INTERNATIONAL JOURNAL OF BIOPHARMACEUTICAL SCIENCES (ISSN:2517-7338)

Sphingolipids represent one of the major classes of bioactive lipids. Studies of Sphingolipids have intensified in the past several years, revealing their roles in nearly all cell biological processes. In addition, epigenetic regulation has gained substantial interest due to its role in controlling gene expression and activity without changing the genetic code. In this review, we first introduce a brief background on sphingolipid biology, highlighting its role in pathophysiology. We then illustrate the concept of epigenetic regulation, focusing on how it affects the metabolism of sphingolipids. We further discuss the roles of bioactive sphingolipids as epigenetic regulators themselves. Overall, a better understanding of the relationship between epigenetics and sphingolipid metabolism may help to improve the development of sphingolipid-targeted therapeutics. 

Sphingolipids; Epigenetics; Therapeutics

Sphingolipid Biology
phingolipid Biology Sphingolipids are bioactive lipids that contain a sphingoid base backbone, they are a class of aliphatic amino alcohols that include sphingosine, and are found in all cellular and subcellular membranes [1]. Sphingolipids can be formed by de novo synthesis or can be created through salvage pathways from existing sphingolipids via various enzymatic pathways, resulting in a large variety of sphingolipids including ceramides and dihydroceramides, phytoceramides, sphingosine and dihydrosphingosine (sphingaine), various phosphorylated sphingolipids, sphingomyelins, and many glycosphingolipids such as gangliosides, cerebrosides, and sulfatides (Figure 1). For example, ceramide can be deacetylated to form sphingosine, and subsequently phosphorylated to form Sphingosine-1-Phosphate (S1P), which can have dramatic impacts to cellular fate. Alternatively, ceramide can be glycosylated to form glycosphingolipids or acquire a phosphocholine group to form sphingomyelin. The synthesis and degradation of sphingolipids are finely tuned and closely regulated to maintain cellular homeostasis. On the cell surface, they can function as adhesion sites for extracellular proteins and serve both as structural lipids and signaling molecules [2]. In 1996, the “Sphingolipid rheostat” model was proposed, hypothesizing that ceramide and S1P differentially regulate cellular signaling pathways involved in cellular death and proliferation/ survival, respectively. Signaling molecules including growth factors, cellular stress and inflammatory mediators may alter the balance between ceramide and S1P to control cell fate [3].

Pathophysiological Roles of Sphingolipids
Structurally diverse Sphingolipids are bioactive and have roles in multiple biological processes, including cell growth, cell death, cell differentiation, inflammation, apoptosis and angiogenesis [4]. Consequently, abnormalities in sphingolipid homeostasis are implicated in aberrant pathophysiological processes such as metabolic disorders, cancers, inflammation, neurological syndromes and cardiovascular disease [4-8].

Metabolic disorders: Sphingolipids influence glucose metabolism in a variety of tissues. Changes in sphingolipid levels have consequential pathophysiological effects in metabolic syndromes [6-8]. It has been observed that diabetic patients have elevated plasma ceramide levels, which improve following Roux-en-Y bypass surgery and correlates to weight loss and restoration of insulin sensitivity [6]. Furthermore, increased levels of ceramide results in attenuation of insulin action, most likely via inhibition of AKT [4,7,8]. In a similar manner, other sphingolipids, including sphingomyelin, glucosylceramide, and other glycosphingolipids, have been shown to inhibit insulin action [8-10]. Beyond its ability to blunt insulin signaling, ceramide has also been shown to cause beta islet apoptosis through a process known as lipotoxicity and to reduce the expression of the insulin mRNA transcript [6-8]. On the contrary, extracellular S1P carried by high-density lipoprotein can increase insulin levels by protecting pancreatic islet cells from apoptosis [11].

Cancer: Sphingolipids have a strong influence in the development and progression of cancer and are clearly implicated in cell death, proliferation, and multi-drug resistance. Ceramide can function as a tumor suppressor lipid by potentiating signaling events that drive apoptosis, autophagic responses and cell cycle arrest [12]. Increases in total plasma ceramide during a course of radiation therapy correlated with disease response in a phase II trial of metastatic colorectal cancer [13]. However, defects in ceramide generation and metabolism in cancer cells contribute to tumor cell survival and resistance to chemotherapy likely by neutralizing ceramide or generating pro-survival and mitogenic sphingolipids such as S1P [12]. Increased S1P signaling via S1P receptor (S1PR) 1 and 2 in endothelial cells are implicated in angiogenesis and inflammatory pathways, potentially contributing to tumor growth, invasion and metastasis [4,7].

Inflammation: During inflammation, immune cells are recruited to sites of infection or injury and produce a cytokine response necessary to protect the host. However, unchecked inflammation can lead to numerous pathophysiological states such as autoimmune disorders. Recently, it has been demonstrated that circulating S1P plays a critical role in lymphocyte egress [14,15]. S1P is enriched in lymph and blood compared to interstitial fluids. Furthermore, the majority (~65%) of plasma S1P is complexed with Apolipoprotein M (ApoM) [16]. ApoM–S1P is dispensable for lymphocyte trafficking yet restrains lymphopoiesis by activating S1PR1 on bone marrow lymphocyte progenitors. Thus, the signaling axis of ApoM–S1P–S1PR1 regulates the lymphocyte emigration from draining lymph nodes and adaptive immune responses [16]. High levels of S1P have also been found in patients with rheumatoid arthritis and with ulcerative colitis [17,18]. Altogether, these roles for S1P signaling in inflammatory and immune disease have led to clinical trials and approval of various S1PR modulators [17]. This is highlighted by approval of the S1PR1 modulator GilenyaTM (FTY720, fingolimod) for the treatment of relapsing-remitting multiple sclerosis, which works to sequester lymphocytes and prevent autoimmune-mediated damage to myelin in the central nervous system [19].

