| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Other Original Articles |
Department of Medicine (S.E.K., B.M., W.H.), Division of Metabolism, Endocrinology, and Nutrition, VA Puget Sound Health Care System and University of Washington, Seattle, Washington 98108; and Novartis Pharmaceuticals Corp. (M.L.-S., C.-H.H., D.D., J.F.M., A.H., J.E.F.), East Hanover, New Jersey 07936
Address all correspondence and requests for reprints to: Steven E. Kahn, M.B., Ch.B., VA Puget Sound Health Care System (151), 1660 South Columbian Way, Seattle, Washington 98108. E-mail: skahn{at}u.washington.edu
Abstract
Insulin secretion is impaired in type 2 diabetes with the early response being essentially absent. The loss of this early insulin secretion is hypothesized to be important in the deterioration of glucose tolerance. To determine whether enhancement of the early-phase insulin response can enhance glucose tolerance, we administered 1) 120 mg nateglinide, an insulinotropic agent that enhances early insulin secretion; 2) 10 mg glyburide, which enhances the later phases of insulin secretion; or 3) placebo in random order to 21 subjects with type 2 diabetes (14 males and 7 females; aged 59.2 ± 2.1 yr, x ± SEM; body mass index 29.7 ± 1.0 kg/m2; fasting plasma glucose 8.1 ± 0.1 mM). ß-Cell function was quantified as the incremental area under the curve for different time periods for the 5 h following iv glucose administration and glucose tolerance as the glucose disappearance constant (Kg) from 10 to 60 min. Insulin release commenced immediately after nateglinide administration, even before glucose injection, but this was not observed with glyburide. Both nateglinide and glyburide enhanced glucose-induced insulin release, compared with placebo (area under the curve -15–300 min: nateglinide 23,595 ± 11,212 pM/min, glyburide 54,556 ± 15,253 pM/min, placebo 10,242 ± 2,414 pM/min). The profiles of insulin release demonstrated significant enhancement of release between -15 and 30 min for nateglinide, compared with glyburide and between 60 and 300 min for glyburide over nateglinide. Kg increased by 15% with nateglinide (0.87 ± 0.04%/min), but it did not increase significantly with glyburide (0.79 ± 0.04%/min), compared with placebo (0.76 ± 0.04%/min). The enhancement of insulin release by glyburide resulted in a lower minimal glucose concentration with glyburide (3.8 ± 0.2 mM), compared with nateglinide (5.0 ± 0.2 mM) and placebo (5.9 ± 0.2 mM). Thus, enhancement of the early phase of insulin secretion improves iv glucose tolerance, whereas delaying it by 30 min results in a slower rate of glucose disappearance for the first 2 h after iv glucose administration. Further, the differences in the kinetics of nateglinide and glyburide action results in continued insulin release with glyburide despite the fact that glucose levels have returned to basal, thus resulting in a further reduction in glucose levels and a lower nadir.
IMPAIRMENTS IN ß-CELL function, which include alterations in the dynamics of insulin release, are an essential feature of type 2 diabetes (1). When considering these changes, it has been clearly demonstrated that the early phase of insulin release is reduced in response to both oral and iv glucose administration in subjects with this disease (2, 3, 4, 5). In studying these phenomena, a large number of studies have used approaches involving iv administration of secretagogues, which obviates the role of the gastrointestinal tract. If the oral route is used, interpretation of insulin secretion data is more complex because variables such as the rate of glucose absorption, neural activation, and incretins come into play.
With the iv approach, two distinct phases of insulin secretion are apparent in individuals with normal glucose tolerance (6, 7). The reduction in the very early insulin response to glucose, also known as the first-phase insulin response, is more or less absent in individuals with type 2 diabetes (4, 8). This response is glucose level independent (9) and can be demonstrated to be absent before the diagnosis of the disease (10). In contrast, the later or second-phase of glucose-induced insulin secretion is glucose level dependent such that when determined at the prevailing glucose level, the insulin response often appears similar between individuals with and without type 2 diabetes (11). However, when the glucose level is matched, the second-phase insulin response is diminished in subjects with type 2 diabetes (2), with the magnitude of this reduction being related to the degree of hyperglycemia (12). This reduction in insulin release has also been demonstrated when examining the effect of glucose to potentiate the ß-cell’s response to nonglucose secretagogues such as amino acids, incretins, ß-agonists, and sulfonylureas (11, 13).
