Satiety and satiety-regulating gut hormone levels are abnormal in hyperglycemic individuals. We aimed to determine whether these abnormalities are secondary to hyperglycemia. Ten healthy overweight/obese subjects (age: 56 ± 3 yr; BMI: 30.3 ± 1.2 kg/m2) received three equicaloric meals at t = 0, 4, and 8 h in the absence (control trial) and presence of experimental hyperglycemia (hyperglycemia trial; 5.4 mM above basal). Circulating levels of glucose, insulin, ghrelin, and peptide YY (PYY)3–36 and visual analog scale ratings of satiety were measured throughout each trial. In the control trial, glucose, insulin, PYY3–36, and the feeling of fullness were increased in the postprandial periods, whereas ghrelin was decreased. In the hyperglycemia trial, in which plasma glucose was increased to 11.2 ± 0.1 mmol/l, postprandial meal responses (AUC: 0–2, 4–6, and 8–10 h) of PYY3–36 were lower (meal 1, P < 0.0001; meal 2, P < 0.001; meal 3, P < 0.05), whereas insulin (meal 1, P < 0.01; meal 2, P < 0.001; meal 3, P < 0.05) and ghrelin (meal 1, P < 0.05; meal 2, P > 0.05; meal 3, P > 0.05) were higher compared with the control trial. Furthermore, the incremental (Δ0–0.5, 4–4.5, and 8–8.5 h) ghrelin response to the first and third meals was higher in the hyperglycemia trial in contrast to control (Δ: 2.3 ± 8.0, P = 0.05; Δ: 14.4 ± 2.5, P < 0.05). Also, meal-induced fullness was prevented (meal 1, P = 0.06; meal 2, P = 0.01; meal 3, P = 0.08) by experimental hyperglycemia. Furthermore, trends in ghrelin, PYY3–36, and fullness were described by different polynomial functions between the trials. In conclusion, hyperglycemia abolishes meal-induced satiety and dysregulates postprandial responses of the gut hormones PYY3–36 and ghrelin in overweight/obese healthy humans.
- gut hormones
obesity is associated with abnormal levels of hormones that regulate energy balance and satiety, including insulin, ghrelin, and peptide YY (PYY)3–36 (10, 20, 36, 42), which can lead to a hyperphagic state (48), causing positive energy balance and further weight gain. Dysregulation of these nutrient-responsive hormones is also present in hyperglycemia-related diseases such as type 2 diabetes (T2D) (9, 10, 17, 30, 42, 43), potentially explaining the reduced postprandial satiety that may exist in this patient group (19). However, it is unclear whether impairments in satiety and gut hormone regulation are secondary to the diabetic state reflected in hyperglycemia.
As seen in obese individuals, fasting and postprandial ghrelin levels are decreased in patients with T2D (10, 17, 19, 30, 42). In fact, Erdmann et al. (10) demonstrated that overweight patients with T2D had lower ghrelin levels compared with their healthy overweight controls. Our recent data confirmed this finding and furthermore demonstrated that the suppressed ghrelin levels were accompanied by a reduced postprandial satiety (19). Additionally, although levels of PYY3–36 are also suppressed in obesity, levels have been found elevated in T2D subjects (9, 43). These data suggest that T2D could be a contributing factor to impairments in gut hormone regulation and postprandial satiety.
Hyperglycemia is proposed to elicit cellular glucotoxicity, which may explain endocrine dysfunctions in T2D (29, 38). Intestinal L cell secretion of the gut hormone glucagon-like peptide-1 (GLP-1) and pancreatic β-cell secretion of insulin are impaired in hyperglycemic T2D individuals (14, 41, 45). Recently, we demonstrated that 24 h of experimental hyperglycemia impaired GLP-1-mediated insulin secretion in healthy overweight/obese subjects (38). Furthermore, reducing hyperglycemia in obese Zucker diabetic fatty rats increased GLP-1 secretion, suggesting a direct effect of hyperglycemia on intestinal L cell function (32). This is of relevance to the current study because L cells also secrete PYY3–36, which allows us to speculate that hyperglycemia may also interfere with normal PYY3–36 secretion. Furthermore, acute short-term hyperglycemia induced by an oral or intravenous glucose load suppresses ghrelin levels in both healthy and T2D subjects (27, 36).
