Dietary lactitol fermentation increases circulating peptide YY and glucagon-like peptide-1 in rats and humans

Nutrition 21 (2005) 1036 –1043 www.elsevier.com/locate/nut

Basic nutritional investigation

Dietary lactitol fermentation increases circulating peptide YY and glucagon-like peptide-1 in rats and humans Jennifer M. Gee, Ph.D., and Ian T. Johnson, Ph.D.ⴱ Nutrition Division, Institute of Food Research, Norwich, United Kingdom Manuscript received December 21, 2004; accepted March 7, 2005.

Abstract

Objective: Recently peptide YY (PYY) has attracted interest as a possible regulator of food intake. Release of PYY by nutrients in the distal small intestine is thought to contribute to the so-called ileal brake by inhibiting motility and secretion in the foregut. Our objective was to establish whether plasma concentrations of the gut peptides PYY and glucagon-like peptide-1 in rats and humans change in response to intake of a non-absorbable but fermentable carbohydrate. Methods: The acute response was determined in rats by killing animals 0, 5, 10, and 24 h after a single meal with or without lactitol (100 g/kg of semisynthetic diet) and measuring PYY and glucan-like peptide-1 concentrations in plasma. Food intake, body mass, and plasma peptide levels were also determined in rats fed the same diet for 10 d. Healthy human volunteers consumed lactitol or sucrose as a fruit-flavored drink. Breath hydrogen levels were measured at 45-min intervals over the next 7.5 h and plasma peptide concentrations were assessed after 0 and 5 h. Volunteers were also asked to complete a questionnaire to record satiety and well-being. Results: Ingestion of lactitol significantly increased the acute postprandial PYY response in rats, and prolonged consumption decreased weight gain in growing rats. In humans given a single dose of lactitol, the effects on PYY were much less marked but the postprandial decrease in circulating concentrations of PYY was attenuated. There was no effect on plasma glucan-like peptide-1. Conclusion: Our observations are consistent with a role for fermentation products in the release of gastrointestinal peptides in the rat and, to a lesser extent, in humans. © 2005 Elsevier Inc. All rights reserved.

Keywords:

Gut peptides; Ileal brake; Prebiotic; Appetite

Introduction Intestinal regulatory peptides, including glucagon-like peptide-1 (GLP-1) [1] and peptide YY (PYY) [2], are secreted by the enteroendocrine L cells of the small and large intestine. The various functions of these peptides are slowly emerging. GLP-1 regulates insulin secretion, and GLP-1 and PYY are involved in the control of upper gastrointestinal secretion and motility [3]. GLP-2 has been shown to support intestinal epithelial proliferation and to suppress

This work was supported by the Office of Science and Technology (United Kingdom) through the Core Strategic Grant of the Biotechnology and Biological Sciences Research Council. ⴱ Corresponding author. Tel.: ⫹44-1603-255-330; fax: ⫹44-1603-255038. E-mail address: [email protected] (I.T. Johnson). 0899-9007/05/$ – see front matter © 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.nut.2005.03.002

apoptosis [4]. The complex regulatory effects of pre-proglucagon– derived peptides have been shown in rodents to extend to the regulation of food intake [5], and this has raised the possibility of exploiting them for the control of human obesity [6]. Recently PYY has become a particular focus of attention. Intravenous infusion of the PYY fragment PYY3-36 has been shown to significantly suppress weight gain in rodents [7,8] and to decrease 24-h food intake in healthy human subjects [9], where, unlike leptin, it is as effective in obese subjects as in individuals of normal weight. These effects appear to be mediated through a gut– hypothalamic pathway in which PYY3-36 acts directly on Y2R receptors in the arcuate nucleus [10]. More recently, Halatchev et al. [11] reported that this mechanism is independent of melanocortin-4 receptor. Pharmacologic applications in the treatment of morbid obesity have been suggested [12], but these new findings also focus attention

