Unbalanced expression of sphingosine 1-phosphate receptors in diabetic nephropathy

Abstract

Sphingosine 1-phosphate (Sph-1-P) regulates vascular homeostasis through its receptors like S1P1 and S1P2. While S1P1 works to protect vasculature, S1P2 works antagonistically against it. Therefore, the balance of S1P1 and S1P2 determines the regulation of vascular permeability. In diabetic nephropathy, one of the typical pathological changes is endothelial injury possibly as a result of changes in vascular permeability. Therefore, we hypothesized that the balance of S1P1 and S1P2 expression becomes inappropriate in glomeruli of diabetic nephropathy. To verify the hypothesis, five SD rats with diabetes induced by streptozotocin injection and six control rats injected with only the vehicle were analyzed one year after injection. The glomeruli of the diabetic rats exhibited endothelial injuries. The analysis by real-time PCR revealed that the ratio of S1P2/S1P1 mRNA in the renal cortex of the diabetic rats was significantly higher than that in the non-diabetic control group. Immunohistochemistry revealed that S1P1 was expressed by endothelial and mesangial cells, while S1P2 was mainly expressed by mesangial cells in glomeruli. Furthermore, the ratio of the staining intensity of S1P2 to that of S1P1 in the glomeruli was significantly higher in the diabetic rats. The number of cells expressing PDGF-B, which enhances S1P2 expression, was also higher in the glomeruli of the diabetic rats than in the controls. In conclusion, Sph-1-P signals are preferentially transmitted through S1P2, rather than S1P1, in the glomeruli of rats with diabetic nephropathy. Such unbalanced delivery of the Sph-1-P signals might be involved in the pathogenesis of endothelial injuries.

Introduction

Diabetic nephropathy is the most common cause of chronic renal failure requiring renal-replacement therapy. Diabetic nephropathy is characterized by endothelial injuries resulting in the so-called exudative lesions in the glomeruli (Fioretto and Mauer, 2007). In addition to such microvascular damages, diabetes mellitus is also associated with the development of macrovascular injuries; the combination of both micro- and macrovascular complications explains early mortality (Tuomilehto et al., 1998). Evidence of widespread endothelial dysfunction has been found in patients with diabetes (Hink et al., 2001).

The importance of sphingosine 1-phosphate (Sph-1-P) as an intercellular lipid mediator has been established (Yatomi et al., 2000; Yatomi, 2006). Sph-1-P is synthesized by the sphingosine kinase-catalyzed phosphorylation of sphingosine. Blood platelets store Sph-1-P abundantly, although erythrocytes have recently attracted attention as another source of blood Sph-1-P (Pappu et al., 2007). Vascular endothelial cells and smooth muscle cells (SMCs) respond dramatically to this bioactive lipid mainly through the family of G protein-coupled Sph-1-P receptors named S1P1, 2, 3, 4, and 5, which were originally referred to as endothelial differentiation gene (EDG)-1, -5, -3, -6, and -8, respectively. The site of expression and the manner of signal transduction vary among the different isoforms of these Sph-1-P receptors. S1P1 and S1P2 are the two most important S1P receptors expressed in vascular cells (Inoki et al., 2006; Sanchez et al., 2007), with contrasting characteristics. S1P1 is mainly expressed in endothelial cells and participates in maintaining vascular homeostasis through assembly of adherens junctions (Lee et al., 1999), enhancement of survival potency (Hisano et al., 1999), promotion of barrier integrity (Garcia et al., 2001; Singleton et al., 2006), successful maturation (Liu et al., 2000), NO production (Kwon et al., 2001; Levine et al., 2007), cooperation with activated protein C (Feistritzer and Riewald, 2005; Finigan et al., 2005), and other studied mechanisms (Armulik et al., 2005). On the other hand, adult smooth muscle cells express S1P2, rather than S1P1 (Kluk and Hla, 2001). In addition, S1P2 seems to have adverse effects against vascular homeostasis because the signal through S1P2 results in suppression of eNOS (Skoura et al., 2007), SMC contraction (Ohmori et al., 2003), or the inhibition of SMC migration (Ryu et al., 2002). Therefore, S1P2 has an essential role in pathological angiogenesis (Skoura et al., 2007). In this manner, the responses to Sph-1-P differ between S1P1 and S1P2 (Donati et al., 2005; Okamoto et al., 2000; Sugimoto et al., 2003).

Furthermore, in in vitro models, Sph-1-P was reported to stimulate mesangial proliferation (Hanafusa et al., 2002), which should be associated with signal transduction through S1P2 and S1P3 (Katsuma et al., 2005). We previously reported the involvement of Sph-1-P in the contraction of mesangial cells through the activity of S1P2 (Osada et al., 2007). In addition, it was recently reported that the balance of S1P1 and S1P2 determined the regulation of vascular permeability (Sanchez et al., 2007). These previous data suggest that Sph-1-P signals through S1P1 and S1P2 play important roles in the etiopathogenesis of glomerular changes, especially in diabetic nephropathy. However, expression patterns of S1P1 and S1P2 in diabetic nephropathy remain uncertain. Here, we revealed that balance of vasculoprotective S1P1 and antagonistic S1P2 was disturbed in the glomeruli of a rat model of diabetic nephropathy, suggesting that such expression patterns may be involved in the dysfunction of glomerular endothelial cells.