- Email: [email protected]
A house divided: Ceramide, sphingosine, and sphingosine-1-phosphate in programmed cell death
Biochimica et Biophysica Acta 1758 (2006) 2027 – 2036 www.elsevier.com/locate/bbamem
Review
A house divided: Ceramide, sphingosine, and sphingosine-1-phosphate in programmed cell death Tarek A. Taha a,b , Thomas D. Mullen a,b , Lina M. Obeid a,b,⁎ a
Division of General Internal Medicine, Ralph H. Johnson Veterans Administration Hospital, Charleston, SC 29401, USA b Department of Medicine, Medical University of South Carolina, Charleston, SC 29425, USA Received 18 July 2006; received in revised form 25 October 2006; accepted 28 October 2006 Available online 1 November 2006
Abstract Programmed cell death is an important physiological response to many forms of cellular stress. The signaling cascades that result in programmed cell death are as elaborate as those that promote cell survival, and it is clear that coordination of both protein- and lipid-mediated signals is crucial for proper cell execution. Sphingolipids are a large class of lipids whose diverse members share the common feature of a longchain sphingoid base, e.g., sphingosine. Many sphingolipids have been shown to play essential roles in both death signaling and survival. Ceramide, an N-acylsphingosine, has been implicated in cell death following a myriad of cellular stresses. Sphingosine itself can induce cell death but via pathways both similar and dissimilar to those of ceramide. Sphingosine-1-phosphate, on the other hand, is an anti-apoptotic molecule that mediates a host of cellular effects antagonistic to those of its pro-apoptotic sphingolipid siblings. Extraordinarily, these lipid mediators are metabolically juxtaposed, suggesting that the regulation of their metabolism is of the utmost importance in determining cell fate. In this review, we briefly examine the role of ceramide, sphingosine, and sphingosine-1-phosphate in programmed cell death and highlight the potential roles that these lipids play in the pathway to apoptosis. © 2006 Elsevier B.V. All rights reserved. Keywords: Programmed cell death; Ceramide; Sphingosine; Sphingosine-1-phosphate (S1P); Sphingolipid; Apoptosis
Contents 1. 2.
Overview of programmed cell death . . Ceramide and cell death . . . . . . . . 2.1. Ceramide and p53 . . . . . . . . 2.2. Ceramide and the Bcl-2 Family . 2.3. Ceramide and the mitochondrion 2.4. Ceramide and proteases . . . . . 2.5. Ceramide and autophagy . . . . 3. Sphingosine and cell death . . . . . . . 4. SK, S1P, and cell death. . . . . . . . . 5. Concluding remarks . . . . . . . . . . Acknowledgments . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
⁎ Corresponding author. Department of Medicine, Medical University of South Carolina, 114 Doughty St., P.O. Box 250779, Charleston, SC 29425, USA. E-mail address: [email protected] (L.M. Obeid). 0005-2736/$ - see front matter © 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.bbamem.2006.10.018
2028 2029 2030 2030 2031 2031 2031 2032 2033 2034 2034 2034
2028
T.A. Taha et al. / Biochimica et Biophysica Acta 1758 (2006) 2027–2036
1. Overview of programmed cell death Ongoing research in the field of cell death shows that the mechanisms by which cells die are as complex (if not more complex) than those by which cells survive. A common feature of all physiologic forms of death is that they are programmed events governed by specific biochemical pathways, hence the name programmed cell death or PCD [1]. While PCD is driven by complex pathways involving multiple players, it can be broken down into three stages for pragmatic purposes: initiation, commitment, and execution. The initiation step often occurs at or in the proximity of the cellular compartment where the stress is induced (Fig. 1) [2]. Disruption of calcium homeostasis at the ER, for example, results in the activation of calcium-mediated cell death [3]. Genotoxic stress in the nucleus induces p53 and drives p53-activated pathways of cell death [4–7]. Lysosome membrane disturbance causes the release of lysosomal proteases, which have also been implicated in PCD [8,9]. Activation of death receptors at the plasma membrane is another means of inducing PCD, which may involve the downstream activation of one or more organelle-mediated pro-apoptotic pathways [1,10]. Regardless of the initiation step, once a death signal is sensed by specific sensors, biochemical pathways are activated, that eventually converge onto the commitment step, or the point of no return, which in many instances, occurs at the level of the mitochondrion [11]. Once the death signal is relayed to the mitochondrion, the organelle incurs a major change: mitochondrial membrane permeabilization (MMP). MMP of the outer membrane causes
the release of several proteins from the intermembrane space into the cytosol. Cytochrome c, the most studied of these proteins, drives oligomerization of the adapter molecule APAF1, which then activates caspase 9, a key cysteine protease in the mitochondrial pathway of cell death [12]. Caspase 9 can then activate caspases 3 and 7, which belong to the execution phase of PCD, since they cleave proteins that are vital to normal cell function [13]. Another protein released from the mitochondrion is Smac/Diablo, which binds to and inhibits cellular proteins called IAPs (inhibitors of apoptosis) that normally keep caspases from undergoing erratic activation in the healthy state. Following release of cytochrome c and Smac, cells display several features of apoptotic PCD, the most prominent being chromatin condensation morphologically and cleavage of PARP and externalization of phosphatidylserine (PS) biochemically. It should be noted that caspase 9 activation and PS exposure on the outer leaflet of the cell membrane are ATPrequiring processes, and that depletion of ATP from cells can switch the death from an apoptotic form to a necrotic form [14]. In addition to the classical apoptotic form of PCD which involves the activation of caspases, the mitochondrion harbors other proteins, which upon release can drive alternative forms of cell death that are entirely caspase-independent. Following MMP, apoptosis inducing factor (AIF) [15] and endonuclease G (Endo G) [16] translocate from the mitochondrion to the nucleus and mediate apoptosis-like PCD, so called because they cause loose chromatin condensation that is caspase independent. HtrA2/Omi is also released from the mitochondrion and has a dual role in cell death, one dependent on its ability to
Fig. 1. Multiple pathways of cellular stress lead to MMP and apoptosis. Genotoxic stress activates p53 and “BH-3 only” proteins such as Noxa and Puma that promote MMP and the release of death factors from the mitochondrion. Cathepsins are released during lysosomal stress and result in the cleavage of Bid, which translocates to the mitochondrion to promote MMP. ER stress can be manifested by dysregulation of calcium homeostasis and subsequent accumulation of calcium in the mitochondria, leading to MMP. Finally, death receptor activation can lead to a multifaceted pathway that may involve lysosomal disruption, caspase 8-mediated Bid cleavage, mitochondrial disruption, and caspase activation.
T.A. Taha et al. / Biochimica et Biophysica Acta 1758 (2006) 2027–2036
inactivate IAPs and cause caspase-dependent PCD, and another due to its serine protease activity [17]. Since the mitochondrion constitutes an essential role in cell death, it is not surprising that the cell has developed an entire protein family to control the function of this organelle in the healthy vs. the apoptotic state. Briefly, members of the Bcl-2 family of proteins belong to one of three different classes: the Bcl-2 like survival factors, and these include Bcl-2 and Bcl-xL (in addition to other less characterized members); the Bax-like death factors, such as Bax and Bak; and the BH3-only death factors, such as Noxa, Puma, Bid, Bad, and Bik [18]. The most well characterized site of action of these proteins is the mitochondrion itself, although actions at other organelles are beginning to be proposed as well. Bcl-2-like proteins maintain mitochondrial integrity and are regarded as guardians of this organelle. In fact, overexpression of Bcl-2 and Bcl-xL rescues cells from death induced by several stimuli that normally induce MMP. Bax-like and BH3-only proteins, on the other hand, compromise mitochondrial function and essentially mediate the effect of several death inducers that cause MMP. Therefore, the integrity of the mitochondrion (and the survival of the cell) depends on a critical balance between the proapoptotic and the antiapoptotic members of the Bcl-2 family. Of note is that Bcl-2 proteins not only are key players in apoptosis, but also in the alternative forms of PCD, where MMP occurs. Classically, apoptotic pathways have been described as either intrinsic or extrinsic. The former indicates that the death signal is generated
2029
from within the cell, such as in the case of genotoxic stress causing cell death. The extrinsic pathway, on the other hand, indicates that the death signal is generated by binding of a death ligand to a cell surface receptor, which then activates the cascade of apoptosis. MMP is essentially a key feature of the intrinsic pathway. In cases where death ligands bind to their surface receptors, cell death can proceed independent of the mitochondrion and without MMP. Of note though is that the extrinsic pathway often cross talks with the intrinsic pathway of PCD, making the mitochondrion and the Bcl-2 proteins key player also in death ligand-mediated PCD. 2. Ceramide and cell death Given its central function, ceramide is the most studied sphingolipid species. Several lines of investigation have clearly established this lipid as a mediator of the eukaryotic stress response (Fig. 2) [19]. Ceramide accumulation can occur via: (i) activation of de novo synthesis mediated by serine-palmitoyl transferase (SPT) and ceramide synthase [20,21]; (ii) activation of sphingomyelin hydrolysis [22,23]; (iii) inhibition of ceramide hydrolysis; and/or (iv) stimulation of glucosylceramide hydrolysis or inhibition of its synthesis [24] (Fig. 1-1A). Some agents activate a multitude of these pathways, such as irradiation [25]. The effects of ceramide are pleiotropic, but for the most part growth inhibiting. The molecule has been implicated in differentiation [26,27], cell cycle arrest [28–30],
Fig. 2. Ceramide is a central mediator of many apoptotic pathways. Following genotoxic stress, ceramide accumulates, but it is still unclear whether the increase is dependent or independent of p53. Signaling through the lysosomes involves increases in ceramide via aSMase. Ceramide can directly activate the lysosomal protease cathepsin D which can subsequently cleave and activate Bid. Another direct target of ceramide is protein phosphatase 2A, which is known to negatively regulate the prosurvival kinase Akt/PKB. Both PP2A and Akt/PKB regulate the Bax/Bcl-2 rheostat; if the PP2A prevails, as is the case in ceramide signaling, Bax prevails over Bcl-2 and the mitochondrion is permeabilized. Death receptor activation can lead to ceramide formation via nSMase and caspase 8-mediated cleavage of Bid; lysosomes may also be activated downstream of death receptors, with both signals converging on MMP. Additionally, generation of ceramide from sphingomyelin hydrolysis in mitochondria results in Bax translocation and MMP.
2030
T.A. Taha et al. / Biochimica et Biophysica Acta 1758 (2006) 2027–2036
apoptosis [31], and senescence [32] in several cell types. It induces cell cycle arrest through the dephosphorylation of the Retinoblastoma gene product (Rb), activation of the cyclin dependent kinase inhibitor p21, and inhibition of the cyclin dependent kinase 2 (CDK2) [29,30,33]. These studies have shown that ceramide is upstream of cell cycle regulators, and that the generation of ceramide is necessary for growth arrest in response to stimuli that induce this arrest. The elevation of ceramide has also been shown to occur in senescent cells, possibly via the activation of neutral sphingomyelinase (nSMase) [34]. The senescent phenotype may be attributed to a defect in the phospholipase D/protein kinase C (PLD/PKC) pathway, and ceramide can inhibit both PLD and PKC [35–37]. One of the most studied roles of ceramide pertains to its function as a proapoptotic molecule. The accumulation of ceramide following treatment of cells with apoptotic agents has implicated this lipid in the biological responses of these agents [25]. Because of its apoptosis-inducing effects in cancer cells, ceramide has been termed the “tumor suppressor lipid” [38]. Several studies have attempted to define further the specific role of ceramide in the events of cell death. The tumor suppressor protein p53, the Bcl-2 family of proteins and several protease classes are all key components of the tumor response to stress insults, and ceramide has been linked to each of these mediators. Many studies, however, report variable associations among these proteins and ceramide, suggesting that different cells have dissimilar networks, and that deciphering the relevance of in vitro findings to in vivo tumor behavior is far from understood. 2.1. Ceramide and p53 Several studies have generated conflicting data on the relationship between ceramide and p53, which is virtually dysfunctional in all human tumors. While some reports show that p53 is upstream of ceramide in tumor stress responses [39,40], other studies implicate p53 as a downstream target [41–43], yet these latter studies utilize exogenous ceramide as the stress inducer, whereas the former use agents that generate endogenous ceramide. Still, other observations place ceramide and p53 in two separate and independent pathways in the apoptotic process [44,45]. In the latter case, treatment with chemotherapeutic agents generates ceramide in p53+/+ as well as p53−/− cells. One study has demonstrated that ceramide can be formed by acid sphingomyelinase (aSMase) in response to genotoxic stress only in cells lacking functional p53, suggesting that p53 can inhibit aSMase activation and ceramide generation [46]. Studies by Santana et al. also proposed a role for aSMase derived ceramide in the apoptotic response and highlighted that the response is different from that derived by p53 driven cellular responses [47]. Therefore, it remains unclear how ceramide and p53 are linked in PCD and there appears to be variations in the link between the two messengers in different cell death models. 2.2. Ceramide and the Bcl-2 Family An equally wide array of studies has investigated the relationship between ceramide and the Bcl-2 family of proteins,
both following treatment with genotoxic and nongenotoxic agents. The most studied connections pertain to Bcl-2 itself, Bcl-xL, and Bax. The position of ceramide with respect to Bcl-2 is variably reported. A number of groups have shown that ceramide is upstream of Bcl-2 in the apoptotic pathway since Bcl-2 overexpression rescues from cell death induced by ceramide [48] or by ceramidase inhibitors [49]. Moreover, ElAssaad et al. have shown that Bcl-2 and Bcl-xL define two different points of regulation of ceramide responses. While both proteins rescue from TNF mediated cell death, only Bcl-xL abrogates ceramide generation, while Bcl-2 does not, implying a pathway where ceramide is downstream of Bcl-xL but upstream of Bcl-2 [50]. Moreover, in C6 glioma cells, etoposide induces ceramide formation by neutral sphingomyelinase activation, which then increases the Bax/Bcl-2 ratio [51], and in A549 cells, exogenous ceramide and endogenous ceramide produced from gemcitabine treatment enhance the expression of proapoptotic Bcl-x (BCl-xs) and caspase 9 splice variants, demonstrating regulation of Bcl-2 family protein expression levels and splicing patterns by the bioactive lipid [20]. On the other hand, other studies have shown that Bcl-2 and or Bcl-xL overexpression attenuates ceramide accumulation following DNA damaging stimuli [52–54], thus implicating that Bcl-2 can also act upstream of the lipid. Of note, however, is that many studies of ceramide regulation in cell death have measured total ceramide levels. Recently, it has become evident that the fatty acid chain length of ceramides is an important determinant of the biological effect mediated by the bioactive lipid. Marchesini et al. reported that nSMase activation in confluent MCF-7 cells causes cell cycle arrest but not apoptosis, and that this effect is mediated by very long chain C24-ceramide species [55]. Kroesen et al., on the other hand, showed that cross linking of the B-cell receptor generates C16-ceramide upstream of the mitochondrion in a caspase independent manner, and that inhibition of C16 ceramide generation rescues from cell death [56]. In the same study, C24 ceramide is generated downstream of mitochondrial dysfunction in a caspase dependent manner. Therefore, it is clear from these data that not all ceramide species generate the same responses, and that the different biological effects caused by ceramide may be mediated by distinct molecular species of the lipid. Moreover, it should be noted that nonmitochondrial sites of action have been reported for the Bcl-2 family of proteins, warranting that conclusions made with respect to these proteins may not always be extrapolated to the mitochondrion. The connection between ceramide and Bax has also been reported by several studies. One study has shown that Bax overexpression does not affect ceramide production in response to etoposide [54], yet ceramide treatment in a prostate and colorectal cancer cell line induces apoptosis only when Bax is overexpressed, suggesting a link between the lipid and the protein [57]. Furthermore, Birbes et al. have recently shown that localized production of ceramide in the mitochondrion induces Bax oligomerization and drives cell death [58]. This study is in agreement with the observation that ceramide treatment of isolated mitochondria potentiates the Bax-mediated induction of mitochondrial permeability transition (MPT) [59]. Further-
T.A. Taha et al. / Biochimica et Biophysica Acta 1758 (2006) 2027–2036
2031
more, a study has shown that C16-ceramide production via acidic sphingomyelinase (aSMase) mediates Bax conformational change and induces MMP in vivo and in vitro [60]. Ceramide has also been implicated in oocyte death as its levels increase in oocytes prior to increases in Bax protein levels and the induction of apoptosis [61]. Endothelial cells of Bax and aSMase deficient mice were also recently shown to have higher resistance to radiation induced cell death, indicating that ceramide generation by aSMase may also be important in microvasculature homeostasis in addition to cancer cell homeostasis [62,63]. Ceramide may regulate Bcl-2 and Bax simultaneously by modulating their phosphorylation states. It was established several years ago that Bcl-2, whose phosphorylation by PKCα at serine 70 is required for its anti-apoptotic function, becomes dephosphorylated in response to ceramide [64,65]. Recent work suggests that the ceramide-activated protein phosphatase PP2A may inhibit Bcl-2 via its dephosphorylation and proteasomal degradation [66]. On the other hand, Bax may also be regulated via PP2A, because Xin et al. showed that Bax is dephosphorylated following ceramide treatment in a PP2A-dependent manner [67]. Dephosphorylation of Bax was associated with its conformational change and subsequent release of cytochrome c from the mitochondria. Therefore, these studies display a proximal connection between ceramide, Bcl-2, and Bax, and implicate that ceramide may be a regulator of Bcl-2 family members and their control over cell death pathways.
