Biochemical Pharmacology 92 (2014) 642–650
Contents lists available at ScienceDirect
Biochemical Pharmacology journal homepage: www.elsevier.com/locate/biochempharm
Phosphatidic acid inhibits ceramide 1-phosphate-stimulated macrophage migration ˜ ez a, Ana Gomez-Larrauri a, Alberto Ouro a, Lide Arana a, Io-Guane´ Rivera a, Marta Ordon a a a Natalia Presa , Jorge Simo´n , Miguel Trueba , Patricia Gangoiti a, Robert Bittman b, ˜ oz a,* Antonio Gomez-Mun a b
Department of Biochemistry and Molecular Biology, Faculty of Science and Technology, University of the Basque Country (UPV/EHU), 48080 Bilbao, Spain Department of Chemistry and Biochemistry, Queens College of The City University of New York, Flushing, NY 11367-1597, USA
A R T I C L E I N F O
A B S T R A C T
Article history: Received 1 August 2014 Accepted 6 October 2014 Available online 18 October 2014
Ceramide 1-phosphate (C1P) was recently demonstrated to potently induce cell migration. This action could only be observed when C1P was applied exogenously to cells in culture, and was inhibited by pertussis toxin. However, the mechanisms involved in this process are poorly understood. In this work, we found that phosphatidic acid (PA), which is structurally related to C1P, displaced radiolabeled C1P from its membrane-binding site and inhibited C1P-stimulated macrophage migration. This effect was independent of the saturated fatty acid chain length or the presence of a double bond in each of the fatty acyl chains of PA. Treatment of RAW264.7 macrophages with exogenous phospholipase D (PLD), an enzyme that produces PA from membrane phospholipids, also inhibited C1P-stimulated cell migration. Likewise, PA or exogenous PLD inhibited C1P-stimulated extracellularly regulated kinases (ERK) 1 and 2 phosphorylation, leading to inhibition of cell migration. However, PA did not inhibit C1P-stimulated Akt phosphorylation. It is concluded that PA is a physiological regulator of C1P-stimulated macrophage migration. These actions of PA may have important implications in the control of pathophysiological functions that are regulated by C1P, including inﬂammation and various cellular processes associated with cell migration such as organogenesis or tumor metastasis. ß 2014 Elsevier Inc. All rights reserved.
Chemical compounds studied in this article: N-Hexadecanoyl-D-erythro-sphingosine-1phosphate (C16:0-ceramide 1-phosphate) 4-Bromo-5-hydroxy-2-nitrobenzhydryl (BHNB)-ceramide 1-phosphate (BHNB-C1P) Phosphatidic acid (PA) Lysophosphatidic acid (LPA) sn-1,2-Diacylglycerol (DAG) Keywords: Ceramide 1-phosphate Phosphatidic acid Cell migration Macrophages Sphingolipids
1. Introduction Cell migration is fundamental for regulation of cell and tissue homeostasis. It is particularly important in physiological processes such as embryogenesis, or immunity, as well as in inﬂammatory responses, and tumor metastasis. One of the most important processes in the inﬂammatory response is the migration of blood cells, including macrophages, from the blood stream to the sites of infection or tissue damage. This process involves the participation of a variety of cytokines, extracellular matrix proteins, and
Abbreviations: BSA, bovine serum albumin; C1P, ceramide 1-phosphate; PA, phosphatidic acid; S1P, sphingosine 1-phosphate; PLD, phospholipase D; SM, sphingomyelin; Akt (PKB), protein kinase B; ERK, extracellularly regulated kinases; FBS, fetal bovine serum; PTX, pertussis toxin; PI3K, phosphatidylinositol 3-kinase; BHNB, 4-bromo-5-hydroxy-2-nitrobenzhydryl. * Corresponding author. ˜ oz). E-mail address: [email protected]
(A. Gomez-Mun http://dx.doi.org/10.1016/j.bcp.2014.10.005 0006-2952/ß 2014 Elsevier Inc. All rights reserved.
integrins. Some of these molecules play crucial roles in regulating adhesion and migration of macrophages and neutrophils. However, the signaling pathways responsible for coordinating these processes are not well deﬁned. It is known that lipids, including some glycerophospholipids, regulate cell migration. In particular, phosphatidic acid (PA) and lyso-PA (LPA) have been reported to be key regulators of chemotaxis in different cell types [1,2]. In addition, accumulating evidence suggests that sphingolipids play key roles in the regulation of cell migration. For instance, it was reported that decreasing the levels of sphingomyelin (SM) through activation of a neutral sphingomyelinase (N-SMase) associated with the plasma membrane positively regulated migration of polymorphonuclear neutrophils , suggesting that SM-derived metabolites may be key factors for regulation of chemotaxis. The products of N-SMase activity are phosphocholine, which is biologically inert, and ceramide, which is metabolized by a variety of enzymes to produce complex sphingolipids or by the action of speciﬁc
A. Ouro et al. / Biochemical Pharmacology 92 (2014) 642–650
ceramidases to yield sphingosine. Sphingosine, in turn, can be converted to sphingosine 1-phosphate (S1P) by sphingosine kinases. Interestingly, S1P was previously found to act as chemoattractant for various cell types, as a result of binding of S1P to speciﬁc plasma membrane receptors . A key metabolite of ceramide is ceramide 1-phosphate (C1P), which is formed in cells by the action of ceramide kinase. C1P has been involved in the regulation of vital cellular functions including cell growth and survival, as well as inﬂammation . We demonstrated that C1P stimulates macrophage migration [6,7], and it was reported that both S1P and C1P are involved in the promotion of inﬂammation [4,8–12], a complex combination of processes that implicates migration of immune cells to the inﬂamed tissue. Noteworthy, C1P was only able to stimulate cell migration when applied exogenously to cells maintained in culture, and not by increasing the intracellular concentration of C1P. In this connection, the existence of extracellular pools of C1P was recently reported. Speciﬁcally, C1P was found in plasma at 0.5 mM, although its concentration varies depending on the nutritional state of the organism . In addition, C1P is secreted by macrophages and by leaky damaged cells so that local concentrations in vivo can be much higher than basal levels under certain circumstances . It was reported recently that basal levels of C1P can increase by about 10-fold in murine bone marrow cells after g-irradiation . The existence of relatively high levels of exogenous C1P led to identiﬁcation