| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Studies |
Department of Endocrinology (P.A., J.B., E.H.-T.), Huddinge University Hospital, CME M63, SE-141 86 Stockholm, Sweden; and Department of Endocrinology (E.B.), Nutrition, and Metabolism, Reims University Hospital, 51092 Reims Cedex, France
Address all correspondence and requests for reprints to: Eva Hagström-Toft, Center of Metabolism and Endocrinology, M63, Huddinge University Hospital, SE-141 86 Stockholm, Sweden. E-mail: eva.toft{at}mkdiv.hs.sll.se
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
It is well known that catecholamines stimulate lipolysis via their ß-adrenoceptor subtypes, both in adipose tissue and in muscle (1, 2). It is possible that glucagon and/or glucagon-like-peptide-1 (GLP-1) also modulate lipolysis, because they also belong to the superfamily of seven transmembrane domain G protein-coupled receptors that positively regulate intracellular cAMP levels via adenylate cyclase. In humans, glucagon and GLP-1 are generated from a single proglucagon gene that gives rise to an identical proglucagon RNA transcript that is translated and processed differently in pancreatic islets and intestine.
GLP-1 is an incretin hormone that is secreted throughout the day and in increased amounts after meals, to directly stimulate insulin secretion. Multiple immunoreactive forms of GLP-1 are liberated in vivo, but GLP-1-(7-36) amide is the major form of circulating GLP-1. The biological activity of GLP-1-(7-36) amide depends on the presence of histidine at position 7 in the N-terminus (3, 4). The multiple biological effects of GLP-1, especially on pancreatic cells, makes it difficult to analyze the impact of the hormone on peripheral tissues. Thus, inconclusive results have been obtained when analyzing the impact of this hormone on glucose disposal in humans (5, 6, 7, 8, 9, 10). Studies of GLP-1 effects on lipolysis have been conducted only in vitro, in rodents and in 3T3-L1 adipocytes, and have given conflicting results (11, 12). No data on lipolysis are available in humans, although GLP-1 receptors have been detected in adipose tissue (13). It is not yet known whether GLP-1 receptors are present in human muscle, although functional receptors are present in rodent muscle. In rodents, GLP-1 receptors can both stimulate and inhibit muscle cAMP production (14, 15, 16, 17). Therefore, an additional GLP-1 receptor isoform, which negatively regulates adenylate cyclase, has been proposed for adipose tissue and muscle in rodent (12, 18).
Glucagon, which plays an important role in maintaining glucose homeostasis for many years, has been considered to have a lipolytic effect in humans (19). This hypothesis is further supported by the demonstration of specific glucagon receptors in human adipose tissue (13) and by in vitro studies showing lipolytic effects (20, 21). However, if glucagon is an important regulator of lipolysis, then convincing in vivo effects should prevail. Although the impact of glucagon on lipolysis in vivo is well recognized in several species, in vivo studies in humans have shown conflicting results (22, 23, 24, 25). In such studies, glucagon was administered iv. This discrepancy could be attributable to the low sensitivity of human adipocytes to glucagon and also to its pancreatic effects (26). Indeed, insulin secretion is never totally abolished by somatostatin administration during pancreatic clamps aimed to distinguish the peripheral effects of glucagon.
The purpose of the present study was to test directly, in vivo, whether GLP-1-(7-36) amide and glucagon have actions on lipolysis in human adipose or skeletal muscle tissue. To avoid the interfering effects of pancreatic stimulation on peripheral tissues, we used the microdialysis technique to determine in situ the effects of these hormones. Furthermore, the action of the two hormones on lipolysis in isolated human sc adipocytes was also investigated in vitro.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The study group comprised nine healthy, nonobese, Caucasian volunteers of both genders (six men and three women). None of the subjects were on regular medication. Their mean age (± SEM) was 26.3 ± 1.3 yr, and their body mass index averaged 23.3 ± 0.6 kg/m2. For the in vitro experiments, abdominal sc adipose tissue was obtained perioperatively from otherwise healthy subjects undergoing elective gallbladder surgery. The study was approved by the Ethics Committee of the Karolinska Institute. The subjects were given a detailed description of the study, and their consent was obtained.
