Endocrinology, 2021, Vol. 162, No.

7, 1–10
doi:10.1210/endocr/bqab065
Mini-Review

Mini-Review

The Role of Incretins on Insulin Function and

Glucose Homeostasis

Downloaded from https://academic.oup.com/endo/article/162/7/bqab065/6199910 by guest on 23 January 2023

Jens Juul Holst,1 Lærke Smidt Gasbjerg,1 and Mette Marie Rosenkilde1
1
Department of Biomedical Sciences and the NovoNordisk Center for Basic Metabolic Research, Faculty
of Health Sciences, University of Copenhagen, The Panum Institute, Copenhagen N, DK-2200 Denmark
ORCiD numbers: 0000-0001-6853-3805 (J. J. Holst); 0000-0002-7880-8515 (L. S. Gasbjerg); 0000-0001-9600-3254 (M. M.
Rosenkilde).

Abbreviations: Ex-9, exendin 9-39; GIGD, gastrointestinally mediated glucose disposal; GIP, glucose-dependent insulinotropic
polypeptide; GLP-1, glucagon-like peptide-1; IV, intravenous
Received: 26 February 2021; Editorial Decision: 22 March 2021; First Published Online: 30 March 2021; Corrected and Typeset:
1 June 2021.

Abstract
The incretin effect—the amplification of insulin secretion after oral vs intravenous ad-
ministration of glucose as a mean to improve glucose tolerance—was suspected even
before insulin was discovered, and today we know that the effect is due to the secretion
of 2 insulinotropic peptides, glucose-dependent insulinotropic polypeptide (GIP) and
glucagon-like peptide-1 (GLP-1). But how important is it? Physiological experiments have
shown that, because of the incretin effect, we can ingest increasing amounts of amounts
of glucose (carbohydrates) without increasing postprandial glucose excursions, which
otherwise might have severe consequences. The mechanism behind this is incretin-
stimulated insulin secretion. The availability of antagonists for GLP-1 and most recently
also for GIP has made it possible to directly estimate the individual contributions to post-
prandial insulin secretion of a) glucose itself: 26%; b) GIP: 45%; and c) GLP-1: 29%. Thus,
in healthy individuals, GIP is the champion. When the action of both incretins is pre-
vented, glucose tolerance is pathologically impaired. Thus, after 100 years of research,
we now know that insulinotropic hormones from the gut are indispensable for normal
glucose tolerance. The loss of the incretin effect in type 2 diabetes, therefore, contributes
greatly to the impaired postprandial glucose control.
Key Words: glucagon-like peptide-1 (GLP-1), glucose-dependent insulinotropic polypeptide (GIP), hormone antagon-
ists, hormone coagonists, exendin 9-39

Bayliss and Starling, who in 1902 discovered secretin (1),
the very first hormone, were convinced that not only the
exocrine, but also the endocrine secretion from the pan-
creas, discovered by von Mering and Minkowski in 1889
(2), was regulated by intestinal factors, and both they
and other scientists investigated the hypoglycemic effects
of intestinal extracts and secretin preparations. Thus, the
incretin concept, the idea that the gut regulates glucose
levels via effects on the endocrine pancreas, is older than
insulin itself. Although Zunz and La Barre, like Bayliss

ISSN Online 1945-7170

© The Author(s) 2021. Published by Oxford University Press on behalf of the Endocrine Society.
This is an Open Access article distributed under the terms of the Creative Commons Attribution-NonCommercial-
NoDerivs licence (http://creativecommons.org/licenses/by-nc-nd/4.0/), which permits non-commercial reproduction https://academic.oup.com/endo   1
and distribution of the work, in any medium, provided the original work is not altered or transformed in any way, and
that the work is properly cited. For commercial re-use, please contact journals.permissions@oup.com
2
and Starling, were convinced that secretin was the most
important hypoglycemic substance (3), they nevertheless
tried to distinguish between “excretins” (stimulating the
exocrine secretion from the pancreas) and “incretins” with
predominant hypoglycemic effects. In those days, the physi-
ologists made many impressive and daring (and ethically
problematic) experiments such as anastomosing a pancre-
atic vein from a donor dog to the jugular vein of a recipient
dog, to demonstrate that blood glucose falls in the latter
after injection of a secretin preparation into the donor dog,
but it was only after the development of the radioimmuno-
assay for insulin (4) that it became possible to quantify
and determine the importance of the incretin effect for in-
sulin secretion. In 1964, McIntyre and colleagues (5) re-
ported that intestinal administration of glucose elicited a
much greater insulin response than intravenous (IV) in-
jections and they correctly deduced that intestinal factors,
incretins, were responsible. In 1971, Brown and Dryburgh
had sequenced a new peptide, gastric inhibitory polypep-
tide, GIP, isolated from impurities of cholecystokinin on
the basis of its inhibitory effects on acid secretion from
canine gastric pouches (6), but in 1973 Dupre and Brown
identified the insulinotropic effects of GIP (7) (which, be-
cause the inhibitory effects on gastric secretion turned out
not to be physiologically relevant, was renamed “glucose-
dependent insulinotropic polypeptide,” retaining the nice
acronym, GIP). Many subsequent studies confirmed its
insulinotropic actions, and in 1989, Nauck et al could
show in mimicry experiments that GIP fulfilled rigorous
criteria for being a physiologically important incretin hor-
mone (8). Thus, GIP, infused intravenously in amounts that
copied its plasma concentrations after oral glucose, had ef-
fects on insulin secretion that were similar to those elicited
by the oral glucose.
But what about secretin? Secretin had been isolated and
sequenced in Stockholm in 1970 by Mutt and colleagues
(from extracts of intestines from 10 000 hogs) (9), and it
was now possible to synthesize the peptide and develop
radioimmunoassays. Infusion experiments clearly con-
firmed the old observations that secretin would stimulate
insulin secretion (actually even in people with type 2 dia-
betes [10]), but the circulating concentrations of secretin
were extremely low (1-2 pmol/L) and even after maximal
stimulation with intraduodenal acid, they did not reach
levels sufficient to stimulate insulin secretion (11). But
there had to be other incretin(s). With the insulin radio-
immunoassay, it was possible to quantify accurately the
incretin effect by comparing insulin responses to IV and
oral administrations of glucose given in amounts that re-
sulted in identical glucose concentrations (isoglycemia),
and it could be calculated that up to 70% of the response
to oral glucose could be due to the incretin effect (12, 13).
Endocrinology, 2021, Vol. 162, No. 7
However, in animal experiments, immunoneutralization
of GIP with potent antiserum samples did not eliminate
the incretin effect (14), and in people with resections of
different parts of the small intestine, there was no correl-
ation between the incretin effect and the GIP responses
(15). Clearly, something was missing. The missing hormone
was glucagon-like peptide-1 (16), a product of intestinal
expression of proglucagon, the precursor of the pancre-
atic hormone glucagon. The gene encoding proglucagon is
also expressed in the so-called L cells, endocrine cells of
the intestinal epithelium (17), where the precursor is pro-
Downloaded from https://academic.oup.com/endo/article/162/7/bqab065/6199910 by guest on 23 January 2023
cessed to release 2 glucagon-containing peptides, glicentin
and oxyntomodulin, and 2 glucagon-like peptides, GLP-1
and GLP-2 (18, 19). Of these products, oxyntomodulin and
GLP-1 both are insulinotropic, but whereas the concentra-
tions of oxyntomodulin normally are too low to stimulate
insulin secretion (20), the 100-fold more potent GLP-1 pro-
vides a very powerful stimulus (21) and in mimicry experi-
ments clearly qualified as a new incretin; actually, at first
sight, more potent than GIP (22). Today, there is agreement
that GLP-1 and GIP are the most important incretin hor-
mones, and it was determined early on that in mice with
deletions of the receptors for both GIP and GLP-1, very
little incretin effect remained (23). However, it is also clear
that these mouse experiments do not tell us much about the
importance of the incretin effect. After oral glucose gavage,
there are observable differences between the glucose excur-
sions in animals with and without receptor deletions, but
they are small and barely significant. In addition, it is pos-
sible that rodents, not the least mice, depend less on the
incretin effect in their regulation of glucose homeostasis
than humans and more on neural regulation (24, 25), and
it is therefore questionable how much these experiments
can tell us about human conditions.
The Importance of the Incretin Effect
So, how important is the incretin effect in people? This
is really an important question, because one of the char-
acteristics of type 2 diabetes is a more or less complete
loss of the incretin effect (13). This raises the ques-
tion whether type 2 diabetes is an incretin disease, al-
though the loss of the incretin effect does not usually
top the list of pathogenic factors. The importance of the
incretin effect in healthy individuals can be illustrated
in experiments in which the so-called gastrointestinally
mediated glucose disposal (GIGD [26]) is determined.
The approach is simple: One measures the total amount
of IV glucose that is required to mimic the plasma ex-
cursions after a given oral dose and compares the two.
Using this technique, Nauck et al (27) first showed
that GIGD strongly depends on the amount of glucose
Endocrinology, 2021, Vol. 162, No. 7 3

