Diabetologia (1999) 42: 1324–1331
`
`(cid:211) Springer-Verlag 1999
`
`Inhibition of dipeptidyl peptidase IV with NVP-DPP728 increases
`plasma GLP-1 (7–36 amide) concentrations and improves oral
`glucose tolerance in obese Zucker rats
`
`B. Balkan1, L. Kwasnik1, R. Miserendino1, J. J. Holst2, X. Li1
`
`1 Novartis Institute for Biomedical Research Summit, New Jersey, USA
`2 Department of Medical Physiology, The Panum Institute, University of Copenhagen, Copenhagen, Denmark
`
`Abstract
`
`Aims/hypothesis. The potent incretin hormone gluca-
`gon-like peptide 1 (GLP-1) plays a pivotal role in
`prandial insulin secretion. In the circulation GLP-1
`(7–36) amide is, however, rapidly (t1/2:1–2 min) inacti-
`vated by the protease dipeptidyl peptidase IV (DPP-
`IV). We therefore investigated whether DPP-IV inhi-
`bition is a feasible approach to improve glucose ho-
`meostasis in insulin resistant, glucose intolerant fatty
`Zucker rats, a model of mild Type II (non-insulin-de-
`pendent) diabetes mellitus.
`Methods. An oral glucose tolerance test was done in
`lean and obese male Zucker rats while plasma DPP-
`IV was inhibited by the specific and selective inhibi-
`tor NVP-DPP728 given orally.
`Results. Inhibition of DPP-IV resulted in a signifi-
`cantly amplified early phase of the insulin response
`to an oral glucose load in obese fa/fa rats and restora-
`tion of glucose excursions to normal. In contrast,
`DPP-IV inhibition produced only minor effects in
`
`lean FA/? rats. Inactivation of GLP-1 (7–36) amide
`was completely prevented by DPP-IV inhibition sug-
`gesting that the effects of this compound on oral glu-
`cose tolerance are mediated by increased circulating
`concentrations of GLP-1 (7–36) amide. Reduced gas-
`tric emptying, as monitored by paracetamol appear-
`ance in the circulation after an oral bolus, did not ap-
`pear to have contributed to the reduced glucose ex-
`cursion.
`Conclusion/interpretation. It is concluded that NVP-
`DPP728 inhibits DPP-IV and improves insulin secre-
`tion and glucose tolerance, probably through aug-
`mentation of the effects of endogenous GLP-1. The
`improvement observed in prandial glucose homeo-
`stasis during DPP-IV inhibition suggests that inhibi-
`tion of this enzyme is a promising treatment for
`Type II diabetes. [Diabetologia (1999) 42: 1324–1331]
`
`Keywords NVP-DPP728, DPP-IV, GLP-1, insulin se-
`cretion, gastric emptying.
`
`Control of prandial plasma concentrations of sub-
`strates, particularly glucose, is aided by incretins. In-
`cretins are hormones released by the digestive tract
`in response to ingested nutrients [1]. The role of in-
`cretins is to sensitize beta cells to stimulation by glu-
`
`Received: 13 April 1999 and in revised form: 1 July 1999
`
`Corresponding author: B. Balkan, Novartis Institute for Bio-
`medical Research, LSB 3517, 556 Morris Ave Summit, New
`Jersey 07901, USA.
`Abbreviations: AMC, 7-Amino-4-methylcoumarin; CMC, car-
`boxymethylcellulose; DPP-IV, dipeptidyl peptidase IV; GIP,
`gastric inhibitory polypeptide/glucose-dependent
`insulino-
`tropic polypeptide; GLP-1, glucagon-like peptide 1.
`
`cose, leading to an accelerated and augmented insulin
`response to absorbed glucose. The two most promi-
`nent incretins are gastric inhibitory polypeptide/glu-
`cose-dependent
`insulinotropic polypeptide (GIP)
`and glucagon-like peptide-1 (GLP-1). The first is re-
`leased by K cells in the duodenum and jejunum [2]
`whereas GLP-1 is produced in L cells in the distal
`small intestine and in the colon [3]. Both GLP-1 and
`GIP augment glucose-stimulated insulin secretion [4,
`5] and increase intracellular cAMP concentrations in
`beta cells [6]. Although some controversy exists re-
`garding the relative importance of each of these in-
`cretins, both hormones have strong priming effects
`on the beta cells [7]. Several studies have shown that
`
`Merck Exhibit 2245, Page 1
`Mylan Pharmaceuticals Inc. v. Merck Sharp & Dohme Corp.