of solutes into the cytosol [74]. Perhaps ceramide can mediate multiple different pathways to release pro-apoptotic factors from the mitochondrion. Lorusso and colleagues showed that ceramide could induce cytochrome c release from isolated mitochondria in both a PTP-dependent and independent manner [75]. Ceramide also maintains cytochrome c in an oxidized state, a status which appears to be necessary for reducing mitochondrial oxygen consumption, dissipating the mitochondrial membrane potential (Δψm), and releasing calcium from the mitochondria [76]. Hence, all these studies point to a direct effect of ceramide on the mitochondrion. While the above in vitro studies provide a direct link between ceramide and the mitochondrion, studies clearly defining the role that ceramide plays in cells have been very scarce, mostly owing to the difficulty in tracking the localization of ceramide. It is clear from the study by Zhang et al. that the topology of ceramide generation is essential for the lipid to mediate its downstream effects, since generation of the lipid at the plasma membrane is insufficient to induce cell death, whereas its generation within the cell is [77]. Further investigation into organelle-specific sphingomyelin hydrolysis has shown that only ceramide generation at the mitochondrion can drive cell death, whereas its formation in other organelles does not [78]. These studies demonstrate that once ceramide reaches the mitochondrion, it can induce the effects that the bioactive lipid exerts on the isolated organelle, making the in vitro observations relevant in vivo as well.
2.3. Ceramide and the mitochondrion
2.4. Ceramide and proteases
In addition to its link to the Bcl-2 family of proteins, an intimate connection is becoming more evident between ceramide and the mitochondrion. First, several enzymes involved in sphingolipid metabolism have been identified in subcellular fractions containing mitochondria. Bionda et al. have reported ceramide synthase and reverse ceramidase activities in the mitochondria and/or ER-related mitochondria associated membranes (MAMs) of rat liver [68]. Ceramide synthase has also been partially purified from bovine liver mitochondria [69]. Furthermore, a mammalian ceramidase localized in a similar compartment has recently been cloned [70], and sphingosine kinase type 2 has been described as a BH3-only like protein [71]. Therefore, more evidence is pointing towards machinery for sphingolipid biosynthesis at the level of the mitochondrion. The localization of these activities has been suggestive of the importance of generating the right lipid in the right place to induce its specific effects. In vitro experiments have revealed that ceramide can induce several changes in isolated mitochondria. Gudz et al. have shown that C2-ceramide can inhibit oxidative phosphorylation by interfering with the electron transport system at the level of complex III [72]. Ceramide also increases the permeability of the outer mitochondrial membrane to several proteins, including cytochrome c [73]. A recent study reported the ability of ceramide to enhance permeability of the inner mitochondrial membrane, via action on the permeability transition pore (PTP) and the electrogenic proton channel, which enhances the release
Other important components of PCD pathways are the proteases, which fall under two general categories: caspases and non-caspase proteases, both of which can be activated by ceramide. While the mediators linking initiator death signals to the activation of the mitochondrion are still poorly characterized, the sequence of events involved in the activation of caspases downstream of the mitochondrion is fairly well established. Extensive work shows that ceramide accumulates in apoptotic cells prior to the activation of execution caspases, but downstream of initiator caspases [40,79,80]. Ceramide also activates the non-caspase protease, cathepsin D [81] (Fig. 2), a mediator of cell death in several p53-dependent death model systems [82]. The formation of ceramide via acidic sphingomyelinase has also been shown to mediate the cleavage of Bid and downstream caspase activation in response to TNF [83]. Also, ceramide generation by aSmases has been suggested to mediate cell death by caspase dependent and independent mechanisms depending on the death stimulus [84]. 2.5. Ceramide and autophagy The role of ceramide in cell death pathways has been recently extended to encompass autophagy or Type II PCD (in contrast to apoptosis or Type I PCD). C2-ceramide upregulates the autophagy genes BNIP3 [85] and Beclin-1 [86] and de novo ceramide synthesis inhibitors can attenuate the induction of autophagy by Tamoxifen in MCF-7 cells [86]. Therefore, the
2032
T.A. Taha et al. / Biochimica et Biophysica Acta 1758 (2006) 2027–2036
role of ceramide as a universal mediator of various forms of PCD may be gaining more ground. 3. Sphingosine and cell death Several studies have pointed to a specific role for sphingosine in the induction of cell death independent of ceramide (Fig. 3A) [87]. In HL-60 cells, both ceramide and sphingosine accumulate before the features of cell death appear, yet their order in the pathway of cell death may be different [88]. While inhibition of caspases with Z-VAD-fmk rescues from apoptosis induced by both ceramide and sphingosine, z-IETD-fmk, the caspase-8 inhibitor and Z-AAD-fmk, the granzyme B inhibitor, are only effective against sphingosine but not ceramide [88]. The ceramide synthase inhibitor fumonisin B1 also does not inhibit sphingosine induced apoptosis in HL-60 and Jurkat T cells, indicating that the conversion of sphingosine to ceramide may not be required for the induction of cell death by sphingosine [88,89]. In MCF-7 cells, however, FB1 or D-MAPP, an inhibitor of alkaline ceramidase, are both effective in reducing apoptosis caused by exogenous sphingosine, suggesting that conversion to a ceramide intermediate may be important for the execution of sphingosine effects in some cells [90]. A recent study has reported that apoptosis induction by FTY720 is due to the ability of this compound to inhibit SK in vitro and accumulate sphingosine in vivo [91]. Concentrations of exogenous sphingosine that produce similar levels of intracellular sphingosine as FTY720 cause apoptotic features as well, implicating sphingosine as an active molecule in cell death [91]. The mechanisms mediating sphingosine-driven cell death are multiple, and in some cases do not coincide with those mediating ceramide death pathways [90]. Both lipids have been shown to induce apoptosis in U937 human monoblastic leukemia cells, yet ceramide-associated lethality involves the strong activation of JNK and weak inhibition of ERK, whereas sphingosine-driven death is associated with weak JNK activation and strong ERK inhibition [92]. Sphingosine can also inhibit Akt/Protein kinase B (PKB), and overexpression of
constitutively active myristoylated Akt partially rescues from sphingosine-induced death [93]. Sphingosine-mediated apoptosis has been shown to occur via caspase-dependent pathways. Sphingosine can induce Bid cleavage, mitochondrial cytochrome c release as well as the activation of downstream effector caspases 3 or 7 and the cleavage of PARP [89,90,94]. The involvement of the mitochondrial pathway of PCD is further illustrated by the effects of sphingosine on the Bcl-2 family of proteins. Studies have shown that sphingosine can reduce the expression levels of Bcl-2 and Bcl-xL, major gatekeepers of the mitochondrion [95,96]. Sphingosine-induced cell death has also been proposed to occur even in Bcl-2 transfected cells and to involve cleavage of Bax, and the association of the cleaved (and full length) Bax with Bcl-2 [97]. Sphingosine can also exert its effects via regulation of PKC. Initially, the bioactive lipid was found to inhibit PKC [98,99], an important prosurvival signal in the cell [100]. Recently, a sphingosine-dependent protein kinase (SDK1) has been described as a truncated version of PKCδ (a proapoptotic PKC), which forms by a caspase-3 dependent cleavage of the fulllength PKCδ [101]. While full-length PKCδ is inhibited by sphingosine and DMS, SDK1 activity is stimulated by the two lysophospholipids [101]. Among the substrates for SDK1 are the 14-3-3 proteins, specifically isoforms β, η, and ζ but not τ or σ [102]. 14-3-3 proteins are chaperones for several cellular proteins, including members of the Bcl-2 family, such as Bax [103,104] and Bad [105–108] (Bad requires phosphorylation itself to bind to 14-3-3 [109]). Phosphorylation of 14-3-3 is often associated with the release of the proapoptotic proteins, allowing them to exert their cell killing effects. Even though inhibition of PKC is a mechanism of sphingosine-mediated cell death, sphingosine can still induce apoptosis in cell systems were PKC is stimulated [94], suggesting the existence of sphingosine-responsive PKC-insensitive pathways of cell death. Multiple sphingosine activated kinases (other than SDK1) have also been suggested to exist, yet the identification of these awaits further studies [110].
Fig. 3. Sphingosine and S1P in apoptosis. (A) Sphingosine mediates cell death through the inhibition of the prosurvival factors PKC and Akt/PKB and induces death via SDK1, which leads to Bax activation through inhibition of 14-3-3 proteins. (B) S1P inhibits apoptosis by activating prosurvival factors such as Akt/PKB, NF-κB, ERK, and NO. Ceramide-induced death may also be inhibited by S1P through the inactivation JNK or other factors that promote PCD.
