Elimination of the Action of Glucagon-like Peptide 1 Causes an Impairment of
`Glucose Tolerance after Nutrient Ingestion by Healthy Baboons
`David A. D’Alessio, Robin Vogel,* Ron Prigeon, Ellen Laschansky, Donna Koerker,
` John Eng, and John W. Ensinck
`‡
`Division of Endocrinology, Metabolism and Nutrition, Departments of Medicine and Physiology, University of Washington,
`‡
`*
`Veterans Affairs Medical Center, Seattle, Washington 98195; and
`Veterans Affairs Medical Center, New York 10468
`
`Abstract
`
`Introduction
`
`Glucagon-like peptide 1 (GLP-1) is an insulinotropic hor-
`mone released after nutrient ingestion which is known to
`augment insulin secretion, inhibit glucagon release, and
`promote insulin-independent glucose disposition. To deter-
`mine the overall effect of GLP-1 on glucose disposition after
`a meal we studied a group of healthy, conscious baboons be-
`fore and after intragastric glucose administration during in-
`fusions of saline, and two treatments to eliminate the action
`
`a) exendin-[9-39] (Ex-9), a peptide receptor an-
`of GLP-1: (
`
`b) an anti–GLP-1 mAb. Fasting con-
`tagonist of GLP-1; or (
`centrations of glucose were higher during infusion of Ex-9
`⫾
`⫾
`⬍
`0.05 mM, P
`0.05 vs. 4.16
`
` 0.01),
`than during saline (4.44
`coincident with an elevation in the levels of circulating glu-
`⫾
`⫾
`⬍
`
`10 vs. 59
`3 ng/liter,
`P
` 0.02). The postpran-
`cagon (96
`dial glycemic excursions during administration of Ex-9 and
`mAb were greater than during the control studies (Ex-9
`⫾
`⫾
`⫽
`
`2.0 vs. saline 10.0
`0.8 mM,
`P
` 0.07; and mAb
`13.7
`⫾
`⫾
`⫽
`1.2 vs. saline 10.6
`0.9 mM,
`
` 0.044). The incre-
`P
`13.6
`ments in insulin levels throughout the absorption of the glu-
`cose meal were not different for the experimental and con-
`trol conditions, but the insulin response in the first 30 min
`after the glucose meal was diminished significantly during
`⫾
`⫾
`139 vs. saline 1,089
`166
`treatment with Ex-9 (Ex-9 761
`⫽
`
`P
` 0.044) and was delayed in three of the four animals
`pM,
`⫾
`262 vs. saline
`given the neutralizing antibody (mAb 946
`⫾
`340 pM). Thus, elimination of the action of GLP-1
`1,146
`impaired the disposition of an intragastric glucose meal and
`this was at least partly attributable to diminished early insu-
`lin release. In addition to these postprandial effects, the
`concurrent elevation in fasting glucose and glucagon during
`GLP-1 antagonism suggests that GLP-1 may have a tonic
`inhibitory effect on glucagon output. These findings demon-
`strate the important role of GLP-1 in the assimilation of
`J. Clin. Invest.
`1996. 97:
`glucose absorbed from the gut. (
` enteroinsular axis
`133–138.) Key words: incretin hormone
`•
`
`
`
`glucose tolerance
`insulin secretion
`glucagon
`•
`•
`•
`
`Portions of this work were presented as abstracts at the American Di-
`abetes Association Scientific Sessions on 11–14 June 1994 in New Or-
`leans, LA and at the 10th International Symposium on Gastrointesti-
`nal Hormones on 27–31 August 1994 in Santa Barbara, CA.
`Address correspondence to David D’Alessio, M.D., Dept. of
`Medicine, RC-14, 1959 Pacific Ave., University of Washington, Seat-
`tle, WA 98195. Phone: 206-548-4703; FAX: 206-548-6987; E-mail:
`dalessio@u.washington.edu
`Received for publication 7 April 1995 and accepted in revised form
`13 September 1995.
`
`The Journal of Clinical Investigation
`Volume 97, Number 1, January 1996, 133–138
`
`1
`Glucagon-like peptide 1(7-36)amide (GLP-1),
` synthesized in
`and released from the mammalian intestinal tract, has a variety
`of actions consistent with a role in the regulation of carbohy-
`drate metabolism (1, 2). GLP-1 is a potent insulinotropin (3,
`4), and because its release is stimulated by ingested nutrients
`(5, 6), it has been proposed as a mediator of the incretin effect,
`namely the augmented insulin release by oral as compared
`with intravenous glucose (7–9). In addition, GLP-1 decreases
`glucagon secretion in vitro and in vivo (10, 11). These findings
`suggest that GLP-1 coordinates hormone secretion from the
`pancreatic islet in a manner that favors glucose anabolism.
