The Effect Of Liraglutide On The Metabolism Of Glucose And Lipid In Prediabetic OLETF Rats And Related Mechanisms

1. Background and ObjectiveNow, with the development of economy and the acceleration of industrialization process in the world, non-communicable diseases increasingly threaten human health and the prevalence rate of diabetes increases rapidly. Diabetes is not only a "rich diasease" in developed countries, but also becomes a major disaster diasease in developing countries including China. In2008, the epidemiology of diabetes showed the age-standardized prevalences of total diabetes and prediabetes were9.7%and15.5%in chinese adults aged20or olders, accounting for148million with prediabetes. Prediabetes is at hight risk of diabetes, approximate40%of worldwide subjects with prediabetes progress to diabetes over5～10years, in some newly identified prediabetes patients even less than3years. Prediabetes is not only easy to develop diabetes but also have a higher risk of cardiovascular disease. The relative risk of cardiovascular events in prediabetes is three times more than the normar person. Therefore, an early intervention during prediabetes is desirable in order to delay or prevent the onset of diabetes and reduce the mortality due to related cardiovascular disease.Type2diabetes mellitus(T2DM) primary prevention goal is to prevent the occurrence of diabetes. The occurrence of T2DM mainly depends on some risk factors that can not be changed such as age, family history and genetic predisposition, ect, and other risk factors that can be changed such as diet, environment, impaired glucose tolerance and/or impaired fasting glucose, and so on. For the latter, in the last two decades, some studies have been conducted regarding the effect of lifestyle modification and pharmacotherapy on prediabetes. Several large clinical trials have shown the beneficial effect of lifestyle modification on diabetes prevention, cardiovascular disease events, and mortality. However, the effective application of lifestyle modification is a challenge for many patients. Additional pharmacotherapy may be necessary, especially for individuals at high risk. The treatment with metformin, acarbose, thiazolidinediones(TZDs), or inhibitors of the renin-angiotensin system (RAS) has been evaluated in subjects with prediabetes. But, drugs that had been discussed to prevent diabetes through the different mechanisms which might partly improve the pathophysiology in a certain have their own shortcomings, such as gastrointestinal reaction, osteoporosis, weight gain and edema, and aggravate heart function failure, and so on. More importanly, these oral drugs work mainly through improving the body’s pathophysiological environment but not directly protecting the key pathological factors that islet function has gradually failure in diabetes. Therefore, the drug that can directly protect beta cell function, increase the beta cell mass, inhibit apoptosis and improve insulin secretion in prediabetes might become a ideal drug for diabetes preventionRecent studies show the intestinal hormones--glucagon-like peptide-1(GLP-1) can protect the beta cells through different mechanisms and have the broad application prospect for diabetes treatment. Liraglutide is a GLP-1analogues.It can stimulate insulin secretion, inhibit glucagon release, protect beta cell, improve the early-phase insulin secretion and insulin sensitivity, and potentially protect the heart and cardiovascular. Moreover, unlike other glucose-lowering drugs, liraglutide might directly promote beta cell proliferation, induce cell regeneration and inhibit apoptosis, leading to expansion of beta cell mass. It can improve postprandial blood glucose by stimulating glucose dependent insulin secretion, thereby reducing the incidence