ingested. Thus, with 25-g glucose given orally, an infu-
sion of about 20 g of glucose was required to copy the
plasma glucose response. However, with 50 g of glucose,
it also took about 20 g to copy the curve, and with 75 g
it also took 20 g to copy the curve. It follows that the
plasma excursions after the 3 doses of oral glucose were
virtually identical. In other words, the body is capable of
disposing of orally ingested glucose so rapidly and effi-
ciently that the plasma glucose excursions remain con-
stant and low in spite of widely different amounts taken
in. Thus, this gastrointestinal mechanism is capable of
some early studies (29); it was shown that a 25-g dose
of glucose resulted in almost the same insulin secretion
(measured as C-peptide) whether it was given intraven-
ously or orally, and it was concluded that the “incretin
effect” was a hoax. But we now know that a) a dose of
glucose as small as 25 g elicits a GIGD of only 20% (be-
cause it took 20 g of IV glucose to mimic the 25-g oral
dose—see earlier), and b) IV administration gives a much
larger plasma glucose response than oral dosing, ruining
the principle of isoglycemia, and therefore the possibility
to gauge the incretin response properly.

Downloaded from https://academic.oup.com/endo/article/162/7/bqab065/6199910 by guest on 23 January 2023

removing up to three-quarters of the absorbed glucose
from the circulation. This shows how efficiently we nor-
mally regulate our postprandial glucose concentrations.
But what is the mechanism behind this remarkable glu-
cose disposal? it turns out to be insulin secretion, which
increases progressively with increasing glucose doses
(27). In other words, the incretin effect secures normal
and healthy postprandial glucose excursions almost no
matter how large the dosing is (125 g has also been in-
vestigated with similar results [28]). Funnily enough, the
importance of the incretin effect was nearly missed in
The incretin hormone responses were also deter-
mined in the cited dose-response experiments (Fig. 1) and
show some interesting features (27, 28). GLP-1 secretion
showed a dose response relationship that corresponded
more or less with the insulin responses. GIP secretion
was a bit different, showing an early rise to a peak value,
which was more or less the same with all doses, and a
subsequent plateau, the duration of which was propor-
tional to the glucose dose. This profile probably reflects
the gastric emptying of glucose, which is regulated so as
to ensure a relatively constant emptying rate of nutrients,