These existing data led us to investigate whether abnormal levels of gut hormones and defects in satiety regulation found in individuals with T2D mellitus may be secondary to the hyperglycemic state. Although prior work in lean young healthy men showed no effect of 48 h of experimental hyperglycemia at prediabetic levels on postprandial hormone responses, satiety, or food intake (40), the effects of experimental hyperglycemia on these variables have not been examined in overweight subjects. In this study, we investigated satiety regulation and gut hormone secretion during acute prolonged experimental hyperglycemia at diabetic levels in overweight/obese healthy subjects. We hypothesized that hyperglycemia would blunt meal-induced satiety and dysregulate gut hormone secretion.
All subjects were recruited from the Copenhagen, Denmark, area via response to an advertisement on www.forsoegsperson.dk and underwent a full medical screening (medical history, physical exam, and blood profile). Normal glucose tolerance was confirmed by oral glucose tolerance test in accordance with American Diabetes Association guidelines (2, 19). Individuals were excluded from participation if they 1) were weight unstable (>5 kg in previous 6 mo), 2) had BMI <25 or >40 kg/m2, 3) had an illness that contraindicated physical activity, 4) demonstrated any evidence of current or previous hematological, renal, hepatic, cardiovascular, or pulmonary disease, or 5) were taking medications known to affect our primary outcome variables. The final study population was n = 10 healthy overweight/obese subjects. Enrolled subjects represented a subset of a study population described previously by our group (38). They underwent dual-energy X-ray absorptiometry scans to determine whole body adiposity and fat-free mass and an incremental workload exercise test on a bicycle ergometer to determine maximal aerobic capacity (V̇o2max) and filled out recall habitual physical activity questionnaires. The study was approved by the Scientific Ethics Committee of the Capital Region of Denmark (file no. H-3-2010-127) in accordance with the Helsinki Declaration. Subjects provided informed written consent to participate. The study was registered on www.clinicaltrials.gov (NCT01375270).
To examine the effects of hyperglycemia on postprandial ratings of satiety and gut hormone responses, healthy individuals underwent two trials in a randomized crossover design separated by 1–2 wk: a control trial and an experimental hyperglycemia trial. Three equicaloric liquid meals were provided during the trials, and visual analog scale (VAS) ratings of satiety and plasma glucose, serum insulin, plasma ghrelin, and plasma PYY3–36 were measured. Subjects kept dietary records for 3 days prior to each trial; these were analyzed to confirm that no differences in energy or macronutrient intake were present prior to the trials. Also, subjects were informed to restrain from physical activity.
For the control trial, subjects arrived at the laboratory at 8 AM following an overnight fast. An antecubital venous line was placed, baseline blood samples were collected, and a VAS satiety questionnaire was completed. At t = 0, 4, and 8 h, subjects ingested equicaloric mixed-nutrient liquid meals (Resource Complete, Nestle, Switzerland) within 10–15 min. The total caloric load of the three meals was estimated to provide subjects with their estimated daily caloric requirement (16). Each of the three meals provided a third of the total daily requirement, with 55, 30, and 15% of calories derived from carbohydrate, fat, and protein, respectively. Blood samples for the measurement of glucose, insulin, total ghrelin, and PYY3–36 were collected at regular intervals (t = 0, 0.25, 0.5, 1, 1.5, 2, and 3 h after each meal). Subjects completed VAS satiety questionnaires every 2 h. The trial ended at t = 11 h. Water was provided ad libitum.
