Cyclooxygenase-2 Deficiency in Macrophages Leads to Defective p110γ PI3K Signaling and Impairs Cell Adhesion and Migration

Manuel D. Díaz-Muñoz, Inés C. Osma-García, Miguel A. Íñiguez and Manuel Fresno
J Immunol July 1, 2013, 191 (1) 395-406; DOI: https://doi.org/10.4049/jimmunol.1202002

Abstract

Cyclooxygenase (Cox)-2 dependent PGs modulate several functions in many pathophysiological processes, including migration of immune cells. In this study, we addressed the role of Cox-2 in macrophage migration by using in vivo and in vitro models. Upon thioglycolate challenge, CD11b+ F4/80+ macrophages showed a diminished ability to migrate to the peritoneal cavity in cox-2−/− mice. In vivo migration of cox-2−/− macrophages from the peritoneal cavity to lymph nodes, as well as cell adhesion to the mesothelium, was reduced in response to LPS. In vitro migration of cox-2−/− macrophages toward MCP-1, RANTES, MIP-1α, or MIP-1β, as well as cell adhesion to ICAM-1 or fibronectin, was impaired. Defects in cell migration were not due to changes in chemokine receptor expression. Remarkably, cox-2−/− macrophages showed a deficiency in focal adhesion formation, with reduced phosphorylation of paxillin (Tyr188). Interestingly, expression of the p110γ catalytic subunit of PI3K was severely reduced in the absence of Cox-2, leading to defective Akt phosphorylation, as well as cdc42 and Rac-1 activation. Our results indicate that the paxillin/p110γ-PI3K/Cdc42/Rac1 axis is defective in cox-2−/− macrophages, which results in impaired cell adhesion and migration.

Introduction

Leukocyte migration to the site of infection is a well-characterized process required to build the immune response (1, 2). Less is known about the egression of immune cells from tissues. Emigration is a key component of normal physiology required for leukocyte recirculation and immune surveillance (3, 4). Egression of leukocytes, including neutrophils, monocytes, and dendritic cells (DCs), is indispensable during the adaptive immune response to transport and present Ags in secondary immune organs (5–8). Furthermore, clearance of inflammatory conditions, such as atherosclerosis (5, 9) or acute peritonitis, occurs, in part, as a result of macrophage emigration (10, 11). Nonetheless, very little is known about the molecular basis controlling this mechanism.

PGs are a group of bioactive lipid mediators that regulate important responses in many physiological and pathological processes, including inflammation, cancer, angiogenesis, and cardiovascular diseases. PGs have also been implicated in lymphocyte development and function (12–15). They are generated by cyclooxygenases (Cox-1 and Cox-2) that metabolize arachidonic acid into PGH2, a substrate of different PG synthases, to produce prostanoids (PGE2, PGF2α, PGI2, PGD2 and TXA2). Cyclooxygenases are the targets of nonsteroidal anti-inflammatory drugs. Cox-1 is primarily involved in cell homeostasis, whereas Cox-2 is tightly associated with PG production during inflammation and cancer. In macrophages, Cox-2 is upregulated in response to several proinflammatory stimuli, such as LPS, IFN-γ, and TNF-α, among others (16–18).

Multiple lines of evidence support the role of Cox-2 and Cox-2–derived prostanoids as key modulators of the immune response. Cox-2–knockout mice show reduced inflammation, fever, and illness in response to infection by bacteria and virus (19, 20). Cox-2 deficiency enhances antitumor responses by promoting Foxp3 expression and CD4+ CD25+ T regulatory cells (21, 22). PGE2, PGF2α, and PGD2 were reported to modulate cell migration of neutrophils, monocytes, DCs, and T cells. PGD2 is a chemotactic agent for DCs and Th2 cells, whereas PGF2α was shown to induce neutrophil migration (23, 24). PGE2 enhances hematopoietic cell homing and modulates monocyte response to chemokines (25, 26). PGE2 can also favor migration of DCs to lymph nodes, likely by inducing MMP9 expression (27–29).

Expression of chemokine receptors on the cell membrane regulates cell migration. Migration of macrophages is primarily modulated by CCR1, CCR2, CXCR31, and CCR5, whereas DC migration is primarily mediated by CCR7 (30). PGs can regulate chemokine and chemokine receptor expression, although the role of PGs in cell migration might depend on the cell type analyzed. PGE2 is an inducer of CCR7 expression and DC migration in response to CCL19 and CCL21 (31), whereas it downregulates CCR5 (26) in macrophages. Moreover, control of migration of immune cells by PGs is involved in the progression of some diseases. Thus, decreased migration of Langerhans cells in the absence of PGE2 signaling leads to decreased contact hypersensitivity (32).

Integrin interaction with their ligands (i.e., ICAM-1) induces reorganization of the actin cytoskeleton and cell polarization, which are indispensable for cell migration (33). Actin polymerization is modulated by GTPases, including Cdc42, Rho, Ras, and Rac. Deficient activation of GTPases results in defective actin polymerization and failure in cell migration (1, 34, 35). Recently, it was shown that PGE2 modulates podosome stability in DCs (36). PGE2 signals through the EP2/EP4 receptors and switches Cdc42 on and Rac-1 off. This switch promotes podosome dissociation and focal adhesion formation (37). Cdc42, Rac, and Rho are direct targets of PI3K (38, 39). Among the PI3K family of proteins, class I PI3Ks were associated with leukocyte migration (40). In particular, the p110γ catalytic subunit plays an important role in regulating the migration of macrophages (41, 42).

In this study, we analyzed the impact of Cox-2 deficiency on macrophage migration and adhesion. Our results suggest that Cox-2 may play an important role in these processes, likely by modulating p110γ PI3K–mediated cell signaling. These results may help to clarify the role of Cox-2 during the onset and progression of inflammation, especially as a modulator of cell trafficking.

Materials and Methods

Animals and reagents

B6;129S7-Ptgs2tm1Jed/J (cox-2−/−) mice were purchased from The Jackson Laboratory. B6/129S wild type (cox-2+/+) mice were obtained by breeding heterozygote pairs. C57BL/6 mice were purchased from Harlan Laboratories. Thioglycolate peritonitis was induced in 8–12-wk-old mice by inoculation of sterile Brewer’s thioglycolate (1 ml 10% p/v; DIFCO). Peritoneal lavage with ice-cold PBS was carried out 4 d later. The different immune cell populations were analyzed by flow cytometry. Peritoneal exudates were cultured at 37°C and 5% CO2 in RPMI 1640 medium (Invitrogen) supplemented with 5% FCS (BioWhittaker-Lonza) and 100 U/ml penicillin, 100 μg/ml streptomycin, 1000 U/ml gentamicin, 2 mM l-glutamine, and 0.1 mM nonessential amino acids. Nonadherent cells were washed out with PBS. Ninety percent of adherent cells in culture were CD11b+ F4/80+ macrophages.

MCP-1, RANTES, MIP-1α, and MIP-1β chemokines were purchased from R&D Systems. Pan-specific PI3K inhibitor LY294002 and the p110γ-specific inhibitor AS252424 were from Sigma-Aldrich. The selective Cox-2 inhibitors celecoxib and NS398 were from Alexis Biochemicals. LPS from Escherichia coli (026:B6) was purchased from Sigma-Aldrich. The doses of the above inhibitors and chemicals were selected based on previous studies (43) or after dose-response tests. The dose of LY294002 and AS252424 inhibitors was selected based on the literature and after testing cell viability (Supplemental Fig. 1).

FACS analysis

Immune cell populations from the peritoneal cavity of cox-2+/+ or cox-2−/− mice were stained for flow cytometry using mAbs against F4/80, CCR7, CD29, ICAM-1 (CD54), and Gr1 (eBioscience). mAbs against CD45R/B220, CD11a, CD11b, CD11c, NK1.1, CD3, and CD18 were from BD Biosciences. CCR2 and CCR5 polyclonal Abs were a gift of Dr. Matthias Mack (Institute for Technical Microbiology Biotechnology, Mannheim, Germany). F-actin was visualized using phalloidin (Invitrogen). All samples were acquired in a FACSCalibur cytometer (BD Biosciences) and analyzed using FlowJo 4.1 software (TreeStar).

In vivo adhesion and migration assays

Thioglycolate-elicited peritoneal macrophages from cox-2+/+ or cox-2−/− donor mice were stained with a PKH26 Red Fluorescent Cell Linker kit from Sigma-Aldrich. A total of 5 × 106 cells in 1 ml PBS was injected into the peritoneal cavity of C57BL/6 recipient mice that had been inoculated with Brewer’s thioglycolate 4 d earlier. One microgram of LPS (E. coli 026:B6; Sigma-Aldrich) in 300 μl PBS was injected i.p. 30 min later. Control animals were injected only with PBS. Animals were killed after 5 min to analyze macrophage adhesion to the peritoneal membrane or after 4 h to study macrophage migration to inguinal lymph nodes. Peritoneal cavity was washed with ice-cold PBS to determine the number of PKH26+ cells. To study the adhesion of PKH26-stained macrophages in vivo, the peritoneal membrane was fixed with 4% PFA, and images were analyzed. Inguinal lymph nodes were frozen in OCT freezing medium. Five-micron cryosections were collected for fluorescence microscopy. Image acquisition was carried out with a CCD Leica camera using a 40× objective.

In vitro migration assays

Peritoneal macrophages were cultured in the top chamber of 5- or 8-μm transwell plates (Corning) for 3 h using complete RPMI 1640 medium with 5% FCS. Nonadherent cells were removed by washing the chambers with 2% FCS, RPMI 1640 medium. The chemokine was placed in the bottom chamber, and cell migration was analyzed after 4 or 24 h at 37°C. Cells were treated with the pan-specific PI3K inhibitor LY294002 or with the p110γ-specific inhibitor AS252424 for 30 min prior to induction of cell migration by RANTES. Nonmigrated cells in the top chamber were removed by rubbing with a cotton stick. Migrated cells in the bottom of the transwell filter were fixed with 1% paraformaldehyde (PFA) and stained with Violet Crystal (Sigma-Aldrich). Images of eight fields of the transwell filter were taken using a 20× objective to quantify the number of migrated cells. In vitro migration of peritoneal macrophages from individual animals (n > 3 per group) was tested in duplicate. Data are shown as relative migration (number of cells in the presence of chemokine/number of cells in the absence of any chemokine) ± SD.