of a speciﬁc receptor through which C1P stimulates cell migration. The receptor is coupled to Gi proteins as C1P was able to increase GTPgS binding to macrophage membranes, and pertussis toxin (PTX) blocked C1P-induced macrophages migration. Ligation of the receptor with C1P caused rapid phosphorylation of extracellularly regulated kinases 1 and 2 (ERK1/2) and Akt (also known as protein kinase B, PKB) upon treatment with C1P, and inhibition of either of these pathways completely abolished C1P-stimulated macrophage migration. Moreover, C1P stimulated the DNA binding activity of nuclear factor kappa B (NF-kB), and blockade of this transcription factor also resulted in complete inhibition of macrophage migration. Recently, we observed that C1P-stimulated macrophage migration is mediated by macrophage chemoattractant protein-1 (MCP-1) . The stimulation of MCP-1 secretion by C1P involved activation of the MEK/ERK, p38, and PI3K/Akt (PKB) pathways whereas c-Jun N-terminal kinase (JNK) was not involved. Interestingly, blockade of secreted MCP-1 with a neutralizing antibody or with speciﬁc siRNA to knock down MCP-1 or the MCP-1 receptor CCR2b inhibited C1P-stimulated macrophage migration . The present work was undertaken to evaluate whether the glycerophospholipid PA, which is structurally related to the sphingophospholipid C1P, could modulate C1P-stimulated macrophage migration and to deﬁne some of the mechanisms involved in this action. 2. Materials and methods 2.1. Materials N-Hexadecanoyl-D-erythro-sphingosine-1-phosphate (C16:0ceramide 1-phosphate) (C1P) was supplied by Avanti Polar Lipids (Alabaster, AL, USA) and Matreya (Pleasant, PA, USA). [33P]ATP (3000 Ci/mmol) was purchased from PerkinElmer (Waltham, MA, USA). [3H]C1P (50–60 Ci/mmol) was supplied by American Radiolabeled Chemicals (ARC) (St. Louis, MO, USA). Dulbecco’s Modiﬁed Eagle’s Medium (DMEM) was from Lonza (Muenchensteinerstrasse, Basel, Switzerland). Exogenous phospholipase D (exPLD), PA from egg yolk, synthetic PAs (dioctanoyl-PA, dimyristoyl-PA, dipalmitoyl-PA, distearoyl-PA, and dioleoyl-PA), and
protease inhibitor cocktail were from Sigma–Aldrich (St. Louis, MO, USA). BHNB-C1P was synthesized as described previously . Nitrocellulose membranes, protein markers, and BCA assay reagents were purchased from Bio-Rad (Alcobendas, Madrid, Spain). Antibodies to phospho-Akt, Akt, p-ERK, and ERK were from Cell Signaling (Schuttersveld, Leiden, The Netherlands). All of the other chemicals and reagents were of the highest grade available. The transilluminator (Darkreader DR-45M) used to expose the cells to light was from Clare Chemical Research (Dolores, CO, USA). The plate reader (Biotek Powerwave XS with Gen 5.1 software) used to measure protein concentrations was from Biotek Instruments (Winooski, VT, USA). 2.2. Cell culture RAW264.7 and J774.A1 macrophages, and 3T3-L1 mouse embryonic ﬁbroblasts were purchased from the American Type Culture Collection (Manassas, VA, USA). C2C12 myoblasts were kindly donated by Dr. Paola Bruni (University of Florence, Italy). Cells were cultured in DMEM supplemented with 10% heatinactivated FBS (or CBS for 3T3-L1 cells), 50 mg/l of gentamicine, and 2 mM L-glutamine at 37 8C in a humidiﬁed atmosphere containing CO2, as previously described [6,7]. 2.3. Delivery of C1P and PA to cells in culture An aqueous dispersion of C1P was added to cultured macrophages, as reported previously [17–19]. Brieﬂy, stock solutions were prepared by sonicating C1P (1.66 mg) in sterile nanopure water (1 ml) on ice using a probe sonicator for six to nine cycles of 15 s on and 10 s off each until a clear dispersion was obtained. The ﬁnal concentration of C1P in the stock solution was approximately 2.6 mM. This procedure is considered preferable to using dispersions prepared by adding C1P in organic solvents to cells because lipid droplet formation is minimized and exposure of cells to organic solvents such as dodecane is avoided. We also delivered C1P to cells by using the photolabile caged C1P analog, 4-bromo-5hydroxy-2-nitrobenzhydryl (BHNB)-C1P , which was dissolved in ethanol at 1.62 mM in the dark. The cells were preincubated with (BHNB)-C1P in the dark for 30 min. The ﬁnal ethanol concentration was <0.06%. The cells were exposed to 400– 500 nm light in a transilluminator equipped with a 9 W lamp for 60 min at a distance of 1.5 cm at 37 8C, so as to release the C1P into the cytosol. PA was also delivered sonicated in water to the cells in culture. Stock solutions for the different PA species were in the range of 1.5–2.5 mM. 2.4. Determination of cell migration Macrophage migration was measured using a Boyden chamberbased cell migration assay, which is also called a trans-well migration assay. Twenty four-well chemotaxis chambers (Transwell, Corning Costar, Amsterdam, The Netherlands) were used for the experiments and before starting experiments, they were coated with 30 mg of ﬁbronectine to allow cell attachment. Cell suspensions (100 ml, 105 cells) were added to the upper compartment of 24-well chemotaxis chambers of 5 mm pores. Agonists were added to the lower wells in 300 ml of medium supplemented with 0.2% of fatty acid-free bovine serum albumin (BSA) and pretreated with activated carbon, and incubated at 37 8C to initiate migration. When used, competitive lipids were added to the upper and lower compartments and preincubated for 30 min prior to agonist addition. Non-migrated cells were removed with a cotton swab, and ﬁlters were ﬁxed with formaldehyde 5% for 30 min. Then formaldehyde was removed and the ﬁlters were stained with hematoxylin for 2 h. After removal of excess hematoxilyn with
A. Ouro et al. / Biochemical Pharmacology 92 (2014) 642–650