The hyperglycemia trial was identical to the control trial but with the addition of exogenous glucose infusion throughout the 11-h trial. In brief, a second antecubital venous line was placed in a dorsal hand vein in retrograde fashion on the contralateral hand to the sampling line for glucose infusion. The sampling hand was warmed to ∼60°C so as to arterialize the blood. At t = −15 min, a priming dose of exogenous glucose was infused to raise plasma glucose levels 5.4 mM above basal. Plasma glucose was measured at bedside every 5–30 min throughout, and the exogenous glucose infusion rate was thus adjusted to maintain a hyperglycemic level at 5.4 mM above individual baseline glucose level. From t = 0 h, meals, blood sampling, and the use of VAS satiety questionnaires proceeded exactly as described in the control trial.
At baseline and every 2 h during the trials, subjects marked their answers to the each of the following five questions on a 100-mm VAS. How hungry do you feel? How thirsty do you feel? How much could you eat right now? How nauseous do you feel? How full do you feel? On the VAS, 0 mm was considered “not at all” and 100 mm was considered “extremely.” Answers are reported as arbitrary units derived from the distance (mm) recorded out of 100 mm. This questionnaire has been validated previously (12).
Blood samples for plasma glucose measurements were collected into heparinized syringes and analyzed immediately at bedside using an automated glucose oxidase method (ABL700; Radiometer, Copenhagen, Denmark). Blood samples for plasma collection were immediately placed on ice and subsequently centrifuged (3,500 g, 15 min, 4°C). Samples were then aliquoted and stored at −80°C until analyses. Blood samples for ghrelin and PYY3–36 analysis were collected into EDTA-containing blood tubes preserved with 200 mM of 4-(2-aminoethyl) benzenesulfonyl fluoride hydrochloride (Sigma, Copenhagen, Denmark). Plasma for ghrelin analysis was acidified to prevent hydrolysis. Plasma total ghrelin (Millipore) and plasma PYY3–36 concentrations (Phoenix Pharmaceuticals, Burlingame, CA) were determined by radioimmunoassay according to the manufacturer's instructions. A dipeptidyl peptidase (DPP)-IV inhibitor was not added to the samples collected for PYY3–36 determination, and therefore, ongoing cleavage of PYY1–36 was possible. However, to minimize this issue, we ensured that plasma was separated and frozen immediately following blood sampling. Hb A1c was determined by high-performance liquid chromatography (Tosoh G7 Analyzer, San Francisco, CA) on whole blood collected into EDTA-containing blood tubes. Blood samples for serum insulin measurements were collected into plain tubes and allowed to clot at room temperature for 30 min before the serum was isolated by centrifugation. Serum insulin was determined by electrochemiluminescent immunoassay (E-Modular; Roche).
Total postprandial responses of ghrelin, PYY3–36, and insulin were determined during the first 2 h after each of the three meals and were reported as the area under the curve (AUC). The acute postprandial changes in ghrelin, PYY3–36, and insulin were determined during the first 30 min after each meal intake and reported as Δ-values before (t = 0, 4, and 8 h, respectively) and 30 min after each of the meals (t = 0.5, 4.5, and 8.5 h).
Data were analyzed using two-way (trial × time) repeated-measures ANOVA. In the event of significant trial × time interactions, Bonferroni post hoc tests were used to examine specific differences between means. Trial differences in satiety ratings, AUC, and Δ-values were analyzed with a paired t-test. These analyses were conducted using Prism version 4 (GraphPad Software, San Diego, CA). Additionally, to determine whether experimental hyperglycemia induced different trends (response patterns) in satiety and gut hormones, polynomial trend analysis was performed using SPSS version 20 (IBM). Between-trial differences in fit of polynomial function indicated a between-trial difference in the variable response. All data are represented as means ± SE, and statistical significance was accepted when P < 0.05.
Subject characteristics are presented in Table 1 and were published previously by our group (38). Healthy overweight/obese subjects (n = 10) were matched with respect to age, BMI, and V̇o2max, and all had normal fasting glucose and insulin levels as well as normal glucose tolerance.
