Overexpression of lysophospholipid acyltransferase, LPLAT10/LPCAT4/LPEAT2, in the mouse liver increases glucose‐stimulated insulin secretion

Postprandial hyperglycemia is an early indicator of impaired glucose tolerance that leads to type 2 diabetes mellitus (T2DM). Alterations in the fatty acid composition of phospholipids have been implicated in diseases such as T2DM and nonalcoholic fatty liver disease. Lysophospholipid acyltransferase 10 (LPLAT10, also called LPCAT4 and LPEAT2) plays a role in remodeling fatty acyl chains of phospholipids; however, its relationship with metabolic diseases has not been fully elucidated. LPLAT10 expression is low in the liver, the main organ that regulates metabolism, under normal conditions. Here, we investigated whether overexpression of LPLAT10 in the liver leads to improved glucose metabolism. For overexpression, we generated an LPLAT10‐expressing adenovirus (Ad) vector (Ad‐LPLAT10) using an improved Ad vector. Postprandial hyperglycemia was suppressed by the induction of glucose‐stimulated insulin secretion in Ad‐LPLAT10‐treated mice compared with that in control Ad vector‐treated mice. Hepatic and serum levels of phosphatidylcholine 40:7, containing C18:1 and C22:6, were increased in Ad‐LPLAT10‐treated mice. Serum from Ad‐LPLAT10‐treated mice showed increased glucose‐stimulated insulin secretion in mouse insulinoma MIN6 cells. These results indicate that changes in hepatic phosphatidylcholine species due to liver‐specific LPLAT10 overexpression affect the pancreas and increase glucose‐stimulated insulin secretion. Our findings highlight LPLAT10 as a potential novel therapeutic target for T2DM.


| INTRODUCTION
The worldwide prevalence of diabetes mellitus has increased in recent years.The global prevalence of this disease in 20-79-year-old individuals in 2021 was estimated to be 536.6 million. 1Type 2 diabetes mellitus (T2DM), which accounts for approximately 90%-95% of all diabetes mellitus cases, is characterized by decreased insulin secretion, insulin resistance, or both.In the postprandial state, insulin is secreted by the pancreatic β cells to decrease glucose production and increase the uptake of blood glucose into the peripheral tissues, which results in decreased blood glucose levels.In T2DM, decreased insulin release or suppression of insulin action leads to decreased glucose uptake by the peripheral tissues, resulting in increased blood glucose levels. 2 Insulin is released from the pancreatic β cells in a biphasic manner in response to increased glucose levels.The first-phase insulin response involves a brief spike, lasting approximately 10 min, followed by a second phase in which insulin levels reach a plateau in 2-3 h. 3 Impairment of the first-phase insulin response is a major factor contributing to postprandial hyperglycemia in individuals with impaired glucose tolerance or in the early stages of T2DM. 4,5Although several treatments have been developed for T2DM, the quality of life of people with T2DM tends to be lower than that of individuals without T2DM, warranting the development of new treatments and prevention methods.
Phospholipids, consisting of two fatty acids and one polar head group linked to a glycerol backbone, are the major components of cellular membranes.They are further subdivided into phosphatidic acid (PA), phosphatidylcholine (PC), phosphatidylethanolamine (PE), phosphatidylserine (PS), phosphatidylinositol (PI), phosphatidylglycerol (PG), and cardiolipin.][8] Phospholipids are subjected to fatty acyl chain remodeling reactions in a pathway known as the Lands cycle. 91][12] To date, 14 mammalian LPLATs have been identified.Tanaka et al.  showed that hepatic depletion of LPLAT11, also known as lysophosphatidylinositol acyltransferase 1 (LPIAT1) and membrane-bound O-acyltransferase domain-containing 7 (MBOAT7), causes accumulation of triacylglycerol fueled by elevated phosphatidylinositol turnover. 135][16] LPLATs are involved in various biological processes and pathological conditions.
Lysophosphatidylcholine (LPC) is a major lysophospholipid generated from PC by phospholipase A and is remodeled back to PC by LPCAT.LPC is involved in hepatocyte toxicity. 17,18Several studies have demonstrated increased LPC content in the liver of db/db type 2 diabetic mice 19 and patients with NAFLD. 17,20The conversion of LPC to PC by LPCAT in the liver may improve glucose and lipid metabolism.Several LPLATs also exhibit LPCAT activity.LPLAT8, also called LPCAT1, is expressed in the lungs, retina, and other tissues, where it generates palmitic acid (C16)-containing PC and platelet-activating For overexpression, we generated an LPLAT10-expressing adenovirus (Ad) vector (Ad-LPLAT10) using an improved Ad vector.Postprandial hyperglycemia was suppressed by the induction of glucose-stimulated insulin secretion in Ad-LPLAT10-treated mice compared with that in control Ad vector-treated mice.
Hepatic and serum levels of phosphatidylcholine 40:7, containing C18:1 and C22:6, were increased in Ad-LPLAT10-treated mice.Serum from Ad-LPLAT10treated mice showed increased glucose-stimulated insulin secretion in mouse insulinoma MIN6 cells.These results indicate that changes in hepatic phosphatidylcholine species due to liver-specific LPLAT10 overexpression affect the pancreas and increase glucose-stimulated insulin secretion.Our findings highlight LPLAT10 as a potential novel therapeutic target for T2DM.

