U.S. flag

An official website of the United States government

NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

Food Forum; Food and Nutrition Board; Institute of Medicine. Relationships Among the Brain, the Digestive System, and Eating Behavior: Workshop Summary. Washington (DC): National Academies Press (US); 2015 Feb 27.

Cover of Relationships Among the Brain, the Digestive System, and Eating Behavior

Relationships Among the Brain, the Digestive System, and Eating Behavior: Workshop Summary.

Show details

2Interaction Between the Brain and the Digestive System

When food enters the mouth and passes through the digestive system, it sends a multitude of interacting signals to the brain, loaded with sensory, nutritive, and other information. In the first session of the workshop, moderated by Danielle Greenberg1 of PepsiCo, participants discussed how those signals are triggered and how the feedback they provide to the brain impacts further food intake. Workshop participants also considered how higher cognitive functions in the brain, as well as developmental, familial, and environmental factors, influence this complex signaling and feedback system. This chapter summarizes the presentations and discussion of this session, key points from which are highlighted in Box 2-1.

Box Icon

BOX 2-1

Key Points Made by Individual Speakers. Timothy Moran explained that the brain receives much of its information about nutrient content and the volume of food consumed from signals sent via the vagus nerve. Vagal fibers in the stomach respond mainly to (more...)


In his overview of interactions between the brain and digestive system, Timothy Moran focused mainly on signals sent from the gastrointestinal (GI) tract to the brain via the vagus nerve and described how GI peptides released in response to the presence of nutrients trigger vagal nerve activity.

Innervation of the GI Tract

The GI tract is innervated both intrinsically and extrinsically. The intrinsic, or enteric, nervous system is embedded in the wall of the digestive tract and is localized primarily in the myenteric plexus and submucosal plexus.3 The enteric nervous system contributes to overall gastrointestinal motility, nutrient handling, gastric acid secretion, and other functions within the GI tract (Furness, 2012; Mawe and Hoffman, 2013). It is important to note, Moran observed, that when external inputs are cut, that is, when the extrinsic system is denervated, the enteric nervous system still functions and can regulate overall GI function—not in a normal, coordinated way, but in such a way that there is ongoing digestive activity. Thus, the enteric nervous system is not completely dependent on extrinsic input and can operate in isolation.

Enteric neurons extend across the GI tract and are activated by the presence of nutrients in what Moran described as a “somewhat nutrient-specific” manner, with different nutrients triggering different patterns of activity. Using c-fos, a stainable marker of neural activity, researchers have demonstrated nutrient-induced intrinsic neural activation under a variety of circumstances (Sayegh et al., 2004). Because the same neurons can also be activated by extrinsic activity, with stimulation of vagal afferent fibers (i.e., extrinsic neurons innervating the intestine) producing similar c-fos activation, it is unclear whether nutrient-induced intrinsic effects are an altogether local phenomenon or are dependent in part on stimulation activated by signals from the brain (Zheng and Berthoud, 2000).

Moran emphasized that while intrinsic neural stimulation can be demonstrated under a variety of conditions, it is unclear whether intrinsic neural regulation plays a role in controlling food intake.

Most information received by the brain about GI contents is transmitted via vagal afferent feedback signals. The vagus is one of two major extrinsic innervation sources, the other being the spinal cord (Sengupta, 2006). Spinal cord afferents appear to play more of a role in mediating GI pain than in providing feedback to the brain about nutrient contents, according to Moran.

Vagal Afferent Feedback Signals

Moran described the work of Hans-Rudolf Berthoud and colleagues, which has been instrumental in providing researchers with an understanding of the GI tract's overall vagal afferent innervation. Berthoud and Neuhuber (2000) described three types of vagal afferent endings, each providing a different type of information to the brain: (1) intramuscular array (providing “stretch” information), (2) intraganglionic laminar endings (providing “tension” information), and (3) mucosal terminals (providing “nutrient” information). According to Moran, it has been fairly well demonstrated that intramuscular array terminals measure stretch; that is, as the stomach begins to fill and food enters the small intestine, the presence of that food causes a stretch in the surrounding muscle fibers that activates vagal afferent neurons with intramuscular array endings (Phillips and Powley, 2000). That activation is transmitted to the brain. The intraganglionic laminar endings, which are found primarily in the stomach and in the proximal duodenum, have been hypothesized to measure tension (Phillips and Powley, 2000). The difference between measuring stretch and measuring tension can be confusing, Moran remarked. Stretch is a change in volume, while tension is a change in the surrounding musculature with no change in volume. Many vagal afferents have both “stretch” and “tension” endings and are able to respond to both stimuli simultaneously. The third type of vagal afferent ending, the mucosal terminal, is located mainly in the intestine, with the endings in close proximity to where nutrients are being absorbed and where various kinds of endocrine cells are releasing their products. “Nutrient” vagal afferents respond to overall nutrient character, not load volume (Berthoud and Neuhuber, 2000; Dockray and Burdyga, 2011).

Vagal afferent innervation of the stomach is different from that of the intestine. In the stomach, the vagal fibers respond primarily to load volume, not chemical composition. According to Moran, it has been demonstrated that single vagal afferent fibers from the stomach become activated when volume is introduced into the stomach (de Lartigue, 2014; Schwartz et al., 1991). Although vagal afferent fibers in the stomach exhibit a range of activity, with some fibers becoming activated in response to small volumes and others requiring larger volumes, together they show a dose–response relationship with load volume: the greater the load volume, the greater the vagal activation. The dose–response relationship is the same regardless of gastric contents (Dockray, 2013; Li, 2007; Mathis et al., 1998; Schwartz and Moran, 1998). Cells innervating the duodenum, on the other hand, show a very different pattern of results, with different contents producing different amounts of activity (e.g., saline versus glucose versus protein).

Gut Peptide Signaling

The small intestine releases a variety of peptides in response to the presence of nutrients. Using cholecystokinin (CCK) as an example, Moran explored the role played by many gut peptides in regulating food intake by mediating the response between nutrient activity in the small intestine and vagal afferent signaling (see Figure 2-1).

FIGURE 2-1. Much of the feedback received by the brain from the gastrointestinal tract (GI) tract is mediated by peptides released in the gut in response to the presence of nutrients.


Much of the feedback received by the brain from the gastrointestinal tract (GI) tract is mediated by peptides released in the gut in response to the presence of nutrients. NOTE: CCK = cholecystokinin; GLP-1 = glucagon-like peptide-1; PYY = peptide tyrosine (more...)

