* To whom correspondence should be addressed.
Received August 18, 1997
Recent evidence suggests that branching pathways of sphingolipid metabolism may mediate either apoptotic or mitogenic responses depending on the cell type and the nature of the stimulus. While ceramide has been shown to be an important regulatory component of apoptosis induced by tumor necrosis factor alpha and the Fas ligand, sphingosine-1-phosphate (SPP), a further metabolite of ceramide, has been implicated as a second messenger in cellular proliferation and survival induced by platelet-derived growth factor, neuronal growth factor, and serum. SPP protects cells from apoptosis resulting from elevations of ceramide. Inflammatory cytokines stimulate sphingomyelinase, but not ceramidase, leading to accumulation of ceramide, whereas growth signals also stimulate ceramidase and sphingosine kinase leading to increased SPP levels. We propose that the dynamic balance between levels of sphingolipid metabolites, ceramide, and SPP and consequent regulation of different members of the mitogen-activated protein kinases (JNK versus ERK) family is an important factor that determines whether a cell survives or dies.
KEY WORDS: sphingosine-1-phosphate, apoptosis, cell growth, signal transduction
Abbreviations: DAG) diacylglycerol; DMS) N,N-dimethylsphingosine; EGF) epidermal growth factor; MAPK) mitogen-activated protein kinases; NGF) nerve growth factor; PDGF) platelet-derived growth factor; PLD) phospholipase D; PKC, PKA) protein kinases C and A, respectively; SPP) sphingosine-1-phosphate; TNF) tumor necrosis factor; TPA) 12-O-tetradecanoylphorbol 13-acetate; SAPK) stress-activated protein kinases; SMC) smooth muscle cells.
SPHINGOSINE-1-PHOSPHATE IN CELL GROWTH REGULATION
Sphingolipid metabolites can mediate either mitogenic or apoptotic effects depending on the cell type and the nature of the stimulus [1-3]. Ceramide has emerged as a regulatory component of apoptosis induced by ionizing radiation and by members of the TNF (tumor necrosis factor) superfamily, TNF-alpha and Fas ligand [4-9]. However, a further metabolite of ceramide, sphingosine-1-phosphate (SPP), has been shown to mediate mitogenesis in several mammalian cell lines, including quiescent Swiss 3T3 fibroblasts , Rat-1 fibroblasts , airway smooth muscle cells (SMC) , and arterial SMCs . The basal level of SPP in cells is very low (10-30 pmoles/106 cells) and can increase rapidly and transiently in response to various mitogenic stimuli in Swiss 3T3 fibroblasts , arterial SMCs  and airway SMCs . Platelet-derived growth factor (PDGF) and serum increased SPP in Swiss 3T3 cells to levels similar to the amount taken up when these cells were treated with mitogenic concentrations of SPP. PDGF also increases levels of sphingosine, the precursor of SPP, in Swiss 3T3 fibroblasts , vascular SMCs  and glomerular mesangial cells . This response is specific for certain growth promoting agents, since other potent mitogens, such as epidermal growth factor (EGF), do not induce significant changes in SPP levels . The level of SPP in cells is determined by the relative contributions of its formation, mediated by sphingosine kinase , and its degradation, catalyzed by a pyridoxal phosphate-dependent lyase located in the endoplasmic reticulum . Recently, we have identified a specific endoplasmic reticulum-associated SPP-phosphatase which may also be important for degradation of SPP (S. Mandala et al., unpublished). PDGF, serum , and other mitogens, such as the B-subunit of cholera toxin, and the phorbol ester 12-O-tetradecanoylphorbol 13-acetate (TPA), significantly increase sphingosine kinase activity [3, 19, 20], whereas some mitogens, including bombesin, bradykinin, insulin, and EGF have little or no effect. D,L-Threo-dihydrosphingosine and N,N-dimethylsphingosine (DMS), competitive inhibitors of sphingosine kinase (J. Van Brocklyn et al., unpublished), completely eliminate formation of SPP and selectively block cellular proliferation induced by PDGF or serum but not by EGF , further supporting a role for SPP in cellular proliferation. In agreement, L-cycloserine, an inhibitor of sphingolipid synthesis, and N-oleoylethanolamine, an inhibitor of ceramidase activity, inhibited PDGF-stimulated DNA synthesis in vascular SMCs and in rat glomerular mesangial cells, respectively [15, 16]. Moreover, addition of sphingosine or SPP abolished cell growth arrest by the immunosuppressant ISP-1/myriocin, an inhibitor of serine palmitoyltransferase and sphingolipid metabolism . Thus, sphingosine kinase seems to be an important growth regulatory enzyme.