`Studies in diabetic and healthy humans indicate that GLP-1
`also promotes glucose disposition independent of islet hor-
`mone secretion (12, 13). In this context, transcription of the
`GLP-1 receptor gene in several nonislet tissues, including mus-
`cle, adipose tissue, and liver (14, 15), makes it plausible that
`GLP-1 has direct effects on glucose uptake and/or glucose pro-
`duction. Recent in vitro work demonstrating that GLP-1 in-
`creases glycogen synthesis in isolated hepatocytes and skeletal
`muscle supports this hypothesis (16, 17).
`Although current information suggests that GLP-1 acts at
`multiple sites, the integrated effect of GLP-1 on carbohydrate
`metabolism has not been determined. This issue is important
`both in understanding normal and pathologic fuel metabolism.
`One of the classic paradigms of endocrinology has been the
`deduction of physiologic actions from the abnormalities ob-
`served in naturally occurring or induced hormone deficiency
`states. Recently, a GLP-1 receptor antagonist, exendin-[9-39]
`(Ex-9), has been shown to specifically block the actions of
`GLP-1 in tissue culture systems (18, 19). Two groups have
`used this peptide to inhibit the biologic effects of GLP-1 in rats
`before and for short periods after intraduodenal or oral glu-
`cose loads (20, 21). Insulin secretion was decreased and post-
`prandial glycemia increased 30–45 min after the glucose load
`during treatment with Ex-9. Contemporaneous with these
`studies, we had independently initiated experiments to create
`functional deficiencies of GLP-1 in healthy conscious baboons
`to evaluate the role of this peptide in normal fuel homeostasis.
`We report herein the separate use of Ex-9 to block GLP-1 re-
`ceptors, and a specific monoclonal antibody to immunoneu-
`tralize circulating GLP-1, to characterize the effects of elimina-
`tion of this hormone on oral glucose tolerance.
`
`, acute insulin response to
` AIR
`1.
`Abbreviations used in this paper:
`g
`glucose; Ex-9, exendin-[9-39]; GLP-1, glucagon-like peptide 1;
`IVGTT, intravenous glucose tolerance test;
`, glucose disappearance
`k
`g
`constant; S
`, glucose effectiveness; S
`, insulin sensitivity index.
`G
`I
`
`Glucagon-like Peptide 1 and Glucose Tolerance
`
`133
`
`Downloaded from http://www.jci.org on May 21, 2015. http://dx.doi.org/10.1172/JCI118380
`
`SANOFI-AVENTIS Exhibit 1020 - Page 133
`
`IPR for Patent No. 8,951,962
`
`

`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`Methods
`
`1(1-36)amide (27), and the principal metabolite GLP-1(9-36)amide
`(28, 29). Ex-9 did not cross-react with either of the antisera used in
`the glucagon and GLP-1 RIAs. During the studies with the anti–
`GLP-1 mAb, plasma was extracted immediately into 70% ethanol to
`remove the circulating GLP-1–Ab complexes and to enable measure-
`ment of free GLP-1.
`Fasting values were computed as the mean of the
`Data analyses.
`21 premeal samples obtained in the Ex-9 and saline studies, and the 6
`premeal samples taken after the bolus administration of mAb 26.1.
`The postprandial responses were defined by the incremental areas,
`above fasting levels, under the curves of glucose and insulin after the
`intragastric meal. These areas were calculated for each study by the
`trapezoidal method using values greater than fasting throughout
`the 180 min after intragastric glucose to calculate differences. Insulin
`increments after the glucose meal were also separated into 0–30- and
`0–180-min time intervals to derive estimates of early and total insulin
`secretion throughout the meal disposition. The acute insulin response
`) was taken as the average incremental rise in insulin
`to glucose (AIR
`g
`in the samples obtained from 2–10 min after the intravenous glucose
`bolus. The glucose disappearance constant (
`) was calculated by de-
`k
`g
`termining the slope of the natural logarithm of the glucose values at
`10, 12, 14, 16, and 19 min after the intravenous glucose. Insulin and
`glucose values obtained during the IVGTT were analyzed using the
`minimal model of glucose kinetics (30) to obtain the insulin sensitiv-
`ity index (S
`) and glucose effectiveness (S
`).
