Published online before print August 29, 2006
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Departments of
* Molecular Cell Biology and
Blood Cell Research, Sanquin Research and Landsteiner Laboratory, Academic Medical Center, University of Amsterdam, Amsterdam, The Netherlands
1 Correspondence: Sanquin Research and Landsteiner Laboratory, Academic Medical Center, University of Amsterdam, Plesmanlaan 125, 1066 CX Amsterdam, The Netherlands. E-mail: p.hordijk{at}sanquin.nl
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: cAMP • serotonin • polarization
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
In lymphocytes, the small guanosinetriphosphatase (GTPase) Rap1 plays a crucial role in the stimulation of integrin-mediated adhesion, cell polarization, and cell motility [2 ]. Rap1 is activated upon GTP binding, which is induced by specific guanine nucleotide exchange factors (GEFs). Several GEFs for Rap1 have been identified, including the recently found exchange protein directly activated by cAMP (Epac) [3 ]. Two Epac isoforms, Epac1 and -2, have been described. Epac1 was found to be expressed in most tissues, and Epac2 is expressed in the adrenal gland and in the brain [4 ]. Attempts to detect Epac expression in leukocytes showed the presence of Epac1 mRNA only in B cells and U937 cells, and Epac1 protein was detected in macrophages. Epac2 was undetectable in all hematopoietic cell types [5 ].
The second messenger cAMP is crucially involved in multiple cellular processes. Until Epac was identified, cAMP-dependent signaling was thought to be carried out by protein kinase A (PKA). Currently, a growing number of studies implicate Epac1 in the regulation of several cAMP-dependent effects including substrate adhesion and cell-cell adhesion in adherent cells [6
7
8
9
10
], Ca++-induced exocytosis [11
, 12
], neurite extension [13
], and Fc
receptor (FcR)-mediated phagocytosis in macrophages [14
]. However, the role of Epac in leukocyte adhesion and chemotaxis has not been established.
Epac becomes activated by stimuli that bind to receptors signaling via the heterotrimeric Gs proteins, which induce a rise in cAMP levels through the activation of adenylate cyclase. These stimuli include serotonin, PGs, and β2adrenergic agonists [6 , 15 16 17 ]. In neurons, the Epac-Rap pathway regulates serotonin-induced secretion of amyloid precursor protein as well as ERK activation [15 , 16 ]. It is interesting that in addition to being a neurotransmitter, serotonin plays an important role in inflammation. It is secreted by mast cells and platelets and induces chemotaxis of eosinophils, lymphocytes, and macrophages [18 19 20 ].
In this study, we demonstrate that Epac1 is expressed in leukocytes, platelets, and hematopoietic cells, and we investigate its functionality in monocytic U937 cells and in primary monocytes. Our results show that activation of Epac1 promotes cell adhesion and polarization and enhances chemokine-induced migration.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
was purchased from Boehringer Mannheim (Germany).
Cell culture
All cell lines were purchased from American Type Culture Collection (ATCC; Manassas, VA) and were cultured at 37°C and 5% CO2. U937 cells (monocytic cell line) were maintained in RPMI-1640 medium (Gibco, Grand Island, NY) containing 10% heat-inactivated FCS (Gibco), 2 mM L-glutamine, 100 U/ml penicillin, and 100 µg/ml streptomycin. HL-60 cells (leukemic cell line) and Chinese hamster ovary (CHO) cells were cultured in IMDM (BioWhittaker, Brussels, Belgium) containing 10% heat-inactivated FCS (Gibco), 2 mM L-glutamine, 100 U/ml penicillin, and 100 µg/ml streptomycin. HUVEC were isolated from umbilical cord veins as described previously [21
].Cells were cultured to confluency in M199 medium (Gibco) containing 20% heat-inactivated FCS (Gibco), 2 mM L-glutamine, 100 U/ml penicillin, 100 µg/ml streptomycin, 100 µg/ml heparin, and 50 µg/ml endothelial mitogen (Sanbio BV, The Netherlands). Endothelial cells of second or third passage were used. HUVEC monolayers were stimulated for 16 h with 100 U/ml TNF- prior to the perfusion experiments.
Cell isolation
Blood was obtained from healthy volunteers. Granulocytes and PBMC were isolated from buffy coats of 500 ml blood by density gradient centrifugation over isotonic Percoll (Pharmacia Biotech, Uppsala, Sweden) with a specific gravity of 1.076 g/ml [22
]. For further purification of monocytes and lymphocytes, the PBMC fraction was incubated with magnetic beads coated with anti-CD14 antibodies, and monocytes were purified with a MACS separation system according to the manufacturer’s instructions (Miltenyi Biotec GmgH, Bergisch Gladbach, Germany). Previous experiments performed in our lab have shown that this isolation protocol does not induce monocyte activation [23
]. The remaining monocyte-depleted fraction of PBMC was labeled with anti-CD3 and anti-CD19 antibodies, and B and T cells were subsequently sorted with a Mo Flo sorter (Dako Cytomation, Denver, CO).
