Plasma Sphingosine-1-Phosphate Is Elevated in Obesity

Abstract

Background

Dysfunctional lipid metabolism is a hallmark of obesity and insulin resistance and a risk factor for various cardiovascular and metabolic complications. In addition to the well known increase in plasma triglycerides and free fatty acids, recent work in humans and rodents has shown that obesity is associated with elevations in the bioactive class of sphingolipids known as ceramides. However, in obesity little is known about the plasma concentrations of sphinogsine-1-phosphate (S1P), the breakdown product of ceramide, which is an important signaling molecule in mammalian biology. Therefore, the purpose of this study was to examine the impact of obesity on circulating S1P concentration and its relationship with markers of glucose metabolism and insulin sensitivity.

Methodology/Principal Findings

Plasma S1P levels were determined in high-fat diet (HFD)-induced and genetically obese (ob/ob) mice along with obese humans. Circulating S1P was elevated in both obese mouse models and in obese humans compared with lean healthy controls. Furthermore, in humans, plasma S1P positively correlated with total body fat percentage, body mass index (BMI), waist circumference, fasting insulin, HOMA-IR, HbA1c (%), total and LDL cholesterol. In addition, fasting increased plasma S1P levels in lean healthy mice.

Conclusion

We show that elevations in plasma S1P are a feature of both human and rodent obesity and correlate with metabolic abnormalities such as adiposity and insulin resistance.

About the Authors:

Greg M. Kowalski

* E-mail: [email protected]

Affiliation: Integrative Physiology and Metabolism Laboratory, Department of Physiology, Monash University, Clayton, Victoria, Australia

Andrew L. Carey

Affiliation: Metabolic and Vascular Physiology Laboratory, Baker IDI Heart and Diabetes Institute, Melbourne, Victoria, Australia

Ahrathy Selathurai

Affiliation: Integrative Physiology and Metabolism Laboratory, Department of Physiology, Monash University, Clayton, Victoria, Australia

Bronwyn A. Kingwell

Affiliation: Metabolic and Vascular Physiology Laboratory, Baker IDI Heart and Diabetes Institute, Melbourne, Victoria, Australia

Clinton R. Bruce

Affiliation: Integrative Physiology and Metabolism Laboratory, Department of Physiology, Monash University, Clayton, Victoria, Australia

Introduction

The global prevalence of obesity is growing at an alarming rate. This is of particular concern especially given that obesity is associated with an array of metabolic complications, including insulin resistance and type 2 diabetes mellitus [1]. In addition to defects in glucose metabolism, these conditions are also closely associated with dyslipidaemia, evident by an increase in circulating lipids including triglycerides and free fatty acids [2]. The dyslipidaemia is thought to arise from an oversupply of both exogenous dietary chylomicron derived lipids along with endogenously synthesized lipids such as those released from the liver (ie. cholesterol, triacylglycerol) and adipose tissue (ie. free fatty acids). Traditionally, elevations of these commonly measured circulating lipids have been used clinically to assess the risk of metabolic and heart disease [3]. While this approach has been useful, recently there has been an emergence of research attempting to identify other circulating lipid species that may prove to be more accurate biomarkers of disease onset and progression [4]–[7]. In this regard, circulating sphingolipids have emerged as potential biomarkers and mediators of disease development [8]–[12].

Sphingolipids are a complex class of lipids that play important structural, signaling and cell recognition roles [13] and have been implicated in the development of metabolic diseases including insulin resistance and type 2 diabetes [4], [8], [9], [12], [14], [15]. Plasma ceramide levels are elevated in type 2 diabetic patients [8], [14] and are inversely associated with measures of insulin sensitivity [8]. Moreover, we recently demonstrated that circulating ceramide is taken up and deposited in skeletal muscle promoting the development of insulin resistance [14]. While we are beginning to develop an appreciation of the importance of circulating ceramide [4], [8], [9], [12], [14], [15], the role of other sphingolipids, in particular sphingosine-1-phosphate (S1P), has received little attention with respect to its relationship with obesity, insulin resistance and type 2 diabetes. S1P is a breakdown product of ceramide metabolism and circulates in mammals at high levels (>200 nM), whereas in the interstitial fluid the levels are very low, thus creating an S1P compartment gradient [16], [17]. It is thought that both haematopoietic and vascular endothelial cells are the major contributors to the plasma S1P pool [16], [18], with approximately 65% of the plasma S1P being bound to lipoproteins, of which ∼50% is associated with HDL, while the remainder is bound to albumin [17], [19]. S1P has effects on many biological processes, with some of the most well characterized effects being on the immune and vascular systems, particularly in the process of inflammation [13], [20]. S1P has both physiological and pathophysiological roles, regulating cell signaling, proliferation, migration, survival, differentiation and metabolism, and its effects can be traced to essentially all organ systems [13], [20]. Interestingly, plasma S1P levels are increased in the genetically obese ob/ob mouse [12] and in rodent models of type 1 diabetes [21]. Furthermore, in vitro studies using C2C12 myoblasts have shown that S1P can stimulate basal glucose uptake by trans-phosphorylation of the insulin receptor [22], while in rat primary adipocytes S1P has been shown to stimulate lipolysis [23]. These findings suggest that S1P levels may either effect or be affected by the metabolic status of the individual. However, there is currently little known about the circulating levels of S1P in obesity and insulin resistance. Therefore, the purpose of this study was to examine the impact of obesity on circulating S1P levels and to determine whether there is a relationship between its concentrations and clinical indices of metabolic dysfunction.