Figure 1. Localization of glucose-dependent insulinotropic polypeptide (GIP)- and glucagon-like peptide-1 (GLP-1)-secreting cells in the gastrointes-
tinal tract (left) and secretion profiles of GIP, GLP-1, and insulin in response to ingestion of low (25 g), medium (50-75 g) or high (100-125 g) amounts
of glucose (right). Curves adapted from Nauck et al (27) and Bagger et al (28).
4
corresponding, in the case of glucose, to about 4 kcal/min
(30, 31). Since this translates into a more or less constant
exposure to glucose of the duodenal mucosa, from where
most of the GIP is secreted, an enhanced secretion of GIP
will be maintained as long as there is glucose left in the
stomach. GLP-1 secretion occurs from slightly more distal
gut segments, which means that most of the L cells are
likely to escape stimulation because of early absorption
of glucose. Actually, the grossly exaggerated GLP-1 re-
sponses to more distal glucose exposure in patients with
gastric bypass operations (32) is another illustration of
this. Both the L cells and the GIP-secreting K cells ex-
press the sodium-coupled glucose transporter, SGLT1,
and their secretory responses can be demonstrated to de-
pend on this transportation. In this way, the secretion of
the incretin hormones becomes proportional to the rate of
glucose absorption in the gut segment, where this occurs.
(33, 34).
The conclusion is that normal postprandial glucose tol-
erance to a very large extent is determined by the incretin
effect, which in turn is due to the secretion of the incretin
hormones. Thus, the stimulatory effect of glucose alone
on insulin secretion is far from sufficient. As briefly men-
tioned earlier, the incretin effect is very much reduced or
frankly missing in patients with type 2 diabetes, and one
may ask why that is. When it became possible to measure
the plasma concentrations of GIP and GLP-1, it was nat-
ural to look for secretion of these hormones in diabetes.
There are strong indications that the postprandial secretion
of GLP-1 is impaired in type 2 diabetes (35); perhaps not
the early phase of meal responses (36), but particularly the
late phase of the response, and the impairment is probably
more related to the negative influence of obesity than it is
related to diabetes. However, it is also clear that impaired
secretion is not the most important explanation for the
decreased incretin effect. Rather, the insulinotropic action
of both hormones is impaired (37). The impaired incretin
effect is a very early event in type 2 diabetes, but it is the
prevailing concept that the defect develops in response to
Figure 2. Central events in the history of the 2 incretin receptor antagonists, exendin(9-39)NH2 and GIP(3-30)NH2.
Endocrinology, 2021, Vol. 162, No. 7
diabetes rather than being a primary deficiency (38). More
about this later.
The evidence regarding the importance of the incretin
hormones for glucose tolerance presented so far is based on
the principles of the mimicry experiments: If we mimic the
plasma profiles observed in vivo with infusions of glucose/
hormones, do we get the same responses? Although such
experiments represent necessary conditions for conclu-
sions on causality, they are not sufficient. Other hormones
could easily be involved and neural, renal, hepatic and, in
fact, gut mechanisms could be overlooked. The double-
Downloaded from https://academic.oup.com/endo/article/162/7/bqab065/6199910 by guest on 23 January 2023
knockout experiments referred to previously could be con-
sidered sufficient evidence (with the trivial reservation that
the receptors were truly knocked out and that knockout
did not result in compensatory adaptations, which is more
likely than not), but were not carried out in the relevant
species. To solve the problem, the application of hormone
receptor antagonists would be a useful approach. Such
antagonists must be acceptable for human use, which is a
major hindrance, but this has recently become possible for
both hormones (Fig. 2).
Glucagon-Like Peptide-1 Receptor
Antagonism
In 1992, Eng et al isolated a 39 amino acid peptide,
exendin-4, from the saliva of the Gila monster (Heloderma
suspectum) (39), which was capable of activating receptors
om pancreatic acinar cells, but also found that a truncated
version, exendin 9-39 (40), could antagonize the actions
of exendin-4 (39). Very soon after its discovery, exendin-4
was demonstrate to be a full and potent agonist for the
GLP-1 receptor, which had just been cloned, and the trun-
cated form, exendin 9-39 (Ex-9), was demonstrated to an-
tagonize the actions of exendin-4 as well as GLP-1 on the
GLP-1 receptor (41) (see Fig. 2). Interestingly, exendin-4,
which has about 50% sequence homology with mamma-
lian GLP-1 (regarding the first 30 amino acids) is not the
GLP-1 of the Gila monster—this poisonous lizard has its
Endocrinology, 2021, Vol. 162, No. 7
own GLP-1 (which is much closer to mammalian GLP-1)
(42). Back to the antagonist: Two laboratories soon car-
ried out infusion experiments in humans of Ex-9 to elu-
cidate the role of the endogenous GLP-1 by blocking its
receptor (43, 44). Regarding safety, these experiments
went well, which is ironic considering that the use of syn-
thetic exendin-4 (exenatide) for diabetes therapy was
subsequently suspected of causing acute pancreatitis (45)—
leading a Food and Drug Administration officer to remark
in the New England Journal of Medicine (46) that this was
an expected result considering the toxicity of being bitten
by a Gila monster (the GLP-1 receptor agonists, including
exendin-4, were later completely acquitted with respect to
the pancreatitis suspicion [47]). Since then numerous ex-
periments have been carried out with Ex-9. Indeed, it is cur-
rently being developed as a therapeutic agent (avexitide) to
prevent reactive postprandial hypoglycemia after bariatric
surgery, which appears to be due to exaggerated GLP-1 se-
cretion and similarly exaggerated insulin responses, suffi-
cient to cause temporary hypoglycemia (ClinicalTrials.gov
