Structure-activity relationships of glucose-dependent insulinotropic polypeptide (GIP)

Six GIP(1-NH2) analogs were synthesized with modifications (de-protonation, N-methylation, reversed chirality, and substitution) at positions 1, 3, and 4 of the N-terminus, and additionally, a cyclized GIP derivative was synthesized. The relationship between altered structure to biological activity was assessed by measuring receptor binding affinity and ability to stimulate adenylyl cyclase in CHO-K1 cells transfected with the wild-type GIP receptor (wtGIPR). These structure-activity relationship studies demonstrate the importance of the GIP N-terminus and highlight structural constraints that can be introduced in GIP analogs. These analogs may be useful starting points for design of peptides with enhanced in vivo bioactivity.     

Evolution of the vertebrate glucose-dependent insulinotropic polypeptide (GIP) gene

The glucose-dependent insulinotropic polypeptide (GIP) gene is believed to have originated from a gene duplication event very early in vertebrate evolution that also produced the proglucagon gene, yet so far GIP has only been described within mammals. Here we report the identification of GIP genes in chicken, frogs, and zebrafish. The chicken and frog genes are organized in a similar fashion to mammalian GIP genes and contain 6 exons and 5 introns in homologous locations. These genes can also potentially be proteolytically processed in identical patterns as observed in the mammalian sequences that would yield a GIP hormone that is only one amino shorter than the mammalian sequences due to the removal of an extra basic residue by carboxypeptidase E. The zebrafish GIP gene and precursor protein is shorter than other vertebrate GIP genes and is missing exon 5. The predicted zebrafish GIP hormone is also shorter, being only 31 amino acids in length. The zebrafish GIP hormone is similar in length to the proglucagon-derived peptide hormones, peptides encoded from the gene most closely related to GIP. We suggest that the structure of zebrafish GIP is more similar to the ancestral gene, and that tetrapod GIP has been extended. The mammalian GIP hormone has also undergone a period of rapid sequence evolution early in mammalian evolution. The discovery of a conserved GIP in diverse vertebrate suggests that it has an essential role in physiology in diverse vertebrates, although it may have only recently evolved a role as an incretin hormone.     

Insulin modulates glucose-dependent insulinotropic polypeptide (GIP) secretion from enteroendocrine K cells in rats

Effects of insulin excess and deficiency on glucose-dependent insulinotropic polypeptide (GIP) was examined in rats following insulinoma transplantation or streptozotocin (STZ) administration. Over 14 days, food intake was increased (p < 0.001) in both groups of rats, with decreased body weight (p < 0.01) in STZ rats. Non-fasting plasma glucose levels were decreased (p < 0.01) and plasma insulin levels increased (p < 0.001) in insulinoma-bearing rats, whereas STZ treatment elevated glucose (p < 0.001) and decreased insulin (p < 0.01). Circulating GIP concentrations were elevated (p < 0.01) in both animal models. At 14 days, oral glucose resulted in a decreased glycaemic excursion (p < 0.05) with concomitant elevations in insulin release (p < 0.001) in insulinoma-bearing rats, whereas STZ-treated rats displayed similar glucose-lowering effects but reduced insulin levels (p < 0.01). GIP concentrations were augmented in STZ rats (p < 0.05) following oral glucose. Plasma glucose and insulin concentrations were not affected by oral fat, but fat-induced GIP secretion was particularly (p < 0.05) increased in insulinoma-bearing rats. Exogenous GIP enhanced (p < 0.05) glucose-lowering in all groups of rats accompanied by insulin releasing (p < 0.001) effects in insulinoma-bearing and control rats. Both rat models exhibited increased (p < 0.001) intestinal weight but decreased intestinal GIP concentrations. These data suggest that circulating insulin has direct and indirect effects on the synthesis and secretion of GIP.     