`IPR2020-00040
`
`

`

`B. Balkan et al.: DPP-IV inhibition improves oral glucose tolerance
`
`1325
`
`blockade of GLP-1 action with the GLP-1 receptor
`antagonist exendin (9–39) amide or with neutralizing
`antibodies results in increased glucose excursions af-
`ter oral or intraduodenal glucose or meals while de-
`laying or reducing the insulin responses [8–10]. This
`picture is reminiscent of the defect in insulin secre-
`tion in subjects with Type II (non-insulin-dependent)
`diabetes mellitus during an oral or intravenous glu-
`cose challenge and can be improved by replacing the
`early phase of insulin secretion [11, 12, 13]. The en-
`hancement of the insulin response to oral nutrients
`by GLP-1 and GIP make GLP-1 and GIP attractive
`potential treatments for Type II diabetes.
`In addition to its insulinotropic action, GLP-1 re-
`duces gastric emptying [14] and inhibits glucagon se-
`cretion [15, 16], both further reducing the prandial
`glucose excursion. In Type II diabetic patients, GLP-
`1 has been successfully used to normalize fasting and
`prandial glycaemia [15, 17–20]. As GLP-1 is a pep-
`tide, the application has, however, been limited to ex-
`ploratory studies with intravenous infusion or buccal
`tablets [21].
`Active GLP-1 (7–36) amide is rapidly transformed
`by dipeptidyl peptidase IV (CD26) into GLP-1
`(9–36) amide which is inactive or antagonistic [22].
`This cleavage leads to a half-life of the active GLP-1
`(7–36) of 1 to 1.5 min in the circulation [23]. Recently,
`it has been shown that a substantial part of GLP-1
`(7–36) amide released by the intestinal L cells is de-
`graded in the surrounding capillaries [24]. This rapid
`inactivation is also responsible for the relative inef-
`fectiveness of subcutaneously injected GLP-1 in the
`control of glycaemia in Type II diabetic patients [25]
`and will remain a major obstacle to any therapy rely-
`ing on GLP-1 (7–36) amide.
`We hypothesized that the inhibition of dipeptidyl
`peptidase IV(DPP-IV) would increase the plasma
`residence time of natural secreted GLP-1 [7–36)
`amide substantially and would thereby improve gly-
`caemic control during a challenge. To test this hy-
`pothesis, we gave glucose orally to lean and obese
`Zucker rats after treatment with the DPP-IV inhibi-
`tor 1-[2-[(5-cyanopyridin-2-yl) amino] ethylamino]
`acetyl-2-cyano-(S)-pyrrolidine monohydrochloride
`salt (NVP-DPP728) and monitored their plasma
`GLP-1 response, and glucose and insulin excursions.
`NVP-DPP728 is a potent and selective orally avail-
`able DPP-IV inhibitor (Fig. 1).
`
`Materials and methods
`
`Animals and surgery. Lean (FA/?) and obese (fa/fa) male
`Zucker rats (Charles River, Wilmington, Mass., USA) were
`housed under a reversed light cycle (lights on 2000 hours to
`0800 hours) with free access to tap water and standard rodent
`chow (Purina Labs, Richmond, Ind.,USA). The animals were
`aseptically implanted with a silastic catheter in the right jugu-
`lar vein under Ketamine/Rompun/Acepromazin anaesthesia.
`The catheter was externalized in the nape of the neck and was
`filled with a solution of heparin and polyvinylpyrrolidone.
`The animals were allowed to recover from the surgery (regain
`of lost body weight) before the experiments.
`
`Procedures. The animals were fasted for approximately 16 h
`(food removed 1700 hours on the day before the experiment).
`At –30 min on the day of the experiment, animals were orally
`dosed with the DPP-IV inhibitor NVP-DPP728 (10 mmol/kg,
`the dose had been chosen from pilot studies) in vehicle (0.5 %
`carboxymethylcellulose (CMC) with 0.2 % Tween 80) or vehi-
`cle alone. The NVP-DPP728 [26] was kindly supplied by E.
`Villhauer and J. Brinkman. The cannulas were then connected
`to sampling tubing. At –10 and 0 min two basal samples (500ml)
`were withdrawn. Glucose (1 g/kg) was given by gavage after
`the second sample. Additional samples were withdrawn at 1,
`3, 5, 10, 15, 20, 30, 45, and 60 min. All samples were replaced
`by donor blood containing citrate and sodium-citrate as antico-
`agulant. Blood samples were collected in chilled Eppendorf
`tubes containing EDTA and trasylol to achieve final concen-
`trations of 2.5 mg EDTA and 1000 KIU trasylol per ml of
`blood. Samples were centrifuged and plasma was stored at
`–20(cid:176)C until analyses.