Neurological Disease: In humans, sphingolipid concentrations are highest in the nervous system where their dysfunction contributes to neurological diseases. Key sphingolipid species are critical for the normal development and function of the brain. Gangliosides, which are heavily glycosylated sphingolipids, are major components of neuronal membranes and account for 10-12% of the lipid content in the nervous system [20]. There is evidence that gangliosides can contribute to the initiation and progression of Alzheimer’s disease by facilitating plaque formation [21]. Conversely, defective ganglioside biosynthesis is associated with the development of epilepsy through a loss-of-function mutation in the monosialodihexosylganglioside 3 (GM3) synthase gene [22].

Cardiovascular Disease: Recent data suggests that specific ceramide species are linked to cardiovascular disease [23]. Higher plasma levels of sphingomyelin have been associated with increased atherosclerosis and have been proposed as independent risk factors for coronary heart disease in humans [24]. Similarly, plasma ceramides have found utilization in predicting cardiovascular death in patients with stable coronary artery disease and acute coronary syndromes [25]. Inhibition of de novo ceramide synthesis has been explored as a post-ischemic strategy to reduce myocardial reperfusion injury [26]. In contrast, plasma S1P is believed to be cardioprotective. Low S1P levels have been associated with impaired cell signaling and ischemic heart disease, although the exact relation between the two has not been conclusively addressed [27].

Sphingolipidoses: Several inherited disorders exist effecting sphingolipid catabolic pathways, and result in the aberrant cellular accumulation of sphingolipids. These disorders are often referred to as lipid storage disorders, and can result in severe pathological manifestations that may include developmental, neurological, hematological, cardiac, and immunological effects. Sphingolipid storage diseases can result from defects influencing  acid ceramidase (Farber) [28], acid sphingomyelinase or certain lipid transporters (Niemann-Pick) [29], acid glucocerebrosidase (Gaucher) [30], galactosylcerebrosidase (Krabbe) [31], alpha galactosidase A (Fabry) [32], beta-hexosaminidase A (Tay-Sachs) [33], or arylsulfatase A (metachromatic leukodystrophy) [34], as well as various other glycosphingolipid pathways. These sphingolipidoses are recognized as Mendelian disorders, and epigenetic influences have generally not been recognized or characterized. However, recent evidence suggests that epigenetic factors may account for some phenotypic differences that present between patients with the same disorders [35]. Current treatment approaches for many of these sphingolipidoses utilize either substrate reduction therapy or enzyme replacement therapy [28- 35].

The human genome is an expanse of genomic loci that contain protein-coding genes and their regulatory elements. Genes are only actively expressed when they are accessible to regulatory factors and transcriptional machinery, otherwise they are suppressed within compact and inaccessible structures [36]. Epigenetics describes inherited traits that are functionally relevant changes to the genome which cannot be identified by changes in the nucleotide sequence (Figure 2). Covalent modifications to histones and DNA without altering the DNA nucleotide sequence facilitate gene expression patterns [37]. These changes can be maintained through multiple cell divisions or may persist for multiple generations as non- genetic factors leading to quasi-permanent changes in gene expression [38]. Epigenetic regulation of gene expression is essential for normal development and cellular function. Changes in chromatin configuration may cause a wide range of diseases by altering the activity of specific genes [39]. Currently, three major mechanisms are involved in the epigenetic regulation of gene expression patterns: DNA methylation, histone modification and non-coding RNAs.

DNA Methylation
DNA methylation refers to methylation of cytosine at the 5’-position, converting it to 5- methylcytosine. DNA methylation is mainly catalyzed by three DNA methyltransferases (DNMTs): DNMT3A and DNMT3B primarily establish new sites of methylation whereas DNMT1 maintains existing methylation patterns [40]. Cytosine methylation within DNA sequences that serve as recognition sites for transcription factors can directly interfere with their binding or recruit transcriptional repressor complexes, which can repress transcription of genes. Most of our DNA is heavily methylated, thus DNA methylation is critical for silencing of transcription and resulting in changes to numerous cellular processes [41]. Abnormalities in DNA methylation play a role in pathological processes such as carcinogenesis and oncogenesis. As a result, DNMT inhibitors are being explored as potential cancer therapies. These inhibitors are cytidine analogues and work by incorporating into replicating DNA, covalently binding to the catalytic sites, and inhibiting the enzymatic activities of DNMTs [41]. To date, two DNMT inhibitors - azacitidine (also known as 5-azacytidine) and decitabine (also known as 5-aza-2ʹ-deoxycytidine), have been approved by the FDA for the clinical treatment of myelodysplastic syndrome [42].