In vitro cell-adhesion assays

Cell adhesion to culture plates precoated with fibronectin (2 μg/ml; Sigma-Aldrich) and ICAM-1 (1 μg/ml; R&D Systems) was tested after labeling peritoneal macrophages with BCECF AM cell tracker (0.1 μg; Invitrogen) for 30 min at 37°C. Macrophage adhesion was performed at 37° for 20 min. Fluorescence intensity from the total number of cells was measured using FLUOstar Optima (BMG LABTECH). Then, nonadherent cells were removed by washing several times with PBS, and fluorescence of remaining cells (adherent cells) was measured. Cell adhesion to fibronectin or ICAM-1 was calculated as the mean fluorescence of adherent cells divided by the mean fluorescence of total cells and is represented as a percentage.

Fluorescence microscopy

To study the structure of the actin cytoskeleton, peritoneal macrophages were cultured on cover slips for 20 min, at an early stage, or for 18 h to evaluate podosome and focal adhesion formation. Cox-2 enzymatic activity was blocked in these experiments by adding celecoxib 1 h before treating the cells with LPS or PGE2. Then, cells were washed with PBS and fixed with 4% PFA for 15 min. Cover slips were blocked with 1% BSA in PBS and incubated with an anti-vinculin mAb (Sigma-Aldrich) and phalloidin–Alexa Fluor 488 (Invitrogen). Vinculin was detected using a donkey anti-mouse Ab coupled to Alexa Fluor 555 (Invitrogen). Cover slips were washed three times with PBS and then once with H2O and 70% ethanol, dried for 5 min, and mounted using Mowiol (Calbiochem). Images from three independent experiments were captured with a Radiance 2000 Confocal System (Bio-Rad) or a Confocal LSM510 Meta (Zeiss) coupled to an inverted microscope Axiovert S100 TV (Zeiss). Abs against CD11b (biotin conjugated) and focal adhesion kinase (FAK; pTyr397) were purchased from eBioscience and Invitrogen, respectively.

SDS-PAGE and Western blotting

Total protein extracts were obtained, and cell lysates were subjected to Western blot analysis using conventional SDS-PAGE, followed by protein transfer to nitrocellulose membranes, as described (43). Abs against paxillin, pTyr118 Paxillin (BD Bioscience), vinculin (Sigma-Aldrich), FAK (Invitrogen), ERK1/2, p-ERK1/2, p-JNK, JNK, Akt, p-Akt Ser473, PI3K p85, PI3K p101 (Cell Signaling), PI3K p110γ, PI3K p110δ, and β-actin (Santa Cruz Biotechnology) were used.

Rac and Cdc42 activation

Rac- and Cdc42-activation assays were carried out by pull-down assays using a GST fusion protein containing the p21-binding domain (PBD) of p21-activated kinase (PAK) for Rac and Cdc42 (10 μg; Millipore). Briefly, a subconfluent monolayer of peritoneal macrophages was lysed with 1 ml ice-cold cell lysis buffer (25 mM HEPES, 150 mM NaCl, 1% IGEPAL CA-630, 10% glycerol, 25 mM NaF, 10 mM MgCl2, 1 mM EDTA, and protease and phosphatase inhibitor mixture; Roche). Cell lysates were immediately centrifuged at 10,000 × g for 20 min at 4°C. Protein concentration was determined by the bicinchoninic acid method (Thermo Scientific). The same concentration of protein was incubated for 1 h at 4°C with PAK-1 PBD (Upstate) with mild agitation. Beads were washed three times with lysis buffer. Bound proteins were eluted with Laemmli sample buffer and separated by SDS-PAGE. Rac and Cdc42 were detected by Western blot with specific Abs (anti-Cdc42 Ab from Santa Cruz Biotechnology; anti-Rac1 Ab [clone 23A8] from Millipore).

Statistics

Statistical analysis of the experimental data was carried out using GraphPad 5. The Student t test (unpaired) and the Mann–Whitney nonparametric test were performed to compare different assay groups. The statistical test done for each particular experiment is described in the figure legends. The p values <0.05 were considered statistically significant.

Results

Cox-2 deficiency affects cell migration to the peritoneal cavity

Thioglycolate-induced peritonitis was used as experimental model to evaluate the role of Cox-2 expression in cell migration to the peritoneum. Interestingly, we found a significant reduction in the total number of cells migrating to the peritoneal cavity in cox-2−/− mice compared with cox-2+/+ mice (Fig. 1A). FACS analysis of peritoneal exudates showed that CD11b+ CD11c+ F4/80+ macrophages (44) are the main cell population in the peritoneal cavity at day 4 after peritonitis induction with sodium thioglycolate (Fig. 1B). The decrease in the total number of cells in cox-2−/− mice was almost exclusively due to a reduction in the migration of F4/80+ CD11c+ CD11b+ macrophages (∼50% reduction in total cell number, p < 0.05). In contrast, the number of CD3+ cells (T lymphocytes), CD45R/B220+ cells (mainly B lymphocytes), NK1.1+ cells (NK cells), and Gr1+ cells (granulocytes) recruited to the peritoneal cavity was similar in cox-2+/+ and cox-2−/− mice (Fig. 1C). Cell survival was not compromised in the peritoneal cavity of cox-2−/− mice (Supplemental Fig. 1A). Moreover, this effect cannot be ascribed to significant alterations in circulating monocytes in cox-2−/− mice (data not shown).

FIGURE 1.
FIGURE 1.

Macrophage recruitment to the peritoneal cavity is diminished in cox-2−/− mice. (A) Quantification of the total number of cells migrated to the peritoneal cavity of cox-2+/+ or cox-2−/− mice 4 d after i.p. injection of sodium thioglycolate (10% p/v). Data were collected from three independent experiments, in which three mice/group were analyzed. Each symbol represent an individual mouse, and the horizontal lines represent the mean. (B and C) Analysis of cell populations in the peritoneal cavity of cox-2+/+ and cox-2−/− mice by flow cytometry. Percentage of cells (B) or total number of cells recovered (C) is shown as mean ± SEM (n = 9 mice/group from data collected in three independent experiments). Mann–Whitney nonparametric test was performed with data shown in (A) and (C); statistical significance is indicated. CD3, T cells; CD11b, macrophages; CD11c, macrophages/DCs; CD45R/B220, B cells; F4/80 and CD11b, macrophages; Gr-1, granulocytes; NK1.1, NKT cells.

In vitro cell migration of cox-2−/− macrophages in response to chemokines is defective

The reduced number of cox-2−/− cells recruited into the peritoneal cavity may indicate a defect in the migratory capacity of macrophages. Thus, we tested in vitro migration of thioglycolate-elicited peritoneal macrophages from cox-2+/+ or cox-2−/− mice in response to CCR2- and CCR5-dependent chemokines (MCP-1, RANTES, MIP-1α, and MIP-1β). The absence of Cox-2 in peritoneal macrophages significantly reduced in vitro cell migration in response to these chemokines (Fig. 2A), whereas cox-2−/− bone marrow–derived DCs migrated normally in response to MIP-3β, one of the ligands of CCR7 that mediates the migration of DCs (Supplemental Fig. 2). Then, we looked for alterations in the expression of different chemokine receptors by FACS. Elicited peritoneal macrophages express CCR2 and CCR5 but lack cell surface expression of CCR7 and CXCR4, in agreement with previous reports (30, 34) (Fig. 2B, data not shown). We observed similar levels of CCR2 and CCR5 in thioglycolate-elicited peritoneal macrophages from cox-2+/+ and cox-2−/− mice, suggesting that the cell migration defect cannot be ascribed to differences in chemokine receptor expression.

FIGURE 2.
FIGURE 2.

Impaired in vitro migration of Cox-2–deficient macrophages in response to chemokines. (A) Macrophage migration in response to chemokines. Thioglycolate-elicited peritoneal macrophages from cox-2+/+ or cox-2−/− mice were cultured over transwell plates in the presence of the indicated concentrations of MCP-1, RANTES, MIP-1α, or MIP-1β in the bottom chamber. Cell migration was quantified after 4 h and is presented as relative migration (number of cells in the presence of chemokine/number of cells in the absence of any chemokine) ± SD. Data pooled from two independent experiments are shown. Thioglycolate-elicited macrophages from two or three mice were used separately in each independent experiment. *p < 0.05, unpaired t test with a total of four or five mice/group. (B) Cell surface expression of CCR2 and CCR5 chemokine receptors in cox-2+/+ or cox-2−/− macrophages was analyzed by FACS. A control isotype Ab was used as a negative control. Results shown are representative of three independent experiments. (C) The Cox-2–specific inhibitor NS-398 blocks PGE2 production by peritoneal macrophages. NS-398 (0.1 μM) was added to the cells 1 h prior to stimulation with LPS (1 μg/ml). PGE2 production was measured by ELISA 18 h later. (D) Cox-2 enzymatic inhibition impairs macrophage migration in response to RANTES (10 ng/ml). Results shown are representative of three independent experiments. **p < 0.005, unpaired t test.

To determine whether effects on the migration of cox-2−/−-null macrophages are related to the absence of Cox-2 activity, we analyzed the effect of the inhibition of Cox-2 enzymatic activity on cell migration of wild type macrophages. Cox-2 is expressed at low levels on freshly isolated thioglycolate-elicited peritoneal macrophages, and it is rapidly induced after LPS or PGE2 stimulation (43). NS-398 is a Cox-2–specific inhibitor that blocked PGE2 production by macrophages after LPS stimulation at a dose of 0.1 μM (Fig. 1C, Supplemental Fig. 1C). In vitro migration of cox-2+/+ macrophages treated with NS-398 1 h before inducing cell migration in response to RANTES was impaired, suggesting that Cox-2 enzymatic activity and PG synthesis are required for macrophage migration (Fig. 2D).