J.M. Gee and I.T. Johnson / Nutrition 21 (2005) 1036 –1043

on the possibility of developing dietary strategies to exploit the appetite-suppressing effects of PYY and other intestinal peptides for the control of food intake in mildly obese subjects. Intestinal L cells are localized in the distal small intestine and colon of several monogastric species, including humans [13], and are thus strategically located to respond to the composition of the intestinal contents [14]. Luminal infusion of nutrients, including protein, peptides, and lipids, into the upper gastrointestinal tract stimulates the release of PYY from the distal ileum in rats and dogs [15] and human volunteers [16]. However, infusion of casein hydrolysate, glucose, and lipid into the human colon does not cause a release of PYY, whereas deoxycholate at physiologically relevant concentrations evokes release of PYY in all regions of the colon [17]. The mechanism of release appears to involve a neural pathway initiated by nutrients in the proximal small intestine and direct exposure of L cells to nutrients in the distal ileum [18]. In humans, concentrations of PYY in the circulation are low during fasting and increase after ingestion of food in proportion to calories ingested [19]. It has also been suggested that chronic intake of a high-fat diet blunts the PYY-stimulated intestinal synthesis of the satiety mediator apolipoprotein A-IV and may be an underlying cause of obesity [20]. As in the rat and other species, these effects of feeding are likely to be mediated by mechanisms involving the small intestine, but the role of the human colon is uncertain. The large bowel is the site of extensive bacterial activity. In more proximal regions, saccharolytic fermentations predominate and yield considerable quantities of short-chain fatty acids (SCFAs) and lactic acid. In the rat, infusion of SCFA into the colon, but not the ileum, stimulates the release of PYY [18,21], showing that fermentation products can in turn modulate the secretion of this gut peptide. Having previously shown that colonic fermentation products regulate plasma concentrations of the intestinal peptide hormone enteroglucagon in the rat, we investigated their effects on two other circulating gut peptides, PYY and GLP-1, in rats and in humans. In the present study we used our previously described meal-feeding protocol [22] to determine whether ingestion of the non-digestible but fermentable carbohydrate lactitol, 4-O-␤-D-galactopyranosyl-Dglucitol, which is used as an artificial sweetener and an osmotic laxative, increases circulating PYY and GLP-1 concentrations in the rat. We also undertook a human feeding trial to search for a similar effect in healthy human volunteers and to assess affects on satiety.

Materials and methods Animal experiments Male Wistar rats were obtained from a licensed animal supplier and housed in wire-bottom cages in a purpose-

1037

Table 1 Composition of diets (g/kg) Component

Control

Test

Sucrose Lactitol Starch Casein Corn oil Cellulose Vitamin mixture* Mineral mixture* DL-Methionine

100 0 458 200 80 100 20 40 2

0 100 458 200 80 100 20 40 2

* The concentrations of vitamins and minerals in the complete diets have been described previously [31].

built, air-conditioned animal facility with an ambient temperature of 21°C and a 12-h light, 12-h dark cycle. The U.K. Home Office approved routine animal care and all experimental procedures. Rats were allowed water ad libitum and fed a commercial pellet diet for a brief acclimatization period (3 to 4 d) before being transferred to a powdered starch-based semisynthetic control diet (Table 1). Acute challenge Over a period of 1 wk, 35 rats (230 to 250 g) were subjected to a reversal of the normal diurnal light/dark cycle (12 h/12 h) and trained to consume all their food (20 g/d of semisynthetic diet) within 5 h from the start of the dark cycle at 8 AM. On the eighth day, rats were randomly allocated to two groups of 15 and one group of 5 animals. The latter received no food on the test day and were killed by intraperitoneal injection of sodium barbiturate (Euthatal, Merial Animal Health Ltd, Harlow, UK; 0.1 mL/100 g of body mass) at 8 AM to provide fasting blood samples. Of the remaining 30 animals, 15 were given a single meal (20 g) of control diet and the remaining animals were fed the test diet in which sucrose (100 g/kg) was replaced by lactitol (Table 1). Five animals from each group were deeply anesthetized, as described above, at 5, 10, and 24 h after the start of the dark cycle and a blood sample (⬃5 mL) was taken from the vena cava. Animals were then killed, the abdomen was opened, and the pH of cecal contents was measured by insertion of a microelectrode through a slit in the intestinal wall. Rat feeding study In a second experiment, 10 male Wistar rats (150 to 180 g) were housed as above and fed ad libitum the control diet (n ⫽ 5) or a test diet (n ⫽ 5) in which sucrose was replaced by lactitol (100 g/kg) for 10 d after acclimatization. The dark cycle, during which rats feed, was between 8 PM and 8 AM. In this case food intake and weight gain were monitored regularly throughout the experimental period. After 10 d animals were deeply anesthetized 2 to 4 h after

