Sphingosine Kinase-1 Pathway Mediates High Glucose-Induced Fibronectin Expression in Glomerular Mesangial Cells

Abstract

Diabetic nephropathy is characterized by accumulation of glomerular extracellular matrix proteins, such as fibronectin (FN). Here, we investigated whether sphingosine kinase (SphK)1 pathway is responsible for the elevated FN expression in diabetic nephropathy. The SphK1 pathway and FN expression were examined in streptozotocin-induced diabetic rat kidney and glomerular mesangial cells (GMC) exposed to high glucose (HG). FN up-regulation was concomitant with activation of the SphK1 pathway as reflected in an increase in the expression and activity of SphK1 and sphingosine 1-phosphate (S1P) production in both diabetic kidney and HG-treated GMC. Overexpression of wild-type SphK1 (SphK^WT) significantly induced FN expression, whereas treatment with a SphK inhibitor, N,N-dimethylsphingosine, or transfection of SphK1 small interference RNA or dominant-negative SphK1 (SphK^G82D) abolished HG-induced FN expression. Furthermore, addition of exogenous S1P significantly induced FN expression in GMC with an induction of activator protein 1 (AP-1) activity. Inhibition of AP-1 activity by curcumin attenuated the S1P-induced FN expression. Finally, by inhibiting SphK1 activity, both N,N-dimethylsphingosine and SphK^G82D markedly attenuated the HG-induced AP-1 activity. Taken together, these results demonstrated that the SphK1 pathway plays a critical role in matrix accumulation in GMC under diabetic condition, suggesting that the SphK1 pathway could be a potential therapeutic target for diabetic nephropathy.

Diabetic nephropathy, one of the major microvascular complications of diabetes, occurs in 30–40% of diabetic patients, and becomes the main cause of end-stage renal disease (1). The pathologic changes of diabetic nephropathy are characterized by early glomerular hypertrophy and accumulation of extracellular matrix (ECM) components, such as fibronectin (FN), and later glomerulosclerosis and tubuloinsterstitial fibrosis (2–4), leading to glomerular hyperfiltration and microalbuminuria (5). Glomerular mesangial cells (GMC) play important roles in physiological and pathological processes of kidney. GMC has been postulated to be a key contributor to the glomerulosclerotic lesion in diabetic patients (6). GMC proliferation and hypertrophy, ECM accumulation, and consequent renal fibrosis have been recognized as major pathogenic events in the progression of renal failure in diabetic nephropathy (2–5). Hyperglycemia is a primary risk factor in diabetic renal disease. However, although intensive control of hyperglycemia reduces the risks of diabetic nephropathy, it does not sufficiently prevent the disease progression. The exact molecular mechanism of diabetic nephropathy remains to be identified.

Sphingosine 1-phosphate (S1P) is one of sphingolipid metabolites that functions as both extracellular and intracellular signaling mediators in the regulation of diverse biological processes, such as cell proliferation, differentiation, migration, adhesion molecules expression, angiogenesis, and vascular lesion, resulting in various chronic vascular diseases, including diabetic nephropathy (7–14). Sphingosine kinase (SphK) catalyzes the synthesis of S1P via phosphorylation of sphingosine (15). Two isozymes of SphK (type 1 and 2) have been cloned and have opposite effects on cell survival. Diverse external stimuli, particularly growth and survival factors, stimulate SphK1 to generate S1P, exerting mitogenic and antiapoptotic effects. Instead, SphK2 suppressed growth and enhanced apoptosis (15–21). The SphK1 pathway has been proposed to be potentially involved in multiple physiological and pathophysiological processes, such as immunity (22, 23), inflammation (24), oncogenesis (19, 25, 26), vascular maturation, and angiogenesis (27). Recently, the SphK1 pathway has also gained considerable attention in its potential role in the pathogenesis of diabetic nephropathy (9, 10). Hyperglycemia, advanced glycation end products (AGE), and oxidative stress can activate SphK1 and increase the intracellular levels of S1P in GMC, leading to induction of GMC proliferation (7–9). The effects of short-term diabetes [4 d after induction by streptozotocin (STZ)] on rat GMC proliferation was found to be associated with a concomitant increase of SphK activity and accumulation of S1P in diabetic glomeruli (10). Furthermore, we have recently observed that SphK1 pathway was activated in diabetic kidney with abnormal glomerular structure and reduced renal function, after 12 wk of alloxan induction in mice (28), suggesting that the activation of SphK1 pathway might be implicated in the pathogenesis of diabetic nephropathy. However, the mechanism by which the activation of SphK1 pathway mediates diabetic renal injury has not yet been elucidated.