Insulin release is an important determinant of glucose tolerance. In keeping with this observation, we found in a cross-sectional study of a group of apparently healthy subjects that the early insulin response to glucose is an important determinant of iv glucose tolerance (14). We have also recently demonstrated cross-sectionally that the magnitude of the early-phase response measured 30 min after oral glucose ingestion is an important determinant of the subsequent increment in plasma glucose (5). By studying groups of subjects with normal, impaired, and diabetic glucose tolerance, we found that the relationship between this early-phase response and glucose tolerance is nonlinear in nature (5). The nature of this relationship would predict that in individuals with impaired glucose tolerance and diabetes, small differences in this early response could have dramatic effects on overall glucose disappearance. These observations also suggest that the effect of a similar increase in secretion on glucose lowering may be less effective if it occurs later.
The development of the D-phenylalanine derivative nateglinide (15), an insulinotropic agent with a rapid onset of action, provided us with an opportunity to examine the question of whether enhancement of the early phases of insulin secretion results in an improvement in glucose tolerance. On the basis of the cross-sectional observation that small changes in insulin secretion could have dramatic effects in subjects with reduced glucose tolerance, we chose to study subjects with type 2 diabetes with mild hyperglycemia to determine whether increasing early insulin secretion is associated with an improvement in iv glucose tolerance. To create the experimental scenario in which late insulin secretion is preferentially stimulated, we used the sulfonylurea glyburide. This agent is known to have a slower onset of action than nateglinide, despite the fact that the mechanisms of action of both include binding to and depolarization of the ß-cell ATP-sensitive potassium channel (16, 17).
Subjects and Methods
Study design
The study was a randomized, placebo-controlled, open-label, crossover design in which subjects were randomized to receive five different treatments separated by at least 7 but not more than 14 d. Treatments comprised oral administration of 120 mg nateglinide, 10 mg glyburide, or placebo followed by iv glucose or saline administration. In the case of the active drugs, each was administered twice, once with glucose and once with saline. With the placebo, only glucose was given.
Subjects
The study group comprised 21 individuals (14 males and 7 females) with type 2 diabetes who were not receiving medications to control their hyperglycemia. All participants had been diagnosed as having diabetes at least 6 months before being studied or, if more recent than that, were glutamic acid decarboxylase-antibody negative. In addition, to qualify for the study, they had to be in otherwise good health with a fasting plasma glucose concentration between 7 and 10 mM or a fasting plasma glucose level between 6.1 and 7 mM if accompanied by a 2-h plasma glucose level >11.1 mM during an oral glucose tolerance test.
All subjects received iv glucose plus medication or placebo, and 20 also underwent the studies involving nateglinide and glyburide administration with saline.
All subjects gave written informed consent to participate in the study, which had been reviewed and approved by the Human Subjects Review Committee at the University of Washington.
Study assessments
All studies were performed following a 10-h overnight fast during which the subjects also refrained from smoking. On each study day, each subject received a single oral dose of study medication (120 mg nateglinide, 10 mg glyburide, or placebo). These medications were then followed by a single iv bolus dose of either a 50% glucose solution at a dose of 300 mg/kg body weight or an equal volume of saline administered over 2 min. These injections were commenced 15 min after nateglinide or 30 min after glyburide administration.
All blood samples were collected through a catheter that was indwelling in a forearm vein. The forearm was heated to ensure arterialization of the samples (18). These samples were obtained at -30, -15, 0, 2, 3, 4, 5, 6, 8, 10, 15, 20, 30, 40, 50, 60, 90, 120, 150, 240, and 300 min relative to glucose or saline administration and were immediately placed on ice before separation. All plasma samples were stored at -20 C until assayed for glucose and insulin. On the glyburide/glucose study day, two subjects experienced an episode of symptomatic hypoglycemia between 150 and 240 min after glucose administration that required the discontinuation of the study. One of these subjects also experienced symptomatic hypoglycemia between 240 and 300 min on the glyburide/saline study day that necessitated terminating the study early. They were given orange juice, and no further samples were drawn for pharmacodynamic measurements.
