Cell. Mol. Life Sci. (2015) 72:959–969 DOI 10.1007/s00018-014-1778-x
Cellular and Molecular Life Sciences
REVIEW
Sphingolipid metabolism and its role in the skeletal tissues Zohreh Khavandgar • Monzur Murshed
Received: 3 June 2014 / Revised: 28 October 2014 / Accepted: 10 November 2014 / Published online: 26 November 2014 Ó Springer Basel 2014
Abstract The regulators affecting skeletal tissue formation and its maintenance include a wide array of molecules with very diverse functions. More recently, sphingolipids have been added to this growing list of regulatory molecules in the skeletal tissues. Sphingolipids are integral parts of various lipid membranes present in the cells and organelles. For a long time, these macromolecules were considered as inert structural elements. This view, however, has radically changed in recent years as sphingolipids are now recognized as important second messengers for signal-transduction pathways that affect cell growth, differentiation, stress responses and programmed death. In the current review, we discuss the available data showing the roles of various sphingolipids in three different skeletal cell types—chondrocytes in cartilage and osteoblasts and osteoclasts in bone. We provide an overview of the biology of sphingomyelin phosphodiesterase 3 (SMPD3), an important regulator of sphingolipid metabolism in the skeleton. SMPD3 is localized in the plasma membrane and has been shown to cleave sphingomyelin to generate ceramide, a bioactive lipid second messenger, and phosphocholine, an essential nutrient. SMPD3 deficiency in mice impairs the mineralization in both cartilage and
Z. Khavandgar M. Murshed (&) Faculty of Dentistry, McGill University, Montreal, Quebec, Canada e-mail: [email protected] M. Murshed Department of Medicine, McGill University, Montreal, Quebec, Canada M. Murshed Shriners Hospital for Children, McGill University, Montreal, Quebec, Canada
bone extracellular matrices leading to severe skeletal deformities. A detailed understanding of SMPD3 function may provide a novel insight on the role of sphingolipids in the skeletal tissues. Keywords Sphingolipids SMPD3 Neutral sphingomyelinase Ceramide Phosphocholine Sphingosine 1 phosphate Matrix vesicles Skeletal development Bone mineralization fro/fro
Background Skeletal tissues not only provide structural and biomechanical supports to the vertebrate body, but also maintain the homeostasis of two essential chemical species, inorganic phosphate and calcium ions, which are required for many critical physiological functions. In addition to its role as a mineral reservoir, more recently the skeleton has been recognized as an endocrine organ that secretes hormone(s) to regulate energy metabolism [1, 2]. These myriad roles of the skeleton make it one of the most dynamic tissues in the body that achieves its complex structure and unique cell-extracellular matrix composition through a multistep process during development. Development of the skeleton in vertebrates involves concerted functions of three major cell types—chondrocytes in cartilage, and osteoclasts and osteoblasts in bone. The spatiotemporal distribution of the precursor stem cells that give rise to these cell types, their proliferation, differentiation and programmed death in the developing skeleton determines the growth, shape and load-bearing capacity of the future skeleton [3–5]. Developmentally, bones can be categorized into two types: endochondral and intramembranous bones. A cartilage ‘precast’ (anlagen) is
123
960
essential for endochondral bone formation, which is not needed for the formation of intramembranous bones. Formation of both bone types however, initiates with the condensation of mesenchymal stem cells (MSC) that first proliferate and differentiate directly into osteoblasts in the case of intramembranous bones or chondrocytes in the case of endochondral bones. Chondrocytes within the core of the developing endochondral bones eventually differentiate to form two distinct growth plates at both ends. Each of these growth plates has four distinct zones: (1) a resting zone of chondrocyte precursors; (2) a proliferative zone with chondrocytes that synthesize extracellular matrix (ECM) proteins including type II collagen; and (3) prehypertrophic and (4) hypertrophic zones. The hypertrophic zone carries terminally differentiated chondrocytes that go through programmed death after secreting a cartilaginous matrix rich in type X collagen. This matrix becomes mineralized and vascularized, and eventually replaced by a type I collagen-rich bone matrix through the concerted resorptive and formative activities of osteoclasts and osteoblasts, respectively [3–5]. With the recent advancements of molecular genetics and the availability of a plethora of genetically modified mouse models, many of the genetic regulators of skeletal development have been identified. These regulators include a wide array of proteins with very diverse functions including signaling molecules/hormones (e.g. IHH, BMPs, FGFs, Leptin, PTH and PTHrP) and their receptors, membrane proteins (e.g. PHEX), transcription factors (e.g. RUNX2, OSX and ATF4), intracellular and extracellular enzymes (e.g. PHOSPHO1 and ALPL) and matrix proteins (e.g. Type I collagen, osteocalcin) [1, 6–21]. More recently, sphingolipids have been added to this growing list of molecules regulating skeletal development and tissue homeostasis [22–25]. Sphingolipids are an integral part of various lipid membranes present in the cells and organelles and serve as important second messengers for signaltransduction pathways affecting cell growth, differentiation, stress responses and death [26].
Sphingolipid biosynthesis Sphingolipids carry long-chain aliphatic-amine backbones containing two or three hydroxyl groups which are known as sphingoid bases [27–29]. The most abundant sphingoid base in animal tissues is sphingosine. Additionally, tissues contain a saturated analog of sphingosine known as dihydrosphingosine or sphinganine. Enzymatic modifications of these simple sphingoid bases generate a wide array of sphingolipids with diverse structural and functional properties (Fig. 1). The most common functional sphingolipids involve more than hundred species of ceramide,
123
Z. Khavandgar, M. Murshed
Fig. 1 Chemical structures of sphingosine and its derivatives ceramide and sphingomyelin. A ceramide molecule is composed of sphingosine and a fatty acid (green) linked by an amide bond. An enzymatic addition of a phosphocholine head group (blue) to a ceramide molecule leads to the synthesis of sphingomyelin
phosphorylated ceramides, sphingoid bases and sphingoid base phosphates among others [27–29]. The current review will discuss the role of some of these sphingolipids in various skeletal cell types. De novo synthesis of sphingolipids Sphingolipids are primarily synthesized in the endoplasmic reticulum (ER), but are transported to the plasma membrane and endosomes, where they perform many of their functions [12, 26, 30]. In the ER, sphingolipids can be synthesized de novo using simple precursors, such as serine and palmitoyl CoA. In a condensation reaction catalyzed by a ubiquitous enzyme called serine palmitoyltransferase these two substrates first generate 3-ketosphinganine, which is converted to sphinganine by ketosphinganine reductase [12, 26, 30]. A dihydroxyceramide synthase then acetylates sphinganine to generate dihydroxyceramide. Finally, a desaturase-catalyzed reaction converts an inactive dihydroxyceramide to ceramide, an important lipid second messenger that regulates a variety of cellular activities (Fig. 2). Ceramide serves as a substrate for one of three enzymes that produce sphingolipids, such as glucosyl ceramide, galactosyl ceramide and sphingomyelin. Ceramide can also be phosphorylated by a family of ceramide kinases. The hydrolysis of ceramide by ceramidase generates sphingosine, which can be phosphorylated by two sphingosine kinases to generate sphingosine-1-phosphate (S1P)—a bioactive lipid with signaling properties. S1P can also be dephosphorylated by a class of lipid-phosphate phosphatases (LPP1-3). These transmembrane enzymes with extracellular catalytic sites play an important regulatory role in S1P-medated signaling events.
Sphingolipid and skeleton
961
Fig. 2 De novo and sphingomyelinase pathways showing the metabolism of ceramide, sphingomyelin and sphingosine
Synthesis of sphingolipids by sphingomyelinases
Role of sphingolipids in skeletal cells: in vitro findings
In addition to the de novo pathway described above, various sphingomyelinases (SMase) play a critical role in sphingolipid biosynthesis (Fig. 2). Sphingomyelinases belong to a class of enzymes, which cleave sphingomyelin to generate ceramide and phosphocholine. Depending on their pH optima, these enzymes have been classified into three categories—acidic, alkaline and neutral sphingomyelinases [25, 31]. Acidic sphingomyelinases are the most extensively studied enzymes among the three subclasses. Mutations in acidic sphingomyelin phosphodiesterase 1 gene (Smpd1) cause Niemann-Pick disease, which is characterized by abnormal lipid accumulation in the spleen, liver, lung, bone marrow and brain [25]. Recently, this enzyme has been implicated in ceramide-mediated signaling events [32]. Alkaline sphingomyelinase hydrolyzes dietary sphingomyelin and generates sphingolipid metabolites in the gut [33]. Impaired spingomyelin metabolism in the intestine might have implications in colon cancer development as dietary supplementation with sphingomyelin and ceramide analogs was found to inhibit the development of chemically-induced colon cancer in animal models [34]. Neutral sphingomyelinases are the key enzymes involved in ceramide-mediated signaling events. So far, three sphingomyelin phosphodiesterases (SMPD2, 3 and 4) have been identified that belong to this subclass [25]. Although all these isoforms are present as membrane-bound forms, they differ in their tissue distribution. SMPD2 and 4 are ubiquitously expressed, while SMPD3 expression is largely restricted to the bone, brain and cartilage [24].