Peptides
Synthetic human GLP-1-(7-36) amide (molecular weight, 3.298K) was obtained from Sigma-Aldrich Corp. Laboratory (St. Louis, MO). Synthetic human glucagon (molecular weight, 3.485K) was obtained from Novo Nordisk A/S (Bagsvaerd, Denmark). The peptides were dissolved at 10-5 mol/L in NaCl 0.9%, aliquoted, and stored frozen at -30 C. All the glucagon and GLP-1-(7-36) amide used in this study was from the same batch. The high-performance liquid chromatography profiles (provided by the manufacturer) showed more than 99% purity of the preparations.
Microdialysis device
The principles of the microdialysis technique for lipolysis studies (27) and the microdialysis device (28) have been described in detail. The microdialysis catheter (CMA/60; CMA Microdialysis, Stockholm, Sweden) consists of a semipermeable polyamide membrane (30 x 0.62 mm; molecular weight cut-off, 20K), which is glued to the end of a double-lumen polyurethane tube. The microdialysis catheter is inserted percutaneously and continuously perfused. The solution enters the device through the outer lumen and streams by the membrane. An exchange of substances in the interstitial fluid takes place over the membrane, so that the composition of the dialysate will mirror that of the interstitial fluid (29). Furthermore, substances added to the ingoing solvent will migrate out from the probe to the surrounding tissue. The dialysate leaves the catheter through the inner lumen, from which it is collected in capped microvials (CMA Microdialysis). The tubing is connected to a high-precision perfusion pump (CMA/107; CMA Microdialysis).
Qualitative estimates of blood flow variations are obtained by adding the flow marker ethanol, which is not locally degraded and does not affect the local tissue metabolism, to the perfusate. The ethanol concentration is determined in the ingoing and outgoing solvents, respectively, and changes in the ethanol concentration ratio (outgoing-to-ingoing ethanol concentration) reflect changes in the local blood flow (30). This technique has been validated by comparing it with the 133Xe clearance method in both skeletal muscle (31) and adipose tissue (32), and it is the only method available that allows the detection of tissue flow changes in the immediate area surrounding the microdialysis catheter.
Study protocol
All subjects were investigated in the supine position after an
overnight fast (see schematic figure of the study procedure; Fig. 1
). After superficial skin anesthesia
(EMLA; Astra L|$$|Adakemedel AB, Södertälje,
Sweden), microdialysis catheters were inserted percutaneously, using a
steel guide cannula, into the abdominal periumbilical sc adipose tissue
region and into the medial portion of the gastrocnemius muscle. The
location of the catheter in the muscle was confirmed by the presence of
involuntary muscular twitches during insertion. The distance between
the two catheters always exceeded 30 mm. On the whole, six catheters
were inserted in each subject, one catheter (control catheter) on the
right side and two catheters (for GLP-1 and glucagon
perfusion) on the left side in the muscle (legs) and adipose tissue
(abdomen), respectively. The microdialysis catheters were continuously
perfused at a speed of 2 µL/min with Ringer’s solution (147 mmol/L
Na+, 4 mmol/L K+, 2.20
mmol/L Ca2+, 156 mmol/L
Cl-) supplemented with 1% human albumin
(Pharmacia & Upjohn Sverige AB, Stockholm, Sweden)
and 50 mmol/L ethanol. Albumin was added to the perfusate to prevent
peptides sticking to either the tubing or the membrane.
|
The control catheters (one in muscle and one in adipose tissue) were perfused with Ringer-Albumin-Ethanol solution throughout the 220-min study period to detect spontaneous variations in glycerol levels or tissue blood flow (reference data).
The only way to test the effect of GLP-1 and glucagon on lipolysis, in the present experimental conditions, was to switch perfusate syringes from peptide-containing ones to isoprenaline-containing ones. Because switching could influence the glycerol and ethanol concentrations in the immediate next fraction, this was discarded. Only results obtained from the following three samples for each period were used for calculations.
