Ceramide kinase deficiency impairs microendothelial cell angiogenesis in vitro

Microvascular Research 77 (2009) 389–393

Contents lists available at ScienceDirect

Microvascular Research j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / y m v r e

Short Communication

Ceramide kinase deficiency impairs microendothelial cell angiogenesis in vitro Satoru Niwa, Christine Graf, Frédéric Bornancin ⁎ Novartis Institutes for BioMedical Research, Brunnerstrasse 59, A-1235 Vienna, Austria

a r t i c l e

i n f o

Article history: Received 5 September 2008 Revised 18 December 2008 Accepted 14 January 2009 Available online 6 February 2009 Keywords: Angiogenesis Ceramide Ceramide kinase Ceramide-1-phosphate Sphingosine-1-phosphate VEGF

a b s t r a c t The recent generation of ceramide kinase (CerK)-deficient (Cerk −/−) mice as well as the identification of the potent CerK inhibitor NVP-231 have provided unprecedented opportunities to better understand CerK biology. Here we used skin dermal microendothelial cells (DMECs) and we show that CerK activity regulates their neovascularization in a matrigel environment in vitro. Capillary-like tube formation was significantly impaired in CerK-deficient cells or in wild-type (WT) cells treated with NVP-231 as compared with untreated WT cells. This was not the result of compromised proliferation or survival because Cerk −/− endothelial cells were able to migrate out of dermal fragments and grow in monolayer culture as well as their WT counterpart. Vascular endothelial growth factor, fibroblast growth factor or tumor necrosis factor could not rescue the angiogenesis defect observed in Cerk −/− DMEMs. Moreover, CerK ablation increased serum ceramide levels at the expense of dihydroceramide levels without affecting sphingosine, dihydrosphingosine, sphingosine-1phosphate or dihydrosphingosine-1-phosphate levels. These observations collectively suggest that CerKcatalyzed formation of C1P may regulate angiogenesis by a novel mechanism that is independent of S1P formation and signaling. © 2009 Elsevier Inc. All rights reserved.

Introduction Sphingolipids have been implicated in a myriad of physiological processes including cell growth or death, cell adhesion, cell migration, inflammation, and angiogenesis (Hannun and Obeid, 2008; Takabe et al., 2008). Angiogenesis, the sprouting of new blood vessels from existing ones, is a key process for progression of proliferative and autoimmune disorders. Over the last years, evidence has accumulated pointing to an important role of sphingolipids in regulation of angiogenesis. One of the foremost studied sphingolipid is sphingosine-1-phosphate (S1P). Together with one of its producing enzyme, sphingosine kinase 1 (SphK1) and the G-protein coupled receptors that bind S1P (S1PR), the SphK1–S1P–S1PR axis has proved key to multiple signaling pathways, in particular in relation to vascular biology. S1P interacts with S1P receptors (Lee et al., 1998) and cooperates with vascular endothelial growth factor (VEGF) and fibroblast growth factor (FGF) to regulate angiogenesis, by inducing formation of adherens junctions and by regulating endothelial cell morphogenesis (Lee et al., 1999a, 1999b). S1P is also a recognized mediator of vascular tone. SphK1 plays a role in endothelial cell survival (Limaye et al., 2005) and is also known to regulate vessel contractility (Bolz et al., 2003). The

Abbreviations: Cer, ceramide; CerK, ceramide kinase; C1P, ceramide-1-phosphate; DMB-Cer, N-(5-(5,7-dimethyl BODIPY)-l-pentanoyl)-D-erythro-sphingosine; DMECs, dermal microvascular endothelial cells; FCS, fetal calf serum; S1P, sphingosine-1-phosphate; S1PR, S1P receptor; SphK, sphingosine kinase; VEGF, vascular endothelial growth factor. ⁎ Corresponding author. Novartis Institutes for BioMedical Research, Forum 1, CH4056 Basle, Switzerland. Fax: +41 61 32 40562. E-mail address: [email protected] (F. Bornancin). 0026-2862/$ – see front matter © 2009 Elsevier Inc. All rights reserved. doi:10.1016/j.mvr.2009.01.006