Histone Modification
Histones are the chief protein components of chromatin that act as spools for DNA. Histones affect gene transcription by changing the accessibility of chromatin to transcription factors and gene expression machinery. They are subject to a wide range of modifications, including methylation, acetylation, phosphorylation, and ubiquitination [43]. Most studies regarding transcriptional changes with respect to histone modification have focused on acetylation and methylation [44]. Histone acetylation primarily occurs at lysine residues H3K18 and H3K27. The acetylation alters the net charge of the histone, weakening its interaction with DNA, leading to a more open chromatin state. The enzymes responsible for this type of modification are Histone Acetyl Transferases (HATs) and are antagonized by Histone Deacetylases (HDACs). Over expression of HDACs are observed in many cancer types, including prostate, gastric, and endometrial cancers [44]. Currently, three HDAC inhibitors are approved by the FDA: vorinostat (also known as Suberanilohydroxamic Acid or SAHA) for the treatment of cutaneous T cell lymphoma [44], belinostat for the treatment of peripheral T cell lymphoma[45] and romidepsin for the treatment of both cutaneous and peripheral T-cell lymphoma [46]. Histone methylation occurs on both lysine and arginine residues. Methylation of residue H3K4 acts to enhance gene expression whereas methylation of residues H3K9 or H3K27 represses gene expression. Methylations of histones are catalyzed by histone methyltransferases and are removed by histone demethylases. Specifically, methylation of H3K27 is regulated by Polycomb-group proteins (PcG), which are important in many aspects of development like homeotic gene regulation and X chromosome inactivation [47,48]. The catalytic enzyme in the PcG, EZH2, is upregulated in numerous cancers. Currently, inhibitors of several histone methyltransferases are in clinical trials, such as inhibitors of EZH2 and DOT1L, although none have been approved [49].

Non-coding RNAs
Non-coding RNA transcripts, including microRNAs (miRNAs), piwi-interacting RNAs (piRNAs), circular RNAs (circRNAs), and long noncoding RNAs (lncRNAs), have a profound epigenetic influence on transcriptional output of the genome [50]. Approximately 50% of miRNA genes are repressed by epigenetic methylation of their own promoter while other miRNAs are derived from spliced intronic sequences of coding mRNA [50]. After being processed into mature miRNA and incorporated into the RNA-Induced Silencing Complex (RISC) miRNAs bind to the 3’- UTR of target mRNAs for degradation. A single miRNA is capable of binding and degrading multiple mRNAs [51]. Like miRNAs, piRNA associate with the RISC complex and function as a gene silencing mechanism, especially by silencing transposons. Originally, piRNAs were thought to be expressed exclusively in germline tissue and only play a biological role in development. Recent evidence has implicated piRNA to be a functional epigenetic modifier in both germ and somatic stem cell self-renewal as well as tumorgenesis [51]. Conversely, ciRNA antagonize gene silencing by acting as miRNA “sponges” or as a decoy to miRNA’s normal target genes, and therefore can play a key role in disease progression, especially cancer [52]. Lastly, lncRNA have the most dynamic range of epigenetic regulatory mechanisms that have been characterized to date. This includes functional roles in chromatin remodeling, X-chromosome inactivation, RNA splicing, genomic imprinting and DNA methylation [40,52-55].

Beyond being lipid components of cellular membranes, sphingolipids mediate signaling functions in both physiological and pathological processes. Multiple factors including development, environmental chemicals, drugs and diet affect sphingolipid homeostasis through epigenetic mechanisms [38,56].

Development
Gangliosides are a class of glycosylated sphingolipids with one or more sialic acids bound to a carbohydrate and are particularly abundant in the nervous system [57]. During brain development gangliosides shift from simple to complex species through a process mainly regulated by the staged expression of genes that encode the glycosyltrasferase enzymes [58]. In mice, developmental alteration of gangliosides is regulated through histone acetylation of the glycosyltransferase enzymes. Furthermore, HDAC inhibitors can effectively change the expression patterns of gangliosides and glycosyltransferases in neuroepithelial cells [59].

Environmental Factors
Hypoxia occurs from stress at high altitude or an otherwise oxygen-depleted environment. In tissues and cells, hypoxia can occur when cellular oxygen demand exceeds that which can be supplied often during times of limited blood flow [60]. During hypoxia, hypoxia-inducible transcription factor-1α (HIF-1α) is expressed to enhance energy metabolism and autophagy. Recent studies found that hypoxia induces the expression of HIF 1α through increased activity of sphingosine kinase (SPHK). This effect can be blocked by administration of a SPHK inhibitor or S1PR antagonists [61]. Stabilization of HIF-1α further increases the expression of distinct histone lysine demethylases to promote gene expression [60]. Thus, sphingolipid metabolism is closely intertwined with epigenetic changes in chromatin during hypoxia. Similarly, cigarette smoke has long been proven to induce epigenetic modifications such as aberrant methylation [62]. It is also the major cause of chronic obstructive pulmonary disease (COPD) where alveolar macrophages are defective in their ability to phagocytose apoptotic cells [63]. Cigarette smoke extract significantly increases expression of S1PR5. In alveolar macrophages from COPD patients, the expression levels of S1PR5 is significantly increased and is correlated with defective efferocytosis [64]. Furthermore, this dysregulation of the S1P signaling system is caused by reduced DNA methylation of the S1PR5 gene. In this way, decreased promoter methylation releases gene transcription of S1PR5, which is essential for S1P-mediated clearance of apoptotic cells [64].

