Plasma levels of HDL, and especially HDL2, are generally considered to be atheroprotective primarily through their role in reverse cholesterol transport. Recent evidence suggests that HDL may contribute additionally through its role as the primary transport lipoprotein for sphingosine-1-phosphate (S1P), a key, bioactive lipid which regulates a variety of beneficial cellular functions. However, there also is emerging evidence that S1P may mediate pro-atherogenic metabolism and in support, we now show that S1P, as well as HDL3, which in normal subjects contains higher concentrations of S1P than HDL2, both stimulate the release of plasminogen activator inhibitor-1 (PAI-1) from adipocytes and, importantly, that circulating S1P concentrations are significantly increased in patients with diabetes and correlate with plasma PAI-1 level. PAI-1 is the negative modulator of fibrinolysis and elevated circulating PAI-1 levels are associated with several risk factors for cardiovascular disease. Recent evidence suggests that adipocytes are a major source of PAI-1, especially in obesity. Although little is known about the mechanisms by which S1P in HDL3 stimulate PAI-1 secretion from adipocytes, we now demonstrate that both the adipocyte S1Pr2 and SR-BI, the HDL binding receptor, are required. These observations have contributed to our hypotheses that the increased total plasma S1P concentration in diabetic patients is increased additionally in patients exhibiting the microvascular or macrovascular complications that can accompany diabetes, and that the increased S1P is transported primarily by HDL. We further postulate that S1P carried by HDL3 leads to increased PAI-1 levels in plasma, that plasma S1P and PAI-1 levels are positively correlated, and that S1Pr2 and SR-BI interact at the molecular level to mediate S1P-HDL3-induced secretion of PAI-1 by adipocytes. We will investigate these hypotheses through the conduct of studies with three Specific Aims.
Specific Aim A is epidemiologic in approach and we will analyze samples collected previously as part of our Program Project from Type 1 diabetic patients in the DCCT/EDIC cohort and from Type 2 diabetic patients in the VA Cooperative Study #465 (VADT) to investigate plasma S1P total concentration and relative distribution between the HDL and albumin pools in patients with diabetes complications compared to those without, and determine the association of S1P concentration and distribution with plasma PAI-1 levels.
In Specific Aim B, we will employ sophisticated high-performance liquid chromatography-tandem mass spectrometry to quantitate ceramide and sphingolipid subspecies in VLDL, LDL, HDL2, and HDL3 prepared from Type 2 diabetic patients with and without albuminuria. We will determine if the concentration and distribution in lipoproteins of these additional bioactive lipids is altered in diabetic patients with macroalbuminuria and if they correlate with plasma PAI-1 levels. Purified lipoprotein fractions will be incubated with adipocytes and the sphingolipid profile in the lipoproteins will be correlated with PAI-1 secretion into the medium. We also will incubate adipocytes with recombinant HDL models (rHDL) supplemented with C16-sphingomyelin and/or C24-ceramide to determine the effects of other HDL sphingolipids on S1P- stimulated PAI-1 release.
In Specific Aim C we will determine if S1P transported in HDL3 can activate S1Pr2 using confocal microscopy to monitor GFP-S1Pr2 metabolism. Using BRET analyses, we will determine if S1Pr2 interacts with SR-BI at the molecular level during this activation. Additionally, we will determine if sphingosine kinase activity is increased by HDL3 binding to adipocytes and if the resulting increase in cellular endogenous S1P level contributes to HDL3-stimulated PAI-1 release from adipocytes. We will determine if cholesteryl ester transfer protein (CETP) and the phospholipid transfer protein (PLTP) mediate S1P movement between lipoproteins. Collectively, these studies will elucidate for the first time the patterns of S1P metabolism in lipoproteins and in diabetes, and define the cellular and plasma mechanisms which mediate this metabolism.

Public Health Relevance

Fibrinolysis controls the breakdown of clots in blood and the blood protein called plasminogen activator inhibitor (PAI) inhibits this process. Sphingosine-1-phosphate (S1P) is a very reactive, small fat molecule which has been associated with many beneficial effects in the body, but there is growing evidence that it may also associate with less beneficial metabolism. We have shown that S1P, and especially the S1P found in small high density lipoproteins (HDL), increases the amount of PAI produced by fat cells and, which therefore, could inhibit fibrinolysis. A decrease in fibrinolysis is frequently related to an increase in strokes, heart attacks, or deaths in people. Our studies will investigate for the first time if the amount of S1P transported in HDL is altered in diabetic patients and in diabetes patients suffering from complications of diabetes. We also will study how the S1P in HDL alters PAI made by fat cells. This knowledge may alter the pattern of treatment used to care for Veteran patients with diabetes to try to improve the quality of the lives of these Veterans.

National Institute of Health (NIH)
Veterans Affairs (VA)
Non-HHS Research Projects (I01)
Project #
Application #
Study Section
Endocriniology A (ENDA)
Project Start
Project End
Budget Start
Budget End
Support Year
Fiscal Year
Total Cost
Indirect Cost
Ralph H Johnson VA Medical Center
United States
Zip Code