S1P1 receptor is highly expressed on endothelial cells and is known to play a role in angiogenesis. Knockout of S1P1R in the mouse resulted in embryonic lethality due to poor vascular development (Liu et al., 2000). Moreover, the knockout of both SphK1 and SphK2 resulted in a similar phenotype (Mizugishi et al., 2005). These studies provided compelling evidence for a master role of S1P in vascular biology. In addition to S1P, there may be other sphingolipids whose regulation impacts on vascular processes. Lactosylceramide and lactosyl ceramide synthase for instance have recently been shown to regulate angiogenesis in vitro (Rajesh et al 2005). Ceramide-1phosphate (C1P) which is produced by ceramide kinase (CerK) ― and an as yet unidentified activity (Graf et al., 2008a; Boath et al., 2008) ― is an intriguing sphingolipid metabolite. CerK-deficient (Cerk −/−) mice displayed decreased neutrophil counts and dysregulated ceramide and dihydroceramide levels together with decreased C1P levels (Graf et al., 2008a). Whether deficiency in CerK may trigger additional vascular defects has not been investigated yet. Using skin dermal microvascular endothelial cells (DMECs) from WT and Cerk −/− animals (Graf et al., 2008a) as well as the potent CerK inhibitor NVP231 (Graf et al., 2008b) we now present evidence that production of C1P by CerK may play a role in angiogenesis.

Materials and methods Materials NVP-231 was synthesized at Novartis Institutes for BioMedical Research, Vienna (Graf et al., 2008b). Growth factor-reduced Matrigel,

390

S. Niwa et al. / Microvascular Research 77 (2009) 389–393

dispase and type I collagen coated flasks were purchased from BD Pharmingen. Mouse endothelial cell growth medium (mECGM) was obtained from Promocell. Mouse vascular endothelial growth factor (VEGF), mouse basic fibroblast growth factor (bFGF), and mouse tumor necrosis factor alpha (TNFα) were obtained from R&D systems. DMB-Cer was from Molecular Probes. Isolation and culture of dermal microvascular endothelial cells (DMECs) Isolation and purification of mouse DMECs was achieved according to the method reported by Cha et al. (2005), with some modifications. Newly born mice at 4 to 7 days of age were sacrificed and skin from the torso region was removed and placed in RPMI 1640 medium (Gibco) for preservation. Fat tissue from the skin was carefully discarded and tissue samples were incubated in dispase for 30 min until the epidermal layer could be peeled and discarded. Skin samples treated with dispase were minced with a scalpel in mECGM supplemented with 2 mM L-glutamine, 10 mM HEPES, 100 U/ml penicillin– streptomycin and 0.1% of 2-mercaptoethanol, and were passed through a 200 μm nylon mesh (Spectrum Laboratories, #146487). The suspension was spun down at 4 °C, 1200 rpm for 10 min, and was washed twice with mECGM. The final suspension ― pooled preparation from 4 to 6 mice ― was seeded onto type I collagen-coated plates, and cultivated at 37 °C in a 5% CO2/95% atmosphere over night. After one day the plate was washed twice with calcium and magnesium-free HBSS (HBSS−) (Gibco) and cells were further cultivated in mECGM until confluence. Confluent cells were detached with 0.25% TrypsinEDTA (Gibco). The cell suspension washed twice with HBSS− was layered onto a sterile gradient solution separately prepared by centrifuging 8 ml of a 35% Percoll solution (Amersham Pharmacia Biotech) at 30,000 ×g for 15 min at 4 °C, and centrifuged for 10 min at 400 ×g at room temperature. In parallel to the gradient tube containing the cells, another gradient tube was used as density reference by substituting the sample with 200 μl of a mixture of colored density marker beads (GE Healthcare). Endothelial cells lay in the gradient between densities 1.033 g/ml and 1.047 g/ml, whereas non-endothelial cells have a density greater than 1.065 g/ml (Cha et al., 2005). The cell band at the appropriate density ― representing ≥90% pure DMECs ― was then collected, washed in HBSS− twice, counted and plated at 7500 cells/cm2 on type I collagen-coated plates. DMECs were grown until sub-confluence and used for tube formation assays within four passages. CerK activity assay DMECs at passage 3 were seeded at 3 × 105 cells per well of a 6-well plate. Twenty four hours later, 5 μM DMB-Cer was added to the medium and incubation was allowed to proceed for 1 h. After removal of the medium, lipid extraction and running of the full lipid extract on thin layer chromatography plates was performed, as described in Boath et al. (2008). NVP-231, when used, was added to the culture medium at the same time as DMB-Cer. Capillary-like tube formation assay Three hundred μl of sterile growth factor-reduced Matrigel containing 5 U/ml of heparin was homogenized and layered into a sterile 8-chamber glass slide wells (BD Falcon) on a cooled planar surface. Matrigel in the chambers was allowed to solidify at 37 °C. Mouse DMECs were detached with trypsin-EDTA, washed with HBSS− twice, and re-suspended in RPMI 1640 medium supplemented with 0.1% of BSA. Mouse DMECs were seeded into chambers at 1 × 105 cells/ well, and cultivated at 37 °C atmosphere with 5% CO2/95% air over night. On the following day DMECs were fixed using 4% of formaldehyde solution, and black and white images of chambers were captured under a light microscope (Nikon S2000 with attached

