Glycosylation in gut-brain communication: from the intestinal barrier and microbiota to immune and neural signaling

Abstract

The gut-brain axis is a complex bidirectional communication network linking the gastrointestinal tract and the central nervous system. Its dysregulation is closely associated with a wide range of gastrointestinal, metabolic, and neuropsychiatric disorders. Glycosylation, one of the most prevalent and structurally diverse post-translational modifications of proteins and lipids, is increasingly recognized as a regulator of multiple processes within the gut-brain axis. In this review, we summarize evidence that glycosylation influences several steps of gut-brain communication, including intestinal barrier function, microbial glycan metabolism, immune signaling, enteric and vagal pathways, blood-brain barrier integrity, and neural responses. We distinguish direct mechanistic findings from associative observations and discuss the more limited evidence for brain-to-gut regulation of intestinal glycosylation. In addition, we discuss the potential of glycosylation-targeted nutritional and microbiota-based interventions and highlight emerging opportunities to integrate glycomics with other omics approaches to dissect the complex regulatory networks underlying the gut-brain axis. In conclusion, elucidating how glycosylation shapes signaling along the gut-brain axis may open new avenues for understanding disease pathogenesis and for developing targeted therapeutic strategies.

Graphical abstract

Glycosylation in gut-brain signaling. Host glycans, including mucin O-glycans and human milk oligosaccharides (HMOs), shape microbial glycan utilization, colonization, and cross-feeding within the intestinal ecosystem. These processes generate or release gut-derived signals, including short-chain fatty acids (SCFAs), 1,2-propanediol (1,2-PD), and L-fucose. Circulating metabolites, immune signaling, and vagal pathways may connect glycan-dependent intestinal processes with brain responses, including blood-brain barrier (BBB) integrity, neuroinflammation, and microbiota-associated glycoproteome remodeling. Brain states, particularly stress-related signals, may also influence intestinal glycosylation, although evidence for this brain-to-gut direction remains comparatively limited. The diagram summarizes representative and partly independent processes rather than a single established sequential pathway. Created with BioRender.com.

1. Introduction: glycosylation in the gut-brain axis

Glycosylation is an enzyme-catalyzed covalent attachment of glycans to proteins or lipids. As a major post-translational modification, it occurs primarily in the endoplasmic reticulum and Golgi apparatus, yielding a wide range of glycoproteins and glycolipids. By modulating protein folding, stability, and molecular interactions, glycosylation is involved in fundamental biological processes, including cell recognition, signal transduction, and immune responses. ^{1 - 3} At the structural and biosynthetic levels, glycosylation is mainly classified into N-linked glycosylation, which occurs on asparagine residues, and O-linked glycosylation, which modifies serine or threonine residues. During biosynthesis, glycans can undergo further elongation and terminal modifications, such as sialylation and fucosylation, generating a high degree of structural diversity. This finely tuned and dynamic process is orchestrated by the coordinated actions of highly specific glycosyltransferases and glycosidases. ^{4 - 6} The gut-brain axis encompasses complex interactions among the intestinal epithelial barrier, the gut microbiota, local and systemic immune systems, and the central nervous system. Accumulating evidence suggests that glycosylation influences host-microbe interactions and modulates immune and neuro-related signaling pathways, indicating that glycosylation likely contributes to signal regulation along the gut-brain axis through multiple interacting pathways. ^{1 , 7}

In the gut-brain axis, glycosylation regulates signaling at multiple levels, with the intestine serving as a primary site of action. Highly O-glycosylated mucins secreted by intestinal goblet cells form the mucus layer, a key physical barrier that separates luminal microbes from the epithelium. ^{8 - 10} In parallel, glycosylation of intestinal epithelial cells is closely associated with barrier permeability and selectivity. ^{11 - 14} Beyond its structural role, mucin glycosylation contributes to intestinal homeostasis by regulating microbial colonization and metabolic activity. ^{15 - 17} Glycosylated structures, such as mucin O-glycans and human milk oligosaccharides (HMOs), not only reinforce the barrier but also serve as major nutrient sources and adhesion substrates for gut microbes, thereby shaping the composition and function of the intestinal ecosystem. In turn, gut microorganisms express specific glycosidases to degrade these glycans, generating signaling metabolites including short-chain fatty acids (SCFAs) and secondary bile acids. These microbe-derived molecules may influence brain function either directly or indirectly through immune, endocrine, and vagal pathways. ^{18 - 20} In addition, emerging evidence suggests that central nervous system states, such as psychological stress, can reciprocally influence intestinal glycosylation and microbial metabolic environments, highlighting a bidirectional regulatory component within the gut-brain axis. ²¹ To provide an integrated overview of these bidirectional interactions, Figure 1 summarizes representative glycosylation-related processes involved in bidirectional gut-brain communication.

