Am J Physiol Endocrinol Metab 4: E E6,. First published April, ; doi:.5/ajpendo.48.. Ceramide -phosphate induces macrophage chemoattractant protein- release: involvement in ceramide -phosphate-stimulated cell migration Lide Arana, Marta Ordoñez, Alberto Ouro, Io-Guané Rivera, Patricia Gangoiti, Miguel Trueba, and Antonio Gomez-Muñoz Department of Biochemistry and Molecular Biology, Faculty of Science and Technology, University of the Basque Country (UPV/EHU), Bilbao, Spain Submitted 5 September ; accepted in final form 6 March Arana L, Ordoñez M, Ouro A, Rivera I-G, Gangoiti P, Trueba M, Gomez-Muñoz A. Ceramide -phosphate induces macrophage chemoattractant protein- release: involvement in ceramide -phosphate-stimulated cell migration. Am J Physiol Endocrinol Metab 4: E E6,. First published April, ; doi:.5/ajpendo.48.. The bioactive sphingolipid ceramide -phosphate (CP) is implicated in inflammatory responses and was recently shown to promote cell migration. However, the mechanisms involved in these actions are poorly described. Using J774A. macrophages, we have now discovered a new biological activity of CP: stimulation of monocyte chemoattractant protein- (MCP-) release. This novel effect of CP was pertussis toxin (PTX) sensitive, suggesting the intervention of G i protein-coupled receptors. Treatment of the macrophages with CP caused activation of the phosphatidylinositol -kinase (PIK)/Akt, mitogen-activated protein kinase kinase (MEK)/extracellularly regulated kinases (ERK), and p8 pathways. Inhibition of these kinases using selective inhibitors or specific sirna blocked the stimulation of MCP- release by CP. CP stimulated nuclear factor- B activity, and blockade of this transcription factor also resulted in complete inhibition of MCP- release. Also, CP stimulated MCP- release and cell migration in human THP- monocytes and T-L preadipocytes. A key observation was that sequestration of MCP- with a neutralizing antibody or treatment with MCP- sirna abolished CP-stimulated cell migration. Also, inhibition of the pathways involved in CP-stimulated MCP- release completely blocked the stimulation of cell migration by CP. It can be concluded that CP promotes MCP- release in different cell types and that this chemokine is a major mediator of CP-stimulated cell migration. The PIK/Akt, MEK/ ERK, and p8 pathways are important downstream effectors in this action. ceramide -phosphate; monocyte chemoattractant protein- release; macrophage migration; sphingosine -phosphate Address for reprint requests and other correspondence: A. Gómez-Muñoz, Dept. of Biochemistry and Molecular Biology, Faculty of Science and Technology, Univ. of the Basque Country (UPV/EHU), P.O. Box 644, 488, Bilbao, Spain (e-mail: antonio.gomez@ehu.es). REGULATION OF CELL MIGRATION is a complex process involving hundreds of molecules. It is necessary for tissue homeostasis and is crucial for regulation of vital biological processes including embryogenesis, organogenesis, or regeneration (reviewed in Refs. 8, 48). Also, cell migration is fundamental to inflammatory responses (, 49), but inadequate migratory signals may induce the migration of the wrong cell type to the wrong place, which may have severe effects in the organism. Some examples include autoimmune syndromes, angiogenesis, or the process of metastasis. Although cell migration is a subject of intense investigation, the mechanisms involved in controlling cell movements are incompletely understood. In this connection, growing evidence suggests that some sphingolipids are key metabolites for controlling chemotaxis (5). Our group recently reported that ceramide -phosphate (CP), a sphingolipid metabolite that is present in serum or plasma (5), and which we found to regulate cell growth and survival (5, 7, 8,, ), is a chemoattractant molecule for macrophages (). This finding is particularly relevant when considering that macrophages play critical roles in chronic and acute inflammation and that these cells facilitate cancer cell migration (, 46). In addition, Barth et al. (4) recently showed that ceramide kinase, the enzyme responsible for the phosphorylation of ceramide to CP, regulates NADPH oxidase activity and eicosanoid biosynthesis in neuroblastoma cells, an action that is consistent with our recent finding that CP increases the production of reactive oxygen species through activation of NADPH oxidase in macrophages (). Besides these proinflammatory actions, increasing evidence suggests that CP can aslo function as an anti-inflammatory agent under different experimental settings. Specifically, Hankins et al. (6) recently reported that CP reduces lipopolysaccharide (LPS)-mediated nuclear factor- B (NF- B) activation and cytokine secretion, and Goldsmith et al. (6) reported that phosphoceramide analog- simultaneously suppresses TNF and induces the production of anti-inflammatory interleukin- in activated macrophages. The most important chemokine that regulates the migration and infiltration of monocytes/macrophages and works as a key factor in initiating the various inflammatory responses is monocyte chemoattractant protein- [MCP-, also known as chemokine (C-C motif) ligand (CCL)] (). MCP--induced migration of monocytes is mediated through interaction with its receptor CCR, a G i protein-coupled receptor (). There are two subtypes of MCP- receptors, CCRa and -b, with the CCRb receptor isoform being about fivefold more sensitive to induction of chemotaxis by MCP- than CCRa (5). Furthermore, monocytes and activated NK cells express predominantly the CCRb isoform (). The present study was undertaken to examine whether CP could stimulate the release of MCP- by macrophages and to assess whether this cytokine mediates the chemotactic effect of CP in macrophages. MATERIALS AND METHODS Materials. N-hexadecanoyl-D-erythrosphingosine- phosphate (C6: -ceramide -phosphate; CP) was supplied by Matreya. Culture medium Dulbecco s modified Eagle s medium (DMEM) and RPMI (Roswell Park Memorial Institute) were from Lonza. Fibronectin, JE (MCP-) from mouse, RS-895, LY-94, PD-9859, pertussis toxin (PTX), and SP-65 were from Sigma-Aldrich. Fetal bovine serum (FBS) and opti-mem were from GIBCO. Nitrocellulose mem- Downloaded from http://ajpendo.physiology.org/ by...6 on December, 7 http://www.ajpendo.org 9-849/ Copyright the American Physiological Society E