Alterations of glucose-dependent insulinotropic polypeptide (GIP) during cold acclimation

Cold acclimation is initially associated with shivering thermogenesis in skeletal muscle followed by adaptive non-shivering thermogenesis, particularly in brown adipose tissue (BAT). In response, hyperphagia occurs to meet increased metabolic demand and thermoregulation. The present study investigates the effects of cold (4 ± 1 °C) acclimation and hyperphagia on circulating and intestinal levels of gastric inhibitory polypeptide (GIP) in rats. Pair fed animals were used as additional controls in some experiments. Cold acclimation for 42 days significantly (p<0.01) increased daily food intake. There was no corresponding change in body weight. However, body weights of pair fed cold exposed rats were significantly (p<0.01) reduced compared to controls and ad libitum fed cold exposed rats. By day 42, non-fasting plasma glucose was increased (p<0.05) by chronic cold exposure regardless of food intake. Corresponding plasma insulin concentrations were significantly (p<0.01) lower in pair fed cold exposed rats. Circulating GIP levels were elevated (p<0.05) in ad libitum fed cold acclimated rats on days 18 and 24, but returned to normal levels by the end of the study. The glycaemic response to oral glucose was improved (p<0.01) in all cold exposed rats, with significantly (p<0.05) elevated GIP responses in ad libitum fed rats and significantly (p<0.05) reduced insulin responses in pair fed rats. In keeping with this, insulin sensitivity was enhanced (p<0.05) in cold exposed rats compared to controls. By the end of the study, cold acclimated rats had significantly (p<0.01) increased BAT mass and intestinal concentrations of GIP and GLP-1 compared to controls, independent of food intake. These data indicate that changes in the secretion and actions of GIP may be involved in the metabolic adaptations to cold acclimation in rats.     

Influence of gastric inhibitory polypeptide (GIP) and glucose on the regulation of glucagon secretion by pancreatic alpha cells

The effects of glucose and GIP on glucagon secretion were studied in perifused microdissected murine pancreatic islets. Glucagon levels were determined in effluent samples collected at 1-min intervals by radioimmunoassay using the glucagon-specific antibody, 30 K. There was no significant difference in the total amount (7740 +/- 212 pg vs 8630 +/- 36 pg, n = 10) of glucagon secreted over a 20 min period when the glucose concentration was alternately shifted between 5.5 mM and 11.1 mM, respectively. However, 22.2 mM glucose profoundly suppressed glucagon secretion. The suppressive effect of high glucose on glucagon release was partially, yet significantly, reversed by the presence of GIP, as glucagon secretion increased from a non-detectable level at 22.2 mM glucose alone to 10,175 +/- 145 pg, n = 10 (P less than 0.01). The glucagonotropic effect of GIP was dose-dependent in the range of 2 x 10(-9) - 2 x 10(-7) M, at 11.1 mM glucose. Our data show that GIP is able to substantially reverse the suppressive effect of a high glucose load on glucagon secretion.     

Glucose dependent insulinotropic polypeptide (GIP) infused intravenously is insulinotropic in the fasting state in type 2 (non-insulin dependent) diabetes mellitus

The effects of an intravenous infusion of porcine GIP on beta-cell secretion in patients with untreated type 2 diabetes mellitus have been studied. The subjects were studied on two separate days. After a 10 h overnight fast and a further 120 min basal period they were given an intravenous infusion of porcine GIP (2 pmol.kg-1.min-1) or control solution in random order from 120-140 min. Frequent plasma glucose, insulin, C-peptide and GIP measurements were made throughout and the study was continued until 200 min. Plasma glucose levels were similar throughout both tests. During the GIP infusion there was an early significant rise in insulin concentration from 0.058 +/- 0.006 nmol/l to 0.106 +/- 0.007 nmol/l (P less than 0.01) within 6 min of commencing the GIP infusion and insulin levels reached a peak of 0.131 +/- 0.011 nmol/l at 10 min (P less than 0.01). Insulin levels remained significantly elevated during the rest of the GIP infusion (P less than 0.01-0.001) and returned to basal values 20 min post infusion. No change in basal insulin values was seen during the control infusion. C-peptide levels were similarly raised during the GIP infusion and the increase was significant just 4 min after commencing the GIP infusion (P less than 0.05). GIP levels increased from 16 +/- 3 pmol/l prior to the infusion to a peak of 286 +/- 24 pmol/l 20 min later. At 4 min when a significant beta-cell response was observed GIP levels were well within the physiological range.(ABSTRACT TRUNCATED AT 250 WORDS)     

Interaction of gastric inhibitory polypeptide (GIP) with the insulin-secreting pancreatic beta cell line, In lll: characteristics of GIP binding sites

