Vagal sensory gut-brain pathways that control eating - satiety and beyond

Abstract

The vagus nerve is the body’s primary sensory conduit from gut to brain, traditionally viewed as a passive relay for satiety signals. However, emerging evidence reveals a far more complex system—one that actively encodes diverse aspects of meal-related information, from mechanical stretch to nutrient content, metabolic state, and even microbial metabolites. This review challenges the view of vagal afferent neurons (VANs) as simple meal-termination sensors and highlights their specialized subpopulations, diverse sensory modalities, and downstream brain circuits, which shape feeding behavior, metabolism, and cognition. We integrate recent advances from single-cell transcriptomics, neural circuit mapping, and functional imaging to examine how VANs contribute to gut-brain communication beyond satiety, including their roles in food reward and memory formation. By synthesizing the latest research and highlighting emerging directions for the field, this review provides a comprehensive update on vagal sensory pathways and their role as integrators of meal information.

Graphical Abstract

The vagus nerve is a key conduit for gut-brain communication, integrating mechanical, chemical, and hormonal signals to regulate feeding, metabolism, and cognition. This review explores how vagal afferent neurons shape satiety, nutrient sensing, and reward, providing new insights into their broader physiological roles beyond meal termination

1. Introduction

What does the gut tell the brain? One of the earliest and most compelling demonstrations of gut-brain communication came from classical sham feeding experiments (Young et al. 1974). These studies revealed that the mere presence of food in the stomach signals the brain to stop eating, highlighting the stomach’s critical role in controlling food intake. Interestingly, this cessation of eating was further suppressed when nutrients were directly infused into the intestine, suggesting that, beyond the stomach, the gut plays an essential role in providing feedback to the brain. Follow-up studies showed that disrupting gut-brain communication by means of vagotomy attenuated this nutrient-induced suppression of sham feeding (Yox, Stokesberry, and Ritter 1991), implicating the vagus nerve as a key mediator of these signals. While this early work established the importance of the vagus in meal termination (i.e., satiation), it is now clear that the gut’s communication with the brain extends far beyond signaling fullness. Over the last 15 years, increasing evidence has revealed that the vagus is not only central to controlling satiation, but also contributes to cognitive and physiological aspects of feeding behavior (Figure 1).

Figure 1. Vagal afferent neurons (VANs) regulate many aspects of feeding behavior.

Gut-brain communication through vagal afferent neurons regulates physiological (blue shade) and cognitive (pink shade) aspects of feeding, including satiety, satiation, gut motility and secretion, glucose metabolism, osmotic balance, thermogenesis, food memory, and food reward. Created in https://BioRender.com

The vagus nerve derives its name from its “wandering” nature. Unlike the spinal sensorimotor system, which is neatly organized along the spinal cord, the vagus nerve arises in the brainstem and branches extensively across the thoracic and abdominal cavities, innervating organs as distal as the colon. This unique anatomy allows the vagus to serve as a critical conduit between the brain and visceral organs. Functionally, the vagus carries both motor (efferent) and sensory (afferent) fibers (for extensive review on the anatomy of the vagus nerve, see (Berthoud and Neuhuber 2000)). Motor fibers originate in the brainstem, specifically from neurons in the dorsal motor nucleus (DMV) and nucleus ambiguus, and travel to the thoracic and abdominal cavities to regulate organ function (Browning and Travagli 2019). In contrast, vagal afferent neurons (VANs) located in the two nodose ganglia (left and right), act as sensory sentinels. These pseudounipolar neurons extend one branch to peripheral visceral organs, and another to the brainstem, where they synapse in the nucleus of the solitary tract (NTS). In the GI tract, VANs form specialized endings suited to distinct roles. Mucosal endings detect chemosensory stimuli, such as nutrients and gastric hormones; whereas intraganglionic laminar endings (IGLEs), located in the myenteric plexus, and intramuscular arrays (IMAs), located between muscle layers, respond primarily to mechanical stimuli like distension (Wang, de Lartigue, and Page 2020). Importantly, each VAN appears to be highly specific (Table 1); recent studies injecting retrograde tracers into several organs have found very little overlap of nodose ganglia neurons (Bai et al. 2019; Han et al. 2018; Williams et al. 2016; Zhao et al. 2022). Furthermore, these neurons exhibit distinct terminal fields in the NTS (Han et al. 2018), suggesting that each neuron innervates only one organ, reinforcing a 1:1 connectivity pattern.

Table 1.

Genetically-identified Vagal Sensory Neuron Populations and their Target Regions in the Gut.

Primary Marker	Subpopulation Markers	Innervation Terminals	Site of Innervation	Response to Stimuli	Function	References
Calca+ (CGRP, Feeding-Modulating Neurons)	General Calca+	Mucosal endings, IGLEs	Stomach, small intestine	Responds to nutrient detection and stretch	Modulates feeding behavior, reduces food intake upon activation	(Bai et al. 2019; W. S. Kim et al. 2021)
	Calca+, Gfra1+	IMAs	Stomach	Transient response to stretch	Mechanosensation, detecting muscle tension	(Zhao et al. 2022)
	Calca+, Gfra1+, P2ry1+	IMAs	Stomach	Transient response to stretch	Gut-brain feedback	(Zhao et al. 2022)
	Calca+, Gfra1+, Drd2+	IMAs	Stomach	Transient response to stretch	Dopamine-mediated gut-brain signaling	(Zhao et al. 2022)
Trpa1+ (Nociceptive, Fat-Sensing, and Inflammatory Neurons)	General Trpa1+	Mucosal endings, submucosal & myenteric plexus	Stomach, colon	Responds to irritants, inflammatory pain	Nociception, gut-brain protective reflexes	(Bai et al. 2019; Patil et al. 2023)
	Trpa1+, Npas1+	IMAs	Stomach	Transient response to stretch	Mechanosensation, detecting muscle tension	(Zhao et al. 2022)
	Trpa1+, Drd2+, Car8+	IMAs	Stomach	Transient response to stretch	Dopamine-related gut-brain interactions	(Zhao et al. 2022)
	Trpa1+ (Fat-Sensing)	Mucosal endings	Small intestine	Responds to fat infusion	Drives fat preference behavior	(M. Li et al. 2022a)
	Trpa1+ (Inflammation-Responsive)	Mucosal endings, nodose ganglion	Gut	Responds to inflammatory signals	Modulates immune responses	(Ichiki et al. 2022)
Piezo2+ (Mechanosensory, Gastric Stretch Neurons)	General Piezo2+	IGLEs	Stomach, intestine	Sustained response to stretch	Detects gastric distension, regulates satiety	(Bai et al. 2019; Zhao et al. 2022)
	Piezo2+, Glp1r+, Rbp4+	IGLEs	Stomach, duodenum	Sustained response to stretch	Mechanosensation, satiety regulation	(Zhao et al. 2022)
	Piezo2+, Lamp5+, Slit2+	IGLEs	Stomach	Sustained response to stretch	Gastric accommodation, stomach filling regulation	(Zhao et al. 2022)
Htr3a+ (Serotonin-Sensitive, Nausea-Modulating Neurons)	General Htr3a+	Mucosal endings	Stomach, intestine	Responds to serotonin release from enterochromaffin cells	Controls nausea, emesis, and gut motility	(Bai et al. 2019; S.-H. Kim et al. 2020)
	Htr3a+, Cckar+, Vip+, Glp1r+ or Uts2b+	Mucosal terminals	Duodenum	Responds to luminal contents	Chemosensation, nutrient detection	(Zhao et al. 2022)
	Htr3a+, Cckar+, Ctxn2+, Oxtr+	IGLEs	Duodenum	Responds to stretch, possibly polymodal	Mechanosensation and chemosensation	(Zhao et al. 2022)
Tac1+ (Substance P, Osmolality-Sensitive Neurons)	General Tac1+	Mucosal endings	Primarily jugular ganglion, few in nodose	Nociceptive	Pain-related signaling, gut-brain protective reflexes	(D. Kim et al. 2020)
	Tac1+ (Osmolality-Sensitive)	Mucosal endings, hepatic portal area	Gut	Responds to hypotonic stimuli	Detection of gut osmolality changes, regulates thirst	(Ichiki et al. 2022)
Gpr65+ (Acid-Sensing, Distension-Sensitive Neurons)	General Gpr65+	IGLEs	Stomach, intestine	Detects gastric distension	Mechanosensation, satiety regulation	(Bai et al. 2019)
	Gpr65+, Tmem233+, P2ry1+	Mucosal endings	Stomach	Responds to chemical stimuli	Chemosensation, detecting luminal contents	(Zhao et al. 2022)
Glp1r+ (Metabolic and Satiety-Regulating Neurons)	General Glp1r+	IGLEs	Stomach, intestine	Detects gastric stretch, metabolic signals	Mechanosensation, feeding behavior	(Bai et al. 2019; Brierley et al. 2021; E. K. K. Williams et al. 2016; J.-P. Krieger et al. 2016)
	Glp1r+, Piezo2+, Rbp4+	IGLEs	Stomach, Duodenum	Sustained response to stretch	Mechanosensation, satiety regulation	(Zhao et al. 2022)
	Glp1+, Vip+, Ust2b+	Mucosal endings	Intestine	Responds to chemical stimuli	Chemosensation, detecting luminal contents	(J. P. Krieger 2020)
Oxtr+ (Mechanosensory, Socially Modulating Neurons)	General Oxtr+	IGLEs	Stomach, intestine	Responds to stretch & metabolic cues	Mechanosensation, modulates feeding & social behaviors	(Bai et al. 2019; Scott et al. 2025)
	Oxtr+, Cckar+, Htr3a+	IGLEs	Duodenum	Responds to stretch	Mechanosensation and chemosensation	(Zhao et al. 2022)
Sst+ (Digestive Regulatory Neurons)	General Sst+	Mucosal endings	Small intestine	Responds to nutrient content	Chemosensation, digestion regulation	(Bai et al. 2019)
Vip+ (Nutrient-Sensing, Gut-Brain Signaling Neurons)	General Vip+	Mucosal endings	Small intestine	Responds to nutrient detection	Modulates gut-brain metabolic responses	(Bai et al. 2019)
	Vip+, Cckar+, Glp1r+, Htr3a+	Mucosal terminals	Duodenum	Responds to luminal contents	Chemosensation, nutrient detection	(Zhao et al. 2022)
Npy2r+ (Energy Balance-Regulating Neurons)	General Npy2r+	IGLEs & mucosal endings	Stomach, intestine	Inhibits vagal activity during fasting	Energy balance regulation, satiety	(Bai et al. 2019)
Cart+ (Gut-Brain Signaling Neurons)	Cart+, Oxtr+, Ctxn2+	IGLEs	Stomach and intestine	Responds to intestinal stretch	Inhibit food intake, slow gastric emptying, increase BAT thermogenesis	(Singh et al. 2021; de Araujo et al 2023)
	Cart+, Sst+	Mucosal endings	Stomach	Presumably response to stroking	TBD	(Singh et al. 2021)
	Cart+, Vip+, utsb2+	Mucosal endings	Intestine	Respond to luminal nutrients	Increase preference via memory, motivation and reinforcement learning	(Singh et al. 2021; de Araujo et al 2023)
	Cart+, Calca+	Mucosal endings	Stomach	Presumably transient response to stretch	TBD	(Singh et al. 2021)