After lysis of the erythrocytes with ice-cold lysis buffer containing 155 mM NH4Cl, 10 mM KHCO3, and 0.1 mM EDTA (pH 7.4), the granulocyte fraction was incubated with an anti-CD16 antibody. Subsequently, neutrophils were sorted with a Mo Flo sorter. Eosinophils were isolated by means of the fMLP method [24 ]. In brief, granulocytes were incubated for 30 min at 37°C to restore initial cell density. Cells were then washed and resuspended in PBS containing 0.5% human serum albumin (CLB, Amsterdam, The Netherlands) and 13 mM trisodium citrate and incubated for 10 min at 37°C after the addition of 10 nM fMLP to the cell suspension. Eosinophil and neutrophil fractions were separated by centrifugation (20 min., 1000 g) over isotonic Percoll (1.082 g/ml, pH 7.4). Platelets were isolated as described previously [23 ]. CD34+ hematopoietic progenitor cells were isolated from cord blood by density gradient centrifugation over Ficoll-paque (1.077 g/ml; Pharmacia Biotech) as described previously [25 ].
The purity of all the isolated cell populations was greater than 95%.
RT-PCR
RNA was isolated from purified leukocyte fractions, platelets, and CD34+ cells and the cell lines U937 and HL-60 by lysis of the cells in a solution containing guanidine isothiocyanate followed by centrifugation over a layer of CsCl (5.7 M) in a Beckman OptimaTM L-100 XP ultracentrifuge (Beckman Instruments Inc., Palo Alto, CA) with an SW41 rotor. cDNA was synthesized using 2.5 µM oligo-dT primers and 10 U/ml Superscript III RT (Invitrogen, Carlsbad, CA; 5 min, 65°C). Quantitative PCR was done with the FastStart DNA Master Plus SYBR Green I kit in the LightCycler instrument (Hoffman-La Roche, Basel, Switzerland). Epac1 RNA was amplified by PCR with the following protocol: 10 min at 95°C, followed by 50 cycles of 5 s at 95°C, 30 s at 65°C, and 15 s at 72°C. The following Epac1-specific primers were used: forward primer, 5'TTGGAGAATGGCTGTGGGAATGCATC3' (exon 14); reverse primer, 5'CCGAGTTGCTGAGGCCAAACATGAC3' (exon 19). The mRNA of the housekeeping gene β-glucuronidase was amplified as internal control using specific primers [26
]. Amplified Epac1-specific cDNA was compared with the standard β-glucuronidase. The specificity of the PCR products was checked by DNA sequencing with the Big-dye Terminator sequencing kit v1.1 and analyzed on a Genetic Analyzer 3100 platform (both from Applied Biosystems, Foster City, CA).
Western blot analysis
Isolated, primary leukocytes, platelets, and cell lines were lysed in Laemmli sample buffer containing a protease inhibitor cocktail (Roche, Indianapolis, IN) and incubated for 15 min at 95°C. Cell lysates were separated on 10% SDS-PAGE gels and transferred onto polyvinylidene difluoride membranes (BioRad Laboratories, Hercules, CA). Epac1 was detected with a rabbit polyclonal anti-Epac1 antibody (Upstate, Lake Placid, NY). Actin was detected with a mouse mAb against actin (Ab1, Oncogene, Darmstadt, Germany).
Integrin activation
U937 cells were resuspended in IMDM containing 0.25% BSA and stimulated with 8CPT-2Me-cAMP for various time periods at 37°C. Stimulation was stopped by the addition of ice-cold PBS containing 0.5% BSA, and cells were pelleted by centrifugation (490 g, 5 min, 4°C). Cells were then resuspended in ice-cold PBS containing 0.5% BSA and incubated with a mouse mAb against the activated conformation of β1-integrins (12G10, Imgen, The Netherlands; ref. [27
]) or β2-integrins (CBRM1/5, a kind gift from Dr. Kevin L. Moore, University of Oklahoma, Norman; ref. [28
]) for 30 min at 4°C. Total surface expression of β1- and β2-integrins was detected with mouse mAb specific for β1- or β2-integrins (CLB). After washing with ice-cold PBS-0.5% BSA, bound mAb were detected with PE-labeled goat anti-mouse-Ig (Dako Cytomation, Denmark) for 30 min at 4°C. The fluorescence intensity of labeled cells was measured with a FACScan flow cytometer (Becton Dickinson, San Jose, CA). Integrin activation was calculated by correcting for the amount of total integrins in every sample and was expressed as fold increase over control-unstimulated cells.
Adhesion assay
Flat-bottom 96-well plates (Maxi Sorp Nunc, Denmark) were coated with 20 µg/ml human fibronectin (Sigma Chemical Co.) for 16 h at 4°C and blocked with 0.5% BSA (Sigma Chemical Co.) for at least 1 h at 37°C. U937 cells were washed in IMDM containing 0.25% BSA, labeled with Calcein-AM, according to the manufacturer’s instructions (Molecular Probes, Leiden, The Netherlands), for 30 min at 37°C, and washed twice with IMDM-BSA. Labeled U937 cells were preincubated or not with 8CPT-2Me-cAMP for 15 min and subsequently added to the fibronectin-coated 96-well plates (2x105 cells per well). Plates were centrifuged at 40 g for 1 min, and stimuli were immediately added. After a further incubation for 30 min at 37°C, nonadhered cells were removed by washing three times with warm PBS. Adhered cells were lysed with 0.5% Triton X-100, and fluorescence was measured on a GENios Plus plate reader (Tecan, Salzburg, Austria). The percentage of adhesion was calculated by dividing the measured fluorescence intensity by the fluorescence intensity of the input cells (set to 100%).