1038

J.M. Gee and I.T. Johnson / Nutrition 21 (2005) 1036 –1043

the end of a 12-h dark cycle, and a blood sample was drawn from the vena cava as described above. This time point was approximately 15 h after the start of the nocturnal feeding period. The order of killing was one from each group sequentially to avoid bias due to time-dependent effects. Human feeding studies Each study received prior approval from the Institute of Food Research ethics committee, and subjects gave their informed consent to participate. The study was of singleblind design, with participants being unaware of which intervention they received on each occasion. In an initial experiment eight adult fasting volunteers (four men and four women, mean age 38.9 ⫾ 3.4 y, mean body mass index 27.4 ⫾ 1.7 kg/m2) consumed 10 g of sucrose or 10 g of lactitol as an orange-flavored drink for breakfast (8 to 9 AM). Volunteers provided breath samples before and at 1.5, 3, 3.75, 4.5, 5.25, 6, 6.75, and 7.5 h after consumption of the drink. At each sampling time, resting volunteers were asked to breathe out through a T-piece and collect alveolar air into a 20-mL syringe. This was assessed for hydrogen content by using a portable breath hydrogen monitor (GMI, Renfrew, Scotland) calibrated with a standard gas mixture (⬃100 ppm in air). Subjects were allowed to drink unsweetened tea or coffee and were offered a low-residue lunch of cooked breast of chicken (150 g) and boiled rice (300 g) 4 h after consumption of the test drink, which they consumed to satiety. In a second study, 12 apparently healthy human subjects (three men and nine women, mean age 40.2 ⫾ 2.0 y; mean body mass index 25.0 ⫾ 0.6 kg/m2) who had fasted overnight (ⱖ10 h) consumed, on separate occasions, lactitol (15 or 20 g) or sucrose (15 g) as single liquid test meals, in the form of an orange-flavored drink. Venous blood samples were taken from an in-dwelling brachial cannula before (t ⫽ 0 h) and after (t ⫽ 5 h) the test meals, and plasma fractions were prepared as described in the next section. Gastrointestinal peptide concentrations were determined by radioimmunoassay as described below. Exhaled air samples were also collected 15 min before and at approximately 45-min intervals after consumption of the drink until at least 5 h had elapsed or the peak response had passed and were analyzed as described previously. Individuals were asked to complete questionnaires to assess their perception of satiety and wellbeing before taking the drink, at the end of the experimental period, before the evening meal, and on the following morning before breakfast. Volunteers were asked if they felt hungry, thirsty, full, or had any desire to eat. They were also asked if they felt flushed, nauseous, jittery or shaky, faint or had a headache, pounding heart, stomach ache, or chest pain. This they did by placing a mark on a horizontal line, the limits of which represented opposing responses to each question (visual assessment scale). The positions of the marks on each line were measured and averaged and the significance of differences was assessed as described below.