In vivo macrophage migration is reduced in the absence of Cox-2

Macrophages are able to emigrate from the peritoneal cavity to adjacent lymph nodes in response to LPS (11). To address whether emigration of cox-2−/− peritoneal macrophages to lymph nodes was also affected, we transferred peritoneal-elicited macrophages freshly isolated from cox-2+/+ and cox-2−/− mice into the peritoneal cavity of C57BL/6 mice. To track macrophages after injection, we prestained them using the red fluorescent membrane linker PKH26. As shown in Fig. 3A, the number of PKH26+ cells remaining in the peritoneal cavity of mice 4 h after i.p. injection of LPS was significantly lower than in mice injected with PBS, indicating macrophage emigration. Interestingly, the number of cox-2−/− macrophages (PHK26+ cells) remaining in the peritoneal cavity of C57/BL6 mice after LPS challenge was significantly higher than in those who received cox-2+/+ macrophages, suggesting a reduced ability of cox-2−/− macrophages to migrate out of the peritoneum. Cell survival was not compromised in the peritoneal cavity of these mice after LPS treatment (Supplemental Fig. 1B). To corroborate that emigration of Cox-2–deficient macrophages was affected, we isolated inguinal lymph nodes from those animals. We detected PKH26+ cells in inguinal lymph nodes that were F4/80+ CD11b+ CD11c+ macrophages (Supplemental Fig. 3). Detection of PKH26+ macrophages by fluorescent microscopy showed a significant reduction in the number of cox-2−/− macrophages in lymph nodes compared with the number of cox-2+/+ macrophages (Fig. 3B). Taken together, these results suggest that Cox-2 is important for in vivo migration of peritoneal macrophages to lymph nodes in response to LPS.

FIGURE 3.
FIGURE 3.

In vivo migration and adhesion of macrophages to the peritoneal membrane are reduced in the absence of Cox-2. Peritoneal macrophages from cox-2+/+ or cox-2−/− mice were isolated after 4 d of thioglycolate injection. Cells were labeled with PKH26 before transfer into the peritoneal cavity of C57BL/6 mice to track cell emigration from the peritoneal cavity to lymph nodes. (A) Analysis of peritoneal exudates from C57BL/6 mice after LPS challenge. Cox-2+/+ and cox-2−/− macrophages were quantified by FACS 4 h after LPS injection (1 μg in PBS, i.p.). Data from two independent experiments (n = 2–3 mice/group) were pooled and are shown as percentage (± SD) of PKH26+ cells (n = 5–6 mice/group). *p = 0.0147, unpaired t test. (B) Fluorescence detection of PKH26+ macrophages in inguinal lymph nodes. The number of PKH26+ macrophages in 5-μm cryosections was quantified. Four fields (40× objective lens) were analyzed per tissue section. Data are shown as mean ± SEM (n = 5–6/group). *p = 0.0484, unpaired t test. (C) FACS analysis of peritoneal macrophages (CD11b+/ PKH26+) from cox-2+/+ or cox-2−/− mice that remained in peritoneal exudates of C57BL/6 mice. Cox-2+/+ and cox-2−/− macrophages were stained with PKH26 prior to cell transfer. Cell adhesion was induced by i.p. injection of LPS or PBS (negative control) 5 min before cell recovery. Bar graph shows the mean percentage (± SEM) of PKH26+ cells that remained in the peritoneal exudates. Data are from three independent experiments (n = 12–13 mice/group). *p = 0.0056, unpaired t test. (D) Immunofluorescence detection of PKH26+ cells attached to the peritoneal membrane. Four fields (40× objective lens) were analyzed for each tissue sample to quantify the number of PKH26+ cells. Data from three independent experiments are shown as mean ± SEM (n = 12 mice/group). *p = 0.0006, unpaired t test.

Cox-2 expression is required for macrophage adhesion to the mesothelium

Cell migration is closely linked to cell adhesion. Thus, we tested the effect of Cox-2 deficiency on macrophage adhesion to the peritoneal membrane in our in vivo experimental model. We transferred PKH26-labeled cox-2+/+ and cox-2−/− thioglycolate-elicited peritoneal macrophages into C57/BL6 mice before injecting them with LPS. After 5 min, the peritoneal cavity was washed with ice-cold PBS. LPS treatment allows macrophage adhesion to the mesothelium, as shown by the significant reduction in the number of cox-2+/+ macrophages (PKH26+ cells) recovered in the peritoneal exudates of LPS-treated mice compared with PBS-injected control mice. Interestingly, this reduction was not observed in mice transferred with cox-2−/− macrophages (Fig. 3C). Next, we analyzed cell adhesion to the peritoneal mesothelium by fluorescent microscopy. Quantification of PKH26+ macrophages from several fields of view showed increased adhesion of cox-2+/+ macrophages to the peritoneal membrane upon LPS injection. In contrast, cox-2−/− macrophages failed to attach to the mesothelium, remaining instead in the peritoneal cavity (Fig. 3D). Thus, a reduced ability of cells to adhere to the mesothelium would explain the decrease in migration of cox-2−/− macrophages in response to LPS.

Macrophage adhesion is reduced in the absence of Cox-2

To study in vitro changes in cell adhesion of thioglycolate-elicited peritoneal macrophages from cox-2−/− mice, we performed cell-adhesion assays in plates coated with ICAM-1 or fibronectin. Compared with cox-2+/+ macrophages, there was a significant decrease in the number of cox-2−/− macrophages that adhered to either ICAM-1 or fibronectin (Fig. 4A, 4B). Furthermore, cell adhesion to fibronectin was significantly reduced when cox-2+/+ macrophages were pretreated with a specific Cox-2 inhibitor (Celecoxib), showing that Cox-2–dependent PG production plays an important role in cell adhesion (Fig. 4C).

FIGURE 4.
FIGURE 4.

Cell surface expression of integrins is normal in Cox-2–deficient macrophages. Adhesion of thioglycolate-elicited peritoneal macrophages from cox-2+/+ and cox-2−/− mice to plates coated with ICAM-1 (A) or fibronectin (B). *p < 0.05, **p < 0.005, unpaired t test. (C) Effect of Cox-2 inhibition on cell adhesion to fibronectin. Macrophages were pretreated for 1 h with a specific inhibitor of Cox-2 (celecoxib, 100 nM). *p < 0.05. (D) Cell surface expression of F4/80, CD11a, CD11b, CD18, CD29, and ICAM-1 in peritoneal macrophages from cox-2+/+ and cox-2−/− mice. Black graphs represent data from cox-2+/+ macrophages. Gray line represents data from cox-2−/− macrophages. The isotype Ab control is shown as a gray graph. Bar plots show mean fluorescence intensity (MFI) ± SEM. All results are representative of three independent experiments.

β1 and β2 integrins are essential for firm adhesion of macrophages to the mesothelium (45). Thus, we analyzed cell surface expression of several integrins in thioglycolate-elicited macrophages from cox-2+/+ and cox-2−/− mice. We did not find any differences in the cell surface expression of integrins CD11a, CD11b, CD18, or CD29 or the adhesion molecule ICAM-1 between cox-2+/+ and cox-2−/− cells (Fig. 4D).

Cox-2–deficient macrophages show an altered actin cytoskeleton organization

Upon integrin-mediated anchoring, firm cell adhesion requires polarization of the actin cytoskeleton, which is key for cell migration (33). Actin cytoskeleton is organized as filaments generating two well-distinguished structures: podosomes and focal adhesions. Podosomes are involved in cell diapedesis, whereas focal adhesions are related more to cell anchoring and interstitial migration (46). To understand the role of Cox-2 in actin cytoskeleton reorganization, we performed experiments with thioglycolate-elicited peritoneal macrophages from cox-2+/+ and cox-2−/− mice. We examined actin cytoskeleton remodeling during short-term (20 min) and long-term (18 h) cell adhesion by staining actin filaments with phalloidin conjugated to Alexa Fluor 488. Detection of vinculin by immunofluorescence, one of several proteins involved in anchoring F-actin to the membrane, was performed in parallel to further characterize podosomes and focal adhesions.

Cox-2+/+ macrophages cultured on fibronectin-coated coverslips for 20 min showed a peripheral distribution of F-actin associated with vinculin. On the contrary, cox-2−/− peritoneal macrophages failed to reorganize the actin cytoskeleton. Neither podosomes nor focal adhesions were clearly seen at this early stage of cell adhesion (Fig. 5A). Further study of some of the proteins involved in focal adhesion, including vinculin, paxillin, FAK, and actin, showed no difference in the total protein expression levels between cox-2+/+ and cox-2−/− peritoneal macrophages (Fig. 5B). F-actin levels were equivalent in both types of macrophages, indicating that altered reorganization of the cytoskeleton was not due to changes in actin expression or polymerization at this early stage of cell adhesion (Fig. 5C). Tyrosine phosphorylation of paxillin is involved in the regulation of dynamics of actin cytoskeletal organization in motile cells (47). Analysis of the levels of phospho-paxillin (Tyr118) in cox-2+/+ and cox-2−/− macrophages by Western blot showed decreased levels of phospho-paxillin in Cox-2–deficient cells (Fig. 5D). This result prompted us to study cell adhesion of cox-2+/+ and cox-2−/− peritoneal macrophages after long-term cell culture.

FIGURE 5.
FIGURE 5.

Cox-2−/− macrophages show abnormal cytoskeleton remodeling at early stages of cell adhesion. (A) Visualization of actin cytoskeleton by confocal fluorescent microscopy (60× objective lens) at early stages of cell adhesion. Adhesion of thioglycolate-elicited peritoneal macrophages from cox-2+/+ or cox-2−/− mice was performed at 37°C for 20 min. Actin cytoskeleton remodeling was analyzed using phalloidin coupled to Alexa Fluor 488 and a monoclonal anti-vinculin Ab (secondary Ab coupled to Alexa Fluor 555). Representative images from three biological replicates/genotype are shown. (B) Analysis of paxillin, vinculin, FAK, and actin protein levels by Western blot. Cell extracts were isolated from thioglycolate-elicited peritoneal macrophages from cox-2+/+ or cox-2−/− mice. Quantification of protein levels relative to actin is indicated below each Western blot panel. (C) FACS analysis of actin polymerization. F-actin was stained using phalloidin. A representative plot of three biological replicates/genotype is shown. (D) Western blot analysis of the levels of phospho-paxillin (Tyr118) in cox-2+/+ or cox-2−/− macrophages. Cells were adhered to plates for 90 min before preparation of cell protein lysates. The amount of phospho-paxillin (Tyr118) was quantified relative to the total amount of paxillin and is indicated below each Western blot panel. A representative experiment of three is shown.

After 18 h of in vitro culture of thioglycolate-elicited peritoneal macrophages on coverslips coated with fibronectin, podosomes were detected as a dense core of actin surrounded by vinculin and integrins distributed in a ring shape (Fig. 6A, Supplemental Fig. 4). The number of cox-2−/− macrophages containing podosomes was significantly diminished compared with cox-2+/+ macrophages. Inhibition of Cox-2 enzymatic activity on cox-2+/+ macrophages by celecoxib showed a similar reduction in the number of cells containing podosomes. Cox-2−/− macrophages treated with celecoxib showed no changes in the number of cells with podosomes compared with nontreated cox-2−/− cells (Fig. 6B).