The functions of two major sphingolipids, ceramide and S1P, in skeletal cells have been discussed below. Ceramide Obeid et al. [35] used synthetic C2-ceramide first to demonstrate the role of sphingolipids in apoptosis. This study showed that this ceramide analog promotes DNA fragmentation and apoptosis in U937 monoblast leukemia cells via upregulation of Tumor Necrosis Factor-alpha (TNF-alpha). Since then numerous studies have reported the pro-apoptotic role of ceramide in many different cell types. These studies described multiple signaling intermediates/pathways including ceramide-activated protein kinase and the activation of mitogen-activated protein kinase to induce cellular apoptosis. For example, ceramide-induced apoptosis may involve the activation protein kinase C-zeta to promote c-Jun N-terminal kinase (JNK) activation and the nuclear translocation of transcription factor nuclear factor-kappa B (NF-kappa B) [35]. It has been suggested that the membrane contact sites (MCS) between ER and mitochondria facilitate ceramideinduced apoptosis. Ceramide produced by SMPD3 in the ER membrane may translocate to the outer mitochondrial membrane to form ceramide channels, which may in turn release pro-apoptotic molecule cytochrome C [36]. A recent study proposed that ceramide serves as a precursor for S1P and hexadecenal which in turn regulates apoptosis. These molecules activate apoptotic regulators Bak and Bax, promoting the release of cytochrome C from the
123
962
mitochondria, which then activates caspases, enzymes essential for apoptotic cell death [37]. So far, only a handful of studies have been performed to examine the effects of ceramide on the apoptosis of bone cells in vitro. Treatment of MC3T3-E1 preosteoblasts with the nitric oxide (NO) donor sodium nitroprusside causes apoptosis [38]. This study reports an increase of long-chain intracellular ceramides, C22 and C24, upon this treatment. Interestingly, NO has been recently shown to promote the release of cytochrome C from the mitochondria [39]. It is therefore likely that NO increases the release of long-chain ceramides, which in turn results in the release of cytochrome C from the mitochondria turning on the intrinsic apoptotic pathways in cultured osteoblasts. In another cell-culture study, murine MC3T3-E1 preosteoblasts were treated with TNF-alpha to induce apoptosis. Within 3 min of TNF-alpha treatment, detectable increase of endogenous ceramide levels was apparent, which reached to its peak by 30 min. TNF-alpha treatment of MC3T3-E1 cells also promoted the nuclear translocation and activation of NF-kappa B. Interestingly, treatment of the preosteoblasts with a ceramide analog alone was sufficient to affect the nuclear translocation/activation of NFkappa B in a similar fashion. This observation lead to the conclusion that in osteoblasts, TNF-alpha modulates NFkappa B localization and function through the upregulation of ceramide synthesis [40]. A separate study showed that the treatment of MC3T3-E1 cells with dexamethasone, a known immunosuppressant, reduces TNF-alpha-induced ceramide production and dampens not only the activation of NF-kappa B, but also the activities of two pro-apoptotic enzymes, JNK and caspase-3-like extracellular protease. Additionally, there was a redistribution of cytochrome C in these cells [41]. This later observation further confirms that mitochondrial pathways might be relevant in the apoptosis of bone cells. Collectively, these findings indicate that inflammation-induced ceramide production in osteoblasts, may result in apoptosis through multiple mechanisms. A more recent study reported that although high-dose treatment (2 9 10-6 M) of mouse primary osteoblasts with C2-ceramide-induced apoptosis, the low-dose treatment (10-7 M) actually resulted in increased cell survival through the protein kinase C activity. It is possible that there might be separate pathways that promote ceramidemediated pro-apoptotic and pro-survival signaling events [42]. As is the case with the cultured osteoblasts, a high dose (3 9 10-5 to 10-4 M) of C2-ceramide treatment induced apoptosis of chondrocytes in rabbit articular cartilage explants. Also, C2-ceramide treatment increased matrix metalloproteinase activity in these explants. These data provide evidence that chronic inflammatory responses that increase matrix remodeling in osteoarthritis might be
123
Z. Khavandgar, M. Murshed
caused by an upregulation of ceramide synthesis [43]. C2ceramide induced apoptosis in murine chondrogenic ATDC5 cells without affecting the expression of chondrogenic markers Sox9, Col2a1 and Col10a1 [44]. Although C2-ceramide treatments lead to increased apoptosis in both osteoblasts and chondrocytes, there was no effect of this cell-permeable synthetic ceramide on the apoptosis of rabbit mature osteoclasts [45]. However, treatment with ceramide inhibited F-actin ring formation within the sealing zone of osteoclasts and impaired their resorption capacities. A similar effect was observed when these cells were treated with exogenous sphingomyelinase [45]. Interestingly, TNF-alpha treatment that promotes apoptosis in osteoblasts by upregulating ceramide synthesis, actually favors the survival of cultured murine osteoclasts [46]. As shown by this study, TNF-alpha treatment activates prosurvival pathways in osteoclasts by engaging phosphatidylinositol 3-kinase, Akt and MEK/ERK signaling. Several studies suggest that ceramide and its derivatives play an important role in osteoclastogenesis. Lactosylceramide, a derivative of ceramide, has been shown to induce the expression of the receptor-activator of NF-kappa B (RANK) in bone marrow-derived osteoclasts [47]. Moreover, this study showed that the inhibition of glucosylceramide synthase by D-threo-1-phenyl-2-decanoylamino-3-morpholino-1-propanol (D-PDM) prevented granulocyte–macrophage colony-stimulating factor and RANK ligand-mediated osteoclastogenesis. In a subsequent study, it was shown that the D-PDM-mediated inhibition of osteoclastogenesis could be rescued by treating the osteoclasts with lactosylceramide [48]. Sphingosine and its derivatives Sphingosine and its derivatives have been implicated in a variety of cell functions in osteoblasts. In MC3T3-E1 preosteoblast cultures, S1P induces the synthesis of heatshock protein 27 (HSP27), a molecular chaperon that acts as a stress-response protein; HSP27 in turn negatively regulates the synthesis of osteocalcin, a marker expressed by functional osteoblasts [49, 50]. Interestingly, treatment of myogenic C2C12 cells with bone morphogenetic protein-2 (BMP-2) in combination with S1P or FTY720 (an agonist of S1P receptor) significantly increased the expression of osteocalcin and several other osteogenic markers [51]. Although the latter study did not report osteocalcin protein levels, this apparent discrepancy in osteocalcin expression might be caused by the differences in the differentiation stages and cell types used for these two studies. Exposure of MC3T3-E1 cells to sphingosine, S1P and sphingosylphosphorylcholine promoted intracellular calcium release increasing cytosolic calcium levels [52, 53]. Also, S1P treatments of rat calvarial osteoblasts
Sphingolipid and skeleton
963
Fig. 3 The major effects of ceramide (a) and Sphingosine 1 phosphate (S1P) (b) on osteoblasts, chondrocytes and osteoclasts. The filled up arrows indicate activation and the filled down arrows indicate inhibition
and human SaOS-2 osteosarcoma cells have been shown to prevent apoptosis of these cells that was initiated by serum deprivation [54]. S1P signaling has been implicated in osteoblast migration; S1P acts as a chemorepellent for the undifferentiated preosteoblasts [55]. As opposed to S1P function, it was shown that platelet-derived growth factor (PDGF) that plays a role in bone remodeling and fracture healing, acts as a chemoattractant for preosteoblasts. Upon treatment with BMP-2, these cells differentiate to mature osteoblasts retaining their abilities to respond to PDGF but not to S1P. Roelofsen et al. [56] suggested that migration of osteoblasts is controlled by the balance between PDGF and S1P allowing only differentiated osteoblasts to travel to the site of bone formation. Although interesting, this notion has been challenged by a more recent study identifying S1P as a chemoattractant for MSCs that give rise to cells of the osteoblast lineage. According to this study, bone-resorbing osteoclasts release S1P to induce the migration of MSCs and thereby promote the recruitment of the osteoblast precursors at the sites undergoing bone remodeling. It is, however, possible that osteoblasts at different stages of differentiation may respond differently to S1P. S1P has been shown to promote the proliferation of rat primary chondrocytes [57]. S1P treatment induced phospholipase C-mediated cytosolic calcium release in these cells. Also, extracellular signal-regulated kinase (ERK) and p38 mitogen-activated kinases were induced. When treated with ERK inhibitor PD98059 the cell proliferation was almost completely blocked, while p38 kinase inhibitor SB203580 did not have any effect. These data suggest a cardinal role for ERK in S1P-mediated chondrocyte proliferation. In articular chondrocytes, S1P signaling has
been implicated in cartilage matrix degradation, upregulation of vascular endothelial growth factor and the progression of arthritis [58, 59]. S1P signaling has also been shown to regulate the migration of osteoclast precursors. Two S1P receptors, S1PR1 and S1PR2, play a critical role in this process. S1PR1 promotes the chemotaxis towards S1P in bone marrow where S1P level is normally low, while S1PR2 mediates the S1P chemorepellant properties in circulation where S1P level is high. This mechanism may facilitate the infiltration of osteoclast precursors from the circulation to the bone marrow [60]. The major effects of ceramide and S1P on the three major cell types in the skeleton have been depicted in Fig. 3.
Role of sphingolipids in skeletal tissues: in vivo findings The role of osteoclast-derived S1P on bone remodeling has been elaborated in an in vivo study by Lotinun et al.[61]. In this study, osteoclast-specific inactivation of cathepsin K, an extracellular matrix degrading enzyme, resulted in an increase of S1P secretion by these cells. Secreted S1P promoted increased bone formation by osteoblasts primarily by enhancing the differentiation of osteoblast precursors to functional osteoblasts. As described by Quint et al. [56], osteoclasts through S1P release may also promote the recruitment of MSCs that give rise to osteoblast precursors and thereby contribute to the observed high bone mass phenotype. A recently published study showed that calcitonin, a thyroid-derived hormone, affects bone formation by regulating the release of S1P by osteoclasts [62].
123
964
Emerging data from the analysis of genetically altered mouse models suggest that sphingolipid metabolism during embryonic development plays a critical role in normal skeletogenesis. This insight originally came from two different mouse models lacking a neutral sphingomyelinase, nSMase2, also known as sphingomyelin phosphodiesterase 3 (SMPD3). Currently, there are two SMPD3-deficient mouse models available; one carries a chemically induced deletion of 1,758 base pairs encompassing part of intron 8 and most of exon 9 (fro) in the Smpd3 locus (fro/fro model) and the other was generated by gene targeting (Smpd3-/model) [22, 63, 64]. So far no skeletal abnormalities in mice lacking SMPD1 and SMPD2 activity have been reported. The fro mutation does not affect the membrane localization of SMPD3 but completely abolishes its enzymatic activity. There was a significant reduction of neutral sphingomyelinase activity in the brain and skeletal tissues of fro/fro mice where Smpd3 is highly expressed [24]. These mice show severe skeletal dysplasia hall marked by abnormal cartilage development and bone mineralization defects. Histological sections of the humeri of late-gestation embryos suggested delayed apoptosis of hypertrophic chondrocytes and poor mineralization of the cortical bones. Both of these anomalies appear to contribute to the skeletal dysplasia seen in the mutant mice [24]. The skeletal deformities seen in the fro/fro mice become less apparent with age although the bone mineralization defects have been detected in mice that are more than 6 month old (unpublished data). fro/fro mice also show delay in mantle dentin mineralization and subsequent delay in enamel formation during early odontogenesis. These abnormalities were largely corrected as these mice grew older [23]. Both fro/fro and the gene targeted Smpd3-/- mice share similar gross skeletal abnormalities, i.e. short-limb dwarfism, deformation of long bones and abnormally formed rib cages. However, Stoffel et al. [65] did not observe the bone and tooth mineralization abnormalities documented in fro/fro mice in their gene targeted model. The phenotypic variations in fro/fro and Smpd3-/- mice cannot be attributable to the presence of any additional genetic alterations in the former strain as such mutation(s), if not closely associated, would have been fully segregated during the propagation of fro/fro mice in multiple laboratories over the past four decades. This inference, together with the observation that the fully penetrant fro/fro phenotype including the severe hypomineralization defect, is always associated with the Smpd3 deletion mutation identified by Aubin et al, strongly suggests that the loss of SMPD3 function is indeed the cause of the fro/fro phenotype [22]. Indeed, two recently published studies describing the complete rescue of the bone and tooth mineralization abnormalities in fro/fro; Col1a1-Smpd3 mice expressing
123
Z. Khavandgar, M. Murshed
Smpd3 in osteoblasts and odontoblasts ruled out the possibility of the involvement of any additional mutation(s) in the fro/fro phenotype [23, 24]. These data provide unambiguous demonstration of a direct role for SMPD3 in osteoblast/bone mineralization. Differences in analytical methods used to characterize the Smpd3-/- and fro/fro mice might lead to different interpretations about the apparent phenotype in the same tissue. For example, Khavandgar et al. analyzed the hard tissues in fro/fro mice using histomorphometric techniques commonly used to detect mineralization defects in bone samples. Stoffel et al. however, analyzed the mineralization status of the Smpd3-/- mice solely by bone mineral density analysis, which determines the total bone mineral content and is not suitable for detecting an increase in unmineralized bone matrix.
SMPD3, a key regulator of sphingolipid functions in bone: possible mode of action Although the role of SMPD3 in the developing skeleton— both in bone and cartilage, is now well established, it’s mode of action is still poorly understood. Based on the data obtained from the animal models, it appears that by far SMPD3 is the only lipid metabolizing enzyme that affects bone mineralization and chondrocyte hypertrophy in vivo. SMPD3, an integral membrane protein that is attached to the inner leaflet of the cell membrane, cleaves sphingomyelin to generate many different species of ceramide that regulate a myriad of cell function. The SMPD3-catalyzed reaction also liberates an essential nutrient, phosphocholine [66]. At this point it is not clear whether both ceramide and phosphocholine or only one of them is involved in skeletal tissue morphogenesis. However, the findings from mutant mouse models suggest that these two bioactive metabolites may have distinct tissue-specific roles. The first in vivo proof suggesting that SMPD3 might be involved in apoptosis of chondrocytes came from the observation of Khavandgar et al. [24]. A delayed hypertrophy and impaired apoptosis of terminally differentiated chondrocytes in these mice results in an abnormal accumulation of chondrocytes in the mid-shaft region of the developing bone (Fig. 4). The fro/fro mice show reduced ceramide levels in their skeletal tissues. Considering the extensive experimental evidence showing a role for ceramide in the regulation of cellular apoptosis, the chondrocyte phenotype in fro/fro mice can be attributed to impaired ceramide metabolism in this strain. This study however, did not fully rule out the possible effects of reduced phosphocholine levels caused by the loss of SMPD3 function on the death of hypertrophic chondrocytes. Indeed, in a recent study, Li et al. [67] reported that
Sphingolipid and skeleton
965
Fig. 4 Abnormal presence of cartilage matrix and chondrocytes asterisk in a fro/ fro developing humerus from an E15.5 embryo. The histological sections were stained with Safranin O
mice lacking choline kinase beta which showed reduced phosphocholine levels, there was an expansion of the hypertrophic zone. Nevertheless, based on the numerous in vitro and in vivo experiments on ceramide-mediated cell death, it is likely that the reduced ceramide level is responsible, at least in part, for the observed delay in apoptosis of hypertrophic chondrocytes in the developing skeleton of fro/fro mice. Interestingly, a recent study suggested that GW4869-mediated inhibition of neutral sphingomyelinase function accelerates the mineralization of chondrogenic ATDC5 cultures [68]. This conclusion was surprising as it contradicts the in vivo findings in SMPD3-deficient mice. Although the role of ceramide in apoptosis is supported by numerous studies, at present no experimental data directly links it to ECM mineralization. As mentioned above, various species of ceramides in the cells are mainly generated by two different mechanisms—the de novo and the sphingomyelinase pathways (Fig. 2). In the de novo pathway, dihydroceramide desaturase 1 (DES1) is the final enzyme that converts dihydroceramide to ceramide. Des1 null mice show multiple physiologic anomalies including hypophagia, weight loss and overall growth impairment and die prematurely [69]. Although these mice are smaller in size, unlike fro/fro or Smpd3 knockout mice they show a normal skeletal structure. They do not show any bone mineralization defects (unpublished data). This later finding has two implications: it is possible that ceramides generated from two different pathways contribute to two distinct pools with separate cellular functions; alternatively, it is possible that ceramides do not play any role in the process of bone mineralization. Interestingly, Des1 null mice like fro/fro mice show a reduced cellular/tissue ceramide levels and a similar metabolic phenotype (68 and unpublished data). This observation suggests that ceramides generated from the de novo and sphingomyelinase pathways act in concert to affect the same
biological functions. If this notion is correct then the absence of bone mineralization defects in Des1 knockout mice may imply that the maintenance of ceramide levels is not essential for normal bone mineralization. This may also point towards the involvement of phosphocholine, the other product of the SMPD3-catalyzed reaction, in the regulation of bone mineralization. The importance of phosphocholine in bone mineralization has been convincingly shown by animal experiments [17, 70]. Phosphocholine is generated from sphingomyelin by SMPD3, or by cytosolic choline kinases from choline, which can be cleaved by PHOSPHO1, an intracellular enzyme [71, 72]. PHOSPHO1 deficiency in mice causes a similar albeit milder bone mineralization defects as seen in fro/fro mice [17, 70]. At present it is not clear how SMPD3, a cell membranebound enzyme with an intracellular catalytic domain, may affect ECM mineralization. One possible mechanism may involve matrix vesicles (MVs) that are released from the matrix synthesizing cells in the mineralized tissues [73]. Although there is no general consensus, it is believed that the initial nucleation of minerals may first occur within these extracellular microstructures. Wu et al. [74] showed that extensive phospholipid degradation occurs in the mineralizing MVs with a concomitant increase of free fatty acids, indicative of the presence of phospholipase activity. This study provide the early evidence that linked phospholipid metabolism to the initiation of mineralization. Phospholipase activity generates phosphocholine and phosphoethanolamine, which can be cleaved by PHOSPHO1 releasing free phosphate inside the MVs and promoting mineral nucleation. SMPD3 may cleave sphingomyelin present in the MV membrane to generate additional phosphocholine to further facilitate this process. This later possibility is supported by the observation that both PHOSPHO1 and SMPD3 are present in MV preparations [75].