identifier: NCT03373435). In fact, avexitide was granted
breakthrough therapy designation by the Food and Drug
Administration in 2019 for this indication (postbariatric
hypoglycemia).
In human experiments involving intraduodenal glu-
cose infusions, Ex-9 increases plasma glucose excursions,
decreases insulin responses, and increases glucagon re-
sponses, which would be expected based on the actions
of exogenous GLP-1 (43, 48). The results of Ex-9 infu-
sion in the fasting state and during simple oral glucose
administration are less easy to understand, however. The
infusions invariably elevate glucose concentrations and in-
crease glucagon concentrations (44), but insulin responses
are more variable, and decreases as well as increases have
been found. The increases in glucagon would be assumed
to arise because of antagonism of GLP-1 receptors, but
where? The α cells do not seem to express GLP-1 receptors
(very few at least), and most people agree that the inhib-
ition would be due to a relief of the paracrine inhibition
within the islets by somatostatin produced by neighboring
δ cells, which express stimulatory GLP-1 receptors (49-
51). But in the fasting state there is not much GLP-1 se-
cretion. It has been shown that Ex-9 may act as an inverse
agonist (52, 53), inhibiting the activity of the receptor it
binds to, and in that case it is conceivable that Ex-9 may
stimulate glucagon secretion by inhibiting the tonic activity
of the δ cells (because of some constitutive activity of the
GLP-1 receptor) to which the α cells are coupled in a para-
crine feedback loop (54). Stimulated insulin secretion after
Ex-9 may then be due to a paracrine action of glucagon
on the β cells via their glucagon receptors, potentiated
by the increasing glucose levels (Fig. 3). This is not pure
5
Downloaded from https://academic.oup.com/endo/article/162/7/bqab065/6199910 by guest on 23 January 2023
Figure 3. Exendin(9-39)NH2 actions on pancreatic δ (delta), α (alpha),
and β (beta) cells, which can result in increased insulin secretion.
Glucagon-like peptide-1 (GLP-1) receptors are red, glucagon recep-
tors are blue, and somatostatin receptors black. The yellow peptide
is exendin(9-39)NH2, GLP-1 is red, glucagon is blue, and somatostatin
black. Blockage of the GLP-1R on δ cells by exendin(9-39)NH2 prevents
somatostatin inhibition of α cells. Increasing glucagon secretion may
stimulate insulin secretion via the glucagon receptor in the β cells.
Exendin(9-39)NH2 may also inhibit GLP-1 action on beta cells or act as
inverse agonist.
speculation: Glucose levels and glucagon concentration do
increase during Ex-9 administration. In previously diabetic
individuals who undergo gastric bypass surgery, Ex-9 in-
variably reduces meal-induced insulin responses and im-
pairs glucose tolerance, while increasing GLP-1 responses
further (55); the GLP-1 increase seems to be due to inter-
ference with another paracrine interaction, this time in the
gut between L cells and somatostatin-producing δ cells or
somatostatinergic neurons (56), which are known to be
stimulated by the increased GLP-1 secretion. When this
effect of GLP-1 is blocked by Ex-9, somatostatin release
falls and GLP-1 secretion increases.
What happens to GIP secretion during Ex-9 infusion?
The answer is: nothing (57). This tells us that the K cells
are not subject to a similar somatostatinergic regulation as
the L cells, and that there is no “compensatory” increase in
GIP secretion, when GLP-1 action is blocked.
Glucose-Dependent Insulinotropic
Polypeptide Receptor Antagonism
Various methods for producing GIP antagonism have been
implemented in animal experiments during previous dec-
ades, but only recently has it been possible to antagonize GIP
action in humans (see Fig. 2). In a systematic search for an-
tagonistic effects of truncations and substitutions of the GIP
sequence, a peptide with a high affinity for the GIP receptors,
which also turned out to be a potent competitive antagonist,
was identified, namely GIP 3-30 NH2 (58, 59). A truncated
6
Figure 4. Contribution of gut-derived factors to postprandial insulin
secretion, expressed in per cent of the total C-peptide secretion (area
under curve, 0-240 minutes) following ingestion of 75-g glucose orally
with or without incretin receptor antagonist infusions in healthy indi-
viduals. Calculations performed by Gasbjerg et al (64). GIP, glucose-
dependent insulinotropic polypeptide; GLP-1, glucagon-like peptide-1.
form of GIP, GIP 1-30 NH2, which is a full and equipotent
agonist of the GIP receptor, had been known to exist for
some time (60), and because GIP, like GLP-1, is substrate for
the proteolytic activity of dipeptidyl peptidase-4 (DPP-4), an
N-terminally truncated form of this molecule, GIP 3-30NH2,
was predicted to exist. GIP 1-30 NH2 is found in the cir-
culation in very low picomolar concentrations (57), so the
concentrations of endogenous GIP 3-30NH2 are likely to be
similarly low. However, infused in sufficient quantities, the
antagonist dose-dependently blocks all the actions of GIP to
a very high degree (> 80%), be it on insulin secretion, stimu-
lation of adipose tissue blood flow and triglyceride uptake, or
inhibition of bone resorption (61, 62). Whereas Ex-9 works
equally well in experimental animals and humans, the GIP
system shows clear differences between species, necessitating
the use of species-specific agonists and antagonists in studies
of GIP receptor–mediated activities (59).
Combinations of Glucose-Dependent
Insulinotropic Polypeptide and Glucagon-
Like Peptide-1 Receptor Antagonism
In a recent study (57), antagonists of GIP and GLP-1 were
combined in an attempt to finally define the incretin actions
of the 2 hormones during an oral glucose tolerance test in
healthy volunteers. Indeed, and as expected, both of the ant-
agonists significantly impaired glucose tolerance, and when
given together their effects were additive, bringing the glu-
cose responses into the range of pathologically impaired
glucose tolerance (peak levels increasing by ~ 3 mmol/L and