K E Y W O R D S
adenovirus vector, diabetes mellitus, glucose-stimulated insulin secretion, LPLAT10, lysophospholipid acyltransferase, phospholipid factor (PAF). 12,21,22 LPLAT9, also known as LPCAT2, is expressed in inflammatory cells and produces PAF. 12,23,24PLAT12, also called LPCAT3, is highly expressed in the liver and is involved in lipid metabolism, as described above. 14LPLAT10, also known as LPCAT4 and lysophosphatidylethanolamine acyltransferase 2 (LPEAT2), is reported to have LPCAT activity. 25,26Several studies have shown that LPLAT10 activity is affected in various diseases, such as cancer 27 and liver injury 28 ; however, the biological functions of LPLAT10 metabolism have not been fully elucidated.
LPCATs play important roles in lipid metabolism and homeostasis by regulating the abundance of various PC species in multiple cell and tissue types. 29The expression of LPLAT10 in the liver, which is one of the central organs in glucose and lipid metabolism, is low.In this study, we investigated whether overexpression of LPLAT10 in the liver could improve abnormalities in glucose metabolism and help treat T2DM.To overexpress LPLAT10 in the liver, we selected an adenovirus (Ad) vector because systemic administration of Ad vectors results in the liver-specific expression of exogenous genes. 30Among the Ad vectors, we utilized an improved Ad vector named Ad-E4-122aT, 31,32 which exhibits higher and longer-term transgene expression and lower hepatotoxicity than conventional Ad vectors.The findings of the present study indicate that overexpression of LPLAT10 in the liver changes hepatic and serum PC species and suppresses postprandial hyperglycemia by increasing postprandial insulin secretion.

| Mice and cells
Male C57BL/6 mice were obtained from Nippon SLC (Hamamatsu, Japan).Male db/db mice were obtained from CLEA Japan (Tokyo, Japan).After acclimation for 1 week, 6-week-old db/db mice were administered the Ad vector at a dose of 5 × 10 9 infectious units/mouse or PBS via the tail vein.C57BL/6 and db/db mice were anesthetized by inhalation of isoflurane (3%) (Pfizer, Tokyo, Japan) and dissected.No relevant adverse effects were observed in the experimental animals.The mice were maintained on a 12/12 h light/dark cycle and provided free access to water and food.Body weight was measured weekly.MIN6 cells (AddexBio, San Diego, CA, USA), a mouse insulinoma cell line, were cultured in Dulbecco's Modified Eagle Medium supplemented with 15% fetal bovine serum (FBS), 55 μM 2-mercaptoethanol, and antibiotics.ICR mouse-derived pancreatic islets were purchased from COSMO BIO (Tokyo, Japan) and cultured according to the protocol.

| Fasting blood glucose levels
Blood samples were obtained from the tail veins of fasted (16 h) mice after Ad vector administration.Fasting blood glucose levels were determined using a Glutest Sensor Neo (Sanwa Kagaku Kenkyusho, Nagoya, Japan).

| Serum insulin levels
Blood samples were collected from mice through retroorbital bleeding, and serum was obtained by centrifuging the samples.Serum insulin levels in db/db and C57BL/6 mice were determined using Mouse Insulin ELISA and Ultrasensitive Mouse Insulin ELISA kits (Mercodia, Uppsala, Sweden), respectively.

| Glucose tolerance tests
Glucose tolerance tests were performed on 16 h-fasted db/ db mice or C57BL/6 mice injected intraperitoneally with 1 or 2 g/kg glucose, respectively.Blood glucose levels were determined immediately before and at the indicated times after injection using the Glutest Sensor Neo.

| Insulin tolerance tests
Insulin tolerance tests were performed on 6 h-fasted db/db mice intraperitoneally injected with insulin (0.75 units/ kg).Blood glucose levels were determined using the Glutest Sensor Neo immediately before and at the indicated time point after injection.

