| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Institute for Immunology, Hannover Medical School, Hanover, Germany
1Correspondence: Institute for Immunology, Hannover Medical School, Bldg. K11, OE 9422, Carl-Neuberg-Str. 1, 30625 Hanover, Germany. E-mail: graeler.markus{at}mh-hannover.de
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: HPLC • plasma • sphingolipid • thrombocyte
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
2
3)
. The latter attracts special interest because the S1P1 receptor appears to be critical for lymphocyte egress from thymus and secondary lymphoid organs, a process that is still not well understood (4
5
6)
. The immunosuppressant Fingolimod (2-amino-2-(2-[4-octylphenyl]ethyl]-1,3-propanediol, FTY720) induces lymphopenia in humans and rodents by down-regulating the S1P1 receptor on lymphocytes (5
, 7)
. Whereas lymphocyte-targeted gene deletion in two different mouse mutants clearly indicates the pivotal role of the S1P1 receptor for maintaining lymphocyte circulation, our knowledge regarding regulation and distribution of its endogenous ligand S1P is quite limited (4
, 5)
. Reported serum S1P concentrations range from 100 nM to 4 µM, depending on the detection method used and the species analyzed (8
9
10)
. Mouse serum, for example, was calculated to contain either 840 nM or >4 µM S1P in two different studies (9
, 11)
. Human serum was reported to contain 340 nM or 1 µM S1P (8
, 10)
. These data indicate that concentrations differ not only between different species, but also between different individuals of the same species. S1P not only stimulates the S1P1 receptor, but also mediates activation-induced receptor down-regulation (12)
. The extent of stimulation and receptor internalization is dependent on the actual S1P concentration. To understand the underlying mechanism of lymphocyte egress control, it is critical to know how the concentration and distribution of endogenous S1P are regulated.
Naive lymphocytes use the systemic vasculature and lymphatics to frequently circulate between different lymphoid and nonlymphoid organs (3)
. It was recently proposed that high S1P levels in blood and lymph establish a gradient for lymphocytes that reside in lymphoid organs with only little extracellular matrix (ECM) -associated S1P (11
, 13)
. Lymphocytes that express the S1P1 receptor on their cell surface chemotactically respond to S1P and migrate along this supposed gradient. They leave the S1P-low environment of lymphoid tissues and enter S1P-high surrounding lymph and blood. Treatment of mice with the vitamin B6 antagonist 4-deoxypyridoxine (DOP) or the caramel color III component 2-acetyl-4,(5)-tetrahydroxybutylimidazole (THI) induces a pathological increase of S1P levels in lymphoid organs (11)
. Supposedly this increase in tissue S1P levels abolishes the native S1P gradient to lymph and blood, which may account for the observed lymphopenia in DOP- and THI-treated mice.
Although the hypothesis of a developing gradient between lymphoid tissues and connecting lymph and blood is appealing, not all researchers follow this argumentation. Prolonged intravenously (i.v.) perfusion of S1P in rats, for example, also results in lymphopenia although the proposed gradient should be pronounced in these animals (14)
. Single i.v. injections, however, do not alter blood lymphocyte counts (15)
. Another strong argument is that so far only S1P1 receptor agonists but not antagonists induce lymphopenia in rodents (6
, 16
, 17)
. Therefore, receptor stimulation seems to be required for the onset of lymphopenia. The above-mentioned S1P gradient hypothesis, however, would propose the reverse—namely, S1P1 antagonists inducing lymphopenia.
To resolve these controversies, we started to investigate the elemental regulation of blood-borne S1P, which plays a critical role in all theories mentioned for lymphocyte exit and recirculation. As a result, we identified erythrocytes as the main blood cell population that is capable of incorporating, protecting, storing, and releasing S1P.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Blood and plasma preparation
Blood was drawn from healthy volunteers, collected in heparinized syringes, and put onto ice immediately. Plasma and blood cells were separated by centrifugation at 1200 g for 5 min at 4°C. Blood cells were washed twice with ice-cold PBS to remove plasma residues.