FIGURE 6.
FIGURE 6.

Formation of podosomes during long-term cell adhesion is reduced in the absence of Cox-2 enzymatic activity. (A) Visualization of actin cytoskeleton by confocal fluorescent microscopy (60× objective lens) at late stages of cell adhesion. Macrophage adhesion to fibronectin was performed at 37°C for 18 h. Where indicated, cox-2+/+ and cox-2−/− macrophages were treated with celecoxib (100 nM) 1 h earlier. Actin cytoskeleton was stained using phalloidin coupled to Alexa Fluor 488. Podosomes and focal adhesions were identified with a monoclonal anti-vinculin Ab. A donkey anti-mouse Ab coupled to Alexa Fluor 555 was used for detection. Representative images are shown. Podosomes are denoted by arrows. (B) Genetic or chemical inhibition of Cox-2 activity decreases the number of cells with podosomes. (C) Quantification of the number of cells with focal adhesions. Each symbol indicates the percentage of cells with podosomes (B) or focal adhesions (C) in one field of view. Two independent immunofluorescence stainings were analyzed per condition for each of the at least three biological replicates/genotype. The unpaired t test was used to compare two groups.

In contrast, focal adhesions were detected as dot-like accumulations of actin that are often found at the end of actin stress fibers. These focal adhesions also contained vinculin and phospho-FAK (pTyr397), as reported previously (48) (Supplemental Fig. 4). Comparison of cox-2+/+ and cox-2−/− macrophages showed no difference in the number of cells that had any focal adhesion. Furthermore, macrophage treatment with celecoxib had no effect on focal adhesion formation (Fig. 6C).

PGE2 modulates actin cytoskeleton organization and restores macrophage adhesion in the absence of Cox-2

PGE2 produced by DCs in response to LPS and TNF-α modulates actin cytoskeleton organization. PGE2 dissembles podosomes, promoting migration of DCs (36). Less clear is the role of PGE2 in focal adhesion formation. Different reports suggest that PGE2 can either promote or inhibit focal adhesion formation, depending on the cell type (37, 49). To better understand the role of PGE2 in the organization of actin cytoskeleton in macrophages, we examined cells, using confocal microscopy, that were treated or not with PGE2 (2 μM) for 18 h. Visualization of actin filaments (stained with phalloidin) and vinculin showed a significant reduction in the number of cells containing podosomes after treatment with PGE2 (Fig. 7A, 7B). On the contrary, a higher number of cells containing actin stress fibers and focal adhesions was observed in macrophages treated with PGE2 compared with nontreated cells (Fig. 7A, 7C). Previously, we showed that inhibition of Cox-2 enzymatic activity with celecoxib diminished macrophage adhesion to ICAM-1. To test whether exogenous PGE2 rescues cell adhesion of thioglycolate-elicited cox-2−/− macrophages, we treated them with 2 μM PGE2 1 h before assaying macrophage adhesion to plates coated with ICAM-1. PGE2 treatment of cox-2+/+ macrophages significantly increased cell adhesion to ICAM-1. Adhesion of cox-2−/− macrophages to ICAM-1 was significantly reduced in the absence of exogenous PGE2, as expected. In contrast, the treatment of cox-2−/− macrophages with PGE2 partially recovered cell adhesion to ICAM-1 (Fig. 7D).

FIGURE 7.
FIGURE 7.

PGE2 promotes macrophage adhesion. (A) Analysis of podosomes and focal adhesion formation after treatment with PGE2. PGE2 (2 μM) was added before triggering cell adhesion of thioglycolate-elicited peritoneal macrophages to fibronectin for 18 h. Images were acquired using 60× objective lens. Arrows denote podosomes or focal adhesions (green = phalloidin, red = vinculin). (B) PGE2 decreases the number of cells with podosomes. (C) PGE2 increases focal adhesion formation. Quantification of the percentage of cells with podosomes (B) and focal adhesions (C) was done by analyzing several fields of view from two independent experiments with cells from three individual mice stained independently. Each symbol represents the percentage of cells in one field of view. All data were pooled, and the unpaired t test was used to compare the two groups. (D) Exogenous PGE2 rescues cell adhesion of cox-2−/− macrophages. Thioglycolate-elicited peritoneal macrophages from cox-2+/+ and cox-2−/− mice were treated or not with PGE2 1 h before assaying macrophage adhesion to plates coated with ICAM-1. Cells from four biological replicates were assayed per genotype in triplicate. The unpaired t test was used to compare the different groups, as indicated.

Next, we tested whether LPS affected focal adhesion formation in macrophages. Production of endogenous PGE2 by macrophages is Cox-2 dependent and increases after LPS activation (Fig. 2C, Supplemental Fig. 1). Analysis of actin and vinculin distribution in thioglycolate-elicited macrophages adhered to fibronectin after LPS stimulation showed a significant increase in the formation of actin stress fibers (Fig. 8A). The number of macrophages in which focal adhesions were detected increased significantly after LPS activation (Fig. 8B). Inhibition of Cox-2 enzymatic activity by celecoxib did not change the number of macrophages in which focal adhesions were detected compared with macrophages that were not activated by LPS. On the contrary, celecoxib blocked focal adhesion formation in response to LPS. Finally, we evaluated cell adhesion of thioglycolate-elicited macrophages to fibronectin (Fig. 8C). Celecoxib significantly decreased macrophage adhesion to fibronectin, whereas macrophage stimulation with LPS increased it. Cox-2 inhibition by celecoxib significantly reduced the adhesion capacity of LPS-activated macrophages, which was now comparable to the cell adhesion capacity of nontreated macrophages.

FIGURE 8.
FIGURE 8.

PGE2 production by macrophages after LPS stimulation mediates focal adhesion formation. (A) Analysis of focal adhesions by confocal microscopy in thioglycolate-elicited macrophages treated with LPS (green = phalloidin, red = vinculin). Celecoxib (100 nM) was added, where indicated, to block Cox-2 enzymatic activity. Macrophage adhesion to fibronectin was assayed after 18 h. Images were acquired using 60× objective lens. (B) Quantification of the percentage of cells with focal adhesions. Each symbol represents the percentage of cells in one field of view. Data are from two independent experiments. The unpaired t test was used to compare the different groups, as indicated. (C) Analysis of macrophage adhesion to fibronectin. Macrophages were treated with celecoxib and/or LPS, as indicated. Cells from three independent experiments were assayed in triplicate. All data were pooled together, and the unpaired t test was used to compare the different groups.

Altogether, the data showed that Cox-2 enzymatic activity and endogenous production of PGE2 were required for actin cytoskeleton organization and macrophage adhesion.

Cox-2 expression is required for PI3K-dependent cell signaling

Macrophage activation through TLRs is intimately related to cell adhesion. LPS stimulation of monocytes and macrophages activates PI3K, which regulates tyrosine phosphorylation of paxillin (50, 51). Lower levels of phospho-paxillin in cox-2−/− macrophages suggested that cell signaling activation may be altered in the absence of Cox-2. β2 integrins are known to activate multiple signaling cascades required for macrophage adhesion and migration, including ERK1/2, JNK, and PI3K (30, 34). We next analyzed the status of some key molecules after adhesion of thioglycolate-elicited macrophages. Cell adhesion for 90 min induced phosphorylation of JNK, ERK1/2, and Akt. The levels of p-ERK1/2 and p-JNK were similar in Cox-2–deficient macrophages compared with wild type ones. Interestingly, lower levels of p-Akt were observed in cox-2−/− peritoneal macrophages (Fig. 9A). The fact that p-Akt levels were reduced suggested a defect in PI3K-mediated signaling. Thus, we analyzed the levels of the PI3K regulatory subunits p85 and p101, along with the PI3K catalytic subunits p110γ and p110δ, which were described previously as being key molecules for leukocyte migration (40, 41). p85, p101, and p110δ proteins were expressed at similar levels in cox-2+/+ and cox-2−/− macrophages. Notably, p110γ protein expression was significantly reduced in cox-2−/− peritoneal macrophages (Fig. 9B). Treatment of thioglycolate-elicited macrophages with PGE2 for 90 min increased p110γ protein expression (Fig. 9C). Short-term inhibition of Cox-2 enzymatic activity by celecoxib decreased p110γ protein levels in macrophages, which were restored by adding PGE2 (Fig. 9D).

FIGURE 9.
FIGURE 9.

Cox-2 deficiency impairs PI3K-dependent signaling. (A) Protein phosphorylation of ERK1/2, JNK, and Akt in cox-2+/+ or cox-2−/− macrophages. Cells were adhered to cell culture plates for 90 min before preparation of cell protein lysates. Protein levels of total or phosphorylated proteins were detected with specific Abs. Protein quantification of each phospho isoform is shown as a number below each panel. Phospho protein abundance was quantified after correcting against the total amount of each protein. Then, it was calculated relative to the amount in wild type cells. (B) Protein expression of p110γ is reduced in cox-2−/− macrophages. Protein levels of PI3K p85, p101, p110δ, and p110γ were analyzed by Western blot. Relative quantification of p110γ protein in cox-2−/− macrophages compared with wild type is shown below the panel. (C) PGE2 increases p110γ protein expression. Thioglycolate-elicited macrophages were treated with PGE2 (100 nM) during cell adhesion to ICAM-1. Cell lysates were prepared after 90 min. Hsp90 was used as a loading control and for relative protein quantification (number is shown below the panel). (D) Inhibition of Cox-2 enzymatic activity decreases p110γ protein expression in macrophages. Thioglycolate-elicited macrophages were incubated with celecoxib for 1 h before setting up a cell adhesion assay to ICAM-1. Exogenous PGE2 was added right before setting up a cell adhesion to ICAM-1. Cell lysates were prepared after 90 min. Hsp90 was used as a loading control and for relative protein quantification (number is shown below the panel). (E and F) Rac-1 and Cdc42 activation is impaired in cox-2−/− macrophages. Rac and Cdc42 proteins were pulled down using PAK-1 PBD. The presence of the activated GTP-Rac and GTP-Cdc42 was analyzed by Western blot. Actin was used as a loading control and for relative protein quantification (number is shown below each panel). (G) The nonspecific PI3K inhibitor LY294002 and the p110γ-selective inhibitor AS252424 reduce macrophage adhesion to ICAM-1. (H) Migration of macrophages is regulated by p110γ. Wild type macrophages were incubated with 2 μM of LY294002 or AS252424 for 30 min prior to induction of cell migration toward RANTES (50 ng/ml). At least three independent experiments were performed for each experiment. *p < 0.05 versus cells not treated with inhibitor, **p < 0.005 versus migrated cells to RANTES not treated with inhibitors, unpaired t test.