| Phospholipid analysis
A comprehensive lipid analysis was performed as described previously. 39,40Briefly, total lipids were extracted from samples using the Bligh-Dyer method. 41Samples of the organic layer were dried under a gentle stream of nitrogen and dissolved in an analytical solvent for LC-MS/ MS.
LC-MS/MS analysis was performed using an UltiMate 3000 LC system (Thermo Fisher Scientific) equipped with an HTC PAL autosampler (CTC Analytics).The lipids were separated on a Waters X-Bridge C18 column (3.5 μm, 150 mm × 1.0 mm i.d.) at room temperature (25°C) using a gradient solvent system as follows: mobile phase A (isopropanol/methanol/water (5/1/4 v/v/v) supplemented with 5 mM ammonium formate and 0.05% ammonium hydroxide (28% in water))/mobile phase B (isopropanol supplemented with 5 mM ammonium formate and 0.05% ammonium hydroxide (28% in water)) ratios of 60%/40% (0 min), 40%/60% (0-1 min), 20%/80% (1-9 min), 5%/95% (9-11 min), 5%/95% (11-30 min), 95%/5% (30-31 min), 95%/5% (31-35 min), and 60%/40% (35-45 min).The flow rate was maintained at 25 μL/ min.Lipid species were measured using selected reaction monitoring in the positive ion mode with a triplestage quadrupole mass spectrometer (TSQ Vantage AM, Thermo Fisher Scientific).Post-processing of raw data files was performed using the Xcalibur 4.2.47 software (Thermo Fisher Scientific).The PC molecular species were carefully annotated manually after the interpretation of molecular substructures by checking the typical product ions derived from the PC head group and the two fatty acyl groups in both negative and positive ion modes.The annotation method used in this study corresponds to "Fatty Acyl Level" defined by the Lipidomics Standard Initiative. 42In this method, the effects of the biological matrix cannot be normalized for all the detected peaks because it is not possible to prepare appropriate internal standards corresponding to all these peaks.The relative concentration of each analyte was calculated using the ratio of the chromatographic peak area for the analyte to that of the internal standard.

| Insulin secretion assays in vitro
For assays using serum, MIN6 cells were seeded in 24well plates at a density of 5 × 10 4 cells/well and cultured for 2 days.The medium was replaced with a medium containing 5% FBS and 10% serum from Ad-LPLAT10or Ad-Luc-treated mice.After 16-18 h of incubation, the cells were washed with HEPES-buffered Krebs-Ringer buffer containing 130 mM NaCl, 4.7 mM KCl, 1.2 mM KH 2 PO 4 , 1.2 mM MgSO 4 , 1.5 mM CaCl 2 , 10 mM HEPES, and 0.1% bovine serum albumin (BSA), supplemented with 2.8 mM glucose.For assays using fatty acids, MIN6 cells were seeded in 12-well plates at a density of 1 × 10 5 cells/well and cultured for 2 days.The medium was replaced with a medium containing oleic acid (OA), docosahexaenoic acid (DHA), or ethanol as a control.After 14 h of incubation, the cells were washed with HEPESbuffered Krebs-Ringer buffer containing 130 mM NaCl, 4.7 mM KCl, 1.2 mM KH 2 PO 4 , 1.2 mM MgSO 4 , 1.5 mM CaCl 2 , 10 mM HEPES, and 0.1% BSA, supplemented with 2.8 mM glucose.
For glucose-stimulated insulin secretion assay, MIN6 cells were preincubated in HEPES-buffered Krebs-Ringer buffer with 0.1% BSA and 2.8 mM glucose for 1 h at 37°C, washed, and then incubated for 1 h with 2.8 or 20 mM glucose.The incubation medium was then collected and centrifuged.The amount of insulin in the medium was measured using the LBIS Mouse Insulin ELISA Kit (H-type; FUJIFILM Wako Shibayagi Corporation).
For glucose-stimulated insulin secretion assay using primary mouse pancreatic islets, islets were cultured in a medium containing OA, DHA, or ethanol for 16 h.After washing, seven formation-matched islets were preincubated in RPMI medium with 1% BSA and 2.8 mM glucose for 1 h.Subsequently, the islets were incubated for 1 h with 25 mM glucose.The amount of insulin in the medium was measured using the LBIS Mouse Insulin ELISA Kit (U-type; FUJIFILM Wako Shibayagi Corporation).

| Statistical analysis
Statistical analyses were performed using BellCurve for Excel (Social Survey Research Information Co., Ltd., Tokyo, Japan).The Mann-Whitney U test was used to compare the differences between two independent groups, and one-way ANOVA with Dunnett's post hoc test was used for multiple comparisons.Data are presented as mean ± standard error (SE), and a p-value <.05 indicated a statistically significant difference.