Blood cell separation
Ficoll separation was performed according to the manufacturer’s protocol (Amersham Pharmacia Biotech, Freiburg, Germany). After the centrifugation platelets were extracted from the upper layer, white blood cells (WBC) from the intermediate layer, and erythrocytes from the bottom. Each cell population was washed twice with PBS before incubation.
Recovery of highly purified erythrocyte and thrombocyte preparations
Erythrocyte and thrombocyte concentrates were obtained from the Transfusion Department of the Hannover Medical School. Erythrocyte concentrates were leukocyte depleted by whole blood inline filtration. Thrombocyte concentrates were leukocyte depleted by zytapheresis. Cell concentrations in erythrocyte concentrates ranged from 3.8 to 5.4 x 106 cells/µl, those in thrombocyte concentrates ranged from 1.5 to 3.5 x 105 cells/µl.
Lipid extraction
Biological samples (1 ml medium, 50 µl erythrocyte cell pellet, 100 µl serum or plasma) were adjusted to 1 ml sample volume with 1 M NaCl and transferred into a glass centrifuge tube. After addition of 1 ml methanol (Baker; Griesheim, Germany) and 100 µl of 37% hydrochloric acid, the samples were vortexed. Chloroform (2 ml) was added, and the samples were mixed for 30 min at 50 rpm in a test tube rotator. Samples were centrifuged for 3 min at 1900 g, and the lower chloroform phase was transferred into a new glass centrifuge tube. After repeating the lipid extraction with another 2 ml of chloroform, the two chloroform phases were combined and vacuum-dried in a SpeedVac for 45 min at 48°C.
Derivatization of sphingolipids with 9-fluorenylmethyl chloroformiate (FMOC-Cl)
18 mg FMOC-Cl were dissolved in 5 ml dioxane. Vacuum-dried samples were dissolved in 200 µl dioxane with subsequent addition of 200 µl of 70 mM dipotassiumhydrogenphosphate in H2O and 200 µl FMOC-Cl solution.
HPLC analysis of FMOC-Cl labeled sphingolipids
Chromatographic detection of sphingolipids was performed as described (18)
using the Merck-Hitachi Elite LaChrom System (VWR; Darmstadt, Germany). The injection pump delivery rate was 1.3 ml/min. The eluent was 82 to 95% methanol, 5 to 0% 70 mM dipotassiumhydrogenphosphate, and 13 to 5% H2O, forming a gradient over a period of 68 min. A 10 µl sample volume was injected using the cut injection method. Separation of sphingolipids with reversed-phase chromatography was done using a 250 x 4.6 mm Kromasil 100–5 C18 separation column and a 17 x 4 mm Kromasil 100–5 C18 precolumn (CS Chromatographie Service; Langerwehe, Germany). Column temperature was 35°C; detection was performed with a fluorescence detector at 263 nm excitation and 316 nm emission wave length.
Blood cell incubations
Blood, erythrocytes, thrombocytes, or WBC were incubated in 24-well plates at 37°C, with 5% CO2. Samples were taken after different time points and immediately put on ice. Cells were separated from the supernatant by centrifugation at 1200 g for 5 min at 4°C. The cells were washed twice with ice-cold PBS. All samples were stored at –20°C.
Blocking of S1P synthesis
Whole human blood was incubated for up to 6 h at 37°C and 5% CO2. Blocking reagents were applied as follows: 10 µM myriocin (MYR) (Sigma, Taufkirchen, Germany); 1 mM N-oleoylethanolamine (NOE) (Sigma); 1 mM (1S,2R)-D-erythro-2-(N-myristolamino)-1-phenyl-1-propanol (DMAPP) (Biotrend, Köln, Germany); 20 µM fumonisin B1 (FUM) (VWR, Darmstadt, Germany); 10 µM N,N-dimethylsphingosine (DMS) (Alexis, Grünberg, Germany); 0.5 mM DOP (Sigma); 0.5 mM THI (Exclusive Chemistry, Obninsk, Russia).
