点击显示 收起
【摘要】
Objective- Tumor necrosis factor (TNF- ), a key proinflammatory cytokine acting on the endothelium, activates endothelial nitric oxide synthase (eNOS). We have examined the signaling pathway leading to this activation and its biological role in endothelium, which are still unknown.
Methods and Results- In human endothelial cells, we found that eNOS activation by TNF- is time dependent and requires activation of Akt, a known eNOS activator. eNOS activation was preceded by sequential activation of neutral-sphingomyelinase-2 (N-SMase2) and sphingosine-kinase-1 (SK1) and generation of sphingosine-1-phosphate (Sph1P). Inhibition of N-SMase2 inhibited Sph1P formation, whereas inhibition of SK1 did not affect N-SMase2 activation by TNF-. Blockade of N-SMase2, SK1, or the Sph1P receptors S1P 1 and S1P 3, either by silencing or pharmacological inhibitors, prevented eNOS activation. Thus, eNOS is activated by TNF- via S1P receptors, activated by Sph1P generated through N-SMase2 and SK1 activation. We found that nitric oxide generated through this pathway has a biological role, because it inhibits the expression of E-selectin and the adhesion of dendritic cells to the endothelium stimulated by TNF-.
Conclusions- This study establishes a previously undescribed link among TNF-, Sph1P, and eNOS in a same signaling pathway of biological relevance in the process of endothelial cell activation by TNF-.
How tumor necrosis factor (TNF)-activates endothelial NO synthase to generate NO in the endothelium was not known. This study shows that TNF- activates endothelial NO synthase via Sph1P. Because Sph1P and NO interact to regulate important aspects of endothelial cell activation by TNF-, their link with TNF- in a same signaling pathway is of biological relevance.
【关键词】 endothelial NO synthase TNF neutral sphingomyelinase sphingosine kinase sphingosine phosphate
Introduction
Nitric oxide (NO) generated in the endothelium by the endothelial NO synthase (eNOS) plays crucial roles not only in physiology but also in regulating specific inflammatory events involving the endothelium, such as the expression of adhesion molecules and leukocyte adhesion. 1,2 Indeed, eNOS -/- animals, when exposed to inflammatory stimuli, display significant increases in extravasated neutrophils and poor wound healing and are more prone to diseases in which inflammation is significant, including atherosclerosis and diabetes. 1,2
eNOS regulates the effects on migration, angiogenesis, vasodilation, endothelial barrier permeability, apoptosis, and expression of inflammatory and adhesion molecules exerted by tumor necrosis factor (TNF- ), a key proinflammatory cytokine acting on the endothelium. 3-9 In some cases, the effects of NO appeared directly associated with eNOS activity; 8,9 in others, they were consistent with the involvement of this enzyme, because they were too rapid to be explained by synthesis of the inducible NO synthase (NOS) 1,2 or because they took place in human cells 4,7 that do not express inducible NOS when exposed to TNF- alone. 8,10 Despite the relevance to TNF- action, eNOS activation by the cytokine in endothelial cells had been investigated in only 2 studies 8,11 that did not provide evidence of the pathway involved, apart from the identification of a phosphatidylinositol 3-kinase (PI3K)/Akt-dependent step, 11 a signaling event common to many eNOS-activating stimuli. 2
In a recent study in HeLa cells, we reported that eNOS activation by TNF- is mediated by the activation of neutral sphingomyelinase (N-SMase), with generation of the sphingolipid ceramide. 12 Sphingolipids play important functions in vascular biology 13 and interact with NO signaling in the regulation of key biological events; 14 a detailed analysis of how they are involved in eNOS activation in the endothelium might lead to a new understanding of specific aspects of endothelial cell pathophysiology.
Here we define the molecular steps in the signaling pathway linking TNF- to eNOS activation in human umbilical vein (HUVEC) and microvascular (HMVEC-C) endothelial cells, identifying a key role for generation of sphingosine-1-phosphate (Sph1P), a sphingolipid involved in proliferation, survival, migration, and differentiation of these cells. 13,15 In addition, we demonstrate that the link, unsuspected before, among TNF-, Sph1P, NO, and their generating enzymes within a same signaling cascade is relevant for the regulation of adhesion of dendritic cells (DCs) to the endothelium.
Methods
The procedures and the sources of the Materials are described in the expanded Materials and Methods section in the online data supplement available at http://atvb.ahajournals.org.