Peptides released from endocrine cells in the small intestine can either enter the bloodstream and travel to a distal target cell (i.e., endocrine signaling) or activate a closely located target cell (i.e., paracrine signaling) (Dockray, 2013; Krstic, 1984). Berthoud and colleagues have demonstrated that vagal afferent nerve fibers innervating the gut are located near endocrine cells in the intestinal villae that release CCK in response to the presence of nutrients in the intestinal lumen (Berthoud and Patterson, 1996; Berthoud et al., 1995; Dockray, 2012; Patterson et al., 2002). According to Moran, it has been further demonstrated that, indeed, activation of those very closely located vagal fibers is mediated by CCK release.

Although gastric fibers respond mainly to gastric load, not nutrient content, and intestinal fibers to nutrient content, not load, it has been demonstrated that nutrient activity in the intestine also results in increased activity in fibers innervating the stomach (Schwartz and Moran, 1998; Schwartz et al., 1993) (see Figure 2-2). According to Moran, it is unclear whether that response in the stomach is due to a vagovagal reflex, with signals sent from the intestine to the hindbrain altering gastric tone, or to nutrient-induced release of a peptide in the intestine that circulates and activates the gastric fibers. The effects of load volume and CCK in the stomach appear to be additive, according to Moran. Subthreshold doses of CCK that by themselves are too weak to stimulate vagal afferent activity can stimulate such activity in combination with load volumes.

FIGURE 2-2. Cholecystokinin (CCK), a peptide released in the intestine, magnifies the response to gastric load in gastric mechanosensitive fibers.


Cholecystokinin (CCK), a peptide released in the intestine, magnifies the response to gastric load in gastric mechanosensitive fibers. SOURCE: Reprinted with permission from Schwartz, G. J., P. R. McHugh, and T. H. Moran. 1993. Gastric loads and cholecystokinin synergistically (more...)

The ability of CCK to reduce food intake requires an intact vagus nerve. Scientists have shown in rats that removing or cutting the vagal afferent nerves blocks CCK satiety and under some circumstances increases the volume of food consumed.

Crosstalk in the Hindbrain

The vagus nerve enters the brain in an area of the brain known as the caudal medulla, with the vagal afferents and vagal efferents entering structures that are immediately adjacent to each other: the vagal afferents enter the nucleus of the solitary tract or nucleus tractus solitarii (NTS), while the vagal efferents enter the dorsal motor nucleus. The adjacency of the vagal afferents and efferents is what allows for some of the vagovagal reflexes that Moran suspects may contribute to some vagal afferent responses. A great deal of what he described as “crosstalk” takes place between the NTS and the dorsal motor nucleus (where the vagal efferents enter), with incoming information from the NTS altering the activity of vagal efferent cell bodies located in the dorsal motor nucleus.

The NTS receives not only information from vagal afferents from the stomach and intestine, but also vagal inputs from the liver and from taste receptors in the oral cavity. Curious about whether isolating and stimulating individual components of the complex array of signals converging in the NTS would reflect what normally occurs during a meal, Moran and colleagues compared c-fos activation in a real versus a sham feeding4 situation (Emond et al., 2001). They observed a much greater degree of activation in the taste region of the NTS in the sham feeding situation, suggesting that the brain processes and responds to oral signals differently depending on where in the GI tract nutrients are present. Other kinds of alterations (e.g., nutrients being placed directly into the stomach versus normal feeding through the mouth) have shown similar changes in brain activation (Emond et al., 2001).


Taste cells in the tongue are among the first cells in the GI tract that come into contact with food. Only recently have scientists discovered taste-like cells in the gut as well. Robert Margolskee provided an overview of taste receptors in the oral cavity and discussed recent research on taste-like receptors in the gut.

Taste Receptors in the Oral Cavity

Oral taste buds—collections of about 50 to 100 specialized epithelial cells—are scattered throughout the oral cavity, primarily in papillae6 on the front, sides, and back of the tongue. Although oral taste buds are not neurons, they have a number of neuronal properties. Much of the taste transduction cellular machinery is contained within the fingerlike microvilli coating the apical end of each taste bud cell.

Margolskee explained that scientists have identified several different types of taste receptors in the oral cavity, each having a unique taste receptor molecule or set of molecules underlying the taste response (Lindemann, 2001). Over the past decade, work from Margolskee's laboratory, as well as the laboratories of Linda Buck, Nick Ryba, and Charles Zucker, has led to identification of many of the different taste quality receptors. Today, researchers know that the bitter taste receptors involve a family of about 25 to 30 G protein-coupled receptors7 called the T2Rs (type 2 taste receptors). Sweet receptors, in contrast, involve a dimeric or multimeric combination of T1R2 (type 1 receptor 2) and T1R3 (type 1 receptor 3) receptors, which together respond to a number of sweet compounds, both sugars and noncaloric sweeteners. A related receptor, the umami receptor, involves a combination of T1R1 (type 1 receptor 1) and T1R3 receptors and responds to “savory” tastes such as monosodium glutamate.

The sour and salty taste transduction channels are not as well understood as the bitter, sweet, and umami channels, said Margolskee. Although ENaC8 certainly plays a role in salty taste transduction, it is involved more with low concentrations of salt. There is likely at least one other transduction channel, as yet unidentified, for high concentrations of salt. The sour taste receptor has a number of candidate channels, including acid-sensing ion channels (ASICs), hyperpolarization-activated cyclic nucleotide-gated (HCN) channels, and polycystic kidney disease (PKD) family member channels, but no one channel has yet been definitively identified.

Taste-Like Cells in the Gut (and Pancreas)

As summarized by Margolskee, researchers recently have identified taste-like cells in the gut that play an important role in integrating physiological responses during digestion. Taste-like cells in the gut are not actual taste cells, although they have a number of characteristics in common with true oral taste cells: they are morphologically similar under both light and electron microscopy and produce many of the same taste signaling proteins. Indeed, the signaling process that occurs in certain types of endocrine cells in the gut is very similar to the transduction process that occurs in oral taste cells (Cummings and Overduin, 2007) (see Figure 2-3). In both types of cells, when G protein-coupled receptors at the apical surface of the cell couple with gustducin and other taste-associated G proteins, they initiate a signal transduction cascade involving multiple signaling enzymes, second messengers (e.g., inositol triphosphate), and channels (e.g., the calcium-activated TRPM5 channel), ultimately leading to neurotransmitter or, in the case of taste-like cells, neuropeptide release. Margolskee explained that one of the differences between taste receptors in the oral cavity and taste-like receptors in the gut is that instead of releasing a true neurotransmitter, taste-like receptors in the gut release neuropeptide hormones, such as GLP-1 (glucagon-like peptide-1).

