ARTICLE
Suppression of glucagon secretion is lower after oral glucose
administration than during intravenous glucose
administration in human subjects
J. J. Meier & C. F. Deacon & W. E. Schmidt & J. J. Holst &
M. A. Nauck
Received: 27 September 2006 /Accepted: 27 December 2006 / Published online: 16 February 2007
# Springer-Verlag 2007
Abstract
Aims/hypothesis The incretin effect describes the augmen-
tation of postprandial insulin secretion by gut hormones. It
is not known whether glucagon secretion is also influenced
by an incretin effect. A glucagon suppression deficiency
has been reported in some patients with type 2 diabetes, but
it is unclear whether this abnormality is present prior to
diabetes onset. We therefore addressed the questions: (1) Is
glucagon secretion different after oral and during intrave-
nous glucose administration? (2) If so, is this related to the
secretion of incretin hormones? (3) Is glucagon secretion
abnormal in first-degree relatives of patients with type 2
diabetes?
Materials and methods We examined 16 first-degree
relatives of patients with type 2 diabetes and ten matched
control subjects with an oral glucose load (75 g) and with
an ‘isoglycaemic’ intravenous glucose infusion.
Results Glucagon levels were significantly suppressed by
both oral and intravenous glucose (p<0.0001), but gluca-
gon suppression was more pronounced during intravenous
glucose administration (76±2%) than after oral glucose
administration (48±4%; p<0.001). The differences in the
glucagon responses to oral and i.v. glucose were correlated
with the increments in gastric inhibitory polypeptide (GIP)
(r=0.60, p=0.001) and glucagon-like peptide (GLP)-1
(r=0.46, p<0.05). There were no differences in glucagon
levels between first-degree relatives and control subjects.
Conclusions/interpretation Despite the glucagonostatic
actions of GLP-1, the suppression of glucagon secretion
by glucose is diminished after oral glucose ingestion,
possibly due to the glucagonotropic actions of GIP and
GLP-2. Furthermore, in this group of first-degree relatives,
abnormalities in glucagon secretion did not precede the
development of other defects, such as impaired insulin
secretion.
Keywords First-degree relatives . Gastric inhibitory
polypeptide . GIP. GLP-1 . Glucagon-like peptide 1 .
Glucagon secretion . Incretin effect . Type 2 diabetes
Abbreviations
GIP gastric inhibitory polypeptide
GLP glucagon-like peptide
Introduction
Postprandial glucose homeostasis is tightly controlled by
the interplay of gastric motility, endocrine pancreatic
secretion and modulation of hepatic glucose release [1, 2].
It has long been recognised that oral nutrient ingestion
elicits a greater stimulation of insulin secretion than the
intravenous infusion of a similar amount of glucose [3–5].
This manifestation of the incretin effect has been attributed
Diabetologia (2007) 50:806–813
DOI 10.1007/s00125-007-0598-z
J. J. Meier (*) :W. E. Schmidt
Department of Medicine I, St. Josef-Hospital,
Ruhr-University Bochum,
Gudrunstr. 56,
44791 Bochum, Germany
e-mail: juris.meier@rub.de
C. F. Deacon : J. J. Holst
Department of Medical Physiology, The Panum Institute,
University of Copenhagen,
Copenhagen, Denmark
M. A. Nauck
Diabeteszentrum Bad Lauterberg,
Bad Lauterberg, Germany
to the actions of the gastrointestinal hormones gastric
inhibitory polypeptide (GIP) and glucagon-like peptide
(GLP)-1 [6, 7]. Overall, the incretin effect accounts for
around 50–70% of the postprandial rise in insulin levels,
depending on the amount of glucose ingested [4, 8].
However, while GIP and GLP-1 act in concert to stimulate
glucose-dependent insulin secretion [7], they display
marked differences with regard to their effects on glucagon
secretion. Thus GLP-1 strongly suppresses glucagon
secretion [9, 10], while GIP under certain conditions even
stimulates glucagon release [11, 12]. Furthermore, the other
proglucagon-derived peptide, GLP-2, which is co-secreted
along with GLP-1, has been shown to possess glucagono-
tropic properties as well [13]. The physiological conse-
quences arising from the diverging actions of GIP, GLP-1
and GLP-2 for the postprandial regulation of glucagon
secretion are as yet unknown.
Unlike healthy volunteers, patients with type 2 diabetes
show only small differences in insulin secretion between
oral and intravenous glucose administration [14]. This
reduction in the incretin effect has been linked to
diminished efficacy of GIP as well as a deficit in GLP-1
secretion and action [15–20]. In some patients with type 2
diabetes a reduced suppression of glucagon levels after
meal ingestion has also been reported [21–23]. It is not
clear whether such alterations in glucagon suppression
represent a primary defect or whether they develop as a
consequence of other metabolic defects in type 2 diabetes.