Assays
Plasma glucose concentrations were determined using a glucose oxidase method, and plasma immunoreactive insulin was determined by RIA.
Calculations and statistical analysis
The change in insulin responses from predose was calculated as the incremental area under the curve using the trapezoidal rule. The Kg was determined as the natural log of the glucose concentrations vs. time from samples drawn between 10 and 60 min following glucose injection. The time point at which the glucose level returned to the basal (predose) value was calculated by interpolation.
Data are presented as mean ± SEM. Predose
measurements for each individual were compared across the five periods
to test for carryover effect. Day-to-day variability was assessed using
the coefficient of variation of replicate analysis (19).
Treatment comparisons were performed by ANOVA followed by a
t test to examine for differences among groups. As stated,
the treatment groups were considered separately except for the
assessment of the effect of nateglinide and glyburide on
insulin release before glucose/saline administration when the data from
the two nateglinide study days were pooled as were those from the two
glyburide study days. A P 
0.05 was
considered significant.
Results
Demographics
The mean age of the subjects participating in the study was 59.2 ± 9.6 yr. They had a mean body mass index of 29.7 ± 4.5 kg/m2 with a mean fasting plasma glucose on the first study day of 8.1 ± 1.1 mM (range 6.5–10.2 mM) and a glycosylated hemoglobin of 7.0 ± 0.8% (range 5.6–8.6%).
Stimulation of insulin release before glucose/saline administration by nateglinide and glyburide
Administration of nateglinide 15 min and glyburide 30 min before the injection of glucose or saline resulted in ß-cell stimulation and the release of insulin. This effect was greater with nateglinide being 367 ± 68 pM/min in the 41 individual studies performed on the two nateglinide study days, compared with 11 ± 32 pM/min in these same subjects on the two glyburide study days and 39 ± 9 pM/min on the study day when 20 subjects received placebo. The large difference in the response following nateglinide administration, compared with the lack of a response to glyburide, despite the fact that nateglinide was administered at -15 min while glyburide was administered at -30 min, is compatible with nateglinide having a more rapid onset of action.
Stimulation of insulin release following glucose/saline administration by nateglinide and glyburide
The profiles of the insulin responses to glucose and saline
following nateglinide, glyburide, and placebo
administration are illustrated in Fig. 1
, A and B. Glucose administration in the presence of nateglinide was
associated with an enhancement of insulin release at earlier time
periods than observed with either glyburide or placebo. On
the other hand, glyburide enhanced insulin secretion
during the later time periods of the test. As would be anticipated, the
response to glucose in the presence of placebo was lower than that of
either nateglinide or glyburide. As expected for agents
that bind to the ATP-sensitive potassium channel in the ß cell,
nateglinide and glyburide also stimulated insulin
secretion in the absence of glucose administration (Fig. 1B). The
dynamics of this effect were similar to that observed with glucose in
that nateglinide was associated with an enhancement of early insulin
release whereas glyburide increased the magnitude of the
insulin response at a later stage.
|
, in the defined time periods of 0–10,
10–30, 30–60, 60–120, and 120–300 min, the mean incremental areas
for the different time periods was in keeping with nateglinide
stimulating the release of insulin earlier than glyburide,
which enhanced insulin release at later times following glucose
injection. During the second half-hour following glucose
administration, the responses to glucose were similar on both the
nateglinide and glyburide study days. An insulin response
to glucose could also be detected on the day that placebo was
administered, but this was less than that observed on the study days
when active medication was administered. Over the duration of the
sampling period from -15 to 300 min, the incremental insulin response
was 2.3-fold greater with glyburide than with nateglinide
and 5.3-fold greater than with placebo.