The macronutrient composition of the three experimental meals was identical within and between trials (54.0 ± 0.7, 31.0 ± 0.7, and 15.0 ± 0.0% of calories derived from carbohydrate, fat, and protein respectively). The total caloric load of the three meals was 1,844 ± 110 and 1,825 ± 120 kcal for the control and hyperglycemia trials, respectively. This caloric load was divided evenly between the three meals within trials, providing 6.50 ± 0.05 kcal·kg−1·meal−1.
As shown in Fig. 1A, baseline glucose levels were not different between trials. Two-way ANOVA revealed a main effect of trial, time, and time × trial (P < 0.0001, all), indicating that glucose levels were significantly elevated in the hyperglycemia trial compared with the control trial (11.0 ± 0.1 vs. 6.9 ± 0.1 mmol/l, P < 0.0001). Post hoc analyses showed that plasma glucose was higher in the hyperglycemia trial at all time points except t = 0.5 h (Fig. 1A). Furthermore, glucose levels were clamped, abolishing potentially confounding postprandial glycemic responses.
Baseline insulin levels were not different between trials (Fig. 1B). Two-way ANOVA revealed a main effect of trial, time, and time × trial (P < 0.01, P < 0.0001, and P < 0.0001, respectively), indicating that insulin levels were higher in hyperglycemia trial compared with control trial. Total postprandial insulin responses to each meal (AUC) were also higher in the hyperglycemia trial compared with the control trial in each of the three meals (P < 0.01, P < 0.001, and P < 0.05, respectively; Table 2). Acute postprandial changes (Δ) in insulin levels following each meal increased in both trials; however, they increased to a greater extent in the hyperglycemia trial (Δ: P < 0.01, P < 0.01, and P = 0.12, respectively; Table 2).
Levels of ghrelin during the control and hyperglycemia trials are presented in Fig. 1C. The fasting levels tended to be higher in the hyperglycemia trial (P = 0.057). Two-way ANOVA revealed no effect of trial or time (P > 0.05); however, total responses (AUC) of ghrelin were higher after the first meal in the hyperglycemia trial compared with control trial (AUC: 27,833 ± 2,022 and 24,073 ± 1,377, P < 0.05; Table 2). Furthermore, whereas acute postprandial change (Δ) in ghrelin was negative during each meal in the control trial, Δghrelin was positive following the first and third meals in the hyperglycemia trial (both P < 0.05; Table 2).
Levels of PYY3–36 during the two trials are presented in Fig. 1D. Fasting levels did not differ between trials, but two-way ANOVA revealed a main effect of trial (P < 0.05), indicating that PYY3–36 levels throughout the trial were lower in the hyperglycemia trial than in the control trial. Total postprandial PYY3–36 responses to each of the three meals (AUC) were also lower in hyperglycemia trial compared with the control trial (P < 0.0001, P < 0.001, and P < 0.05, respectively; Table 2). Acute postprandial PYY3–36 responses (Δ) were not different between trials (P > 0.05; Table 2).
Satiety ratings (hunger, thirst, food that could be eaten, nausea, and fullness) during the two trials are presented in Fig. 2. Baseline satiety did not differ between trials. Two-way ANOVA revealed a main effect of time for hunger, food that could be eaten, nausea, and fullness in both trials (P < 0.0001, P < 0.01, P < 0.05, and P < 0.05, respectively), indicating that these satiety parameters change during both the control and hyperglycemia trials. Two-way ANOVA revealed no effect of trial or interactions between time and trial in any of the satiety ratings. However, postprandial fullness was lower after each of the three meals in the hyperglycemia trial compared with the control trial (P = 0.06, P = 0.01, and P = 0.08, respectively; Fig. 2).