VLA-5 and CD11b-blocking experiments were performed by preincubating cells for 30 min with the mAb SAM-1 (CLB) and 44a (ATCC), respectively. To block undesirable FcR activation by the blocking VLA-4 antibody HP 2/1, cells were incubated for 10 min with the anti-FcR antibodies anti-CD32 (Medarex, Annandale, NJ) and anti-CD16b (CLB), prior to the incubation with a mixture of HP 2/1 and anti-FcR antibodies for an additional 30 min.
Perfusion assay
Perfusion experiments were performed as described previously [23
]. In brief, the monocyte suspension (2x106 cell/ml in incubation buffer) was aspirated from a reservoir through plastic tubing and perfused through a chamber (containing the HUVEC monolayer) with a Harvard syringe pump (Harvard Apparatus, Holliston, MA) at flow rate of at 0.8 dyn/cm2. During perfusions, the flow chamber was mounted on a microscope stage (Axiovert 25, Zeiss, Germany), equipped with a black and white charged-coupled device video camera (Sanyo, Osaka, Japan) and coupled to a VHS video recorder [29
]. Video images were evaluated for the number of adherent monocytes and the rolling velocity per cell, and dedicated routines were made in the image analysis software Optimas 6.1 (Media Cybernetics Systems, Silver Spring, MD). The number of surface-adherent monocytes was measured after 5 min of perfusion at a minimum of 25 randomized high-power fields. To automatically determine the velocity of rolling cells, custom-made software was developed in Optimas 6.1. A sequence of 50 frames representing an adjustable time interval (
t, with a minimal interval of 80 ms) was digitally captured. The position of every cell was detected in each frame, and for all subsequent frames, the distance traveled by each cell and the number of images in which a cell appears in focus were measured. The cut-off value to distinguish between rolling and static adherent cells was set at 1 µm/s. With this method, static adherent cells and rolling and free-flowing cells could be distinguished and quantified clearly.
Transmigration assay
Migration assays were performed as described previously [25
]. In brief, Transwells of 6.5 mm diameter, with 5-µm pore size filters (Costar, Cambridge, MA) were coated with 20 µg/ml fibronectin (Sigma Chemical Co.). Before use, cells were washed once with migration medium (IMDM containing 0.25% BSA). At the start of the assay, 105 cells were placed in the upper compartment of the Transwells and allowed to migrate for 1 h at 37°C to chemokine-containing medium added to the lower compartment. Migrated cells were collected from the lower compartment and quantified by flow cytometric analysis in the presence of a fixed amount of control cells labeled with Calcein-AM (Molecular Probes). The percentage of migrated cells was calculated as a fraction of the total cell input as follows: percent of migrated cells = [(number of transmigrated cells/number of input-labeled cells)/(number of not-labeled input cells/number of labeled input cells)] x100%.
For primary monocyte migration, transmigrated cells were collected from the lower compartment of the Transwell as well as from the bottom of the Transwell filter (as a fraction of transmigrated monocytes remains adhered to the bottom of the filter). Monocytes in the lower compartment were quantified by flow cytometric analysis as indicated above. Cells adhering to the bottom side of the filters were counted under a fluorescent microscope (three random fields) after removing cells on the top side of the filters with a cotton swab, followed by fixation and staining with Hoechst (Molecular Probes). The percentage of transmigrated monocytes that adhered to the filter was calculated as follows: percent of migrated cells adhered to the filter = [(number of transmigrated cells counted per field/surface of the field)/(number of input cells/surface of the filter)] x 100%. The calculated percentages of the two fractions were added to give the total percentage of transmigrated monocytes.
Rap1 activation assays
Rap1 pull-down experiments were performed as described previously [30
]. In brief, following stimulation, U937 cells were lysed in ice-cold lysis buffer containing 10 mM Tris-HCl, 150 mM NaCl, 1% Nonidet-P40, 0.5% deoxycholic acid, 0.1% SDS, 1 mM NaF, 2 mM NaVO3, and protease inhibitor cocktail (Roche) for 10 min on ice. Lysates were clarified by centrifugation at 10,000 g for 10 min at 4°C. GST-Ral guanine nucleotide dissociation stimulator (RalGDS)-raf-1 ras-binding domain (RBD) coupled to glutathione-Sepharose beads (Amersham Biosciences, UK) was added to the supernatants and incubated for 1 h at 4°C. Beads were washed three times in lysis buffer, and bound proteins were eluted with Laemmli sample buffer. Rap1 in total cell lysates and precipitates was detected by Western blotting with a mouse anti-Rap1 mAb (Santa Cruz Biotechnology, CA). Densitometric analysis was performed with a CanoScan LiDE20 scanner (Canon) and Gene Tools Analysis software, Version 3.03.03 (SynGene, Cambridge, UK).