123
966
During embryonic development of mice, Smpd3 is highly expressed in the bone, cartilage and brain [24]. This expression pattern is maintained during the perinatal period. However, until now only a limited number of studies investigated the regulation of the Smpd3 promoter in the skeletal tissues/cell types. An in vitro study convincingly showed that treatment with BMP-2 upregulated Smpd3 expression in myogenic C2C12 cells. This study also reported the presence of RUNX2-binding elements in the murine Smpd3 promoter and an upregulation of Smpd3 expression upon transfection of C2C12 cells with a Runx2 expression vector. Later, using an ex vivo organ culture system and chondrogenic ATDC5 cells, Kakoi et al. [68] demonstrated a similar BMP-2-induced upregulation of Smpd3 expression during chondrogenesis. Although Smpd3 expression in bone is higher than any other tissue, until now no other osteoblast-specific transcription factor has been shown to regulate its expression [24]. Coleman et al. [76] have reported abnormal matrix deposition in SMPD3-deficient fro/fro mice. This group did not find any alteration of trabecular bone parameter in the three-month-old mutant mice in comparison to their littermate controls. The mineralization defects in bones and teeth and the growth plate abnormalities in fro/fro mice appear to be less severe with age. Taken together, these findings suggest that SMPD3 activity is required for normal skeletal development but becomes redundant in aged mice. In Smpd3 knockout mice, reduction of several brainderived hormones e.g. thyroid hormone, growth hormone and pituitary hormone has been reported [64]. This observation suggests an important neuroendocrine role for SMPD3 in the nervous system, which is consistent with its high expression in the brain. In addition to its local regulatory roles on skeletal ECM mineralization and apoptosis of hypertrophic chondrocytes, SMPD3 may have a novel systemic mode of action in the developing skeleton. The deficiency of growth hormone and the growth abnormalities such as dwarfism and smaller body size seen in both SMPD3-deficient models further support this notion. However, the precise mechanisms by which SMPD3 affects the circulating levels of various brain-derived hormones are still unknown.
Z. Khavandgar, M. Murshed
could be that most of the mutant mouse models that show altered sphingolipid metabolism do not show any overt skeletal phenotype during their adulthood. Nevertheless, a systematic analysis of these mice during embryonic development using histology and morphometric techniques may provide further clues on the function of sphingolipids in skeletal development. The chemical mutagenesis and targeted mutations in murine SMPD3 gene have given us a unique opportunity to explore the role of this lipid metabolizing enzyme in the skeletal tissues. Further studies involving mouse models with tissue-specific inactivation of Smpd3 are needed to decipher the local versus systemic contribution of this enzyme to skeletal development and growth. More specifically, these studies are expected to elucidate the role of this pleiotropic enzyme in the brain and how its activity might contribute to the hypothalamic regulation of skeletal growth. Additionally, novel mechanistic studies may shed light on the specific roles of the two products of the SMPD3-catalyzed reaction in the skeletal tissues. The de novo pathway of sphingolipid metabolism, more particularly ceramide metabolism, has been shown to regulate energy expenditure and a similar role is expected for the sphingomyelinase pathway of ceramide synthesis [69, 77–79]. An important aspect would be to investigate how various species of ceramides generated by SMPD3 in the osteoblasts and chondrocytes might contribute to overall energy metabolism in the body. This work is particularly relevant as osteocalcin, an osteoblast-specific protein has been recently shown to regulate energy metabolism [2].
Conclusion The fro/fro mouse model is one of the first animal model to demonstrate a role for sphingolipids in skeletal development and homeostasis. Although a cell-autonomous role of SMPD3 in bone and tooth mineralization is now well established, the precise mechanism of its action in hard tissue mineralization is still unknown. SMPD3 metabolites, phosphocholine and different species of ceramides, may have distinct roles in the cells of the developing skeleton. A thorough understanding of the mode of actions of these two metabolites will be required to fully appreciate the complex process that regulates vertebrate skeletogenesis.
Future directions Despite the published data from numerous cell culture studies showing the role of lipid metabolites in chondrocytes, osteoblasts and osteoclasts, the in vivo validation of these studies is still missing. At present, limited in vivo data are available on the roles of sphingolipid metabolizing enzymes in the skeletal tissues. One possible reason for this
123
References 1. Horton WA, Degnin CR (2009) FGFs in endochondral skeletal development. Trends Endocrinol Metab 20(7):341–348 2. Karsenty G (2011) Bone endocrine regulation of energy metabolism and male reproduction. C R Biol 334(10):720–724
Sphingolipid and skeleton 3. Karsenty G (2003) The complexities of skeletal biology. Nature 423(6937):316–318 4. Karsenty G, Kronenberg HM, Settembre C (2009) Genetic control of bone formation. Annu Rev Cell Dev Biol 25:629–648 5. Mackie EJ, Tatarczuch L, Mirams M (2011) The skeleton: a multi-functional complex organ: the growth plate chondrocyte and endochondral ossification. J Endocrinol 211(2):109–121 6. Amano K, Hata K, Sugita A, Takigawa Y, Ono K, Wakabayashi M, Kogo M, Nishimura R, Yoneda T (2009) Sox9 family members negatively regulate maturation and calcification of chondrocytes through up-regulation of parathyroid hormonerelated protein. Mol Biol Cell 20(21):4541–4551 7. Amizuka N, Henderson JE, Hoshi K, Warshawsky H, Ozawa H, Goltzman D, Karaplis AC (1996) Programmed cell death of chondrocytes and aberrant chondrogenesis in mice homozygous for parathyroid hormone-related peptide gene deletion. Endocrinology 137(11):5055–5067 8. Degnin CR, Laederich MB, Horton WA (2010) FGFs in endochondral skeletal development. J Cell Biochem 110(5):1046–1057 9. Ducy P, Zhang R, Geoffroy V, Ridall AL, Karsenty G (1997) Osf2/Cbfa1: a transcriptional activator of osteoblast differentiation. Cell 89(5):747–754 10. Komori T, Yagi H, Nomura S, Yamaguchi A, Sasaki K, Deguchi K, Shimizu Y, Bronson RT, Gao YH, Inada M, Sato M, Okamoto R, Kitamura Y, Yoshiki S, Kishimoto T (1997) Targeted disruption of Cbfa1 results in a complete lack of bone formation owing to maturational arrest of osteoblasts. Cell 89(5):755–764 11. Kozhemyakina E, Cohen T, Yao TP, Lassar AB (2009) Parathyroid hormone-related peptide represses chondrocyte hypertrophy through a protein phosphatase 2A/histone deacetylase 4/MEF2 pathway. Mol Cell Biol 29(21):5751–5762 12. Merrill AH Jr (2002) De novo sphingolipid biosynthesis: a necessary, but dangerous, pathway. J Biol Chem 277(29):25843–25846 13. Nakashima K, Zhou X, Kunkel G, Zhang Z, Deng JM, Behringer RR, de Crombrugghe B (2002) The novel zinc finger-containing transcription factor osterix is required for osteoblast differentiation and bone formation. Cell 108(1):17–29 14. Nesbitt T, Fujiwara I, Thomas R, Xiao ZS, Quarles LD, Drezner MK (1999) Coordinated maturational regulation of PHEX and renal phosphate transport inhibitory activity: evidence for the pathophysiological role of PHEX in X-linked hypophosphatemia. J Bone Miner Res 14(12):2027–2035 15. Ornitz DM, Marie PJ (2002) FGF signaling pathways in endochondral and intramembranous bone development and human genetic disease. Genes Dev 16(12):1446–1465 16. Retting KN, Song B, Yoon BS, Lyons KM (2009) BMP canonical Smad signaling through Smad1 and Smad5 is required for endochondral bone formation. Development 136(7):1093–1104 17. Roberts S, Narisawa S, Harmey D, Millan JL, Farquharson C (2007) Functional involvement of PHOSPHO1 in matrix vesiclemediated skeletal mineralization. J Bone Miner Res 22(4):617–627 18. Stewart AJ, Roberts SJ, Seawright E, Davey MG, Fleming RH, Farquharson C (2006) The presence of PHOSPHO1 in matrix vesicles and its developmental expression prior to skeletal mineralization. Bone 39(5):1000–1007 19. Stonich D, Su Y, Dad S, Reddy S, Mostofi Y, Russell D, Chung TDY, Hedrick NM, Rascon J, Garcia X, Sergienko E, Millan JL, Cosford N 2010 The role of PHOSPHO1 in the initiation of skeletal calcification 20. Whyte MP (1994) Hypophosphatasia and the role of alkaline phosphatase in skeletal mineralization. Endocr Rev 15(4):439–461
967 21. Zhou Z, Xie J, Lee D, Liu Y, Jung J, Zhou L, Xiong S, Mei L, Xiong WC (2010) Neogenin regulation of BMP-induced canonical Smad signaling and endochondral bone formation. Dev Cell 19(1):90–102 22. Aubin I, Adams CP, Opsahl S, Septier D, Bishop CE, Auge N, Salvayre R, Negre-Salvayre A, Goldberg M, Guenet JL, Poirier C (2005) A deletion in the gene encoding sphingomyelin phosphodiesterase 3 (Smpd3) results in osteogenesis and dentinogenesis imperfecta in the mouse. Nat Genet 37(8):803–805 23. Khavandgar Z, Alebrahim S, Eimar H, Tamimi F, McKee MD, Murshed M (2013) Local regulation of tooth mineralization by sphingomyelin phosphodiesterase 3. J Dent Res 92(4):358–364 24. Khavandgar Z, Poirier C, Clarke CJ, Li J, Wang N, McKee MD, Hannun YA, Murshed M (2011) A cell-autonomous requirement for neutral sphingomyelinase 2 in bone mineralization. J Cell Biol 194(2):277–289 25. Stoffel W (1999) Functional analysis of acid and neutral sphingomyelinases in vitro and in vivo. Chem Phys Lipids 102(1–2):107–121 26. Merrill AH Jr, Schmelz EM, Dillehay DL, Spiegel S, Shayman JA, Schroeder JJ, Riley RT, Voss KA, Wang E (1997) Sphingolipids–the enigmatic lipid class: biochemistry, physiology, and pathophysiology. Toxicol Appl Pharmacol 142(1):208–225 27. Airola MV, Hannun YA (2013) Sphingolipid metabolism and neutral sphingomyelinases. Handb Exp Pharmacol 215:57–76 28. Mullen TD, Hannun YA, Obeid LM (2012) Ceramide synthases at the centre of sphingolipid metabolism and biology. Biochem J 441(3):789–802 29. Hannun YA, Obeid LM (2011) Many ceramides. J Biol Chem 286(32):27855–27862 30. Futerman AH, Riezman H (2005) The ins and outs of sphingolipid synthesis. Trends Cell Biol 15(6):312–318 31. Nilsson A, Duan RD (1999) Alkaline sphingomyelinases