The gut-brain axis in Binge Eating Disorder: From microbial dysbiosis to vagal dysfunction and therapeutic intervention

Abstract

Binge Eating Disorder (BED), the most prevalent eating disorder worldwide, is characterized by recurrent episodes of uncontrolled food intake accompanied by marked psychological distress in the absence of regular compensatory behaviors. While behavioral and neurobiological frameworks have traditionally dominated research, converging evidence now positions the gut microbiota as a critical mediator of BED pathophysiology through bidirectional microbiota-gut-brain axis (MGBA) communication. This review synthesizes current knowledge of gut microbial remodeling in BED, elucidating how compositional alterations contribute to pathogenesis through disrupted satiety signaling, altered neurotransmitter metabolism, immune-inflammatory cascades, and vagal dysfunction. Recent evidence highlights BED-specific depletion of fermentative taxa, including Pediococcus, Catenibacterium, and Megasphaera, with depletion of Pediococcus correlated with reduced cardiac vagal tone. The mechanistic landscape encompasses short-chain fatty acid (SCFA)-mediated appetite modulation, α-MSH mimicry by bacterial caseinolytic protease B, serotonin and GABA pathways, and immune activation via lipopolysaccharide (LPS). Emerging microbiota-targeted therapeutic strategies, including prebiotics, probiotics, synbiotics, and fecal microbiota transplantation (FMT), are critically evaluated in light of current evidence from BED and related eating disorders. Despite the promising mechanistic rationale, rigorous clinical trials remain scarce. This review highlights several priorities for future research, and the development of precision medicine strategies integrating established therapeutic approaches with emerging microbiota-targeted interventions.

1. Introduction

Among all eating disorders, BED is the most prevalent, with a worldwide lifetime prevalence estimated between 0.6–6.1% in women and 0.3–0.7% in men, surpassing that of anorexia nervosa (AN) and bulimia nervosa (BN) combined. ¹ Formally recognized as a distinct diagnostic entity in DSM-5 (2013) and ICD-11 (2019), BED is characterized by recurrent episodes of consuming objectively large amounts of food accompanied by an overwhelming sense of loss of control. ^{2 , 3} Contemporary research frames BED not merely as behavioral deregulation, but as a neurobiological disorder involving dysregulated reward circuits, impaired executive function, stress-response system alterations, and disrupted homeostatic mechanisms. ^{4 , 5} Within this evolving conceptual landscape, the gut microbiota has emerged as a compelling biological mediator. The trillions of microorganisms inhabiting the gastrointestinal tract communicate bidirectionally with the central nervous system (CNS) through the microbiota-gut-brain axis (MGBA), influencing appetite regulation, reward sensitivity, stress reactivity, and mood, processes that are central to BED pathophysiology. ^5-7 Nevertheless, the contribution of the gut microbiota to BED remains largely unexplored. The first evidence of BED-associated microbial alterations was reported by Leyrolle et al. in 2021, highlighting the still-emerging nature of this field. ⁸

The term “dysbiosis” is used throughout this manuscript with full awareness of its conceptual limitations. Olesen and Alm ⁹ argued that the term often refers to an implicit “balance” metaphor that presupposes a normative microbial state without empirical justification. Hooks and O'Malley ¹⁰ further claimed that dysbiosis is frequently invoked as an umbrella label rather than a mechanistic explanation, risking circular reasoning when a compositional shift is labeled pathological simply because it co-occurs with disease. Brüssow ¹¹ extended this critique, noting that the term is frequently deployed even in the absence of any clearly identified causal pathology, reducing its explanatory value and inviting vagueness across the literature. Despite these acknowledged shortcomings, dysbiosis is retained here as a necessary shorthand: no better single term currently exists to capture the “allostatic shift” toward an alternative microbial equilibrium that recurs across BED studies, variously described as depletion of fermentative taxa, fermentative guild collapse, synchronized depletion, or compositional alterations. Used with this caveat, the term is not meant to imply a fixed, universally “healthy” baseline microbiota, but rather to signal a directional, functionally consequential reorganization of the gut ecosystem that appears reproducibly associated with BED phenotypes.

This review specifically focuses on the role of the gut microbiota in BED, synthesizing evidence for microbial dysbiosis, mechanistic pathways linking microbiota perturbations to binge eating behavior, and emerging therapeutic strategies targeting the MGBA. We draw on evidence from human studies in BED and BN, animal models of binge-like eating, and related research in obesity and mood disorders. The most recent clinical evidence from 2026, including the first study to directly link a specific microbial taxon (Pediococcus) to autonomic dysfunction (reduced cardiac vagal tone) in BED, is incorporated to highlight the frontier of this rapidly evolving field. ¹²