CFSE labeling of erythrocytes
Blood cells from two male C57BL/6 mice were Ficoll-separated as described above. Erythrocytes were washed twice with ice-cold PBS. Concentrated erythrocytes (400 µl) were incubated overnight at 37°C and 5% CO2 in 3 ml RPMI medium supplemented with 10% FCS, 100 U/ml penicillin G, and 100 µg/ml streptomycin with and without addition of 1 µM S1P. Subsequently S1P-loaded and unloaded erythrocytes were washed twice with ice-cold PBS and labeled with either 25 µM or 100 µM of 5(6)-carboxyfluorescein diacetate, succinimidyl ester (CFSE, Invitrogen, Karlsruhe, Germany) in 6 ml PBS for 10 min. The reaction was stopped with FCS and cells were washed twice with PBS. Labeling efficiency was confirmed by flow cytometry using the FacsCalibur (Becton-Dickinson, Heidelberg, Germany). S1P content was analyzed by high-performance liquid chromatography (HPLC) as described above. Labeled cells were adoptively transferred into the tail vein of syngeneic wild-type recipient mice. Blood was taken from the retro-orbital sinus of recipient mice and analyzed for the occurrence of labeled erythrocytes at different time points.
Biotin labeling of erythrocytes
Blood from six male C57BL/6 mice was collected by heart puncture and Ficoll-separated as described above. Washed erythrocytes were incubated overnight at 37°C and 5% CO2 in 7.5 ml RPMI medium supplemented with 10% FCS, 100 U/ml penicillin G, and 100 µg/ml streptomycin with and without addition of 3 µM S1P. S1P-loaded and unloaded cells were washed twice with ice-cold PBS, and 1 ml of centrifuged erythrocytes (1x10E10 cells/ml) was labeled with 0.6 mg of sulfoNHS-SS-biotin (Pierce, Rockford, IL, USA) in 3 ml PBS for 1 h on ice. Afterward cells were washed three times with ice-cold PBS and S1P content was analyzed by HPLC as described above. Labeled cells were adoptively transferred into the tail vein of two separate syngeneic wild-type recipient mice. After 24 h, blood was taken by heart puncture from recipient mice. Transferred erythrocytes were recovered and separated from endogenous erythrocytes with streptavidin-coupled magnetic beads according to the manufacturer’s protocol (Miltenyi Biotec, Bergisch-Gladbach, Germany). The S1P content of labeled cells and endogenous erythrocytes was analyzed by HPLC as described above.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
A, B). DH-S1P levels followed a similar course and increased from 50 nM to 510 nM within 4 h, then decreased again after 23 h to 240 nM (Fig. 1A, B
).
|
Erythrocytes accounted for the observed S1P and DH-S1P release
We next addressed the cellular source of S1P and DH-S1P in blood plasma. Blood cells were separated into fractions enriched with thrombocytes, WBC, and erythrocytes using the Ficoll-Hypaque technique. Fractionated cells were resuspended with fresh filtered human plasma at a ratio equivalent to their occurrence in human blood and incubated for 1 to 6 h at 37°C. Under these conditions, only the erythrocyte-enriched cell fraction secreted S1P and DH-S1P into human plasma (Fig. 2
). Notably, thrombocytes that were reported to release S1P after activation did not contribute significantly to the S1P release observed (Fig. 2)
. A significant release of 800 nM S1P was observed only with a 100-fold higher concentration of highly purified thrombocytes, which does not reflect physiological conditions (Supplemental Fig. 1). In addition, thrombocyte-derived S1P release was not accompanied by an increase of DH-S1P (Supplemental Fig. 1). WBC did not release S1P at all (Supplemental Fig. 2). Naive WBC degraded cell-bound S1P and plasma S1P after incubation for 6 h at 37°C (Supplemental Fig. 2). Release of S1P was also confirmed with purified erythrocytes (Supplemental Fig. 3).
