- About this Journal
- Abstracting and Indexing
- Aims and Scope
- Article Processing Charges
- Articles in Press
- Author Guidelines
- Bibliographic Information
- Citations to this Journal
- Contact Information
- Editorial Board
- Editorial Workflow
- Free eTOC Alerts
- Publication Ethics
- Reviewers Acknowledgment
- Submit a Manuscript
- Subscription Information
- Table of Contents
Journal of Lipids
Volume 2012 (2012), Article ID 180705, 8 pages
Sphingosine 1-Phosphate Distribution in Human Plasma: Associations with Lipid Profiles
1Department of Regenerative Medicine and Cell Biology, Medical University of South Carolina, Charleston, SC 29425, USA
2Division of Endocrinology, Metabolism, and Medical Genetics, Department of Medicine, Medical University of South Carolina, Charleston, SC 29425, USA
3Research Service, Ralph H. Johnson VA Medical Center, United States Department of Veterans Affairs, Charleston, SC 29401, USA
Received 8 August 2012; Accepted 27 September 2012
Academic Editor: Philip W. Wertz
Copyright © 2012 Samar M. Hammad et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
The physiological significance of sphingosine 1-phosphate (S1P) transport in blood has been debated. We have recently reported a comprehensive sphingolipid profile in human plasma and lipoprotein particles (VLDL, LDL, and HDL) using HPLC-MS/MS (Hammad et al., 2010). We now determined the relative concentrations of sphingolipids including S1P in the plasma subfraction containing lipoproteins compared to those in the remaining plasma proteins. Sphingomyelin and ceramide were predominantly recovered in the lipoprotein-containing fraction. Total plasma S1P concentration was positively correlated with S1P concentration in the protein-containing fraction, but not with S1P concentration in the lipoprotein-containing fraction. The percentage of S1P transported in plasma lipoproteins was positively correlated with HDL cholesterol (HDL-C) concentration; however, S1P transport in lipoproteins was not limited by the concentration of HDL-C in the individual subject. Thus, different plasma pools of S1P may have different contributions to S1P signaling in health and disease.
Sphingolipids have been implicated in diseases such as cancer, obesity, and atherosclerosis; however, efforts addressing blood sphingolipids as biomarkers of disease or targets for therapeutics are still in their infancy. Sphingosine-1-phosphate (S1P) is a bioactive lipid that has been shown to play major roles in immunity, inflammation, and cardiovascular physiology [1–6]. S1P is found in plasma at relatively high concentrations (>200 nM) and is transported in blood bound both to lipoproteins (~65%) and albumin [7–9] with the bulk of the lipoprotein-associated S1P found in HDL (~54%), especially the smaller diameter HDL3 subfraction [10, 11]. Plasma HDL cholesterol levels correlate positively with those of plasma S1P , but HDLs do not appear to be merely inert carriers of S1P as the S1P constituent of HDL is biologically active and contributes to numerous metabolic effects of HDL [10, 11, 13–16].
We have recently reported a comprehensive sphingolipid profile in “normal” human plasma and lipoprotein particles (VLDL, LDL, and HDL) using HPLC-MS/MS . As an extension of this study we now identify the nonlipoprotein (albumin)-associated versus the lipoprotein-associated sphingolipids. It is established that the concentration of S1P in plasma/serum is much higher than the half-maximal concentration of S1P needed to stimulate its receptors. Nevertheless, it was shown for instance that the inositol phosphate response mediated by S1P receptors was much smaller than the response expected from the total amount of S1P introduced to cells . This response to exogenous S1P was markedly attenuated in the presence of lipoprotein-deficient serum, and was associated with the trapping of exogenous S1P . Importantly, HDL and LDL showed a stronger activity for trapping S1P than lipoprotein-deficient serum . As such, these findings suggest that the “tight” binding of S1P to lipoproteins may interfere with the S1P binding to its receptors and thereby attenuate the S1P-receptor-mediated cell responses.
