LPCs have been previously described as potential anti

LPCs have been previously described as potential anti-diabetic factors due to stimulation of secretory activity from the isolated rodent pancreas, L-type cells, ands insulin-producing cell lines, with LPC 18:1 as the main structure of interest (Soga et al., 2005, Sakamoto et al., 2006, Overton et al., 2008, Lan et al., 2009, Lauffer et al., 2009). Although these studies have shed light on LPC as GPR119 agonist, it should be kept in mind that other pancreatic islet receptors might be involved in LPC-induced anti-diabetic activity. We have previously shown that both native LPCs and their phosphorothioate analogues (Fig. 1) have the ability to stimulate the mobilization of intracellular Ca2+ in a murine β-cell line and the amount of Ca2+ flux depends on the length of the saturated acyl residue. Additionally, we have noticed that simultaneous phosphorothioate and 2-OMe substitution at the sn-3 and sn-2 position, respectively, within LPC moieties greatly reduce off-target interactions with the cell membrane and make LPC analogues less toxic compared to their natural counterparts (Rytczak et al., 2013, Drzazga et al., 2015). Here, we explored the complexity of GPR119/GPR55/GPR40 signaling in insulin-secreting MIN6 β-cells affected by unique synthetic phosphorothioate analogues of LPC with strictly defined fatty induced pluripotent stem cell residues. The MIN6 cell line was established from insulinoma by targeted expression of the simian virus 40 T antigen gene in transgenic mice (Miyazaki et al., 1990). The cell line produces insulin, has morphological characteristics of primary pancreatic β-cells, and exhibits glucose-inducible insulin secretion albeit less than in normal primary mouse islet cells. Expression of GPR40, GPR55 and GPR119 has previously been confirmed in MIN6 cells (Liu et al., 2016, Itoh et al., 2003, Zhang et al., 2014). Taking into account the favorable properties of phosphorothioate analogues of LPCs over their natural counterparts we examined a set of LPC analogues bearing medium (C12, C14) to long (C16, C18) fatty acid residues as potential modulators of glucose-stimulated insulin secretion (GSIS). The strongest stimulators of GSIS were subjected to further studies and compared to their native LPC counterparts. We investigated basic intracellular signaling in MIN6 cells, namely cyclic AMP (cAMP) and intracellular calcium ([Ca2+]i) levels since these second messengers play a major role in augmenting GSIS (Soga et al., 2005, Itoh et al., 2003, Moran et al., 2014, McKillop et al., 2013, Shibasaki et al., 2004, Yaekura et al., 1996, Hou et al., 2009, Ning et al., 2008, Landa et al., 2005). Due to ligand promiscuity between GPR119 and GPR55 documented in the case of their natural lipid agonists (Godlewski et al., 2009), we studied possible complex activity in the case of our test compounds. GPR40 was included as a potential target as well since lipids of more complex structure than FFAs were not investigated so far as potential ligands of this receptor and because LPCs contain a long-chain fatty acid moiety that theoretically could bind to the receptor.
Materials and methods
Results
Discussion LPCs are important signaling molecules exerting beneficial effects on glucose metabolism. LPC 18:1 was the first proposed endogenous ligand for GPR119, based on its ability to stimulate glucose-dependent insulin release from both the perfused rat pancreas and the NIT-1 mouse β-cell line as well as increase cAMP in GPR119-transfected cells (Soga et al., 2005). Unfortunately, natural LPCs have limited therapeutic value due to poor stability and relatively short half-life (minutes) (Smani et al., 2015). Another factor limiting application of unmodified LPCs may be intramolecular 1 → 2 migration of an acyl group (Okudaira et al., 2014). We have shown that methylation of the sn-2 position of the LPC analogue results in enhanced metabolic stability (Drzazga et al., 2017). Phosphorothioate LPC analogues with methoxy group in sn-2 position and hydrophilic phosphate head modified by sulfur atom are even more stable than sn-2 methoxy derivatives (Gendaszewska-Darmach and Szustak, 2016). On the other hand, the replacement of one of the two non-bridging oxygen atoms at the phosphate group by a sulfur atom introduces P-S bond (which is 0.5 Å longer than the corresponding P-O linkage) and changes charge distribution with localization of the negative charge on sulfur atom (Frey and Sammons, 1985) likely enhancing compound affinity towards potential targets (Mou and Gray, 2002). Indeed, in this study we present that not only natural LPCs but also LPC phosphorothioate analogues stimulate insulin secretion in a glucose-dependent manner (Fig. 2) and LPC 16:0 analogue (PT) is the most potent stimulator of GSIS demonstrating 2-fold higher efficiency compared to the natural counterpart (P). We have also proven that all C12-C18 modified LPCs tested significantly evoke insulin secretion in MIN6, where compounds bearing medium acyl chains (LT, MT) are less effective compared to those of long acyl chains (PT, ST, OT). Similar GSIS results obtained for all unsaturated molecules containing 18:1 residue (OEA, O and OT) could be related with geometry of the oleoyl acyl and suggest that effectiveness of LPCs is easier to manipulate with structure modifications when the acyl is saturated. Importantly, unlike in the case of the LPC-based compounds, significant augmentation of insulin secretion at low 2 mM glucose concentrations was observed in the case of OEA (consistent with results obtained with BRIN-BD11 insulinoma cell line (Moran et al., 2014)), which suggests that the OEA-evoked effect is partially glucose-independent.