Vagal neuroimmune circuits in inflammation and immunity

Physiological homeostasis depends on continuous communication between the nervous and immune systems. While the immune system orchestrates tissue homeostasis and cellular and molecular responses to infection and tissue injury, the nervous system provides real-time surveillance and reflexive control, ensuring that these immune responses remain proportional, localized, and reversible (Chavan et al., 2017; Udit et al., 2022). This neuroimmune cross talk occurs through multiple anatomical pathways in the peripheral nervous system, each contributing distinct mechanisms for detecting, encoding, and modulating inflammatory responses (Pavlov and Tracey, 2012; Udit et al., 2022). Among these peripheral neuroimmune pathways, the vagus nerve (cranial nerve X) serves as a principal neural conduit linking visceral organs to the brain, integrating visceral physiology with immune homeostasis. Over the past two decades, vagal sensory and autonomic circuits have emerged as a core component of the neuroimmune network that governs inflammatory tone across multiple organ systems. Vagal afferent sensory fibers convey visceral information to the brain while efferent motor fibers modulate peripheral immune activity, together forming a body–brain axis that regulates inflammation and immunity.

Pioneering studies in the 1990s by Linda Watkins and colleagues established the vagus nerve as a route for cytokine-to-brain signaling: subdiaphragmatic vagotomy abolished fever, sickness behavior, and hypothalamic–pituitary–adrenal (HPA) axis activation in response to peripheral IL-1β and LPS, implicating afferent vagal fibers as the responsible substrate (Maier and Watkins, 1998). Subsequent work defining the inflammatory reflex, a vagus nerve-mediated circuit that regulates cytokine production and limits tissue injury during endotoxic shock, extended this framework to efferent immune control (Borovikova et al., 2000; Pavlov and Tracey, 2017; Tracey, 2002). Together, these discoveries established the inflammatory reflex and its efferent arm—the cholinergic anti-inflammatory pathway (CAIP), by which vagal output limits cytokine production via α7 nicotinic acetylcholine receptors (α7nAChR) on macrophages (Chavan et al., 2017). These discoveries framed the neural regulation of inflammation as a reflexive process mediated by specific neural circuits, rather than a purely humoral process. The initial input to these neural reflex circuits comes from peripheral sensory neurons, which themselves act as sentinels of infection and injury. Both the nervous system and immune system have evolved a common set of receptors to detect pathogens and tissue damage, including pattern-recognition receptors (PRRs) such as TLRs, and receptors for damage-associated molecular patterns (Hosoi et al., 2005; Li et al., 2005). Additionally, peripheral neuroimmune communication occurs through the secretion of neuroimmune modulators, such as cytokines, prostaglandins, neurotransmitters, and neuropeptides, which modulate function in each respective system (Klein Wolterink et al., 2022; Pinho-Ribeiro et al., 2017). Vagal sensory neurons are highly diverse and express a broad complement of cytokine and immunomodulator receptors (Kupari et al., 2019; Wang et al., 2017). Similarly, immune cells, including macrophages, dendritic cells, T cells, mast cells, and others, express receptors for canonical neuromodulators such as acetylcholine, norepinephrine, serotonin, and neuropeptides such as vasoactive intestinal peptide (VIP), substance P (SP), and calcitonin gene–related peptide (CGRP) (Gonzalez-Rey et al., 2010; Kawashima et al., 2012; Mashimo et al., 2022; Thoppil et al., 2023; Wang et al., 2003; Klein Wolterink et al., 2022). Immune cells have also been shown to produce and secrete classical neuromodulators such as acetylcholine, serotonin, dopamine, glutamate, and catecholamines, which can result in both pro- and anti-inflammatory responses (Flierl et al., 2007; Hodo et al., 2020). It is through this combination of neuroanatomical circuits, paracrine signaling, and molecular receptor interactions that the vagus nerve orchestrates sophisticated neuroimmune regulation across multiple physiological contexts.

In this review, we examine the organization of vagal neuroimmune circuits from peripheral sensing to central integration and immune regulation. We first describe the anatomy and cellular diversity of vagal afferent and efferent pathways. We then consider how inflammatory signals are detected and represented by vagal sensory neurons, how central circuits transform these signals into autonomic and behavioral responses, and how vagal pathways influence immune function in different organs and disease states. Finally, we evaluate bioelectronic strategies that target these circuits and identify key challenges for the field.

The vagus nerve is the longest cranial nerve and the principal conduit linking the brainstem to thoracic and abdominal viscera. Its fibers comprise ∼80% afferent sensory neurons and 20% efferent motor neurons, enabling bidirectional information exchange between peripheral tissues and central autonomic nuclei (Berthoud and Neuhuber, 2000; Matteoli and Boeckxstaens, 2013). The sensory afferent and motor efferent pathways arise from two distinct but complementary structures: the vagal ganglia complex, which contains the cell bodies of vagal sensory neurons, and the dorsal motor nucleus of the vagus (DMN) in the brainstem, which provides preganglionic efferent output to parasympathetic postganglionic neurons in target organs. The brainstem nucleus tractus solitarius (NTS) serves as the primary central relay for vagal afferents, integrating immune and visceral signals with autonomic and neuroendocrine responses (Gasparini et al., 2020; Ran et al., 2022; Travagli et al., 2006). Efferent fibers originate in the DMN and nucleus ambiguus, projecting toward postganglionic neurons embedded within target tissues involved in cardiorespiration (heart and lungs), digestion (gastrointestinal tract, pancreas, and liver), and immune function (Coverdell et al., 2024; Pavlov and Tracey, 2012; Williams et al., 2016). Although this review focuses on visceral circuits, the vagus nerve also carries a somatic sensory component. The auricular branch of the vagus nerve supplies the cymba conchae and external auditory canal, providing an anatomical basis for transcutaneous auricular stimulation approaches discussed later (Butt et al., 2020). Recent evidence indicates that this somatic vagal field is not simply a convenient access point for stimulation, but a functional afferent territory whose activation can shape immune responses in visceral organs (Shibuya et al., 2026).

