Abstract
Administration of an α-glucosidase inhibitor, voglibose, increases the secretion of glucagon-like peptide (GLP)-1, a key modulator of pancreatic islet hormone secretion and glucose homeostasis. In the present study, novel mechanisms by which voglibose increases active GLP-1 circulation were evaluated. Voglibose (0.001 and 0.005%) was administered in the diet to ob/ob mice for 1 day or 3 to 4 weeks to determine effects on incretin profiles and plasma activity of dipeptidyl peptidase-4 (DPP-4), an enzyme responsible for GLP-1 degradation. Voglibose showed no direct inhibitory effect against DPP-4 in vitro (DPP-4 inhibitor alogliptin, IC50 < 10 nM). Likewise, 1-day treatment with voglibose did not change plasma DPP-4 activity; however, it increased plasma active GLP-1 by 1.6- to 3.4-fold. After chronic treatment, voglibose stimulated GLP-1 secretion, as evidenced by the 1.3- to 1.5-fold increase in plasma active plus inactive amidated GLP-1 levels. Plasma DPP-4 activity was decreased unexpectedly by 40 to 51%, resulting from reduced plasma DPP-4 concentrations in voglibose-treated mice. Voglibose increased GLP-1 content by 1.5- to 1.6-fold and 1.4- to 1.6-fold in the lower intestine and colon, respectively. The increased GLP-1 content in the colon was associated with elevated expression of gut glucagon gene. Chronic treatment with voglibose resulted in 1.9- to 4.1-fold increase in active GLP-1 circulation, which was higher than 1-day treatment. A similar treatment with pioglitazone (0.03%), an insulin sensitizer, did not affect plasma DPP-4 activity or GLP-1 levels. These results suggest that increased GLP-1 secretion, decreased DPP-4 activity, and increased gut GLP-1 content may have contributed to increased active GLP-1 circulation after chronic treatment with voglibose in a glucose control-independent manner in ob/ob mice.
Gastrointestinal hormone incretins, glucose-dependent insulinotropic polypeptide (GIP), and glucagon-like peptide (GLP)-1, are physiologically important regulators of metabolic control (Meier and Nauck, 2006; Drucker, 2007). When nutrients are orally ingested, GIP and GLP-1 are secreted from the upper gut K- and lower gut L-cells, respectively, and enhance insulin release in a glucose-dependent manner from pancreatic β-cells. The augmentation of insulin secretion by gut-derived factors was termed the “incretin effect” (Creutzfeldt, 1974; Meier and Nauck, 2006; Drucker, 2007); recently, the incretin effect has been demonstrated to be largely impaired in patients with type 2 diabetes (Meier and Nauck, 2006). This reduced incretin effect has been attributed to a small but significant reduction in postprandial secretion of GLP-1 and a large impairment of insulinotropic action of GIP (Meier and Nauck, 2006). Thus, considerable interest has emerged on the pharmacological regulation of incretin actions for type 2 diabetes (Meier and Nauck, 2006; Drucker, 2007).
Because the incretin effect of GLP-1 remains relatively preserved in diabetic patients compared with those of GIP (Meier and Nauck, 2006), most efforts directed at potentiating incretin action have focused on GLP-1 agonism for the treatment of type 2 diabetes (Meier and Nauck, 2006). GLP-1 analogs have been demonstrated recently to improve glycemic control in patients with type 2 diabetes (Chia and Egan, 2008; Madsbad et al., 2008). Another approach for the pharmacological regulation of incretin action is inhibiting activity of dipeptidyl peptidase-4 (DPP-4), which is an enzyme responsible for N-terminal cleavage of intact GIP and GLP-1 (Lambeir et al., 2003). DPP-4 is a ubiquitous membrane-spanning cell surface aminopeptidase widely expressed in many tissues (Lambeir et al., 2003). The extracellular domain of DPP-4 can also be cleaved from its membrane-anchored form and circulates in plasma, where it retains its full enzymatic activity (Lambeir et al., 2003). Thus, the biological activities of GIP and GLP-1 are rapidly inactivated by DPP-4 in vivo (Lambeir et al., 2003). DPP-4 inhibitors have been demonstrated recently to extend the half-life of endogenously secreted GIP and GLP-1 in blood, resulting in augmentation of their action and improved blood glucose control in patients with type 2 diabetes (Ahrén, 2007; Madsbad et al., 2008).