|
Effect of nateglinide and glyburide administration on glucose disappearance
The profiles for glucose disappearance following glucose
administration are illustrated in Fig. 2, A and B and demonstrate the expected exponential decline in glucose
levels over time. The effect of nateglinide and glyburide
on iv glucose tolerance was quantified only on the study days on which
glucose was administered. Intravenous glucose tolerance, assessed as
the Kg up to 1 h following glucose injection, increased on the
study day when nateglinide was administered (Fig. 3). This measure was 0.87 ±
0.04%/min with nateglinide, compared with 0.79 ± 0.04%/min with
glyburide (P < 0.05 vs.
nateglinide) and 0.76 ± 0.04%/min with placebo
(P < 0.005 vs. nateglinide).
|
|
Because glucose tolerance was enhanced with nateglinide and insulin release in the presence of this agent diminished as glucose levels declined while insulin release continued with glyburide despite the fact that glucose levels were declining, the minimum glucose concentration achieved differed with the different treatments being lower with glyburide administration (3.8 ± 0.2 mM) than with either nateglinide (5.0 ± 0.2 mM; P < 0.001) or placebo (5.9 ± 0.2 mM; P < 0.001). The times taken to reach these minimum glucose concentrations were 230 ± 15 min with nateglinide, compared with 270 ± 15 min with glyburide (P < 0.05) and 281 ± 9 min with placebo (P < 0.05).
Reproducibility of study days
To ensure that there was no carryover effect or difference in glucose stimuli between study days, we determined the fasting glucose and insulin levels before and the mean increment in the glucose concentration 10 min following commencement of the glucose injection. The fasting plasma glucose levels on the nateglinide/glucose, glyburide/glucose, placebo/glucose, nateglinide/saline, and glyburide/saline days were 8.1 ± 0.2, 8.1 ± 0.2, 8.1 ± 0.3, 8.0 ± 0.2, and 8.2 ± 0.3 mM, respectively, and the plasma insulin levels were 126 ± 13, 133 ± 16, 122 ± 28, 129 ± 15, and 141 ± 20 pM, respectively. When glucose mixing was complete at 10 min following the commencement of the glucose injection, the mean increment in the glucose concentration was 8.8 ± 0.2, 8.5 ± 0.2, and 8.7 ± 0.2 mM for the nateglinide, glyburide, and placebo treatments, respectively. Day-to-day variability quantified as the coefficient of variation of replicate analysis was 10.3% for fasting plasma glucose, 15.3% for fasting plasma insulin, and 9.1% for the mean increment in glucose. Thus, the subjects would appear to have been studied under near identical conditions.
Discussion
We have used two different agents to stimulate insulin secretion from the ß-cell, nateglinide, which enhances insulin secretion early, and glyburide, which enhances secretion late, as a means to test the impact of early insulin secretion on iv glucose tolerance. We have found that enhancing the early insulin response is associated with an improvement in glucose tolerance, which results in it taking less time for the glucose concentration to return to the level prevailing before glucose administration. Interestingly, the greatest consequence of enhancing early vs. later insulin secretion with these agents is that the total amount of insulin secreted over the 5-h observation period is more than twice as much when this enhancement occurred later rather than earlier. All of this difference in the amount of insulin released came during the last 3 h of the observation period when glucose levels had returned to the levels present before the iv bolus, resulting in a nadir that brings subjects into the hypoglycemic range.
Both insulin secretion and insulin sensitivity are important to glucose disposal (14). The data from the present study demonstrate that even in the presence of insulin resistance, which is likely to be present in these subjects because they have type 2 diabetes (20), increases in early insulin output can have profound effects on glucose tolerance. Thus, the 15% increase in glucose disappearance within the first hour following glucose administration occurred in the face of a 246% increase in insulin secretion with nateglinide over that with placebo during the time interval of -15 min to 30 min. Similarly, insulin release with glyburide increased by 143% over the same time period, but this was not associated with an improvement in glucose disappearance during the first hour. There was a similar enhancement of insulin secretion with both agents between 30 and 60 min following glucose administration. This failure to observe an increase in glucose disappearance with glyburide despite the comparable increase in insulin release with nateglinide between 30 and 60 min after the iv glucose bolus may be owing in part to the fact that there is a delay between insulin release and action because of the need for insulin to traverse the endothelial barrier into the interstitial space (21, 22). It must be noted that in making this comparison, we have assumed that insulin sensitivity did not improve following administration of the nateglinide or glyburide 15 and 30 min before glucose, respectively. This assumption seems reasonable because these agents appear to act directly on the ß cell, and to our knowledge no direct acute effects in the peripheral tissues have been reported. Furthermore, because fasting insulin is a good surrogate marker of insulin sensitivity (23) and the fasting insulin levels did not differ on the different study days, it seems reasonable to conclude that insulin sensitivity was similar in the subjects on the different study days.
