With the increasing prevalence of obesity and a possible association with increasing sucrose consumption, nonnutritive sweeteners are gaining popularity. Given that some studies indicate that artificial sweeteners might have adverse effects, alternative solutions are sought. Xylitol and erythritol have been known for a long time and their beneficial effects on caries prevention and potential health benefits in diabetic patients have been demonstrated in several studies. Glucagon-like peptide-1 (GLP-1) and cholecystokinin (CCK) are released from the gut in response to food intake, promote satiation, reduce gastric emptying (GE), and modulate glucose homeostasis. Although glucose ingestion stimulates sweet taste receptors in the gut and leads to incretin and gastrointestinal hormone release, the effects of xylitol and erythritol have not been well studied. Ten lean and 10 obese volunteers were given 75 g of glucose, 50 g of xylitol, or 75 g of erythritol in 300 ml of water or placebo (water) by a nasogastric tube. We examined plasma glucose, insulin, active GLP-1, CCK, and GE with a [13C]sodium acetate breath test and assessed subjective feelings of satiation. Xylitol and erythritol led to a marked increase in CCK and GLP-1, whereas insulin and plasma glucose were not (erythritol) or only slightly (xylitol) affected. Both xylitol and erythritol induced a significant retardation in GE. Subjective feelings of appetite were not significantly different after carbohydrate intake compared with placebo. In conclusion, acute ingestion of erythritol and xylitol stimulates gut hormone release and slows down gastric emptying, whereas there is no or only little effect on insulin release.
- gastric emptying
obesity has increased significantly worldwide (7). Sugar consumption, in the form of sucrose or high-fructose corn syrup (HFCS), has contributed partly to the dramatic rise in obesity, metabolic syndrome, and diabetes (15, 35). Research on the effects of dietary sugars on health has focused recently on fructose, given the striking parallel increases in obesity and in fructose intake over the past decades (5). Fructose intake in diets originates mostly from sucrose (containing 50% fructose and 50% glucose) and soft drinks containing high-fructose corn syrup (HFCS) (39). Patients with nonalcoholic fatty liver disease consume twofold more calories of HFCS from beverages than healthy patients (26). The increasing evidence of the detrimental role of sucrose and fructose justifies a reduction in intake and substitution of sugar by alternative dietary sweeteners. However, several human- and animal-based studies have reported that chemically originated sugar substitutes or artificial, nonnutritive sweeteners (including saccharine, aspartame, neotame, sucralose, and acesulfame-K) have either short- or long-term side effects (2, 38).
Xylitol and erythritol are sweeteners found naturally in low concentrations in fruits and vegetables and can be extracted from fibrous material such as birch. In particular, xylitol has gained popularity, as several studies were able to show a dental caries preventive effect, which was also demonstrated for erythritol (13). Apart from the proven anticariogenic properties, xylitol seems to be effective in reducing the accumulation of visceral fat, and in animal models xylitol improves glycemia (1, 6, 16, 27). Polyol metabolism requires little or no insulin (20, 33). The effects in animal studies include antidiabetic properties such as improved pancreatic islet morphology and blood glucose-lowering effects in heathy and diabetic rats (17, 27). In pilot studies of patients with diabetes, daily intake of 36 g of erythritol resulted in improvement of endothelial function and reduced central aortic stiffness (9). Taken together, these studies support the concept that polyols, especially erythritol, might be attractive nonnutritive sweeteners for the dietary management of diabetes mellitus. Appropriately used, these products might be helpful in both weight management and glycemic control. In conclusion, there is emerging evidence to indicate a beneficial role for dietary polyols in modulating either insulin release or related factors, including gut hormones and attenuating factors associated with the metabolic syndrome, and other potential health benefits warrant further investigation (20).
In 1987, Shafer et al. (34) showed that gastric emptying of a solid meal was markedly prolonged if 25 g of xylitol had been ingested prior to a meal. Shafer could also show that a preload of 25 g of xylitol significantly suppressed subsequent food intake from a buffet compared with a placebo preload or 250 g of aspartame, both of which had no effect at all (34). A decrease in gastric emptying after ingestion of a 30-g xylitol solution was also shown by scintigraphy in 1989 by Salminen et al. (32). In this study, the investigators also measured GIP, insulin, and motilin and demonstrated that xylitol leads to motilin secretion but no GIP release. However, temporal correlation with gastric emptying and other important satiation hormones such as glucagon-like peptide-1 (GLP-1) and cholecystokinin (CCK) were not measured (32). No data describing the effect of erythritol on incretins and gastric emptying were found.