Drugs
Neutral Sphingomyelinase-2 (nSMase2) is a key ceramideproducing enzyme in cellular stress responses. Tretinoin, also known as All-Trans Retinoic Acid (ATRA), is used in the treatment of Acute Promyelocytic Leukemia (APL) [65,66]. Although the mechanism of action is still unknown, studies have demonstrated that tretinoin encourages progenitor cells in APL patients to differentiate terminally which thereby ameliorates the progression of the disease [65,66]. A recent study identified nSMase2 as an early ATRA-induced gene and implicated nSMase2 as having a role in growth arrest after ATRA treatment [67]. Furthermore, ATRA regulates nSMase2 directly through modulation of histone acetylation in a CREB-binding protein- and p300-dependent manner [67].

Mesenchymal stem cells (MSCs) are multipotent progenitor cells that can be used to improve tissue repair. Current MSC therapies are often primed with bioactive lipids such as S1P and ceramide-1-phosphate, which can enhance in vivo engraftment and survival of transplanted cells [68,69]. A recent study found that the HDAC inhibitor Valproic Acid (VPA) enforced the priming effect of S1P at a low dosage in human umbilical cord-derived MSCs through the activation of the MAPK and AKT signaling pathways [70]. This study indicates that the combinatory effects of epigenetic modulators and S1P-priming strategies can improve the therapeutic potency of stem cells. Similarly, VPA induces the generation of glycosphingolipid GM3, a known suppressor of epidermal growth factor receptor phosphorylation, to suppress cancer cell proliferation [71]. Additionally, another HDAC inhibitor, AR-42, exerts an anti-colon cancer activity partially through de novo ceramide production [72].

Toxins
Fumonisin B1 (FB1) is toxin that comes from the Fusarium family of molds, which commonly are found in maize and wheat-based foods and feed products [73-76]. Hepatotoxicity, nephrotoxicity and carcinogenicity can occur following dietary exposure to FB1, which is a well-recognized inhibitor of de novo ceramide synthesis due to its ability to inhibit (dihydro) ceramide synthase activity [73-76]. Therefore, exposure to FB1 mostly results in the accumulation of dihydrosphingosine (sphinganine) and dihydrosphingosine-1-phosphate (dhS1P). Interestingly, it has been shown that FB1 exposure may lead to liver tumorigenesis by disrupting DNA methylation and chromatin modifications which causes chromatin instability in HepG2 cells [73,74]. This indicates that FB1 may act as an epigenetic regulator, or alternatively, either dihydrosphingosine (sphinganine) or dhS1P. Importantly, studies have shown that FB1-breakdown products do not exert the same toxicities or alterations to sphingolipid metabolism [75,76]. This reinforces that FB1, or the sphingolipids it regulates, are specifically involved in its mechanisms of toxicity.

As mentioned, bioactive sphingolipids serve both as structural lipids and signaling molecules. Aside from prominent roles associated with localization at the plasma membrane, sphingolipids and their metabolizing enzymes are also found in the nucleus, and associated with the nuclear matrix and chromatin. Their metabolism in the nucleus is linked to the remodeling of chromatin and epigenetic regulation of gene expression [77].

S1P
S1P is the first identified endogenous nuclear regulator of HDACs [78]. In human MCF-7 breast cancer cells, nuclear S1P was found to specifically bind to HDAC1 and HDAC2 and inhibited their enzymatic activity [78]. Another study in dystrophic mice revealed that expression of S1P increased histone acetylation of specific genes, in which inflammatory genes are down regulated while metabolic genes are upregulated, helping muscle cells maintain energy metabolism [79]. Additionally, in a drosophila model of Duchenne muscular dystrophy, S1P was found to increase the capacity of the muscle cell to use fatty acids as an energy source by inhibiting the acetylation of specific histone residues [79]. Thus, nuclear S1P is directly involved in epigenetic regulation of gene expression by modulating histone acetylation. In a somewhat contrasting manner, the degradation of S1P by the S1P lyase has been attributed to the regulation of sepsis-mediated inflammation. This is attributed to S1P degradation products serving as regulators of histone acetylation [80].

Ceramide 
Inhibitor 2 of Protein Phosphatase 2A (I2PP2A) is a biological inhibitor of the cellular serine/threonine protein phosphatase PP2A [81,82]. I2PP2A has been suggested to be oncogenic and is overexpressed in many tumor cell types. I2PP2A also has other targets besides PP2A, e.g., DNA exonucleases and the modification of histone acetylation [82]. Ceramide could potentially disrupt the association between PP2A and I2PP2A. Additionally, ceramide inhibits I2PP2A’s upregulation of c-Myc and its downregulation of histone acetylation in prostate cancer cells, blocking the epigenetic action of I2PP2A [82]. Conversely, in Madin- Darby canine kidney cells, sphingomyelin-derived ceramide has been shown to inhibit the deacetylation of microtubules which induces formation of a ciliogenic lipid-protein complex that can sustain primary cilia [83]. This may not be directly epigenetic in nature, but involves the regulation of a post-translational modification that also plays a role in epigenetics.