`I
`G
`The data from the multiple Ex-9 and saline studies were averaged
`for each animal. Comparisons of fasting, postprandial, and post-
`IVGTT values for the five animals receiving Ex-9 and the four ani-
`mals given mAb 26.1 were made with their control (saline) studies us-
`ing the
` test for paired samples and the Wilcoxon signed ranks test;
`t
`two-tailed analyses are reported in all cases. ANOVA for repeated
`measures was used to analyze GLP-1 levels before and after intragas-
`tric glucose in the different experimental groups. Data are presented
`as mean
`SEM.
`⫾
`
`Results
`
`Fasting concentrations of GLP-1 were similar on the days the
`baboons received Ex-9 or saline/HSA (5.0
`0.5 vs. 6.1
`0.5
`⫾
`⫾
`
`Figure 1. Mean concentrations⫾SEM of circulating GLP-1 before
`and after an intragastric glucose meal in five baboons treated with in-
`travenous Ex-9 (boxes) or saline (circles). *P ⬍ 0.05 compared with
`fasting.
`
`Ex-9 was synthesized by solid phase methods and purified
`Materials.
`by HPLC (18). Authenticity was determined by amino acid sequenc-
`ing. A monoclonal antibody designated mAb 26.1 was generated
`from a hybridoma cell line that was a gift from Scios Nova Inc.
`(Mountain View, CA). This antibody specifically recognizes an
`epitope at the NH
` terminus of GLP-1(7-36)amide (Jan Scardina,
`2
`Scios Nova, personal communication). The immunoglobulin was par-
`tially purified by (NH
`)
`SO
` precipitation, and the anti–GLP-1 anti-
`4
`2
`4
`body titer was determined. The GLP-1–immunoneutralizing capacity
`of mAb 26.1 was tested in rat islet monolayer cultures. 50
`g mAb
`␮
`8
`⫺
`26.1 abolished the
` cell stimulatory effect of 10
` M GLP-1.
`␤
`Five healthy male baboons, weighing 10–21 kg, were
`Animals.
`anesthetized with ketamine and halothane, and catheters were intro-
`duced into their femoral or subclavian venous system and into their
`stomach. The proximal ends of these catheters were tunneled subcu-
`taneously and exited the skin on the animal’s upper backs. The ba-
`boons were placed in jackets, and the catheters were passed through a
`tether which attached to the top of each animal’s cage (1
` 1
` 2 m).
`⫻
`⫻
`The catheters eventuated in a swivel apparatus where the venous
`lines were connected to 150 mM saline and were accessed for blood
`sampling, and the gastric line was capped and available for infusions.
`This system permitted repeated studies in the baboons while they
`were awake and moving about their cages. The cages were housed in
`a room in the Regional Primate Research Center at the University of
`Washington. Animal care was provided by staff veterinarians and
`skilled animal technicians. Animal health was monitored by behavior,
`food intake, and intermittent measurements of electrolytes, blood
`cell counts, and blood cultures. The protocols were approved by the
`Institutional Animal Welfare Committee.
`Animals were studied after an overnight fast. In ran-
`Protocols.
`dom order and on separate days, five baboons received continuous
`infusions of: (
`) Ex-9 (150 nmol/kg/h) in 150 mM saline/0.1% HSA;
`a
`or (
`) saline/0.1% HSA, for the duration of the 450-min experiment.
`b
`Three animals had the Ex-9 and the saline studies repeated three
`times each, and two animals twice each. At least 3 wk after the com-
`pletion of the experiments with Ex-9 and saline, four of the animals
`were studied after receiving a 220-mg bolus of the anti–GLP-1 mAb;
`the fifth baboon had died of a surgical complication and could not be
`used in this protocol. Starting 20 min after the initiation of each treat-
`ment (Ex-9, mAb, or saline), basal blood samples were drawn every
`2 min from
`40 to 0 min. At 0 min an intragastric infusion of glucose
`⫺
`(1.5 grams/kg) and
`-xylose (0.5 gram) was given over 5 min. Post-
`prandial blood samples were withdrawn periodically over the next
`200 min. At 210 min a frequently sampled intravenous glucose toler-
`ance test (IVGTT) was performed. The baboons received an intrave-
`nous infusion of glucose (3 grams/kg) followed 20 min later by an in-
`travenous bolus of tolbutamide (125 mg/kg) and blood sampled for
`210–390 min as previously described (22). The duration of each ex-
`periment was 7.5 h, and 90–100 ml of blood was removed from the an-
`imals on each day of study. The experiments were separated by at
`least 1 wk and no more than three experiments were performed on
`one animal per month.