Materials and Methods

Ethics Statement

All animal experiments were approved by the Monash University Animal Research Platform Animal Ethics Committee and were in accordance with the National Health and Medical Research Council of Australia Guidelines on Animal Experimentation. The human study was approved by The Alfred Human Research Ethics Committee, performed in accordance with the Declaration of Helsinki (fifth revision, 2000) and each participant provided written informed consent.

Animal Studies

Male C57BL/6 mice at 8 weeks of age were fed a control chow (∼9% energy as fat, Barastok Rat & Mouse, Ridley AgriProducts, Melbourne, Australia) or high-fat diet (∼42% energy as fat, Specialty Feeds SF4-001, Glen Forrest, WA, Australia) for 6 weeks. Genetically obese male ob/ob mice on a pure C57BL/6 background and lean littermate controls maintained on a standard chow diet were studied at 20 weeks of age. Metabolic characteristics of the mice are presented in Table 1 and 2. Blood collection was performed after a 5 h fast via the tail vein. For the fed and fasting study, 8 week old male C57BL/6 mice on a chow diet had tail vein blood collected at 06∶00 (fed condition) after which food was removed until 18∶00 (fasted state) when blood was again collected for biochemical analysis.

[Figure omitted. See PDF.]

Table 1. Metabolic characteristics of the chow and high-fat diet fed mice.

https://doi.org/10.1371/journal.pone.0072449.t001

[Figure omitted. See PDF.]

Table 2. Metabolic characteristics of the lean and ob/ob mice.

https://doi.org/10.1371/journal.pone.0072449.t002

Human Studies

Fifteen lean and 10 obese (BMI >30 kg/m2) otherwise healthy young men were studied. These individuals were recruited as part of a larger study which has recently been published [24]. The characteristics of all participants are presented in Table 3. Participants arrived at the laboratory following an overnight fast and the physical characteristics (height, weight, waist circumference brachial blood pressure) of each subject were measured. An 18-G cannula was inserted into an antecubital vein and a venous blood sample was obtained for measurement of HbA1c, blood glucose, plasma insulin and lipids. An oral glucose tolerance test (OGTT) was then performed. Subjects ingested a 75 g glucose load and blood was sampled at 60 and 120 min for the measurement of blood glucose. Body composition was measured using dual energy x-ray absorptiometry (DEXA).

[Figure omitted. See PDF.]

Table 3. Participant characteristics.

https://doi.org/10.1371/journal.pone.0072449.t003

Biochemical Analyses

Venous blood samples obtained from the human subjects were centrifuged and plasma immediately frozen. Plasma was subsequently analyzed for total cholesterol, HDL-cholesterol, LDL-cholesterol, triacylglycerol (enzymatic methods, automated analyser), insulin (chemiluminescent microparticle immunoassay) and HbA1c (boronate-affinity HPLC) (all conducted according to clinical diagnostic standards by Alfred Hospital Pathology). For the animal studies, plasma measurements were performed on blood collected from the tail vein and centrifuged at 6000 rpm for 10 min at 4°C. Insulin was measured by ELISA (Linco Research, St. Louis, MO, USA). Plasma non-esterified fatty acids (NEFA) were determined using a colorimetric kit (Wako Pure Chemical Industries, Osaka, Japan). Free plasma glycerol levels were determined using a commercially available kit (Sigma Aldrich, St. Louis, MO, USA). Plasma S1P was measured with a commercially available competitive ELISA designed for plasma/serum determination of S1P concentration (Echelon, Salt Lake City, UT). The S1P ELISA measures total S1P concentrations in plasma/serum without distinction between lipoprotein/albumin bound S1P fractions.