Figure 1.

Schematic overview of glycosylation-related processes in gut-brain communication. In the intestine, goblet cells secrete heavily O-glycosylated mucins to form the mucus barrier, which serves both as a physical shield and as a major source of glycans for commensal bacteria. Host-derived glycans, including mucin O-glycans, shape microbial composition and function by acting as nutrient sources and adhesion substrates. Microbial metabolism of these glycans generates a diverse repertoire of bioactive metabolites, such as short-chain fatty acids (SCFAs), secondary bile acids, and neurotransmitter precursors (e.g., tryptamine and serotonin precursors), which signal to the host through immune, endocrine and neural pathways. At the brain interface, glycosylation plays a critical role in maintaining blood-brain barrier (BBB) integrity. The endothelial glycocalyx, composed of proteoglycans, glycoproteins and associated glycans, regulates vascular permeability and mediates interactions with perivascular cells, including astrocytes, pericytes and microglia. Although not depicted in detail in this schematic, glycosylation broadly modulates neuronal development, synaptic function and neural plasticity. Together, these processes illustrate how glycosylation can contribute to interactions among the gut microbiota, immune system, and central nervous system. The dashed arrow from the brain to the intestine indicates potential brain-to-gut influences (e.g., in response to stress) rather than direct structural control. Created with BioRender.com.

At the brain end of the gut-brain axis, glycosylation likewise plays important roles in multiple key physiological processes. These include maintenance of blood-brain barrier (BBB) integrity through the endothelial glycocalyx, which is primarily composed of proteoglycans and highly glycosylated mucin-like molecules, ²² as well as regulation of neurodevelopment, synaptic plasticity, and learning and memory via specific glycosylated proteins such as polysialylated neural cell adhesion molecule (PSA-NCAM). ^{23 , 24} In addition, glycosylation of neurotransmitter receptors contributes to their membrane localization and signal transduction, underscoring the widespread and functionally relevant presence of glycosylation in the central nervous system. ^{24 , 25} Advances in high-resolution mass spectrometry, particularly when integrated with artificial intelligence and multi-omics strategies, have provided essential technical foundations for systematically dissecting the regulatory roles of glycosylation within the gut-brain axis. ^{1 , 26} Despite these advances, whether and how the gut microbiota directly regulates protein glycosylation in distal brain regions has long remained unclear. Previous studies have largely focused on indirect signaling mechanisms mediated by microbial metabolites, such as SCFAs. ^{19 , 20} Recently, a high-resolution glycoproteomics study employing a newly developed deep quantitative glycoproteome (DQGlyco) approach systematically profiled glycosylation in the adult mouse brain. This analysis revealed that specific gut microbial communities can remodel a large number of site-specific N-glycosylation events across the brain, with changes that cannot be explained solely by alterations in protein abundance and that are preferentially enriched in pathways related to axon guidance and neurotransmitter signaling. ²⁷ Together, these findings suggest that the gut microbiota may serve as an extrinsic regulator of the brain glyco-code, offering a new perspective on the molecular basis of gut-brain communication.

Current evidence indicates that glycosylation can modify several steps of gut-brain communication. In the intestine, glycan structures regulate mucus organization, microbial access to host substrates, and the production of microbial metabolites. These changes may influence the brain through circulating immune and metabolic signals or through enteric and vagal pathways. In the brain, glycosylation regulates BBB properties and neural responses, although the links between many brain-intrinsic glycosylation mechanisms and intestinal events remain uncertain. From this perspective, glycan structures and glycosylation enzymes are examined as molecular regulators across barrier, microbial, immune, and neural processes, complementing microbiota- and immune-centered models of gut-brain communication. Accordingly, this review distinguishes direct mechanistic evidence from associative observations and focuses on the routes linking glycosylation-related changes in the gut to neural outcomes.

2. Glycosylation and intestinal barrier homeostasis: molecular basis of structure and defense

The intestinal barrier is an early site at which glycosylation can influence gut-brain communication. By maintaining mucus organization and epithelial integrity, glycosylation limits the translocation of microbial products and regulates systemic exposure to gut-derived metabolites and inflammatory mediators. This chapter first describes the local barrier functions of glycosylation and then evaluates how barrier disruption has been linked to distal brain alterations.