`A second experiment was done in obese Zucker rats to ob-
`tain samples for analysis of plasma concentrations of GLP-1
`(7–36) amide and total GLP-1. The procedure was identical to
`the one described above except that larger samples (1600ml)
`were collected fewer times (0, 5, 10, 20, 30, 60 min) and that
`paracetamol (4-acetomidophenol, Sigma, St. Louis, Mo.,
`USA) was given (at 100 mg/kg i. e. 0.66 mmol/Kg) together
`with the oral glucose load, to evaluate effects on gastric empty-
`ing. In addition to the six samples described above, smaller
`samples (100 ml) collected at all the same times as in the first
`study were sampled to follow the appearance of the paraceta-
`mol in the plasma and to verify the glucose excursions from
`the first study.
`
`Plasma analyses. Plasma glucose was analysed using a modi-
`fied Sigma Diagnostics glucose oxidase kit (Sigma). Plasma
`immunoreactive insulin (IRI) concentrations were assayed by
`a double antibody RIA method using a specific anti-rat insulin
`antibody from Linco Research (St Louis, Mo., USA). The as-
`say has a lower limit of detection of 30 pmol/l with intra-assay
`and inter-assay variations of less than 5 %.
`Plasma GLP-1 concentrations were determined using ra-
`dioimmunoassays specific for each end of the molecule. Ami-
`no-terminal immunoreactivity was measured using antiserum
`93 242 [21], which has a cross-reactivity of about 10 % with
`GLP-1 (1–36) amide and of less than 0.1 % with GLP-1 (8–36)
`amide and GLP-1 (9–36) amide. The detection limit is 5 pmol/
`l. High-performance liquid chromatography (HPLC) supports
`the use of RIAs with this specificity for determination of intact
`GLP-1 [22]. COOH-terminal
`immunoreactivity was deter-
`mined using antiserum 89 390 [27], which has an absolute re-
`quirement for the intact amidated COOH-terminus of GLP-1
`(7–36) amide and cross-reacts less than 0.01 % with COOH-
`terminally truncated fragments and 83 % with GLP-1 (9–36)
`amide. For all assays, the intra-assay coefficient of variation
`
`N
`
`O
`
`N
`
`NH
`
`N
`
`N
`
`N
`H
`
`HCl
`Fig. 1. Chemical structure of NVP-DPP728
`
`Merck Exhibit 2245, Page 2
`Mylan Pharmaceuticals Inc. v. Merck Sharp & Dohme Corp.
`IPR2020-00040
`
`

`

`1326
`
`B. Balkan et al.: DPP-IV inhibition improves oral glucose tolerance
`
`Table 1. Basal plasma variables of obese (fa/fa) and lean (FA/?) Zucker rats after vehicle (CMC) or compound (10 mmol/kg) treat-
`ment
`Strain
`
`Treatment
`
`Body weight
`(g)
`385 – 17
`8
`Vehicle
`fa/fa
`373 – 19
`8
`NVP-DPP 728
`fa/fa
`288 – 10a
`6
`Vehicle
`FA/?
`286 – 7
`7
`NVP-DPP 728
`FA/?
`Data are absolute values – SEM of 0 min. samples in the re-
`spective unit as indicated by the column header. Data were
`analysed by two-way ANOVA for strain effects (a: p < 0.01)
`and treatment effects (b: p < 0.05, c: p < 0.01, vs, the respective
`
`n
`
`DPP-IV activity
`Plasma IRI
`Plasma glucose
`(mU/ml)
`(pmol/l)
`(mmol/l)
`5.8 – 1.4
`2298 – 534
`6.4 – 0.4
`5.1 – 0.4b
`0 – 0c
`1674 – 462
`3.7 – 0.8
`209 – 24a
`4.4 – 0.1a
`0 – 0b
`147 – 14
`4.1 – 0.2
`vehicle-treated control group by Bonferroni correction). Va-
`lues that differ at the p < 0.01 level are marked c and d, respec-
`tively. The number of animals per group is given in column n.
`Vehicle or compound was given orally at (cid:159)30 min
`
`was less than 6 %. Plasma samples were extracted with 70 %
`ethanol (vol/vol, final concentration) before assay, giving re-
`coveries of 75 % [28]. The assay for GLP-1 (7–36) amide re-
`sulted in higher values than the COOH-terminal assay. For
`this reason all data for GLP-1 responses are depicted as change
`from basal levels.