water, the ﬁlters were immersed in an acid alcohol solution (70% ethanol/35% HCl, 50:1, v/v) for a few seconds, and they were then submerged in blueing agent for 2 min. Next, ﬁlters were dehydrated with ethanol and stained with eosin for 2 min. After hematoxilyn–eosin staining, the ﬁlters were placed on microscope slides using mineral oil (avoiding bubbles between slides and coverslips). Cell migration was measured by counting the number of migrated cells in a Nikon Eclipse 90i microscope equipped with NIS-Elements 3.0 software. 2.5. Synthesis of [33P]C1P [33P]-C1P was prepared from C16:0-ceramide enzymatically essentially as described previously for unlabeled C1P . Brieﬂy, C16:0-ceramide (from Avanti Polar Lipids, 0.5 mg/tube) was solubilized by sonication in 5 mM cardiolipin, 7.5% octyl bglucopyranoside, and 1 mM diethylenetriaminepentaacetic acid, and then resuspended in a mixture containing 50 mM imidazole, pH 6.6, 50 mM NaCl, 100 mM MgCl2, 1 mM EGTA, and 0.4 U/ml diacylglycerol kinase (DAGK). The reaction was started with 25 mM ATP with [g-33P]ATP (3000 Ci/mmol) and left overnight in a shaking water bath at 37 8C. The reaction was stopped by extraction of the lipids as previously described . The organic phase was dried under N2 and the lipids were analyzed by thinlayer chromatography (TLC) on 0.2-mm silica gel 60 G glass plates. The plates were developed sequentially with chloroform/methanol/NH4OH (65:35:7.5, v/v/v) and chloroform/acetone/acetic acid/ methanol/water (10:4:3:2:1, v/v/v). Radiolabeled C1P was eluted with three washes of 3 ml of chloroform/methanol/acetic acid/ water (50:39:1:1, v/v/v). Silica was removed by centrifugation, and the supernatants were pooled. One ml of water was added to separate the phases. The organic phase was washed once with 1 ml of methanol/water (1:1, v/v), and then dried under a N2 stream. The lipid extract was redissolved in chloroform/methanol (1:1, v/v) and then applied to the TLC plate. C1P was identiﬁed as a single spot at Rf 0.60 by autoradiography. The concentration of C1P was determined by phosphate analysis. [33P]C1P was stored as a stock solution in chloroform/methanol (1:1, v/v) at 20 8C until used. 2.6. Preparation of cell membranes for C1P radioligand-binding assays RAW264.7 macrophages were incubated in 100-mm diameter dishes at 1 107 cells/dish and were grown in DMEM cointaining 10% FBS. They were the incubated in homogenization buffer (10 mM Tris–HCl, 3 mM EDTA, 3 mM EGTA, 1 mM NaF, pH 7.5) containing 1 ml/ml protease inhibitor cocktail and 1 mM phenylmethylsulfonyl ﬂuoride for 30 min on ice. The cells were lysed with a Dounce homogenizer and the remaining intact cells and nuclei were removed by centrifugation at 500 g for 5 min. Cell membranes were pelleted by centrifugation at 100,000 g for 30 min and resuspended in binding buffer (50 mM Tris–HCl, 150 mM NaCl, 0.8% fatty acid-free BSA, 1 ml/ml protease inhibitor cocktail, and 0.2 mM phenylmethylsulfonyl ﬂuoride, pH 7.5). Only freshly prepared membranes were used in experiments. [33P]C1P or [3H]C1P at 8 mM and other sphingolipids at different concentrations were sonicated in fatty acid-free BSA binding buffer (see above) and mixed with membranes (50 mg/tube) in a total volume of 150 ml, in borosilicate tubes. Competition assays were performed as previously described . Binding was performed at 37 8C for 30 min with gentle mixing, and terminated by collecting the membranes on GF/C ﬁlters with a 1225 Sampling Manifold from Millipore. The ﬁlters were then washed rapidly three times with 350 ml of ice-cold washing buffer containing 10 mM Tris–HCl and 15 mM NaCl, pH 7.5. Radioactivity of
ﬁlter-bound radionuclides was quantiﬁed by liquid scintillation counting. 2.7. Western blotting Macrophages were harvested and lysed in ice-cold homogenization buffer as described . Protein (20–40 mg) from each sample was loaded and separated by SDS-PAGE, using 12% separating gels. Proteins were transferred onto nitrocellulose paper and blocked for 1 h with 5% skim milk in Tris-buffered saline (TBS) containing 0.01% NaN3 and 0.1% Tween 20, pH 7.6, and then incubated overnight with the primary antibody in TBS/0.1% Tween at 4 8C. After three washes with TBS/0.1% Tween 20, the nitrocellulose membranes were incubated with horseradish peroxidase-conjugated secondary antibody at 1:4000 dilution for 1 h. Bands were visualized by enhanced chemiluminescence. 2.8. Statistical analyses Results are expressed as mean SEM of three independent experiments performed in triplicate, unless indicated otherwise. Statistical analyses were performed using the two-tailed, paired Student’s t-test, where p < 0.05 was considered to be signiﬁcant (GraphPad Prism software, San Diego, CA).
3. Results We previously observed that treatment of macrophages with exogenous C1P stimulated cell migration [6,7] whereas incubation of the cells with agonists capable of stimulating ceramide kinase (CerK), including interleukin 1-b (IL-1b) or the calcium ionophore A23187 to increase the intracellular levels of C1P, failed to alter chemotaxis . Nonetheless, both IL-1b and Ca2+ can affect a variety of metabolic or cell signaling pathways, other than CerK. Therefore, in this study we have used a cell-permeable photosensitive caged-C1P analog (4-bromo-5-hydroxy-2-nitrobenzhydrylC1P, denoted as BHNB-C1P) to increase C1P levels in the cytosol in the absence of activation of other signaling events. In contrast to exogenous C1P, which causes extensive cell migration, increasing the intracellular concentration of C1P using this caged analog did not cause macrophage migration (Fig. 1), suggesting that C1P requires the implication of mechanisms associated with interactions with cell membranes. Using radiolabeled C1P ([33P]C1P or [3H]C1P) and structurally related lipids including ceramides, SM, or S1P it was concluded that C1P binding to cell membranes was speciﬁc. Further characterization of the membrane C1P binding site was possible using pertussis toxin (PTX), which inhibits the action of Gi proteins, and GTPgS . The latter studies, together with recent observations using S1P agonists and antagonists, led to the conclusion that C1P promotes cell migration through interaction with a speciﬁc cell membrane receptor that is coupled to Gi proteins [6,7]. However no ligand, other than C1P, was found to bind to the C1P receptor, and so regulation of the actions elicited through this receptor have so far remained elusive. In the present study, we have evaluated whether PA, a glycerophospholipid that is structurally closely related to C1P, could affect the stimulation of cell migration by C1P. We found that PA, but not S1P, which can bind to ﬁve different Gi protein-coupled receptors, displaces [33P]C1P from binding to macrophage membranes, suggesting that PA can interact with the putative C1P receptor (Fig. 2A). The IC50 value for PA was 24.63 6.9 mM (mean SEM of three experiments performed in triplicate). Displacement of [33P]C1P from its membrane binding site was independent of the fatty acid chain length of PA or the presence of a double bond in each of the fatty acyl chains, as shown in Fig. 2B.