2. BED: Diagnostic framework and neurobiological underpinnings

2.1. Diagnostic criteria and clinical manifestations

According to the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-5), and the International Classification of Diseases, 11th Revision (ICD-11), BED is characterized by recurrent episodes of consuming an amount of food that is markedly larger than most individuals would eat under comparable circumstances, accompanied by a subjective sense of loss of control over eating. Episodes are marked by eating rapidly, eating until uncomfortably full, eating without physical hunger, eating alone due to embarrassment, and experiencing post-binge guilt or self-disgust. ^{2 , 3} Crucially, unlike BN, these episodes are not followed by compensatory purging, fasting, or excessive exercise. ² The diagnosis requires clinically significant distress and that episodes occur on average at least once per week for three months. Severity is graded by episode frequency: mild (1–3/week), moderate (4–7/week), severe (8–13/week), and extreme (≥14/week). ² Despite its high global prevalence, BED remains largely underdiagnosed; ¹³ a survey by Cossrow et al. ¹⁴ found that only 3.2% of individuals meeting DSM-5 criteria had ever been formally diagnosed. Physical comorbidities are both diverse and frequent: obesity (30–45%), hypertension (31%), type 2 diabetes (14%), cardiovascular disease (17%). ⁴ Moreover, critically relevant to microbiota research, upper gastrointestinal symptoms (reflux, bloating, dysphagia and upper abdominal pain) and lower gastrointestinal symptoms (diarrhea, urgency and constipation) were significantly associated with BED. ¹⁵ Psychiatric comorbidities are equally pervasive, with mood disorders occurring in up to 70% of BED individuals, and Major Depressive Disorder (MDD) being the most common comorbidity studied in microbiome contexts. ^{5 , 16}

2.2. Neurobiological mechanisms

The neurobiological architecture of BED involves three converging systems. First, corticostriatal reward circuits are fundamentally dysregulated in BED, with consistent neuroimaging evidence of hyperactivation in the ventral striatum, orbitofrontal cortex (OFC), nucleus accumbens (NAc), and anterior cingulate cortex (ACC) in response to food cues. ¹⁷ These patterns resemble those classically observed in substance use disorders, reflecting dopaminergic reward sensitization and heightened cue-reactivity. Dopaminergic hyperactivation in striatal regions may create an insatiable drive for food, akin to drug craving. ¹⁸ Second, impulsivity and executive dysfunction are core features of BED. Individuals with BED show elevated impulsivity scores, diminished self-control, impaired cognitive flexibility, and reduced gray matter in dorsal and ventral striatum and prefrontal structures. ^19-21 Decreased inhibitory control relates to hypoactivation of the insula, ventromedial prefrontal cortex, and inferior frontal gyrus. These executive deficits facilitate binge initiation and resist extinction once consolidated into habitual behavior patterns driven by stimulus-response conditioning. ^{17 , 22} Third, stress-response system dysregulation is a hallmark of BED. Chronic stress activates the hypothalamic-pituitary-adrenal (HPA) axis, elevates cortisol and catecholamines, and drives consumption of highly palatable foods as a maladaptive coping strategy. ^{23 , 24} HPA axis dysfunction has been extensively documented in BED, and stress directly triggers binge episodes while correlating with their frequency. ²⁵ Critically, the same HPA axis dysregulation also shapes gut microbial composition, providing a biological bridge between psychological stress, microbiota perturbation, and binge behavior. ^{26 , 27}

3. Gut microbiota dysbiosis in BED: State of the art

3.1. Microbial alterations associated with BED

Research into gut microbiota alterations in BED remains in its early stages, yet available evidence reveals consistent, if heterogeneous, patterns of dysbiosis. Current findings suggest a tendency toward decreased alpha diversity, increased Firmicutes and Enterobacteriaceae, in BED. ²⁸ A cross-sectional study comparing 42 obese BED patients with 59 obese non-BED controls using 16S rDNA sequencing found up-regulated Anaerostipes, Roseburia, and Bifidobacterium alongside down-regulated Akkermansia and Intestinimonas in BED. Depletion of Akkermansia, a species strongly linked to gut barrier function and metabolic health, is particularly noteworthy given BED's metabolic comorbidity burden. ⁸ Moreover, Akkermansia muciniphila may contribute to the stimulation of endogenous Glucagon-Like Peptide-1 (GLP-1) secretion through the production of short-chain fatty acids and, potentially, through the secretion of the P9 protein, which has been shown to stimulate GLP-1 release from enteroendocrine L cells via Intercellular Adhesion Molecule 2 (ICAM-2) in preclinical models. ^{29 , 30} A further study of patients engaging in binge eating found reduced Bifidobacterium, Collinsella, Blautia, and Eubacterium hallii group, alongside elevated Prevotella compared to healthy controls. ³¹ Moreover, a pivotal 2023 study using 16S rDNA sequencing and metabolomic analysis in mice with binge-like eating revealed augmented Bacteroidaceae and Lachnospiraceae with diminished Lactobacillaceae and Ruminococcaceae. At genus level, increases in Bacteroides, Roseburia, and Alistipes were accompanied by loss of Lactobacillus and Ruminococcaceae-UCG-014. In the same study, a parallel clinical analysis of 11 BN patients demonstrated significant reduction of Faecalibacterium prausnitzii, a major anti-inflammatory butyrate producer, and this taxon was identified as making the most significant contribution to microbiota differences between binge patients and healthy controls. ³² A very recent and methodologically rigorous human study compared 19 women with BED and comorbid MDD against 38 age- and BMI-matched MDD-only controls, specifically designed to isolate BED-specific microbiota alterations from depression-driven changes. Using differential abundance analysis based on 16S rRNA gene sequencing, the study identified a synchronized depletion of a fermentative guild in BED, including Catenibacterium, Acidaminococcus, Pediococcus, Fusobacterium, Megasphaera, and Prevotella. These taxa span multiple families, Erysipelotrichaceae, Acidaminococcus, Lactobacillaceae, and Fusobacteriaceae, and share a specialized metabolic niche through fermentation of dietary carbohydrates and amino acids into organic acids, including lactate and SCFAs. ¹² Pediococcus strains may support host metabolism by regulating lipid and cholesterol levels and, as lactic acid bacteria, synthesizing B group vitamins including folate. ^{33 , 34} These functions make their depletion potentially relevant to BED, although the effects are strain-specific and have not been demonstrated directly in patients. Human evidence further indicates that binge-eating-related phenotypes are associated with distinct microbial profiles. Achamrah et al. reported reduced gut microbiota richness and distinct taxonomic alterations in patients with typical or atypical BED, intermittent compulsive eating, or Night Eating Syndrome compared with healthy controls and other eating-disorder groups. ³⁵ The BED group showed lower abundances of Akkermansia, Intestinimonas, Oscillibacter, and Sporobacter, together with higher abundances of Actinomyces, Eggerthella, Proteus, and Streptococcus. Enrichment of potentially pro-inflammatory taxa such as Proteus may contribute to the low-grade intestinal inflammation reported in BED. Moreover, Acidaminococcus abundance was inversely associated with anxiety severity, whereas Eggerthella and Streptococcus were positively associated with depressive symptoms. These findings support a distinct BED-associated microbial profile and emphasize the importance of stratifying eating-disorder phenotypes; however, comparisons across studies remain limited by differences in populations, control groups, and psychiatric assessments.

