Sphingosine Kinase 2 (SphK2) Deficiency Does Not Lower but Instead Surprisingly Increases the Level of Sphingosine-1-Phosphate (S1P) in Peripheral Blood, Providing a Novel Strategy for Stimulating Hematopoietic Stem Cell Mobilization By Inhibiting SphK2 Activity

Background. Sphingosine-1-phosphate (S1P) is a bioactive lipid involved in cell signaling and, if released from cells, also plays a crucial role in regulating the trafficking of lympho-hematopoietic cells [ Cell 2007; 131: 994-1008 ] and primitive hematopoietic stem/progenitor cells (HSPCs) [ Leukemia 2010; 24: 976-985 ]. S1P is highly expressed in tissue fluids, such as blood lymph and peripheral blood (PB), and its most important sources are red blood cells, albumin, and apoM bound to high-density lipoproteins (HDL). The high concentration of S1P in PB creates a crucial chemotactic gradient across the endothelium between BM and PB that directs the egress of HSPCs during mobilization [ Leukemia 2010; 24: 976-985 ]. The process of sphingosine phosphorylation to produce S1P is governed by sphingosine kinases. Two forms of sphingosine kinase-type 1 (Sphk1) and type 2 (SphK2)-have been described.Sphk1 is found in the cytosol of eukaryotic cells and shifts to the plasma membrane upon activation, while SphK2 is localized in the cell nucleus. As previously demonstrated, mice with Sphk1 knockout are poor mobilizers [ Blood 2012; 119: 2478-2488 ] and have impaired homing of HSPCs after transplantation [ Oncotarget 2015; 6: 18819-18828 ]. The S1P level can also be downregulated in a sphingosine phosphate lyase (Sgpl1)-dependent manner, and inhibition of Sgpl1 with tetrahydroxybutylimidazole (THI), which is a small-molecule inhibitor of this enzyme, results in an increase in S1P levels in PB. Hypothesis. We hypothesized that, like SphK1-KO mice, SphK2-KO mice are poor mobilizers . Materials and Methods. To better address the role of the S1P gradient in the mobilization of HSPCs, we first analyzed hematopoiesis in mice deficient in sphingosine kinase 2 (SphK2-KO mice) and studied the effect of this mutation on plasma S1P levels. Next, normal control mice and Sphk2-KO mice were mobilized with G-CSF or AMD3100. In some of our mobilization studies we also employed a small-molecule inhibitor of SphK2 (SLM6041434) or an inhibitor of the S1P-degrading enzyme Sgpl1 (tetrahydroxybutylimidazole, THI). Results. We obtained the surprising result that, in contrast to SphK1-KO mice, SphK2-KO mice have normal hematopoiesis and that the circulating S1P level is highly elevated in these animals. This was corroborated by the finding that, also in contrast to SphK1-KO animals, SphK2-KO animals show enhanced mobilization of HSPCs. These results were recapitulated in wild type (WT) animals by employing the SphK2 inhibitor SLM6041434. We also observed that administration of THI to WT mice enhanced the mobilization of HSPCs. Conclusions. In sum, our results suggest a crucial role for S1P gradients in blood plasma in the mobilization process and indicate that small-molecule inhibitors of SphK2 and Sgpl1 could be employed as novel mobilization-facilitating compounds. At the same time, further studies are needed to explain the unexpected effect of SphK2 ablation or inhibition on increasing S1P levels in plasma. Our working hypothesis is that given the rapidity with which S1P blood levels can change is that SphK2 may also surprisingly function in clearing S1P from blood and/or a decrease in SphK2 activity leads to enhanced activity of SphK1. These possibilities are currently investigated in our laboratories.