2.1. Mucin glycosylation in the maintenance of the intestinal mucus barrier

Mucins secreted by intestinal goblet cells, such as mucin 2 (MUC2), are the principal components of the intestinal mucus layer. Dense O-linked glycosylation on serine and threonine residues is essential for proper mucin folding, elongation, and the formation of a highly ordered three-dimensional architecture. ²⁸ Beyond shaping the physical properties of the mucus layer, these glycan modifications directly contribute to its defensive function. Specifically, site-specific O-glycans can protect the MUC2 protein backbone from cleavage by bacterial proteases, thereby enhancing the mucus layer’s resistance to microbial enzymatic degradation. ²⁹

From a biosynthetic perspective, mucin O-glycans are structurally heterogeneous and can be initiated from distinct core structures (cores 1-4), conferring substantial molecular diversity to the mucus layer. Among these, core 3- and core 4-derived O-glycans are highly enriched in the intestinal epithelium, particularly in the colon, where they exhibit pronounced tissue specificity. ³⁰ This distribution suggests that specific core structures may have specialized roles in maintaining colonic mucus barrier function. Core 3 O-glycans are synthesized by core 3 β1,3-N-acetylglucosaminyltransferase (C3GnT), encoded by B3GNT6. Genetic studies have shown that loss of C3GnT markedly alters the O-glycosylation profile of colonic mucins and disrupts the supramolecular organization of the mucus layer. Accordingly, C3GnT^-/^- mice display a thinner and disorganized colonic mucus layer with increased intestinal permeability, exhibit heightened susceptibility to pathogenic bacterial invasion, and show increased vulnerability in chemically induced models of colitis and colorectal cancer. ¹¹ While O-glycosylation is the predominant modification shaping mucin structure, N-glycan modifications also contribute to mucin regulation. For instance, core fucosylation mediated by α1,6-fucosyltransferase (FUT8) has been shown to control both the quantity and composition of secreted mucins, thereby influencing mucus viscoelasticity and epithelial barrier properties in the context of intestinal inflammation. ³¹

Together, these studies demonstrate that both O-glycan core structures and N-glycan modifications contribute to the physicochemical properties of mucins. These glycosylation-dependent features are essential for maintaining the structural integrity and protective function of the intestinal mucus barrier, highlighting glycosylation as a critical component of the gut’s first line of defense.

2.2. Regulatory roles of fucosylation in intestinal epithelial stress responses

Glycosylation in intestinal epithelial cells is not merely a structural feature but also a highly dynamic, stress-responsive system in which fucosylation plays a prominent regulatory role. α1,2-Fucosyltransferase (FUT2) catalyzes the addition of terminal fucose residues to glycans on intestinal epithelial cells and secreted mucins. ³² Both the expression and activity of FUT2 exhibit marked plasticity during inflammation, infection, or physical injury, enabling fucosylation to contribute to local barrier protection and repair in response to specific immune and metabolic cues.

For example, group 3 innate lymphoid cells secrete interleukin-22 (IL-22) in response to tissue damage, which directly induces Fut2 expression in intestinal epithelial cells and promotes epithelial fucosylation. ¹⁴ Dietary factors can also engage this regulatory axis. Polysaccharides from Lentinula edodes activate the dendritic cell-associated C-type lectin-1 receptor on macrophages, stimulate IL-22 production, and thereby indirectly drive epithelial fucosylation to accelerate barrier repair. ³³ In addition, exogenous L-fucose supplementation further enhances FUT2-mediated fucosylation, alleviates dextran sulfate sodium-induced colitis, and preserves tight junction integrity. ¹² Mechanistically, these effects involve activation of the aryl hydrocarbon receptor (AhR)/IL-22 pathway to promote intestinal stem cell proliferation and differentiation, ³⁴ as well as stabilization of FUT2-dependent fucosylation of hypoxia upregulated protein 1, protecting it from aberrant degradation under inflammatory conditions. ¹³

Beyond local inflammatory and dietary cues, intestinal epithelial fucosylation can also respond to psychological stress. Chronic social defeat stress (CSDS) significantly reduces Fut1 and Fut2 expression and decreases α1,2-fucosylation in the distal small intestine, ²¹ indicating that host stress states can remodel the intestinal glycan environment. Large-scale genetic studies have also identified FUT2 as a shared susceptibility locus for gastrointestinal and neuropsychiatric disorders. ³⁵ Although these genetic associations do not establish the direction or mechanism of gut-brain communication, they support the relevance of FUT2 to phenotypes involving both intestinal and neural systems.

These findings indicate that FUT2-mediated α1,2-fucosylation in intestinal epithelial cells is dynamically regulated by local inflammatory, infectious, physical, and dietary stimuli, and can also be altered by psychological stress. Local intestinal stimuli regulate FUT2 expression through relatively well-defined immune and metabolic pathways, whereas the upstream mediators of stress-associated changes in intestinal fucosylation remain less clear. Together, these observations suggest that epithelial fucosylation may link intestinal barrier adaptation with broader host stress responses relevant to gut-brain communication. Future studies should determine how these distinct signals converge on FUT2 transcription and identify the specific fucosylated proteins that mediate barrier-protective effects. Prospective human studies are also needed to determine whether loss-of-function FUT2 variants modify the relationships among psychological stress, intestinal glycosylation, microbial metabolism, and neuropsychiatric symptoms.