Fig. 1. CerK activity in Cerk −/− and NVP-231-treated DMECs. Mouse DMECs grown in monolayer were incubated with 5 μM DMB-Cer for 1 h. Lipids were extracted and analyzed on thin layer chromatography. Left, WT DMECs ± 100 nM NVP-231; Right, Cerk −/− DMECs ± 100 nM NVP-231. DMB-Cer and DMB-C1P standards are shown on each panel. (⁎) denotes an uncharacterized lipid species.

SFN1 Nikon digital camera) at ×4 magnification. Three images of each chamber well were randomly taken for quantitative analysis of tubeformed area. Images were then analyzed by counting the tube area using Scion Image, NIH image software (Windows version alpha 4.0.3.2.). After elimination of background signals, the area of capillaries was quantified by counting the number of white pixels. Results Elimination of CerK activity in DMECs by genetic ablation or treatment with NVP-231 CerK activity in DMECs was analyzed using a validated assay that specifically traces C1P produced by CerK (Boath et al., 2008). In this assay, a cell-permeable fluorescently labeled short chain ceramide (here DMB-Cer) is applied exogenously to cells and is delivered to intracellular organelles where it can be metabolized by various enzymes including CerK. DMB-C1P was readily detected in WT DMECs incubated with DMB-Cer (Fig. 1). In Cerk −/− DMECs, DMB-C1P was not formed (Fig. 1, right). NVP-231 is a recently identified inhibitor of CerK that potently inhibits CerK-catalyzed production of C1P both in vitro and in cellbased assays (Graf et al., 2008b). Treatment of WT DMECs with 100 nM NVP-231 completely abolished CerK activity (Fig. 1, left). Thus, by genetic means or use of an inhibitor such as NVP-231, it is possible to obtain DMECs that are permanently or transiently deficient in CerK activity, respectively. Defective in vitro angiogenesis in Cerk

−/−

DMECs

Characterization of Cerk −/− mice so far has provided evidence for a role of CerK in lung bacterial clearance (Graf et al., 2008a) and emotional behavior (Mitsutake et al., 2007). Using DMECs we now addressed the capacity of Cerk −/− cells in a neovascularisation assay in vitro as compared with WT cells. WT cells seeded on Matrigel were able to build a capillary-like network after 24 h. Under the same conditions, the tubule network obtained with Cerk −/− cells was significantly reduced, as can be seen in the limited number of branching tubules (Fig. 2A). Of note, the lower capacity of Cerk −/− cells to neovascularize did not result from decreased survival because the reduced tubular network was paralleled by an increase in cell monolayer area (Figs. 2A and B). Thus, the capillary/monolayer ratio was twice as high in WT cells as compared to Cerk −/− cells (Table 1). Vascular endothelial cell growth factor (VEGF) is instrumental in the regulation of angiogenesis and has been shown to control many processes including basement membrane degradation, migration and proliferation of vascular endothelial cells. VEGF is also critically involved in the regulation of S1P signaling pathway through up-