Statistical Analysis

Data are reported as mean ± SEM. Comparisons between data from two groups were made using students two-tailed un-paired or paired (fed/fasting experiments) t-tests. Associations between variables were investigated using Spearman correlation coefficient with GraphPad Prism software. Statistical significance was accepted at P<0.05.

Results

Plasma S1P is Elevated in Mouse Models of Obesity and Insulin Resistance

To determine if circulating S1P levels are altered in obesity and insulin resistance, we studied two commonly used mouse models of obesity and insulin resistance; the chronically HFD fed and genetically obese chow fed ob/ob mouse. The metabolic characteristics of these models are presented in Table 1 and 2. As expected, mice maintained on the HFD displayed key metabolic disturbances associated with obesity such as fasting hyperglycaemia, hyperinsulinaemia, and elevated plasma triglycerides (Table 1). Similarly, the ob/ob mice displayed severe obesity and fasting hyperglycaemia when compared with their lean counterparts (Table 2). Despite the differences in the aetiology of obesity in these mouse models (i.e. one consumes excess fat and sugar calories while the other over-eats a standard laboratory chow diet), both HFD induced and genetically obese ob/ob mice had significantly elevated plasma S1P concentrations (Fig. 1A & B).

[Figure omitted. See PDF.]

Figure 1. Plasma S1P concentrations are elevated in obesity.

Plasma S1P concentrations in chow (N = 13) and HFD (N = 14) fed mice (A); plasma S1P levels in lean (N = 6) and ob/ob (N = 8) mice (B) and circulating S1P in lean (N = 15) and obese (N = 10) humans (C). Data are mean ± SEM. **P<0.01.

https://doi.org/10.1371/journal.pone.0072449.g001

Plasma S1P is Elevated in Obese Humans and Correlates Positively with Indices of Metabolic Dysfunction

To determine whether our findings of elevated plasma S1P concentrations in obese mice were also conserved in humans, we measured the plasma S1P levels in a cohort of well characterised obese (with characteristics of the metabolic syndrome) and lean control subjects. The clinical characteristics of the human subjects are presented in Table 3. Subjects did not differ in age and height and were completely un-medicated. Body mass, BMI, body fat percentage and waist circumference were higher in the obese subjects. While fasting blood glucose levels were normal in the obese individuals, fasting plasma insulin levels and HOMA-IR were elevated compared to the control subjects. Furthermore, obese subjects had significantly higher blood glucose levels at 60 and 120 min during the OGTT, indicating the presence of glucose intolerance and insulin resistance. In relation to plasma lipids, the obese subjects had significantly higher plasma triglycerides and NEFAs and significantly lower HDL cholesterol while there was a non-significant trend for LDL cholesterol (P = 0.059) to be higher in the obese subjects. Consistent with our finding in the rodent models of obesity, plasma S1P levels were also higher (∼28%; P<0.01) in the obese subjects when compared with the lean controls (Fig. 1C). Next, we performed correlation analysis to examine if a relationship exists between the plasma S1P levels and any clinical indices of the metabolic syndrome. Plasma S1P levels were positively correlated with body fat percentage (Fig. 2A), BMI (Fig. 2B), waist circumference (Fig. 2C), fasting plasma insulin (Fig. 2D), HOMA-IR (Fig. 2E), HbA1c (Fig. 2F), total and LDL cholesterol (Fig. 2G & H). Interestingly, we were unable to detect any correlation between plasma S1P levels and HDL cholesterol, NEFAs, triglycerides or glycerol (data not shown).

[Figure omitted. See PDF.]

Figure 2. Plasma S1P concentrations correlate with clinical indices of metabolic dysfunction in humans.