T.A. Taha et al. / Biochimica et Biophysica Acta 1758 (2006) 2027–2036
Besides its action through the classical mitochondrial pathway, sphingosine can also exert effects on other organelles. It has been shown that HeLa cells exposed to ceramide undergo massive Golgi fragmentation and inhibition of β1 integrin glycosylation and transport to the cell surface prior to the induction of apoptosis [111]. Exogenous sphingosine results in similar effects as exogenous ceramide, suggesting that the ceramide effects are due to its conversion to sphingosine in the Golgi complex [111]. The resulting loss of integrin signaling then leads to cell death by anoikis [111]. Almost a decade before that, the work of Rosenwald and Pagano also suggested a role for ceramide in slowing transport of viral glycoprotein across the Golgi apparatus, which was also disrupted by the lipid as well [112]. In that study, sphingosine had no effect, suggesting that the lipids may have slightly variable effects in different cells. There appears to be specific lipid mediated responses though because stereoisomers of ceramide were unable to cause changes in the Golgi complex [111]. Phytosphingosine, a saturated close relative of sphingosine having a hydroxyl group at the carbon-4 position (Fig. 1-4), is also a mediator of cell death. In combination with gamma irradiation, phytosphingosine enhances apoptosis of radiation resistant T cells by promoting reactive oxygen species (ROS) formation and AIF release, in a caspase independent manner [113]. The mechanism of phytosphingosine-induced cell death has been described to involve the inhibition of ERK, which drives caspase-8 activation, and the stimulation of p38 MAPK, which mediates Bax translocation and mitochondrial apoptosis [114]. 4. SK, S1P, and cell death Of particular interest in the biology of sphingolipids are the antagonistic roles that ceramide and sphingosine, on one hand, and sphingosine-1-phosphate, on the other, appear to exert in a variety of cells. Ceramide and sphingosine mediate apoptosis, cell cycle arrest, and differentiation, whereas S1P promotes proliferation, survival, and inhibition of apoptosis [115]. These findings have lead to the so-called sphingolipid rheostat which proposes that the relative levels of these lipids (especially in response to stressors or growth stimuli) are important determinants of cell fate [115]. One of the earliest observations implicating S1P as a progrowth lipid was the ability of PDGF and fetal calf serum to induce growth via the activation of sphingosine kinase and the formation of S1P, then called an intracellular second messenger [116]. Several growth stimuli were then shown to activate the SK/S1P pathway, which in turn appears to be required for the activation of downstream signaling by these stimuli. In addition, overexpression of SK1 in itself decreases ceramide and sphingosine levels and increases S1P [117], a sphingolipid conversion that is associated with enhanced proliferation, G1-S transition, and increased DNA synthesis [118]. The role of SK and S1P in PCD has come from several studies showing that these two mediators can attenuate death and growth inhibitory effects caused by several stressors. As mentioned earlier, ceramide accumulates under many forms of
2033
stress. A key study consolidating the “sphingolipid rheostat” is the report that S1P can inhibit PCD induced by ceramide [119]. In fact, the killing effects of several agents that enhance ceramide formation in the cell are significantly attenuated when SK is stimulated by an agonist or by overexpression of the enzyme. Conversely, knockdown of SK1 causes cellular accumulation of ceramide, activation of the intrinsic pathway, and leads to cell death [120]. The mechanisms by which SK and S1P exert their proliferative and survival effects appear to involve several key players in cell survival and cell death pathways (Fig. 3B). SK activation and S1P production mediate the resistance of HUVEC cells to TNF induced proapoptotic effects through the activation of NF-kB, a known anti-apoptotic transcription factor [121]. Other pro-survival mediators shown to be activated by SK/S1P include Akt [122], nitric oxide [123], and ERK. SK overexpression can also inhibit apoptotic pathways, as shown in PC12 cells, where the enzyme rescues from apoptosis through attenuation of JNK activation by serum starvation or ceramide [124]. In leukemia cells, S1P inhibits the release of mitochondrial proapoptotic factors caused by serum deprivation, antiFas, TNF, or ceramide [125]. Interestingly, the SK/S1P pathway appears also to be involved in anoikis. Nakamura et al. have shown that cells dying by anoikis have lower SK activity than adherent cells, and that inhibition of SK in the adherent cells causes cell death, whereas supplementation of S1P rescues nonadherent cells from anoikis [126]. SK is the only degradative pathway for the clearance of ceramide and sphingosine from the cell. The reaction catalyzed by sphingosine phosphate lyase (SPL) is the only irreversible catalytic reaction in the sphingolipid pathway in which the sphingosine backbone is hydrolyzed. Thus, the lyase reaction is the exit point of sphingolipids to the general cellular lipid pool, and any sphingolipid to be hydrolyzed has to be converted to S1P first, highlighting indirectly the importance of SK. Therefore, one straight forward explanation for the ability of SK overexpression to attenuate cell death caused by stressors that elevate ceramide may be due to increased clearance of a toxic lipid. However, the ability of S1P itself to overcome ceramide effects, also suggests that enhanced S1P production by SK activation is itself important for these biological effects. Even more interesting is the potential for S1P to feedback on de novo ceramide biosynthesis. Van Echten-Deckert et al. have shown that the conversion of sphingosine to S1P can inhibit SPT activity [127]. Moreover, in SPPase-1 overexpressing cells, S1P is dephosphorylated and ceramide synthesis is enhanced [128], suggesting that SK/S1P not only modulate clearance of a toxic lipid (ceramide), but they may also reduce its synthesis. These studies therefore, broaden the mechanisms by which SK/S1P can modulate cell death. The regulation of SK1 has also been shown to occur in the apoptotic pathway by a mechanism involving proteases. DNA damage and TNF both induce the loss of SK1 through a mechanism that is sensitive to inhibitors of the lysosomal protease cathepsin B [129,130]. Given the emerging role of the lysosome in PCD, these data add a further dimension to the mechanism of SK1 regulation, potentially implicating the
2034
T.A. Taha et al. / Biochimica et Biophysica Acta 1758 (2006) 2027–2036
enzyme as a potential mediator of lysosome-driven pathways of cell death. Despite the majority of studies demonstrating anti-apoptotic roles for SK and S1P, cloning of SK2 has yielded unexpected results. Although it produces the same lipid (S1P) as SK1, SK2 appears to be proapoptotic and growth inhibitory. SK2 has a putative BH3-only motif, which may be involved in its proapoptotic effects through the mitochondrial pathway [71]. Moreover, SK2 localizes in the nucleus of some cells under certain conditions, where it may inhibit DNA synthesis [131]. Whether these effects are dependent on S1P production or on the activity of SK2 remains to be determined. 5. Concluding remarks
[12] [13] [14] [15]
[16] [17]
[18] [19] [20] [21] [22]
Although the studies addressing the role of sphingolipids in cell death are numerous, many of the specific functions that these mediators regulate are still elusive. The evidence implicating ceramide in cell death and SK1 and S1P in cell survival is consistent, yet there appears to be differences in the mechanisms of cell signaling in different cell types as well as the signals generated by different stimuli. This suggests a versatile role for these lipids in cell function that can potentially be tailored for a therapeutic advantage. The sphingolipid rheostat appears to be a critical point in determining cell fate. Once more specific molecular targets are identified for the various lipid mediators, the exact mechanisms of action may become even clearer, and such interactions can be exploited for more efficient cell killing.
[23] [24] [25] [26] [27] [28] [29] [30]
[31]
Acknowledgments [32]
This work is supported by NIH grants R01 AG016583, R01 GM062887, and P01 CA097132. This material is also based upon work supported (or supported in part) in part by a MERIT Award to LMO by the Office of Research and Development, Department of Veterans Affairs, Ralph H. Johnson VA Medical Center, Charleston, South Carolina. References [1] M. Jaattela, Oncogene 23 (2004) 2746–2756. [2] X. Feng, Y.A. Hannun, J. Biol. Chem. 273 (1998) 26870–26874. [3] D.G. Breckenridge, M. Germain, J.P. Mathai, M. Nguyen, G.C. Shore, Oncogene 22 (2003) 8608–8618. [4] T. Miyashita, J.C. Reed, Cell 80 (1995) 293–299. [5] E. Oda, R. Ohki, H. Murasawa, J. Nemoto, T. Shibue, T. Yamashita, T. Tokino, T. Taniguchi, N. Tanaka, Science 288 (2000) 1053–1058. [6] J. Yu, L. Zhang, P.M. Hwang, K.W. Kinzler, B. Vogelstein, Mol. Cell. 7 (2001) 673–682. [7] K. Nakano, K.H. Vousden, Mol. Cell 7 (2001) 683–694. [8] V. Stoka, B. Turk, S.L. Schendel, T.H. Kim, T. Cirman, S.J. Snipas, L.M. Ellerby, D. Bredesen, H. Freeze, M. Abrahamson, D. Bromme, S. Krajewski, J.C. Reed, X.M. Yin, V. Turk, G.S. Salvesen, J. Biol. Chem. 276 (2001) 3149–3157. [9] L. Foghsgaard, D. Wissing, D. Mauch, U. Lademann, L. Bastholm, M. Boes, F. Elling, M. Leist, M. Jaattela, J. Cell Biol. 153 (2001) 999–1010. [10] B.C. Barnhart, E.C. Alappat, M.E. Peter, Semin. Immunol. 15 (2003) 185–193. [11] K.F. Ferri, G. Kroemer, Nat. Cell Biol. 3 (2001) E255–E263.
[33] [34] [35] [36] [37] [38] [39] [40] [41] [42] [43] [44] [45] [46] [47]
W.R. Sellers, D.E. Fisher, J. Clin. Invest. 104 (1999) 1655–1661. B.B. Wolf, D.R. Green, J. Biol. Chem. 274 (1999) 20049–20052. P. Nicotera, G. Melino, Oncogene 23 (2004) 2757–2765. M.A. Gallego, B. Joseph, T.H. Hemstrom, S. Tamiji, L. Mortier, G. Kroemer, P. Formstecher, B. Zhivotovsky, P. Marchetti, Oncogene 23 (2004) 6282–6291. H.P. Hahn, M. Pang, J. He, J.D. Hernandez, R.Y. Yang, L.Y. Li, X. Wang, F.T. Liu, L.G. Baum, Cell Death Differ. 11 (2004) 1277–1286. R. Hegde, S.M. Srinivasula, Z. Zhang, R. Wassell, R. Mukattash, L. Cilenti, G. DuBois, Y. Lazebnik, A.S. Zervos, T. Fernandes-Alnemri, E.S. Alnemri, J. Biol. Chem. 277 (2002) 432–438. C. Borner, Mol. Immunol. 39 (2003) 615–647. Y.A. Hannun, Science 274 (1996) 1855–1859. C.E. Chalfant, K. Rathman, R.L. Pinkerman, R.E. Wood, L.M. Obeid, B. Ogretmen, Y.A. Hannun, J. Biol. Chem. 277 (2002) 12587–12595. D.K. Perry, J. Carton, A.K. Shah, F. Meredith, D.J. Uhlinger, Y.A. Hannun, J. Biol. Chem. 275 (2000) 9078–9084. A. Lucci, T.Y. Han, Y.Y. Liu, A.E. Giuliano, M.C. Cabot, Int. J. Oncol. 15 (1999) 541–546. J.C. Strum, G.W. Small, S.B. Pauig, L.W. Daniel, J. Biol. Chem. 269 (1994) 15493–15497. A. Senchenkov, D.A. Litvak, M.C. Cabot, J. Natl. Cancer Inst. 93 (2001) 347–357. N.S. Radin, Biochem. J. 371 (2003) 243–256. S. Furuya, J. Mitoma, A. Makino, Y. Hirabayashi, J. Neurochem. 71 (1998) 366–377. E.M. Jung, R.D. Griner, R. Mann-Blakeney, W.B. Bollag, J. Invest. Dermatol. 110 (1998) 318–323. X.F. Zhu, Z.C. Liu, B.F. Xie, G.K. Feng, Y.X. Zeng, Cancer Lett. 193 (2003) 149–154. J.Y. Lee, L.G. Leonhardt, L.M. Obeid, Biochem. J. 334 (Pt 2) (1998) 457–461. G.S. Dbaibo, M.Y. Pushkareva, S. Jayadev, J.K. Schwarz, J.M. Horowitz, L.M. Obeid, Y.A. Hannun, Proc. Natl. Acad Sci. U S A 92 (1995) 1347–1351. Y. Uchida, A.D. Nardo, V. Collins, P.M. Elias, W.M. Holleran, J. Invest. Dermatol. 120 (2003) 662–669. M.E. Venable, J.Y. Lee, M.J. Smyth, A. Bielawska, L.M. Obeid, J. Biol. Chem. 270 (1995) 30701–30708. J.Y. Lee, A.E. Bielawska, L.M. Obeid, Exp. Cell Res. 261 (2000) 303–311. C.J. Miller, G.H. Stein, J. Gerontol., A Biol. Sci. Med. Sci. 56 (2001) B8–B19. M.E. Venable, A. Bielawska, L.M. Obeid, J. Biol. Chem. 271 (1996) 24800–24805. M.E. Venable, G.C. Blobe, L.M. Obeid, J. Biol. Chem. 269 (1994) 26040–26044. J.Y. Lee, Y.A. Hannun, L.M. Obeid, J. Biol. Chem. 271 (1996) 13169–13174. Y.A. Hannun, Adv. Exp. Med. Biol. 400A (1997) 305–312. G.S. Dbaibo, M.Y. Pushkareva, R.A. Rachid, N. Alter, M.J. Smyth, L.M. Obeid, Y.A. Hannun, J. Clin. Invest. 102 (1998) 329–339. M. Sawada, T. Kiyono, S. Nakashima, J. Shinoda, T. Naganawa, S. Hara, T. Iwama, N. Sakai, Cell Death Differ. 11 (2004) 997–1008. M. Pruschy, H. Resch, Y.Q. Shi, N. Aalame, C. Glanzmann, S. Bodis, Br. J. Cancer 80 (1999) 693–698. S.S. Kim, H.S. Chae, J.H. Bach, M.W. Lee, K.Y. Kim, W.B. Lee, Y.M. Jung, J.V. Bonventre, Y.H. Suh, Oncogene 21 (2002) 2020–2028. A. Bras, J.P. Albar, E. Leonardo, G.G. de Buitrago, A.C. Martinez, Cell Death Differ. 7 (2000) 262–271. J. Yang, P.J. Duerksen-Hughes, J. Biol. Chem. 276 (2001) 27129–27135. Y.Q. Shi, F.E. Wuergler, H. Blattmann, N.E. Crompton, Radiat. Environ. Biophys. 40 (2001) 301–308. S. Hara, S. Nakashima, T. Kiyono, M. Sawada, S. Yoshimura, T. Iwama, Y. Banno, J. Shinoda, N. Sakai, Cell Death Differ. 11 (2004) 853–861. P. Santana, L.A. Pena, A. Haimovitz-Friedman, S. Martin, D. Green, M. McLoughlin, C. Cordon-Cardo, E.H. Schuchman, Z. Fuks, R. Kolesnick, Cell 86 (1996) 189–199.