Within mucosal tissues of visceral organs, including the gastrointestinal tract, airways, and lungs, vagal afferents display remarkable structural and functional diversity (Kupari et al., 2019; Zhao et al., 2022). In the gut, vagal afferents form an extensive network of mechanosensory and chemosensory endings throughout the mucosa, muscularis, and lamina propria. Berthoud and Powley (1992) first characterized the principal ending types: intraganglionic laminar endings (IGLEs), which synapse onto enteric ganglia; intramuscular arrays, which detect mechanical stretch in smooth muscle; and finely branching mucosal endings, ideally situated to sample luminal contents (Phillips and Powley, 2000). Within the intestinal mucosa, vagal fibers terminate in regions populated by tissue-resident sentinels, including enteroendocrine cells, resident mast cells, and macrophages (Williams et al., 1997). Recent studies suggest that neuropod cells, a subset of enteroendocrine cells, can communicate rapidly with vagal afferents to convey gut luminal chemical information to the brainstem (Kaelberer et al., 2018). Whether this communication requires direct synaptic contacts in all gut regions remains debated, as recent anatomical studies support paracrine rather than synaptic signaling for some enteroendocrine cell–vagal afferent interactions (Spencer et al., 2024). This neural architecture positions the vagus nerve to detect cytokines, microbial metabolites, and inflammatory mediators released during intestinal immune responses, potentially participating in local immune regulation.

A similar organizational logic applies to airway tissues, where the lungs are primarily innervated by vagal afferent fibers that penetrate the airway mucosa, with terminals residing in the epithelial layer (Mazzone and Undem, 2016; Su et al., 2022; Kim et al., 2022). These fibers are distinct nociceptor subsets closely associated with pulmonary neuroendocrine cells and resident immune populations, enabling rapid sensory transduction in response to inhaled irritants, pathogens, or local inflammatory signals (Darcey et al., 2025; Su et al., 2022). At the molecular level, single-cell transcriptomic analyses of vagal ganglia neurons have identified multiple molecularly distinct subsets of vagal sensory neurons specialized for visceral surveillance. These include populations expressing transient receptor potential (TRP) vanilloid 1 (TRPV1), TRP ankyrin 1 (TRPA1), and Mas-related G-protein–coupled receptor member D, as well as chemosensory subtypes enriched for IL-1 receptors (IL-1R1), TNF receptors, IL-10 receptors (IL-10R), IL-6ST, and other cytokine receptors (Prescott et al., 2020; Kupari et al., 2019; Zhao et al., 2022; Jin et al., 2024). Expression of these receptors in vagal sensory neurons suggests that defined afferent subsets are positioned to respond to specific cytokines, although the relative contribution of direct neuronal sensing versus indirect signaling through intermediary cells remains incompletely resolved. Importantly, inflammation and infection within the airways have also been shown to induce de novo expression of specific genes, further complicating the interpretation of signaling pathways (Undem and Taylor-Clark, 2014; Zaccone and Undem, 2016). Transcriptionally distinct subsets exist between different visceral organs and within different layers of the same organ (Zhao et al., 2022). This level of transcriptional segregation has been used to delineate functional specificity for vagal subsets and the behavioral circuits they control. Although most clearly defined for homeostatic functions such as feeding, where glucagon-like peptide-1 receptor+ and oxytocin receptor gene+ IGLE mechanoreceptors regulate feeding behavior, this organizational principle provides a framework for asking whether immune-responsive vagal subsets are similarly specialized by organ, tissue layer, or inflammatory context (Bai et al., 2019).

Vagal efferent projections, arising from the DMN, play a reciprocal role in maintaining immune homeostasis in mucosal organs. In the gastrointestinal tract, vagal efferents form synaptic connections with enteric neurons to regulate gastric and small intestinal motility and communicate with intestinal macrophages to modulate cytokine release and barrier function (Matteoli et al., 2014). Acetylcholine released by vagal terminals or enteric cholinergic neurons engages α7nAChR on muscularis macrophages, promoting an anti-inflammatory phenotype and facilitating tissue repair (Matteoli et al., 2014). The proximity of vagal fibers to lymphoid aggregates, Peyer’s patches, and mesenteric lymph nodes further suggests anatomical routes through which autonomic and immune signals may interact along the gut–brain axis, although functional roles for these associations require circuit-level validation (Zhang et al., 2022).

The spleen has been considered a central organ in neural regulation of systemic immunity. Although direct vagal innervation of the spleen remains anatomically debated, functional evidence supports the existence of a vagal-splenic neural circuit that mediates cytokine production (Zhu et al., 2024). The prevailing model posits that efferent vagal signals are relayed via the celiac–splenic sympathetic pathway: preganglionic vagal fibers synapse on neurons in the celiac ganglion, which project noradrenergic terminals to the spleen (Rosas-Ballina et al., 2011). These terminals, in turn, modulate splenic choline acetyltransferase (ChAT)+ T cells, which release acetylcholine that acts on α7nAChR-expressing macrophages to inhibit TNF and other proinflammatory cytokines (Rosas-Ballina et al., 2011; Olofsson et al., 2012). Functional mapping studies support this relay model, showing that cholinergic neurons in the DMN project to the celiac-superior mesenteric ganglionic complex, that activation of these neurons increases splenic nerve activity, and that this pathway suppresses systemic TNF production (Kressel et al., 2020). These findings support the canonical CAIP while leaving open how this pathway interacts with other sympathetic routes of immune regulation.

Electrophysiological and genetic evidence support functional coupling between the vagus nerve and systemic immune activation. Electrical vagus nerve stimulation (VNS) reduces systemic TNF release and improves survival during endotoxemia, an effect abolished by either splenectomy or α7nAChR deletion (Borovikova et al., 2000; Wang et al., 2003). Similarly, selective activation of DMN neurons or sympathetic neurons in the superior mesenteric/celiac ganglion complex can engage splenic neuroimmune pathways that suppress TNF production (Murray et al., 2019; Kressel et al., 2020). Together, these findings support the spleen as a principal downstream site through which vagal and sympathetic pathways can regulate systemic cytokine production. Additional studies suggest that this vagal–splenic network is bidirectional. Vagal afferents innervating the upper gastrointestinal tract and hepatic portal region can respond to circulating cytokines and metabolites that indirectly reflect splenic immune activity (Goehler et al., 2000). Moreover, the spleen communicates with central autonomic nuclei via humoral and cytokine-mediated pathways, creating a feedback loop that links splenic immune activation to brainstem circuits. Together, these findings highlight that the vagal and splenic axes function as a coupled system, wherein vagal efferents modulate immune tone, and afferents relay systemic inflammatory status to the brain.