The interaction of GIP with its receptors in the hamster pancreatic insulin-secreting beta cell line, In lll, has been analyzed. 125I-labelled GIP used as tracer showed the same affinity as native GIP for the GIP binding sites. Binding of the tracer was time, temperature and cell concentration dependent. It was saturable, reversible and highly specific. Under equilibrium conditions, i.e. 2 hours at 13 degrees C, 20% and 25% of the tracer and of GIP binding sites were inactivated, respectively. Native GIP inhibited binding of 125I-labelled GIP in a dose-dependent manner, saturation of the GIP binding sites being obtained at 3. 10(-7) M peptide. Two types of GIP binding sites were found by Scatchard analysis, a small population with a high affinity for GIP (KD = 7 nM) and a large population with a low affinity (KD = 800 nM). The biphasic dissociation process confirmed the GIP binding sites heterogeneity. Apart from GIP, no peptide tested influenced the binding of the 125I-labelled GIP. The present data represents the first analysis of functionally relevant GIP binding sites in a insulin-secreting cell.     

Effects of gastric inhibitory polypeptide (GIP) and related analogues on glucagon release at normo- and hyperglycaemia in Wistar rats and isolated islets

Glucose-dependent insulinotropic polypeptide (GIP) is an incretin hormone secreted by endocrine K-cells in response to nutrient absorption. This study has utilised numerous well-characterised dipeptidyl peptidase IV-resistant GIP analogues to evaluate the glucagonotropic actions of GIP in Wistar rats and isolated rat islets. Intraperitoneal administration of GIP analogues (25 nmol/kg body weight) in combination with glucose had no effect on circulating glucagon concentrations compared to controls in Wistar rats. However, plasma glucose concentrations were significantly (p<0.05 to p<0.001) lowered by the GIP-receptor agonists, N-AcGIP, GIP(Lys37)PAL and N-AcGIP(Lys37)PAL. The GIP antagonist, (Pro3)GIP, caused a significant (p<0.05) reduction in glucagon levels following concurrent administration with saline in Wistar rats. In isolated rat islets native GIP induced a significant (p<0.01) enhancement of glucagon release at basal glucose concentrations, which was completely annulled by (Pro3)GIP. Furthermore, glucagon release in the presence of GLP-1, GIP(Lys37)PAL, N-AcGIP(Lys37)PAL and (Pro3)GIP was significantly (p<0.05 to p<0.001) decreased compared to native GIP in isolated rat islets. These data indicate a modest effect of GIP on glucagon secretion from isolated rat islets, which was not observed in vivo. However, the GIP agonists N-AcGIP, GIP(Lys37)PAL and N-AcGIP(Lys37)PAL had no effect on glucagon release demonstrating an improved therapeutic potential for the treatment of type 2 diabetes.     

Effect of exogenous or endogenous gastric inhibitory polypeptide (GIP) on plasma triglyceride responses in rats.

We examined the effects of exogenous and endogenous GIP on plasma triglyceride levels in rats, pretreated with a fat-enriched diet, during intraduodenal infusion of a lipid test meal (Lipomul, 8 ml/h). Following the fat load the plasma triglyceride levels increased nearly linearly from a fasting value of 0.621 +/- 0.031 mmol/l to 3.32 +/- 0.403 mmol/l at 150 min. Simultaneously, the plasma GIP levels rose from 47.1 +/- 5.1 at fasting to a peak value of 268.4 +/- 32.2 pmol/l at 120 min. When porcine GIP was infused intravenously during the fat load, the plasma triglyceride increments were significantly smaller (control 1.64 +/- 0.264 mmol/l versus 0.949 +/- 0.114 mmol/l during GIP infusion at 60 min; p less than 0.002). GIP infusion in the absence of the fat load did not change fasting triglyceride levels. The effect of endogenous GIP was investigated by neutralization of GIP by injection of GIP antiserum (0.3 ml). Rats pretreated with the antiserum exhibited a significantly greater triglyceride increment late in the time course of the fat load. These data demonstrate that exogenous and endogenous GIP are able to lower the plasma triglyceride response to a fat load. Both, inhibition of fat absorption or stimulation of triglyceride uptake by peripheral tissues may be responsible for the GIP effects. The gut peptide GIP seems to represent an important hormonal regulator of postprandial triglyceride response.     