T.A. Taha et al. / Biochimica et Biophysica Acta 1758 (2006) 2027–2036 [48] J. Zhang, N. Alter, J.C. Reed, C. Borner, L.M. Obeid, Y.A. Hannun, Proc. Natl. Acad Sci. U. S. A. 93 (1996) 5325–5328. [49] M. Raisova, G. Goltz, M. Bektas, A. Bielawska, C. Riebeling, A.M. Hossini, J. Eberle, Y.A. Hannun, C.E. Orfanos, C.C. Geilen, FEBS Lett. 516 (2002) 47–52. [50] W. El-Assaad, M. El-Sabban, C. Awaraji, N. Abboushi, G.S. Dbaibo, Biochem. J. 336 (Pt 3) (1998) 735–741. [51] M. Sawada, S. Nakashima, Y. Banno, H. Yamakawa, K. Hayashi, K. Takenaka, Y. Nishimura, N. Sakai, Y. Nozawa, Cell Death Differ. 7 (2000) 761–772. [52] A.D. Tepper, E. de Vries, W.J. van Blitterswijk, J. Borst, J. Clin. Invest. 103 (1999) 971–978. [53] W. El-Assaad, L. Kozhaya, S. Araysi, S. Panjarian, F.F. Bitar, E. Baz, M.E. El-Sabban, G.S. Dbaibo, Biochem. J. 336 (1998) 735–741. [54] M. Sawada, S. Nakashima, Y. Banno, H. Yamakawa, K. Takenaka, J. Shinoda, Y. Nishimura, N. Sakai, Y. Nozawa, Oncogene 19 (2000) 3508–3520. [55] N. Marchesini, W. Osta, J. Bielawski, C. Luberto, L.M. Obeid, Y.A. Hannun, J. Biol. Chem. 279 (2004) 25101–25111. [56] B.J. Kroesen, S. Jacobs, B.J. Pettus, H. Sietsma, J.W. Kok, Y.A. Hannun, L.F. de Leij, J. Biol. Chem. 278 (2003) 14723–14731. [57] C. von Haefen, T. Wieder, B. Gillissen, L. Starck, V. Graupner, B. Dorken, P.T. Daniel, Oncogene 21 (2002) 4009–4019. [58] H. Birbes, C. Luberto, Y.T. Hsu, S.E. Bawab, Y.A. Hannun, L.M. Obeid, Biochem. J. 386 (2005) 445–451. [59] J.G. Pastorino, M. Tafani, R.J. Rothman, A. Marcinkeviciute, J.B. Hoek, J.L. Farber, J. Biol. Chem. 274 (1999) 31734–31739. [60] H. Kashkar, K. Wiegmann, B. Yazdanpanah, D. Haubert, M. Kroenke, J. Biol. Chem. 280 (2005) 20804–20813. [61] G.I. Perez, A. Jurisicova, T. Matikainen, T. Moriyama, M.R. Kim, Y. Takai, J.K. Pru, R.N. Kolesnick, J.L. Tilly, FASEB J. 19 (2005) 860–862. [62] M. Garcia-Barros, F. Paris, C. Cordon-Cardo, D. Lyden, S. Rafii, A. Haimovitz-Friedman, Z. Fuks, R. Kolesnick, Science 300 (2003) 1155–1159. [63] F. Paris, Z. Fuks, A. Kang, P. Capodieci, G. Juan, D. Ehleiter, A. Haimovitz-Friedman, C. Cordon-Cardo, R. Kolesnick, Science 293 (2001) 293–297. [64] T. Ito, X. Deng, B. Carr, W.S. May, J. Biol. Chem. 272 (1997) 11671–11673. [65] P.P. Ruvolo, X. Deng, T. Ito, B.K. Carr, W.S. May, J. Biol. Chem. 274 (1999) 20296–20300. [66] S.S. Lin, M.C. Bassik, H. Suh, M. Nishino, J.D. Arroyo, W.C. Hahn, S.J. Korsmeyer, T.M. Roberts, J. Biol. Chem. 281 (2006) 23003–23012. [67] M. Xin, X. Deng, J. Biol. Chem. 281 (2006) 28859–28867. [68] C. Bionda, J. Portoukalian, D. Schmitt, C. Rodriguez-Lafrasse, D. Ardail, Biochem. J. 382 (2004) 527–533. [69] H. Shimeno, S. Soeda, M. Sakamoto, T. Kouchi, T. Kowakame, T. Kihara, Lipids 33 (1998) 601–605. [70] S. El Bawab, P. Roddy, T. Qian, A. Bielawska, J.J. Lemasters, Y.A. Hannun, J. Biol. Chem. 275 (2000) 21508–21513. [71] H. Liu, R.E. Toman, S. Goparaju, M. Maceyka, V.E. Nava, H. Sankala, S.G. Payne, M. Bektas, I. Ishii, J. Chun, S. Milstien, S. Spiegel, J. Biol. Chem. 278 (2006) 40330–40336. [72] T.I. Gudz, K.Y. Tserng, C.L. Hoppel, J. Biol. Chem. 272 (1997) 24154–24158. [73] L.J. Siskind, R.N. Kolesnick, M. Colombini, J. Biol. Chem. 277 (2002) 26796–26803. [74] S.A. Novgorodov, Z.M. Szulc, C. Luberto, J.A. Jones, J. Bielawski, A. Bielawska, Y.A. Hannun, L.M. Obeid, J. Biol. Chem. 280 (2006) 16096–16105. [75] M. Di Paola, P. Zaccagnino, G. Montedoro, T. Cocco, M. Lorusso, J. Bioenerg. Biomembr. 36 (2004) 165–170. [76] P. Ghafourifar, S.D. Klein, O. Schucht, U. Schenk, M. Pruschy, S. Rocha, C. Richter, J. Biol. Chem. 274 (1999) 6080–6084. [77] P. Zhang, B. Liu, G.M. Jenkins, Y.A. Hannun, L.M. Obeid, J. Biol. Chem. 272 (1997) 9609–9612. [78] H. Birbes, S. El Bawab, Y.A. Hannun, L.M. Obeid, FASEB J. 15 (2001) 2669–2679.
2035
[79] A.D. Tepper, E. de Vries, W.J. van Blitterswijk, J. Borst, J. Clin. Invest. 103 (1999) 971–978. [80] Y.A. Hannun, C. Luberto, Trends Cell Biol. 10 (2000) 73–80. [81] M. Heinrich, M. Wickel, W. Schneider-Brachert, C. Sandberg, J. Gahr, R. Schwandner, T. Weber, P. Saftig, C. Peters, J. Brunner, M. Kronke, S. Schutze, EMBO J. 18 (1999) 5252–5263. [82] G.S. Wu, P. Saftig, C. Peters, W.S. El-Deiry, Oncogene 16 (1998) 2177–2183. [83] M. Heinrich, J. Neumeyer, M. Jakob, C. Hallas, V. Tchikov, S. WinotoMorbach, M. Wickel, W. Schneider-Brachert, A. Trauzold, A. Hethke, S. Schutze, Cell Death Differ. 11 (2004) 550–563. [84] J.A. Rotolo, J. Zhang, M. Donepudi, H. Lee, Z. Fuks, R. Kolesnick, J. Biol. Chem. 280 (2005) 26425–26434. [85] S. Daido, T. Kanzawa, A. Yamamoto, H. Takeuchi, Y. Kondo, S. Kondo, Cancer Res. 64 (2004) 4286–4293. [86] F. Scarlatti, C. Bauvy, A. Ventruti, G. Sala, F. Cluzeaud, A. Vandewalle, R. Ghidoni, P. Codogno, J. Biol. Chem. 279 (2004) 18384–18391. [87] O. Cuvillier, Biochim. Biophys. Acta 1585 (2002) 153–162. [88] E.A. Sweeney, J. Inokuchi, Y. Igarashi, FEBS Lett. 425 (1998) 61–65. [89] O. Cuvillier, L. Edsall, S. Spiegel, J. Biol. Chem. 275 (2000) 15691–15700. [90] O. Cuvillier, V.E. Nava, S.K. Murthy, L.C. Edsall, T. Levade, S. Milstien, S. Spiegel, Cell Death Differ. 8 (2001) 162–171. [91] W.J. Lee, H.S. Yoo, P.G. Suh, S. Oh, J.S. Lim, Y.M. Lee, Exp. Mol. Med. 36 (2004) 420–427. [92] W.D. Jarvis, F.A. Fornari Jr., K.L. Auer, A.J. Freemerman, E. Szabo, M.J. Birrer, C.R. Johnson, S.E. Barbour, P. Dent, S. Grant, Mol. Pharmacol. 52 (1997) 935–947. [93] H.C. Chang, L.H. Tsai, L.Y. Chuang, W.C. Hung, J. Cell Physiol. 188 (2001) 188–193. [94] W.C. Hung, H.C. Chang, L.Y. Chuang, Biochem. J. 338 (Pt 1) (1999) 161–166. [95] T. Shirahama, C. Sakakura, E.A. Sweeney, M. Ozawa, M. Takemoto, K. Nishiyama, Y. Ohi, Y. Igarashi, FEBS Lett. 407 (1997) 97–100. [96] C. Sakakura, E.A. Sweeney, T. Shirahama, S. Hakomori, Y. Igarashi, FEBS Lett. 379 (1996) 177–180. [97] C. Isogai, T. Murate, K. Tamiya-Koizumi, S. Yoshida, T. Ito, H. Nagai, T. Kinoshita, Y. Kagami, T. Hotta, M. Hamaguchi, H. Saito, Exp. Hematol. 26 (1998) 1118–1125. [98] Y.A. Hannun, R.M. Bell, Clin. Chim. Acta 185 (1989) 333–345. [99] Y.A. Hannun, R.M. Bell, Science 235 (1987) 670–674. [100] W.D. Jarvis, F.A. Fornari Jr., R.M. Tombes, R.K. Erukulla, R. Bittman, G.K. Schwartz, P. Dent, S. Grant, Mol. Pharmacol. 54 (1998) 844–856. [101] A. Hamaguchi, E. Suzuki, K. Murayama, T. Fujimura, T. Hikita, K. Iwabuchi, K. Handa, D.A. Withers, S.C. Masters, H. Fu, S. Hakomori, J. Biol. Chem. 278 (2003) 41557–41565. [102] T. Megidish, J. Cooper, L. Zhang, H. Fu, S. Hakomori, J. Biol. Chem. 273 (1998) 21834–21845. [103] F. Tsuruta, J. Sunayama, Y. Mori, S. Hattori, S. Shimizu, Y. Tsujimoto, K. Yoshioka, N. Masuyama, Y. Gotoh, EMBO J. 23 (2004) 1889–1899. [104] M. Nomura, S. Shimizu, T. Sugiyama, M. Narita, T. Ito, H. Matsuda, Y. Tsujimoto, J. Biol. Chem. 278 (2003) 2058–2065. [105] S.C. Masters, H. Yang, S.R. Datta, M.E. Greenberg, H. Fu, Mol. Pharmacol. 60 (2001) 1325–1331. [106] H. Yang, S.C. Masters, H. Wang, H. Fu, Biochim. Biophys. Acta 1547 (2001) 313–319. [107] C.W. Chiang, G. Harris, C. Ellig, S.C. Masters, R. Subramanian, S. Shenolikar, B.E. Wadzinski, E. Yang, Blood 97 (2001) 1289–1297. [108] H.G. Wang, N. Pathan, I.M. Ethell, S. Krajewski, Y. Yamaguchi, F. Shibasaki, F. McKeon, T. Bobo, T.F. Franke, J.C. Reed, Science 284 (1999) 339–343. [109] J. Zha, H. Harada, E. Yang, J. Jockel, S.J. Korsmeyer, Cell 87 (1996) 619–628. [110] M. Pushkareva, W.A. Khan, A.V. Alessenko, N. Sahyoun, Y.A. Hannun, J. Biol. Chem. 267 (1992) 15246–15251. [111] W. Hu, R. Xu, G. Zhang, J. Jin, Z.M. Szulc, J. Bielawski, Y.A. Hannun, L.M. Obeid, C. Mao, Mol. Biol. Cell 16 (2005) 1555–1567. [112] A.G. Rosenwald, R.E. Pagano, J. Biol. Chem. 268 (1993) 4577–4579. [113] M.T. Park, M.J. Kim, Y.H. Kang, S.Y. Choi, J.H. Lee, J.A. Choi, C.M.
2036
[114]
[115] [116] [117] [118] [119] [120] [121] [122]
T.A. Taha et al. / Biochimica et Biophysica Acta 1758 (2006) 2027–2036 Kang, C.K. Cho, S. Kang, S. Bae, Y.S. Lee, H.Y. Chung, S.J. Lee, Blood 105 (2005) 1724–1733. M.T. Park, J.A. Choi, M.J. Kim, H.D. Um, S. Bae, C.M. Kang, C.K. Cho, S. Kang, H.Y. Chung, Y.S. Lee, S.J. Lee, J. Biol. Chem. 278 (2003) 50624–50634. S. Spiegel, S. Milstien, Nat. Rev., Mol. Cell Biol. 4 (2003) 397–407. A. Olivera, S. Spiegel, Nature 365 (1993) 557–560. T. Kohama, A. Olivera, L. Edsall, M.M. Nagiec, R. Dickson, S. Spiegel, J. Biol. Chem. 273 (1998) 23722–23728. A. Olivera, T. Kohama, L. Edsall, V. Nava, O. Cuvillier, S. Poulton, S. Spiegel, J. Cell Biol. 147 (1999) 545–558. O. Cuvillier, G. Pirianov, B. Kleuser, P.G. Vanek, O.A. Coso, S. Gutkind, S. Spiegel, Nature 381 (1996) 800–803. T.A. Taha, K. Kitatani, M. El-Alwani, J. Bielawski, Y.A. Hannun, L.M. Obeid, FASEB J. 20 (2006) 482–484. P. Xia, L. Wang, J.R. Gamble, M.A. Vadas, J. Biol. Chem. 274 (1999) 34499–34505. Y. Osawa, Y. Banno, M. Nagaki, D.A. Brenner, T. Naiki, Y. Nozawa, S. Nakashima, H. Moriwaki, J. Immunol. 167 (2001) 173–180.
[123] Y.G. Kwon, J.K. Min, K.M. Kim, D.J. Lee, T.R. Billiar, Y.M. Kim, J. Biol. Chem. 276 (2001) 10627–10633. [124] L.C. Edsall, O. Cuvillier, S. Twitty, S. Spiegel, S. Milstien, J. Neurochem. 76 (2001) 1573–1584. [125] O. Cuvillier, T. Levade, Blood 98 (2001) 2828–2836. [126] H. Nakamura, T. Oda, K. Hamada, T. Hirano, N. Shimizu, H. Utiyama, J. Biol. Chem. 273 (1998) 15345–15351. [127] G. van Echten-Deckert, A. Zschoche, T. Bar, R.R. Schmidt, A. Raths, T. Heinemann, K. Sandhoff, J. Biol. Chem. 272 (1997) 15825–15833. [128] H. Le Stunff, I. Galve-Roperh, C. Peterson, S. Milstien, S. Spiegel, J. Cell Biol. 158 (2002) 1039–1049. [129] T.A. Taha, W. Osta, L. Kozhaya, J. Bielawski, K.R. Johnson, W.E. Gillanders, G.S. Dbaibo, Y.A. Hannun, L.M. Obeid, J. Biol. Chem. 279 (2004) 20546–20554. [130] T.A. Taha, K. Kitatani, J. Bielawski, W. Cho, Y.A. Hannun, L.M. Obeid, J. Biol. Chem. 280 (2005) 17196–171202. [131] N. Igarashi, T. Okada, S. Hayashi, T. Fujita, S. Jahangeer, S. Nakamura, J. Biol. Chem. 278 (2003) 46832–46839.