Electroporation and immunofluorescence
A plasmid containing hemagglutinin (HA)-Epac1 (pMT2SM-HA-Epac1, 30 µg) [3
] was added to U937 cells (12x106 cells) resuspended in RPMI medium. Cells were subsequently electroporated with a BioRad Gene Pulser II electroporator (950 µF, 250 V) and cultured in RPMI medium containing 20% FCS for 48 h. Subsequently, cells were collected by centrifugation, washed, and resuspended in IMDM containing 0.25% BSA. Transfected cells were allowed to adhere to fibronectin-coated coverslips for 10 min at 37°C, followed by a 20-min incubation in the presence of stimuli. Cells were then washed with PBS containing 0.5% BSA, fixed with 3.7% formaldehyde for 10 min at room temperature, and permeabilized with 0.1% Triton X-100 for 5 min. For immunofluorescence staining of HA-Epac-1, cells were incubated with a mouse mAb to HA (12CA5, Boehringer Mannheim Corp., Indianapolis, IN), followed by a goat anti-mouse-Ig antibody labeled with Alexa 488 (Molecular Probes). F-actin was visualized with Texas Red-labeled phalloidin (Molecular Probes). Images were recorded with a Zeiss LSM 510 confocal laser-scanning microscope. Fluorescence distribution profiles were created with Zeiss LSM 510 confocal laser-scanning microscope software.
Statistical analysis
All results were expressed as a mean ± SEM of at least three independent experiments. Where applicable, values were compared with paired two-tailed Student’s t-test. Multiple comparisons were analyzed with a two-way ANOVA test. A P value lower than 0.05 was considered significant. All statistical analyses were performed with GraphPad Prism, Version 3.0, software.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
 View larger version (35K): [in a new window] |
Figure 1. Expression of Epac1 in primary leukocytes and leukocytic cell lines. (A) RT-PCR for Epac1 of cDNA derived from purified human primary neutrophils, monocytes, lymphocytes, CD34+ cells, eosinophils, platelets, and the leukemic cell lines U937 and HL60. The size of the amplified PCR fragment was 735 bp. As a control, cDNA for β-glucuronidase was amplified. Representative results of three independent experiments are shown. (B) Western blot detection of Epac1 in cell lysates from purified human primary neutrophils, monocytes, B cells, T cells, CD34+ cells, eosinophils, platelets, and the leukocytic cell lines U937 and HL60. CHO cells were used as a positive control for Epac1 protein expression. Bands corresponding to Epac1 are  110 kDa. β-actin was used as a control for equal protein loading. Representative results of four independent experiments are shown.
|
|
View this table: [in a new window] |
Table 1. CT Values of Epac1 and β-Glucuronidase
|
 View larger version (17K): [in a new window] |
Figure 2. Epac1 activates Rap1 in U937 cells and promotes cell adhesion to fibronectin and endothelial cells. (A) U937 cells were treated with 100 µM 8CPT-2Me-cAMP for the indicated time periods, and GTP-bound Rap1 was precipitated with GST-Ral-GDS-RBD, followed by Western blotting and Rap1 detection by immunoblotting (upper panel). Total levels of Rap1 in whole cell lysates are shown (lower panel). Representative results of three independent experiments are shown. Bar graph represents densitometric analysis of Rap1 activation. Data are mean of three independent experiments (±SEM). (B) U937 cells were pretreated or not with 100 µM 8CPT-2Me-cAMP for 15 min and placed in a fibronectin-coated plate for 30 min. As a positive control, PMA (100 ng/ml) was added at the time of plating. The percentage of adhesion was determined as described in Materials and Methods. Data are mean (±SEM) of five independent experiments performed in triplicate. *, P < 0.05; **, P < 0.005. (C) Primary monocytes were pretreated or not with 100 µM 8CPT-2Me-cAMP for 30 min and perfused over TNF--stimulated monolayers of HUVEC for 5 min at 0.8 dyn/cm2. Video images were evaluated for the number of adherent monocytes and cell rolling as described in Materials and Methods. Data are mean (±SEM) of five independent experiments performed in duplicate. *, P < 0.05. ctrl, Control.
|
To gain insight into the mechanism by which Epac1 enhances U937 cell adhesion to fibronectin, we analyzed the surface expression of total and activated β1- and β2-integrins upon treatment with 8CPT-2Me-cAMP. Epac1 activation resulted in a rapid and significant increase of activated β1-integrins on the cell surface followed by their down-modulation (Fig. 3A ). In contrast, no changes were observed for β2-integrin activation (data not shown). Our results show a rapid but transient activation of integrins by Epac, and the effects on cell adhesion are prolonged for at least 30 min. This may be explained by the fact that integrin activation was measured in cells in suspension, where no integrin engagement takes place. Upon seeding, stable engagement of the activated integrins by fibronectin may induce a prolonged effect on adhesion.
