralleling decreased adhesion, diminished expression of active VLA-4, and lamellipodia formation. The notion that BBB impairment is associated with leukocyte migration into the central nervous system is well accepted.
Disengagement of inflammatory cells from brain endothelium via GKS3β inhibition may prove beneficial in multiple sclerosis, stroke, and infection-driven encephalitis, among other inflammatory conditions.
The leukocyte cytoskeleton undergoes significant changes, depending on whether the cell is circulating within the bloodstream or migrating through tissues. Cytoskeleton-inhibiting proteins, such as cofilin, regulate rearrangements for these different cell states.
Serine 3 phosphorylation of cofilin, an actin-depolymerizing protein, promotes the disassembly of F-actin, thereby rendering cofilin unable to depolymerize F-actin and stabilizing the cytoskeleton.
Therefore, we investigated the effect of GSK3β suppression on levels of phosphorylated/total cofilin in U937 cells and showed that GSK3β inhibition increased the amount of phosphorylated cofilin, without a significant change in total cofilin (Figure 7). The specific Rac1 inhibitor demonstrated the same changes.
VASP is involved in filopodia formation and regulation of actin cytoskeleton formation. Human VASP harbors three phosphorylation sites, including S157, S239, and T278. The last two sites are associated with impaired filament formation,
Dephosphorylation of VASP allows its binding to talin, which, in turn, binds to integrin cytoplasmic domains, leading to a separation of the integrin cytoplasmic domains and inside-out activation.
Our results indicate that two of the three GSK3β inhibitors tested increased phosphorylation of VASP at S157. VASP phosphorylation of S157 has inhibited neutrophil migration and is associated with integrin β2 regulation.
Additional studies are necessary to prove whether GKS3β inhibition regulates active conformation of integrin β1 via VASP phosphorylation. Figure 9 summarizes potential intracellular pathways associated with GSK3β inhibition in monocytes, leading to the decreased ability to adhere/migrate and to engage brain endothelium. These observations suggest that GSK3β inhibition in leukocytes could be a target for therapeutic intervention of neuroinflammatory disorders.
Figure 9Cellular signaling events halting the ability of monocytes to engage brain microvascular endothelial cells and migrate across the BBB after GSK3β inhibition. On interaction with ligand, integrins α and β change their conformation from a nonactive (low-affinity) form to an active (high-affinity) form and consequently mediate extracellular signals to cytoskeletal elements and signaling factors via Rac1 GTPase proteins. Paxillin, talin, and vinculin are cytoskeleton (actin)-interacting proteins that are involved in inside-out integrin signaling. Inhibition of GSK3β affects inside-out and outside-in integrin signaling via diminished Rac 1 GTPase activity, thereby affecting activation of VASP and cofilin and actin rearrangements. As a result of these events, lamellipodia formation, adhesion, and migration of monocytes through the brain endothelial monolayer are affected. PAK, p21-activated kinase.
We thank the NIH National NeuroAIDS Consortium for the brain tissue specimens, Mary Olson for assistance with Figure 9 and Dionne Tyler for technical assistance with preparation of paraffin sections of human brain tissues.
GSK3β inhibitors do not affect calcein-AM staining or cause toxic effects in U937 monocytic cells. A: U937 cells (pretreated with 20 mmol/L LiCl, 1 μmol/L I3′M, 5 μmol/L SB2, 1 μmol/L AR, or 75 mmol/L NSC for 24 hours) are stained with 5 μmol/L calcein-AM for 30 minutes. Cells are rinsed with PBS and resuspended in FACS buffer (1% bovine serum albumin in PBS), and the amount of fluorescently labeled cells is measured by FACS. The percentage of labeled cells is calculated based on a comparison of total amount of cells versus stained cells. The results are shown as the mean ± SEM value from two independent experiments. B: U937 cells (pretreated with 20 mmol/L LiCl, 1 μmol/L I3′M, 5 μmol/L SB2, 1 μmol/L AR, or 75 mmol/L NSC for 24 hours) are rinsed with PBS and resuspended in FACS buffer (1% bovine serum albumin in PBS). The viability of cells is measured using the Live/Dead Cell Viability Assay (Invitrogen/Life Technologies) and quantified by FACS. Cells treated with 0.01% saponin are used as a positive control for dead cells. The total amount of cells is assigned the value of 100%. The percentage of live cells is calculated by the regression curve calculation tool of Prism version 5 software (GraphPad Software Inc.). The results are shown as the mean ± SEM value from two independent experiments. N.S., nonsignificant.
References
Ley K.
Laudanna C.
Cybulsky M.I.
Nourshargh S.
Getting to the site of inflammation: the leukocyte adhesion cascade updated.
Activation of peroxisome proliferator-activated receptor gamma (PPARgamma) suppresses Rho GTPases in human brain microvascular endothelial cells and inhibits adhesion and transendothelial migration of HIV-1 infected monocytes.
VLA-4 integrin cross-linking on human monocytic THP-1 cells induces tissue factor expression by a mechanism involving mitogen-activated protein kinase.
Real time analysis of the affinity regulation of alpha 4-integrin: the physiologically activated receptor is intermediate in affinity between resting and Mn(2+) or antibody activation.
Conformational regulation of alpha 4 beta 1-integrin affinity by reducing agents: “inside-out” signaling is independent of and additive to reduction-regulated integrin activation.
Glycogen synthase kinase 3 activation is important for anthrax edema toxin-induced dendritic cell maturation and anthrax toxin receptor 2 expression in macrophages.
00493-2&id=gr1.jpg){kind=link}
00493-2&id=gr2.jpg){kind=link}
00493-2&id=gr3.jpg){kind=link}
00493-2&id=gr4.jpg){kind=link}
00493-2&id=gr5.jpg){kind=link}
00493-2&id=gr6.jpg){kind=link}
00493-2&id=gr7.jpg){kind=link}
00493-2&id=gr8.jpg){kind=link}
00493-2&id=gr9.jpg){kind=link}