`Paracetamol concentrations were determined in plasma
`samples using a fluorometric kit (No 430-A, Sigma). Briefly,
`plasma proteins are removed with trichloroacetic acid, fol-
`lowed by reaction of paracetamol with nitrous acid, yielding ni-
`tro-derivatives which assume a deep yellow colour in alkaline
`medium. The sample is read in a spectrophotometer at
`430 mm and the absorbance is proportional to the paracetamol
`concentration. Plasma DPP-IV activity was measured using an
`assay based on a modification described previously [29] of a
`published method [30]. Briefly, aliquots of plasma were incu-
`bated with substrate (Gly-Pro-AMC, where AMC is 7-amino-
`4-methylcoumarin (Bachem, King of Prussia, Pa., USA). Free
`AMC generated proportionally to DPP-IV activity was mea-
`sured by fluorimetry. Catalytic DPP-IV activity in plasma is ex-
`pressed as mUnit/ml (1 Unit = 1 umol substrate cleaved per
`min). The study was approved by the Novartis Animal Care
`and Use Committee, and is in accordance with the NIH guide-
`lines for laboratory animal care.
`Data are expressed as means  SEM. Statistical analysis
`was by t test. All statistical significances reflect two-tailed test-
`ing unless stated otherwise. Because insulin responses were ex-
`pected to be increased and glucose excursions decreased in an-
`imals treated with drugs, one-tailed t tests were done for area
`under the curve (AUC) results. Basal data for the animals in
`the different groups (Table 1) were tested by two-way ANO-
`VA followed by Bonferroni correction.
`A difference was considered statistically significant when p
`was less than 0.05.
`
`Results
`
`Basal variables of the animals in this study are de-
`scribed in Table 1. Obese (fa/fa) rats were heavier
`than their lean counterparts. The obese animals dis-
`played a moderate increase of fasting glycaemia,
`compared with the lean rats, in the presence of pro-
`nounced hyperinsulinaemia. This is indicative of pro-
`found insulin resistance in these animals. Inhibition
`of DPP-IV with NVP-DPP728 lowered basal plasma
`glucose concentrations
`in obese rats by 20 %
`(p < 0.05) but had no effect in the lean rats. Inhibition
`
`of plasma DPP-IV activity did reduce basal plasma
`insulin concentrations by about 30 % in both lean
`and obese rats but this was not statistically significant
`due to large interanimal variability. Treatment with
`NVP-DPP728 led to complete inhibition of plasma
`DPP-IV activity.
`Giving 1 g/kg glucose orally to lean and obese
`Zucker rats evoked a pronounced glucose excursion in
`both groups of animals (Fig. 2). Obese rats treated
`with vehicle reached, however, higher peak glucose
`concentrations (mean of individual peak glucose con-
`centrations
`fa/fa: 15.3  0.4 mmol/l, FA/?: 9.3 
`0.3 mmol/l, p < 0.0001) with a glucose excursion of ap-
`proximately twice the size of lean animals similarly
`treated (Fig. 3). This was in spite of a considerably in-
`creased insulin response (Fig. 2). When plasma DPP-
`IV activity was fully inhibited (Table 1, Fig. 4), the in-
`sulin response to the oral glucose challenge was in-
`creased in obese and to a lesser extent in lean Zucker
`rats (Fig. 2). Concomitantly the plasma glucose con-
`centrations were lower in DPP-IV-inhibited obese
`Zucker rats, with lower peak glucose concentrations
`than in fa/fa rats
`treated with vehicle (12.1 
`1.0 mmol/l vs 15.3  0.4 mmol/l, p < 0.01). The change
`in the glucose profile occurred predominantly in the
`latter part of the test with glucose excursions signifi-
`cantly different between fa/fa vehicle and fa/fa NVP-
`DPP728 at times 20, 30 and 45 min (p < 0.05). The im-
`provement in glucose tolerance was less pronounced
`in the lean Zucker rats but the pattern of the response
`to DPP-IV inhibition was similar to that in obese ani-
`mals. These results are also evident from the integrat-
`ed insulin responses and glucose excursions (Fig. 3).
`When plasma insulin concentrations were ex-
`pressed as a percent of individual basal concentra-
`tions, a method that takes account of the compensat-
`ed insulin resistance in normoglycaemic animals and
`reflects the secretory response itself, it is apparent
`that the obese rats treated with vehicle have a re-
`duced insulin response to the oral glucose load com-
`pared with the lean control rats (Fig. 2). After treat-
`ment with NVPDPP728 the relative insulin response
`in lean and obese animals was, however, similarly in-
`creased.
`
`Merck Exhibit 2245, Page 3
`Mylan Pharmaceuticals Inc. v. Merck Sharp & Dohme Corp.
`IPR2020-00040
`
`

`

`B. Balkan et al.: DPP-IV inhibition improves oral glucose tolerance
`
`1327
`
`In the second part of the study, the experiment was
`repeated in obese fa/fa rats to obtain samples for plas-
`ma GLP-1 determination and to obtain an index of
`gastric emptying through measurement of paraceta-
`mol appearance in the circulation. During this study
`plasma glucose concentrations were measured in all
`samples to assure that the study closely resembled
`the first experiment. Basal plasma GLP-1 (7–36 and
`9–36) amide concentrations did not differ between fa/
`fa rats treated with vehicle (6.2  2.4 pmol/l) and those
`treated with DPP-IV inhibitor (3.4  1.0 pmol/l, NS).
