Group 1B Phospholipase A2eCMediated Lysophospholipid Absorption Directly Contributes to Postprandial Hyperglycemia
http://www.100md.com
糖尿病学杂志 2006年第4期
Department of Pathology, Genome Research Institute, University of Cincinnati, Cincinnati, Ohio
G6Pase, glucose-6-phosphatase; JNK, c-Jun NH2-terminal kinase; LPC, lysophosphatidylcholine; PKC-, protein kinase C-; Pla2g1b, group 1B phospholipase A2
ABSTRACT
Postprandial hyperglycemia is an early indicator of abnormality in glucose metabolism leading to type 2 diabetes. However, mechanisms that contribute to postprandial hyperglycemia have not been identified. This study showed that mice with targeted inactivation of the group 1B phospholipase A2 (Pla2g1b) gene displayed lower postprandial glycemia than that observed in wild-type mice after being fed a glucose-rich meal. The difference was caused by enhanced postprandial glucose uptake by the liver, heart, and muscle tissues as well as altered postprandial hepatic glucose metabolism in the Pla2g1beC/eC mice. These differences were attributed to a fivefold decrease in the amount of dietary phospholipids absorbed as lysophospholipids in Pla2g1beC/eC mice compared with that observed in Pla2g1b+/+ mice. Elevating plasma lysophospholipid levels in Pla2g1beC/eC mice via intraperitoneal injection resulted in glucose intolerance similar to that exhibited by Pla2g1b+/+ mice. Studies with cultured hepatoma cells revealed that lysophospholipids dose-dependently suppressed insulin-stimulated glycogen synthesis. These results demonstrated that reduction of lysophospholipid absorption enhances insulin-mediated glucose metabolism and is protective against postprandial hyperglycemia.
Postprandial hyperglycemia is an early indicator of abnormality in glucose metabolism leading to type 2 diabetes. However, mechanisms responsible for postprandial hyperglycemia have not been identified to date. Previously, we reported that the Pla2g1beC/eC mice defective in expression of pancreatic group 1B phospholipase A2 (Pla2g1b) were resistant to diet-induced obesity and diabetes (1). The marginal decrease in lipid absorption efficiency cannot account for the resistance to diet-induced obesity and the increased insulin sensitivity in Pla2g1beC/eC mice because more fat was absorbed by fat-fed Pla2g1beC/eC mice than chow-fed Pla2g1b+/+ mice, yet their weights and plasma glucose levels were similar after 4 months of being fed their respective diets (1). Moreover, increased insulin sensitivity was observed not only in fat-fed Pla2g1beC/eC mice compared with Pla2g1b+/+ mice, but also in Pla2g1beC/eC mice fed in basal low-fat/low-cholesterol dietary conditions, when lipid absorption efficiency was identical to that of Pla2g1b+/+ mice. These results suggested that Pla2g1b has a direct contributory role in diet-induced obesity and diabetes.
Pla2g1b is a lipolytic enzyme that is secreted by the pancreas in response to the ingestion of a meal. Because of its abundant availability, stability, and ease of isolation, Pla2g1b is one of the most extensively studied enzymes in terms of structure and mechanism of action. However, despite the wealth of information known about the biochemical and structural characteristics of this protein (2,3), the exact physiological function of Pla2g1b has not been completely delineated. Based on its enzymatic activity against phospholipids and its abundant presence in the digestive tract, Pla2g1b has generally been thought to function only in nutrient absorption, catalyzing the hydrolysis of phospholipids in the intestinal lumen to facilitate dietary lipid absorption (4). However, our recent results with Pla2g1beC/eC mice showed that although phospholipid digestion in the intestinal lumen is a prerequisite for dietary lipid absorption (1,5), additional enzyme(s) in the digestive tract can compensate for the lack of Pla2g1b for phospholipid digestion in the intestinal lumen of Pla2g1beC/eC mice, and Pla2g1b is not required for normal lipid absorption under basal dietary conditions (5). Nevertheless, the Pla2g1beC/eC mice were more insulin sensitive and resistant to diet-induced obesity and diabetes than their wild-type counterparts. The current study was undertaken to identify the mechanism by which Pla2g1b contributes to diet-induced insulin resistance. Because Pla2g1b is secreted to the intestinal lumen in response to meal feeding, where it catalyzes phospholipid hydrolysis, we explored the possibility that lysophospholipids, the product formed by Pla2g1b hydrolysis of phospholipids, may promote postprandial hyperglycemia and contribute to diet-induced insulin resistance.
RESEARCH DESIGN AND METHODS
Pla2g1beC/eC mice were generated by homologous recombination in embryonic stem cells and backcrossed 10 times into the C57BL/6 background as described previously (5). Wild-type C57BL/6j mice were used as Pla2g1b+/+ controls. All animals were maintained in a temperature- and humidity-controlled room with a 12-h light/dark cycle. The animals were fed a rodent chow (LM485; Harlan-Teklad, Madison, WI) with free access to water. All animal protocols used in this study were approved by the institutional animal care and use committee at the University of Cincinnati.
Glucose tolerance tests.
Pla2g1b+/+ and Pla2g1beC/eC mice were fed with either basal low-fat/low-glucose diet (D12328; Research Diets, New Brunswick, NJ) or diabetogenic high-fat/high-carbohydrate diet (D12331; Research Diets) for 1 week and fasted overnight for 12 h before experiments. On the morning of the study, Pla2g1b+/+ and Pla2g1beC/eC mice were fed 4 ml/kg body wt of a bolus glucose-lipid mixed meal containing 50% glucose, 2.6 mmol/l egg phosphatidylcholine, 13.33 mmol/l triolein, and 2.6 mmol/l cholesterol in PBS for oral glucose tolerance tests. Glucose tolerance tests were also conducted by injecting a 50% glucose-PBS solution (2 g glucose/kg of body wt) into the intraperitoneal cavity. Blood was obtained by tail vein bleeding before and at different times after administration of the test meal. Blood glucose and insulin levels were measured by colorimetric assay and radioimmunoassay kits obtained from Wako Chemicals (Richmond, VA) and Linco Research (St. Charles, MO), respectively. For glucose tolerance tests involving testing the effect of lysophosphatidylcholine (LPC), saline with or without LPC (Sigma-Aldrich, St. Louis, MO) was injected (32 mg/kg of body wt i.p.) 5 min before glucose bolus administration.
Tissue glucose uptake assay.
Pla2g1b+/+ and Pla2g1beC/eC mice fed with the basal low-fat/low-glucose diet were injected with glucose (2 g/kg body wt i.p.) containing 5 e藽i 2-deoxy-[3H]glucose after an overnight fast. Animals were killed after 30 min, and tissues were collected and homogenized. The amount of radiolabeled 2-deoxyglucose taken up by each tissue was quantified by liquid scintillation.
Phospholipid absorption studies.
Mice were fasted overnight and then fed by gavage 100 e蘬 of the glucose-lipid mixed meal containing trace amounts of phosphatidyl-[3H]choline. The mice were anesthetized with ketamine/xylazine after 2 h. Blood was collected from the mesenteric portal vein, and the liver was removed. Lipid was extracted with CHCl3:methanol (2:1, vol/vol), and radioactivity was measured to determine the amount of phosphatidyl-[3H]choline absorbed as lysophosphatidyl-[3H]choline and/or phosphatidyl-[3H]choline.
LPC assay.
The concentrations of LPC in plasma and liver homogenates were determined by a modification of the enzymatic assay procedure described by Kishimoto et al. (6). Briefly, 8 e蘬 of sample was preincubated for 5 min in 240 e蘬 of reagent A (0.1 mol/l Tris-HCl, pH 8.0, 0.01% Triton X100, 1 mmol/l CaCl2, 3 mmol/l N-ethyl-N-(2-hydroxy-3-sulfopropyl)-3-methylaniline sodium dihydrate; 10 kU/l peroxidase, 0.1 kU/l glycerophosphorylcholine phosphodiesterase, 1 kU/l choline oxidase) in 96-well plates. The initial absorbance at 600 nm was recorded before initiating the reaction with the addition of 80 e蘬 of reagent B (0.1 mol/l Tris-HCl, pH 8.0, 0.01% Triton X100, 5 mmol/l 4-aminoantipyrine, and 1 kU/l phospholipase B). The incubation was continued for 30 min at 37°C, and the final absorbance at 600 nm was recorded. Total LPC concentration in each sample was calculated by comparing the change in absorbance between the standard and samples after subtraction of the background absorbance recorded before initiation of the reaction. The accuracy of this method was confirmed by comparing concentrations determined by this enzymatic assay with high-performance liquid chromatography analysis.
Cell culture experiments.