The former alternative has been supported by previous
studies showing a deficient suppression of glucagon
secretion and, consequently, a diminished reduction in
hepatic glucose output in subjects with IGT [24, 25].
First-degree relatives of patients with type 2 diabetes are
another cohort at high risk of developing type 2 diabetes
later during their life, the average risk being about 50%
[26]. In previous studies we have shown that despite an
approximately 50% reduction of the insulinotropic effect of
GIP in a group of first-degree relatives of Europid origin,
the quantitative contribution of the incretin effect to overall
insulin secretion in these individuals was normal [17, 27].
In the present studies we sought to elucidate the potential
impact of an incretin effect on the suppression of glucagon
secretion in these individuals.
Therefore, glucagon measurements from the same study
were analysed to address the following questions: (1) Is
there a difference in the suppression of glucagon levels
between oral and intravenous glucose administration? (2) If
so, can such differences be attributed to the secretion of
incretin hormones? (3) Are there any differences in the
suppression of glucagon secretion after oral and during
intravenous glucose administration between first-degree
relatives of patients with type 2 diabetes and control
subjects?
Subjects and methods
Study protocol The study protocol was approved by the
ethics committee of the medical faculty of the Ruhr-
University, Bochum, prior to the experiments. Written
informed consent was obtained from all participants. Parts
of these studies relating to the differences in insulin
secretion between first-degree relatives of patients with
type 2 diabetes and control subjects have been reported in a
previous paper [27].
Subjects We studied 16 first-degree relatives of patients
with type 2 diabetes and ten control subjects without a
family history of type 2 diabetes. First-degree relatives were
recruited from the initial group of 21 subjects studied
previously with the intravenous infusion of GIP [17]. In the
control group of the initial cohort four members agreed to
participate in the present study and six additional subjects
were recruited. Since our previous study revealed a bimodal
pattern of insulin secretion in the first-degree relatives [17],
a higher number of first-degree relatives than controls were
studied. The groups were matched for age, sex and obesity.
Detailed subject characteristics are presented in Table 1.
Control subjects with any first- or second-degree relatives
with type 2 diabetes were excluded after collation of their
family health history.
Blood was drawn from all participants in the fasting state
for measurements of standard haematological and clinical
chemistry parameters. Spot urine was sampled for the
determination of albumin, protein and creatinine by standard
methods. Subjects with anaemia (haemoglobin<120 g/l),
with elevation in liver enzyme (alanine aminotransferase,
aspartate aminotransferase, alkaline phosphatase, γ-glutamyl
transferase) activities to greater than double the respective
normal value or with elevated creatinine concentrations
(>114 μmol/l [1.5 mg/dl]) were excluded. Body height and
weight were determined and waist and hip circumference
were measured in order to calculate BMI and the waist-to-
hip ratio, respectively (Table 1).
Table 1 Characteristics of first-degree relatives of patients with type
2 diabetes and of control subjects participating in oral glucose and
‘isoglycaemic’ intravenous clamp tests
Variable First-degree relatives
of type 2 diabetic patients
Control
subjects
p value a
Sex (men/women) 4/12 6/4 0.11
Age (years) 50±12 45±13 0.30
BMI (kg/m2) 26.1±3.8 26.1±4.2 0.98
WHR 0.83±0.09 0.88±0.09 0.19
HbA1c (%) 5.1±0.3 5.4±0.6 0.08
Mean±SD
aANOVA/χ2 test
Diabetologia (2007) 50:806–813 807
Study design All participants were first invited for a
screening visit. A general clinical examination was per-
formed and laboratory parameters were screened. If subjects
met the inclusion criteria, they were recruited for the
following tests: (1) an oral glucose challenge (75 g glucose
and/or low-molecular-mass glucose oligomers [O.G.T.;
Roche Diagnostics, Mannheim, Germany]), performed
with blood being drawn over 240 min from an indwelling
venous cannula; (2) an ‘isoglycaemic’ intravenous glucose
infusion, performed to duplicate the plasma glucose
profile determined in the same individual after the oral
glucose challenge, and with venous blood drawn over
240 min.
Experimental procedures The tests were performed in the
morning after an overnight fast with subjects in a supine
position throughout the experiments and the upper body
lifted by approximately 30°. One or two forearm veins were
punctured with a Teflon cannula (Moskito 123, 18 gauge;
Vygon, Aachen, Germany) and kept patent using 0.9%
NaCl (for blood sampling and for infusions, respectively).
Both ear lobes were made hyperaemic using Finalgon
(Nonivamid 4 mg/g, Nicoboxil 25 mg/g; Boehringer
Ingelheim Pharma, Ingelheim, Germany).