Results
Activation of N-SMase and sphingosine-kinase (SK) and changes in ceramide, sphingosine, and Sph1P levels induced by TNF- are time dependent and precede activation of Akt and eNOS in HUVECs.
We compared the time course of activation of eNOS with those of N-SMase, SK, and Akt in HUVECs treated with TNF- (50 ng/mL). Because eNOS is the only isoform expressed by HUVECs 6 hours after treatment with TNF- ( Figure 1 C, inset), it accounts for the whole NOS activity in these cells in the time span of the experiments described in this study. Activation of N-SMase, SK, Akt, and eNOS by TNF- was time dependent and transient; the activation of N-SMase and SK ( Figure 1A and 1 B) preceded that of eNOS and Akt, the latter activated in a time course consistent with previous results ( Figure 1C and 1 D). 16
Figure 1. Activation of N-SMase, SK, eNOS, and Akt by TNF- in HUVEC. A through C, N-SMase, SK, and eNOS activities, expressed as percentage over activities measured in untreated controls ( n =5). Inset in C, expression of eNOS, inducible, and neuronal NOS (iNOS and nNOS) after 6-hour exposure of HUVEC to TNF-. p.c. indicates positive controls (human aortic endothelial cells, rat pituitary cells, and activated RAW 264.7 macrophages). D, Western blot analysis of active, phosphorylated Akt (P-Akt), and eNOS (P-eNOS) and total, active, and inactive enzymes. Shown are both representative images and quantitative values, expressed as percentage of the mean ratio ± SEM of densitometric values of P-eNOS and P-Akt vs total eNOS and Akt, respectively ( n =4). *, **, ***, statistical probability vs control ( P <0.05, 0.01, and 0.001, respectively).
In parallel experiments, we measured the time course of the generation of ceramide, sphingosine, and Sph1P, that is, products and substrates in the enzymatic reactions catalyzed by N-SMase and SK. To assess the generation of NO by eNOS, we measured the formation of cGMP, a proxy for NO, because soluble guanylate cyclase is activated by nanomolar concentrations of the gas. 1,2 Treatment with TNF- (50 ng/mL) resulted in time-dependent changes in ceramide, sphingosine, and Sph1P that were consistent with the time course of activation of N-SMase and SK activities and occurred earlier than activation of eNOS and Akt and generation of NO/cGMP (Figure I, available online at http://atvb.ahajournals.org).
eNOS Activation by TNF- Requires Sequential Activation of N-SMase2 and SK1 With Generation of Sph1P in HUVECs
We studied the role of N-SMase and SK in the activation of Akt and eNOS and investigated the molecular identity of the enzymes involved, focusing on the recently cloned N-SMase2 and SK1, which account for N-SMase and SK activities in a variety of cells. 17-20 In these experiments, we used well-characterized tools, namely a small interfering RNA (siRNA) specific for N-SMase2 (N-SMase2 siRNA) and a dominant-negative SK1 (SK1-DN). 17,18 mRNA levels of N-SMase2 were reduced with respect to those observed in cells treated with the control-scrambled siRNA ( Figure 2 A, inset). Expression of SK1-DN was checked by detecting its FLAG epitope by Western blotting and FACS analyses ( Figure 2 B, insets). The efficacy of the constructs to inhibit N-SMase and SK activities in HUVECs was tested 5 minutes after administration of TNF-, that is, when both enzymes were active (see Figure 1 ), and compared with that of the N-SMase inhibitor manumycin A and the SK inhibitor dimethylsphingosine. 12 In cells transfected with the N-SMase2 siRNA, activation of N-SMase by TNF- (50 ng/mL) was inhibited by 78±8.5% with respect to that of cells transfected with a control siRNA ( n =6). Manumycin A (5 µmol/L) inhibited N-SMase by 91±5.2% ( n =5). In cells transfected with SK1-DN, activation of SK was inhibited by 73±7.2% with respect to that of cells transfected with the empty vector ( n =6). Dimethylsphingosine (5 µmol/L) inhibited SK activity by 89±7.5% ( n =6). Basal N-SMase and SK activities were not significantly affected by any of these treatments. The comparison with the pharmacological inhibitors indicates that N-SMase2 siRNA and SK1-DN are efficient inhibitors in HUVECs and that N-SMase2 and SK1 account for most of the respective enzymatic activities stimulated by TNF- in these cells.