 View larger version (28K): [in a new window] |
Figure 3. Epac1 induces β1-integrin activation and β1-integrin-mediated adhesion. (A) U937 cells were treated with 100 µM 8CPT-2Me-cAMP for the indicated time periods. Surface expression of activated β1-integrins was determined by flow cytometry using a mouse mAb that recognizes the activated state of β1-integrins (12G10). The activation of β1-integrins is expressed as fold increase over control-untreated cells and is corrected for total β1-integrin surface expression. Data are mean (±SEM) of three independent experiments. *, P < 0.05. (B–D) U937 cells were preincubated with integrin-blocking mAb to VLA-5 (B), VLA-4 (C), or membrane-activated complex-1 (Mac-1; D) for 30 min before addition of the cells to fibronectin-coated plates. Data are mean (±SEM) of three independent experiments performed in triplicate. *, P < 0.05.
|
5β1) antibody completely abrogated adhesion of control and 8CPT-2Me-cAMP-treated cells (Fig. 3B)
, indicating that VLA-5 is the main integrin involved in U937 cell attachment to fibronectin. However, an anti-VLA-4 (4β1) antibody specifically reduced an 8CPT-2Me-cAMP-induced increase of adhesion, which returned to control levels (Fig. 3C)
, whereas an anti-Mac1 (mβ2) antibody showed no inhibitory effect (Fig. 3D)
. From these experiments, we conclude that Epac1 activation triggers inside-out signaling, resulting in the activation of β1-integrins, leading to VLA-4- and likely, VLA-5-mediated adhesion of U937 cells to fibronectin.
Epac1 induces cell polarization and localizes to the uropod
Cell polarization plays a crucial role in directional cell movement. Polarized cells develop a leading edge, where membrane extension and lamellipodia formation occur, and a retracting rear (uropod). As Rap1 is proposed to play a central role in lymphocyte polarization [37
], we analyzed whether Epac1 activation triggers polarization of U937 cells. Fibronectin-adherent U937 cells were treated with 8CPT-2Me-cAMP, fixed, and stained for F-actin. The percentage of polarized cells was quantified by microscopy, according to morphological criteria: nonpolarized cells are round in shape, whereas polarized cells have a morphologically defined leading edge and a uropod (cell images in Fig. 4A
). Quantitative analysis indicated that 8CPT-2Me-cAMP induced a twofold increase in the number of polarized cells (Fig. 4A)
. We next investigated the intracellular distribution of Epac1 in polarized versus nonpolarized cells. U937 cells were transfected with HA-tagged Epac1, seeded on fibronectin, and stained with HA antibodies for microscopy analysis. Epac1 localized at the cell periphery in nonpolarized cells, whereas it concentrated at the uropod of polarized cells (Fig. 4B)
. These observations suggest that Epac1 activates Rap1 at a perinuclear location.
 View larger version (21K): [in a new window] |
Figure 4. Epac1 induces cell polarization and redistributes to the uropod of polarized cells. (A) U937 cells were allowed to adhere to fibronectin-coated coverslips and were stimulated or not with 100 µM 8CPT-2Me-cAMP for 20 min, fixed, stained for F-actin, and analyzed by confocal microscopy. Data represent the percentage of polarized cells, scored on the basis of morphology, of a total of 100–120 cells per condition (left panel). Representative images of a polarized and a nonpolarized cell stained for F-actin are shown (right panel, arrow indicates direction of migration). Data are mean (±SEM) of three independent experiments. (B) U937 cells were transiently transfected with HA-tagged Epac1 and allowed to adhere to fibronectin-coated coverslips. Thereafter, U937 cells were stimulated or not with 100 µM 8CPT-2Me-cAMP for 20 min, fixed, and stained for Epac1. Fluorescence intensity profiles along the indicated dashed line between the points a and b are shown. White arrows on images indicate direction of cell polarization. Images are representative of three independent experiments. Original bars, 5 µm. *, P < 0.05.
|
 View larger version (17K): [in a new window] |
Figure 5. Epac1 promotes chemotaxis of U937 cells and primary monocytes. U937 cells (left panel) or primary monocytes (right panel) were allowed to migrate for 1 h to 100 ng/ml CXCL12 (SDF-1) or 10 ng/ml CCL2 (MCP-1), respectively, in the presence or absence of 100 µM 8CPT-2Me-cAMP added to the cell suspension in the upper compartment of a Transwell system. When no chemoattractant was present, 1–2% U937 cells or primary monocytes migrated, regardless the presence of 8CPT-2Me-cAMP. In the presence of chemoattractant, 30–40% of untreated U937 cells and 6–7% of untreated monocytes migrated. Data represent the percentage of cell migration compared with the untreated cells (set at 100%). Data are mean (±SEM) of three to five independent experiments. *, P < 0.05; **, P <0.005.