Beyond neurons, the nodose ganglia also harbor non-neuronal cells that may play essential, yet poorly understood roles in vagal function. Satellite glial cells, which envelop neuronal cell bodies (Feldman-Goriachnik, Belzer, and Hanani 2015; Retamal et al. 2014; Shoji, Yamaguchi-Yamada, and Yamamoto 2010; Burdyga et al. 2008), are thought to modulate neuronal excitability and maintain homeostasis, while microglia-like cells may contribute to immune surveillance and inflammatory responses (Cawthon et al. 2020; Vaughn et al. 2017; Waise et al. 2015). Although these glial populations are well-characterized in the central nervous system, their contributions to vagal sensory processing remain largely unexplored. Understanding their interactions with VANs could provide new insights into how gut-derived signals are fine-tuned before reaching the brain.

In the central nervous system, VANs synapse in the NTS, using glutamate as their primary neurotransmitter. Furthermore, VANs express a wide range of neuropeptides that could potentially modulate vagal-NTS communication (Zhuo, Ichikawa, and Helke 1997), with CART and MCH being the most studied (de Lartigue 2014). The NTS serves as the first relay for gut-derived sensory information, connecting to the DMV to mediate rapid vago-vagal reflexes while also transmitting signals to higher brain regions, including the hypothalamus, hippocampus, and reward circuitry (de Araujo et al. 2023; Han et al. 2018). Through these pathways, information from the gut influences many aspects of feeding behavior beyond meal termination, including energy balance, food reward, and memory.

This review aims to dissect the mechanisms by which gut-brain circuits, particularly those mediated by VANs, control feeding behavior and meal-related physiology. We primarily focus on insights gained from rodent models, which have been instrumental in uncovering the functional diversity of VAN subtypes and their causal roles in physiology. We begin by outlining the diverse gastrointestinal signals that VANs detect, highlighting their specialized sensory modalities. Next, we explore how vagal gut-brain communication contributes to satiety and satiation, emphasizing the hindbrain circuits that inhibit food intake as food moves through the gastrointestinal tract. Beyond meal termination, we delve into the broader physiological roles of VANs, such as their contribution to gastrointestinal motility and secretion, thermogenesis, glucose metabolism, and drinking behavior. Finally, we examine cognitive aspects of feeding, focusing on how VAN-mediated signaling shapes reward processing and memory, offering insights into how gut-derived cues influence higher-order aspects of food-related decision-making.

2. Gut sensing

2.1. Mechanosensing.

VANs are key mechanosensors of the gastrointestinal (GI) tract, responding to distension and other mechanical stimuli. In vivo calcium imaging of the nodose ganglia and electrical recordings of vagal afferent fibers have demonstrated that VANs are activated by distension of the esophagus, stomach, and duodenum (de Araujo et al. 2023; Kentish et al. 2016; Williams et al. 2016; Zhao et al. 2022; Lowenstein et al. 2023). This mechanosensitivity is further supported by molecular studies showing that VANs express the mechanotransduction channels PIEZO1 and PIEZO2 (de Araujo et al. 2023; Kupari et al. 2019; Scott et al. 2025; Servin-Vences et al. 2023; Lowenstein et al. 2023). In the upper gut, a population of nodose neurons expressing the transcription factors Prox2 and Runx3 is highly enriched for PIEZO2. These neurons form extensive projections to the esophagus and stomach, where they terminate in IGLEs, reinforcing their role in mechanosensation and esophageal transit (Lowenstein et al. 2023). Interestingly, PIEZO 2 is also expressed in a population of VAN marked by oxytocin receptor (OXTR) expression (Scott et al. 2025). Some of these OXTR neurons form IGLEs in the stomach (Scott et al. 2025), intestine (Bai et al. 2019), and esophagus (Lowenstein et al. 2023), further supporting their role in gastrointestinal mechanotransduction.

In addition to OXTR neurons, VANs expressing the Glucagon-Like-Peptide1 Receptor (GLP-1R) also form IGLEs in the stomach and respond to distension (Bai et al. 2019; Williams et al. 2016), suggesting that GLP-1R VANs also contribute to mechanosensation. While these findings indicate that OXTR and GLP-1R VANs detect gastrointestinal distension, the direct influence of oxytocin and glucagon-like peptide-1 (GLP-1) on their mechanosensitive properties remains to be explored.

Beyond distension, vagal afferent fibers also respond to stroking of the gastric mucosa (Kentish et al. 2013a; Page, Martin, and Blackshaw 2002). Some mucosal-stroking-responsive fibers may also respond to bile (Page, Martin, and Blackshaw 2002) or distension (tension-mucosal receptors in the ferret; Page & Blackshaw, 1998), suggesting these neurons may have multimodal sensory capabilities. The genetic identity of these neurons is still unknown, but Bai et al. (2019) reported a subset of Calca-expressing VANs innervating the gastric mucosa that could correspond to these mucosal mechanosensory neurons. Stimulation of these neurons results in significant reduction in food intake (Kim et al. 2021). More research is needed to understand the role of gastric mucosal fibers in mechanosensation and to define the molecular mechanisms underlying their sensory modalities.

2.2. Chemosensing.

2.2.1. Nutrient sensing

In addition to detecting mechanical stimuli, VANs also respond to nutrients infused into the intestine, including amino acids, sugars, and fats (Jordi et al. 2013; M. Li et al. 2022a; McDougle et al. 2024; Tan et al. 2020; Tomé et al. 2009). However, the precise molecular mechanisms of nutrient sensing are still debated.

2.2.1.2. Glucose sensing

Electrical recordings of vagal afferent fibers indicate that glucose administration in the hepatic portal vein activates VANs (Niijima 1982). Similarly, intestinal infusion of glucose, sucrose, and galactose increases VAN activity as measured by calcium imaging experiments (McDougle et al. 2024; Tan et al. 2020). The administration of a sodium-glucose cotransporter 1 (SGLT1) blocker diminished glucose-induced VAN activation, suggesting that this transporter is necessary for glucose sensing (Tan et al. 2020). However, because SGLT1 is primarily involved in glucose absorption by enterocytes, it remains unclear from these experiments whether VANs detect glucose directly following its absorption by enterocytes or if enteroendocrine cells (EECs) are required to mediate vagal activation.