Figure 2. TNF- activates eNOS through sequential activation of N-SMase2 and SK1. HUVECs were transfetced with SK1-DN, empty vector (e.v.), N-SMase2 siRNA, or scrambled siRNA (Sc). Inset in A, reduction of N-SMase2 mRNA levels induced by the N-SMase2 siRNA vs Sc and lack of effects on the expression of the glyceraldehyde 3-phosphate dehydrogenase (GAPDH) mRNA. p.c. indicates positive control (MCF-7 cells); NT, untreated HUVECs. Inset in B, expression of SK1-DN, detected using an anti-FLAG antibody, both by Western blot and by FACS analyses ( n =6). Relative fluorescence intensity (RFI) values represent the efficiency of transfection ± SEM. Cells were incubated for 24 hours with (A and B) or without (C through F) [ 3 H]-sphingosine and then treated with or without TNF- for 5 minutes (A and B) or 30 minutes (C through F). A and B, cellular Sph1P, sphingosine (Sph), and ceramide (Cer; n =4). C and D, eNOS activity ( n =4). E and F, analysis of active, phosphorylated Akt (P-Akt) and eNOS (P-eNOS) and total, active, and inactive enzymes as described in Figure 1 ( n =4). *** and xxx, statistical probability vs control and cells treated with only TNF-, respectively ( P <0.001).
Then we measured the effect of N-SMase2 and SK1 inhibition on the levels of cellular sphingolipids. Cell transfection with N-SMase2 siRNA inhibited the changes induced by TNF- in the levels not only of ceramide but also of sphingosine and Sph1P ( Figure 2 A), whereas transfection with SK1-DN blocked generation of Sph1P with accumulation of sphingosine and ceramide ( Figure 2 B). Previous studies showed that TNF- directly activates both sphingomyelin turnover and SK 21 without investigating their possible link in Sph1P generation; our results indicate that N-SMase2 and SK1 operate in sequence in a same signaling pathway, leading to the generation of Sph1P by TNF-. Of importance, transfection with either N-SMase2 siRNA or SK1-DN inhibited Akt and eNOS phosphorylation ( Figure 2E and 2 F) and NOS activity ( Figure 2C and 2 D) triggered by TNF-, indicating that activation of the N-SMase2-to-SK1 signaling pathway is necessary for Akt and eNOS activation by the cytokine.
eNOS Activation by TNF- Requires Stimulation of S1P 1 and S1P 3 by Sph1P Upstream to PI3K/Akt in HUVECs
In endothelial cells, Sph1P is generated by SK1 both intracellularly and extracellularly 22 and acts as an intracellular messenger or as an extracellular agonist of S1P receptors. 19 Consistent with previous reports, 23,24 Western blotting analyses confirmed that, of the known S1P receptors, HUVECs express S1P 1 and S1P 3 (data not shown). Both cell pretreatment with pertussis toxin (PTx, 100 ng/mL), which inhibits Gi-coupled receptor signaling and eNOS activation by exogenous Sph1P, 25,26 and cell transfection with antisense oligonucleotides, which specifically inhibit expression of S1P 1 and S1P 3 receptors ( Figure 3 B, inset), inhibited Akt and eNOS activation by TNF- (50 ng/mL; Figure 3 A through 3D). These inhibitions were not reversible by exogenously administered Sph1P (100 nmol/L). When administered alone, however, exogenous Sph1P triggered Akt and eNOS activation ( Figure 3A through 3 D). These results indicate that the activation of Akt and eNOS by TNF- is mediated via an extracellular action of Sph1P on S1P 1 and S1P 3.
Figure 3. TNF- activates eNOS through S1P 1 and S1P 3 membrane receptors. HUVECs were transfected with specific antisense oligonucleotides (AS-S1P 1 and AS-S1P 3 ), scrambled oligonucleotides (Sc), or incubated overnight with PTx or vehicle (NT). Inset in B, S1P 1 and S1P 3 expression by Western blot analysis. The various cell transfectants were treated with or without TNF- or Sph1P for 30 minutes. A and B, eNOS activity ( n =4). C and D, analysis of active, phosphorylated Akt (P-Akt) and eNOS (P-eNOS), total, active, and inactive enzymes and actin, as described in Figure 1 ( n =4). *** and xxx, statistical probability vs NT and cells treated with only TNF-, respectively ( P <0.001).