E4 branes, protein markers, and BCA assay reagents were purchased from Bio-Rad. MCP- neutralizing antibody (cell tested) was purchased from ebioscience. -Actin, GAPDH, sphingosine -phosphate receptor (SPR; EDG-5), and SPR4 (EDG-6) antibodies were from Santa Cruz Biotechnology. SPR, SPR, and SPR5 antibodies were supplied by Abcam. Other antibodies were from Cell Signaling. Mouse CCL (MCP-) ELISA Ready-SET-Go! was supplied by ebioscience. Oligofectamine reagent was from Molecular Probes (Invitrogen). SC-54 and -DEBC were from Tocris. Akt sirna, Mapk (ERK) sirna, Pikr (PIK) sirna, Mapk8 (JNK) sirna, Mapk9 (JNK) sirna, Mapk (JNK) sirna, Mapk4 (p8 ) sirna, SPR (spr) sirna, SPR (spr) sirna and SPR4 (spr4) sirna were from Applied Biosystems (Ambion) and CKR- sirna, negative sirna, SPR (EDG-5) sirna, SPR5 (EDG-8) sirna, and MCP- sirna were supplied by Santa Cruz Biotechnology. All of the other chemicals and reagents were of the highest grade available. Cell culture. The J774A., THP-, and T-L cell lines used in this work were purchased from ATCC (Manassas, VA) and were cultured following the manufacturer s indications. J774A. macrophages were grown in DMEM supplemented with % heat-inactivated FBS, 5 mg/l gentamicin, and M L-glutamine. Cells were incubated at constant temperature (7 C) in a humidified atmosphere containing 5% CO. The THP- cells were grown in RPMI supplemented with % heat-inactivated FBS, 5 mg/l gentamicin, and M L-glutamine. The cells were incubated at constant temperature (7 C) in a humidified atmosphere containing 5% CO. The T-L cell line was grown in DMEM supplemented with % heat-inactivated FCS, 4.5 g/l D-glucose, 5 mg/l gentamicin, and M L-glutamine. Cells were incubated at constant temperature (7 C) in a humidified atmosphere containing 5% CO. Bone marrow-derived macrophages (BMDM) were isolated from femurs of 6- to 8-wk-old female CD- mice as described (4). Cells were plated for 4 h in RPMI 64 medium containing % FBS and % L cell conditioned medium as the source of macrophage colony-stimulating factor (MCSF) (7). The nonadherent cells were removed and cultured for 4 6 days in the same medium until 8% confluence was reached. Delivery of CP to cells in culture. An aqueous dispersion (in the form of liposomes) of CP was added to cultured macrophages as previously described (5, 9, ). Specifically, stock solutions were prepared by sonicating CP (5 mg) in sterile nanopure water ( ml) on ice using a probe sonicator until a clear dispersion was obtained. The final concentration of CP in the stock solution was.6 mm. This procedure is considered preferable to dispersions prepared by adding CP in organic solvents, because droplet formation is minimized and exposure of cells to alcohols or dodecane is avoided. Measurement of MCP- concentration. J774A. macrophages were seeded in 4-well plates ( 4 cells/well) and incubated in.55 ml of DMEM containing % FBS overnight. The next day, the cells were washed twice with PBS, and the medium was replaced by fresh DMEM containing % FBS. Macrophages were incubated for two additional hours, and agonists were then added as required. After incubation, the cell media were collected into microcentrifuge tubes, and cells were scrapped for counting. The T-L preadipocytes were seeded in 4-well plates ( 4 cells/well) and incubated in DMEM containing % FCS overnight. The next day, the cells were washed twice with PBS, and the medium was replaced by fresh DMEM in the absence of FCS. Cells were incubated for two additional hours, and agonists were then added as required. Then, the media were collected, and cells were trypsinized for counting. THP- cells were seeded in 4-well plates ( 5 cells/well) and incubated in.5 ml of RPMI. After h, agonists were added as required. Then, cell media along with monocytes were collected into microcentrifuge tubes, and cells were counted. The harvested media were centrifuged at, g for 5 min at 4 C, and supernatants were diluted for determination of MCP- concentration. The amount of MCP- in the medium was determined using a Mouse CCL (MCP-) ELISA Ready-SET-Go! kit according to the manufacturer s instructions. Sample concentration values (pg/ ml) obtained by the calibration curve were normalized by the number of cells counted in each well or by the amount of protein in each well. Determination of cell migration. Cell migration was measured using a Boyden chamber-based cell migration assay, as described (). Twenty-four-well chemotaxis chambers (Transwell, Corning Costar) precoated with g of fibronectin were used for the experiments. For J774A., T-L, and BMDM cell migration studies, 8 m pore size was used, and for THP- cells the pore size was 5 m. Cell suspensions ( l, 5 4 cells for J774A.; l, 5 4 cells for T-L; 5 5 cells for BMDM; and l, 5 cells for THP-) were then added to the upper wells of 4-well chemotaxis chambers. Agonists were added to the lower wells diluted in l of medium supplemented with.% fatty acid-free bovine serum albumin (BSA) and.% FBS treated with activated carbon. When used, inhibitors were added to the upper and lower wells and Fig.. Ceramide -phosphate (CP) induces monocyte chemoattractant protein- (MCP-) release in J774A. macrophages. A: cells were seeded in 4-well plates ( 4 cells/well) and incubated overnight in DMEM supplemented with % FBS. Cells were then washed, and medium was replaced with DMEM supplemented with % FBS. After hofincubation, macrophages were further incubated for 4 h with indicated concentrations of CP. MCP- concentration was measured by ELISA, as indicated in MATERIALS AND METHODS. MCP- values were normalized to the total cell number, and results are expressed as means SE of 5 independent experiments performed in duplicate. B: cells were treated as in A and incubated with M CP for indicated periods of time. MCP- concentration was determined as indicated in MATERIALS AND METHODS. Values were normalized to total cell number, and results are expressed as means SE of 4 independent experiments performed in duplicate (P.5, P.). Downloaded from http://ajpendo.physiology.org/ by...6 on December, 7