`Both groups of rats had a clear GLP-1 response to
`the oral glucose challenge (Fig. 5). Although the incre-
`ment in DPP-IV inhibited rats was almost twice that in
`control animals, this effect was not statistically signifi-
`cant due to a large degree of interanimal variation.
`The basal concentrations of active GLP-1 (7–36)
`amide in rats treated with NVP-DPP728 (20.3  2.3
`pmol/l) were higher than in control rats (11.8  2.2
`pmol/l, p < 0.05). Whereas the GLP-1 (7–36) amide
`response to the oral challenge in control fa/fa rats was
`blunted compared with the total GLP-1 release, the
`rise in GLP-1 (7–36) amide in animals treated with
`DPP-IV inhibitor was pronounced. In fa/fa rats treat-
`ed with vehicle 24  17 % of total GLP-1 (7–36) amide
`at the peak (5–10 min) is present in the active form
`(calculated from the mean of values obtained at 5
`GLP-1 (7(cid:255) 36)
`(cid:148) 100 %). In contrast all
`and 10 min
`GLP-1
`of the GLP-1 is present as the intact form in rats treat-
`ed with NVP-DPP728 (109  20 %, p = 0.01).
`During the second study, paracetamol was given
`orally together with the glucose solution to obtain an
`index of gastric emptying. Plasma paracetamol con-
`centrations in the fa/fa rats are depicted in Figure 6.
`The curves for vehicle and fa/fa rats treated with
`DPP-IV inhibitor are superimposable except for a
`significant difference (p = 0.03) at 20 min after giving
`glucose and paracetamol. The area under the parace-
`tamol curve was not different between NVP-
`DPP728-treated (AUC 39.5  1.3 mmol/l (cid:215) 120 min)
`and fa/fa rats treated with vehicle (42.6  1.3 mmol/
`l (cid:215) 120 min, NS).
`
`Discussion
`
`Our aim was to study the effects of DPP-IV inhibition
`on insulin secretion, glycaemic control and GLP-1
`concentrations during an oral glucose challenge in a
`model of impaired glucose tolerance. The results in-
`dicate that treatment with NVP-DPP728, an inhibitor
`of DPP-IV, preserves the active GLP-1 (7–36) amide
`in circulation and augments early insulin response to
`an oral glucose challenge. As a consequence, the glu-
`cose excursion in obese Zucker is nearly restored to
`normal in fa/fa rats. The effects of DPP-IV inhibition
`were much less pronounced in lean Zucker rats.
`
`A
`
`B
`
`C F
`
`ig. 2 A–C. Changes in plasma (A) glucose concentrations
`(B) immunoreactive insulin (IRI) concentrations and (C) rela-
`tive change in plasma IRI concentrations in obese (fa/fa) and
`lean (FA/?) Zucker rats (n = 6–8) after vehicle (CMC) or com-
`pound (10 mmol/kg), was given at (cid:159) 30 min. All animals re-
`ceived 1 g/kg glucose orally at 0 min. The data are displayed
`as means  SEM of individual delta plasma glucose concentra-
`tions from basal (pre-glucose, 0 min.) levels. In C (cid:190)~(cid:190) FA/?
`vehicle (cid:190)~(cid:190) FA/? NVP-DPP728 (cid:190)*(cid:190) fa/fa vehicle (cid:190)*(cid:190)
`fa/fa NVP-DPP728. The data are displayed as mean  SEM
`per cent of individual basal (pre-glucose, 0 min.) plasma IRI
`concentrations
`
`Merck Exhibit 2245, Page 4
`Mylan Pharmaceuticals Inc. v. Merck Sharp & Dohme Corp.
`IPR2020-00040
`
`

`

`1328
`
`B. Balkan et al.: DPP-IV inhibition improves oral glucose tolerance
`
`Fig. 3 A, B. Integrated changes in plasma immunoreactive in-
`sulin (A) and plasma glucose concentrations (B) during an
`oral glucose tolerance test in obese (fa/fa) and lean (FA/?)
`Zucker rats (n = 6–8) after vehicle (CMC) or compound
`(NVP-DPP728, 10 mmol/kg given at (cid:159)30 min. All animals re-
`ceived 1 g/kg glucose orally at 0 min. The data are displayed
`as means  SEM of
`incremental areas under the curves
`(AUC) calculated by the trapezoidal rule. Insulin AUC re-
`flects the early phase of insulin secretion (0–15 min.), glucose
`AUC 0–45 min. Statistically significant differences are presen-
`ted as *: p < 0.05 vs fa/fa vehicle, =: p < 0.05 vs, FA/? vehicle
`, vehicle; NVP-DPP728
`and #: p = 0.06 vs FA/? vehicle.