Plasma S1P concentrations correlate with percent body fat (A), BMI (B), waist circumference (C), fasting plasma insulin (D), HOMA-IR (E), HbA1c (F), total (G) and LDL (H) cholesterol.

https://doi.org/10.1371/journal.pone.0072449.g002

Plasma S1P is Increased with Fasting in Mice

Given obesity is associated with increased lipid flux and plasma S1P was elevated in obesity, we next investigated whether an acute fast which increases lipid flux, would also alter plasma S1P concentrations. Plasma S1P levels were measured in both the fed and fasted conditions. Compared to the fed condition, fasting (12 h) resulted in a significant loss of bodyweight (5.5% of starting bodyweight; Fig. 3A) which was paralleled by a significant increase in plasma NEFA levels (Fig. 3B). Interestingly, there was a robust increase (∼60%) in plasma S1P concentration with fasting (Fig. 3C).

[Figure omitted. See PDF.]

Figure 3. Plasma S1P concentrations in mice are elevated in response to fasting.

Changes in body mass (A), blood glucose (B), plasma NEFA (C) and plasma S1P concentrations (D) in response to fasting. Data are mean ± SEM. *P<0.05; **P<0.01, N = 11–12.

https://doi.org/10.1371/journal.pone.0072449.g003

Discussion

Traditionally dyslipidaemia has been characterized by an elevation in plasma triglycerides, total cholesterol and reduction in HDL cholesterol. However, it is becoming increasingly evident that obesity and type 2 diabetes mellitus are characterized by alterations in other circulating lipid species including the bioactive sphingolipid ceramide [4], [8], [14]. Such bioactive lipids may play a more causal mechanistic role in the development of insulin resistance along with other associated complications such as atherosclerosis [8]–[12]. Here we extend those findings by demonstrating that another sphingolipid metabolite, S1P, is significantly elevated in the plasma of obese mice and humans. Furthermore, plasma S1P concentrations in humans positively correlate with clinical indices of metabolic and cardiovascular dysfunction including fasting insulin, HOMA-IR, HbA1c, total and LDL cholesterol, BMI, waist circumference and body fat percentage. We also report that plasma S1P levels in normal healthy mice are robustly increased under acute fasting conditions. Thus it seems that circulating S1P levels are sensitive to changes in metabolic status. Indeed, it is possible that increased lipid flux and/or availability (NEFAs, triglycerides, cholesterol) which are associated with obesity and insulin resistance may govern the production and ultimately the rise in plasma S1P.

It has previously been shown that plasma S1P levels are increased in the genetically obese ob/ob mouse [12]. Here we extend upon these findings by showing that in addition to the obese ob/ob mouse model, plasma S1P is also increased in HFD fed obese mice and more importantly in obese humans. Furthermore, S1P is positively correlated with clinical indices of the metabolic syndrome. However, whether S1P is simply a biomarker of the metabolic syndrome or whether it plays a mechanistic role in the development of obesity related pathologies such as insulin resistance is not known. This is challenging to resolve especially given the complex and sometimes paradoxical roles S1P plays in mammalian biology [13], [20], [25]. In the context of the metabolic syndrome and cardiovascular disease, S1P has been shown to have pro-survival (anti-apoptotic) [26] and anti-inflammatory actions on hepatocytes [27], kidney podocytes [28], protective actions on the cardiovascular system [29], including improvements in blood flow recovery following ischemia [30], and has been implicated in the metabolically protective actions of adiponectin [31]. S1P has also been shown to have anti-inflammatory actions on macrophages [32]. This may be particularly relevant as obesity is characterized by chronic low-grade adipose tissue inflammation, thought to be predominantly mediated by macrophages that infiltrate obese adipose tissue which secrete pro-inflammatory cytokines [33]. However, in contrast to these findings in macrophages, treatment of adipocytes with S1P has been shown to have pro-inflammatory/pro-thrombotic actions, leading to increased gene expression of plasminogen activator inhibitor-1, TNF-α, interleukin-6, and keratinocyte-derived chemokine [12], thus adding further complexity to its role in obesity and metabolic dysfunction. While S1P has been shown to have many protective actions that improve cell survival under pathological stress settings, these same actions, particularly those related to its effects on anti-apoptotic pathways and angiogenesis have however strongly implicated S1P in the development of cancer [34]. Given obesity is associated with increased prevalence of certain cancers [35], it is therefore difficult to ascertain whether the increase in S1P would be beneficial or detrimental in the setting of obesity. Nonetheless, future research should focus on large population studies to determine if there is a relationship between the incidence of cancer and high circulating S1P concentrations.