Our study here shows that SphK1 pathway plays a pivotal role in high glucose (HG)-induced FN expression in GMC, suggesting that inhibition of SphK1 pathway offers therapeutic potential for diabetic nephropathy.

Results

Renal injury in STZ-induced diabetic rats

The characteristics of diabetic rats at the end of the experimental period are presented in Table 1. In diabetic group, fasting blood glucose levels were significantly higher, whereas the body weights of diabetic rats were less than that of control rats. Although the average size of kidney in diabetic rats was slightly reduced compared with control rats, the difference was not statistically significantly. However, the ratio of kidney to body weight was significantly increased in diabetic rats compared with control rats. No significant difference in serum creatinine levels was detected between the diabetic and control groups. However, increased protein excretion, a hallmark of early diabetic nephropathy, was observed in the diabetic group.

Table 1.

Biochemical parameters in control and diabetic rats after 12 wk of STZ induction

Parameters	Control	Diabetic
Glucose (mmol/liter−1)	5.1 ± 0.18	22.46 ± 1.27b
Body weight (g)	426.3 ± 13.3	165.0 ± 6.0b
Kidney/body weight (mg/g−1)	6.03 ± 0.21	15.09 ± 0.92a
Serum creatinine (μmol//liter−1)	30.0 ± 2.18	31.86 ± 2.34
Albuminuria (mg/24 h−1)	1.79 ± 0.15	12.27 ± 1.09a

Data are mean ± se, n = 8.

^a P < 0.05.

^b P < 0.01 vs. control.

Glomerular injury in diabetic rats was characterized by glomerular hypertrophy and mesangial ECM expansion. As illustrated in Fig. 1, A–D, compared with age-matched control rats, the predominant form of glomerular alteration at 12 wk of induction in STZ-induced diabetic rats included glomerular hypertrophy and focal segmental glomerulosclerosis, with increased glomerular size (glomerular tuft area) and mesangial matrix areas as well as regional adhesion of the glomerular tuft to Bowman's capsule. Furthermore, the mesangial staining markedly increased in diabetic kidney (Fig. 1, A–D). In addition, the accumulation of ECM represented by increased FN expression in kidney was observed in the diabetic group (Fig. 1F). Collectively, these results, consistent with the literature (28), confirmed the renal injury as characterized by renal hypertrophy, ECM accumulation, glomerulosclerosis, and renal dysfunction in STZ-induced diabetic rats.

Fig. 1.

Glomerular injury and activation of SphK1 pathway in STZ-induced diabetic kidney. A, Glomerular histopathology analysis by Periodic Acid-Schiff (PAS) staining. The pictures display representative glomeruli of PAS-stained sections in control and diabetic group at an original magnification of ×200. B, Glomerular size (tuft area) was measured by tracing the tuft. *, P < 0.01 vs. control. C, Mesangial matrix area was defined as the PAS-positive area (red area). *, P < 0.01 vs. control. D, The mesangial matrix index represented the ratio of mesangial matrix area divided by tuft area. *, P < 0.01 vs. control. E, Relative levels of SphK1 and SphK2 mRNA in control and diabetic kidney were analyzed by real-time PCR. *, P < 0.01 vs. control. F, The protein expression levels of SphK1 and FN in the rat kidneys were detected by Western blot analysis. SphK activity (G) and S1P levels (H) were measured by LC-MS/MS assays. *, P < 0.01 vs. control.

Activation of SphK1 pathway in diabetic kidney

To investigate whether SphK1 pathway is involved in diabetic nephropathy, we determined the expression and activity of SphK as well as S1P levels in diabetic kidney. As shown in Fig. 1E, levels of SphK1 mRNA examined by real-time PCR were increased to 3-fold in diabetic kidney in comparison with the controls. However, there were no differences in SphK2 mRNA levels between the two groups (Fig. 1E). Correspondingly, the protein levels of SphK1 in diabetic kidney were significantly increased compared with the control (Fig. 1F). The expression levels of FN were also elevated concomitantly with the increased protein expression of SphK1 in diabetic kidney (Fig. 1F). SphK1 activity and S1P levels in diabetic kidney was substantially increased relative to the controls (Fig. 1, G and H). Consistent with our previous results from the mouse model of alloxan-induced diabetes (28), these findings further demonstrate that the SphK1 pathway is activated in diabetic kidney and suggest a potential role of this pathway in diabetic renal injury.

