|Home | About | Journals | Submit | Contact Us | Français|
The bone marrow is a rich reservoir of cells that are mobilized in response to physiologic stress signals (hypoxia, inflammation) or pathological states (cancer, chronic inflammation) 1. Although mobilization of myeloid cells that participate in innate immune responses and inflammation has been appreciated for some time, recent studies have focused on so called bone marrow-derived progenitor cells (BMPC) 1. It is now appreciated that distinct progenitor population which can differentiate into cells with both myeloid and vascular markers exit the bone marrow and enter into tissues in response to chemokine cues. Indeed, BMPC may circulate widely in the body; recent work has shown that sphingosine 1-phosphate (S1P) signaling is used as a egress signal for these cells to re-enter the lymphatic system to eventually return to the bone marrow2. Once in the tissue, BMPC respond to extracellular signals to differentiate into cells that are closely associated with the vascular system. Although opinions are divided whether BMPC are incorporated into the vascular tree or are perivascular, it is clear that their presence is functionally important for inflammation, neoplasia and angiogenesis3. Indeed, several clinical trials are currently testing the efficacy of bone marrow cell therapy in ischemic tissues4, 5.
Such bone marrow-derived cells modulate vascular and tissue responses by elaborating cytokines, chemokines and lipid mediators1, 3. Interestingly, the lipid mediator S1P has been the subject of recent interest as a regulator of vascular and immune systems 6. It is produced by the metabolism of sphingomyelin, an abundant phospholipid essential for the formation of membrane domains such as rafts and caveolae 7. Sphingomyelinase hydrolyze sphingomyelin to form ceramide, which is further degraded by ceramidase to form sphingosine. Sphingosine levels are kept low since sphingosine kinase (Sphk) catalyzes the phosphorylation into S1P. Subsequently, S1P lyase catalyzes the irreversible degradation of S1P into hexadecenal and phosphoenthanolamine, intermediates in the biosynthesis of phospholipids. Many of the steps in the S1P metabolic pathway are reversible. For example, S1P is converted by S1P phosphatase enzymes into sphingosine, which itself is converted back into ceramide by ceramide synthases and eventually into sphingomyelin by sphingomyelin synthases. S1P is released into the extracellular environment by various cells, including those of the hematopoietic system 8. It is highly enriched in the circulatory and lymphatic systems but is lower in interstitial fluids of tissues, thus creating a gradient. This gradient is essential as a chemotactic cue in the traverse of various hematopoietic cells into lymphatic and vascular channels 9.
S1P exerts powerful effects on the vascular endothelium 6. The prototypical S1P receptor (S1P1) was originally cloned as an adundant and inducible mRNA from human endothelial cells 10. Activation of S1P receptors in endothelial cells result in the redistribution of adherens junction proteins into areas of cell-cell contact and the their assembly. For example, treatment of human endothelial cells in vitro with S1P results in the rapid translocation of VE-cadherin, β-catenin and α-cateninto areas of cell-cell contact 11. Indeed Triton X-100 solubility of adherens junction proteins decreases with S1P treatment, suggesting that signaling from S1P receptors induce the assembly of adherens junction in endothelial cells. This signaling event requires the heterotrimeric Gi protein, small GTPases Rac and Cdc42 11, 12. This effect of S1P (via the S1P1 receptor) results in tightening of junctions and increase in transmonolayer electrical resistance in vitro 13. Further, acute agonism of S1P1 with the pharmacological agent FTY720 results in profound suppression of vascular endothelial growth factor-induced vascular permeability in vivo 14. In addition, S1P treatment suppressed lipopolysaccharide (LPS)-induced pulmonary vascular permeability in a canine and murine models 15. Indeed, inhibition of S1P1 signaling with a specific pharmacological antagonist resulted in the increased vascular permeability in the lung and the skin tissues 16. Moreover, a recent study using conditional Sphk knockout mice showed that reduction in plasma S1P resulted in increased basal vascular permeability in the lung and decreased survival during platelet activating factor-induced anaphylactic shock 17. These data form the basis for the emerging concept that tonic signaling of the endothelial cell S1P1 receptor is needed for maintenance of the homeostatic barrier property of the vascular system. In addition, during infection and inflammation, S1P1 receptor system is required for the restoration of normal vascular barrier function. This is clinically important since increased fluid retention in the lung due to abnormal vascular permeability is a significant clinical problem in infectious diseases.