Activation of the PI3K/Akt pathway is a critical event in eNOS activation by Sph1P; 25,26 to clarify its role, we used 2 unrelated PI3K inhibitors, LY 294002 (5 µmol/L) and wortmannin (100 nmol/L). 12 These inhibitors prevented activation by TNF- (50 ng/mL) of both Akt and eNOS in ways not reversible by addition of Sph1P ( Figure II, available online at http://atvb.ahajournals.org). The results shown in Figures 3 and II indicate the obligatory role of PI3K/Akt in eNOS activation by TNF- and that PI3K/Akt acts downstream to Sph1P generation and S1P activation.
Although increases in [Ca 2+ ] c play a role in eNOS activation by Sph1P, 25 activation of eNOS by TNF- occurred without a concomitant rise in [Ca 2+ ] c (data not shown), ruling out a major involvement of the cation in the cytokine-induced pathway of eNOS activation. As a control of specificity of the various strategies used to define the pathway of eNOS activation by TNF-, we assessed whether they influenced eNOS activation by ATP, that is, a stimulus that triggers enzyme activity in a Ca 2+ -dependent, PI3K/Akt-independent way. 12 Five-minute stimulation with ATP (100 µmol/L) triggered eNOS activity (L-citrulline formation was 0.010±0.001 and 0.083±0.003 pmol/mg per min -1 before and after ATP administration, respectively; n =4). Cell transfection with SK1-DN, N-SMase 2 siRNA, or antisense oligonucleotides to S1Ps and cell treatment with PTx or wortmannin did not modify significantly the effect of ATP (data not shown).
Generation of Sph1P and NO Regulates Adhesion of DCs and Expression of Adhesion Molecules Induced by TNF- in HUVECs
We investigated the role of eNOS activation and Sph1P generation in the adhesion of immature DCs and the expression of adhesion molecules. HUVEC confluent monolayers transfected or not with SK1-DN were incubated for 4 hours in the presence or absence of Sph1P, TNF-, and the NOS inhibitor N -nitro-l-arginine methyl ester (l-NAME; 2 mmol/L), in various combinations, before a 2-hour coincubation with DCs. TNF- stimulated adhesion of DCs to HUVECs in a concentration-dependent way ( Figure 4A, 4B, and 4 D). Expression of SK1-DN or exposure to l-NAME increased the effect of the cytokine. Also, these effects were dependent on the concentration of TNF- and correlated directly with the concentration-dependency of the activation of Akt and eNOS induced by the cytokine ( Figure 4C and 4 D). No additional increases were observed after the administration of l-NAME in SK1-DN-transfected cells. Exogenous Sph1P, when administered alone, did not trigger DC adhesion, whereas, when administered to SK1-DN-expressing cells, it reduced the effect of TNF-. Sph1P, however, did not reduce the effect of TNF- in L-NAME-treated cells ( Figure 4A and 4 B). Administration of NO, using the NO donor DETA-NO (50 µmol/L), reverted the effects of SK1 inhibition. Thus, generation of NO through the SK1/Sph1P pathway regulates in an inhibitory fashion the ability TNF- to stimulate adhesion of DCs to endothelial cells. The role of this regulation depends on the concentration of TNF-.
Figure 4. Role of endogenously generated NO and Sph1P on adhesion of DCs to HUVECs. HUVECs transfected with SK1-DN or empty vector (e.v.) were incubated for 4 hours with or without TNF-, Sph1P, or l-NAME in various combinations as detailed in the keys. A, B, and D, cells were then incubated for an additional 2 hours with immature DCs loaded with the fluorescent, vital dye 5-chlormethylfluorescein. Shown are both representative images (A) and quantitative values (B and D), expressed as mean±SEM of DC adhesion to the endothelial monolayers ( n =4). C, eNOS and Akt activation, measured as described in Figure 1 (n=3). ***, xxx, and +++, statistical probability vs control, cells treated with only TNF-, and SK1-DN-expressing cells treated with TNF-, respectively ( P <0.001).
Sph1P and NO generation also affected the expression of adhesion molecules stimulated by 4-hour treatment with TNF- (50 ng/mL). In particular, the TNF- -triggered expression of E-selectin and vascular cell adhesion molecule (VCAM) 1 was prevented by cell transfection with SK1-DN, whereas no effect was observed on intercellular adhesion molecule-1 (ICAM-1) expression (Figure III, available online at http://atvb.ahajournals.org). Sph1P, although necessary to expression of E-selectin and VCAM-1 by TNF-, did not have a direct effect on adhesion molecules in the absence of the cytokine, at least at the concentration used (100 nmol/L). l-NAME did not change the levels of expression of ICAM-1 and VCAM-1 induced by TNF- but increased significantly those of E-selectin.