A. Ouro et al. / Biochemical Pharmacology 92 (2014) 642–650
Fig. 1. Intracellular ceramide 1-phosphate does not stimulate macrophage migration. Macrophage migration was measured by the Boyden chamber-based cell migration assay. RAW264.7 macrophages were seeded in the upper wells of 24well chamber coated with ﬁbronectin at 1 105 cells/well. Cells were incubated in DMEM supplemented with 0.1% of fatty acid-free BSA and 0.2% of FBS (Ctrl), with C1P (30 mM), or BHNB-C1P (1 mM), as indicated. The cells were exposed to 400500 nm light for 1 h using a transilluminator (white bars) or incubated in the dark (black bars), and then were incubated for 4 h, which is the optimal time for induction of cell migration by C1P in these cells. Results are expressed relative to control (Ctrl) values, with or without light, respectively, and are the mean SEM of three independent experiments (*p < 0.05, **p < 0.01).
. The speciﬁcity of PA to block C1P-stimulated macrophage migration was examined using macrophage chemoattractant protein-1 (MCP-1) as an agonist. MCP-1 is a potent chemotactic agent for macrophages, and is known to function through interaction with its speciﬁc plasma membrane receptor CCR2. Fig. 6 shows that PA was unable to block MCP-1-stimulated macrophage migration, suggesting that the effect of PA on inhibition of C1P-stimulated cell migration is speciﬁc. We demonstrated previously that the mechanism whereby C1P stimulates cell migration involves activation of the MEK/ERK1–2 and PI3K/Akt pathways [6,7]. Therefore, we hypothesized that PA could cause inhibition of either or both of these pathways to block cell migration. We found that exogenous PA, or PA that was generated in situ at the plasma membrane by the action of exogenous PLD, increased phosphorylation of Akt, which is in agreement with other work , and that both PA and PLD enhanced C1P-stimulated phosphorylation of this kinase (Fig. 7). Thus, these observations rule out a possible involvement of PI3K/ Akt inhibition in the blockade of C1P-stimulated cell migration by PA. By contrast, although PA can increase phosphorylation of ERK to some extent (Fig. 8) [28,29], PA or exogenous PLD potently blocked C1P-stimulated ERK phosphorylation, which indicates that inhibition of C1P-stimulated macrophage migration by PA occurs through blockade of the MEK/ERK1–2 pathway. 4. Discussion
The above observations led us to hypothesize that PA could either mimic or antagonize the stimulatory effect of C1P on macrophage migration. In this connection, we ﬁrst observed that PA or exogenously added bacterial PLD, the enzyme that produces PA from membrane phospholipids, did not signiﬁcantly stimulate macrophage migration (Fig. 3A). However, although increasing concentrations of PA had no effect on cell migration (Fig. 3B), they caused gradual inhibition of C1P-stimulated macrophage migration. Maximal inhibitory effect was attained at 30 mM PA (Fig. 3C). Fig. 3D shows that the inhibitory effect of PA on C1P-stimulated cell migration was independent of the fatty acid chain length of natural or synthetic saturated PA species or the presence of double bonds in the fatty acid moiety of PA. Fig. 3D also shows that treatment of macrophages with exogenous PLD almost completely blocked the stimulation of cell migration by C1P. These observations indicate that PA is an antagonist of C1P in the regulation of macrophage migration. Like for RAW264.7 macrophages, PA failed to stimulate migration in J774.A1 macrophages and was able to inhibit the stimulatory effect of C1P on these cells (Fig. 4A). However, contrary to macrophages, PA by itself stimulated migration of C2C12 myoblasts and 3T3-L1 mouse embryonic ﬁbroblasts (Fig. 4B), a ﬁnding that is in agreement with other work showing that PA can induce chemotaxis in some cell types, including human breast cancer cell [22,23], and lung epithelial cells . Since PA can be metabolized to lysophosphatidic acid (LPA) by A-type phospholipases, we examined whether this closely related metabolite could mimic the effect of PA. Fig. 5 shows that LPA did not antagonize the stimulatory effect of C1P on macrophage migration, a ﬁnding that is consistent with failure of LPA to displace [3H]C1P from its membrane binding site in C2C12 myoblasts (Ouro et al., unpublished work). Another important metabolite derived from PA is diacylglycerol (DAG), which can be generated in the plasma membrane by the action of lipid phosphate phosphatases . However, contrary to PA, DAG stimulated migration of RAW264.7 macrophages buy about 2.9 0.5 (mean SEM of three independent experiments performed in triplicate, p < 0.05) at 20 mM. Although the mechanism by which exogenous DAG can exert this effect is uncertain, it is known that DAG causes disruption of cell membranes leading to non-speciﬁc effects
C1P is a well-established regulator of cell growth and death, and has both pro- and anti-inﬂammatory properties (reviewed in ). More recently, it was observed that exogenous C1P stimulated cell migration [6,7], an action that is key to regulation of cell and tissue homeostasis and is associated with inﬂammatory responses and tumor dissemination. However, agonist stimulation of ceramide kinase (CerK), the enzyme that produces C1P intracellularly, did not alter cell chemotaxis, suggesting that cell migration required the interaction of C1P with the plasma membrane of cells. The incapability of intracellular C1P to stimulate cell migration was established in this work using the photosensitive caged-C1P analog BHNB-C1P. The ‘‘caging/uncaging’’ strategy allows investigation of the cellular activity of cell-impermeable bioactive molecules. The bond between the ligand and the cage masks the charged phosphate headgroup in the ligand, thereby making the compound cell permeable. This facilitates cell delivery of the compound and prevents any interaction with cell-surface receptors. The bioactive molecule is released into the cytosol on photolysis using light (400–500 nm wavelength) that does not damage cellular components , as indicated in Section 2. Increasing the intracellular concentration of C1P using this cell permeable analog did not cause cell migration, suggesting that C1P acts extracellularly to trigger this process. These observations are consistent with the existence of a putative Gi protein-coupled (PTX-sensitive) receptor speciﬁc for C1P, which we previously identiﬁed . By contrast, intracellular delivery of C1P using the same concentration (1 mM) of the photolabile analog BHNB-C1P resulted in stimulation of macrophage proliferation , a ﬁnding that supports the notion that cell proliferation is regulated by intracellular C1P in a PTX-insensitive manner [16,31,32]. The promotion of cell migration by extracellular C1P has been recently conﬁrmed in other cell types. For instance, Kim and co-workers found that extracellular C1P potently stimulated migration of hematopoietic stem progenitor cells , multipotent stromal cells, and endothelial progenitor cells . In addition, Karapetyan and co-workers reported that extracellular C1P stimulated migration of bone marrow derived stem cells in patients suffering from acute myocardial infarction . Also, we observed that exogenous C1P stimulated migration of 3T3-L1
A. Ouro et al. / Biochemical Pharmacology 92 (2014) 642–650
Fig. 2. Displacement of membrane-bound radiolabeled ceramide 1-phosphate by phosphatidic acid. (A) Cell membranes were incubated with [33P]C1P radioligand at 8 mM in the absence or in the presence of various concentrations of unlabeled of C1P, PA, or S1P, as indicated. Displacement assays were performed as indicated in Section 2. Results are the mean SEM of three independent experiments performed in triplicate. (B) Effect of different PA species on membrane-bound [3H]C1P (8 mM). Binding assays were performed as in (A) in the presence of different PA species (400 mM), as follows: PA mix from egg yolk (EY), dioctanoyl-PA (C8:0), dimyristoyl-PA(C14:0), dipalmitoyl-PA (C16:0), distearoyl-PA (C18:0), or dioleoyl-PA (C18:1). Palmitoyl-C1P (C1P) and octanoyl-C1P (C8-C1P) were used as positive controls in the experiments. Results are expressed as [3H]C1P bound relative to control (Ctrl), and are the mean of three independent experiments performed in duplicate (***p < 0.001; **p < 0.01;*p < 0.05).
