Biochemical Pharmacology 95 (2015) 46–57
Contents lists available at ScienceDirect
Biochemical Pharmacology journal homepage: www.elsevier.com/locate/biochempharm
Impact of alogliptin and pioglitazone on lipid metabolism in islets of prediabetic and diabetic Zucker Diabetic Fatty rats Ying Cai a, Todd A. Lydic b, Thomas Turkette a, Gavin E. Reid b,c,1, L. Karl Olson a,* a
Department of Physiology, Michigan State University, East Lansing, MI 48824, USA Department of Biochemistry and Molecular Biology, Michigan State University, East Lansing, MI 48824, USA c Department of Chemistry, Michigan State University, East Lansing, MI 48824 USA b
A R T I C L E I N F O
A B S T R A C T
Article history: Received 26 January 2015 Accepted 13 March 2015 Available online 20 March 2015
Prolonged exposure of pancreatic beta (b) cells to elevated glucose and free fatty acids (FFA) as occurs in type 2 diabetes results in loss of b cell function and survival. In Zucker Diabetic Fatty (ZDF) rats, b cell failure is associated with increased triacylglyceride (TAG) synthesis and disruption of the glycerolipid/ FFA (GL/FFA) cycle, a critical arm of glucose-stimulated insulin secretion (GSIS). The aim of this study was to determine the impact of activation of PPARg and increased incretin action via dipeptidyl-peptidase inhibition using pioglitazone and/or alogliptin, respectively, on islet lipid metabolism in prediabetic and diabetic ZDF rats. Transition of control prediabetic ZDF rats to diabetes was associated with reduced plasma insulin levels, reduced islet insulin content and GSIS, reduced stearoyl-CoA desaturase 2 (SCD 2) expression, and increased islet TAG, diacylglyceride (DAG) and ceramides species containing saturated FA. Treatment of prediabetic ZDF rats with a combination of pioglitazone and alogliptin, but not individually, prevented the transition to diabetes and was associated with marked lowering of islet TAG and DAG levels. Pioglitazone and alogliptin, however, did not restore SCD2 expression, the degree of FA saturation in TAG, DAG or ceramides, islet insulin content, or lower ceramide levels. These findings are consistent with activation of PPARg and increased incretin action working in concert to restore GL/FFA cycle in b cells of ZDF rats. Restoration of the GL/FFA cycle without correcting islet FA desaturation, production of islet ceramides, and/or insulin sensitivity, however, may place these islets at risk for b cell failure. ß 2015 Elsevier Inc. All rights reserved.
Keywords: Lipotoxicity Peroxisome proliferator-activated receptor g (PPARg) Dipeptidyl-peptidase 4 (DPP-4) Triacylglyceride Diacylglyceride Insulin secretion
1. Introduction Type 2 diabetes (T2D) develops when pancreatic beta (b) cells fail to secrete sufficient amounts of insulin necessary to overcome peripheral insulin resistance and maintain glucose homeostasis. Exposure of b cells to excessive levels of glucose and free fatty acids (FFA), as occurs during obesity and T2D, causes b cell failure through depletion of intracellular insulin stores [1], ER stress [2,3], oxidative stress [4,5], inflammation [6], and excessive production
* Corresponding author at: Department of Physiology, Michigan State University, Biomedical Physical Science, 567 Wilson Rd, East Lansing, MI 48824-3320, USA. Tel.: +1 517 884 5116. E-mail addresses: [email protected] (Y. Cai), [email protected] (T.A. Lydic), [email protected] (T. Turkette), [email protected] (G.E. Reid), [email protected] (L.K. Olson). 1 Present address: School of Chemistry, Department of Biochemistry and Molecular Biology, Bio21 Molecular Science and Biotechnology Institute, University of Melbourne, Victoria 3010, Australia. http://dx.doi.org/10.1016/j.bcp.2015.03.010 0006-2952/ß 2015 Elsevier Inc. All rights reserved.