`
`received a large
`Glucagon-like peptide-1 has
`amount of attention in recent years because of its po-
`tential as a treatment for Type II diabetes [31]. This is
`mainly based on its potent sensitization of pancreatic
`beta cells to stimulation with glucose and its inhibi-
`tion of glucagon secretion [4, 5]. Since impaired insu-
`lin secretion to a glucose challenge is a hallmark of
`Type II diabetes [32], GLP-1-based treatment would
`target one of the fundamental defects of the disease.
`A GLP-1-based therapy is also attractive because
`the responsiveness of beta cells to GLP-1 is main-
`tained in Type II diabetic subjects [33]. Although
`many studies have shown the benefit of GLP-1 treat-
`ment on glycaemic control in Type II and Type I dia-
`betic patients, poor bioavailability is a major draw-
`back for the clinical use of GLP-1.
`Furthermore, because the half-life of active GLP-1
`in the circulation is in the order of 1 to 2 min [22, 34]
`the feasibility of GLP-1 itself as a therapy is limited.
`The short half-life is exclusively dependent on the
`dipeptidylpeptidase IV-mediated cleavage of
`two
`amino-terminal amino acids of GLP-1 (7–36) amide
`to generate the inactive, or antagonistic [35], frag-
`ment GLP-1 (9–36) amide. This rapid degradation of
`GLP-1 (7–36) amide leads to the prediction that
`only approximately 20 % of endogenous or exoge-
`nous GLP-1 is circulating in the active form. This esti-
`mate is supported by this and other studies [21–23,
`36]. Active GLP-1 (7–36) amide concentrations nec-
`essary to achieve improvements in glycaemic control
`are in the range of 20–30 pmol/l, the concentrations
`found postprandially for total GLP-1. We hypothe-
`
`sized therefore that inhibition of DPP-IV activity
`would raise the low circulating concentrations of ac-
`tive GLP-1 (7–36) amide after a nutrient challenge
`to those that are more effective and would thereby
`improve glycaemic control.
`The effects of NVP-DPP728, an inhibitor of DPP-
`IV, was investigated in obese Zucker rats (fa/fa), a
`well-characterized model of obesity and insulin re-
`sistance. The primary genetic defect resides in the re-
`ceptor for the ob gene product (leptin, [37]). As a
`consequence, the feedback loop from the adipocytes
`to the central nervous system and peripheral organs
`is impaired, leading to hyperphagia, increased adipo-
`genesis, hyperinsulinaemia, obesity, impaired glucose
`tolerance, and insulin resistance [38–40]. Our study
`confirmed these characteristics and further indicates
`that the obese Zucker rats fail to increase insulin se-
`cretion in proportion to the prevailing insulin resist-
`ance (as also indicated by the tenfold hyperinsulin-
`aemia found in the basal state compared with lean
`rats, Fig. 4). Consequently, plasma glucose concentra-
`tions after the oral glucose load reach higher peak
`values than in the lean control group and remain
`high for a prolonged period of time, both characteris-
`tics of human impaired glucose tolerance.
`Complete DPP-IV inhibition augmented the insu-
`lin response to an oral glucose challenge in glucose
`intolerant, obese Zucker rats. It also resulted in a
`pronounced reduction in the glucose excursion. Two
`details are noteworthy. Firstly, the statistically signifi-
`cant increase in insulin concentrations could first be
`seen 5 min after giving glucose in animals treated
`with DPP-IV inhibitor. Second, no differences be-
`tween fa/fa rats treated with vehicle or inhibitor could
`be found in the glucose concentrations until approxi-
`mately 20 min after the glucose bolus. These data
`are congruent with the hypothesis that the augment-
`ed insulin response is causal to the reduction in the
`glucose excursions. Furthermore, because the rise in
`glucose concentration is unaffected by DPP-IV inhi-
`bition, this suggests that the rate of gastric emptying
`or absorption is not likely to have affected the glu-
`cose appearance. To corroborate these conclusions
`we added paracetamol to the glucose load in the sec-
`ond study. Paracetamol is absorbed in the duodenum
`but not in the stomach [41]. As such it is a suitable
`
`Merck Exhibit 2245, Page 5
`Mylan Pharmaceuticals Inc. v. Merck Sharp & Dohme Corp.
`IPR2020-00040
`
`

`

`B. Balkan et al.: DPP-IV inhibition improves oral glucose tolerance
`
`1329
`
`Fig. 4. Plasma DPP-IV activity in obese (fa/fa) and lean (FA/?)