Ghrelin levels throughout the trial were described by an eighth-order polynomial function in the control trial (P = 0.01), whereas it was fit by a seventh-order function (P = 0.04) in the hyperglycemia trial. Also, PYY3–36 levels throughout the trial were fit by a sixth-order polynomial function in the control trial (P = 0.004), whereas they were described by a ninth-order function (P = 0.03). This indicates that trends in ghrelin and PYY3–36 responses differed between trials. As for the satiety markers, the feeling of fullness throughout the trial was described by a fifth-order polynomial function during the control trial but not during the hyperglycemia trial (P = 0.007 vs. P = 0.54), indicating that trends in meal-induced fullness were different between trials. Also, the feeling of nausea was described only by a quadratic function in the control trial (P = 0.031 vs. P = 0.29). Hunger as well as food that could be eaten fit a fifth-order polynomial function during both the control (P = 0.01 and P = 0.05, respectively) and hyperglycemia (P = 0.007 and P = 0.08) trials, respectively, and thirst did not fit any function in either trial, all of which was indicative of no between-trial differences in these satiety markers.
No correlations between any of the satiety rating and hormone responses were found.
Under standard conditions, meal ingestion acutely suppressed plasma ghrelin levels and increased serum insulin levels, plasma PYY3–36 levels, and the feeling of fullness, reflecting normal meal responses (5, 8, 28, 47). However, the novel finding of this study was that experimental hyperglycemia decreased postprandial levels of PYY3–36, increased levels of ghrelin, induced exaggerated insulin levels, and blunted meal-induced suppression of ghrelin and the feeling of fullness. These data suggest that hyperglycemia may abolish satiety in healthy overweight/obese subjects and that this is accompanied by abnormal levels of nutrient-responsive gut hormones.
Recently, we demonstrated that the postprandial increment in the feeling of fullness is impaired in subjects with T2D compared with their BMI-matched controls (19). Our present finding that meal-induced fullness was abolished during hyperglycemia at a diabetic level (∼11 mmol/l) in overweight/obese healthy subjects implies that the impairment in satiety in T2D may be a secondary phenomenon of the prevailing hyperglycemia. High plasma glucose slows down the intestinal absorption of glucose, thus delaying the emptying of glucose from the gastrointestinal tract (34), which expectedly should reduce appetite and stimulate satiety (6). Therefore, we suggest that the reduced satiety is not just a response to an elevated glucose level but may also be due to a direct effect of hyperglycemia upon enteroendocrine cells, potentially due to glucotoxicity.
A rise in ghrelin levels and a drop in levels of PYY3–36 stimulate appetite and food intake in the healthy state (5, 47), explaining the finding that elevated ghrelin levels and/or suppressed levels of PYY3–36 would decrease fullness and satiety. Le Roux et al. (20) have linked attenuated PYY3–36 release to reduced satiety in obese subjects, and weight loss-induced elevation of ghrelin levels (13, 15) has been associated with improved appetite control (24). Interpreting our findings together with these data suggests that the abnormal postprandial levels of ghrelin and PYY3–36 induced by hyperglycemia in the present study may explain the demonstrated blunted meal-induced satiety. In the control trial, meal ingestion acutely suppressed plasma ghrelin and elevated PYY3–36 levels in all of the subjects, reflecting normal meal responses (5, 8, 28, 47). However, during experimental hyperglycemia, these normal postprandial responses to the meals were partly lost; the acute postprandial suppression of ghrelin was diminished, and levels actually rose in response to two of the three meals. This further supports the idea that hyperglycemia-induced glucotoxicity may disrupt the ghrelin-mediated regulation of satiety.
In contrast to our present findings, Teff et al. (40) found no effects on postprandial hormone responses following 48 h of experimental hyperglycemia at prediabetic levels (∼6 mmol/l) in young, lean, healthy subjects. We did not include a lean healthy control group, which could be perceived as a limitation of the study. However, the satiety-regulating hormones PYY3–36 and ghrelin were not measured in this previous study. Also, the experimental level of hyperglycemia used in the study by Teff et al. (40) was not at a diabetic level. Finally, fasting ghrelin values in our overweight/obese healthy subjects in the present study (220.3 ± 7.7 pmol/l) are comparable with values found previously in lean healthy subjects (∼245 pmol/l) (36). However, it should be noted that a comparison with ghrelin values measured in different laboratories should be made with caution. Altogether, this argues that hyperglycemia rather than body weight is a contributing factor for the results in the present study.