SphK1 pathway mediates HG-induced FN expression in GMC

Mesangial cells are one of the major constituents of renal glomerulus, and FN is one of key matrix proteins of ECM accumulation. To gain insight into the implication of SphK1 pathway in the pathogenesis of diabetic nephropathy, we examined changes in SphK1 pathway and FN expression in GMC exposed to HG. We firstly examined the effect of HG on SphK1 expression in GMC. Treatment of GMC with 22 mm glucose increased SphK1 protein expression in a time-dependent manner (Fig. 2A). The glucose-induced SphK1 expression was also in a concentration-dependent manner (Fig. 2B). Similarly, HG-induced FN expression in GMC in a time- and concentration-dependent manner.

Fig. 2.

Time- and concentration-dependent effects of HG on SphK1 and FN expression in GMC. A, GMC were cultured by HG (22 mm glucose) for 0–48 h. B, GMC were cultured by 5.5–30 mm glucose for 48 h. At the end of the incubation period, cells were lysed, and protein expression was determined by Western blot analysis. *, P < 0.05; **, P < 0.01 vs. control for SphK1; #, P < 0.05; ##, P < 0.01 vs. control for FN.

To further confirm the effect of HG on SphK1 expression in GMC, cells were treated with 5.5 or 22 mm glucose, or 5.5 mm glucose plus 16.5 mm mannitol (Mtol) for 48 h. As shown in Fig. 3A, SphK1 mRNA was up-regulated 2-fold in response to HG (P < 0.01), whereas SphK2 mRNA was unaltered, which was consistent with the in vivo data (Fig. 1E).

Fig. 3.

HG activates SphK1 pathway in GMC. GMC were serum starved for 24 h, then exposed to media containing 5.5 mm glucose (NG), 22 mm glucose (HG), or NG plus 16.5 mm Mtol for 48 h. A, Real-time PCR was conducted to determine relative mRNA expression levels of SphK1 and SphK2 in GMC. *, P < 0.01 vs. NG. B, SphK1 protein expression was evaluated by Western blot analysis. *, P < 0.01 vs. NG. C, SphK activity was determined by LC-MS/MS. *, P < 0.01 vs. NG. D, Levels of S1P were measured by LC-MS/MS. *, P < 0.01 vs. NG.

In view of HG-induced increases in SphK1 expression, SphK1 activity was measured by liquid chromatography tandem-mass spectrometry (LC-MS/MS). As shown in Fig. 3C, HG markedly induced SphK1 activity (∼2-fold increase, P < 0.01). Consistent with the increase in enzyme activity, intracellular S1P production was increased by 78% in cells exposed to HG (Fig. 3D). In contrast, Mtol, served as a hyperosmotic control, has no effect on the levels of mRNA, protein and activity of SphK1, as well as S1P formation (Fig. 3, A–D). Taken together, these data suggested that SphK1 pathway was activated in GMC exposed to HG, concomitant with FN up-regulation.

Reactive oxygen species (ROS) and AGE mediate HG-induced SphK1 activation

Because HG induces oxidative stress and AGE formation contributing to diabetic nephropathy, we investigated whether ROS and AGE are involved in HG-induced SphK1 activation. Strikingly, the activity and expression of SphK1 induced by HG were significantly inhibited by antioxidant N-acetylcysteine (NAC) and the AGE inhibitor, pyridoxamine (Fig. 4, A and B). These data suggest that oxidative stress and AGE formation could account for HG-induced SphK1 activation in GMC.

Fig. 4.

Roles of ROS and AGE in HG-induced SphK1 activation. GMC were cultured in NG and HG in the presence or absence of NAC (5 mm) or pyridoxamine (PM) (5 mm) for 48 h. A, LC-MS/MS analysis of SphK1 activity. B, Western blot analysis of SphK1 expression. *, P < 0.01 vs. NG; #, P < 0.01 vs. HG.

SphK inhibitor attenuates HG-induced FN expression in GMC

We next sought to explore whether activation of SphK1 pathway is associated with FN up-regulation. To this end, we examined the effects of SphK inhibitor, N,N-dimethylsphingosine (DMS) on HG-induced FN expression in GMC. As shown in Fig. 5A, DMS caused a significant inhibition of SphK activity in GMC stimulated by HG. Exposure of GMC to HG for 48 h resulted in significant increase in the FN expression as evaluated by confocal microscopic analysis. Interestingly, HG-induced FN expression was abolished in the presence of DMS (Fig. 5B). Likewise, Western blot analysis also indicated that HG-induced increases in FN protein expression were significantly reduced by DMS treatment (Fig. 5C). Together, these results demonstrated that SphK inhibitor significantly diminished HG-induced FN expression in GMC, suggesting that SphK1 pathway might be involved in HG-induced FN expression.

Fig. 5.