of hypoglycaemia following antidiabetes treatment. Researches show GLP-1can stimulate beta cell proliferation and insulin secretion, and pancreatic duodenal homeobox-1(Pdx-1) is a key factor for pancreatic growth and insulin genetic transcription. The expression of Pdx-1mRNA and protein increased several times in pancreas and islet after GLP-1treatment. Gene MafA is an important regular fator for insulin secretion and pancreatic islet structure maintenance. So, whether liraglutide can regulate this two genes in living animal is one of our concerned problems.Based on the hypothesis that liraglutide treatment might be effective to prevent diabetes, we chose OLETF rats, a spontaneous animal model of T2DM, to process this experimental research. We treated prediabetic rats with liraglutide to investigate their food intake, weight, blood glucose, serum insulin, blood lipid and inflammatory markers, caculate the number of diabetes rats in each group, evaluate early-phase insulin secretion and insulin sensitivity index, and assess the homeostasis model of beta cell function and insulin resistance index. At the end of the experiment, we observed the morphology of pancreatic tissue and detect the expression of islet Pdx-1、 MafA and apoptosis-related genes with molecular biology research technique to seek the related mechanisms how liraglutide to protect islet beta cell function in the piediabetes.2. Methods2.1Animals and Composition of Experimental GroupsFour-week-old male OLETF rats and age-matched non-diabetic control male LETO rats were generously provided by the Tokushima Research Institute, Otsuka Pharmaceutical Co., Ltd.(Tokushima, Japan). All rats were kept4-6rats/cage in polycarbonate cages with free access to standard rodent chow and tap water in a controlled SPF environment with temperature (21±2℃), humidity(55±10)%and a12:12h light-dark cycle(lights on7:00a.m. to7:00p.m.). We used12weeks old male OLETF rats as a prediabetic model and treated them with three doses of liraglutide (50,100, or200μg/kg, Novo Nordisk Pharmaceuticals Co., Ltd) or0.9%saline intraperitoneally, twice daily for12weeks,8rats kept individually in each group. In the meantime, eight LETO rats served as normal controls with saline treatment. During the experiment, three rats died from gavage miss accidentally (including two100μg/kg group and one200μg/kg). After12week-intervention, all rats were deprived of food for15h before being killed by intraperitoneally administration of pentobarbital (50mg/kg), the abdomen was opened to remove the whole pancreas. A splenic portion of the pancreas was performed by islet isolation and frozen at-80℃for RT-PCR and Westernblot assays. A duodenal portion of the pancreas was used for histological examination.2.2Oral Glucose Tolerance Test (OGTT)OGTT was performed in all rats at0,2,4,6,8,10, and12weeks of experiment. Briefly, a50%glucose solution (2g/kg body weight) was orally administrated with an18-gauge gavage needle after overnight fast for15h, and then blood samples were collected from the tail vein using a OneTouch UltraVue glucose meter (LifeScan, Inc., Milpitas, CA) in a conscious state at0,30,60, and120minutes following the glucose challenge to determine plasma glucose levels.2.3Diagnosis of Clinical Diabetes and PrediabetesThe diagnosis of diabetes and prediabetes in the present study follows experimental research on OLETF rats by the Tokushima Research Institute. Plasma glucose was measured at0,30,60,90and120min. Rats were classified based on OGTT data using the following criteria:(1) a peak plasma glucose level of more than16.7mmol/L and a level of plasma glucose at120min of more than11.1mmol/L were diagnosed as diabetes mellitus(DM), but in that either of the above two conditions was diagnosed as impaired glucose