The authors have declared that no competing interests exist.
Conceived and designed the experiments: MPM WWL RSP. Performed the experiments: RSP. Analyzed the data: RSP MPM. Contributed reagents/materials/analysis tools: RSP MPM. Wrote the paper: MPM RSP.
Glucagon levels are often moderately elevated in diabetes. It is known that glucagon leads to a decrease in hepatic glutathione (GSH) synthesis that in turn is associated with decreased postprandial insulin sensitivity. Given that cAMP pathway controls GSH levels we tested whether insulin sensitivity decreases after intraportal (ipv) administration of a cAMP analog (DBcAMP), and investigated whether glucagon promotes insulin resistance through decreasing hepatic GSH levels.Insulin sensitivity was determined in fed male Sprague-Dawley rats using a modified euglycemic hyperinsulinemic clamp in the postprandial state upon ipv administration of DBcAMP as well as glucagon infusion. Glucagon effects on insulin sensitivity was assessed in the presence or absence of postprandial insulin sensitivity inhibition by administration of L-NMMA. Hepatic GSH and NO content and plasma levels of NO were measured after acute ipv glucagon infusion. Insulin sensitivity was assessed in the fed state and after ipv glucagon infusion in the presence of GSH-E. We founf that DBcAMP and glucagon produce a decrease of insulin sensitivity, in a dose-dependent manner. Glucagon-induced decrease of postprandial insulin sensitivity correlated with decreased hepatic GSH content and was restored by administration of GSH-E. Furthermore, inhibition of postprandial decrease of insulin sensitivity L-NMMA was not overcome by glucagon, but glucagon did not affect hepatic and plasma levels of NO. These results show that glucagon decreases postprandial insulin sensitivity through reducing hepatic GSH levels, an effect that is mimicked by increasing cAMP hepatic levels and requires physiological NO levels. These observations support the hypothesis that glucagon acts via adenylate cyclase to decrease hepatic GSH levels and induce insulin resistance. We suggest that the glucagon-cAMP-GSH axis is a potential therapeutic target to address insulin resistance in pathological conditions.
The prandial status modulates the physiology of whole-body insulin-stimulated glucose disposal which reaches a maximum after a meal and decreases by about 55% after a 24h fasting period [
Glucagon is a pancreatic hormone released in the fasted state in order to maintain an adequate blood glucose level. According to Lu and colleagues glucagon also regulates cAMP that acts in the liver to decrease hepatic GSH levels [
Here, we tested the hypothesis that glucagon modulates hepatic GSH content, through the activation of the adenylate cyclase pathway, resulting in a state of postprandial insulin resistance.
Male Sprague-Dawley rats weighing 319.4±7.6g (9-weeks old) from Charles River, St. Constant, Quebec, Canada were maintained in the animal house under controlled conditions (22±1°C) on a 12h light/dark cycle. Rats had
The animals were kept anesthetised during the experiment and at the end of the protocols they were euthanized with a lethal injection of sodium pentobarbital in accordance with the guidelines of the Canadian Council on Animal Care (CCAC).
Rats were anesthetized with an intraperitoneal injection of sodium pentobarbital (65mg/kg) and anesthesia was maintained throughout the experiment by continuous infusion into the femoral vein (10mg/h/kg). During all the surgical procedures, body temperature was monitored with a rectal probe and kept at 37.0±0.5°C, by means of a heated surgical table (Harvard Apparatus, Kent, England) and overhead lamp.
All the animals were treated according to the guidelines of the CCAC, and the ethics committee on animal care at the University of Manitoba approved all protocols.
The trachea was cannulated (polyethylene tubing, PE 240, Becton Dickinson, USA) to allow spontaneous respiration. The femoral artery and the femoral vein were cannulated (polyethylene tubing, PE 50, Becton Dickinson, USA) to establish an arterial-venous shunt [
Rats were allowed to stabilize from the surgical intervention for at least 30minutes before any procedures were carried out. After stabilization, arterial blood samples (25μl) were collected every 5minutes, and glucose concentration was immediately determined, by the glucose oxidase method using a glucose analyser (1500 YSI Sport, Yellow Springs Instruments, USA), until three successive stable glucose measures were obtained. The mean of these three values is referred to as the basal glucose level.