Preclinical studies provide evidence that the gut microbiota can influence eating behavior, rather than merely reflecting it. ^{36 , 37} Experimental microbiota depletion increased spontaneous intake, motivation for high-fat, high-sucrose foods and binge-eating behavior in mice, whereas microbiota restoration reversed these effects. ^{38 , 39} These behavioral changes were accompanied by increased activation of reward-related regions, including the nucleus accumbens and ventral tegmental area, suggesting altered central processing of food reward. ⁴⁰ Members of the S24-7 family within Bacteroidales and Lactobacillus johnsonii, were identified as potential suppressors of excessive palatable food intake. ⁴⁰ Similarly, administration of Bacteroides uniformis CECT-7771 reduced binge eating and anxiety-like behavior in a rat model of intermittent-fasting-induced binge eating, alongside changes in neurotransmitter systems and dopamine-receptor expression in reward- and control-related regions. ⁴¹ Collectively, these findings support a microbiota–brain contribution to compulsive overeating, although the relevance of these specific taxa and mechanisms to human BED remains to be established.

3.2. Proposed mechanisms driving dysbiosis in BED

Distinct mechanisms may contribute to the substantial physiological perturbations that characterize the gastrointestinal environment associated with BED (Figure 1).

Figure 1.

The self-perpetuating pathophysiological cascade involving the bidirectional microbiota-gut-brain axis in Binge Eating Disorder (BED). Gut microbiome dysbiosis is here illustrated by depletion of key fermentative and anti-inflammatory taxa (Faecalibacterium prausnitzii, Pediococcus, Megasphaera) and consequent reductions in short-chain fatty acids (SCFA/butyrate), kynurenic acid (KYNA), and increased circulating lipopolysaccharide (LPS), reflecting functional and metabolic impairment of the microbial ecosystem. These microbial and metabolic changes increase intestinal permeability and immune activation, altering vagal tone and disrupting BED-relevant brain circuits, including the paraventricular thalamus (PVT), prefrontal cortex (PFC), insula (In), nucleus tractus solitarius (NTS), and nucleus accumbens (NAc), which together govern visceral signal integration and satiety, reward processing and motivation, and energy homeostasis and inhibitory control. This circuit disruption manifests behaviorally and autonomically as altered satiety, reward system dysregulation, heightened food preoccupation, stress hypersensitivity, and autonomic dysfunction (including HPA axis dysregulation), collectively increasing the likelihood of a subsequent binge episode and perpetuating the cycle.

Recurrent cycles of dietary restriction and binge eating expose the intestinal ecosystem to large and intermittent nutrient loads, particularly rapidly fermentable carbohydrates, and may thereby alter osmotic conditions, microbial substrate availability, and fermentation dynamics. ^42-44 Intestinal pH is influenced by interacting host factors (including gastric acid and HCO₃ ^- secretion, and intestinal transit), together with diet, microbial production of acid or alkaline metabolites, and the buffering capacity of the luminal environment. ⁴⁵ Although intestinal pH has not been directly characterized in patients with BED, Liang et al. ¹² proposed a plausible ecological mechanism whereby carbohydrate-rich binge episodes could stimulate intense lactic acid production by primary fermenters such as Pediococcus. If the resulting organic-acid production exceeds local buffering and microbial utilization capacity, luminal pH may decrease. Such acidification could inhibit acid-sensitive lactate-utilizing taxa such as Megasphaera, ⁴⁶ which normally convert lactate into weaker short-chain fatty acids, including propionate, butyrate, and valerate. ⁴⁷ Reduced lactate utilization could, in turn, promote further lactate accumulation and disrupt metabolic cross-feeding between primary and secondary fermenters. ³⁴ The synchronized depletion of Pediococcus and Megasphaera observed in BED is therefore consistent with impaired fermentative cooperation and ecological destabilization rather than independent taxonomic losses. However, this proposed acidification–cross-feeding mechanism remains hypothetical and requires direct validation through longitudinal measurements of intestinal pH, microbial metabolites, and community dynamics across binge-eating episodes.