The findings of our study show that circulating S1P is elevated in obesity, however, we can only speculate regarding the mechanisms. Recently, studies in mice have shown that S1P regulates glucose stimulated insulin secretion in pancreatic β-cells [36], while its clinically approved therapeutic analogue FTY720 (Fingolimod), which is a S1P receptor agonist, normalizes blood glucose levels in type 2 diabetic db/db mice by increasing β-cell mass and circulating insulin levels [37]. S1P has also been shown to increase basal glucose uptake in C2C12 muscle cells [22], however, the consequences of increased plasma S1P on in vivo glucose metabolism are not known and whether S1P affects insulin-stimulated glucose uptake is not clear. In addition, our own recent work has shown that FTY720 treatment reduced muscle ceramide, diacylglycerol and triglyceride levels in HFD induced obese mice, ultimately leading to improved whole body glucose homeostasis [38]. In support of the potentially protective role of S1P in obesity, endurance trained humans, who are remarkably insulin sensitive [39], [40], have been shown to have ∼40% higher plasma S1P concentrations than untrained controls [41]. Similarly, one single exercise bout in untrained human subjects, which is a physiological situation that is accompanied by a simultaneous increase in lipid flux [42] (ie. adipose tissue lipolysis, skeletal muscle lipid uptake and oxidation) and insulin sensitivity [43], increases plasma S1P levels by ∼42%, an effect that is accompanied by decreased erythrocyte ceramide levels [41]. This is somewhat similar to our fasting experiment in mice, which like exercise, would be expected to improve insulin sensitivity while also increasing lipid flux [44], resulting in a significant rise in plasma S1P levels (Fig. 3D). Thus it is plausible that an increase in lipid flux which is evident by elevated rates of lipolysis and NEFA release into the circulation, as seen in obesity [45], type 2 diabetes [46], exercise [42] and fasting [47] may result in increased S1P synthesis and secretion into the plasma. In relation to fasting however, previous research in humans has shown that a 10 h overnight fast did not result in altered S1P concentrations when compared to the fed state [48]. This may possibly be explained by the difference in the relative severity of the fast in mice compared to humans. For example, a 10 h fast in humans has relatively minor effects on hepatic glycogen content [49], [50], while in mice, a 12 h fast results in almost complete depletion of hepatic glycogen [44]. Based on the collective results of the abovementioned studies, we speculate that S1P may elicit potentially protective effects by promoting pancreatic β-cell regeneration/survival during the forced compensatory hyperinsulinaemia that is required to maintain euglycaemia in an insulin resistant obese state. Additionally, based on the fact that we have previously shown that the S1P analogue FTY720 prevents muscle lipid accumulation in HFD mice [38], it is possible that elevated plasma S1P may play a role in trying to minimise excessive bioactive lipid build-up in tissues such as liver, pancreatic β-cells and skeletal muscle, an effect most likely mediated by its ability to signal via the S1P receptors [38]. Although the origin of triacylglycerol and diacylglycerol is distinct to that of ceramide, it suggests that alterations in sphingolipid metabolism can have pronounced effects on numerous lipid classes, including those not directly linked to ceramide. Clearly more work is required to further elucidate the biological role of circulating S1P, particularly in the regulation of pancreatic β-cell function and energy metabolism.

While it is well known that the majority (∼54%) of the plasma S1P pool is bound to HDL cholesterol [17], [19], we did not find an association between plasma S1P levels and the total HDL cholesterol fraction (r = −0.128, P = 0.54). In support of this lack of a correlation with HDL, recent studies by Hammad and colleagues have shown that S1P transport in lipoproteins was not limited by the concentration of HDL cholesterol in individual subjects due to the high variability of S1P content in HDL particles [17]. Additionally, the relative ‘per particle’ S1P content was actually highest in the larger VLDL particles [48], thus possibly explaining our positive correlations between plasma S1P with total and LDL but not HDL cholesterol (Fig. 2G & H). It is also worth noting that some of the cardio-protective functions of HDL have been attributed to its high S1P content [51]. The S1P cargo on HDL particles may be cardio-protective due to its cell survival actions on cardiomyocytes and effects on the vascular endothelium that enhance blood flow/angiogenesis [51], [52].

The source of the increased S1P levels in obesity is unclear and requires further investigation. Given the bulk of the circulating S1P pool comes from erythrocytes, leukocytes and vascular endothelial cells [16], [18], there is a distinct possibility that these cells may also be over-secreting this molecule in obesity. However, the possibility remains that the source could be other cells that are more directly affected by obesity such as adipocytes and/or hepatocytes, although it has been reported that hepatocytes do not secrete S1P, and that hepatic derived S1P is of endothelial origin [18]. Furthermore, the increased plasma S1P levels seen in obesity could also potentially result from decreased cellular/tissue S1P degradation by the cleavage enzyme S1P lyase. If this was to be the case, decreased S1P clearance rather than increased secretion could be the reason circulating S1P is increased in obesity.