Drugs were administered either intravenously (iv), by puncturing the shunt on the venous side or through the jugular catheter, or intraportally (ipv), through a portal vein catheter.
The methodology chosen to evaluate insulin sensitivity was the Rapid Insulin Sensitivity Test (RIST), previously described [
The medial and the right lobes of the rat liver were harvested before sacrificed the animal, and immediately frozen using dry ice, for glutathione quantification. The liver samples were wrapped and stored in -80°C freezer. The hepatic glutathione was measure using the glutathione assay kit BIOTECH GSH-420, which is a quantitative and colorimetric kit for determination of total glutathione (BIOXYTECH and OxisResearch,OXIS International, Inc. Portland).
Plasma nitric oxide levels were determined by the chemiluminescence technique, using a Sievers 280 NO Analyzer (Sievers Instruments) as previously described [
After a control RIST in the fed state, DBcAMP (N6,2’-O-dibutyryladenosine 3’,5’-cyclic monophosphate), a cAMP analog, was infused ipv at different doses ranging from 0.01 to 1.0mg/kg, for 10minutes at an infusion rate of 0.04ml/min. The second RIST was carried out after DBcAMP ipv infusion.
This protocol was divided into 2 different series. In the first series, after a control fed RIST, glucagon was infused ipv for 10minutes at an infusion rate of 0.04ml/min, up to a final dose that ranged from 0.5ng/kg to 20μg/kg,. A second RIST was carried out after glucagon infusion.
In the second series, after a control fed RIST, L-NMMA (N-monomethyl-L-arginine) was infused ipv at a dose of 0.73mg/kg for 10minutes at an infusion rate of 0.04ml/min. The second RIST was carried out after L-NMMA ipv infusion. Then, glucagon was infused ipv at 200ng/kg for 10minutes at an infusion rate of 0.04ml/min, and a third RIST was carried out.
In one set of experiments, the livers of fed and 24h fasted animals were used as controls, for hepatic GSH measurement. In other set of experiments, glucagon was infused ipv at 200ng/kg for 10minutes at an infusion rate of 0.04ml/min and 30minutes after glucagon infusion the liver was harvested, for hepatic GSH measurement. The liver samples were wrapped and stored in -80°C freezer until further analysis.
In one set of experiments, the livers and plasma of fed animals were used as controls for NO measurement. In other set of experiments, glucagon was infused ipv at 200ng/kg for 10minutes at an infusion rate of 0.04ml/min and 30minutes after glucagon infusion the liver was harvested and blood was collected for hepatic and plasma NO measurement, respectively. The liver samples were wrapped and plasma samples were stored in -80°C freezer until further analysis.
After a control fed RIST, GSH-E (Glutathione monoethylester) 1mmol/kg was administered ipv as a 10minutes bolus. After a 20min period of stabilization, glucagon was infused ipv at 200ng/kg for 10minutes at an infusion rate of 0.04ml/min. Thirty minutes after glucagon infusion, a second RIST was performed.
Sodium pentobarbital (Somnotol) was obtained from Biomeda-MTC Animal Health Inc., Cambridge, Ontario. Human insulin (NovolingeToronto) was purchased from Novo Nordisk (Mississauga, ON, Canada). Heparin was purchased from Pharmaceutical Partners of Canada, Richmond Hill, Ontario and saline from Baxter Corporation, Toronto, Ontario, Canada. D-Glucose, L-NMMA, DBcAMP and glucagon were purchased from Sigma Chemical Co. (St. Louis, MO, USA). GSH-E was purchased from Bachem, Switzerland. Tissue adhesive was acquired from GluStich Inc., Canada. All chemicals were of the highest degree of purity on the market. All the solutions for
The statistical analysis was performed by paired t-student test, two-tailed. One way ANOVA, repeated measures ANOVA, followed by Tukey's Multiple Comparison Test. Differences were accepted as statistically significant at
Hepatic cAMP regulates GSH levels that in turn control insulin sensitivity. We tested whether intraportal administration of a cAMP analogue (DBcAMP) impacts on postprandial insulin sensitivity in rats.