| Study approval
All animal procedures were approved by the Institutional Animal Care and Use Committee of Osaka Ohtani University (approval IDs: 1701 and 2002) and were performed in accordance with the Institutional Guidelines and Regulations for Animal Experiments at Osaka Ohtani University.

| RESULTS
3.1 | Tissue distribution of LPLAT10 and its liver-specific overexpression using an Ad vector LPLAT10 mRNA levels were determined in different tissues of male C57BL/6 mice.LPLAT10 mRNA was primarily expressed in the cerebrum and testis, whereas its expression was very low in the liver (Figure 1A).
We performed liver-specific overexpression of LPLAT10 in db/db mice, a mouse model of T2DM using an LPLAT10-expressing Ad vector (Ad-LPLAT10).A luciferase-expressing Ad vector (Ad-Luc) was used as a control.The administration of Ad-LPLAT10 in db/ db mice increased the hepatic expression of LPLAT10 by 370-fold compared with that in Ad-Luc-treated mice (Figure 1B).LPLAT10 protein levels were considerably higher in the liver of mice transduced with Ad-LPLAT10 than in the liver of Ad-Luc-or PBS-treated mice (Figure 1C).To examine whether intravenous administration of Ad-LPLAT10 leads to liver-specific overexpression of LPLAT10 in mice, LPLAT10 mRNA levels in the skeletal muscle and epididymal adipose tissue were determined using quantitative RT-PCR.LPLAT10 mRNA levels in the skeletal muscle and epididymal adipose tissue were similar in Ad-LPLAT10-and Ad-Luctreated mice (Figure S1A,B).

| Metabolic phenotype of mice treated with Ad-LPLAT10
To investigate the effect of LPLAT10 on the metabolic profile in vivo, the body weights of db/db mice were monitored for 8 weeks.The increase in body weight was similar among Ad-LPLAT10-, Ad-Luc-, and PBS-treated db/db mice (Figure 2A).Next, we monitored blood glucose and insulin levels under fasting conditions 2 weeks after the administration of Ad-LPLAT10, Ad-Luc, or PBS.Fasting blood glucose levels in Ad-Luc-treated db/db mice were significantly higher than those in PBS-treated C57BL/6 mice (Figure 2B).No significant differences in blood glucose and insulin levels were observed between Ad-LPLAT10-and Ad-Luc-treated db/db mice (Figure 2B,C).Fasting blood glucose levels were not significantly different between Ad-LPLAT10-and Ad-Luc-treated C57BL/6 mice, as well as db/ db mice (Figure S2).Hepatic mRNA levels of G6Pase, the rate-limiting enzyme in gluconeogenesis, were not significantly different between Ad-LPLAT10-and Ad-Luc-treated db/db mice (Figure 2D).
Glucose tolerance tests were performed to explore the effects of LPLAT10 on glucose metabolism.Blood glucose levels after glucose injection in Ad-Luc-treated db/ db mice were significantly higher in PBS-treated C57BL/6 mice and similar to those in PBS-treated db/db mice (Figure 3A).Glucose levels, 30 min after glucose injection, were significantly lower in Ad-LPLAT10-treated db/ db mice than in Ad-Luc-treated db/db mice (Figure 3A).Next, we performed insulin tolerance tests.The rates of decrease in blood glucose levels relative to that at 0 min after insulin injection in Ad-LPLAT10-treated db/db mice were slightly lower than those in Ad-Luc-treated db/db mice that received the same treatment (Figure 3B).To examine the levels of insulin secreted after glucose loading, C57BL/6 mice were subjected to a glucose tolerance test.Glucose levels, at 30 and 60 min after glucose injection, in Ad-LPLAT10-treated mice were significantly lower than those in Ad-Luc-treated mice (Figure 3C).The insulin secretion levels in Ad-LPLAT10-treated mice were higher than those in Ad-Luc-treated mice at 15 and 60 min after glucose injection (Figure 3D).These results indicate that the liver-specific overexpression of LPLAT10 suppresses postprandial hyperglycemia by increasing postprandial insulin secretion.
F I G U R E 1 Tissue distribution of LPLAT10 mRNA expression and liver-specific overexpression of LPLAT10 using an adenovirus (Ad) vector.(A) LPLAT10 mRNA levels in C57BL/6 mouse tissues measured using quantitative RT-PCR.LPLAT10 mRNA levels in the cerebrum were set as 100.Hepatic LPLAT10 (B) mRNA and (C) protein levels in mice, 2 weeks after administration of Ad-LPLAT10, Ad-Luc, or PBS, as determined using quantitative RT-PCR and western blot analysis, respectively.Hepatic LPLAT10 mRNA levels in Ad-Luc-treated db/db mice were set as 1.0.Two independent experiments were performed and yielded similar results.The data are expressed as mean ± SE values (n = 4-5).Arrowhead indicates LPLAT10 protein.