of ceramides and esterified lipids [7,8]—so called glucolipotoxicity. Recently, it has been recognized that lipid and glucose metabolic flux through the glycerol lipid/free fatty acid (GL/FFA) cycle plays a critical role in many aspects of b cell function including insulin secretion, b cell growth, compensation for insulin resistance, and detoxification of excessive fuel (reviewed in refs. [9,10]). Case in point, in obese insulin-resistant yet diabetes-resistant Zucker Fatty (ZF) rats, b cells compensate for insulin resistance through increased b cell mass and enhanced glucose and lipid metabolism, which is accompanied by hypersecretion of insulin [11,12]. This compensation is associated with increased FFA esterification into GL (e.g. TAG and DAG) and glucose-sensitive lipolysis of GL, presumably resulting in the generation of nutrient-secretion coupling factors [10] and no buildup of islet GL [12]. In contrast, obese insulin-resistant yet diabetes-prone Zucker Diabetic Fatty (ZDF) rats have enhanced islet lipogenesis and lipid esterification, but defective lipolysis of GL, resulting in increased islet TAG levels and b cell dysfunction and apoptosis [13–16]. Therapeutic approaches targeted to improve glucose and lipid flux through
Y. Cai et al. / Biochemical Pharmacology 95 (2015) 46–57
the b cell GL/FFA cycle would hence be predicted to improve b cell compensation in states of insulin resistance and T2D, and protect b cells from excessive nutrient fuels. Current pharmacological strategies for treatment of T2D include reducing insulin resistance and/or increasing insulin secretion. Thiazolidinediones (TZD), which function through activation of peroxisome proliferator-activated receptor-g (PPARg), predominantly increase insulin sensitivity by promoting adipose lipid storage and lowering circulatory lipid levels, thereby reducing inappropriate lipid storage in non-adipose tissues including skeletal muscle and liver (reviewed in ref. [17]). TZD have also been shown to have protective effects on b cell function and islet structure in rodent models of T2D [18–22]. Although in vivo evidence is limited, ex vivo studies suggest that TZD activation of PPARg protects against lipotoxicity [14,23] and regulates transcription of islet specific genes ([24] and references contained within). Other studies, however, have suggested that PPARg plays little role in normal b cell function [24], and that TZD such as pioglitazone acutely suppresses GSIS and the GL/FFA cycle through PPARg-independent mechanisms [25,26]. Dipeptidyl peptidase-4 (DPP-4) inhibitors, which function to elevate incretin levels (e.g. GLP-1 and GIP), have also been shown to have protective effects on b cell function, and islet structure and gene expression in rodent models of T2D [27–31]. Whether incretin therapies or DPP4 inhibitors influence b cell function and islet structure in T2D by modulation of the GL/FFA cycle has not been established. In support of this possibility, GLP-1 and its stable analog, exendin-4, have been shown to increase lipolysis in HIT-T15 b cells and islets isolated from high fat diet fed mice [32,33], and GLP-1 has been shown to stimulate glucose-derived de novo FA synthesis and elongation that is utilized for TAG and membrane synthesis during b cell differentiation [34]. Combined treatment of db/db and ob/ob mice with pioglitazone and the selective DPP4 inhibitor alogliptin has been shown to improve glycemic control, plasma lipid levels, b cell function, and islet gene expression [28,35,36], but little has been reported on the impact of this drug combination on islet lipids or the GL/FFA cycle. In this study, the impact of pioglitazone (Pio) and/or alogliptin (Alo) on islet lipid metabolism was characterized during the transition from prediabetes to diabetes in ZDF rats in vivo. ZDF rats were chosen for these studies because the progression from prediabetes to diabetes occurs in a predictable manner between 7 and 13 weeks of age and is associated with impaired GL/FFA cycle, increased islet TAG levels, decreased mRNA levels of FA desaturases and elongases, loss of islet insulin content and structure, and b cell apoptosis [3,13–16]. Outcomes of this study show that combined treatment of ZDF rats with Pio plus Alo attenuates loss of glucose tolerance and progression to T2D, and
47