|
Erythrocytes did not synthesize S1P and DH-S1P
To find out more about the source of released S1P, several different toxins were used to block the de novo and intermediate synthesis of S1P by purified human erythrocytes. MYR was used to block the serine palmitoyl transferase, the first step for de novo generation of S1P (Fig. 3
A). N-Oleoylethanolamine and (1S,2R)-D-erythro-2-(N-myristoylamino)-1-phenyl-1-propanol (DMAPP) were used to inhibit the intermediate production of S1P by acid and alkaline ceramidases, respectively. Ceramide synthase was inhibited by FUM to block the competitive synthesis of ceramide. N,N-Dimethylsphingosine blocked sphingosine kinases, which catalyze the phosphorylation of sphingosine to S1P, and DOP and THI inhibited the S1P-lyase that degrades S1P to hexadecenal and phosphoethanolamine (Fig. 3A
). None of the toxins mentioned had a significant effect on the S1P release from human erythrocytes in plasma (Fig. 3B
). To investigate whether S1P synthesis occurred, blood cells were analyzed regarding their S1P and DH-S1P content at different time points during plasma incubation. Resulting cellular S1P and DH-S1P levels revealed a negative correlation between blood cell and plasma S1P and DH-S1P levels. Cellular S1P and DH-S1P levels were decreasing when S1P and DH-S1P were released into plasma (Fig. 3C
). This result was also confirmed with purified erythrocytes (Supplemental Fig. 3). Notably, purified erythrocytes that were suitable for blood transfusions did release lower amounts of S1P than did freshly prepared erythrocytes (Fig. 2
, Supplemental Fig. 3). This was obviously due to the loss of cellular S1P during the purification steps (Supplemental Fig. 3). In line with these results, the amount of S1P in whole human blood, including plasma and cells, was 1.5 µM and did not change significantly within 6 h at 37°C (Supplemental Fig. 4). Although sphingosine kinases did not contribute to the observed release of S1P, erythrocytes were able to phosphorylate exogenously added sphingosine completely (Supplemental Fig. 5). The S1P-lyase was not active in erythrocytes because extracellular S1P in plasma was stable for >19 h in the presence of purified erythrocytes (Fig. 3C
).
|
S1P release was dependent on plasma and on the amount of erythrocytes
Erythrocytes function as an S1P carrier, but not as an S1P producer. Consequently the maximal amount of released S1P should be dependent on the ratio of erythrocytes and plasma. Incubation of plasma with decreasing amounts of erythrocytes for 6 h showed a diminished release of S1P and DH-S1P with lower amounts of erythrocytes (Fig. 4
A). The rate of S1P release and the maximally released amount of S1P in plasma were both reduced with a decreasing proportion of erythrocytes (Fig. 4A
). However, not only erythrocytes determined the amount and pace of S1P release; plasma also played an important role. Erythrocytes did not release S1P in serum- and plasma-free media. The amount of serum or plasma present in the medium determined the maximal amount of S1P released (Fig. 4B
). Pure plasma was necessary to obtain maximal S1P release (Fig. 4B
). This also applied to release of DH-S1P (Fig. 4B
).
|
Erythrocytes incorporate, store, and protect S1P
Erythrocytes release S1P in the presence of plasma but not in the presence of serum- and plasma-free medium (Fig. 5
A). When S1P was added to serum-free medium, erythrocytes quantitatively incorporated this exogenously added S1P (Fig. 5A
). Significant incorporation of S1P was seen in neither WBC (Supplemental Fig. 2) nor thrombocytes (Supplemental Fig. 6). Therefore, only erythrocytes were able to take up significant amounts of S1P from the medium in a serum- and plasma-free environment. Release and uptake of S1P by erythrocytes were temperature dependent and did not efficiently occur at 4°C (Supplemental Fig. 7). Both processes could be confirmed with either heparin- or EDTA-treated blood (Supplemental Fig. 8). Only the use of citrate as an anticoagulant resulted in diminished S1P release and uptake with more scattered values (Supplemental Fig. 8). Coculture of erythrocytes with HUVECs in the presence of S1P in serum-free medium also resulted in the uptake of S1P by erythrocytes. Erythrocyte-associated S1P was protected from degradation by HUVECs (Fig. 5B
). Coculture of erythrocytes and HUVECs in the presence of plasma, however, caused release and subsequent degradation of S1P by HUVECs (Fig. 5B
). The environment in which erythrocytes resided therefore determined whether they released or incorporated S1P. Erythrocyte-associated S1P was not accessible for metabolic enzymes of other cells.