Collectively, the gastrointestinal and splenic vagal axes thus represent complementary arms of a unified vagal neuroimmune network. The afferent component detects peripheral immune perturbations, particularly within the gut mucosa, and transmits this information to the NTS, where it is integrated with metabolic, visceral, and hormonal inputs. The efferent component, via cholinergic and sympathetic intermediates, exerts targeted control over macrophage and lymphocyte function to restore homeostasis. This bidirectional communication mediated by the vagus nerve underlies a reflexive control system that maintains immune balance while preserving essential host-defense mechanisms. Anatomically distributed yet functionally coordinated, these circuits exemplify how the vagus nerve serves as an adaptable interface between the immune system and the brain.

The vagus nerve modulates immune responses at both local and systemic levels. Acting primarily through cholinergic signaling, vagal efferents influence the activation state, cytokine release, and phenotype of immune cells in the gut, spleen, and other visceral organs. Vagal afferents modulate local immune responses primarily through neuropeptide release, including VIP, CGRP, SP, and somatostatin (SST). Sympathetic postganglionic neurons, which interface with vagal circuits, additionally release neuropeptide Y to regulate immune function. Collectively, these actions limit excessive inflammation, promote resolution, and support tissue repair.

A well-defined efferent mechanism of vagal neuro-immunomodulation is the CAIP, first characterized in models of endotoxemia and sepsis (Borovikova et al., 2000; Pavlov and Tracey, 2012; Tracey, 2002). The circuit architecture of this pathway is described above: efferent cholinergic vagal fibers descending from the DMN engage neurons of the celiac–superior mesenteric ganglionic complex, whose noradrenergic terminals in the spleen stimulate β₂-adrenergic receptors on ChAT+ T cells, which in turn release acetylcholine onto α7nAChR-expressing macrophages (Rosas-Ballina et al., 2011; Kressel et al., 2020). Electrical or optogenetic activation of vagal efferents increases splenic nerve activity and norepinephrine release, completing a multiorgan relay that translates vagal nerve activity into systemic suppression of TNF and IL-6 (Borovikova et al., 2000; Rosas-Ballina et al., 2011; Kressel et al., 2020). At the level of the macrophage, engagement of α7nAChR activates the JAK2-STAT3 signaling cascade, inhibits nuclear translocation of NF-κB, and suppresses translation of proinflammatory cytokines such as TNF, IL-1β, and HMGB1 (Chavan et al., 2017; Wang et al., 2003). Cholinergic signaling through this receptor also promotes IL-10 production (Chavan et al., 2017; Tracey, 2024). This pathway is one of the best-characterized examples of neuroimmune relay circuitry, in which distinct neuronal and immune elements cooperate to execute a reflexive anti-inflammatory response. Its reach extends beyond systemic endotoxemia: in preclinical studies, CAIP activation by electrical VNS reduces inflammation in several models of colitis, including 2,4,6-trinitrobenzene sulfonic acid– and DSS-induced colitis, thereby improving epithelial integrity and reducing monocyte/macrophage activation (Meregnani et al., 2011; Meroni et al., 2021).

Vagal efferents can also reach immune targets without a splenic intermediate. In the context of postoperative ileus, the Boeckxstaens group established a direct circuit of efferent vagal innervation to the gut. In this pathway, vagal efferents synapse with enteric neurons in the intestinal wall that release acetylcholine onto local α7nAChR-expressing macrophages, thereby resolving intestinal inflammation (Matteoli and Boeckxstaens, 2013). Activation of this direct vagal efferent circuit to the gut was sufficient to prevent postoperative ileus in mice (The et al., 2007). Together, these two anti-inflammatory pathways illustrate how efferent vagal activity regulates immunity both locally within tissues and systemically throughout the body (Fig. 1).

Vagal sensory neurons and their peripheral nerve terminals can also release a variety of neuropeptides, including VIP, CGRP, SP, and SST, each exerting distinct effects on immune cell function depending on context and tissue localization (Chiu et al., 2012; Erdogan et al., 2025). This local neuropeptide release occurs through an axon reflex mechanism, in which calcium influx and back-propagating action potentials trigger the exocytosis of dense-core vesicles throughout the peripheral arbor of these neurons (Chiu et al., 2012). Among the neuropeptides vagal sensory neurons can produce, CGRP is a major immunomodulatory mediator. CGRP signaling has been shown to attenuate IL-13 production by type 2 innate lymphoid cells (ILCs), promote tissue repair through reprogramming of macrophage transcriptional states, suppress neutrophil recruitment, and suppress lung inflammation (Baral et al., 2018; Nagashima et al., 2019; Tamari et al., 2024; Wallrapp et al., 2019). CGRP is abundantly expressed by a subset of vagal sensory neurons, particularly those originating from the jugular ganglion, and has been traced along vagal projections innervating airways, the heart, and gastrointestinal organs (Hayakawa et al., 2011; Hayakawa et al., 2014; Springall et al., 1987). While there is likely a mixed contribution of CGRP secretion from vagal and spinal dorsal root ganglia (DRG) afferents in the gut, the lung is known to be densely innervated by the sensory vagus nerve with less DRG input. Additionally, vagal afferents innervate mucosal tissue, and their terminals are often found near barrier lymphocytes (Williams et al., 1997). This organization allows bidirectional paracrine signaling between vagal fibers and immune cells, with important consequences for inflammatory responses and tissue homeostasis. VIP, widely expressed in vagal and enteric neurons, acts via VPAC1 and VPAC2 receptors on macrophages, dendritic cells, and T lymphocytes to suppress proinflammatory cytokine production while promoting T helper 2 (Th2) polarization and regulatory T cell differentiation. In the gut, VIP maintains epithelial barrier function by limiting myeloid activation and stimulating IL-22 production by ILCs (ILC3s), thereby enhancing mucosal defense without provoking tissue damage. SP is a potent proinflammatory neuropeptide that acts on the neurokinin 1 receptor to promote vascular permeability and pain and also contributes to airway inflammation by inducing mucus production and bronchoconstriction (O’Connor et al., 2004; Steinhoff et al., 2014; Li et al., 2022). SP is an established mediator of neurogenic inflammation through multiple mechanisms, including effects on vascular permeability and smooth muscle tone. In addition, SP can directly activate mast cells to degranulate via the Mgpcr MrgprB2 (MRGPRX2 in humans) (Nagamine et al., 2024; Chompunud Na Ayudhya et al., 2021). SST may play a more immune-inhibitory role, acting through its receptors SSTR1–5 to inhibit T cell proliferation and cytokine production (ten Bokum et al., 2000; Pintér et al., 2006). The balance between opposing neuropeptide signals shapes local immune microenvironments. For instance, during colitis, increased SP release amplifies mucosal inflammation, whereas immunomodulatory neuropeptides such as CGRP and SST help suppress inflammation and restore homeostasis (Reinshagen et al., 1998; Patel et al., 2020; Erdogan et al., 2025).

