The phosphatidylethanolamine derivative diDCP-LA-PE mimics intracellular insulin signaling

Insulin facilitates glucose uptake into cells by translocating the glucose transporter GLUT4 towards the cell surface through a pathway along an insulin receptor (IR)/IR substrate 1 (IRS-1)/phosphatidylinositol 3 kinase (PI3K)/3-phosphoinositide-dependent protein kinase-1 (PDK1)/Akt axis. The newly synthesized phosphatidylethanolamine derivative 1,2-O-bis-[8-{2-(2-pentyl-cyclopropylmethyl)-cyclopropyl}-octanoyl]-sn-glycero-3-phosphatidylethanolamine (diDCP-LA-PE) has the potential to inhibit protein tyrosine phosphatase 1B (PTP1B) and to directly activate PKCζ, an atypical isozyme, and PKCε, a novel isozyme. PTP1B inhibition enhanced insulin signaling cascades downstream IR/IRS-1 by preventing tyrosine dephosphorylation. PKCζ and PKCε directly activated Akt2 by phosphorylating at Thr309 and Ser474, respectively. diDCP-LA-PE increased cell surface localization of GLUT4 and stimulated glucose uptake into differentiated 3T3-L1 adipocytes, still with knocking-down IR or in the absence of insulin. Moreover, diDCP-LA-PE effectively reduced serum glucose levels in type 1 diabetes (DM) model mice. diDCP-LA-PE, thus, may enable type 1 DM therapy without insulin injection.

PTP1B inhibition would cause a persistent tyrosine phosphorylation of IR/IRS-1, thereby enhancing the ensuing downstream signaling. Indeed, diDCP-LA-PE enhanced IR phosphorylation at Tyr1185 and IRS-1 phosphorylation at Tyr1222, although the latter was not significant. PKC, on the other hand, is implicated in the regulation of GLUT4 trafficking through an insulin signaling pathway 3 . In response to insulin PKCλ /ι and -ζ , are activated and promote GLUT4 translocation towards the cell surface [4][5][6][7][8] . Insulin activates PKCα , -β II, and -δ as well, which regulate GLUT4 trafficking 9,10 . Moreover, PKCε may also participate in the regulation of GLUT4 trafficking 11 . We, in the light of these facts, postulated that diDCP-LA-PE might mimic intracellular insulin signaling.
The present study was conducted to prove this hypothesis. We show here that diDCP-LA-PE facilitates glucose uptake into differentiated 3T3-L1 adipocytes still with knocking-down IR or in the absence of insulin and reduces serum glucose levels in type 1 DM model mice without insulin injection.
Monitoring of GLUT4 trafficking. GLUT4 trafficking in differentiated 3T3-L1-GLUT4myc adipocytes was monitored by the methods as described previously 1 . Differentiated 3T3-L1 adipocytes were incubated in Krebs-Ringer-HEPES buffer (136 mM NaCl, 4.7 mM KCl, 1.25 mM CaCl 2 , 1.25 mM MgSO 4 and 20 mM HEPES, pH 7.5) containing 0.2% (w/v) bovine serum albumin supplemented with 10 mM glucose for 1 h at 37 °C. Cells were treated with insulin or a variety of lipids in the presence and absence of inhibitors for 20 min. Then, cells were homogenized by sonication in an ice-cold mitochondrial buffer [210 mM mannitol, 70 mM sucrose, and 1 mM EDTA, 10 mM HEPES, pH 7.5] containing 1% (v/v) protease inhibitor cocktail (Nacalai Tesque, Kyoto, Japan) and subsequently, homogenates were centrifuged at 3,000 rpm for 5 min at 4 °C. The supernatants were centrifuged at 11,000 rpm for 15 min at 4 °C and further, the collected supernatants were ultracentrifuged at 100,000 g for 60 min at 4 °C to separate the cytosolic and plasma membrane fractions. The supernatants and pellets were used as the cytosolic and plasma membrane fractions, respectively. Whether the cytosolic and plasma membrane components were successfully separated was confirmed in the Western blot analysis using antibodies against the cytosolic marker lactate dehydrogenase (LDH) and the plasma membrane marker cadherin. The cytosolic fraction contains GLUT4 in transport vesicles as well as in intracellular compartments such as the endosomes and the trans-Golgi network, and the plasma membrane fraction otherwise contains GLUT4 on the plasma membrane, but not in a partial pool near the plasma membrane.
Protein concentrations for each fraction were determined using a BCA protein assay kit (Thermo Fisher Scientific, Waltham, MA, USA). Proteins in the plasma membrane fraction were resuspended in the mitochondrial buffer containing 1% (w/v) sodium dodecyl sulfate (SDS). Proteins for each fraction were separated by SDS-polyacrylamide gel electrophoresis (SDS-PAGE) and transferred to polyvinylidene difluoride (PVDF) membranes. After blocking with TBS-T [150 mM NaCl, 0.1% (v/v) Tween-20, and 20 mM Tris, pH 7.5] containing 5% (w/v) bovine serum albumin (BSA), blotting membranes were reacted with an anti-c-myc antibody (Merck Millipore, Darmstadt, Germany) followed by a horseradish peroxidase (HRP)-conjugated goat anti-mouse IgG antibody. Immunoreactivity was detected with an ECL kit (Invitrogen, Carlsbad, CA, USA) and visualized using a chemiluminescence detection system (GE Healthcare, Piscataway, NJ, USA). Signal density was measured with an ImageQuant software (GE Healthcare).