and ceramidases of the gastrointestinal tract. Chem Phys Lipids 102(1–2):97–105 32. Kirschnek S, Paris F, Weller M, Grassme H, Ferlinz K, Riehle A, Fuks Z, Kolesnick R, Gulbins E (2000) CD95-mediated apoptosis in vivo involves acid sphingomyelinase. J Biol Chem 275(35):27316–27323 33. Duan RD, Nyberg L, Nilsson A (1995) Alkaline sphingomyelinase activity in rat gastrointestinal tract: distribution and characteristics. Biochim Biophys Acta 1259(1):49–55 34. Duan RD (2006) Alkaline sphingomyelinase: an old enzyme with novel implications. Biochim Biophys Acta 1761(3):281–291 35. Herr I, Debatin KM (2001) Cellular stress response and apoptosis in cancer therapy. Blood 98(9):2603–2614 36. Siskind LJ, Kolesnick RN, Colombini M (2006) Ceramide forms channels in mitochondrial outer membranes at physiologically relevant concentrations. Mitochondrion 6(3):118–125 37. Chipuk JE, McStay GP, Bharti A, Kuwana T, Clarke CJ, Siskind LJ, Obeid LM, Green DR (2012) Sphingolipid metabolism cooperates with BAK and BAX to promote the mitochondrial pathway of apoptosis. Cell 148(5):988–1000 38. Olivier S, Fillet M, Malaise M, Piette J, Bours V, Merville MP, Franchimont N (2005) Sodium nitroprusside-induced osteoblast apoptosis is mediated by long chain ceramide and is decreased by raloxifene. Biochem Pharmacol 69(6):891–901 39. Snyder CM, Shroff EH, Liu J, Chandel NS (2009) Nitric oxide induces cell death by regulating anti-apoptotic BCL-2 family members. PLoS One 4(9):e7059 40. Kitajima I, Soejima Y, Takasaki I, Beppu H, Tokioka T, Maruyama I (1996) Ceramide-induced nuclear translocation of NFkappa B is a potential mediator of the apoptotic response to TNFalpha in murine clonal osteoblasts. Bone 19(3):263–270
123
968 41. Chae HJ, Chae SW, Kang JS, Bang BG, Cho SB, Park RK, So HS, Kim YK, Kim HM, Kim HR (2000) Dexamethasone suppresses tumor necrosis factor-alpha-induced apoptosis in osteoblasts: possible role for ceramide. Endocrinology 141(8):2904–2913 42. Hill PA, Tumber A (2010) Ceramide-induced cell death/survival in murine osteoblasts. J Endocrinol 206(2):225–233 43. Sabatini M, Rolland G, Leonce S, Thomas M, Lesur C, Perez V, de Nanteuil G, Bonnet J (2000) Effects of ceramide on apoptosis, proteoglycan degradation, and matrix metalloproteinase expression in rabbit articular cartilage. Biochem Biophys Res Commun 267(1):438–444 44. MacRae VE, Burdon T, Ahmed SF, Farquharson C (2006) Ceramide inhibition of chondrocyte proliferation and bone growth is IGF-I independent. J Endocrinol 191(2):369–377 45. Takeda H, Ozaki K, Yasuda H, Ishida M, Kitano S, Hanazawa S (1998) Sphingomyelinase and ceramide inhibit formation of F-actin ring in and bone resorption by rabbit mature osteoclasts. FEBS Lett 422(2):255–258 46. Lee SE, Chung WJ, Kwak HB, Chung CH, Kwack KB, Lee ZH, Kim HH (2001) Tumor necrosis factor-alpha supports the survival of osteoclasts through the activation of Akt and ERK. J Biol Chem 276(52):49343–49349 47. Iwamoto T, Fukumoto S, Kanaoka K, Sakai E, Shibata M, Fukumoto E, Inokuchi Ji J, Takamiya K, Furukawa K, Kato Y, Mizuno A (2001) Lactosylceramide is essential for the osteoclastogenesis mediated by macrophage-colony-stimulating factor and receptor activator of nuclear factor-kappa B ligand. J Biol Chem 276(49):46031–46038 48. Fukumoto S, Iwamoto T, Sakai E, Yuasa K, Fukumoto E, Yamada A, Hasegawa T, Nonaka K, Kato Y (2006) Current topics in pharmacological research on bone metabolism: osteoclast differentiation regulated by glycosphingolipids. J Pharmacol Sci 100(3):195–200 49. Kato K, Adachi S, Matsushima-Nishiwaki R, Minamitani C, Natsume H, Katagiri Y, Hirose Y, Mizutani J, Tokuda H, Kozawa O, Otsuka T (2011) Regulation by heat shock protein 27 of osteocalcin synthesis in osteoblasts. Endocrinology 152(5):1872–1882 50. Kozawa O, Niwa M, Matsuno H, Tokuda H, Miwa M, Ito H, Kato K, Uematsu T (1999) Sphingosine 1-phosphate induces heat shock protein 27 via p38 mitogen-activated protein kinase activation in osteoblasts. J Bone Miner Res 14(10):1761–1767 51. Sato C, Iwasaki T, Kitano S, Tsunemi S, Sano H (2012) Sphingosine 1-phosphate receptor activation enhances BMP-2-induced osteoblast differentiation. Biochem Biophys Res Commun 423(1):200–205 52. Liu R, Farach-Carson MC, Karin NJ (1995) Effects of sphingosine derivatives on MC3T3-E1 pre-osteoblasts: psychosine elicits release of calcium from intracellular stores. Biochem Biophys Res Commun 214(2):676–684 53. Lyons JM, Karin NJ (2001) A role for G protein-coupled lysophospholipid receptors in sphingolipid-induced Ca2? signaling in MC3T3-E1 osteoblastic cells. J Bone Miner Res 16(11):2035–2042 54. Grey A, Chen Q, Callon K, Xu X, Reid IR, Cornish J (2002) The phospholipids sphingosine-1-phosphate and lysophosphatidic acid prevent apoptosis in osteoblastic cells via a signaling pathway involving G(i) proteins and phosphatidylinositol-3 kinase. Endocrinology 143(12):4755–4763 55. Roelofsen T, Akkers R, Beumer W, Apotheker M, Steeghs I, van de Ven J, Gelderblom C, Garritsen A, Dechering K (2008) Sphingosine-1-phosphate acts as a developmental stage specific inhibitor of platelet-derived growth factor-induced chemotaxis of osteoblasts. J Cell Biochem 105(4):1128–1138 56. Quint P, Ruan M, Pederson L, Kassem M, Westendorf JJ, Khosla S, Oursler MJ (2013) Sphingosine 1-phosphate (S1P) receptors 1 and 2 coordinately induce mesenchymal cell migration through
123
Z. Khavandgar, M. Murshed
57.
58.
59.
60.
61.
62.
63.
64.
65.
66.
67.
68.
69.
70.
71.
S1P activation of complementary kinase pathways. J Biol Chem 288(8):5398–5406 Kim MK, Lee HY, Kwak JY, Park JI, Yun J, Bae YS (2006) Sphingosine-1-phosphate stimulates rat primary chondrocyte proliferation. Biochem Biophys Res Commun 345(1):67–73 Masuko K, Murata M, Beppu M, Nakamura H, Kato T, Yudoh K (2012) Sphingosine-1-phosphate modulates expression of vascular endothelial growth factor in human articular chondrocytes: a possible new role in arthritis. Int J Rheum Dis 15(4):366–373 Masuko K, Murata M, Nakamura H, Yudoh K, Nishioka K, Kato T (2007) Sphingosine-1-phosphate attenuates proteoglycan aggrecan expression via production of prostaglandin E2 from human articular chondrocytes. BMC Musculoskelet Disord 8:29 Ishii M (1831) Kikuta J 2013 Sphingosine-1-phosphate signaling controlling osteoclasts and bone homeostasis. Biochim Biophys Acta 1:223–227 Lotinun S, Kiviranta R, Matsubara T, Alzate JA, Neff L, Luth A, Koskivirta I, Kleuser B, Vacher J, Vuorio E, Horne WC, Baron R (2013) Osteoclast-specific cathepsin K deletion stimulates S1Pdependent bone formation. J Clin Invest 123(2):666–681 Keller J, Catala-Lehnen P, Huebner AK, Jeschke A, Heckt T, Lueth A, Krause M, Koehne T, Albers J, Schulze J, Schilling S, Haberland M, Denninger H, Neven M, Hermans-Borgmeyer I, Streichert T, Breer S, Barvencik F, Levkau B, Rathkolb B, Wolf E, Calzada-Wack J, Neff F, Gailus-Durner V, Fuchs H, de Angelis MH, Klutmann S, Tsourdi E, Hofbauer LC, Kleuser B, Chun J, Schinke T, Amling M (2014) Calcitonin controls bone formation by inhibiting the release of sphingosine 1-phosphate from osteoclasts. Nat Commun 21(5), p 5215 Guenet JL, Stanescu R, Maroteaux P, Stanescu V (1981) Fragilitas ossium: a new autosomal recessive mutation in the mouse. J Hered 72(6):440–441 Stoffel W, Jenke B, Block B, Zumbansen M, Koebke J (2005) Neutral sphingomyelinase 2 (smpd3) in the control of postnatal growth and development. Proc Natl Acad Sci USA 102(12):4554–4559 Stoffel W, Jenke B, Holz B, Binczek E, Gunter RH, Knifka J, Koebke J, Niehoff A (2007) Neutral sphingomyelinase (SMPD3) deficiency causes a novel form of chondrodysplasia and dwarfism that is rescued by Col2A1-driven smpd3 transgene expression. Am J Pathol 171(1):153–161 Hannun YA, Obeid LM (2002) The Ceramide-centric universe of lipid-mediated cell regulation: stress encounters of the lipid kind. J Biol Chem 277(29):25847–25850 Li Z, Wu G, Sher RB, Khavandgar Z, Hermansson M, Cox GA, Doschak MR, Murshed M, Beier F, Vance DE (2014) Choline kinase beta is required for normal endochondral bone formation. Biochim Biophys Acta 1840(7):2112–2122 Kakoi H, Maeda S, Shinohara N, Matsuyama K, Imamura K, Kawamura I, Nagano S, Setoguchi T, Yokouchi M, Ishidou Y, Komiya S (2014) Bone morphogenic protein (BMP) signaling upregulates neutral sphingomyelinase 2 to suppress chondrocyte maturation via the Akt protein signaling pathway as a negative feedback mechanism. J Biol Chem 289(12):8135–8150 Holland WL, Brozinick JT, Wang LP, Hawkins ED, Sargent KM, Liu Y, Narra K, Hoehn KL, Knotts TA, Siesky A, Nelson DH, Karathanasis SK, Fontenot GK, Birnbaum MJ, Summers SA (2007) Inhibition of ceramide synthesis ameliorates glucocorticoid-, saturated-fat-, and obesity-induced insulin resistance. Cell Metab 5(3):167–179 Yadav MC, Simao AM, Narisawa S, Huesa C, McKee MD, Farquharson C, Millan JL (2011) Loss of skeletal mineralization by the simultaneous ablation of PHOSPHO1 and alkaline phosphatase function: a unified model of the mechanisms of initiation of skeletal calcification. J Bone Miner Res 26(2):286–297 Brophy PJ, Choy PC, Toone JR, Vance DE (1977) Choline kinase and ethanolamine kinase are separate, soluble enzymes in rat liver. Eur J Biochem 78(2):491–495
Sphingolipid and skeleton 72. Gallego-Ortega D, Ramirez de Molina A, Ramos MA, ValdesMora F, Barderas MG, Sarmentero-Estrada J, Lacal JC (2009) Differential role of human choline kinase alpha and beta enzymes in lipid metabolism: implications in cancer onset and treatment. PLoS One 4(11):e7819 73. Anderson HC (2003) Matrix vesicles and calcification. Curr Rheumatol Rep 5(3):222–226 74. Wu LN, Genge BR, Kang MW, Arsenault AL, Wuthier RE (2002) Changes in phospholipid extractability and composition accompany mineralization of chicken growth plate cartilage matrix vesicles. J Biol Chem 277(7):5126–5133 75. Mebarek S, Abousalham A, Magne D, le Do D, BandorowiczPikula J, Pikula S, Buchet R (2013) Phospholipases of mineralization competent cells and matrix vesicles: roles in physiological and pathological mineralizations. Int J Mol Sci 14(3):5036–5129
969 76. Coleman RM, Aguilera L, Quinones L, Lukashova L, Poirier C, Boskey A (2012) Comparison of bone tissue properties in mouse models with collagenous and non-collagenous genetic mutations using FTIRI. Bone 51(5):920–928 77. Lipina C, Hundal HS (2011) Sphingolipids: agents provocateurs in the pathogenesis of insulin resistance. Diabetologia 54(7):1596–1607 78. Samad F, Badeanlou L, Shah C, Yang G (2011) Adipose tissue and ceramide biosynthesis in the pathogenesis of obesity. Adv Exp Med Biol 721:67–86 79. Yang G, Badeanlou L, Bielawski J, Roberts AJ, Hannun YA, Samad F (2009) Central role of ceramide biosynthesis in body weight regulation, energy metabolism, and the metabolic syndrome. Am J Physiol Endocrinol Metab 297(1):E211–E224