Although several reports have provided evidence that some of the effects of SPP are mediated through a cell surface receptor (see below), the experiments discussed above suggest that SPP acts as an intracellular second messenger in growth factor signaling pathways. Further evidence that the mitogenic effect of SPP is not mediated through a cell surface receptor comes from the use of the polyanionic drug suramin, a broad-specificity inhibitor of ligand receptor interactions. Suramin has no effect on SPP-induced cytoskeletal remodeling, tyrosine phosphorylation of focal adhesion kinase (p125FAK) or mitogenesis, but inhibits these response to lysophosphatidic acid, which is structurally similar to SPP and is known to act through cell surface receptors . Thus, while SPP mediates the activation of a variety of mitogenic signaling pathways, a more complete understanding of how SPP functions in mitogenesis awaits the identification of its proximal downstream target(s).
DOWNSTREAM SIGNALING OF SPHINGOSINE-1-PHOSPHATE
Although the mechanism by which SPP modulates cell growth is still not completely understood, several studies have recently begun to unravel the signal transduction pathways involved. In Swiss 3T3 fibroblasts, D,L-threo-dihydrosphingosine blocked PDGF-, but not EGF-induced activation of two cyclin-dependent kinases (p34cdc2 and Cdk2), the mitogen-activated protein kinase Erk2, tyrosine phosphorylation of the adaptor protein Crk, and DNA binding activity of AP-1, but had no effect on PDGF receptor autophosphorylation, Shc phosphorylation, or phosphatidylinositol 3-kinase activation . Inhibition of PDGF-stimulated Cdk activation and DNA synthesis was reversed by SPP, demonstrating that the effects of D,L-threo-dihydrosphingosine were due to sphingosine kinase inhibition . Crk, which links growth factor receptors to Ras activation, was tyrosine phosphorylated in response to SPP treatment in NIH 3T3 cells . In addition, Crk overexpression enhanced SPP-induced mitogenesis concomitantly with increased expression of Crk . SPP also affects cell growth by stimulation of the mitogen-activated protein kinase (MAPK) pathway and inhibition of the SAPK/JNK pathway (see below). These results suggest that SPP may mediate mitogenesis in part by the regulation of tyrosine kinases.
In quiescent Swiss 3T3 fibroblasts, SPP induces increased proliferation by activation of phospholipase D (PLD) leading to increases in levels of the second messenger phosphatidic acid . Activation of PLD by SPP occurs independently of protein kinase C (PKC) and Gi/0 protein-linked receptor pathways. In many cell types, SPP uniformly has a stimulatory effect on phosphatidic acid levels, whereas, ceramide inhibits PLD [26, 27]. Thus, increased levels of ceramide may exhibit growth arresting effects by blocking the PLD signaling pathway, while the mitogenic effects of SPP may be due to stimulation of PLD. The regulation of phosphatidic acid levels by sphingolipid metabolites indicates that the sphingolipid cycle could regulate the glycerophospholipid cycle. Cross-talk between these pathways adds another level of complexity to signaling pathways utilizing lipid metabolites and additional sites of regulation.
Another important downstream effect of SPP is calcium mobilization. In most cell lines tested, SPP-mediated calcium mobilization occurs by an inositol trisphosphate (InsP3)-independent pathway [28, 29] from stores within the endoplasmic reticulum which also contain sphingosine kinase . Similarly, calcium mobilization resulting from FceR1 engagement requires activation of sphingosine kinase leading to generation of SPP . Inhibitors of sphingosine kinase, such as D,L-threo-dihydrosphingosine, block FceR1-mediated calcium mobilization as well as partially inhibit histamine release. However, in HL-60 and thyroid cells, SPP-stimulated calcium mobilization requires phospholipase C .