α-Glucosidase inhibitors (α-GIs) are clinically used for the treatment of patients with type 2 diabetes (Scheen, 2003). α-Glucosidase is an intestinal enzyme required for carbohydrate digestion and glucose absorption (Baba, 1994). Thus, primary pharmacological action of α-GIs is to delay the absorption of carbohydrates from the small intestine by inhibiting its degradation, resulting in lowering both postprandial glucose and insulin levels (Scheen, 2003). It is interesting that α-GIs have been shown to decrease GIP levels but increase GLP-1 secretion when acutely administered in patients with type 2 diabetes (Fukase et al., 1992; Göke et al., 1995; Qualmann et al., 1995; Ranganath et al., 1998; Seifarth et al., 1998; Enç et al., 2001; Lee et al., 2002). However, the effects of α-GIs on GLP-1 regulation, in particular when chronically administered, remain poorly understood.
Voglibose is a clinically available member of the α-GIs (Horii et al., 1986; Odaka et al., 1992; Vichayanrat et al., 2002; Yasuda et al., 2003). Voglibose is 190- to 3900-fold and 23- to 33-fold more potent in inhibiting rat and porcine intestinal disaccharases, respectively, than the same class drug acarbose (Matsuo et al., 1992). Acarbose has a much greater inhibitory activity on pancreatic α-amylase, which catalyzes the first step in the breakdown of polysaccharides, such as starch, whereas voglibose has almost no effect on this enzyme (Matsuo et al., 1992); thus, at clinical dosages, voglibose is a selective disaccharidase inhibitor.
To clarify the action of chronic treatment with α-GIs on GLP-1 regulation, the present study was designed to evaluate the effects of the α-GI, voglibose, on GLP-1 secretion, plasma DPP-4 levels, gut glucagon (Gcg) gene expression and GLP-1 content, and active GLP-1 circulation in obese diabetic ob/ob mice. All parameters induced by chronic treatment with voglibose were compared with those by the peroxisome proliferator-activated receptor γ agonist pioglitazone (Sohda et al., 1990; Pfützner et al., 2007) that improves glycemic control as much as voglibose in this model.
Materials and Methods
Test Materials. Voglibose and pioglitazone hydrochloride were synthesized by Takeda Pharmaceutical Company Limited (Osaka, Japan). Alogliptin benzoate was synthesized by Albany Molecular Research Institute (Albany, NY). The dose of pioglitazone is expressed as the free base equivalent. All reagents were purchased from Wako Pure Chemicals (Osaka, Japan) or Sigma-Aldrich (St. Louis, MO), unless otherwise indicated.
Mice. Male Lepob/Lepob (ob/ob; B6.V-Lepob/J) mice and their nondiabetic, untyped (?/+; B6.V-Lepob/J) littermates were obtained from Charles River Japan (Yokohama, Japan). Male C57BL/6J mice were obtained from CLEA Japan (Tokyo, Japan). All mice were housed in individual metal cages in a room with controlled temperature (23°C), humidity (55%), and lighting (lights on from 7:30 AM–7:30 PM) and were maintained on a laboratory chow diet (CE-2; CLEA Japan). The care and use of the animals and the experimental protocols used in this research were approved by the Experimental Animal Care and Use Committee of Takeda Pharmaceutical Company Limited. All animal studies were conducted after a more than 6-day acclimation period.
One-Day Study with Voglibose. Seven-week-old ob/ob mice were divided into three groups (seven mice per group) based on body weight and food consumption and fed a CE-2 diet containing 0.001 or 0.005% voglibose for 1 day. Control diabetic ob/ob and nondiabetic?/+ mice (seven mice per group) were fed a drug-free CE-2 diet (vehicle). Blood samples (200 μl, 8:00 AM in the nonfasting state) for measuring plasma DPP-4 activity and active GLP-1 levels were collected from the orbital vein via capillary pipette after 1 day of treatment, centrifuged (12,000g) at 4°C for 5 min, and kept in a deep freezer (-80°C) until measurement.