DMS attenuates HG-induced FN expression in GMC. GMC were treated with HG in the presence or absence of DMS (2.5 μm) for 48 h. A, SphK activity was measured by LC-MS/MS after 48 h of exposure to NG and HG. *, P < 0.01 vs. NG; #, P < 0.01 vs. HG. B, FN expression in GMC was evaluated by confocal microscopic analysis. C, Representative Western blotting for FN expression in GMC. *, P < 0.01 vs. NG; #, P < 0.01 vs. HG.

SphK1, but not SphK2, is responsible for HG-induced FN expression in GMC

To exclude the off-target effects of DMS and examine the specificity of the effects from two isoforms of SphK, we used SphK1-small interference RNA (siRNA) to specifically silence SphK1 expression in GMC and examined its effects on HG-induced FN expression. Levels of SphK1 mRNA and protein expression were markedly decreased by approximately 80% after treatment with SphK1-specific siRNA for up to 72 h (Fig. 6, A and B), whereas SphK2 mRNA levels were not altered (Fig. 6A). Under the condition that cells were treated with SphK1-siRNA for 24 h and cotreated with 22 mm HG for additional 48 h, SphK1-specific siRNA not only resulted in significant attenuation in the basal levels and activity of SphK1 but also completely inhibited HG-induced increases in SphK1 activity (Fig. 6, A–C). Likewise, cells treated with SphK1-specific siRNA showed an approximately 70% reduction of S1P levels under normal glucose (NG) conditions, and HG-induced S1P formation was completely abrogated by the SphK1-siRNA (Fig. 6D). Remarkably, treatment with SphK1-specific siRNA resulted in a significant down-regulation in basal levels of FN expression and completely abrogated HG-induced FN up-regulation (Fig. 6E). Taken together, these results indicated that SphK1 in GMC was responsible for not only the HG-induced increases in SphK activity and S1P formation but also the induction of FN expression by HG.

Fig. 6.

SphK1-siRNA blocks HG-induced FN expression in GMC. A, Real-time PCR shows relative mRNA expression of SphK1 and SphK2 in GMC transfected with SphK1 or control siRNA. *, P < 0.01 vs. control. B, The protein expression of SphK1 in GMC transfected with SphK1 or control siRNA was evaluated by Western blot analysis. *, P < 0.01 vs. control. LC-MS/MS assay was used to analyze SphK activity (C) and S1P levels (D) in GMC transfected with SphK1 or control siRNA, followed by incubation with NG or HG for 48 h. *, P < 0.01. E, Western blotting was performed to evaluate expression of SphK1 (upper panel) and FN (lower panel) in GMC transfected SphK1 or control siRNA, followed by incubation with NG or HG for 48 h. *, P < 0.01.

Overexpression of SphK1 enhances HG-induced FN expression in GMC

GMC transfected with wild-type SphK1 (SphK^WT) resulted in a 2-fold increase in SphK1 activity, which was further enhanced by HG treatment to a similar extent as observed in the parental GMC (Fig. 7A). Consistently, overexpression of SphK^WT significantly increased S1P levels and further exerted the synergistic induction of S1P by HG (Fig. 7B). In contrast, overexpression of dominant-negative SphK1 (SphK^G82D) almost completely abolished the HG-induced SphK activation and S1P formation in GMC (Fig. 7, A and B), confirming the dominant-negative role of SphK^G82D in the transfected GMC. FN expression was markedly enhanced by overexpression of SphK^WT in GMC under normal and HG conditions (Fig. 7C), whereas SphK^G82D significantly inhibited HG-induced FN expression (Fig. 7C). Collectively, these findings further indicated that the activation of SphK1 and the generation of S1P mediated the induction of FN expression in GMC by HG.

Fig. 7.

Effects of SphK1 overexpression on FN expression in GMC. GMC overexpressing empty vector (vector), SphK^WT, or dominant-negative SphK1 (SphK^G82D) were exposed to NG or HG for 48 h. A, SphK activity was analyzed by LC-MS/MS. B, S1P levels were analyzed by LC-MS/MS. C, The expression of SphK^WT, SphK^G82D (upper panel), and FN (lower panel) in the transfected GMC was determined by Western blotting. *, P < 0.01.

S1P stimulates FN expression in GMC

Because S1P is the unique product of SphK1 and accounts for the functional property of SphK1, we examined whether S1P mediates FN expression in GMC. As shown in Fig. 8A, stimulation of GMC with S1P leaded to rapid increases in FN expression. A substantial increase in FN expression occurred after 20 min of S1P stimulation and peaked at 30–60 min of the treatment (Fig. 8A). Figure 8B shows the dose-dependent effect of S1P on FN expression in GMC, further confirming the role of S1P in FN expression.