Beyond dietary instability, the elevated affective burden in BED is accompanied by a hyperactivation of the HPA axis and the sympathetic-adrenal-medullary (SAM) system. Chronic stress and elevated cortisol levels can significantly alter gut microbial communities, by reducing diversity and promoting the growth of potentially pathogenic species. ⁴⁸ These stress-induced microbial changes contribute to altered gut barrier function and mucosal immunity. Increased intestinal permeability promotes translocation of lipopolysaccharide (LPS) into the systemic circulation, triggering systemic inflammation and further HPA axis dysregulation. ⁴⁹ This bidirectional relationship suggests that interventions targeting the gut microbiota may have beneficial effects on stress reactivity and HPA axis function. Interestingly, Pediococcus has emerged as a bacterial genus with potential stress-modulating properties. In particular, recent studies have identified Pediococcus acidilactici as a promising psychobiotic candidate, owing to its reported ability to alleviate depressive-like behaviors, mitigate neuroinflammation, and improve autonomic and behavioral outcomes (Figure 1). ^{50 , 51}

Antibiotics and gastrointestinal diseases are also factors able to alter the gut microbiota composition. In one study it was shown that compared to healthy controls, BED and BN patients were prescribed significant higher amounts of antimicrobial medications during the five years preceding the onset of binge eating episodes. Even though the elevated number of infections in the BN and BED groups might be a consequence of poor diet or body weight gain, it cannot be excluded that changes in intestinal microbiota following these frequent antibacterial courses may have contributed to the onset of the eating disorder. ⁵²

4. Mechanistic pathways linking gut microbiota to BED pathophysiology

4.1. Short-chain fatty acids and appetite regulation

Short-chain fatty acids (SCFAs)—principally acetate, propionate, and butyrate—are the primary bacterial fermentation products and represent the most extensively characterized MGBA signaling molecules. SCFAs modulate appetite through G-protein-coupled receptors (FFAR2/GPR43 and FFAR3/GPR41) on enteroendocrine cells and vagal afferent neurons. ⁵³ Through FFAR2 activation, SCFAs stimulate release of peptide YY (PYY) and GLP-1, which potently suppress appetite. Conversely, SCFA activation of FFAR3 (GPR41) in pancreatic β-cells/islets can inhibit glucose-stimulated insulin secretion, highlighting the context-dependent effects of SCFAs on metabolic and energy homeostasis. ⁵⁴ Critically, SCFAs can traverse the blood-brain barrier (BBB) and directly influence hypothalamic appetite circuits: intraperitoneal acetate upregulates anorexigenic pro-opiomelanocortin (POMC) and downregulates agouti-related peptide (AgRP) expression. ⁵⁵ In women with BED and comorbid Major Depressive Disorder (MDD), a recent study reported a coordinated depletion of Catenibacterium, Acidaminococcus, Pediococcus, Prevotella, and Megasphaera, relative to MDD-only controls. ¹² These genera share a fermentative niche, converting carbohydrates and amino acids into lactate and SCFAs. Their loss could therefore reduce SCFA-dependent satiety and vagal signaling. Reduced SCFA production could also lower butyrate availability. Butyrate is a major energy substrate for colonocytes and has anti-inflammatory effects, including through histone deacetylase inhibition. ⁵⁶ Butyrate depletion might therefore contribute to the low-grade inflammatory phenotype reported in some BED cohorts. However, these mechanisms remain hypothetical. SCFA and butyrate levels have not been measured directly in BED, and the inflammatory profile of BED is still inconsistent (see Section 4.5).

4.2. Appetite-regulating peptides and bacterial mimicry

The gut microbiota influences appetite through neuroendocrine peptide axes in ways extending beyond SCFA-mediated signaling. Indeed ghrelin and PYY exhibit altered postprandial responses in BN (and to a lesser extent in BED), contributing to perpetuation of binge eating. ⁵⁷ Another particularly intriguing mechanism involves bacterial molecular mimicry of appetite-regulating neuropeptides. Caseinolytic protease B (ClpB), produced by Escherichia coli and other Enterobacteriaceae, shares conformational homology with α-melanocyte-stimulating hormone (α-MSH), the primary anorexigenic peptide of the arcuate nucleus. ^{58 , 59} Elevated plasma ClpB concentrations have been detected in patients with AN, BN, and BED. ⁶⁰ Moreover, in the blood of these patients autoantibodies (IgG) generated against ClpB can form immunocomplexes with α-MSH, either potentiating or inhibiting its anorexigenic signaling through MC4R depending on immunoglobulin class and concentration, thereby directly dysregulating homeostatic food intake. ^61-63 This immune-mediated modulation of appetite represents a compelling mechanism by which microbiota compositional shifts translate into altered eating behavior in BED, and a potential biomarker for disorder stratification.