Given the importance of S1P signaling in pulmonary edema, Zhao et al. in this issue of Circulation Research reported the role of Sphk enzyme in the BMPC in the restoration of vascular permeability in a murine model of LPS-induced lung injury 18. They showed that intravenous therapy with BMPC improved pulmonary edema and lethality from LPS. Interestingly, BMPC from Sphk1−/− mice did not provide similar protection, suggesting that secretion of S1P from the BMPC in the local environment and activation of S1P1 receptors on the pulmonary vasculature protected the lung tissue from excessive vascular leak. In support of this, the authors showed that BMCP from wild-type mice secreted S1P and induced transendothelial monolayer resistance in a S1P1 dependent manner. Indeed, the ability of BMPC to induce adherens junctions in endothelial cell monolayers required signaling by small GTPases Rac and Cdc42, which is very similar to signal transduction pathways induced by S1P-induced activation of its receptor on endothelial cells 12. These data are of importance in the understanding of pathogenesis of pulmonary edema induced during infections. In addition, from a therapeutic standpoint, activation of S1P1 receptors by agonists should aid in the alleviation of pulmonary vascular leak. Indeed, S1P receptor modulators have been tested for potential clinical application in the control of autoimmune reactions in multiple sclerosis 19. However, the prototypical S1P receptor modulator, FTY720 has a complex mode of action – although it is a potent agonist on S1P1, it acts as a functional antagonist since downregulates S1P1 receptors and induces ubiquitinylation-dependent proteosomal degradation of the receptor6. Thus, an agonist of S1P1 that does not induce receptor downregulation or degradation is needed to inhibit pathologic vascular permeability in the lung. In the absence of such a pharmacologic tool, BMPC therapy appears promising.
Despite these promising findings, data in the report suggest that more complex mechanisms may be at play. Characterization of BMPC from wild-type and and Sphk1−/− mice indicated significant differences in cell surface marker expression. Interestingly, progenitor/stem cell markers were reduced and differentiated vascular and hematopoietic markers were elevated in Sphk1−/− BMPC, suggesting that lack of Sphk1 enzyme facilitates cellular differentiation. Indeed, work from the laboratory of Jennifer Gamble has shown that Sphk1 regulates the rate of endothelial progenitor cell differentiation 20. The mechanism by which Sphk1 regulates renewal and/or differentiation of BM stem cells and progenitors is not known, but GPCR-independent mechanisms may be involved. Indeed, since Sphk1 occupies a central position in the sphingolipid metabolic pathway, lack of this enzyme may alter cell fate in rapidly proliferating progenitor populations that require tight control of membrane turnover. Alternatively, intracellular signaling function of sphingolipid metabolites, such as sphingosine, ceramide or S1P itself may be involved. Thus, functional differences in BMPC from wild-type vs. Sphk1−/− mice may be involved in the differential protective functions of these two populations. Secondly, Zhao et al. noted that pulmonary retention of wild-type and Sphk1−/− BMPC are very different. While it is not known why injected Sphk1−/− BMPC are not retained in the lungs, differential cellular adhesion or survival may be responsible. Thus, in addition to released S1P, functional differences between BMPC from wild-type and Sphk1−/− mice may explain the differences in pulmonary protective property.
Several critical questions are highlighted by interesting results of Zhao et al. For example, how is Sphk1 activated in BMPC, what is the functional role of Sphk1 in BMPC, what cytokines are elaborated by Sphk1−/− BMPC and does S1P signaling play a role in migration and/or egress of BMPC in the lung? Ultimately one would like to know if BMPC are recruited physiologically during infection to help restore pulmonary microvascular function. Nevertheless, these results also suggest exciting opportunities for BMPC-based therapies to control infectious diseases. In addition, Sphk1and S1P1 function were shown to be critical for human ES cell proliferation 21. Therefore, this system may be a fundamental signaling pathway in stem/progenitor cell biology.
Sources of Funding: This work is supported by NIH grants HL67330 and HL89934.
Disclosures: The author has no conflicts to declare.