Chronic Study with Voglibose and Pioglitazone. Seven-week-old ob/ob mice were divided into four groups (six mice per group) based on glycosylated hemoglobin, plasma glucose, plasma insulin, plasma DPP-4 activity, and body weight and fed a CE-2 diet containing 0.001% (1.4 mg/kg/day) or 0.005% (6.5 mg/kg/day) voglibose and 0.03% pioglitazone (46.8 mg/kg/day) during the experimental period. Control diabetic ob/ob and nondiabetic?/+ mice (six and five mice, respectively) were fed a drug-free CE-2 diet (vehicle). After 6, 13, and 20 days of treatment, blood samples (100 μl each) were collected at 8:00 AM in the nonfasting state, and plasma metabolic parameters and DPP-4 activity were determined. After 23 days of treatment, samples (250 μl each) were collected at 8:00 AM in the nonfasting state, and plasma total GIP levels, total amidated GLP-1 [GLP-1 (7–36) amide + GLP-1 (9–36) amide], active forms of GLP-1, and DPP-4 concentration were determined. Throughout the chronic study, blood samples were collected from the orbital vein via capillary pipette. All samples were kept on ice until measurement. For measuring plasma levels of total GIP, total amidated GLP-1 [GLP-1 (7–36) amide + GLP-1 (9–36) amide], active forms of GLP-1, and DPP-4 concentration, collected samples were centrifuged (12,000g) at 4°C for 5 min, and plasma samples were kept in a deep freezer (-80°C) until measurement. After 28 days of treatment, all mice were fasted for 17 h, and then the whole gut was isolated for the measurement of GIP and GLP-1 content and gene expression analysis (after a study period of 29 days). Body weight and food consumption were measured at regular intervals. Average food consumption was calculated using the following formula: [(total weight of added food) - (total weight of remaining food)]/experimental day.
Assays for Plasma Metabolic Parameters. Glycosylated hemoglobin levels were analyzed by an HLC-723 G7 (Tosoh Corporation, Tokyo, Japan). Plasma metabolic parameters were determined using an AutoAnalyzer 7080 (Hitachi, Tokyo, Japan). Plasma insulin (Millipore Corporation, Billerica, MA; Shibayagi, Gunma, Japan), total amidated GLP-1 [GLP-1 (7–36) amide + GLP-1 (9–36) amide; Yanaihara, Shizuoka, Japan], total GIP (Millipore Corporation), and active GLP-1 (Millipore Corporation) were determined by an enzyme-linked immunosorbent assay (ELISA).
Assays for DPP-4 Activity and DPP-4 Concentration. To determine the potential of direct inhibitory activity against DPP-4, voglibose, pioglitazone, and a DPP-4 inhibitor, alogliptin (Feng et al., 2007; Moritoh et al., 2008, 2009), were prepared at final concentrations ranging from 10 pM to 100 μM in dimethylformamide (n = 2). In this assay, 10 μl of plasma, which was prepared from 7-week-old male C57BL/6J mice, was used as a DPP-4 enzymatic source; 10 μl of plasma was mixed with 1 μl of compound preparation and 89 μl of assay buffer [25 mM HEPES, 140 mM NaCl, 100 μM H-Gly-Pro-7-amino-4-methylcoumarin (AMC; Bachem, Bubendorf, Switzerland), 1 mg/ml bovine serum albumin]. For determining plasma DPP-4 activity in the chronic study, 10 μl of plasma was mixed with 90 μl of assay buffer. Both assays were reacted for 10 to 30 min at room temperature, and the released AMC was determined fluorometrically using a Fuluoroskan Ascent FL (375-nm excitation and 460-nm emission; Thermo Fisher Scientific, Waltham, MA), and DPP-4 activity was determined with an AMC standard curve. The concentration that caused 50% inhibition (IC50) against DPP-4 was calculated from the enzyme reaction curves using the SAS system version 8.2 (SAS Institute, Cary, NC). Plasma DPP-4 concentrations were determined by ELISA for mouse DPP-4 (R&D Systems, Minneapolis, MN).
Gut Isolation and Measurement of Incretin Content. After 28 days of treatment, the mice were fasted for 17 h and sacrificed with carbon dioxide. The whole gut was isolated, washed with phosphate-buffered saline, and cut into three segments [upper intestine (10 cm below the pylorus), lower intestine (10 cm above the cecum), and colon (5.5 and 4 cm below the cecum for ob/ob and?/+ mice, respectively)]. The tissue was homogenized in acid-ethanol containing 74% ethanol with 0.15 M HCl for the determination of GIP and GLP-1 concentrations. The homogenized tissues were extracted overnight at 4°C and centrifuged at 12,000g for 10 min. The resultant supernatants were then diluted with phosphate-buffered saline containing 1 mg/ml bovine serum albumin and total GIP (Millipore Corporation), and active GLP-1 levels (Millipore Corporation) in the supernatants were determined by ELISA.