|
 View larger version (28K): [in a new window] |
Figure 6. Serotonin activates Rap1, induces U937 cell adhesion, and increases CXCL12-induced chemotaxis. (A) U937 cells were allowed to migrate across fibronectin-coated filters toward CXCL12 (10 ng/ml), serotonin (5-HT; 1 µM), or a combination of both for the indicated time periods. The percentage of migration was determined as described in Material and Methods. Data are mean (±SEM) of three independent experiments. The percentage of migration toward CXCL12 was compared with the percentage of migration toward the combination of 5-HT and CXCL12 by a two-way ANOVA test (P<0.05). (B) U937 cells were assayed for adhesion to fibronectin in the presence of CXCL12 (100 ng/ml), 5-HT (10 µM), or a combination of both. The percentage of adhesion was determined as described in Materials and Methods. Data are mean (±SEM) of five independent experiments performed in triplicate. *, P < 0.05; **, P < 0.005. (C) U937 cells were allowed to adhere to fibronectin-coated coverslips and were stimulated or not with 10 µM serotonin for 30 min, fixed, stained for F-actin, and analyzed by confocal microscopy. Data represent the percentage of polarized cells, scored on the basis of morphology, of a total of 100–120 cells per condition. (D) U937 cells were stimulated with 5-HT (10 µM), CXCL12 (100 ng/ml), or a combination of both stimuli for the indicated time periods, and Rap1 GTP-loading was assayed (upper panel). Total levels of Rap1 in whole cell lysates are shown (lower panel). Representative results of three independent experiments are shown. Bar graph represents densitometric analysis of Rap1 activation. Data are mean of three independent experiments (±SEM).
|
To unequivocally implicate Epac in the mediation of serotonin effects on adhesion and chemotaxis, we transfected U937 cells with a dominant-negative Epac mutant. However, no quantitative data could be obtained as a result of low transfection efficiencies. We then followed an indirect approach by demonstrating that serotonin-induced adhesion requires cAMP production but is PKA-independent. U937 cells were treated with two different adenylate cyclase inhibitors, SQ22536 and 2'-5'-deoxyadenosine, to block cAMP generation upon stimulation with serotonin. Treatment with either inhibitor abrogated the differences in adhesion between serotonin-stimulated and unstimulated cells and between CXCL12-stimulated and CXCL12/serotonin-stimulated cells (Fig. 7A ). These data suggest that cAMP production is required for the enhancing effects that serotonin has on adhesion. As cAMP can activate Epac1 and PKA, we excluded the possibility that serotonin-induced Rap1 activation and cell adhesion were mediated by PKA by using the specific PKA inhibitor H89. Pretreatment of cells with H89 did not inhibit serotonin-induced Rap1 activation (Fig. 7B) or prevent the effects of serotonin on cell adhesion (Fig. 7C) . Similar results were obtained with the competitive PKA inhibitor Rp-cAMPS (data not shown). These data indicate that PKA is not involved in serotonin-induced Rap1 activation and adhesion of U937 cells to fibronectin.
 View larger version (19K): [in a new window] |
Figure 7. Serotonin increases adhesion in a cAMP-dependent, PKA-independent manner. (A) U937 cells were preincubated with the adenylate cyclase inhibitors SQ22536 or 2'5'-deoxyadenosime (2'-5'-dd-Ado) for 1 h before addition of the cells to fibronectin-coated plates. The percentage of adhesion was determined as described in Materials and Methods. Data are mean (±SEM) of three independent experiments performed in triplicate. (B) U937 cells were pretreated or not with 10 µM H89 for 30 min and stimulated with 5-HT (10 µM), and Rap1 GTP-loading was assayed (upper panel). Total levels of Rap1 in whole cell lysates are shown (lower panel). Representative results of three independent experiments are shown. Bar graph represents densitometric analysis of Rap1 activation. Data are mean of three independent experiments (±SEM). (C) U937 cells, pretreated or not with H89 (10 µM) for 30 min, were assayed for adhesion to fibronectin in the presence of CXCL12 (100 ng/ml) and 5-HT (10 µM). The percentage of adhesion was determined as described in Materials and Methods. Data are mean (±SEM) of five independent experiments performed in triplicate. *, P < 0.05; ***, P < 0.0005.
|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The small GTPase Rap1 is activated by almost all receptor types and regulates adhesion-related functions such as cell-cell contact and integrin-mediated adhesion [42 , 43 ]. In lymphocytes, Rap1 plays a crucial role in integrin-mediated adhesion, polarization, and transendothelial migration downstream of chemokine receptors [38 , 39 ]. Recently, Epac1 was identified as a Rap1 exchange factor directly activated by cAMP and shown to be involved in integrin-mediated adhesion in ovarian carcinoma cells through Rap1 activation [6 ]. Although Epac1 is expressed in most tissues, previous studies failed to show Epac1 expression in primary leukocytes with the exception of B cells and macrophages [5 ]. Here, we have used optimized RT-PCR conditions to detect low copy number transcripts and found Epac1 mRNA in circulating leukocytes (monocytes, eosinophils, neutrophils, and B and T cells), platelets, CD34-positive hematopoietic cells, and the myelocytic cell lines U937 and HL60. It is important that we found Epac1 protein expression in all leukocytes with the exception of neutrophils. Epac1 protein levels did not always correlate directly with Epac1 mRNA levels, e.g., in eosinophils and platelets. This might be a result of differential, post-translational regulation of Epac1 expression in different types of leukocytes and may have functional consequences during the response of different leukocyte types to cAMP-elevating agents.