4.3. Neurotransmitter pathways: Serotonin and GABA

Serotonin (5-HT) is produced predominantly (95%) by intestinal enterochromaffin cells (ECs), with its synthesis partly regulated by gut microbiota-derived short-chain fatty acids and specific bacteria including spore-forming Clostridia. ⁶⁴ 5-HT suppresses appetite through multiple receptors: 5-HT2cR activation in nucleus tractus solitarius (NTS) reduces food intake, while 5-HTergic projections from the dorsal raphe nucleus (DRN) to the arcuate nucleus (ARH) exert inhibitory control over hunger-driven feeding. ⁶⁵ Activation of 5-HT2cR in dopamine neurons also inhibits binge-like eating in intermittent high-fat diet models. ⁶⁶ In BED patients, diminished tryptophan hydroxylase-2 activity would reduce brain 5-HT synthesis, and acute tryptophan depletion would provoke intensified binge craving, directly implicating serotonergic dysregulation in the neurobiology of binge episodes. ⁶⁷

GABA, another gut microbiota-modulated neurotransmitter produced by Lactobacillus species, regulates gastrointestinal motility and appetite-controlling neuropeptide secretion. ⁶⁸ Centrally, GABA inhibits orexigenic NPY/AgRP neurons in the hypothalamus, and acute activation of GABAergic neurons in the zona incerta induces rapid binge-like eating behaviors in mice. ^{69 , 70} In AN patients, reduced fecal GABA concentrations have been documented, though direct evidence in BED remains limited. ⁷¹ Given that BED-associated dysbiosis depletes Lactobacillus species—key GABA producers—disrupted gut-derived GABAergic signaling may contribute to the disinhibition of appetite circuits in BED.

4.4. The vagal pathway

The gut and the brain can bidirectionally communicate through the autonomic nervous system. The vagus nerve represents the primary parasympathetic link in this circuitry, characterized by a predominance of afferent fibers (80%) relative to its efferent projections (20%). While efferent fibers influence gut physiology by synapsing with enteric neurons to enhance peristalsis, and stimulate gastric secretion, afferent fibers transmit gut-derived signals to brainstem nuclei including the NTS, which projects to higher brain regions governing feeding control, reward, and stress. ⁷² Vagal afferent terminals in the intestinal villi express a diverse array of G-protein-coupled receptors (GPR65, GPR174, GPR35, S1pr3) that respond to microbial metabolites. ⁷³ Kynurenic acid (KYNA), a tryptophan metabolite whose luminal concentration is depleted in BN patients and binge-eating mice, normally binds to NMDA receptors on ileal vagal terminals, tonically inhibiting gut-brain axis activity. KYNA depletion disinhibits the vagus-NTS-paraventricular thalamus (PVT) pathway, inducing hyperactivation of food reward circuits and precipitating binge eating syndrome. ³²

Notably, vagal tone plays a central role in the regulation of both hypothalamic–pituitary–adrenal (HPA) axis activity and peripheral inflammatory responses. ⁷⁴ Neurons within the NTS, the primary central relay for vagal afferent signaling, project to the paraventricular nucleus (PVN) of the hypothalamus, where they modulate activation of the HPA axis and the subsequent release of cortisol. Although cortisol exerts potent immunoregulatory effects, vagal signaling also functions as an intrinsic negative feedback mechanism, with higher vagal tone attenuating excessive HPA axis activation and limiting the detrimental consequences of chronic stress. ⁷² Furthermore, vagal afferent fibers continuously monitor peripheral inflammatory signals and convey this information to the NTS, which integrates these inputs and engages efferent projections originating from the dorsal motor nucleus of the vagus (DMN). Activation of these efferent pathways triggers the release of acetylcholine at peripheral synapses, where it binds to α7 nicotinic acetylcholine receptors (α7nAChRs) expressed on macrophages and other immune cells. ⁷⁵ This mechanism, commonly referred to as the cholinergic anti-inflammatory (CAP) pathway, suppresses NF-κB signaling and consequently reduces the production of pro-inflammatory cytokines, thereby contributing to the maintenance of systemic immune homeostasis. ⁷⁶

The most direct clinical evidence linking specific microbial taxa to vagal dysfunction in BED comes from Liang et al. ¹² . Notably, in patients with comorbid BED and MDD, depletion of Pediococcus was specifically associated with reduced cardiac vagal tone, as measured by heart rate variability (HRV). This association was not detected in individuals with MDD alone, suggesting a BED-specific host–microbiota interaction. The Lactobacillaceae family, to which Pediococcus belongs, is well documented for its capacity to stimulate vagal nerve activity and enhance parasympathetic tone—potentially through SCFA-mediated FFAR3 activation on vagal afferents. ⁷⁷