123
Cellular and Molecular Life Sciences
REVIEW
Sphingolipid metabolism and its role in the skeletal tissues Zohreh Khavandgar • Monzur Murshed
Received: 3 June 2014 / Revised: 28 October 2014 / Accepted: 10 November 2014 / Published online: 26 November 2014 Ó Springer Basel 2014
Abstract The regulators affecting skeletal tissue formation and its maintenance include a wide array of molecules with very diverse functions. More recently, sphingolipids have been added to this growing list of regulatory molecules in the skeletal tissues. Sphingolipids are integral parts of various lipid membranes present in the cells and organelles. For a long time, these macromolecules were considered as inert structural elements. This view, however, has radically changed in recent years as sphingolipids are now recognized as important second messengers for signal-transduction pathways that affect cell growth, differentiation, stress responses and programmed death. In the current review, we discuss the available data showing the roles of various sphingolipids in three different skeletal cell types—chondrocytes in cartilage and osteoblasts and osteoclasts in bone. We provide an overview of the biology of sphingomyelin phosphodiesterase 3 (SMPD3), an important regulator of sphingolipid metabolism in the skeleton. SMPD3 is localized in the plasma membrane and has been shown to cleave sphingomyelin to generate ceramide, a bioactive lipid second messenger, and phosphocholine, an essential nutrient. SMPD3 deficiency in mice impairs the mineralization in both cartilage and
Z. Khavandgar M. Murshed (&) Faculty of Dentistry, McGill University, Montreal, Quebec, Canada e-mail: [email protected] M. Murshed Department of Medicine, McGill University, Montreal, Quebec, Canada M. Murshed Shriners Hospital for Children, McGill University, Montreal, Quebec, Canada
bone extracellular matrices leading to severe skeletal deformities. A detailed understanding of SMPD3 function may provide a novel insight on the role of sphingolipids in the skeletal tissues. Keywords Sphingolipids SMPD3 Neutral sphingomyelinase Ceramide Phosphocholine Sphingosine 1 phosphate Matrix vesicles Skeletal development Bone mineralization fro/fro
Background Skeletal tissues not only provide structural and biomechanical supports to the vertebrate body, but also maintain the homeostasis of two essential chemical species, inorganic phosphate and calcium ions, which are required for many critical physiological functions. In addition to its role as a mineral reservoir, more recently the skeleton has been recognized as an endocrine organ that secretes hormone(s) to regulate energy metabolism [1, 2]. These myriad roles of the skeleton make it one of the most dynamic tissues in the body that achieves its complex structure and unique cell-extracellular matrix composition through a multistep process during development. Development of the skeleton in vertebrates involves concerted functions of three major cell types—chondrocytes in cartilage, and osteoclasts and osteoblasts in bone. The spatiotemporal distribution of the precursor stem cells that give rise to these cell types, their proliferation, differentiation and programmed death in the developing skeleton determines the growth, shape and load-bearing capacity of the future skeleton [3–5]. Developmentally, bones can be categorized into two types: endochondral and intramembranous bones. A cartilage ‘precast’ (anlagen) is
123
960
essential for endochondral bone formation, which is not needed for the formation of intramembranous bones. Formation of both bone types however, initiates with the condensation of mesenchymal stem cells (MSC) that first proliferate and differentiate directly into osteoblasts in the case of intramembranous bones or chondrocytes in the case of endochondral bones. Chondrocytes within the core of the developing endochondral bones eventually differentiate to form two distinct growth plates at both ends. Each of these growth plates has four distinct zones: (1) a resting zone of chondrocyte precursors; (2) a proliferative zone with chondrocytes that synthesize extracellular matrix (ECM) proteins including type II collagen; and (3) prehypertrophic and (4) hypertrophic zones. The hypertrophic zone carries terminally differentiated chondrocytes that go through programmed death after secreting a cartilaginous matrix rich in type X collagen. This matrix becomes mineralized and vascularized, and eventually replaced by a type I collagen-rich bone matrix through the concerted resorptive and formative activities of osteoclasts and osteoblasts, respectively [3–5]. With the recent advancements of molecular genetics and the availability of a plethora of genetically modified mouse models, many of the genetic regulators of skeletal development have been identified. These regulators include a wide array of proteins with very diverse functions including signaling molecules/hormones (e.g. IHH, BMPs, FGFs, Leptin, PTH and PTHrP) and their receptors, membrane proteins (e.g. PHEX), transcription factors (e.g. RUNX2, OSX and ATF4), intracellular and extracellular enzymes (e.g. PHOSPHO1 and ALPL) and matrix proteins (e.g. Type I collagen, osteocalcin) [1, 6–21]. More recently, sphingolipids have been added to this growing list of molecules regulating skeletal development and tissue homeostasis [22–25]. Sphingolipids are an integral part of various lipid membranes present in the cells and organelles and serve as important second messengers for signaltransduction pathways affecting cell growth, differentiation, stress responses and death [26].
Sphingolipid biosynthesis Sphingolipids carry long-chain aliphatic-amine backbones containing two or three hydroxyl groups which are known as sphingoid bases [27–29]. The most abundant sphingoid base in animal tissues is sphingosine. Additionally, tissues contain a saturated analog of sphingosine known as dihydrosphingosine or sphinganine. Enzymatic modifications of these simple sphingoid bases generate a wide array of sphingolipids with diverse structural and functional properties (Fig. 1). The most common functional sphingolipids involve more than hundred species of ceramide,
123
Z. Khavandgar, M. Murshed
Fig. 1 Chemical structures of sphingosine and its derivatives ceramide and sphingomyelin. A ceramide molecule is composed of sphingosine and a fatty acid (green) linked by an amide bond. An enzymatic addition of a phosphocholine head group (blue) to a ceramide molecule leads to the synthesis of sphingomyelin
phosphorylated ceramides, sphingoid bases and sphingoid base phosphates among others [27–29]. The current review will discuss the role of some of these sphingolipids in various skeletal cell types. De novo synthesis of sphingolipids Sphingolipids are primarily synthesized in the endoplasmic reticulum (ER), but are transported to the plasma membrane and endosomes, where they perform many of their functions [12, 26, 30]. In the ER, sphingolipids can be synthesized de novo using simple precursors, such as serine and palmitoyl CoA. In a condensation reaction catalyzed by a ubiquitous enzyme called serine palmitoyltransferase these two substrates first generate 3-ketosphinganine, which is converted to sphinganine by ketosphinganine reductase [12, 26, 30]. A dihydroxyceramide synthase then acetylates sphinganine to generate dihydroxyceramide. Finally, a desaturase-catalyzed reaction converts an inactive dihydroxyceramide to ceramide, an important lipid second messenger that regulates a variety of cellular activities (Fig. 2). Ceramide serves as a substrate for one of three enzymes that produce sphingolipids, such as glucosyl ceramide, galactosyl ceramide and sphingomyelin. Ceramide can also be phosphorylated by a family of ceramide kinases. The hydrolysis of ceramide by ceramidase generates sphingosine, which can be phosphorylated by two sphingosine kinases to generate sphingosine-1-phosphate (S1P)—a bioactive lipid with signaling properties. S1P can also be dephosphorylated by a class of lipid-phosphate phosphatases (LPP1-3). These transmembrane enzymes with extracellular catalytic sites play an important regulatory role in S1P-medated signaling events.
Sphingolipid and skeleton
961
Fig. 2 De novo and sphingomyelinase pathways showing the metabolism of ceramide, sphingomyelin and sphingosine
Synthesis of sphingolipids by sphingomyelinases
Role of sphingolipids in skeletal cells: in vitro findings
In addition to the de novo pathway described above, various sphingomyelinases (SMase) play a critical role in sphingolipid biosynthesis (Fig. 2). Sphingomyelinases belong to a class of enzymes, which cleave sphingomyelin to generate ceramide and phosphocholine. Depending on their pH optima, these enzymes have been classified into three categories—acidic, alkaline and neutral sphingomyelinases [25, 31]. Acidic sphingomyelinases are the most extensively studied enzymes among the three subclasses. Mutations in acidic sphingomyelin phosphodiesterase 1 gene (Smpd1) cause Niemann-Pick disease, which is characterized by abnormal lipid accumulation in the spleen, liver, lung, bone marrow and brain [25]. Recently, this enzyme has been implicated in ceramide-mediated signaling events [32]. Alkaline sphingomyelinase hydrolyzes dietary sphingomyelin and generates sphingolipid metabolites in the gut [33]. Impaired spingomyelin metabolism in the intestine might have implications in colon cancer development as dietary supplementation with sphingomyelin and ceramide analogs was found to inhibit the development of chemically-induced colon cancer in animal models [34]. Neutral sphingomyelinases are the key enzymes involved in ceramide-mediated signaling events. So far, three sphingomyelin phosphodiesterases (SMPD2, 3 and 4) have been identified that belong to this subclass [25]. Although all these isoforms are present as membrane-bound forms, they differ in their tissue distribution. SMPD2 and 4 are ubiquitously expressed, while SMPD3 expression is largely restricted to the bone, brain and cartilage [24].