At the same time that microdialysis catheters were inserted, one forearm vein was punctured with a Teflon cannula and kept patent using 0.9% saline. Plasma samples for the determination of glucose, glycerol, and insulin were then collected at the beginning of the dialysate sampling and at the end of the experiments.
Membrane recovery of the catheters for glucagon and GLP-1-(7-36) amide
The in vitro recovery across the probe membrane was studied to determine whether our experimental protocol was relevant for obtaining the extracellular levels of the peptides tested, necessary for studying a potential physiological impact on tissue lipolysis or blood flow. Because the molecular weight of glucagon exceeds that of GLP-1 and the two peptides share a similar structure, we assumed that the in vitro membrane recovery for glucagon would be lower than that of GLP-1. Thus, we only tested the in vitro membrane recovery for glucagon.
Microdialysis probes (n = 6) were placed into separate vials containing the Ringer-Albumin-Ethanol solution supplemented with glucagon 10-6 mol/L. Probes were perfused at 2.0 µL/min with the Ringer-Albumin-Ethanol solution. After a 30-min equilibration period, three 20-min samples were collected. The concentration of glucagon was determined from each dialysate sample and from the vial at the beginning and at the end of the perfusion period. At the end of the experiment, all the samples were stored at -70 C until the measurements of glucagon could be performed simultaneously. Recovery was calculated using the equation: relative recovery (%) = 100 x mean[peptidedialysate]/mean[peptidevial].
In vitro lipolysis
Surgical biopsies of abdominal sc adipose tissue were used. Tissue was obtained during elective surgery (cholecystectomy, hysterectomy, gastric banding for obesity) at the beginning of the procedure. The subjects were otherwise healthy, and no one was on regular medication. The subjects fasted overnight, and only saline was administrated iv until the fat biopsy was taken. Adipocytes were prepared exactly as described previously (35) and incubated exactly as described previously (35) with glucagon, GLP-1, and isoprenaline for 2 h (n = 4). The concentrations used were 10-10–10-6 mol/L for glucagon and GLP-1 and 10-10–10-5 mol/L for isoprenaline. The glycerol concentration in the medium was determined and applied as an indicator of lipolysis.
Chemical analysis
Dialysate glycerol was determined with an enzymatic fluorometric method, with a tissue sample analyzer (CMA/60; CMA Microdialysis). This method was shown in methodological experiments to give values almost identical to those of a bioluminescence method (36, 37). Dialysate ethanol was determined with an enzymatic spectrophotometric method (38), and the outgoing vs. ingoing perfusate ratio was calculated. Measurements of glucagon were performed using a solid-phase 125I-glucagon RIA kit (Euro-Diagnostica, Malmö, Sweden). The intraassay variation assessed by the coefficient of variation (CV) was, respectively, 2.5% at 31.1 pmol/L and 4.8% at 51.2 pmol/L. Medium glycerol concentration in the in vitro experiments was determined by bioluminescence (37).
Statistical analysis
Data are presented as means ± SEM. The CV (SD divided by the mean) was used to define the variability of sampling data in the basal state. A one-factor repeated-measures ANOVA was used to assess the time point variation within the experimental period.
The mean concentration of the last three dialysate samples was calculated for each sampling period. Differences between time segments were evaluated with the Student’s paired t test, in control catheters, to determine the spontaneous variation of the studied parameters. The Student’s two-way unpaired t test was used to analyze differences between muscle and adipose tissues in the basal state. The local glycerol responses and tissue blood flow variations to the different drugs, over time, were analyzed by calculating the relative variations in the mean values for glycerol and ethanol outflow vs. inflow ratio between the different periods of dialysate sampling (second-to-first period or third-to-second period ratio). Then, a one-factor ANOVA was used to compare these relative variations in the test catheters to those in the reference control catheter. During this analysis, a two-factor ANOVA was used to screen for a gender-modulating effect in the drug responses. A software package was used for all statistical calculations (Statview SE + Graphics, V. 1.05; Abacus Concepts, Berkeley, CA).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
During the first period, which corresponded to the basal state, the CV for catheters was 5.3 ± 0.5% for dialysate glycerol and 9.2 ± 1.3% for the ethanol ratio. These CV values were not significantly different between muscle and adipose tissue or between the three groups of catheters (control and test catheters for the respective peptides).