Collectively, these observations reveal that vagal output does not uniformly suppress immunity but rather modulates it contextually, favoring resolution and tissue protection when needed while preserving host defense. The functional outcome depends on which neurochemical mediators are released, the receptors expressed by local immune populations, and the tissue state in which these interactions occur. Recent findings have begun to reveal how specific ion channels expressed by vagal afferent subsets are crucial for mucosal immunity. For example, co-expression of TRPA1 and IL-1R by vagal afferents was shown to be necessary for promoting the afferent arm of the inflammatory reflex (Silverman et al., 2023). Another study showed that TRPV1⁺ vagal afferents are vital for suppressing damaging lung myeloid cell responses after influenza virus infection (Almanzar et al., 2025). Further research using targeted genetic approaches to dissect vagal afferent subsets will continue to elucidate the role of these ion channel-specific neuronal populations in shaping tissue-specific immune responses throughout the body, potentially identifying novel therapeutic targets for precision treatment.

Vagal sensory afferents convey visceral information, including signals about immune mediators, along the body–brain axis to the brainstem. Classical physiological studies have demonstrated the breadth of vagus nerve contributions to autonomic function, including respiration, heart rate, gastrointestinal motility, and blood pressure regulation. Given the range of internal organs innervated by the vagus nerve, recent work has revealed tremendous diversity among vagal sensory neuron types, which may partially explain this functional diversity (Prescott and Liberles, 2022).

The vagus nerve has been implicated in sensing immune mediators and initiating sickness behavior (Dantzer and Kelley, 2007; Thayer and Sternberg, 2010). Vagal encoding of immune stimuli provides a faster route of communication and a more anatomically specific alternative to the traditional endocrine route via the blood to the central nervous system (CNS) (Huerta et al., 2025; Goehler et al., 2000). By detecting inflammatory mediators, either directly or through intermediary cells, and conveying this information to central nuclei, vagal sensory neurons provide the first step in the reflexive regulation of immune responses. Vagal sensory neurons exhibit molecular and functional specialization that allows them to discriminate between distinct immune states through distinct patterns of neural activity (Fig. 2).

Electrophysiological analyses have demonstrated that subsets of vagal afferents generate distinct compound action potentials or “neurograms” in response to specific cytokines (Steinberg et al., 2016; Zanos et al., 2018). This research presented evidence that vagal sensory neurons can discriminate between different inflammatory cues, such as systemic administration of IL-1β and TNF, at the population level. To gain insight into vagal cytokine sensing at the individual neuron level, Jin et al. (2024) combined in vivo calcium imaging with molecular profiling to show that vagal sensory neurons express a variety of immune sensors, including cytokine receptors, polymodal chemosensors (TRP channels), and TLRs, and that discrete nodose neuron subsets display distinct activation patterns after systemic administration of inflammatory stimuli such as IL-10, IL-1β, and IL-6. Vagal sensory afferent subsets express receptors for these cytokines, including IL-10R for IL-10 and IL-6ST for IL-6, but whether each cytokine acts mainly directly on vagal terminals in vivo or through intermediary epithelial, immune, or stromal cells is likely context-dependent and remains to be fully resolved. This work highlighted a dynamic body–brain axis, in which cytokine-responsive vagal neurons relay immune information to the brainstem, activating dopamine β-hydroxylase (DBH+) neurons that can be captured using TRAP2 methodology and subsequently reactivated to either enhance anti-inflammatory responses or shut down immune regulation, leading to unregulated inflammation (Jin et al., 2024). Similar work by Huerta et al. (2025) demonstrated that individual neurons in the nodose ganglion respond differently to specific cytokines when exposed directly to the vagus nerve and adapt their responses depending on the severity of colitis. Although this work suggests that disease state may also lead to plasticity through altered transcriptional pathways, the mechanism of change remains unclear.

It is currently unknown whether vagal encoding of immune signals is organ-specific. Evidence so far suggests that there is no clear somatotopic organization in the vagal ganglia, but responses in the brainstem NTS have been shown to be organotopically mapped (Ran et al., 2022). This discrepancy between an apparently unstructured organization in peripheral ganglia and highly organized mapping in the brainstem raises important questions about how the body sorts and interprets these signals. One hypothesis is that the unique features of vagal afferent fibers, such as their sensitivity to certain cytokines or danger signals, only become functionally relevant once they reach the brainstem, where central circuits sort these inputs to produce organ-specific responses. Another possibility is that encoding occurs at the level of nodose neurons, such that vagal sensory neurons are intrinsically able to discriminate different stimuli based on their molecular receptor profiles. Another major question is how the vagus nerve balances the plethora of immune and visceral signals through its relatively limited number of fibers. It remains unknown whether sensing of inflammation interferes with signaling of homeostatic processes, or whether these functions are encoded by distinct fiber populations.