Gene Expression Analysis. For quantitative analysis of mRNA expression of Gcg and neurogenic differentiation 1 (Neurod1) and actin, β (Actb), real-time quantitative polymerase chain reaction was performed. Total RNAs of the colon were obtained using Isogen reagents (Wako Pure Chemicals) and purified using an RNeasy 96 Kit with DNase treatment (Invitrogen, Carlsbad, CA). Next, the first strand cDNA was synthesized from 1 μg of total RNA using Super-ScriptIII reverse transcriptase at 60°C (Invitrogen). The Universal Probe Library and polymerase chain reaction primers, which were designed using the Probe Library Assay Design Center (Roche Diagnostics, Basel, Switzerland), were used to assess gene expression. Primer sequences were as follows: Gcg forward, 5′-cacgcccttcaagacacag-3′; Gcg reverse, 5′-gtcctcatgcgcttctgtc-3′ using the universal probe 33; Neurod1 forward, 5′-cgcagaaggcaaggtgtc-3′; and Neurod1 reverse, 5′-tttggtcatgtttccacttcc-3′ using the universal probe 1. For Actb expression, Universal Probe Library Reference Gene Assays were used (Roche Diagnostics). Amplification was performed under the following conditions: initial denaturation at 95°C for 10 s, followed by 40 cycles of denaturation at 95°C for 5 s and annealing and extension at 60°C for 30 s on the ABI PRISM 7900HT Sequence Detector (Applied Biosystems, Foster City, CA) using Premix Ex Taq polymerase (Takara, Kyoto, Japan). A standard curve was generated by amplifying known concentrations of synthetic oligonucleotides, and the copy number of the target gene was calculated. Then, the relative mRNA expression of each gene against Actb was determined. All primers and oligonucleotide were synthesized by Sigma-Aldrich.
Statistical Analysis. Statistical analysis was performed using the SAS system version 8.2 (SAS Institute). To evaluate the effect of voglibose in the 1-day and chronic studies, statistical significance was analyzed using Bartlett's test, which was used for testing the homogeneity of variances, followed by the Williams' test (P > 0.05 by Bartlett's test) or Shirley-Williams test (P ≤ 0.05 by Bartlett's test). To evaluate the effect of pioglitazone in the chronic study, statistical significance was analyzed using the F test, which was used for testing the homogeneity of variances, followed by Student's t test (P > 0.2 by F test) or Aspin-Welch test (P ≤ 0.2 by F test). Williams' test and the Shirley-Williams test were conducted at the one-tailed significance level of 2.5% (0.025), and the other tests were conducted at the two-tailed significance levels of 5% (0.05). All data were presented as mean and S.D.
Results
In Vitro DPP-4 Inhibition Assay for Voglibose and Pioglitazone. Chemical structures of voglibose, pioglitazone, and a selective DPP-4 inhibitor alogliptin are shown in Fig. 1. To investigate whether voglibose and pioglitazone have a direct inhibitory effect against plasma DPP-4 activity, in vitro DPP-4 inhibition assay was performed. Alogliptin exhibited an IC50 value of <10 nM in inhibiting DPP-4 activity derived from plasma of 7-week-old C57BL/6J mice, whereas voglibose and pioglitazone did not inhibit DPP-4 activity in the plasma (Fig. 2).
Effects of 1-Day Treatment with Voglibose on Plasma DPP-4 Activity and Active GLP-1 Levels. To investigate whether short term treatment with voglibose affects plasma DPP-4 activity or stimulates GLP-1 secretion, mice were administered voglibose at a dose of 0.001 or 0.005% in the diet for 1 day (Fig. 3). Acute treatment with either dose of voglibose did not change plasma DPP-4 activity in ob/ob mice (Fig. 3A). On the other hand, plasma active GLP-1 levels were significantly (P ≤ 0.025) increased by 1.6- and 3.4-fold, respectively, in 0.001 and 0.005% voglibose-treated ob/ob mice, compared with vehicle-treated ob/ob mice (Fig. 3B).