The TNF- -Mediated eNOS Regulation Is Operative in HMVEC-Cs
We investigated whether the pathway of activation of eNOS by TNF- and its biological consequences were also operative in other types of endothelial cells. We studied HMVEC-C because of the role of microvasculature in inflammation. 3 TNF- (50 ng/mL) triggered eNOS activation that was dependent on N-SMase2, SK1, and PI3K and inhibited by PTx (Figure IVA, available online at http://atvb.ahajournals.org). TNF- stimulated DC adhesion to HMVEC-Cs (Figure IVB). Expression of SK1-DN or exposure to l-NAME increased the effect of the cytokine to similar extents. Administration of exogenous Sph1P to SK1-DN-expressing cells reduced the effect of TNF-. Sph1P, however, did not reduce the effect of TNF- in l-NAME-treated cells (data not shown). TNF- triggered expression of E-selectin (Figure IVC). The effect of the cytokine was prevented by cell transfection with SK1-DN, restored by Sph1P, and increased by l-NAME. The results obtained in HMVEC-Cs are consistent with those observed in HUVECs.
Discussion
In this study, we describe how TNF- activates eNOS in endothelial cells, showing that it occurs through a pathway undescribed before for other eNOS-activating stimuli, which includes sequential stimulation of N-SMase2 and SK1. The product of SK1, Sph1P, activates Akt through its S1P 1 and S1P 3 receptors, and Akt, in turn, activates eNOS. Those identified here are most of the events leading to activation of eNOS by TNF-; our results, however, do not exclude that other enzymes and messenger molecules contribute to intermediate steps. Generation of Sph1P by TNF- was accompanied by decreases in the concentrations of sphingosine, implying the involvement of a ceramidase activity in the pathway of metabolic conversion of sphingolipids. The identity of the responsible enzyme(s) remains to be established. Similarly, it remains to be established how generation of Sph1P leads to S1P receptor activation, that is, whether the sphingolipid is generated by an extracellular action of SK1 22 or intracellularly and then exported through as-yet-unidentified pathways. 13
Sph1P and NO regulate in an inhibitory fashion key aspects of TNF- signaling in the endothelium, such as apoptosis, leukocyte adhesion, and expression of adhesion molecules. 5,6,15,19,21,24,27,28 Up to now, however, the role of Sph1P and NO had been investigated separately. We have now established a link between Sph1P and NO as messengers generated within the same signaling cascade, and we have shown that this link is of biological relevance. We found that the interplay between Sph1P and NO is important for at least 2 aspects, that is, adhesion of DCs and expression of adhesion molecules. In fact, NO and Sph1P did not have effects in the absence of TNF-. However, adhesion of DCs, key actors in the initiation of immune responses, 29 to TNF- -activated endothelium was regulated in an inhibitory fashion by the Sph1P generated by the cytokine itself, and this regulation was attributable entirely to eNOS activation and NO generation. The effect of this regulation was dependent on the concentration of TNF- in a way consistent with the concentration dependency of Akt and eNOS activation. Thus, Sph1P and NO act in an autocrine loop switched on by TNF- to regulate in an inhibitory fashion its own effects on adhesion. The dependency on TNF- concentration suggests that the protective role of this pathway depends on the level of inflammation.