The effect of SPP on calcium mobilization closely resembles the calcium-releasing effect of InsP3, which is mediated by direct calcium channel activation. However, the SPP-specific channel remains to be identified, which will no doubt greatly facilitate our understanding of SPP signaling.
In addition to calcium mobilization, SPP also regulates levels of another second messenger, cAMP . However, the relevance of changes in cAMP levels to mitogenic effects of SPP is still unclear. In most cell types, SPP treatment results in inactivation of adenyl cyclase and protein kinase A (PKA) through a Gi protein [31, 33, 34]. However, in SMCs, SPP-mediated calcium mobilization results in increased cAMP and activation of PKA, leading to actin disassembly and altered cell motility . Because calcium and cAMP affect polymerization/depolymerization of actin filaments, the involvement of SPP in cytoskeletal remodeling, cellular motility, and metastatic invasiveness is of great interest . In arterial smooth muscle cells, SPP inhibits PDGF-induced chemotaxis by favoring actin filament disassembly and inhibiting focal adhesion formation .
In contrast to its effects on Swiss 3T3 fibroblasts, SPP induces calcium mobilization and subsequent cAMP increases and PKA activation. These responses result in disruption of the equilibrium of assembly/disassembly of actin filaments, inhibiting cell spreading, extension of the leading lamellae, and chemotaxis toward PDGF. However, in Swiss 3T3 fibroblasts, SPP induced rapid reorganization of the actin cytoskeleton resulting in stress fiber formation with concomitant focal adhesion assembly and tyrosine phosphorylation of p125FAK and paxilline. The exoenzyme C3 transferase, which inactivates the small GTP-binding protein p21rho (Rho) by ADP-ribosylation, inhibited both protein tyrosine phosphorylation and stress fiber formation, indicating a novel, Rho-mediated, signaling pathway modulated by SPP .
SPHINGOSINE-1-PHOSPHATE INHIBITS CERAMIDE-MEDIATED PROGRAMMED CELL DEATH
Apoptosis or programmed cell death is an evolutionarily conserved mechanism playing an important role in early development and homeostasis of adult tissues. Ceramide is emerging as a crucial component of apoptosis [1, 36]. Exposure to TNF-alpha, Fas ligand, irradiation, or chemotherapeutic agents, leads to an increased cellular level of ceramide, which in turn, triggers the hallmarks of apoptosis [4, 5, 7, 8, 37-39]. Activation of PKC by TPA or diacylglycerol (DAG) protects against various apoptotic insults, suggesting that PKC counteracts the ceramide-mediated apoptosis pathway [19, 36, 37, 40]. Although the mechanism by which PKC opposes the apoptosis effect of ceramide is not well understood, we noticed that in many cell types, activation of PKC stimulates sphingosine kinase activity, suggesting that this effect is mediated by SPP. Indeed, recently, we have shown that SPP prevents apoptosis resulting from elevated levels of ceramide induced by TNF-alpha, anti-Fas antibody, sphingomyelinase, or cell permeable ceramide analogs . Furthermore, inhibition of PKC, as well as inhibition of sphingosine kinase by DMS, induces apoptosis, which can be overcome by the addition of SPP . Our results suggest that an important target of PKC in the suppression of apoptosis is the activation of sphingosine kinase resulting in increased cellular levels of SPP. Surprisingly, TPA also completely blocked TNF-alpha-induced increases in ceramide levels. Conversely, TNF-alpha reduced sphingosine kinase activity and markedly decreased SPP. Thus, TPA, growth factors, and TNF-alpha, have opposing effects on intracellular levels of ceramide and SPP and can significantly alter their relative intracellular ratio. Although this ceramide/SPP rheostat may be an inherent characteristic of each cell type, external stimuli can reset the ratio. Thus, dissociation of growth factor-induced mitogenesis from cytokine-mediated apoptosis is a consequence of distinct sphingolipid-derived second messengers. In agreement, in rat glomerular mesangial cells and in Swiss 3T3 fibroblasts, PDGF, but not cytokines, mediates proliferation in part through ceramidase-regulated sphingosine and SPP formation [14, 16], whereas inflammatory cytokines stimulate