The aim of this study was to examine the effects of these two naturally occurring, nonnutritive sweeteners on incretin release and gastric emptying.
MATERIALS AND METHODS
The protocol was approved by the Ethics Committee of Basel, Switzerland (EKNZ: 2014/072), and conducted in accordance with the principles of the Declaration of Helsinki of 1975 as revised in 1983. Subjects were recruited by word of mouth over a period of 4 mo (from February to May 2014). All patients gave written informed consent. The trial is registered in the Clinical Trials Registry of the National Institutes of Health (NCT 02563847) and was funded by the Swiss National Science Foundation (Marie Heim-Voegtlin subsidy: PMPDP3-145486/1).
A total of 10 lean (mean BMI 21.7 ± 0.5 kg/m2, range 19.9–24.3 kg/m2, 5 men and 5 women; mean age 24.6 ± 0.2 yr, range 24–26 yr) and 10 obese (mean BMI 40.0 ± 1.4 kg/m2, range 33.8–48.2 kg/m2, 5 men and 5 women; mean age 27.2 ± 2.8 yr, range 20–48 yr) volunteers were recruited. Inclusion criteria were general good health, age between 18 and 50 yr and BMI <18 and >25 kg/m2 in the lean group and >30 kg/m2 in the obese group. Exclusions included smoking, substance abuse, regular intake of medications, psychiatric or medical illness, and any abnormalities detected by physical examination or laboratory screening. None of the subjects had a history of gastrointestinal disorders, food allergies, or dietary restrictions. Anthropometric measurements, including weight, height, and BMI, as well as heart rate and blood pressure, were recorded for all participants. Subjects were instructed to abstain from alcohol, caffeine, black and green tea, Coke, chocolate, and strenuous exercise for 24 h before each treatment and, furthermore, to abstain from sprouts, broccoli, and grapefruit for the entire study duration.
Study design and experimental procedures.
The study was conducted as a randomized, double-blind, placebo-controlled, crossover trial. Randomization was computer-generated (computer-generated random order of treatment sessions). The day before each study day, subjects consumed a restricted, simple-carbohydrate standard dinner before 8 PM and fasted from 12 AM (midnight) onward. On each study day, subjects were admitted to the Phase 1 Research Unit of the University Hospital Basel at 8 AM. An antecubital catheter was inserted into a forearm vein for blood collection. Subjects swallowed a polyvinyl feeding tube (external diameter 8 French). The tube was placed through an anesthetized nostril; its intragastric position was confirmed by rapid injection of 10 ml of air and auscultation of the upper abdomen. The test trials were identical in design, except for the test solutions containing 50 g of xylitol dissolved in 300 ml of tap water, 75 g of erythritol dissolved in 300 ml of tap water, 75 g of glucose dissolved in 300 ml of tap water (positive control), and 300 ml of tap water (negative control).
Concentrations were chosen based on the following considerations: 75 g of glucose as in a standard oral glucose tolerance test (with known effects on plasma insulin, plasma glucose, and gastric emptying), 50 g of xylitol, and 75 g of erythritol as the sweetness of the xylitol and erythritol concentrations corresponded to ∼75 g of glucose, resulting in equisweet loads. Each test solution was labeled with 50 mg of [13C]sodium acetate for determination of gastric emptying. Glucose was purchased from Haenseler (Switzerland), xylitol and erythritol were purchased from Mithana (Switzerland), and [13C]sodium acetate was from ReseaChem (Switzerland). The intragastric infusions were freshly prepared each morning of the study and were at room temperature when administered. To maintain the blind, differing persons prepared and administered the treatment. After taking two fasting blood samples (t = −10 and −1 min) and a fasting breath sample (t = −1 min), subjects received the test solution via the feeding tube within 2 min (t = 0–2 min). Blood samples were taken at regular time intervals (15, 30, 45, 60, 90, 120, and 180 min) on ice into tubes containing EDTA (6 μmol/l), a protease inhibitor cocktail (Complete, EDTA-free, 1 tablet/50 ml of blood; Roche, Mannheim, Germany), and a dipeptidylpeptidase IV inhibitor (10 μl/ml; Millipore, St. Charles, MO). Tubes were centrifuged at 4°C at 3,000 rpm for 10 min, and plasma samples were stored at −70°C until analysis of plasma glucose, insulin, active GLP-1, and CCK was performed. For determining gastric emptying rates, end-expiratory breath samples were taken at fixed time intervals (15, 30, 45, 60, 75, 90, 105, 120, 150, 180, 210, and 240 min) after instillation of the test solution. The subject's vital signs (blood pressure and heart rate) were measured before and after each study intervention. Appetite perceptions (feelings of hunger, satiety, fullness, and prospective food consumption) were assessed by visual analog scales (8). Visual analog scales consisted of a horizontal, unstructured, 10-cm line representing the minimum (0.0 points) to the maximum rating (10.0 points). Subjects assigned a vertical mark across the line to indicate the magnitude of their subjective sensation at the present time point. The measurement was quantified by the distance from the left end of the line (minimum rating) to the subject's vertical mark.