When considering the effect of this enhancement of the early phase of insulin secretion to improve glucose tolerance, it may be argued that this effect is relatively small. However, one must consider that glucose disappearance is not solely dependent on insulin action, which is mediated by the combined effect of insulin sensitivity in the liver and peripheral tissues and the level of circulating insulin released from the ß cell. Rather, the effect of glucose to mediate its own disposal independent of insulin by enhancing uptake in the peripheral tissues and suppressing hepatic glucose production, also known as glucose effectiveness, is an important determinant of glucose tolerance and accounts for a major proportion of glucose disposal (24). Thus, although in the present study the quantified increase in glucose disappearance owing to the enhancement of early insulin release was about 15%, because glucose effectiveness is not likely to have been changed by the interventions, this increase represents a larger proportionate increase in the insulin-mediated component.
The findings in this study are in keeping with the observations we made in a group of 93 apparently healthy subjects in whom we evaluated the importance of ß-cell function, insulin sensitivity, and glucose effectiveness on glucose disappearance (14). In that study we found that both insulin-independent and insulin-dependent factors are important in determining the rate of glucose disappearance. Glucose effectiveness was an important determinant accounting for 49% of the variance in the Kg when this latter measure was quantified from 10 to 19 min following glucose injection. Insulin action, which represents the net effect of insulin to enhance glucose disposal, was quantified as the product of insulin sensitivity and insulin secretion, a measure also known as the disposition index, accounted for 55% of the variance (25). Because insulin action is dependent on both insulin sensitivity and ß-cell function, we also examined these two factors separately and found that whereas insulin secretion was a significant determinant of iv glucose tolerance, insulin sensitivity was not. These data thus further support the findings of the present study in that they demonstrate that the early disappearance of glucose following iv glucose administration is critically dependent on the ß cell being capable of releasing insulin. It is also important to recognize that in our study in healthy subjects, we quantified glucose disappearance over the first 19 min following glucose injection, compared with 60 min in the present analysis. Thus, because insulin action is delayed because it is dependent on insulin transport across the endothelium (21, 22), it is likely that insulin action and particularly insulin secretion may become even more important at later time points following glucose administration.
We have recently observed that the relationship between oral glucose tolerance and the early phase of insulin secretion is nonlinear in nature (5). The implication of this relationship is that under circumstances where ß-cell function is compromised, as observed in impaired glucose tolerance and diabetes, small changes in insulin release can have dramatic effects on glucose tolerance. On the other hand, in individuals with normal glucose tolerance and ß-cell function, the impact of changes in insulin secretion are not likely to have a marked effect on overall glucose metabolism. Although the findings of the present study cannot be directly extrapolated to that expected with the oral glucose tolerance test, they are compatible with the observations in the cross-sectional study and imply that restoration of early insulin secretion during the oral glucose tolerance test will enhance glucose tolerance. In fact, another study in which nateglinide enhanced early insulin secretion following oral glucose loading was associated with an improvement in glucose tolerance (26). Further support for this concept comes from two studies using insulin administration in type 2 diabetes. In the first, infusion of insulin over different time periods following nutrient ingestion demonstrated that the earlier the insulin was presented, the more physiologic was the postprandial glucose rise (27). The second study demonstrated that a rapid acting insulin analog reduced the glucose excursion following glucose ingestion by producing a greater suppression of hepatic glucose production, compared with one that had kinetics that were associated with a slower onset of action (28).