`Blood was collected into tubes containing hep-
`Plasma analyses.
`arin for analysis of glucose, insulin, and
`-xylose, a benzamidine-
`based antiproteolytic cocktail (23) for glucagon measurement, and
`0.5 M EDTA/500 KIU/ml aprotinin for assay of GLP-1. Samples
`were centrifuged immediately, and the plasma was removed and
`stored at
`20
`C. Plasma glucose was measured using a glucose oxi-
`⫺
`⬚
`dase method, and
`-xylose by a colorimetric assay (24). Insulin and
`glucagon values were measured by previously described RIA (23, 25).
`GLP-1 was measured by RIA using antiserum 89390 (kindly provided
`by Dr. Jens Juul Holst, Paanum Institute, Copenhagen, Denmark), as
`described by Orskov et al. (26, 27), from ethanol extracts of plasma.
`Antiserum 89390 recognizes the COOH-terminal amidated arginine
`of GLP-1–related peptides and thus recognizes GLP-1(7-36)amide,
`the predominant bioactive moiety (27), but also the precursor GLP-
`
`d
`
`d
`
`d
`
`134
`
`D’Alessio, Vogel, Prigeon, Laschansky, Koerker, Eng, and Ensinck
`
`Downloaded from http://www.jci.org on May 21, 2015. http://dx.doi.org/10.1172/JCI118380
`
`SANOFI-AVENTIS Exhibit 1020 - Page 134
`
`IPR for Patent No. 8,951,962
`
`

`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`Table I. Fasting Concentrations of Insulin, Glucose, and
`Glucagon in Studies with Ex-9 and Saline Infusions
`
`Glucose
`
`mM
`
`4.16
`0.05
`⫾
`4.44
`0.05*
`⫾
`
`Insulin
`
`pM
`
`126
`⫾
`102
`⫾
`
`12
`12
`
`Glucagon
`
`ng/liter
`
`59
`⫾
`96
`⫾
`
`3
`10*
`
`Saline
`Ex-9
`
`Results are expressed as mean
`SEM. *
`⫾
`of studies.
`
`
`P
`
`⬍
`
`0.02 vs. saline.
`
`
`n
`
` 5 pairs
`⫽
`
`pM). GLP-1 levels rose promptly after intragastric glucose in
`both sets of studies, remained significantly elevated for 90 min
`after the meal, and returned to preprandial concentrations by
`120 min (Fig. 1). There were no differences in GLP-1 levels be-
`tween the Ex-9 and saline experiments before or after glucose
`administration. Fasting GLP-1 concentrations in the four ani-
`mals given mAb were slightly, but not statistically, higher than
`those during the saline studies in these animals (9.7
`1.2 vs.
`⫾
`7.0
`1.1,
`
` 0.125). Postprandial GLP-1 levels were not dif-
`⫾
`⫽
`P
`ferent than fasting after administration of mAb, with changes
`of 2.0
`2.6,
`0.4
`1.9, and 0.1
`1.8 pM at 30, 60, and 90 min,
`⫾
`⫺
`⫾
`⫾
`
`respectively. In contrast, GLP-1 concentrations in these ba-
`boons after intragastric glucose were significantly higher than
`fasting levels at 30, 60, and 90 min during saline infusion, with
`postprandial increments of 3.6
`1.1, 2.5
`0.8, and 2.8
`1.1 pM
`⫾
`⫾
`⫾
`(
`
` 0.05 at each time point). Thus, while there was a signifi-
`⬍
`P
`cant postprandial increment in GLP-1 with saline infusions,
`there was no detectable change in plasma GLP-1 in response
`to the glucose meal after mAb. This suggests that bioactive
`GLP-1(7-36)amide, the GLP-1 species released by nutrient
`stimulation (27), was removed from the circulation by the
`mAb.
`During the infusion of Ex-9, mean fasting glucose concen-
`trations were significantly higher in comparison with the controls
`(Table I). In addition, fasting glucagon levels were elevated
`when Ex-9 was infused. The fasting insulin concentrations
`were not different during the Ex-9 and control studies. Fasting
`insulin and glucose concentrations did not differ before and af-
`ter the administration of mAb 26.1 to four baboons; glucagon
`concentrations were not measured in these studies.
`After the glucose meal in the Ex-9 and saline studies the
`plasma glucose profiles rose to maximal levels in the first 40
`120 min (Fig. 2
`).