FIGURE 2-3. Oral taste cells (“taste cell”) and gut taste-like cells (“endocrine cell”) share similar signaling processes.


Oral taste cells (“taste cell”) and gut taste-like cells (“endocrine cell”) share similar signaling processes. SOURCE: Modified from Cummings and Overduin, 2007. Reprinted with permission of the American Society for Clincial (more...)

Margolskee went on to explain that the idea that taste signaling molecules exist in the gut dates back to the mid-1990s, when Dirk Höfer discovered alpha-gustducin (the alpha subunit of the heterotrimeric gustducin protein) being expressed in stomach and intestinal cells that had the general appearance of taste receptor cells (Höfer et al., 1996). Subsequently, Enrique Rozengurt's group identified a number of T2R bitter taste receptors in the stomach and small intestine (Wu et al., 2002). Later, Soraya Shirazi-Beechey found T1R receptors in the gut (Dyer et al., 2005).

In more detailed microscopic studies, Shirazi-Beechey and Margolskee collaborated and found that both T1R2 and T1R3, the two components of the sweet receptor, are present in a small subset of cells lining the small intestinal mucosa and that the cells have the typical appearance of enteroendocrine cells (Margolskee et al., 2007). Margolskee and his team also collaborated with Josephine Egan at the National Institutes of Health and identified several taste signaling proteins in both human and mouse tissues. They also found essentially the entire taste transduction pathway as it was known to exist in oral taste cells, in gut endocrine cells, and particularly in L cells expressing GLP-1 (Jang et al., 2007).

More recently, Yan Li in Margolskee's laboratory examined co-expression of gustducin-positive endocrine cells from various locations in the small and large intestines and found a roughly equal level of L, K, and L/K co-expression9 with gustducin in the colon but mainly only K or L/K cells co-expressing with gustducin in other areas (Li et al., 2013). Li also found a number of short chain fatty acids co-expressed with alpha-gustducin in endocrine cells in the colon, including cells activated by the G protein-coupled receptors GPR43 and GPR41. Curious about the potential physiological role of gustducin in the colon, she turned to gustducin knockout mice and found that short chain fatty acid–stimulated GLP-1 secretion from colon endocrine cells requires alpha-gustducin.

In other collaborative work between Margolskee's laboratory and Shirazi-Beechey's group, the researchers examined SGLT1 (sodium glucose co-transporter 1) expression in two types of knockout mice (Margolskee et al., 2007). SGLT1 is a protein that co-transports glucose and sodium from the gut lumen across the absorptive enterocytes and into the epithelial cells. According to Margolskee, this is typically the rate-limiting step for glucose uptake in the small intestine. Margolskee, Shirazi-Beechey, and colleagues found that SGLT1 mRNA (messenger RNA), SGLT1 protein expression, and glucose uptake activity in wild-type mice all increased when the mice were treated with a high-carbohydrate diet compared with a low-carbohydrate diet. But in knockout mice missing T1R3, a component of both the sweet and umami receptors, there was no difference in SGLT1 between the low- and high-carbohydrate diets. Likewise with gustducin knockout mice, the research revealed no difference in SGLT1 mRNA or protein or glucose uptake activity between the low- and high-carbohydrate diets. According to Margolskee, the evidence suggests that both T1R3 and gustducin are necessary to elicit an increase in SGLT1 in response to dietary carbohydrate and a subsequent increase in glucose uptake activity.

Margolskee described a similar effect observed in knockout mice fed either a low-carbohydrate diet alone or a low-carbohydrate diet supplemented with a noncaloric sweetener (i.e., sucralose) (Margolskee et al., 2007). Wild-type mice showed an increase in SGLT1 mRNA, SGLT1 protein, and glucose uptake activity when their low-carbohydrate diet was supplemented with a noncaloric sweetener, but knockout mice did not. These results indicate a chemosensory detection pathway in the gut that responds to luminal sugars and luminal sweeteners and leads to the up-regulation of SGLT1 and an increase in glucose uptake activity across the gut.

Margolskee and others have found taste-like receptors not just in the stomach and intestine but also in the pancreas. Margolskee described unpublished data showing the expression of gustducin in pancreatic islet alpha cells and the expression of T1R3 in both alpha and beta cells. The function of these pancreas taste-like receptors is unclear. However, both in vitro data and data from wild-type versus T1R3 knockout mice suggest that these receptors play a role in sweetener-enhanced insulin release.

Oral Taste Cells and the Expression of Gut Proteins

Margolskee noted that researchers have observed a number of gut hormones, including GLP1, GIP (gastric inhibitory peptide), and CCK, expressed in multiple types of oral taste cells. Oral taste cells also express intestinal sugar sensors, such as SGLT1, and pancreatic metabolic sensors (Yee et al., 2011).

Margolskee gave an example of the expression of gut proteins in the oral cavity. Based on studies with T1R3 knockout mice showing a loss of response to noncaloric sweeteners but not to sugars (Damak et al., 2003), he and his colleagues suspected that something else in the oral cavity besides the oral sweet receptor, a T1R2 and T1R3 heterodimer, responds to sugars. They hypothesized the presence of a glucose transport pathway similar to what has been observed in pancreatic beta cells. Indeed, they found that a number of the same pancreatic pathway components were present in oral taste tissue (Yee et al., 2011). Margolskee speculated that gut-like glucose transporters in taste cells may help people and animals distinguish caloric from noncaloric sweeteners.

Taste-Like Receptors in the Gut and Pancreas: Summary of the Science

In summary, Margolskee noted that researchers have identified whole taste signaling pathways in both the gut and pancreas and in both the proximal and distal gut. In the gut, taste elements are expressed in L, K, and L/K cells. In the pancreas, both pancreatic islet alpha and beta cells express taste elements. Gustducin and T1R3 in the gut are involved in the release of GLP-1 and GIP in response to sweeteners and, in the proximal gut, in the regulation of SGLT1 levels. In the colon, gustducin appears to be involved in the release of GLP-1 and GIP in response to short-chain fatty acids. With regard to the role of taste signaling molecules in the pancreas, preliminary evidence suggests that gustducin and T1R3 are involved in sweetener detection and, under some circumstances, insulin secretion.


Robert Ritter elaborated on information and ideas presented earlier by Timothy Moran and explored in more detail how GI peptides, CCK in particular, provide the brain with information that contributes to the process of satiation and reduces food intake. He focused on CCK because scientists know more about how it modulates vagal afferent activity compared with what is known about other GI peptides.