S. Niwa et al. / Microvascular Research 77 (2009) 389–393

391

Fig. 2. Tube formation assay with WT and Cerk −/− DMECs. DMECs were seeded and cultivated overnight on Matrigel chambers with or without 100 ng/ml mouse VEGF. The data show representative images in (A) (4-fold magnification; 28-fold for the area within the white squares that were expanded and are shown on the right) as well as the quantitative analyses that express the average ± SEM of three chamber wells for each condition in (B). Statistical significance was evaluated using a t-test, with Kruskal–Wallis corrections for multiple comparisons (⁎⁎: P b 0.01, ⁎: P b 0.05). This is one of three independent experiments with similar results. The black arrows point to differences in diameter of the tubes formed by WT and Cerk −/− DMECs respectively, suggesting a higher immaturity index (Linz-McGillem et al., 2004) for Cerk −/− DMEC-derived tubes.

regulation of S1P1 receptor on vascular endothelial cells (Igarashi et al., 2003). Addition of VEGF increased tube formation both with WT and Cerk −/− cells but did not change the propensity of Cerk −/− cells to accumulate in monolayer areas at the expense of tubule areas (Figs. 2A, B and Table 1). Moreover, even in the presence of VEGF, the tubes formed by Cerk −/− cells showed evidence for immaturity as seen in their increased thickness compared with those produced by WT cells (Fig. 2A, right). Basic FGF and TNFα are other potent pro-angiogenic factors particularly relevant for development of inflammatory and autoimmune disorders (Yoshida et al., 1997). Similarly to VEGF, neither bFGF nor TNFα could rescue the impaired tube formation displayed by Cerk −/− cells (Fig. 3).

of monolayer patches that remained under these conditions (Figs. 4A, B and Table 1). Discussion It is noteworthy that CerK deficiency or CerK inhibition did not reduce cell viability because C1P has been characterized as a

Table 1 Regulation of capillary formation by CerK Ratios

Capillary vs monolayer

WT vs Cerk

No VEGF

WT Cerk −/− WT Cerk −/− WT WT + NVP-231

2.08 ± 0.60

+ VEGF

0.81 ± 0.13 0.39 ± 0.07 3.02 ± 0.66a 1.28 ± 0.33 9.58 ± 1.70a 1.60 ± 0.34

−/−

WT vs WT + NVP-231

2.36 ± 0.90

NVP-231 inhibits DMEC angiogenesis in vitro

+ VEGF

Cerk −/− cells have an established deficiency in CerK that may trigger compensatory pathways to adapt to reduced C1P or enhanced ceramide levels. To evaluate the impact of acute inhibition of CerK we performed the in vitro neovascularisation assay using WT mDMECs treated with NVP-231. Consistent with the observations in Cerk −/− cells, addition of 100 nM NVP-231 impaired tube formation by WT DMECs as seen in the poor network produced as well as in the number

Data presented in Figs. 2 and 4 were analyzed to compare the ratios of capillary versus monolayer growth in WT and Cerk −/− DMECs as well as in WT and NVP-231-treated DMECs. Ratio values ± SEM are indicated. VEGF increased by more than 3-fold on average the capillary/monolayer ratio in both WT and Cerk −/− DMECs. a The experiments that compared WT with NVP-231-treated WT DMECs were performed at a different time point than those that compared WT with Cerk–/– DMECs; distinct matrigel batches were used that did not support tube formation to the same extent.