A moderate, but significant, change in dialysate glycerol content and ethanol ratio (P < 0.01) was seen during the whole sampling time with the control catheters, with a similar continuous decrease of these parameters being seen in both tissues. The relative variations in glycerol and ethanol ratio between the first and the third period were, respectively, -12.0 ± 3.9% and -13.8 ± 3.6%. Because there was a significant decrease after each period in the control catheter, the results were expressed as changes with respect to the previous period.
Lipolysis
In the first period, the mean glycerol level was lower in muscle than in adipose tissue (P < 0.0001). Mean glycerol levels were not significantly different among the control, GLP-1, and glucagon catheters, either for muscle (45.6 ± 2.7, 43.0 ± 5.3 and 44.7 ± 4.6 µmol/L) or for adipose tissue (82.8 ± 14.1, 92.6 ± 9.1 and 86.7 ± 8.8 µmol/L), respectively.
Similarly, no significant effect of GLP-1 or glucagon
could be detected during the second period, when peptides were
perfused. The relative variation (period 2-to-period 1 data ratio) in
the mean glycerol level was similar between the GLP-1 or
glucagon catheters and control catheters, whichever tissue was analyzed
(Fig. 2).
|
). The CV for the three consecutive data samples within each of the three sampling periods did not significantly differ among the three groups of catheters, whichever tissue was investigated.
Gender did not influence the relative variations in glycerol level among the different sampling periods (values not shown).
Blood flow
The mean ethanol ratio was lower in muscle than in adipose tissue (19.6 ± 1.3% vs. 52.6 ± 3.0%, P < 0.0001) in the basal state (indicating a higher blood flow in muscle). The ethanol ratio, which was similar among the three groups of catheters for each tissue in the first sampling period, was not significantly changed by the perfusion with GLP-1 or glucagon, when compared with the spontaneous variation in the control catheters.
However, isoprenaline induced a significant decrease in the ethanol
ratio, reflecting an increase in the tissue blood flow, both in muscle
and adipose tissue (Fig. 3). This
decrease in the ethanol ratio was similar between the
GLP-1 and glucagon catheters. The decrease in the ethanol
ratio was -26% and -36% in muscle and adipose tissue, respectively
(not statistically significant).
|
During the sampling period of the in vitro experiment, the CV for the glucagon content recovered in the dialysate of the probes was 13.3 ± 3.1%. There was no significant difference in the mean of the three dialysate samples collected, indicating a stable recovery during the period of analysis. The relative membrane recovery in vitro was 5.2 ± 1.2%.
In vitro lipolysis
Only isoprenaline resulted in increased adipocyte medium glycerol
concentrations (doubled at
10-6 mol/L), whereas
neither of the peptide hormones induced any changes in medium glycerol
at any concentration (Table 1).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Our results did not show a significant effect of glucagon on lipolysis in human adipose tissue in vivo. This is in contrast to previous experimental studies in other species, and to earlier in vitro analyses on human adipocytes, although negative in vitro results have also been reported (19). However, the fact that human adipocytes are less sensitive to glucagon than adipocytes from other species has been clearly demonstrated previously (39). This could explain the positive results on lipolysis reported when pharmacological doses of the hormone were used, as well as the conflicting data obtained at glucagon concentrations near physiological levels (22, 23, 24, 25). The stimulating action of high concentrations of glucagon on catecholamine (40) and GH (41) release could also confound the interpretation of previous studies.