this is associated with correction of islet GL levels and possibly restoration of the GL/FFA cycle. 2. Materials and methods 2.1. Materials Alogliptin and pioglitazone were received from Takeda Pharmaceutical North America, Inc. All solvents used were high performance liquid chromatography (HPLC) grade. Isopropanol, methanol, and chloroform were purchased from Macron Chemical (St. Louis, Missouri). HPLC water was from Fisher (Hampton, New Hampshire). Ammonium formate was purchased from Alfa Aesar (Ward Hill, Massachusetts). Internal standards used for relative quantitation of lipidomic data were obtained from Avanti Polar Lipids (Alabaster, AL) and included dimyristoyl (14:0/14:0) species of phosphatidylcholine (PC), phosphatidylethanolamine (PE), and phosphatidylserine (PS). RPMI-1640 medium, penicillin, streptomycin, and fetal bovine serum were from Life Technologies (Grand Island, New York). 2.2. Rats The Institutional Animal Care and Use committee at Michigan State University approved all animal treatments and procedures. Rats were kept on a 12-h light:12-h dark cycle with food and water ad libitum. Male Zucker Diabetic Fatty (ZDF-Leprfa/Crl) and control (+/?) rats were received at 5 weeks of age from Charles River Laboratories (Wilmington, Massachusetts) and maintained on Purina 5008 diet. At 6 weeks of age, ZDF rats were split into four drug treatment groups including: (1) control, (2) pioglitazone (2.5 mg per kg), (3) alogliptin (10 mg per kg), and (4) pioglitazone (2.5 mg per kg) plus alogliptin (10 mg per kg). Alogliptin and pioglitazone were prepared in 0.5% methylcellulose and delivered daily by oral gavage. Control groups also received daily oral gavage of 0.5% methylcellulose. Pioglitazone (3 mg per kg) was previously shown to increase insulin sensitivity and lower plasma TAG and FFA in Zucker Fatty rats [37,38], while not affecting insulin sensitivity or producing toxicity in control lean rats [39]. Alogliptin (10 mg per kg) was previously shown to inhibit DPP-4 activity 70% 24 h after treatment, while increasing plasma GLP-1 and insulin levels, and glucose tolerance in glucose-challenged Zucker Fatty rats [40]. Body weights were measured daily (Table 1). At 8 and 13 weeks of age (i.e. after 2 and 7 weeks of drug treatment), rats were given an oral glucose tolerance test (1 g glucose per kg body weight). Total area under the curve (AUC0-120) was determined from time 0 to 120 min using baseline fasting glucose concentrations of control (+/?) rats. Two to four days later, rats were
Table 1 Mean body weights and fold body weight change of ZDF rats treated with vehicle, pioglitazone and/or alogliptin for 2 or 6 weeks, respectively.
a
Animals
Treatment
Duration
Initial weighta
Final weight
Fold changeb
+/? (n = 12) ZDF (n = 12) ZDF (n = 12) ZDF (n = 12) ZDF (n = 12)
Vehicle Vehicle Pio Alo Pio + Alo
14 14 14 14 14
days days days days days
139.2 3.0 179.8 3.3 190.3 4.1 191.1 4.7 178.0 2.7
198.5 3.3 264.2 2.8 292.8 8.0* 279.1 4.5 282.8 2.9*
1.43 0.03 1.48 0.03 1.54 0.02 1.46 0.02# 1.59 0.01*,#,§
+/? (n = 8) ZDF (n = 8) ZDF (n = 8) ZDF (n = 8) ZDF (n = 8)
Vehicle Vehicle Pio Alo Pio + Alo
42 42 42 42 42
days days days days days
143.0 3.6 181.4 5.0 189.0 4.2 181.1 4.2 186.1 1.8
303.3 4.7 363.9 10.6 453.2 9.6* 402.3 7.4*,# 431.9 6.9*
2.13 0.03 2.01 0.02 2.40 0.05* 2.23 0.06* 2.32 0.05*
Initial body weights are from 6-week old rats at the beginning of treatment. Values are the mean SEM of the fold weight gain for each individual rat. * p < 0.05 versus vehicle treated ZDF. # p < 0.05 versus Pio treated ZDF. § p < 0.05 versus Alo treated ZDF b
48
Y. Cai et al. / Biochemical Pharmacology 95 (2015) 46–57
anesthetized and blood and islets were procured for analysis of fasting insulin levels, glucose-stimulated insulin secretion (GSIS), islet gene expression, and islet lipid composition. To avoid acute effects of diet on plasma lipid composition, insulin content and GSIS, and islet gene expression, samples for these analyses were procured from fasted rats. Islet lipid composition was determined on samples procured from fed rats to avoid fasting induced changes in islet lipid levels.
source (Advion Ithaca, New York) with a spray voltage of 1.4 kV and a gas pressure of 0.3 psi. The TSQ Vantage was operated with a capillary temperature of 225 8C, Q1 and Q3 isolation widths of 0.5 Da, and the collision cell set to 0.5 mTorr of argon. Samples were subjected to positive ion mode precursor ion scanning of m/z 184 for detection of PC lipids, and neutral loss scanning of 141 or 185 m/z for detection of PE and PS lipids, respectively. Collision energies were optimized for each individual lipid class of interest.
2.3. Islet isolation and glucose-stimulated insulin release
2.5.4. Detection of neutral lipids and sphingolipids by high resolution/ accurate mass MS Diluted islet samples were introduced to a high resolution/ accurate mass Thermo Scientific model LTQ Orbitrap Velos mass spectrometer operating in positive ion mode using a Triversa Nanomate with nESI conditions as described above. The ion source interface settings of the LTQ Orbitrap Velos (inlet temperature of 100 8C and S-Lens value of 50%) were optimized to maximize the sensitivity for precursor ions while minimizing in-source fragmentation. High resolution mass spectra were acquired using the FT analyzer operating at 100,000 resolving power, across the range of m/z from 200 to 1500, and were signal averaged for 2 min. All mass spectra were recalibrated offline using XCalibur software (Thermo Scientific, San Jose, California). Selected ions of interest were confirmed by targeted high resolution MS/MS analyses using the FT analyzer at 100,000 resolving power.