Volunteers received a standard lunch consisting of soup, two filled rolls or filled jacket potato or pizza, fruit pie, and coffee. The evening meal on the experimental day and breakfast the following day were self-selected and eaten at home. Preparation and analysis of blood samples Blood samples were immediately transferred to tubes containing ethylene-diaminetetra-acetic acid disodium salt (100 ␮g/mL of blood) and aprotinin (600 KIU/mL of blood) as minimal volumes of aqueous solutions and then centrifuged (⬃5000g). Plasma fractions were removed and stored in fresh tubes containing aprotinin (600 KIU/mL of plasma) at ⫺20°C before extraction and analysis for PYY and GLP-1. In both cases radioimmunoassay kits (Peninsula Laboratories, San Carlos, CA, USA) were used to quantify gut hormones in extracts prepared according the manufacturer’s protocol. Briefly, plasma (1 mL) was acidified with an equal volume of trifluoracetic acid (10 g/L in water) and centrifuged at 10 000g for 20 min at 4°C. The supernatant was then applied to a washed and equilibrated C18 SEPCOLUMN1. Impurities in the sample were removed by washing the column with trifluoracetic acid and the peptide was eluted with acetonitrile (60% in 10 g/L of trifluoracetic acid). The extract was evaporated to dryness by using a centrifugal concentrator and redissolved in a reduced volume of radioimmunoassay buffer. The extracted peptide was assayed by radioimmunoassay based on the competitive binding of an iodine-125–labeled standard peptide to a limited quantity of specific antibody. Kits containing appropriate antisera specific for rat or human PYY or GLP-1 were used. Cross-reactivity with other peptides was negligible (0% or ⬍0.1%). The assay dynamic range was 1 to 128 pg/mL, and typical sensitivity was 7 pg/tube. Controls, standards, and samples were analyzed in duplicate. Data collection and statistical analysis In the case of human studies samples and questionnaires were coded to protect volunteer identity and mask treatment groups during the initial stages of analysis. Significance of differences between means were assessed by one-way analysis of variance (ANOVA), with use of Tukey’s comparison to assess the significance of differences between groups, or by two-way ANOVA, where appropriate, using Minitab statistical software (Minitab, State College, Pennsylvania, USA). Tests were also performed to ensure that the data satisfied the assumptions of parametric testing and, in cases where variance was unequal, data were logged to satisfy this condition. Where values were not suitable for ANOVA modeling, data transformation and/or non-parametric methods were employed. Results are expressed as untransformed means ⫾ standard error of the mean.

J.M. Gee and I.T. Johnson / Nutrition 21 (2005) 1036 –1043

1039

Results Animal experiments Acute challenge. Two-way ANOVA and an interaction plot showed that consumption of lactitol significantly decreased cecal pH (P ⬍ 0.001), and that this change was consistent over the 24-h period. Five hours after consumption of the meal, intracecal pH in lactitol-fed rats had decreased to 6.14 ⫾ 0.05 compared with 7.06 ⫾ 0.12 in control animals. Similar values were seen after 10 h followed by an increase to 6.60 ⫾ 0.24 at 24 h. In animals fed the control diet there was a small but significant postprandial increase in circulating PYY, with the concentration increasing transiently to 48.4 pg/mL (5 h) compared with the fasting value of 17.7 pg/mL (P ⬍ 0.01); there were no significant differences at later time points (Fig. 1a). Animals fed the control diet also showed a very small but significant increase in plasma GLP-1 (P ⬍ 0.05) after the meal, but not until the 10-h time point (Fig. 1b). In complete contrast, PYY concentrations in rats fed the lactitol diet were significantly increased after 5 h (P ⬍ 0.001) and increased to eight times the initial value at the 10-h time point (P ⬍ 0.01) and decreased thereafter (Fig. 1a). In the case of GLP-1 plasma concentrations remained higher than those of fasting rats 5, 10, and 24 h after feeding (Fig. 1b), and two-way ANOVA of logtransformed data showed this effect to be significant (P ⬍ 0.001). Interaction plots of the PYY and GLP-1 data indicated a highly significant effect of lactitol treatment between 5 and 24 h (P ⬍ 0.001). Feeding study. Rats fed the lactitol diet had a significantly lower average total food intake over 10 d (222.3 ⫾ 19.8 g/rat) compared with the sucrose-fed control group (267.6 ⫾ 17.0 g/rat; P ⬍ 0.05). Average body weight gain of lactitol-fed rats (53.2 ⫾ 6.2 g) was also significantly lower compared with the sucrose-fed control group (74.2 ⫾ 4.8 g; P ⬍ 0.05). PYY concentrations in plasma approximately 15 h after the start of nocturnal feeding were somewhat lower (6.7 ⫾ 0.4 pg/mL for the control group versus 5.5 ⫾ 0.3 pg/mL for the lactitol-fed group; no significant difference [NSD]) than those seen in fasting rats in the acute feeding study, and the same was true of GLP-1 (60.8 ⫾ 1.4 pg/mL for the control group versus 68.3 ⫾ 2.1 pg/mL for the lactitol-fed group; NSD). Human studies There was no significant change in breath hydrogen after the control drink, with concentrations being maintained below the initial fasting value of 10 ppm throughout the experiment (Fig. 2), despite the consumption of lunch. Breath hydrogen concentrations peaked 3 to 5 h after con-