To determine whether lower levels of p-Akt in cox-2−/− macrophages resulted in deficient signaling downstream of this kinase, we studied the activation of GTPases involved in cell adhesion and migration. Rac-1 and cdc42 were pulled down, after cell adhesion mediated by ICAM-1, using the PBD of PAK. The presence of active Rac-1–GTP and Cdc42-GTP was severely reduced in cox-2−/− macrophages compared with cox-2+/+ macrophages (Fig. 9E, 9F).

PI3K signaling was proved to be fundamental for cell migration. Class I PI3K p110γ and p110δ are key for regulating the migration of immune cells. In particular, p110γ−/− mice have a defective accumulation of macrophages in a septic peritonitis model due to reduced macrophage migration toward several chemokines, although less is known about its affect on cell adhesion (41). We performed cell-adhesion assays to ICAM-1 in the presence of a nonspecific PI3K inhibitor, LY294002, or a p110γ-specific inhibitor, AS252424. Pretreatment of peritoneal macrophages with these inhibitors diminished cell adhesion to ICAM-1 (Fig. 9G) without affecting cell survival (Supplemental Fig. 1D). Furthermore, PI3K and p110γ inhibitors severely decreased macrophage migration in response to RANTES, thus highlighting the importance of PI3K p110γ catalytic activity in macrophage cell adhesion and migration (Fig. 9H).

Discussion

Migration of leukocytes to the inflamed tissues is a hallmark of inflammation (2). Previous studies reported that Cox-derived PGs can modulate leukocyte migration to the inflamed tissue (52). Mice deficient in Cox-2 or mPGES-1, a downstream PG synthase, have a reduced recruitment of leukocytes to the inflammatory foci, as shown in different mouse models, including thioglycolate-induced peritonitis (53, 54). Nevertheless, the underlying mechanisms remain unclear. Our results point to the importance of Cox-2 in macrophage adhesion and migration, which are markedly impaired in cox-2−/− mice. Moreover, analysis of the signaling pathways involved in these effects indicates that Cox-2 is required to maintain normal protein levels of p110γ in these cells. The paxillin/p110γ-PI3K/Cdc42/Rac1 axis is defective in cox-2−/− macrophages, resulting in altered cell adhesion and migration.

In recent years, the view of macrophages merely as phagocytes has been expanded with the description of new macrophage populations with differentiated effector functions, including them as APCs to activate the adaptive immune response (55, 56). New data show that monocyte–macrophage turnover involves cell migration in and out of the inflammatory focus with important physiological consequences (57, 58). However, macrophage emigration from the inflammatory focus is not completely understood.

We found that, in the thioglycolate-elicited peritonitis model, Cox-2 deficiency leads to a significant selective reduction in the migration of CD11b+ F4/80+ CD11c+ myeloid cells to the peritoneal cavity, in line with previous data obtained from mPGES1-null mice (53, 54). More than 90% of CD11b+ cells were positive for the macrophage-specific Ag F4/80, which is expressed by mature macrophages (44), indicating a defect in the migration of those cells to the peritoneum. Using a model of thioglycolate-elicited peritonitis to study macrophage emigration described previously (10, 11), we show in this study that macrophage emigration to lymph nodes is reduced after LPS challenge in the absence of Cox-2. Cox-2−/− macrophages were defective with regard to migration toward CCR2- and CCR5-binding chemokines, as were cox-2+/+ macrophages treated with Cox-2–specific inhibitors, whereas cox-2−/− bone marrow–derived DCs migrated normally in response to MIP-3β (a ligand of the CCR7 receptor). The migration defect was not due to alterations in CCR2 and CCR5 surface expression on cox-2−/− macrophages. Nevertheless, taking into account the results showing altered signaling in cox-2−/− macrophages, it must be considered that chemokine receptor signaling could be also impaired.

Impaired macrophage migration can be explained by the significant reduction in cell adhesion to the mesothelium (in vivo assays) or to ICAM-1 or fibronectin (in vitro assays). Integrins are known to activate multiple signaling cascades required for macrophage migration (2, 30, 34). The “integrin adhesome” involves >100 proteins linked together by >500 interactions (59). Recently, we found that integrin-mediated signaling induces Cox-2 upregulation, which is associated with FAK activation in mesangial cells (60). In this study, we found that cox-2−/− macrophages have a defect in cell adhesion that is due to defective actin cytoskeleton reorganization. A reduced number of cox-2−/− macrophages present podosomes after adhesion. Podosomes were described as short-lived adhesion organelles with a core of actin surrounded by adhesion molecules, including integrins and metalloproteinases, which are able to digest extracellular matrix and, thus, are important for interstitial migration (61). F-actin polymerization was not affected in cox-2−/− macrophages, suggesting that these cells may have a defect in cell-signaling transduction. Activation of FAK and Rho-family GTPases is required for cytoskeleton remodeling and cell adhesion (62). FAK phosphorylates downstream-signaling molecules, such as paxillin (Tyr118), which is required for migration of cancer cells (63). The role of Cox-2–derived PGs in FAK activation and cell migration was suggested previously. HCT-116 cells overexpressing Cox-2 show an enhanced phosphorylation of FAK, resulting in elevated kinase activity (64). Moreover, PGE2 induces FAK and paxillin phosphorylation and subsequent migration in cancer cells (65). P-paxillin regulates adhesion dynamics associated with focal adhesions (47). We show in this study that exogenous PGE2 increases focal adhesion formation in macrophages. LPS induces Cox-2 expression and PGE2 synthesis in macrophages, which are also associated with actin cytoskeleton remodeling. LPS dissociates podosomes in DCs, as shown previously (36). We observed a similar effect on focal adhesion assembly in macrophages. This effect is Cox-2 dependent, and it can be blocked by pretreating macrophages with celecoxib.

Phosphorylation of paxillin on Tyr118 is reduced significantly in cox-2−/− macrophages. Although some reports showed the ability of different PGs to modulate cell adhesion, the molecular mechanism is not well understood. A remarkable finding of our study is the link established between Cox-2 expression and PI3K signaling. PI3K inhibition blocks tyrosine phosphorylation of Pyk2 and paxillin, affects actin cytoskeleton remodeling, and decreases monocyte and macrophage adhesion (50, 51). Analysis of some of the main cell-signaling cascades (ERK1/2, JNK, and PI3K) in cox-2−/− macrophages after cell adhesion showed that phosphorylation of Akt is reduced in the absence of Cox-2. The role of PI3K in cell adhesion and migration has been studied extensively in the last few decades. PI3K activation is required for macrophage spreading and migration (66). PI3K is activated through chemokine receptors, as well as LPS and PGE2 (67, 68). The PI3K family of proteins is classified into three groups: class I, class II, and class III. Class I PI3Ks are subdivided into two groups: catalytic and regulatory PI3Ks, with four and six isoforms, respectively. Class I PI3K p110γ and p110δ have been described as key for regulating the migration of leukocytes (40). Cox-2−/− macrophages synthesize lower levels of p110γ compared with wild type macrophages. Interestingly, treatment of macrophages with exogenous PGE2 for a short period of time increases p110γ protein levels, whereas inhibition of Cox-2 enzymatic activity by celecoxib reduces p110γ expression. p110γ protein levels are restored by exogenous PGE2.

p110γ−/− macrophages show a reduced capacity to migrate toward several chemokines (41, 42), an observation similar to what we described in cox-2−/− macrophages. Reduced expression of p110γ and reduced phosphorylation of Akt likely leads to the observed impaired activation of Rac and Cdc42 in cox-2−/− macrophages upon adhesion. PI3K regulates the activation of Rac and Cdc42 (69-71). The Rho family of GTPases transduces signals generated from chemokine receptors and adhesion molecules, serving as cross-talk points between different signaling pathways involved in migration (72). Our results point out that Cox-2 expression is required to maintain the PI3K–GTPase axis for cell adhesion and migration. PI3K isoforms have redundant functions in leukocytes, but Cox-2–deficient macrophages have a selective defective expression of p110γ but not p110δ. The differential expression of the PI3K isoforms was shown to be important in a cell- and time-dependent manner, being a tightly regulated process (40, 73). Our data reinforce this idea. PI3K isoforms can have unique functions, and the relevance of Cox-2 in regulating cell migration by selectively affecting p110γ could be cell dependent.

Interestingly, we also found that PGE2 increases the adhesion of peritoneal cox-2+/+ macrophages and restores the cell adhesion defect of cox-2−/− macrophages. This is associated with a change in cell morphology and a normalized actin cytoskeleton organization. Our hypothesis is that autocrine PGE2 secretion by activated macrophages is required to maintain macrophages in a promigratory state by maintaining high levels of p110γ, which is required for adhesion/migration. How this takes place is currently under study in our laboratory. Previous results from our laboratory indicate that macrophages’ autocrine PGE2 secretion is likely required to maintain gene expression of cytokines, Cox-2, and mPGES-1 (74). This might be equally true for signaling kinases like p110γ. EP4 deficiency decreases macrophage survival, compromising PI3K/Akt and NF-κB pathways (75). These pathways are key in the development of the inflammatory response in early atherosclerosis; thus, EP4 deficiency significantly reduced the early development of atheroma plaques in the aortic valves. Missing Cox-2 during any stage of macrophage activation will affect the final outcome of the inflammatory response. Low production of PGE2 in Cox-2−/− macrophages, in response to an inflammatory stimulus, results in less mature/active macrophages with a differential ability to migrate. This hypothesis is in line with the proposed role for PGE2 in the migration of monocytes as a booster of chemokine-induced migration (26).

In summary, our results suggest that the paxillin/p110γ PI3K/Cdc42/Rac1 axis is defective in cox-2−/− macrophages. Likely, this leads to defective actin polymerization and focal adhesion formation in those cells, as well as impaired cell adhesion and migration. Experiments in progress will help to identify which PGs produced by Cox-2 are required for the observed effect on macrophage migration, although previous in vitro studies suggested that Cox-2–mediated production of both PGE2 and PGD2 is responsible for macrophage migration in response to LPS (76). Together, our results may be important to further understand the role of Cox-2 in inflammatory diseases, especially those in which an essential role for macrophage emigration is emerging, such as atherosclerosis (57, 58, 77, 78).