Fig. 8.

Time- and concentration-dependent effects of S1P on FN expression in GMC. FN expression levels were measured in (A) GMC treated with S1P (1 μm) for the indicated time periods or (B) treated for 60 min with S1P at the indicated concentrations. *, P < 0.05; **, P < 0.01 vs. control.

Effect of S1P receptor 2 (S1P₂) on HG-induced FN expression in GMC

S1P is a signaling molecule that stimulates S1P receptors (S1P_1–5) to control complex physiological and pathophysiological processes. Real-time PCR analysis indicated that of five S1P receptors, S1P₂ mRNA levels increased greatest in diabetic kidney (data not shown). Subsequently, we used the potent and selective S1P₂ antagonist, JTE-013, to test whether FN induction by potent and selective S1P and HG through S1P₂. We found that N-(2,6-dichloro-4-pyridinyl)-2-[1,3-dimethyl-4-(1-methylethyl)-1H-pyrazolo[3,4-b]pyridin-6-yl]-hydrazinecarboxamide (JTE-013) block FN induction by S1P and HG, respectively (Fig. 9). Together, these evidences support that exogenous S1P induces FN expression mainly through binding with S1P₂ to elicit downstream signaling.

Fig. 9.

Effects of S1P₂ receptor antagonist JTE-013 on FN induction by S1P and HG in GMC. Cells treated with 10 μm JTE-013 for 24 h in the presence of S1P. Data are represented as mean ± sd. *, P < 0.05 vs. control; #, P < 0.05 vs. S1P alone.

Activator protein 1 (AP-1) activation is required for S1P-mediated FN up-regulation

FN gene has AP-1 binding site on its promoter region and could be positively regulated by AP-1 (29). Therefore, we sought to examine whether AP-1 is responsible for SphK1-S1P-mediated FN up-regulation in GMC under HG conditions. To this end, we firstly determined the effect of S1P on AP-1 DNA-binding activity in GMC. As shown in Fig. 10A, treatment of GMC with S1P significantly induced AP-1 activity. Incubation of cells with AP-1 inhibitor curcumin (20 μm) significantly prevented the induction of AP-1 activity by S1P. Correspondingly, S1P induced a significant increase in levels of FN mRNA (Fig. 10B) and protein expression (Fig. 10C), whereas curcumin significantly suppressed the induction of FN by S1P (Fig. 10, B and C). These data indicated that S1P-induced FN expression is mediated by AP-1 activity.

Fig. 10.

Effects of exogenous S1P on AP-1 activity and FN expression in GMC. A, GMC were incubated with or without curcumin (20 μm) for 3 h, followed by stimulation with S1P for 1 h. Nuclear proteins (5 μg) were subjected to EMSA. The arrow shows the specific binding of AP-1. *, P < 0.01. B, FN mRNA levels were measured by real-time PCR in GMC preincubated with or without curcumin (20 μm) for 3 h, followed by stimulation with S1P for 1 h. *, P < 0.01. C, FN expression was evaluated in GMC preincubated with or without curcumin (20 μm) for 3 h, followed by stimulation with S1P for 1 h. *, P < 0.01.

HG-induced AP-1 activation is dependent on SphK1-S1P signaling pathway

Having demonstrated that S1P increases AP-1 DNA-binding activity and FN up-regulation, we wanted to determine whether SphK1-S1P pathway is involved in HG-induced AP-1 activation. Using EMSA, we showed that treatment of GMC with HG led to a significant increase in AP-1 activity and that the maximal effect occurred at 6 h of the treatment (Fig. 11A). The specificity of AP-1 DNA-binding complex was confirmed by competition and supershift assays. As shown in Fig. 11B, HG-induced AP-1 activity was completely abrogated by unlabeled AP-1 oligonucleotide and antibodies against c-Fos and c-Jun. In contrast, the unlabeled nuclear factor κB oligonucleotide did not alter HG-induced AP-1 activation, confirming the specificity of these EMSA. Interestingly, HG-induced AP-1 activity was significantly suppressed by the SphK inhibitor DMS (Fig. 11C). Furthermore, HG was incapable of activating AP-1 in the GMC overexpressing SphK^G82D (Fig. 11D). Taken together, these data indicated that SphK1 pathway played a pivotal role in AP-1 activation, which mediates the HG-induced FN expression in GMC.

Fig. 11.