Chronic Effects of Voglibose and Pioglitazone on Metabolic Profiles. Seven-week-old ob/ob mice were treated with 0.001% (1.4 mg/kg/day) or 0.005% (6.5 mg/kg/day) voglibose, 0.03% (46.8 mg/kg/day) pioglitazone, or vehicle for 4 weeks, and metabolic parameters were analyzed (Table 1). Average food consumption was significantly (P ≤ 0.025) decreased by 20 and 34% in 0.001 and 0.005% voglibose-treated ob/ob mice, whereas no significant change was observed in pioglitazone-treated ob/ob mice (-7%), compared with vehicle-treated ob/ob mice. After 23 days of treatment, body weight was decreased by 7% (N.S.) and 24% (P ≤ 0.025) in 0.001 and 0.005% voglibose-treated ob/ob mice, respectively, whereas no significant change was observed in pioglitazone-treated ob/ob mice (+2%), compared with vehicle-treated ob/ob mice. After 20 days of treatment, 0.001 and 0.005% voglibose significantly (P ≤ 0.025) decreased glycosylated hemoglobin by 1.6 and 2.6%, respectively, and significantly (P ≤ 0.025) reduced plasma glucose levels by 31 and 56%, respectively, compared with vehicle-treated ob/ob mice. Pioglitazone also significantly (P ≤ 0.01) decreased glycosylated hemoglobin and plasma glucose by 2.4 and 67%, respectively, compared with vehicle-treated ob/ob mice. Plasma triglyceride levels were decreased by 5 (N.S.) and 45% (P = 0.028) in 0.001 and 0.005% voglibose-treated ob/ob mice, respectively, and by 92% (P ≤ 0.01) in pioglitazone-treated ob/ob mice, compared with vehicle-treated ob/ob mice. Although 0.001% voglibose did not decrease plasma total cholesterol levels, 0.005% voglibose and pioglitazone did by 31% (P ≤ 0.025) and 22% (P ≤ 0.01), respectively, compared with vehicle-treated ob/ob mice. Only pioglitazone significantly (P ≤ 0.01) decreased plasma nonesterified fatty acid levels by 53% compared with vehicle-treated ob/ob mice. Treatment with 0.001 and 0.005% voglibose decreased plasma insulin levels by 11% (N.S.) and 81% (P ≤ 0.025), respectively, compared with vehicle-treated ob/ob mice. Pioglitazone also significantly (P ≤ 0.01) decreased plasma insulin levels by 87% compared with vehicle-treated ob/ob mice. Overall effects of 0.005% voglibose and 0.03% pioglitazone on glycemic control were similar in ob/ob mice.
Chronic Effects of Voglibose and Pioglitazone on Plasma Levels of DPP-4 Activity and DPP-4 Concentration. The DPP-4 enzyme plays a critical role in the regulation of incretin activity; thus, changes in plasma DPP-4 activity were investigated. As shown in Fig. 4, vehicle-treated ob/ob mice exhibited elevated plasma DPP-4 activity compared with?/+ mice. It is interesting that treatment with 0.001 and 0.005% voglibose significantly (P ≤ 0.025) decreased plasma DPP-4 activity by 40 and 51%, respectively, compared with vehicle-treated ob/ob mice after 20 days of treatment (Fig. 4). Because voglibose had no direct inhibition against plasma DPP-4 activity in vitro, the reduced plasma DPP-4 activity observed in voglibose-treated ob/ob mice was speculated to be a result of changes in plasma concentration of the DPP-4 protein. Thus, plasma concentration of circulating DPP-4 was measured via ELISA. It is notable that treatment with 0.001 and 0.005% voglibose significantly (P ≤ 0.025) decreased plasma DPP-4 concentrations by 31% (138 ± 11 ng/ml) and 43% (114 ± 14 ng/ml), respectively, compared with vehicle-treated ob/ob mice (202 ± 23 ng/ml), whereas pioglitazone did not (196 ± 15 ng/ml) after 23 days of treatment.