Disclosures

The authors have no financial conflicts of interest.

Acknowledgments

We thank B. Barrocal, H. Salgado, M. Chorro, and M. Maza for excellent technical assistance.

Footnotes

  • This work was supported in part by grants from Comunidad Autónoma de Madrid (S2010/BMD-2332), the Cardiovascular Red Temática de Investigación en Enfermedades Cardiovasculares and Red de Investigación Cooperativa en Enfermedades Tropicales Networks of the Instituto de Salud Carlos III (RD06/0014/1013 and RD06/0021/0016), the European Union (EICOSANOX LSH-CT-2004-005033), and Ministerio de Ciencia e Innovación (SAF2007-61716 and SAF2010-18733 to M.F. and BFU2010-21055 and SAF2011-23971 to M.A.I.). I.C.O.-G. holds a predoctoral fellowship from Fondo de Investigación Sanitaria. Centro de Biología Molecular Severo Ochoa receives institutional funding from the Fundación Ramón Areces.

  • The online version of this article contains supplemental material.

  • Abbreviations used in this article:

    Cox
    cyclooxygenase
    DC
    dendritic cell
    FAK
    focal adhesion kinase
    PAK
    p21-activated kinase
    PBD
    p21-binding domain
    PFA
    paraformaldehyde.

  • Received July 27, 2012.
  • Accepted April 23, 2013.