Plasma glucose concentration was measured by a glucose oxidase method (Rothen Medizinische Laboratorien, Basel, Switzerland). The intra- and interassay coefficients of variation are <2.9 and 3.9%, respectively. Plasma insulin was measured with a commercially available electrochemiluminescence immunoassay (Cobas/Roche Diagnostics, Mannheim, Germany). The intra- and interassay coefficients of variation for this assay are <2.0 and 2.8%, respectively. Plasma active GLP-1 was measured with a commercially available ELISA kit (Millipore, St. Charles, MO). The intra- and interassay variabilities are <9.0 and 13.0%, respectively.
Plasma CCK concentrations were measured with a sensitive radioimmunoassay using a highly specific antiserum (no. 92128), (29). The intra- and interassay variabilities are <15% for both.
Assessment of Gastric Emptying
The gastric emptying rate was determined using a [13C]sodium acetate breath test, an accurate, noninvasive method for measuring gastric emptying, without radiation exposure, and a reliable alternative to scintigraphy, the current “gold standard” (10). Test solutions were labeled with 50 mg of [13C]sodium acetate, an isotope absorbed readily in the proximal small intestine, next transported to the liver, where it is metabolized to 13CO2, which is then exhaled rapidly (10). At fixed time intervals, end-expiratory breath samples were taken into a 100-ml foil bag. The 13C exhalation was determined by nondispersive infrared spectroscopy using an isotope ratio mass spectrophotometer (IRIS; Wagner Analysen Technik, Bremen, Germany) and expressed as the relative difference (δ%) from the universal reference standard (carbon from Pee Dee Belemnite limestone). 13C enrichment was defined as the difference between preprandial 13C exhalation and postprandial 13C exhalation at defined time points, δ over basal (%). Δvalues were converted into atom percent excess and then into percent of administered dose of 13C excreted per hour. In this last conversion, the CO2 production of the subjects was used, which is assumed to be 300 mmol/h multiplied by the body surface area. The body surface area was calculated by the weight height formula of Haycock et al. (11).
The purpose of this study was to gain basic information on the physiological role of the aforementioned doses of xylitol and erythritol on incretin release and gastric emptying. The sample size of this study was chosen on the basis of practical considerations rather than statistical estimation. However, according to our experience, a sample size of eight to 12 subjects would most likely allow us to detect large differences in parameters (>50%) between the treatments groups. Descriptive statistics were used for demographic variables such as age, weight, height and BMI. Hormone and glucose profiles were analyzed by calculating the area under the concentration-time curve (AUC) from baseline values. The parameters were tested for normality by the Shapiro-Wilk test method. General linear model repeated-measures ANOVA was applied to describe differences between lean subjects and obese participants in the different treatment groups (50 g of xylitol, 75 g of erythritol, and 75 g of glucose), where obesity status (yes or no) was used as between-subject factor in this analysis. Pairwise post hoc within-subject comparisons were done with the Šidak multicomparison test and between-subject comparisons by univariate ANOVA. All statistical analysis was done using the statistical software package SPSS for Windows, version 23.0 (SPSS, Chicago, IL). Values were reported as means ± SE. Differences were considered to be significant when P < 0.05. Prevalence of diarrhea associated with either polyol intake was compared by use of Fisher's exact test.
Fifty grams of xylitol ingestion led to bloating and diarrhea in 70% of all subjects, and 75 g of erythritol had the same side effects in 60% of all subjects (P = 0.741). There was no statistically significant difference between obese and lean subjects (obese vs. lean: xylitol P = 1.0 and erythritol P = 1.0) or between the two polyols (xylitol vs. erythritol; lean: P = 1.0; obese: P = 1.0) concerning side effects. Despite diarrhea (which usually stopped after 1–2 bowel movements), no study session had to be terminated prematurely. There were no dropouts, and complete data from 20 subjects (10 lean and 10 obese) were available for analysis.