In conclusion, we demonstrate that plasma S1P is elevated in obese humans and is positively correlated with clinical features of metabolic dysfunction, particularly fasting insulin, HOMA-IR, HbA1c, total and LDL cholesterol, BMI, waist circumference and body fat percentage. Furthermore, circulating S1P is also elevated in HFD induced and genetically obese ob/ob mice. Additionally, we also show that an acute period of fasting (12 h) in mice robustly increases plasma S1P levels. Collectively these results show that plasma S1P concentrations are altered in association with metabolic status, possibly via an increase in lipid flux and/or availability. Future research should focus on determining the role that circulating S1P may have on nutrient metabolism and whether it contributes to or alleviates metabolic abnormalities such as insulin resistance and defective glucose homoeostasis.

Author Contributions

Conceived and designed the experiments: GK AC BK CB. Performed the experiments: GK AC CB AS. Analyzed the data: GK AC CB. Wrote the paper: GK AC BK CB AS.

Citation: Kowalski GM, Carey AL, Selathurai A, Kingwell BA, Bruce CR (2013) Plasma Sphingosine-1-Phosphate Is Elevated in Obesity. PLoS ONE 8(9): e72449. https://doi.org/10.1371/journal.pone.0072449

References

1. Shaw JE, Sicree RA, Zimmet PZ (2010) Global estimates of the prevalence of diabetes for 2010 and 2030. Diabetes Res Clin Pract 87: 4–14.

2. Boden G (2008) Obesity and free fatty acids. Endocrinol Metab Clin North Am 37: 635–646, viii-ix.

3. Nguyen NT, Magno CP, Lane KT, Hinojosa MW, Lane JS (2008) Association of hypertension, diabetes, dyslipidemia, and metabolic syndrome with obesity: findings from the National Health and Nutrition Examination Survey, 1999 to 2004. J Am Coll Surg 207: 928–934.

4. Barber MN, Risis S, Yang C, Meikle PJ, Staples M, et al. (2012) Plasma lysophosphatidylcholine levels are reduced in obesity and type 2 diabetes. PLoS One 7: e41456.

5. Rhee EP, Cheng S, Larson MG, Walford GA, Lewis GD, et al. (2011) Lipid profiling identifies a triacylglycerol signature of insulin resistance and improves diabetes prediction in humans. J Clin Invest 121: 1402–1411.

6. Schwab U, Seppanen-Laakso T, Yetukuri L, Agren J, Kolehmainen M, et al. (2008) Triacylglycerol fatty acid composition in diet-induced weight loss in subjects with abnormal glucose metabolism–the GENOBIN study. PLoS One 3: e2630.

7. Pietilainen KH, Sysi-Aho M, Rissanen A, Seppanen-Laakso T, Yki-Jarvinen H, et al. (2007) Acquired obesity is associated with changes in the serum lipidomic profile independent of genetic effects–a monozygotic twin study. PLoS One 2: e218.

8. Haus JM, Kashyap SR, Kasumov T, Zhang R, Kelly KR, et al. (2009) Plasma ceramides are elevated in obese subjects with type 2 diabetes and correlate with the severity of insulin resistance. Diabetes 58: 337–343.

9. Watt MJ, Barnett AC, Bruce CR, Schenk S, Horowitz JF, et al. (2012) Regulation of plasma ceramide levels with fatty acid oversupply: evidence that the liver detects and secretes de novo synthesised ceramide. Diabetologia 55: 2741–2746.

10. de Mello VD, Lankinen M, Schwab U, Kolehmainen M, Lehto S, et al. (2009) Link between plasma ceramides, inflammation and insulin resistance: association with serum IL-6 concentration in patients with coronary heart disease. Diabetologia 52: 2612–2615.

11. Ichi I, Nakahara K, Miyashita Y, Hidaka A, Kutsukake S, et al. (2006) Association of ceramides in human plasma with risk factors of atherosclerosis. Lipids 41: 859–863.

12. Samad F, Hester KD, Yang G, Hannun YA, Bielawski J (2006) Altered adipose and plasma sphingolipid metabolism in obesity: a potential mechanism for cardiovascular and metabolic risk. Diabetes 55: 2579–2587.