Previously, we and others determined that levels of gut hormones are abnormal in hyperglycemic T2D individuals (9, 10, 17, 19, 30, 42, 43). Whereas PYY3–36 levels are suppressed in obese subjects (20), levels have been found elevated in individuals with T2D compared with their BMI-matched healthy controls (9). Also, we and others have found that T2D subjects have even lower ghrelin levels both at baseline and postprandial compared with healthy BMI-matched controls (10, 19). These findings suggest that the diabetic state may be a contributing factor to the abnormal levels of gut hormones found in T2D subjects compared with nondiabetic subjects. However, no prior study has investigated whether the diabetic state, reflected as hyperglycemia, is causative of these dysregulated pathways. For the first time, our data demonstrate that hyperglycemia induces abnormal postprandial ghrelin (elevated) and PYY3–36 (suppressed) responses in healthy overweight/obese subjects. However, the direction of change in these variables following experimental hyperglycemia seems contradictory to the levels of PYY3–36 (elevated) and ghrelin (suppressed) documented in patients with T2D (9, 10, 17, 19, 30, 42, 43), eliminating the possibility that hyperglycemia can explain the dysregulations of gut hormones present in this patient group. Still, the phagic hormone profiles and reversed meal responses demonstrated in the present study argue that our findings are a result not just of an elevated steady-state glucose level but of an effect of hyperglycemia, which is due potentially to glucotoxicity. However, although we have previously been able to confirm a glucotoxic effect on pancreatic endocrine function, as seen in T2D with the present hyperglycemic model (38), our data bring us to speculate that the model perhaps reflects a more acute state of hyperglycemia than is present in T2D individuals. Altogether, this suggests that the effects of acute hyperglycemia do not reflect the metabolism in T2D patients with chronic hyperglycemia, and therefore, it remains uncertain whether abnormal gut hormone levels present in T2D subjects are an adaptation to long-term exposure to hyperglycemia.
It is beyond the scope of this study to elucidate the underlying mechanisms; however, previous studies may help explain the disruptive effect of hyperglycemia in our study. The suppressed PYY3–36 levels may be due to decreased release or increased degradation. Increased degradation is unlikely because elevated activity of DPP-IV, which degrades PYY3–36, among other intestinal hormones (25), has been reported only in T2D patients with Hb A1c >8.5%, i.e., in patients with chronic hyperglycemia (23). Therefore, our acute model of hyperglycemia is unlikely to alter DPP-IV activity. However, an alteration in secretion is more likely. PYY3–36 is secreted from intestinal L cells, from which GLP-1 is also released. Interestingly, reduced postprandial circulating GLP-1 in T2D is suggested to be a consequence of chronic hyperglycemia (18), and we have shown previously that 24 h of experimental hyperglycemia reduced GLP-1-mediated insulin secretion in humans (38). Furthermore, Reimer et al. (32) have shown that reduction of hyperglycemia in obese Zucker diabetic fatty rats increased L cell secretion of GLP-1. Since PYY3–36 is secreted from the same cells as GLP-1, we speculate that our experimental hyperglycemia in this study induces a glucotoxic effect in intestinal L cells, thus lowering PYY3–36 levels.