4.5. Immune and inflammatory pathways

Gut microbial dysbiosis can promote systemic inflammation through multiple immune mechanisms. LPS, the outer membrane component of Gram-negative bacteria, activates Toll-like receptor 4 (TLR4) on enteroendocrine cells and immune cells, triggering IL-1β, IL-6, and TNF-α secretion and activating the HPA axis. ⁷⁸ Butler et al. ⁷⁹ specifically highlights the role of gut dysbiosis, peripheral inflammation, and neuroinflammation in eating disorder pathophysiology. Increased gut permeability, driven by SCFA depletion and LPS-mediated tight junction disruption, facilitates translocation of bacterial products into the systemic circulation and potentially the CNS. Circulating proinflammatory cytokines can traverse the BBB, activating microglia and perpetuating neuroinflammatory states that impair serotonergic, dopaminergic, and HPA axis function. ⁷⁹

Despite the high prevalence of BED, evidence regarding its inflammatory profile remains limited and inconsistent, precluding robust conclusions. Studies investigating circulating pro-inflammatory cytokines, including TNF-α and IL-6, have yielded conflicting findings, and a recent meta-analysis was also unable to establish a clear association because of substantial inter-study heterogeneity. ^80-84 However, reduced levels of the anti-inflammatory cytokine IL-10 have been reported in BED. ⁸¹ Biomarkers related to low-grade systemic inflammation and metabolic dysfunction, such as high-sensitivity C-reactive protein (hs-CRP), appear to be more consistently elevated in individuals with BED or loss-of-control eating, whereas adiponectin levels have been reported to be reduced. ^{82 , 84 , 85} Together, these findings point to a possible low-grade pro-inflammatory state in BED, but the few available studies are heterogeneous and the evidence remains preliminary. Whether this inflammation contributes to psychiatric comorbidity, or could be modified by microbiota-targeted interventions, remains to be determined.

5. Microbiota-targeted therapeutic interventions

5.1. Probiotics, prebiotics, and fecal microbial transplantation

Probiotics (live microorganisms that confer health benefits when administered in adequate amounts) represent the most directly applicable microbiota-based therapeutic strategy for BED. ⁸⁶ Bifidobacterium, Lactobacillus, and Enterococcus species are the primary studied genera, valued for their capacity to synthesize SCFAs, GABA, and 5-HT, stimulate immune regulation, and reinforce intestinal barrier integrity. ⁸⁷ In a 90-day randomized controlled trial (RCT) of 101 patients following bariatric surgery, probiotic supplementation (Lactobacillus + Bifidobacterium) versus placebo significantly reduced binge eating scores and attenuated food addiction symptoms, with benefits persisting at one-year follow-up. ⁸⁸ Of direct mechanistic relevance to BED-specific dysbiosis, oral supplementation of Faecalibacterium prausnitzii in binge-eating-model mice significantly alleviated overeating behaviors, comparable to the effect of oral KYNA administration. ³² Pediococcus acidilactici, depleted in BED patients, has been shown experimentally to alleviate chronic stress-induced depression by reducing neuroinflammation and modulating the gut-brain axis through indole-3-lactic acid metabolite production. ¹² Clinical trials specifically targeting these strains in BED populations represent an urgent research priority. Probiotic supplementation has additionally demonstrated benefits for anxiety and depressive symptoms in systematic reviews and meta-analyzes—directly addressing the psychiatric comorbidities that both characterize and amplify BED. ⁸⁹

Prebiotics (selectively fermented substrates conferring benefit through targeted microbiome modulation) offer a complementary approach by promoting proliferation of endogenous beneficial taxa. ⁹⁰ Inulin and fructooligosaccharides are the most extensively studied, selectively promoting Bifidobacterium and SCFA-producing bacteria while reducing luminal pH and inhibiting potential pathogens. In obesity and metabolic syndrome—conditions frequently comorbid with BED—prebiotic supplementation has demonstrated consistent improvements in appetite regulation, reduced caloric intake, and attenuated GLP-1 responses. ^{91 , 92} The restoration of SCFA-producing capacity through prebiotic fermentation is mechanistically relevant to BED: augmented propionate delivery to the human colon reduces appetite and energy intake in controlled trials, partly through PYY and GLP-1 secretion. ⁹³ No clinical trials have yet specifically investigated prebiotics in BED populations, representing a significant evidence gap. However, the metabolic and gut microbial profile of BED, characterized by depleted fermentative capacity, reduced SCFA production, and impaired satiety signaling, provides a compelling mechanistic rationale for prebiotic intervention as adjunct to established psychotherapy.