Day profiles of glucose dependent insulinotropic polypeptide (GIP) in normal subjects

IRGIP release results from nutrient absorption, the major stimulants being fat and carbohydrate. Little is known, however, about its diurnal profile in response to serial meals. The purpose of this study was to determine the plasma IRGIP day profile in normal subjects following four isocaloric meals administered serially throughout the day. Five healthy normal weight (67-77 kg) male volunteers aged 38-49 years were investigated following a 10 hour overnight fast on two days. On each day, isocaloric non-identical test meals were consumed at 09.00, 13.00, 16.00 and 19.00 hours. Plasma glucose, insulin (IRI), IRC-peptide and IRGIP levels were measured half-hourly from 08.30 to 21.00 hours. Peak IRGIP levels occurred within 2 hours of the commencement of each meal and then decreased gradually but never returned to fasting levels. Compared with the first meal, the subsequent pre-prandial IRGIP levels were significantly higher (P less than 0.05) which was consistent for the two study days. The highest mean IRGIP levels occurred after breakfast and tea which were the meals containing the greater proportion of fat. Plasma IRGIP levels correlated (P less than 0.001) with the concentrations of both insulin and IRC-peptide. In conclusion, plasma IRGIP levels increased following ingestion of serial mixed meals but the levels did not return to fasting concentration throughout the day. There was a gradual upward trend of each subsequent pre-prandial IRGIP value. The physiological importance of this observation requires further exploration.     

Tyr1 and Ile7 of glucose-dependent insulinotropic polypeptide (GIP) confer differential ligand selectivity toward GIP and glucagon-like peptide-1 receptors

Glucagon like peptide-1 (GLP-1) and glucose-dependent insulinotropic polypeptide (GIP) are incretin hormones released in response to food intake and potentiate insulin secretion from pancreatic β cells through their distinct yet related G protein-coupled receptors, GLP1R and GIPR. While GLP-1 and GIP exhibit similarity in their N-terminal sequence and overall α-helical structure, GLP-1 does not bind to GIPR and vice versa. To determine which amino acid residues of these peptide ligands are responsible for specific interaction with their respective receptors, we generated mutant GIP in which several GLP-1-specific amino acid residues were substituted for the original amino acids. The potency of the mutant ligands was examined using HEK293 cells transfected with GLP1R or GIPR expression plasmids together with a cAMP-responsive element-driven luciferase (CRE-luc) reporter plasmid. A mutated GIP peptide in which Tyr¹, Ile⁷, Asp¹⁵, and His¹⁸ were replaced by His, Thr, Glu, and Ala, respectively, was able to activate both GLP1R and GIPR with moderate potency. Replacing the original Tyr¹ and/or Ile⁷ in the N-terminal moiety of this mutant peptide allowed full activation of GIPR but not of GLP1R. However, reintroducing Asp¹⁵ and/or His¹⁸ in the central α-helical region did not significantly alter the ligand potency. These results suggest that Tyr/His¹ and Ile/Thr⁷ of GIP/GLP-1 peptides confer differential ligand selectivity toward GIPR and GLP1R.     

Effect of endogenous gastric inhibitory polypeptide (GIP) on the removal of triacylglycerol in dogs

In order to clarify the effect of endogenous gastric inhibitory polypeptide (GIP) upon lipid metabolism, the removal of intravenously administered triacylglycerol was investigated following an oral glucose or galactose load in dogs. After an overnight fast, the triacylglycerol emulsion was infused at a constant rate of 1 ml/min for 90 min, and glucose, galactose or tap water was orally administered at 30 min. Blood glucose increased after the glucose load but it did not change following the galactose load or water ingestion. Plasma insulin increased after the glucose load but did not change after galactose or tap water ingestion. Plasma glucagon did not show any discernible change in the three experimental groups. Plasma GIP increased following the glucose or galactose load to 4360 or 1653 pg/ml, respectively. Plasma triacylglycerol increased to the same levels at 30 min in the three experimental groups. The peak levels of plasma triacylglycerol and integrated plasma triacylglycerol for 150 min did not differ in the three groups. Moreover, there was no difference in the removal rate of plasma triacylglycerol following the withdrawal of the fat emulsion. It is concluded from the present study that endogenously released GIP does not elicit any effect upon triacylglycerol removal.     

