| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
ORIGINAL RESEARCH COMMUNICATION |
1 From the University of Adelaide Discipline of Medicine (AK, RC, SD, JMW, KLJ, MH, and CKR), and the Department of Nuclear Medicine PET and Bone Densitometry (MB and FDB), Royal Adelaide Hospital, Adelaide, South Australia, Australia
2 Supported by a University of Adelaide Faculty of Health Sciences Research Establishment Grant (to CKR). RC's salary was provided by a scholarship from the Thai Government and the Faculty of Medicine, Siriraj Hospital, Mahidol University, Bangkok, Thailand. KLJ's salary was funded jointly by Diabetes Australia and the National Health and Medical Research Council of Australia. 3 Address reprint requests to CK Rayner, Discipline of Medicine, Royal Adelaide Hospital, North Terrace, Adelaide, SA, Australia 5000. E-mail: chris.rayner{at}adelaide.edu.au.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Objective: The aim of the current study was to evaluate the mechanisms by which protein supplementation lowers the blood glucose response to oral glucose.
Design: Nine healthy men were studied on 3 d each in a random order. Subjects consumed 300-mL drinks containing either 50 g glucose (Glucose), 30 g gelatin (Protein), or 50 g glucose with 30 g gelatin (Glucose + Protein) in water labeled with 150 mg [13C]acetate. Blood and breath samples were subsequently collected for 3 h to measure blood glucose and plasma insulin, glucagon-like peptide 1 (GLP-1), and glucose-dependent insulinotropic polypeptide (GIP) concentrations and gastric half-emptying time, which was calculated from 13CO2 excretion.
Results: The blood glucose response was less after Glucose + Protein than after Glucose (P < 0.005); GIP was lower (P < 0.005), and there were no significant differences in plasma insulin or GLP-1. Protein alone stimulated insulin, GLP-1, and GIP (P < 0.05 for each) without elevating blood glucose. The gastric half-emptying time was greater after Glucose + Protein than after Glucose (P < 0.05) and tended to be greater for Glucose than for Protein (P = 0.06).
Conclusions: In healthy humans, the addition of protein to oral glucose lowers postprandial blood glucose concentrations acutely, predominantly by slowing gastric emptying, although protein also stimulates incretin hormones and non-glucose-dependent insulin release.
Key Words: Glucose-dependent insulinotropic polypeptide • glucagon-like peptide 1 • breath test
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The capacity of protein to enhance insulin secretion has been recognized for many years. Gannon et al (7) reported that the addition of 25 g protein (from sources such as meat, fish, or gelatin) to a 50-g oral glucose load reduced the subsequent blood glucose response in type 2 diabetic patients (by 
30% with gelatin), whereas it increased plasma insulin (by 2.5 times with gelatin) when compared with glucose alone.
It is now established that the rate of gastric emptying is a major determinant of postprandial glycemia, so that even modest changes may have a substantial effect on the magnitude and timing of postprandial increases in blood glucose and insulin (8, 9). Gastric emptying is normally regulated at a relatively constant rate of 2–3 kcal/min, predominantly by feedback arising from the small intestine (10, 11). The addition of energy in the form of protein to an oral glucose load would be expected, therefore, to slow gastric emptying of glucose and thereby reduce the glycemic response. In this circumstance, it may be expected that there would be a reduction, rather than an increase, in the plasma insulin response. Previous studies that evaluated the effects of protein supplementation on oral glucose tolerance (7, 12, 13) did not measure the rate of gastric emptying.