Several studies suggest that vagal glucose sensing may rely, at least in part, on enteroendocrine cells (EECs) as intermediaries. A subset of EECs expressing cholecystokinin (CCK), known as neuropods, has been shown to detect glucose in the intestine and form synapse-like connections with VANs (Kaelberer et al. 2018). These cells require SGLT1 for glucose detection and use glutamate as a neurotransmitter to activate vagal afferents. Notably, neuropods do not respond to non-caloric sweeteners (Buchanan et al. 2022), highlighting their role in transmitting caloric information rather than taste perception. While these findings support an EEC-mediated pathway, it remains unclear whether VANs also possess an independent mechanism for glucose sensing following its absorption by enterocytes.

2.2.1.3. Fatty acid sensing

Fatty acids, on the other hand, can be detected by the G-coupled receptor 40 (GPR40, also known as free fatty acid receptor 40), as observed in cultured nodose ganglion neurons (Darling et al. 2014). However, in vivo experiments using GPR40 knockout animals still show activation of VANs upon intestinal lipid infusion, suggesting GPR40 is not the only fat sensor (Li et al. 2022a). Conversely, knocking out both GRP40 and GPR120 ablates VAN activation, suggesting that both receptors are essential for detecting intestinal lipids in vivo. Importantly, since this study used global GP40/120 knockout mice, it is unclear whether these receptors are directly located in VANs or their effects are mediated through an intestinal mediator cell. EECs also express GPR40 and 120 and secrete hormones such as CCK, Peptide YY (PYY), and GLP-1 in response to increased intracellular concentrations of long-chain fatty acids (Hand et al. 2013; McLaughlin et al. 1998; Raka et al. 2019), suggesting that VANs may respond to these hormones rather than the fatty acids themselves. More work is needed to evaluate the role of medium-chain fatty acid and triglyceride signaling in VANs.

2.2.1.3. Amino acid and protein sensing

Finally, the vagus nerve also fires upon the infusion of proteins and amino acids (Jordi et al. 2013; Niijima 1982; Tomé et al. 2009). It is still unknown whether VANs have specific receptors for dietary amino acids, but there is ample evidence that dietary proteins trigger the release of gut hormones such as CCK, GLP-1, and PYY from EECs, providing a potential pathway for vagal protein sensing.

2.2.2. Direct vs EEC-mediated nutrient sensing

As mentioned earlier, there are potentially three sensory pathways that allow VANs to detect nutrients: 1) direct sensing by VANs, which would rely on the presence of nutrient receptors on VAN terminals; 2) EEC-mediated sensing, in which cells such as neuropods form synapse-like connections with vagal afferent terminals (Kaelberer et al. 2018); and 3) indirect hormone sensing, which would rely on gut hormones secreted by EECs acting on hormone receptors on vagal terminals. Establishing the relative contributions of each system is essential for understanding how VANs integrate multiple chemical cues to regulate feeding behavior and metabolism. These pathways not only operate on different timescales but also convey distinct types of information about the availability and composition of nutrients in the gut. 1) Direct sensing could provide a dynamic measure of nutrient availability as it enters circulation, serving as an energy availability monitor that coordinates metabolic responses in real time. 2) Neuropod cells would allow fast, direct neurotransmission of luminal nutrient information (Kaelberer et al. 2018) before nutrients are absorbed, enabling rapid neural feedback about meal content. 3) Hormones would provide a longer-lasting signal that reflects the quantity and macronutrient content of the meal (Hand et al. 2013; McLaughlin et al. 1998; Raka et al. 2019; Buchanan et al. 2022), shaping meal termination and postprandial metabolic responses. The interplay between these three pathways enables a multi-layered sensory system, where VANs receive information about nutrients at different stages—before absorption, during absorption, and post-absorption. Further research is needed to determine how these pathways interact and whether they function redundantly or convey complementary signals to regulate food intake and energy homeostasis. Understanding this integration will be critical for defining the role of vagal chemosensation in metabolic control and its potential dysfunction in conditions such as obesity and metabolic disorders.

2.2.3. Hormone sensing.

As indicated previously, the presence of food in the gut triggers the release of many gut hormones, including CCK, PYY, and glucagon-like peptide (GLP-1). These hormones may act on several targets in the periphery and brain to regulate feeding and meal-related physiology; 1) they can act in a paracrine manner on local enteric circuits to regulate GI motility and nutrient absorption, 2) they can act on afferent nerve fibers such as those in VANs, or 3) they can be transported to the bloodstream and sensed directly by certain brain regions (Bakar, Reimann, and Gribble 2023). VANs express most gut hormone receptors, and many of the effects of gut hormones on satiation are vagally mediated, as will be discussed in the next section. Since vagal fibers directly innervate the gastrointestinal mucosa, where hormones are produced, this area has been assumed as their primary site of action. However, the possibility that gut hormones are also detected by receptors located in the cell bodies themselves, or even throughout the vagal fibers, cannot be discarded, especially since imaging specific receptors in vagal nerve terminals has proven difficult (Patterson, Zheng, and Berthoud 2002).

Studies using isolated VANs in vitro have shown that hormones such as CCK may directly depolarize these cells (Girardi et al. 2023), but there is still limited evidence of how this occurs in vivo. Furthermore, it remains uncertain whether all vagally expressed hormone receptors mediate direct neuronal activation. For example, VANs clearly respond to CCK, leptin, serotonin, and ghrelin (Daly et al. 2011; Peters, Ritter, and Simasko 2006; Davis et al. 2020), but calcium imaging studies have failed to detect direct activation of VANs by GLP-1. While this has led to some speculation that GLP-1 does not activate VANs in a paracrine manner, this conclusion may be premature. GLP-1R are Gαs-coupled (Girada et al. 2017), and their activation primarily modulates neuronal excitability via cyclic AMP (cAMP) (Kuna et al. 2013), which may not always produce detectable calcium transients. Thus, the absence of a calcium response does not necessarily imply that VANs are unresponsive to GLP-1. Given that the deletion of GLP-1R on vagal sensory neurons influences feeding and glucose homeostasis (Krieger et al. 2016), further research is needed to fully resolve the mechanism of GLP-1 signaling on VANs.

In addition, VANs may express several hormone receptors simultaneously, allowing for the integration of multiple signals. For instance, the effects of GLP-1 on satiation depend on CCKR-expressing VANs (Diepenbroek et al. 2017; Vana et al. 2022). Furthermore, gut hormone signaling can modify VAN sensitivity to other stimuli, such as mechanical distension. For example, ghrelin, a gut hormone related to hunger, decreases vagal afferent mechanosensitivity (Page et al. 2007), whereas CCK, a postprandial hormone, enhances the vagal response to gastric distension (Schwartz, McHugh, and Moran 1991; van de Wall, Duffy, and Ritter 2005). This highlights that VANs are not merely highways of gut-brain communication but already integrate multiple sensory signals from the gut.

2.2.4. Sensing of microbial products.

The composition and metabolism of the gut microbiome influence many physiological processes, including feeding behavior, metabolism, and immune function (Cook et al. 2021). Growing evidence suggests that microbial-derived signals can modulate host physiology, in part, through direct communication with VANs. Recent studies indicate that microbial metabolites can rapidly engage the vagus nerve. For example, intestinal infusion of short-chain fatty acids (SCFAs), a major class of microbial fermentation products, robustly activates nodose ganglia neurons (Jameson et al. 2025). Moreover, mice with a conventional gut microbiota exhibit higher vagal afferent firing than germ-free or antibiotic-treated mice, suggesting that microbial signals are continuously communicated to the brain via VANs (Jameson et al. 2025). This persistent vagal activation underscores the microbiota’s role as a dynamic regulator of gut-brain signaling.

VANs detect microbial products through multiple receptor pathways. They express pattern recognition receptors, such as Toll-like receptor (TLRs), which recognize bacterial components like lipopolysaccharide (LPS) from gram-negative bacteria (Hosoi et al. 2005; de La Serre, de Lartigue, and Raybould 2015). Additionally, VANs express receptors for microbial metabolites, including free fatty acid receptors (FFARs) that detect SCFA such as acetate, propionate, and butyrate (Cook et al. 2021). Beyond SCFAs, gut bacteria also produce gamma-aminobutyric acid (GABA) (Barrett et al. 2012), and VANs express GABA Receptor A (Ashworth-Preece et al. 1997). Similarly, VANs respond to serotonin secreted from gut microbes (de Araujo et al. 2024).