`min and returned to fasting values over
`A
`The postprandial glucose response was greater when the ani-
`mals received Ex-9 compared with controls (Table II). Insulin
`
`ⵑ
`
`Figure 2. Mean levels⫾SEM of
`plasma glucose before and after
`a glucose meal in five baboons
`infused with saline (circles) or
`Ex-9 (boxes) (A) and in four ba-
`boons given anti–GLP-1 mAb
`26.1 (boxes) or saline (circles)
`(B).
`
`Figure 3. Mean plasma insulin
`concentrations⫾ SEM before
`and after a glucose meal in five
`baboons treated with either Ex-9
`(boxes) or saline (circles) (A)
`and in four baboons given anti–
`GLP-1 mAb 26.1 (boxes) or sa-
`line (circles) (B).
`
`Glucagon-like Peptide 1 and Glucose Tolerance
`
`135
`
`Downloaded from http://www.jci.org on May 21, 2015. http://dx.doi.org/10.1172/JCI118380
`
`SANOFI-AVENTIS Exhibit 1020 - Page 135
`
`IPR for Patent No. 8,951,962
`
`

`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`Table II. Postprandial Increments of Glucose and Insulin in
`Studies with Ex-9, mAb 26.1, and Controls (Saline)
`
`Table III. Parameters of Intravenous Glucose Tolerance in
`Five Baboons Studied with and without Ex-9 and Four
`Baboons Studied with and without Anti–GLP-1 Ab (mAb)
`
`Glucose
`
`Insulin (0–30)
`
`Insulin (0–180)
`
`mM
`
`pM/30 min
`
`pM/180 min
`
`Ex-9
`Control
`mAb
`Control
`
`13.7⫾2.0*
`10.0⫾0.8
`13.6⫾1.2‡
`10.6⫾0.9
`
`761⫾139‡
`1132⫾272
`946⫾262
`1146⫾340
`
`4270⫾1337
`4952⫾1528
`4302⫾1254
`3648⫾1002
`
`kg
`
`%/min
`
`AIRg
`
`pM
`
`SI
`
`SG
`
`⫻10⫺5/min/pM
`
`⫻10⫺2/min
`
`Ex-9
`Saline
`mAb
`Saline
`
`2.45⫾0.03
`2.54⫾0.1
`2.69⫾0.3
`2.68⫾0.1
`
`48.6⫾5.7
`49.4⫾5.9
`46.5⫾12.1
`49.5⫾7.0
`
`13.5⫾6.8
`9.1⫾4.1
`10.4⫾3.8
`10.0⫾5.2
`
`3.9⫾0.8
`4.7⫾0.7
`4.2⫾0.6
`4.5⫾0.3
`
`*P ⫽ 0.07, ‡P ⬍ 0.05 compared with controls.
`
`Results are expressed as mean⫾SEM.
`
`secretion followed a similar time course to glucose after the
`), but in contrast to the differences in glycemic
`meal (Fig. 3
`A
`responses, the total incremental changes in insulin concentra-
`tion were similar on the days the animals got Ex-9 and saline
`(Table II). However, the rise in insulin secretion was delayed
`when Ex-9 was administered as reflected in the smaller insulin
`increment in the first 30 min after the glucose meal (Table II).
`After administration of the intragastric glucose in the
`paired mAb and control experiments, plasma glucose rose to
`peak levels within 30–40 min and returned to basal within 70
`). The glycemic response was
`35% greater after
`min (Fig. 2
`B
`immunoneutralization of GLP-1 than during the matched con-
`trol experiments (Table II). Corresponding with this accentu-
`ated glycemic profile, the total insulin response was higher, but
`not significantly different, in the mAb versus control (Fig. 3
`B
`and Table II). However, insulin release in the first 30 min after
`intragastric glucose was lower in three of the four baboons on
`the days they were immunoneutralized compared with control
`(Table II); the fourth animal had a very rapid rise in prandial
`glucose and a delayed insulin response was not detectable.
`There were no differences in the plasma profiles of
`-xylose
`
`ⵑ
`
`d
`
`Figure 4. Plasma concentrations of d-xylose⫾SEM after an intragas-
`tric glucose/d-xylose meal in baboons during treatment with saline
`(circles), Ex-9 (boxes), or anti–GLP-1 mAb 26.1 (triangles).
`
`between the Ex-9/control and mAb 26.1/control studies (Fig.
`4), indicating that intestinal absorption of glucose was not sig-
`nificantly affected by the experimental conditions.