The putative effects of the two peptides on muscle were also investigated because recent studies have demonstrated that lipolysis is also present in this tissue, which could be differentially regulated by glucagon when compared with the adipose tissue. However, as seen in adipose tissue, neither GLP-1 nor glucagon modified the lipolysis rate. In contrast, isoprenaline (a potent lipolytic nonselective ß-adrenergic agonist) increased the glycerol levels in muscle, although to a lesser extent than seen in adipose tissue. This finding confirms that isoprenaline has in vivo lipolytic effects in both tissues and also demonstrates the quality of the equipment we used as well as the ability to detect changes in lipolysis rate using our experimental protocol. In the same way, isoprenaline induced a significant decrease in tissue blood flow, as assessed by the ethanol outflow-to-inflow ratio measurement, leading to a relative underestimation of its impact on lipolysis. The lack of effect of glucagon and GLP-1 on tissue blood flow in our experiment indicates that the hormone has no vascular impact in peripheral tissues, despite a possible modulation of splanchnic vascular flow rate (42). Because none of the peptides altered either glycerol levels or tissue blood flow, it is evident that they had no effect on lipolysis in muscle or adipose tissue. These results were supported by the findings of the in vitro experiments of adipocytes, where only isoprenaline resulted in increased lipolysis, whereas the peptide hormones did not affect the medium glycerol levels at all.
These negative results cannot be explained by a poor membrane recovery (diffusion through the membrane). When assessed for glucagon, we could expect that local delivery of the peptide by microdialysis gave intercellular space concentrations of the hormone that were higher than those seen under physiological conditions. The same conclusion can be made for GLP-1-(7-36) amide, because glucagon and GLP-1-(7-36) amide are very similar in shape and structure, and GLP-1-(7-36) amide has a lower molecular weight than glucagon. From in vitro experiments, we obtained a relative membrane recovery of 5.2%, which corresponds to a peptide concentration of about 10-8 mol/L outside the membrane when perfusing 10-6 mol/L in the catheter. Even if the membrane recovery in vivo is lower, compared with that in vitro, the difference in the in vitro and in vivo recovery cannot exceed a factor of 10, as indicated by previous methodological studies (43).
However, clearance of peptides in the extracellular space also has to be considered. Enzymatic degradation of the peptides inside the catheter is probably not possible because the 20K-Da probes do not allow the recovery of the large-sized enzymes. Therefore, the peptide half-life would be higher in the extracellular space when infusing peptides at a similar concentration than in plasma. Concerning the issues of peptide dilution and tissue blood flow, we can refer to the mean data for the ethanol ratio, which was about 20% in muscle and 50% in adipose tissue, which was not influenced by perfusion of the peptides. We can then assume that there is a maximal one-tenth decrease in peptide level attributable to these two factors in the extracellular space around the membrane. Altogether, these data show that the minimum respective concentrations of the peptides would be 10-10 mol/L; this value is much higher than the physiological concentration of the two peptides in vivo (between 20 and 50 pmol/L; Refs. 4 and 19).
The mechanism behind the lack of effect is partly clarified by this study. Because both receptors have been identified in adipose tissue in vitro, it can only be speculated that the failure of effect could be attributable to nonfunctional receptors at physiological levels. This notion is supported by the present in vitro experiments with isolated sc fat cells. Lipolysis could not be stimulated by glucagon or GLP-1, even at the supraphysiological levels used in the in vitro experiments. However, isoprenaline was lipolytical in vitro. For GLP-1, it has been suggested that different variants of the receptor exist, having diverging effects on cAMP levels after hormone stimulation (3, 12, 18). At present, we do not know whether glucagon or GLP-1 exerts lipolytic effects in human skeletal muscle in vitro, because no methods to study this event are developed.
In conclusion, the present data demonstrate that neither glucagon nor GLP-1)-(7-36) has a direct effect on the lipolysis rate in either human sc adipose tissue or skeletal muscle.
|
Footnotes |
|---|
Received July 17, 2000.
Revised November 14, 2000.