E5 Downloaded from http://ajpendo.physiology.org/ by...6 on December, 7 Fig.. The PIK/Akt pathway is involved in CP-stimulated MCP- release. Cells were seeded and treated as in Fig.. A: cells were preincubated with M LY-94 (a selective PIK inhibitor; filled bars) or with vehicle (open bars) for min prior to stimulation with M CP. Cells were then incubated further for 4 h, and MCP- concentration was measured by ELISA, as indicated in MATERIALS AND METHODS. MCP- values were normalized to total cell number, and results are expressed as means SE of 5 independent experiments performed in duplicate. B: cells were preincubated with M -DEBC (a selective Akt inhibitor; filled bars) or with vehicle (open bars) for min and then treated with M CP for 4 h. MCP- concentration was determined by ELISA as indicated in MATERIALS AND METHODS. Values were normalized to total cell number, and results are expressed as means SE of 5 independent experiments performed in duplicate. C: cells were seeded in 6-mm dishes ( 5 cells/dish) and treated with vehicle (open bars), negative sirna (filled bars), or PIK sirna (hatched bars), as described in MATERIALS AND METHODS. Vehicle (control) or M CP was then added for 4 h, as indicated. Cells were then scraped and treated as in A. MCP- concentration was determined by ELISA and normalized to total cell number. D: cells were incubated and treated as in C, except that sirna Akt was used instead of PIK sirna. Data are expressed as means SE of 4 independent experiments performed in duplicate. E: PIK sirna inhibitory efficiency was confirmed by Western blotting using specific antibodies against PIK. Equal loading of protein was monitored using a specific antibody to GAPDH. Similar results were obtained in each of independent experiments. F: results of scanning densitometry of exposed film. Data are expressed as arbitrary units of intensity and are means range of independent experiments. G: Akt sirna inhibitory efficiency was confirmed by Western blotting using specific antibodies against Akt. Equal loading of protein was monitored using a specific antibody to GAPDH. Similar results were obtained in each of independent experiments. H: results of scanning densitometry of exposed film. Data are expressed as arbitrary units of intensity and are means range of independent experiments. (n.s., not significant; P.5, P.).
E6 preincubated for h prior to agonist addition. Nonmigrated cells were removed with a cotton swab, and the filters were fixed with formaldehyde (5% in PBS) and stained with hematoxylin-eosin. Cell migration was assessed by counting the number of migrated cells in a Nikon Eclipse 9i microscope equipped with NIS-Elements. software. Cells were counted in six randomly selected microscopy fields per well at magnification. Treatment of cells with sirna. sirna transfection protocols were performed following the manufacturer s instructions. Cells were seeded in 6-mm diameter plates ( 5 cells/well) in DMEM containing % FBS. The medium was replaced by.6 ml of opti- MEM, and cells were then incubated for 4 h. sirna ( pmol/ml) in.4 ml of opti-mem was added into each well. Cells were then incubated for 4 5 h, and ml of opti-mem containing % FBS was then added to the wells without removal of the transfection mixture. The cells were incubated further for 4 h, and the medium was then replaced by fresh DMEM containing % FBS. Cells were scrapped and counted and then used for experiments. Western blotting. Cells (.5 6 cells/plate) were seeded in 6-mm diameter plates in DMEM containing % FBS and incubated overnight. The next day, the cells were washed, and medium was replaced by DMEM containing % FBS. After h, agonists were added, and cells were further incubated for the indicated periods of time. Macrophages were harvested and lysed in ice-cold homogenization buffer as described (4). About 4 g of protein from each sample was loaded and separated by sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE), using % separating gels. Proteins were transferred onto nitrocellulose paper and blocked for h with 5% skim milk in Tris-buffered saline (TBS) containing.% NaN and.% Tween, and then incubated overnight with the primary antibody in TBS-.% Tween at 4 C. After three washes with the TBS-.% Tween, membranes were incubated with horseradish peroxidase-conjugated secondary antibody at :4, dilution for h. Thereafter, the proteins were visualized by enhanced chemiluminescence. Statistical analyses. Results are expressed as means SE of three to six independent experiments performed in triplicate unless indicated otherwise. Confirmation for sirna inhibition of each particular sirna-targeted protein is expressed as the mean range of two independent experiments. Statistical analyses were performed using the two-tailed, paired Student s t-test, where P.5 was considered to be significant (GraphPad Prism software, San Diego, CA). RESULTS CP stimulates MCP- release in J774A. macrophages. Leukocyte chemotaxis toward sites of inflammation or tumorigenesis is primarily mediated by chemokine signaling (7), and we () have demonstrated that CP stimulates macrophage migration. A major chemokine that regulates the migration and infiltration of macrophages is MCP-. Figure shows that CP significantly stimulated the release of MCP- by J774A. macrophages in a concentration- (Fig. A) and timedependent (Fig. B) manner. Optimal MCP- release was attained at M CP after 4 h of incubation. A similar fold increase of MCP- release was still observed after 48 h of incubation with M CP (Fig. B). Although this optimal concentration of CP is relatively high compared with plasma levels (.5 M) (5), it was shown that CP concentrations vary according to the nutritional state of the organism and can be secreted by macrophages (5, 5) or by leaky damaged cells (9), so that local concentrations of CP in vivo can be much higher than.5 M. In addition, it was reported that ATP at physiological concentrations (. mm) can elevate intracellular Downloaded from http://ajpendo.physiology.org/ by...6 on December, 7 Fig.. The MEK/ERK pathway is involved in stimulation of MCP- release by CP. A: macrophages were seeded and treated as in Fig. and preincubated for min with vehicle (open bars) or with M PD-9859 to inhibit MEK (filled bars) prior to stimulation with M CP for 4 h, as indicated. MCP- concentration in the medium was measured by ELISA (see MATERIALS AND METHODS), and results are means SE of 6 independent experiments performed in duplicate. B: cells were seeded in 6-mm dishes ( 5 cells/dish) and treated with vehicle (open bars), negative sirna (filled bars), or ERK sirna (hatched bars), as described in MATERIALS AND METHODS. Cells were then incubated with vehicle (Control) or with M CP for 4 h, as indicated. MCP- concentration was normalized to total cell number, and data are expressed as means SE of 5 independent experiments performed in duplicate. C: ERK sirna inhibitory efficiency was confirmed by Western blotting using specific antibodies against ERK. Equal loading of protein was monitored using a specific antibody to GAPDH. Similar results were obtained in each of independent experiments. D: results of scanning densitometry of exposed film. Data are expressed as arbitrary units of intensity and are means range of independent experiments (P.5, P.).