The physiological significance of S1P transport in HDL versus bound to albumin has been debated and this has led to the suggestion that S1P may be atheroprotective when bound to HDL but proatherogenic when localized to the plasma protein fraction of blood bound to albumin [1, 7, 16–18]. Most importantly, the distribution of S1P between the plasma lipoprotein and protein fractions of blood is reportedly altered in patients with coronary artery disease . We determined the concentrations of sphingolipids in total plasma, in the subfraction of plasma containing all the lipoprotein fractions, and in the plasma subfraction containing the remaining plasma proteins which were separated using ultracentrifugation of plasma. In addition, we investigated the distribution of S1P between the lipoprotein and plasma protein S1P pools and the influence of plasma lipoprotein and lipid concentrations on this distribution.
2. Materials and Methods
2.1. Study Subjects
Subjects aged 30–58 years were screened before participating in the study and those with known heart disease, kidney disease, diabetes, cancer, or serious current illness were excluded. Smokers, pregnant females, and subjects who take daily multivitamins or antioxidants were also excluded. A conventional lipid panel (total cholesterol (C), HDL-C, LDL-C, VLDL-C, and triglycerides) was determined in each subject (Cholestech LDX, Cholestech Corporation, Hayward, CA, USA). The plasma lipid and lipoprotein profiles for each subject are summarized in Table 1. Participating study subjects () were asked to fast overnight for at least 10 h before collecting the blood sample for analysis. The study was approved by the Institutional Review Board (IRB) at the Medical University of South Carolina (MUSC) and proper consent was obtained from each subject.
2.2. Blood Sample Collection
Blood was collected in the presence of a lipoprotein preservative cocktail EDTA (0.1% w/v), chloramphenicol (20 μg/ml), gentamycin sulfate (50 μg/ml), epsilon aminocaproic acid (0.13% w/v), and dithiobisnitrobenzoic acid (0.04% w/v) to inhibit LCAT activity (final concentrations). Blood was then centrifuged (2,400 g, 20 min, 4°C) to obtain plasma. Plasma samples were stored at 4°C until analyzed further.
2.3. Conventional Plasma Lipoprotein Profiles
Plasma conventional lipid profiles were determined by Cholestech LDX (Cholestech Corporation, Hayward, CA, USA).
2.4. Separation of Lipoproteins from Plasma Proteins
The fraction of plasma containing all the lipoproteins (VLDL, LDL, and HDL) was separated from the plasma subfraction containing the nonlipoprotein, plasma proteins using isopycnic density ultracentrifugation. Plasma solvent density was adjusted to g/ml with solid potassium bromide (KBr) and all the lipoproteins in plasma were isolated by preparative ultracentrifugation of the plasma samples (10 ml) using a type SW41Ti rotor (Beckman Coulter, Fullerton, CA, USA) spun at 41,000 rpm (288,000 g) for 40 h at 4°C in an Optima TM XL-100K ultracentrifuge. The floating lipoprotein subfraction in the ultracentrifugal supernatant was separated from the sedimented plasma lipoproteins in the subnatant after slicing each tube exactly 1.25 inch from the meniscus. The lipoprotein supernatant and plasma protein subnatant fractions were quantitatively harvested and the volume of each subfraction was adjusted to 10 ml using saline/EDTA (150 mM NaCl, 300 μM EDTA, pH 7.4) and the fractions were stored at 4°C until used. Cholesterol levels in the total lipoprotein preparation in the ultracentrifugal supernatant and in the subnatant containing the nonlipoprotein plasma proteins were measured using gas chromatography as described previously . Whole plasma and the ultracentrifugally prepared plasma subfractions were analyzed using electrophoresis in agarose gels to assess the purity of the subfractions (Beckman Paragon LIPOEPG, Beckman-Coulter, Fullerton, CA, USA).