Fig. 1. Rat acute plasma PYY (a) and GLP-1 (b) responses to consumption of semisynthetic diets with and without 10% lactitol. Values are mean ⫾ standard error (n ⫽ 5). *P ⬍ 0.05, **P ⬍ 0.01 versus 0 time value for the same treatment (one-way analysis of variance). Two-way analysis of variance showed a highly significant effect of lactitol on both gut hormones (P ⬍ 0.001) but no significant effect of time and no evidence of interactions between 5 and 24 h. GLP-1, glucagon-like peptide-1; PYY, peptide YY

sumption of lactitol (Fig. 2). No adverse effects were reported after consumption of lactitol except at the largest dose (20 g) when 4 of the 12 volunteers recorded an increase in sensations of nausea, generally associated with mild stomach pain, at the 5-h time point. Visual assessment scale data was not significantly different for any of the questions asked at any time point in response to any of the treatments, although there was a trend toward a decrease in appetite 2 to 5 h after consumption of 15 g of lactitol (Fig. 3). This trend was not upheld at the larger dose of 20 g when adverse effects, such as stomach ache and nausea, also became evident (Fig. 3). After consumption of the control drink, plasma PYY concentrations decreased significantly in the 5-h postprandial period (Fig. 4a; P ⬍ 0.01). This decrease was not seen

1040

J.M. Gee and I.T. Johnson / Nutrition 21 (2005) 1036 –1043

ing circulating levels of GLP-1 by dietary means could thus provide an attractive alternative to drug therapy by injection. The release of gastrointestinal peptides in response to the fermentative activity of colorectal microbiota can be viewed as a potential feedback mechanism with the capacity to inhibit excess consumption of calories. Ropert et al. [26] studied the effect of fermentable carbohydrate ingestion on gastric tone in humans and observed a significant decrease, which correlated with the breath hydrogen response to a test meal containing lactose. However, they saw no evidence of any effect of fermentation products on circulating enteroglucagon, GLP-1, or PYY concentrations. In contrast, Cuche et al. [27] demonFig. 2. Breath hydrogen response in fasting human volunteers to consumption of a drink with and without 10 g of lactitol. Values are mean ⫾ standard error of the mean (n ⫽ 12).

after consumption of 15- or 20-g doses of lactitol (Fig. 4a). Analysis of these data by two-way ANOVA (additive model) confirmed that consumption of 15 or 20 g of lactitol had highly significant effects on plasma PYY concentrations (P ⬍ 0.001), apparently by eliminating the postprandial decrease seen under control conditions. Data were assessed by regression diagnostics to ascertain their suitability for this model and were shown to be acceptable. There was no change in plasma GLP-1 concentration after consumption of the control drink, and no significant effects of lactitol were observed (Fig. 4b).

Discussion Bacterial fermentation of carbohydrate in the colon yields SCFAs in monogastric species. In previous work we reported that the substitution of the sugar alcohol lactitol (100 g/kg) for non-fermentable cellulose in rat diets led to a sustained release of pre-proglucagon derivatives in the rat [22]. Ghatei et al. [23] noted a similar effect in response to dietary fiber in rats, and Massimo et al. [24] showed that fermentable carbohydrate increases transcription of the proglucagon gene and increases postprandial GLP-1 concentrations in dogs. Potentially this is of great interest because this endocrine response may exert a regulatory influence on various aspects of gastric and duodenal function, with implications for the modulation of blood sugar concentrations and satiety. Knudsen [25] recently proposed that, due to their glucose-dependent insulinotropic activity, peptides with GLP-1–like activity could be used as a novel treatment regime for type 2 diabetics. The advantage of this approach is that, in addition to stimulating secretion in a glucose-dependent manner, the insulinotropic activity of GLP-1 may help to restore the sensitivity of pancreatic islets to glucose and increase ␤-cell mass. The possibility of increas-

Fig. 3. VAS data from questionnaires completed by human volunteers 2, 5, 10 (approximately), and 24 h after a drink containing (a) sucrose, (b) 15 g of lactitol, or (c) 20 g of lactitol. Values are mean ⫾ standard error of the mean (n ⫽ 12). VAS, visual assessment scale.