13. Hla T, Dannenberg AJ (2012) Sphingolipid signaling in metabolic disorders. Cell Metab 16: 420–434.

14. Boon J, Hoy AJ, Stark R, Brown RD, Meex RC, et al.. (2012) Ceramides Contained in LDL Are Elevated in Type 2 Diabetes and Promote Inflammation and Skeletal Muscle Insulin Resistance. Diabetes.

15. Huang H, Kasumov T, Gatmaitan P, Heneghan HM, Kashyap SR, et al. (2011) Gastric bypass surgery reduces plasma ceramide subspecies and improves insulin sensitivity in severely obese patients. Obesity (Silver Spring) 19: 2235–2240.

16. Pappu R, Schwab SR, Cornelissen I, Pereira JP, Regard JB, et al. (2007) Promotion of lymphocyte egress into blood and lymph by distinct sources of sphingosine-1-phosphate. Science 316: 295–298.

17. Hammad SM, Al Gadban MM, Semler AJ, Klein RL (2012) Sphingosine 1-phosphate distribution in human plasma: associations with lipid profiles. J Lipids 2012: 180705.

18. Venkataraman K, Lee YM, Michaud J, Thangada S, Ai Y, et al. (2008) Vascular endothelium as a contributor of plasma sphingosine 1-phosphate. Circ Res 102: 669–676.

19. Sattler K, Levkau B (2009) Sphingosine-1-phosphate as a mediator of high-density lipoprotein effects in cardiovascular protection. Cardiovasc Res 82: 201–211.

20. Pyne S, Pyne NJ (2011) Translational aspects of sphingosine 1-phosphate biology. Trends Mol Med 17: 463–472.

21. Fox TE, Bewley MC, Unrath KA, Pedersen MM, Anderson RE, et al. (2011) Circulating sphingolipid biomarkers in models of type 1 diabetes. J Lipid Res 52: 509–517.

22. Rapizzi E, Taddei ML, Fiaschi T, Donati C, Bruni P, et al. (2009) Sphingosine 1-phosphate increases glucose uptake through trans-activation of insulin receptor. Cell Mol Life Sci 66: 3207–3218.

23. Jun DJ, Lee JH, Choi BH, Koh TK, Ha DC, et al. (2006) Sphingosine-1-phosphate modulates both lipolysis and leptin production in differentiated rat white adipocytes. Endocrinology 147: 5835–5844.

24. Carey AL, Formosa MF, Van Every B, Bertovic D, Eikelis N, et al. (2013) Ephedrine activates brown adipose tissue in lean but not obese humans. Diabetologia 56: 147–155.

25. Maceyka M, Harikumar KB, Milstien S, Spiegel S (2012) Sphingosine-1-phosphate signaling and its role in disease. Trends Cell Biol 22: 50–60.

26. Cuvillier O, Pirianov G, Kleuser B, Vanek PG, Coso OA, et al. (1996) Suppression of ceramide-mediated programmed cell death by sphingosine-1-phosphate. Nature 381: 800–803.

27. Osawa Y, Uchinami H, Bielawski J, Schwabe RF, Hannun YA, et al. (2005) Roles for C16-ceramide and sphingosine 1-phosphate in regulating hepatocyte apoptosis in response to tumor necrosis factor-alpha. J Biol Chem 280: 27879–27887.

28. Awad AS, Rouse MD, Khutsishvili K, Huang L, Bolton WK, et al. (2011) Chronic sphingosine 1-phosphate 1 receptor activation attenuates early-stage diabetic nephropathy independent of lymphocytes. Kidney Int 79: 1090–1098.

29. Knapp M (2011) Cardioprotective role of sphingosine-1-phosphate. J Physiol Pharmacol 62: 601–607.

30. Oyama O, Sugimoto N, Qi X, Takuwa N, Mizugishi K, et al. (2008) The lysophospholipid mediator sphingosine-1-phosphate promotes angiogenesis in vivo in ischaemic hindlimbs of mice. Cardiovasc Res 78: 301–307.

31. Holland WL, Miller RA, Wang ZV, Sun K, Barth BM, et al. (2011) Receptor-mediated activation of ceramidase activity initiates the pleiotropic actions of adiponectin. Nat Med 17: 55–63.

32. Hughes JE, Srinivasan S, Lynch KR, Proia RL, Ferdek P, et al. (2008) Sphingosine-1-phosphate induces an antiinflammatory phenotype in macrophages. Circ Res 102: 950–958.