In addition to microbial metabolites, immune-derived signals serve as another route of microbiome-vagus communication. VANs express receptors for cytokines and chemokines (Wang et al. 2017), which can be produced by the immune system upon several microbe-host interactions. In agreement with this, a new study showed that VANs show calcium responses to both proinflammatory and anti-inflammatory cytokines, and activating these neurons engages a body-brain circuit to modulate inflammatory responses (Jin et al. 2024). This pathway suggests that microbial signals can influence vagal activity indirectly through immune system activation, providing a potential link between gut dysbiosis, inflammation, and altered vagal signaling (Kim et al. 2020; Sen et al. 2017). By integrating microbial, metabolic, and immune-derived cues, VANs act as sentinels of microbial activity, continuously monitoring the gut environment and relaying signals that influence feeding behavior, metabolism, and homeostasis

2.3. Summary of gut sensing by VANs

In sum, VANs are equipped with a diverse array of receptors that enable them to detect different chemical and mechanical stimuli, many of which are only beginning to be characterized. In addition, VANs integrate information from multiple gastrointestinal inputs, underscoring the complexity of gut-brain signaling. Therefore, the regulation of food intake by the vagus relies on a diverse repertoire of gut-brain signals that extend beyond meal size and composition. In the following section, we explore the multifaceted roles of VANs in feeding behavior. First, we examine the physiological aspects of feeding controlled by the brainstem and hypothalamic regions, focusing on satiation and satiety.

3. Vagal gut-brain communication controlling satiation and satiety

Food intake is a complex process controlled by the decision to initiate and terminate a meal. Satiation dictates when to stop eating during a meal, while satiety determines how long one remains full until the next meal. VANs influence these key aspects of eating behavior as they can detect GI signals about the quantity and quality of food.

3.1. Role of gastric-projecting VANs.

During feeding, food enters the stomach, where it is mixed, ground, and temporarily stored before moving to the intestines for nutrient absorption. Vagal mechanosensors, specifically IGLEs located within the myenteric plexus and IMAs located in the muscular layers of the stomach wall, are ideally positioned to detect tension and tactile stimuli as the stomach accommodates food (Figure 2) (Berthoud 2008; Grundy and Scratcherd 2011). Decades of research show that food in the stomach activates vagal afferent fibers (Schwartz, McHugh, and Moran 1991), VANs (Williams et al. 2016), and their target neurons in the NTS (Ran et al. 2022) in a volume-dependent manner. Interestingly, mechanosensitive vagal afferent fiber firing increases dose-dependently irrespective of whether saline, carbohydrate, protein, or fat is infused (Mathis, Moran, and Schwartz 1998). These data, combined with vagotomy studies (Phillips and Powley 1996), have established that the stomach provides negative feedback to the brain to control food intake based on the volume rather than the type of food ingested. Furthermore, genetic-based studies have strengthened this idea by identifying a population of mechanosensory IGLEs expressing glucagon-like peptide 1 receptor (GLP1R) that exclusively innervates the stomach ( Williams et al. 2016; Bai et al. 2019; Zhao et al. 2022). As described above, this vagal subtype is selectively activated by mechanical distension, rather than nutrients (Williams et al. 2016), and its artificial activation reduces food intake (Bai et al., 2019). Similarly, activation of the mechanosensory OXTR neurons significantly decreases food intake (Scott et al. 2025), highlighting the role of mechanosensory VANs in the regulation of satiation.

Figure 2. Vagal gut-brain circuits control satiety and satiation.

A depiction of the anatomy of vagal afferent nerves and their primary sensory responses within the gut and their central neural pathways. HYP: Hypothalamus, PBN: parabrachial nucleus, NTS: nucleus of the solitary tract, NG: nodose ganglia, CCK: Cholecystokinin, GIP: Gastric inhibitory polypeptide, 5H-T: serotonin, GLP-1: Glucagon-like peptide-1, PYY: Peptide YY, EEC: enteroendocrine cells, SM: submucosal, MP: myenteric plexus, LM: longitudinal muscle, IMA: intramuscular arrays, IGLE: intraganglionic laminar endings, ME: mucosal ending. Created in https://BioRender.com

The lack of food can also affect VAN gastric mechanosensitivity. The hunger hormone ghrelin, released by PD/ D1 EEcs in the stomach (Al Massadi et al. 2010), reduces vagal afferent fiber sensitivity to gastric tension (Kentish et al. 2012) and inhibits vagally mediated effects of cholecystokinin (de Lartigue et al. 2010a), suggesting that VANs also influence satiety.

In addition to its role in the mechanical processing of food, the stomach is essential for the chemical breakdown of nutrients. As indicated previously, gastric mucosal VAN fibers respond to stroking (Page, Martin, and Blackshaw 2002) or changes in pH, signaling high concentrations of gastric acid to the brainstem (Clarke and Davison 1978; Iggo 1957; Michl et al. 2001). Although little is known about these mucosal fibers and their role in food intake, a subpopulation of VANs exclusively expressing chemosensitive mucosal endings in the gastric corpus has more recently been genetically identified as Calca+ neurons (Bai et al., 2019). Optogenetic activation of the mucosal terminals of these Calca-expressing VANs using a wireless optoelectronic device implanted in the stomach robustly suppressed food intake, highlighting their role in satiation (Kim et al. 2021).

The location of mucosal terminals within the villi and their proximity to the gastric lamina propria positions them to respond to hormones released from EECs in the stomach, suggesting a role for stomach hormones in vagal satiation signaling. For instance, gastric leptin released upon food intake (Cammisotto and Bendayan 2012) regulates meal size by enhancing the sensitivity of VANs to CCK (de Lartigue et al. 2010b) and mucosal stroking (Kentish et al. 2013a). Ablation of upper gut VANs expressing the CCK receptor affects satiety in chow-fed animals by increasing meal number and frequency. In HFD-fed animals, it alters both satiety and satiation, leading to an increase in cumulative food intake, meal size, and meal duration (McDougle et al. 2021). These data highlight that VANs integrate gut hormonal inputs from the stomach to regulate satiety and satiation.

3.2. Role of intestinal-projecting VANs.

After leaving the stomach, digested food enters the small intestine, the primary site for nutrient absorption. The inner lining contains millions of tiny, finger-like projections called villi, which greatly increase surface area and facilitate the efficient absorption of nutrients—such as amino acids, glucose, fatty acids, vitamins, and minerals—by enterocytes. Furthermore, EECs sense intestinal nutrients and release hormones that activate nearby vagal afferent fibers terminating in the submucosa (Lal et al. 2001;. Williams, Baskin, and Schwartz 2009; Berthoud et al. 1995; Berthoud and Patterson 1996). This activation provides negative feedback to the brain for satiation (Raybould 2010). As indicated above, fatty acids and amino acids cause CCK release from EECs, which activates CCKA receptor-expressing VANs to promote satiation (Moran and Kinzig 2004). Surgical and chemical lesioning of the vagus nerve abolishes the effects of CCK on food intake, highlighting the essential role of VANs on CCK-mediated satiation (MacLean 1985; Le Sauter, Goldberg, and Geary 1988; Raybould and Taché 1988).

Although most vagal afferent fibers in the intestine are mucosal endings, there is also evidence of intestinal mechanosensing in the control of food intake (Berthoud et al. 1995). As mentioned earlier, mechanosensory OXTR+ neurons project to the intestine, and their optogenetic activation dramatically reduces food intake. However, this effect was rapidly reversible, resulting in a compensatory increase in feeding once the opto stimulation ceased (Bai et al. 2019). This suggests that intestinal-projecting OXTR+ VANs are important for mediating satiation but not satiety.

Further along the gastrointestinal tract, VANs have sparse projections to the distal small intestine (Berthoud 2008); therefore, their role in feeding regulation remains relatively unexplored. Additionally, under normal conditions, most macronutrients are absorbed before reaching the distal small intestine, resulting in minimal vagal activation by nutrients or mechanical stimuli in that region (Meyer et al. 1998). However, these vagal neurons may influence feeding under conditions that alter normal digestion patterns, such as bariatric surgery or duodenal ulcers, where faster gastric emptying and increased intestinal transit rate might allow nutrients to reach distal regions like the ileum. Research has shown that ileal vagal afferent terminals are indeed sensitive to nutrients (Randich et al. 2000), and that a celiac branch vagotomy (which ablates vagal branches to the small and large intestine while preserving those to the stomach and upper gut) abolishes the suppressive effects of intestinal nutrients on food intake (Walls et al. 1995). Furthermore, selective ablation of vagal neurons disrupts the effects of GLP-1 and PYY, two satiety hormones primarily secreted by L-cells in the ileum (Brierley and de Lartigue 2022; Barakat et al. 2024). Further research is needed to clarify the functions of vagal neurons innervating the distal small intestine and their potential role in regulating feeding.