5.99 ± 1.93

392

S. Niwa et al. / Microvascular Research 77 (2009) 389–393

Fig. 3. Effect of VEGF, bFGF or TNFα on the impaired angiogenesis of Cerk −/− DMECs. The tube formation assay was performed as described in the legend of Fig. 2 in the presence of either 100 ng/ml VEGF, 100 ng/ml bFGF, or 30 ng/ml TNFα.

proliferative/pro-survival lipid metabolite (reviewed in GomezMunoz, 2006). In both Cerk −/− and NVP-231-treated cells, the density of the cell monolayer area was in fact increased at the expense of the tube forming area (Figs. 2–4 and Table 1). These observations

are consistent with previous findings showing that bone-marrow derived macrophages or mast cells isolated from Cerk −/− mice do survive as well as WT cells in culture. It is unknown yet if the survival of CerK-deficient cells results from sufficient residual C1P levels (Graf et al., 2008a) or from the existence of compensatory or even redundant pathways used for cell survival. The preparation of endothelial cells for the matrigel-based tubular assay requires first that the cells migrate out of dermal fragments and cover the plate where there are let to grow for 2 to 3 weeks. Neither in their ability to migrate nor in their ability to expand as a monolayer did we observe any defect in Cerk −/− cells. Altogether, this indicates that the effects of CerK deficiency or CerK inhibition on the differentiation into capillary like structures are probably not resulting from compromised cell proliferation, cell survival or ability to migrate. What hinders tubule formation under conditions of CerK inhibition may rather result from impairment to commit to tubular assembly or from inefficiency during the assembly process itself, or even from inefficient tubule maturation; this will need to be further investigated. S1P levels were not decreased in the serum of Cerk −/− mice (Table 2), suggesting that CerK ablation does not interfere with S1P regulation. In fact, the observation that Cerk −/− DMECs survive as well as WT cells contrasts with what is known about the capacity of SphK/S1P to regulate endothelial cell survival (Peters and Alewijnse, 2007; Limaye, 2008). It is therefore likely that S1P and C1P use distinct mechanisms. S1P produced by SphK1 is known to regulate angiogenesis after extracellular release and subsequent binding to S1P receptors (reviewed in Takabe et al., 2008). When we looked if C1P might activate S1P receptors we found that ≥5 μM amounts are required, depending on the ceramide chain length (F. Bassilana, D. Guerini, K. Seuwen and F. Bornancin, unpublished observations); these

Table 2 Sph and S1P levels in the serum of Cerk

Fig. 4. NVP-231 inhibits VEGF-stimulated DMEC tube formation in vitro. DMECs were seeded and cultivated overnight on Matrigel chamber with 100 ng/ml mouse VEGF in the presence or absence of 100 nM of NVP-231. Representative images (48-fold magnification), quantitative analysis and statistical significance were obtained as described in the legend of Fig. 2. This is one of three independent experiments with similar results.

−/−

and WT animals −/−

Lipid

WT

Cerk

DH-Sph DH-S1P Sph S1P DH-Cer Cer

0.05 ± 0.01 0.78 ± 0.17 0.10 ± 0.06 2.69 ± 0.43 0.537 ± 0.187 0.906 ± 0.319

0.05 ± 0.03ns 0.81 ± 0.15ns 0.06 ± 0.01ns 2.46 ± 0.14ns 0.127 ± 0.053⁎ 1.512 ± 0.324⁎

p 0.48 0.44 0.16 0.23 0.017 0.044

Sphingolipid levels in the serum (expressed in μM) were measured by liquid chromatography followed by mass spectrometry analysis, as described in Zemann et al. (2007), and Graf et al. (2008a). There is no significant difference (ns) in the levels of DH-Sph, DH-S1P, Sph and S1P levels between Cerk −/− and WT animals. DH-Cer and Cer levels represent the sum of all species with the following chain length: C12, C16, C18, C18:1, C24 and C24:1. DH-Cer and Cer levels are significantly (⁎) reduced and enhanced, respectively, in Cerk −/− animals compared with WT animals. Data represent the mean ± SD of levels measured in 4 WT and 4 Cerk −/− animals. Statistical significance was obtained using t tests. DH, dehydro.