Chronic Effects of Voglibose and Pioglitazone on Plasma Total GIP, Total Amidated GLP-1, and Active GLP-1 Levels. Given the stimulatory effect of voglibose on GLP-1 secretion in humans, plasma incretin profiles were investigated after chronic treatment in ob/ob mice. L-cells are present in the lower gut and secrete intact forms of GLP-1 (7–36) amide and GLP-1 (7–37), both of which have identical incretin activity, into the circulation. These active forms of GLP-1 are metabolized by DPP-4 to inactivated forms of GLP-1 (9–36) amide and GLP-1 (9–37). Thus, by measuring total levels of GLP-1 (7–36) amide plus GLP-1 (9–36) amide, the rate of GLP-1 secretion can be indirectly estimated. After 23 days of treatment, 0.001 and 0.005% voglibose significantly (P ≤ 0.025) decreased plasma GIP levels by 38 and 54%, respectively, compared with vehicle-treated ob/ob mice (Fig. 5A). On the other hand, plasma total amidated GLP-1 levels were increased by 1.3-fold (N.S.) and 1.5-fold (P ≤ 0.025) in 0.001 and 0.005% voglibose-treated ob/ob mice, respectively, compared with vehicle-treated ob/ob mice (Fig. 5B). Furthermore, 0.001 and 0.005% voglibose significantly (P ≤ 0.025) increased plasma active GLP-1 levels by 1.9- and 4.1-fold, respectively, compared with vehicle-treated ob/ob mice (Fig. 5C). Plasma active GLP-1 levels after chronic treatment were higher by 1.4- and 1.5-fold in 0.001 and 0.005% voglibose-treated ob/ob mice, respectively, compared with those after 1 day of treatment. Pioglitazone showed no effect on these parameters.
Chronic Effects of Voglibose and Pioglitazone on Gut Incretin Content. To further characterize the effects of voglibose on incretin profiles in vivo, gut incretin content was determined after 28 days of treatment followed by 17 h of fasting. Consistent with a previous report, GIP content was higher in the upper gut compared with the lower gut, whereas GLP-1 content was higher in the lower gut compared with the upper gut (Fig. 6, A and B). Gut GIP content was slightly increased through the overall gut in ob/ob mice compared with?/+ mice (+11% for upper intestine, +16% for lower intestine in ob/ob mice). Although GLP-1 content in ob/ob mice was lower by 46 and 23% in the upper and lower intestines, respectively, it was higher by 14% in the colon, compared with?/+ mice. It is interesting that treatment with 0.001 and 0.005% voglibose significantly (P ≤ 0.025) decreased GLP-1 content by 20 and 49%, respectively, in the upper intestine, compared with vehicle-treated ob/ob mice (Fig. 6B). In contrast, treatment with 0.001 and 0.005% voglibose significantly (P ≤ 0.025) increased GLP-1 content by 1.5- and 1.6-fold, respectively, in the lower intestine and 1.4- and 1.6-fold, respectively, in the colon, compared with vehicle-treated ob/ob mice (Fig. 6B). Similar to the observed GLP-1 content, treatment with 0.001 and 0.005% voglibose slightly decreased GIP content by 3% (N.S.) and 13% (N.S.), respectively, in the upper intestine and significantly (P ≤ 0.025) increased it by 1.6-fold in the lower intestine, compared with vehicle-treated ob/ob mice (Fig. 6A). Pioglitazone had no significant effect on gut incretin content investigated, except for a slight but significant (P ≤ 0.05) decrease of GIP content in the lower intestine.
Chronic Effects of Voglibose and Pioglitazone on Gcg and Neurod1 Gene Expression in the Colon. Because of the ability of voglibose to increase GLP-1 content in the lower gut, expression of enteroendocrine cell-related genes was measured in the colon. Although gut Gcg gene expression levels in ob/ob mice were similar to those of?/+ mice, 0.001 and 0.005% voglibose significantly (P ≤ 0.025) increased Gcg gene expression in the colon region by 2.6- and 3.1-fold, respectively, compared with vehicle-treated ob/ob mice (Fig. 7A). Besides elevated Gcg gene expression, 0.001 and 0.005% voglibose increased Neurod1 gene expression by 1.3-fold (N.S.) and 1.4-fold (P ≤ 0.025), respectively, compared with vehicle-treated ob/ob mice (Fig. 7B). Pioglitazone had no significant effect on expression of these genes.
Discussion
In the present study, we characterized the effects of the α-GI voglibose on plasma incretin profiles, DPP-4 activity, gut incretin content, and gene expression in the colon of ob/ob mice, an obese rodent model with type 2 diabetes. Voglibose had no direct inhibitory effect against DPP-4 in vitro, and 1-day treatment with voglibose did not change plasma DPP-4 activity in ob/ob mice. On the other hand, 1-day treatment with voglibose increased plasma active GLP-1 levels. When chronically administered for 3 to 4 weeks, voglibose increased total amidated GLP-1 circulation. It is interesting that chronic treatment with voglibose decreased plasma DPP-4 activity by reducing plasma DPP-4 concentration. A subsequent analysis revealed that chronic treatment with voglibose increased gut GLP-1 content, which was associated with elevated Gcg gene expression. After chronic treatment, voglibose increased plasma active GLP-1 levels, compared with those after 1-day treatment in ob/ob mice.