33. Olefsky JM, Glass CK (2010) Macrophages, inflammation, and insulin resistance. Annu Rev Physiol 72: 219–246.

34. Pyne NJ, Pyne S (2010) Sphingosine 1-phosphate and cancer. Nat Rev Cancer 10: 489–503.

35. Calle EE, Thun MJ (2004) Obesity and cancer. Oncogene 23: 6365–6378.

36. Cantrell Stanford J, Morris AJ, Sunkara M, Popa GJ, Larson KL, et al. (2012) Sphingosine 1-phosphate (S1P) regulates glucose-stimulated insulin secretion in pancreatic beta cells. J Biol Chem 287: 13457–13464.

37. Zhao Z, Choi J, Zhao C, Ma ZA (2012) FTY720 normalizes hyperglycemia by stimulating beta-cell in vivo regeneration in db/db mice through regulation of cyclin D3 and p57(KIP2). J Biol Chem 287: 5562–5573.

38. Bruce CR, Risis S, Babb JR, Yang C, Lee-Young RS, et al. (2013) The sphingosine-1-phosphate analog FTY720 reduces muscle ceramide content and improves glucose tolerance in high fat-fed male mice. Endocrinology 154: 65–76.

39. Ebeling P, Bourey R, Koranyi L, Tuominen JA, Groop LC, et al. (1993) Mechanism of enhanced insulin sensitivity in athletes. Increased blood flow, muscle glucose transport protein (GLUT-4) concentration, and glycogen synthase activity. J Clin Invest 92: 1623–1631.

40. Bruce CR, Anderson MJ, Carey AL, Newman DG, Bonen A, et al. (2003) Muscle oxidative capacity is a better predictor of insulin sensitivity than lipid status. J Clin Endocrinol Metab 88: 5444–5451.

41. Baranowski M, Charmas M, Dlugolecka B, Gorski J (2011) Exercise increases plasma levels of sphingoid base-1 phosphates in humans. Acta Physiol (Oxf) 203: 373–380.

42. Horowitz JF, Klein S (2000) Lipid metabolism during endurance exercise. Am J Clin Nutr 72: 558S–563S.

43. Devlin JT, Hirshman M, Horton ED, Horton ES (1987) Enhanced peripheral and splanchnic insulin sensitivity in NIDDM men after single bout of exercise. Diabetes 36: 434–439.

44. Ayala JE, Bracy DP, McGuinness OP, Wasserman DH (2006) Considerations in the design of hyperinsulinemic-euglycemic clamps in the conscious mouse. Diabetes 55: 390–397.

45. Horowitz JF, Klein S (2000) Whole body and abdominal lipolytic sensitivity to epinephrine is suppressed in upper body obese women. Am J Physiol Endocrinol Metab 278: E1144–1152.

46. Nurjhan N, Consoli A, Gerich J (1992) Increased lipolysis and its consequences on gluconeogenesis in non-insulin-dependent diabetes mellitus. J Clin Invest 89: 169–175.

47. Jensen MD, Haymond MW, Gerich JE, Cryer PE, Miles JM (1987) Lipolysis during fasting. Decreased suppression by insulin and increased stimulation by epinephrine. J Clin Invest 79: 207–213.

48. Hammad SM, Pierce JS, Soodavar F, Smith KJ, Al Gadban MM, et al. (2010) Blood sphingolipidomics in healthy humans: impact of sample collection methodology. J Lipid Res 51: 3074–3087.

49. Rothman DL, Magnusson I, Katz LD, Shulman RG, Shulman GI (1991) Quantitation of hepatic glycogenolysis and gluconeogenesis in fasting humans with 13C NMR. Science 254: 573–576.

50. Konig M, Bulik S, Holzhutter HG (2012) Quantifying the contribution of the liver to glucose homeostasis: a detailed kinetic model of human hepatic glucose metabolism. PLoS Comput Biol 8: e1002577.

51. Egom EE, Mamas MA, Soran H (2013) HDL quality or cholesterol cargo: what really matters - spotlight on sphingosine-1-phosphate-rich HDL. Curr Opin Lipidol.

52. Garcia JG, Liu F, Verin AD, Birukova A, Dechert MA, et al. (2001) Sphingosine 1-phosphate promotes endothelial cell barrier integrity by Edg-dependent cytoskeletal rearrangement. J Clin Invest 108: 689–701.