Firstly, we check that the DBcAMP doses used did not change mean arterial pressure (
DBcAMP dose(mg/kg) | Arterial pressure before DBcAMP infusion (mmHg) | Arterial pressure after DBcAMP infusion (mmHg) | Glycemia before DBcAMP infusion (mg/dl) | Glycemia after DBcAMP infusion (mg/dl) |
---|---|---|---|---|
0.01 | 89.5±8.0 | 90.8±8.0 | 119.3±5.8 | 122.03±7.2 |
0.1 | 113.0±4.0 | 114.7±2.2 | 116.3±10.5 |
126.7±14.2 |
1 | 93.0±4.0 | 98.0±13.0 | 122.6±6.1 |
187.8±25.3 |
Results are means±SEM, n = 10,
*p<0.05.
We found that increasing doses of ipv DBcAMP (n = 8) lead to decrease peripheral insulin sensitivity (DBcAMP 0.01mg/kg: from 172.3±6.3mg glucose/kg bw to 125.7±8.3mg glucose/kg bw, p<0.01; DBcAMP 0.1mg/kg: from 165.7±26.2 mg glucose/kg bw to 77.0±7.5mg glucose/kg bw, p<0.05; DBcAMP 1mg/kg: from 173.2±24.0mg glucose/kg bw to 98.1±38.0mg glucose/kg bw, p<0.05) (
Insulin sensitivity decreases after DBcAMP 0.01, 0.1 and 1mg/kg ipv infusion. Results are means±SEM. Paired t-test. ** = p<0.01, * = p<0.05.
Since glucagon increases hepatic cAMP and affects glucose and insulin metabolism, we sought to evaluate the role of glucagon in postprandial insulin sensitivity. A dose-response infusion experiment was performed to evaluate the effect of different doses of glucagon (0.5, 1, 2.5, 5, 10, 200ng/kg, 2 and 20±g/kg, ipv, n = 14) on arterial pressure and arterial glycemia. The values of mean arterial pressure were not significantly changed by any of the glucagon doses infused (data not shown). Basal fed glycemia levels had a negligible increase in glycemia after glucagon 0.5, 1, 2.5, 5, 10 and 200ng/kg ipv infusion (
Insulin sensitivity decreases after glucagon 200ng/kg ipv infusion(A) and insulin sensitivity increases in the fed state (B). Results are means±SEM, n = 4. Paired t-test. *** = p<0.001, * = p<0.05.
Insulin sensitivity in the fed state is dependent on GSH and NO and we tested whether glucagon is affected by hepatic NO levels in decreasing insulin sensitivity. Therefore we evaluated if after inhibition by L-NMMA (a nitric oxide synthase specific inhibitor), glucagon was able to further affect postprandial insulin sensitivity. In this series of experiments, the control fed RIST index was 177.1±1.6mg glucose/kg bw and after ipv L-NMMA infusion the RIST index decreased to 81.0±8.3mg glucose/kg bw (n = 5, p<0.001), corresponding to a full strong inhibition of postprandial insulin sensitivity. Glucagon infusion did not further alter the RIST index (80.4±5.9mg glucose/kg bw, n = 5, p<0.001,
The insulin sensitivity decreased after ipv L-NMMA 0.73mg/kg infusion and did not change after ipv glucagon 200ng/kg infusion. Results are means±SEM, n = 5. Repeated measures ANOVA, followed by the Tukey-Kramer multiple-comparison test. *** = p<0.001 Control
On the other hand, the hepatic and plasma levels of NO were not altered after acute ipv glucagon 200ng/kg infusion (hepatic levels: from 171.3±23.1 to 200.6±23.5μM/g liver, n = 7; plasma levels: from 11.7±1.4 to 10.3±0.7μM, n = 7, Fig
The hepatic NO content did not change after glucagon ipv 200ng/kg infusion in the liver (A) nor in the circulation (B). Results are means±SEM, n = 7. Unpaired t-test.