The effect of α-GIs on incretin secretion has been a matter of investigation. With α-GIs, GIP secretion was suppressed, but GLP-1 secretion was enhanced and prolonged in clinical studies (Fukase et al., 1992; Göke et al., 1995; Qualmann et al., 1995; Ranganath et al., 1998; Seifarth et al., 1998; Enç et al., 2001; Lee et al., 2002). In the present study, voglibose did not inhibit DPP-4 activity in plasma in vitro, and 1-day treatment with voglibose did not decrease plasma DPP-4 activity. On the other hand, 1-day treatment with voglibose increased plasma active GLP-1 levels. In addition, plasma total amidated GLP-1 levels (active plus inactive forms of amidated GLP-1) were increased after the chronic treatment with voglibose. These observations indicate that voglibose may have a stimulatory effect on GLP-1 secretion in ob/ob mice. In contrast to the increased GLP-1 circulation, plasma total GIP levels were decreased by voglibose treatment in ob/ob mice. As proposed earlier, a delayed absorption of carbohydrates might be responsible for decreased GIP circulation and increased GLP-1 secretion in this model. Although little is known about the mechanism by which nutrients stimulate GIP and GLP-1 secretion, carbohydrates are effective stimulants for these incretins (Baggio and Drucker, 2007). For GIP secretion, the rate of nutrient absorption rather than the mere presence of nutrients in the intestine seems to be critical (Wachters-Hagedoorn et al., 2006; Baggio and Drucker, 2007). As shown in Fig. 6, GIP- and GLP-1-secreting cells were abundant in the upper and lower guts, respectively, in ob/ob mice. Taken together with the previous observation that the majority of glucose is absorbed in the upper gut under normal feeding conditions (Ferraris et al., 1990), voglibose-induced delayed absorption of carbohydrates may have contributed to attenuated GIP secretion in the upper gut and stimulated GLP-1 secretion in the lower gut in these mice.
Voglibose unexpectedly decreased plasma DPP-4 activity after the chronic treatment in ob/ob mice, and this reduction in plasma DPP-4 activity was associated with decreased plasma DPP-4 protein levels. Serum DPP-4 activity has been correlated positively with glycosylated hemoglobin levels in patients with type 2 diabetes (Mannucci et al., 2005; Ryskjaer et al., 2006), suggesting that glucose levels may influence plasma DPP-4 activity. However, considering that pioglitazone had no effect on plasma DPP-4 activity but strongly decreased glucose levels, glycemic control itself may not be a primary contributor to decreased plasma DPP-4 activity in this model. Thus, the mechanisms by which voglibose decreases DPP-4 circulation remain to be determined in this study.
Studies have shown that the gut incretin content is altered in animal models of type 2 diabetes (Pinto et al., 1995; Berghöfer et al., 1997), although this has not yet been reported in human diabetic patients. For example, GLP-1 content was decreased in the intestinal ileum but was increased in the colon of Zucker diabetic fatty rats compared with normal nondiabetic rats (Berghöfer et al., 1997). Consistent with Zucker diabetic fatty rats, in the ob/ob mice in this study, GLP-1 content was decreased in the upper and lower intestines (-46 and -23%, respectively) but slightly increased in the colon (+14%) compared with normal?/+ mice. It is interesting that chronic treatment with voglibose affected incretin content in each segment of the intestine. Voglibose slightly decreased GIP and GLP-1 content in the upper intestine, whereas it increased these levels in the lower intestine, indicating that the function and/or differentiation of GIP- and GLP-1-producing enteroendocrine cells may have been attenuated in the upper intestine and stimulated in the lower intestine by voglibose.