Review
A house divided: Ceramide, sphingosine, and sphingosine-1-phosphate in programmed cell death Tarek A. Taha a,b , Thomas D. Mullen a,b , Lina M. Obeid a,b,⁎ a
Division of General Internal Medicine, Ralph H. Johnson Veterans Administration Hospital, Charleston, SC 29401, USA b Department of Medicine, Medical University of South Carolina, Charleston, SC 29425, USA Received 18 July 2006; received in revised form 25 October 2006; accepted 28 October 2006 Available online 1 November 2006
Abstract Programmed cell death is an important physiological response to many forms of cellular stress. The signaling cascades that result in programmed cell death are as elaborate as those that promote cell survival, and it is clear that coordination of both protein- and lipid-mediated signals is crucial for proper cell execution. Sphingolipids are a large class of lipids whose diverse members share the common feature of a longchain sphingoid base, e.g., sphingosine. Many sphingolipids have been shown to play essential roles in both death signaling and survival. Ceramide, an N-acylsphingosine, has been implicated in cell death following a myriad of cellular stresses. Sphingosine itself can induce cell death but via pathways both similar and dissimilar to those of ceramide. Sphingosine-1-phosphate, on the other hand, is an anti-apoptotic molecule that mediates a host of cellular effects antagonistic to those of its pro-apoptotic sphingolipid siblings. Extraordinarily, these lipid mediators are metabolically juxtaposed, suggesting that the regulation of their metabolism is of the utmost importance in determining cell fate. In this review, we briefly examine the role of ceramide, sphingosine, and sphingosine-1-phosphate in programmed cell death and highlight the potential roles that these lipids play in the pathway to apoptosis. © 2006 Elsevier B.V. All rights reserved. Keywords: Programmed cell death; Ceramide; Sphingosine; Sphingosine-1-phosphate (S1P); Sphingolipid; Apoptosis
Contents 1. 2.
Overview of programmed cell death . . Ceramide and cell death . . . . . . . . 2.1. Ceramide and p53 . . . . . . . . 2.2. Ceramide and the Bcl-2 Family . 2.3. Ceramide and the mitochondrion 2.4. Ceramide and proteases . . . . . 2.5. Ceramide and autophagy . . . . 3. Sphingosine and cell death . . . . . . . 4. SK, S1P, and cell death. . . . . . . . . 5. Concluding remarks . . . . . . . . . . Acknowledgments . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
⁎ Corresponding author. Department of Medicine, Medical University of South Carolina, 114 Doughty St., P.O. Box 250779, Charleston, SC 29425, USA. E-mail address: [email protected] (L.M. Obeid). 0005-2736/$ - see front matter © 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.bbamem.2006.10.018
2028 2029 2030 2030 2031 2031 2031 2032 2033 2034 2034 2034
2028
T.A. Taha et al. / Biochimica et Biophysica Acta 1758 (2006) 2027–2036
1. Overview of programmed cell death Ongoing research in the field of cell death shows that the mechanisms by which cells die are as complex (if not more complex) than those by which cells survive. A common feature of all physiologic forms of death is that they are programmed events governed by specific biochemical pathways, hence the name programmed cell death or PCD [1]. While PCD is driven by complex pathways involving multiple players, it can be broken down into three stages for pragmatic purposes: initiation, commitment, and execution. The initiation step often occurs at or in the proximity of the cellular compartment where the stress is induced (Fig. 1) [2]. Disruption of calcium homeostasis at the ER, for example, results in the activation of calcium-mediated cell death [3]. Genotoxic stress in the nucleus induces p53 and drives p53-activated pathways of cell death [4–7]. Lysosome membrane disturbance causes the release of lysosomal proteases, which have also been implicated in PCD [8,9]. Activation of death receptors at the plasma membrane is another means of inducing PCD, which may involve the downstream activation of one or more organelle-mediated pro-apoptotic pathways [1,10]. Regardless of the initiation step, once a death signal is sensed by specific sensors, biochemical pathways are activated, that eventually converge onto the commitment step, or the point of no return, which in many instances, occurs at the level of the mitochondrion [11]. Once the death signal is relayed to the mitochondrion, the organelle incurs a major change: mitochondrial membrane permeabilization (MMP). MMP of the outer membrane causes
the release of several proteins from the intermembrane space into the cytosol. Cytochrome c, the most studied of these proteins, drives oligomerization of the adapter molecule APAF1, which then activates caspase 9, a key cysteine protease in the mitochondrial pathway of cell death [12]. Caspase 9 can then activate caspases 3 and 7, which belong to the execution phase of PCD, since they cleave proteins that are vital to normal cell function [13]. Another protein released from the mitochondrion is Smac/Diablo, which binds to and inhibits cellular proteins called IAPs (inhibitors of apoptosis) that normally keep caspases from undergoing erratic activation in the healthy state. Following release of cytochrome c and Smac, cells display several features of apoptotic PCD, the most prominent being chromatin condensation morphologically and cleavage of PARP and externalization of phosphatidylserine (PS) biochemically. It should be noted that caspase 9 activation and PS exposure on the outer leaflet of the cell membrane are ATPrequiring processes, and that depletion of ATP from cells can switch the death from an apoptotic form to a necrotic form [14]. In addition to the classical apoptotic form of PCD which involves the activation of caspases, the mitochondrion harbors other proteins, which upon release can drive alternative forms of cell death that are entirely caspase-independent. Following MMP, apoptosis inducing factor (AIF) [15] and endonuclease G (Endo G) [16] translocate from the mitochondrion to the nucleus and mediate apoptosis-like PCD, so called because they cause loose chromatin condensation that is caspase independent. HtrA2/Omi is also released from the mitochondrion and has a dual role in cell death, one dependent on its ability to
Fig. 1. Multiple pathways of cellular stress lead to MMP and apoptosis. Genotoxic stress activates p53 and “BH-3 only” proteins such as Noxa and Puma that promote MMP and the release of death factors from the mitochondrion. Cathepsins are released during lysosomal stress and result in the cleavage of Bid, which translocates to the mitochondrion to promote MMP. ER stress can be manifested by dysregulation of calcium homeostasis and subsequent accumulation of calcium in the mitochondria, leading to MMP. Finally, death receptor activation can lead to a multifaceted pathway that may involve lysosomal disruption, caspase 8-mediated Bid cleavage, mitochondrial disruption, and caspase activation.
T.A. Taha et al. / Biochimica et Biophysica Acta 1758 (2006) 2027–2036
inactivate IAPs and cause caspase-dependent PCD, and another due to its serine protease activity [17]. Since the mitochondrion constitutes an essential role in cell death, it is not surprising that the cell has developed an entire protein family to control the function of this organelle in the healthy vs. the apoptotic state. Briefly, members of the Bcl-2 family of proteins belong to one of three different classes: the Bcl-2 like survival factors, and these include Bcl-2 and Bcl-xL (in addition to other less characterized members); the Bax-like death factors, such as Bax and Bak; and the BH3-only death factors, such as Noxa, Puma, Bid, Bad, and Bik [18]. The most well characterized site of action of these proteins is the mitochondrion itself, although actions at other organelles are beginning to be proposed as well. Bcl-2-like proteins maintain mitochondrial integrity and are regarded as guardians of this organelle. In fact, overexpression of Bcl-2 and Bcl-xL rescues cells from death induced by several stimuli that normally induce MMP. Bax-like and BH3-only proteins, on the other hand, compromise mitochondrial function and essentially mediate the effect of several death inducers that cause MMP. Therefore, the integrity of the mitochondrion (and the survival of the cell) depends on a critical balance between the proapoptotic and the antiapoptotic members of the Bcl-2 family. Of note is that Bcl-2 proteins not only are key players in apoptosis, but also in the alternative forms of PCD, where MMP occurs. Classically, apoptotic pathways have been described as either intrinsic or extrinsic. The former indicates that the death signal is generated
2029
from within the cell, such as in the case of genotoxic stress causing cell death. The extrinsic pathway, on the other hand, indicates that the death signal is generated by binding of a death ligand to a cell surface receptor, which then activates the cascade of apoptosis. MMP is essentially a key feature of the intrinsic pathway. In cases where death ligands bind to their surface receptors, cell death can proceed independent of the mitochondrion and without MMP. Of note though is that the extrinsic pathway often cross talks with the intrinsic pathway of PCD, making the mitochondrion and the Bcl-2 proteins key player also in death ligand-mediated PCD. 2. Ceramide and cell death Given its central function, ceramide is the most studied sphingolipid species. Several lines of investigation have clearly established this lipid as a mediator of the eukaryotic stress response (Fig. 2) [19]. Ceramide accumulation can occur via: (i) activation of de novo synthesis mediated by serine-palmitoyl transferase (SPT) and ceramide synthase [20,21]; (ii) activation of sphingomyelin hydrolysis [22,23]; (iii) inhibition of ceramide hydrolysis; and/or (iv) stimulation of glucosylceramide hydrolysis or inhibition of its synthesis [24] (Fig. 1-1A). Some agents activate a multitude of these pathways, such as irradiation [25]. The effects of ceramide are pleiotropic, but for the most part growth inhibiting. The molecule has been implicated in differentiation [26,27], cell cycle arrest [28–30],
Fig. 2. Ceramide is a central mediator of many apoptotic pathways. Following genotoxic stress, ceramide accumulates, but it is still unclear whether the increase is dependent or independent of p53. Signaling through the lysosomes involves increases in ceramide via aSMase. Ceramide can directly activate the lysosomal protease cathepsin D which can subsequently cleave and activate Bid. Another direct target of ceramide is protein phosphatase 2A, which is known to negatively regulate the prosurvival kinase Akt/PKB. Both PP2A and Akt/PKB regulate the Bax/Bcl-2 rheostat; if the PP2A prevails, as is the case in ceramide signaling, Bax prevails over Bcl-2 and the mitochondrion is permeabilized. Death receptor activation can lead to ceramide formation via nSMase and caspase 8-mediated cleavage of Bid; lysosomes may also be activated downstream of death receptors, with both signals converging on MMP. Additionally, generation of ceramide from sphingomyelin hydrolysis in mitochondria results in Bax translocation and MMP.
2030
T.A. Taha et al. / Biochimica et Biophysica Acta 1758 (2006) 2027–2036
apoptosis [31], and senescence [32] in several cell types. It induces cell cycle arrest through the dephosphorylation of the Retinoblastoma gene product (Rb), activation of the cyclin dependent kinase inhibitor p21, and inhibition of the cyclin dependent kinase 2 (CDK2) [29,30,33]. These studies have shown that ceramide is upstream of cell cycle regulators, and that the generation of ceramide is necessary for growth arrest in response to stimuli that induce this arrest. The elevation of ceramide has also been shown to occur in senescent cells, possibly via the activation of neutral sphingomyelinase (nSMase) [34]. The senescent phenotype may be attributed to a defect in the phospholipase D/protein kinase C (PLD/PKC) pathway, and ceramide can inhibit both PLD and PKC [35–37]. One of the most studied roles of ceramide pertains to its function as a proapoptotic molecule. The accumulation of ceramide following treatment of cells with apoptotic agents has implicated this lipid in the biological responses of these agents [25]. Because of its apoptosis-inducing effects in cancer cells, ceramide has been termed the “tumor suppressor lipid” [38]. Several studies have attempted to define further the specific role of ceramide in the events of cell death. The tumor suppressor protein p53, the Bcl-2 family of proteins and several protease classes are all key components of the tumor response to stress insults, and ceramide has been linked to each of these mediators. Many studies, however, report variable associations among these proteins and ceramide, suggesting that different cells have dissimilar networks, and that deciphering the relevance of in vitro findings to in vivo tumor behavior is far from understood. 2.1. Ceramide and p53 Several studies have generated conflicting data on the relationship between ceramide and p53, which is virtually dysfunctional in all human tumors. While some reports show that p53 is upstream of ceramide in tumor stress responses [39,40], other studies implicate p53 as a downstream target [41–43], yet these latter studies utilize exogenous ceramide as the stress inducer, whereas the former use agents that generate endogenous ceramide. Still, other observations place ceramide and p53 in two separate and independent pathways in the apoptotic process [44,45]. In the latter case, treatment with chemotherapeutic agents generates ceramide in p53+/+ as well as p53−/− cells. One study has demonstrated that ceramide can be formed by acid sphingomyelinase (aSMase) in response to genotoxic stress only in cells lacking functional p53, suggesting that p53 can inhibit aSMase activation and ceramide generation [46]. Studies by Santana et al. also proposed a role for aSMase derived ceramide in the apoptotic response and highlighted that the response is different from that derived by p53 driven cellular responses [47]. Therefore, it remains unclear how ceramide and p53 are linked in PCD and there appears to be variations in the link between the two messengers in different cell death models. 2.2. Ceramide and the Bcl-2 Family An equally wide array of studies has investigated the relationship between ceramide and the Bcl-2 family of proteins,
both following treatment with genotoxic and nongenotoxic agents. The most studied connections pertain to Bcl-2 itself, Bcl-xL, and Bax. The position of ceramide with respect to Bcl-2 is variably reported. A number of groups have shown that ceramide is upstream of Bcl-2 in the apoptotic pathway since Bcl-2 overexpression rescues from cell death induced by ceramide [48] or by ceramidase inhibitors [49]. Moreover, ElAssaad et al. have shown that Bcl-2 and Bcl-xL define two different points of regulation of ceramide responses. While both proteins rescue from TNF mediated cell death, only Bcl-xL abrogates ceramide generation, while Bcl-2 does not, implying a pathway where ceramide is downstream of Bcl-xL but upstream of Bcl-2 [50]. Moreover, in C6 glioma cells, etoposide induces ceramide formation by neutral sphingomyelinase activation, which then increases the Bax/Bcl-2 ratio [51], and in A549 cells, exogenous ceramide and endogenous ceramide produced from gemcitabine treatment enhance the expression of proapoptotic Bcl-x (BCl-xs) and caspase 9 splice variants, demonstrating regulation of Bcl-2 family protein expression levels and splicing patterns by the bioactive lipid [20]. On the other hand, other studies have shown that Bcl-2 and or Bcl-xL overexpression attenuates ceramide accumulation following DNA damaging stimuli [52–54], thus implicating that Bcl-2 can also act upstream of the lipid. Of note, however, is that many studies of ceramide regulation in cell death have measured total ceramide levels. Recently, it has become evident that the fatty acid chain length of ceramides is an important determinant of the biological effect mediated by the bioactive lipid. Marchesini et al. reported that nSMase activation in confluent MCF-7 cells causes cell cycle arrest but not apoptosis, and that this effect is mediated by very long chain C24-ceramide species [55]. Kroesen et al., on the other hand, showed that cross linking of the B-cell receptor generates C16-ceramide upstream of the mitochondrion in a caspase independent manner, and that inhibition of C16 ceramide generation rescues from cell death [56]. In the same study, C24 ceramide is generated downstream of mitochondrial dysfunction in a caspase dependent manner. Therefore, it is clear from these data that not all ceramide species generate the same responses, and that the different biological effects caused by ceramide may be mediated by distinct molecular species of the lipid. Moreover, it should be noted that nonmitochondrial sites of action have been reported for the Bcl-2 family of proteins, warranting that conclusions made with respect to these proteins may not always be extrapolated to the mitochondrion. The connection between ceramide and Bax has also been reported by several studies. One study has shown that Bax overexpression does not affect ceramide production in response to etoposide [54], yet ceramide treatment in a prostate and colorectal cancer cell line induces apoptosis only when Bax is overexpressed, suggesting a link between the lipid and the protein [57]. Furthermore, Birbes et al. have recently shown that localized production of ceramide in the mitochondrion induces Bax oligomerization and drives cell death [58]. This study is in agreement with the observation that ceramide treatment of isolated mitochondria potentiates the Bax-mediated induction of mitochondrial permeability transition (MPT) [59]. Further-
T.A. Taha et al. / Biochimica et Biophysica Acta 1758 (2006) 2027–2036
2031
more, a study has shown that C16-ceramide production via acidic sphingomyelinase (aSMase) mediates Bax conformational change and induces MMP in vivo and in vitro [60]. Ceramide has also been implicated in oocyte death as its levels increase in oocytes prior to increases in Bax protein levels and the induction of apoptosis [61]. Endothelial cells of Bax and aSMase deficient mice were also recently shown to have higher resistance to radiation induced cell death, indicating that ceramide generation by aSMase may also be important in microvasculature homeostasis in addition to cancer cell homeostasis [62,63]. Ceramide may regulate Bcl-2 and Bax simultaneously by modulating their phosphorylation states. It was established several years ago that Bcl-2, whose phosphorylation by PKCα at serine 70 is required for its anti-apoptotic function, becomes dephosphorylated in response to ceramide [64,65]. Recent work suggests that the ceramide-activated protein phosphatase PP2A may inhibit Bcl-2 via its dephosphorylation and proteasomal degradation [66]. On the other hand, Bax may also be regulated via PP2A, because Xin et al. showed that Bax is dephosphorylated following ceramide treatment in a PP2A-dependent manner [67]. Dephosphorylation of Bax was associated with its conformational change and subsequent release of cytochrome c from the mitochondria. Therefore, these studies display a proximal connection between ceramide, Bcl-2, and Bax, and implicate that ceramide may be a regulator of Bcl-2 family members and their control over cell death pathways.