`The intravenous glucose tolerance (
`kg) was not different
`from controls during the infusion of Ex-9 or after the adminis-
`tration of the monoclonal antibody (Table III). Similarly,
`AIRg, SI, and SG did not differ among the experimental and
`control situations.
`
`Discussion
`
`These data are in keeping with an important role for GLP-1 in
`the disposition of glucose after eating. There were significant
`increases in postprandial glycemia during both competitive an-
`tagonism of GLP-1 receptors with Ex-9 and after immunoneu-
`tralization with mAb 26.1. We infer that circulating GLP-1-
`(7-36)amide was neutralized in the circulation by the mAb
`because it is the primary secreted GLP-1 peptide (27), and
`there was no change in the postprandial GLP immunoreactiv-
`ity in these experiments. The presence of fasting amounts of
`GLP-1 immunoreactivity in the circulation after mAb treat-
`ment is likely due to other GLP-1 species, such as GLP-1(9-36)
`(28, 29), which are not recognized by mAb 26.1. We treated
`the animals with amounts of Ex-9 that were estimated to give a
`significant excess relative to circulating GLP-1. The concor-
`dance of our results, using two separate means to create func-
`tional GLP-1 deficiency, supports the conclusion that post-
`prandial release of GLP-1 is an important factor regulating
`glucose disposition after nutrient ingestion. Moreover, the sim-
`ilarities in primate physiology make it likely that these obser-
`vations are applicable to humans.
`The alteration in intragastric glucose tolerance seen with
`both experimental methods to achieve functional GLP-1 defi-
`ciency could have been due to blunting of the insulinotropic ef-
`fect of GLP-1 (3, 4), impairment of its extraislet action (12, 13),
`or a combination of the two. Based on this study it is not possi-
`ble to distinguish between the relative contributions of these
`two putative mechanisms. Despite the greater glycemia with
`both Ex-9 and mAb, there were no quantitative differences in
`total postprandial insulin output between these experimental
`conditions and controls. It is likely that the comparable abso-
`lute amounts of insulin secreted reflect a compensatory ␤ cell
`secretory response to the higher glucose levels occurring when
`GLP-1 activity was inhibited. Thus, in the absence of incretin-
`stimulated insulin secretion, postprandial glucose levels rose
`excessively, driving insulin release. This was best seen during
`immunoneutralization (Figs. 2 B and 3 B) when an accentu-
`
`136
`
`D’Alessio, Vogel, Prigeon, Laschansky, Koerker, Eng, and Ensinck
`
`Downloaded from http://www.jci.org on May 21, 2015. http://dx.doi.org/10.1172/JCI118380
`
`SANOFI-AVENTIS Exhibit 1020 - Page 136
`
`IPR for Patent No. 8,951,962
`
`

`
`ated glucose excursion preceded a peak insulin release that
`was actually greater than that observed in the control studies.
`Insulin release was delayed, as reflected by the lower insu-
`lin increments in the 30 min after intragastric glucose, in all
`five of the Ex-9 and three of four mAb experiments relative to
`controls. This is one potential explanation for the worsening of
`glucose tolerance during these studies. Insulin secretion in the
`first 30 min after oral glucose has been related to the acute in-
`sulin response to intravenous glucose (31, 32), and decrements
`in this early insulin response have long been used as a marker
`of insulin secretory abnormalities (32–34). The importance of
`the rate of insulin secretion in the regulation of postmeal gly-
`cemia has been emphasized frequently in the past (34–36) and
`is supported by our present results. Previous demonstrations
`of the rapid release of GLP-1 after meals (6, 37) are consistent
`with its acting in the early phases of nutrient disposition, and,
`taken together with the current findings, emphasize the contri-
`bution of GLP-1 to early insulin release in response to a meal.
`In vitro data indicate that GLP-1 induces glucose competence
`in ␤ cells, heightening their response to increases in ambient
`glucose concentration (38). Thus, as has been reported for
`glucose-dependent insulinotropic polypeptide (GIP), GLP-1
`may lower the threshold for glucose initiation of insulin secre-
`tion (39). It is reasonable to further propose that GLP-1 mini-
`mizes shifts in postprandial glycemia by amplifying the insulin
`response to absorbed nutrient stimuli.
`Ex-9 has been used in two recent studies to block the action
`of GLP-1 during oral or intraduodenal glucose administration
`to rats (20, 21). In both of these reports, competitive antago-
`nism of GLP-1 receptor binding caused an increased glycemic
`excursion in response to the carbohydrate meal and a significant
`reduction in total insulin release. The duration of the measure-
`ments was only for 30 and 45 min postcibum, so the decrease in
`insulin release was consistent with that seen in the 0–30-min
`postmeal in our baboons. It is possible that had these experi-
`ments been extended, the pattern of insulin release and the
`overall effect on glucose tolerance would have been similar to
`those reported here.