Oral glucose stimulates insulin secretion to a much greater degree than does an equivalent intravenous glucose load (14). This "incretin effect" is mediated by the small intestinal peptides glucagon-like peptide 1 (GLP-1) and glucose-dependent insulinotropic polypeptide (GIP). Both carbohydrate and fat are potent stimuli for GLP-1 and GIP secretions(15), but the response to protein is more variable; amino acids (16-18) and casein hydrolysates (19) tend to stimulate GIP and GLP-1 more potently than do intact proteins, at least in healthy humans (15, 20). Strategies to enhance the incretin response are of particular interest in the management of patients with type 2 diabetes, in whom both the secretion of GLP-1 (21) and the ß cell response to GIP (22) appear to be impaired. We have now examined the potential contributions of gastric emptying, incretin peptides, and insulin in mediating the reduction in the glycemic response to oral glucose by the addition of protein, in healthy humans.
|
SUBJECTS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
3 d. Each subject gave written informed consent in accordance with the guidelines of the Declaration of Helsinki. The protocol was approved by the Research Ethics Committee of the Royal Adelaide Hospital.
Protocol
The order of the studies was randomized and single-blinded. An intravenous cannula was placed in a forearm vein 30 min before each study to allow repeated blood samples to be drawn.
On each day, a 300-mL drink was consumed, which contained 50 g glucose ("Glucose" drink; 200 kcal), 30 g powdered gelatin (incorporating 25 g protein; Davis Gelatin; Gelita NZ Ltd, Christchurch, New Zealand; "Protein" drink; 100 kcal), or both glucose and gelatin ("Glucose + Protein" drink; 300 kcal), dissolved in water. Each drink contained 150 mg [13C]acetate and nonnutritive fruit flavoring, was served at 37 °C, and was consumed within 3 min.
Blood samples were collected at –5, 0, 15, 30, 45, 60, 90, 120, 150, and 180 min for measurements of blood glucose and plasma insulin, GLP-1, and GIP concentrations. To calculate the rate of gastric emptying, breath samples were obtained at 5-min intervals from 0 to 60 min and then at 15-min intervals from 60 to 180 min.
Measurements
Blood glucose, plasma insulin, GLP-1, and GIP concentrations
Blood samples for the determinations of plasma insulin, GLP-1, and GIP were collected in ice-chilled tubes containing EDTA and 400 kIU aprotonin (Trasylol; Bayer Australia Ltd, Pymble, Australia) per liter of blood. Plasma was separated by centrifugation and stored at –70 °C for subsequent analysis. Blood glucose concentrations were determined immediately by using a portable glucometer (Medisense Precision QID; Abbott Laboratories, Bedford, MA). The accuracy of this method was confirmed by our laboratory with the use of the hexokinase technique (8). Plasma insulin was measured by enzyme-linked immunosorbent assay (Diagnostics Systems Laboratories Inc, Webster, TX); the intraassay CV was 2.6%, and the interassay CV was 6.2% (23). Plasma GLP-1 concentrations were measured by radioimmunoassay; the intraassay CV was 17%, and the interassay CV was 18% (24). Plasma GIP was also measured by radioimmunoassay; intra- and interassay CVs were both 15% (24).
Gastric emptying
13CO2 enrichment in the breath samples was measured by mass spectroscopy (ABCA 20–20 mass spectrometer; Europa Scientific, Crewe, United Kingdom) to determine the percentage 13CO2 recovery per hour and the cumulative percentage 13CO2 recovery over 3 h (25). The method of Ghoos et al (26) was applied to calculate the gastric half-emptying time and gastric emptying coefficient. Breath tests that use the [13C]acetate label were validated against the gold standard of scintigraphy for the measurement of liquid gastric emptying (27).