Control and ZDF islets were dissociated from pancreatic tissue by collagenase digestion and isolated by hand. Freshly isolated islets were used for islet gene expression and lipid analysis. For assessing GSIS, freshly isolated islets were allowed to recover for 2 h at 37 8C in RPMI-1640 (11.1 mM glucose) supplemented with 10% FBS, 100 units/ml penicillin, and 100 mg/ml streptomycin. After recovery, groups of 10 islets from individual rats were incubated twice in Krebs–Ringer bicarbonate buffer (KRBB) pH 7.4 containing 2.8 mM glucose and 0.1% BSA for 30 min, after which islets were incubated for 1 h in KRBB-0.1% BSA containing 2.8 or 16.7 mM glucose. Insulin released into the KRBB during the final incubation was normalized to intracellular insulin extracted from islets with acidified-ethanol (1.5% HCl in 75% ethanol). Plasma insulin and insulin from the GSIS assays were measured using a rat insulin radioimmunoassay kit (Millipore, Billerica, Massachusetts). 2.4. RNA analysis Levels of mRNA in islets were determined by quantitative real-time RT-PCR (qPCR) as previously described [41,42]. Primers used for qPCR are: ATGL 50 -CGGCATTTCAGACAACTTGCCACT-30 , 50 -GCAGGTTGAATTGGATGCTGGTGT-30 ; DGAT1 50 -AGACTGGGCAGCAACAAATGGATG-30 , 50 -CCACACAGCTGCATTGCCATAGTT30 ; SCD2 50 -ATGCCGGCTCACATACTG-30 , 50 -GACCAGTGTGATCCCGTACA-30 ; CPT1a 50 -AGACCGTGAGGAACTCAAACCCAT-30 , 50 -CACAACAATGTGCCTGCTGTCCTT-3 0 ; LCAD 50 -AATGGGAGAAAGCCGGAGAAGTGA-30 , 50 -GAAACCAGGGCCTGTGCAATTTGA; and cyclophilin 50 -CTTCTTGCTGGTCTTGCCATTCCT-30 , 50 -TGGATGGCAAGCATGTGGTCTTTG-30 . Gene expression data are reported relative to cyclophilin mRNA levels. 2.5. Methods for islet and plasma lipid analysis 2.5.1. Lipid extraction Islet lipids were extracted by a modified Folch method as previously described [43]. 2.5.2. Sample preparation Immediately prior to mass spectrometry analysis, islet lipids were centrifuged, loaded into Whatman multichem 96-well plates (Sigma–Aldrich, St. Louis, Missouri), dried under nitrogen, and reconstituted in 10 volumes of isopropanol/methanol/chloroform (4:2:1, v:v:v) containing 20 mM ammonium formate, 250 nM PC (14:0/14:0), 500 nM PE (14:0/14:0), and 500 nM PS(14:0/14:0). The 96-well plates were sealed with Teflon Ultra-Thin Sealing Tape (Analytical Sales and Services, Prompton Plains, New Jersey). 2.5.3. Detection of phospholipids by precursor ion and neutral loss scanning MS/MS Relative abundances of PC, PE, and PS lipids were determined as previously described [44]. Diluted islet samples were directly infused into a Thermo Scientific model TSQ Vantage triple quadrupole mass spectrometer (San Jose, California) using an Advion Triversa Nanomate nano-electrospray ionization (nESI)
2.5.5. Peak finding and quantitation of lipidomic data Automated peak finding, correction for 13C isotope effects, and quantitation of lipid molecular species were performed using Lipid Mass Spectrum Analysis (LIMSA) software version 1.0 [45]. All ion abundances for lipids analyzed on the Orbitrap Velos were quantitated against a single internal standard (dimyristoyl PC) [46], while phospholipid data obtained on the TSQ Vantage was quantitated against an internal standard of the appropriate lipid class. As no attempts were made to correct for differences in ionization efficiency among individual molecular species of various lipid classes, lipid molecular species are presented only as a fraction of the total normalized abundance of each lipid class. 2.6. Statistical analysis Islet studies used 4–6 animals per group. Statistical significance was determined using one-way ANOVA or two-way ANOVA followed by Tukey’s multiple comparison test using GraphPad Prism. p-Values
Contents lists available at ScienceDirect