2-hour values from 6.4 to 7.5 mmol/L), clearly confirming
the importance of the incretin hormones for postprandial
glucose tolerance. Their immediate effect on insulin secre-
tion was difficult to gauge because of the increasing blood
glucose concentrations (compared to the saline control
Endocrinology, 2021, Vol. 162, No. 7
infusion). However, when C-peptide responses (as a more
accurate measure of β-cell secretion) were adjusted for the
different glucose responses, a clear impairment of glucose-
induced insulin secretion became apparent, more with the
GIP antagonist GIP(3-30)NH2 than with Ex9, and again
roughly additively with the combination of the 2 antag-
onists. In an accompanying editorial, Nauck and Meier
calculated the individual contributions to the effects (63)
and concluded that the incretin effect, deduced from the
contributions of each of the components, was distributed
with 26% from glucose alone (similar to the value found
Downloaded from https://academic.oup.com/endo/article/162/7/bqab065/6199910 by guest on 23 January 2023
in traditional estimates of the incretin effect) and with 45%
from GIP and 29% from GLP-1. These figures were subse-
quently confirmed by calculations from the individual data
(Fig. 4) (64). Thus, it may be concluded that in healthy in-
dividuals, GIP is the most important incretin, perhaps con-
sistent with the proximal location of the K cells (duodenum
and proximal jejunum) and their immediate dependence on
the outflow of carbohydrate from the stomach, whereas the
GLP-1 effect sets in when ingested glucose reaches a bit
farther down into the intestine. Indeed, after a gastric by-
pass operation, as alluded to earlier, the nutritional load
is pushed farther down the small intestine, and the GLP-1
responses are huge (typically 10-fold increased), whereas
GIP responses are unchanged, and almost all the increased
insulin secretion disappears with Ex-9 (55). It should not
be forgotten, as discussed previously, that Ex-9 is a difficult
tool to work with; therefore, the quantitative data obtained
may be somewhat misleading.
The Incretins and Type 2 Diabetes
As mentioned, patients with type 2 diabetes have an im-
paired or absent incretin effect, and in an experiment in
patients with rather long-standing diabetes, infusions of
the 2 hormones in amounts recapitulating normal meal
responses, there was absolutely no effect on insulin se-
cretion during the conditions of a 15-mmol/L hyper-
glycemic clamp (65). In agreement with these findings,
infusion of the GIP antagonist in patients with type 2 dia-
betes had no effect on postprandial glucose levels (66),
whereas Ex-9 slightly increases postprandial glucose (67).
Likewise, whereas even very high rates of GIP infusion
had very little effects on insulin secretion and glucose
turnover in additional individuals with type 2 diabetes
(68), supraphysiological GLP-1 infusions were capable of
restoring insulin secretion levels to values similar to those
observed in healthy controls during glucose infusion
alone (whereas glucose alone was similarly ineffective
in the patients). The reason for this difference between
the 2 incretin hormones is still unknown. In a careful
Endocrinology, 2021, Vol. 162, No. 7
dose-response study involving increasing infusion rates of
GLP-1 on top of a step-wise clamping of plasma glucose
at increasing levels, the slopes of the GLP-1 effects on
the β-cell responsiveness to glucose (the latter calculated
from cross-correlations of insulin secretion rates and glu-
cose clamp levels) could be determined (69). This slope
was clearly impaired in patients compared to controls,
even if adjusted to the estimated maximal β-cell perform-
ance. This clearly illustrated the decreased responsiveness
of the diabetic β cells to GLP-1, but it also showed that
GLP-1, even at relatively low infusion rates, was capable
of restoring β-cell function to the same levels as those ob-
served in the controls given glucose alone (69). In other
words, in patients with type 2 diabetes, supraphysiological
administration of GLP-1 may improve β-cell secretion to
levels similar to those of healthy controls in response to
glucose alone (whose responses to combinations of glu-
cose and GLP-1 would, of course, be much larger). This
effect of GLP-1 forms the background for its therapeutic
use in patients with type 2 diabetes, in whom a GLP-1
infusion may actually restore fasting glucose levels to
completely normal levels (70). It must be emphasized,
however, that the β-cell responses of the individual pa-
tient with type 2 diabetes depend strongly on the residual
β-cell function, and in patients with very poor β-cell re-
serve, the effect of GLP-1 may not be sufficient to nor-
malize glucose levels (71). Thus, in patients with fasting
glucose levels between 7 and 10 mmol/L, a “standard”
1.2-pmol/kg × min infusion of GLP-1 was capable of
completely normalizing glucose levels (72); in those with
fasting glucose levels between 10 and 15 mmol/L, glucose
levels were much improved and reached 6 to 7 mmol/L
within 4 hours; and in those with fasting values greater
than 15 mmol/L there was a decrease in glucose levels
of up to 10 mmol/L, but in spite of extended infusion,
levels did not fall below 8 mmol/L, indicating that even
the slightest meal stimulus would bring glucose levels up
again, so it would never be possible to cause a sufficient
and lasting improvement in glycemic control in this way
(72). In such a patient, a combination of basal insulin and
GLP-1 works wonders (73).
As is evident from the process described earlier, the
focus has been on the development of GLP-1 analogues for
type 2 diabetes, whereas GIP was considered unattractive.
Some experiments indicated that improved glycemic con-
trol in the patients might lead to recovery of some of the
insulinotropic effects of GIP (74). But the truth is that
these improvements were very small and far from levels
that would be therapeutically relevant. However, more re-
cently monomolecular GIP/GLP-1 coagonists have been