In this study, we have investigated the function of the Epac1-Rap1 pathway in the promonocytic cell line U937 and in primary monocytes. We found that a cAMP analog (8CPT-2Me-cAMP), which specifically activates Epac1 and not PKA, was able to induce cell adhesion, polarization, and chemotaxis. In U937 cells, Epac1 activation induced β1-integrin activation and VLA-4-mediated adhesion to fibronectin. Thus, Epac1 appears to have a similar role in leukocytes as in adherent cells, namely, the activation of Rap1 and the consequent "inside-out" signaling toward β1-integrins [6 , 7 ].
In line with previous studies showing that Rap1 activation induces T cell polarization [37 ], we show that activation of the Epac1-Rap1 pathway induces polarization of U937 cells. In addition, ectopically expressed Epac1 redistributes from the cell periphery to the perinuclear area upon cell polarization. This suggests that Epac1 activates Rap1 at a perinuclear location and that activated Rap1 subsequently translocates to the plasma membrane. Accordingly, wild-type (active and inactive) Rap1 was shown to localize to a perinuclear vesicular compartment and to the plasma membrane in T cells, and activated GTP-bound Rap1 was found exclusively at the plasma membrane [44 ].
We have used monocytic U937 cells for most of our studies; however, we have demonstrated that the Epac pathway is functional in primary human monocytes. We show that Epac1 activation up-regulates adhesion of freshly isolated monocytes to endothelial cells under flow as well as monocyte migration toward CCL2 (MCP-1). This chemokine is a potent monocyte chemoattractant, which has a key role in the recruitment of monocytes to atherosclerotic lesions. Based on these data, we postulate that Epac1 activation by cAMP-raising agonists plays a role in the pathophysiology of atherosclerosis. Supporting this, we have shown that Epac1 activation induced β1-integrin-mediated adhesion to fibronectin. β1-integrins mediate the arrest and initial adhesion of monocytes to the endothelium [35 ]. In addition, β1-integrins can bind to the alternatively spliced connecting segment-1 (CS-1) domain of fibronectin, which contributes to monocyte-endothelium interactions [45 , 46 ]. Endothelial β1-integrins and CS-1-containing fibronectin are suggested to play a crucial role in atherogenesis through the recruitment of circulating monocytes [46 47 48 ]. In addition, minimally modified low-density lipoprotein (MM-LDL) induces the deposition of CS-1 on the endothelial surface and the induction of β1-integrin-mediated monocyte binding to this integrin fragment [46 ]. It is interesting that treatment of endothelial cells with MM-LDL was demonstrated to cause a rapid increase in cAMP, which was necessary for the induction of monocyte binding by MM-LDL [49 ]. The same study showed that other cAMP-elevating agents were also inducing monocyte but not neutrophil binding to endothelium. This is interesting, as we show here that neutrophils do not contain Epac1 protein. Together, these data suggest a model in which local concentrations of cAMP-elevating agonists in atherosclerotic lesions induce endothelial activation and monocyte recruitment, which is mediated by Epac1-induced activation of monocyte β1-integrin-mediated adhesion to fibronectin.
Serotonin is a cAMP-elevating agonist secreted by activated platelets and mast cells, and increased plasma levels of serotonin are associated with the pathophysiology of atherosclerosis and asthma [19 , 50 ]. It is interesting that serotonin was recently shown to be a chemotactic factor for eosinophils [18 ] and to modulate cytokine and chemokine release by monocytes [51 ]. We show here that serotonin is able to induce adhesion, polarization, and chemokinesis of U937 cells similarly to Epac activation. Although we could not directly implicate Epac1 in these effects, we show that serotonin-induced adhesion requires cAMP but is PKA-independent, which suggests that Epac1 activation by cAMP mediates serotonin-induced adhesion of monocytes to fibronectin. Accordingly, Epac1 has been previously shown to be activated by serotonin receptors in neuronal cells [15 , 16 , 52 ]. In the concentrations used in our study, serotonin did not show chemotactic properties for U937 cells, similar to the Epac activator 8CPT-2Me-cAMP. However, both agents increased CXCL12-induced chemotaxis, indicating that other signaling pathways engaged by chemokines are likely to be required for cell movement [53 54 55 ]. The enhancing effects of serotonin on migration could be a result of its ability to induce a more sustained Rap1 activation than CXCL12. This may result in the improvement of Rap1-mediated functions such as polarization and adhesion, which are prerequisites for directional migration. In conclusion, our data support a proinflammatory role for serotonin as an enhancer of monocyte adhesion and chemotaxis.