Statistical analyses
Data were evaluated by use of analysis of variance for comparisons of peak blood glucose concentrations, gastric half-emptying time, and gastric emptying coefficient. Repeated-measures analysis of variance was used to compare blood glucose and plasma peptide data between 0 and 180 min, with post hoc comparisons in the event of a treatment x time interaction. The difference in the blood glucose increment at 15 min after Glucose + Protein and Glucose was calculated, as was the difference in gastric half-emptying time after these drinks, and the correlation between these variables was examined. Analyses were performed with the use of STATVIEW 5 software (SAS Institute, Cary, NC). P values < 0.05 were deemed significant.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Blood glucose, plasma insulin, GLP-1, and GIP concentrations
Blood glucose, plasma insulin, GLP-1, and GIP concentrations are shown in Figure 1
. Blood glucose concentrations increased from baseline after the Glucose and Glucose + Protein drinks (P < 0.05 for each), but not after the Protein drink, and the blood glucose profile differed significantly between the 3 study days (P < 0.005 for both treatment effect and treatment x time interaction). Blood glucose concentrations were lower after the Glucose + Protein drink than after the Glucose drink (P < 0.05), with a lower peak blood glucose value for Glucose + Protein (8.0 ± 0.4 mmol/L) than for Glucose (9.4 ± 0.3 mmol/L) (P < 0.05). For the drinks containing glucose, blood glucose concentrations fell below baseline values late in the study, so that blood glucose was lower for Glucose + Protein than for Protein at 120 and 150 min and was also lower for Glucose than for Protein at 150 min (P < 0.05 for all).
|
Plasma GLP-1 concentrations increased on all 3 d (P < 0.05), with a relatively early peak followed by a steady decline. There was no significant difference between the 3 drinks.
Plasma GIP concentrations are shown in Figure 1
D. Plasma GIP increased on all 3 d (P < 0.05 for each), and GIP concentrations differed between study days (P < 0.005 for treatment effect; P < 0.001 for treatment x time interaction). Plasma GIP concentrations were greater after Glucose than after each of the other drinks (P < 0.05 for each) and were greater after Glucose + Protein than after Protein (P < 0.05). As for insulin, GIP remained elevated above baseline beyond the return of blood glucose to baseline.
Gastric emptying
Gastric half-emptying time differed between the 3 study days (P < 0.0005) and was greater for Glucose + Protein (51.2 ± 1.4 min) than for either the Glucose (45.6 ± 1.3 min; P < 0.05) or the Protein (42.1 ± 1.3 min; P < 0.005) drinks (Figure 2). Half-emptying time tended to be greater for Glucose than for Protein (P = 0.06). The gastric emptying coefficient tended to differ between the 3 study days (P = 0.05) and on direct comparison was greater for Glucose (3.82 ± 0.07) than for Glucose + Protein (3.61 ± 0.06) (P < 0.05), but it did not differ significantly between Protein (3.98 ± 0.16) and Glucose (P = 0.29) or between Protein and Glucose + Protein (P = 0.07).
|
), there was an inverse correlation between the change in blood glucose increment at 15 min and the change in gastric half-emptying time (r = –0.77, P < 0.05).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Protein did not stimulate additional GLP-1, GIP, or insulin release, and, indeed, plasma GIP concentrations were lower after Glucose + Protein than after the Glucose drink alone. This occurred despite the fact that protein, when consumed alone, stimulated both incretin and insulin release. The increase in plasma insulin after the Protein drink was not associated with any elevation in blood glucose, which suggests that it was mediated by a mechanism other than GLP-1 and GIP, because these peptides stimulate insulin secretion in a glucose-dependent manner (14). A possible alternative mechanism for this effect was the stimulation of ß cells by amino acids (29). The fact that the blood glucose did not decrease in response to the release of insulin could be explained by concurrent glucagon release, although we did not measure this.
The loads of glucose and protein administered in this study were chosen to match those evaluated previously in patients with type 2 diabetes (7). The observed lowering of glycemia after the addition of protein to the glucose drink is in accordance with the report of Gannon et al (7). In contrast with our findings, these authors observed a 2.5-fold stimulation of insulin secretion. However, the insulin response to the glucose drink alone in their type 2 diabetic patients was markedly less than in our healthy subjects. The same investigators then studied the effects of adding glycine (12) or proline (13), both of which are found in gelatin, to oral glucose in healthy volunteers and reported a reduction in blood glucose but no increment in the insulin response. These observations are consistent with the evidence that type 2 diabetes is associated with an impaired early-phase insulin secretion to oral glucose and a diminished glucose-sensing ability of the ß cell, whereas the insulin responses to other stimuli may be relatively intact (30, 31), at least in the early stages of the disease.