J.M. Gee and I.T. Johnson / Nutrition 21 (2005) 1036 –1043

Fig. 4. Human acute plasma PYY (a) and GLP-1 (b) responses to drinks containing sucrose (15 g; control) and 15 and 20 g of lactitol. Values are mean ⫾ standard error of the mean (n ⫽ 12, with subjects acting as their own controls). *P ⬍ 0.05 versus control value for the same treatment (one-way analysis of variance). Two-way analysis of variance showed a significant effect of lactitol in preventing the postprandial decrease in PYY after the control drink (P ⬍ 0.001), as indicated by bars annotated with a and b. GLP-1, glucan-like peptide-1; PYY, peptide YY.

strated that infusion of pig ileal loops with acetate in vivo inhibits gastric motility, implicating a humoral factor such as PYY. Olesen et al. [28] have since reported increased levels of circulating GLP-1 and PYY in patients with short bowel in response to carbohydrate ingestion, which they attributed to glucose absorption in the distal bowel rather than carbohydrate fermentation per se. In the present study with rats, consumption of a meal containing fermentable carbohydrate was associated with significant increases in plasma PYY and GLP-1 concentrations. Prolonged consumption of the same diet ad libitum, which in rats is characterized by intermittent nocturnal feeding, also resulted in a suppression of food intake and decreased weight gain. However, plasma concentrations of PYY and GLP-1 at death, 14 to 16 h after the start of the feeding period, were not significantly different. It is possible

1041

that secretion of these gut hormones in response to food had peaked before this time point, despite intake occurring over a more lengthy period. A slower delivery of fermentable carbohydrate to the colon, as a result of ad libitum feeding, could also be a contributory factor, thus changing the nature and the magnitude of the gut hormone response. In fasting human subjects who consumed a dose of lactitol that was probably close to the maximum tolerable for routine use, the large postprandial PYY response observed in the rat did not occur, but there was a statistically significant inhibition of the decrease in fasting PYY concentrations observed under control conditions. There was also a trend toward a mild suppression of satiety within 2 h of a single dose of 15 g of lactitol. There was no such effect after consumption of 20 g of lactitol, but there was evidence of adverse symptoms at this larger dose. Frost et al. [29], who carried out oral glucose tolerance tests in human volunteers, found no effect of background colonic lactulose fermentation on circulating levels of glucose, insulin, free fatty acid, and GLP-1. However, in their study, the last dose of lactulose was consumed 12 h before the oral glucose tolerance test, which is considerably later than the fermentative peak, and the lack of GLP-1 response is perhaps not surprising. Frost et al. [30] also reported that consumption of psyllium as part of a test meal did not change the acute GLP-1 response in humans. Their choice of sampling time, just 4 h after the meal, may have been too early to reflect a colonic response, especially because fiber tends to slow transit time. Using a three-meal protocol, Robertson et al. [31] observed that plasma levels of this gut hormone began to increase approximately 3 h after consumption of complex carbohydrate and were still increasing at the 8-h time point, demonstrating that colonic L-cell stimulation positively contributes to circulating GLP-1. In our initial human study, where a drink was administered without food, breath hydrogen data indicated that the liquid had reached the colon within 3 h and that carbohydrate fermentation peaked at around 4.5 to 5 h. This provided the rationale for the relatively early sampling time in the second study. The effect of soluble but non-digestible carbohydrates on satiety has been explored previously. For example, Lavin and Read [32] gave drinks incorporating the viscous nonstarch polysaccharide guar gum to human volunteers and observed some evidence of a transient increase in satiety during the following 2 h, which was reversed later. It is well established that viscous dietary components that increase the unstirred layer adjacent to the mucosal epithelium in the small bowel slow glucose transport [33]. Lavin and Read proposed that this effect led to increased contact between intraluminal carbohydrate and receptors in the small intestine, thus enhancing the release of putative satiety peptides at that site. However, relatively little consideration has been given to the possibility that peptide release as a result of