It was described by Lu et al. [
The hepatic GSH content decreases after glucagon ipv 200ng/kg infusion (A) to levels obtained in the control fast situation (B). Results are means±SEM, n = 7. Unpaired t-test. ** = p<0.01, * = p<0.05.
The insulin sensitivity did not change when ipv glucagon 200ng/kg infusion was given after GSH-E 1mmol/kg ipv infusion. Results are means±SEM, n = 3.
Results are means±SEM.
In summary, we showed that glucagon 200ng/kg showed a complete inhibition of insulin sensitivity while GSH-E administration after ipv glucagon infusion leads to a complete restoration of insulin sensitivity. This work identifies a role for glucagon in reducing postprandial insulin sensitivity and unravels that it operates through decreasing GSH hepatic levels.
The increase in postprandial insulin sensitivity is dependent on the activation of the hepatic parasympathetic nerves, leading to an increase in hepatic NO and GSH levels [
Our results showed that glucagon infusion causes a state of decreased postprandial insulin sensitivity possibly mimicking the fasting state with respect to hepatic cAMP and GSH levels (
The effect of glucose on insulin secretion can be amplified by signaling pathways involving inositol trisphosphate and diacylglycerol, derived from activation of phospholipase C [
It has been suggested that decreased GSH levels operates to increase insulin sensitivity in the muscle in the fasting conditions [
The finely tuned balance of the two major pancreatic hormones, insulin and glucagon, is impaired in type 2 diabetic subjects. This is in agreement with an imbalance of the insulin:glucagon molar ratio, since this ratio mainly affects hepatic glucose production [
Type 2 diabetes patients feature a bihormonal disorder where either absolute insulin insufficiency or relative lack of insulin is present alongside fasting and postprandial hyperglucagonemia [
In the past decades increasing evidence, including various interventions targeting glucagon secretion, glucagon’s receptor and glucagon clearance, has emerged to unequivocally support the role of fasting and postprandial hyperglucagonemia, as major contributing factor for the elevated levels of blood glucose, a hallmark of diabetes [
This work supports for the first time the notion that glucagon controls postprandial insulin sensitivity through its inhibitory action on hepatic GSH formation. Furthermore, we suggest this effect is mediated through increasing cAMP levels. Insulin resistance is an early feature of diabetes progression and our results call attention for further studies addressing the axis glucagon-cAMP-GSH as new therapeutic target for the treatment of insulin resistant states, a hallmark of type 2 diabetes and obesity (
The cAMP analog and glucagon promote an increase in hepatic cAMP levels that are related to the decrease of hepatic GSH synthesis, leading to an impairment of peripheral postprandial insulin sensitivity. On the other hand, in the presence of increased GSH levels the glucagon is abrogated and peripheral insulin sensitivity is restored. DBcAMP: N6,2‟-O-dibutyryladenosine 3‟,5‟-cyclicmonophosphate; cAMP: 3‟,5‟-cyclic adenosine 5‟-monophosphate; GSH:gluthatione.
Glycemic profiles at specific time points determined after saline and glucagon 200ng/kg ipv infusion (n = 4). Results are means±SEM.
(TIF)
Glycemic profiles at specific time points determined after saline and DBcAMP 0.01, 0.1 and 1mg/kg ipv infusion (n = 10). Results are means±SEM.
(TIF)
Insulin levels were not altered by ipv DBcAMP infusion (0.01mg/kg). Results are means±SEM. One-way ANOVA, followed by the Tukey-Kramer multiple-comparison test.
(TIF)
Insulin levels were not altered by ipv DBcAMP infusion (0.1mg/kg). Results are means±SEM. One-way ANOVA, followed by the Tukey-Kramer multiple-comparison test.
(TIF)
Insulin levels were not altered by ipv DBcAMP infusion (1mg/kg). Results are means±SEM. One-way ANOVA, followed by the Tukey-Kramer multiple-comparison test.
(TIF)
The lowest ipv glucagon doses had a minimal or negligible effect on glycemia
(TIF)
We thank Lima IS, Fernandes AB, Gaspar JM and Martins FO for helpful scientific discussions.