Ning et al showed that LPC appeared to
Ning et al. showed that LPC 18:1 appeared to increase insulin release through both GPR119-dependent and -independent mechanisms (Ning et al., 2008). In our study, we have shown that not only GPR119 but also GPR40 and GPR55 participate in LPC-stimulated insulin secretion (Fig. 4, Fig. 5) and target preferences were structure-dependent (Fig. 4). Involvement of GPR55 and GPR40 in modulation of insulin secretion has already been demonstrated. Whereas OEA and LPI are the main natural GPR55 agonists studied so far (Liu et al., 2016, McKillop et al., 2013, Romero-Zerbo et al., 2011), GPR40 has turned out to be a key protein mediating FFA potentiation of insulin secretion (Itoh et al., 2003, Itoh and Hinuma, 2005, Briscoe et al., 2003). Findings concerning GPR40-mediated insulinotropic activity of more complex lipids are limited. Consistent with previous studies, apparent selectivity may be seen in the case of OEA, which affected GSIS via GPR55 (McKillop et al., 2013) and GPR119 (Soga et al., 2005, Overton et al., 2006, Sakamoto et al., 2006, Moran et al., 2014, Ning et al., 2008), but not via GPR40. Targeted activity of OEA and P appeared in GSIS experiments (Fig. 4) suggests that the applied antagonists (Hu et al., 2009, Kargl et al., 2013, Heynen-Genel et al., 2010, Briscoe et al., 2006, McClure et al., 2011) are highly receptor-selective and that all the three receptors are functional in the cell model despite vast predominance of GPR40 over the latter two targets. LPC 18:1 and LPC 16:0 were proven to activate GPR119-Gs pathways in RH7777 rat hepatoma N6-Methyl-dATP stably expressing human GPR119 (Soga et al., 2005), which is consistent with the results obtained in our study (Fig. 6). Importantly, PT did not lead to significant [cAMP]i increase, even though it was found to activate GPR119 in the MIN6 cell model. Similar effect was observed in the case of the unsaturated phosphorothioate LPC (OT, data not shown), therefore, it could be suspected that modification of the glycerol backbone at the sn-2 and sn-3 positions (Fig. 1) not only increases ligand-protein affinity (Frey and Sammons, 1985) but is also the reason for differences in triggered cAMP signaling and possible biased signaling shifted towards other G proteins (Hassing et al., 2016). Opposite to GPR119, treatment with GPR55 and GPR40 antagonists did not affect unmodified LPC-triggered cAMP signal. GPR40 has not been previously studied as a receptor for LPCs, but appeared to be affected by both native and modified LPCs (P and PT). GPR40 has been demonstrated to couple with Gq protein in MIN6 cells, when stimulated with FFAs (Itoh et al., 2003, Itoh and Hinuma, 2005), but Gq/Gs agonists of GPR40 are also known (Hauge et al., 2015). Here, none of the compounds tested evoked cAMP accumulation in GPR40-related pathway (Fig. 6), confirming GPR40-Gq coupling in MIN6 cells. Exposing MIN6 cells with P or PT together with GPR40 antagonists resulted in lowering of the total [Ca2+]i and GSIS. GPR55 is a well-documented target for lysophosphatidylinositol (LPI) (Liu et al., 2016, Kotsikorou et al., 2011, Oka et al., 2007, Oka et al., 2009), which structurally resembles LPC (Gendaszewska-Darmach and Drzazga, 2014). Although GPR55 is documented to be coupled to various G-proteins (G12/13 or Gq (Henstridge et al., 2009, Lauckner et al., 2008)) leading to elevation of [Ca2+]i in various cell lines (Moreno-Navarrete et al., 2012), LPI was noted to preferably induce the G13-Rho pathway (Obara et al., 2011). We have shown very recently that various molecular species of LPC (P, PT, O, OT) elicit intracellular Ca2+ mobilization in a GPR55-dependent manner in PC-3 human prostate carcinoma cells (Drzazga et al., 2017). In this study we demonstrate GPR55 involvement in [Ca2+]i mobilization and GSIS induced by phosphorothioate LPC analogues. In summary, we demonstrate that phosphorothioate analogues of LPC are potent enhancers of glucose-stimulated insulin secretion out of which PT is the strongest one. When compared with its native counterpart, PT demonstrated 2-fold higher efficiency in potentiation of insulin secretion as well as [Ca2+]i signaling, but did not stimulate cAMP accumulation above the basal level. The same cAMP signaling was seen in the case of O and OT. The effect on calcium flux as well as GSIS evoked by PT was weakened by GPR40, GPR55 and GPR119 antagonists suggesting that Ca2+ signaling plays pivotal role in insulin secretion stimulated by phosphorothioate analogues of LPCs. It may be claimed that the introduced chemical modifications modulated affinity of LPC analogues towards receptor targets, which resulted in increased biological activity compared to the natural compound. We conclude that chemically modified LPCs have qualities as insulin secretagogues that may make them suitable as treatment options in situations when enhanced insulin secretion is called upon.