Cell-free kinase assay. PKC activity in the cell-free systems was quantified by the method as previously described 2,12 . Briefly, synthetic PKC substrate peptide (Pyr-Lys-Arg-Pro-Ser-Gln-Arg-Ser-Lys-Tyr-Leu; MW, 1,374) (Peptide Institute Inc., Osaka, Japan) (10 μ M) was reacted with human recombinant PKCα , -β I, -β II, -γ , -δ , -ε , -λ /ι or -ζ in a medium containing 20 mM Tris-HCl (pH 7.5), 5 mM Mg-acetate, 10 μ M ATP, and diDCP-LA-PE in the absence of phosphatidylserine and diacylglycerol at 30 °C for 5 min. Activity for novel PKCs such as PKCδ and -ε was assayed in Ca 2+ -free medium and activity for the other PKC isozymes in the medium containing 100 μ M CaCl 2 . After loading on a reversed phase high performance liquid chromatography (LC-10ATvp; Shimadzu Co., Kyoto, Japan), a substrate peptide peak and a new product peak were detected at an absorbance of 214 nm. Areas for non-phosphorylated and phosphorylated PKC substrate peptide were measured (total area corresponds to concentration of PKC substrate peptide used here), and the amount of phosphorylated substrate peptide was calculated. The amount of phosphorylated substrate peptide (pmol/1 min) was used as an index of PKC activity.
Glucose uptake assay. Glucose uptake assay was carried out by the method as described previously 1,13,14 .
Differentiated 3T3-L1-GLUT4myc adipocytes without and with IR knock-down were incubated in a Krebs-Ringer-HEPES buffer containing 0.2% (w/v) BSA supplemented with 10 mM glucose at 37 °C for 1 h. Then, cells were not treated and treated with diDCP-LA-PE or insulin in phosphate-buffered saline supplemented with 10 mM glucose at 37 °C for 2 h. After treatment, extracellular solution was collected and glucose was labeled with p-aminobenzoic ethyl ester (ABEE), followed by HPLC. Glucose concentration taken up into cells was calculated by subtracting extracellular glucose concentration from initial extracellular glucose concentration (10 mM).
Oral glucose tolerance test (OGTT). Streptozotocin, which exerts its cytotoxic effect on pancreatic β cells, is a chemical inducer of experimental DM in rodents. C57BL/6J mice (male, 8 weeks of age) (Japan SLC Inc., Shizuoka, Japan) were intraperitoneally injected with a single streptozotocin (250 mg/kg) and used as a type 1 DM model mice 4 days after injection. For normal control group, mice were injected with saline.
C57BL/KsJ-leprdb/leprdb mice are a well-established genetic model of type 2 DM, which have characteristics similar to human type 2 DM including obesity, hyperglycemia, and extreme insulin resistance. The mice are obese and hyperinsulinemic up to 1 month of age, then insulin resistance worsens with the appearance of hyperglycemia. C57BL/KsJ-leprdb/leprdb and wild-type C57BL/6J mice (female, 8 weeks) were purchased from CLEA Japan (Tokyo, Japan) and used as a type 2 DM model mice and normal control mice, respectively.
In OGTT, mice were fasted for 12 h, followed by oral administration with diDCP-LA-PE using a feeding needle or intraperitoneal injection with insulin 30 min prior to loading glucose. After collection of blood (10 μ L) from the tail vein, the serum was labeled with ABEE and loaded onto the HPLC system and glucose concentrations were calculated from the peak area/concentration calibration curve made before using a standard glucose solution.
Statistical analysis. Statistical analysis was carried out using unpaired t-test, analysis of variance (ANOVA) followed by a Bonferroni correction and ANOVA followed by Fisher's protected least significant difference (PLSD) test.

Cooperation of PKCζ and PKCε directly and fully activates Akt2. To obtain evidence for interaction
of PKCζ and PKCε with Akt, we performed cell-free kinase assay. diDCP-LA-PE phosphorylated Akt2 at Thr309 in the presence of PKCζ in a concentration (1-100 μ M)-dependent manner (Fig. 3A). diDCP-LA-PE, alternatively, phosphorylated Akt2 at Ser474 in the presence of PKCε in a concentration (1-100 μ M)-dependent manner (Fig. 3B). Like in the presence of PKCζ diDCP-LA-PE phosphorylated Akt2 at Thr309 in the presence of PKCλ /ι (Fig. 3C). In contrast, no phosphorylation was induced in the presence of PKCγ (Fig. 3D). Overall, these results indicate that diDCP-LA-PE is capable of directly activating Akt2 by cooperation of PKCζ (or PKCλ /ι ) and PKCε , regardless of a pathway along an IR/IRS-1/PI3K/PDK1/Akt axis.