GI Peptides

GI peptides are localized in specialized enteroendocrine cells scattered among the cells of the absorptive and secretory mucosa of the GI tract, from the stomach through the colon. Nerve fibers pass through the extracellular space beneath the mucosa, into which GI peptide secretion occurs, creating the opportunity for both endocrine and neuronal peptide actions. According to Ritter, although the actions of some GI peptides were discovered in the early 20th century (e.g., 1902 for secretin and 1905 for gastrin), none of the GI peptides were identified as peptides per se until the 1960s and 1970s, when they were synthesized and sequenced. A dozen or more GI peptides have been identified to date. Several are involved in control of food intake, including CCK, which is secreted in the proximal small intestine, and GLP-1, PYY 3-36 (peptide tyrosine tyrosine), and oxyntomodulin, all of which are secreted by L cells in the more distal small intestine and large intestine. CCK, GLP-1, PYY 3-36, and oxyntomodulin all reduce food intake (e.g., Chelikani et al., 2005; Ritter, 2010). Ghrelin, which is released from cells in the gastric mucosa, increases food intake.

Ritter went on to explain that after their secretion from enteroendocrine cells, GI peptides in the blood can broadcast a signal to any tissue with a matching receptor, including tissues in GI organs where the peptides help coordinate digestive function. Early during the digestive process, they contribute to slowing gastric emptying and stimulating pancreatic secretion of enzymes and bicarbonate. Later they facilitate secretion of insulin and the postabsorptive assimilation of nutrients (see the review by Rehfeld, 2011). GI peptides also play an important role in limiting food intake. In Ritter's opinion, food intake can be viewed as yet another part of the digestive process, given that reducing food intake limits the inflow of food into the digestive tract during a meal and thereby facilitates the efficient digestion and absorption of what has been eaten. In addition to their impact on GI tissues, GI peptides act on the brain and innervation of the GI tract (see reviews by Banks, 2008; de Lartigue, 2014; and Schwartz, 2010).

According to Ritter, a hallmark of GI peptides is that their secretion and levels in circulation are controlled by nutrients in the GI tract during a meal. When a meal is eaten, levels of GI peptides in the blood rise dramatically (Ellrichmann et al., 2008). Initially, upon entry of nutrients into the intestine, CCK levels rise rapidly to six or seven times their fasting level. Soon thereafter, GLP-1 and PYY 3-36 levels rise as well. The initial rapid rise in CCK levels has been shown to facilitate the release of the other peptides in anticipation of actual direct stimulation of their secretion by nutrients as food moves down through the intestine.

Another hallmark of GI peptides, according to Ritter, is that their impact on the control of food intake is focused on limiting the size and duration of an ingested meal. CCK, GLP-1, and PYY 3-36 all reduce food intake, primarily by reducing meal size and meal duration rather than by decreasing the number of meals initiated (see the review by Ritter, 2010).

The Cellular Mechanisms by Which GI Peptides Modulate Vagal Afferent Activity

Ritter elaborated on what Moran had discussed about CCK reducing food intake through its effect on vagal afferent neurons. According to Ritter, a vagal mode of action characterizes not only CCK but most other GI peptides as well; in fact, their ability to reduce food intake is attenuated or virtually abolished when the abdominal vagus nerve is cut. For ghrelin, however, the stimulatory effect on food intake is more complicated. According to Ritter, ghrelin appears to antagonize the excitatory effects of some of the other GI peptides on vagal afferent firing, although a role for the vagus in actually mediating the increase in food intake through ghrelin is doubtful.

All vagal afferents release glutamate, a neurotransmitter, in the hindbrain. Thus, not surprisingly in Ritter's opinion, CCK-induced reduction of food intake has been shown to be sensitive to antagonism of glutamate receptors in the hindbrain. In fact, antagonism of NMDA-type (N-methyl-D-aspartate) glutamate receptors with selective receptor antagonists injected directly into the hindbrain reverses or prevents reduction of food intake by exogenously administered CCK (Wright et al., 2011).

An interesting feature of vagal afferent fibers, according to Ritter, is their very quick release of all available neurotransmitters and failure over time. Susan Appleyard has shown that upon stimulation of vagal afferent inputs, postsynaptic cells fire but then fail; however, their failure can be reversed by local application of CCK (Appleyard et al., 2005).

In terms of the specific cellular mechanism by which CCK enhances vagal afferent transmission, Ritter has found that CCK activates an enzyme, an extracellular receptor kinase, that phosphorylates synapsin. Synapsins are proteins that bind synaptic vesicles to the cytoskeleton of the neuron; they help control the availability of neurotransmitters for release. When phosphorylated, synaptic vesicles are freed from the cytoskeleton and the availability of transmitters for release is increased. When dephosphorylated, the vesicles remain bound to the cytoskeleton of the neuron and fewer transmitters are available for release (Cesca et al., 2010). Normally, CCK reduces food intake for only a short period of time, about 30 minutes, but inhibiting dephosphorylation of synapsin can extend and enhance the ability of CCK to reduce meal size (Campos et al., 2013). According to Ritter, it is not yet known whether other GI peptides operate in a similar way.

The Impact of Non-GI Proteins on Food Intake

Ritter emphasized that the GI signals controlling food intake are directly related to food that has just been consumed and is in the process of being digested and absorbed. However, other parts of the physiology of an organism provide the brain with indirect information about metabolism that can also impact food intake. Notable among these, said Ritter, is leptin, a protein produced by adipose tissue. Injection of leptin into rats and mice dramatically reduces food intake by reducing meal size, with administration over days or weeks leading to weight loss (Kahler et al., 1998).

Given that leptin acts on the brain to produce reductions in meal size in a manner very similar to that of feedback signals from GI tract hormones such as CCK, Ritter and his colleagues were driven to ask whether vagal afferent function is modulated in any way by leptin. Indeed, interaction between leptin and gut hormones begins in the GI tract, at the peripheral vagal afferents. About 45 percent of vagal afferents that innervate the stomach and small intestine express both CCK and leptin receptors (Peters et al., 2006). It has been shown that leptin and CCK can enhance each other's action, with the combined administration of subthreshold doses of both substances resulting in reduced meal size (i.e., when administered alone, subthreshold doses of either do not reduce meal size) (Peters et al., 2005).

Nevertheless, according to Ritter, there is good evidence that leptin produces major effects on food intake by acting on the hypothalamus, where it activates what are known as POMC (pro-opiomelanocortin) neurons and increases release of alpha-melanocyte-stimulating hormone (alpha-MSH), which then acts on the melanocortin-4 (MC4) receptor (see the review by Ellacot and Cone, 2004). Of interest, Ritter noted, antagonism of the MC4 receptor also attenuates the response to CCK (Sutton et al., 2005; van Swieten et al., 2014).