Glucose and both polyols lead to a significant CCK release. There was no statistically significant difference between the two polyols and glucose (Table 1).
Only xylitol treatment increased AUC0–180 min of CCK compared with placebo due to a higher variability. However, the pattern was the same as in lean subjects (Table 1).
If all subjects were taken together (lean + obese; n = 20), glucose and both polyols led to a significant CCK release [F(3, 15) = 16.15, P < 0.001], and there was no statistically significant difference between the two polyols and glucose (Fig. 1 and Table 1).
Lean vs. obese.
Basal CCK concentrations were higher in obese vs. lean subjects (obese 1.4 ± 0.2 vs. lean 0.9 ± 0.1 mmol/l; P = 0.044), but there were no statistically significant differences in integrated CCK responses [AUC0–180 min; F(1, 17) = 0.009, P = 0.925].
Glucose ingestion as well as polyol intake stimulated GLP-1 release. However, this increase was numerically smaller with polyols, only borderline significant for polyols compared with placebo treatment (xylitol: P = 0.081; erythritol: P = 0.08), and only significantly different for glucose administration compared with placebo (AUC0–180 min; P = 0.004). Comparing glucose with xylitol administration, GLP-1 release was significantly lower after xylitol (AUC0–180 min; P = 0.027; Table 1).
Glucose ingestion as well as polyol intake stimulated GLP-1 release. Only glucose compared with placebo treatment was statistically significant (AUC0–180 min; P = 0.002; Table 1).
If all subjects were taken together, glucose and both polyols led to a significant GLP-1 release [F(3, 15) = 15.95, P < 0.001], and no statistically significant difference between the two polyols was found (P = 0.276; Fig. 1 and Table 1).
Lean vs. obese.
Basal GLP-1 concentrations were similar in both lean and obese groups. The integrated GLP-1 response to glucose administration (AUC0–180 min) was significantly higher in lean subjects [AUC0–180 min in lean, 862.3 ± 104.6 pmol·min−1·l−1; and in obese, 437.1 ± 62.6 pmol·min−1·l−1; F(1, 17) = 12.775, P = 0.002, respectively], whereas there were no differences after polyol intake.
Glucose administration increased glucose AUC0–180 min significantly (P = 0.045), and xylitol and erythritol compared with placebo showed no statistically significant effect (Table 2).
Glucose ingestion led to a statistically significant increase in plasma glucose AUC0–180 min (P = 0.008). Plasma glucose response (AUC0–180 min) was slightly but significantly increased after administrations of xylitol (P = 0.002) but also erythritol (P = 0.001) compared with placebo. We hypothesize that this is due to a decrease in plasma glucose over time after placebo rather than a small increase of plasma glucose after erythritol ingestion (Table 2).
If all subjects were taken together, then glucose, xylitol, and erythritol led to statistically significant changes in plasma glucose [F(1.1, 19.73) = 27.97, P < 0.001], and obesity status (yes/no) modified these responses significantly [F(1, 18) = 6.79, P = 0.018] (Fig. 1 and Table 2). However, compared with placebo, the increases in plasma glucose after xylitol and erythritol ingestion were minimal, albeit statistically significant (P = 0.004 and P = 0.01, respectively). There was no statistically significant difference between the two polyols.
Lean vs. obese.
Fasting glucose concentrations where higher in obese compared with lean subjects [5.2 ± 0.0 vs. 4.7 ± 0.1 mmol/l, respectively, F(1, 79) = 28.5, P < 0.001]; glucose excursions showed a higher Cmax for all carbohydrate treatments in the obese group compared with lean group [6.6 ± 0.3 vs. 5.6 ± 0.2 mmol/l, F(1, 79) = 20.2, P = 0.009; Cmax xylitol lean vs. obese: F(1, 19) = 10.2, P = 0.005; Cmax erythritol lean vs. obese: F(1, 19) = 7.97, P = 0.011]. AUC0–180 min was significantly higher in the obese compared with lean subjects after glucose treatment only [F(1, 19) = 6.19, P = 0.023].
Glucose ingestion led to an increase in insulin (P < 0.001). Xylitol had a minimal but statistically significant (P < 0.001) enhancing effect on insulin AUC0–180 min. In contrast to xylitol, erythritol treatment did not stimulate insulin release. However, comparing the integrated insulin response (AUC0–180 min) after erythritol treatment to placebo, there was a statistically significant difference (P = 0.037), as insulin decreased over time after the placebo treatment, whereas insulin concentration remained stable after erythritol treatment (Table 2).