Voglibose also increased GLP-1 content in the colon, again indicating that voglibose may have contributed to the function and/or differentiation of GLP-1-producing cells in the colon. The subsequent gene expression analysis revealed increased Gcg expression in the colon by voglibose treatment, which was associated with elevated GLP-1 content in the colon. Cani et al. (2007) have reported recently that oligofructose, a dietary nondigestible carbohydrate, promoted L-cell differentiation as evidenced by the increased number of L-cells, elevated Gcg gene expression, and doubled GLP-1 content in the proximal colon and increased levels of portal GLP-1 concentration in rats. Thus, undigested carbohydrates, which were generated by voglibose, may have contributed to increased Gcg gene expression and GLP-1 content in the lower gut, thereby contributing to increased GLP-1 circulation in ob/ob mice. Although relatively little is known about how enteroendocrine cell lineages are determined from their precursor cells, the bHLH class transcription factor Neurod1 is known to be a pivotal regulator for enteroendocrine cell differentiation (Schonhoff et al., 2004). In the present study, voglibose increased Neurod1 expression in the colon, supporting the presumption that voglibose may have increased differentiation of enteroendocrine cells.
Chronic treatment with voglibose resulted in more increased plasma active GLP-1 levels than those after 1 day of treatment. It is highly plausible that increased GLP-1 secretion, decreased plasma DPP-4 activity, and increased GLP-1 content together have contributed to increased active GLP-1 circulation after chronic treatment with voglibose.
GIP plays a more important role in maintaining body weight than GLP-1 in obesity (Miyawaki et al., 2002; Flatt, 2008). In this study, voglibose decreased food intake, plasma levels of glucose, insulin, GIP, and body weight. Acarbose, another α-GI, at doses of 0.05 and 0.5% in the diet had almost the same efficacy in lowering glucose and insulin levels as 0.001 and 0.005% voglibose in the diet in ob/ob mice (data not shown). It is interesting that acarbose-treated ob/ob mice showed unchanged food consumption and more substantial decrease in plasma GIP (64–72% decrease) than voglibose (38–54% decrease) (data not shown). However the decrease in body weight of acarbose-treated ob/ob mice was lower (7–13% decrease) compared with voglibose (7–24% decrease) (data not shown). Taken together, decreased food intake, which may be a result of abdominal symptoms caused by voglibose administration, rather than reduced plasma GIP levels, may play a more dominant role at decreasing body weight in ob/ob mice.
Contrary to its effect on plasma GIP, voglibose had more substantial effects in increasing total amidated GLP-1 and active GLP-1 levels (1.3–1.5-fold and 1.9–4.1-fold, respectively), compared with acarbose (1.1–1.2-fold and 1.5-fold, respectively, by 0.05–0.5% acarbose in the diet; data not shown) in ob/ob mice. Taking into account that voglibose virtually lacks inhibitory activity against α-amylase (Matsuo et al., 1992), the composition of digested carbohydrates may play a role in the modulation of incretin secretion by α-GIs; however, this hypothesis awaits further investigation.
Pioglitazone did not induce any significant effects on incretin-related parameters, except for a slightly decreased GIP content in the lower gut, whereas glycemia was controlled equally to that achieved by voglibose. This fact may eliminate the conjecture that glycemic control itself has the greatest influence over the incretin profile in ob/ob mice.
In conclusion, our study provides a new insight into the mechanisms by which voglibose increases plasma active GLP-1 levels after chronic treatment. Chronic treatment with voglibose stimulated GLP-1 secretion, decreased plasma DPP-4 activity by reducing DPP-4 circulation, increased gut Gcg gene expression, and elevated GLP-1 content, resulting in higher active GLP-1 circulating levels compared with those achieved by 1-day treatment. These results suggest that increasing GLP-1 secretion, decreasing plasma DPP-4 activity, and increasing gut GLP-1 content may likely have contributed to increased circulation of active GLP-1 after chronic treatment with voglibose. Clinical trials may be of interest if the present preclinical results are reproduced in humans.
Acknowledgments
We thank Michelle Kujawski and Nicholas Hird for writing support and comments on the manuscript and Shoichi Asano for technical assistance.
Footnotes
-
This work was supported by Takeda Pharmaceutical Company Limited.
-
Article, publication date, and citation information can be found at http://jpet.aspetjournals.org.
-
doi:10.1124/jpet.108.148056.
-
ABBREVIATIONS: GIP, glucose-dependent insulinotropic polypeptide; GLP, glucagon-like peptide; DPP-4, dipeptidyl peptidase-4; Gcg, glucagon; α-GI, α-glucosidase inhibitor; ELISA, enzyme-linked immunosorbent assay; AMC, 7-amino-4-methylcoumarin; Neurod1, neurogenic differentiation 1; Actb, actin, β.
- Received October 30, 2008.
- Accepted February 9, 2009.
- The American Society for Pharmacology and Experimental Therapeutics