tolerance(IGT).2.4Generally Observation and Biochemical DetectionBody weight and food intake were monitored at the same time once a week. Blood samples were taken from the tail vein for overnight fasting blood glucose assessment at0,2,4,6,8,10, and12weeks after drug intervention using a glucose meter. Plasma glucose area under a curve (G-AUC) during OGTT in each group was derived according to the trapezoidal rule and the differences between groups were compared. The blood samples at0and12weeks were also used for the measurement of lipid profile, including total cholesterol (TC), triglycerides (TG), low-density lipoprotein cholesterol (LDL-c), and HDL-c with an automatic biochemical analyzer (Aeroset, American) and inflammatory markers, including serum high-sensitivity C-reactive protein (Hs-CRP), tumor necrosis factor-a (TNF-a), interleukin-6(IL-6), fibrinogen (Fib), and plasminogen activator inhibitor-1(PAI-1) with a mouse enzyme-linked immunosorbent assay (ELISA)(Uscnlife, Wuhan EIAab Science Co., Ltd, Wuhan, China). Blood samples at0and12weeks were also used for the measurement of fasting serum insulin (FINS) level, and serum insulin level at OGTT30min (INS30min) by ELISA.2.5Assessment of Insulin Sensitivity and Islet β-cell FunctionThe indices for insulin sensitivity and beta cell function were calculated subsequently, including insulin sensitivity index (ISI)=1/FPGxFINS, the homeostasis model assessment of insulin resistance(HOMA-IR)=(FPGxFINS)/22.5, the homeostasis model assessment of B-cell function HOMA-p=20×FIns/(FBG-3.5), and the index of early-phase insulin secretion using the ratio of insulin incremental value to glucose incremental value at30min after the meal Ins30/ΔGlu30=(Ins30-Ins0)/(Glu30-Glu0).2.6Islet Isolation, Purification and CertificationRats were anesthetized by intraperitoneally administration of pentobarbital (50mg/kg). D-Hank’s liquid was adversely infused into common bile duct. Remove the whole pancreas quickly and move into the collagenase P liquid for digesting. After the digestion, pancreas was joined and mixed with25%Ficoll, and then joined with23%,20%and11%Ficoll-400solution in turn. Islets were hand-picked under the microscope after4℃centrifugation and stored at-80℃.2.7Quantitative Real-time PCRTotal RNA was extracted from rat pancreas tissue using Trizol reagent kit. First-strand cDNA synthesis was performed using MMLV reverse transcriptase with oligo dT primers. Real-time PCR products were detected using an ABI7500system.20μL of reaction volume was used according to the SYBR Premix ExTaqTM kit. Primers to detect Pdx-1, MafA, Bax, Bcl-2, Caspase3and IL-1β were obtained from Invitrogen. For each assay, a standard curve was obtained by analyzing a dilution series of pooled cDNA samples for the relevant gene. The program consisted of initial denaturation for5minutes at94℃followed by30cycles of94℃for39seconds,60℃for30seconds, and72℃for7minutes.2.8Westernblot Analysis Immunoblots were probed with rabbit anti-Pdx-1、 rabbit anti-MafA、 rabbit anti-caspase3and rabbit anti-IL-1β polyclonal antibodies. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis and Western blots were carried out according to the standard protocols. Signals were detected by a goat anti-rabbit IgG secondary antibody conjugated to horseradish peroxidase. Protein bands were imaged and quantified with the Universal hood Ⅱ gel imaging system.2.9Pancreas HistologyA duodenal portion of the pancreas was fixed overnight with4%paraformaldehyde and then embeded with paraffin. Paraffin sections (5μm) were cut for Hematoxylin and Eosin staining (HE) and immunostaining. Immunohistochemical was performed with three-step SABC methods.3. Statistical AnalysisSPSS16.0software was used for statistical analysis. Data throughout were stated as x±SEM unless otherwise