Ritter and his colleagues have hypothesized that the modulatory effect of leptin occurs at the vagal afferent terminal itself. Evidence to this effect includes MC4 receptor expression by vagal afferents (Wan et al., 2008) and close interaction between vagal afferent neurons and POMC fibers in the hindbrain. Indeed, Campos and colleagues (2014) demonstrated that POMC neurons act at receptors at the first presynaptic element in the visceral afferent communication pathway and that administration of an MC4 agonist into the hindbrain can elevate phosphorylation of synapsin for hours. The ultimate effect, Ritter explained, is that leptin-initiated activation of MC4 enhances vagal afferent transmission and normally, transmission from the vagal afferents to the hindbrain experiences about a 70 percent failure rate. Activation of the MC4 receptor cuts that rate in half. It also decreases the rate of decline of the amplitude of postsynaptic depolarizations that occur in response to vagal stimulation. Essentially, then, MC4 activation increases the fidelity and strength of vagal afferent transmission.


Based on this growing body of evidence, Ritter proposed a model in need of further study: CCK and other gut peptides activate vagal afferents and provide the primary signal for satiation, but the signal is modulated by leptin and perhaps other endocrine signals. Ritter described the vagal afferent ending as a “paintbrush that paints the … sensory process of satiety … on the hindbrain.”

Ritter concluded by emphasizing that several GI peptides are involved with food intake and that they all interact with each other as well as with relevant non-GI hormones to reduce food intake. One of the places where they interact is the first visceral afferent synapse in the nucleus of the solitary tract of the hindbrain, which, he said, is where the experience of satiation begins.


Laurette Dubé considered the different levels of context within which brain-digestive system interactions operate. Specifically, she considered how “higher-level” brain systems and mental processes (i.e., attention, cognition, and free will); the fetal environment and lifelong programming; parenting and other familial influences; and the broader social, commercial, and cultural food environment can impact eating behavior.

Impact of “Higher-Level” Brain Systems and Mental Processes on Eating Behavior

“Higher-level” brain systems bearing on cognitive, reward learning, and executive control processes serve as the first-level context within which brain-digestive system interactions operate. Dubé referred workshop participants to two recent reviews of scientists' understanding of that context: (1) Dagher (2012), on brain regions activated during functional magnetic resonance imaging (fMRI) studies of food cue reactivity, and (2) Vainik et al. (2013), on neural behavioral correlates with eating behavior and body mass index (BMI).

Dubé then described in detail two empirical studies she and her colleagues conducted based on the Dutch Eating Behavior Questionnaire (DEBQ), used to assess three types of eating behaviors: restrained, emotional, and external (van Strien et al., 1986). External eating involves a predisposition to ignoring homeostatic signals and reacting primarily to external hedonic cues (Burton et al., 2007; Rodin and Slochower, 1976). Together, the results of these two studies suggest to Dubé that as scientists move forward in their quest to understand eating behavior, they need to study more closely the interactions among the rewarding, executive, and homeostatic control regions of the brain and their psychological and behavioral correlates.

In the first study (unpublished) Dubé described, her research team asked participants to come to the laboratory and work on a puzzle. While working on the puzzle, the participants were interrupted six times to eat chocolate. Some participants were instructed to remain focused on the experience of eating chocolate, others to continue working on the puzzle. The researchers evaluated impact on consumption by measuring self-reported hunger before and after consumption. They found that high-external eaters behaved as expected based on reports in the literature; that is, they experienced a much more intense hedonic response and only a small change in hunger before and after consumption. Low-external eaters, in contrast, experienced a significant decline in hunger before and after consumption when distracted by the puzzle task and not focused on the sensory experience of eating chocolate. This finding reflects their individual predisposition to rely on biological processes more than on environmental cues. When asked to focus on the chocolate, however, low-external eaters experienced no decrease in hunger, their attention to sensory cues seemingly interfering with usual biological signals.

In the second study (Lebel et al., 2008), Dubé and colleagues evaluated change in hunger and fullness before and after consumption among “high schematics” versus “low schematics.” High-schematic eaters score high for all three DEBQ types of eating and are driven by both emotion and external cues, but also show cognitive restraint. In other words, their eating behavior is driven by a full array of mental schemata, attempting to overrule biological processes. Low-schematic eaters score low on all three types of eating. Participants were asked to provide self-reports of hunger and fullness both before and after consuming “comfort food.” The researchers found no difference in hunger between the high and low schematics either before or after consumption. However, they did find significant differences in preconsumption fullness and change in fullness (between pre- and postconsumption), with high schematics reporting greater preconsumption fullness and a smaller change in fullness upon eating, and low schematics reporting being less full before consumption and experiencing a greater change in fullness upon eating.

In a third study, Finkelstein and Fishbach (2010) provided participants with a chocolate bar and framed the food as either “healthy” (i.e., chocolate health bar) or “tasty” (i.e., chocolate candy bar). Participants also were either told that their job was to taste the bar (imposed consumption) or asked whether they would like to try it (free choice). The researchers found that participants who were told that the bar was tasty reported similar levels of hunger after consumption regardless of whether consumption was imposed or they were given free choice. In contrast, participants who were told that the bar was healthy reported significantly greater hunger after consumption when consumption was imposed compared with when they were given free choice. Again, for Dubé, these results highlight the need for scientific study of eating behavior and the complex interplay among different brain systems within a broader behavioral context.

Impact of the Fetal Environment on Eating Behavior

Dubé characterized the fetal environment as a key context in biology and behavior. She pointed to the Barker hypothesis as an example. Barker (1990) hypothesized that low birth weight is associated with increased risk of metabolic syndrome, diabetes, and obesity later in life. Dubé pointed workshop participants to a forthcoming review in the Annals of the New York Academy of Sciences on intrauterine growth restriction (IUGR) and its impact later in life.

In fact, researchers are finding correlations between IUGR and eating behavior not just later in life but early on as well. A study of 24-year-old women who had been observed over their lifetime showed that low-birth-weight women were consuming more carbohydrates and had higher BMIs (Barbieri et al., 2009). Meanwhile, a study of 27-week-old preterm newborn babies showed that low-birth-weight babies reacted less to sensitivity tests, postulated as being due to increased need, compared with non-low-birth-weight babies of the same gestational age (Ayres et al., 2012). Numerous other studies have found similar correlations across a wide range of ages (e.g., Crume et al., 2014; Escobar et al., 2014; Kaseva et al., 2013; Lussana et al., 2008; Perälä et al., 2012; Stein et al., 2009).