Glucose ingestion led to an increase in insulin (P = 0.005), whereas xylitol had a minimal but statistically significant effect (P = 0.047). In contrast to xylitol, erythritol treatment did not stimulate insulin release (P = 0.98), (Table 2).
If all subjects were taken together, treatments led to significant changes in insulin release [F(1.1, 19.9) = 33.4, P < 0.001] that were significantly different between lean and obese subjects [F(1, 18) = 12.0, P = 0.003; Fig. 1 and Table 2]. In particular, glucose and xylitol significantly increased insulin release (P < 0.001 and P = 0.001, respectively), whereas erythritol had no effect on insulin release (p = 0.57).
Lean vs. obese.
Basal insulin concentrations were higher in obese compared with lean subjects [21.9 ± 2.1 vs. 6.8 ± 0.4 μU/ml, F(1, 79) 50.72, P < 0.001, respectively], and insulin excursions showed a higher Cmax [78.8 ± 15.2 vs. 22.8 ± 3.4 μU/ml, F(1, 79) 12.89, P = 0.001] after all treatments in obese subjects. The integrated insulin response (AUC0–180 min) was significantly higher in the obese persons after the glucose treatment [AUC0–180 min; lean vs. obese, F(1, 19) = 11.78, P = 0.003].
Glucose (given as positive control) compared with placebo (negative control) slowed gastric emptying (AUC0–60 min, P < 0.001), and both polyols had a decelerating effect as well (AUC0–60 min xylitol P = 0.001, erythritol P = 0.008). No statistically significant difference was seen between the two polyols (P = 0.683). The effect of both polyols was slightly smaller compared with glucose, and there was a statistically significant difference in AUC0–60 min between erythritol and glucose (P = 0.036) but not between xylitol and glucose (P = 0.361) (Fig. 2 and Table 3).
Glucose and both polyols compared with placebo slowed gastric emptying within the first hour (AUC0–60 min glucose P < 0.001, xylitol P = 0.004, and erythritol P = 0.001). No statistically significant difference was seen between the two polyols or between glucose and each polyol (Fig. 2 and Table 3).
If all subjects were taken together, glucose and both polyols slowed gastric emptying during the first 60 min [F(3, 54) = 46.1, P < 0.001], with no significant effect between lean and obese subjects (Fig. 2 and Table 3). There was no statistically significant difference between glucose and either polyol.
Baseline assessments were not equivalent across all study sessions. Therefore, we used relative values (posttreatment values minus pretreatment value) representing changes in appetite perception. Over time, feelings of satiety and fullness decreased, whereas feelings of hunger and prospective food consumption increased. There were no statistically significant differences between the four treatments or between lean and obese subjects (Fig. 3).
The objectives of this trial were to investigate whether 1) polyols can stimulate GLP-1 and CCK release, 2) gastric emptying is affected, and 3) whether polyols show these effects not only in lean but also in obese patients with impaired glucose tolerance, the “target group” for sugar substitutes.
Polyols such as xylitol and erythritol are natural sugar substitutes and have a long history of use in a wide variety of foods. Xylitol and erythritol are not completely absorbed, as most of the ingested xylitol passes through the small intestine and is fermented by bacteria in the large intestine, whereas erythritol is mostly absorbed (>90%) but then excreted by the kidneys (3, 4, 12). As a consequence, erythritol is better tolerated than xylitol, provoking less gastrointestinal side effects such as diarrhea and bloating. However, when erythritol is consumed as a single oral bolus exceeding 35 g, undesirable effects, including nausea and borborygmi, are common (18, 19, 25, 37). Repetitive exposure appears to lead to increased tolerance through adaptive processes (23). In our trial, subjects who had not been exposed to polyols before received high loads of glucose, xylitol, and erythritol to achieve equisweet conditions. After polyol treatments, the majority of participants had diarrhea irrespective of which polyol was used.