developed, and one of these, tirzepatide, showed very
7
strong antidiabetic properties in phase 2 studies, exceeding
those of an established, long-acting GlP-1 receptor agonist
(dulaglutide), and it was suggested that its effectiveness was
due to a combination of the GIP and the GLP-1 effect (75).
The coagonist also has strong inhibitory effects on appetite
and food intake (and eventually body weight), something
that has never been observed in human studies with acute
GIP administration (and it has not yet been excluded that
the effect could be due solely to GLP-1 receptor agonism).
In contrast, in other experimental studies, GIP antagon-
ists lowered body weight, particularly in combination with
Downloaded from https://academic.oup.com/endo/article/162/7/bqab065/6199910 by guest on 23 January 2023
GLP-1 (76). The surprising effects of the coagonism are
currently under intense investigation, and some studies in-
dicate that differential internalization of the GIP and GLP-1
receptors after exposure to the coagonist may explain its
unexpected effect (77). At any rate, the phase 3 studies with
tirzepatide (78) support its suitability for therapy of type
2 diabetes (according to press releases from Eli Lilly), and
future studies will probably show whether GIP agonism or
antagonism is preferable for metabolic interventions.
However, the incretin story presented here clearly shows
that in healthy individuals the β cells do not work alone—for
normal glucose tolerance, the gastrointestinal tract and the
incretin hormones are absolutely indispensable. Although
not expected originally, the incretin system also plays an im-
portant role in the regulation of food intake, and the most
recent research has shown that GLP-1 receptor agonists
and surprisingly also GIP/GLP-1 receptor coagonists have
powerful effects on the treatment of not only diabetes but
also obesity. The incretins certainly have come of age.
Additional Information
Correspondence: Jens Juul Holst, MD, University of Copen-
hagen, Department of Biomedical Sciences, The Panum Institute,
3 Blegdamsvej, Copenhagen, DK-2200 Denmark. Email: jjholst@
sund.ku.dk.
Disclosures: J.J.H. appears on advisory boards for NovoNordisk
and MSD, and J.J.H. and M.M.R. have collaborated with Lilly on
coagonist projects. J.J.H. and M.M.R. are founders of Antag Phar-
maceuticals. L.S.G. has nothing to disclose.
Data Availability: No new data have been included in this review.
References
1. Mering Jv,Minkowski O. Diabetes mellitus nach Pankreasexstirpation.
Archiv f experiment Pathol u Pharmakol. 1890;26:371-387.
2. von Mering J, Minkowski O. Diabetes mellitus and pancreas ex-
tirpation. Arch Exp Path Pharmak. 1889;21:371-390.
3. Zunz E, La Barre J. Contributions a l’étude des variation
physiologiques de la sécrétion interne de pancréas: relations
entre les secretions externe et interne du pancréas. Archs Int
Physiol Biochim. 1929;31(1):20-44.
8
4. Berson SA, Yalow RS. Quantitative aspects of the reaction
between insulin and insulin-binding antibody. J Clin Invest.
1959;38(11):1996-2016.
5. McIntyre N, Holdsworth CD, Turner DS. New interpretation of
oral glucose tolerance. Lancet. 1964;2(7349):20-21.
6. Brown JC, Dryburgh JR. A gastric inhibitory polypep-
tide. II. The complete amino acid sequence. Can J Biochem.
1971;49(8):867-872.
7. Dupre J, Ross SA, Watson D, Brown JC. Stimulation of in-
sulin secretion by gastric inhibitory polypeptide in man. J Clin
Endocrinol Metab. 1973;37(5):826-828.
8. Nauck M, Schmidt WE, Ebert R, et al. Insulinotropic proper-
ties of synthetic human gastric inhibitory polypeptide in man:
interactions with glucose, phenylalanine, and cholecystokinin-8.
J Clin Endocrinol Metab. 1989;69(3):654-662.
9. Mutt V, Jorpes JE, Magnusson S. Structure of porcine secretin.
The amino acid sequence. Eur J Biochem. 1970;15(3):513-519.
10. Deckert T. Stimulation of insulin secretion by glucagon and se-
cretin. Acta Endocrinol (Copenh). 1968;57(4):578-584.
11. Fahrenkrug J, Schaffalitzky de Muckadell OB, Kühl C. Effect
of secretin on basal- and glucose-stimulated insulin secretion in
man. Diabetologia. 1978;14(4):229-234.
12. Perley MJ, Kipnis DM. Plasma insulin responses to oral and
intravenous glucose: studies in normal and diabetic subjects. J
Clin Invest. 1967;46(12):1954-1962.
13. Nauck M, Stöckmann F, Ebert R, Creutzfeldt W. Reduced
incretin effect in type 2 (non-insulin-dependent) diabetes.
Diabetologia. 1986;29(1):46-52.
14. Ebert R, Unger H, Creutzfeldt W. Preservation of incretin ac-
tivity after removal of gastric inhibitory polypeptide (GIP)
from rat gut extracts by immunoadsorption. Diabetologia.
1983;24(6):449-454.
15. Lauritsen KB, Moody AJ, Christensen KC, Lindkaer Jensen S.
Gastric inhibitory polypeptide (GIP) and insulin release
after small-bowel resection in man. Scand J Gastroenterol.
1980;15(7):833-840.
16. Holst JJ. The physiology of glucagon-like peptide 1. Physiol
Rev. 2007;87(4):1409-1439.
17. Eissele R, Göke R, Willemer S, et al. Glucagon-like peptide-1
cells in the gastrointestinal tract and pancreas of rat, pig and
man. Eur J Clin Invest. 1992;22(4):283-291.
18. Orskov C, Holst JJ, Knuhtsen S, Baldissera FG, Poulsen SS,
Nielsen OV. Glucagon-like peptides GLP-1 and GLP-2, pre-
dicted products of the glucagon gene, are secreted separately
from pig small intestine but not pancreas. Endocrinology.
1986;119(4):1467-1475.
19. Mojsov S, Heinrich G, Wilson IB, Ravazzola M, Orci L,
Habener JF. Preproglucagon gene expression in pancreas and
intestine diversifies at the level of post-translational processing.
J Biol Chem. 1986;261(25):11880-11889.
20. Holst JJ, Albrechtsen NJW, Gabe MBN, Rosenkilde MM.
Oxyntomodulin: actions and role in diabetes. Peptides.
2018;100:48-53.
21. Holst JJ, Orskov C, Nielsen OV, Schwartz TW. Truncated
glucagon-like peptide I, an insulin-releasing hormone from the
distal gut. FEBS Lett. 1987;211(2):169-174.
22. Kreymann B, Williams G, Ghatei MA, Bloom SR. Glucagon-
like peptide-1 7-36: a physiological incretin in man. Lancet.
1987;2(8571):1300-1304.