Human HepG2 and rat H4-2E hepatoma cell lines were obtained from the American Type Culture Collection (Manassas, VA) and maintained at 37°C under 5% CO2 atmosphere in low-glucose Dulbecco’s modified Eagle’s medium containing 10% fetal bovine serum. Confluent cultures were serum-starved with Dulbecco’s modified Eagle’s medium, and subsequent additions of 5 mmol/l [14C]glucose (0.25 mCi/l), BSA, and LPC were made for a 30-min preincubation. Insulin or empty vehicle was then added to the samples indicated, and incubation was continued for 2 h at 37°C. At the end of the incubation period, medium was removed, and cells were washed extensively with ice-cold PBS and then lysed with 30% KOH containing 5 mg/ml carrier glycogen. The lysate was incubated for 1 h at 60°C, and the glycogen was precipitated with four volumes of ice-cold 95% ethanol. The precipitated glycogen was dissolved in 0.2 N HCl, and the amount of [14C]glucose incorporated into glycogen was determined by liquid scintillation counting. For RNA isolation and quantification, we used cells in duplicate wells treated identically except without radioactive glucose.
RNA quantification.
Total RNA was isolated from liver, using a micro-to-midi spin column purification system from Invitrogen (Carlsbad, CA). Then, 5 e蘥 of total RNA was reverse-transcribed with oligo(dT), using a Superscript first-strand synthesis system obtained from Invitrogen. Real-time PCR assays were performed in a BioRad iCycler, using 50 ng of the prepared cDNA with 0.2 e蘭ol/l of each primer indicated in Table 1, 1 x PCR buffer (Invitrogen), 0.2 mmol/l dNTPs, 0.5 x SYBR Green (Molecular Probes), 10 nmol/l fluorescein, and 1 unit Taq polymerase (Invitrogen). Each reaction also contained Mg2+ at various concentrations, as indicated in Table 1. The samples were denatured at 95°C for 10 min, followed by 40 cycles of 95°C for 30 s and then 30 s at the temperature indicated in Table 1. Amplification of specific transcripts was confirmed by product melt curve analysis. Results are reported as arbitrary units and are normalized to the expression level of cyclophilin mRNA in each sample. The expression levels of each mRNA in wild-type mice under fasting conditions were set to equal 1.0. Statistical analyses were performed, using Student’s t test.
RESULTS
Reduced lysophospholipid absorption in Pla2g1beC/eC mice.
Although we previously showed no difference between Pla2g1b+/+ and Pla2g1beC/eC mice in phospholipid digestion and absorption of the associated fatty acyl chains through the gastrointestinal tract, it remains to be clarified whether the phospholipids were digested and absorbed as nonesterified fatty acids and glycerol 3-phosphorylcholine or as fatty acids and lysophospholipids. The difference in glycerol 3-phosphorylcholine versus lysophospholipid absorption and transport to the liver may have significant impact on regulation of hepatic glucose metabolism. Therefore, additional experiments were performed to test this possibility. A glucose-lipid mixed meal containing phosphatidyl-[3H]choline was prepared and fed to Pla2g1b+/+ and Pla2g1beC/eC mice after a 12-h fast. Blood was collected from the portal vein after 2 h, and lipids were extracted for scintillation counting to determine the amount of the [3H]choline head group absorbed intact with the lipid backbone, likely as lysophosphosphatidyl-[3H]choline. Results showed that a significant amount of the [3H]choline was absorbed as lipid moiety in portal blood of both Pla2g1b+/+ and Pla2g1beC/eC mice (Fig. 1A). Importantly, the lipid-associated [3H]choline in portal blood of Pla2g1beC/eC mice was fivefold lower than that in the Pla2g1b+/+ mice (Fig. 1A).
Hepatic and circulating lysophospholipid levels in Pla2g1b+/+ and Pla2g1beC/eC mice.
The difference between Pla2g1b+/+ and Pla2g1beC/eC mice in the absorption of luminal phospholipids as lysophospholipids was confirmed by measuring lysophospholipid concentrations in plasma and in the liver of these animals under fasting conditions as well as after feeding a glucose-lipid mixed meal. No significant difference in either plasma or liver lysopholipid concentrations was observed between Pla2g1b+/+ and Pla2g1beC/eC mice under fasting conditions (data not shown). Feeding of a glucose-lipid mixed meal resulted in the elevation of lysophospholipid levels in the plasma and liver of Pla2g1b+/+ mice fed control diet (Fig. 1B and C, respectively). In contrast, lysophospholipid levels in plasma and liver of Pla2g1beC/eC mice fed control diet remained at fasting levels after feeding the glucose-lipid mixed meal. Furthermore, whereas both mouse lines fed diabetogenic diet had increased hepatic LPC levels, an exaggerated decrease in circulating LPC levels was displayed in mice lacking Pla2g1b. Thus, the meal-induced increase in lysophospholipid levels is most likely caused by Pla2g1b digestion of the dietary and biliary phospholipids in the intestinal lumen and the absorption of the digested product lysophospholipid through the portal blood. The lack of postprandial increase in plasma and liver lysophospholipid levels in the Pla2g1beC/eC mice is consistent with this interpretation.
Glucose tolerance in Pla2g1b+/+ and Pla2g1beC/eC mice.
Our previous studies showed that Pla2g1b+/+ and Pla2g1beC/eC mice were similar in oral glucose tolerance when maintained on a low-fat chow diet. However, the Pla2g1beC/eC mice were more insulin sensitive and displayed a more rapid glucose clearance than Pla2g1b+/+ mice after feeding a high-fat/high-cholesterol Western-type diet for 15 weeks (1). In these studies, glucose tolerance tests were administered after intraperitoneal injection of a bolus load of glucose in saline solution. Because postprandial hyperglycemia is associated with type 2 diabetes after ingestion of a high-carbohydrate/high-lipid mixed meal (7), we compared plasma glucose levels in chow-fed Pla2g1b+/+ and Pla2g1beC/eC mice after feeding them an oral glucose meal (2 g/kg body wt) containing 2.6 mmol/l phosphatidylcholine, 2.6 mmol/l cholesterol, and 13.33 mmol/l triglyceride. In contrast to previous results showing no difference in glucose tolerance between chow-fed Pla2g1b+/+ and Pla2g1beC/eC mice (1), the inclusion of fat in the oral glucose test meal resulted in the Pla2g1beC/eC mice displaying significantly less postprandial hyperglycemia compared with that observed in the wild-type Pla2g1b+/+ mice (Fig. 2A). Furthermore, the difference in glucose tolerance after mixed-meal feeding was still observed after chronic feeding of the animals with a high-fat/high-glucose diet (data not shown). The age of the mice in which diet-induced hyperglycemia has been achieved is also known to affect glucose tolerance. As such, younger adult 13-week-old mice were tested for sensitivity to glucose via intraperitoneal injection of a bolus load of glucose in saline solution. Postprandial hyperglycemia was found to be reduced in Pla2g1beC/eC mice compared with Pla2g1b+/+ mice, in contrast to previous studies utilizing much older mice, despite no accompanying oral administration of a lipid bolus (Fig. 2B). These results suggest that postprandial hyperglycemia resistance observed in Pla2g1beC/eC mice is not limited to a state of extended high-fat feeding, as previously reported, but is also readily apparent in younger chow-fed animals.
Altered luminal phospholipid absorption and decreased postprandial hepatic lysophospholipids may be contributing to the increased postprandial hyperglycemia resistance phenotype observed in Pla2g1beC/eC mice. To test this possibility, intraperitoneal injections of saline with and without LPC (32 mg/kg body wt) were administered to Pla2g1beC/eC mice 5 min before testing glucose tolerance with either oral or intraperitoneal delivery of a glucose bolus (2 g/kg body wt). Blood glucose levels were elevated after glucose ingestion in Pla2g1beC/eC mice injected with LPC in saline compared with mice injected with saline alone (Fig. 2C and D). Plasma glucose levels in Pla2g1beC/eC mice injected with LPC were similar to those of Pla2g1b+/+ mice (Fig. 2A). Therefore, regardless of whether glucose is delivered orally or via intraperitoneal injection, the elevation of LPC in the circulation of Pla2g1beC/eC mice, including hepatic portal circulation, results in a glucose tolerance reminiscent of Pla2g1b+/+ mice. The reintroduction of LPC into postprandial Pla2g1beC/eC mice and the associated elimination of the hyperglycemia resistance phenotype suggest a link between circulating LPC levels and glucose intolerance.