Previous reports have shown that agents that increase cAMP such as forskolin, 3-isobutyl-1-methylxanthine, or PGE2 inhibit chemokine-induced monocyte adhesion and migration [56 , 57 ]. However, other studies demonstrated that urokinase-type plasminogen activator and relaxin stimulate monocyte adhesion and migration through cAMP-dependent pathways [58 , 59 ]. An explanation for these contradictory observations may be the compartmentalization of cAMP signaling in cells [60 ]. Different signaling receptors activate differentially located members of the adenylate cyclase family, and specific phosphodiesterases degrade cAMP to prevent its diffusion. This results in the formation of cAMP "clouds" at discrete sites within the cell, which activate only nearby located effectors. Thus, it may be that cAMP activates PKA or Epac1 more potently, depending on the stimulus, resulting in different outcomes for adhesion and migration. It is interesting that cAMP is known to consistently inhibit neutrophil migration, which may be explained by the fact that they do not express Epac1 protein as shown here.
In conclusion, our work reveals a previously unrecognized cAMP-dependent signaling pathway in monocytes regulating cell adhesion, polarization, and chemotaxis through the activation of Epac1. Finally, our data suggest that cAMP-elevating receptor agonists may regulate inflammatory processes through the activation of Epac1-Rap1 signaling in monocytes.
Received May 23, 2006; revised July 13, 2006; accepted July 14, 2006.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
3β1 integrin but not the 6β4 integrin J. Biol. Chem. 279,44889-44896 Nat. Cell Biol. 5,633-639[CrossRef][Medline]/β2 integrin in Rap1-induced adhesion and migration Mol. Biol. Cell 14,2570-2582This article has been cited by other articles:
 |
|
 |
|
  K. M. Hoque, O. M. Woodward, D. B. van Rossum, N. C. Zachos, L. Chen, G. P.H. Leung, W. B. Guggino, S. E. Guggino, and C.-M. Tse Epac1 mediates protein kinase A-independent mechanism of forskolin-activated intestinal chloride secretion J. Gen. Physiol., January 1, 2010; 135(1): 43 - 58. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. M. Kreckler, E. Gizewski, T. C. Wan, and J. A. Auchampach Adenosine Suppresses Lipopolysaccharide-Induced Tumor Necrosis Factor-{alpha} Production by Murine Macrophages through a Protein Kinase A- and Exchange Protein Activated by cAMP-Independent Signaling Pathway J. Pharmacol. Exp. Ther., December 1, 2009; 331(3): 1051 - 1061. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. Baljinnyam, K. Iwatsubo, R. Kurotani, X. Wang, C. Ulucan, M. Iwatsubo, D. Lagunoff, and Y. Ishikawa Epac increases melanoma cell migration by a heparan sulfate-related mechanism Am J Physiol Cell Physiol, October 1, 2009; 297(4): C802 - C813. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Stefanini, R. C. Roden, and W. Bergmeier CalDAG-GEFI is at the nexus of calcium-dependent platelet activation Blood, September 17, 2009; 114(12): 2506 - 2514. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. M. van Gils, J. J. Zwaginga, and P. L. Hordijk Molecular and functional interactions among monocytes, platelets, and endothelial cells and their relevance for cardiovascular diseases J. Leukoc. Biol., February 1, 2009; 85(2): 195 - 204. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Parkinson, P. Bolourani, D. Traynor, N. L. Aldren, R. R. Kay, G. Weeks, and C. R. L. Thompson Regulation of Rap1 activity is required for differential adhesion, cell-type patterning and morphogenesis in Dictyostelium J. Cell Sci., February 1, 2009; 122(3): 335 - 344. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. Carmona, S. Gottig, A. Orlandi, J. Scheele, T. Bauerle, M. Jugold, F. Kiessling, R. Henschler, A. M. Zeiher, S. Dimmeler, et al. Role of the small GTPase Rap1 for integrin activity regulation in endothelial cells and angiogenesis Blood, January 8, 2009; 113(2): 488 - 497. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. M. Cifuni, D. D. Wagner, and W. Bergmeier CalDAG-GEFI and protein kinase C represent alternative pathways leading to activation of integrin {alpha}IIb{beta}3 in platelets Blood, September 1, 2008; 112(5): 1696 - 1703. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. El Zein, B. Badran, and E. Sariban VIP differentially activates {beta}2 integrins, CR1, and matrix metalloproteinase-9 in human monocytes through cAMP/PKA, EPAC, and PI-3K signaling pathways via VIP receptor type 1 and FPRL1 J. Leukoc. Biol., April 1, 2008; 83(4): 972 - 981. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. Carmona, E. Chavakis, U. Koehl, A. M. Zeiher, and S. Dimmeler Activation of Epac stimulates integrin-dependent homing of progenitor cells Blood, March 1, 2008; 111(5): 2640 - 2646. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. E. Eckert and S. L. Jones Regulation of VASP serine 157 phosphorylation in human neutrophils after stimulation by a chemoattractant J. Leukoc. Biol., November 1, 2007; 82(5): 1311 - 1321. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. J. Lorenowicz, M. Fernandez-Borja, and P. L. Hordijk cAMP Signaling in Leukocyte Transendothelial Migration Arterioscler Thromb Vasc Biol, May 1, 2007; 27(5): 1014 - 1022. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. J. Gerdin and L. E. Eiden Regulation of PC12 Cell Differentiation by cAMP Signaling to ERK Independent of PKA: Do All the Connections Add Up? Sci. Signal., April 17, 2007; 2007(382): pe15 - pe15. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