of solutes into the cytosol [74]. Perhaps ceramide can mediate multiple different pathways to release pro-apoptotic factors from the mitochondrion. Lorusso and colleagues showed that ceramide could induce cytochrome c release from isolated mitochondria in both a PTP-dependent and independent manner [75]. Ceramide also maintains cytochrome c in an oxidized state, a status which appears to be necessary for reducing mitochondrial oxygen consumption, dissipating the mitochondrial membrane potential (Δψm), and releasing calcium from the mitochondria [76]. Hence, all these studies point to a direct effect of ceramide on the mitochondrion. While the above in vitro studies provide a direct link between ceramide and the mitochondrion, studies clearly defining the role that ceramide plays in cells have been very scarce, mostly owing to the difficulty in tracking the localization of ceramide. It is clear from the study by Zhang et al. that the topology of ceramide generation is essential for the lipid to mediate its downstream effects, since generation of the lipid at the plasma membrane is insufficient to induce cell death, whereas its generation within the cell is [77]. Further investigation into organelle-specific sphingomyelin hydrolysis has shown that only ceramide generation at the mitochondrion can drive cell death, whereas its formation in other organelles does not [78]. These studies demonstrate that once ceramide reaches the mitochondrion, it can induce the effects that the bioactive lipid exerts on the isolated organelle, making the in vitro observations relevant in vivo as well.
2.3. Ceramide and the mitochondrion
2.4. Ceramide and proteases
In addition to its link to the Bcl-2 family of proteins, an intimate connection is becoming more evident between ceramide and the mitochondrion. First, several enzymes involved in sphingolipid metabolism have been identified in subcellular fractions containing mitochondria. Bionda et al. have reported ceramide synthase and reverse ceramidase activities in the mitochondria and/or ER-related mitochondria associated membranes (MAMs) of rat liver [68]. Ceramide synthase has also been partially purified from bovine liver mitochondria [69]. Furthermore, a mammalian ceramidase localized in a similar compartment has recently been cloned [70], and sphingosine kinase type 2 has been described as a BH3-only like protein [71]. Therefore, more evidence is pointing towards machinery for sphingolipid biosynthesis at the level of the mitochondrion. The localization of these activities has been suggestive of the importance of generating the right lipid in the right place to induce its specific effects. In vitro experiments have revealed that ceramide can induce several changes in isolated mitochondria. Gudz et al. have shown that C2-ceramide can inhibit oxidative phosphorylation by interfering with the electron transport system at the level of complex III [72]. Ceramide also increases the permeability of the outer mitochondrial membrane to several proteins, including cytochrome c [73]. A recent study reported the ability of ceramide to enhance permeability of the inner mitochondrial membrane, via action on the permeability transition pore (PTP) and the electrogenic proton channel, which enhances the release
Other important components of PCD pathways are the proteases, which fall under two general categories: caspases and non-caspase proteases, both of which can be activated by ceramide. While the mediators linking initiator death signals to the activation of the mitochondrion are still poorly characterized, the sequence of events involved in the activation of caspases downstream of the mitochondrion is fairly well established. Extensive work shows that ceramide accumulates in apoptotic cells prior to the activation of execution caspases, but downstream of initiator caspases [40,79,80]. Ceramide also activates the non-caspase protease, cathepsin D [81] (Fig. 2), a mediator of cell death in several p53-dependent death model systems [82]. The formation of ceramide via acidic sphingomyelinase has also been shown to mediate the cleavage of Bid and downstream caspase activation in response to TNF [83]. Also, ceramide generation by aSmases has been suggested to mediate cell death by caspase dependent and independent mechanisms depending on the death stimulus [84]. 2.5. Ceramide and autophagy The role of ceramide in cell death pathways has been recently extended to encompass autophagy or Type II PCD (in contrast to apoptosis or Type I PCD). C2-ceramide upregulates the autophagy genes BNIP3 [85] and Beclin-1 [86] and de novo ceramide synthesis inhibitors can attenuate the induction of autophagy by Tamoxifen in MCF-7 cells [86]. Therefore, the
2032
T.A. Taha et al. / Biochimica et Biophysica Acta 1758 (2006) 2027–2036
role of ceramide as a universal mediator of various forms of PCD may be gaining more ground. 3. Sphingosine and cell death Several studies have pointed to a specific role for sphingosine in the induction of cell death independent of ceramide (Fig. 3A) [87]. In HL-60 cells, both ceramide and sphingosine accumulate before the features of cell death appear, yet their order in the pathway of cell death may be different [88]. While inhibition of caspases with Z-VAD-fmk rescues from apoptosis induced by both ceramide and sphingosine, z-IETD-fmk, the caspase-8 inhibitor and Z-AAD-fmk, the granzyme B inhibitor, are only effective against sphingosine but not ceramide [88]. The ceramide synthase inhibitor fumonisin B1 also does not inhibit sphingosine induced apoptosis in HL-60 and Jurkat T cells, indicating that the conversion of sphingosine to ceramide may not be required for the induction of cell death by sphingosine [88,89]. In MCF-7 cells, however, FB1 or D-MAPP, an inhibitor of alkaline ceramidase, are both effective in reducing apoptosis caused by exogenous sphingosine, suggesting that conversion to a ceramide intermediate may be important for the execution of sphingosine effects in some cells [90]. A recent study has reported that apoptosis induction by FTY720 is due to the ability of this compound to inhibit SK in vitro and accumulate sphingosine in vivo [91]. Concentrations of exogenous sphingosine that produce similar levels of intracellular sphingosine as FTY720 cause apoptotic features as well, implicating sphingosine as an active molecule in cell death [91]. The mechanisms mediating sphingosine-driven cell death are multiple, and in some cases do not coincide with those mediating ceramide death pathways [90]. Both lipids have been shown to induce apoptosis in U937 human monoblastic leukemia cells, yet ceramide-associated lethality involves the strong activation of JNK and weak inhibition of ERK, whereas sphingosine-driven death is associated with weak JNK activation and strong ERK inhibition [92]. Sphingosine can also inhibit Akt/Protein kinase B (PKB), and overexpression of
constitutively active myristoylated Akt partially rescues from sphingosine-induced death [93]. Sphingosine-mediated apoptosis has been shown to occur via caspase-dependent pathways. Sphingosine can induce Bid cleavage, mitochondrial cytochrome c release as well as the activation of downstream effector caspases 3 or 7 and the cleavage of PARP [89,90,94]. The involvement of the mitochondrial pathway of PCD is further illustrated by the effects of sphingosine on the Bcl-2 family of proteins. Studies have shown that sphingosine can reduce the expression levels of Bcl-2 and Bcl-xL, major gatekeepers of the mitochondrion [95,96]. Sphingosine-induced cell death has also been proposed to occur even in Bcl-2 transfected cells and to involve cleavage of Bax, and the association of the cleaved (and full length) Bax with Bcl-2 [97]. Sphingosine can also exert its effects via regulation of PKC. Initially, the bioactive lipid was found to inhibit PKC [98,99], an important prosurvival signal in the cell [100]. Recently, a sphingosine-dependent protein kinase (SDK1) has been described as a truncated version of PKCδ (a proapoptotic PKC), which forms by a caspase-3 dependent cleavage of the fulllength PKCδ [101]. While full-length PKCδ is inhibited by sphingosine and DMS, SDK1 activity is stimulated by the two lysophospholipids [101]. Among the substrates for SDK1 are the 14-3-3 proteins, specifically isoforms β, η, and ζ but not τ or σ [102]. 14-3-3 proteins are chaperones for several cellular proteins, including members of the Bcl-2 family, such as Bax [103,104] and Bad [105–108] (Bad requires phosphorylation itself to bind to 14-3-3 [109]). Phosphorylation of 14-3-3 is often associated with the release of the proapoptotic proteins, allowing them to exert their cell killing effects. Even though inhibition of PKC is a mechanism of sphingosine-mediated cell death, sphingosine can still induce apoptosis in cell systems were PKC is stimulated [94], suggesting the existence of sphingosine-responsive PKC-insensitive pathways of cell death. Multiple sphingosine activated kinases (other than SDK1) have also been suggested to exist, yet the identification of these awaits further studies [110].
Fig. 3. Sphingosine and S1P in apoptosis. (A) Sphingosine mediates cell death through the inhibition of the prosurvival factors PKC and Akt/PKB and induces death via SDK1, which leads to Bax activation through inhibition of 14-3-3 proteins. (B) S1P inhibits apoptosis by activating prosurvival factors such as Akt/PKB, NF-κB, ERK, and NO. Ceramide-induced death may also be inhibited by S1P through the inactivation JNK or other factors that promote PCD.