diDCP-LA-PE increases cell surface localization of GLUT4 in an insulin-independent manner.
If diDCP-LA-PE activates Akt, then the drug should stimulate GLUT4 translocation towards the cell surface. To address this point, we next examined the effect of diDCP-LA-PE on GLUT4 trafficking. Like insulin diDCP-LA-PE increased cell surface localization of GLUT4 in a concentration (0.1-50 μ M)-dependent manner in differentiated 3T3-L1 adipocytes (Fig. 4A,B). diDCP-LA-PS increased cell surface localization of GLUT4 to a lesser extent, but no effect was obtained with diDCP-LA-PC, diDCP-LA-PI, DCP-LA or 1,2-dilinoleoyl-sn-glycero-3-p hosphoethanolamine (DL-PE) (Fig. 4C). This indicates that of the investigated lipids diDCP-LA-PE has the highest potential to translocate GLUT4 towards the cell surface.
diDCP-LA-PE stimulates glucose uptake into differentiated 3T3-L1 adipocytes and reduces serum glucose levels in type 1 DM model mice. We finally examined the effect of diDCP-LA-PE on glucose uptake into differentiated 3T3-L1 adipocytes and serum glucose levels in type 1 DM model mice. diDCP-LA-PE stimulated glucose uptake into differentiated 3T3-L1 adipocytes in a concentration (0.1-50 μ M)dependent manner (Fig. 6A). diDCP-LA-PE significantly promoted glucose uptake into cells still with IR knock-down, while no significant effect was obtained with insulin (Fig. 6B). This implies that diDCP-LA-PE could stimulate glucose uptake in an insulin/IR-independent manner.
In the OGTT using type 1 DM model mice, oral administration with diDCP-LA-PE significantly reduced serum glucose levels as compared with that for saline-administered control mice (Fig. 6C).
Like insulin diDCP-LA-PE promoted translocation of GLUT4 towards the cell surface in differentiated 3T3-L1 adipocytes in a concentration-dependent manner. Such effect was not found with DL-PE. This implies that diDCP-LA-PE exhibits stable bioactivities in cells. diDCP-LA-PE-induced GLUT4 translocation was inhibited by an inhibitor of tyrosine kinase, PI3K, PDK1, or Akt and knocking-down PI3K, PDK1, or Akt1/2. This indicates that diDCP-LA-PE stimulates GLUT4 translocation towards the cell surface through a well-recognized IRS-1/PI3K/PDK1/Akt pathway (Fig. 7A). Moreover, this, in the light of the finding that diDCP-LA-PE-induced Akt1/2 activation in differentiated 3T3-L1 adipocytes was not significantly inhibited by knocking-down PI3K or PDK1, raises the possibility that PI3K or PDK1 each by itself directly regulates GLUT4 translocation in an Akt-independent manner (Fig. 7).
Of particular interest is the finding that diDCP-LA-PE-induced Akt1/2 activation and GLUT4 translocation in differentiated 3T3-L1 adipocytes were not affected by knocking-down IR. Moreover, diDCP-LA-PE significantly increased glucose uptake into differentiated 3T3-L1 adipocytes with IR knock-down. Collectively, these findings indicate that diDCP-LA-PE is capable of mimicking intracellular insulin signaling, i.e., diDCP-LA-PE has the potential to promote GLUT4 translocation towards the cell surface and stimulate glucose uptake still into cells lacking IR or in the absence of insulin. In the OGTT, oral administration with diDCP-LA-PE significantly reduced serum glucose levels in type 1 DM model mice. This raises the possibility that diDCP-LA-PE could control serum glucose levels in type 1 DM patients without insulin injection. Type 1 DM is caused by little/no insulin production in pancreas β cells, and insulin injection is indispensable for type 1 DM therapy. The patients, therefore, suffer physical and mental pain everyday, which would last till the end of their lives. We have been challenging a new therapy for type 1 DM without insulin injection. Consequently, we have devised diDCP-LA-PE, that must become a promising drug for type 1 DM and provide a new hope to relieve the distress for the patients.
In summary, the results of the present study demonstrate that diDCP-LA-PE promotes GLUT4 translocation towards the cell surface and stimulates glucose uptake into differentiated 3T3-L1 adipocytes through PKCζ /-ε -cooperated direct Akt2 activation and in part through PTP1B inhibition-associated activation of a PI3K/ PDK1/Akt pathway (Fig. 7B) and that diDCP-LA-PE facilitates glucose uptake still into differentiated 3T3-L1 adipocytes lacking IR or in the absence of insulin and reduces serum glucose levels in type 1 DM model mice. Insulin signal mimetic diDCP-LA-PE, thus, may shed bright light on type 1 DM therapy without insulin injection.