1042

J.M. Gee and I.T. Johnson / Nutrition 21 (2005) 1036 –1043

fermentable carbohydrates reaching the colon may contribute to this mechanism. The ability of intestinal tissue to adapt at the cellular level to constant ingestion of fermentable non-starch polysaccharides has been investigated in rats [34] and dogs [24], and Cherbut et al. [21] demonstrated that peptide release can occur as a result of exposure of colorectal L cells to SCFA. Increased expression of proglucagon messenger RNA and increased concentrations of the proglucagon-derived peptide GLP-1 were observed in distal enterocytes of both species and in colonocytes of rats. The investigators suggested that SCFA from fermentation of carbohydrate may act as a mediator in the adaptive response of the intestine. In conclusion, the concept of using soluble fermentable polysaccharides to enhance secretion of gut peptides can be seen as an extension of the “ileal brake” principle, whereby nutrients reaching the distal small intestine inhibit gastric motility via endocrine feedback. The present study suggests that, in the rat, fermentable carbohydrate may exert a significant degree of physiologic control over upper gastrointestinal tract function via modulation of L-type endocrine cell activity. The human PYY response to fermentable carbohydrate in this study was small in comparison with that of the rat, and effects on satiety were not confirmed. However, further studies are warranted to explore the extent to which delivery of fermentable carbohydrates to the colon might contribute to a sequence of physiologic events providing feedback control of gastric function and perhaps also of human eating behavior.

Acknowledgments The authors acknowledge Wendy Adams for technical assistance, Wendy Lee-Finglas and Simon Deakin for expert animal care, and Robert Foxall, M.D., for statistical advice.

References [1] Drucker DJ. Minireview: the glucagon-like peptides. Endocrinology 2001;142:521–7. [2] Cerda-Reverter JM, Larhammar D. Neuropeptide Y family of peptides: structure, anatomical expression, function, and molecular evolution. Biochem Cell Biol 2000;78:371–92. [3] Drucker DJ. Biological actions and therapeutic potential of the glucagon-like peptides. Gastroenterology 2002;122:531– 44. [4] Druker D. Glucagon-like peptides: regulators of cell proliferation, differentiation, and apoptosis. Mol Endocrinol 2003;17:161–71. [5] Larsen PJ, Fledelius C, Knudsen LB, Tang-Christensen M. Systemic administration of the long-acting GLP-1 derivative NN2211 induces lasting and reversible weight loss in both normal and obese rats. Diabetes 2001;50:2530 –9. [6] Larsen PJ, Vrang N, Tang-Christensen M. Central pre-proglucagon derived peptides: opportunities for treatment of obesity. Curr Pharm Des 2003;9:1373– 82.