sphingomyelinase but not ceramidase leading to accumulation of ceramide [14, 16]. Thus, regulation of sphingomyelinase, ceramidase, and sphingosine kinase may determine whether signals flow through ceramide or SPP, altering the balance between apoptotic and mitogenic lipid signals. Although neither SPP nor ceramide have direct effects on DAG-responsive PKC isozymes, both have been shown to cross-talk with signaling pathways modulating the levels of DAG. Ceramide inhibits DAG generation through inhibition of phospholipase(s), while SPP uniformly has a stimulatory effect on PLD. Thus, SPP and ceramide act in feedback regulatory loops to regulate PKC either positively or negatively, respectively. Because activation of PKC results in increased SPP and decreased ceramide levels, these feedback loops amplify the effects of PKC on the balance between the concentrations of ceramide and SPP.
An important question is: How do sphingolipid metabolites regulate life and death? Cell survival and death may be determined by the dynamic balance between different mitogen-activated protein kinases (MAPKs) present in distinct but related signaling pathways. Activation of extracellular signal regulated kinases (ERK-1 and ERK-2) and suppression of stress-activated protein kinases (SAPKs), also known as c-Jun amino terminal kinases (JNKs), as well as p38 signaling, is crucial for cell survival and proliferation. In contrast, activation of SAPK/JNK and concurrent inhibition of ERK promote cell death . It is of particular interest to note that ceramide and SPP have opposing effects on these pathways. In many cell types, ceramide-induced apoptosis has been correlated with stimulation of JNK and overexpression of dominant-negative constituents of the SAPK/JNK pathway abrogates ceramide-mediated apoptosis . Conversely, SPP not only stimulates ERKs  but also prevents SAPK/JNK activation by ceramide and consequent apoptosis . Thus, regulation of the ceramide/SPP rheostat and subsequent effects on ERK and SAPK/JNK is important in determining cell fate .
ROLE OF SPHINGOSINE 1-PHOSPHATE IN NEUROTROPHIC ACTIONS OF NGF
Nerve growth factor (NGF), a known trophic factor for survival and differentiation of neuronal cells, mediates its effects on the central and peripheral nervous system through two classes of cell surface receptors, the high affinity tyrosine kinase trkA receptor, and p75NGFR, which has structural similarity to the TNF family of receptors. Survival and differentiative effects of NGF are predominantly mediated by tyrosine kinase trkA phosphorylation . In contrast, binding of NGF to p75NGFR has been observed to play a distinct role in neuronal cell death resulting from activation of sphingomyelinase and production of ceramide . Interestingly, enhanced ceramide production was abolished in the presence of trkA. Moreover, inhibition of the tyrosine kinase activity of trkA with K252a restored the ability of NGF to induce sphingomyelin hydrolysis , suggesting that cross-talk may occur between these two NGF receptors.
Recently, we have observed that binding of NGF to trkA increases cellular levels of SPP in pheochromocytoma PC12 cells by activation of sphingosine kinase . Moreover, SPP inhibits cell death induced by serum withdrawal leading to increased ceramide levels . Furthermore, inhibition of SPP production by DMS enhanced DNA fragmentation and reduced the cytoprotective effect of NGF which was restored by SPP. Moreover, DMS inhibited NGF neurite outgrowth and neurofilament expression, whereas SPP enhanced neurofilament expression elicited by suboptimal doses of NGF . When growth arrest and neurite retraction are required, it is anticipated that trkA functions would be inhibited and engagement of p75NGFR would activate sphingomyelinase activity leading to increased ceramide levels, and conversely, when survival and neurite extensions are required, trkA would not only suppress the ability of p75NGFR to stimulate sphingomyelinase, it would also stimulate sphingosine kinase and increase SPP levels. Our results suggest that cross-talk between the two types of NGF receptors may reciprocally modulate the production of ceramide and SPP, thereby affecting pathways involved in the maintenance and differentiation of neuronal cells.