specified. Repeated data was analyzed by two-factor (time and treatment) repeated-measures ANOVA followed by a Bonferroni post hoc test. General effects at the same time were tested using one way ANOVA followed by Bonferroni or Dunnett’s test for individual comparisons of means. Differences between before and after treatment were analysed by use of Paired-Samples T tests. A two-tailed Pearson test or Spearman test was performed for correlation analysis between the variables. P<0.05was considered statistically significant.4. Results4.1The effect of liraglutide on food intake and body weight in IGT-OLETF ratsFood intake in IGT-OLETF rats (37.01±1.17g/d) was significantly more than that of LETO rats (27.14±0.75g/d) before liraglutide treatment (P<0.0001).After12-week treatment, the amounts of food intake (35.12±1.21,34.78±1.67,35.16±1.00g/d) in the three liraglutide(50,100,200μg/kg) groups were similar and significantly more than that (25.01±0.5g/d)of LETO rats (P<0.0001). They were slightly reduced but not statistically significant from that (36.77±0.40g/d)of saline-treated OLETF rats(all P>0.05). Body weight in OLETF rats (416.74±7.21g) was significantly greater than that of LETO rats (331.91±7.80g) before treatment (P<0.0001),which was significantly reduced with liraglutide (50μg/kg group,380.16±4.09g,100μg/kg group,365.18±2.97g,200μg/kg group,361.28±7.80g) treatment within the first week compared with that (426.69±7.61g) of saline-treated OLETF rats (all P<0.05).This then gradually increased (50μg/kg,408.81±3.18g,100μg/k,395.55±5.15g,200μg/kg,395.27±9.58g) from the second week parallel to saline-treated animals and normal controls, which were significantly greater than LETO rats (all P<0.01)but still less than that (439.14±4.96g) of saline-treated OLETF rats (all P<0.05) despite similar levels of food intak.4.2The effect of liraglutide on glycemic metabolism and insulin sensitivity in IGT-OLETF ratsAt the end of the experimental period,7of8(87.5%) saline-treated OLETF rats progressed to diabetes characterized by significantly increased FPGand2h-PG levels, however IGT levels were reversed to normal in9of21(42.9%) liraglutide-treated OLETF rats whilst none of liraglutide-treated OLETF rats progressed to diabetes. Before treatment, FPG (6.68±0.07mmol/L),2h-PG (7.21±0.19mmol/L), and G-AUC (666.21±40.59Mm*min/L) levels in saline-treated OLETF rats were higher than in LETO rats (FPG,5.23±0.10mmol/L,2hPG,6.64±0.14mmol/L, G-AUC,306.00±27.37Mm*min/L), and then FPG,2h-PG, and G-AUC maintained stablely in LETO rats but all increased gradually in saline-treated OLETF rats. After two weeks of liraglutide (50,100,200μg/kg) treatment, FPG (5.75±0.28mmol/L,5.33±0.13mmol/L,5.43±0.23mmol/L),2h-PG (6.82±0.18mmol/L,6.73±0.28mmol/L,7.09±0.34mmol/L), and G-AUC (306.55±41.97Mm*min/L,292.92±42.09Mm*min/L,306.19±57.80Mm*min/L) significantly decreased (FPG,6.51±0.08mmol/L,2hPG,7.44±0.22mmol/L, G-AUC,573.14±49.59Mm*min/L, P<0.05compared with OLETF-saline). In addition, FPG was independent of food intake (2w, R=-0.307, P=0.460;12w, R=0.369, P=0.369) and bodyweight gain (2w, R=-0.245, P=0.559;12w, R=0.153,P=0.718) in saline-treated OLETF rats during the study, but2h-PG was positively correlated to bodyweight gain (R=0.784, P=0.021)after the12-week treatment.Before the study, compared with LETO rats (FINS,27.21±0.36μIU/mL, HOMR-IR,6.32±0.17, ISI,7.06±0.18, HOMA-p,323.46±20.63, ΔIns30/ΔGlu30,11.23±0.85), IGT-OLETF rats had increased FINS37.49±0.78μIU/mL and HOMA-IR10.83±0.31(all P<0.0001), and reduced ISI4.12±0.12(all P<0.05), HOMA-β251.92±9.82and ΔIns30/ΔGlu305.21±0.40levels were also significantly decreased (all P<0.05); INS30min in OGTT increased3-fold in LETO rats compared to2.5-fold increase in IGT-OLETF rats. FINS (51.03±1.12μIU/mL) and HOMR-IR (19.55±0.87) were furtherly