|
Cellular S1P level of erythrocytes was dynamically regulated in vivo
To investigate the in vivo regulation of erythrocyte-associated S1P, mouse erythrocytes were incubated with and without S1P in serum-free medium for 20 h at 37°C, resulting in erythrocyte populations with relatively low and high cellular S1P content, respectively (Fig. 6
A). During this incubation period, the S1P concentration of the supplemented medium went down to 50 nM from an original 6.7 µM (Fig. 6A
). Cellular S1P levels in these samples went up from 1 µM to 5.6 µM (Fig. 6A
). At the same time, S1P levels of erythrocytes that were incubated without S1P in the medium decreased from 1.1 µM to 0.7 µM (Fig. 6A
). These cells were subsequently labeled with two different concentrations of CFSE, mixed, and adoptively transferred into syngeneic C57BL/6 mice. The recovery rate of labeled cells from blood was determined at different time points between 1 and 50 days. All four different cell populations (CFSEhigh/S1Phigh, CFSElow/S1Plow, CFSElow/S1Phigh, CFSEhigh/S1Plow) displayed a similar decrease in blood of adoptively transferred mice (Fig. 6B
). The one-time charge process of cellular S1P therefore did not affect the life time of erythrocytes.
|
Loading erythrocytes with S1P was easily performed by incubating them in serum- and plasma-free medium supplemented with S1P (Fig. 6A
). Depleting erythrocyte-bound S1P could also be managed during culture in freshly isolated plasma, which stimulates its release (Supplemental Fig. 3). To test whether both processes were also true in vivo, S1P-loaded and unloaded mouse erythrocytes were biotinylated and adoptively transferred into syngeneic C57BL/6 mice. 24 h later biotinylated cells were isolated with streptavidin-conjugated magnetic beads and analyzed by HPLC for their S1P content. This experiment revealed that cellular S1P levels of adoptively transferred erythrocytes with high and low S1P concentrations equalized again in vivo (Fig. 6C
). Thus, both S1P-loading and unloading processes of erythrocytes must have taken place in the mice.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. Although it is known that S1P is present in human and rodent blood at high nanomolar levels, its role in this special environment is unclear (4
, 8
, 10
, 11)
. Lymphocytes are frequently exposed to blood-borne S1P during circulation. Thus, S1P may play a role in the control of lymphocyte circulation (7)
. The S1P1 receptor was recently shown to be involved in thymocyte and lymphocyte egress from lymphoid organs; S1P in blood therefore was suggested to serve as the endogenous stimulus for exiting lymphocytes (4
, 5
, 7)
. To prove this hypothesis, it is envisaged to specifically alter S1P levels in blood. As a first step, we describe here the basic regulatory system for S1P release in blood.
Several attempts have been made to find the physiological source for S1P in vivo. So far, activated thrombocytes and mast cells were reported to secrete S1P, although several reviews suggested that many other cells should be able to secrete S1P after stimulation (20
, 21)
. The S1P release observed in whole blood (Fig. 1)
was not mediated by S1P production from thrombocytes or leukocytes, but by S1P release from erythrocytes (Fig. 2
, Supplemental Figs. 2, 3, 6). This finding adds a novel function to erythrocytes and emphasizes the special environment of blood, since erythrocytes are found exclusively in blood. It contains comparatively high levels of S1P whereas tissue levels are generally low (11)
. Erythrocytes may therefore be part of a regulatory system that maintains high levels of S1P in blood.