Biochemical Pharmacology journal homepage: www.elsevier.com/locate/biochempharm
Impact of alogliptin and pioglitazone on lipid metabolism in islets of prediabetic and diabetic Zucker Diabetic Fatty rats Ying Cai a, Todd A. Lydic b, Thomas Turkette a, Gavin E. Reid b,c,1, L. Karl Olson a,* a
Department of Physiology, Michigan State University, East Lansing, MI 48824, USA Department of Biochemistry and Molecular Biology, Michigan State University, East Lansing, MI 48824, USA c Department of Chemistry, Michigan State University, East Lansing, MI 48824 USA b
A R T I C L E I N F O
A B S T R A C T
Article history: Received 26 January 2015 Accepted 13 March 2015 Available online 20 March 2015
Prolonged exposure of pancreatic beta (b) cells to elevated glucose and free fatty acids (FFA) as occurs in type 2 diabetes results in loss of b cell function and survival. In Zucker Diabetic Fatty (ZDF) rats, b cell failure is associated with increased triacylglyceride (TAG) synthesis and disruption of the glycerolipid/ FFA (GL/FFA) cycle, a critical arm of glucose-stimulated insulin secretion (GSIS). The aim of this study was to determine the impact of activation of PPARg and increased incretin action via dipeptidyl-peptidase inhibition using pioglitazone and/or alogliptin, respectively, on islet lipid metabolism in prediabetic and diabetic ZDF rats. Transition of control prediabetic ZDF rats to diabetes was associated with reduced plasma insulin levels, reduced islet insulin content and GSIS, reduced stearoyl-CoA desaturase 2 (SCD 2) expression, and increased islet TAG, diacylglyceride (DAG) and ceramides species containing saturated FA. Treatment of prediabetic ZDF rats with a combination of pioglitazone and alogliptin, but not individually, prevented the transition to diabetes and was associated with marked lowering of islet TAG and DAG levels. Pioglitazone and alogliptin, however, did not restore SCD2 expression, the degree of FA saturation in TAG, DAG or ceramides, islet insulin content, or lower ceramide levels. These findings are consistent with activation of PPARg and increased incretin action working in concert to restore GL/FFA cycle in b cells of ZDF rats. Restoration of the GL/FFA cycle without correcting islet FA desaturation, production of islet ceramides, and/or insulin sensitivity, however, may place these islets at risk for b cell failure. ß 2015 Elsevier Inc. All rights reserved.
Keywords: Lipotoxicity Peroxisome proliferator-activated receptor g (PPARg) Dipeptidyl-peptidase 4 (DPP-4) Triacylglyceride Diacylglyceride Insulin secretion
1. Introduction Type 2 diabetes (T2D) develops when pancreatic beta (b) cells fail to secrete sufficient amounts of insulin necessary to overcome peripheral insulin resistance and maintain glucose homeostasis. Exposure of b cells to excessive levels of glucose and free fatty acids (FFA), as occurs during obesity and T2D, causes b cell failure through depletion of intracellular insulin stores [1], ER stress [2,3], oxidative stress [4,5], inflammation [6], and excessive production
* Corresponding author at: Department of Physiology, Michigan State University, Biomedical Physical Science, 567 Wilson Rd, East Lansing, MI 48824-3320, USA. Tel.: +1 517 884 5116. E-mail addresses: [email protected] (Y. Cai), [email protected] (T.A. Lydic), [email protected] (T. Turkette), [email protected] (G.E. Reid), [email protected] (L.K. Olson). 1 Present address: School of Chemistry, Department of Biochemistry and Molecular Biology, Bio21 Molecular Science and Biotechnology Institute, University of Melbourne, Victoria 3010, Australia. http://dx.doi.org/10.1016/j.bcp.2015.03.010 0006-2952/ß 2015 Elsevier Inc. All rights reserved.