`Competitive antagonism of GLP-1 with Ex-9 was associ-
`ated with increased glucagon concentrations in the fasting
`state which coincided with a small increase in basal glucose
`concentrations, suggesting that the primary effect was an alter-
`ation of ␣ cell output. Infusion of GLP-1 in physiologic
`amounts has been shown to decrease glucagon levels during
`hyperglycemic clamps (11), but the glucagonostatic effect of
`GLP-1 has not been shown at fasting concentrations. To our
`knowledge, this is the first demonstration of regulation of the
`islet by basal concentrations of a gastrointestinal hormone,
`and it raises the possibility that secretion of GLP-1 that occurs
`independently of acute nutrient ingestion may modulate re-
`lease of an islet hormone without a concurrent change in sub-
`strate. Because we did not measure postprandial glucagon lev-
`els in these studies, we cannot comment on the possibility that
`augmented postmeal glucagon release contributed to the wors-
`ening of the glucose tolerance seen with the administration of
`Ex-9 and mAb. However, it is plausible that an alteration in
`␣ cell output caused by impaired GLP-1 action may have con-
`tributed to the higher glycemia in these studies.
`There was no detectable effect of either Ex-9 or mAb on
`intravenous glucose tolerance. We have shown previously that,
`in healthy humans, infusion of GLP-1 to achieve either supra-
`physiologic or postprandial levels increases glucose disappear-
`
`ance and glucose effectiveness (13, 40). However, the IVGTTs
`in this study were performed at concentrations of GLP-1 that
`had returned to fasting levels (Fig. 1). Based on these data, it
`seems likely that the insulin-independent effect of GLP-1 to
`promote glucose disposition (12, 13, 40), either by stimulating
`glucose uptake or inhibiting hepatic glucose output, occurs
`only at the higher GLP-1 concentrations present after stimu-
`lated secretion. Furthermore, there was no evidence that Ex-9,
`nor mAb 26.1, has direct effects on insulin secretion or glucose
`disposition.
`Previous work has indicated that GLP-1 may alter gastric
`emptying (41), an action that could contribute to its effect on
`glucose tolerance. It is unlikely that such an effect explains our
`results. First, the peak increase in plasma glucose after the
`meal was nearly identical in the Ex-9, mAb 26.1, and control
`studies, occurring at ⵑ30–40 min postcibum. In addition, the
`plasma levels of d-xylose were not different among the con-
`trol, Ex-9, and mAb conditions, suggesting that passage of the
`liquid meal, reflected in the absorption of this nonmetabolized
`sugar, was not increased when the GLP-1 effect was abolished.
`Despite these observations, we cannot conclude that GLP-1
`does not play a role in gastric motility after eating. It is possi-
`ble that had we interfered with the actions of GLP-1 after oral
`ingestion of a solid meal, and measured gastric motility di-
`rectly, a difference from controls would have been seen.
`Because GLP-1 affects several key sites in carbohydrate
`metabolism, there has been some enthusiasm for the therapeu-
`tic potential of this compound (42, 43). Although the present
`study examines physiologic effects of GLP-1, the results may
`have some applicability to the treatment of diabetes. It has
`been shown previously that subjects with type II diabetes are
`responsive to the insulinotropic action of GLP-1 (42, 43), so
`that delivery of this peptide before a meal might provoke the
`otherwise sluggish insulin response of these patients to in-
`gested carbohydrate. Likewise, the action of GLP-1 to inhibit
`fasting glucagon secretion could be useful to restrain the high
`␣ cell secretion present in persons with type II diabetes (44).
`The data presented here also raise the question of whether or
`not abnormalities in GLP-1 secretion are involved in glucose
`intolerance. In one previous report, type II diabetic subjects
`were noted to have a higher GLP-1 response to ingested glu-
`cose than nondiabetic controls (45), but a subsequent study
`found no difference in the secretion of GLP-1 among similar
`groups (11). Little is known about the mechanisms triggering
`GLP-1 secretion, and although there is some variability in
`measured levels among individuals before and after meals nat-
`urally occurring GLP-1 deficiency states in humans have not
`been described. These important questions will require further
`study.