As far as the expression of adhesion molecules, we found that Sph1P, administered at nanomolar concentrations, did not exert significant effects in the absence of TNF-; however, the inhibition of its generation reduced expression of VCAM-1 and E-selectin induced by the cytokine. These results confirm previous findings with inhibitors of Sph1P generation and in S1P 1 knockout animals 19,21,24 indicating the importance of Sph1P, in the context of TNF- signaling, as a regulator of the expression of these adhesion molecules. We did not find effects of Sph1P in the absence of TNF-, consistent with previously reported actions of Sph1P at nanomolar concentrations on its receptors. 15 High concentrations of Sph1P, however, may induce expression of VCAM-1 and E-selectin directly, through pathways at least in part independent of S1P receptors. 21 The possibility that the effects of Sph1P generated within TNF- signaling on endothelial cell activation synergize with those triggered by Sph1P when present at high concentrations remains open. Additional complexity arises from the existence of various isoforms of PI3K and Akt, some of which might be activated by TNF- independent of S1P receptors to counterbalance the effect of Sph1P on E-selectin. 16 In addition, Sph1P generated by SK may increase the expression of PECAM-1, an effect independent of S1P receptors. 30
In the context of the TNF- /Sph1P effects detailed above, the generation of NO contributes a specific role, because it regulates in a negative fashion the expression of E-selectin while having no effects on the expression of VCAM-1 and ICAM-1. E-selectin contributes to the initial adhesion of DCs to endothelial cells. 31 The correlation between the effects of NO on E-selectin expression and DC adhesion suggests that NO plays an important role in the process of adhesion of DCs to the endothelium stimulated by TNF-. This effect might synergize with the actions of NO taking place in DCs during their maturation process and contribute to regulate their antigen-presenting function. 32
The overall role of endogenous NO in regulating the expression of adhesion molecules triggered by TNF- is still debated, because some studies showed that NO downregulates the ability of TNF- to increase the expression of adhesion molecules, whereas, in others, downregulation is limited to some of these molecules or to none of them. 5,6,33 It must be noted, however, that these studies were carried out on different kinds of endothelial cells and expression of adhesion molecules measured at different time points. The variability of NO action might, thus, depend on the cell type and time at which the protein expression levels were measured. In this respect, it is important to consider that TNF- affects eNOS expression in a time-dependent way, by inducing first an increase and then a downregulation. 8 In addition, TNF- regulates the synthesis of tetrahydrobiopterin, which influences eNOS activity. 34
The complex and sometime contrasting effects of NO, Sph1P, and their interaction on adhesion and adhesion molecules described above is not surprising, because adhesion is a complex process involving a variety of signaling events both at the plasma membrane and inside the cell, variously activated depending on the triggering stimulus. 35 Elucidation of the role of NO and Sph1P and of their interactions on these other signaling events will be necessary to fully understand their role in regulating adhesion.
In conclusion, the link we have revealed among NO, sphingolipid metabolism, and TNF- appears of great biological significance, also because it occurs in cells as diverse as HUVECs and HMVEC-Cs, and deserves to be additionally investigated, especially in view of the emerging role of the Sph1P-to-NO signaling in the endothelium. 25,26 The observation that this protective pathway is operative in microvascular endothelial cells may also help in elucidating the complex array of signals occurring in inflammation. 3 In this respect, it is worth noting that eNOS may be activated by inflammatory stimuli other than TNF-, and, thus, the signaling events activated by NO here described may also apply to these other molecules in integrated signaling systems tuning inflammation. 1,2
Acknowledgments
This work was supported by grants from the Ministero dell?Istruzione, dell?Università e della Ricerca cofinanziamento 2003, and the Italian Association of Cancer Research (to E.C). We thank Dr. Stuart M. Pitson, University of Adelaide, for providing the SK1 G82D mutant, and Cesare Covino for his help with the microscopy analyses. We thank also Jacopo Meldolesi for helpful discussions and for critically revising the manuscript.
【参考文献】
Albrecht EWJA, Stegeman CA, Heeringa P, Henning RH, van Goor H. Protective role of endothelial nitric oxide synthase. J Pathol. 2003; 199: 8-17.
Cirino G, Fiorucci S, Sessa WC. Endothelial nitric oxide synthase: the Cinderella of inflammation? Trends Pharmacol Sci. 2003; 24: 91-95.
Pober JS. Endothelial activation: intracellular signaling pathways. Arthritis Res. 2002; 4: S109-S116.
Kilbourn RG, Gross SS, Jubran A, Adams J, Griffith OW, Levi R, Lodato RF. N -methyl-L-arginine inhibits tumor necrosis factor-induced hypotension: implications for the involvement of nitric oxide. Proc Natl Acad Sci U S A. 1990; 87: 3629-3632.
De Caterina R, Libby P, Peng HB, Thannickal VJ, Rajavashisth TB, Gimbrone MA Jr, Shin WS, Liao JK. Nitric oxide decreases cytokine-induced endothelial activation. Nitric oxide selectively reduces endothelial expression of adhesion molecules and proinflammatory cytokines. J Clin Invest. 1995; 96: 60-68.
Khan BV, Harrison DG, Olbrych MT, Alexander RW, Medford RM. Nitric oxide regulates vascular cell adhesion molecule 1 gene expression and redox-sensitive transcriptional events in human vascular endothelial cells. Proc Natl Acad Sci U S A. 1996; 93: 9114-9118.