SphK1 pathway mediates HG-induced activation of AP-1 in GMC. AP-1 DNA-binding activity was assayed by EMSA in GMC (A) exposed to HG for the indicated time periods. B, The specific binding of consensus AP-1 was verified by competition assays and the subunit components for AP-1 dimers identified by supershift assays. C, GMC were treated with HG in the presence or absence of the DMS (2.5 μmol/liter) for 6 h. *, P < 0.01 vs. NG; #, P < 0.01 vs. HG. D, GMC were transfected with SphK^G82D or empty vector followed by incubation with HG for 6 h. The arrows indicate the specific binding of AP-1. *, P < 0.01 vs. vector + NG; #, P < 0.01 vs. vector + HG.

Discussion

Previous experiments performed on STZ-induced diabetic rats have suggested the potential relevance of the SphK1 pathway to diabetic nephropathy (10). However, it is still unclear whether activation of SphK1 pathway is associated with advanced renal injury in diabetes. Our recent study with the alloxan-induced diabetic mouse model revealed that SphK1 pathway was activated in the diabetic kidney after 12 wk of diabetes induction when the renal structure and function were damaged (28). In the current study, we found that, concomitant with renal dysfunction and glomerular ECM accumulation, SphK1 pathway was also activated in diabetic kidney in the STZ-induced diabetes model. These observations suggested a potential role of SphK1 pathway in the pathogenesis of diabetic renal injury.

Long-term hyperglycemia, AGE, and oxidative stress have been shown to activate SphK1, which converts sphingosine to S1P (7–10), eliciting expression of proinflammatory adhesion molecules in endothelial cells (7); vascular smooth muscle cells proliferation (8); and GMC proliferation (9, 10). In the present study, we found that both AGE and oxidative stress can mediate HG-induced increases in the activity and expression of SphK1, suggesting a new pathway underlying the adverse effect of HG on kidney. Recently, Clavreul et al. (30) reported that (pro)rennin promotes fibrosis gene expression in HEK cells through a reduced nicotinamide adenine dinucleotide phosphate oxidase (Nox)4-dependent mechanism, suggesting that Nox4 pathway is important in the regulation of FN expression. Various ROS-producing systems exist in cells, and among them is reduced nicotinamide adenine dinucleotide phosphate oxidase that produces superoxides when it is in an activated state (31). Although all isoforms are present in kidney cells, the constitutively active isoform Nox4 is predominantly expressed, especially in mesangial cells and participates in the process of renal fibrosis through oxidative stress (30, 32). Furthermore, oxidative stress is one of the major factors stimulating SphK1 pathway. The study (30) mainly elucidated the upstream mechanism of SphK1 activation. In addition, our result that antioxidant (NAC) significantly attenuates HG-induced activity and expression of SphK1 in GMC further demonstrates that oxidative stress mediates the activation of SphK1 pathway. Multiple lines of evidence have suggested that S1P contributes importantly to the pathogenesis of mesangium proliferative diseases (33–35). Exogenous S1P has been shown to induce GMC proliferation (33–35), leading to renal diseases. The finding that HG treatment resulted in significant increases in expression and activity of SphK1 as well as S1P levels in GMC indicates that hyperglycemia is a key factor responsible for the activation of SphK1 pathway in diabetic kidney. HG has no effect on SphK2, suggesting a specificity of SphK1 in this pathway. In addition, the SphK1 pathway was not activated by Mtol, indicating that the effect of HG was not due to high osmotility.

The pathogenic effect of HG in kidney has been recognized attributable to up-regulation of glomerular ECM proteins, such as FN. The finding that SphK inhibitor DMS significantly inhibited HG-induced FN expression in GMC suggests an important role of SphK1 activity in this pathogenic event. SphK1-siRNA not only significantly down-regulated basal levels of FN but also markedly abolished HG-induced FN expression, demonstrating that endogenous SphK1 is a key regulator in HG-induced FN expression in GMC. This notion was further supported by the experimental data from GMC overexpressing SphK^WT or SphK^G82D. Overexpression of SphK^WT resulting in an up-regulation of SphK1 activity and S1P production caused a significant increase in FN expression under NG conditions, which was similar to that observed in the parental GMC exposed to HG. Interestingly, cells overexpressing SphK^WT treated with HG exerted the synergistic induction of SphK activity, S1P formation, and FN expression. In the SphK^G82D-transfected GMC, HG-induced SphK activity and S1P formation were completely blocked. Strikingly, the HG-induced FN expression was almost abolished by SphK^G82D. These results strongly suggested that the activation of SphK1 was critical for HG-induced FN up-regulation in GMC. Furthermore, we have observed for the first time that S1P stimulated the FN expression in GMC in a time- and dose-dependent fashion, revealing a new mediator of FN expression. Taken together, our results illustrated a key role of the SphK1 pathway in FN expression by GMC under diabetic status.