Synbiotics combine prebiotics with specific probiotic strains, providing direct fermentation substrates to transplanted beneficial bacteria while supporting endogenous microbial communities. ⁸⁶ The strategic pairing of prebiotic substrates with strains selected for their SCFA-producing, GABA-synthesizing, or vagus-activating properties is theoretically superior to either approach alone, particularly for restoring the functionally inter-dependent fermentative guild observed to be depleted in BED. ⁹⁴ Galactooligosaccharides combined with Bifidobacterium and Lactobacillus strains have demonstrated mood benefits in healthy volunteers, with reductions in cortisol awakening responses and attentional bias to negative stimuli, suggesting HPA axis modulation relevant to BED's stress-response dysregulation. ⁸⁶

Fecal microbiota transplantation (FMT) provides comprehensive microbiome restoration that cannot be achieved through probiotics or prebiotics alone, capitalizing on the functional complexity and diversity of the donor microbiome. Preclinical FMT studies in anorexia nervosa models demonstrate that AN-associated microbiota can transfer disease-relevant eating, metabolic, and behavioral phenotypes to recipient mice, supporting a potential causal contribution of the gut microbiota to eating behavior. ^{95 , 96} For BED specifically, no controlled FMT trials exist to date. The metabolic syndrome context of BED provides partial guidance: FMT has demonstrated improvements in insulin sensitivity and glucose metabolism in overweight individuals, though results are heterogeneous and long-term efficacy data are lacking. ⁹⁷ The prospect of targeting the complete fermentative guild collapse identified in BED—restoring not just single taxa but the inter-dependent consortium of fermenters and their cross-feeding architecture—positions FMT as a theoretically compelling but practically challenging intervention requiring rigorous safety evaluation and standardized protocols before clinical translation in BED (Figure 2).

Figure 2.

Therapeutic targeting of the microbiota-gut-brain axis in Binge Eating Disorder (BED): from established to emerging treatments. Established, evidence-based interventions for BED include psychotherapy (cognitive behavioral therapy, CBT; dialectical behavior therapy, DBT; interpersonal psychotherapy, IPT) and pharmacological therapy (selective serotonin reuptake inhibitors, SSRI; lisdexamfetamine, LDX; glucagon-like peptide-1 receptor agonists, GLP-1). Emerging, microbiota-targeted interventions—prebiotics (e.g., fructo-oligosaccharides (FOS)/inulin), probiotics, synbiotics (e.g., galacto-oligosaccharides (GOS) combined with Bifidobacterium and Lactobacillus), and fecal microbiota transplantation (FMT)—act through proposed mechanisms including increased short-chain fatty acid (SCFA)-producing bacteria and downstream acetate, propionate, and butyrate production, elevated GABA synthesis, improved vagal tone (assessed via heart rate variability, HRV), and reduced peripheral inflammation (IL-6, TNF-α, hs-CRP). Both treatment categories are hypothesized to converge on shared clinical outcomes, including reduced binge frequency, improved body weight and satiety regulation, better autonomic and HPA axis regulation, reduced systemic and central inflammation, and improved metabolic parameters (insulin sensitivity, lipid profile). Given the current scarcity of BED-specific randomized controlled trials, outcomes associated with emerging microbiota-targeted interventions should be interpreted as hypothesis-generating, largely extrapolated from studies in obesity, related eating disorders, and mood disorders.

5.2. Integration with established treatments

Current international guidelines recommend cognitive-behavioral therapy (CBT), dialectical behavior therapy (DBT), and interpersonal psychotherapy (IPT) as first-line treatments for BED. ^98-100 Lisdexamfetamine dimesylate is the only FDA-approved pharmacotherapy. ¹⁰¹ Selective serotonin reuptake inhibitors (SSRIs) reduce binge frequency but efficacy diminishes after 3–6 months. ¹⁰² Moreover, recent clinical investigations identify glucagon-like peptide-1 receptor agonists (GLP-1RAs) as promising treatments to reduce binge eating behavior and improve psychiatric outcomes in BED and bulimia nervosa. ¹⁰³ Microbiota-based interventions are not proposed as replacements for these evidence-based approaches but as mechanistically complementary adjuncts targeting one of the biological substrate (i.e. the gut-brain axis dysfunction) that psychological and pharmacological treatments do not directly address. A multi-target model is emerging: psychological therapy addresses cognitive and behavioral patterns; pharmacotherapy modulates monoaminergic and dopaminergic circuits; and microbiota interventions restore the SCFA-vagus axis, reduce neuroinflammation, and normalize satiety signaling. The overlap between GLP-1 receptor agonist pharmacology and microbiota-derived GLP-1 stimulation represents a particularly promising convergence point: GLP-1 agonists reduce binge eating behavior and craving, while prebiotic- and probiotic-driven GLP-1 secretion from enteroendocrine L-cells may amplify and sustain these effects through physiological rather than pharmacological routes (Figure 2).