E7 Fig. 4. p8 MAPK is implicated in CP-stimulated MCP- release. A: macrophages were seeded and treated as in Fig.. Cells were preincubated with vehicle (open bars) or with M SB 9 (a selective inhibitor of p8; filled bars) for min prior to stimulation with M CP for 4 h, as indicated. MCP- was measured by ELISA and normalized to total cell number. Results are means SE of independent experiments performed in duplicate. B: cells were seeded in 6-mm dishes ( 5 cells/dish) and treated with vehicle (open bars), negative sirna (filled bars), or p8 sirna (hatched bars), as described in MATERIALS AND METHODS. Cells were then incubated with vehicle (Control) or with M CP for 4 h, as indicated. MCP- concentration was normalized, and data are expressed as means SE of 5 independent experiments performed in duplicate. C: p8 sirna inhibitory efficiency was confirmed by Western blotting using specific antibodies against p8. Equal loading of protein was monitored using a specific antibody to GAPDH. Similar results were obtained in each of independent experiments. D: results of scanning densitometry of exposed film. Data are expressed as arbitrary units of intensity and are means range of independent experiments (P.5). Downloaded from http://ajpendo.physiology.org/ by...6 on December, 7 Fig. 5. JNK is not implicated in CP-stimulated MCP- release. J774A. cells were seeded in 6-mm dishes ( 5 cells/dish), and sirna treatment to silence JNK (A), JNK (B), and JNK (C) was performed as described in MATERIALS AND METHODS. Cells were seeded and treated with vehicle (Control) or with M CP for 4 h, as indicated. MCP- concentration was normalized, and data are expressed as means SE of 5 independent experiments performed in duplicate. sirna inhibitory efficiency was confirmed by Western blotting using specific antibodies against JNK (D), JNK (E), and JNK (F). Equal loading of protein was monitored using a specific antibody to GAPDH. Similar results were obtained in each of independent experiments. Exposed films were scanned, and data obtained by densitometry are expressed as arbitrary units of intensity (JNK G, JNK H, and JNK I) and are means range of independent experiments.
E8 Fig. 6. CP induces phosphorylation of ERK/, p8, and Akt. Macrophages were seeded in 6-mm dishes (.5 5 cells/dish) and treated as indicated in MATERIALS AND METHODS. Cells were then treated with M CP for various times as indicated. A: phosphorylated proteins were detected by Western blotting using specific antibodies against each phosphorylated kinase. Equal loading of protein was monitored using a specific antibody to total protein of each kinase. Similar results were obtained in each of 4 independent experiments. B: results of scanning densitometry of exposed film. Data are expressed as arbitrary units of intensity and are means SE of 4 independent experiments (P.5). CP up to 4 5 nmol per million cells (). So, if this amount of CP were released into the extracellular milieu, local concentrations of 5 M, or even higher, would be easily achievable right after secretion. Also, in this work, CP was added to the cells in the form of liposomes (sonicated in water), and so the actual concentration of CP in touch with the cells was much lower than what was added to the culture medium. Therefore, the effective concentrations of CP used in these experiments can be considered to be within or very close to physiological levels. CP-stimulated MCP- release is mediated by the PIK/Akt (PKB), MEK/ERK/, and p8 pathways. We showed previously that CP was able to stimulate PIK/Akt (PKB) and MEK/ERK/ and that these pathways were involved in the mitogenic (5) and chemotactic () effects of CP. Therefore, we tested to see whether these pathways were implicated in the release of MCP- by CP. First, selective inhibitors of PIK and Akt, as well as specific sirna to silence the expression of the genes encoding these kinases, were used. Figure, A and B, shows that the PIK inhibitor LY-94 ( M) or -DEBC ( M), an inhibitor of Akt, completely inhibited CP-stimulated MCP- release. Likewise, preincubation of these cells with specific sirnas to silence PIK or Akt completely blocked the stimulation of MCP- release by CP (Fig., C and D). Western blotting demonstrated potent knockdown of PIK (Fig., E and F), or Akt (Fig., G and H) with pmol/ml PIK-targeted or Akt-targeted sirna, respectively. To evaluate the possible involvement of MAP kinases in this process, the MEK inhibitor PD-9859 and specific sirna against ERK were used. Both of these agents completely blocked CP-stimulated MCP- release (Fig., A and B). Western-blotting demonstrated potent knockdown of ERK (Fig., C and D) with pmol/ml ERK-targeted sirna. In addition, the p8 inhibitor SB 9 and sirna against p8 significantly decreased the stimulation of MCP- release by CP (Fig. 4, A and B). Western blotting demonstrated potent knockdown of p8 (Fig. 4, C and D) with pmol/ml p8 -targeted sirna. However, specific sirnas against the three different JNK isoforms (JNK ), although potently inhibited, did not significantly alter CP-stimulated MCP- release, suggesting that this kinase is not involved in this process (Fig. 5). The effect of CP on stimulation of ERK/, p8, and Akt phosphorylation is shown in Fig. 6. A well-known downstream target of Akt and ERK/ is NF- B, a transcription factor that plays an essential role in the induction of inflammatory mediators. The activation of NF- B by CP in the J774A. macrophages was evaluated by determining its phosphorylation state after treatment with CP. Figure 7, A and B, shows that CP increases phosphorylation of NF- B in a time-dependent manner, which is consistent with NF- B activation. The implication of NF- B in CP-stimulated MCP- release was evaluated using SC-54, a selective inhibitor of this transcription factor. This inhibitor completely blocked the release of MCP- that was stimulated by CP (Fig. 7C). MCP- is essential for stimulation of cell migration by CP. We reported recently () that CP promotes RAW 64.7 leukemia monocytic cell migration and that the PIK/Akt and MEK/ERK pathways were involved in this process. These observations together with the results shown above led us to hypothesize that MCP- might be the key mediator of CPstimulated macrophage migration. The ability of MCP- to stimulate macrophage migration was studied by incubating the cells with increasing concentrations of MCP-. Figure 8 shows that MCP- stimulates macrophage migration in a concentration- (Fig. 8A) and time-dependent (Fig. 8B) manner. Of interest, MCP- was as potent as CP at stimulating cell migration (Fig. 8, A and C). To evaluate whether the release of MCP- was required for the stimulation of macrophage migration by CP, two different experimental approaches were used. First, the cells were preincubated with a specific monoclonal antibody against MCP- prior to stimulation with CP, and second, the macrophages were pretreated with specific MCP- sirna. Both of these treatments completely blocked CPstimulated macrophage migration (Fig. 9), thereby demonstrating that MCP- release is absolutely required in this process. Western blotting demonstrated potent knockdown of MCP- (Fig. 9, C and D) with pmol/ml MCP--targeted sirna. Two key observations that further supported this hypothesis were that RS-895, a selective inhibitor of the MCP- receptor CCRb (Fig. A), and sirna against this receptor (Fig. B), completely blocked both MCP-- and CP-stimulated macrophage migration (Fig. ). Western blotting demonstrated potent knockdown of CCRb (Fig., C and D) with pmol/ml CCRb-targeted sirna. Downloaded from http://ajpendo.physiology.org/ by...6 on December, 7