preadipocytes and human THP-1 monocytes . Therefore, it can be concluded that only extracellular C1P, but not C1P that was generated intracellularly, is responsible for stimulation of cell migration in macrophages. Interestingly, it has been recently reported that ceramide kinase-derived C1P is required for primary mouse embryonic ﬁbroblast migration , thereby supporting the concept that C1P may be a universal regulator of cell migration. The putative C1P receptor was demonstrated to be coupled to Gi proteins and was found to be of moderately low afﬁnity, with an apparent Kd value of 7.7 mM and a maximum binding capacity of 1269 pmol/mg protein . A key observation in the present study is that PA, a natural and structurally related glycerophospholipid C1P analog, was able to displace [33P]C1P from binding to cell membranes, suggesting that PA binds to the putative C1P receptor. Although exogenous PA or bacterial PLD, the enzyme that produces PA from membrane phospholipids, did not signiﬁcantly stimulate macrophage migration, both of them inhibited C1P-stimulated macrophage migration, clearly indicating that PA is an antagonist of C1P in the
regulation of cell chemotaxis. In this connection, it has been recently reported that PA levels ﬂuctuate during cell migration and membrane process extension, suggesting that PA is a relevant regulator of chemotaxis . However, LPA, which binds to a variety of Gi protein-coupled cell membrane receptors, did not affect C1P-stimulated macrophage migration, which is in agreement with the incapability of LPA to displace radiolabeled C1P from its membrane binding site in myoblasts (Ouro et al., unpublished work). We demonstrated previously that the mechanism whereby C1P stimulates cell migration involves activation of the MEK/ERK1–2 and PI3K/Akt pathways [6,7]. Therefore, it was hypothesized that PA could cause inhibition of either or both of these pathways to block cell migration. We found that exogenous PA, or PA that was generated at the plasma membrane by the action of exogenous PLD, increased phosphorylation of Akt, which is in agreement with other work . Furthermore, both PA and PLD enhanced C1P-stimulated phosphorylation of this kinase, ruling out a possible involvement of PI3K/Akt inhibition in the blockade of
A. Ouro et al. / Biochemical Pharmacology 92 (2014) 642–650
Fig. 3. Effect of phosphatidic acid on cell migration. (A) PA or exogenous PLD do not alter macrophage migration. Macrophage migration was measured by the Boyden chamber-based cell migration assay. RAW264.7 macrophages were seeded in the upper wells of 24-well chamber coated with ﬁbronectin at 1 105 cells/well. Cells were incubated in DMEM supplemented with 0.1% of fatty acid-free BSA and 0.2% of FBS (Ctrl), with C1P (30 mM), or with a PA mixture from egg yolk (EY); dioctanoyl-PA; dimyristoyl-PA; dipalmitoyl-PA; distearoyl-PA; or dioleoyl-PA at 30 mM or exogenous PLD (exPLD) at 500 mU/ml, as indicated. Then, the cells were incubated for 4 h, which is the optimal time for induction of cell migration by C1P in RAW264.7 macrophages. Results are expressed as cell migration relative to the control (Ctrl) value, and are the mean SEM of three independent experiments except for EY, which is the average of six experiments (**p < 0.01). (B) Macrophage migration was measured as in (A). RAW264.7 macrophages were seeded at 1 105 cells/well in the upper wells of 24-well chamber coated with ﬁbronectin. The cells were incubated in DMEM supplemented with 0.1% of fatty acid-free BSA and 0.2% of FBS (Ctrl), with C1P (30 mM), or with increasing concentrations of PA from egg yolk (EY) for 4 h, as indicated. Results are expressed as cell migration relative to the control (Ctrl) value, and are the mean SEM of three independent experiments (**p < 0.01). (C) PA blocked C1P-induced cell migration. Macrophage migration was measured as in (A). The cells were pre-incubated with increasing concentrations of PA from egg yolk (EY) for 1 h, as indicated. Then, the cells were incubated with 30 mM of C1P for 4 h, which are the optimal concentration and time for induction of cell migration by C1P. Results are expressed as cell migration relative to the control (Ctrl) value, and are the mean SEM of three independent experiments (***p < 0.001; control versus C1P-treated cells, #p < 0.05, ##p < 0.01; C1P-treated cells versus C1P-treated cells in the presence of PA). (D) Effect of different PA species or exogenous PLD (exPLD) on C1P-induced cell migration. Macrophage migration was measured as in (A). Cells were pre-incubated in DMEM supplemented with 0.1% of fatty acid-free BSA and 0.2% of FBS (Ctrl) or with a PA mixture from egg yolk (EY); dioctanoyl-PA; dimyristoyl-PA (C14:0); dipalmitoyl-PA (C16:0); distearoyl-PA (C18:0); or dioleoyl-PA (C18:1) at 30 mM for 1 h, or with exPLD at 500 mU/ml for 2 h, as indicated. Then, the cells were incubated for 4 h with C1P at 30 mM. Results are expressed as cell migration relative to the control (Ctrl) value, and are the mean SEM of three independent experiments (***p < 0.001; control versus C1P-treated cells, #p < 0.05, ##p < 0.01; C1Ptreated cells versus C1P-treated cells in the presence of the corresponding PA species or exPLD). The percentage of cells that migrated spontaneously (Ctrl) was 8.3 1.6, mean SEM of six independent experiments performed in triplicate).