Vagal and nociceptor afferents express receptors traditionally associated with immune cells, including TLRs and cytokine receptors such as IL-1R and IL-10R. Studies have shown that peripheral administration of either LPS or IL-1 stimulates vagal responses in the NTS. Jin et al. (2024) found that direct engagement of vagal TLR4 was not required for endotoxin-induced brainstem responses, suggesting that vagal activation was occurring through indirect or secondary immune mechanisms. In that study, there was robust activation of discrete nodose neuron subsets by cytokines, whereas LPS failed to directly activate vagal neurons even at high doses; moreover, LPS-induced brainstem activation and cytokine induction were blunted in Myd88^−/− mice, supporting an indirect mechanism in which LPS engages Myd88-dependent innate immune sensing to generate secondary mediators (e.g., cytokines and lipid mediators) that then activate vagal afferents. By contrast, other studies support direct LPS effects on sensory neurons under specific experimental conditions. In a bacterial pneumonia model, Granton et al. showed that exposed LPS from non-biofilm–producing Gram-negative bacteria was detected through TLR4 in lung TRPV1⁺ sensory neurons, driving sickness behavior, hypothermia, and activation of hypothalamic stress circuits; these effects were attenuated by sensory neuron-specific Tlr4 deletion and by targeted Tlr4 deletion in vagal ganglia neurons (Granton et al., 2024). LPS has also been shown to induce CGRP release from Nav1.8-restricted TLR4-expressing vagal afferents in vitro (Jia et al., 2021). Together, these findings indicate that LPS sensing by vagal and other sensory afferents is context-dependent: in some settings, LPS can be detected directly by defined sensory neuron subsets, whereas in others, including systemic LPS responses in Jin et al., vagal activation may occur indirectly through immune-derived cytokines, lipid mediators, or other secondary inflammatory signals.

Vagal sensing of immune mediators is an integral first step in initiating the inflammatory reflex, where proinflammatory cytokines are sensed by vagal afferents, leading to reduction of inflammation through the CAIP. Functionally, IL-1β–sensitive vagal afferents initiate sickness behavior, fever, and HPA axis activation, whereas TNF- and IL-10–responsive neurons mediate differential coding of pro- and anti-inflammatory states (Maier and Watkins, 1998; Goehler et al., 2000; Jin et al., 2024; Huerta et al., 2025). These findings support a model in which cytokine detection is molecularly compartmentalized across vagal sensory subtypes, enabling high-fidelity representation of the immune milieu.

Cytokine receptor engagement can influence neuronal excitability through intracellular signaling pathways that converge on ion channels, second messengers, and inflammatory mediators. Specific ion channels, such as TRPV1 and potentially TRPA1, can be coupled to these inflammatory signaling pathways, thereby amplifying calcium and cation influx upon receptor activation (Chiu et al., 2012; Erdogan et al., 2025; Silverman et al., 2023). Activation of these neurons elicits the local release of neuropeptides such as CGRP and SP, which can amplify or suppress local immune activity depending on tissue context. Thus, the nervous system not only responds to inflammation but also participates actively in its initiation and resolution. Transcriptomic studies have detected expression of PRRs in nodose ganglion neurons, including TLR4, TLR7, NOD-like receptors, and formyl peptide receptors. While DRG nociceptors have been shown to directly sense bacterial products, such as LPS and N-formylated peptides (Chiu et al., 2012; Meseguer et al., 2014), evidence for direct pathogen-associated molecular pattern activation of vagal afferents remains limited (Reardon et al., 2018). Consistent with the LPS discussion above, current in vivo evidence supports a predominant role for indirect immune-to-neural signaling in LPS-evoked vagal activation, while direct neuronal detection in specific subsets or contexts remains unresolved. Thus, whether vagal sensory neurons act as direct detectors of microbial products in a manner analogous to primary immune sentinels, or instead primarily encode secondary cytokine and lipid mediator signals generated by classical sentinel cells such as macrophages and dendritic cells, remains an important area for future investigation.

Thus, recent discoveries have established a novel role for vagal sensory neurons as integrators of cytokine and microbial information. Distinct subsets of vagal sensory neurons exhibit cytokine-specific neural activity patterns, forming the afferent arm of a body–brain circuit that maps immune signals onto discrete brainstem populations (Huerta et al., 2025; Jin et al., 2024). In parallel, efferent vagal and enteric circuits regulate macrophage and lymphocyte activity through acetylcholine and neuropeptide signaling, establishing local feedback loops that maintain mucosal and systemic balance. Together, these findings delineate the vagal neuroimmune axis as a bidirectional control system integrating immune, metabolic, and behavioral states. Understanding the molecular logic and functional plasticity of these pathways offers new therapeutic opportunities to treat inflammatory and autoimmune disorders.

The CNS integrates vagal immune signals, converting peripheral cues detected by vagal afferents into coordinated brain-driven changes in physiology, inflammation, and behavior. The vagus nerve has long been linked to immune responses associated with infection and sickness behaviors (Maier and Watkins, 1998; Goehler et al., 2000). Once immune signals are detected by vagal afferents, they are transmitted to the NTS, where they are integrated and relayed to further autonomic and neuroendocrine centers to coordinate systemic inflammatory responses (Dantzer and Kelley, 2007; Pavlov and Tracey, 2017). Within the NTS, distinct neuronal populations, including Dbh⁺ neurons, serve as central hubs that convert cytokine-specific vagal input into adaptive efferent responses, modulating inflammation through the HPA axis and parasympathetic motor output (Jin et al., 2024). Selective activation or silencing of this vagal–NTS–forebrain circuitry can bidirectionally modulate systemic cytokine responses, defining a hierarchical neural network through which the brain both perceives and regulates immune activity (Jin et al., 2024). Descending projections regulate parasympathetic immune tone via the DMN. In parallel, the NTS can engage premotor sympathetic circuits, including the rostral ventrolateral medulla, to influence sympathetic outflow to immune-relevant organs such as the spleen (Guyenet and Stornetta, 2022). How these central sympathetic relays interact with the canonical CAIP model remains unresolved. One model emphasizes efferent vagal input to the celiac-superior mesenteric ganglionic complex, which can drive splenic nerve activity and downstream ChAT+ T cell–dependent acetylcholine release. Other studies support afferent-to-sympathetic routes in which vagal sensory input recruits brainstem sympathetic pathways to regulate splenic or renal inflammation. These models are not necessarily mutually exclusive, but their relative contribution likely depends on stimulus modality, disease context, and stimulation parameters. While outside of the scope of this review, there is emerging evidence that higher CNS centers, such as the parabrachial nucleus and insular cortices, encode immune and interoceptive information, linking immune activation to behavioral and motivational changes, which supports adaptive host defense while avoiding excessive inflammation (Koren et al., 2021; Koren and Rolls, 2022). Recent work from our lab has shown that CNS regions, including the bed nucleus of the stria terminalis, are part of precise neural circuits that encode inflammatory mediators and can activate specific stress-induced responses (Hashimoto et al., 2026).