of ceramides and esterified lipids [7,8]—so called glucolipotoxicity. Recently, it has been recognized that lipid and glucose metabolic flux through the glycerol lipid/free fatty acid (GL/FFA) cycle plays a critical role in many aspects of b cell function including insulin secretion, b cell growth, compensation for insulin resistance, and detoxification of excessive fuel (reviewed in refs. [9,10]). Case in point, in obese insulin-resistant yet diabetes-resistant Zucker Fatty (ZF) rats, b cells compensate for insulin resistance through increased b cell mass and enhanced glucose and lipid metabolism, which is accompanied by hypersecretion of insulin [11,12]. This compensation is associated with increased FFA esterification into GL (e.g. TAG and DAG) and glucose-sensitive lipolysis of GL, presumably resulting in the generation of nutrient-secretion coupling factors [10] and no buildup of islet GL [12]. In contrast, obese insulin-resistant yet diabetes-prone Zucker Diabetic Fatty (ZDF) rats have enhanced islet lipogenesis and lipid esterification, but defective lipolysis of GL, resulting in increased islet TAG levels and b cell dysfunction and apoptosis [13–16]. Therapeutic approaches targeted to improve glucose and lipid flux through
Y. Cai et al. / Biochemical Pharmacology 95 (2015) 46–57
the b cell GL/FFA cycle would hence be predicted to improve b cell compensation in states of insulin resistance and T2D, and protect b cells from excessive nutrient fuels. Current pharmacological strategies for treatment of T2D include reducing insulin resistance and/or increasing insulin secretion. Thiazolidinediones (TZD), which function through activation of peroxisome proliferator-activated receptor-g (PPARg), predominantly increase insulin sensitivity by promoting adipose lipid storage and lowering circulatory lipid levels, thereby reducing inappropriate lipid storage in non-adipose tissues including skeletal muscle and liver (reviewed in ref. [17]). TZD have also been shown to have protective effects on b cell function and islet structure in rodent models of T2D [18–22]. Although in vivo evidence is limited, ex vivo studies suggest that TZD activation of PPARg protects against lipotoxicity [14,23] and regulates transcription of islet specific genes ([24] and references contained within). Other studies, however, have suggested that PPARg plays little role in normal b cell function [24], and that TZD such as pioglitazone acutely suppresses GSIS and the GL/FFA cycle through PPARg-independent mechanisms [25,26]. Dipeptidyl peptidase-4 (DPP-4) inhibitors, which function to elevate incretin levels (e.g. GLP-1 and GIP), have also been shown to have protective effects on b cell function, and islet structure and gene expression in rodent models of T2D [27–31]. Whether incretin therapies or DPP4 inhibitors influence b cell function and islet structure in T2D by modulation of the GL/FFA cycle has not been established. In support of this possibility, GLP-1 and its stable analog, exendin-4, have been shown to increase lipolysis in HIT-T15 b cells and islets isolated from high fat diet fed mice [32,33], and GLP-1 has been shown to stimulate glucose-derived de novo FA synthesis and elongation that is utilized for TAG and membrane synthesis during b cell differentiation [34]. Combined treatment of db/db and ob/ob mice with pioglitazone and the selective DPP4 inhibitor alogliptin has been shown to improve glycemic control, plasma lipid levels, b cell function, and islet gene expression [28,35,36], but little has been reported on the impact of this drug combination on islet lipids or the GL/FFA cycle. In this study, the impact of pioglitazone (Pio) and/or alogliptin (Alo) on islet lipid metabolism was characterized during the transition from prediabetes to diabetes in ZDF rats in vivo. ZDF rats were chosen for these studies because the progression from prediabetes to diabetes occurs in a predictable manner between 7 and 13 weeks of age and is associated with impaired GL/FFA cycle, increased islet TAG levels, decreased mRNA levels of FA desaturases and elongases, loss of islet insulin content and structure, and b cell apoptosis [3,13–16]. Outcomes of this study show that combined treatment of ZDF rats with Pio plus Alo attenuates loss of glucose tolerance and progression to T2D, and
47