3.3. Role of VANs projecting to the large intestine and the microbiome.

Large intestine vagal innervation has received much attention due to the increasing evidence that the microbiome, which mainly resides in this intestinal region, plays an important role in controlling satiety and metabolism. Changes in gut microbiome functionality, composition, and diversity have been associated with the development and restoration of metabolic diseases, such as obesity (Kim et al. 2012; David et al. 2014; Rutsch, Kantsjö, and Ronchi 2020; Klingbeil et al. 2024). For example, diet-induced obese (DIO) mice reduce their body weight and food intake after taking a probiotic supplement (Bagarolli et al. 2017). However, the role of VANs in this process is still incompletely understood.

VANs innervate the colon and cecum (Bai et al. 2019; Berthoud, Carlson, and Powley 1991), and approximately 8% of nodose ganglion neurons innervate the distal colon (Osman et al. 2023). Injecting Lactobacillus johnsonii into the intestine enhances vagal activity (Yu and Hsiao 2021), and oral administration of Campylobacter jejuni in mice activates the NTS (Tanida et al. 2005). Furthermore, in the large intestine, the gut microbiome produces metabolites such as SCFAs, indoles, and deconjugated bile acids, which further promote the secretion of the satiety hormone GLP-1 (Tolhurst et al. 2012; Trabelsi et al. 2015). These metabolites can also increase vagal activation to reduce food intake (Richards, Thornberry, and Pinto 2021; Mamedova, Árting, and Rekling 2022; Jameson et al. 2025).

Taken together, the passage of food throughout the GI tract produces many signals that may be detected by VANs to modify satiety and satiation, including nutrient sensing, mechanosensing, sensing of bacterial products, and hormone signaling. In this manner, the vagus can relay meal-related signals to the brain in order to terminate meals and regulate physiology in the postprandial state.

3.4. The NTS, the first relay of VANs for the regulation of satiety and satiation.

The NTS in the brainstem is a crucial region for food intake control. As the first brain region that receives direct inputs from gut-derived signals through VANs (Ritter 2004; Yu, Xu, and Chang 2020), the NTS plays a well-established role in satiation. Studies on decerebrated rats, in which communication between the hindbrain and forebrain is severed, have shown that these animals can still terminate a meal (Grill and Norgren 1978; Hayes, Skibicka, and Grill 2008), underscoring the sufficiency of the hindbrain circuitry for meal termination. Moreover, artificial activation of vagal postsynaptic neurons in the NTS reduces food intake (de Araujo et al. 2023). NTS neurons respond to multiple GI stimuli that are known to activate VANs, such as gastric distention, hormones, and nutrients (Maniscalco and Rinaman 2018; Chen et al. 2020; Rinaman 1998). Importantly, vagotomy ablates those effects (Chen et al. 2020; Winzenried et al. 2024), further emphasizing the critical role of vagal inputs on NTS function.

Beyond its established role in satiation, the NTS may have additional, less-understood roles in eating behavior. One study identified a subpopulation of catecholaminergic NTS neurons that co-express Neuropeptide Y (NPY) and receive input from the vagus. These neurons show increased Fos expression, a marker of neuronal activity, following a 15-hour fast and appear to promote food intake (Chen et al. 2020). While these findings suggest that NTS-NPY neurons may be activated during hunger, their precise function in driving food intake remains to be elucidated.

In addition to directly promoting satiation, the NTS functions as a central hub for vagal signals, relaying gastrointestinal information to higher-order forebrain neurons that also regulate feeding, energy homeostasis (Grill and Hayes 2012), and reward pathways (Han et al. 2018; Tellez et al. 2013). It is worth noting that the NTS integrates meal-related information from other sensory systems, such as spinal afferents and the area postrema (Fry and Ferguson 2007; Holt 2022; Münzberg, Berthoud, and Neuhuber 2023), highlighting its role as a multimodal center for interoceptive signals. To fully understand the broader implications of vagal signaling, we next examine the downstream central circuits engaged by the NTS, including the parabrachial nucleus (PBN) and hypothalamus, which refine and extend the integration of meal-related signals.

3. 5. NTS relays to the PBN and hypothalamus involved in satiety and satiation

Downstream of the NTS, the PBN and hypothalamic nuclei are important for mediating vagal-induced satiation. These regions integrate meal-related information into the broader brain network that controls energy balance and different aspects of feeding behavior (for review, see (Alcantara et al. 2022; Rossi and Stuber 2018).

The PBN is a major hub for feeding circuits, including those mediating satiation, aversive feeding suppression, food reward, and hunger (Zhao et al., 2023). Several subpopulations of PBN neurons regulating food intake have been characterized, including those expressing the calcitonin gene-related peptide (CGRP), prodynorphin (Pdyn), and Calca, which are involved in satiation and aversive feeding suppression (Condon et al. 2024; Kim et al. 2020; Palmiter 2018; Roman, Derkach, and Palmiter 2016; Carter et al. 2013). While anatomical studies provide clear evidence for a VAN-NTS-PBN circuit (Han et al. 2018), fewer studies have evaluated the direct contribution of gut manipulations in PBN-driven feeding behavior. One study found that dynorphin neurons in the PBN receive information from VANs via the NTS in response to gastrointestinal distension (Kim et al. 2020). In hungry animals, activating this vagal-NTS-PBN pathway inhibits food intake, while inhibiting this pathway in satiated animals does not have the opposite effect (i.e., increasing food intake), indicating the specificity of this pathway for satiation (Kim et al. 2020). Furthermore, direct stimulation of the mechanosensory OXTR+ VANs increased activation of the PBN, in particular, that of Calca+ neurons (Bai et al. 2019), which are known to inhibit food intake (Roman, Derkach, and Palmiter 2016). Importantly, the role of PBN neurons in gut-brain communication is not limited to satiation; the PBN also acts as a relay to communicate nutrient information from VANs to dopaminergic nigrostriatal pathways (Han et al. 2018); discussed in more detail in section 5.1 below).

The paraventricular nucleus of the hypothalamus (PVH) also plays a key role in integrating vagally mediated gastrointestinal signals relayed through the NTS (Fawley et al. 2021; Han et al. 2018). Multiple studies have supported the involvement of NTS-PVH pathways in the control of food intake (Kim et al. 2023; Liu et al. 2017; Murphy et al. 2023; Sayar-Atasoy et al. 2023). For example, CCK+ NTS neurons project to PVH neurons expressing the melanocortin 4 receptor (MC4R), and activation of this pathway strongly and transiently inhibits food intake (D’Agostino et al. 2016). In addition, AgRP neurons in the hypothalamus, known as “hunger neurons” (Aponte, Atasoy, and Sternson 2011; Atasoy et al. 2012; Gropp et al. 2005; Krashes et al. 2011) are strongly inhibited by gastric fat infusion through a vagal mechanism (Goldstein et al. 2021). Importantly, these neurons are also strongly inhibited by chemogenetic activation of GLP1R+ or OXTR+ neurons in the nodose ganglia, which, as previously discussed, detect gastric distension (Bai et al. 2019). These data support the notion that hypothalamic AgRP neurons receive meal-related signals from VANs to regulate hunger.

Other hypothalamic nuclei involved in the control of food intake, such as the dorsomedial and lateral hypothalamic nuclei, also receive meal-related information from VANs (Zhang et al. 2003; Zhu et al. 2007). While the role of these vagal gut-brain pathways remains less explored, these regions are known to be critical for feeding behavior and energy balance. Thus, the PBN and hypothalamus integrate vagal inputs to influence feeding behaviors, energy homeostasis, and reward signaling. These central circuits ensure that meal-related signals from the gut drive appropriate behavioral and physiological outcomes. However, the gut-brain axis is not purely sensory; it also encompasses efferent pathways that regulate postprandial gastrointestinal and metabolic functions.

4. Vagal gut-brain communication regulating physiological responses to a meal

The vagus nerve’s efferent pathways form a critical part of the bidirectional gut-brain axis, coordinating physiological responses to a meal (Browning and Travagli 2019). While upstream circuits like the PBN and hypothalamus process sensory information to influence behavior, vagal efferent reflexes directly regulate gut motility, secretion, and glucose metabolism. These reflexes optimize digestion and nutrient absorption, highlighting the bidirectional interplay of the vagal system.