This work was supported by Research Grants RO1 CA61774 and RO1 CA61774 from the National Institutes of Health and BE-275 from the American Cancer Society.
1.Hannun, Y. (1996) Science, 274,
2.Kolesnick, R., and Fuks, Z. (1995) J. Exp. Med., 181, 1949-1952.
3.Spiegel, S., and Milstien, S. (1995) J. Membr. Biol., 146, 225-237.
4.Hannun, Y. A. (1994) J. Biol. Chem., 269, 3125-3128.
5.Kolesnick, R., and Golde, D. W. (1994) Cell, 77, 325-328.
6.Heller, R. A., and Kronke, M. (1994) J. Cell Biol., 126, 5-9.
7.Cifone, M. G., De Maria, R., Roncaioli, P., Rippo, M. R., Azuma, M., Lanier, L. L., Santoni, A., and Testi, R. (1994) J. Exp. Med., 180, 1547-1552.
8.Haimovitz-Friedman, A., Kan, C., Ehleiter, D., Persaud, R. S., McLoughlin, M., Fuks, Z., and Kolesnick, R. N. (1994) J. Exp. Med., 180, 525-535.
9.Gulbins, E., Bissonnette, R., Mahboubi, A., Martin, S., Nishioka, W., Brunner, T., Baier, G., Baier-Bitterlich, G., Byrd, C., Lang, F., Kolesnick, R., Altman, A., and Green, D. (1995) Immunity, 2, 341-351.
10.Zhang, H., Desai, N. N., Olivera, A., Seki, T., Brooker, G., and Spiegel, S. (1991) J. Cell Biol., 114, 155-167.
11.Gomez-Munoz, A., Waggoner, D. W., O'Brien, L., and Brindley, D. N. (1995) J. Biol. Chem., 270, 26318-26325.
12.Pyne, S., and Pyne, N. J. (1996) Biochem. J., 315, 917-923.
13.Bornfeldt, K. E., Graves, L. M., Raines, E. W., Igarashi, Y., Wayman, G., Yamamura, S., Yatomi, Y., Sidhu, J. S., Krebs, E. G., Hakomori, S., and Ross, R. (1995) J. Cell Biol., 130, 193-206.
14.Olivera, A., and Spiegel, S. (1993) Nature, 365, 557-560.
15.Jacobs, L. S., and Kester, M. (1993) Am. J. Physiol., 265, 740-747.
16.Coroneos, E., Martinez, M., McKenna, S., and Kester, M. (1995) J. Biol. Chem., 270, 23305-23309.
17.Stoffel, W., Hellenbroich, B., and Heimann, G. (1973) Hoppe-Seyler's Z. Physiol. Chem., 354, 1311-1316.
18.Van Veldhoven, P. P., and Mannaerts, G. P. (1991) J. Biol. Chem., 266, 12502-12507.
19.Cuvillier, O., Pirianov, G., Kleuser, B., Vanek, P. J., Coso, O. A., Gutkind, J. S., and Spiegel, S. (1996) Nature, 381, 800-803.
20.Mazurek, N., Megidish, T., Hakomori, S.-I., and Igarashi, Y. (1994) Biochem. Biophys. Res. Commun., 198, 1-9.
21.Miyake, Y., Kozutsumi, Y., Nakamura, S., Fujita, T., and Kawasaki, T. (1995) Biochem. Biophys. Res. Commun., 211, 396-403.
22.Wang, F., Nobes, C. D., Hall, A., and Spiegel, S. (1997) Biochem. J., 324, 481-488.
23.Rani, C. S., Berger, A., Wu, J., Sturgill, T. W., Beitner-Johnson, D., LeRoith, D., Varticovski, L., and Spiegel, S. (1997) J. Biol. Chem., 272, 10777-10783.