| Serum obtained from Ad-LPLAT10- treated mice increased glucose-stimulated insulin secretion
Next, we attempted to elucidate the mechanism by which LPLAT10 overexpression in the liver increased postprandial insulin secretion.We hypothesized that liver-specific overexpression of LPLAT10 alters phospholipid components in the liver and that the altered phospholipids are secreted from the liver.Altered phospholipids would act on the pancreas via the blood and induce glucosestimulated insulin secretion.First, to investigate whether the liver-specific overexpression of LPLAT10 alters hepatic PCs and LPCs, we examined their profiles using LC-MS/MS analysis.Although the total amounts of PC and LPC in the liver of Ad-LPLAT10-and Ad-Luc-treated mice were comparable (Figure 4A,C), significant changes were observed in PC and LPC species.The amount of PC 40:7 in the liver of Ad-LPLAT10-treated mice was 2.4-fold higher than that in the liver of Ad-Luc mice (Figure 4B).Second, to investigate whether changes in PC and LPC in the liver cause changes in PC and LPC in the blood, we examined the PC and LPC species in the serum.No significant difference in the total amounts of serum PC and LPC was observed between the Ad-LPLAT10 and Ad-Luc groups (Figure 5A,C).The amount of PC 40:7 in the serum of Ad-LPLAT10-treated mice was 2.2-fold higher than F I G U R E 2 Body weight and glucose metabolism.(A) Body weights of db/db mice treated with Ad-LPLAT10, Ad-Luc, or PBS.(B) Fasting glucose and (C) insulin levels in mice, 2 weeks after Ad-LPLAT10, Ad-Luc, or PBS treatment.(D) Hepatic G6Pase mRNA levels in mice, 2 weeks after Ad-LPLAT10, Ad-Luc, or PBS treatment measured using quantitative RT-PCR.Hepatic G6Pase mRNA levels in Ad-Luctreated db/db mice were set as 1.0.One-way ANOVA with Dunnett's post hoc tests was used for multiple comparisons.Two independent experiments were performed and yielded similar results.The data are expressed as mean ± SE (n = 4-5).N.S.; not significant.that in the serum of Ad-Luc-treated mice (Figure 5B).The changes in PC and LPC levels in the serum of Ad-LPLAT10-treated mice were similar to those in the liver (Figure 5B,D).These results indicate that liver-specific overexpression of LPLAT10 leads to changes in PC and LPC species in the liver and serum.
Third, we cultured MIN6 cells, a mouse insulinoma cell line, in a medium containing the serum of Ad-LPLAT10or Ad-Luc-treated mice, and measured the amount of insulin secreted from MIN6 cells under high glucose stimulation.Significant insulin secretion was observed in MIN6 cells cultured in a medium containing serum collected from mice treated with Ad-LPLAT10 (Figure 6A).Glucagon-like peptide 1 (GLP-1) is secreted from the L cells of the gastrointestinal mucosa and induces glucosestimulated insulin secretion. 43Serum GLP-1 levels in Ad-Luc-treated db/db mice were significantly higher than those in PBS-treated C57BL/6 mice and similar to those in Ad-LPLAT10-treated db/db mice (Figure 6B).These results indicate that overexpression of LPLAT10 in the liver mediates glucose-stimulated insulin secretion via the blood.Insulin secretion in MIN6 cells treated with OA was not significantly increased compared with that in ethanoltreated cells, used as a control, in 2.8 and 20 mM glucose (Figure 7A).However, DHA treatment significantly increased insulin secretion in MIN6 cells in the presence of 20 mM glucose (Figure 7B).The increase in insulin secretion from DHA-treated MIN6 cells was concentrationdependent.Furthermore, insulin secretion in primary mouse pancreatic islets treated with OA did not increase significantly compared with that in ethanol-treated cells, F I G U R E 5 Analysis of serum phosphatidylcholine (PC) and lysophosphatidylcholine (LPC).(A) Total PC, (B) PC species, (C) total LPC, and (D) LPC species in the serum of db/db mice, 2 weeks after the administration of Ad-LPLAT10 or Ad-Luc, as determined using LC-MS/ MS.The Mann-Whitney U test was used to compare differences between two independent groups.The data are expressed as mean ± SE (n = 4).*p < .05compared with the Ad-Luc treatment.N.D.; not detected.insulin secretion in DHA-treated islets was twofold higher than that in ethanol-treated islets under high glucose conditions (Figure 8).These results indicate that, among the PC subspecies altered by LPLAT10, DHA, a component of PC 40:7, is important for glucose-stimulated insulin secretion.