Additional experiments were also performed to evaluate the tissues responsible for increased glucose uptake in Pla2g1beC/eC mice. In these experiments, Pla2g1b+/+ and Pla2g1beC/eC mice were injected intraperitoneally with a glucose load (2 g/kg body wt) containing 2-deoxy-[3H]glucose after an overnight fast. Various high-energy metabolism tissues were harvested from the animals after 30 min to determine tissue partitioning of the deoxy-[3H]glucose. Results showed that the Pla2g1beC/eC mice accumulated significantly more radiolabeled 2-deoxyglucose in liver, heart, and muscle tissues compared with that observed in Pla2g1b+/+ mice (Fig. 3A). The difference is less obvious in white adipose tissues, but a trend toward higher deoxyglucose uptake was also observed in the Pla2g1beC/eC mice. The increased glucose uptake by various tissues in Pla2g1beC/eC mice was not caused by increased insulin secretion by the pancreas because postprandial insulin concentrations in plasma of Pla2g1beC/eC mice were consistently lower but not statistically different from those observed in Pla2g1b+/+ mice (Fig. 3B). Thus, the Pla2g1beC/eC mice appeared to be more insulin sensitive, achieving greater glucose tolerance with similar or lower levels of postprandial insulin release than the Pla2g1b+/+ mice.
Lysophospholipid-dependent reduction of insulin-stimulated glycogen synthesis in hepatoma cell lines.
The mechanistic relationship between differences in postprandial lysophospholipid levels and hyperglycemia in Pla2g1b+/+ and Pla2g1beC/eC mice was explored by examining the influence of LPC on glucose metabolism in hepatoma cell lines. The inclusion of LPC with BSA in the incubation medium had no effect on the ability of HepG2 cells to utilize glucose for glycogen biosynthesis under basal conditions (Fig. 4A). However, insulin-stimulated glycogen biosynthesis increased approximately twofold in the absence of LPC, and the inclusion of LPC in the incubation medium reduced insulin-stimulated glycogen biosynthesis in a dose-dependent manner. Similar results were found in rat hepatoma H4-2E cells, except that higher concentrations of LPC lowered glycogen synthesis below basal levels (Fig. 4B). Analysis of RNA isolated from these cells revealed LPC suppression of glucose kinase gene expression under both basal and insulin-stimulated conditions, but had no effect on glucose-6-phosphatase (G6Pase) gene expression under both stimulated and unstimulated conditions (Fig. 5A and C). These results suggest that LPC may play a role in suppressing insulin-induced glycogen synthesis.
Expression of hepatic glucose kinase and G6Pase in Pla2g1b+/+ and Pla2g1beC/eC mice.
Increasing recent evidence has indicated that one mechanism contributing to abnormal postprandial glycemia is the failure to suppress hepatic glucose production (7,8). Although the in vitro cell culture data showed no difference in G6Pase mRNA levels in cells incubated with or without LPC, it remains possible that the higher postprandial LPC level in Pla2g1b+/+ mice resulted in less suppression of hepatic glucose production compared with that observed in Pla2g1beC/eC mice. We isolated RNA from the liver of both fasting and glucose/lipid-fed mice to test this possibility. Results showed no difference between Pla2g1b+/+ and Pla2g1beC/eC mice in hepatic expression of glucose kinase and G6Pase under fasting conditions (Fig. 5B and D). The ingestion of a bolus glucose-lipid mixed meal did not alter hepatic expression of the gluconeogenic enzyme G6Pase in Pla2g1b+/+ mice (Fig. 5B). However, hepatic G6Pase expression was reduced twofold in Pla2g1beC/eC mice after feeding them the glucose-lipid mixed meal. In contrast, expression of glucose kinase was induced in both Pla2g1b+/+ and Pla2g1beC/eC mice after feeding the glucose-lipid mixed meal (Fig. 5D). Interestingly, the induction of glucose kinase expression was significantly greater in the Pla2g1beC/eC mice compared with that observed in Pla2g1b+/+ mice. These results suggest increased glucose uptake and/or utilization for glycogen biosynthesis in the Pla2g1beC/eC mice compared with Pla2g1b+/+ mice. The expression levels of glucose kinase and G6Pase in the livers of Pla2g1beC/eC mice in both fasting and postprandial states are consistent with the hyperglycemia resistance phenotype reported.
DISCUSSION
Phospholipid hydrolysis in the intestinal lumen is required before triglyceride digestion and absorption of dietary fat and cholesterol through brush border membranes (9eC13). The major enzyme for phospholipid hydrolysis in the intestinal lumen is Pla2g1b secreted by the pancreas in response to meal feeding. However, other lipolytic enzyme(s) can serve a compensatory function in catalyzing phospholipid hydrolysis in the absence of Pla2g1b to sustain fat and cholesterol absorption (5). The difference in phospholipid hydrolysis between Pla2g1b+/+ and Pla2g1beC/eC mice, as shown in this study, is the products formed and absorbed by the animals. The major phospholipid digestive products absorbed by Pla2g1b+/+ mice are nonesterified fatty acids and LPC. The absorption of dietary phospholipids as lysophospholipids is consistent with previous reports demonstrating >90% of dietary phosphatidyl-[3H]choline is normally converted to lysophosphatidyl-[3H]choline and transported directly to the liver via the portal circulation (14). However, in animals lacking a functional Pla2g1b gene, significantly less dietary phosphatidylcholine was absorbed as LPC (Fig. 1A), even though they absorbed similar levels of the fatty acyl moiety of dietary phospholipids as the Pla2g1b+/+ mice. Thus, the compensatory enzyme for phospholipid digestion in the absence of Pla2g1b is likely to be phospholipase B, a distal intestinal enzyme that hydrolyzes phospholipids at both the sn-1 and sn-2 positions to produce nonesterified fatty acids and glycerophosphocholine (15,16).
The physiological significance of LPC generated during luminal phospholipid digestion and absorption has not been investigated previously. The availability of Pla2g1b+/+ and Pla2g1beC/eC mice, with different capabilities of LPC absorption and transport, provided the opportunity to evaluate the impact of LPC absorption in physiology and pathophysiology. The results of this study showed that LPC absorbed after Pla2g1b digestion of phospholipids in the intestinal lumen contributes to postprandial hyperglycemia by inhibiting glucose uptake by the liver, heart, and muscle tissues. The reduced level of LPC transported to the liver of Pla2g1beC/eC mice also resulted in increased postprandial glucose kinase expression (Fig. 5). The increased expression of the glucose utilization enzyme glucose kinase implies that postprandial hepatic glucose uptake and its utilization for glycogen synthesis are more efficient in Pla2g1beC/eC mice than in Pla2g1b+/+ mice. The data also showed decreased expression of G6Pase after meal feeding in the Pla2g1beC/eC mice, suggesting that the Pla2g1beC/eC mice were capable of suppressing hepatic glucose production in response to the glucose-lipid mixed meal. In contrast, postprandial suppression of G6Pase expression was not observed in Pla2g1b+/+ mice. This failure to suppress G6Pase expression after a glucose-lipid mixed meal may account for postprandial hyperglycemia and susceptibility to diet-induced insulin resistance and diabetes in Pla2g1b+/+ mice. Likewise, the ability to downregulate hepatic glucose production, as well as elevated glucose uptake by the liver, heart, and muscle tissues, after glucose feeding may account for the resistance of the Pla2g1beC/eC mice to diet-induced insulin resistance and diabetes. The increased glucose uptake in tissues of Pla2g1beC/eC mice also provided evidence that LPC may serve as a negative regulator of this insulin-stimulated process (Fig. 3A). Previous studies have shown that LPC can directly inhibit protein-mediated glucose uptake via interaction with glucose transport proteins on cell membranes (17,18). Additionally, LPC may inhibit insulin-stimulated glucose kinase expression in the liver, as demonstrated in cell culture of both HepG2 and H4-2E hepatoma cell lines (Fig. 4), in promoting postprandial hyperglycemia. The LPC-induced decrease in glucose kinase mRNA levels in HepG2 cells, along with elevated glucose kinase mRNA levels observed in lipid/glucose-fed Pla2g1beC/eC mice compared with those in Pla2g1b+/+ mice, would suggest a direct role of LPC in glucose kinase gene regulation. Interestingly, despite the lack of an LPC effect on G6Pase mRNA levels in cell culture experiments, G6Pase mRNA levels were found to be lower in lipid/glucose-fed Pla2g1beC/eC mice compared with those in Pla2g1b+/+ mice. Thus, the reduced G6Pase mRNA levels observed in vivo may not be a direct effect of LPC, but possibly an indirect effect via communication with other cell types or organs in the intact animal.