T.A. Taha et al. / Biochimica et Biophysica Acta 1758 (2006) 2027–2036
Besides its action through the classical mitochondrial pathway, sphingosine can also exert effects on other organelles. It has been shown that HeLa cells exposed to ceramide undergo massive Golgi fragmentation and inhibition of β1 integrin glycosylation and transport to the cell surface prior to the induction of apoptosis [111]. Exogenous sphingosine results in similar effects as exogenous ceramide, suggesting that the ceramide effects are due to its conversion to sphingosine in the Golgi complex [111]. The resulting loss of integrin signaling then leads to cell death by anoikis [111]. Almost a decade before that, the work of Rosenwald and Pagano also suggested a role for ceramide in slowing transport of viral glycoprotein across the Golgi apparatus, which was also disrupted by the lipid as well [112]. In that study, sphingosine had no effect, suggesting that the lipids may have slightly variable effects in different cells. There appears to be specific lipid mediated responses though because stereoisomers of ceramide were unable to cause changes in the Golgi complex [111]. Phytosphingosine, a saturated close relative of sphingosine having a hydroxyl group at the carbon-4 position (Fig. 1-4), is also a mediator of cell death. In combination with gamma irradiation, phytosphingosine enhances apoptosis of radiation resistant T cells by promoting reactive oxygen species (ROS) formation and AIF release, in a caspase independent manner [113]. The mechanism of phytosphingosine-induced cell death has been described to involve the inhibition of ERK, which drives caspase-8 activation, and the stimulation of p38 MAPK, which mediates Bax translocation and mitochondrial apoptosis [114]. 4. SK, S1P, and cell death Of particular interest in the biology of sphingolipids are the antagonistic roles that ceramide and sphingosine, on one hand, and sphingosine-1-phosphate, on the other, appear to exert in a variety of cells. Ceramide and sphingosine mediate apoptosis, cell cycle arrest, and differentiation, whereas S1P promotes proliferation, survival, and inhibition of apoptosis [115]. These findings have lead to the so-called sphingolipid rheostat which proposes that the relative levels of these lipids (especially in response to stressors or growth stimuli) are important determinants of cell fate [115]. One of the earliest observations implicating S1P as a progrowth lipid was the ability of PDGF and fetal calf serum to induce growth via the activation of sphingosine kinase and the formation of S1P, then called an intracellular second messenger [116]. Several growth stimuli were then shown to activate the SK/S1P pathway, which in turn appears to be required for the activation of downstream signaling by these stimuli. In addition, overexpression of SK1 in itself decreases ceramide and sphingosine levels and increases S1P [117], a sphingolipid conversion that is associated with enhanced proliferation, G1-S transition, and increased DNA synthesis [118]. The role of SK and S1P in PCD has come from several studies showing that these two mediators can attenuate death and growth inhibitory effects caused by several stressors. As mentioned earlier, ceramide accumulates under many forms of
2033
stress. A key study consolidating the “sphingolipid rheostat” is the report that S1P can inhibit PCD induced by ceramide [119]. In fact, the killing effects of several agents that enhance ceramide formation in the cell are significantly attenuated when SK is stimulated by an agonist or by overexpression of the enzyme. Conversely, knockdown of SK1 causes cellular accumulation of ceramide, activation of the intrinsic pathway, and leads to cell death [120]. The mechanisms by which SK and S1P exert their proliferative and survival effects appear to involve several key players in cell survival and cell death pathways (Fig. 3B). SK activation and S1P production mediate the resistance of HUVEC cells to TNF induced proapoptotic effects through the activation of NF-kB, a known anti-apoptotic transcription factor [121]. Other pro-survival mediators shown to be activated by SK/S1P include Akt [122], nitric oxide [123], and ERK. SK overexpression can also inhibit apoptotic pathways, as shown in PC12 cells, where the enzyme rescues from apoptosis through attenuation of JNK activation by serum starvation or ceramide [124]. In leukemia cells, S1P inhibits the release of mitochondrial proapoptotic factors caused by serum deprivation, antiFas, TNF, or ceramide [125]. Interestingly, the SK/S1P pathway appears also to be involved in anoikis. Nakamura et al. have shown that cells dying by anoikis have lower SK activity than adherent cells, and that inhibition of SK in the adherent cells causes cell death, whereas supplementation of S1P rescues nonadherent cells from anoikis [126]. SK is the only degradative pathway for the clearance of ceramide and sphingosine from the cell. The reaction catalyzed by sphingosine phosphate lyase (SPL) is the only irreversible catalytic reaction in the sphingolipid pathway in which the sphingosine backbone is hydrolyzed. Thus, the lyase reaction is the exit point of sphingolipids to the general cellular lipid pool, and any sphingolipid to be hydrolyzed has to be converted to S1P first, highlighting indirectly the importance of SK. Therefore, one straight forward explanation for the ability of SK overexpression to attenuate cell death caused by stressors that elevate ceramide may be due to increased clearance of a toxic lipid. However, the ability of S1P itself to overcome ceramide effects, also suggests that enhanced S1P production by SK activation is itself important for these biological effects. Even more interesting is the potential for S1P to feedback on de novo ceramide biosynthesis. Van Echten-Deckert et al. have shown that the conversion of sphingosine to S1P can inhibit SPT activity [127]. Moreover, in SPPase-1 overexpressing cells, S1P is dephosphorylated and ceramide synthesis is enhanced [128], suggesting that SK/S1P not only modulate clearance of a toxic lipid (ceramide), but they may also reduce its synthesis. These studies therefore, broaden the mechanisms by which SK/S1P can modulate cell death. The regulation of SK1 has also been shown to occur in the apoptotic pathway by a mechanism involving proteases. DNA damage and TNF both induce the loss of SK1 through a mechanism that is sensitive to inhibitors of the lysosomal protease cathepsin B [129,130]. Given the emerging role of the lysosome in PCD, these data add a further dimension to the mechanism of SK1 regulation, potentially implicating the
2034
T.A. Taha et al. / Biochimica et Biophysica Acta 1758 (2006) 2027–2036
enzyme as a potential mediator of lysosome-driven pathways of cell death. Despite the majority of studies demonstrating anti-apoptotic roles for SK and S1P, cloning of SK2 has yielded unexpected results. Although it produces the same lipid (S1P) as SK1, SK2 appears to be proapoptotic and growth inhibitory. SK2 has a putative BH3-only motif, which may be involved in its proapoptotic effects through the mitochondrial pathway [71]. Moreover, SK2 localizes in the nucleus of some cells under certain conditions, where it may inhibit DNA synthesis [131]. Whether these effects are dependent on S1P production or on the activity of SK2 remains to be determined. 5. Concluding remarks
[12] [13] [14] [15]
[16] [17]
[18] [19] [20] [21] [22]
Although the studies addressing the role of sphingolipids in cell death are numerous, many of the specific functions that these mediators regulate are still elusive. The evidence implicating ceramide in cell death and SK1 and S1P in cell survival is consistent, yet there appears to be differences in the mechanisms of cell signaling in different cell types as well as the signals generated by different stimuli. This suggests a versatile role for these lipids in cell function that can potentially be tailored for a therapeutic advantage. The sphingolipid rheostat appears to be a critical point in determining cell fate. Once more specific molecular targets are identified for the various lipid mediators, the exact mechanisms of action may become even clearer, and such interactions can be exploited for more efficient cell killing.
[23] [24] [25] [26] [27] [28] [29] [30]
[31]
Acknowledgments [32]
This work is supported by NIH grants R01 AG016583, R01 GM062887, and P01 CA097132. This material is also based upon work supported (or supported in part) in part by a MERIT Award to LMO by the Office of Research and Development, Department of Veterans Affairs, Ralph H. Johnson VA Medical Center, Charleston, South Carolina. References [1] M. Jaattela, Oncogene 23 (2004) 2746–2756. [2] X. Feng, Y.A. Hannun, J. Biol. Chem. 273 (1998) 26870–26874. [3] D.G. Breckenridge, M. Germain, J.P. Mathai, M. Nguyen, G.C. Shore, Oncogene 22 (2003) 8608–8618. [4] T. Miyashita, J.C. Reed, Cell 80 (1995) 293–299. [5] E. Oda, R. Ohki, H. Murasawa, J. Nemoto, T. Shibue, T. Yamashita, T. Tokino, T. Taniguchi, N. Tanaka, Science 288 (2000) 1053–1058. [6] J. Yu, L. Zhang, P.M. Hwang, K.W. Kinzler, B. Vogelstein, Mol. Cell. 7 (2001) 673–682. [7] K. Nakano, K.H. Vousden, Mol. Cell 7 (2001) 683–694. [8] V. Stoka, B. Turk, S.L. Schendel, T.H. Kim, T. Cirman, S.J. Snipas, L.M. Ellerby, D. Bredesen, H. Freeze, M. Abrahamson, D. Bromme, S. Krajewski, J.C. Reed, X.M. Yin, V. Turk, G.S. Salvesen, J. Biol. Chem. 276 (2001) 3149–3157. [9] L. Foghsgaard, D. Wissing, D. Mauch, U. Lademann, L. Bastholm, M. Boes, F. Elling, M. Leist, M. Jaattela, J. Cell Biol. 153 (2001) 999–1010. [10] B.C. Barnhart, E.C. Alappat, M.E. Peter, Semin. Immunol. 15 (2003) 185–193. [11] K.F. Ferri, G. Kroemer, Nat. Cell Biol. 3 (2001) E255–E263.
[33] [34] [35] [36] [37] [38] [39] [40] [41] [42] [43] [44] [45] [46] [47]
W.R. Sellers, D.E. Fisher, J. Clin. Invest. 104 (1999) 1655–1661. B.B. Wolf, D.R. Green, J. Biol. Chem. 274 (1999) 20049–20052. P. Nicotera, G. Melino, Oncogene 23 (2004) 2757–2765. M.A. Gallego, B. Joseph, T.H. Hemstrom, S. Tamiji, L. Mortier, G. Kroemer, P. Formstecher, B. Zhivotovsky, P. Marchetti, Oncogene 23 (2004) 6282–6291. H.P. Hahn, M. Pang, J. He, J.D. Hernandez, R.Y. Yang, L.Y. Li, X. Wang, F.T. Liu, L.G. Baum, Cell Death Differ. 11 (2004) 1277–1286. R. Hegde, S.M. Srinivasula, Z. Zhang, R. Wassell, R. Mukattash, L. Cilenti, G. DuBois, Y. Lazebnik, A.S. Zervos, T. Fernandes-Alnemri, E.S. Alnemri, J. Biol. Chem. 277 (2002) 432–438. C. Borner, Mol. Immunol. 39 (2003) 615–647. Y.A. Hannun, Science 274 (1996) 1855–1859. C.E. Chalfant, K. Rathman, R.L. Pinkerman, R.E. Wood, L.M. Obeid, B. Ogretmen, Y.A. Hannun, J. Biol. Chem. 277 (2002) 12587–12595. D.K. Perry, J. Carton, A.K. Shah, F. Meredith, D.J. Uhlinger, Y.A. Hannun, J. Biol. Chem. 275 (2000) 9078–9084. A. Lucci, T.Y. Han, Y.Y. Liu, A.E. Giuliano, M.C. Cabot, Int. J. Oncol. 15 (1999) 541–546. J.C. Strum, G.W. Small, S.B. Pauig, L.W. Daniel, J. Biol. Chem. 269 (1994) 15493–15497. A. Senchenkov, D.A. Litvak, M.C. Cabot, J. Natl. Cancer Inst. 93 (2001) 347–357. N.S. Radin, Biochem. J. 371 (2003) 243–256. S. Furuya, J. Mitoma, A. Makino, Y. Hirabayashi, J. Neurochem. 71 (1998) 366–377. E.M. Jung, R.D. Griner, R. Mann-Blakeney, W.B. Bollag, J. Invest. Dermatol. 110 (1998) 318–323. X.F. Zhu, Z.C. Liu, B.F. Xie, G.K. Feng, Y.X. Zeng, Cancer Lett. 193 (2003) 149–154. J.Y. Lee, L.G. Leonhardt, L.M. Obeid, Biochem. J. 334 (Pt 2) (1998) 457–461. G.S. Dbaibo, M.Y. Pushkareva, S. Jayadev, J.K. Schwarz, J.M. Horowitz, L.M. Obeid, Y.A. Hannun, Proc. Natl. Acad Sci. U S A 92 (1995) 1347–1351. Y. Uchida, A.D. Nardo, V. Collins, P.M. Elias, W.M. Holleran, J. Invest. Dermatol. 120 (2003) 662–669. M.E. Venable, J.Y. Lee, M.J. Smyth, A. Bielawska, L.M. Obeid, J. Biol. Chem. 270 (1995) 30701–30708. J.Y. Lee, A.E. Bielawska, L.M. Obeid, Exp. Cell Res. 261 (2000) 303–311. C.J. Miller, G.H. Stein, J. Gerontol., A Biol. Sci. Med. Sci. 56 (2001) B8–B19. M.E. Venable, A. Bielawska, L.M. Obeid, J. Biol. Chem. 271 (1996) 24800–24805. M.E. Venable, G.C. Blobe, L.M. Obeid, J. Biol. Chem. 269 (1994) 26040–26044. J.Y. Lee, Y.A. Hannun, L.M. Obeid, J. Biol. Chem. 271 (1996) 13169–13174. Y.A. Hannun, Adv. Exp. Med. Biol. 400A (1997) 305–312. G.S. Dbaibo, M.Y. Pushkareva, R.A. Rachid, N. Alter, M.J. Smyth, L.M. Obeid, Y.A. Hannun, J. Clin. Invest. 102 (1998) 329–339. M. Sawada, T. Kiyono, S. Nakashima, J. Shinoda, T. Naganawa, S. Hara, T. Iwama, N. Sakai, Cell Death Differ. 11 (2004) 997–1008. M. Pruschy, H. Resch, Y.Q. Shi, N. Aalame, C. Glanzmann, S. Bodis, Br. J. Cancer 80 (1999) 693–698. S.S. Kim, H.S. Chae, J.H. Bach, M.W. Lee, K.Y. Kim, W.B. Lee, Y.M. Jung, J.V. Bonventre, Y.H. Suh, Oncogene 21 (2002) 2020–2028. A. Bras, J.P. Albar, E. Leonardo, G.G. de Buitrago, A.C. Martinez, Cell Death Differ. 7 (2000) 262–271. J. Yang, P.J. Duerksen-Hughes, J. Biol. Chem. 276 (2001) 27129–27135. Y.Q. Shi, F.E. Wuergler, H. Blattmann, N.E. Crompton, Radiat. Environ. Biophys. 40 (2001) 301–308. S. Hara, S. Nakashima, T. Kiyono, M. Sawada, S. Yoshimura, T. Iwama, Y. Banno, J. Shinoda, N. Sakai, Cell Death Differ. 11 (2004) 853–861. P. Santana, L.A. Pena, A. Haimovitz-Friedman, S. Martin, D. Green, M. McLoughlin, C. Cordon-Cardo, E.H. Schuchman, Z. Fuks, R. Kolesnick, Cell 86 (1996) 189–199.