`Zucker rats (n= 6–8) after vehicle (CMC) or compound (NVP-
`DPP728, 10 mmol/kg) given at (cid:159)30 min. All animals received
`1 g/kg glucose orally at 0 min. The data are displayed as means
` SEM of individual activity. (cid:190)*(cid:190) fa/fa vehicle (cid:190)*(cid:190) fa/fa
`NVP-DPP728 (cid:190)~(cid:190) FA/? vehicle (cid:190)~(cid:190) FA/? NVP-DPP728
`
`marker for gastric emptying of liquids [42]. The lack
`of any major delay in the paracetamol appearance in-
`dicates that DPP-IV inhibition had a very small ef-
`fect, if any, on gastric emptying and that the reduction
`in glucose excursion is not likely to be the result of re-
`duced absorption. Indeed in a control experiment in
`which cholecystokinin was given to delay gastric
`emptying the paracetamol appearance (and the rise
`in plasma glucose) was considerably delayed (data
`not shown).
`These results differ from those in studies in which
`GLP-1 was given exogenously which report effects on
`gastric emptying [35, 43, 44]. Furthermore, the degra-
`dation product GLP-1 (9–36) amide has been reported
`to antagonize GLP-1 (7–36) amide-induced inhibition
`of gastric motility [35]. Thus, it could be expected that
`during DPP-IV inhibition, with higher GLP-1 (7–36)
`and lower antagonist concentrations the delayed gas-
`tric emptying effects would be prominent.
`Three theoretical explanations for this difference
`can be given. Firstly, humans and rodents possibly
`rely to a different degree on GLP-1 for regulating
`gastric emptying and with different sensitivity. We re-
`cently observed that supra pharmacological doses of
`GLP-1 (7–36) amide injected intravenously at the
`time of the oral glucose challenge substantially delay
`paracetamol appearance [45], suggesting that in the
`present study GLP-1 did not reach sufficient concen-
`trations to result in delayed gastric emptying. Second-
`ly, basal gastric motility is relatively low, and may not
`be required for emptying of a solely liquid nutrient
`load. Third, most of the glucose solution may have
`been in the intestine by the time glucose-stimulated
`GLP-1 release could affect gastric motility. Thus, the
`lack of effect of DPP-IV inhibition on gastric empty-
`
`Fig. 5 A, B. Changes in plasma immunoreactive GLP-1 (7–36)
`amide (A) and total plasma GLP-1 immunoreactivity (B) dur-
`ing an oral glucose tolerance test in obese (fa/fa) Zucker rats
`(n = 8) after vehicle (CMC) or compound (NVP-DPP728,
`10 mmol/kg) given at (cid:159)30 min. All animals received 1 g/kg glu-
`cose orally at 0 min. The data are displayed as means + SEM
`of individual delta plasma GLP-1 concentrations from basal
`(pre-glucose, 0 min.) levels. (cid:190)*(cid:190) vehicle (cid:190)*(cid:190) NVP DPP
`728
`
`ing in the current study could well be limited to the
`applied nutritional challenge.
`The results of this study confirm the observations
`by ourselves and others [46, 47] that DPP-IV inhibi-
`tion considerably improves oral glucose tolerance in
`Zucker rats. We extend this finding by showing that
`endogenously secreted GLP-1 (7–36) amide is highly
`preserved in the circulation during DPP-IV inhibi-
`tion, suggesting that the improved insulin response
`and glucose tolerance is a result of increased GLP-1
`action on the pancreas. The data further suggest that
`the improved glucose tolerance is not a result of re-
`duced gastric emptying. The paracetamol data fur-
`ther indicate that a large part of the glucose load is
`emptied into the small
`intestine within the first
`20 min after the glucose load is given. This is support-
`ed by the profile of total GLP-1 in the plasma. Peak
`concentrations of GLP-1 are achieved 5 min after
`glucose is given and total plasma GLP-1 concentra-
`
`Merck Exhibit 2245, Page 6
`Mylan Pharmaceuticals Inc. v. Merck Sharp & Dohme Corp.
`IPR2020-00040
`
`

`

`1330
`
`B. Balkan et al.: DPP-IV inhibition improves oral glucose tolerance
`
`caemia is greatest when traditional insulin secreta-
`gogues, such as sulphonylureas are used.
`
`Acknowledgements. Expert assistance of L. Albaek, A. Bolat,
`C. Crisafi, T. Hughes, M. Mone, J. Olejarczyk-Gurkan is grate-
`fully acknowledged.