The phenotype of resistance to diet-induced obesity and diabetes observed in Pla2g1beC/eC mice (1) is similar to that observed in JNKeC/eC mice with targeted inactivation of the c-Jun NH2-terminal kinase (JNK) gene (19). Because LPC has been shown to activate JNK in fibroblasts and HeLa cells (20), and JNK inhibits insulin signaling cascade by inhibiting tyrosine phosphorylation of insulin receptor substrate-1 via serine phosphorylation (21), it is possible that LPC inhibition of glucose uptake by liver, heart, and muscle tissues as well as its suppression of hepatic glucose metabolism are also mediated through activation of JNK. Alternatively, LPC may also inhibit insulin-stimulated glucose metabolism by activation of protein kinase C- (PKC-). Previously, LPC activation of PKC- resulting in the inhibition of insulin-induced protein kinase B/Akt phosphorylation was observed in smooth muscle (22) and endothelial cells (23). Insulin signaling through phosphatidylinositol 3-kinase and protein kinase B/Akt activation was also enhanced in cells with an inactivated PKC- gene (24,25). Interestingly, the mechanism by which PKC- inhibits insulin signaling is also related to its ability to induce serine/threonine phosphorylation of insulin receptor substrates and block their tyrosine phosphorylation (26). Thus, LPC may inhibit postprandial insulin signaling in tissues via inhibiting tyrosine phosphorylation of insulin receptor substrates by activating either JNK or PKC- pathways. It must be noted, however, that these two pathways need not be mutually exclusive. Our future effort will focus on determining whether LPC inhibition of insulin signaling is mediated through either one or both of these pathways.
Regardless of the signaling mechanism(s) by which LPC inhibits postprandial hyperglycemia, results of the current study have major clinical implications. Previous studies have already demonstrated that failure to suppress endogenous glucose production after a glucose-lipid mixed meal is a major contributing factor toward hyperglycemia and type 2 diabetes (7). The results of this study further document that the Pla2g1b-digested product LPC is directly responsible for suppression of hepatic glucose metabolism. In view of studies showing increased LPC levels in patients with type 2 diabetes (27,28), the current study suggests that inhibition of Pla2g1b may be a viable strategy to prevent the onset of type 2 diabetes in individuals consuming a high-glucose/high-fat diet.
ACKNOWLEDGMENTS
This work was supported by Grants PO1 DK54504 and RO1 DK69967 from the National Institutes of Health (to D.Y.H.), National Research Service Award DK10065 (to K.W.H.), and Postdoctoral Fellowship 0525340B from the Ohio Affiliates of the American Heart Association (to E.D.L.).
We thank Josh Bashford and Todd Greer for their superb technical assistance and Dr. Susanna Hofmann for helpful discussions and advice.
FOOTNOTES
E.D.L. and R.J.K. contributed equally to this work.
K.W.H. is currently affiliated with the Department of Nutrition and Food Science, Auburn University, Auburn, Alabama.
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
REFERENCES
Huggins KW, Boileau AC, Hui DY: Protection against diet-induced obesity and obesity-related insulin resistance in group 1B PLA2-deficient mice. Am J Physiol 283:E994eCE1001, 2002
Dennis EA: Diversity of group types, regulation, and function of phospholipase A2. J Biol Chem 269:13057eC13060, 1994
Dijkstra BW, Renetseder R, Kalk KH, Hol WG, Drenth J: Structure of porcine pancreatic phospholipase A2 at 2.6 resolution and comparison with bovine phospholipase A2. J Mol Biol 168:163eC179, 1983
Carey MC, Small DM, Bliss CM: Lipid digestion and absorption. Annu Rev Physiol 45:651eC677, 1983
Richmond BL, Boileau AC, Zheng S, Huggins KW, Granholm NA, Tso P, Hui DY: Compensatory phospholipid digestion is required for cholesterol absorption in pancreatic phospholipase A2 deficient mice. Gastroenterology 120:1193eC1202, 2001
Kishimoto T, Soda Y, Matsuyama Y, Mizuno K: An enzymatic assay for lysophosphatidylcholine concentration in human serum and plasma. Clin Biochem 35:411eC416, 2002
Singhal P, Caumo A, Carey PE, Cobelli C, Taylor R: Regulation of endogenous glucose production after a mixed meal in type 2 diabetes. Am J Physiol 283:E275eCE283, 2002
Staehr P, Hother-Nielsen O, Beck-Nielsen H: The role of the liver in type 2 diabetes. Rev Endo Metab Disord 5:105eC110, 2004
Beil FU, Grundy SM: Studies on plasma lipoproteins during absorption of exogenous lecithin in man. J Lipid Res 21:525eC536, 1980
Homan R, Hamelehle KL: Phospholipase A2 relieves phosphatidylcholine inhibition of micellar cholesterol absorption and transport by human intestinal cell line Caco-2. J Lipid Res 39:1197eC1209, 1998
Mackay K, Starr JR, Lawn RM, Ellsworth JL: Phosphatidylcholine hydrolysis is required for pancreatic cholesterol esterase- and phospholipase A2-facilitated cholesterol uptake into intestinal Caco-2 cells. J Biol Chem 272:13380eC13389, 1997
Rampone AJ, Machida CM: Mode of action of lecithin in suppressing cholesterol absorption. J Lipid Res 22:744eC752, 1981
Young SC, Hui DY: Pancreatic lipase-colipase mediated triglyceride hydrolysis is required for cholesterol transport from lipid emulsions to intestinal cells. Biochem J 339:615eC620, 1999
Ikeda I, Imaizumi K, Sugano M: Absorption and transport of base moieties of phosphatidylcholine and phosphatidylethanolamine in rats. Biochim Biophys Acta 921:245eC253, 1987
Pind S, Kuksis A: Further characterization of a novel phospholipase B (phospholipase A2eClysophospholipase) from intestinal brush-border membranes. Biochem Cell Biol 69:346eC357, 1991
Tojo H, Ichida T, Okamoto M: Purification and characterization of a catalytic domain of rat intestinal phospholipase B/lipase associated with brush border membranes. J Biol Chem 273:2214eC2221, 1998
Melchior DL, Carruthers A, Makriyannis A, Duclos RI, Abdel-Mageed OH: Alterations in red blood cell sugar transport by nanomolar concentrations of alkyl lysophospholipid. Biochim Biophys Acta 1028:1eC8, 1990
Naderi S, Carruthers A, Melchior DL: Modulation of red blood cell sugar transport by lyso-lipid. Biochim Biophys Acta 985:173eC181, 1998
Hirosumi J, Tuncman G, Chang L, Gorgun CZ, Uysal KT, Maeda K, Karin M, Hotamisligil GS: A central role for JNK in obesity and insulin resistance. Nature 420:333eC336, 2002
Fang X, Gibson S, Flowers M, Furui T, Best RCJ, Mills GB: Lysophosphatidylcholine stimulates activator protein 1 and the c-Jun N-terminal kinase activity. J Biol Chem 272:13683eC13689, 1997
Lee YH, Giraud J, Davis RJ, White MF: c-Jun N-terminal kinase (JNK) mediates feedback inhibition of the insulin signaling cascade. J Biol Chem 278:2896eC2902, 2003
Motley ED, Kabir SM, Gardner CD, Eguchi K, Frank GD, Kuroki T, Ohba M, Yamakawa T, Eguchi S: Lysophosphatidylcholine inhibits insulin-induced akt activation through protein kinase C-alpha in vascular muscle cells. Hypertension 39:508eC512, 2002
Thors B, Hallorsson H, Clarke GD, Thorgeirsson G: Inhibition of Akt phosphorylation by thrombin, histamine and lysophosphatidylcholine in endothelial cells: differential role of protein kinase C. Atherosclerosis 168:245eC253, 2003
Leitges M, Plomann M, Standaert ML, Bandyopadhyay G, Sajan MP, Kanoh Y, Farese RV: Knockout of PKC alpha enhances insulin signaling through PI3K. Mol Endocrinol 16:847eC858, 2002
Standaert ML, Bandyopadhyay G, Galloway L, Soto J, Ono Y, Kikkawa U, Farese RV: Effects of knockout of the PKB-beta gene on glucose transport and glucose homeostasis. Endocrinology 140:4470eC4477, 1999
Farese RV: Insulin-sensitive phospholipid signaling systems and glucose transport. Update II. Exp Biol Med 226:283eC295, 2001
Bleich D, Polak M, Chen S, Swiderek KM, Levy-Marchal C: Sera from children with type 1 diabetes mellitus react against a new group of antigens composed of lysophospholipids. Hormone Res 52:86eC94, 1999
Rabini RA, Galassi R, Fumelli P, Dousset N, Solera ML, Valdiguie P, Curatola G, Ferretti G, Taus M, Mazzanti L: Reduced Na+-K+-ATPase activity and plasma lysophosphatidylcholine concentrations in diabetic patients. Diabetes 43:915eC919, 1994(Eric D. Labontee, R. Jaso)
G6Pase, glucose-6-phosphatase; JNK, c-Jun NH2-terminal kinase; LPC, lysophosphatidylcholine; PKC-, protein kinase C-; Pla2g1b, group 1B phospholipase A2
ABSTRACT
Postprandial hyperglycemia is an early indicator of abnormality in glucose metabolism leading to type 2 diabetes. However, mechanisms that contribute to postprandial hyperglycemia have not been identified. This study showed that mice with targeted inactivation of the group 1B phospholipase A2 (Pla2g1b) gene displayed lower postprandial glycemia than that observed in wild-type mice after being fed a glucose-rich meal. The difference was caused by enhanced postprandial glucose uptake by the liver, heart, and muscle tissues as well as altered postprandial hepatic glucose metabolism in the Pla2g1beC/eC mice. These differences were attributed to a fivefold decrease in the amount of dietary phospholipids absorbed as lysophospholipids in Pla2g1beC/eC mice compared with that observed in Pla2g1b+/+ mice. Elevating plasma lysophospholipid levels in Pla2g1beC/eC mice via intraperitoneal injection resulted in glucose intolerance similar to that exhibited by Pla2g1b+/+ mice. Studies with cultured hepatoma cells revealed that lysophospholipids dose-dependently suppressed insulin-stimulated glycogen synthesis. These results demonstrated that reduction of lysophospholipid absorption enhances insulin-mediated glucose metabolism and is protective against postprandial hyperglycemia.