increased (all P<0.001), while ISI (2.37±0.09), HOMAβ (205.75±4.20) and ΔIns30/ΔGlu30(3.85±0.19) were furtherly decreased (all P<0.05). After liraglutide (50,100,200μg/kg) treatment, FINS (39.86±0.94,38.19±1.00,37.03±1.21μIU/mL), HOM-IR (11.27±0.36,10.54±0.28,10.40±0.48), ISI (4.06±0.11,4.34±0.11,4.73±0.20), HOMA-β(290.99±12.79,288.62±11.23,294.41±14.24) and ΔIns30/ΔGlu30(9.00±0.35,8.90±0.37,9.55±0.25) were significantly ameliorated (compared with OLETF-saline, all P<0.05), especially ΔIns30/ΔGlu30was greatly reversed (compared with before treatment, P<0.05). INS30min in OGTT increased2-fold in saline-treated OLETF rats compared to3-fold increase in LETO rats and liraglutide-treated OLETF rats.4.3The effect of liraglutide on lipid profiles in IGT-OLETF ratsCompared with LETO rats, serum level of TC was significantly increased in OLETF rats at the age of12weeks (LETO rat,2.59±0.05vs OLETF rat,3.01±0.05mmol/L, P=0.003) and24weeks (LETO rat,2.02±0.07vs OLETF rat,2.56±0.08mmol/L, P=0.006). Serum level of TG was also significantly increased at the age of24weeks (LETO rat,0.36±0.04vs OLETF rat,1.09±0.06mmol/L, P<0.0001). After12weeks of liraglutide (50,100,200μg/kg) treatment, compared with that before treatment (TC,2.84±0.04,3.05±0.12,2.94±0.02mmol/L, TG,0.55±0.04,0.70±0.10,0.64±0.02mmol/L), serum level of TC (2.21±0.02,2.24.±0.05,2.13±0.02mmol/L) but not TG (0.71±0.04,0.40±0.13,0.51±0.03mmol/L) was significantly reduced (all P<0.01); and compared with OLETF-saline group, both TC (2.56±0.08mmol/L, P=0.006) and TG (1.09±0.06mmol/L, P<0.0001) were significantly decreased. There was no difference in terms of HDL and LDL between groups or before and after liraglutide treatment (all P>0.05).4.4The effect of liraglutide on inflammatory state in IGT-OLETF ratsThe serum levels of fibrinogen (74.04±1.66U/mL), TNF-α (48.90±2.62ng/mL), Hs-CRP (5.53±0.17ng/mL), IL-6(8.89±0.27pg/mL), and PAI-1(48.67±3.01ng/mL) were significantly higher in OLETF rats compared with LETO rats (Fib,55.38±1.92U/mL, TNF-α,31.70±3.00ng/mL, Hs-CRP,3.56±0.16ng/mL, IL-6,5.11±0.38pg/mL, PAI-1,26.45±1.54ng/mL) at the age of12weeks (all P<0.001). PAI-1and fibrinogen were further increased in vehicle-treated OLETF rats at the age of24weeks (PAI-1:before treatment,48.67±3.01vs after treatment,63.37±0.99, P=0.026; Fib:before treatment,74.04±1.66vs after treatment,94.20±1.32, P=0.003). After12weeks of treatment, IL-6was significantly reduced (50μg/kg group, before treatment,8.18±0.20vs after treatment,5.63±0.32pg/mL,100μg/kg group, before treatment,8.18±0.56vs after treatment,4.79±0.90pg/mL,200μg/kg group, before treatment,8.30±0.22vs after treatment,5.11±0.26g/mL, all P<0.05) by all doses of liraglutide (50,100,200μg/kg), while Hs-CRP was only reduced by liraglutide200μg/kg group (before treatment,5.39±0.07vs after treatment,3.68±0.12ng/mL, P=0.006) and PAI-1by liraglutide50μg/kg group (before treatment,50.42±2.04vs after treatment,27.74±3.19ng/mL, P=0.026). And compared with the OLETF-saline group, Fib, TNF-a, Hs-CRP, IL-6and PAI-1were all significantly reduced after treatment by all doses of liraglutide (all P<0.01), Table4-4.Moreover, a positive correlation was found between the bodyweight gain and serum level of Hs-CRP (R=0.714, P=0.047), and between the serum level of TG and PAI-1(R=0.929, P=0.001) in24weeks old vehicle-treated OLETF rats. Such relationships did not appear in the liraglutide-treated animals.4.5The effect of liraglutide on the expression of islet transcription factors and apoptotic factors in IGT-OLETF ratsAt the end of the experiment, i.e., the age of24weeks, RT-PCR analysis revealed that islet Pdx-1(2.19±0.44), MafA (0.62±0.01) and Bcl2(8.22±1.62) mRNA levels were significantly decreased in vehicle-treated OLETF rats compared with LETO (Pdx-1,7.06±1.07, MafA,3.18±0.42, Bcl2,29.92±1.72, Bax,5.49±0.53, Caspase3,84.48±3.39) rats (all P<0.05), whereas Bax (29.74±3.74) and Caspase3(406.63±62.88) mRNA levels were significantly increased (P<0.05).However, concomitant administration of liraglutide partly restored Pdx-1(3.60±0.88,5.87±1.03,3.77±1.05), MafA (0.76±0.04,1.54±0.06,3.20±0.13), and Bcl2(15.13±1.41,16.35±1.14,19.96±1.80) mRNA levels, and restrained the Bax (15.16±2.26,7.22±0.80,7.18±0.32) and Caspase3(205.58±5.40,163.90±32.81,153.41±7.91) mRNA levels in OLETF rats.The protein expressions of Pdx-1(0.21±0.03) and MafA (0.14±0.03) in islet of vehicle-treated OLETF rats were also significantly reduced compared to those (Pdx-1,0.62±0.09, MafA,0.60±0.05, Caspase3,0.31±0.09) of LETO rats (all P<0.05), whereas that of Caspase3(0.87±0.07) was significantly decreased (P<0.05).However, the former (Pdx-1,0.42±0.01,0.45±0.04,0.44±0.04, MafA,0.30±0.04,0.42±0.02,0.51±0.05) was up-regulated and the latter (Caspase3,0.42±0.02,0.46±0.07,0.36±0.04) down-regulated by liraglutide therapy.4.6The effect of liraglutide on the expression of inflammatory factor IL-1β in IGT-OLETF ratsAt the end of the experiment, RT-PCR and Westernblot analysis revealed that the mRNA (15.39±0.60vs31.78±1.92) and protein (0.24±0.05vs0.78±0.02) levels of islet IL-1β were significantly increased in vehicle-treated OLETF rats compared with LETO rats (all P<0.0001), which (IL-1βmRNA,20.56±2.27,16.38±1.06,15.60±1.17, IL-1βprotein,0.39±0.012,0.27±0.02,0.26±0.01) were down-regulatedby liraglutide (50,100,200μg/kg) therapy in a dose-independent manner.4.7The effect of liraglutide on islet cell morphology in IGT-OLETF ratsUnder the light microscopy, the islet margins of LETO rats was quite regular and homogeneous. Whereas OLETF rats had characteristically islet cell hyperplasia hypertrophy and inflammatory cell infiltration around, and interstitial fiber cell hyperplasia. Similar to LETO rats, the liraglutide-treated OLETF rats had islet margins characteristically regular and homogeneous without inflammatory cell infiltration. Fig4-7. 4.8The expression of insulin and IL-1βin islet under light microscopyUnder the light microscopy, both insulin and IL-1β were located to cytoplasm. The expression of insulin in vehicle-treated OLETF rats were also significantly reduced compared to those of LETO rats (P<0.05), whereas that of IL-1β was significantly decreased (P<0.05). However, the former was up-regulated and the latter down-regulated by liraglutide therapy. Fig4-7.5Conclusions(1) OLETF rats had slightly hyperglycemia, obesity, hyperinsulinemia, insulin resistance, impaired beta cell function, hyperlipidaemia, and high inflammatory state at the age of12weeks. Therefore,12-week OLETF rats appears to be a valid model for studying prediabetes.(2) Liraglutide only had an acute effect on food intake but its beneficial effect on weight loss was sustained and independent of food intake.(3) Liraglutide treatment suppressed IGT, insulin resistance and islet beta cell function, improved serum hyperlipidaemia and inflammatory state and also decreased islet protein and mRNA expressions of IL-1β.These indicated that liraglutide might improve insulin sensitivity and islet cell function throught lowering the level of triglyceride and inflammation factors and inhibiting the islet inflammatory cytokine IL-1β expression in prediabetes, and then prevent diabetes progression.(4) Liraglutide increased protein and mRNA expressions of pancreatic transcription factors Pdx-1and Mafa and antiapoptotic factor Bcl-2mRNA and reduced proapoptotic factor Bax mRNA level. These indicated liraglutide might maintain islet (3-cell mass and functions partly through its effects on transcription factors Pdx-1and MafA, apoptotic factors Bcl-2and Bax and apoptosis protease Caspase3in prediabetes.