| DISCUSSION
T2DM is characterized by the dysregulation of glucose homeostasis.Glucose-stimulated insulin secretion is central to the normal control of metabolic fuel homeostasis, and its impairment is a key factor in the failure of β cells in T2DM. 44In particular, the impairment of first-phase insulin secretion has been identified as a major contributor to postprandial hyperglycemia, leading to T2DM.In this study, we observed that liver-specific overexpression of LPLAT10 suppressed postprandial hyperglycemia by stimulating glucose-stimulated insulin secretion.Overexpression of LPLAT10 in the liver changed PC and LPC species in the liver and serum.Our results indicate that changes in hepatic PC species by LPLAT10 increase glucose-stimulated insulin secretion from the pancreatic β cells.
Improvement in hepatic insulin resistance suppresses gluconeogenesis, leading to a reduction in fasting blood glucose levels. 45The overexpression of LPLAT10 in the liver did not alter the levels of fasting blood glucose, insulin, or hepatic G6Pase mRNA, suggesting that hepatic LPLAT10 overexpression did not improve insulin resistance.It did, however, suppress the increase in blood glucose levels early after glucose loading and increased insulin secretion, indicating that the first-phase insulin secretion was increased.
The diversity of phospholipids due to variations in the polar head group and fatty acids, as determined by LPLATs, has been reported 39,[46][47][48] ; however, phospholipid profiles determined by LPLAT10 in vivo have not been elucidated.Our study is the first to reveal PC and LPC profiles determined by LPLAT10 in the mouse liver using an Ad vector-mediated overexpression system.Various PC and LPC species were altered by LPLAT10, with the increase in PC 40:7 being of special interest.PC 40:7 contains C18:1 (OA) and C22:6 (DHA).Our data are consistent with those of a previous study showing that LPLAT10 has endogenous activity of incorporating both C18:1 and C22:6 into LPE and LPC in vitro. 12,25,49Further studies are needed to understand in vivo substrate preferences of LPLAT10 in the brain and testis (Figure 1).
In this study, the overexpression of LPLAT10 in the liver and DHA supplementation in MIN6 cells induced PC 40:7 and insulin secretion, respectively.Because LPLAT3 is a potent LPLAT that generates DHA-containing phospholipids, 39,50,51 it is possible that LPLAT3 also plays a similar role in regulating insulin secretion.LPLAT3 biosynthesizes PA, containing DHA, which is converted into all phospholipid types.Differences in enzymatic products (phospholipid species) may affect their biological roles.Except for LPLAT10, LPLATs exhibit LPCAT activity.Further studies are needed to evaluate the effects of other LPLATs on insulin secretion in the liver and pancreas.
Liver-specific overexpression of LPLAT10 leads to changes in the PC and LPC profiles in the liver and affects insulin secretion from the pancreatic β cells.Changes in the PC and LPC profiles in the liver tended to be the same F I G U R E 6 Increase in glucose-stimulated insulin secretion by serum obtained from Ad-LPLAT10-treated mice.(A) Insulin secretion from MIN6 cells incubated for 1 h with 20 mM glucose.MIN6 cells were cultured with a medium containing serum from Ad-LPLAT10 or Ad-Luc-treated mice for 16-18 h.After washing, the cells were preincubated in HEPES-buffered Krebs-Ringer buffer with 0.1% BSA and 2.8 mM glucose for 1 h.After washing, MIN6 cells were incubated for 1 h with 20 mM glucose.The amount of insulin in the medium was measured using ELISA assays.(B) Fasting serum GLP-1 levels in mice, 2 weeks after Ad-LPLAT10, Ad-Luc, or PBS treatment.The Mann-Whitney U test was used to compare differences between two independent groups (A).One-way ANOVA with Dunnett's post hoc tests was used for multiple comparisons (B).The data are expressed as mean ± SE (A: n = 3, B: n = 4-5).*p < .05compared with the Ad-Luc treatment.
as those in the serum, suggesting PC and LPC in the serum originated from the liver.Phospholipids and lysophospholipids converted by the overexpression or suppression of various LPLAT in the liver can act on pancreas-like humoral factors via the bloodstream, leading to a potential therapeutic target for treating glucose metabolism.
Insulin secretion in response to blood glucose stimulation occurs in a biphasic manner.The first-phase insulin secretion is rapid but lasts for approximately 10 min, followed by a second phase which plateaus at 2-3 h. 3 Compared to Ad-Luc-treated mice, insulin levels in Ad-LPLAT10-treated mice showed no difference at the 30-min time point, while both 15-and 60-min time points showed a significant difference (Figure 3D).The overexpression of LPLAT10 in the liver may promote first-and second-phase insulin secretion.The firstphase insulin secretion occurs from the rapid fusion of insulin granules that are pre-docked at the plasma membrane.The second-phase insulin secretion involves the mobilization of intracellular insulin granules to the plasma membrane to enable distal docking and fusion during insulin exocytosis. 52In patients with T2DM, both first-and second-phase insulin secretion are lost. 53 previous study shows that a negatively charged phospholipid enhances the interaction between proteins in insulin granules and plasma membranes, resulting in key roles during insulin exocytosis and leading to an association between phospholipid and insulin secretion. 54n our study, changes in the fatty acid composition of PC and LPC in the liver and serum by LPLAT10 may have F I G U R E 7 DHA increased glucose-stimulated insulin secretion in MIN6 cells.Insulin secretion from (A) OA-treated or (B) DHAtreated MIN6 cells.MIN6 cells were cultured with a medium containing OA, DHA, or ethanol for 14 h.After washing, the cells were preincubated in HEPES-buffered Krebs-Ringer buffer with 0.1% BSA and 2.8 mM glucose for 1 h.After washing, the cells were incubated for 1 h with 2.8 or 20 mM glucose.The amount of insulin in the medium was measured using ELISA assays.One-way ANOVA with Dunnett's post hoc tests was used for multiple comparisons.The data are expressed as mean ± SE (n = 4).*p < .05compared with EtOH in the presence of 20 mM glucose.EtOH; ethanol, OA; oleic acid, DHA; docosahexaenoic acid.insulin secretion from pancreatic β cells, the negatively charged phospholipids.Therefore, further studies are needed.
The overexpression of LPLAT10 in the liver increased hepatic and serum levels of PC 40:7, containing OA and DHA.n-3 polyunsaturated fatty acids (PUFAs), especially αlinolenic acid, eicosapentaenoic acid, and DHA, positively affect insulin secretion. 44,55The present study and a previous one showed that the supplementation of the culture medium with DHA increased insulin secretion under high-glucose stimulation in MIN6 cells. 56Some of the proposed mechanisms for the effect of n-3 PUFAs on insulin secretion are the induction of GLP-1 by chronic dietary n-3 fatty acid intake, 57 binding to G-protein-coupled receptors (GPRs), such as GPR40, and lipid raft structure and function modulation. 55Serum levels of GLP-1, a gut-derived incretin hormone secreted from intestinal L cells, and induction of glucose-stimulated insulin secretion 43 were similar between Ad-LPLAT10-and control Ad-treated mice.GPR40 is expressed in pancreatic β cells and enteroendocrine cells, 58 and DHA is a GPR40 ligand.GPR40 activation enhances Ca 2+ release from the endoplasmic reticulum by activating inositol 1,4,5-triphosphate receptors, leading to insulin secretion. 59Lipid rafts are associated with the cellular glucose transport activity of glucose transporter 2. 60 n-3 PUFAs regulate membrane structure and properties, and influence insulin secretion. 55The endogenous increase in DHA in PC 40:7 due to the hepatic overexpression of LPLAT10 may have promoted glucose-stimulated insulin secretion via GPR40 and changes in lipid rafts.Further studies are required to elucidate the underlying mechanisms.
In addition to the liver, skeletal muscle and adipose tissues are important as they affect insulin sensitivity. 61PC and LPC changes (e.g., OA and DHA) in the liver might contribute to the improvement of insulin sensitivity in skeletal muscle and adipose tissue, such as an increase in glucose-stimulated insulin secretion from the pancreas.We performed the insulin tolerance test, which primarily measures the insulin-stimulated uptake of glucose by skeletal muscles and is a marker of whole-body insulin sensitivity.The rates of decrease in blood glucose levels compared to 0 min of insulin injection of Ad-LPLAT10treated mice were slightly lower than those in Ad-Luc mice.These results suggest that changes in PC and LPC may partially contribute to the improvement of insulin sensitivity in the whole body.Further research is needed to elucidate the precise contribution to insulin sensitivity.
At the beginning of this study, we attempted to decrease the total LPC and increase the total PC in the liver using LPLAT10.The total amounts of LPC and PC in the liver of Ad-LPLAT10-and Ad-Luc-treated mice were comparable.However, hepatic LPC 18:0, LPC 20:3, LPC 20:4, and LPC 20:5 levels in Ad-LPLAT10-treated mice were significantly lower than those in Ad-Luc-treated mice.C18:0 (stearic acid) is toxic to pancreatic cells. 62,63The hepatic levels of PC 36:5 (containing C16:0 and C20:5) and PC 38:4 (containing C18:0 and C20:4) were also significantly lower in Ad-LPLAT10-treated mice than in Ad-Luc-treated mice.C16:0 (palmitic acid) is toxic to β cells and suppresses glucose-stimulated insulin secretion. 64,65hanges in these PC and LPC showed similar trends in Ad-LPLAT10-treated mouse liver and serum.In addition to the increase in PC 40:7, LPLAT10-mediated decreases in specific PC-and LPC-containing fatty acids that are toxic to β cells, such as palmitic acid and stearic acid, may have resulted in the enhancement of glucose-stimulated insulin secretion.
In summary, our study showed that the overexpression of LPLAT10 in the mouse liver increased glucosestimulated insulin secretion by increasing hepatic and serum PC 40:7 (C18:1 and C22:6) levels, leading to the attenuation of postprandial hyperglycemia.Thus, LPLAT10 may serve as a new therapeutic target for the treatment of glucose metabolism disorders, including T2DM.