T.A. Taha et al. / Biochimica et Biophysica Acta 1758 (2006) 2027–2036 [48] J. Zhang, N. Alter, J.C. Reed, C. Borner, L.M. Obeid, Y.A. Hannun, Proc. Natl. Acad Sci. U. S. A. 93 (1996) 5325–5328. [49] M. Raisova, G. Goltz, M. Bektas, A. Bielawska, C. Riebeling, A.M. Hossini, J. Eberle, Y.A. Hannun, C.E. Orfanos, C.C. Geilen, FEBS Lett. 516 (2002) 47–52. [50] W. El-Assaad, M. El-Sabban, C. Awaraji, N. Abboushi, G.S. Dbaibo, Biochem. J. 336 (Pt 3) (1998) 735–741. [51] M. Sawada, S. Nakashima, Y. Banno, H. Yamakawa, K. Hayashi, K. Takenaka, Y. Nishimura, N. Sakai, Y. Nozawa, Cell Death Differ. 7 (2000) 761–772. [52] A.D. Tepper, E. de Vries, W.J. van Blitterswijk, J. Borst, J. Clin. Invest. 103 (1999) 971–978. [53] W. El-Assaad, L. Kozhaya, S. Araysi, S. Panjarian, F.F. Bitar, E. Baz, M.E. El-Sabban, G.S. Dbaibo, Biochem. J. 336 (1998) 735–741. [54] M. Sawada, S. Nakashima, Y. Banno, H. Yamakawa, K. Takenaka, J. Shinoda, Y. Nishimura, N. Sakai, Y. Nozawa, Oncogene 19 (2000) 3508–3520. [55] N. Marchesini, W. Osta, J. Bielawski, C. Luberto, L.M. Obeid, Y.A. Hannun, J. Biol. Chem. 279 (2004) 25101–25111. [56] B.J. Kroesen, S. Jacobs, B.J. Pettus, H. Sietsma, J.W. Kok, Y.A. Hannun, L.F. de Leij, J. Biol. Chem. 278 (2003) 14723–14731. [57] C. von Haefen, T. Wieder, B. Gillissen, L. Starck, V. Graupner, B. Dorken, P.T. Daniel, Oncogene 21 (2002) 4009–4019. [58] H. Birbes, C. Luberto, Y.T. Hsu, S.E. Bawab, Y.A. Hannun, L.M. Obeid, Biochem. J. 386 (2005) 445–451. [59] J.G. Pastorino, M. Tafani, R.J. Rothman, A. Marcinkeviciute, J.B. Hoek, J.L. Farber, J. Biol. Chem. 274 (1999) 31734–31739. [60] H. Kashkar, K. Wiegmann, B. Yazdanpanah, D. Haubert, M. Kroenke, J. Biol. Chem. 280 (2005) 20804–20813. [61] G.I. Perez, A. Jurisicova, T. Matikainen, T. Moriyama, M.R. Kim, Y. Takai, J.K. Pru, R.N. Kolesnick, J.L. Tilly, FASEB J. 19 (2005) 860–862. [62] M. Garcia-Barros, F. Paris, C. Cordon-Cardo, D. Lyden, S. Rafii, A. Haimovitz-Friedman, Z. Fuks, R. Kolesnick, Science 300 (2003) 1155–1159. [63] F. Paris, Z. Fuks, A. Kang, P. Capodieci, G. Juan, D. Ehleiter, A. Haimovitz-Friedman, C. Cordon-Cardo, R. Kolesnick, Science 293 (2001) 293–297. [64] T. Ito, X. Deng, B. Carr, W.S. May, J. Biol. Chem. 272 (1997) 11671–11673. [65] P.P. Ruvolo, X. Deng, T. Ito, B.K. Carr, W.S. May, J. Biol. Chem. 274 (1999) 20296–20300. [66] S.S. Lin, M.C. Bassik, H. Suh, M. Nishino, J.D. Arroyo, W.C. Hahn, S.J. Korsmeyer, T.M. Roberts, J. Biol. Chem. 281 (2006) 23003–23012. [67] M. Xin, X. Deng, J. Biol. Chem. 281 (2006) 28859–28867. [68] C. Bionda, J. Portoukalian, D. Schmitt, C. Rodriguez-Lafrasse, D. Ardail, Biochem. J. 382 (2004) 527–533. [69] H. Shimeno, S. Soeda, M. Sakamoto, T. Kouchi, T. Kowakame, T. Kihara, Lipids 33 (1998) 601–605. [70] S. El Bawab, P. Roddy, T. Qian, A. Bielawska, J.J. Lemasters, Y.A. Hannun, J. Biol. Chem. 275 (2000) 21508–21513. [71] H. Liu, R.E. Toman, S. Goparaju, M. Maceyka, V.E. Nava, H. Sankala, S.G. Payne, M. Bektas, I. Ishii, J. Chun, S. Milstien, S. Spiegel, J. Biol. Chem. 278 (2006) 40330–40336. [72] T.I. Gudz, K.Y. Tserng, C.L. Hoppel, J. Biol. Chem. 272 (1997) 24154–24158. [73] L.J. Siskind, R.N. Kolesnick, M. Colombini, J. Biol. Chem. 277 (2002) 26796–26803. [74] S.A. Novgorodov, Z.M. Szulc, C. Luberto, J.A. Jones, J. Bielawski, A. Bielawska, Y.A. Hannun, L.M. Obeid, J. Biol. Chem. 280 (2006) 16096–16105. [75] M. Di Paola, P. Zaccagnino, G. Montedoro, T. Cocco, M. Lorusso, J. Bioenerg. Biomembr. 36 (2004) 165–170. [76] P. Ghafourifar, S.D. Klein, O. Schucht, U. Schenk, M. Pruschy, S. Rocha, C. Richter, J. Biol. Chem. 274 (1999) 6080–6084. [77] P. Zhang, B. Liu, G.M. Jenkins, Y.A. Hannun, L.M. Obeid, J. Biol. Chem. 272 (1997) 9609–9612. [78] H. Birbes, S. El Bawab, Y.A. Hannun, L.M. Obeid, FASEB J. 15 (2001) 2669–2679.
2035
[79] A.D. Tepper, E. de Vries, W.J. van Blitterswijk, J. Borst, J. Clin. Invest. 103 (1999) 971–978. [80] Y.A. Hannun, C. Luberto, Trends Cell Biol. 10 (2000) 73–80. [81] M. Heinrich, M. Wickel, W. Schneider-Brachert, C. Sandberg, J. Gahr, R. Schwandner, T. Weber, P. Saftig, C. Peters, J. Brunner, M. Kronke, S. Schutze, EMBO J. 18 (1999) 5252–5263. [82] G.S. Wu, P. Saftig, C. Peters, W.S. El-Deiry, Oncogene 16 (1998) 2177–2183. [83] M. Heinrich, J. Neumeyer, M. Jakob, C. Hallas, V. Tchikov, S. WinotoMorbach, M. Wickel, W. Schneider-Brachert, A. Trauzold, A. Hethke, S. Schutze, Cell Death Differ. 11 (2004) 550–563. [84] J.A. Rotolo, J. Zhang, M. Donepudi, H. Lee, Z. Fuks, R. Kolesnick, J. Biol. Chem. 280 (2005) 26425–26434. [85] S. Daido, T. Kanzawa, A. Yamamoto, H. Takeuchi, Y. Kondo, S. Kondo, Cancer Res. 64 (2004) 4286–4293. [86] F. Scarlatti, C. Bauvy, A. Ventruti, G. Sala, F. Cluzeaud, A. Vandewalle, R. Ghidoni, P. Codogno, J. Biol. Chem. 279 (2004) 18384–18391. [87] O. Cuvillier, Biochim. Biophys. Acta 1585 (2002) 153–162. [88] E.A. Sweeney, J. Inokuchi, Y. Igarashi, FEBS Lett. 425 (1998) 61–65. [89] O. Cuvillier, L. Edsall, S. Spiegel, J. Biol. Chem. 275 (2000) 15691–15700. [90] O. Cuvillier, V.E. Nava, S.K. Murthy, L.C. Edsall, T. Levade, S. Milstien, S. Spiegel, Cell Death Differ. 8 (2001) 162–171. [91] W.J. Lee, H.S. Yoo, P.G. Suh, S. Oh, J.S. Lim, Y.M. Lee, Exp. Mol. Med. 36 (2004) 420–427. [92] W.D. Jarvis, F.A. Fornari Jr., K.L. Auer, A.J. Freemerman, E. Szabo, M.J. Birrer, C.R. Johnson, S.E. Barbour, P. Dent, S. Grant, Mol. Pharmacol. 52 (1997) 935–947. [93] H.C. Chang, L.H. Tsai, L.Y. Chuang, W.C. Hung, J. Cell Physiol. 188 (2001) 188–193. [94] W.C. Hung, H.C. Chang, L.Y. Chuang, Biochem. J. 338 (Pt 1) (1999) 161–166. [95] T. Shirahama, C. Sakakura, E.A. Sweeney, M. Ozawa, M. Takemoto, K. Nishiyama, Y. Ohi, Y. Igarashi, FEBS Lett. 407 (1997) 97–100. [96] C. Sakakura, E.A. Sweeney, T. Shirahama, S. Hakomori, Y. Igarashi, FEBS Lett. 379 (1996) 177–180. [97] C. Isogai, T. Murate, K. Tamiya-Koizumi, S. Yoshida, T. Ito, H. Nagai, T. Kinoshita, Y. Kagami, T. Hotta, M. Hamaguchi, H. Saito, Exp. Hematol. 26 (1998) 1118–1125. [98] Y.A. Hannun, R.M. Bell, Clin. Chim. Acta 185 (1989) 333–345. [99] Y.A. Hannun, R.M. Bell, Science 235 (1987) 670–674. [100] W.D. Jarvis, F.A. Fornari Jr., R.M. Tombes, R.K. Erukulla, R. Bittman, G.K. Schwartz, P. Dent, S. Grant, Mol. Pharmacol. 54 (1998) 844–856. [101] A. Hamaguchi, E. Suzuki, K. Murayama, T. Fujimura, T. Hikita, K. Iwabuchi, K. Handa, D.A. Withers, S.C. Masters, H. Fu, S. Hakomori, J. Biol. Chem. 278 (2003) 41557–41565. [102] T. Megidish, J. Cooper, L. Zhang, H. Fu, S. Hakomori, J. Biol. Chem. 273 (1998) 21834–21845. [103] F. Tsuruta, J. Sunayama, Y. Mori, S. Hattori, S. Shimizu, Y. Tsujimoto, K. Yoshioka, N. Masuyama, Y. Gotoh, EMBO J. 23 (2004) 1889–1899. [104] M. Nomura, S. Shimizu, T. Sugiyama, M. Narita, T. Ito, H. Matsuda, Y. Tsujimoto, J. Biol. Chem. 278 (2003) 2058–2065. [105] S.C. Masters, H. Yang, S.R. Datta, M.E. Greenberg, H. Fu, Mol. Pharmacol. 60 (2001) 1325–1331. [106] H. Yang, S.C. Masters, H. Wang, H. Fu, Biochim. Biophys. Acta 1547 (2001) 313–319. [107] C.W. Chiang, G. Harris, C. Ellig, S.C. Masters, R. Subramanian, S. Shenolikar, B.E. Wadzinski, E. Yang, Blood 97 (2001) 1289–1297. [108] H.G. Wang, N. Pathan, I.M. Ethell, S. Krajewski, Y. Yamaguchi, F. Shibasaki, F. McKeon, T. Bobo, T.F. Franke, J.C. Reed, Science 284 (1999) 339–343. [109] J. Zha, H. Harada, E. Yang, J. Jockel, S.J. Korsmeyer, Cell 87 (1996) 619–628. [110] M. Pushkareva, W.A. Khan, A.V. Alessenko, N. Sahyoun, Y.A. Hannun, J. Biol. Chem. 267 (1992) 15246–15251. [111] W. Hu, R. Xu, G. Zhang, J. Jin, Z.M. Szulc, J. Bielawski, Y.A. Hannun, L.M. Obeid, C. Mao, Mol. Biol. Cell 16 (2005) 1555–1567. [112] A.G. Rosenwald, R.E. Pagano, J. Biol. Chem. 268 (1993) 4577–4579. [113] M.T. Park, M.J. Kim, Y.H. Kang, S.Y. Choi, J.H. Lee, J.A. Choi, C.M.
2036
[114]
[115] [116] [117] [118] [119] [120] [121] [122]
T.A. Taha et al. / Biochimica et Biophysica Acta 1758 (2006) 2027–2036 Kang, C.K. Cho, S. Kang, S. Bae, Y.S. Lee, H.Y. Chung, S.J. Lee, Blood 105 (2005) 1724–1733. M.T. Park, J.A. Choi, M.J. Kim, H.D. Um, S. Bae, C.M. Kang, C.K. Cho, S. Kang, H.Y. Chung, Y.S. Lee, S.J. Lee, J. Biol. Chem. 278 (2003) 50624–50634. S. Spiegel, S. Milstien, Nat. Rev., Mol. Cell Biol. 4 (2003) 397–407. A. Olivera, S. Spiegel, Nature 365 (1993) 557–560. T. Kohama, A. Olivera, L. Edsall, M.M. Nagiec, R. Dickson, S. Spiegel, J. Biol. Chem. 273 (1998) 23722–23728. A. Olivera, T. Kohama, L. Edsall, V. Nava, O. Cuvillier, S. Poulton, S. Spiegel, J. Cell Biol. 147 (1999) 545–558. O. Cuvillier, G. Pirianov, B. Kleuser, P.G. Vanek, O.A. Coso, S. Gutkind, S. Spiegel, Nature 381 (1996) 800–803. T.A. Taha, K. Kitatani, M. El-Alwani, J. Bielawski, Y.A. Hannun, L.M. Obeid, FASEB J. 20 (2006) 482–484. P. Xia, L. Wang, J.R. Gamble, M.A. Vadas, J. Biol. Chem. 274 (1999) 34499–34505. Y. Osawa, Y. Banno, M. Nagaki, D.A. Brenner, T. Naiki, Y. Nozawa, S. Nakashima, H. Moriwaki, J. Immunol. 167 (2001) 173–180.
[123] Y.G. Kwon, J.K. Min, K.M. Kim, D.J. Lee, T.R. Billiar, Y.M. Kim, J. Biol. Chem. 276 (2001) 10627–10633. [124] L.C. Edsall, O. Cuvillier, S. Twitty, S. Spiegel, S. Milstien, J. Neurochem. 76 (2001) 1573–1584. [125] O. Cuvillier, T. Levade, Blood 98 (2001) 2828–2836. [126] H. Nakamura, T. Oda, K. Hamada, T. Hirano, N. Shimizu, H. Utiyama, J. Biol. Chem. 273 (1998) 15345–15351. [127] G. van Echten-Deckert, A. Zschoche, T. Bar, R.R. Schmidt, A. Raths, T. Heinemann, K. Sandhoff, J. Biol. Chem. 272 (1997) 15825–15833. [128] H. Le Stunff, I. Galve-Roperh, C. Peterson, S. Milstien, S. Spiegel, J. Cell Biol. 158 (2002) 1039–1049. [129] T.A. Taha, W. Osta, L. Kozhaya, J. Bielawski, K.R. Johnson, W.E. Gillanders, G.S. Dbaibo, Y.A. Hannun, L.M. Obeid, J. Biol. Chem. 279 (2004) 20546–20554. [130] T.A. Taha, K. Kitatani, J. Bielawski, W. Cho, Y.A. Hannun, L.M. Obeid, J. Biol. Chem. 280 (2005) 17196–171202. [131] N. Igarashi, T. Okada, S. Hayashi, T. Fujita, S. Jahangeer, S. Nakamura, J. Biol. Chem. 278 (2003) 46832–46839.