Accepted November 29, 2000.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  J. J. Holst The Physiology of Glucagon-like Peptide 1 Physiol Rev, October 1, 2007; 87(4): 1409 - 1439. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. C. Carpentier, F. Frisch, P. Brassard, F. Lavoie, A. Bourbonnais, D. Cyr, R. Giguere, and J.-P. Baillargeon Mechanism of insulin-stimulated clearance of plasma nonesterified fatty acids in humans Am J Physiol Endocrinol Metab, March 1, 2007; 292(3): E693 - E701. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. T. Gunnarsson, M. S. Winzell, C. F. Deacon, M. O. Larsen, K. Jelic, R. D. Carr, and B. Ahren Glucose-Induced Incretin Hormone Release and Inactivation Are Differently Modulated by Oral Fat and Protein in Mice Endocrinology, July 1, 2006; 147(7): 3173 - 3180. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. A. Nafikov, B. N. Ametaj, G. Bobe, K. J. Koehler, J. W. Young, and D. C. Beitz Prevention of Fatty Liver in Transition Dairy Cows by Subcutaneous Injections of Glucagon J Dairy Sci, May 1, 2006; 89(5): 1533 - 1545. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. C. Carpentier, F. Frisch, D. Cyr, P. Genereux, B. W. Patterson, R. Giguere, and J.-P. Baillargeon On the suppression of plasma nonesterified fatty acids by insulin during enhanced intravascular lipolysis in humans Am J Physiol Endocrinol Metab, November 1, 2005; 289(5): E849 - E856. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M D. Robertson, A. S Bickerton, A L. Dennis, H. Vidal, and K. N Frayn Insulin-sensitizing effects of dietary resistant starch and effects on skeletal muscle and adipose tissue metabolism Am. J. Clinical Nutrition, September 1, 2005; 82(3): 559 - 567. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Enoksson, E. Hagstrom-Toft, J. Nordahl, K. Hultenby, N. Pettersson, B. Isaksson, J. Permert, R. Wibom, C. Holm, J. Bolinder, et al. Marked Reutilization of Free Fatty Acids during Activated Lipolysis in Human Skeletal Muscle J. Clin. Endocrinol. Metab., February 1, 2005; 90(2): 1189 - 1195. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Akesson, B. Ahren, V. C. Manganiello, L. S. Holst, G. Edgren, and E. Degerman Dual Effects of Pituitary Adenylate Cyclase-Activating Polypeptide and Isoproterenol on Lipid Metabolism and Signaling in Primary Rat Adipocytes Endocrinology, December 1, 2003; 144(12): 5293 - 5299. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 U. Holmback, A. Lowden, T. Akerfeldt, M. Lennernas, L. Hambraeus, J. Forslund, T. Akerstedt, M. Stridsberg, and A. Forslund The Human Body May Buffer Small Differences in Meal Size and Timing during a 24-h Wake Period Provided Energy Balance Is Maintained J. Nutr., September 1, 2003; 133(9): 2748 - 2755. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. M. Gustavson, C. A. Chu, M. Nishizawa, B. Farmer, D. Neal, Y. Yang, E. P. Donahue, P. Flakoll, and A. D. Cherrington Interaction of glucagon and epinephrine in the control of hepatic glucose production in the conscious dog Am J Physiol Endocrinol Metab, April 1, 2003; 284(4): E695 - E707. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Thomsen, H. Storm, J. J Holst, and K. Hermansen Differential effects of saturated and monounsaturated fats on postprandial lipemia and glucagon-like peptide 1 responses in patients with type 2 diabetes Am. J. Clinical Nutrition, March 1, 2003; 77(3): 605 - 611. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. Karpe, B. A. Fielding, J.-L. Ardilouze, V. Ilic, I. A. Macdonald, and K. N. Frayn Effects of insulin on adipose tissue blood flow in man J. Physiol., May 1, 2002; 540(3): 1087 - 1093. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
, Control catheter
(perfusion with Ringer solution during the periods 1, 2, and 3); 
,
GLP-1-(7-36) amide catheter (perfusion with Ringer
solution during the period 1; Ringer + GLP-1,
10-6 mol/L, during the period 2; Ringer +
isoprenaline, 10-6 mol/L, during the period
3); 
, glucagon catheter (perfusion with Ringer solution during the
period 1; Ringer + glucagon, 10-6 mol/L,
during the period 2; Ringer + isoprenaline,
10-6 mol/L, during the period 3). 
,
P < 0.0001 for one-factor ANOVA.