E9 latter work, Fig. B shows that inhibition of NF- B blocks CP-stimulated macrophage migration. In addition, we show in this work that CP-stimulated MCP- release and macrophage migration are inhibited by PTX (Fig. ), which is consistent with our previous studies suggesting the involvement of a G i protein-coupled receptor in the stimulation of RAW 64.7 monocytic leukemia cell migration by CP (). Fig. 7. NF- B is implicated in CP-induced MCP- release. A: macrophages were seeded in 6-mm dishes (.5 5 cells/dish) and treated as in Fig. 6. Cells were stimulated with M CP for various times as indicated. Phosphorylation of the p65 subunit of NF- B was detected by Western blotting using a specific antibody to phospho-p65. Equal loading of protein was monitored using a specific antibody to -actin. Similar results were obtained in each of independent experiments B: results of scanning densitometry of exposed film. Data are expressed as arbitrary units of intensity and are means SE of independent experiments. C: J774A. cells were seeded and treated as in Fig. and preincubated with vehicle (open bars) or with NF- B inhibitor SC-54 (5 M) (filled bars) for min prior to stimulation with M CP as indicated. After 4 h, MCP- concentration was measured by ELISA, as described in MATERIALS AND METHODS and normalized to total cell number. Results are means SE of 5 independent experiments performed in duplicate (P.). CP-stimulated migration of J774A. macrophages involves activation of the PIK/Akt, MEK/ERK/, and p8 pathways. The observation that MCP- release is essential for the stimulation of macrophage migration by CP suggested that the pathways involved in CP-stimulated MCP- release might also be implicated in cell migration. This was examined using selective inhibitors of these pathways (PD-9859 to inhibit MEK, -DEBC to inhibit Akt, and SB 9 to inhibit p8), as well as specific sirnas to silence the genes encoding these kinases. All of these inhibitory agents abrogated the induction of macrophage migration by CP (Fig., A and B), suggesting that these pathways are crucial in this process. We previously reported that NF- B is a downstream target of Akt and ERK and that CP stimulated the DNA binding activity of this transcription factor in macrophages (5). In agreement with the A B C 5 4 7.5 75 5 CP [MCP-] (ng/ml) 4 Time (h) 7.5 5..5. 5 [CP] (μm) Fig. 8. Stimulation of cell migration by MCP- and CP. Macrophage migration was measured using the Boyden chamber-based cell migration assay, as described in MATERIALS AND METHODS. Cells (5 4 cells/well) were plated in the upper wells of 4-well chambers coated with fibronectin and incubated for h. A: cells were stimulated with increasing concentrations of MCP- or with M CP as indicated and incubated further for 4 h. Results are expressed relative to control value and are means SE of 6 independent experiments performed in duplicate. B: cells were treated with 5 ng/ml MCP- for indicated time periods. Results are expressed relative to the value at h and are means SE of 5 independent experiments performed in duplicate. C: cells were stimulated with increasing concentrations of CP as indicated and incubated further for 4 h. Results are expressed relative to time at h and are means SE of independent experiments performed in duplicate (P.5, P.). Downloaded from http://ajpendo.physiology.org/ by...6 on December, 7
E Fig. 9. MCP- is essential for CP-stimulated cell migration. A: macrophages were seeded and treated as in Fig. 8 and incubated with indicated concentrations of anti-mcp- antibody in the absence (squares) or presence (triangles) of M CP. Results are means SE of 4 independent experiments performed in duplicate. B: cells were seeded in 6-mm dishes ( 5 cells/dish) and treated with vehicle (open bars), negative sirna (filled bars), or MCP- sirna (hatched bars) as indicated in MATERIALS AND METHODS. Vehicle (Control) or M CP was then added to the cells, and they were incubated further for 4 h. Results are expressed relative to control value and are means SE of 4 independent experiments performed in duplicate. C: MCP- sirna inhibitory efficiency was confirmed by Western blotting using specific antibodies against MCP-. Equal loading of protein was monitored using a specific antibody to GAPDH. Similar results were obtained in each of independent experiments. D: results of scanning densitometry of exposed film. Data are expressed as arbitrary units of intensity and are means range of independent experiments (P.5, P.). Downloaded from http://ajpendo.physiology.org/ by...6 on December, 7 Fig.. MCP- receptor (CCRb) is involved in CP-stimulated cell migration. A: macrophage migration was measured as indicated in Fig. 8. Cells were preincubated for h with vehicle (open bars) or with nm RS-895 (a selective inhibitor of CCRb; filled bars) and then treated with M CP or 5 ng/ml MCP- for 4 h as indicated. Results are expressed relative to control value and are means SE of 4 independent experiments performed in duplicate. B: cells were seeded in 6-mm dishes ( 5 cells/dish) and treated with vehicle (open bars), negative sirna (filled bars), or CCRb sirna (hatched bars) as described in MATERIALS AND METHODS. Vehicle (Control) or M CP was then added to the cells as indicated, and they were incubated further for 4 h. Results are expressed relative to control values and are means SE of 4 independent experiments performed in duplicate. C: CCRb sirna inhibitory efficiency was confirmed by Western blotting using specific antibodies against CCRb. Equal loading of protein was monitored using a specific antibody to GAPDH. Similar results were obtained in each of independent experiments. D: results of scanning densitometry of exposed film. Data are expressed as arbitrary units of intensity and are means range of independent experiments (P.5, P., #P.5, cells treated with MCP- vs. cells treated with MCP- and RS-895).