Postprandial hyperglycemia is an early indicator of abnormality in glucose metabolism leading to type 2 diabetes. However, mechanisms responsible for postprandial hyperglycemia have not been identified to date. Previously, we reported that the Pla2g1beC/eC mice defective in expression of pancreatic group 1B phospholipase A2 (Pla2g1b) were resistant to diet-induced obesity and diabetes (1). The marginal decrease in lipid absorption efficiency cannot account for the resistance to diet-induced obesity and the increased insulin sensitivity in Pla2g1beC/eC mice because more fat was absorbed by fat-fed Pla2g1beC/eC mice than chow-fed Pla2g1b+/+ mice, yet their weights and plasma glucose levels were similar after 4 months of being fed their respective diets (1). Moreover, increased insulin sensitivity was observed not only in fat-fed Pla2g1beC/eC mice compared with Pla2g1b+/+ mice, but also in Pla2g1beC/eC mice fed in basal low-fat/low-cholesterol dietary conditions, when lipid absorption efficiency was identical to that of Pla2g1b+/+ mice. These results suggested that Pla2g1b has a direct contributory role in diet-induced obesity and diabetes.
Pla2g1b is a lipolytic enzyme that is secreted by the pancreas in response to the ingestion of a meal. Because of its abundant availability, stability, and ease of isolation, Pla2g1b is one of the most extensively studied enzymes in terms of structure and mechanism of action. However, despite the wealth of information known about the biochemical and structural characteristics of this protein (2,3), the exact physiological function of Pla2g1b has not been completely delineated. Based on its enzymatic activity against phospholipids and its abundant presence in the digestive tract, Pla2g1b has generally been thought to function only in nutrient absorption, catalyzing the hydrolysis of phospholipids in the intestinal lumen to facilitate dietary lipid absorption (4). However, our recent results with Pla2g1beC/eC mice showed that although phospholipid digestion in the intestinal lumen is a prerequisite for dietary lipid absorption (1,5), additional enzyme(s) in the digestive tract can compensate for the lack of Pla2g1b for phospholipid digestion in the intestinal lumen of Pla2g1beC/eC mice, and Pla2g1b is not required for normal lipid absorption under basal dietary conditions (5). Nevertheless, the Pla2g1beC/eC mice were more insulin sensitive and resistant to diet-induced obesity and diabetes than their wild-type counterparts. The current study was undertaken to identify the mechanism by which Pla2g1b contributes to diet-induced insulin resistance. Because Pla2g1b is secreted to the intestinal lumen in response to meal feeding, where it catalyzes phospholipid hydrolysis, we explored the possibility that lysophospholipids, the product formed by Pla2g1b hydrolysis of phospholipids, may promote postprandial hyperglycemia and contribute to diet-induced insulin resistance.
RESEARCH DESIGN AND METHODS
Pla2g1beC/eC mice were generated by homologous recombination in embryonic stem cells and backcrossed 10 times into the C57BL/6 background as described previously (5). Wild-type C57BL/6j mice were used as Pla2g1b+/+ controls. All animals were maintained in a temperature- and humidity-controlled room with a 12-h light/dark cycle. The animals were fed a rodent chow (LM485; Harlan-Teklad, Madison, WI) with free access to water. All animal protocols used in this study were approved by the institutional animal care and use committee at the University of Cincinnati.
Glucose tolerance tests.
Pla2g1b+/+ and Pla2g1beC/eC mice were fed with either basal low-fat/low-glucose diet (D12328; Research Diets, New Brunswick, NJ) or diabetogenic high-fat/high-carbohydrate diet (D12331; Research Diets) for 1 week and fasted overnight for 12 h before experiments. On the morning of the study, Pla2g1b+/+ and Pla2g1beC/eC mice were fed 4 ml/kg body wt of a bolus glucose-lipid mixed meal containing 50% glucose, 2.6 mmol/l egg phosphatidylcholine, 13.33 mmol/l triolein, and 2.6 mmol/l cholesterol in PBS for oral glucose tolerance tests. Glucose tolerance tests were also conducted by injecting a 50% glucose-PBS solution (2 g glucose/kg of body wt) into the intraperitoneal cavity. Blood was obtained by tail vein bleeding before and at different times after administration of the test meal. Blood glucose and insulin levels were measured by colorimetric assay and radioimmunoassay kits obtained from Wako Chemicals (Richmond, VA) and Linco Research (St. Charles, MO), respectively. For glucose tolerance tests involving testing the effect of lysophosphatidylcholine (LPC), saline with or without LPC (Sigma-Aldrich, St. Louis, MO) was injected (32 mg/kg of body wt i.p.) 5 min before glucose bolus administration.
Tissue glucose uptake assay.
Pla2g1b+/+ and Pla2g1beC/eC mice fed with the basal low-fat/low-glucose diet were injected with glucose (2 g/kg body wt i.p.) containing 5 e藽i 2-deoxy-[3H]glucose after an overnight fast. Animals were killed after 30 min, and tissues were collected and homogenized. The amount of radiolabeled 2-deoxyglucose taken up by each tissue was quantified by liquid scintillation.
Phospholipid absorption studies.
Mice were fasted overnight and then fed by gavage 100 e蘬 of the glucose-lipid mixed meal containing trace amounts of phosphatidyl-[3H]choline. The mice were anesthetized with ketamine/xylazine after 2 h. Blood was collected from the mesenteric portal vein, and the liver was removed. Lipid was extracted with CHCl3:methanol (2:1, vol/vol), and radioactivity was measured to determine the amount of phosphatidyl-[3H]choline absorbed as lysophosphatidyl-[3H]choline and/or phosphatidyl-[3H]choline.
LPC assay.
The concentrations of LPC in plasma and liver homogenates were determined by a modification of the enzymatic assay procedure described by Kishimoto et al. (6). Briefly, 8 e蘬 of sample was preincubated for 5 min in 240 e蘬 of reagent A (0.1 mol/l Tris-HCl, pH 8.0, 0.01% Triton X100, 1 mmol/l CaCl2, 3 mmol/l N-ethyl-N-(2-hydroxy-3-sulfopropyl)-3-methylaniline sodium dihydrate; 10 kU/l peroxidase, 0.1 kU/l glycerophosphorylcholine phosphodiesterase, 1 kU/l choline oxidase) in 96-well plates. The initial absorbance at 600 nm was recorded before initiating the reaction with the addition of 80 e蘬 of reagent B (0.1 mol/l Tris-HCl, pH 8.0, 0.01% Triton X100, 5 mmol/l 4-aminoantipyrine, and 1 kU/l phospholipase B). The incubation was continued for 30 min at 37°C, and the final absorbance at 600 nm was recorded. Total LPC concentration in each sample was calculated by comparing the change in absorbance between the standard and samples after subtraction of the background absorbance recorded before initiation of the reaction. The accuracy of this method was confirmed by comparing concentrations determined by this enzymatic assay with high-performance liquid chromatography analysis.
Cell culture experiments.