C1P-stimulated cell migration by PA. By contrast, PA or exogenous PLD potently blocked C1P-stimulated ERK phosphorylation, indicating that inhibition of C1P-stimulated macrophage migration by PA occurs through blockade of the MEK/ERK1–2 pathway. Interestingly, Asano and co-workers  recently reported that lowering SM levels facilitates ERK1–2 activation, and Sitrin et al.  showed that decreasing SM levels by activation of a neutral sphingomyelinase (N-SMase) associated with the plasma membrane positively regulated migration of polymorphonuclear neutrophils. These ﬁndings are consistent with the notion that C1P is synthesized from SM-derived ceramide . Although the optimal concentration of C1P to stimulate migration of the macrophages is relatively high compared with the plasma levels found in vivo (0.5 mM) , it was shown that C1P levels vary according to the nutritional state of the organism, and that C1P can be secreted by macrophages [13,14,38];
therefore, local concentrations of C1P in vivo can be much higher than 0.5 mM. In fact, it was reported that physiological concentrations of ATP (0.1 mM) can elevate intracellular C1P up to about 4–5 nmol per 106 cells . Therefore, if this amount of C1P were released into the extracellular milieu, local concentrations of 5– 10 mM, or even higher, would be achievable immediately after secretion. In addition, Ratajczak and co-workers reported that C1P levels increased by about 10-fold after g-radiation of the bone marrow in mice, also suggesting that local concentrations of C1P can be much higher than basal C1P levels . Thus the effective concentrations of C1P to induce migration in cultured macrophage are within or close to physiological levels. All of these ﬁndings are particularly relevant when considering that macrophages play critical roles in chronic and acute inﬂammation, and that these cells promote cancer cell invasion . With regards to PA, regulation of PA levels is complex, exhibiting important differences
A. Ouro et al. / Biochemical Pharmacology 92 (2014) 642–650
Fig. 5. Lysophosphatidic acid does not inhibit C1P-stimulated cell migration. Macrophages were incubated and treated with PA or LPA at 30 mM as in Fig. 3. Results are expressed as cell migration relative to the control value (Ctrl), and are the mean SEM of three independent experiments (***p < 0.001; control versus C1P-treated cells, or C1P-treated cells versus C1P-treated cells in the presence of LPA; # p < 0.05).
Fig. 4. Effect of phosphatidic acid on cell migration in J774.A1 macrophages, C2C12 myoblasts, and 3T3-L1 mouse embryonic ﬁbroblasts. (A) Macrophage migration was measured as in (Fig. 3). J774A.1 macrophages were seeded at 2.5 104 cells/ well in the upper wells of 24-well chamber coated with ﬁbronectin. The cells were pre-incubated with PA at 20 mM for 1 h, as indicated. Then, the cells were incubated with 20 mM C1P for 24 h, which are the optimal concentration and time for induction of cell migration by C1P in these cells (see Ref. ). Results are expressed as cell migration relative to the control (Ctrl) value, and are the mean SEM of three independent experiments (*p < 0.05; control versus C1P-treated cells, #p < 0.05 C1Ptreated cells versus C1P-treated cells in the presence of PA). (B) Myoblast and ﬁbroblast migration was measured as in Fig. 3. The cells were seeded at 2.5 104 cells/well in the upper wells of 24-well chamber coated with ﬁbronectin. The cells were incubated with C1P or PA at 10 mM for 24 h, which are the optimal concentrations and time for induction of cell migration (see Ref. ). Results are expressed as cell migration relative to the control (Ctrl) value, and are the mean SEM of three independent experiments (**p < 0.01; *p < 0.05; control versus C1P). The percentages of cells that migrated spontaneously for each particular cell line (Ctrl values) were 19.8 6.6 for J774.A1 macrophages, 6.6 0.3 for C2C12 myoblasts, and 17.8 9.9 for 3T3-L1 mouse embryonic ﬁbroblasts (results expressed as mean SEM of three independent experiments performed in triplicate).
among cell types. There are high levels of different PLD activities in serum or plasma, and the plasma membrane of cells, namely glycosylphosphatidylinositol-speciﬁc phospholipase D (GPI-PLD) and PLD2. These enzymes can act on different phospholipids at the plasma membrane of cells (namely glycosylphosphatidylinositol and phosphatidylcholine, respectively) to release PA in situ [41,42]. In addition, PA can be generated through phosphorylation of diacylglycerol (DAG) by DAG kinase, and Lipid phosphate phosphatases can convert PA back to DAG . Although the concentration of PA at the plasma membrane of cells has not been reported, relatively high concentrations of PA may be achieved
depending upon the extracellular levels and degree of activation of these PLD and DAG kinase activities, under certain pathological (i.e. PLD secreted by bacteria after infection) or physiological conditions. Of interest, important PA ﬂuctuations have been observed during cell migration  thereby supporting the notion that PA is an important regulator of chemotaxis. In conclusion, this is the ﬁrst report showing that PA antagonizes the extracellular effects of C1P, suggesting that this glycerophospholipid is a key factor for regulation of receptor mediated C1P actions. Speciﬁcally, PA is a potent inhibitor of C1Pstimulated macrophage migration through a mechanism involving both displacement of C1P from its membrane binding site and inhibition of C1P-stimulated ERK phosphorylation. These actions of PA may have important implications in the control of physiological cell functions such as organogenesis or tissue regeneration, and may also impact pathological processes that are associated with cell migration such as inﬂammation and tumor metastasis.
Fig. 6. Phosphatidic acid does not inhibit MCP-1-stimulated macrophage migration. Cell migration was measured as in Fig. 3. The cells were seeded at 0.5 105 cells/ well in the upper wells of 24-well chamber coated with ﬁbronectin. RAW264.7 macrophages were pre-incubated with PA at 30 mM for 1 h, as indicated. Then, the cells were incubated with 150 ng/ml of MCP-1 for 24 h, which are the optimal concentration and time for induction of cell migration by MCP-1 (see ref. 7). Results are expressed as cell migration relative to the control (Ctrl) value, and are the mean SEM of three independent experiments (*p < 0.05; control versus MCP-1).