Dubé and colleagues recently collected self-reported birth weight data for 616 children aged 6 to 12 years from both the children and their mothers and used the DEBQ to measure eating behaviors and daily food consumption. They found that low-birth-weight children showed the same pattern as in the previous literature, with higher consumption of fat and sugar (manuscript under review). They also examined high-birth-weight children—that is, children born with high BMIs—and found that the high-birth-weight children showed more restrained eating and more emotional eating (as defined by the DEBQ) compared with controls, but no difference in external eating (manuscript under review). According to Dubé, both increased restricted eating and increased emotional eating are associated with obesity and high BMIs.

Impact of the Parental/Familial/Home Environment on Eating Behavior

In the same cohort of 616 children aged 6 to 12 years discussed above, Dubé and colleagues also measured attachment (Muris et al., 2001). Attachment is an extensively studied construct in both animals and humans, Dubé explained, with a measure of attachment providing information about the role of the primary caregiver in defining how an animal or person decides to explore beyond what has been programmed at birth. More secure attachment allows child and adult alike to engage with confidence in novel activities, including exploring alternatives to biological programming such as an innate liking for sugar (typical of high-calorie food) and dislike of bitter foods (which typically encompass many nutritious foods, including vegetables). Using 24-hour recall not just for food but also for other healthy and unhealthy eating-related habits, Dubé and colleagues found that children with insecure attachment experienced high eating schematicity for all three DEBQ eating behaviors; greater consumption of salty snacks; lower consumption of water and fruit; and greater likelihood of skipping breakfast, eating out, and eating in front of the television during weekdays. In Dubé's opinion, these findings suggest that more attention should be paid, in both research and practice, to exploring how the early home environment influences a life course of eating behavior.

Other relevant findings include Puhl and Schwartz's (2003) report that parental food rules can influence eating behavior, with some parents using food to reward or punish and encourage or discourage good or bad noneating behavior. Parents applying a control food rule typically use high-calorie food to encourage good behavior. Dubé cited a study showing higher caloric content, fat, and sugar in the diets of children exposed to parental food control rules. This effect was stronger for children (in particular boys) with an individual predisposition to being responsive to rewarding environmental cues as indexed by the Behavioral Activation System (BAS) scale, with children scoring on the high end of the scale tending to be more sensitive to reward (Carver and White, 1994). Dubé cited Côté and Moskowitz (1998), Lu et al. (2011), and Stroebele and De Castro (2004) as additional relevant studies.

Impact of the Broader Social, Commercial, and Cultural Environment on Eating Behavior

Dubé explained that plentiful correlational evidence collected at the population level over the past few decades links changes in eating behavior and BMI with various changes in the food environment. Examples are the increased availability of processed food, typically with high fat, sugar, and salt content, and increased food advertising (Buijzen et al., 2008; Dhar and Baylis, 2011; Foster et al., 2014; Franco et al., 2009; Kunkel et al., 2004; Powell and Bao, 2009; Powell et al., 2007; Scott et al., 2008).

Dubé argued that it is necessary to examine the effects of the food environment on individual and social processes. She reported results of a study conducted in the Montreal metropolitan area (Buckeridge et al., 2014) that found a correlation between area-level carbonated soft drink sales and median personal income. A decrease of $10,000 in income was associated with almost a five-fold increase in soft drink sales. In another study conducted in Montreal, an individual food environment was defined by a buffer zone around a person's residential address (Paquet et al., 2010). That study demonstrated interactive effects between the density of fast-food restaurants and eating behavior. Individuals scoring low on the BAS were not influenced by the density of fast-food restaurants, while those scoring high on the BAS consumed more fast food when exposed to a higher density of fast-food restaurants. Dubé urged more such studies. She encouraged the use of geographic information systems (GISs) to aid in examining multiple layers of data for the same geographic area.

The Brain-to-Society Model of Eating Behavior

For almost 10 years now, Dubé has been leading a network of McGill University and other scientists in studying eating behavior in its broader context. Together, they developed the Brain-to-Society (BtS) model of eating behavior (Dubé et al., 2008, 2010). The BtS model is based on the premise that eating is a neurobehavior that operates in contexts on different sectoral, temporal, and geographic scales. Not only does each contextual level need to be studied by itself in depth, Dubé opined, but the different levels also need to be studied in combination through a systems science framework (Dubé et al., 2012; Hammond and Dubé, 2012).


Following Dubé's presentation, the speakers in session 1 participated in a panel discussion with the audience. Questions from the audience spanned a wide range of topics.

Nutrient-Specific Signaling: What Does the Science Say?

During his presentation, Moran had emphasized that vagal afferents innervating the stomach were stimulated by load volume, not content. A member of the audience observed that Moran had presented gastric load data from experiments using glucose and asked whether other macronutrients produced the same effect. Moran replied that he and his research team compared glucose and casein and observed no difference. Additionally, in experiments using pyloric cuffs,12 no differences in subsequent food intake were observed across loads of different nutrient characters (Phillips and Powley, 1996). Moran reiterated that in the stomach, the reduced food intake response is a response to gastric volume. He pointed to work showing that in the intestine, on the other hand, nutrient content can be sensed and can guide behavior (Sclafani and Akroff, 2012).

While the discussion was on the topic of nutrient-specific responses, Margolskee was asked whether any other macronutrients produce taste-like responses similar to what he and his colleagues observed with the sweet taste-like receptor and response. Margolskee replied that he and his team observed responses in the proximal gut to sugars and sweeteners, triggering the release of the gut hormones GLP-1 and GIP. But in the distal gut, where one would not expect sugars to be reaching, they observed at least some association with short-chain fatty acid responses leading to release of GLP-1 (Li et al., 2013). In Margolskee's opinion, then, different macronutrients do in fact trigger taste-like responses depending on where in the gut the GLP-1–producing L and GIP-producing K cells are located.

With respect to bitter taste, Margolskee said, the evidence for expression of the bitter T2R receptor in the gut is weaker than the evidence for the sweet taste receptor molecules, as is the evidence for a physiological role for bitter taste-like receptors in the gut. With respect to salt, there is good evidence that ENaC is involved in a low sodium concentration response in the oral cavity. Also in the oral cavity, there is likely a different, still unidentified receptor involved in a high sodium concentration response. But according to Margolskee, it is unclear how what is happening in the oral cavity relates to salt-responding cells in the gut.

Taste and Taste-Like Cells: What Does the Science Say?