AUTHOR CONTRIBUTIONS
Kahori Shimizu conceived and designed the project, performed the experiments, analyzed the data, and wrote the manuscript.Moe Ono, Takenari Mikamoto, Yuya Urayama, Sena Yoshida, Tomomi Hase, and Shotaro Michinaga performed the experiments.Hiroki Nakanishi F I G U R E 8 DHA increased glucose-stimulated insulin secretion in primary mouse pancreatic islets.Primary mouse pancreatic islets were cultured in a medium containing OA, DHA, or ethanol for 16 h.After washing, seven formation-matched islets were preincubated in the RPMI medium supplemented with 1% BSA and 2.8 mM glucose for 1 h.The islets were incubated for 1 h with 25 mM glucose.The amount of insulin in the medium was measured using ELISA assays.One-way ANOVA with Dunnett's post hoc tests was used for multiple comparisons.The data are expressed as mean ± SE (n = 3).*p < .05compared with EtOH.and analyzed the lipid measurements.Miho Tomoyuki Terada, Fuminori Sakurai, and Hiroyuki Mizuguchi provided the materials.Hideo Shindou revised the manuscript and supervised the study.Koji Tomita analyzed the data, revised the manuscript, and supervised the project.Toru Nishinaka provided materials and supervised the project.All authors approved the manuscript.