In this study, we used nateglinide and glyburide as tools to produce different insulin profiles so that we could examine the effects of these differences on glucose disappearance. The study was not designed as a comparator of the pharmacokinetics of the two medications; therefore, we have not excluded the possibility that the dynamics of the insulin response to glyburide would have been different if we had delayed glucose administration for a longer period of time following glyburide ingestion.
In conclusion, we have demonstrated that partial restoration of the early phase of insulin secretion is associated with an enhancement of iv glucose tolerance. This effect appears to be related to the early phase of insulin release, as an enhancement of the later stages of insulin release does not have a similar glucose lowering effect. Thus, targeting the earlier phases of insulin secretion may be a very valuable approach in attempts to improve overall glucose disappearance in individuals with disturbances of glucose tolerance.
Acknowledgments
We thank the staff of the Diabetes Research Group for their help with the performance of these studies.
Footnotes
This work was supported in part by NIH Grants DK-02654 and DK-17047, the Medical Research Service of the Department of Veterans Affairs, and a grant from Novartis Pharmaceuticals Corp.
Abbreviations: Kg, Glucose disappearance constant.
Received July 30, 2001.
Accepted September 12, 2001.
References
This article has been cited by other articles:
 |
|
 |
|
  K. L. Teff Visceral Nerves: Vagal and Sympathetic Innervation JPEN J Parenter Enteral Nutr, September 1, 2008; 32(5): 569 - 571. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Gustavsson, Y. Lao, A. Maximov, J.-C. Chuang, E. Kostromina, J. J. Repa, C. Li, G. K. Radda, T. C. Sudhof, and W. Han Impaired insulin secretion and glucose intolerance in synaptotagmin-7 null mutant mice PNAS, March 11, 2008; 105(10): 3992 - 3997. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Tuomi, E. H. Honkanen, B. Isomaa, L. Sarelin, and L. C. Groop Improved Prandial Glucose Control With Lower Risk of Hypoglycemia With Nateglinide Than With Glibenclamide in Patients With Maturity-Onset Diabetes of the Young Type 3 Diabetes Care, February 1, 2006; 29(2): 189 - 194. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Abletshauser, P. Brunel, K.-H. Usadel, and M. Hanefeld Effect of nateglinide and glimepiride in reducing postprandial hyperglycaemia in patients with type 2 diabetes mellitus The British Journal of Diabetes & Vascular Disease, March 1, 2005; 5(2): 93 - 99. [Abstract] [PDF]
|
|
|
|
 |
|
 M. E. Doyle and J. M. Egan Pharmacological Agents That Directly Modulate Insulin Secretion Pharmacol. Rev., March 1, 2003; 55(1): 105 - 131. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. Ferreira, E. Filiputti, V. C. Arantes, L. F. Stoppiglia, E. P. Araujo, V. Delghingaro-Augusto, M. Q. Latorraca, M. H. Toyama, A. C. Boschero, and E. M. Carneiro Decreased Cholinergic Stimulation of Insulin Secretion by Islets from Rats Fed a Low Protein Diet Is Associated with Reduced Protein Kinase C{alpha} Expression J. Nutr., March 1, 2003; 133(3): 695 - 699. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. F. Carroll, A. Izard, K. Riboni, M. R. Burge, and D. S. Schade Control of Postprandial Hyperglycemia: Optimal use of short-acting insulin secretagogues Diabetes Care, December 1, 2002; 25(12): 2147 - 2152. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Rosenstock, S. G. Shen, M. R. Gatlin, and J. E. Foley Combination Therapy With Nateglinide and a Thiazolidinedione Improves Glycemic Control in Type 2 Diabetes Diabetes Care, September 1, 2002; 25(9): 1529 - 1533. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. M. Cohen and B. A. Ramlo-Halsted How Do the New Insulin Secretagogues Compare? Diabetes Care, August 1, 2002; 25(8): 1472 - 1473. [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
), glyburide 10 mg (O), or placebo
(
) or saline injection (B) together with nateglinide 120 mg (
) or
glyburide 10 mg (X) in subjects with type 2 diabetes.
Nateglinide and glyburide were administered orally 15 and
30 min before glucose or saline administration. Data are mean ±
SEM.