Taste-signaling mechanisms identified in the oral cavity are also present in the gut and play a role in both locations for sugar detection; activation of sweet taste receptors trigger regulatory circuits, which in turn are important in the control of eating behavior and the regulation of energy homeostasis. In the gut, nutrient detection is controlled mainly by enteroendocrine cells; upon sensing nutrients, a cascade of physiological phenomena is activated, including secretion of insulin, CCK (cholecystokinin), and GLP-1 (glucagon-like peptide-1) as well as inhibition of gastric emptying and reduction in food intake (28, 30). Colocalization of GLP-1, GIP (glucose-dependent insulinotropic peptide), PYY (peptide tyrosine tyrosine), and CCK with taste-signaling elements such as the sweet taste receptor T1R2-T1R3 is found in human intestinal endocrine L cells, explaining part of this phenomenon (14, 31). Because both caloric sweeteners (e.g., glucose, fructose, and sucrose) and nonnutritive, artificial sweeteners (e.g., aspartame, acesulfame-K, and sucralose) bind to oral sweet-taste receptors, binding to sweet-taste receptors on enteroendocrine cells is likely to cause signal transduction and downstream actions such as gut peptide release. However, the effect of nonnutritive sweeteners on incretin release seems to be more complicated. Nonnutritive sweeteners seem to be able to stimulate GLP-1 release in vitro (22), but in humans nonnutritive sweetener administration alone had no effect on plasma incretin concentrations (21, 36). In this study, both xylitol and erythritol stimulated GLP-1 release, suggesting an activation of the sweet receptor in the gut, although in vitro support of this finding is currently lacking.
We and others have reported that obese subjects show an attenuated incretin response to meal ingestion compared with lean persons (24, 40). In the present study, GLP-1 and CCK release could be demonstrated after glucose, xylitol, and erythritol treatment both in lean and obese subjects. Whereas the two polyols had similar effects on CCK release in lean and obese persons, the effect on GLP-1 secretion seemed to be reduced in obese persons. This was apparent for glucose and polyol administration; however, only after glucose administration could a statistically significant difference in integrated GLP-1 response be seen. The data are in line with previous studies documenting reduced nutrient-stimulated GLP-1 response in obese subjects (24, 40).
When glucose was ingested, the GLP-1 response in the presence of increased plasma glucose resulted in the expected plasma insulin response. As expected with both erythritol and xylitol when a GLP-1 response is triggered but a significant rise in plasma glucose is not simultaneously present, very little insulin response will follow. The obese subjects in our trial all showed impaired glycemic control, as demonstrated by elevated fasting glucose and insulin concentrations and higher glucose and insulin excursions after all carbohydrates. The effect of the two polyols on plasma glucose concentration and insulin release, although still higher in obese compared with lean subjects, was much smaller than after glucose ingestion, and this patient group might particularly profit from polyols as sugar substitutes.
Gastric emptying is regulated by numerous feedback mechanisms, including gut peptide release such as CCK and GLP-1. Prolonged gastric emptying leads to a feeling of fullness and satiation, which results in meal termination. As we demonstrated in this trial, erythritol and xylitol both lead to a prolonged gastric emptying. We also found a marked increase in GLP-1 and CCK after both polyol treatments. We infer from these observations that the significant retardation in gastric emptying is mediated by those incretins, particularly CCK. Subjective feelings of appetite were not significantly different after glucose, xylitol, or erythritol intake compared with placebo.
In this trial, we studied acute effects of rather high doses of erythritol and xylitol in subjects who were not used to these substances. In future studies, effects of lower doses, which could be used in everyday life, should be examined as well (e.g., 10 and 25 g). Furthermore, effects of long-term exposure on gastric emptying and stimulation of gut hormone release need to be investigated, as adaptive processes cannot be ruled out.
We conclude that acute ingestion of the natural sweeteners erythritol and xylitol lead to stimulation of gut hormone release (CCK and GLP-1) and have a decelerating effect on gastric emptying, whereas there is no (erythritol) or only little (xylitol) effect on insulin release.
B. K. Wölnerhanssen was supported by the Swiss National Science Foundation (Marie Heim-Voegtlin subsidy PMPDP3-145486/1), and C. Beglinger and R. Peterli received grant support from the Swiss National Science Foundation (Grant no. 138 157).
The authors have declared that no conflicts of interest exist.
B.K.W., R.P., C.B., and A.C.M.-G. conception and design of research; B.K.W., L.C., N.K., A.D., J.F.R., and A.C.M.-G. performed experiments; B.K.W. and J.D. analyzed data; B.K.W., A.D., J.D., J.F.R., C.B., and A.C.M.-G. interpreted results of experiments; B.K.W. prepared figures; B.K.W., L.C., N.K., and A.C.M.-G. drafted manuscript; B.K.W., J.F.R., R.P., C.B., and A.C.M.-G. edited and revised manuscript; B.K.W., L.C., N.K., A.D., J.D., J.F.R., R.P., C.B., and A.C.M.-G. approved final version of manuscript.
We thank C. Bläsi, S. Ketterer, M. Falck, and N. Viggiano for technical assistance.
- Copyright © 2016 the American Physiological Society