A B 9 Vehicle Negative sirna 8 ERK sirna 7 PIK sirna Akt sirna 6 p8α sirna 5 4 Control 6 Vehicle PD 9859 5 -DEBC SB 9 SC-54 4 Control n.s. CP CP Fig.. Involvement of PIK/Akt, MEK/ERK/, p8, and NF- B in CPstimulated cell migration. A: cells were seeded in 6-mm dishes ( 5 cells/dish) and treated with vehicle, negative sirna, or specific sirna to inhibit ERK, PIK, Akt, or p8, as indicated (see MATERIALS AND METHODS). Vehicle (Control) or M CP was then added to the cells, and they were incubated further for 4 h. Results are expressed relative to control value and are means SE of 4 independent experiments performed in duplicate. B: macrophages were seeded and treated as in Fig. 8. Cells were then preincubated with M PD-9859, M -DEBC, M SB 9, or 5 M SC-54 for h and then treated with or without M CP, as indicated, and incubated for 4 h. Results are means SE of 4 independent experiments performed in duplicate (P.5). Stimulation of MCP- release and cell migration by CP are independent of interaction with sphingosine -phosphate receptors. Sphingosine -phosphate (SP) is a bioactive sphingolipid that is closely related to CP. In fact, like the CP receptor, SP receptors are also coupled to G i proteins and sensitive to inhibition by PTX (, 5). There are five SP receptors, which are named SPR 5. Therefore, it could be speculated that CP might be able to interact with any of these receptors to stimulate MCP- release and cell migration. To test this possibility, we treated the macrophages with specific sirnas to silence the genes encoding for each SPR. Figure shows that pretreating cells with specific SPR sirnas did not significantly affect CP-stimulated MCP- release (Fig. A) or cell migration (Fig. B), thereby demonstrating that these effects of CP are independent of interaction with SP receptors. Western blotting demonstrated potent knockdown of SPR (Fig., C and D), SPR (Fig., E and F), SPR (Fig., G and H), SPR4 (Fig., I and J), and SPR5 (Fig., K and L) with pmol/ml of each SPR-targeted sirna. CP stimulates MCP- release and cell migration in human THP- monocytes and T-L preadipocytes. In addition to stimulating MCP- release and cell migration in murine macrophages, we found that CP also exerted these actions on other cell types, including human THP- monocytes and T-L preadipocytes. Figure 4 shows that CP promoted MCP- release (Fig. 4A) and cell migration (Fig. 4B) in THP- cells and that migration was inhibited when the cells were incubated in the presence of a specific antibody to MCP- (Fig. 4C), in analogy to the results obtained in J774A. macrophages (Figs., 8, and 9). Likewise, CP was able to stimulate MCP- release (Fig. 5A) and cell migration (Fig. 5B) in T-L preadipocytes, and these effects were also abolished in the presence of a specific antibody to MCP- (Fig. 5C). DISCUSSION We reported recently that CP stimulates cell migration in RAW 64.7 cells (). Although this finding has increased our knowledge on how chemotaxis can be regulated in cells, the mechanisms or signaling pathways involved in this process are only beginning to be elucidated. In the present study, we A MCP- release (ng/ 6 cells) B 5 Vehicle Ptx 75 5 5 4 Control 6 Vehicle 5 Ptx Control CP CP Fig.. Inhibition of CP-stimulated MCP- release and macrophage migration by pertussis toxin (PTX). A: cells were seeded and treated as in Fig.. Macrophages were preincubated for 6 h with vehicle (open bars) or with pg/ml PTX (filled bars). Then, vehicle (Control) or M CP was added and cells were further incubated for 4 h. Medium was collected and MCP- concentration determined by ELISA as indicated in MATERIALS AND METHODS. MCP- values were normalized to total cell number, and results are expressed as means SE of 4 independent experiments performed in duplicate. B: macrophage migration was measured as in Fig. 8 (see MATERIALS AND METHODS). Cells were preincubated for 4 h in the absence (open bars) or presence (filled bars) of pg/ml PTX and then treated with vehicle (Control) or M CP, as indicated, for 4 h. Results are means SE of 4 independent experiments performed in duplicate (P.5, P.). E Downloaded from http://ajpendo.physiology.org/ by...6 on December, 7
E Downloaded from http://ajpendo.physiology.org/ Fig.. Stimulation of MCP- release and cell migration by CP are independent of interaction with sphingosine -phosphate (SP) receptors (SPR). Cells were seeded in 6-mm dishes ( 5 cells/dish) and treated with vehicle, negative sirna, or SPR sirna, SPR sirna, SPR sirna, SPR4 sirna, or SPR5 sirna, as indicated. A: cells were then seeded as in Fig. and treated with vehicle (Control) or M CP for 4 h as indicated. MCP- concentration was normalized, and data are expressed as means SE of 5 independent experiments performed in duplicate, except for studies using SPR5 sirna, whose data are expressed as means SE of independent experiments performed in duplicate. B: macrophage migration was measured as in Fig. 8 (see MATERIALS AND METHODS). Cells were treated with vehicle (Control) or M CP, as indicated, for 4 h. Results are means SE of 5 independent experiments performed in duplicate, except for studies using SPR5 sirna, whose data are expressed as means SE of independent experiments performed in duplicate. C: SPR sirna inhibitory efficiency was confirmed by Western blotting using specific antibodies against SPR. Equal loading of protein was monitored using a specific antibody to GAPDH. Similar results were obtained in each of independent experiments. D: results of scanning densitometry of exposed film. Data are expressed as arbitrary units of intensity and are means range of independent experiments. E: SPR sirna inhibitory efficiency was confirmed by Western blotting using specific antibodies against SPR. Equal loading of protein was monitored using a specific antibody to GAPDH. Similar results were obtained in each of independent experiments. F: results of scanning densitometry of exposed film. Data are expressed as arbitrary units of intensity and are means range of independent experiments. G: SPR sirna inhibitory efficiency was confirmed by Western blotting using specific antibodies against SPR. Equal loading of protein was monitored using a specific antibody to GAPDH. Similar results were obtained in each of independent experiments. H: results of scanning densitometry of exposed film. Data are expressed as arbitrary units of intensity and are means range of independent experiments. I: SPR4 sirna inhibitory efficiency was confirmed by Western blotting using specific antibodies against SPR4. Equal loading of protein was monitored using a specific antibody to GAPDH. Similar results were obtained in each of independent experiments. J: results of scanning densitometry of exposed film. Data are expressed as arbitrary units of intensity and are means range of independent experiments. K: SPR5 sirna inhibitory efficiency was confirmed by Western blotting using specific antibodies against SPR5. Equal loading of protein was monitored using a specific antibody to GAPDH. Similar results were obtained in each of independent experiments. L: results of scanning densitometry of exposed film. Data are expressed as arbitrary units of intensity and are means range of independent experiments. by...6 on December, 7 present strong evidence for a novel action of CP: stimulation of MCP- secretion. This chemokine plays an important role in the recruitment of mononuclear cells into sites of inflammation and has been associated with tumor metastasis (7, 4, 57). This proinflammatory action of CP is consistent with previous work by Pettus and coworkers (4 45), who found that this phosphosphingolipid potently stimulates calcium-dependent cytosolic phospholipase A (cpla ) activity and the subsequent generation of arachidonic acid and prostaglandins leading to inflammation. Although CP can increase synthesis of prostaglandins, it is unlikely that they participate in the stimulation of MCP- production and stimulation of cell migration by CP, as prostaglandins were reported to inhibit MCP- production (5). We have confirmed this latter observation in the J774A. macrophages, and have also observed that CPstimulated macrophage migration is inhibited by prostaglandin
E A MCP- release (pg/ 6 cells) B C 5 4 4 5 [CP] (μm) 4 [CP] (μm) 4 Control Vehicle Anti-MCP- Ab E (unpublished observations). MCP- has also been associated with stimulation of cell proliferation in both healthy and malignant cells (7, 55), and although it did not affect the growth of bladder cancer cells, it mediated migration and CP Fig. 4. CP stimulates MCP- release and cell migration in human THP- monocytes. A: cells were seeded in 4-well plates ( 5 cells/well) in RPMI as described in MATERIALS AND METHODS. After h, monocytes were incubated for 4 h with indicated concentrations of CP. MCP- concentration was measured by ELISA as indicated in MATERIALS AND METHODS. MCP- values were normalized to total cell number, and results are expressed as means SE of 4 independent experiments performed in duplicate. B: macrophage migration was measured by Boyden chamber-based cell migration assay as described in MATERIALS AND METHODS. Cells ( 5 cells/well) were plated in the upper wells of 4-well chambers coated with fibronectin and incubated for h. Cells were then stimulated with indicated concentrations of CP and incubated further for 4 h. Results are expressed relative to control value and are means SE of 4 independent experiments performed in duplicate. C: cells were seeded as in B and incubated without (open bars) or with.5 g/ml anti-mcp- antibody (filled bars) in the absence or presence of M CP as indicated. Results are means SE of 4 independent experiments performed in duplicate (P.5, P.). invasion of those cells (9) and facilitated metastasis to bone, lung, and prostate (6, 6, 47, 57). There are several pathways involved in the regulation of secretion of chemokines, and we showed previously that CP A MCP- release (ng/ 6 cells) B C 75 5 5 75 5 5 [CP] (μm) [CP] (μm).5..5..5..5. Vehicle Anti-MCP- Ab Control CP Fig. 5. CP stimulates MCP- release and cell migration in T-L preadipocytes. A: cells were seeded in 4-well plates ( 4 cells/well) in DMEM supplemented with % FCS overnight. Then, medium was replaced by DMEM without FCS and further incubated for h. Cells were then incubated for 4 h with indicated concentrations of CP. MCP- concentration was measured by ELISA as indicated in MATERIALS AND METHODS. MCP- values were normalized to total cell number, and results are expressed as means SE of 4 independent experiments performed in duplicate. B: T-L migration was measured by Boyden chamber-based cell migration assay as described in MATERIALS AND METHODS. Cells (5 4 cells/well) were plated in the upper wells of 4-well chambers coated with fibronectin and incubated for h. Cells were stimulated with indicated concentrations of CP and incubated further for 4 h. Results are expressed relative to control value and are means SE of 4 independent experiments performed in duplicate. C: cells were seeded as in B and incubated without (open bars) or with.5 g/ml anti-mcp- antibody (filled bars) in the absence or presence of M CP as indicated. Results are means SE of 4 independent experiments performed in duplicate (P.5, P.). Downloaded from http://ajpendo.physiology.org/ by...6 on December, 7