The functions of two major sphingolipids, ceramide and S1P, in skeletal cells have been discussed below. Ceramide Obeid et al. [35] used synthetic C2-ceramide first to demonstrate the role of sphingolipids in apoptosis. This study showed that this ceramide analog promotes DNA fragmentation and apoptosis in U937 monoblast leukemia cells via upregulation of Tumor Necrosis Factor-alpha (TNF-alpha). Since then numerous studies have reported the pro-apoptotic role of ceramide in many different cell types. These studies described multiple signaling intermediates/pathways including ceramide-activated protein kinase and the activation of mitogen-activated protein kinase to induce cellular apoptosis. For example, ceramide-induced apoptosis may involve the activation protein kinase C-zeta to promote c-Jun N-terminal kinase (JNK) activation and the nuclear translocation of transcription factor nuclear factor-kappa B (NF-kappa B) [35]. It has been suggested that the membrane contact sites (MCS) between ER and mitochondria facilitate ceramideinduced apoptosis. Ceramide produced by SMPD3 in the ER membrane may translocate to the outer mitochondrial membrane to form ceramide channels, which may in turn release pro-apoptotic molecule cytochrome C [36]. A recent study proposed that ceramide serves as a precursor for S1P and hexadecenal which in turn regulates apoptosis. These molecules activate apoptotic regulators Bak and Bax, promoting the release of cytochrome C from the
123
962
mitochondria, which then activates caspases, enzymes essential for apoptotic cell death [37]. So far, only a handful of studies have been performed to examine the effects of ceramide on the apoptosis of bone cells in vitro. Treatment of MC3T3-E1 preosteoblasts with the nitric oxide (NO) donor sodium nitroprusside causes apoptosis [38]. This study reports an increase of long-chain intracellular ceramides, C22 and C24, upon this treatment. Interestingly, NO has been recently shown to promote the release of cytochrome C from the mitochondria [39]. It is therefore likely that NO increases the release of long-chain ceramides, which in turn results in the release of cytochrome C from the mitochondria turning on the intrinsic apoptotic pathways in cultured osteoblasts. In another cell-culture study, murine MC3T3-E1 preosteoblasts were treated with TNF-alpha to induce apoptosis. Within 3 min of TNF-alpha treatment, detectable increase of endogenous ceramide levels was apparent, which reached to its peak by 30 min. TNF-alpha treatment of MC3T3-E1 cells also promoted the nuclear translocation and activation of NF-kappa B. Interestingly, treatment of the preosteoblasts with a ceramide analog alone was sufficient to affect the nuclear translocation/activation of NFkappa B in a similar fashion. This observation lead to the conclusion that in osteoblasts, TNF-alpha modulates NFkappa B localization and function through the upregulation of ceramide synthesis [40]. A separate study showed that the treatment of MC3T3-E1 cells with dexamethasone, a known immunosuppressant, reduces TNF-alpha-induced ceramide production and dampens not only the activation of NF-kappa B, but also the activities of two pro-apoptotic enzymes, JNK and caspase-3-like extracellular protease. Additionally, there was a redistribution of cytochrome C in these cells [41]. This later observation further confirms that mitochondrial pathways might be relevant in the apoptosis of bone cells. Collectively, these findings indicate that inflammation-induced ceramide production in osteoblasts, may result in apoptosis through multiple mechanisms. A more recent study reported that although high-dose treatment (2 9 10-6 M) of mouse primary osteoblasts with C2-ceramide-induced apoptosis, the low-dose treatment (10-7 M) actually resulted in increased cell survival through the protein kinase C activity. It is possible that there might be separate pathways that promote ceramidemediated pro-apoptotic and pro-survival signaling events [42]. As is the case with the cultured osteoblasts, a high dose (3 9 10-5 to 10-4 M) of C2-ceramide treatment induced apoptosis of chondrocytes in rabbit articular cartilage explants. Also, C2-ceramide treatment increased matrix metalloproteinase activity in these explants. These data provide evidence that chronic inflammatory responses that increase matrix remodeling in osteoarthritis might be
123
Z. Khavandgar, M. Murshed
caused by an upregulation of ceramide synthesis [43]. C2ceramide induced apoptosis in murine chondrogenic ATDC5 cells without affecting the expression of chondrogenic markers Sox9, Col2a1 and Col10a1 [44]. Although C2-ceramide treatments lead to increased apoptosis in both osteoblasts and chondrocytes, there was no effect of this cell-permeable synthetic ceramide on the apoptosis of rabbit mature osteoclasts [45]. However, treatment with ceramide inhibited F-actin ring formation within the sealing zone of osteoclasts and impaired their resorption capacities. A similar effect was observed when these cells were treated with exogenous sphingomyelinase [45]. Interestingly, TNF-alpha treatment that promotes apoptosis in osteoblasts by upregulating ceramide synthesis, actually favors the survival of cultured murine osteoclasts [46]. As shown by this study, TNF-alpha treatment activates prosurvival pathways in osteoclasts by engaging phosphatidylinositol 3-kinase, Akt and MEK/ERK signaling. Several studies suggest that ceramide and its derivatives play an important role in osteoclastogenesis. Lactosylceramide, a derivative of ceramide, has been shown to induce the expression of the receptor-activator of NF-kappa B (RANK) in bone marrow-derived osteoclasts [47]. Moreover, this study showed that the inhibition of glucosylceramide synthase by D-threo-1-phenyl-2-decanoylamino-3-morpholino-1-propanol (D-PDM) prevented granulocyte–macrophage colony-stimulating factor and RANK ligand-mediated osteoclastogenesis. In a subsequent study, it was shown that the D-PDM-mediated inhibition of osteoclastogenesis could be rescued by treating the osteoclasts with lactosylceramide [48]. Sphingosine and its derivatives Sphingosine and its derivatives have been implicated in a variety of cell functions in osteoblasts. In MC3T3-E1 preosteoblast cultures, S1P induces the synthesis of heatshock protein 27 (HSP27), a molecular chaperon that acts as a stress-response protein; HSP27 in turn negatively regulates the synthesis of osteocalcin, a marker expressed by functional osteoblasts [49, 50]. Interestingly, treatment of myogenic C2C12 cells with bone morphogenetic protein-2 (BMP-2) in combination with S1P or FTY720 (an agonist of S1P receptor) significantly increased the expression of osteocalcin and several other osteogenic markers [51]. Although the latter study did not report osteocalcin protein levels, this apparent discrepancy in osteocalcin expression might be caused by the differences in the differentiation stages and cell types used for these two studies. Exposure of MC3T3-E1 cells to sphingosine, S1P and sphingosylphosphorylcholine promoted intracellular calcium release increasing cytosolic calcium levels [52, 53]. Also, S1P treatments of rat calvarial osteoblasts
Sphingolipid and skeleton
963
Fig. 3 The major effects of ceramide (a) and Sphingosine 1 phosphate (S1P) (b) on osteoblasts, chondrocytes and osteoclasts. The filled up arrows indicate activation and the filled down arrows indicate inhibition
and human SaOS-2 osteosarcoma cells have been shown to prevent apoptosis of these cells that was initiated by serum deprivation [54]. S1P signaling has been implicated in osteoblast migration; S1P acts as a chemorepellent for the undifferentiated preosteoblasts [55]. As opposed to S1P function, it was shown that platelet-derived growth factor (PDGF) that plays a role in bone remodeling and fracture healing, acts as a chemoattractant for preosteoblasts. Upon treatment with BMP-2, these cells differentiate to mature osteoblasts retaining their abilities to respond to PDGF but not to S1P. Roelofsen et al. [56] suggested that migration of osteoblasts is controlled by the balance between PDGF and S1P allowing only differentiated osteoblasts to travel to the site of bone formation. Although interesting, this notion has been challenged by a more recent study identifying S1P as a chemoattractant for MSCs that give rise to cells of the osteoblast lineage. According to this study, bone-resorbing osteoclasts release S1P to induce the migration of MSCs and thereby promote the recruitment of the osteoblast precursors at the sites undergoing bone remodeling. It is, however, possible that osteoblasts at different stages of differentiation may respond differently to S1P. S1P has been shown to promote the proliferation of rat primary chondrocytes [57]. S1P treatment induced phospholipase C-mediated cytosolic calcium release in these cells. Also, extracellular signal-regulated kinase (ERK) and p38 mitogen-activated kinases were induced. When treated with ERK inhibitor PD98059 the cell proliferation was almost completely blocked, while p38 kinase inhibitor SB203580 did not have any effect. These data suggest a cardinal role for ERK in S1P-mediated chondrocyte proliferation. In articular chondrocytes, S1P signaling has
been implicated in cartilage matrix degradation, upregulation of vascular endothelial growth factor and the progression of arthritis [58, 59]. S1P signaling has also been shown to regulate the migration of osteoclast precursors. Two S1P receptors, S1PR1 and S1PR2, play a critical role in this process. S1PR1 promotes the chemotaxis towards S1P in bone marrow where S1P level is normally low, while S1PR2 mediates the S1P chemorepellant properties in circulation where S1P level is high. This mechanism may facilitate the infiltration of osteoclast precursors from the circulation to the bone marrow [60]. The major effects of ceramide and S1P on the three major cell types in the skeleton have been depicted in Fig. 3.
Role of sphingolipids in skeletal tissues: in vivo findings The role of osteoclast-derived S1P on bone remodeling has been elaborated in an in vivo study by Lotinun et al.[61]. In this study, osteoclast-specific inactivation of cathepsin K, an extracellular matrix degrading enzyme, resulted in an increase of S1P secretion by these cells. Secreted S1P promoted increased bone formation by osteoblasts primarily by enhancing the differentiation of osteoblast precursors to functional osteoblasts. As described by Quint et al. [56], osteoclasts through S1P release may also promote the recruitment of MSCs that give rise to osteoblast precursors and thereby contribute to the observed high bone mass phenotype. A recently published study showed that calcitonin, a thyroid-derived hormone, affects bone formation by regulating the release of S1P by osteoclasts [62].
123
964
Emerging data from the analysis of genetically altered mouse models suggest that sphingolipid metabolism during embryonic development plays a critical role in normal skeletogenesis. This insight originally came from two different mouse models lacking a neutral sphingomyelinase, nSMase2, also known as sphingomyelin phosphodiesterase 3 (SMPD3). Currently, there are two SMPD3-deficient mouse models available; one carries a chemically induced deletion of 1,758 base pairs encompassing part of intron 8 and most of exon 9 (fro) in the Smpd3 locus (fro/fro model) and the other was generated by gene targeting (Smpd3-/model) [22, 63, 64]. So far no skeletal abnormalities in mice lacking SMPD1 and SMPD2 activity have been reported. The fro mutation does not affect the membrane localization of SMPD3 but completely abolishes its enzymatic activity. There was a significant reduction of neutral sphingomyelinase activity in the brain and skeletal tissues of fro/fro mice where Smpd3 is highly expressed [24]. These mice show severe skeletal dysplasia hall marked by abnormal cartilage development and bone mineralization defects. Histological sections of the humeri of late-gestation embryos suggested delayed apoptosis of hypertrophic chondrocytes and poor mineralization of the cortical bones. Both of these anomalies appear to contribute to the skeletal dysplasia seen in the mutant mice [24]. The skeletal deformities seen in the fro/fro mice become less apparent with age although the bone mineralization defects have been detected in mice that are more than 6 month old (unpublished data). fro/fro mice also show delay in mantle dentin mineralization and subsequent delay in enamel formation during early odontogenesis. These abnormalities were largely corrected as these mice grew older [23]. Both fro/fro and the gene targeted Smpd3-/- mice share similar gross skeletal abnormalities, i.e. short-limb dwarfism, deformation of long bones and abnormally formed rib cages. However, Stoffel et al. [65] did not observe the bone and tooth mineralization abnormalities documented in fro/fro mice in their gene targeted model. The phenotypic variations in fro/fro and Smpd3-/- mice cannot be attributable to the presence of any additional genetic alterations in the former strain as such mutation(s), if not closely associated, would have been fully segregated during the propagation of fro/fro mice in multiple laboratories over the past four decades. This inference, together with the observation that the fully penetrant fro/fro phenotype including the severe hypomineralization defect, is always associated with the Smpd3 deletion mutation identified by Aubin et al, strongly suggests that the loss of SMPD3 function is indeed the cause of the fro/fro phenotype [22]. Indeed, two recently published studies describing the complete rescue of the bone and tooth mineralization abnormalities in fro/fro; Col1a1-Smpd3 mice expressing
123
Z. Khavandgar, M. Murshed
Smpd3 in osteoblasts and odontoblasts ruled out the possibility of the involvement of any additional mutation(s) in the fro/fro phenotype [23, 24]. These data provide unambiguous demonstration of a direct role for SMPD3 in osteoblast/bone mineralization. Differences in analytical methods used to characterize the Smpd3-/- and fro/fro mice might lead to different interpretations about the apparent phenotype in the same tissue. For example, Khavandgar et al. analyzed the hard tissues in fro/fro mice using histomorphometric techniques commonly used to detect mineralization defects in bone samples. Stoffel et al. however, analyzed the mineralization status of the Smpd3-/- mice solely by bone mineral density analysis, which determines the total bone mineral content and is not suitable for detecting an increase in unmineralized bone matrix.
SMPD3, a key regulator of sphingolipid functions in bone: possible mode of action Although the role of SMPD3 in the developing skeleton— both in bone and cartilage, is now well established, it’s mode of action is still poorly understood. Based on the data obtained from the animal models, it appears that by far SMPD3 is the only lipid metabolizing enzyme that affects bone mineralization and chondrocyte hypertrophy in vivo. SMPD3, an integral membrane protein that is attached to the inner leaflet of the cell membrane, cleaves sphingomyelin to generate many different species of ceramide that regulate a myriad of cell function. The SMPD3-catalyzed reaction also liberates an essential nutrient, phosphocholine [66]. At this point it is not clear whether both ceramide and phosphocholine or only one of them is involved in skeletal tissue morphogenesis. However, the findings from mutant mouse models suggest that these two bioactive metabolites may have distinct tissue-specific roles. The first in vivo proof suggesting that SMPD3 might be involved in apoptosis of chondrocytes came from the observation of Khavandgar et al. [24]. A delayed hypertrophy and impaired apoptosis of terminally differentiated chondrocytes in these mice results in an abnormal accumulation of chondrocytes in the mid-shaft region of the developing bone (Fig. 4). The fro/fro mice show reduced ceramide levels in their skeletal tissues. Considering the extensive experimental evidence showing a role for ceramide in the regulation of cellular apoptosis, the chondrocyte phenotype in fro/fro mice can be attributed to impaired ceramide metabolism in this strain. This study however, did not fully rule out the possible effects of reduced phosphocholine levels caused by the loss of SMPD3 function on the death of hypertrophic chondrocytes. Indeed, in a recent study, Li et al. [67] reported that
Sphingolipid and skeleton
965
Fig. 4 Abnormal presence of cartilage matrix and chondrocytes asterisk in a fro/ fro developing humerus from an E15.5 embryo. The histological sections were stained with Safranin O
mice lacking choline kinase beta which showed reduced phosphocholine levels, there was an expansion of the hypertrophic zone. Nevertheless, based on the numerous in vitro and in vivo experiments on ceramide-mediated cell death, it is likely that the reduced ceramide level is responsible, at least in part, for the observed delay in apoptosis of hypertrophic chondrocytes in the developing skeleton of fro/fro mice. Interestingly, a recent study suggested that GW4869-mediated inhibition of neutral sphingomyelinase function accelerates the mineralization of chondrogenic ATDC5 cultures [68]. This conclusion was surprising as it contradicts the in vivo findings in SMPD3-deficient mice. Although the role of ceramide in apoptosis is supported by numerous studies, at present no experimental data directly links it to ECM mineralization. As mentioned above, various species of ceramides in the cells are mainly generated by two different mechanisms—the de novo and the sphingomyelinase pathways (Fig. 2). In the de novo pathway, dihydroceramide desaturase 1 (DES1) is the final enzyme that converts dihydroceramide to ceramide. Des1 null mice show multiple physiologic anomalies including hypophagia, weight loss and overall growth impairment and die prematurely [69]. Although these mice are smaller in size, unlike fro/fro or Smpd3 knockout mice they show a normal skeletal structure. They do not show any bone mineralization defects (unpublished data). This later finding has two implications: it is possible that ceramides generated from two different pathways contribute to two distinct pools with separate cellular functions; alternatively, it is possible that ceramides do not play any role in the process of bone mineralization. Interestingly, Des1 null mice like fro/fro mice show a reduced cellular/tissue ceramide levels and a similar metabolic phenotype (68 and unpublished data). This observation suggests that ceramides generated from the de novo and sphingomyelinase pathways act in concert to affect the same
biological functions. If this notion is correct then the absence of bone mineralization defects in Des1 knockout mice may imply that the maintenance of ceramide levels is not essential for normal bone mineralization. This may also point towards the involvement of phosphocholine, the other product of the SMPD3-catalyzed reaction, in the regulation of bone mineralization. The importance of phosphocholine in bone mineralization has been convincingly shown by animal experiments [17, 70]. Phosphocholine is generated from sphingomyelin by SMPD3, or by cytosolic choline kinases from choline, which can be cleaved by PHOSPHO1, an intracellular enzyme [71, 72]. PHOSPHO1 deficiency in mice causes a similar albeit milder bone mineralization defects as seen in fro/fro mice [17, 70]. At present it is not clear how SMPD3, a cell membranebound enzyme with an intracellular catalytic domain, may affect ECM mineralization. One possible mechanism may involve matrix vesicles (MVs) that are released from the matrix synthesizing cells in the mineralized tissues [73]. Although there is no general consensus, it is believed that the initial nucleation of minerals may first occur within these extracellular microstructures. Wu et al. [74] showed that extensive phospholipid degradation occurs in the mineralizing MVs with a concomitant increase of free fatty acids, indicative of the presence of phospholipase activity. This study provide the early evidence that linked phospholipid metabolism to the initiation of mineralization. Phospholipase activity generates phosphocholine and phosphoethanolamine, which can be cleaved by PHOSPHO1 releasing free phosphate inside the MVs and promoting mineral nucleation. SMPD3 may cleave sphingomyelin present in the MV membrane to generate additional phosphocholine to further facilitate this process. This later possibility is supported by the observation that both PHOSPHO1 and SMPD3 are present in MV preparations [75].
