Sphingosine kinases (SKs) 1 and 2 produce high concentrations of sphingosine 1-phosphate (S1P) in blood and lymph. In contrast, S1P concentrations in lymphoid tissues are kept low by the S1P-degrading activity of the S1P-lyase. These differences in S1P concentrations drive lymphocyte circulation. Inhibition of the S1P-lyase prevents lymphocyte egress and causes lymphopenia because of increased S1P levels in lymphoid tissues. In this study, we investigated the source of this accumulating S1P in lymphoid tissues by using SK2-deficient (SK2−/−) mice. In contrast to wild-type mice, SK2−/− mice exhibited attenuated lymphopenia after S1P-lyase inhibition by 4-deoxypyridoxine (DOP). Consistently, S1P concentrations were only modestly increased in lymphoid tissues of SK2−/− mice compared with a significantly higher increase in wild-type mice after DOP treatment. Low S1P concentrations in lymphoid tissues of DOP-treated SK2−/− mice were accompanied by higher S1P concentrations in blood, suggesting that SK2−/− mice display defective S1P transport from blood into lymphoid tissues. To investigate this potential new role of SK2, RBCs loaded with traceable C17-S1P were transfused into wild-type and SK2−/− mice, resulting in much higher C17-S1P concentrations in blood of SK2−/− mice compared with wild-type mice 2 h after transfusion. Moreover, cocultures of RBCs with mouse splenocytes and endothelial cells demonstrated that SK2 regulated cellular uptake of S1P from RBCs. Collectively, our data suggest that S1P in lymphoid tissues derives from blood and point to an essential role of SK2 in S1P transport.

Sphingosine 1-phosphate (S1P) initiates T and B cell exit from lymphoid tissues by activating the S1P1 receptor on lymphocytes. To define the mechanistic details of this ligand–receptor interaction, the biological activity of the S1P-blocking Ab Sphingomab was investigated. Treatment of mice with Sphingomab resulted in blood B and T cell lymphopenia. Although Sphingomab blocked S1P1-mediated calcium flux and receptor downregulation by S1P in vitro, plasma from Sphingomab-treated mice demonstrated a 4-fold increase in S1P concentration and largely retained its stimulating activity on S1P receptors. Plasma-borne S1P was obviously not sufficiently inactivated by Sphingomab to account for the observed lymphopenia. Therefore, we addressed the local S1P-blocking activity of Sphingomab in spleen and peripheral lymph nodes (pLNs) as a potential cause of PBL depletion. Transwell chemotaxis assays revealed the migration of freshly isolated splenocytes, but not pLN cells to S1P. However, chemotaxis of pLN cells was regained after culture in S1P-low medium, and pLN cells isolated from Sphingomab-treated mice also revealed enhanced chemotaxis to S1P, indicating substantial local inactivation of S1P in pLN after Sphingomab treatment. We conclude that treatment with the S1P-blocking Ab Sphingomab induces lymphopenia by inactivating S1P locally in pLN and not systemically in plasma. Consequently, the presence of local S1P amounts in secondary lymphoid organs contributes to B and T cell egress.

No abstract