4.1. Vago vagal reflexes in the control of gastrointestinal motility and secretion.

Feeding elicits a plethora of physiological responses to facilitate the accommodation, digestion, distribution, and storage of nutrients. Many of these processes are orchestrated by vago-vagal reflexes, which form a bidirectional communication loop between the brainstem and gastrointestinal organs. These reflexes rely on the integration of signals at the NTS and its projection to the DMV and the nucleus ambiguus (Figure 3A). NTS neurons project to the DMV either directly or through interneurons releasing neurotransmitters such as GABA, glutamate, or norepinephrine (Travagli and Anselmi 2016). In this manner, sensory inputs detected by VANs can modulate vagal efferent signaling, which in turn influences GI function by projecting to specific areas of the GI tract through cholinergic fibers (Tao et al. 2021). While local enteric circuits can independently control gut motility, vago-vagal reflexes allow the brain to exert top-down control over gut motility and secretory processes (Powley 2021). A clear example of this mechanism is the receptive relaxation reflex, triggered by mechanical stimulation as food enters the esophagus. This reflex reduces gastric tone, enabling the stomach to accommodate incoming food with minimal resistance (Arakawa et al. 1997). Similarly, chemical signals, such as the presence of glucose in the intestine, initiate vago-vagal reflexes. Intestinal glucose stimulates enteroendocrine cells to release serotonin and GLP-1 activating VANs and modulating processes like gastric emptying, pancreatic exocrine secretion, and intestinal fluid secretion (Raybould 2010). Importantly, vago-vagal reflexes are not static hardwired circuits, but may also show plasticity depending on the physiological state of the organism. For instance, factors such as fasting, satiety, and stress can influence the sensitivity and output of these reflexes, ensuring they adapt to the body’s current needs (Browning and Travagli 2010).

Figure 3. Vagal gut-brain circuits regulate the physiological responses to a meal.

A) Vagal afferent neurons innervating the GI tract mediate gastrointestinal secretion and motility through vago-vagal circuits on the brainstem encompassing the NTS, the DMV and the Amb. B) Vagal afferent neurons innervate the pancreas, intestine, and hepatic portal area, providing an anatomical pathway for glucose detection. Furthermore, GLP1R-expressing neurons in the nodose ganglia might regulate glucose metabolism in response to gastric distension. Information from VANs is communicated to the NTS, where it could regulate glucose metabolism via at least two pathways: a vago-vagal reflex that promotes insulin release, and a hypothalamic pathway that regulates hepatic glucose production through vagal and/or spinal (sympathetic) efferent systems. C) VANs may detect different signals in the GI tract such as intestinal lipids and gastric distension to regulate thermogenesis. Since the NTS projects to hypothalamic areas involved in BAT thermogenesis, this pathway is probably mediated through a hypothalamus-SNS-BAT axis. D) Vagal afferent neurons detect hypotonic stimuli in the gut and hypertonic in the hepatic portal area and communicate this information to the NTS. In turn, the NTS relays this information to the SFO to mediate drinking behavior. NTS, nucleus of the solitary tract; DMV, dorsal motor nucleus of the vagus; Amb, nucleus ambiguus; HYP, Hypothalamus; SNS, sympathetic nervous system; BAT, brown adipose tissue; SFO, subfornical organ; NPY, neuropeptide; GLP-1R, glucagon-like peptide 1 receptor. Created in https://BioRender.com

4.2. Glucose metabolism.

Proper regulation of glucose metabolism after a meal is essential to maintaining metabolic homeostasis (Wasserman 2009). Most foods contain carbohydrates that are broken down to glucose, which is absorbed into the hepatic portal system. In the liver, the presence of glucose decreases hepatic glucose production, while in the bloodstream, it triggers the release of insulin by the pancreas, promoting glucose uptake in the skeletal muscle and other tissues. Since VANs innervate the intestine, hepatic portal area, and pancreas, they are well positioned to directly sense glucose and insulin and contribute to glycemic regulation after a meal (Figure 3B).

As discussed previously, VANs respond to glucose infusion into the intestine and HPV (McDougle et al. 2024; Niijima 1982; Tan et al. 2020). However, whether vagal signaling plays a direct role in regulating glucose metabolism is less clear. One study found that activating GLP-1 receptor-expressing VANs improved glucose tolerance by promoting glucose uptake by the skeletal muscle and decreasing hepatic glucose production (Borgmann et al., 2021). These effects may involve downstream activation of specific neuronal populations such as CCK and NPY neurons in the NTS and Calca neurons in the PB since these populations increased their activity after activation of GLP-1R vagal neurons (Borgmann et al. 2021). However, it remains unclear whether these neurons respond to glucose itself or other meal-related stimuli. Since GLP-1R VANs have also been identified as mechanosensors, their artificial activation could mimic mechanosensing processes. Supporting this notion, stomach distension has been shown to improve glucose tolerance, and this effect was abolished by capsaicin-mediated denervation, suggesting the involvement of VANs (Ohbayashi et al. 2021). In addition, other meal-derived signals like lipids infused into the intestine diminish glucose production by the liver through vagal afferent signaling (Wang et al. 2008). Gut hormones like GLP-1 are also well-known to regulate glucose metabolism, with some of their effects mediated through VANs (Krieger, Langhans, and Lee 2015). Furthermore, a recent study found that overexpression of the fibroblast growth factor 3, specifically in the nodose, increased glucose-stimulated insulin secretion, although the signaling pathway of this factor is still unclear (Tahiri et al. 2024). Pancreas-projecting VANs could play a role in modulating insulin secretion since another study found that beta cells in the pancreas directly communicate with VANs through serotonin release (Makhmutova et al. 2021). Since serotonin is co-released with insulin, it is possible that pancreas-projecting VANs survey insulin release through this pathway.

These findings suggest that VANs integrate several meal-related signals to modulate glucose metabolism. Despite these insights, significant gaps remain in our understanding of the downstream circuits through which VANs influence whole-body glucose metabolism. One possibility is that VAN activation triggers vago-vagal reflexes, promoting insulin release through cholinergic projections of DMV neurons to the pancreas (Borgmann and Fenselau 2024). Alternatively, meal-related information from VANs could also be transmitted to the hypothalamus or other brain regions known to modulate glucose production by the liver through the autonomic nervous system, either vagal (parasympathetic) or spinal (sympathetic) efferent pathways (Lin, Scott-Solomon, and Kuruvilla 2021; Zsombok, Desmoulins, and Derbenev 2024). These pathways highlight the vagus nerve’s potential as a critical integrator of diverse signals to maintain glucose homeostasis, though further research is needed to elucidate the specific mechanisms involved.

4.3. Temperature regulation.

Feeding is accompanied by postprandial thermogenesis, a process that increases energy expenditure after a meal. A few studies have linked VANs to this phenomenon. For example, lipid infusion into the duodenum increases brown adipose tissue thermogenesis, and this effect is blocked by CCK-R antagonists infused into the NTS or the application of a local anesthetic to duodenal vagal fibers (Blouet and Schwartz 2012). These findings suggest that information about the lipid meal is conveyed via a VAN-NTS pathway to regulate thermogenesis. Similarly, secretin, a gut peptide secreted by the stomach after a meal, also increases BAT thermogenesis. While the authors showed that secretin may directly target the BAT to produce these effects, the potential involvement of the vagus nerve was not explored (Li et al. 2018).

In contrast, other studies suggest that VAN activation may decrease thermogenesis under certain conditions. Madden and colleagues showed that cervical vagal stimulation decreases brown adipose tissue thermogenesis evoked by cold exposure (Madden, Santos da Conceicao, and Morrison 2017). Similarly, our group found that chemogenetically activating OXTR-expressing VANs, which project to both the stomach and the heart, significantly decreases body temperature (Scott et al. 2025). Conversely, ablating GLP-1R expression in VANs in rats increased BAT thermogenesis, especially when animals were fed a high-fat diet, suggesting that GLP-1 signaling in VANs decreases body temperature (Krieger et al. 2018). These contrasting findings suggest that the role of VANs in thermoregulation may depend on the physiological state of the animal and the source of the vagal signals. For example, during the postprandial state, gut-sensing VANs might elicit thermogenic circuits, whereas, under stress conditions such as cold exposure, VANs might engage circuits that reduce body temperature.

The regulation of BAT thermogenesis involves hypothalamic nuclei, including the preoptic area, ventromedial hypothalamus, and paraventricular hypothalamus. These nuclei contain preautonomic neurons that regulate BAT function via the sympathetic system (Caron et al. 2018; Mohammed et al. 2018). Therefore, VANs might provide input to these hypothalamic circuits for thermoregulation (Figure 3C). Indeed, injection of a multisynaptic tracer into the nodose ganglia, labeled presympathetic neurons in the rostral raphe pallidus (rRPa), and rostral ventrolateral medulla (RVLM), as well as hypothalamic neurons known to be involved in controlling thermogenesis (Krieger et al. 2018). Furthermore, injecting a pseudorabies virus into the BAT labeled neurons in the nodose ganglia, providing strong evidence for a multisynaptic VAN-BAT connection that may mediate vagal regulation of BAT thermogenesis (Krieger et al. 2018).