Transcription factor AP-1 is a menagerie of dimeric basic region-leucine zipper proteins, consisting of homodimers of Jun or heterodimers of Fos and Jun (36, 37). A binding site of AP-1 in FN promoter region has been identified and implicated in the regulation of FN gene expression (29). It has been reported that AP-1 is activated by HG in GMC and in diabetic kidneys (38) and that HG-induced FN up-regulation is dependent on AP-1 activation (39). Interestingly, we found that inhibition of AP-1 by curcumin abrogated both S1P and HG-mediated increase in FN expression, further indicating that AP-1 was a key regulator of FN gene expression in GMC. Furthermore, we provide evidence showing that HG-induced AP-1 activation was markedly inhibited by DMS treatment or overexpression of SphK^G82D in GMC. Thus, these findings not only confirmed the key role for SphK1 pathway in the activation of AP-1 but also verified that AP-1 activation was an important downstream signaling component of the SphK1-mediated induction of FN expression by HG. Further studies are warranted to explain the precise signaling mechanism involving how SphK1 pathway induces AP-1 activity.

In summary, the present study illustrates that HG activates SphK1, leading to AP-1 activation and FN up-regulation in GMC, which might be responsible for the pathogenesis of diabetic nephropathy. It provides a novel mechanism underlying the hyperglycemic damage on the kidney and might offer a new therapeutic strategy targeting SphK1 pathway for the prevention and treatment of diabetic nephropathy.

Materials and Methods

Animal experiment

Animal studies were carried out in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals. Male Sprague Dawley (SD) rats (175–200 g) were purchased from Center of Experimental Animals in Sun Yat-sen University and housed under specific pathogen-free conditions. Experimental diabetes was induced by a single tail-vein iv injection of freshly prepared STZ [60 mg/kg⁻¹ in 10 mmol/liter of citrate buffer (pH 4.5)] (Sigma, St. Louis, MO) into SD rats after an overnight fasting. Control rats were injected with equal volume of citrate buffer alone. The levels of fasting blood glucose were determined at 72 h after the injection of STZ or citrate buffer, and rats with fasting blood glucose levels more than 16.7 mmol/liter were considered diabetic.

Biochemical analysis and morphological studies

At the termination of the experiment, rats were weighed, and housed in metabolic cages for 24 h to collect urine. Serum creatinine and 24-h albuminuria were analyzed by the Department of Pathology at the First Affiliated Hospital, Sun Yat-sen University. For histological studies, the sections were examined by light microscopy for the degree of glomerulosclerosis and expansion of the mesangial matrix in the glomeruli. Glomerular cell number was determined by counting hematoxyllin-stained nuclei. Glomerular hypertrophy was calculated from the cross-sectional area of the glomerular tuft. Glomerular tuft areas and mesangial area were measured with image analysis software Image-Pro Plus (Media Cybernetics, Inc., Bethesda, MD). Renal hypertrophy was assessed by kidney-to-body weight ratio (mg/g⁻¹).

Cell culture

Rat GMC were obtained from young SD rat kidneys as described previously (9, 40). For the experimental studies, GMC were grown to confluence in the presence of 10% fetal calf serum and then made quiescent by serum starvation in DMEM for 24 h. Medium was then changed to: 1) DMEM containing 5.5 mm glucose (NG), 2) NG medium supplemented with additional 16.5 mm glucose to a final concentration of 22 mm (HG), or 3) NG medium containing 16.5 mm Mtol.

Plasmids, siRNA, and transient transfection

Vector (pcDNA3 plasmid), FLAG-tagged human wild-type SphK1 cDNA (SphK^WT), and the dominant-negative SphK1 (SphK^G82D) were generated as described previously (7). The Validated Stealth Negative Control and double-stranded SphK1-specific siRNA oligonucleotides were purchased from Invitrogen (Carlsbad, CA) with the following sequences: sense, 5′-AGGGACAGCAGAUUCAUGGGUGACA-3′ and antisense, 5′-UGUCACCCAUGAAUCUGCUGUCCCU-3′ (SphK1). GMC was transfected using Lipofectamine 2000 (Invitrogen) according to the manufacture's protocol and incubated for 72 h to harvest.