After drawing basal blood specimens at −15 and 0 min,
the oral glucose challenges were started by the ingestion of
75 g of oral glucose at 0 min. The ‘isoglycaemic’ clamp
experiments were started by the slow i.v. administration of
a small bolus of 40% glucose at t=0 min, intended to raise
plasma glucose concentrations to levels similar to those
measured after oral glucose ingestion. Subsequently, a
continuous i.v. infusion of 20% glucose was started and
the infusion rate was adjusted every 5 min according to the
respective plasma glucose measurements. Blood samples
were drawn as indicated in Figs. 1 and 2, stored on ice and
processed as described [27].
Laboratory determinations Glucose was measured using a
glucose oxidase method (Glucose Analyser 2; Beckman
Instruments, Munich, Germany). Total GLP-1 concentra-
tions were measured using an RIA (antiserum no. 89390;
all antisera raised in the laboratory of J. J. Holst) that is
specific for the C-terminal of the GLP-1 molecule and reacts
equally with intact GLP-1 and the primary (N-terminally
truncated) metabolite as described [27]. Total GIP was
measured, as described previously, using the C-terminally
directed antiserum R65 [27], which reacts fully with intact
GIP and the N-terminally truncated metabolite.
Immunoreactive glucagon was measured by an RIA
using antibody no. 4305 in ethanol-extracted plasma, as
described [28]. The detection limit was <1 pmol/l. The
intra-assay CV was 6.7% and inter-assay CV was 16%.
Fig. 1 Plasma concentrations of GIP (a, b) and GLP-1 (c, d) after
stimulation with oral glucose (75 g; filled circles) or during ‘isoglycae-
mic’ intravenous glucose infusion (open diamonds) in ten control
subjects (a, c) and 16 first-degree relatives of patients with type 2
diabetes (b, d). Beginning of intravenous infusion is marked by dotted
vertical line. Arrows indicate the time of oral glucose administration.
Means±SEM. Statistics were carried-out using paired repeated measures
ANOVA with the following p values: (1) for differences between the
experiments: p<0.0001 (a, b, d), p=0.0031 (c); (2) for differences over
time: p<0.0001 (a–d); (3) for differences due to the interaction of
experiment and time: p<0.0001 (a–d). * p<0.05 for differences at
individual time points (one-way ANOVA)
808 Diabetologia (2007) 50:806–813
Calculations Integrated plasma concentrations of glucagon
were calculated using the trapezoidal rule. For the calcula-
tion of the maximal suppression of glucagon secretion, the
lowest glucagon concentration between 15 and 240 min
was determined and expressed as a percentage of the
respective basal glucagon levels (mean value of glucagon
concentrations between −15 and 0 min).
Statistical analysis Results are reported as mean±SEM. All
statistical calculations were carried out using repeated-
measures ANOVA and Statistica, version 5.0 (Statsoft
Europe, Hamburg, Germany). Values at single time points
were compared by one-way ANOVA followed by Duncan’s
post hoc test. A two-sided p value<0.05 was taken to
indicate significant differences.
Results
Plasma glucose concentrations were similar after oral glucose
ingestion and during ‘isoglycaemic’ intravenous glucose
infusion (p=0.99). The oral glucose load elicited significant
rises in the plasma concentrations of GIP and GLP-1
(p<0.001), whereas incretin levels remained unchanged
during intravenous glucose infusion (Fig. 1). There were
no differences in the secretion of GIP or GLP-1 between
first-degree relatives of patients with type 2 diabetes and
control subjects (p=0.89 and p=0.99, respectively [27]).
Plasma insulin and C-peptide levels measured after oral
glucose administration were significantly lower in first-degree
relatives than in control subjects (p<0.001 and p=0.011,
respectively, details see [27]). The rise in insulin and C-
peptide concentrations elicited by the oral glucose load was
significantly higher (about threefold) than that evoked by the
‘isoglycaemic’ intravenous glucose infusion (p<0.0001, see
[27]).
Glucagon levels were significantly suppressed both by oral
and by intravenous glucose (p<0.0001; Fig. 2). Interestingly,
glucagon concentrations were significantly lower during the
‘isoglycaemic’ clamp experiments than in the experiments
with oral glucose administration, both in first-degree
relatives and in controls (p<0.01 for the differences between
the experiments; Fig. 2a–c). Moreover, when glucagon levels
were expressed as a percentage of basal concentrations, the
glucose-induced suppression was more pronounced during
intravenous than after oral glucose administration (p<0.0001
for the interaction of group and time; Fig. 2d).