4.4. Osmotic balance.

Blood osmolarity is regulated through different processes including sodium and water intake. These processes are primarily controlled by the hypothalamic subfornical organ (SFO) and lamina terminalis (LT), both of which can directly sense changes in blood osmolarity (Bourque 2008). However, animals change their drinking behavior before blood osmolarity reaches homeostatic levels, suggesting the involvement of other sensory systems (Zimmerman et al. 2016; Augustine et al. 2019). One study showed that the response of the SFO to ingested water or a hypertonic solution is diminished in animals with ablation of VANs (Zimmerman et al. 2019), suggesting that VANs play a critical role in communicating osmotic information to the brain. Supporting this, several studies have shown that VANs are responsive to hyperosmolar solutions infused into the gut (Williams et al. 2016; Bourque 2008; Ichiki et al. 2022). Furthermore, the populations that respond to hypertonic solutions differ from those that respond to low osmolarity or sugars (Ichiki et al. 2022; Tan et al. 2020), further highlighting the specificity of VAN subpopulations for different osmotic stimuli. Expanding on this specificity, VANs that project to the hepatic portal area, where water is absorbed from the GI tract, respond to water but not hypertonic stimuli (Ichiki et al. 2022). Infusion of vasoactive intestinal peptide into this region elicited a similar response, suggesting that VIP may serve as a key signal detected by these VANs (Ichiki et al. 2022). Together, these studies indicate that VANs detect both hypo- and hypertonic stimuli in the gut, contributing to the regulation of drinking behavior and, consequently, osmotic balance. By relaying detailed sensory information to the brain, VANs ensure precise adjustments to maintain homeostasis, complementing the osmosensory functions of the SFO and LT (Figure 3D).

5. Vagal gut-brain communication regulating cognitive aspects of feeding behavior

5.1. Role of VANs in reward.

In the last ten years, the role of VANs in food intake has expanded beyond just physiology with the recognition that VANs also regulate cognitive aspects of feeding behavior (Figure 4). This has been especially important in view of the steady rise in worldwide obesity rates over the last 50 years, emphasizing the importance of understanding how calorie-rich foods elicit reward signals in the brain. The reward pathway contributes to “appetition” (the processes that promote food preference and food intake) (Sclafani 2013), and is thought to contain two distinct functional components that involve distinct brain regions and circuitry: ‘wanting’ (the motivation to seek out pleasurable stimuli) and ‘liking’ (the hedonic pleasure derived from consuming said stimulus) (Berridge 1996; Recio-Román, Recio-Menéndez, and Román-González 2020).

Figure 4. Vagal gut-brain circuits regulate cognitive aspects of feeding behavior.

Distinct populations of vagal afferent neurons innervating the gut and liver respond to post-ingestive fat (blue) and sugar (pink). Neurons of the NTS receive these signals from vagal inputs and relay them to reward (A) and memory (B) pathways, with fats and sugars recruiting separate but parallel populations of neurons at each node within each circuit.

A) In the reward pathway, glutamatergic neurons of the PBN are engaged downstream of the NTS. These neurons project to the SNc, where they activate dopaminergic neurons. Release of dopamine from SNc neurons onto the dorsal striatum underlies reward behaviors in animals and preference for fats and sugars. B) Within the polysynaptic gut-memory circuit, the medial septum is a likely intermediary node between the NTS and HPC. NTS = nucleus of the solitary tract ; PBN = parabrachial nucleus ; SNc = substantia nigra pars compacta ; MS = medial septum ; HPC = hippocampus. Created in https://BioRender.com

Rewarding macronutrients such as sugar or fat engage reward circuitry first at the gustatory level, where palatability cues recruit mesolimbic circuits to release dopamine in the ventral striatum (Tellez et al. 2016a) (Fig 4 A). Subsequently, post-ingestive nutritive cues from the intestine recruit a separate reward circuit resulting in nigrostriatal dopamine release in the dorsal striatum (Tellez et al. 2016b; Thanarajah et al. 2019). Importantly, the reward circuits engaged by a food’s nutritional value are independent of those engaged by its palatability (Tellez et al. 2016b; Ferreira et al. 2012). Animal studies indicate that nutritional value is likely the primary driver of reinforcement because taste receptors are not required for food preference and palatability can be learned as a conditioned response to post-ingestive nutritional cues (Sclafani 2013; de Araujo et al. 2008; Holman 1969; de Araujo 2016).

Accumulating evidence has demonstrated that VANs are important players in post-ingestive signaling in the presence of sugar and fat in the gut. For instance, animals develop robust reinforcement behavior for nutritive sugar but not for artificial sweeteners, which simulate sweet taste but are detected by distinct enteroendocrine receptors in the gut lumen (Liu and Bohórquez 2022; Buchanan et al. 2022). Tan et al identified a subpopulation of VANs that detect glucose in the small intestine and activate the NTS. Silencing these neurons inhibits an animal’s capacity to develop a preference for sugar over artificial sweetener (Tan et al. 2020), suggesting a key role of this VAN-NTS pathway for sugar preference. Although artificial sweeteners may stimulate some VANs (Buchanan et al. 2022), they do not significantly activate brain circuits that increase preference (Tan et al. 2020) and thus do not evoke reinforcement behavior.

Like sugar, VANs are also necessary to detect the presence of fat in the intestine and relay this signal to the brain for preference development (Li et al. 2022b). Our recent study revealed that intragastric fat and sugar are sensed by two distinct VAN populations, which are necessary and sufficient to produce reinforcement for each nutrient by engaging separate but parallel downstream circuits (McDougle et al. 2024) (Figure 4A). These distinct sugar and fat-responsive populations exhibit different organ innervation patterns and independently drive upstream macronutrient-specific flavor preference. Together, these studies demonstrate that post-ingestive signals from sugar and fat are conveyed to the brain via VANs, and this gut-brain pathway is crucial for fat and sugar reinforcement.

Interestingly, nutrients are not the only post-ingestive signals that induce preference. In thirsty mice, intragastric water infusion increases dopamine release in the VTA, likely through a vagal mechanism (Grove et al. 2022; Kim et al. 2020). Importantly, silencing these VTA neurons in thirsty animals decreased the preference of the mice to consume hypotonic solutions that satiated their thirst, indicating that a VAN-VTA pathway might be mediating post-ingestive learning of fluid osmolarity, similar to mediating the preference for nutrient-paired flavors.

Downstream of the VAN-NTS connection, gut information reaches the nigrostriatal reward circuit. Using targeted viral approaches, we found that optogenetic stimulation of upper gut-innervating VANs increased dopamine release in the dorsal striatum and promoted sustained self-stimulation behavior in the animals, indicating the activation of motivational reward circuits (Han et al. 2018). Full mapping of the circuit with anterograde transsynaptic viral tracing revealed a poly-synaptic pathway where glutamatergic neurons of the dorsolateral PBN serve as an intermediary link between vagal neurons of the right nodose ganglion terminating at the NTS and dopaminergic neurons of the substantia nigra (Figure 4A). Separately we also showed that after initial sensing of fats and sugars by VANs, separate neuronal populations are recruited at each node along the gut-reward circuit, including the NTS, PBN, and substantia nigra. Consequently, combining fat and sugar in a single infusion additively recruited both vagal circuits, increasing striatal dopamine and enhancing reinforcement behavior beyond the effects of either nutrient alone, even when matched for calories (McDougle et al. 2024).

Overall, these studies indicate that VANs recruit nigrostriatal dopaminergic reward circuits to promote the consumption of meals with high nutritional content, especially sugar-fat combinations. These experiments are further complemented by studies in humans showing that participants within a food auction paradigm will pay more (ie – exhibit more motivation) for foods containing both fat and carbohydrates and that this behavior is coupled with dorsal striatal brain activity (DiFeliceantonio et al. 2018). Thus, the existence of these gut-brain reward pathways activated by post-ingestive fat and sugar could explain the intrinsic and involuntary drive in both animals and humans to consume obesogenic foods, opposing conscious dietary efforts for weight loss. Nonetheless, future studies are needed in order to determine whether and how other types of nutrients, metabolites, and microbiota products engage reward circuits. Amino acids, for example, have been shown to modulate reward signaling in rodents and in humans (Leidy et al. 2011; M. Li et al. 2022a; Pérez, Ackroff, and Sclafani 1996; Zeeni et al. 2010; Jiyoung S. Kim et al. 2023). However, the role of VANs in this modulation remains unclear. Furthermore, the extent to which gut-brain reward drives meal satisfaction (and termination) versus meal appetition (and prolongation) is unknown. In all likelihood, post-ingestive signals from the gut engage multiple reward circuits in the brain, which holistically contribute to feeding behavior.