A. Ouro et al. / Biochemical Pharmacology 92 (2014) 642–650
Fig. 7. Effect of phosphatidic acid and exogenous PLD on C1P-stimulated Akt phosphorylation. (A) Macrophages were seeded at 1 106 cells/35-mm dish in 1 ml of medium. The next day, the cells were pre-incubated with PA (30 mM) for 1 h or with exPLD (500 mU/ml) for 2 h. Then, the cells were incubated with C1P (30 mM) for 10 min. Macrophages were lysed by sonication and Akt phosphorylation was analyzed by western blotting using phospho-Akt (p-Akt), as described in Section 2. Equal loading of protein was monitored using a speciﬁc antibody to total Akt. Similar results were obtained in each of 3 replicate experiments. (B) Results of scanning densitometry of exposed ﬁlm. Data are expressed as arbitrary units of intensity relative to control value and are the mean SEM of three replicate experiments (*p < 0.05; **p < 0.01; control versus C1Ptreated cells).
Fig. 8. Effect of phosphatidic acid and exogenous PLD on C1P-stimulated ERK1/2 phosphorylation. (A) Macrophages were seeded at 1 106 cells/35-mm dish in 1 ml of medium. Cells were pre-incubated with PA (30 mM) for 1 h or exPLD 500 mU/ml for 2 h, and then the cells were incubated with C1P at 30 mM for 10 min. Macrophages were lysed by sonication, and ERK1/2 phosphorylation was analyzed by western blotting using phospho-ERK (p-ERK), as described in Section 2. Equal loading of protein was monitored using a speciﬁc antibody to total ERK. Similar results were obtained in each of 3 replicate experiments. (B) Results of scanning densitometry of exposed ﬁlm. Data are expressed as arbitrary units of intensity relative to control value and are the mean SEM of three replicate experiments (**p < 0.001, control versus C1P-treated cells, # p < 0.05, C1P-treated cells versus C1P-treated cells in the presence of PA; ##p < 0.01, C1P-treated cells versus C1P-treated cells in the presence of exPLD).
A. Ouro et al. / Biochemical Pharmacology 92 (2014) 642–650
Acknowledgements This work was supported by Grants IT-705-13 from Departamento de Educacio´n, Universidades e Investigacio´n del Gobierno Vasco (GV/EJ, Spain), S-PE12UN040 and S-PE13UN017 from Departamento de Industria, Comercio y Turismo del Gobierno Vasco (Basque Government, GV/EJ, Spain). A. Ouro and L. Arana were recipients of fellowships from the Basque Government. I.-G. Rivera is the recipient of a fellowship from MICINN, and M. ˜ ez is the recipient of a fellowship from the University of the Ordon Basque Country (GV/EJ, Spain). Also, we are grateful to ‘‘Servicios Generales de investigacio´n (SGIker)’’ and Unidad de formacio´n e investigacio´n (UFI) 11/20 (UPV/EHU)’’ for technical support. References  Gomez-Cambronero J. Phosphatidic acid, phospholipase D and tumorigenesis. Adv Biol Regul 2014;54C:197–206.  Willier S, Butt E, Grunewald TG. Lysophosphatidic acid (LPA) signalling in cell migration and cancer invasion: a focussed review and analysis of LPA receptor gene expression on the basis of more than 1700 cancer microarrays. Biol Cell 2013;105:317–33.  Sitrin RG, Sassanella TM, Petty HR. An obligate role for membrane-associated neutral sphingomyelinase activity in orienting chemotactic migration of human neutrophils. Am J Respir Cell Mol Biol 2011;44:205–12.  Spiegel S, Milstien S. The outs and the ins of sphingosine-1-phosphate in immunity. Nat Rev Immunol 2011;11:403–15.  Gangoiti P, Camacho L, Arana L, Ouro A, Granado MH, Brizuela L, et al. Control of metabolism and signaling of simple bioactive sphingolipids: Implications in disease. Prog Lipid Res 2010;49:316–34.  Granado MH, Gangoiti P, Ouro A, Arana L, Gonzalez M, Trueba M, et al. Ceramide 1-phosphate (C1P) promotes cell migration Involvement of a speciﬁc C1P receptor. Cell Signal 2009;21:405–12.  Arana L, Ordonez M, Ouro A, Rivera IG, Gangoiti P, Trueba M, et al. Ceramide 1phosphate (C1P) induces macrophage chemoattractant protein-1 release: involvement in C1P-stimulated cell migration. Am J Physiol Endocrinol Metab 2013;304:213–26.  Lamour NF, Chalfant CE. Ceramide-1-phosphate: the missing’’ link in eicosanoid biosynthesis and inﬂammation. Mol Interv 2005;5:358–67.  Baumruker T, Bornancin F, Billich A. The role of sphingosine and ceramide kinases in inﬂammatory responses. Immunol Lett 2005;96:175–85.  Gomez-Munoz A, Gangoiti P, Granado MH, Arana L, Ouro A. Ceramide-1phosphate in cell survival and inﬂammatory signaling. Adv Exp Med Biol 2010;118–30.  Hammad SM, Crellin HG, Wu BX, Melton J, Anelli V, Obeid LM. Dual and distinct roles for sphingosine kinase 1 and sphingosine 1 phosphate in the response to inﬂammatory stimuli in RAW macrophages. Prostaglandins Other Lipid Mediat 2008;85:107–14.  Maceyka M, Harikumar KB, Milstien S, Spiegel S. Sphingosine-1-phosphate signaling and its role in disease. Trends Cell Biol 2012;22:50–60.  Hammad SM, Pierce JS, Soodavar F, Smith KJ, Al Gadban MM, Rembiesa B, et al. Blood sphingolipidomics in healthy humans: impact of sample collection methodology. J Lipid Res 2010;51:3074–87.  Kim C, Schneider G, Abdel-Latif A, Mierzejewska K, Sunkara M, Borkowska S, et al. Ceramide-1-phosphate regulates migration of multipotent stromal cells and endothelial progenitor cells—implications for tissue regeneration. Stem Cells 2012;31:500–10.  Ratajczak MZ, Jadczyk T, Schneider G, Kakar SS, Kucia M. Induction of a tumormetastasis-receptive microenvironment as an unwanted and underestimated side effect of treatment by chemotherapy or radiotherapy. J Ovarian Res 2013;6:95.  Lankalapalli RS, Ouro A, Arana L, Gomez-Munoz A, Bittman R. Caged ceramide 1-phosphate analogues: synthesis and properties. J Org Chem 2009;74:8844–7.  Gangoiti P, Granado MH, Wang SW, Kong JY, Steinbrecher UP, Gomez-Munoz A. Ceramide 1-phosphate stimulates macrophage proliferation through activation of the PI3-kinase/PKB, JNK and ERK1/2 pathways. Cell Signal 2008;20:726–36.  Gomez-Munoz A, Duffy PA, Martin A, O’Brien L, Byun HS, Bittman R, et al. Short-chain ceramide-1-phosphates are novel stimulators of DNA synthesis and cell division: antagonism by cell-permeable ceramides. Mol Pharmacol 1995;47:833–9.