The identification of vagal circuits as regulators of immunity has driven extensive experimental and translational investigation across diverse disease contexts. From acute endotoxemia to chronic inflammatory and autoimmune disorders, manipulation of vagal activity through electrical, pharmacologic, or behavioral means has revealed the capacity of this pathway to suppress cytokine production, promote resolution, and improve survival. In earlier work, experimental silencing of vagal signaling, primarily by surgical or chemical vagotomy, in disease models ranging from endotoxemia, gastrointestinal inflammation, lung inflammation, acute kidney injury, sepsis, and arthritis, consistently results in exaggerated immune activation and worse outcomes in acute and chronic inflammatory models (Ghia et al., 2006; Huston et al., 2006; Inoue et al., 2016; Sévoz-Couche et al., 2024). Conversely, activation of vagal pathways, whether through pharmacological cholinergic agonists, cholinesterase inhibitors, optogenetic modulation, or electrical stimulation, restores homeostatic immune function and supports the resolution of inflammation in organs as diverse as the gut, lung, kidney, and joints (de Jonge et al., 2005; Meregnani et al., 2011; Miceli and Jacobson, 2003; Murray et al., 2025; Tanaka et al., 2021). Here we synthesize this evidence thematically, spanning acute inflammatory responses, chronic mucosal inflammation, and autoimmune diseases.

Early demonstrations of the inflammatory reflex emerged from models of acute systemic inflammation. In endotoxemic rats, electrical stimulation of the vagus nerve markedly reduced serum TNF levels and prevented lethal shock (Borovikova et al., 2000; Huston et al., 2006; Rosas-Ballina et al., 2008; Wang et al., 2003). These effects were abolished by splenectomy or α7nAChR blockade, confirming that the spleen serves as a critical efferent target of vagal control (Wang et al., 2003). Mechanistic studies indicate that vagal efferent activity attenuates macrophage activation and neutrophil recruitment during acute infection, limiting collateral tissue damage. These data establish that vagal activation is both necessary and sufficient to mitigate the cytokine storm characteristic of acute systemic inflammation. Conversely, vagotomy exacerbates endotoxemia and worsens outcomes in sepsis, demonstrating that intact vagal signaling is required for survival under inflammatory stress (Borovikova et al., 2000; Chavan et al., 2017).

The gastrointestinal tract provides a paradigm for chronic, low-grade inflammation under neural control. In models of colitis, vagal stimulation or pharmacologic enhancement of cholinergic tone ameliorates disease severity, reducing mucosal TNF and IL-1β levels, preserving epithelial integrity, and promoting regulatory macrophage polarization (Bonaz et al., 2017; Bonaz et al., 2019; Matteoli et al., 2014). Mice with selective disruption of α7nAChR signaling exhibit exaggerated inflammatory responses and impaired healing, highlighting the essential role of cholinergic signaling in intestinal homeostasis. Vagal modulation also interacts with the enteric nervous system to regulate local immune microenvironments. During colitis, enteric neurons release acetylcholine and VIP that act on macrophages and ILCs to promote IL-22–mediated epithelial repair. This coordinated vagal–enteric–immune axis ensures that inflammatory resolution is coupled to tissue regeneration.

Acute kidney injury following ischemia-reperfusion involves robust inflammatory responses, including neutrophil infiltration and cytokine release that amplify tissue damage beyond the initial ischemic insult. Vagal circuits provide significant protection against renal ischemia-reperfusion injury through two distinct neuroimmune pathways (Inoue et al., 2016; Tanaka et al., 2021). Using optogenetics to selectively activate vagal populations, Tanaka et al. (2021) identified both the canonical efferent cholinergic pathway and a novel sensory afferent route. In this second pathway, vagal sensory activation triggers brainstem C1 neurons to activate sympathetic outflow via the splenic nerve. This afferent-to-sympathetic relay circuit demonstrates that vagal sensory fibers can indirectly control renal inflammation through central sympathetic intermediates. Both pathways converge to reduce renal cytokine production, limit neutrophil infiltration, and preserve kidney function during ischemic injury. This pathway suggests that activation of sensory vagal fibers may provide renoprotective effects and expand treatment options for inflammatory kidney diseases.

Beyond local inflammation, vagal modulation influences systemic immune balance in chronic autoimmune conditions. In rheumatoid arthritis (RA) models, VNS reduces joint inflammation, bone erosion, and serum TNF levels, effects mediated by splenic α7nAChR+ macrophages and ChAT+ T cells (Levine et al., 2020). Chronic VNS induces sustained reductions in IL-6 and HMGB1 and promotes expansion of regulatory T cells, leading to durable amelioration of disease severity. Similar findings have been reported in systemic lupus erythematosus and multiple sclerosis models, where enhancement of vagal tone shifts immune polarization from Th1/Th17 dominance toward regulatory T and Th2 phenotypes (Pavlov and Tracey, 2012). In metabolic inflammation, vagal activity modulates adipose tissue macrophage polarization and glucose metabolism, linking neural control of inflammation to metabolic homeostasis. Reduced vagal tone, a common comorbidity in obesity, diabetes, and cardiovascular disease, is associated with systemic low-grade inflammation and impaired immune regulation (Bonaz et al., 2016). Restoring vagal tone through stimulation, exercise, or behavioral interventions can partially reverse these immune abnormalities, emphasizing the broad physiological influence of parasympathetic signaling.