24.Blakesley, V. A., Beitner-Johnson, D., Van Brocklyn, J. R., Rani, S., Shen-Orr, Z., Stannard, B. S., Spiegel, S., and LeRoith, D. (1997) J. Biol. Chem., 272, 16211-16215.
25.Desai, N. N., Zhang, H., Olivera, A., Mattie, M. E., and Spiegel, S. (1992) J. Biol. Chem., 267, 23122-23128.
26.Venable, M. E., Blobe, G. C., and Obeid, L. M. (1994) J. Biol. Chem., 269, 26040-26044.
27.Gomez-Munoz, A., Martin, A., O'Brien, L., and Brindley, D. N. (1994) J. Biol. Chem., 269, 8937-8943.
28.Mattie, M., Brooker, G., and Speigel, S. (1994) J. Biol. Chem., 269, 3181-3188.
29.Ghosh, T. K., Bian, J., and Gill, D. L. (1994) J. Biol. Chem., 269, 22628-22635.
30.Choi, O. H., Kim, J.-H., and Kinet, J.-P. (1996) Nature, 380, 634-636.
31.Okajima, F., Tomura, H., Sho, K., Kimura, T., Sato, K., Im, D., Akbar, M., and Kondo, Y. (1997) Endocrinology, 138, 220-229.
32.Spiegel, S., and Merril, A. H., Jr. (1996) FASEB J., 10, 1388-1397.
33.Van Koppen, C. J., Meyer zu Heringdorf, D., Laser, K. T., Zhang, C., and Jakob, K. H. (1996) J. Biol. Chem., 271, 2082-2087.
34.Goodemote, K. A., Mattie, M. E., Berger, A., and Spiegel, S. (1995) J. Biol. Chem., 270, 10272-10277.
35.Spiegel, S., Olivera, A., Zhang, H., Thompson, E. W., Su, Y., and Berger, A. (1994) Breast Cancer Res. Treat., 31, 195-206.
36.Obeid, L. M., Linardic, C. M., Karolak, L. A., and Hannun, Y. A. (1993) Science, 259, 1769-1771.
37.Tepper, C. G., Jayadev, S., Liu, B., Bielawska, A., Wolff, R., Yonehara, S., Hannun, Y. A., and Seldin, M. F. (1995) Proc. Natl. Acad. Sci. USA, 92, 8443-8447.
38.Jaffrhzou, J., Levade, T., Bettaìeb, A., Andrieu, N., Bezombes, C., Maestre, N., Vermeersch, S., Rousse, A., and Laurent, G. (1996) EMBO J., 15, 2417-2424.
39.Skowronski, E. W., Kolesnik, R. N., and Green, D. R. (1996) Cell Death Diff., 3, 171-176.
40.Jarvis, W. D., Fornari, F. A., Browning, J. L., Gewirtz, D. A., Kolesnick, R. N., and Grant, S. (1994) J. Biol. Chem., 269, 31685-31692.
41.Xia, Z., Dickens, M., Raingeaud, J., Davis, R. J., and Greenberg, M. E. (1995) Science, 270, 1326-1331.
42.Verheij, M., Bose, R., Lin, X., Yao, B., Jarvis, W. D., Grant, S., Birrer, M. J., Szabo, E., Zon, L. I., Kyriakis, J. M., Haimovitz-Friedman, A., Fuks, Z., and Kolesnik, R. N. (1996) Nature, 380, 75-79.
43.Wu, J., Spiegel, S., and Sturgill, T. W. (1995) J. Biol. Chem., 270, 11484-11488.
44.Greene, L., and Kaplan, D. (1995) Curr. Opin. Neurobiol., 5, 579-587.
45.Casaccia-Bonnefil, P., and Chao, M. (1996) J. Neurosci. Res., 43, 382-389.
46.Dobrowsky, R., Jenkins, G., and Hannun, Y. (1995) J. Biol. Chem., 270, 22135-22142.
47.Edsall, L. C., Pirianov, G., and Spiegel, S. (1997) J. Neurosci., in press.