2.5. Sphingolipid Extraction and Analysis
Analysis of endogenous sphingoid bases, sphingoid base 1-phosphates, ceramide, and sphingomyelin (SM) species were conducted in the Lipidomics Core Facility at MUSC as previously described . Briefly, 100 μl of each plasma sample, and the ultracentrifuge supernatant and subnatant subfractions of plasma were spiked with internal standards and the sphingolipid complement in each sample was quantitatively extracted as described previously . Analyses of sphingolipids were performed by high-performance liquid chromatography-tandem mass spectrometry (LC-ESI-MS/MS) as described previously . The equipment consisted of a Thermo Scientific Accela Autosampler and Quaternary Pump (Waltham, MA, USA). A Thermo Scientific Quantum Access triple quadrupole mass spectrometer equipped with an Electrospray Ion Source (ESI) operating in multiple reaction monitoring (MRM) positive ion mode was used. Chromatographic separations were obtained under a gradient elution of a Peeke Scientific (Redwood City, CA, USA), Spectra C8SR , 3-μm particle size column. Quantitative analyses were based on calibration curves generated by injecting known amounts of the target analytes and an equal amount of the internal standards. Final concentrations of analytes in samples were determined using the appropriate corrections for sample loss based on internal standard recovery calculations. Sphingolipids with no available standards were quantified using the calibration curve of its closest counterpart. The resulting data were then normalized to the volume of sample analyzed.
Data were analyzed using a commercially available statistical analysis package (SigmaStat v3.0, SPSS, Chicago, IL, USA).
We selectively recruited six subjects whose plasma lipid and lipoprotein levels (Table 1) were representative of normolipidemia (subjects 1–4), combined hyperlipoproteinemia (subject 5), and hypertriglyceridemia (subject 6). Most importantly, these subjects provided the potential to investigate the concentrations and distribution of sphingolipids over a broad range of plasma HDL-C (0.7–2.1 mM) concentrations.
We determined the concentrations of sphingolipids in the fraction of plasma containing all the plasma lipoproteins distinct from plasma fraction containing the remaining plasma proteins. These two plasma fractions were prepared by ultracentrifugation of whole plasma to float all the lipoprotein classes to separate them from the remaining plasma proteins which sedimented to the tube bottom. There were no detectable lipid staining bands with mobility comparable to plasma lipoproteins in the ultracentrifuge subnatant fractions after electrophoresis of the ultracentrifuge super- and subnatants in agarose gels (Figure 1(a)) confirming that plasma lipoproteins had been quantitatively removed from the fraction containing the plasma proteins. We also determined the concentrations of total cholesterol in each plasma fraction using gas chromatography (Figure 1(b)). Cholesterol levels in the plasma protein-containing fraction averaged only percent of the total cholesterol in plasma. Collectively, these data suggest that using the methods described, lipoproteins (lipoprotein cholesterol) were quantitatively removed from the plasma fraction containing the remaining plasma proteins.
The concentrations of the molecular species of ceramides measured in whole plasma, in the fraction of plasma containing all the plasma lipoproteins, and the plasma fraction containing the plasma proteins are shown in Table 2. Ceramide concentrations in the lipoprotein-containing fraction averaged 96 ± 3% of total plasma ceramide content in agreement with their role as an integral lipoprotein structural constituent and further support our conclusion that we quantitatively separated lipoproteins from the remaining plasma proteins. The concentrations of the ceramide species in the lipoprotein- and plasma protein-containing fractions averaged 97 ± 5% of the total ceramide concentration in plasma in support of our conclusion that recovery of sphingolipid in the plasma fractions after ultracentrifugation was quantitative. The ceramides present in highest concentration were the C24, C24:1, C20, and C22 species in agreement with our previous reports [9, 11]. We also analyzed the plasma concentrations of the SM species which averaged (mean ± SD, μM) (C14-SM), (C16-SM), (C18-SM), (C18:1-SM), (C20-SM), (C20:1-SM), (C22-SM), (C22:1-SM), (C24-SM), and (C24:1-SM). SM species distribution in the lipoprotein-containing fraction averaged of total plasma SM in agreement with our findings for ceramide species distribution. This data provides additional support of our conclusion that the separation of lipoproteins from the plasma protein fraction after ultracentrifugation was quantitative.