Endocrinology, 2021, Vol. 162, No. 7
23. Hansotia T, Baggio LL, Delmeire D, et al. Double incretin re-
ceptor knockout (DIRKO) mice reveal an essential role for the
enteroinsular axis in transducing the glucoregulatory actions of
DPP-IV inhibitors. Diabetes. 2004;53(5):1326-1335.
24. Pacini G, Thomaseth K, Ahrén B. Contribution to glucose toler-
ance of insulin-independent vs. insulin-dependent mechanisms
in mice. Am J Physiol Endocrinol Metab. 2001;281(4):E693
-E703.
25. Ahrén B, Pacini G. Glucose effectiveness: lessons from studies
on insulin-independent glucose clearance in mice. J Diabetes
Investig. Published online October 23, 2020. doi:10.1111/
jdi.13446
26. Hare KJ, Vilsbøll T, Holst JJ, Knop FK. Inappropriate glucagon
Downloaded from https://academic.oup.com/endo/article/162/7/bqab065/6199910 by guest on 23 January 2023
response after oral compared with isoglycemic intravenous
glucose administration in patients with type 1 diabetes. Am J
Physiol Endocrinol Metab. 2010;298(4):E832-E837.
27. Nauck MA, Homberger E, Siegel EG, et al. Incretin effects of
increasing glucose loads in man calculated from venous in-
sulin and C-peptide responses. J Clin Endocrinol Metab.
1986;63(2):492-498.
28. Bagger JI, Knop FK, Lund A, Vestergaard H, Holst JJ, Vilsbøll T.
Impaired regulation of the incretin effect in patients with type 2
diabetes. J Clin Endocrinol Metab. 2011;96(3):737-745.
29. Madsbad S, Kehlet H, Hilsted J, Tronier B. Discrepancy between
plasma C-peptide and insulin response to oral and intravenous
glucose. Diabetes. 1983;32(5):436-438.
30. Horowitz M, Edelbroek MA, Wishart JM, Straathof JW.
Relationship between oral glucose tolerance and gastric emptying
in normal healthy subjects. Diabetologia. 1993;36(9):857-862.
31. Holst JJ, Gribble F, Horowitz M, Rayner CK. Roles of the gut in
glucose homeostasis. Diabetes Care. 2016;39(6):884-892.
32. Jørgensen NB, Jacobsen SH, Dirksen C, et al. Acute and
long-term effects of Roux-en-Y gastric bypass on glucose me-
tabolism in subjects with Type 2 diabetes and normal glucose
tolerance. Am J Physiol Endocrinol Metab. 2012;303(1):E122
-E131.
33. Tolhurst G, Reimann F, Gribble FM. Nutritional regu-
lation of glucagon-like peptide-1 secretion. J Physiol.
2009;587(1):27-32.
34. Kuhre RE, Frost CR, Svendsen B, Holst JJ. Molecular mechan-
isms of glucose-stimulated GLP-1 secretion from perfused rat
small intestine. Diabetes. 2015;64(2):370-382.
35. Færch K, Torekov SS, Vistisen D, et al. GLP-1 response to oral
glucose is reduced in prediabetes, screen-detected type 2 dia-
betes, and obesity and influenced by sex: the ADDITION-PRO
study. Diabetes. 2015;64(7):2513-2525.
36. Nauck MA, Vardarli I, Deacon CF, Holst JJ, Meier JJ. Secretion
of glucagon-like peptide-1 (GLP-1) in type 2 diabetes: what is
up, what is down? Diabetologia. 2011;54(1):10-18.
37. Holst JJ, Knop FK, Vilsbøll T, Krarup T, Madsbad S.
Loss of incretin effect is a specific, important, and
early characteristic of type 2 diabetes. Diabetes Care.
2011;34(Suppl 2):S251-S257.
38. Knop FK, Vilsbøll T, Højberg PV, et al. Reduced incretin effect
in type 2 diabetes: cause or consequence of the diabetic state?
Diabetes. 2007;56(8):1951-1959.
39. Eng J, Kleinman WA, Singh L, Singh G, Raufman JP. Isolation
and characterization of exendin-4, an exendin-3 analogue, from
Heloderma suspectum venom. Further evidence for an exendin
Endocrinology, 2021, Vol. 162, No. 7
receptor on dispersed acini from guinea pig pancreas. J Biol
Chem. 1992;267(11):7402-7405.
40. Raufman JP, Singh L, Eng J. Exendin-3, a novel peptide from
Heloderma horridum venom, interacts with vasoactive intestinal
peptide receptors and a newly described receptor on dispersed
acini from guinea pig pancreas. Description of exendin-3(9-39)
amide, a specific exendin receptor antagonist. J Biol Chem.
1991;266(5):2897-2902.
41. Thorens B, Porret A, Bühler L, Deng SP, Morel P, Widmann C.
Cloning and functional expression of the human islet GLP-1
receptor. Demonstration that exendin-4 is an agonist and
exendin-(9-39) an antagonist of the receptor. Diabetes.
1993;42(11):1678-1682.
42. Chen YE, Drucker DJ. Tissue-specific expression of unique
mRNAs that encode proglucagon-derived peptides or exendin
4 in the lizard. J Biol Chem. 1997;272(7):4108-4115.
43. Schirra J, Sturm K, Leicht P, Arnold R, Göke B, Katschinski M.
Exendin(9-39)amide is an antagonist of glucagon-
like peptide-1(7-36)amide in humans. J Clin Invest.
1998;101(7):1421-1430.
44. Edwards CM, Todd JF, Mahmoudi M, et al. Glucagon-like pep-
tide 1 has a physiological role in the control of postprandial
glucose in humans: studies with the antagonist exendin 9-39.
Diabetes. 1999;48(1):86-93.
45. Elashoff M, Matveyenko AV, Gier B, Elashoff R, Butler PC.
Pancreatitis, pancreatic, and thyroid cancer with glucagon-like
peptide-1-based therapies. Gastroenterology. 2011;141(1):150-156.
46. Ahmad SR, Swann J. Exenatide and rare adverse events. N Engl
J Med. 2008;358(18):1970-1972.
47. Nauck MA. A critical analysis of the clinical use of incretin-
based therapies: the benefits by far outweigh the potential risks.
Diabetes Care. 2013;36(7):2126-2132.
48. Salehi M, Vahl TP, D’Alessio DA. Regulation of islet hormone
release and gastric emptying by endogenous glucagon-like
peptide 1 after glucose ingestion. J Clin Endocrinol Metab.
2008;93(12):4909-4916.
49. Adriaenssens AE, Svendsen B, Lam BY, et al. Transcriptomic
profiling of pancreatic alpha, beta and delta cell populations
identifies delta cells as a principal target for ghrelin in mouse
islets. Diabetologia. 2016;59(10):2156-2165.
50. Xu SFS, Andersen DB, Izarzugaza JMG, Kuhre RE, Holst JJ. In
the rat pancreas, somatostatin tonically inhibits glucagon secre-
tion and is required for glucose-induced inhibition of glucagon
secretion. Acta Physiol (Oxf). 2020;229(3):e13464.
51. Kuhre RE, Christiansen CB, Ghiasi SM, et al. Neuromedin U
does not act as a decretin in rats. Cell Metab. 2019;29(3):719-
726.e5.
52. Svendsen B, Larsen O, Gabe MBN, et al. Insulin secre-
tion depends on intra-islet glucagon signaling. Cell Rep.