[7] Batterham RL, Cowley MA, Small CJ, Herzog H, Cohen MA, Dakin CL, et al. Gut hormone PYY(3-36) physiologically inhibits food intake. Nature 2002;418(6898):650 – 4. [8] Pittner RA, Moore CX, Bhavsar SP, Gedulin BR, Smith PA, Jodka CM, et al. Effects of PYY[3-36] in rodent models of diabetes and obesity. Int J Obes Relat Metab Disord 2004;28:963–71. [9] Batterham RL, Cohen MA, Ellis SM, Le Roux CW, Withers DJ, Frost GS, et al. Inhibition of food intake in obese subjects by peptide YY3-36. N Engl J Med 2003;349:941– 8. [10] Challis BG, Pinnock SB, Coll AP, Carter RN, Dickson SL, O’Rahilly S. Acute effects of PYY3-36 on food intake and hypothalamic neuropeptide expression in the mouse. Biochem Biophys Res Commun 2003;311:915–9. [11] Halatchev IG, Ellacott KL, Fan W, Cone RD. Peptide YY3-36 inhibits food intake in mice through a melanocortin-4 receptor-independent mechanism. Endocrinology 2004;145:2585–90. [12] Halford JC, Cooper GD, Dovey TM. The pharmacology of human appetite expression. Curr Drug Targets 2004;5:221– 40. [13] Lundberg JM, Tatemoto K, Terenius L, Hellstrom PM, Mutt V, Hokfelt T, et al. Localisation of peptide YY (PYY) in gastrointestinal endocrine cells and effects on intestinal blood flow and motility. Proc Natl Acad Sci USA 1982;79:4471–5. [14] Ekblad E, Sundler F. Distribution of pancreatic polypeptide and peptide YY. Peptides 2002;23:251– 61. [15] Onaga T, Zabielski R, Kato S. Multiple regulation of peptide YY secretion in the digestive tract. Peptides 2002;23:279 –90. [16] Calbet JA, Holst JJ. Gastric emptying, gastric secretion and enterogastrone response after administration of milk proteins or their peptide hydrolysates in humans. Eur J Nutr 2004;43:127–39. [17] Adrian TE, Ballantyne GH, Longo WE, Bilchik AJ, Graham S, Basson MD, et al. Deoxycholate is an important releaser of peptide YY and enteroglucagon from the human colon. Gut 1993;34:1219 – 24. [18] Fu-Cheng X, Anini Y, Chariot J, Voisin T, Galmiche JP, Roze C. Peptide YY release after intraduodenal, intraileal, and intracolonic administration of nutrients in rats. Pflugers Arch 1995;431:66 –75. [19] Adrian TE, Ferri GL, Bacarese-Hamilton AJ, Fuessl HS, Polak JM, Bloom SR. Human distribution and release of a putative new gut hormone, peptide YY. Gastroenterology 1985;89:1070 –7. [20] Tso P, Liu M. Ingested fat and satiety. Physiol Behav 2004;81:275– 87. [21] Cherbut C, Ferrier L, Roze C, Anini Y, Blottiere H, Lecannu G, et al. Short-chain fatty acids modify colonic motility through nerves and polypeptide YY release in the rat. Am J Physiol 1998;275(6 pt 1):G1415–22. [22] Gee JM, Lee-Finglas W, Johnson IT. Fermentable carbohydrate modulates postprandial enteroglucagon and gastrin release in rats. Br J Nutr 1996;75:757– 66. [23] Ghatei MA, Ratcliffe B, Bloom SR, Goodlad RA. Fermentable dietary fibre, intestinal microflora and plasma hormones in the rat. Clin Sci (Lond) 1997;93:109 –12. [24] Massimino SP, McBurney MI, Field CJ, Thomson AB, Keelan M, Hayek MG, et al. Fermentable dietary fiber increases GLP-1 secretion and improves glucose homeostasis despite increased intestinal glucose transport capacity in healthy dogs. J Nutr 1998;128:1786 –93. [25] Knudsen LB. Glucagon-like peptide-1: the basis of a new class of treatment for type 2 diabetes. J Med Chem 2004;47:4128 –34. [26] Ropert A, Cherbut C, Roze C, Le Quellec A, Holst JJ, Fu-Cheng X, et al. Colonic fermentation and proximal gastric tone in humans. Gastroenterology 1996;111:289 –96. [27] Cuche G, Cuber JC, Malbert CH. Ileal short-chain fatty acids inhibit gastric motility by a humoral pathway. Am J Physiol Gastrointest Liver Physiol 2000;279:G925–30. [28] Olesen M, Gudmand-Hoyer E, Holst JJ, Jorgensen S. Importance of colonic bacterial fermentation in short bowel patients: small intestinal malabsorption of easily digestible carbohydrate. Dig Dis Sci 1999; 44:1914 –23.

J.M. Gee and I.T. Johnson / Nutrition 21 (2005) 1036 –1043 [29] Frost G, Brynes A, Leeds A. Effect of large bowel fermentation on insulin, glucose, free fatty acids, and glucagon-like peptide 1 (7-36) amide in patients with coronary heart disease. Nutrition 1999;15: 183– 8. [30] Frost GS, Brynes AE, Dhillo WS, Bloom SR, McBurney MI. The effects of fiber enrichment of pasta and fat content on gastric emptying, GLP-1, glucose, and insulin responses to a meal. Eur J Clin Nutr 2003;57:293– 8. [31] Robertson MD, Livesey G, Morgan LM, Hampton SM, Mathers JC. The influence of the colon on postprandial glucagon-like peptide 1 (7-36) amide concentration in man. J Endocrinol 1999;161:25–31.

1043

[32] Lavin JH, Read NW. The effect on hunger and satiety of slowing the absorption of glucose: relationship with gastric emptying and postprandial blood glucose and insulin responses. Appetite 1995; 25:89 –96. [33] Johnson IT, Gee JM. Influence of viscous incubation media on the resistance to diffusion of the intestinal unstirred water layer in vitro. Pflugers Arch 1982;393:139 – 43. [34] Reimer RA, McBurney MI. Dietary fiber modulates intestinal proglucagon messenger ribonucleic acid and postprandial secretion of glucagon-like peptide-1 and insulin in rats. Endocrinology 1996;137: 3948 –56.