Human HepG2 and rat H4-2E hepatoma cell lines were obtained from the American Type Culture Collection (Manassas, VA) and maintained at 37°C under 5% CO2 atmosphere in low-glucose Dulbecco’s modified Eagle’s medium containing 10% fetal bovine serum. Confluent cultures were serum-starved with Dulbecco’s modified Eagle’s medium, and subsequent additions of 5 mmol/l [14C]glucose (0.25 mCi/l), BSA, and LPC were made for a 30-min preincubation. Insulin or empty vehicle was then added to the samples indicated, and incubation was continued for 2 h at 37°C. At the end of the incubation period, medium was removed, and cells were washed extensively with ice-cold PBS and then lysed with 30% KOH containing 5 mg/ml carrier glycogen. The lysate was incubated for 1 h at 60°C, and the glycogen was precipitated with four volumes of ice-cold 95% ethanol. The precipitated glycogen was dissolved in 0.2 N HCl, and the amount of [14C]glucose incorporated into glycogen was determined by liquid scintillation counting. For RNA isolation and quantification, we used cells in duplicate wells treated identically except without radioactive glucose.
RNA quantification.
Total RNA was isolated from liver, using a micro-to-midi spin column purification system from Invitrogen (Carlsbad, CA). Then, 5 e蘥 of total RNA was reverse-transcribed with oligo(dT), using a Superscript first-strand synthesis system obtained from Invitrogen. Real-time PCR assays were performed in a BioRad iCycler, using 50 ng of the prepared cDNA with 0.2 e蘭ol/l of each primer indicated in Table 1, 1 x PCR buffer (Invitrogen), 0.2 mmol/l dNTPs, 0.5 x SYBR Green (Molecular Probes), 10 nmol/l fluorescein, and 1 unit Taq polymerase (Invitrogen). Each reaction also contained Mg2+ at various concentrations, as indicated in Table 1. The samples were denatured at 95°C for 10 min, followed by 40 cycles of 95°C for 30 s and then 30 s at the temperature indicated in Table 1. Amplification of specific transcripts was confirmed by product melt curve analysis. Results are reported as arbitrary units and are normalized to the expression level of cyclophilin mRNA in each sample. The expression levels of each mRNA in wild-type mice under fasting conditions were set to equal 1.0. Statistical analyses were performed, using Student’s t test.
RESULTS
Reduced lysophospholipid absorption in Pla2g1beC/eC mice.
Although we previously showed no difference between Pla2g1b+/+ and Pla2g1beC/eC mice in phospholipid digestion and absorption of the associated fatty acyl chains through the gastrointestinal tract, it remains to be clarified whether the phospholipids were digested and absorbed as nonesterified fatty acids and glycerol 3-phosphorylcholine or as fatty acids and lysophospholipids. The difference in glycerol 3-phosphorylcholine versus lysophospholipid absorption and transport to the liver may have significant impact on regulation of hepatic glucose metabolism. Therefore, additional experiments were performed to test this possibility. A glucose-lipid mixed meal containing phosphatidyl-[3H]choline was prepared and fed to Pla2g1b+/+ and Pla2g1beC/eC mice after a 12-h fast. Blood was collected from the portal vein after 2 h, and lipids were extracted for scintillation counting to determine the amount of the [3H]choline head group absorbed intact with the lipid backbone, likely as lysophosphosphatidyl-[3H]choline. Results showed that a significant amount of the [3H]choline was absorbed as lipid moiety in portal blood of both Pla2g1b+/+ and Pla2g1beC/eC mice (Fig. 1A). Importantly, the lipid-associated [3H]choline in portal blood of Pla2g1beC/eC mice was fivefold lower than that in the Pla2g1b+/+ mice (Fig. 1A).
Hepatic and circulating lysophospholipid levels in Pla2g1b+/+ and Pla2g1beC/eC mice.
The difference between Pla2g1b+/+ and Pla2g1beC/eC mice in the absorption of luminal phospholipids as lysophospholipids was confirmed by measuring lysophospholipid concentrations in plasma and in the liver of these animals under fasting conditions as well as after feeding a glucose-lipid mixed meal. No significant difference in either plasma or liver lysopholipid concentrations was observed between Pla2g1b+/+ and Pla2g1beC/eC mice under fasting conditions (data not shown). Feeding of a glucose-lipid mixed meal resulted in the elevation of lysophospholipid levels in the plasma and liver of Pla2g1b+/+ mice fed control diet (Fig. 1B and C, respectively). In contrast, lysophospholipid levels in plasma and liver of Pla2g1beC/eC mice fed control diet remained at fasting levels after feeding the glucose-lipid mixed meal. Furthermore, whereas both mouse lines fed diabetogenic diet had increased hepatic LPC levels, an exaggerated decrease in circulating LPC levels was displayed in mice lacking Pla2g1b. Thus, the meal-induced increase in lysophospholipid levels is most likely caused by Pla2g1b digestion of the dietary and biliary phospholipids in the intestinal lumen and the absorption of the digested product lysophospholipid through the portal blood. The lack of postprandial increase in plasma and liver lysophospholipid levels in the Pla2g1beC/eC mice is consistent with this interpretation.
Glucose tolerance in Pla2g1b+/+ and Pla2g1beC/eC mice.
Our previous studies showed that Pla2g1b+/+ and Pla2g1beC/eC mice were similar in oral glucose tolerance when maintained on a low-fat chow diet. However, the Pla2g1beC/eC mice were more insulin sensitive and displayed a more rapid glucose clearance than Pla2g1b+/+ mice after feeding a high-fat/high-cholesterol Western-type diet for 15 weeks (1). In these studies, glucose tolerance tests were administered after intraperitoneal injection of a bolus load of glucose in saline solution. Because postprandial hyperglycemia is associated with type 2 diabetes after ingestion of a high-carbohydrate/high-lipid mixed meal (7), we compared plasma glucose levels in chow-fed Pla2g1b+/+ and Pla2g1beC/eC mice after feeding them an oral glucose meal (2 g/kg body wt) containing 2.6 mmol/l phosphatidylcholine, 2.6 mmol/l cholesterol, and 13.33 mmol/l triglyceride. In contrast to previous results showing no difference in glucose tolerance between chow-fed Pla2g1b+/+ and Pla2g1beC/eC mice (1), the inclusion of fat in the oral glucose test meal resulted in the Pla2g1beC/eC mice displaying significantly less postprandial hyperglycemia compared with that observed in the wild-type Pla2g1b+/+ mice (Fig. 2A). Furthermore, the difference in glucose tolerance after mixed-meal feeding was still observed after chronic feeding of the animals with a high-fat/high-glucose diet (data not shown). The age of the mice in which diet-induced hyperglycemia has been achieved is also known to affect glucose tolerance. As such, younger adult 13-week-old mice were tested for sensitivity to glucose via intraperitoneal injection of a bolus load of glucose in saline solution. Postprandial hyperglycemia was found to be reduced in Pla2g1beC/eC mice compared with Pla2g1b+/+ mice, in contrast to previous studies utilizing much older mice, despite no accompanying oral administration of a lipid bolus (Fig. 2B). These results suggest that postprandial hyperglycemia resistance observed in Pla2g1beC/eC mice is not limited to a state of extended high-fat feeding, as previously reported, but is also readily apparent in younger chow-fed animals.
Altered luminal phospholipid absorption and decreased postprandial hepatic lysophospholipids may be contributing to the increased postprandial hyperglycemia resistance phenotype observed in Pla2g1beC/eC mice. To test this possibility, intraperitoneal injections of saline with and without LPC (32 mg/kg body wt) were administered to Pla2g1beC/eC mice 5 min before testing glucose tolerance with either oral or intraperitoneal delivery of a glucose bolus (2 g/kg body wt). Blood glucose levels were elevated after glucose ingestion in Pla2g1beC/eC mice injected with LPC in saline compared with mice injected with saline alone (Fig. 2C and D). Plasma glucose levels in Pla2g1beC/eC mice injected with LPC were similar to those of Pla2g1b+/+ mice (Fig. 2A). Therefore, regardless of whether glucose is delivered orally or via intraperitoneal injection, the elevation of LPC in the circulation of Pla2g1beC/eC mice, including hepatic portal circulation, results in a glucose tolerance reminiscent of Pla2g1b+/+ mice. The reintroduction of LPC into postprandial Pla2g1beC/eC mice and the associated elimination of the hyperglycemia resistance phenotype suggest a link between circulating LPC levels and glucose intolerance.