2018;25(5):1127-1134.e2.
53. Serre V, Dolci W, Schaerer E, et al. Exendin-(9-39) is an in-
verse agonist of the murine glucagon-like peptide-1 re-
ceptor: implications for basal intracellular cyclic adenosine
3′,5′-monophosphate levels and β-cell glucose competence.
Endocrinology. 1998;139(11):4448-4454.
54. Svendsen B, Holst JJ. Paracrine regulation of somatostatin se-
cretion by insulin and glucagon in mouse pancreatic islets.
Diabetologia. 2021;64(1):142-151.
9
55. Jørgensen NB, Dirksen C, Bojsen-Møller KN, et al. Exaggerated
glucagon-like peptide 1 response is important for improved
β-cell function and glucose tolerance after Roux-en-Y gas-
tric bypass in patients with type 2 diabetes. Diabetes.
2013;62(9):3044-3052.
56. Jepsen SL, Grunddal KV, Wewer Albrechtsen NJ, et al. Paracrine
crosstalk between intestinal L- and D-cells controls secretion
of glucagon-like peptide-1 in mice. Am J Physiol Endocrinol
Metab. 2019;317(6):E1081-E1093.
57. Gasbjerg LS, Helsted MM, Hartmann B, et al. Separate and
combined glucometabolic effects of endogenous glucose-
dependent insulinotropic polypeptide and glucagon-like peptide
1 in healthy individuals. Diabetes. 2019;68(5):906-917.
Downloaded from https://academic.oup.com/endo/article/162/7/bqab065/6199910 by guest on 23 January 2023
58. Hansen LS, Sparre-Ulrich AH, Christensen M, et al.
N-terminally and C-terminally truncated forms of glucose-
dependent insulinotropic polypeptide are high-affinity competi-
tive antagonists of the human GIP receptor. Br J Pharmacol.
2016;173(5):826-838.
59. Sparre-Ulrich AH, Gabe MN, Gasbjerg LS, et al. GIP(3-30)NH2
is a potent competitive antagonist of the GIP receptor and ef-
fectively inhibits GIP-mediated insulin, glucagon, and somato-
statin release. Biochem Pharmacol. 2017;131:78-88.
60. Yanagimachi T, Fujita Y, Takeda Y, et al. Pancreatic glucose-
dependent insulinotropic polypeptide (GIP) (1-30) expression is
upregulated in diabetes and PEGylated GIP(1-30) can suppress
the progression of low-dose-STZ-induced hyperglycaemia in
mice. Diabetologia. 2016;59(3):533-541.
61. Asmar M, Asmar A, Simonsen L, et al. The gluco- and
liporegulatory and vasodilatory effects of glucose-
dependent insulinotropic polypeptide (GIP) are abolished
by an antagonist of the human GIP receptor. Diabetes.
2017;66(9):2363-2371.
62. Gasbjerg LS, Christensen MB, Hartmann B, et al. GIP(3-30)
NH2 is an efficacious GIP receptor antagonist in humans: a ran-
domised, double-blinded, placebo-controlled, crossover study.
Diabetologia. 2018;61(2):413-423.
63. Nauck MA, Meier JJ. GIP and GLP-1: stepsiblings rather
than monozygotic twins within the incretin family. Diabetes.
2019;68(5):897-900.
64. Gasbjerg LS, Bergmann NC, Stensen S, et al. Evaluation of the
incretin effect in humans using GIP and GLP-1 receptor antag-
onists. Peptides. 2020;125:170183.
65. Højberg PV, Zander M, Vilsbøll T, et al. Near normalisation
of blood glucose improves the potentiating effect of GLP-1 on
glucose-induced insulin secretion in patients with type 2 dia-
betes. Diabetologia. 2008;51(4):632-640.
66. Stensen S, Gasbjerg LS, Krogh LL, et al. Effects of endogenous
GIP in patients with type 2 diabetes. Eur J Endocrinol.
2021;EJE-21-0135.R1.
67. Aulinger BA, Bedorf A, Kutscherauer G, et al. Defining the role
of GLP-1 in the enteroinsulinar axis in type 2 diabetes using
DPP-4 inhibition and GLP-1 receptor blockade. Diabetes.
2014;63(3):1079-1092.
68. Vilsbøll T, Knop FK, Krarup T, et al. The pathophysiology of
diabetes involves a defective amplification of the late-phase in-
sulin response to glucose by glucose-dependent insulinotropic
polypeptide-regardless of etiology and phenotype. J Clin
Endocrinol Metab. 2003;88(10):4897-4903.
10 
69. Kjems LL, Holst JJ, Vølund A, Madsbad S. The influence of
GLP-1 on glucose-stimulated insulin secretion: effects on beta-
cell sensitivity in type 2 and nondiabetic subjects. Diabetes.
2003;52(2):380-386.
70. Nauck MA, Kleine N, Orskov C, Holst JJ, Willms B, Creutzfeldt W.
Normalization of fasting hyperglycaemia by exogenous glucagon-
like peptide 1 (7-36 amide) in type 2 (non-insulin-dependent) dia-
betic patients. Diabetologia. 1993;36(8):741-744.
71. Jones AG, McDonald TJ, Shields BM, et al; PRIBA Study
Group. Markers of β-cell failure predict poor glycemic response
to GLP-1 receptor agonist therapy in type 2 diabetes. Diabetes
Care. 2016;39(2):250-257.
72. Toft-Nielsen MB, Madsbad S, Holst JJ. Determinants of the ef-
fectiveness of glucagon-like peptide-1 in type 2 diabetes. J Clin
Endocrinol Metab. 2001;86(8):3853-3860.
73. Billings LK, Doshi A, Gouet D, et al. Efficacy and safety of
IDegLira versus basal-bolus insulin therapy in patients with
type 2 diabetes uncontrolled on metformin and basal in-
sulin: the DUAL VII randomized clinical trial. Diabetes Care.
2018;41(5):1009-1016.
Endocrinology, 2021, Vol. 162, No. 7
74. Højberg PV,Vilsbøll T, Rabøl R, et al. Four weeks of near-normalisation
of blood glucose improves the insulin response to glucagon-like pep-
tide-1 and glucose-dependent insulinotropic polypeptide in patients
with type 2 diabetes. Diabetologia. 2009;52(2):199-207.
75. Frias JP, Nauck MA, Van J, et al. Efficacy and safety of
LY3298176, a novel dual GIP and GLP-1 receptor agonist, in
patients with type 2 diabetes: a randomised, placebo-controlled
and active comparator-controlled phase 2 trial. Lancet.
2018;392(10160):2180-2193.
76. Killion EA, Wang J, Yie J, et al. Anti-obesity effects
of GIPR antagonists alone and in combination with
GLP-1R agonists in preclinical models. Sci Transl Med
2018;10(472):eaat3392.
Downloaded from https://academic.oup.com/endo/article/162/7/bqab065/6199910 by guest on 23 January 2023
77. Willard FS, Douros JD, Gabe MB, et al. Tirzepatide is an
imbalanced and biased dual GIP and GLP-1 receptor agonist.
JCI Insight 2020;5(17):e140532.
78. Min T, Bain SC. The role of tirzepatide, dual GIP and
GLP-1 receptor agonist, in the management of type 2
diabetes: the SURPASS clinical trials. Diabetes Ther.
2021;12(1):143-157.