Gangliosides
Gangliosides are expressed primarily on the outer leaflet of the plasma membrane of cells in all vertebrates, and are most abundant in the nervous tissues [84]. The ganglioside GM1 is associated with gene regulation in neuronal cells through an interaction with active chromatin via acetylated histones, binding acetylated histones H3 and H4 on the promoters of the GalNAcT and NeuroD1 genes [84,85]. This interaction activates neuronal differentiation in neural stem cells, and can be promoted by exogenous supplementation of ganglioside GM1 [84,85].

Sulfatide
ulfatide Sulfatide also is a galactosphingolipid mostly known for its role in the myelin sheath. In hepatocellular carcinoma cells, it suppresses miR-223 expression in association with reduced recruitment of acetylated histone H3 and C/EBPα to the premiR-223 gene promoter [86]. Furthermore, sulfatide also stimulates the expression of histone deacetylases HDAC9 and HDAC10, and enhances their recruitment to the miR-223 gene promoter [86].

Sphingolipid metabolism and epigenetic modulation are closely intertwined with one another. Epigenetic regulation is an important regulatory mechanism for controlling the expression pattern of sphingolipids. Conversely, sphingolipids in the nucleus can directly modify transcriptional activity via epigenetic modulation. An understanding of both processes is critical. Sphingolipids are integral in both normal development and the pathological processes leading to disease [4]. A deeper investigation in the context of the pathophysiology of disease development of the sphingolipid metabolic influence on epigenetic regulation, and the reciprocal relationship, can potentially lead to new diagnostics and therapeutic targets. Future studies of the relationships between epigenetics and sphingolipid metabolism should address the following questions: 1) what other sphingolipids directly influence epigenetic regulation, 2) how aberrant epigenetic regulation and sphingolipid metabolism directly contribute to the development of cancer and other diseases, and 3) what are the mechanisms that dictate the interplay between epigenetic and sphingolipid metabolic regulation? Investigation of these questions will undoubtedly yield results that inform a new generation of molecular and sphingolipid-targeted therapies.

This review was funded in part by the National Cancer Institute of the National Institutes for Health under award number K22- CA190674 (B.M.B). 