Additional experiments were also performed to evaluate the tissues responsible for increased glucose uptake in Pla2g1beC/eC mice. In these experiments, Pla2g1b+/+ and Pla2g1beC/eC mice were injected intraperitoneally with a glucose load (2 g/kg body wt) containing 2-deoxy-[3H]glucose after an overnight fast. Various high-energy metabolism tissues were harvested from the animals after 30 min to determine tissue partitioning of the deoxy-[3H]glucose. Results showed that the Pla2g1beC/eC mice accumulated significantly more radiolabeled 2-deoxyglucose in liver, heart, and muscle tissues compared with that observed in Pla2g1b+/+ mice (Fig. 3A). The difference is less obvious in white adipose tissues, but a trend toward higher deoxyglucose uptake was also observed in the Pla2g1beC/eC mice. The increased glucose uptake by various tissues in Pla2g1beC/eC mice was not caused by increased insulin secretion by the pancreas because postprandial insulin concentrations in plasma of Pla2g1beC/eC mice were consistently lower but not statistically different from those observed in Pla2g1b+/+ mice (Fig. 3B). Thus, the Pla2g1beC/eC mice appeared to be more insulin sensitive, achieving greater glucose tolerance with similar or lower levels of postprandial insulin release than the Pla2g1b+/+ mice.
Lysophospholipid-dependent reduction of insulin-stimulated glycogen synthesis in hepatoma cell lines.
The mechanistic relationship between differences in postprandial lysophospholipid levels and hyperglycemia in Pla2g1b+/+ and Pla2g1beC/eC mice was explored by examining the influence of LPC on glucose metabolism in hepatoma cell lines. The inclusion of LPC with BSA in the incubation medium had no effect on the ability of HepG2 cells to utilize glucose for glycogen biosynthesis under basal conditions (Fig. 4A). However, insulin-stimulated glycogen biosynthesis increased approximately twofold in the absence of LPC, and the inclusion of LPC in the incubation medium reduced insulin-stimulated glycogen biosynthesis in a dose-dependent manner. Similar results were found in rat hepatoma H4-2E cells, except that higher concentrations of LPC lowered glycogen synthesis below basal levels (Fig. 4B). Analysis of RNA isolated from these cells revealed LPC suppression of glucose kinase gene expression under both basal and insulin-stimulated conditions, but had no effect on glucose-6-phosphatase (G6Pase) gene expression under both stimulated and unstimulated conditions (Fig. 5A and C). These results suggest that LPC may play a role in suppressing insulin-induced glycogen synthesis.
Expression of hepatic glucose kinase and G6Pase in Pla2g1b+/+ and Pla2g1beC/eC mice.
Increasing recent evidence has indicated that one mechanism contributing to abnormal postprandial glycemia is the failure to suppress hepatic glucose production (7,8). Although the in vitro cell culture data showed no difference in G6Pase mRNA levels in cells incubated with or without LPC, it remains possible that the higher postprandial LPC level in Pla2g1b+/+ mice resulted in less suppression of hepatic glucose production compared with that observed in Pla2g1beC/eC mice. We isolated RNA from the liver of both fasting and glucose/lipid-fed mice to test this possibility. Results showed no difference between Pla2g1b+/+ and Pla2g1beC/eC mice in hepatic expression of glucose kinase and G6Pase under fasting conditions (Fig. 5B and D). The ingestion of a bolus glucose-lipid mixed meal did not alter hepatic expression of the gluconeogenic enzyme G6Pase in Pla2g1b+/+ mice (Fig. 5B). However, hepatic G6Pase expression was reduced twofold in Pla2g1beC/eC mice after feeding them the glucose-lipid mixed meal. In contrast, expression of glucose kinase was induced in both Pla2g1b+/+ and Pla2g1beC/eC mice after feeding the glucose-lipid mixed meal (Fig. 5D). Interestingly, the induction of glucose kinase expression was significantly greater in the Pla2g1beC/eC mice compared with that observed in Pla2g1b+/+ mice. These results suggest increased glucose uptake and/or utilization for glycogen biosynthesis in the Pla2g1beC/eC mice compared with Pla2g1b+/+ mice. The expression levels of glucose kinase and G6Pase in the livers of Pla2g1beC/eC mice in both fasting and postprandial states are consistent with the hyperglycemia resistance phenotype reported.
DISCUSSION
Phospholipid hydrolysis in the intestinal lumen is required before triglyceride digestion and absorption of dietary fat and cholesterol through brush border membranes (9eC13). The major enzyme for phospholipid hydrolysis in the intestinal lumen is Pla2g1b secreted by the pancreas in response to meal feeding. However, other lipolytic enzyme(s) can serve a compensatory function in catalyzing phospholipid hydrolysis in the absence of Pla2g1b to sustain fat and cholesterol absorption (5). The difference in phospholipid hydrolysis between Pla2g1b+/+ and Pla2g1beC/eC mice, as shown in this study, is the products formed and absorbed by the animals. The major phospholipid digestive products absorbed by Pla2g1b+/+ mice are nonesterified fatty acids and LPC. The absorption of dietary phospholipids as lysophospholipids is consistent with previous reports demonstrating >90% of dietary phosphatidyl-[3H]choline is normally converted to lysophosphatidyl-[3H]choline and transported directly to the liver via the portal circulation (14). However, in animals lacking a functional Pla2g1b gene, significantly less dietary phosphatidylcholine was absorbed as LPC (Fig. 1A), even though they absorbed similar levels of the fatty acyl moiety of dietary phospholipids as the Pla2g1b+/+ mice. Thus, the compensatory enzyme for phospholipid digestion in the absence of Pla2g1b is likely to be phospholipase B, a distal intestinal enzyme that hydrolyzes phospholipids at both the sn-1 and sn-2 positions to produce nonesterified fatty acids and glycerophosphocholine (15,16).
The physiological significance of LPC generated during luminal phospholipid digestion and absorption has not been investigated previously. The availability of Pla2g1b+/+ and Pla2g1beC/eC mice, with different capabilities of LPC absorption and transport, provided the opportunity to evaluate the impact of LPC absorption in physiology and pathophysiology. The results of this study showed that LPC absorbed after Pla2g1b digestion of phospholipids in the intestinal lumen contributes to postprandial hyperglycemia by inhibiting glucose uptake by the liver, heart, and muscle tissues. The reduced level of LPC transported to the liver of Pla2g1beC/eC mice also resulted in increased postprandial glucose kinase expression (Fig. 5). The increased expression of the glucose utilization enzyme glucose kinase implies that postprandial hepatic glucose uptake and its utilization for glycogen synthesis are more efficient in Pla2g1beC/eC mice than in Pla2g1b+/+ mice. The data also showed decreased expression of G6Pase after meal feeding in the Pla2g1beC/eC mice, suggesting that the Pla2g1beC/eC mice were capable of suppressing hepatic glucose production in response to the glucose-lipid mixed meal. In contrast, postprandial suppression of G6Pase expression was not observed in Pla2g1b+/+ mice. This failure to suppress G6Pase expression after a glucose-lipid mixed meal may account for postprandial hyperglycemia and susceptibility to diet-induced insulin resistance and diabetes in Pla2g1b+/+ mice. Likewise, the ability to downregulate hepatic glucose production, as well as elevated glucose uptake by the liver, heart, and muscle tissues, after glucose feeding may account for the resistance of the Pla2g1beC/eC mice to diet-induced insulin resistance and diabetes. The increased glucose uptake in tissues of Pla2g1beC/eC mice also provided evidence that LPC may serve as a negative regulator of this insulin-stimulated process (Fig. 3A). Previous studies have shown that LPC can directly inhibit protein-mediated glucose uptake via interaction with glucose transport proteins on cell membranes (17,18). Additionally, LPC may inhibit insulin-stimulated glucose kinase expression in the liver, as demonstrated in cell culture of both HepG2 and H4-2E hepatoma cell lines (Fig. 4), in promoting postprandial hyperglycemia. The LPC-induced decrease in glucose kinase mRNA levels in HepG2 cells, along with elevated glucose kinase mRNA levels observed in lipid/glucose-fed Pla2g1beC/eC mice compared with those in Pla2g1b+/+ mice, would suggest a direct role of LPC in glucose kinase gene regulation. Interestingly, despite the lack of an LPC effect on G6Pase mRNA levels in cell culture experiments, G6Pase mRNA levels were found to be lower in lipid/glucose-fed Pla2g1beC/eC mice compared with those in Pla2g1b+/+ mice. Thus, the reduced G6Pase mRNA levels observed in vivo may not be a direct effect of LPC, but possibly an indirect effect via communication with other cell types or organs in the intact animal.