The concentrations of sphingoid bases, sphingosine and dihydrosphingosine, and their 1-phosphates (S1P and dhS1P) measured in whole plasma, in the fraction of plasma containing all the plasma lipoproteins, and in the plasma fraction containing the plasma proteins are summarized in Table 3. The percentage of S1P concentration in plasma localized to the lipoprotein-containing fraction was significantly, positively correlated (; ) with HDL cholesterol concentration in the six subjects (Figure 1(c)). The total concentration of S1P in plasma varied significantly among the six subjects studied (Table 3). We wished to determine if increases in S1P mass in plasma were transported primarily in lipoproteins or were localized to the plasma protein-containing fraction. Increasing concentrations of S1P in plasma were significantly and positively correlated with the concentrations of S1P in the plasma protein-containing fraction (Figure 1(d)) (; ), but not with the S1P concentrations in the lipoprotein-containing fraction (Figure 1(e)). Because the bulk of the lipoprotein-associated S1P is found in HDL and HDL concentration in the six subjects investigated varied by almost 2.9-fold (Table 1), we investigated if the S1P concentration in the lipoprotein-containing fraction in each subject was limited by the HDL concentration in the subject. Therefore, we normalized the S1P concentration in the lipoprotein-containing fraction by the HDL cholesterol concentration in the subject (Table 3). There was a significant, positive association (; ) between total S1P concentration in plasma with the “normalized” amount of S1P in the lipoprotein-containing fraction when the concentration of S1P was normalized by the HDL cholesterol concentration in the subject (Figure 1(f)).
We investigated the associations of plasma lipid and lipoprotein concentrations with the total plasma concentrations of ceramides and sphingoid bases, sphingosine, and dihydrosphingosine, and their 1-phosphates. We also determined the distribution of these sphingolipids between the pool localized to plasma lipoproteins compared to that found in the remaining plasma proteins mainly bound to albumin. We determined that S1P concentration in plasma is highly variable and determined further that S1P in plasma was nonuniformly distributed between the plasma lipoprotein and plasma protein pools of S1P in the six subjects investigated. We determined further that increasing concentrations of S1P in plasma were localized primarily to the plasma protein-containing fraction (Figures 1(d) versus 1(e)) in subjects exhibiting a broad range of plasma lipid concentrations (Table 1).
The results of our studies in these select subjects confirmed and extended those reporting previously  that S1P concentrations in the lipoprotein-containing fraction of plasma were positively correlated with HDL cholesterol concentrations (Figure 1(c)). Because HDL is the primary lipoprotein which transports S1P, and because HDL-C concentration varied widely in the six subjects (Table 1), we also normalized the levels of S1P in the lipoprotein-containing fraction to the HDL-C concentration in each subject (Figure 1(f)). We determined that this “normalized” concentration of S1P transported in the lipoprotein-containing fraction was positively correlated to the total S1P concentration in plasma, which suggests that the S1P content of HDL is variable and that S1P transport in lipoproteins was not limited by the concentration of HDL-C in the individual subject. In fact, we have demonstrated previously that per particle, the larger VLDL particle contains the highest content of S1P, and the smallest lipoprotein particle, HDL3, contains higher S1P levels than LDL particles . Our current data may allow subsequent research on mechanisms that mediate binding of the different S1P carriers to S1P receptors, knowing that only 5% of HDL particles carry S1P molecules .