Montrucchio G, Lupia E, de Martino A, Battaglia E, Arese M, Tizzani A, Bussolino F, Camussi G. Nitric oxide mediates angiogenesis induced in vivo by platelet-activating factor and tumor necrosis factor-. Am J Pathol. 1997; 151: 557-563.
Bove K, Neumann P, Gertzberg N, Johnson A. Role of ecNOS-derived NO in mediating TNF-induced endothelial barrier dysfunction. Am J Physiol. 2001; 280: L914-L922.
Hoffmann J, Haendeler J, Aicher A, Rossig L, Vasa M, Zeiher AM, Dimmeler S. Aging enhances the sensitivity of endothelial cells toward apoptotic stimuli: important role of nitric oxide. Circ Res. 2001; 89: 709-715.
Sala R, Rotoli BM, Colla E, Visigalli R, Parolari A, Bussolati O, Gazzola GC, Dall?Asta V. Two-way arginine transport in human endothelial cells: TNF- stimulation is restricted to system y(+). Am J Physiol. 2002; 282: C134-C143.
Kawanaka H, Jones MK, Szabo IL, Baatar D, Pai R, Tsugawa K, Sugimachi K, Sarfeh IJ, Tarnawski AS. Activation of eNOS in rat portal hypertensive gastric mucosa is mediated by TNF- via the PI 3-kinase-Akt signaling pathway. Hepatology. 2002; 35: 393-402.
Barsacchi R, Perrotta C, Bulotta S, Moncada S, Borgese N, Clementi E. Activation of endothelial nitric-oxide synthase by tumor necrosis factor- : a novel pathway involving sequential activation of neutral sphingomyelinase, phosphatidylinositol-3' kinase, and Akt. Mol Pharmacol. 2003; 63: 886-895.
Levade T, Auge N, Veldman RJ, Cuvillier O, Negre-Salvayre A, Salvayre R. Sphingolipid mediators in cardiovascular cell biology and pathology. Circ Res. 2001; 89: 957-968.
Clementi E, Borgese N, Meldolesi J. Interactions between nitric oxide and sphingolipids and the potential consequences in physiology and pathology. Trends Pharmacol Sci. 2003; 24: 518-523.
Bolick DT, Srinivasan S, Kim KW, Hatley ME, Clemens JJ, Whetzel A, Ferger N, Macdonald TL, Davis MD, Tsao PS, Lynch KR, Hedrick CC. Sphingosine-1-phosphate prevents tumor necrosis factor- -mediated monocyte adhesion to aortic endothelium in mice. Arterioscler Thromb Vasc Biol. 2005; 25: 976-981.
Madge LA, Pober JS. A phosphatidylinositol 3-kinase/Akt pathway, activated by tumor necrosis factor or interleukin-1, inhibits apoptosis but does not activate NFkappaB in human endothelial cells. J Biol Chem. 2000; 275: 15458-15465.
Marchesini N, Osta W, Bielawski J, Luberto C, Obeid LM, Hannun YA. Role for mammalian neutral sphingomyelinase 2 in confluence-induced growth arrest of MCF7 cells. J Biol Chem. 2004; 279: 25101-25111.
Pitson SM, Moretti PA, Zebol JR, Xia P, Gamble JR, Vadas MA, D?Andrea RJ, Wattenberg BW. Expression of a catalytically inactive sphingosine kinase mutant blocks agonist-induced sphingosine kinase activation. A dominant-negative sphingosine kinase. J Biol Chem. 2000; 275: 33945-33950.
Chen XL, Grey JY, Thomas S, Qiu FH, Medford RM, Wasserman MA, Kunsch C. Sphingosine kinase-1 mediates TNF- -induced MCP-1 gene expression in endothelial cells: upregulation by oscillatory flow. Am J Physiol. 2004; 287: H1452-H1458.
Stoffel W, Jenke B, Block B, Zumbansen M, Koebke J. Neutral sphingomyelinase 2 (smpd3) in the control of postnatal growth and development. Proc Natl Acad Sci U S A. 2005; 102: 4554-4559.
Xia P, Gamble JR, Rye KA, Wang L, Hii CS, Cockerill P, Khew-Goodall Y, Bert AG, Barter PJ, Vadas MA. Tumor necrosis factor- induces adhesion molecule expression through the sphingosine kinase pathway. Proc Natl Acad Sci U S A. 1998; 95: 14196-14201.