`
`References
`
`1. Creutzfeldt W (1979) The incretin concept today. Diabetologia
`16: 75–85
`2. Brown JC, Buchan AM, McIntosh CH, Pederson RA (1989)
`Gastric inhibitory polypeptide. Schultz S. G., Makhlouf, G. M.,
`and Rauner, B. B. Handbook of Physiology, Section 6. Beth-
`esda, American Physiological Society, pp 403–430
`3. Holst JJ (1997) Enteroglucagon. Annu Rev Physiol 59:
`257–271
`4. Goeke R, Wagner B, Fehmann HC, Goeke B (1993) Glucose-
`dependency of the insulin stimulatory effect of glucagon- like
`peptide-1 (7–36) amide on the rat pancreas. Res Exp Med
`(Berl) 193: 97–103
`5. Fridolf T, Bottcher G, Sundler F, Ahren B (1991) GLP-1 and
`GLP-1(7–36) amide: influences on basal and stimulated insulin
`and glucagon secretion in the mouse. Pancreas 6: 208–215
`6. Lu M, Wheeler MB, Leng XH, Boyd AE 3rd (1993) The role of
`the free cytosolic calcium level in beta-cell signal transduction
`by gastric inhibitory polypeptide and glucagon-like peptide
`I(7–37). Endocrinology 132: 94–100
`7. Fehmann HC, Goeke R, Goeke B, Bachle R, Wagner B, Ar-
`nold R (1991) Priming effect of glucagon-like peptide-1 (7–36)
`amide, glucose- dependent insulinotropic polypeptide and
`cholecystokinin-8 at the isolated perfused rat pancreas. Bio-
`chim Biophys Acta 1091: 356–363
`8. Kolligs F, Fehmann HC, Goke R, Goke B (1995) Reduction of
`the incretin effect in rats by the glucagon-like peptide 1 recep-
`tor antagonist exendin (9–39) amide. Diabetes 44: 16–19
`9. Wang Z, Wang RM, Owji AA, Smith DM, Ghatei MA, Bloom
`SR (1995) Glucagon-like peptide-1 is a physiological incretin
`in rat. J Clin Invest 95: 417–421
`10. D’Alessio DA, Vogel RL, Prigeon RL et al. (1996) Elimination
`of the action of Glucagon-like Peptide 1 causes an impairment
`of glucose tolerance after nutrient ingestion in healthy ba-
`boons. J Clin Invest 97: 133–138
`11. Seltzer HS, Allen EW, Herron AL, Brennan MT (1998) Insulin
`secretion in response to glycemic stimulus: relation of delayed
`initial release to carbohydrate intolerance in mild diabetes
`mellitus. J Clin Invest 46: 323–335
`12. Bruce DG, Chisholm DJ, Storlien LH, Kraegen EW (1988)
`Physiological importance of deficiency in early prandial insulin
`secretion in non-insulin-dependent diabetes. Diabetes 37:
`736–744
`13. Groop PH, Melander A, Groop LC (1993) The relationship be-
`tween early insulin release and glucose tolerance in healthy
`subjects. Scand J Clin Lab Invest 53: 405–409
`14. Willms B, Werner J, Holst JJ, Oerskov C, Creutzfeldt W,
`Nauck MA (1996) Gastric emptying, glucose responses, and in-
`sulin secretion after a liquid test meal: effects of exogenous glu-
`cagon – like peptide-1 (GLP-1)-(7–36) amide in type 2 (nonin-
`sulin-dependent) diabetic patients. J Clin Endocrinol Metab
`81: 327–332
`15. Gutniak MK, Linde B, Holst JJ, Efendic S (1994) Subcutane-
`ous injection of the incretin hormone glucagon-like peptide 1
`abolishes postprandial glycemia in NIDDM. Diabetes Care
`17: 1039–1044
`16. Ritzel U, Orskov C, Holst JJ, Nauck MA (1995) Pharmacoki-
`netic, insulinotropic, and glucagonostatic properties of GLP-1
`(7–36 amide) after subcutaneous injection in healthy volun-
`teers. Dose-response-relationships. Diabetologia 38: 720–725
`
`Fig. 6. Changes in plasma paracetamol concentrations during
`an oral glucose tolerance test in obese (fa/fa) Zucker rats
`(n = 8) after vehicle (CMC) or compound (NVP-DPP728,
`10 mmol/kg). Animals were administered vehicle or compound
`at (cid:159)30 min. All animals received 1 g/kg glucose and 100 mg/kg
`paracetamol orally at 0 min. The data are displayed as mean-
`s  SEM of individual plasma paracetamol concentrations
`(cid:190)*(cid:190) vehicle (cid:190)*(cid:190) NVP DPP 728
`
`tions are close to basal at 20 min. Considering that
`the half-life of elimination of total GLP-1 from the
`circulation is 3–3.5 min in pigs [29] still severa