5.2. Role of VANs in food memory.

An important cognitive aspect of feeding is the capacity to remember when, where, and what to eat. As indicated in the previous section, post-ingestive signals detected by VANs are important drivers of what to eat. However, less is known about the gut-brain circuits mediating other types of food-related memory. The hippocampus (HPC), a brain region critical for learning and memory, has been shown to respond to vagally mediated gastrointestinal signals, including gastric distension (Min, Tuor, and Chelikani 2011) and nutrient infusion (Min et al. 2011). Subdiaphragmatic vagotomy (SDV) reduces HPC activation in response to nutrient infusions (Yang et al. 2025), highlighting the essential role of VANs in transmitting post-ingestive signals to the HPC. Moreover, disruption of vagal gut-brain signaling using techniques such as SDV or cholecystokinin-conjugated saporin-mediated ablation results in impaired HPC-dependent contextual episodic memory (Onimus et al. 2024; Suarez et al. 2018). Similarly, selective knockdown of the ghrelin receptor (growth hormone secretagogue receptor 1A) in the nodose ganglion leads to significant deficits in HPC-dependent contextual episodic memory (Davis et al. 2020); these behavioral impairments are accompanied by reduced expression of brain-derived neurotrophic factor and the neurogenic marker doublecortin (Suarez et al. 2018; O’Leary et al. 2018), as well as morphological and plasticity-related changes in the HPC (Onimus et al. 2024). These data highlight the importance of vagal signaling for HPC-dependent memory, but their relevance for eating behavior is less understood.

There is evidence that impairing HPC-dependent memory alters feeding behavior in both rodents and humans (Kanoski and Grill 2017). In rodent models, permanent pharmacological lesions of the entire HPC result in increased meal frequency and decreased meal size (Davidson and Jarrard 1993; Clifton, Vickers, and Somerville 1998). In contrast, selective postprandial inactivation of HPC neurons through intrahippocampal muscimol administration yields shortened inter-meal intervals and increased meal size (Henderson, Smith, and Parent 2013). More importantly, studies utilizing advanced transgenic mouse models have revealed crucial insights into HPC regulation of feeding behavior. Both chronic and acute inhibition of hippocampal neurons expressing dopamine D2 receptors (hD2R) significantly increases chow intake (Azevedo et al. 2019). Moreover, distinct populations of hippocampal neurons show selective responsiveness to specific nutrients, and targeted ablation of these nutrient-responsive neurons results in reduced consumption of a high-fat diet and prevented weight gain (Yang et al. 2025). In clinical studies, impaired meal-related memory encoding due to distraction during eating leads to increased subsequent food intake (Higgs and Woodward 2009). Moreover, patients with compromised hippocampal-dependent memory function exhibit profound disruptions in feeding behavior. Specifically, these individuals demonstrate an inability to recall recent meals and will consume additional food when presented with food-related cues, despite having eaten shortly before (Hebben et al. 1985; Rozin et al. 1998; Higgs et al. 2008).

In contrast to the cognitive deficits observed following vagal denervation, enhancement of vagal signaling through electrical stimulation improves learning and memory processes. In rodent models, optogenetic stimulation of right vagal terminals in the duodenum elicits significant HPC activation (de Araujo et al. 2023). In addition, stimulation of nutrient-responsive vagal neurons has been shown to activate HPC neurons (Yang et al. 2025). These effects are likely mediated through a polysynaptic circuit, where vagal sensory fibers terminating in the NTS project via an intermediary node, possibly the medial septum, before ultimately relaying signals to the HPC (Suarez et al. 2018). The functional significance of this vagal-HPC circuit is further demonstrated by chronic vagus nerve stimulation (VNS), which enhances multiple cognitive domains including decision-making (Cao et al. 2016), associative memory Clark et al. 1998; Peña, Engineer, and McIntyre 2013), and recognition memory (Sanders et al. 2019). Notably, even a single session of acute VNS is sufficient to enhance cognitive flexibility, and short-term memory (Driskill et al. 2022). Similar to rodents, VNS also results in robust cognitive enhancement in humans (Vonck et al. 2014). Clinical studies demonstrate that VNS-treated epilepsy patients exhibit enhanced memory processing across distinct phases, particularly consolidation, and retrieval (Clark et al. 1999; Ghacibeh et al. 2006b; Helmstaedter, Hoppe, and Elger 2001) and improvements in higher-order cognitive functions such as decision-making and cognitive flexibility (Ghacibeh et al. 2006a; Martin et al. 2004). Moreover, in patients with Alzheimer’s disease, chronic VNS improves cognitive performance (Sjögren et al. 2002) and ameliorates disease pathology (Merrill et al. 2006). The memory-enhancing effects of VNS are likely due to modulation of neurotransmitter systems (i.e., norepinephrine and serotonin) (Dorr and Debonnel 2006; Roosevelt et al. 2006; Furmaga, Shah, and Frazer 2011), strengthened long-term potentiation (Zuo, Smith, and Jensen 2007), and enhanced HPC plasticity, as evidenced by elevated BDNF expression (Biggio et al. 2009; Follesa et al. 2007; Furmaga, Carreno, and Frazer 2012).

In addition to the benefits in cognition, enhancing HPC-dependent memory significantly alters feeding behavior. Distinct neuronal populations within the HPC exert opposing effects on food intake. Selective stimulation of anorexigenic HPC neurons expressing D2R suppresses standard chow consumption (Azevedo et al. 2019)., consistent with human studies showing that recalling previously consumed meals effectively reduces subsequent food intake (Higgs 2002; Higgs et al. 2008). In contrast, activation of a separate subpopulation of nutrient-responsive HPC neurons produces a distinct orexigenic behavioral phenotype, characterized by increased consumption of both a liquid diet and a high-fat diet (Yang et al. 2025). Overall, these data strongly suggest a role for VAN-HPC communication in cognition and food-related memory.

6. Conclusions and future directions

The use of rodent models has significantly expanded our understanding of VANs, revealing their molecular, anatomical, and functional diversity. Rather than serving as simple conduits of gut-derived sensory information, VANs integrate mechanical, chemical, and hormonal signals to regulate satiety, glucose metabolism, thermogenesis, and even cognitive processes such as memory and reward. While it remains unclear whether specific genetically defined neuronal populations have identical functions in humans as they do in rodents, anatomical and functional studies suggest that vagal pathways are largely conserved across species, offering potential translational value.

Currently, vagal nerve stimulation (VNS) is used clinically for conditions such as epilepsy, depression, and obesity. However, existing VNS techniques stimulate vagal fibers non-selectively, leading to inconsistent efficacy and off-target effects. The fascicular organization of the vagus nerve, the spatial separation of sensory and motor fibers, as well as organ-specific branches, has remained largely unexplored. However, recent advancements in micro-computed tomography imaging and neuroanatomical mapping have provided a fascicle-level understanding of vagal fiber organization, demonstrating that different fiber types and organ-specific pathways are spatially segregated, at least in swine models (Jayaprakash et al. 2023). These findings offer a framework for developing more precise VNS approaches that selectively target functionally distinct vagal fibers while avoiding unwanted side effects.

One promising approach is the development of multi-contact cuff electrodes, which can asymmetrically activate distinct vagal fascicles to achieve selective neuromodulation. Such next-generation bioelectronic therapies could revolutionize VNS by targeting specific vago-vagal reflexes rather than engaging the vagus nerve in a non-specific manner. While these findings represent a significant advance, further research is needed to determine how well these fascicular organizations translate to humans and whether functionally distinct vagal circuits can be selectively engaged for therapeutic benefit.

Beyond electrical stimulation, emerging pharmacological and genetic approaches may allow for even greater specificity. Progress in vagal transcriptomics and circuit tracing has identified molecular markers for functionally distinct VAN populations, raising the possibility of targeted pharmacological modulation of specific vagal subtypes. However, pinpointing which transcriptionally defined vagal subpopulations mediate particular functions remains an ongoing challenge. Advances in intersectional genetic tools, optogenetics, and chemogenetics have begun to refine our ability to manipulate vagal circuits in preclinical models, but translating these findings to human applications will require innovative engineering, improved targeting strategies, and validation in clinical settings.

Finally, understanding vagal sensory mechanisms may unlock entirely new approaches to treating metabolic and neuropsychiatric diseases. The vagus nerve plays a role not only in feeding and metabolism but also in interoception, stress responses, and inflammation, factors implicated in conditions such as obesity, irritable bowel syndrome, neurodegenerative disorders, and mood disorders. Moving forward, integrating basic neuroscience with translational research will be essential for bridging the gap between our growing understanding of vagal function and the development of targeted therapies that harness its full therapeutic potential.

As our understanding of vagal circuits continues to evolve, the hope is that the integration of cutting-edge neuroscience, bioengineering, and translational medicine will pave the way for precise, circuit-specific interventions; transforming the vagus nerve from a broad therapeutic target into a finely tuned instrument for targeted treatment of metabolic, autonomic, and cognitive disorders.