F I G U R E 3
Hepatic LPLAT10 overexpression improved glucose tolerance and glucose-stimulated insulin secretion.(A) Blood glucose levels during the intraperitoneal glucose tolerance test of mice, 1 week after the administration of Ad-LPLAT, Ad-Luc, or PBS.(B) Blood glucose levels during the intraperitoneal insulin tolerance test of db/db mice, 1 week after administration of Ad-LPLAT10 or Ad-Luc.(C) Blood glucose and (D) insulin levels during the intraperitoneal glucose tolerance test of C57BL/6 mice, 5 weeks after the administration of Ad-LPLAT or Ad-Luc.One-way ANOVA with Dunnett's post hoc tests was used for multiple comparisons (A).The Mann-Whitney U test was used to compare differences between two independent groups (B-D).Two independent experiments were performed and yielded similar results.The data are expressed as mean ± SE (n = 5).*p < .05compared with the Ad-Luc treatment.| Fatty acids comprising PC 40:7 induced glucose-stimulated insulin secretionWe sought to determine whether the serum PCs altered by the overexpression of LPLAT10 in the liver act on the pancreas via the bloodstream to enhance glucosestimulated insulin secretion.PC 40:7, containing C18:1 and C22:6, is among the PCs whose levels are greatly increased by the overexpression of LPLAT10 in the liver.To test this hypothesis, MIN6 cells were cultured in a medium containing OA (C18:1) or DHA (C22:6) for 14 h.

F I G U R E 4
Analysis of hepatic phosphatidylcholine (PC) and lysophosphatidylcholine (LPC).(A) Total PC, (B) PC species, (C) total LPC, and (D) LPC species in the liver of db/db mice, 2 weeks after the administration of Ad-LPLAT10 or Ad-Luc, as determined using LC-MS/MS.The Mann-Whitney U test was used to compare differences between two independent groups.The data are expressed as mean ± SE (n = 4).*p < .05compared with the Ad-Luc treatment.N.D.; not detected.