ES cells which have been used frequently for

ES cells, which have been used frequently for research in the field of developmental biology and tried to apply in cell-based regenerative medicine (Murry and Keller, 2008), replicate by the process of self-renewal and several different pathways are believed to be involved in the regulation of self-renewal, although the exact mechanisms remain poorly understood (Avery et al., 2006; Niwa, 2001). Although several studies with regard to this matter have detailed on peculiar in vitro culture protocols for ES cells, with the main effort being to optimize ES cell functions, it is still important to develop defined culture systems for expansion of ES cells in vitro in chemically-defined and animal product-free conditions. Recent findings (Kleger et al., 2007; Rodgers et al., 2009; Schuck et al., 2003) which showed that the bioactive lipid metabolites can regulate the ES cell functions have compelled us to investigate the mechanism by which S1P regulates ES cell proliferation. Thus, in this study, we examined the interaction between both S1P and Flk-1 in regulation of ES cell proliferation and its related signaling molecules.
Materials and methods
Results
Discussion In the present study, we demonstrated that S1P transactivated Flk-1 through S1P1/3-dependent β-arrestin and c-Src activation, which sequentially stimulated proliferation of mouse ES cells. Owing to the variations in S1P functions depending on the cell type, we determined the presence of all S1P receptors in mouse ES cells. In accordance with our results, previous reports demonstrated that mouse ES cells expressed all five S1P receptors, but human ES cells expressed only three S1P receptors (S1P1, S1P2, S1P3) (Inniss and Moore, 2006; Kleger et al., 2007; Pebay et al., 2005). S1P is physiologically detectable at concentrations of about 0.4–1.2μM in plasma and 0.2–0.5μM in serum which largely exceed the Kd values of S1P receptors (Yatomi et al., 1997). In addition, consistent with previous reports (Che et al., 2012; Kim et al., 2011; Mascall et al., 2012; Taniguchi et al., 2012), S1P increased cellular proliferation from 0.1μM onwards and maximum increment in cellular proliferation was observed at 1–10μM in vitro. Therefore, in this study, we chose to use 5μM of S1P. S1P receptors are closely associated with cholesterol-rich lipid rafts of the plasma membrane, and this association with lipid rafts is important for the regulation of S1P functions (Singleton et al., 2006; Zhao et al., 2009). Furthermore, it has been known that the lipid raft acts as a signaling platform, which mediates various stem cell responses such as migration and proliferation (Lee et al., 2010; Sasaki et al., 2011; Shirvaikar et al., 2010). The effect of S1P on buspirone hcl regulation was expected based on previous reports, which demonstrated that the activation of S1P receptors located in lipid rafts significantly increased the cell survival by reducing apoptosis and increasing proliferation of human ES cells and endothelial cells (Inniss and Moore, 2006; Wong et al., 2007; Zhao et al., 2009). In the present study, S1P1–3 were located in both lipid raft and non-lipid raft fractions. The distribution of S1P receptors was different with isotypes; S1P1 and S1P3 were mainly presented in lipid raft but most of S1P2 was located in a non-lipid raft fraction which suggests the possibilities that the role of S1P receptors could be affected by their membrane location. The disruption of lipid raft using MβCD significantly attenuated S1P-induced ES cell proliferation, thereby suggesting that the lipid raft is involved in the regulation of S1P receptor functions. In various types of cells including stem cells, vascular endothelial growth factor (VEGF) induces proliferation primarily through the autophosphorylation of Flk-1, which initially occurs partly in lipid rafts (Oshikawa et al., 2012; Xiao et al., 2007). These reports suggest the possibility that the S1P receptor could transactivate the Flk-1 and stimulate G1-S transition in mouse ES cells. In the present study, the data demonstrated that the binding of S1P to its receptor (S1P1 and S1P3) recruited β-arrestin and sequentially activated c-Src. However, VEGF-A164 did not affect the β-arrestin recruitment to the S1P receptor, thereby suggesting that the S1P-induced activation of β-arrestin and c-Src is regulated by S1P1/3 dependent-manner. After S1P stimulation, Flk-1 expression in the lipid raft fraction was increased, which was blocked by lipid raft disruption. To clarify the complex formation between S1P receptor and Flk-1, we carried out co-immunoprecipitation and in situ PLA, and these results revealed that S1P increased interaction of S1P1 and S1P3 with Flk-1. Next, we examined the increment in Flk-1 phosphorylation in response to S1P. In experiments to examine the increment in Flk-1 phosphorylation in response to S1P, the S1P-elicited activation of β-arrestin and c-Src is important for Flk-1 activation. These data were supported by previous reports that the GPCR activation caused an increase in the assembly of a protein complex containing activated β-arrestin, c-Src, and the receptor (Luttrell and Lefkowitz, 2002). This binding of c-Src and β-arrestin is mediated in part by an interaction between the c-Src homology (SH) 3 domain of the kinase and proline-rich PXXP motifs in the β-arrestin (McDonald et al., 2000). Previously, it has been reported that S1P has no effect on secretion of VEGF in BAECs (Tanimoto et al., 2002). However, a more recent study showed that S1P stimulates VEGF synthesis and secretion in epithelial cells (Sun et al., 2010), thereby suggesting the cell type-specific effect of S1P on VEGF production. In the present study, S1P elicited VEGF production/secretion and both S1P and VEGF-A164 stimulated Flk-1 activation. This suggests the possibilities that the S1P stimulates ES cell proliferation through direct crosstalk between S1P receptors and Flk-1 and/or S1P-induced production/secretion of VEGF activates Flk-1 as a ligand in an autocrine/paracrine manner. Interestingly, S1P-induced Flk-1 phosphorylation occurred in less time than VEGF synthesis in response to S1P. Furthermore, VEGFR2 kinase inhibitor (SU1498) inhibited the S1P-induced proliferation of mouse ES cells, but VEGF-A164 neutralizing antibodies and VEGF specific siRNA partially blocked. These results imply that S1P-induced Flk-1 activation occurred through both a VEGF-dependent and a VEGF-independent manner. To our knowledge, our observations are the first to provide information concerning the role of S1P and its receptors in terms of regulating ES cell proliferation through the interaction with Flk-1.