123
966
During embryonic development of mice, Smpd3 is highly expressed in the bone, cartilage and brain [24]. This expression pattern is maintained during the perinatal period. However, until now only a limited number of studies investigated the regulation of the Smpd3 promoter in the skeletal tissues/cell types. An in vitro study convincingly showed that treatment with BMP-2 upregulated Smpd3 expression in myogenic C2C12 cells. This study also reported the presence of RUNX2-binding elements in the murine Smpd3 promoter and an upregulation of Smpd3 expression upon transfection of C2C12 cells with a Runx2 expression vector. Later, using an ex vivo organ culture system and chondrogenic ATDC5 cells, Kakoi et al. [68] demonstrated a similar BMP-2-induced upregulation of Smpd3 expression during chondrogenesis. Although Smpd3 expression in bone is higher than any other tissue, until now no other osteoblast-specific transcription factor has been shown to regulate its expression [24]. Coleman et al. [76] have reported abnormal matrix deposition in SMPD3-deficient fro/fro mice. This group did not find any alteration of trabecular bone parameter in the three-month-old mutant mice in comparison to their littermate controls. The mineralization defects in bones and teeth and the growth plate abnormalities in fro/fro mice appear to be less severe with age. Taken together, these findings suggest that SMPD3 activity is required for normal skeletal development but becomes redundant in aged mice. In Smpd3 knockout mice, reduction of several brainderived hormones e.g. thyroid hormone, growth hormone and pituitary hormone has been reported [64]. This observation suggests an important neuroendocrine role for SMPD3 in the nervous system, which is consistent with its high expression in the brain. In addition to its local regulatory roles on skeletal ECM mineralization and apoptosis of hypertrophic chondrocytes, SMPD3 may have a novel systemic mode of action in the developing skeleton. The deficiency of growth hormone and the growth abnormalities such as dwarfism and smaller body size seen in both SMPD3-deficient models further support this notion. However, the precise mechanisms by which SMPD3 affects the circulating levels of various brain-derived hormones are still unknown.
Z. Khavandgar, M. Murshed
could be that most of the mutant mouse models that show altered sphingolipid metabolism do not show any overt skeletal phenotype during their adulthood. Nevertheless, a systematic analysis of these mice during embryonic development using histology and morphometric techniques may provide further clues on the function of sphingolipids in skeletal development. The chemical mutagenesis and targeted mutations in murine SMPD3 gene have given us a unique opportunity to explore the role of this lipid metabolizing enzyme in the skeletal tissues. Further studies involving mouse models with tissue-specific inactivation of Smpd3 are needed to decipher the local versus systemic contribution of this enzyme to skeletal development and growth. More specifically, these studies are expected to elucidate the role of this pleiotropic enzyme in the brain and how its activity might contribute to the hypothalamic regulation of skeletal growth. Additionally, novel mechanistic studies may shed light on the specific roles of the two products of the SMPD3-catalyzed reaction in the skeletal tissues. The de novo pathway of sphingolipid metabolism, more particularly ceramide metabolism, has been shown to regulate energy expenditure and a similar role is expected for the sphingomyelinase pathway of ceramide synthesis [69, 77–79]. An important aspect would be to investigate how various species of ceramides generated by SMPD3 in the osteoblasts and chondrocytes might contribute to overall energy metabolism in the body. This work is particularly relevant as osteocalcin, an osteoblast-specific protein has been recently shown to regulate energy metabolism [2].
Conclusion The fro/fro mouse model is one of the first animal model to demonstrate a role for sphingolipids in skeletal development and homeostasis. Although a cell-autonomous role of SMPD3 in bone and tooth mineralization is now well established, the precise mechanism of its action in hard tissue mineralization is still unknown. SMPD3 metabolites, phosphocholine and different species of ceramides, may have distinct roles in the cells of the developing skeleton. A thorough understanding of the mode of actions of these two metabolites will be required to fully appreciate the complex process that regulates vertebrate skeletogenesis.
Future directions Despite the published data from numerous cell culture studies showing the role of lipid metabolites in chondrocytes, osteoblasts and osteoclasts, the in vivo validation of these studies is still missing. At present, limited in vivo data are available on the roles of sphingolipid metabolizing enzymes in the skeletal tissues. One possible reason for this
123
References 1. Horton WA, Degnin CR (2009) FGFs in endochondral skeletal development. Trends Endocrinol Metab 20(7):341–348 2. Karsenty G (2011) Bone endocrine regulation of energy metabolism and male reproduction. C R Biol 334(10):720–724
Sphingolipid and skeleton 3. Karsenty G (2003) The complexities of skeletal biology. Nature 423(6937):316–318 4. Karsenty G, Kronenberg HM, Settembre C (2009) Genetic control of bone formation. Annu Rev Cell Dev Biol 25:629–648 5. Mackie EJ, Tatarczuch L, Mirams M (2011) The skeleton: a multi-functional complex organ: the growth plate chondrocyte and endochondral ossification. J Endocrinol 211(2):109–121 6. Amano K, Hata K, Sugita A, Takigawa Y, Ono K, Wakabayashi M, Kogo M, Nishimura R, Yoneda T (2009) Sox9 family members negatively regulate maturation and calcification of chondrocytes through up-regulation of parathyroid hormonerelated protein. Mol Biol Cell 20(21):4541–4551 7. Amizuka N, Henderson JE, Hoshi K, Warshawsky H, Ozawa H, Goltzman D, Karaplis AC (1996) Programmed cell death of chondrocytes and aberrant chondrogenesis in mice homozygous for parathyroid hormone-related peptide gene deletion. Endocrinology 137(11):5055–5067 8. Degnin CR, Laederich MB, Horton WA (2010) FGFs in endochondral skeletal development. J Cell Biochem 110(5):1046–1057 9. Ducy P, Zhang R, Geoffroy V, Ridall AL, Karsenty G (1997) Osf2/Cbfa1: a transcriptional activator of osteoblast differentiation. Cell 89(5):747–754 10. Komori T, Yagi H, Nomura S, Yamaguchi A, Sasaki K, Deguchi K, Shimizu Y, Bronson RT, Gao YH, Inada M, Sato M, Okamoto R, Kitamura Y, Yoshiki S, Kishimoto T (1997) Targeted disruption of Cbfa1 results in a complete lack of bone formation owing to maturational arrest of osteoblasts. Cell 89(5):755–764 11. Kozhemyakina E, Cohen T, Yao TP, Lassar AB (2009) Parathyroid hormone-related peptide represses chondrocyte hypertrophy through a protein phosphatase 2A/histone deacetylase 4/MEF2 pathway. Mol Cell Biol 29(21):5751–5762 12. Merrill AH Jr (2002) De novo sphingolipid biosynthesis: a necessary, but dangerous, pathway. J Biol Chem 277(29):25843–25846 13. Nakashima K, Zhou X, Kunkel G, Zhang Z, Deng JM, Behringer RR, de Crombrugghe B (2002) The novel zinc finger-containing transcription factor osterix is required for osteoblast differentiation and bone formation. Cell 108(1):17–29 14. Nesbitt T, Fujiwara I, Thomas R, Xiao ZS, Quarles LD, Drezner MK (1999) Coordinated maturational regulation of PHEX and renal phosphate transport inhibitory activity: evidence for the pathophysiological role of PHEX in X-linked hypophosphatemia. J Bone Miner Res 14(12):2027–2035 15. Ornitz DM, Marie PJ (2002) FGF signaling pathways in endochondral and intramembranous bone development and human genetic disease. Genes Dev 16(12):1446–1465 16. Retting KN, Song B, Yoon BS, Lyons KM (2009) BMP canonical Smad signaling through Smad1 and Smad5 is required for endochondral bone formation. Development 136(7):1093–1104 17. Roberts S, Narisawa S, Harmey D, Millan JL, Farquharson C (2007) Functional involvement of PHOSPHO1 in matrix vesiclemediated skeletal mineralization. J Bone Miner Res 22(4):617–627 18. Stewart AJ, Roberts SJ, Seawright E, Davey MG, Fleming RH, Farquharson C (2006) The presence of PHOSPHO1 in matrix vesicles and its developmental expression prior to skeletal mineralization. Bone 39(5):1000–1007 19. Stonich D, Su Y, Dad S, Reddy S, Mostofi Y, Russell D, Chung TDY, Hedrick NM, Rascon J, Garcia X, Sergienko E, Millan JL, Cosford N 2010 The role of PHOSPHO1 in the initiation of skeletal calcification 20. Whyte MP (1994) Hypophosphatasia and the role of alkaline phosphatase in skeletal mineralization. Endocr Rev 15(4):439–461
967 21. Zhou Z, Xie J, Lee D, Liu Y, Jung J, Zhou L, Xiong S, Mei L, Xiong WC (2010) Neogenin regulation of BMP-induced canonical Smad signaling and endochondral bone formation. Dev Cell 19(1):90–102 22. Aubin I, Adams CP, Opsahl S, Septier D, Bishop CE, Auge N, Salvayre R, Negre-Salvayre A, Goldberg M, Guenet JL, Poirier C (2005) A deletion in the gene encoding sphingomyelin phosphodiesterase 3 (Smpd3) results in osteogenesis and dentinogenesis imperfecta in the mouse. Nat Genet 37(8):803–805 23. Khavandgar Z, Alebrahim S, Eimar H, Tamimi F, McKee MD, Murshed M (2013) Local regulation of tooth mineralization by sphingomyelin phosphodiesterase 3. J Dent Res 92(4):358–364 24. Khavandgar Z, Poirier C, Clarke CJ, Li J, Wang N, McKee MD, Hannun YA, Murshed M (2011) A cell-autonomous requirement for neutral sphingomyelinase 2 in bone mineralization. J Cell Biol 194(2):277–289 25. Stoffel W (1999) Functional analysis of acid and neutral sphingomyelinases in vitro and in vivo. Chem Phys Lipids 102(1–2):107–121 26. Merrill AH Jr, Schmelz EM, Dillehay DL, Spiegel S, Shayman JA, Schroeder JJ, Riley RT, Voss KA, Wang E (1997) Sphingolipids–the enigmatic lipid class: biochemistry, physiology, and pathophysiology. Toxicol Appl Pharmacol 142(1):208–225 27. Airola MV, Hannun YA (2013) Sphingolipid metabolism and neutral sphingomyelinases. Handb Exp Pharmacol 215:57–76 28. Mullen TD, Hannun YA, Obeid LM (2012) Ceramide synthases at the centre of sphingolipid metabolism and biology. Biochem J 441(3):789–802 29. Hannun YA, Obeid LM (2011) Many ceramides. J Biol Chem 286(32):27855–27862 30. Futerman AH, Riezman H (2005) The ins and outs of sphingolipid synthesis. Trends Cell Biol 15(6):312–318 31. Nilsson A, Duan RD (1999) Alkaline sphingomyelinases and ceramidases of the gastrointestinal tract. Chem Phys Lipids 102(1–2):97–105 32. Kirschnek S, Paris F, Weller M, Grassme H, Ferlinz K, Riehle A, Fuks Z, Kolesnick R, Gulbins E (2000) CD95-mediated apoptosis in vivo involves acid sphingomyelinase. J Biol Chem 275(35):27316–27323 33. Duan RD, Nyberg L, Nilsson A (1995) Alkaline sphingomyelinase activity in rat gastrointestinal tract: distribution and characteristics. Biochim Biophys Acta 1259(1):49–55 34. Duan RD (2006) Alkaline sphingomyelinase: an old enzyme with novel implications. Biochim Biophys Acta 1761(3):281–291 35. Herr I, Debatin KM (2001) Cellular stress response and apoptosis in cancer therapy. Blood 98(9):2603–2614 36. Siskind LJ, Kolesnick RN, Colombini M (2006) Ceramide forms channels in mitochondrial outer membranes at physiologically relevant concentrations. Mitochondrion 6(3):118–125 37. Chipuk JE, McStay GP, Bharti A, Kuwana T, Clarke CJ, Siskind LJ, Obeid LM, Green DR (2012) Sphingolipid metabolism cooperates with BAK and BAX to promote the mitochondrial pathway of apoptosis. Cell 148(5):988–1000 38. Olivier S, Fillet M, Malaise M, Piette J, Bours V, Merville MP, Franchimont N (2005) Sodium nitroprusside-induced osteoblast apoptosis is mediated by long chain ceramide and is decreased by raloxifene. Biochem Pharmacol 69(6):891–901 39. Snyder CM, Shroff EH, Liu J, Chandel NS (2009) Nitric oxide induces cell death by regulating anti-apoptotic BCL-2 family members. PLoS One 4(9):e7059 40. Kitajima I, Soejima Y, Takasaki I, Beppu H, Tokioka T, Maruyama I (1996) Ceramide-induced nuclear translocation of NFkappa B is a potential mediator of the apoptotic response to TNFalpha in murine clonal osteoblasts. Bone 19(3):263–270