S. Niwa et al. / Microvascular Research 77 (2009) 389–393

concentrations are likely beyond physiological levels. Alternatively, C1P might act on putative C1P receptors. However, there is no evidence to date that natural C1P can be released into the extracellular space. VEGF was able to stimulate capillary formation to the same extent in WT and Cerk −/− cells (Fig. 2 and Table 1). This suggests that CerK may regulate angiogenesis independently of VEGF signaling. This may hold true with bFGF and TNFα as well, based on the results presented in Fig. 3. Because VEGF impacts on S1P signaling by up-regulating S1P1 receptors (Igarashi et al., 2003) this further indicates that the S1P1 receptor is not involved in CerK-mediated regulation of tube formation in vitro. We previously showed that VEGF production in Cerk −/− macrophages that had been stimulated with 5′-N-ethylcarboxamidoadenosine and lipopolysaccharide, was unchanged compared with WT cells (Graf et al., 2008a). Altogether, this suggests that both VEGF signaling and production of VEGF remain unaffected after CerK ablation. In conclusion, we showed evidence that CerK/C1P may represent novel regulators of angiogenesis in vitro. These findings call for further experiments using in vivo settings. Given their dysregulated ceramide metabolism, it may prove valuable to challenge Cerk −/− animals in order to uncover vascular-related phenotypes. Because of the dual function of C1P production ― depletion of Cer and generation of a signaling intermediate ― the underlying principle used by CerK to regulate angiogenesis remains to be clarified. Acknowledgments We thank Claudia Reichel and Manuela Pillinger for expert technical assistance, and Roland Reuschel's laboratory for measurement of lipid levels. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.mvr.2009.01.006. References Boath, A., Graf, C., Lidome, E., Ullrich, T., Nussbaumer, P., Bornancin, F., 2008. Regulation and traffic of ceramide-1-phosphate produced by ceramide kinase: comparative analysis to glucosylceramide and sphingomyelin. J. Biol. Chem. 283, 8517–8526. Bolz, S.S., Vogel, L., Sollinger, D., Derwand, R., Boer, C., Pitson, S.M., Spiegel, S., Pohl, U., 2003. Sphingosine kinase modulates microvascular tone and myogenic responses through activation of RhoA/Rho kinase. Circulation 108, 342–347. Cha, S.T., Talavera, D., Demir, E., Nath, A.K., Sierra-Honigmann, M.R., 2005. A method of isolation and culture of microvascular endothelial cells from mouse skin. Microvasc. Res. 70, 198–204.