References

  1. ↵
    1. Luster A. D.,
    2. R. Alon,
    3. U. H. von Andrian
    . 2005. Immune cell migration in inflammation: present and future therapeutic targets. Nat. Immunol. 6: 1182–1190.
    OpenUrlCrossRefPubMed
  2. ↵
    1. Ley K.,
    2. C. Laudanna,
    3. M. I. Cybulsky,
    4. S. Nourshargh
    . 2007. Getting to the site of inflammation: the leukocyte adhesion cascade updated. Nat. Rev. Immunol. 7: 678–689.
    OpenUrlCrossRefPubMed
  3. ↵
    1. Huttenlocher A.,
    2. M. C. Poznansky
    . 2008. Reverse leukocyte migration can be attractive or repulsive. Trends Cell Biol. 18: 298–306.
    OpenUrlCrossRefPubMed
  4. ↵
    1. Massberg S.,
    2. P. Schaerli,
    3. I. Knezevic-Maramica,
    4. M. Köllnberger,
    5. N. Tubo,
    6. E. A. Moseman,
    7. I. V. Huff,
    8. T. Junt,
    9. A. J. Wagers,
    10. I. B. Mazo,
    11. U. H. von Andrian
    . 2007. Immunosurveillance by hematopoietic progenitor cells trafficking through blood, lymph, and peripheral tissues. Cell 131: 994–1008.
    OpenUrlCrossRefPubMed
  5. ↵
    1. Randolph G. J.
    2008. Emigration of monocyte-derived cells to lymph nodes during resolution of inflammation and its failure in atherosclerosis. Curr. Opin. Lipidol. 19: 462–468.
    OpenUrlCrossRefPubMed
    1. Bertho N.,
    2. H. Adamski,
    3. L. Toujas,
    4. M. Debove,
    5. J. Davoust,
    6. V. Quillien
    . 2005. Efficient migration of dendritic cells toward lymph node chemokines and induction of T(H)1 responses require maturation stimulus and apoptotic cell interaction. Blood 106: 1734–1741.
    OpenUrlAbstract/FREE Full Text
    1. Allan R. S.,
    2. J. Waithman,
    3. S. Bedoui,
    4. C. M. Jones,
    5. J. A. Villadangos,
    6. Y. Zhan,
    7. A. M. Lew,
    8. K. Shortman,
    9. W. R. Heath,
    10. F. R. Carbone
    . 2006. Migratory dendritic cells transfer antigen to a lymph node-resident dendritic cell population for efficient CTL priming. Immunity 25: 153–162.
    OpenUrlCrossRefPubMed
  8. ↵
    1. Chtanova T.,
    2. M. Schaeffer,
    3. S. J. Han,
    4. G. G. van Dooren,
    5. M. Nollmann,
    6. P. Herzmark,
    7. S. W. Chan,
    8. H. Satija,
    9. K. Camfield,
    10. H. Aaron,
    11. et al
    . 2008. Dynamics of neutrophil migration in lymph nodes during infection. Immunity 29: 487–496.
    OpenUrlCrossRefPubMed
  9. ↵
    1. Tabas I.
    2010. Macrophage death and defective inflammation resolution in atherosclerosis. Nat. Rev. Immunol. 10: 36–46.
    OpenUrlCrossRefPubMed
  10. ↵
    1. Bellingan G. J.,
    2. H. Caldwell,
    3. S. E. Howie,
    4. I. Dransfield,
    5. C. Haslett
    . 1996. In vivo fate of the inflammatory macrophage during the resolution of inflammation: inflammatory macrophages do not die locally, but emigrate to the draining lymph nodes. J. Immunol. 157: 2577–2585.
    OpenUrlAbstract
  11. ↵
    1. Broche F.,
    2. J. M. Tellado
    . 2001. Defense mechanisms of the peritoneal cavity. Curr. Opin. Crit. Care 7: 105–116.
    OpenUrlCrossRefPubMed
  12. ↵
    1. Harris S. G.,
    2. J. Padilla,
    3. L. Koumas,
    4. D. Ray,
    5. R. P. Phipps
    . 2002. Prostaglandins as modulators of immunity. Trends Immunol. 23: 144–150.
    OpenUrlCrossRefPubMed
    1. Wang D.,
    2. R. N. Dubois
    . 2006. Prostaglandins and cancer. Gut 55: 115–122.
    OpenUrlFREE Full Text
    1. Iñiguez M. A.,
    2. A. Rodríguez,
    3. O. V. Volpert,
    4. M. Fresno,
    5. J. M. Redondo
    . 2003. Cyclooxygenase-2: a therapeutic target in angiogenesis. Trends Mol. Med. 9: 73–78.
    OpenUrlCrossRefPubMed
  15. ↵
    1. Wymann M. P.,
    2. R. Schneiter
    . 2008. Lipid signalling in disease. Nat. Rev. Mol. Cell Biol. 9: 162–176.
    OpenUrlCrossRefPubMed
  16. ↵
    1. Blanco J. C.,
    2. C. Contursi,
    3. C. A. Salkowski,
    4. D. L. DeWitt,
    5. K. Ozato,
    6. S. N. Vogel
    . 2000. Interferon regulatory factor (IRF)-1 and IRF-2 regulate interferon gamma-dependent cyclooxygenase 2 expression. J. Exp. Med. 191: 2131–2144.
    OpenUrlAbstract/FREE Full Text
    1. Crofford L. J.
    1997. COX-1 and COX-2 tissue expression: implications and predictions. J. Rheumatol. Suppl. 49: 15–19.
    OpenUrlPubMed
  18. ↵
    1. Vila-del Sol V.,
    2. M. Fresno
    . 2005. Involvement of TNF and NF-kappa B in the transcriptional control of cyclooxygenase-2 expression by IFN-gamma in macrophages. J. Immunol. 174: 2825–2833.
    OpenUrlAbstract/FREE Full Text
  19. ↵
    1. Ryan E. P.,
    2. C. M. Malboeuf,
    3. M. Bernard,
    4. R. C. Rose,
    5. R. P. Phipps
    . 2006. Cyclooxygenase-2 inhibition attenuates antibody responses against human papillomavirus-like particles. J. Immunol. 177: 7811–7819.
    OpenUrlAbstract/FREE Full Text
  20. ↵
    1. Carey M. A.,
    2. J. A. Bradbury,
    3. J. M. Seubert,
    4. R. Langenbach,
    5. D. C. Zeldin,
    6. D. R. Germolec
    . 2005. Contrasting effects of cyclooxygenase-1 (COX-1) and COX-2 deficiency on the host response to influenza A viral infection. J. Immunol. 175: 6878–6884.
    OpenUrlAbstract/FREE Full Text
  21. ↵
    1. Sharma S.,
    2. S. C. Yang,
    3. L. Zhu,
    4. K. Reckamp,
    5. B. Gardner,
    6. F. Baratelli,
    7. M. Huang,
    8. R. K. Batra,
    9. S. M. Dubinett
    . 2005. Tumor cyclooxygenase-2/prostaglandin E2-dependent promotion of FOXP3 expression and CD4+ CD25+ T regulatory cell activities in lung cancer. Cancer Res. 65: 5211–5220.
    OpenUrlAbstract/FREE Full Text
  22. ↵
    1. Sharma S.,
    2. L. Zhu,
    3. S. C. Yang,
    4. L. Zhang,
    5. J. Lin,
    6. S. Hillinger,
    7. B. Gardner,
    8. K. Reckamp,
    9. R. M. Strieter,
    10. M. Huang,
    11. et al
    . 2005. Cyclooxygenase 2 inhibition promotes IFN-gamma-dependent enhancement of antitumor responses. J. Immunol. 175: 813–819.
    OpenUrlAbstract/FREE Full Text
  23. ↵
    1. Gosset P.,
    2. M. Pichavant,
    3. C. Faveeuw,
    4. F. Bureau,
    5. A. B. Tonnel,
    6. F. Trottein
    . 2005. Prostaglandin D2 affects the differentiation and functions of human dendritic cells: impact on the T cell response. Eur. J. Immunol. 35: 1491–1500.
    OpenUrlCrossRefPubMed
  24. ↵
    1. Wallace A. E.,
    2. K. J. Sales,
    3. R. D. Catalano,
    4. R. A. Anderson,
    5. A. R. Williams,
    6. M. R. Wilson,
    7. J. Schwarze,
    8. H. Wang,
    9. A. G. Rossi,
    10. H. N. Jabbour
    . 2009. Prostaglandin F2alpha-F-prostanoid receptor signaling promotes neutrophil chemotaxis via chemokine (C-X-C motif) ligand 1 in endometrial adenocarcinoma. Cancer Res. 69: 5726–5733.
    OpenUrlAbstract/FREE Full Text
  25. ↵
    1. Hoggatt J.,
    2. P. Singh,
    3. J. Sampath,
    4. L. M. Pelus
    . 2009. Prostaglandin E2 enhances hematopoietic stem cell homing, survival, and proliferation. Blood 113: 5444–5455.
    OpenUrlAbstract/FREE Full Text
  26. ↵
    1. Panzer U.,
    2. M. Uguccioni
    . 2004. Prostaglandin E2 modulates the functional responsiveness of human monocytes to chemokines. Eur. J. Immunol. 34: 3682–3689.
    OpenUrlCrossRefPubMed
  27. ↵
    1. Luft T.,
    2. M. Jefford,
    3. P. Luetjens,
    4. T. Toy,
    5. H. Hochrein,
    6. K. A. Masterman,
    7. C. Maliszewski,
    8. K. Shortman,
    9. J. Cebon,
    10. E. Maraskovsky
    . 2002. Functionally distinct dendritic cell (DC) populations induced by physiologic stimuli: prostaglandin E(2) regulates the migratory capacity of specific DC subsets. Blood 100: 1362–1372.
    OpenUrlAbstract/FREE Full Text
    1. Côté S. C.,
    2. S. Pasvanis,
    3. S. Bounou,
    4. N. Dumais
    . 2009. CCR7-specific migration to CCL19 and CCL21 is induced by PGE(2) stimulation in human monocytes: Involvement of EP(2)/EP(4) receptors activation. Mol. Immunol. 46: 2682–2693.
    OpenUrlCrossRefPubMed
  29. ↵
    1. Yen J. H.,
    2. T. Khayrullina,
    3. D. Ganea
    . 2008. PGE2-induced metalloproteinase-9 is essential for dendritic cell migration. Blood 111: 260–270.
    OpenUrlAbstract/FREE Full Text
  30. ↵
    1. Imhof B. A.,
    2. M. Aurrand-Lions
    . 2004. Adhesion mechanisms regulating the migration of monocytes. Nat. Rev. Immunol. 4: 432–444.
    OpenUrlCrossRefPubMed
  31. ↵
    1. Scandella E.,
    2. Y. Men,
    3. D. F. Legler,
    4. S. Gillessen,
    5. L. Prikler,
    6. B. Ludewig,
    7. M. Groettrup
    . 2004. CCL19/CCL21-triggered signal transduction and migration of dendritic cells requires prostaglandin E2. Blood 103: 1595–1601.
    OpenUrlAbstract/FREE Full Text
  32. ↵
    1. Matsuoka T.,
    2. S. Narumiya
    . 2007. Prostaglandin receptor signaling in disease. ScientificWorldJournal 7: 1329–1347.
    OpenUrlCrossRefPubMed
  33. ↵
    1. Vicente-Manzanares M.,
    2. D. Sancho,
    3. M. Yáñez-Mó,
    4. F. Sánchez-Madrid
    . 2002. The leukocyte cytoskeleton in cell migration and immune interactions. Int. Rev. Cytol. 216: 233–289.
    OpenUrlCrossRefPubMed
  34. ↵
    1. Worthylake R. A.,
    2. K. Burridge
    . 2001. Leukocyte transendothelial migration: orchestrating the underlying molecular machinery. Curr. Opin. Cell Biol. 13: 569–577.
    OpenUrlCrossRefPubMed
  35. ↵
    1. Pankov R.,
    2. Y. Endo,
    3. S. Even-Ram,
    4. M. Araki,
    5. K. Clark,
    6. E. Cukierman,
    7. K. Matsumoto,
    8. K. M. Yamada
    . 2005. A Rac switch regulates random versus directionally persistent cell migration. J. Cell Biol. 170: 793–802.
    OpenUrlAbstract/FREE Full Text
  36. ↵
    1. van Helden S. F.,
    2. D. J. Krooshoop,
    3. K. C. Broers,
    4. R. A. Raymakers,
    5. C. G. Figdor,
    6. F. N. van Leeuwen
    . 2006. A critical role for prostaglandin E2 in podosome dissolution and induction of high-speed migration during dendritic cell maturation. J. Immunol. 177: 1567–1574.
    OpenUrlAbstract/FREE Full Text
  37. ↵
    1. van Helden S. F.,
    2. M. M. Oud,
    3. B. Joosten,
    4. N. Peterse,
    5. C. G. Figdor,
    6. F. N. van Leeuwen
    . 2008. PGE2-mediated podosome loss in dendritic cells is dependent on actomyosin contraction downstream of the RhoA-Rho-kinase axis. J. Cell Sci. 121: 1096–1106.
    OpenUrlAbstract/FREE Full Text
  38. ↵
    1. Cantrell D. A.
    2001. Phosphoinositide 3-kinase signalling pathways. J. Cell Sci. 114: 1439–1445.
    OpenUrlAbstract/FREE Full Text
  39. ↵
    1. Weiss-Haljiti C.,
    2. C. Pasquali,
    3. H. Ji,
    4. C. Gillieron,
    5. C. Chabert,
    6. M. L. Curchod,
    7. E. Hirsch,
    8. A. J. Ridley,
    9. R. Hooft van Huijsduijnen,
    10. M. Camps,
    11. C. Rommel
    . 2004. Involvement of phosphoinositide 3-kinase gamma, Rac, and PAK signaling in chemokine-induced macrophage migration. J. Biol. Chem. 279: 43273–43284.
    OpenUrlAbstract/FREE Full Text
  40. ↵
    1. Liu L.,
    2. K. D. Puri,
    3. J. M. Penninger,
    4. P. Kubes
    . 2007. Leukocyte PI3Kgamma and PI3Kdelta have temporally distinct roles for leukocyte recruitment in vivo. Blood 110: 1191–1198.
    OpenUrlAbstract/FREE Full Text
  41. ↵
    1. Hirsch E.,
    2. V. L. Katanaev,
    3. C. Garlanda,
    4. O. Azzolino,
    5. L. Pirola,
    6. L. Silengo,
    7. S. Sozzani,
    8. A. Mantovani,
    9. F. Altruda,
    10. M. P. Wymann