Satiety regulation is complex, with the hypothalamus playing a vital role in integrating the hormonal and neural signals triggered during fasting and feeding (35). Glucose levels can modulate neural firing in the hypothalamus (22), and the presence of diabetes (1, 44) as well as hyperglycemia (31) has been found to interfere with hypothalamic signaling. However, since neural signals may be necessary only in the preprandial rise of ghrelin (46), the presence of disrupted neural regulation of satiety is therefore unlikely to explain the absence of a postprandial decline in ghrelin levels in our study. The role of insulin in ghrelin regulation remains controversial; however, with the stimulatory effect of glucose on insulin secretion, this hormone may also play a role in the consequence of hyperglycemia on ghrelin levels. Although several studies have shown that supraphysiological doses of insulin suppress ghrelin levels (11, 21, 26, 37), others have found that when provided in physiological doses, insulin has no effect of on ghrelin (7, 33). Herein, we found that during the hyperglycemia trial, despite progressively larger and exaggerated insulinemic responses to each meal, the postprandial suppression of ghrelin was absent. This is in line with the findings of Erdmann et al. (10), who demonstrated that postprandial ghrelin levels were suppressed equally in obese subjects with normo- or hyperinsulinemia. To explain the lack of association between elevated insulin and the suppression of ghrelin, insulin-dependent modulation of plasma ghrelin concentrations has been found to be less pronounced in hyperglycemic T2D patients (3). As such, it is possible that experimental hyperglycemia disrupts the suppressive effect of insulin on ghrelin secretion. Recent evidence in human subjects also suggests that PYY suppresses circulating concentrations of ghrelin when given peripherally (4). Therefore, it is also possible that the loss of ghrelin regulation in the experimental hyperglycemia trial is a secondary consequence of L cell glucotoxicity that impairs PYY3–36 secretion.
A limitation of our study design is that subjects did not receive a placebo infusion during the control trial and thus were not blinded to the trial. However, to ensure that this did not confound our VAS measurements, whereas the full methods were explained to subjects, the study hypotheses regarding appetite regulation by hyperglycemia were not. Another potential limitation of our design is the lack of earlier and more frequent satiety questionnaires. This may have led to an underestimation of the effects of hyperglycemia in relation to our observed changes in gut hormones. Finally, although acylated ghrelin is considered active, we have found previously that oral glucose induces similar plasma responses of total and acylated ghrelin, indicating that total ghrelin provides a surrogate marker of secretion of the acylated form (19, 39).
We have shown that an experimental hyperglycemia at a diabetic level blunted meal-induced satiety and that this was accompanied by abnormal postprandial gut hormone levels. Abnormal levels may be mediated by a glucotoxic effect on gut hormone-secreting cells, particularly PYY3–36-secreting L cells. Because the abnormal levels of PYY3–36 and ghrelin induced by experimental hyperglycemia in the present study did not reflect levels documented previously in patients with T2D, we suggest that there are likely other mechanisms responsible for the dysregulation of satiety in such patients. The question of how the perturbations in the regulation of PYY3–36 and ghrelin concentrations are secondary to hyperglycemia remains unanswered, and future studies are needed to clarify these mechanisms.
This study was funded by a Paul Langerhans program grant from the European Foundation for the Study of Diabetes (to T. P. J. Solomon). The Centre of Inflammation and Metabolism is supported by a Centre Grant from the Danish National Research Foundation and is part of the UNIK Project: Food, Fitness, and Pharma for Health and Disease, which is supported by the Danish Ministry of Science, Technology, and Innovation.
The authors have nothing to disclose.
S.H.K., K.K., and T.P.S. performed the experiments; S.H.K. and T.P.S. analyzed the data; S.H.K., K.K., and T.P.S. interpreted the results of the experiments; S.H.K. and T.P.S. prepared the figures; S.H.K. drafted the manuscript; S.H.K., K.K., and T.P.S. edited and revised the manuscript; S.H.K., K.K., and T.P.S. approved the final version of the manuscript; T.P.S. conceived and designed the research.
We express our gratitude to Lisbeth Andreasen (Department of Clinical Biochemistry, Rigshospitalet, Denmark) for technical assistance with clinical biochemistry assays.
- Copyright © 2014 the American Physiological Society