Although erythrocytes release S1P in blood, they do not produce it (Figs. 3
, 4
; Supplemental Fig. 4). They were able to incorporate, but not to release, S1P in a plasma- and serum-free environment (Fig. 5)
. Release was therefore restricted to plasma, which again indicates the exclusiveness of blood regarding constitutive S1P secretion. The source of erythrocyte-associated S1P is not known at this time. It is possible that they pick up S1P from a central organ in each round of their circulation. Another possibility would be that erythrocytes incorporate a precursor of S1P like sphingosine and convert it to S1P, since erythrocytes are able to efficiently phosphorylate sphingosine (Supplemental Fig. 5). Sphingosine may therefore be continuously released from one or more organs, then incorporated and phosphorylated by erythrocytes to maintain a constant S1P level in the blood. It is also possible that sphingosine is released by different cells and tissues upon activation. Sphingosine incorporation and phosphorylation by erythrocytes may consequently result in increased S1P concentrations in blood, which likely contributes to a pathological condition. Results from adoptive transfer experiments with S1P-loaded and unloaded erythrocytes indicate that they constantly release and incorporate S1P (Fig. 6C
). A one-time load or unload of erythrocytes with S1P therefore had no sustained influence on their function (Fig. 6B
). But this does not mean that the amount of cell-associated S1P does not play a role at all in their life cycle. The role of S1P for erythrocyte function remains unclear until we find tools to influence incorporation and release of S1P by erythrocytes.
S1P has attracted much attention as a signaling molecule involved in important cellular functions like cell migration and survival (3
, 22)
. Several attempts, including the genetic deletion of important enzymes involved in sphingolipid metabolism like sphingosine kinases, ceramidases, or the S1P-lyase, have been undertaken to modulate the amount of S1P in vivo (9
, 17
, 23
24
25)
. The accordant knockout mice had no or only minor alterations in S1P levels or were lethal at an embryonic stage or soon after birth. The presented data demonstrate that S1P release in blood is highly regulated (Figs. 4
5
6
, Supplemental Fig. 7). This regulatory system may be used to alter S1P levels in endogenous blood plasma, which would be much more specific than any genetic approach that targets metabolic enzymes. This is an important achievement since S1P is not only an extracellular messenger, but also serves as an important intracellular component that is involved in maintaining membrane structures, like lipid rafts, in lipid metabolism and intracellular signaling.
Classical approaches for investigating S1P metabolism include the use of metabolic enzymes and radioactively labeled S1P precursors like sphingosine or ceramide (21)
. Although very sensitive, these experiments do not detect S1P itself, but rather conversions of labeled precursors to the final product S1P. The observed S1P release from erythrocytes does not include any enzymatic conversions, and so would not have been picked up by those experiments. The adopted HPLC-based analysis of phosphorylated and nonphosphorylated sphingosine, sphinganine, and derivatives made it possible to systematically quantify several sphingolipids simultaneously on a routine basis (18)
. This allowed us to identify relevant conversions such as (de-)phosphorylation and (de-)hydrolyzation, and also determine absolute S1P concentrations.
Due to the complexity of sphingolipid metabolism and the difficulty to specifically target enzymes and receptors involved in S1P metabolism and signaling, most studies in this field were performed in mice or with cell lines (4
, 5
, 9
, 11
, 13
, 17
, 23
, 24)
. A limited amount of studies have focused on human pathology (26)
. The data presented focus primarily on the human blood of healthy volunteers and reveal basic principles for S1P storage and release. Our results thus provide an initial insight into a physiologically relevant regulation that may potentially be altered in certain pathological situations. Since S1P release increases exponentially after taking blood samples (Fig. 1
, Supplemental Fig. 8), it is important to immediately cool them and to separate plasma from cells to ensure their comparability. Processed plasma from blood donors, for example, already showed S1P concentrations of 
2 µM (data not shown), whereas plasma from freshly isolated blood contained only 200 to 300 nM S1P (18)
. To avoid false positive results, much attention must be placed on extracting and processing blood samples in order to ensure their comparability. The differences in reported S1P concentrations from human serum and plasma samples may also be explained in this manner (8
, 10)
.