this is associated with correction of islet GL levels and possibly restoration of the GL/FFA cycle. 2. Materials and methods 2.1. Materials Alogliptin and pioglitazone were received from Takeda Pharmaceutical North America, Inc. All solvents used were high performance liquid chromatography (HPLC) grade. Isopropanol, methanol, and chloroform were purchased from Macron Chemical (St. Louis, Missouri). HPLC water was from Fisher (Hampton, New Hampshire). Ammonium formate was purchased from Alfa Aesar (Ward Hill, Massachusetts). Internal standards used for relative quantitation of lipidomic data were obtained from Avanti Polar Lipids (Alabaster, AL) and included dimyristoyl (14:0/14:0) species of phosphatidylcholine (PC), phosphatidylethanolamine (PE), and phosphatidylserine (PS). RPMI-1640 medium, penicillin, streptomycin, and fetal bovine serum were from Life Technologies (Grand Island, New York). 2.2. Rats The Institutional Animal Care and Use committee at Michigan State University approved all animal treatments and procedures. Rats were kept on a 12-h light:12-h dark cycle with food and water ad libitum. Male Zucker Diabetic Fatty (ZDF-Leprfa/Crl) and control (+/?) rats were received at 5 weeks of age from Charles River Laboratories (Wilmington, Massachusetts) and maintained on Purina 5008 diet. At 6 weeks of age, ZDF rats were split into four drug treatment groups including: (1) control, (2) pioglitazone (2.5 mg per kg), (3) alogliptin (10 mg per kg), and (4) pioglitazone (2.5 mg per kg) plus alogliptin (10 mg per kg). Alogliptin and pioglitazone were prepared in 0.5% methylcellulose and delivered daily by oral gavage. Control groups also received daily oral gavage of 0.5% methylcellulose. Pioglitazone (3 mg per kg) was previously shown to increase insulin sensitivity and lower plasma TAG and FFA in Zucker Fatty rats [37,38], while not affecting insulin sensitivity or producing toxicity in control lean rats [39]. Alogliptin (10 mg per kg) was previously shown to inhibit DPP-4 activity 70% 24 h after treatment, while increasing plasma GLP-1 and insulin levels, and glucose tolerance in glucose-challenged Zucker Fatty rats [40]. Body weights were measured daily (Table 1). At 8 and 13 weeks of age (i.e. after 2 and 7 weeks of drug treatment), rats were given an oral glucose tolerance test (1 g glucose per kg body weight). Total area under the curve (AUC0-120) was determined from time 0 to 120 min using baseline fasting glucose concentrations of control (+/?) rats. Two to four days later, rats were
Table 1 Mean body weights and fold body weight change of ZDF rats treated with vehicle, pioglitazone and/or alogliptin for 2 or 6 weeks, respectively.
a
Animals
Treatment
Duration
Initial weighta
Final weight
Fold changeb
+/? (n = 12) ZDF (n = 12) ZDF (n = 12) ZDF (n = 12) ZDF (n = 12)
Vehicle Vehicle Pio Alo Pio + Alo
14 14 14 14 14
days days days days days
139.2 3.0 179.8 3.3 190.3 4.1 191.1 4.7 178.0 2.7
198.5 3.3 264.2 2.8 292.8 8.0* 279.1 4.5 282.8 2.9*
1.43 0.03 1.48 0.03 1.54 0.02 1.46 0.02# 1.59 0.01*,#,§
+/? (n = 8) ZDF (n = 8) ZDF (n = 8) ZDF (n = 8) ZDF (n = 8)
Vehicle Vehicle Pio Alo Pio + Alo
42 42 42 42 42
days days days days days
143.0 3.6 181.4 5.0 189.0 4.2 181.1 4.2 186.1 1.8
303.3 4.7 363.9 10.6 453.2 9.6* 402.3 7.4*,# 431.9 6.9*
2.13 0.03 2.01 0.02 2.40 0.05* 2.23 0.06* 2.32 0.05*
Initial body weights are from 6-week old rats at the beginning of treatment. Values are the mean SEM of the fold weight gain for each individual rat. * p < 0.05 versus vehicle treated ZDF. # p < 0.05 versus Pio treated ZDF. § p < 0.05 versus Alo treated ZDF b
48
Y. Cai et al. / Biochemical Pharmacology 95 (2015) 46–57
anesthetized and blood and islets were procured for analysis of fasting insulin levels, glucose-stimulated insulin secretion (GSIS), islet gene expression, and islet lipid composition. To avoid acute effects of diet on plasma lipid composition, insulin content and GSIS, and islet gene expression, samples for these analyses were procured from fasted rats. Islet lipid composition was determined on samples procured from fed rats to avoid fasting induced changes in islet lipid levels.
source (Advion Ithaca, New York) with a spray voltage of 1.4 kV and a gas pressure of 0.3 psi. The TSQ Vantage was operated with a capillary temperature of 225 8C, Q1 and Q3 isolation widths of 0.5 Da, and the collision cell set to 0.5 mTorr of argon. Samples were subjected to positive ion mode precursor ion scanning of m/z 184 for detection of PC lipids, and neutral loss scanning of 141 or 185 m/z for detection of PE and PS lipids, respectively. Collision energies were optimized for each individual lipid class of interest.
2.3. Islet isolation and glucose-stimulated insulin release
2.5.4. Detection of neutral lipids and sphingolipids by high resolution/ accurate mass MS Diluted islet samples were introduced to a high resolution/ accurate mass Thermo Scientific model LTQ Orbitrap Velos mass spectrometer operating in positive ion mode using a Triversa Nanomate with nESI conditions as described above. The ion source interface settings of the LTQ Orbitrap Velos (inlet temperature of 100 8C and S-Lens value of 50%) were optimized to maximize the sensitivity for precursor ions while minimizing in-source fragmentation. High resolution mass spectra were acquired using the FT analyzer operating at 100,000 resolving power, across the range of m/z from 200 to 1500, and were signal averaged for 2 min. All mass spectra were recalibrated offline using XCalibur software (Thermo Scientific, San Jose, California). Selected ions of interest were confirmed by targeted high resolution MS/MS analyses using the FT analyzer at 100,000 resolving power.