393

Gomez-Munoz, A., 2006. Ceramide 1-phosphate/ceramide, a switch between life and death. Biochim. Biophys. Acta 1758, 2049–2056. Graf, C., Zemann, B., Rovina, P., Urtz, N., Schanzer, A., Reuschel, R., Mechtcheriakova, D., Müller, M., Fischer, E., Reichel, C., Huber, S., Dawson, J., Meingassner, J.G., Billich, A., Niwa, S., Badegruber, R., Van Veldhoven, P.P., Kinzel, B., Baumruker, T., Bornancin, F., 2008a. Neutropenia with impaired immune response to Streptococcus pneumoniae in ceramide kinase-deficient mice. J. Immunol. 180, 3457–3466. Graf, C., Klumpp, M., Habig, M., Rovina, P., Billich, A., Baumruker, T., Oberhauser, B., Bornancin, F., 2008b. Targeting ceramide metabolism with a potent and specific ceramide kinase inhibitor. Mol. Pharmacol. 74, 925–932. Hannun, Y.A., Obeid, L.M., 2008. Principles of bioactive lipid signalling: lessons from sphingolipids. Nat. Rev. Mol. Cell. Biol. 9, 139–150. Igarashi, J., Erwin, P.A., Dantas, A.P.V., Chen, H., Michel, T., 2003. VEGF induces S1P1 receptors in endothelial cells: implications for cross-talk between sphingolipid and growth factor receptors. Proc. Natl. Acad. Sci. U. S. A. 100, 10664–10699. Lee, M.J., Van Brocklyn, J.R., Thangada, S., Liu, C.H., Hand, A.R., Menzeleev, R., Spiegel, S., Hla, T., 1998. Sphingosine-1-phosphate as a ligand for the G protein-coupled receptor EDG-1. Science 279, 1552–1555. Lee, M.J., Thangada, S., Claffey, K.P., Ancellin, N., Liu, C.H., Kluk, M., Volpi, M., Sha'afi, R.I., Hla, T., 1999a. Vascular endothelial cell adherens junction assembly and morphogenesis induced by sphingosine-1-phosphate. Cell 99, 301–312. Lee, O.H., Kim, Y.M., Lee, Y.M., Moon, E.J., Lee, D.J., Kim, J.H., Kim, K.W., Kwon, Y.G., 1999b. Sphingosine 1-phosphate induces angiogenesis: its angiogenic action and signaling mechanism in human umbilical vein endothelial cells. Biochem. Biophys. Res. Commun. 264, 743–750. Limaye, V., 2008. The role of sphingosine kinase and sphingosine-1-phosphate in the regulation of endothelial cell biology. Endothelium 15, 101–112. Limaye, V., Li, X., Hahn, C., Xia, P., Berndt, M.C., Vadas, M.A., Gamble, J.R., 2005. Sphingosine kinase-1 enhances endothelial cell survival through a PECAM-1dependent activation of PI-3K/Akt and regulation of Bcl-2 family members. Blood 105, 3169–3177. Linz-McGillem, L.A., Moitra, J., Garcia, J.G., 2004. Cytoskeletal rearrangement and caspase activation in sphingosine 1-phosphate-induced lung capillary tube formation. Stem Cells Dev. 13, 496–508. Liu, Y., Wada, R., Yamashita, T., Mi, Y., Deng, C.X., Hobson, J.P., Rosenfeldt, H.M., Nava, V.E., Chae, S.S., Lee, M.J., Liu, C.H., Hla, T., Spiegel, S., Proia, R.L., 2000. Edg-1, the G protein-coupled receptor for sphingosine-1-phosphate, is essential for vascular maturation. J. Clin. Invest. 106, 951–961. Mitsutake, S., Yokose, U., Kato, M., Matsuoka, I., Yoo, J.M., Kim, T.J., Yoo, H.S., Fujimoto, K., Ando, Y., Sugiura, M., Kohama, T., Igarashi, Y., 2007. The generation and behavioral analysis of ceramide kinase-null mice, indicating a function in cerebellar Purkinje cells. Biochem. Biophys. Res. Commun. 363, 519–524. Mizugishi, K., Yamashita, T., Olivera, A., Miller, G.F., Spiegel, S., Proia, R.L., 2005. Essential role for sphingosine kinases in neural and vascular development. Mol. Cell. Biol. 25, 11113–11121. Peters, S.L., Alewijnse, A.E., 2007. Sphingosine-1-phosphate signaling in the cardiovascular system. Curr. Opin. Pharmacol. 7, 186–192. Rajesh, M., Kolmakova, A., Chatterjee, S., 2005. Novel role of lactosylceramide in vascular endothelial growth factor-mediated angiogenesis in human endothelial cells. Circ. Res. 97, 796–804. Takabe, K., Paugh, S.W., Milstien, S., Spiegel, S., 2008. “Inside-out” signaling of sphingosine-1-phosphate: therapeutic targets. Pharmacol. Rev. 60, 181–195. Yoshida, S., Ono, M., Shono, T., Izumi, H., Ishibashi, T., Suzuki, H., Kuwano, M., 1997. Involvement of interleukin-8, vascular endothelial growth factor, and basic fibroblast growth factor in tumor necrosis factor alpha-dependent angiogenesis. Mol. Cell. Biol. 17, 4015–4023. Zemann, B., Urtz, N., Reuschel, R., Mechtcheriakova, D., Bornancin, F., Badegruber, R., Baumruker, T., Billich, A., 2007. Normal neutrophil functions in sphingosine kinase type 1 and 2 knockout mice. Immunol. Lett. 109, 56–63.