1. Barth BM, Cabot MC, Kester M. Ceramide-based therapeuticsfor the treatment of cancer. Anticancer Agents Med Chem. 2011Nov;11(9):911-919.
2. Iqbal J, Walsh MT, Hammad SM, Hussain MM. Sphingolipidsand Lipoproteins in Health and Metabolic Disorders. TrendsEndocrinol Metab. 2017 Jul;28(7):506-518.
3. Evangelisti C, Evangelisti C, Buontempo F, Lonetti A, Orsini E,et al. Therapeutic potential of targeting sphingosine kinasesand sphingosine 1-phosphate in hematological malignancies.Leukemia. 2016 Nov;30(11):2142-2151.
4. Hannun YA, Obeid LM. Sphingolipids and their metabolism inphysiology and disease. Nat Rev Mol Cell Biol. 2017 Nov.
5. Schneider-Schaulies J, Schneider-Schaulies S. Sphingolipids inviral infection. Biol Chem. 2015 Jun;396(6-7):585-595.
6. Borodzicz S, Czarzasta K, Kuch M, Cudnoch-JedrzejewskaA. Sphingolipids in cardiovascular diseases and metabolicdisorders. Lipids Health Dis. 2015 Jun;14:55.
7. Hla T, Dannenberg AJ. Sphingolipid signaling in metabolicdisorders. Cell Metab. 2012 Oct;16(4):420-434.
8. Meikle PJ, Summers SA. Sphingolipids and phospholipids ininsulin resistance and related metabolic disorders. Nat RevEndocrinol. 2017 Feb;13(2):79-91.
9. Chavez JA, Siddique MM, Wang ST, Ching J, Shayman JA,et al. Ceramides and glucosylceramides are independentantagonists of insulin signaling. J Biol Chem. 2014 Jan10;289(2):723-734.
10. Aerts JM, Boot RG, van Eijk M, Groener J, Bijl N, et al.Glycosphingolipids and insulin resistance. Adv Exp Med Biol.2011;721:99-119.
11. Rütti S, Ehses JA, Sibler RA, Prazak R, Rohrer L, et al. Low- andhigh-density lipoproteins modulate function, apoptosis, andproliferation of primary human and murine pancreatic beta-cells.Endocrinology. 2009 Oct;150(10):4521-4530.
12. Morad SA, Cabot MC. Ceramide-orchestrated signalling incancer cells. Nat Rev Cancer. 2013 Jan;13(1):51-65.
13. Dubois N, Rio E, Ripoche N, Ferchaud-Roucher V, Gaugler MH, etal. Plasma ceramide, a real-time predictive marker of pulmonaryand hepatic metastases response to stereotactic body radiationtherapy combined with irinotecan. Radiother Oncol. 2016 May;119(2):229-235. 
14. Cyster JG, Schwab SR. Sphingosine-1-phosphate and lymphocyteegress from lymphoid organs. Annu Rev Immunol. 2012;30:69-94.
15. Schwab SR, Cyster JG. Finding a way out: lymphocyte egress fromlymphoid organs. Nat Immunol. 2007 Dec;8(12):1295-301.
16. Blaho VA, Galvani S, Engelbrecht E, Liu C, Swendeman SL, et al.HDL-bound sphingosine-1-phosphate restrains lymphopoiesisand neuroinflammation. Nature. 2015 Jul;523(7560):342-346.
17. Maceyka M, Spiegel S. Sphingolipid metabolites in inflammatorydisease. Nature. 2014 Jun;510(7503):58-67.
18. Kitano M, Hla T, Sekiguchi M, Kawahito Y, Yoshimura R, et al.Sphingosine 1-phosphate/sphingosine 1-phosphate receptor1 signaling in rheumatoid synovium: regulation of synovialproliferation and inflammatory gene expression. ArthritisRheum. 2006 Mar;54(3):742-753.
19. Zécri FJ. From Natural Product to the First Oral Treatment forMultiple Sclerosis: The Discovery of FTY720 (Gilenya™)? CurrOpin Chem Biol. 2016 Jun;32:60-66.
20. Posse de Chaves E, Sipione S. Sphingolipids and gangliosidesof the nervous system in membrane function and dysfunction.FEBS Lett. 2010 May;584(9):1748-1759.
21. Olsen ASB, Færgeman NJ. Sphingolipids: membranemicrodomains in brain development, function and neurologicaldiseases. Open Biol. 2017 May;7(5).
22. Fragaki K, Ait-El-Mkadem S, Chaussenot A, Gire C, Mengual R, etal, Desquiret-Dumas V. Refractory epilepsy and mitochondrialdysfunction due to GM3 synthase deficiency. Eur J Hum Genet.2013 May;21(5):528-534.
23. Chen X, Sun A, Zou Y, Ge J, Lazar JM, et al. Impact of sphingomyelinlevels on coronary heart disease and left ventricular systolicfunction in humans. Nutr Metab (Lond). 2011 Apr;8(1):25.
24. Jiang XC, Paultre F, Pearson TA, Reed RG, Francis CK, et al. Plasmasphingomyelin level as a risk factor for coronary artery disease.Arterioscler Thromb Vasc Biol. 2000 Dec;20(12):2614-2618.
25. Laaksonen R, Ekroos K, Sysi-Aho M, Hilvo M, Vihervaara T, etal. Plasma ceramides predict cardiovascular death in patientswith stable coronary artery disease and acute coronarysyndromes beyond LDL-cholesterol. Eur Heart J. 2016 Jul1;37(25):1967-1976.
26. Reforgiato MR, Milano G, Fabriàs G, Casas J, Gasco P, et al.Inhibition of ceramide de novo synthesis as a postischemicstrategy to reduce myocardial reperfusion injury. Basic ResCardiol. 2016 Mar;111(2):12.
27. Polzin A, Piayda K, Keul P, Dannenberg L, Mohring A, et al. Plasmasphingosine-1-phosphate concentrations are associated withsystolic heart failure in patients with ischemic heart disease. JMol Cell Cardiol. 2017 Sep;110:35-37.
28. Schuchman EH. Acid ceramidase and the treatment of ceramidediseases: The expanding role of enzyme replacement therapy.Biochim Biophys Acta. 2016 Sep;1862(9):1459-1471.
29. Schuchman EH, Desnick RJ. Types A and B Niemann-Pick disease.Mol Genet Metab.2017 Jan - Feb;120(1-2):27-33.