Building on these experimental foundations, bioelectronic medicine has emerged as a strategy to treat inflammatory and autoimmune diseases by targeting vagal neuroimmune circuits. VNS, both invasive (cervical) and noninvasive (transauricular, transcutaneous, or ultrasound), is designed to engage neural pathways that regulate inflammatory tone, with clinical effects that appear to depend on indication, stimulation approach, and patient population (Fig. 3). Clinical studies provide encouraging but still heterogeneous evidence for anti-inflammatory effects of VNS. Implanted VNS devices have reduced disease activity and serum TNF levels in patients with refractory RA (Gaylis et al., 2025; Genovese et al., 2020; Koopman et al., 2016) and pilot studies in Crohn’s disease and ulcerative colitis have reported clinical improvement, changes in inflammatory biomarkers, and variable endoscopic responses (Bonaz et al., 2017; D’Haens et al., 2023; Liu et al., 2025; Mikami et al., 2022; Sinniger et al., 2020). In a preregistered systematic review of 12 VNS clinical trials across a range of autoimmune diseases, more than half of the studies reported reductions in pro-inflammatory cytokines, with IL-6 showing the most consistent decrease across seven studies (Lombo et al., 2025). C-reactive protein and TNF-α were also reduced in the majority of studies measuring these markers. However, responses are not uniform, and systematic analyses have not found consistent anti-inflammatory effects of VNS across human studies (Schiweck et al., 2024). These differences likely reflect variation in disease state, stimulation site, stimulation parameters, outcome measures, and baseline autonomic or inflammatory tone. Mechanistically, clinical effects may involve both afferent-driven changes in central immune processing and efferent activation of anti-inflammatory pathways, with downstream effects on immune cell function, cytokine production, and barrier integrity (Eberhardson et al., 2021).

Recent regulatory milestones have also shaped the field of bioelectronic medicine. The Vivistim System, approved by the U.S. Food and Drug Administration (FDA) in 2021 for chronic ischemic stroke rehabilitation, illustrates the clinical feasibility of paired VNS for activity-dependent neural plasticity, although its relevance to inflammatory disease is indirect (Hays et al., 2025; Khodaparast et al., 2013; Kilgard et al., 2025). More directly relevant to neuroimmune modulation, cervical VNS was approved by the U.S. FDA in 2025 for moderate-to-severe RA in patients who have not responded to or cannot tolerate biologic or targeted synthetic disease-modifying antirheumatic drugs. This approval was based on the RESET-RA study, a randomized, double-blind, sham-controlled clinical trial of patients with moderately to severely active RA (Tesser et al., 2025). This regulatory milestone supports the translational relevance of inflammatory reflex biology, while continued post-approval studies will be important for defining durability, responder characteristics, and comparative effectiveness. Clinical trials are also advancing for other inflammatory conditions. In a prospective open-label study of VNS in biologic-refractory Crohn’s disease, treatment was associated with reductions in clinical disease activity (clinical disease activity index), fecal calprotectin levels, and serum cytokines (TNF and IFN-γ), while paired endoscopic measures showed variable improvement (D’Haens et al., 2023). These results are encouraging, but the open-label design, small sample size, absence of a sham-control group, and incomplete or heterogeneous responses limit interpretation. Together with earlier Bonaz studies in Crohn’s disease, these findings suggest that VNS-based neuroimmune modulation may benefit selected patients with inflammatory bowel disease, while emphasizing the need to identify predictors of response and nonresponse (Bonaz et al., 2017; Meroni et al., 2021; Stakenborg and Boeckxstaens, 2021). The presence of nonresponders in early Crohn’s disease studies may reflect differences in disease stage, inflammatory burden, medication exposure, autonomic tone, or stimulation engagement of relevant vagal fibers.

Closed-loop VNS systems that monitor neural or physiological biomarkers and adjust stimulation parameters in real time may eventually improve the precision of neuroimmune modulation (Ottaviani et al., 2022). However, biomarker selection remains a major challenge. Heart rate variability (HRV), blood pressure, and electrodermal activity provide indirect measures of autonomic state, but they are not inflammation-specific and can be influenced by stress, posture, medications, respiration, and cardiovascular comorbidities. Circulating cytokines and immune cell phenotypes may more directly reflect inflammatory activity, but they are difficult to measure with the temporal resolution needed for real-time closed-loop control. Vagal compound action potentials or branch-specific neural recordings may provide more proximal readouts of stimulation engagement, although their relationship to downstream immune outcomes requires further validation (Patros et al., 2024). Closed-loop stimulation in spinal cord injury rehabilitation, where stimulation is paired with successful movements to enhance neuroplasticity, provides a useful proof-of-principle for adaptive neuromodulation (Ganzer et al., 2018). For inflammatory diseases, comparable systems remain largely aspirational and require biomarkers that can reliably link stimulation, neural engagement, and immune state. Noninvasive transauricular VNS has also shown promise in reducing circulating cytokines and improving depressive symptoms associated with inflammation, supporting a broader role for vagal modulation in immune-brain communication (Lerman et al., 2025; Liu et al., 2024). However, clinical responses are heterogeneous, reflecting differences in electrode placement, stimulation parameters, individual autonomic baseline, and the degree to which stimulation engages disease-relevant vagal fibers.