Recent studies have reported that HDL-associated S1P is bound specifically to apolipoprotein M (apoM) and, furthermore, is transported selectively in apoM-containing HDL particles . We did not measure apoM concentration or distribution in this select subject set and thus cannot infer if differences in apoM metabolism between the subjects were associated with the observed differences in S1P metabolism. Alternatively, we [9, 11] and others [8, 10] have determined that S1P levels in HDL are significantly higher in the smaller sized HDL3 subfraction compared to S1P levels in HDL2. The smaller HDL3 particles were shown also to be enriched in apoM . Despite the very strong evidence from apoM knockout and apoM-transgenic mice that apoM determines S1P concentrations in plasma and HDL [22, 23], there was no statistically significant correlation between S1P and apoM concentrations in human patients with different monogenic disorders of HDL metabolism . Thus, the relative concentrations of HDL subfractions may have influenced the distribution of S1P in the individual subjects.
More recently, the crystal structure of a main S1P G protein-coupled receptor, S1P1, was revealed, and it was found that extracellular access to the binding pocket of this receptor is occluded by the aminoterminus and extracellular loops of the receptor . Interestingly, access is gained by ligands entering laterally between helices I and VII within the transmembrane region of the receptor . In the context of previous findings, we postulate that effects of HDL-bound S1P could be related to HDL particle size and the ability to modulate cell lipid rafts and comprised receptors, S1P content of HDL particles being higher in the smaller HDL particles, and HDL role as scavenger/reservoir for biologically active lipids including S1P.
A confounding factor in recent studies addressing the cardioprotective role of S1P levels in HDL is the assumption that 30% of S1P is recovered in the lipoprotein-deficient serum . As shown in Table 3, S1P distribution in the plasma protein-containing fraction approximated 30% in only three of the six subjects studied. Thus, even in this limited number of subjects, S1P distribution in 50% of the subjects does not support this often cited assumption. Figure 1(e) clearly shows that there is no significant association between total plasma S1P and S1P concentration in plasma lipoprotein-containing fraction, which suggests a crucial pathophysiological role of the S1P bound to plasma proteins. Furthermore, the procedure in which apolipoprotein B-depleted plasma is prepared and S1P concentration is subsequently determined in the “HDL-containing” fraction as performed in two recent studies [21, 26] also does not discriminate between HDL- and albumin-associated S1P. Thus, when this methodology is employed, assumptions regarding S1P distribution in plasma may result in erroneous estimates.
We have shown previously that the secretion of plasminogen activator inhibitor-1 (PAI-1) from cultured adipocytes is significantly increased when the cells were incubated with increasing S1P concentration in the media regardless of whether the S1P is bound to HDL or transported by albumin . Previous studies have determined that triglyceride level was independently related to plasma PAI-1 activity level in both subjects with hypertriglyceridemia and in age-matched normotriglyceridemic subjects . The factor(s) contributing to this phenomenon are still unclear. In this limited study, plasma S1P concentrations were increased in the two subjects exhibiting elevated plasma triglyceride levels. Clearly, additional studies are warranted in this area.
In summary, we determined that S1P concentration in plasma is highly variable, S1P in plasma is nonuniformly distributed between the plasma lipoprotein and plasma protein pools of S1P and increasing concentrations of S1P in plasma are localized primarily to the plasma protein-containing fraction. We also determined that the S1P content of HDL, the major lipoprotein carrier of S1P, is variable and that S1P transport in lipoproteins is not limited by the concentration of HDL-C. The data further show that there is no significant association between total plasma S1P and S1P concentration in plasma lipoprotein-containing fraction, which suggests a crucial pathophysiological role of the S1P bound to plasma proteins.
|VLDL:||very low-density lipoproteins|
|PAI-1:||plasminogen activator inhibitor-1.|
This work was supported by the National Institutes of Health Grants HL-079274 and R0-1HL079274-04S1 (American Recovery and Reinvestment Act), the Southeastern Clinical and Translational Research Institute, the South Carolina Center of Biomedical Research Excellence (COBRE) in Lipidomics and Pathobiology Grant no. P20 RR17677 from National Center for Research Resources (NCCR), and the Pilot Research funding from the Hollings Cancer Center’s Cancer Center Support Grant at MUSC Grant no. P30 CA13813 (to S. M. Hammad), and the Department of Veterans Affairs Merit Review Program (to R. L. Klein). The contents presented herein do not represent the views of the Department of Veterans Affairs or the United States Government. The authors gratefully acknowledge the Lipidomics Shared Resource Facility at MUSC for their continued expert analytical contributions, without which this work could not have been conducted. We are particularly grateful to Mr. Kent Smith for technical assistance.