Integrated glucagon concentrations were 33 ± 3% lower
during intravenous than after oral glucose (p<0.001;
Fig. 3). Likewise, the maximal suppression of glucagon
levels was significantly more pronounced during intravenous
glucose administration than after oral glucose administration
(76±2% vs 48±4%, respectively; p<0.001). These differ-
Fig. 2 Glucagon plasma concentrations, expressed as absolute values
(a–c) and as percentage of basal (d–f) after stimulation with oral
glucose (75 g; filled circles) or during ‘isoglycaemic’ intravenous
glucose infusion (open diamonds) in ten control subjects (b, e) and 16
first-degree relatives of patients with type 2 diabetes (c, f). a, d Results
for both groups combined. Beginning of intravenous infusion is marked
by dotted vertical line. Arrows indicates the time of oral glucose
administration. Means±SEM. Statistics were carried-out using paired
repeated measures ANOVA with the following p values: (1) for
differences between the experiments: p<0.0001 (a, c, d), p=0.0051
(b), p=0.07 (e), p<0.001 (f); (2) for differences over time: p<0.0001
(a–f); (3) for differences due to the interaction of experiment and time:
p=0.013 (a), p=0.59 (b), p=0.022 (c), p=0.0011 (d), p=0.35 (e),
p<0.0001 (f). * p<0.05 for differences at individual time points (one-
way ANOVA)
Diabetologia (2007) 50:806–813 809
ences in the integrated glucagon concentrations and the
maximal suppression of glucagon by glucose administration
were present in control subjects and in first-degree relatives
of patients with type 2 diabetes (Fig. 3).
Glucagon levels were similar in first-degree relatives of
patients with type 2 diabetes and control subjects, both after
oral glucose ingestion (p=0.96) and during intravenous
glucagon infusion (p=0.85).
There was a significant linear relationship between the
basal plasma concentrations of GIP and GLP-1 and the
respective plasma glucagon levels (r=0.73, p<0.0001 and
r=0.58, p=0.002, respectively; Fig. 4a,b). Likewise, the
differences (Δ) in the integrated plasma concentrations of
GIP and GLP-1 between the experiments with oral and
intravenous glucose administration were significantly corre-
lated to the respective differences in glucagon levels
(r=0.60, p=0.001 for GIP; r=0.46, p=0.019 for GLP-1;
Fig. 4c,d).
Discussion
In the present studies we sought to elucidate whether the
suppression of glucagon release is different after oral and
during ‘isoglycaemic’ intravenous glucose administration and
whether normal glucose-tolerant first-degree relatives of
patients with type 2 diabetes already exhibit abnormalities in
glucagon secretion. Interestingly, the suppression of glucagon
secretion was diminished by about 30% after oral glucose
ingestion compared with the intravenous glucose infusion.
These differences in the relative glucagon suppression
between oral and intravenous glucose administration were
positively associated with the secretion of GIP and GLP-1,
consistent with a stimulatory role for the incretin hormones on
alpha cell secretion. However, despite detectable alterations in
insulin secretion in the first-degree relatives [17, 27],
glucagon secretion and its suppression by glucose were
unchanged in this group at high risk of contracting type 2
diabetes.
The differences in the extent of glucagon suppression
between the experiments with oral and with intravenous
glucose are a novel and rather unexpected finding. In fact,
given the potent glucagonostatic effects of exogenous GLP-
1 shown previously in vitro and in vivo [9, 10], glucagon
levels would, if anything, have been expected to be lower
after the oral glucose load. Moreover, in the same experi-
ments the enhancement of insulin secretion was approxi-
mately three times greater after oral glucose ingestion than
with the ‘isoglycaemic’ glucose infusion [27], and prior
studies have shown that this incretin effect on insulin
secretion is specifically accomplished through an amplifi-
cation of insulin pulse mass [29]. Since the intra-islet
pulsatile release of insulin secretion may contribute to the
inhibition of alpha cell secretion [30], the augmentation of
insulin secretion after oral glucose could be expected to
result in a marked inhibition of glucagon release. Therefore,
the question arises as to which factors are responsible for
the dampening of glucagon suppression after oral glucose.
On the basis of the present studies the mechanism
underlying this phenomenon cannot be clarified with
certainty. However, a possible explanation could be the
glucagonotropic effects of GIP and GLP-2 [11–13, 31].
Indeed, GIP has been shown to enhance glucagon release in
isolated pancreatic perfusions and in humans in vivo [11,
12]. Consistent with this, the differences in glucagon
suppression between oral and i.v. glucose were significantly
correlated to the secretion of GIP (Fig. 4). On the other
hand, one might argue that the dampening of glucagon
secretion was also significantly correlated with GLP-1
secretion (Fig. 4), which is well known to inhibit glucagon
release [9, 10]. However, since the secretion of GLP-1 is
tightly linked to that of GLP-2 [32, 33] and GIP [27, 34], it
seems possible that this association was indirectly caused
Fig. 3 Integrated glucagon concentrations (0–240 min) and maximal
suppression of glucagon levels (0–240 min) after stimulation with oral
glucose (75 g) or dur