6. Sex differences and microbiota interactions

Lifetime prevalence of BED in Western studies ranges from 0.6–6.1% in women and 0.3–0.7% in men, with a female-to-male ratio of roughly 2:1 to 4:1 in population-based estimates. ¹

Sex steroid hormones exert profound effects on both eating pathology and microbiota composition. Estradiol directly reduces appetite through modulation of arcuate nucleus circuits, while testosterone promotes protective effects against binge-like eating. ^104-106 Moreover, estrogen and the gut microbiota are linked through a bidirectional relationship. On one hand, estrogens influence gut microbial composition and intestinal barrier function through their immunomodulatory effects on immune cell populations, including dendritic cells and B lymphocytes. ¹⁰⁷ On the other hand, the gut microbiota contributes to the maintenance of systemic estrogen homeostasis by regulating enterohepatic estrogen recycling. Following hepatic glucuronidation, biologically active estrogens are excreted into the intestinal lumen as inactive conjugates. There, bacterial β-glucuronidases, produced by members of genera such as Clostridium, Escherichia, and Bacteroides, hydrolyze these conjugates, regenerating free estrogens that can be reabsorbed into the circulation. Consequently, alterations in the abundance or activity of β-glucuronidase-producing bacteria may disrupt the balance between estrogen excretion and reabsorption, thereby contributing to dysregulated estrogen homeostasis and the development of estrogen-dependent disorders. ¹⁰⁸

In sum, whether sex-dependent host–microbiota interactions contribute to the sex disparity in BED prevalence remains an open question. This hypothesis is biologically plausible, given the bidirectional links between sex steroids and the gut microbiota, but it has not yet been tested directly in BED. Sex-stratified analyzes in future microbiome studies would help address it and could inform precision-medicine strategies.

7. Research gaps, limitations, and future directions

Several critical limitations constrain current understanding. First, most human microbiome studies in BED are cross-sectional, precluding causal inference. The temporal question, whether dysbiosis precedes and contributes to BED onset, or whether binge-restrict cycles drive dysbiosis, remains unresolved. Longitudinal cohort studies with pre-disorder microbiome sampling and prospective behavioral monitoring are essential. Second, confounding variables including obesity, psychiatric comorbidities (especially MDD), medication use (antibiotic exposure is notably elevated in pre-BED onset), dietary patterns, and BMI independently shape microbial landscapes, complicating interpretation of cross-sectional comparisons.

Third, sample sizes in existing BED microbiome studies are uniformly modest: the Liang et al. ¹² study involved only 19 BED participants, necessitating careful interpretation of effect sizes. The ongoing Binge Eating Genetics Initiative (BEGIN), aiming to recruit 1,000 BED/BN participants for integrated gut microbiota, genomic, and behavioral analysis, promises to provide substantially greater statistical power. Fourth, mechanistic studies directly linking specific metabolites (SCFAs, KYNA, LPS) to behavioral outcomes in human BED are lacking; most mechanistic evidence derives from rodent models or BED comorbid human conditions (obesity, depression). Future research priorities include: (1) longitudinal designs tracking microbiome-behavior-autonomic function relationships across binge-restrict cycles; (2) multi-omics integration combining 16S/shotgun metagenomics with metabolomics (fecal SCFA, KYNA), metatranscriptomics, and autonomic phenotyping (HRV) to capture mechanistically relevant functional states; (3) targeted Randomized Controlled Trials (RCTs) of specific probiotic strains implicated by mechanistic data (e.g. F. prausnitzii, Pediococcus acidilactici, and SCFA-producing consortia) as adjuncts to cognitive behavioral therapy (CBT) and dialectical behavior therapy (DBT); (4) investigation of sex-stratified microbiome effects; and (5) standardized gut microbiome assessment protocols enabling cross-study comparison and meta-analysis.

8. Conclusions

BED is a neurobiologically complex disorder in which the gut microbiota-brain axis plays an increasingly well-characterized but still incompletely understood role. The current body of evidence identifies consistent patterns of BED-associated dysbiosis (among which depletion of Faecalibacterium prausnitzii, Akkermansia, and a broader fermentative guild encompassing Pediococcus, Megasphaera, and Catenibacterium) that mechanistically converge on disrupted SCFA production, impaired vagal tone, altered neurotransmitter signaling, and heightened neuroinflammation. The discovery of a BED-specific link between Pediococcus depletion and reduced cardiac vagal tone, absent in MDD controls, represents a significant conceptual advance, identifying the SCFA-vagus axis as a novel, potentially modifiable neurobiological pathway in BED pathophysiology.

Microbiota-targeted interventions—probiotics, prebiotics, synbiotics, and FMT—demonstrate mechanistic rationale and promising preliminary evidence for reducing binge eating frequency and associated comorbidities. However, clinical translation requires rigorous RCTs with standardized microbiome profiling, appropriate control groups matched for depression and BMI, and outcome measures capturing both eating behavior and autonomic/metabolic endpoints. Integration of microbiota-based approaches as adjuncts to CBT, DBT, or pharmacotherapy represents the most clinically tractable immediate strategy. Realizing the full therapeutic potential of microbiota-directed interventions in BED will require coordinated effort across basic microbiology, clinical psychiatry, nutritional science, and regulatory frameworks—an ambitious but achievable goal given the accelerating pace of MGBA research.

Funding Statement

This work was supported by Swiss National Science Foundation grants (310030_185192 and 320030_215615 to BB), Fondation Donase and the Faculty of Biology and Medicine, Lausanne University Hospital, University of Lausanne. Perplexity AI was used to assist with literature synthesis and manuscript restructuring during the preparation of the revised version of the manuscript. The authors reviewed, verified, and edited all AI-assisted content and take full responsibility for the accuracy and integrity of the final manuscript. The figures were created in BioRender.

Disclosure of potential conflicts of interest

No potential conflicts of interest were disclosed.