- A. E. Alewijnse and S. L. M. Peters, “Sphingolipid signalling in the cardiovascular system: good, bad or both?” European Journal of Pharmacology, vol. 585, no. 2-3, pp. 292–302, 2008.
- S. Spiegel and S. Milstien, “Sphingosine-1-phosphate: an enigmatic signalling lipid,” Nature Reviews Molecular Cell Biology, vol. 4, no. 5, pp. 397–407, 2003.
- T. Hla, “Physiological and pathological actions of sphingosine 1-phosphate,” Seminars in Cell and Developmental Biology, vol. 15, no. 5, pp. 513–520, 2004.
- K. Watterson, H. Sankala, S. Milstien, and S. Spiegel, “Pleiotropic actions of sphingosine-1-phosphate,” Progress in Lipid Research, vol. 42, no. 4, pp. 344–357, 2003.
- Y. Takuwa, Y. Okamoto, K. Yoshioka, and N. Takuwa, “Sphingosine-1-phosphate signaling and biological activities in the cardiovascular system,” Biochimica et Biophysica Acta, vol. 1781, no. 9, pp. 483–488, 2008.
- H. Rosen, P. J. Gonzalez-Cabrera, M. G. Sanna, and S. Brown, “Sphingosine 1-phosphate receptor signaling,” Annual Review of Biochemistry, vol. 78, pp. 743–768, 2009.
- F. Okajima, “Plasma lipoproteins behave as carriers of extracellular sphingosine 1-phosphate: is this an atherogenic mediator or an anti-atherogenic mediator?” Biochimica et Biophysica Acta, vol. 1582, no. 1–3, pp. 132–137, 2002.
- N. Murata, K. Sato, J. Kon et al., “Interaction of sphingosine 1-phosphate with plasma components, including lipoproteins, regulates the lipid receptor-mediated actions,” Biochemical Journal, vol. 352, no. 3, pp. 809–815, 2000.
- S. M. Hammad, J. S. Pierce, F. Soodavar et al., “Blood sphingolipidomics in healthy humans: impact of sample collection methodology,” Journal of Lipid Research, vol. 51, no. 10, pp. 3074–3087, 2010.
- A. Kontush, P. Therond, A. Zerrad et al., “Preferential sphingosine-1-phosphate enrichment and sphingomyelin depletion are key features of small dense HDL3 particles: relevance to antiapoptotic and antioxidative activities,” Arteriosclerosis, Thrombosis, and Vascular Biology, vol. 27, no. 8, pp. 1843–1849, 2007.
- M. H. Lee, S. M. Hammad, A. J. Semler, L. M. Luttrell, M. F. Lopes-Virella, and R. L. Klein, “HDL3, but not HDL2, stimulates plasminogen activator inhibitor-1 release from adipocytes: the role of sphingosine-1-phosphate,” Journal of Lipid Research, vol. 51, no. 9, pp. 2619–2628, 2010.
- B. Zhang, H. Tomura, A. Kuwabara et al., “Correlation of high density lipoprotein (HDL)-associated sphingosine 1-phosphate with serum levels of HDL-cholesterol and apolipoproteins,” Atherosclerosis, vol. 178, no. 1, pp. 199–205, 2005.
- J. R. Nofer, M. Van Der Giet, M. Tölle et al., “HDL induces NO-dependent vasorelaxation via the lysophospholipid receptor S1P3,” Journal of Clinical Investigation, vol. 113, no. 4, pp. 569–581, 2004.