Control and ZDF islets were dissociated from pancreatic tissue by collagenase digestion and isolated by hand. Freshly isolated islets were used for islet gene expression and lipid analysis. For assessing GSIS, freshly isolated islets were allowed to recover for 2 h at 37 8C in RPMI-1640 (11.1 mM glucose) supplemented with 10% FBS, 100 units/ml penicillin, and 100 mg/ml streptomycin. After recovery, groups of 10 islets from individual rats were incubated twice in Krebs–Ringer bicarbonate buffer (KRBB) pH 7.4 containing 2.8 mM glucose and 0.1% BSA for 30 min, after which islets were incubated for 1 h in KRBB-0.1% BSA containing 2.8 or 16.7 mM glucose. Insulin released into the KRBB during the final incubation was normalized to intracellular insulin extracted from islets with acidified-ethanol (1.5% HCl in 75% ethanol). Plasma insulin and insulin from the GSIS assays were measured using a rat insulin radioimmunoassay kit (Millipore, Billerica, Massachusetts). 2.4. RNA analysis Levels of mRNA in islets were determined by quantitative real-time RT-PCR (qPCR) as previously described [41,42]. Primers used for qPCR are: ATGL 50 -CGGCATTTCAGACAACTTGCCACT-30 , 50 -GCAGGTTGAATTGGATGCTGGTGT-30 ; DGAT1 50 -AGACTGGGCAGCAACAAATGGATG-30 , 50 -CCACACAGCTGCATTGCCATAGTT30 ; SCD2 50 -ATGCCGGCTCACATACTG-30 , 50 -GACCAGTGTGATCCCGTACA-30 ; CPT1a 50 -AGACCGTGAGGAACTCAAACCCAT-30 , 50 -CACAACAATGTGCCTGCTGTCCTT-3 0 ; LCAD 50 -AATGGGAGAAAGCCGGAGAAGTGA-30 , 50 -GAAACCAGGGCCTGTGCAATTTGA; and cyclophilin 50 -CTTCTTGCTGGTCTTGCCATTCCT-30 , 50 -TGGATGGCAAGCATGTGGTCTTTG-30 . Gene expression data are reported relative to cyclophilin mRNA levels. 2.5. Methods for islet and plasma lipid analysis 2.5.1. Lipid extraction Islet lipids were extracted by a modified Folch method as previously described [43]. 2.5.2. Sample preparation Immediately prior to mass spectrometry analysis, islet lipids were centrifuged, loaded into Whatman multichem 96-well plates (Sigma–Aldrich, St. Louis, Missouri), dried under nitrogen, and reconstituted in 10 volumes of isopropanol/methanol/chloroform (4:2:1, v:v:v) containing 20 mM ammonium formate, 250 nM PC (14:0/14:0), 500 nM PE (14:0/14:0), and 500 nM PS(14:0/14:0). The 96-well plates were sealed with Teflon Ultra-Thin Sealing Tape (Analytical Sales and Services, Prompton Plains, New Jersey). 2.5.3. Detection of phospholipids by precursor ion and neutral loss scanning MS/MS Relative abundances of PC, PE, and PS lipids were determined as previously described [44]. Diluted islet samples were directly infused into a Thermo Scientific model TSQ Vantage triple quadrupole mass spectrometer (San Jose, California) using an Advion Triversa Nanomate nano-electrospray ionization (nESI)
2.5.5. Peak finding and quantitation of lipidomic data Automated peak finding, correction for 13C isotope effects, and quantitation of lipid molecular species were performed using Lipid Mass Spectrum Analysis (LIMSA) software version 1.0 [45]. All ion abundances for lipids analyzed on the Orbitrap Velos were quantitated against a single internal standard (dimyristoyl PC) [46], while phospholipid data obtained on the TSQ Vantage was quantitated against an internal standard of the appropriate lipid class. As no attempts were made to correct for differences in ionization efficiency among individual molecular species of various lipid classes, lipid molecular species are presented only as a fraction of the total normalized abundance of each lipid class. 2.6. Statistical analysis Islet studies used 4–6 animals per group. Statistical significance was determined using one-way ANOVA or two-way ANOVA followed by Tukey’s multiple comparison test using GraphPad Prism. p-Values