Quantitative real-time PCR

Total RNA was extracted from kidney tissue or cells by the TRIzol (Invitrogen) method according to the manufacturer's instructions. Total RNA was subjected to reverse transcription, followed by quantitative real-time PCR using Bio-Rad iCycler Iq system (Bio-Rad, Hercules, CA). The oligonucleotide primers used for cDNA PCR amplifications were as follows: SphK1 forward primer, 5′-GGCAGCGAACCCCACCACTC-3′ and reverse primer, 5′-GCGGGTGTCTGGTGACTGGC-3′; SphK2 forward primer, 5′-CACCTGTGCTGGGTGCGGAG-3′ and reverse primer, 5′-GGCAGCCCAGGCTGAAGTGG-3′; and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) forward primer, 5′-AGGAGTAAGAAACCCTGGAC-3′ and reverse primer, 5′-CTGGGATGGAATTGTGAG-3′. The PCR reactions were performed using SYBR Premix Ex Taq II (TaKaRa Biotechnology Co., Ltd, Dalian, China).

Western blotting

Western blot analysis was performed as described previously (28). Briefly, 30 μg of protein from each sample were blotted with primary antibody as follows: rabbit polyclonal antibody against SphK1 (Abcam, Cambridge, MA), mouse monoclomal antibody against FN (Santa Cruz Biotechnology, Inc., Santa Cruz, CA).

Assay of SphK activity

SphK activity was determined as described previously (41). The reaction was initiated by adding 2.5 μl of 200 μm C17-Sph (dissolved in 5% Triton X-100) and 2.5 μl of 20 mm ATP containing MgCl₂ (200 mm) in a final volume of 50 μl. After incubation at 37 C for 20 min, the reaction was terminated with 5 μl of 1 m HCl followed by 200 μl of chloroform:methanol:HCl (100:200:1, vol/vol), then 10 ng of S1P were added as an internal standard. After vigorous vortexing, 60 μl of chloroform and 60 μl of 2 m KCl were added, and phases were separated by centrifugation at 12,000 × g for 5 min at 4 C. The lower chloroform phase was transferred into a new 1.5-ml Eppendorf tube and vacuum dried in a SpeedVac for 60 min. The dried residue was reconstituted in 100 μl of the mobile phase, and an aliquot (10 μl) of the final solution was injected directly into the LC-MS/MS system (ThermoFinnigan) for analysis.

Assay of S1P level

S1P levels were determined as described previously (42). S1P was extracted with methanol precipitation. The mixture was vortex mixed for 10 sec, followed by centrifugation at 12,000 rpm for 5 min at 4 C; then, the supernatant was transferred to clean glass vials, and 10 μl of the supernatant were injected into the LC-MS/MS system for analysis.

Confocal laser scanning fluorescence microscopy

GMC were grown on glass coverslips. After treatment, the cells were fixed with 4% paraformaldehyde. The cells were incubated with anti-FN antibody (Santa Cruz Biotechnology, Inc.) at room temperature for 2 h and then incubated with the fluorescein isothiocyanate-conjugated secondary antibody (Invitrogen) for 1 h. The coverslips were mounted on glass slides with antifade mounting media (Invitrogen), and images were collected using Zeiss LSM 710 laser confocal fluorescence microscope (Carl Zeiss, Oberkochen, Germany).

Electrophoretic mobility shift assay

Nuclear proteins for EMSA were prepared using the Nuclear Extract kit (Active Motif, Carlsbad, CA) according to manufacturer's instruction. The nuclear proteins (5 μg) were incubated with 1× binding buffer (LightShift Chemiluminescent EMSA kit; Pierce, Rockford, IL) in the presence of 50 ng/μl poly(dI-dC), 0.05% Nonidet P-40, 5 mm MgCl₂, and 2.5% glycerol for 10 min and then incubated at room temperature for additional 20 min with 1 pmol biotin-labeled AP-1 consensus oligonucleotide (Sangon Biotech Co., Ltd., Shanghai, China). The reaction mixture was subjected to a 6% nondenaturing SDS-PAGE, transferred to nylon hybridization transfer membrane (Amersham, Piscataway, NJ), and DNA cross-linked for 10 min and probed with horseradish peroxidase-conjugated streptavidin antibodies (1:300 dilution), then visualized with enhanced chemiluminescence. For supershift analysis, the indicated antibodies (rabbit anti-c-Jun and anti-c-Fos; Santa Cruz Biotechnology, Inc.) were preincubated at 37 C for 30 min before the reactions with the biotin-labeled probes. Competition experiments were performed by the addition of an excess amount of the indicated unlabeled double-stranded oligonucleotide to the reaction mixtures. AP-1 consensus, 5′-CGCTTGATGAGTCAGCCGGAA-3′.

Statistical analysis

Values are expressed as mean ± se. Statistical differences between two groups were analyzed by the unpaired Student's t test, and differences between multiple groups of data were analyzed by one-way ANOVA with Bonferroni correction. P < 0.05 was considered statistically significant.