- C. Rodríguez, M. Gonzalez-Diez, L. Badimon, and J. Martínez-González, “Sphingosine-I-phosphate: a bioactive lipid that confers high-density lipoprotein with vasculoprotection mediated by nitric oxide and prostacyclin,” Thrombosis and Haemostasis, vol. 101, no. 4, pp. 665–673, 2009.
- G. Theilmeier, C. Schmidt, J. Herrmann et al., “High-density lipoproteins and their constituent, sphingosine-1-phosphate, directly protect the heart against ischemia/reperfusion injury in vivo via the S1P3 lysophospholipid receptor,” Circulation, vol. 114, no. 13, pp. 1403–1409, 2006.
- K. Sattler and B. Levkau, “Sphingosine-1-phosphate as a mediator of high-density lipoprotein effects in cardiovascular protection,” Cardiovascular Research, vol. 82, no. 2, pp. 201–211, 2009.
- J. S. Karliner, “Sphingosine kinase and sphingosine 1-phosphate in cardioprotection,” Journal of Cardiovascular Pharmacology, vol. 53, no. 3, pp. 189–197, 2009.
- S. M. Hammad, “Blood sphingolipids in homeostasis and pathobiology,” Sphingolipids and Metabolic Disease, vol. 721, pp. 57–66, 2011.
- K. J. E. Sattler, Ş. Elbasan, P. Keul et al., “Sphingosine 1-phosphate levels in plasma and HDL are altered in coronary artery disease,” Basic Research in Cardiology, vol. 105, no. 6, pp. 821–832, 2010.
- R. L. Klein, T. J. Lyons, and M. F. Lopes-Virella, “Metabolism of very low- and low-density lipoproteins isolated from normolipidaemic Type 2 (non-insulin-dependent) diabetic patients by human monocyte-derived macrophages,” Diabetologia, vol. 33, no. 5, pp. 299–305, 1990.
- R. Karuna, R. Park, A. Othman et al., “Plasma levels of sphingosine-1-phosphate and apolipoprotein M in patients with monogenic disorders of HDL metabolism,” Atherosclerosis, vol. 219, no. 2, pp. 855–863, 2011.
- C. Christoffersen, H. Obinata, S. B. Kumaraswamy et al., “Endothelium-protective sphingosine-1-phosphate provided by HDL-associated apolipoprotein M,” Proceedings of the National Academy of Sciences of the United States of America, vol. 108, no. 23, pp. 9613–9618, 2011.
- W. S. Davidson, R. A. G. D. Silva, S. Chantepie, W. R. Lagor, M. J. Chapman, and A. Kontush, “Proteomic analysis of defined hdl subpopulations reveals particle-specific protein clusters: relevance to antioxidative function,” Arteriosclerosis, Thrombosis, and Vascular Biology, vol. 29, no. 6, pp. 870–876, 2009.
- M. A. Hanson, C. B. Roth, E. Jo et al., “Crystal structure of a lipid G protein-coupled receptor,” Science, vol. 335, no. 6070, pp. 851–855, 2012.
- N. Murata, K. Sato, J. Kon et al., “Interaction of sphingosine 1-phosphate with plasma components, including lipoproteins, regulates the lipid receptor-mediated actions,” Biochemical Journal, vol. 352, no. 3, pp. 809–815, 2000.
- K. M. Argraves, A. A. Sethi, P. J. Gazzolo et al., “S1P, dihydro-S1P and C24:1-ceramide levels in the HDL-containing fraction of serum inversely correlate with occurrence of ischemic heart disease,” Lipids in Health and Disease, vol. 10, p. 70, 2011.
- A. Asplund-Carlson, A. Hamsten, B. Wiman, and L. A. Carlson, “Relationship between plasma plasminogen activator inhibitor-1 activity and VLDL triglyceride concentration, insulin levels and insulin sensitivity: studies in randomly selected normo- and hypertriglyceridaemic men,” Diabetologia, vol. 36, no. 9, pp. 817–825, 1993.