Response of gastric inhibitory polypeptide (GIP) to oral administration of nutrients in normal and pancreatectomized dogs

In order to clarify the response of plasma gastric inhibitory polypeptide (GIP) to various nutrients and to investigate the relationship between the pancreas and GIP secretion, an experimental study was performed using normal and pancreatectomized dogs. Oral administration of glucose (2 g/kg) or butter (2 g/kg) resulted in an increase of plasma GIP in five normal dogs. In contrast, oral administration of arginine (1 g/kg) did not produce any discernible changes in plasma GIP in normal dogs. In a group of nine pancreatectomized dogs, the fasting level of plasma GIP did not differ from that of the control group. Furthermore, glucose ingestion in the pancreatectomized group resulted in the same pattern and the same degree of change in plasma GIP as it did in the normal controls. In contrast, plasma GIP did not change at all following fat loading in the pancreatectomized group. However, butter with pancreatic enzymes elicited a significant rise of plasma GIP in the pancreatectomized dogs. The present study indicates that plasma GIP increases following oral administration of glucose or fat but not arginine. Furthermore, it is demonstrated that GIP secretion following fat ingestion occurs only after fat digestion by pancreatic enzymes. In addition, the findings observed in the present study do not support the existence of feedback effect of insulin on GIP secretion.     

Glucose-dependent insulinotropic polypeptide (GIP) and the enteroinsular axis in equines (Equus caballus).

To investigate the enteroinsular axis (EIA) in equines oral (oGTT) and intravenous (i.v.GTT) glucose tolerance tests (5.6 and 1 mmol glucose/kg BW, respectively) were performed with healthy, normal weight large horses and Shetland ponies. Plasma was analysed for concentrations of glucose, glucose-dependent insulinotropic polypeptide (GIP) and insulin. In all equines plasma GIP concentrations only increased significantly when glucose was administered orally. The insulin glucose ratio (IGR) was significantly higher during the oGTT than during the i.v.GTT in both races. Basal plasma glucose levels were significantly higher in large horses than in ponies in both experiments. During the oGTT maximum glucose values were significantly higher in ponies. Ponies tended to a higher insulin secretion but the IGRs were identical in both races after oral and intravenous glucose administration. One clinically inconspicuous pony showed hyperinsulinaemia and, in case of the oGTT, insulin resistance, glucose intolerance, and GIP hypersecretion. The results of this study indicate the existence of an EIA in equines due to the higher IGRs during the oGTT. Furthermore, the similarity of plasma GIP levels and IGRs in ponies and large horses suggest a comparable activity of the EIA in both races. Regarding the elevated plasma GIP concentrations of the insulin resistant pony the EIA appears to participate in equine hyperinsulinaemia.     

The effect of somatostatin on release and insulinotropic action of gastric inhibitory polypeptide.

Studies were carried out in conscious dogs in which the effect of intravenous somatostatin on immunoreactive gastric inhibitory polypeptide (IR-GIP) release was investigated. In addition, the inhibitory action of somatostatin on the insulin response to pure porcine GIP was assessed. Intravenous administration of somatostatin resulted in a delayed IR-GIP and immunoreactive insulin (IRI) response to oral glucose. Somatostatin also delayed the IR-GIP response to the ingestion of fat. In both types of experiments, initial depression of IRI levels was followed by a sharp rise in IRI release. Intravenous infusion of somatostatin produced 80% inhibition of the IRI response to pure porcine GIP. It was concluded that somatostatin inhibits the physiological release of IR-GIP and the insulinotropic action of exogenous porcine GIP.     

The reduction in hepatic insulin clearance after oral glucose is not mediated by gastric inhibitory polypeptide (GIP)

Since the C-peptide/insulin ratio is reduced after oral glucose ingestion, the incretin hormone gastric inhibitory polypeptide (GIP) has been assumed to decrease hepatic insulin extraction. It was the aim of the present study to evaluate the effects of GIP on insulin extraction. Seventy-eight healthy subjects (27 male, 51 female, 43+/-11 years) were subjected to (a). an oral glucose tolerance test and (b). an intravenous injection of 20 pmol GIP/kg body weight, with capillary and venous blood samples collected over 30 min for insulin, C-peptide and GIP (specific immunoassays). Following GIP administration, plasma concentrations of total and intact GIP reached to peak levels of 80+/-7 and 54+/-5 pmol/l, respectively (p<0.0001). The rise in insulin after oral glucose and after intravenous GIP administration significantly exceeded the rise in C-peptide (p<0.0001). Estimating insulin extraction from the total integrated insulin and C-peptide concentrations (AUCs), only the oral glucose load (p<0.0001), but not the intravenous GIP administration (p=0.18) significantly reduced insulin clearance. Therefore, insulin clearance is reduced after an oral glucose load. This effect does not appear to be mediated by GIP.     