30. Barth BM, Shanmugavelandy SS, Tacelosky DM, Kester M, MoradSA, et al. Gaucher’s disease and cancer: a sphingolipid perspective.Crit Rev Oncog. 2013;18(3):221-234.
31. Wenger DA, Rafi MA, Luzi P. Krabbe disease: One Hundred yearsfrom the bedside to the bench to the bedside. J Neurosci Res.2016 Nov;94(11):982-929.
32. Kubo T. Fabry disease and its cardiac involvement. J Gen Fam Med. 2017 May;18(5):225-229.
33. Patterson MC. Gangliosidoses. Handb Clin Neurol. 2013;113:1707-1708.
34. Van Rappard DF, Boelens JJ, Wolf NI. Metachromatic leukodystrophy: Disease spectrum and approaches fortreatment. Best Pract Res Clin Endocrinol Metab. 2015Mar;29(2):261-273. 
35. Hassan S, Sidransky E, Tayebi N. The role of epigenetics inlysosomal storage disorders: Uncharted territory. Mol GenetMetab. 2017 Nov;122(3):10-18.
36. Mann BS, Johnson JR, Cohen MH, Justice R, Pazdur R. FDAapproval summary: vorinostat for treatment of advanced primarycutaneous T-cell lymphoma. Oncologist. 2007 Oct;12(10):1247-1252.
37. Ronnekleiv-Kelly SM, Sharma A, Ahuja N. Epigenetic therapy andchemosensitization in solid malignancy. Cancer Treat Rev. CancerTreat Rev. 2017 Apr;55:200-208.
38. McCarrey JR. EPIGENETICS. The epigenome--a family affair.Science. 2015 Nov;350(6261):634-635.
39. Brookes E, Shi Y. Diverse epigenetic mechanisms of humandisease. Annu Rev Genet. 2014;48:237-268.
40. Ahuja N, Sharma AR, Baylin SB. Epigenetic Therapeutics: A NewWeapon in the War Against Cancer. Annu Rev Med. 2016;67:73-89.
41. Azad N, Zahnow CA, Rudin CM, Baylin SB. The future of epigenetictherapy in solid tumours--lessons from the past. Nat Rev ClinOncol. 2013 May;10(5):256-266.
42. Pfister SX, Ashworth A. Marked for death: targeting epigenetic changes in cancer. Nat Rev Drug Discov. 2017 Apr;16(4):241-263
43. Tan M, Luo H, Lee S, Jin F, Yang JS, et al. Identification of 67 histonemarks and histone lysine crotonylation as a new type of histonemodification. Cell. 2011 Sep;146(6):1016-1028.
44. Tough DF, Tak PP, Tarakhovsky A, Prinjha RK. Epigenetic drugdiscovery: breaking through the immune barrier. Nat Rev DrugDiscov. 2016 Dec;15(12):835-853.
45. Campbell P, Thomas CM. Belinostat for the treatment of relapsedor refractory peripheral T-cell lymphoma. J Oncol Pharm Pract.2017 Mar;23(2):143-147.
46. Smolewski P, Robak T. The discovery and development ofromidepsin for the treatment of T-cell lymphoma. Expert OpinDrug Discov. 2017 Aug;12(8):859-873.
47. Kassis JA, Kennison JA, Tamkun JW. Polycomb and TrithoraxGroup Genes in Drosophila. Genetics. 2017 Aug;206(4):1699-1725.
48. Maclary E, Hinten M, Harris C, Sethuraman S, Gayen S, et al.PRC2 represses transcribed genes on the imprinted inactive Xchromosome in mice. Genome Biol. 2017 May;18(1):82.
49. Hock H. A complex Polycomb issue: the two faces of EZH2 incancer. Genes Dev. 2012 Apr;26(8):751-755.
50. Wang Z, Yao H, Lin S, Zhu X, Shen Z, et al. Transcriptional andepigenetic regulation of human microRNAs. Cancer Lett. 2013Apr 30;331(1):1-10.
51. Bamezai S, Rawat VP, Buscke C. Concise review: The Piwi-piRNAaxis: pivotal beyond transposon silencing. Stem Cells. 2012Dec;30(12):2603-2611.
52. Di Ruscio A, Ebralidze AK, Benoukraf T, Amabile G, Goff LA, et al.DNMT1-interacting RNAs block gene-specific DNA methylation.Nature. 2013 Nov;503(7476):371-376.
53. Mercer TR, Dinger ME, Mattic JS. Long non-coding RNAs: insightsinto functions. Nat Rev Genet. 2009 Mar;10(3):155-159.
54. Lee JT, Bartolomei MS. X-inactivation, imprinting andlong noncoding RNAs in health and disease. Cell. 2013Mar;152(6):1308-1323.
55. Rinn JL, Kertesz M, Wang JK, Squazzo SL, Xu X, et al. Functionaldemarcation of active and silent chromatin domains in humanHOX loci by noncoding RNAs. Cell. 2007 Jun;129(7):1311-1323.
56. Wegner MS, Schiffmann S, Parnham MJ, Geisslinger G, Grösch S.The enigma of ceramide synthase regulation in mammalian cells.Prog Lipid Res. 2016 Jul;63:93-119. 
57. Yu RK, Nakatani Y, Yanagisawa M. The role of glycosphingolipidmetabolism in the developing brain. J Lipid Res. 2009 Apr;50Suppl:S440-5.
58. Tsai YT, Yu RK. Epigenetic activation of mouse gangliosidesynthase genes: implications for neurogenesis. J Neurochem.2014 Jan;128(1):101-110.
59. Suzuki Y, Yanagisawa M, Ariga T, Yu RK. Histone acetylationmediated glycosyltransferase gene regulation in mouse braindevelopment. J Neurochem. 2011 Mar;116(5):874-880.
60. Salminen A, Kaarniranta K, Kauppinen A. Hypoxia-InducibleHistone Lysine Demethylases: Impact on the Aging Process andAge-Related Diseases. Aging Dis. 2016 Mar;7(2):180-200.
61. Sanagawa A, Iwaki S, Asai M, Sakakibara D, Norimoto H, et al.Sphingosine 1-phosphate induced by hypoxia increases theexpression of PAI‑1 in HepG2 cells via HIF‑1α. Mol Med Rep. 2016Aug;14(2):1841-1848.
62. Zochbauer-Muller S, Lam S, Toyooka S, Virmani AK, ToyookaKO, et al. Aberrant methylation of multiple genes in the upperaerodigestive tract epithelium of heavy smokers. Int J Cancer.2003 Nov;107(4):612-626.