Several questions were raised about taste and taste-like cells. First, an audience member asked whether tastes have differential effects on reward and subsequent eating behavior. For example, would subsequent eating behavior differ if umami were placed in the gut instead of glucose? And do different amino acids placed in the gut have different satiety potency? The audience member cited evidence from Kunio Torii that monosodium glutamate is particularly effective in the gut in producing satiety and controlling dietary-induced obesity (Yasumatsu et al., 2012). Noting that the umami oral taste system in rats appears to be more specifically sensitive to monosodium glutamate relative to other amino acids than is the case in humans, he asked whether the same is true of the umami gut system.

Margolskee remarked that the umami taste system is highly complex, even in the oral cavity. In addition to significant differences in umami receptors, T1R1 and T1R3, in the oral cavity of rodents versus humans, which may explain some sensory differences between rodent and human preferences for particular amino acids, there is good evidence to suggest that other receptors play a role as well. But it is difficult to tease apart which receptors are involved with which amino acids. In Margolskee's opinion, this is likely as true of umami receptors in the gut as of those in the oral cavity. That being said, while taste receptors in the oral cavity are “pretty good” at distinguishing one nutrient from another—that is, sweet from salty from bitter from umami and so on—preliminary evidence suggests that the gut taste-like receptors may not be as sensitive. Some taste-like cells appear to have both sweet and umami receptors, for example, or both sweet and bitter receptors. Margolskee suggested that some taste-like calls in the gut may be more generalist chemosensory cells rather than what he referred to as “segregationist” cells.

During his presentation, Margolskee briefly touched on the existence and role of taste-like receptors in the pancreas. An audience member asked whether the same pancreatic response that has been observed in wild-type mice—that is, that sucralose promotes insulin release—would be expected in mice or rats that are prediabetic or have type 1 diabetes. Margolskee replied that one would expect the same kind of response, but the question has not been studied.

Margolskee also was asked about oral sensory detection of fat and its effects on physiology. Whether fat is a real taste is still controversial, Margolskee said; there are some fat receptor candidates, but the evidence is “complicated” and “messy.” In his opinion, there is a fat taste and probably an appetitive fat taste that is different from the free fatty acid taste responses. He mentioned work he is doing in collaboration with Anthony Sclafani and John Glendinning (Sclafani et al., 2007) on gustducin and TRPM5 knockout mice, suggesting that there may be oral and postingestive gut endocrine fat sensors tied to some taste proteins. “But a lot of work is yet to be done,” he said.

Relative Importance of Examining Tissue-Level Responses Versus Whole-Organism Responses to Food

During his presentation, Ritter emphasized that stimulation of one type of nerve fiber can influence the response of other types of nerve fiber (because of the proximity of different types of nerve endings in the brain). This and other observations led an audience member to ask the panel to comment on whether studying cells or tissues in isolation creates a different impression of brain-digestive system interactions compared with studying whole organisms. Margolskee replied, “Ideally, one would be looking at the whole organism [and] integrative systematic responses. From a practical point of view, we do many reductionist, reduced preparations where we drive the system to be able to see a response, for example, with isolated pancreatic eyelets. We can do things to the eyelets that would be much harder to do in the intact animal model.” He noted the struggle to interpret the importance of some of the observed effects of high-potency noncaloric sweeteners on insulin and GLP-1 responses. Whereas he and his research team have shown that high-potency noncaloric sweeteners can drive changes in insulin levels in isolated eyelets, Rebecca Brown's work with noncaloric sweeteners in human subjects has demonstrated an increase in GLP-1 levels but no change in insulin levels (Brown et al., 2009). So the physiological relevance of what Ritter and his team have observed with respect to changes in insulin is questionable. On the other hand, it may be worth considering the possibility that there are long-term effects of many years of high ingestion of high-potency noncaloric sweeteners. Ritter said, “I tried to be fairly cautious in not overinterpreting or overextending from the data, but I think there is something there worth noting and worth considering.”

In contrast to the questionable insulin responses, Margolskee said there are some clear systematic physiological responses. He noted Steven Munger's work demonstrating that cephalic phase13 responses can be driven in isolation with extracted tissues (Geraedts and Munger, 2013). Some cephalic phase responses appear to be “hardwired” into endocrine cells in the gut, Margolskee observed. In sum, he said, “It's a very complex system where we need to understand each of the parts and understand how it functions in totality.”

The Role of Animal Models in Understanding Human Eating Behavior

When asked about the use of animal models to study human eating behavior, Dubé opined that many processes can be studied with rats even at the presymbolic level of decision making. She pointed to Peter Shizgal's work on decision making in rats, which has documented how multiple sensory information and biological needs are integrated into a common currency driving the nature and quantity of food choices (Shizgal and Conover, 1996). However, many layers of complexity and diversity must be added in accounting for human choice. Dubé said, “If you want to study human behavior, you need to have all the pieces, but you also need to get the real world context…. It's not an either-or. It's a portfolio.”

Moran agreed that rats can be used to study more than physiological responses; they also can be used to study dietary preferences. He noted that some of the same fetal outcomes described by Dubé in her presentation can also be shown in rats. He said, “It's likely that a number of these long-term effects are mediated through epigenetic changes, and rodent models really provide a very good vehicle for getting at just what those kinds of specifics are.”

The Challenge of Studying Overall Control of Eating: Integrating Homeostatic and Reward Responses to Food Stimuli

An audience member asked how researchers plan to integrate what is known about the homeostatic processes described thus far—that is, all of the various neural mechanisms mediated largely through the vagal nerve and mainly in the hindbrain—with what is known about reward-related dopamine responses in the brain. For example, how can one integrate what is known about individual peptides involved in the control of meal size with what is known about reward processing that goes on in the brain in response to food stimuli? Dubé emphasized that all areas of study contribute information. In her opinion, methodological interfaces could be developed to integrate those pieces of information. When McGill University hosted the first BtS model think tank in 2005, Dubé was struck by the disconnect in people's thinking about the different processes and parts of the brain associated with eating. The situation has changed since then, she said. Still, she encouraged development of an interface protocol and stressed the importance of having a sense of the system as whole while studying single pieces. Having knowledge of the individual pieces is “absolutely necessary,” she said, but it is also necessary to understand those pieces within their larger context. She urged a systems-level approach to moving forward.

Ritter opined that food intake is controlled largely by sensory experiences. Those experiences, he said, “ascend” into the reward areas of the brain and likely influence responses from the GI tract in a “descending” manner. He, too, stressed the importance of gaining a better understanding of the individual pieces and then studying them in their broader context.

Moran added that many of the hormones being studied for their peripheral pathway effects have the ability to cross the blood-brain barrier and directly impact brain reward pathways.