 Gomez-Munoz A, Kong JY, Salh B, Steinbrecher UP. Ceramide-1-phosphate blocks apoptosis through inhibition of acid sphingomyelinase in macrophages. J Lipid Res 2004;45:99–105.  Dressler KA, Kolesnick RN. Ceramide 1-phosphate, a novel phospholipid in human leukemia (HL-60) cells. Synthesis via ceramide from sphingomyelin. J Biol Chem 1990;265:14917–21.  Hundal RS, Gomez-Munoz A, Kong JY, Salh BS, Marotta A, Duronio V, et al. Oxidized low density lipoprotein inhibits macrophage apoptosis by blocking ceramide generation, thereby maintaining protein kinase B activation and BclXL levels. J Biol Chem 2003;278:24399–408.  Sliva D, Mason R, Xiao H, English D. Enhancement of the migration of metastatic human breast cancer cells by phosphatidic acid. Biochem Biophys Res Commun 2000;268:471–9.  Sliva D, Harvey K, Mason R, Lloyd Jr F, English D. Effect of phosphatidic acid on human breast cancer cells exposed to doxorubicin. Cancer Invest 2001;19: 783–90.  Shao ZR, Wang Q, Xu XF, Zhang Z, Lu YB, Shen G, et al. Phospholipase D participates in H(2)O(2)-induced A549 alveolar epithelial cell migration. Exp Lung Res 2012;38:427–33.  Brindley DN, Pilquil C. Lipid phosphate phosphatases and signaling. J Lipid Res 2009;50(Suppl.):S225–30.  Leikin S, Kozlov MM, Fuller NL, Rand RP. Measured effects of diacylglycerol on structural and elastic properties of phospholipid membranes. Biophys J 1996;71:2623–32.  Lim HK, Choi YA, Park W, Lee T, Ryu SH, Kim SY, et al. Phosphatidic acid regulates systemic inﬂammatory responses by modulating the Akt-mammalian target of rapamycin-p70 S6 kinase 1 pathway. J Biol Chem 2003;278:45117–27.  Kraft CA, Garrido JL, Fluharty E, Leiva-Vega L, Romero G. Role of phosphatidic acid in the coupling of the ERK cascade. J Biol Chem 2008;283:36636–45.  Andresen BT, Rizzo MA, Shome K, Romero G. The role of phosphatidic acid in the regulation of the Ras/MEK/Erk signaling cascade. FEBS Lett 2002;531: 65–8.  Gomez-Munoz A, Gangoiti P, Arana L, Ouro A, Rivera IG, Ordonez M, et al. New insights on the role of ceramide 1-phosphate in inﬂammation. Biochim Biophys Acta 2013;1831:1060–6.  Gangoiti P, Bernacchioni C, Donati C, Cencetti F, Ouro A, Gomez-Munoz A, et al. Ceramide 1-phosphate stimulates proliferation of C2C12 myoblasts. Biochimie 2012;94:597–607.  Gomez-Munoz A, Kong JY, Parhar K, Wang SW, Gangoiti P, Gonzalez M, et al. Ceramide-1-phosphate promotes cell survival through activation of the phosphatidylinositol 3-kinase/protein kinase B pathway. FEBS Lett 2005;579: 3744–50.  Kim CH, Wu W, Wysoczynski M, Abdel-Latif A, Sunkara M, Morris A, et al. Conditioning for hematopoietic transplantation activates the complement cascade and induces a proteolytic environment in bone marrow: a novel role for bioactive lipids and soluble C5b-C9 as homing factors. Leukemia 2011;26:106–16.  Karapetyan AV, Klyachkin YM, Selim S, Sunkara M, Ziada KM, Cohen DA, et al. Bioactive lipids and cationic antimicrobial peptides as new potential regulators for trafﬁcking of bone marrow-derived stem cells in patients with acute myocardial infarction. Stem Cells Dev 2013;22:1645–56.  Wijesinghe DS, Brentnall M, Mietla JA, Hoeferlin LA, Diegelmann RF, Boise LH, et al. Ceramide kinase is required for a normal eicosanoid response and the subsequent orderly migration of ﬁbroblasts. J Lipid Res 2014;55:1298–309.  Ferraz-Nogueira JP, Diez-Guerra FJ, Llopis J. Visualization of phosphatidic acid ﬂuctuations in the plasma membrane of living cells. PLoS ONE 2014;9: e102526.  Asano S, Kitatani K, Taniguchi M, Hashimoto M, Zama K, Mitsutake S, et al. Regulation of cell migration by sphingomyelin synthases: sphingomyelin in lipid rafts decreases responsiveness to signaling by the CXCL12/CXCR4 pathway. Mol Cell Biol 2012;32:3242–52.  Boath A, Graf C, Lidome E, Ullrich T, Nussbaumer P, Bornancin F. Regulation and trafﬁc of ceramide 1-phosphate produced by ceramide kinase: comparative analysis to glucosylceramide and sphingomyelin. J Biol Chem 2008;283: 8517–26.  Lamour NF, Stahelin RV, Wijesinghe DS, Maceyka M, Wang E, Allegood JC, et al. Ceramide kinase uses ceramide provided by ceramide transport protein: localization to organelles of eicosanoid synthesis. J Lipid Res 2007;48: 1293–304.  Goswami S, Sahai E, Wyckoff JB, Cammer M, Cox D, Pixley FJ, et al. Macrophages promote the invasion of breast carcinoma cells via a colony-stimulating factor-1/epidermal growth factor paracrine loop. Cancer Res 2005;65: 5278–83.  Deeg MA, Bierman EL, Cheung MC. GPI-speciﬁc phospholipase D associates with an apoA-I- and apoA-IV-containing complex. J Lipid Res 2001;42:442–51.  Du G, Huang P, Liang BT, Frohman MA. Phospholipase D2 localizes to the plasma membrane and regulates angiotensin II receptor endocytosis. Mol Biol Cell 2004;15:1024–30.