Several groups have investigated the insulin response to ingested protein and found that not only the presence of diabetes, but also the relative quantities of protein and carbohydrate and the presence of free amino acids (as opposed to intact protein) are important variables. When ingested in isocaloric quantities, glucose and protein from lean beef have additive effects on insulin secretion in healthy subjects (32), whereas if the amount of protein is less than that of glucose, the increase in insulin is substantially less than additive (33). Furthermore, the insulin response correlates with plasma amino acid concentrations; the insulin response to intact protein is less than that observed with protein hydrolysates, which presumably reflects the more rapid increase in plasma amino acid concentrations with the latter (34).
Little information is available on the effects of supplementing carbohydrate with protein on the incretin response. Frid et al (35) reported that the addition of whey protein to carbohydrate in type 2 diabetic patients increased the insulin response and decreased the glycemic profile more than did the addition of lean ham. The GIP response was greater with whey (area under the curve increased by 30% over 120 min), but GLP-1 concentrations did not significantly differ. Gunnarsson et al (36) recently reported that in gastric gavage-fed mice, the insulin response to glucose (ie, area under the curve) was trebled by the addition of an equal weight of whey protein, and this was associated with a substantial decrease in plasma glucose. The total plasma GLP-1 concentration increased, but GIP did not. However, analysis of intact (or active) incretins indicated an increase in intact GIP, and the authors suggested that fragments of whey protein could inhibit dipeptidyl peptidase IV, the enzyme that degrades the incretin peptides, in the small intestine. As in our study, gastric emptying was slowed by the addition of whey protein to glucose. The fact that the addition of protein in the present study did not increase plasma GLP-1 and decreased GIP is likely to reflect the relatively smaller load and the nature of the protein used. Although we did not measure intact, as opposed to total, GIP, gelatin may be less easily digested to fragments that block dipeptidyl peptidase IV than is whey protein.
We evaluated the acute effects of modest supplementation with protein in healthy individuals. Acute and longer-term studies would be of interest in patients with type 2 diabetes to evaluate the effect of protein supplementation on glycemic control, and the load and type of protein could be optimized for a greater incretin response (34). An alternative approach, eg, the ingestion of a protein "preload" at a given interval before meals, could also be evaluated to determine whether the stimulation of incretins and insulin could improve glycemia after the subsequent meal (37).
In summary, we confirmed the capacity of protein supplementation to improve the glycemic response to glucose in healthy humans, and we showed that the slowing of gastric emptying makes an important contribution to this effect. Further studies aimed at optimizing this phenomenon are indicated in patients with type 2 diabetes, who may derive additional benefit, in terms of postprandial glycemia, from the stimulation of non–glucose-dependent insulin secretion.
|
ACKNOWLEDGMENTS |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  R. Hageman, C. Severijnen, B. J. M. van de Heijning, H. Bouritius, N. van Wijk, K. van Laere, and E. M. van der Beek A Specific Blend of Intact Protein Rich in Aspartate Has Strong Postprandial Glucose Attenuating Properties in Rats J. Nutr., September 1, 2008; 138(9): 1634 - 1640. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
), 25 g oral protein (Protein drink; 
), and 50 g oral glucose with 25 g oral protein (Glucose + Protein drink; •). n = 9 subjects. Blood glucose, plasma insulin, and plasma GIP differed significantly between study days (P < 0.005 for all, repeated-measures ANOVA treatment effects and treatment x time interactions), but there was no significant difference in GLP-1 between treatments. 
Significant difference between Glucose + Protein and Glucose drinks, P < 0.05. *Significant difference between Glucose and Protein drinks, P < 0.05. 
Significant difference between Glucose + Protein and Protein drinks, P < 0.05.
Significant difference between Glucose + Protein and Protein drinks, P < 0.005.