Gut Peptides: Hormone Versus Paracrine Signaling and the Impact of Overall Metabolic State

An audience member noted that many studies have found no relationship between levels of gut hormones circulating in the blood and subjective measures of appetite or food intake. She asked, “Are we going down the wrong path looking at those gut hormones as opposed to knowing that there's that direct effect that's happening in the gut and in the brain?” Moran replied that 20 years ago, one of the arguments against a physiological role for gut hormones in contributing to satiety was a lack of that type of correlational data. However, antagonist experiments have made clear that, rather than a hormonal role, many of these peptides likely play a paracrine role in contributing to satiety. Also, a number of studies have shown that hormones released during one meal do in fact have an effect on meal termination in subsequent meals. Moran explained that many of the positive correlations being seen today are in bariatric surgery conditions, where hormone release is greatly exaggerated because of the anatomical changes associated with the surgery and nutrients accumulate in high concentrations in areas where they normally would not accumulate.

Ritter identified two relevant areas in need of further investigation. The first is the way G protein-coupled receptors display constitutive activity, that is, activity even with very low levels of agonist present, which suggests that the very existence of G protein-coupled receptors facilitates signaling of the nerve. Second is the effect of the metabolic state of an animal or person on hormonal and behavioral response.

Another audience member asked whether some of the “nonsatiety” proteins, such as adiponectin and glucagon, should be studied for their potential role in food intake signaling. She also asked about the role of the liver in appetite regulation and the importance of considering overall metabolic state. Regarding the latter, might some of the lack of correlation between hormone level in the blood and food intake be related to lack of consideration of overall metabolic state? Ritter agreed that there is much to be learned about the role of multiple cytokines in regulating food intake during both illness and health. He reiterated that overall metabolic state, as well as other contextual and neural factors, needs to be considered when evaluating food intake, overeating, and obesity.

Moran agreed that from a therapeutic standpoint it will be important to understand not just individual signals but the range of signals and their interactions, and how those interactions change across different metabolic states.

Loss of Vagal Afferent Feedback: Obesity and Dementia

Questions were raised about whether loss of vagal afferent feedback may in any way contribute to either obesity or dementia. First, given that obesity can be considered a state of overconsumption, is there any evidence that loss of nutrient-activated vagal afferent feedback contributes to obesity? Ritter replied, “The short answer is yes.” Studies in both animals and humans indicate that down-regulation of leptin sensitivity, for example, leads to an increase in meal size. More generally, type of diet (e.g., cafeteria diet) can induce changes in vagal afferent signaling that lead to decreased nutrient sensing and decreased caloric feedback. When asked whether overstimulation or macronutrient content drives decreased sensitivity, Ritter replied that there is evidence for both mechanisms.

An audience member mentioned that her father suffered a brain injury and then gained about 100 pounds in 100 days. In her opinion, “there was a feedback loop that just wasn't working.” She asked whether similar malfunctioning feedback loops may contribute to Alzheimer's disease, given its hypothesized relationship with insulin sensitivity (i.e., it has been dubbed by some experts as a “type 3” diabetes). Moran replied that various kinds of brain injuries are known to produce excessive weight, generally in response to excessive food intake. With respect to the relationship between Alzheimer's and a malfunctioning food intake feedback loop, he noted the well-documented relationship between a drop in insulin sensitivity and Alzheimer's and other forms of dementia and the known effect that a drop in insulin sensitivity has on food intake. He said, “There is a deficit in the brain's ability to get the kind of glucose that it needs for normal functioning.”

Childhood Development and Obesity

Dubé was asked about the importance of considering upbringing when conducting cross-sectional studies on food intake comparing lean versus obese individuals. She replied that examining children and their relationships with both food and reward is key to understanding food intake and obesity. This is especially true when applying the addiction model to food. In Dubé's opinion, not only does an excessive focus on addiction processes in food intake fail to account fully for the complex interplay with metabolism and energy balance that is less relevant with, for example, cocaine; it also neglects reinforcement learning processes that are core to reward learning (with or without addiction) and, in the context of food, may certainly be as important at cognitive learning. Dubé emphasized the importance of early exposure to high-calorie, high-fat, and high-sugar foods—or to healthier alternatives—in setting a life course of reinforcement learning that impacts what children learn to like.

Studying the Social Context of Eating Behavior

Dubé also was asked how population-level data, such as the soft drink consumption data that she presented, could be used to generate hypotheses about eating behavior and how better-quality population data could be collected. She replied by emphasizing that the very rigorous standard for collecting population data for epidemiological study needs to be applied in the study of food environments, accounting for sampling and other research methods. She noted that the soft drink data she used in her analysis were predictions based on available private data and that she and her research team used predicted rather than actual data in order to derive population-level estimates and draw inferences at the population level.



Daniel Greenberg, Ph.D., F.A.C.N., is a Food Forum member and was a member of the workshop planning committee.


This section summarizes the presentation of Timothy Moran, Ph.D., Johns Hopkins University School of Medicine, Baltimore, Maryland.


The myenteric plexus and submucosal plexus are networks of neurons located in different areas of the wall of the digestive tract. The myenteric plexus is located between the layers of longitudinal and circular muscle (two layers of muscle involved with propulsive activity within the intestine), while the submucosal plexus is located between the circular muscle layer and the mucosa.


Sham feeding involves providing an animal with a liquid diet, which descends through the esophagus but immediately drains out from the stomach, thereby eliminating gastric stretch and intestinal stimulation.


This section summarizes the presentation of Robert Margolskee, M.D., Ph.D., Monell Chemical Senses Center, Philadelphia, Pennsylvania.


Papillae are small structures on the upper surface of the tongue.


G protein-coupled receptors are proteins located in the cell membrane that bind extracellular substances and transmit signals from those substances to an intracellular molecule known as a G protein.


ENaC is the epithelial sodium channel, a membrane-bound channel permeable to sodium ions and other substances.


L and K cells are types of intestinal enteroendocrine cells. L cells secrete GLP-1; K cells secrete gastric inhibitory peptide (GIP).


This section summarizes the presentation of Robert Ritter, V.M.D., Ph.D., Washington State University, Pullman.


This section summarizes the presentation of Laurette Dubé, Ph.D., M.P.S., M.B.A., McGill University, Montreal, Quebec, Canada.


A pyloric cuff is a device used to tighten the pylorus and prevent food from leaving the stomach, allowing researchers to separate gastric from intestinal factors.


The cephalic phase is a phase of gastric secretion that occurs before food enters the stomach.

Copyright 2015 by the National Academy of Sciences. All rights reserved.
Bookshelf ID: NBK279994


  • PubReader
  • Print View
  • Cite this Page
  • PDF version of this title (1.5M)

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...