`In summary, removal of the GLP-1 effect, either by com-
`petitive receptor blockade or immunoneutralization, causes a
`deterioration of intragastric glucose tolerance in nonhuman
`primates. This demonstrates that GLP-1 has a significant role
`in the disposition of glucose absorbed from the gut in a species
`closely related to humans. In addition, it appears that circulat-
`ing GLP-1 has a regulatory effect on basal islet output of glu-
`cagon and consequently fasting glycemia. This finding raises
`the possibility that differences in basal and stimulated GLP-1
`secretion among individuals may account for some of the vari-
`ation in levels of glucose before, as well as after, eating. The
`results described here amplify the endocrine role of the gas-
`trointestinal tract in fuel metabolism.
`
`Glucagon-like Peptide 1 and Glucose Tolerance
`
`137
`
`Downloaded from http://www.jci.org on May 21, 2015. http://dx.doi.org/10.1172/JCI118380
`
`SANOFI-AVENTIS Exhibit 1020 - Page 137
`
`IPR for Patent No. 8,951,962
`
`

`
`Acknowledgments
`
`We thank Don Martin for analytic and statistical assistance, and John
`Balch for his invaluable expertise in the care of the animals.
`This work was carried out at the Regional Primate Center at the
`University of Washington (National Institutes of Health [NIH] 5P21
`RR00166). A portion of the peptide analysis was performed in the
`Diabetes and Endocrine Research Center (NIH P30 DK-17047). It
`was supported by United States Public Health Service grants DK-
`34397, DK-30992, and RR-00037. David D’Alessio is a recipient of a
`New Investigator Award from the Clinical Nutrition Research Unit
`(grant DK-35816).
`
`References
`
`1. Drucker, D. 1990. Glucagon and the glucagon-like peptides. Pancreas. 5:
`484–488.
`2. Goke, R., H.-C. Fehmann, and B. Goke. 1991. Glucagon-like peptide-1
`(7-36) amide is a new incretin/enterogastrone candidate. Eur. J. Clin. Invest. 21:
`135–144.
`3. Mojsov, S., G. C. Weir, and J. F. Habener. 1987. Insulinotropin: gluca-
`gon-like peptide 1 (7-37) coencoded in the glucagon gene is a potent stimulator
`of insulin release in the perfused rat pancreas. J. Clin. Invest. 79:616–619.
`4. Kreymann, B., M. A. Ghatei, G. Williams, and S. R. Bloom. 1987. Gluca-
`gon-like peptide-1 7-36: a physiological incretin in man. Lancet. 2:1300–1303.
`5. Holst, J. J. 1994. Glucagonlike peptide 1: a newly discovered gastrointes-
`tinal hormone. Gastroenterology. 107:1848–1855.
`6. Marks, R. M., L. M. Morgan, J. A. Tredger, S. Deacon, J. Wright, and V.
`Marks. 1993. Glucagon-like peptide-1(7-36)amide and glucose-dependent in-
`sulinotropic polypeptide secretion in response to nutrient ingestion in man:
`acute post-prandial and 24-h secretion patterns. J. Endocrinol. 138:159–166.
`7. McIntyre, N., C. D. Holdsworth, and D. S. Turner. 1965. Intestinal fac-
`tors in the control of insulin secretion. J. Clin. Endocrinol. 25:1317–1324.
`8. Perley, M. J., and D. M. Kipnis. 1967. Plasma insulin responses to oral
`and intravenous glucose. Studies in normal and diabetic subjects. J. Clin. Invest.
`46:1954–1962.
`9. Creutzfelt, W., and R. Ebert. 1985. New developments in the incretin
`concept. Diabetologia. 28:565–573.
`10. Komatsu, R., T. Matsuyama, M. Namba, N. Watanabe, H. Itoh, N.
`Kono, and S. Tarui. 1989. Glucagonostatic and insulinotropic action of glu-
`cagonlike peptide 1-(7-36)amide. Diabetes. 38:902–905.
`11. Nauck, M. A., M. M. Heimesaat, C. Orskov, J. J. Holst, R. Ebert, and
`W. Creutzfeldt. 1993. Preserved incretin activity of glucagon-like peptide 1 (7-
`36 amide) but not of synthetic human gastric inhibitory polypeptide in patients
`with type-2 diabetes mellitus. J. Clin. Invest. 91:301–307.
`12. Gutniak, M., C. Orskov, J. Holst, B. Ahren, and S. Efendic. 1992. An-
`tidiabetogenic effect of glucagon-like peptide-1(7-36)amide in normal subjects
`and patients with diabetes mellitus. N. Engl. J. Med.