|
ACKNOWLEDGMENTS |
|---|
Received for publication September 27, 2006. Accepted for publication November 9, 2006.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  J.-J. Liao, M.-C. Huang, K. Fast, K. Gundling, M. Yadav, J. R. Van Brocklyn, M. R. Wabl, and E. J. Goetzl Immunosuppressive human anti-lymphocyte autoantibodies specific for the type 1 sphingosine 1-phosphate receptor FASEB J, June 1, 2009; 23(6): 1786 - 1796. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Skoura and T. Hla Regulation of vascular physiology and pathology by the S1P2 receptor subtype Cardiovasc Res, May 1, 2009; 82(2): 221 - 228. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Maceyka, S. Milstien, and S. Spiegel Sphingosine-1-phosphate: the Swiss army knife of sphingolipid signaling J. Lipid Res., April 1, 2009; 50(Supplement): S272 - S276. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. M. Whetzel, D. T. Bolick, and C. C. Hedrick Sphingosine-1-phosphate inhibits high glucose-mediated ERK1/2 action in endothelium through induction of MAP kinase phosphatase-3 Am J Physiol Cell Physiol, February 1, 2009; 296(2): C339 - C345. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. M. Scanu and C. Edelstein HDL: bridging past and present with a look at the future FASEB J, December 1, 2008; 22(12): 4044 - 4054. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 X. Ye Lysophospholipid signaling in the function and pathology of the reproductive system Hum. Reprod. Update, September 1, 2008; 14(5): 519 - 536. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Ohkawa, K. Nakamura, S. Okubo, S. Hosogaya, Y. Ozaki, M. Tozuka, N. Osima, H. Yokota, H. Ikeda, and Y. Yatomi Plasma sphingosine-1-phosphate measurement in healthy subjects: close correlation with red blood cell parameters Ann Clin Biochem, July 1, 2008; 45(4): 356 - 363. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Takabe, S. W. Paugh, S. Milstien, and S. Spiegel "Inside-Out" Signaling of Sphingosine-1-Phosphate: Therapeutic Targets Pharmacol. Rev., June 1, 2008; 60(2): 181 - 195. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Venkataraman, Y.-M. Lee, J. Michaud, S. Thangada, Y. Ai, H. L. Bonkovsky, N. S. Parikh, C. Habrukowich, and T. Hla Vascular Endothelium As a Contributor of Plasma Sphingosine 1-Phosphate Circ. Res., March 28, 2008; 102(6): 669 - 676. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 V. Anelli, C. R. Gault, A. B. Cheng, and L. M. Obeid Sphingosine Kinase 1 Is Up-regulated during Hypoxia in U87MG Glioma Cells: ROLE OF HYPOXIA-INDUCIBLE FACTORS 1 AND 2 J. Biol. Chem., February 8, 2008; 283(6): 3365 - 3375. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. L. Allende, D. Zhou, D. N. Kalkofen, S. Benhamed, G. Tuymetova, C. Borowski, A. Bendelac, and R. L. Proia S1P1 receptor expression regulates emergence of NKT cells in peripheral tissues FASEB J, January 1, 2008; 22(1): 307 - 315. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. M. Argraves and W. S. Argraves HDL serves as a S1P signaling platform mediating a multitude of cardiovascular effects J. Lipid Res., November 1, 2007; 48(11): 2325 - 2333. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S.-H. Goh, M. Josleyn, Y. T. Lee, R. L. Danner, R. B. Gherman, M. C. Cam, and J. L. Miller The human reticulocyte transcriptome Physiol Genomics, July 18, 2007; 30(2): 172 - 178. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