E4 could activate some of these pathways (, 4). In particular, we found that CP was able to stimulate PIK/Akt (PKB) and MEK/ERK/, and that these pathways were involved in the mitogenic (5) and chemotactic () effects of CP. In the present study, we demonstrate that CP-stimulated MCP- release is mediated by these pathways. In addition, we found that CP promoted phosphorylation of the MAP kinase p8 and that blockade of this enzyme activity using sirna or chemical inhibition completely abrogated CP-stimulated MCP- release. Moreover, CP-stimulated MCP- release was completely abolished by PTX, an action that is compatible with the existence of a specific G i protein-coupled receptor for CP (). The implication of Akt, ERK/, and p8 in this CP action is consistent with previous work by other groups showing that MCP- upregulated these kinases in different cell types, including rat aortic smooth muscle cells (55), proximal tubular cells (5), T-L preadipocytes (56), vascular smooth muscle cells (), and monocytes (). Nonetheless, although MCP- release has been associated with activation of p8 in many instances (8,, 56; and our unpublished observations), other studies have indicated that this kinase is not involved in this process (, 5), and therefore the role of p8 in stimulation of MCP- secretion remains controversial. Of note, two related phosphosphingolipid mediators, sphingosine phosphorylcholine (SPC) and SP, also induced secretion of proinflammatory MCP- in human umbilical vein endothelial cells (4) and vascular smooth muscle cells (54), and human mast cells (4), respectively. As for CP, the stimulating effect of SPC on MCP- secretion was also dependent on activation of a specific, albeit still not well-defined, receptor (4). However, unlike CP, the stimulation of MCP- release by SP was not dependent on receptor activation (4). In addition, we demonstrate here that CP-stimulated MCP- release and cell migration are independent of interaction of CP with SP receptors. A key observation in this work was that incubation of the macrophages with an MCP--neutralizing antibody or with MCP- sirna or the MCP- receptor antagonist RS-895 or CCRb-targeted sirna abrogated CP-stimulated cell migration, thereby demonstrating that MCP- is the principal mediator of CP-stimulated macrophage migration. These findings are consistent with recent work by Mitsutake et al. (9), who recently demonstrated that MCP--induced cell migration was significantly reduced in BMDM from mice lacking ceramide kinase, the enzyme responsible for CP generation. We also observed that CP was able to stimulate migration of BMDM, but this effect was only marginal compared with the extent of activation seen in J774A. macrophages, in agreement with Mitsutake et al. (data not shown). Although the molecular identity of the target receptor whose activation by CP leads to MCP- release and subsequent migration of the macrophages is not yet clear, we show here that a PTX-sensitive G protein-coupled receptor is likely to be involved in mediating these effects of CP. In fact, PTX blocked CP-stimulated macrophage migration, which is consistent with the inhibition of CP-stimulated MCP- release by this toxin and with our previous work on RAW 64.7 cells (). These observations suggest that MCP- release depends on the interaction of CP with its putative receptor to trigger macrophage migration. In terms of the signaling events downstream of CP stimulation in macrophages, our previous work showed that PIK/ Akt, MEK/ERK/, and JNK are involved in the mitogenic or antiapoptotic effects of CP, and these pathways are also known to act downstream of the CP receptor (). In the present work, we found that PIK/Akt, MEK/ERK, and their downstream target NF- B are involved in the stimulation of cell migration by CP. However, inhibition of JNK did not alter macrophage migration. Of interest, and unlike PIK/Akt and MEK/ERK, JNK was also not involved in upregulation of MCP- expression in endometrial stroma cells (5). Finally, it should be emphasized that the effects of CP on MCP- release and cell migration are not restricted to murine macrophages, as CP also promoted these actions in human THP- monocytes and T-L preadipocytes. Moreover, recent work by Ratajczak s group demonstrates that CP also stimulates migration of hematopoietic stem progenitor cells (), multipotent stromal cells, and endothelial progenitor cells (9), as well as migration of bone marrow-derived stem cells in patients suffering from acute myocardial infarction (8). In conclusion, we have demonstrated here that CP promotes MCP- release and cell migration in macrophages. This stimulatory effect of CP is accomplished through activation of the PIK/Akt, MEK/ERK/, and p8 pathways, but not JNK, via PTX-sensitive G proteins, leading to NF- B activation. CP also stimulated MCP- release and migration in human THP- monocytes and T-L preadipocytes. Because MCP- is proinflammatory and may be involved in tumorigenesis and tumor metastasis, CP and its putative plasma membrane receptor may have potential as therapeutic targets to combat inflammation and cancer. ACKNOWLEDGMENTS We are grateful to Servicios Generales de investigación (SGIker) and Unidad de formación e investigación (UFI) / (UPV/EHU) for technical support. GRANTS This work was supported by Grants BFU9-4/BFI from Ministerio de Ciencia e Innovación (MICINN) (Madrid, Spain), IT-75- from Departamento de Educación, Universidades e Investigación del Gobierno Vasco (GV/EJ, Spain), S-PEUN7 and S-PEUN4 from Departamento de Industria, Comercio y Turismo del Gobierno Vasco (Basque Government, GV/EJ, Spain). L. Arana and A. Ouro are recipients of fellowships from the Basque Government. I.-G. Rivera is the recipient of a fellowship from MICINN, and M. Ordoñez is the recipient of a fellowship from the University of the Basque Country (GV/EJ, Spain). DISCLOSURES No conflicts of interest, financial or otherwise, are declared by the author(s). AUTHOR CONTRIBUTIONS Author contributions: L.A., M.O., A.O., I.-G.R., and P.G. performed experiments; L.A., M.O., A.O., I.-G.R., P.G., M.T., and A.G.-M. analyzed data; L.A., M.O., A.O., I.-G.R., P.G., M.T., and A.G.-M. interpreted results of experiments; L.A. prepared figures; L.A. and A.G.-M. drafted manuscript; L.A., M.O., P.G., M.T., and A.G.-M. edited and revised manuscript; L.A., M.O., A.O., I.-G.R., P.G., M.T., and A.G.-M. approved final version of manuscript; A.G.-M. conception and design of research. REFERENCES. Aman A, Piotrowski T. Cell migration during morphogenesis. Dev Biol 4:, 9.. Arana L, Gangoiti P, Ouro A, Rivera IG, Ordonez M, Trueba M, Lankalapalli RS, Bittman R, Gomez-Munoz A. Generation of reactive Downloaded from http://ajpendo.physiology.org/ by...6 on December, 7