63. Barnawi J, Jersmann H, Haberberger R, Hodge S, Meech R.Reduced DNA methylation of sphingosine-1 phosphate receptor5 in alveolar macrophages in COPD: A potential link to failedefferocytosis. Respirology. 2017 Feb;22(2):315-321.
64. Barnawi J, Tran H, Jersmann J, Pitson S, Hodge G, et al. Potentiallink between the sphingosine- 1-phosphate (S1P) system anddefective alveolar macrophage phagocytic function in chronicobstructive pulmonary disease (COPD). PLoS One. 2015Oct;10(10):e0122771
65. Abaza Y, Kantarjian H, Garcia-Manero G, Estey E, Borthakur G, etal. Long-term outcome of acute promyelocytic leukemia treatedwith all-trans-retinoic acid, arsenic trioxide, and gemtuzumab.Blood. 2017 Mar 9;129(10):1275-1283.
66. Nitto T, Sawaki K. Molecular mechanisms of the antileukemiaactivities of retinoid and arsenic. J Pharmacol Sci.2014;126(3):179-185.
67. Clarke CJ, Shamseddine AA, Jacob JJ, Khalife G, Burns TA, et al. ATRAtranscriptionally induces nSMase2 through CBP/p300-mediatedhistone acetylation. J Lipid Res. 2016 May;57(5):868-881.
68. Ratajczak MZ, Lee H, Wysoczynski M, Wan W, Marlicz W, et al.Novel insight into stem cell mobilization-plasma sphingosine1-phosphate is a major chemoattractant that directs the egressof hematopoietic stem progenitor cells from the bone marrowand its level in peripheral blood increases during mobilizationdue to activation of complement cascade/membrane attackcomplex. Leukemia. 2010 May;24(5):976-985.
69. Kim C, Schneider G, Abdel-Latif A, Mierzejewska K, Sunkara M,et al. Ceramide-1-phosphate regulates migration of multipotentstromal cells and endothelial progenitor cells--implications fortissue regeneration. Stem Cells. 2013 Mar;31(3):500-510.
70. Lim J, Lee S, Ju H, Kim Y, Heo J, et al. Valproic acid enforcesthe priming effect of sphingosine-1 phosphate on humanmesenchymal stem cells. Int J Mol Med. 2017 Sep;40(3):739-747.
71. Kawashima N, Nishimiya Y, Takahata S, Nakayama KI. Induction ofGlycosphingolipid GM3 Expression by Valproic Acid SuppressesCancer Cell Growth. J Biol Chem. 2016 Oct 7;291(41):21424-21433.
72. Xu W, Xu B, Yao Y, Y u X, Shen J. The novel HDAC inhibitor AR42-induced anti-colon cancer cell activity is associated withceramide production. Biochem Biophys Res Commun. 2015 Aug7;463(4):545-550.
73. Demirel G, Alpertunga B, Ozden S. Role of fumonisin B1 on DNAmethylation changes in rat kidney and liver cells. Pharm Biol.2015;53(9):1302-1310. 74. Chuturgoon A, Phulukdaree A, Moodley D. Fumonisin B1 inducesglobal DNA hypomethylation in HepG2 cells - An alternativemechanism of action. Toxicology. 2014 Jan;315:65-69.
75. Hahn I, Nagl V, Schwartz-Zimmermann HE, Varga E, Schwarz C,et al. Effects of orally administered fumonisin B₁(FB₁), partiallyhydrolysed FB₁, hydrolysed FB₁ and N-(1-deoxy-D-fructos-1-yl)FB₁ on the sphingolipid metabolism in rats. Food Chem Toxicol.2015 Feb;76:11-18.
76. Grenier B, Bracarense AP, Schwartz HE, Trumel C, Cossalter AM, etal. The low intestinal and hepatic toxicity of hydrolyzed fumonisinB₁ correlates with its inability to alter the metabolism ofsphingolipids. Biochem Pharmacol. 2012 May;83(10):1465-1473.
77. Spiegel S, Milstien S, Grant S. Endogenous modulators andpharmacological inhibitors of histone deacetylases in cancertherapy. Oncogene. 2012 Feb;31(5):537-551.
78. Hait NC, Allegood J, Maceyka M, Strub GM, Harikumar KB, et al.Regulation of histone acetylation in the nucleus by sphingosine1-phosphate. Science. 2009 Sep;325(5945):1254-1257.
79. Nguyen-Tran DH, Hait NC, Sperber H, Qi J, Fischer K, et al. Molecularmechanism of sphingosine-1-phosphate action in Duchennemuscular dystrophy. Dis Model Mech. 2014 Jan;7(1):41-54.
80. Ebenezer DL, Fu P, Suryadevara V, Zhao Y, Natarajan V. Epigeneticregulation of pro- inflammatory cytokine secretion bysphingosine 1-phosphate (S1P) in acute lung injury: Role of S1Plyase. Adv Biol Regul. 2017 Jan;63:156-166.
81. Dent P. Ceramide in the prostate. Cancer Biol Ther. 2013Oct;14(10):881-882.
82. Mukhopadhyay A, Tabanor K, Chaguturu R, Aldrich JV.Targeting inhibitor 2 of protein phosphatase 2A as atherapeutic strategy for prostate cancer treatment.Cancer BiolTher. 2013 Oct;14(10):962-972.
83. He Q, Wang G, Dasgupta S, Dinkins M, Zhu G, et al. Characterizationof an apical ceramide-enriched compartment regulatingciliogenesis. Mol Biol Cell. 2012 Aug;23(16):3156-3166.
84. Tsai YT, Itokazu Y, Yu RK. GM1 Ganglioside is involved inEpigenetic Activation Loci of Neuronal Cells. Neurochem Res.2016 Feb;41(1-2):107-115.
85. Itokazu Y, Tsai YT, Yu RK. Epigenetic regulation of gangliosideexpression in neural stem cells and neuronal cells. Glycoconj J.2017 Dec;34(6):749-756. 86. Dong YW, Wang R, Cai QQ, Qi B, Wu W, et al. Sulfatideepigenetically regulates miR-223 and promotes the migrationof human hepatocellular carcinoma cells. J Hepatol. 2014Apr;60(4):792-801.