Ancellin N, Colmont C, Su J, Li Q, Mittereder N, Chae SS, Stefansson S, Liau G, Hla T. Extracellular export of sphingosine kinase-1 enzyme. Sphingosine 1-phosphate generation and the induction of angiogenic vascular maturation. J Biol Chem. 2002; 277: 6667-6675.
Osada M, Yatomi Y, Ohmori T, Hosogaya S, Ozaki Y. Modulation of sphingosine 1-phosphate/EDG signaling by tumor necrosis factor- in vascular endothelial cells. Thromb Res. 2002; 108: 169-174.
Krump-Konvalinkova V, Yasuda S, Rubic T, Makarova N, Mages J, Erl W, Vosseler C, Kirkpatrick CJ, Tigyi G, Siess W. Stable knock-down of the sphingosine 1-phosphate receptor S1P1 influences multiple functions of human endothelial cells. Arterioscler Thromb Vasc Biol. 2005; 25: 1-7.
Igarashi J, Bernier SG, Michel T. Sphingosine 1-phosphate and activation of endothelial nitric-oxide synthase. Differential regulation of Akt and MAP kinase pathways by EDG and bradykinin receptors in vascular endothelial cells. J Biol Chem. 2001; 276: 12420-12426.
Morales-Ruiz M, Lee MJ, Zollner S, Gratton JP, Scotland R, Shiojima I, Walsh K, Hla T, Sessa WC. Sphingosine 1-phosphate activates Akt, nitric oxide production, and chemotaxis through a Gi protein/phosphoinositide 3-kinase pathway in endothelial cells. J Biol Chem. 2001; 276: 19672-19677.
Xia P, Wang L, Gamble JR, Vadas MA. Activation of sphingosine kinase by tumor necrosis factor- inhibits apoptosis in human endothelial cells. J Biol Chem. 1999; 274: 34499-34505.
Weis M, Schlichting CL, Engleman EG, Cooke JP. Endothelial determinants of dendritic cell adhesion and migration: new implications for vascular diseases. Arterioscler Thromb Vasc Biol. 2002; 22: 1817-1820.
Banchereau J, Steinman RM. Dendritic cells and the control of immunity. Nature. 1998; 392: 245-252.
Limaye V, Li X, Hahn C, Xia P, Berndt MC, Vadas MA, Gamble JR. Sphingosine kinase-1 enhances endothelial cell survival through a PECAM-1-dependent activation of PI-3K/Akt and regulation of Bcl-2 family members. Blood. 2005; 105: 3169-3177.
Robert C, Fuhlbrigge RC, Kieffer JD, Ayehunie S, Hynes RO, Cheng G, Grabbe S, von Andrian UH, Kupper TS. Interaction of dendritic cells with skin endothelium: A new perspective on immunosurveillance. J Exp Med. 1999; 189: 627-633.
Paolucci C, Burastero SE, Rovere-Querini P, De Palma C, Falcone S, Perrotta C, Capobianco A, Manfredi AA, Clementi E. Synergism of nitric oxide and maturation signals on human dendritic cells occurs through a cyclic GMP-dependent pathway. J Leukoc Biol. 2003; 73: 253-262.
Kuhlencordt PJ, Rosel E, Gerszten RE, Morales-Ruiz M, Dombkowski D, Atkinson WJ, Han F, Preffer F, Rosenzweig A, Sessa WC, Gimbrone MA Jr, Ertl G, Huang PL. Role of endothelial nitric oxide synthase in endothelial activation: insights from eNOS knockout endothelial cells. Am J Physiol. 2004; 286: C1195-C1202.
Rosenkranz-Weiss P, Sessa WC, Milstien S, Kaufman S, Watson CA, Pober JS. Regulation of nitric oxide synthesis by proinflammatory cytokines in human umbilical vein endothelial cells. Elevations in tetrahydrobiopterin levels enhance endothelial nitric oxide synthase specific activity. J Clin Invest. 1994; 93: 2236-2243.
Imhof BA, Aurrand-Lions M. Adhesion mechanisms regulating the migration of monocytes. Nat Rev Immunol. 2004; 4: 432-444.
作者单位：Clara De Palma; Elisabetta Meacci; Cristiana Perrotta; Paola Bruni; Emilio ClementiFrom the Department of Pharmaco-Biology (C.D.P., C.P.), University of Calabria, Rende; Stem Cell Research Institute (C.D.P., C.P., E.C.), DIBIT-H San Raffaele Institute, Milan; the Department of Biochemical Sciences (