    . 2000. Central role for G protein-coupled phosphoinositide 3-kinase gamma in inflammation. Science 287: 1049–1053.
    OpenUrlAbstract/FREE Full Text
  42. ↵
    1. Chang J. D.,
    2. G. K. Sukhova,
    3. P. Libby,
    4. E. Schvartz,
    5. A. H. Lichtenstein,
    6. S. J. Field,
    7. C. Kennedy,
    8. S. Madhavarapu,
    9. J. Luo,
    10. D. Wu,
    11. L. C. Cantley
    . 2007. Deletion of the phosphoinositide 3-kinase p110gamma gene attenuates murine atherosclerosis. Proc. Natl. Acad. Sci. USA 104: 8077–8082.
    OpenUrlAbstract/FREE Full Text
  43. ↵
    1. Díaz-Muñoz M. D.,
    2. I. C. Osma-García,
    3. C. Cacheiro-Llaguno,
    4. M. Fresno,
    5. M. A. Iñiguez
    . 2010. Coordinated up-regulation of cyclooxygenase-2 and microsomal prostaglandin E synthase 1 transcription by nuclear factor kappa B and early growth response-1 in macrophages. Cell. Signal. 22: 1427–1436.
    OpenUrlCrossRefPubMed
  44. ↵
    1. Hume D. A.
    2008. Macrophages as APC and the dendritic cell myth. J. Immunol. 181: 5829–5835.
    OpenUrlAbstract/FREE Full Text
  45. ↵
    1. Bellingan G. J.,
    2. P. Xu,
    3. H. Cooksley,
    4. H. Cauldwell,
    5. A. Shock,
    6. S. Bottoms,
    7. C. Haslett,
    8. S. E. Mutsaers,
    9. G. J. Laurent
    . 2002. Adhesion molecule-dependent mechanisms regulate the rate of macrophage clearance during the resolution of peritoneal inflammation. J. Exp. Med. 196: 1515–1521.
    OpenUrlAbstract/FREE Full Text
  46. ↵
    1. Pixley F. J.
    2012. Macrophage Migration and Its Regulation by CSF-1. Int. J. Cell Biol. 2012: 501962.
    OpenUrlCrossRefPubMed
  47. ↵
    1. Zaidel-Bar R.,
    2. R. Milo,
    3. Z. Kam,
    4. B. Geiger
    . 2007. A paxillin tyrosine phosphorylation switch regulates the assembly and form of cell-matrix adhesions. J. Cell Sci. 120: 137–148.
    OpenUrlAbstract/FREE Full Text
  48. ↵
    1. Li S.,
    2. P. Butler,
    3. Y. Wang,
    4. Y. Hu,
    5. D. C. Han,
    6. S. Usami,
    7. J. L. Guan,
    8. S. Chien
    . 2002. The role of the dynamics of focal adhesion kinase in the mechanotaxis of endothelial cells. Proc. Natl. Acad. Sci. USA 99: 3546–3551.
    OpenUrlAbstract/FREE Full Text
  49. ↵
    1. Thomas P. E.,
    2. M. Peters-Golden,
    3. E. S. White,
    4. V. J. Thannickal,
    5. B. B. Moore
    . 2007. PGE(2) inhibition of TGF-beta1-induced myofibroblast differentiation is Smad-independent but involves cell shape and adhesion-dependent signaling. Am. J. Physiol. Lung Cell. Mol. Physiol. 293: L417–L428.
    OpenUrlAbstract/FREE Full Text
  50. ↵
    1. Williams L. M.,
    2. A. J. Ridley
    . 2000. Lipopolysaccharide induces actin reorganization and tyrosine phosphorylation of Pyk2 and paxillin in monocytes and macrophages. J. Immunol. 164: 2028–2036.
    OpenUrlAbstract/FREE Full Text
  51. ↵
    1. Hazeki K.,
    2. N. Masuda,
    3. K. Funami,
    4. N. Sukenobu,
    5. M. Matsumoto,
    6. S. Akira,
    7. K. Takeda,
    8. T. Seya,
    9. O. Hazeki
    . 2003. Toll-like receptor-mediated tyrosine phosphorylation of paxillin via MyD88-dependent and -independent pathways. Eur. J. Immunol. 33: 740–747.
    OpenUrlCrossRefPubMed
  52. ↵
    1. Choi S. H.,
    2. S. Aid,
    3. U. Choi,
    4. F. Bosetti
    . 2010. Cyclooxygenases-1 and -2 differentially modulate leukocyte recruitment into the inflamed brain. Pharmacogenomics J. 10: 448–457.
    OpenUrlCrossRefPubMed
  53. ↵
    1. Dinchuk J. E.,
    2. B. D. Car,
    3. R. J. Focht,
    4. J. J. Johnston,
    5. B. D. Jaffee,
    6. M. B. Covington,
    7. N. R. Contel,
    8. V. M. Eng,
    9. R. J. Collins,
    10. P. M. Czerniak,
    11. et al
    . 1995. Renal abnormalities and an altered inflammatory response in mice lacking cyclooxygenase II. Nature 378: 406–409.
    OpenUrlCrossRefPubMed
  54. ↵
    1. Kamei D.,
    2. K. Yamakawa,
    3. Y. Takegoshi,
    4. M. Mikami-Nakanishi,
    5. Y. Nakatani,
    6. S. Oh-Ishi,
    7. H. Yasui,
    8. Y. Azuma,
    9. N. Hirasawa,
    10. K. Ohuchi,
    11. et al
    . 2004. Reduced pain hypersensitivity and inflammation in mice lacking microsomal prostaglandin e synthase-1. J. Biol. Chem. 279: 33684–33695.
    OpenUrlAbstract/FREE Full Text
  55. ↵
    1. Galli S. J.,
    2. N. Borregaard,
    3. T. A. Wynn
    . 2011. Phenotypic and functional plasticity of cells of innate immunity: macrophages, mast cells and neutrophils. Nat. Immunol. 12: 1035–1044.
    OpenUrlCrossRefPubMed
  56. ↵
    1. Gordon S.,
    2. P. R. Taylor
    . 2005. Monocyte and macrophage heterogeneity. Nat. Rev. Immunol. 5: 953–964.
    OpenUrlCrossRefPubMed
  57. ↵
    1. Potteaux S.,
    2. E. L. Gautier,
    3. S. B. Hutchison,
    4. N. van Rooijen,
    5. D. J. Rader,
    6. M. J. Thomas,
    7. M. G. Sorci-Thomas,
    8. G. J. Randolph
    . 2011. Suppressed monocyte recruitment drives macrophage removal from atherosclerotic plaques of Apoe-/- mice during disease regression. J. Clin. Invest. 121: 2025–2036.
    OpenUrlCrossRefPubMed
  58. ↵
    1. van Gils J. M.,
    2. M. C. Derby,
    3. L. R. Fernandes,
    4. B. Ramkhelawon,
    5. T. D. Ray,
    6. K. J. Rayner,
    7. S. Parathath,
    8. E. Distel,
    9. J. L. Feig,
    10. J. I. Alvarez-Leite,
    11. et al
    . 2012. The neuroimmune guidance cue netrin-1 promotes atherosclerosis by inhibiting the emigration of macrophages from plaques. Nat. Immunol. 13: 136–143.
    OpenUrlCrossRefPubMed
  59. ↵
    1. Zaidel-Bar R.,
    2. S. Itzkovitz,
    3. A. Ma’ayan,
    4. R. Iyengar,
    5. B. Geiger
    . 2007. Functional atlas of the integrin adhesome. Nat. Cell Biol. 9: 858–867.
    OpenUrlCrossRefPubMed
  60. ↵
    1. Alique M.,
    2. L. Calleros,
    3. A. Luengo,
    4. M. Griera,
    5. M. A. Iñiguez,
    6. C. Punzón,
    7. M. Fresno,
    8. M. Rodríguez-Puyol,
    9. D. Rodríguez-Puyol
    . 2011. Changes in extracellular matrix composition regulate cyclooxygenase-2 expression in human mesangial cells. Am. J. Physiol. Cell Physiol. 300: C907–C918.
    OpenUrlAbstract/FREE Full Text
  61. ↵
    1. Calle Y.,
    2. S. Burns,
    3. A. J. Thrasher,
    4. G. E. Jones
    . 2006. The leukocyte podosome. Eur. J. Cell Biol. 85: 151–157.
    OpenUrlCrossRefPubMed
  62. ↵
    1. Tomar A.,
    2. D. D. Schlaepfer
    . 2009. Focal adhesion kinase: switching between GAPs and GEFs in the regulation of cell motility. Curr. Opin. Cell Biol. 21: 676–683.
    OpenUrlCrossRefPubMed
  63. ↵
    1. McLean G. W.,
    2. N. O. Carragher,
    3. E. Avizienyte,
    4. J. Evans,
    5. V. G. Brunton,
    6. M. C. Frame
    . 2005. The role of focal-adhesion kinase in cancer - a new therapeutic opportunity. Nat. Rev. Cancer 5: 505–515.
    OpenUrlCrossRefPubMed
  64. ↵
    1. Choi C. K.,
    2. M. Sukhthankar,
    3. C. H. Kim,
    4. S. H. Lee,
    5. A. English,
    6. K. D. Kihm,
    7. S. J. Baek
    . 2010. Cell adhesion property affected by cyclooxygenase and lipoxygenase: Opto-electric approach. Biochem. Biophys. Res. Commun. 391: 1385–1389.
    OpenUrlCrossRefPubMed
  65. ↵
    1. Bai X. M.,
    2. W. Zhang,
    3. N. B. Liu,
    4. H. Jiang,
    5. K. X. Lou,
    6. T. Peng,
    7. J. Ma,
    8. L. Zhang,
    9. H. Zhang,
    10. J. Leng
    . 2009. Focal adhesion kinase: important to prostaglandin E2-mediated adhesion, migration and invasion in hepatocellular carcinoma cells. Oncol. Rep. 21: 129–136.
    OpenUrlPubMed
  66. ↵
    1. Meng F.,
    2. C. A. Lowell
    . 1998. A beta 1 integrin signaling pathway involving Src-family kinases, Cbl and PI-3 kinase is required for macrophage spreading and migration. EMBO J. 17: 4391–4403.
    OpenUrlAbstract
  67. ↵
    1. Wong M. M.,
    2. E. N. Fish
    . 2003. Chemokines: attractive mediators of the immune response. Semin. Immunol. 15: 5–14.
    OpenUrlCrossRefPubMed
  68. ↵
    1. Alfranca A.,
    2. M. A. Iñiguez,
    3. M. Fresno,
    4. J. M. Redondo
    . 2006. Prostanoid signal transduction and gene expression in the endothelium: role in cardiovascular diseases. Cardiovasc. Res. 70: 446–456.
    OpenUrlAbstract/FREE Full Text
  69. ↵
    1. Welch H. C.,
    2. W. J. Coadwell,
    3. L. R. Stephens,
    4. P. T. Hawkins
    . 2003. Phosphoinositide 3-kinase-dependent activation of Rac. FEBS Lett. 546: 93–97.
    OpenUrlCrossRefPubMed
    1. Beemiller P.,
    2. Y. Zhang,
    3. S. Mohan,
    4. E. Levinsohn,
    5. I. Gaeta,
    6. A. D. Hoppe,
    7. J. A. Swanson
    . 2010. A Cdc42 activation cycle coordinated by PI 3-kinase during Fc receptor-mediated phagocytosis. Mol. Biol. Cell 21: 470–480.
    OpenUrlAbstract/FREE Full Text
  71. ↵
    1. Larson Y.,
    2. J. Liu,
    3. P. D. Stevens,
    4. X. Li,
    5. J. Li,
    6. B. M. Evers,
    7. T. Gao
    . 2010. Tuberous sclerosis complex 2 (TSC2) regulates cell migration and polarity through activation of CDC42 and RAC1. J. Biol. Chem. 285: 24987–24998.
    OpenUrlAbstract/FREE Full Text
  72. ↵
    1. Tybulewicz V. L.,
    2. R. B. Henderson
    . 2009. Rho family GTPases and their regulators in lymphocytes. Nat. Rev. Immunol. 9: 630–644.
    OpenUrlCrossRefPubMed
  73. ↵
    1. Vadas O.,
    2. J. E. Burke,
    3. X. Zhang,
    4. A. Berndt,
    5. R. L. Williams
    . 2011. Structural basis for activation and inhibition of class I phosphoinositide 3-kinases. Sci. Signal. 4: re2.
    OpenUrlAbstract/FREE Full Text
  74. ↵
    1. Díaz-Muñoz M. D.,
    2. I. C. Osma-García,
    3. M. Fresno,
    4. M. A. Iñiguez
    . 2012. Involvement of PGE2 and the cAMP signalling pathway in the up-regulation of COX-2 and mPGES-1 expression in LPS-activated macrophages. Biochem. J. 443: 451–461.
    OpenUrlAbstract/FREE Full Text
  75. ↵
    1. Babaev V. R.,
    2. J. D. Chew,
    3. L. Ding,
    4. S. Davis,
    5. M. D. Breyer,
    6. R. M. Breyer,
    7. J. A. Oates,
    8. S. Fazio,
    9. M. F. Linton
    . 2008. Macrophage EP4 deficiency increases apoptosis and suppresses early atherosclerosis. Cell Metab. 8: 492–501.
    OpenUrlCrossRefPubMed
  76. ↵
    1. Tajima T.,
    2. T. Murata,
    3. K. Aritake,
    4. Y. Urade,
    5. H. Hirai,
    6. M. Nakamura,
    7. H. Ozaki,
    8. M. Hori
    . 2008. Lipopolysaccharide induces macrophage migration via prostaglandin D(2) and prostaglandin E(2). J. Pharmacol. Exp. Ther. 326: 493–501.
    OpenUrlAbstract/FREE Full Text
  77. ↵
    1. Feig J. E.,
    2. I. Pineda-Torra,
    3. M. Sanson,
    4. M. N. Bradley,
    5. Y. Vengrenyuk,
    6. D. Bogunovic,
    7. E. L. Gautier,
    8. D. Rubinstein,
    9. C. Hong,
    10. J. Liu,
    11. et al
    . 2010. LXR promotes the maximal egress of monocyte-derived cells from mouse aortic plaques during atherosclerosis regression. J. Clin. Invest. 120: 4415–4424.
    OpenUrlCrossRefPubMed
  78. ↵
    1. Llodrá J.,
    2. V. Angeli,
    3. J. Liu,
    4. E. Trogan,
    5. E. A. Fisher,
    6. G. J. Randolph
    . 2004. Emigration of monocyte-derived cells from atherosclerotic lesions characterizes regressive, but not progressive, plaques. Proc. Natl. Acad. Sci. USA 101: 11779–11784.
    OpenUrlAbstract/FREE Full Text
Back to top

In this issue

Print
Download PDF
Article Alerts
Email Article
Citation Tools

Jump to section