123
968 41. Chae HJ, Chae SW, Kang JS, Bang BG, Cho SB, Park RK, So HS, Kim YK, Kim HM, Kim HR (2000) Dexamethasone suppresses tumor necrosis factor-alpha-induced apoptosis in osteoblasts: possible role for ceramide. Endocrinology 141(8):2904–2913 42. Hill PA, Tumber A (2010) Ceramide-induced cell death/survival in murine osteoblasts. J Endocrinol 206(2):225–233 43. Sabatini M, Rolland G, Leonce S, Thomas M, Lesur C, Perez V, de Nanteuil G, Bonnet J (2000) Effects of ceramide on apoptosis, proteoglycan degradation, and matrix metalloproteinase expression in rabbit articular cartilage. Biochem Biophys Res Commun 267(1):438–444 44. MacRae VE, Burdon T, Ahmed SF, Farquharson C (2006) Ceramide inhibition of chondrocyte proliferation and bone growth is IGF-I independent. J Endocrinol 191(2):369–377 45. Takeda H, Ozaki K, Yasuda H, Ishida M, Kitano S, Hanazawa S (1998) Sphingomyelinase and ceramide inhibit formation of F-actin ring in and bone resorption by rabbit mature osteoclasts. FEBS Lett 422(2):255–258 46. Lee SE, Chung WJ, Kwak HB, Chung CH, Kwack KB, Lee ZH, Kim HH (2001) Tumor necrosis factor-alpha supports the survival of osteoclasts through the activation of Akt and ERK. J Biol Chem 276(52):49343–49349 47. Iwamoto T, Fukumoto S, Kanaoka K, Sakai E, Shibata M, Fukumoto E, Inokuchi Ji J, Takamiya K, Furukawa K, Kato Y, Mizuno A (2001) Lactosylceramide is essential for the osteoclastogenesis mediated by macrophage-colony-stimulating factor and receptor activator of nuclear factor-kappa B ligand. J Biol Chem 276(49):46031–46038 48. Fukumoto S, Iwamoto T, Sakai E, Yuasa K, Fukumoto E, Yamada A, Hasegawa T, Nonaka K, Kato Y (2006) Current topics in pharmacological research on bone metabolism: osteoclast differentiation regulated by glycosphingolipids. J Pharmacol Sci 100(3):195–200 49. Kato K, Adachi S, Matsushima-Nishiwaki R, Minamitani C, Natsume H, Katagiri Y, Hirose Y, Mizutani J, Tokuda H, Kozawa O, Otsuka T (2011) Regulation by heat shock protein 27 of osteocalcin synthesis in osteoblasts. Endocrinology 152(5):1872–1882 50. Kozawa O, Niwa M, Matsuno H, Tokuda H, Miwa M, Ito H, Kato K, Uematsu T (1999) Sphingosine 1-phosphate induces heat shock protein 27 via p38 mitogen-activated protein kinase activation in osteoblasts. J Bone Miner Res 14(10):1761–1767 51. Sato C, Iwasaki T, Kitano S, Tsunemi S, Sano H (2012) Sphingosine 1-phosphate receptor activation enhances BMP-2-induced osteoblast differentiation. Biochem Biophys Res Commun 423(1):200–205 52. Liu R, Farach-Carson MC, Karin NJ (1995) Effects of sphingosine derivatives on MC3T3-E1 pre-osteoblasts: psychosine elicits release of calcium from intracellular stores. Biochem Biophys Res Commun 214(2):676–684 53. Lyons JM, Karin NJ (2001) A role for G protein-coupled lysophospholipid receptors in sphingolipid-induced Ca2? signaling in MC3T3-E1 osteoblastic cells. J Bone Miner Res 16(11):2035–2042 54. Grey A, Chen Q, Callon K, Xu X, Reid IR, Cornish J (2002) The phospholipids sphingosine-1-phosphate and lysophosphatidic acid prevent apoptosis in osteoblastic cells via a signaling pathway involving G(i) proteins and phosphatidylinositol-3 kinase. Endocrinology 143(12):4755–4763 55. Roelofsen T, Akkers R, Beumer W, Apotheker M, Steeghs I, van de Ven J, Gelderblom C, Garritsen A, Dechering K (2008) Sphingosine-1-phosphate acts as a developmental stage specific inhibitor of platelet-derived growth factor-induced chemotaxis of osteoblasts. J Cell Biochem 105(4):1128–1138 56. Quint P, Ruan M, Pederson L, Kassem M, Westendorf JJ, Khosla S, Oursler MJ (2013) Sphingosine 1-phosphate (S1P) receptors 1 and 2 coordinately induce mesenchymal cell migration through
123
Z. Khavandgar, M. Murshed
57.
58.
59.
60.
61.
62.
63.
64.
65.
66.
67.
68.
69.
70.
71.
S1P activation of complementary kinase pathways. J Biol Chem 288(8):5398–5406 Kim MK, Lee HY, Kwak JY, Park JI, Yun J, Bae YS (2006) Sphingosine-1-phosphate stimulates rat primary chondrocyte proliferation. Biochem Biophys Res Commun 345(1):67–73 Masuko K, Murata M, Beppu M, Nakamura H, Kato T, Yudoh K (2012) Sphingosine-1-phosphate modulates expression of vascular endothelial growth factor in human articular chondrocytes: a possible new role in arthritis. Int J Rheum Dis 15(4):366–373 Masuko K, Murata M, Nakamura H, Yudoh K, Nishioka K, Kato T (2007) Sphingosine-1-phosphate attenuates proteoglycan aggrecan expression via production of prostaglandin E2 from human articular chondrocytes. BMC Musculoskelet Disord 8:29 Ishii M (1831) Kikuta J 2013 Sphingosine-1-phosphate signaling controlling osteoclasts and bone homeostasis. Biochim Biophys Acta 1:223–227 Lotinun S, Kiviranta R, Matsubara T, Alzate JA, Neff L, Luth A, Koskivirta I, Kleuser B, Vacher J, Vuorio E, Horne WC, Baron R (2013) Osteoclast-specific cathepsin K deletion stimulates S1Pdependent bone formation. J Clin Invest 123(2):666–681 Keller J, Catala-Lehnen P, Huebner AK, Jeschke A, Heckt T, Lueth A, Krause M, Koehne T, Albers J, Schulze J, Schilling S, Haberland M, Denninger H, Neven M, Hermans-Borgmeyer I, Streichert T, Breer S, Barvencik F, Levkau B, Rathkolb B, Wolf E, Calzada-Wack J, Neff F, Gailus-Durner V, Fuchs H, de Angelis MH, Klutmann S, Tsourdi E, Hofbauer LC, Kleuser B, Chun J, Schinke T, Amling M (2014) Calcitonin controls bone formation by inhibiting the release of sphingosine 1-phosphate from osteoclasts. Nat Commun 21(5), p 5215 Guenet JL, Stanescu R, Maroteaux P, Stanescu V (1981) Fragilitas ossium: a new autosomal recessive mutation in the mouse. J Hered 72(6):440–441 Stoffel W, Jenke B, Block B, Zumbansen M, Koebke J (2005) Neutral sphingomyelinase 2 (smpd3) in the control of postnatal growth and development. Proc Natl Acad Sci USA 102(12):4554–4559 Stoffel W, Jenke B, Holz B, Binczek E, Gunter RH, Knifka J, Koebke J, Niehoff A (2007) Neutral sphingomyelinase (SMPD3) deficiency causes a novel form of chondrodysplasia and dwarfism that is rescued by Col2A1-driven smpd3 transgene expression. Am J Pathol 171(1):153–161 Hannun YA, Obeid LM (2002) The Ceramide-centric universe of lipid-mediated cell regulation: stress encounters of the lipid kind. J Biol Chem 277(29):25847–25850 Li Z, Wu G, Sher RB, Khavandgar Z, Hermansson M, Cox GA, Doschak MR, Murshed M, Beier F, Vance DE (2014) Choline kinase beta is required for normal endochondral bone formation. Biochim Biophys Acta 1840(7):2112–2122 Kakoi H, Maeda S, Shinohara N, Matsuyama K, Imamura K, Kawamura I, Nagano S, Setoguchi T, Yokouchi M, Ishidou Y, Komiya S (2014) Bone morphogenic protein (BMP) signaling upregulates neutral sphingomyelinase 2 to suppress chondrocyte maturation via the Akt protein signaling pathway as a negative feedback mechanism. J Biol Chem 289(12):8135–8150 Holland WL, Brozinick JT, Wang LP, Hawkins ED, Sargent KM, Liu Y, Narra K, Hoehn KL, Knotts TA, Siesky A, Nelson DH, Karathanasis SK, Fontenot GK, Birnbaum MJ, Summers SA (2007) Inhibition of ceramide synthesis ameliorates glucocorticoid-, saturated-fat-, and obesity-induced insulin resistance. Cell Metab 5(3):167–179 Yadav MC, Simao AM, Narisawa S, Huesa C, McKee MD, Farquharson C, Millan JL (2011) Loss of skeletal mineralization by the simultaneous ablation of PHOSPHO1 and alkaline phosphatase function: a unified model of the mechanisms of initiation of skeletal calcification. J Bone Miner Res 26(2):286–297 Brophy PJ, Choy PC, Toone JR, Vance DE (1977) Choline kinase and ethanolamine kinase are separate, soluble enzymes in rat liver. Eur J Biochem 78(2):491–495
Sphingolipid and skeleton 72. Gallego-Ortega D, Ramirez de Molina A, Ramos MA, ValdesMora F, Barderas MG, Sarmentero-Estrada J, Lacal JC (2009) Differential role of human choline kinase alpha and beta enzymes in lipid metabolism: implications in cancer onset and treatment. PLoS One 4(11):e7819 73. Anderson HC (2003) Matrix vesicles and calcification. Curr Rheumatol Rep 5(3):222–226 74. Wu LN, Genge BR, Kang MW, Arsenault AL, Wuthier RE (2002) Changes in phospholipid extractability and composition accompany mineralization of chicken growth plate cartilage matrix vesicles. J Biol Chem 277(7):5126–5133 75. Mebarek S, Abousalham A, Magne D, le Do D, BandorowiczPikula J, Pikula S, Buchet R (2013) Phospholipases of mineralization competent cells and matrix vesicles: roles in physiological and pathological mineralizations. Int J Mol Sci 14(3):5036–5129
969 76. Coleman RM, Aguilera L, Quinones L, Lukashova L, Poirier C, Boskey A (2012) Comparison of bone tissue properties in mouse models with collagenous and non-collagenous genetic mutations using FTIRI. Bone 51(5):920–928 77. Lipina C, Hundal HS (2011) Sphingolipids: agents provocateurs in the pathogenesis of insulin resistance. Diabetologia 54(7):1596–1607 78. Samad F, Badeanlou L, Shah C, Yang G (2011) Adipose tissue and ceramide biosynthesis in the pathogenesis of obesity. Adv Exp Med Biol 721:67–86 79. Yang G, Badeanlou L, Bielawski J, Roberts AJ, Hannun YA, Samad F (2009) Central role of ceramide biosynthesis in body weight regulation, energy metabolism, and the metabolic syndrome. Am J Physiol Endocrinol Metab 297(1):E211–E224
123