The phenotype of resistance to diet-induced obesity and diabetes observed in Pla2g1beC/eC mice (1) is similar to that observed in JNKeC/eC mice with targeted inactivation of the c-Jun NH2-terminal kinase (JNK) gene (19). Because LPC has been shown to activate JNK in fibroblasts and HeLa cells (20), and JNK inhibits insulin signaling cascade by inhibiting tyrosine phosphorylation of insulin receptor substrate-1 via serine phosphorylation (21), it is possible that LPC inhibition of glucose uptake by liver, heart, and muscle tissues as well as its suppression of hepatic glucose metabolism are also mediated through activation of JNK. Alternatively, LPC may also inhibit insulin-stimulated glucose metabolism by activation of protein kinase C- (PKC-). Previously, LPC activation of PKC- resulting in the inhibition of insulin-induced protein kinase B/Akt phosphorylation was observed in smooth muscle (22) and endothelial cells (23). Insulin signaling through phosphatidylinositol 3-kinase and protein kinase B/Akt activation was also enhanced in cells with an inactivated PKC- gene (24,25). Interestingly, the mechanism by which PKC- inhibits insulin signaling is also related to its ability to induce serine/threonine phosphorylation of insulin receptor substrates and block their tyrosine phosphorylation (26). Thus, LPC may inhibit postprandial insulin signaling in tissues via inhibiting tyrosine phosphorylation of insulin receptor substrates by activating either JNK or PKC- pathways. It must be noted, however, that these two pathways need not be mutually exclusive. Our future effort will focus on determining whether LPC inhibition of insulin signaling is mediated through either one or both of these pathways.
Regardless of the signaling mechanism(s) by which LPC inhibits postprandial hyperglycemia, results of the current study have major clinical implications. Previous studies have already demonstrated that failure to suppress endogenous glucose production after a glucose-lipid mixed meal is a major contributing factor toward hyperglycemia and type 2 diabetes (7). The results of this study further document that the Pla2g1b-digested product LPC is directly responsible for suppression of hepatic glucose metabolism. In view of studies showing increased LPC levels in patients with type 2 diabetes (27,28), the current study suggests that inhibition of Pla2g1b may be a viable strategy to prevent the onset of type 2 diabetes in individuals consuming a high-glucose/high-fat diet.
ACKNOWLEDGMENTS
This work was supported by Grants PO1 DK54504 and RO1 DK69967 from the National Institutes of Health (to D.Y.H.), National Research Service Award DK10065 (to K.W.H.), and Postdoctoral Fellowship 0525340B from the Ohio Affiliates of the American Heart Association (to E.D.L.).
We thank Josh Bashford and Todd Greer for their superb technical assistance and Dr. Susanna Hofmann for helpful discussions and advice.
FOOTNOTES
E.D.L. and R.J.K. contributed equally to this work.
K.W.H. is currently affiliated with the Department of Nutrition and Food Science, Auburn University, Auburn, Alabama.
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
REFERENCES
Huggins KW, Boileau AC, Hui DY: Protection against diet-induced obesity and obesity-related insulin resistance in group 1B PLA2-deficient mice. Am J Physiol 283:E994eCE1001, 2002
Dennis EA: Diversity of group types, regulation, and function of phospholipase A2. J Biol Chem 269:13057eC13060, 1994
Dijkstra BW, Renetseder R, Kalk KH, Hol WG, Drenth J: Structure of porcine pancreatic phospholipase A2 at 2.6 resolution and comparison with bovine phospholipase A2. J Mol Biol 168:163eC179, 1983
Carey MC, Small DM, Bliss CM: Lipid digestion and absorption. Annu Rev Physiol 45:651eC677, 1983
Richmond BL, Boileau AC, Zheng S, Huggins KW, Granholm NA, Tso P, Hui DY: Compensatory phospholipid digestion is required for cholesterol absorption in pancreatic phospholipase A2 deficient mice. Gastroenterology 120:1193eC1202, 2001
Kishimoto T, Soda Y, Matsuyama Y, Mizuno K: An enzymatic assay for lysophosphatidylcholine concentration in human serum and plasma. Clin Biochem 35:411eC416, 2002
Singhal P, Caumo A, Carey PE, Cobelli C, Taylor R: Regulation of endogenous glucose production after a mixed meal in type 2 diabetes. Am J Physiol 283:E275eCE283, 2002
Staehr P, Hother-Nielsen O, Beck-Nielsen H: The role of the liver in type 2 diabetes. Rev Endo Metab Disord 5:105eC110, 2004
Beil FU, Grundy SM: Studies on plasma lipoproteins during absorption of exogenous lecithin in man. J Lipid Res 21:525eC536, 1980
Homan R, Hamelehle KL: Phospholipase A2 relieves phosphatidylcholine inhibition of micellar cholesterol absorption and transport by human intestinal cell line Caco-2. J Lipid Res 39:1197eC1209, 1998
Mackay K, Starr JR, Lawn RM, Ellsworth JL: Phosphatidylcholine hydrolysis is required for pancreatic cholesterol esterase- and phospholipase A2-facilitated cholesterol uptake into intestinal Caco-2 cells. J Biol Chem 272:13380eC13389, 1997
Rampone AJ, Machida CM: Mode of action of lecithin in suppressing cholesterol absorption. J Lipid Res 22:744eC752, 1981
Young SC, Hui DY: Pancreatic lipase-colipase mediated triglyceride hydrolysis is required for cholesterol transport from lipid emulsions to intestinal cells. Biochem J 339:615eC620, 1999
Ikeda I, Imaizumi K, Sugano M: Absorption and transport of base moieties of phosphatidylcholine and phosphatidylethanolamine in rats. Biochim Biophys Acta 921:245eC253, 1987
Pind S, Kuksis A: Further characterization of a novel phospholipase B (phospholipase A2eClysophospholipase) from intestinal brush-border membranes. Biochem Cell Biol 69:346eC357, 1991
Tojo H, Ichida T, Okamoto M: Purification and characterization of a catalytic domain of rat intestinal phospholipase B/lipase associated with brush border membranes. J Biol Chem 273:2214eC2221, 1998
Melchior DL, Carruthers A, Makriyannis A, Duclos RI, Abdel-Mageed OH: Alterations in red blood cell sugar transport by nanomolar concentrations of alkyl lysophospholipid. Biochim Biophys Acta 1028:1eC8, 1990
Naderi S, Carruthers A, Melchior DL: Modulation of red blood cell sugar transport by lyso-lipid. Biochim Biophys Acta 985:173eC181, 1998
Hirosumi J, Tuncman G, Chang L, Gorgun CZ, Uysal KT, Maeda K, Karin M, Hotamisligil GS: A central role for JNK in obesity and insulin resistance. Nature 420:333eC336, 2002
Fang X, Gibson S, Flowers M, Furui T, Best RCJ, Mills GB: Lysophosphatidylcholine stimulates activator protein 1 and the c-Jun N-terminal kinase activity. J Biol Chem 272:13683eC13689, 1997
Lee YH, Giraud J, Davis RJ, White MF: c-Jun N-terminal kinase (JNK) mediates feedback inhibition of the insulin signaling cascade. J Biol Chem 278:2896eC2902, 2003
Motley ED, Kabir SM, Gardner CD, Eguchi K, Frank GD, Kuroki T, Ohba M, Yamakawa T, Eguchi S: Lysophosphatidylcholine inhibits insulin-induced akt activation through protein kinase C-alpha in vascular muscle cells. Hypertension 39:508eC512, 2002
Thors B, Hallorsson H, Clarke GD, Thorgeirsson G: Inhibition of Akt phosphorylation by thrombin, histamine and lysophosphatidylcholine in endothelial cells: differential role of protein kinase C. Atherosclerosis 168:245eC253, 2003
Leitges M, Plomann M, Standaert ML, Bandyopadhyay G, Sajan MP, Kanoh Y, Farese RV: Knockout of PKC alpha enhances insulin signaling through PI3K. Mol Endocrinol 16:847eC858, 2002
Standaert ML, Bandyopadhyay G, Galloway L, Soto J, Ono Y, Kikkawa U, Farese RV: Effects of knockout of the PKB-beta gene on glucose transport and glucose homeostasis. Endocrinology 140:4470eC4477, 1999
Farese RV: Insulin-sensitive phospholipid signaling systems and glucose transport. Update II. Exp Biol Med 226:283eC295, 2001
Bleich D, Polak M, Chen S, Swiderek KM, Levy-Marchal C: Sera from children with type 1 diabetes mellitus react against a new group of antigens composed of lysophospholipids. Hormone Res 52:86eC94, 1999
Rabini RA, Galassi R, Fumelli P, Dousset N, Solera ML, Valdiguie P, Curatola G, Ferretti G, Taus M, Mazzanti L: Reduced Na+-K+-ATPase activity and plasma lysophosphatidylcholine concentrations in diabetic patients. Diabetes 43:915eC919, 1994(Eric D. Labontee, R. Jaso)