Despite rapid advances in our understanding of vagal neuroimmune circuits and their therapeutic potential, several important questions and technological challenges remain. A central challenge is defining which vagal afferent and efferent subsets regulate specific inflammatory processes, and how these circuits change across organs and disease states. This question is particularly important because systemic inflammation can expose multiple vagal territories to shared circulating cytokines, raising the question of how organ specificity is maintained during widespread immune activation. Specificity may arise at several levels, including the distribution of peripheral terminals, local intermediary cells, receptor expression patterns, stimulus timing, and central sorting of afferent inputs in the NTS and higher brain regions. As these circuit principles become better defined, vagal neuroimmune research may also extend beyond classical inflammatory disorders to diseases in which autonomic dysfunction and chronic inflammation intersect, including metabolic, neurodegenerative, psychiatric, and oncologic conditions (Kelly et al., 2022; Balasubramanian et al., 2024; Erdogan et al., 2025). A recent synthesis of neurocardiac and neuroimmune biology highlighted how chronic sympathetic overdrive and parasympathetic withdrawal in cardiometabolic disease amplify myocardial inflammation and metabolic dysfunction through neuroimmune circuits involving the brain, bone marrow, and spleen—positioning vagal and splenic nerve modulation as an emerging therapeutic frontier for cardiovascular inflammatory disease (Ziegler et al., 2025). Dissecting the contribution of discrete neuronal subsets, such as those defined by unique neuropeptide, ion channel, or receptor expression patterns, to specific immune processes will be crucial for identifying new therapeutic targets.

A critical gap is the spatial organization and neural coding logic of immune-responsive vagal neurons. Single-cell transcriptomic studies have revealed extensive molecular diversity, yet the functional significance of most transcriptionally defined subsets remains unknown. Do specific vagal neuron subtypes respond exclusively to particular cytokines or immune challenges? How is specificity achieved at the molecular level—is it through differential receptor expression, intracellular signaling cascades, or synaptic connectivity? Another open question concerns the mechanisms of vagal plasticity during chronic inflammation. Studies have shown that vagal afferents undergo remodeling in response to intestinal inflammation, diabetes, high-fat diet, and neurotrauma, but the molecular drivers and functional consequences of this plasticity are poorly understood. Such plasticity may include changes in cytokine receptor expression, ion channel abundance, intrinsic excitability, rheobase, neuropeptide expression, peripheral terminal density, synaptic strength in the NTS, or recruitment of central autonomic circuits. These changes could be adaptive, by enhancing immune surveillance during tissue injury or infection, or maladaptive, by disrupting homeostatic reflexes and amplifying inflammatory signaling. Understanding these dynamics will require longitudinal approaches that combine transcriptomics, circuit mapping, electrophysiology, and in vivo measurements of vagal activity during immune responses and disease states (Alhadeff, 2021; Huerta et al., 2023; Huerta et al., 2025; Ran et al., 2022).

The development of biomarkers to guide VNS therapy represents another critical need. While HRV provides a general measure of autonomic tone, more specific biomarkers reflecting vagal immune signaling or peripheral immune cell phenotypes could enable personalized, closed-loop control strategies. Machine learning approaches may help identify such biomarkers by integrating multimodal physiological signals, such as electrocardiogram–derived HRV, respiration, blood pressure, activity, nerve recordings, and inflammatory biomarkers, to classify inflammatory state, predict response to stimulation, and optimize stimulation parameters in individual patients (Jafari et al., 2023). Emerging technologies such as focused ultrasound stimulation may further expand therapeutic precision by enabling noninvasive, spatially targeted modulation of specific vagal branches in relevant end-organs without surgical implantation. Direct mechanistic evidence for this noninvasive approach was recently shown by Shimoyama et al., who demonstrated that abdominal ultrasound activates afferent vagal fibers and induces c-Fos expression in the NTS, with effects abolished by subdiaphragmatic vagotomy and afferent blockade (Shimoyama et al., 2026). These results support the idea that noninvasive organ-targeted ultrasound can activate the afferent arm of the inflammatory reflex to suppress systemic inflammation (Huerta et al., 2021; Zanos et al., 2023). As scientific understanding of vagal neuroimmune communication advances, bioelectronic medicine stands at the frontier of precision, circuit-based treatments. It can be used in conjunction with standard-of-care pharmacological therapies or as an alternative therapy for treatment-resistant disease.

Vagal neuroimmune circuits form a dynamic communication network at the interface of the nervous and immune systems, in which specialized neuronal subtypes coordinate distinct aspects of host defense and inflammation. Cytokine encoding by vagal sensory neurons, combined with powerful efferent anti-inflammatory pathways, positions the vagus nerve as an important regulator of visceral immunity. Recent advances in molecular profiling, circuit mapping, and neural coding have revealed an unexpected complexity in how vagal pathways integrate signals from barrier tissues and convey them to the brain, shaping both acute immune responses and adaptation to changes in the organ microenvironment. The precise molecular and functional heterogeneity of vagal afferent and efferent subpopulations is only beginning to be unraveled, particularly in the context of immune regulation across different organs.

As we continue to map the complexity of vagal neuroimmune circuits, integrating basic neuroscience, immunology, and clinical medicine will pave the way for transformative therapies that harness the body’s own neural circuits to restore immune balance and treat disease. The successful clinical translation of VNS for RA validates decades of mechanistic research and opens new therapeutic avenues for other inflammatory disorders. As the fields of neuroimmune interactions and bioelectronic medicine advance, integration of our growing understanding of sensory neuron diversity, neural coding mechanisms, and disease-induced plasticity will enable increasingly precise targeting of specific neuroimmune circuits. The convergence of molecular profiling, systems-level circuit mapping, and increasingly sophisticated bioelectronic technologies will establish a new era in which targeted vagal neuromodulation becomes a foundational therapeutic strategy for inflammatory and autoimmune disorders.

This work was supported in part by grants from the National Institutes of Health/National Institute of General Medical Sciences (NIGMS) R35GM118182 (to K.J. Tracey), NIGMS R01GM143362 (to E.H. Chang), and National Institute of Allergy and Infectious Diseases 5R01AI168005 (to I.M. Chiu). I.M. Chiu also receives funding from the Kenneth Rainin Foundation, Food Allergy Science Initiative, and the Gene Lay Institute of Immunology and Inflammation.

Author contributions: Tomás S. Huerta: conceptualization, investigation, visualization, and writing—original draft, review, and editing. Kevin J. Tracey: conceptualization, funding acquisition, and writing—review and editing. Isaac M. Chiu: conceptualization, resources, supervision, and writing—review and editing. Eric H. Chang: conceptualization, funding acquisition, project administration, supervision, and writing—original draft, review, and editing.