The effects of gastric inhibitory polypeptide (GIP) on the release of anterior pituitary hormones

Several members of the secretin family of hormones have been demonstrated to alter anterior pituitary hormone secretion. Here we report the action of gastric inhibitory polypeptide (GIP) on gonadotropin and somatotropin release. Intraventricular injection of 1 microgram (0.2 nmole) GIP (2.5 microliters) produced a significant decrease in plasma FSH at 30 (p less than 0.02) and 60 min after its injection (p less than 0.01). The FSH-lowering effect of a higher dose of 5 micrograms (1 nmole) of GIP was already developed at 15 min (p less than 0.01) and was prolonged until the end of the experiment (60 min, p less than 0.05). No change in plasma LH was detected at any time during the experimental period. If 5 micrograms of estradiol-benzoate were given SC 48 hr prior to experiment, the initial values of FSH and LH were markedly decreased. In these animals GIP failed to influence plasma FSH and LH. When dispersed anterior pituitary cells from OVX rats were cultured overnight and incubated in vitro with GIP, the peptide was found to induce both FSH and LH release. Highly significant release occurred with the lowest dose tested of 10(-7) M and there was a dose-response effect for both hormones. The slope of the dose-response curve was similar for both FSH and LH release. GIP was less potent than LHRH which produced a greater stimulation of both FSH and LH release at a dose of 10(-9) M than did 10(-7) M GIP. The two peptides had an additive effect on the release of both FSH and LH.(ABSTRACT TRUNCATED AT 250 WORDS)     

Electronimmunocytochemical evidence for the K cell localization of gastric inhibitory polypeptide (GIP) im man

Summary Application of the semithin-thin section technique indicates that the previously proposed identification of the ultrastructurally-defined K cell with the immunocytochemically-defined GIP cell is essentially correct.The K cell is established as a distinct entity and the way is open for an explanation of its role in the physiology and pathology of the gastroenteropancreatic system.     

Electronimmunocytochemical evidence for the K cell localization of gastric inhibitory polypeptide (GIP) in man.

Application of the semithin-thin section technique indicates that the previously proposed identification of the ultrastructurally-defined K cell with the immunocytochemically-defined GIP cell is essentially correct. The K cell is established as a distinct entity and the way is open for an explanation of its role in the physiology and pathology of the gastroenteropancreatic system.     

Sitagliptin decreases ventricular arrhythmias by attenuated glucose-dependent insulinotropic polypeptide (GIP)-dependent resistin signalling in infarcted rats

Myocardial infarction (MI) was associated with insulin resistance, in which resistin acts as a critical mediator. We aimed to determine whether sitagliptin, a dipeptidyl peptidase-4 (DPP-4) inhibitor, can attenuate arrhythmias by regulating resistin-dependent nerve growth factor (NGF) expression in postinfarcted rats. Normoglycaemic male Wistar rats after ligating coronary artery were randomized to either vehicle or sitagliptin for 4 weeks starting 24 h after operation. Post-infarction was associated with increased myocardial noradrenaline [norepinephrine (NE)] levels and sympathetic hyperinnervation. Compared with vehicle, sympathetic innervation was blunted after administering sitagliptin, as assessed by immunofluorescent analysis of tyrosine hydroxylase, growth-associated factor 43 and neurofilament and western blotting and real-time quantitative RT-PCR of NGF. Arrhythmic scores in the sitagliptin-treated infarcted rats were significantly lower than those in vehicle. Furthermore, sitagliptin was associated with reduced resistin expression and increased Akt activity. Ex vivo studies showed that glucose-dependent insulinotropic polypeptide (GIP) infusion, but not glucagon-like peptide-1 (GLP-1), produced similar reduction in resistin levels to sitagliptin in postinfarcted rats. Furthermore, the attenuated effects of sitagliptin on NGF levels can be reversed by wortmannin (a phosphatidylinositol 3-kinase antagonist) and exogenous resistin infusion. Sitagliptin protects ventricular arrhythmias by attenuating sympathetic innervation in the non-diabetic infarcted rats. Sitagliptin attenuated resistin expression via the GIP-dependent pathway, which inhibited sympathetic innervation through a signalling pathway involving phosphatidylinositol 3-kinase (PI3K) and Akt protein.