2.3. Cross-barrier signaling and distal effects: linking intestinal barrier dysfunction to brain alterations

The glycosylation status of the intestinal barrier is closely associated with local homeostasis and may be linked to alterations in distal organs, including the brain. Evidence for this comes from C3GnT^-/^- mouse models. These mice display pronounced defects in intestinal barrier integrity and marked alterations in the microbial metabolite profiles of cecal contents. Several of the altered metabolites, including dimethylglycine and N-acetyltyrosine, are known to be associated with brain function and neurotransmitter metabolism. ¹¹ In addition to metabolic changes, structural alterations are also observed in the brain. In the dentate gyrus of the hippocampus, a brain region closely linked to learning and memory, C3GnT^-/^- mice display abnormal morphology of PSA-NCAM-positive immature granule neurons (Figure 2). These cells are characterized by shortened and disorganized dendritic projections, together with a trend toward reduced cell numbers. These alterations are accompanied by impaired object recognition memory. ¹¹

Figure 2.

Core 3 O-glycans regulate mucin architecture and may contribute to barrier-mediated gut-brain signaling. Under homeostatic conditions, core 3 O-glycans support the structural organization of mucins secreted by goblet cells, enabling the formation of a dense and stratified mucus layer that spatially separates luminal bacteria from the intestinal epithelium. This organized mucin architecture limits microbial contact with epithelial cells and maintains barrier integrity. Loss of core 3 O-glycans disrupts mucin organization, resulting in a thinner and less structured mucus layer that permits bacterial encroachment toward the epithelial surface. This compromised barrier is associated with increased epithelial permeability and altered luminal metabolite profiles, including molecules such as dimethylglycine and N-acetyltyrosine. Changes in the barrier are associated with alterations in the brain, particularly in the dentate gyrus of the hippocampus. Under normal conditions, polysialylated neural cell adhesion molecule (PSA-NCAM)⁺ immature neurons display well-developed dendritic structures and balanced distribution. In contrast, barrier disruption is associated with reduced neuronal number and simplified, disorganized dendritic morphology. This schematic is adapted from findings in C3GnT^-/^- murine models ¹¹ . Created with BioRender.com.

Together, these observations support a potential link between intestinal glycosylation defects and a “leaky gut” phenotype. Compromised barrier function facilitates the translocation of microbes and their metabolites across the epithelial layer, which may amplify local and systemic low-grade inflammation. Barrier disruption may increase the systemic exposure to microbial products, metabolites, and inflammatory mediators. These signals may influence the central nervous system through circulating immune and metabolic pathways or through neural routes such as vagal signaling, although the relative contribution of each route has not been resolved in this model. ³⁶ Such mechanisms, initially defined in animal models, also provide a plausible explanation for clinical observations that inflammatory bowel disease (IBD) is frequently accompanied by anxiety and depressive symptoms, as well as for gastrointestinal dysfunctions that precede the onset of neurodegenerative disorders.

In this context, intestinal glycosylation provides a molecular basis for host-microbiota interactions and may contribute to the association between barrier dysfunction and brain alterations along the gut-brain axis. Intestinal glycosylation is also sensitive to systemic metabolic conditions. Hyperglycemic conditions suppress global N-glycan biosynthetic pathways in intestinal epithelial cells, impair barrier integrity, and are accompanied by enhanced systemic and central inflammatory responses. ³⁷ In contrast, the plant polyphenol-derived metabolite urolithin A restores epithelial N-glycosylation, improves intestinal barrier function, and concomitantly alleviates neuroinflammation and cognitive impairment. ³⁷ However, it remains unclear whether these effects are mediated indirectly through the gut or reflect direct systemic actions of urolithin A on the brain. These findings suggest that targeting glycosylation pathways may be a promising strategy for ameliorating gut-brain axis dysfunction associated with metabolic diseases.

The key unresolved issue is whether circulating metabolites or inflammatory signals alone are sufficient to drive the hippocampal changes observed in this model. It will also be important to define the specific routes by which local intestinal inflammatory signals reach the brain, including immune cell trafficking, vagus nerve signaling, and other humoral mechanisms. In addition, experimental models will need to test whether these gut-brain axis pathways are causal. In particular, intestine-specific conditional knockout models of key glycosylation enzymes will be essential to distinguish local barrier-driven effects from systemic or developmental confounders and to establish causal links along the gut-brain axis. Complementary approaches, such as supplementation with specific metabolites or anti-inflammatory interventions, may further help to evaluate the functional relevance of these pathways. These strategies will be important for determining whether targeting these mechanisms has therapeutic potential.

3. Glycosylation and the gut microbiota in gut-brain axis regulation: mechanistic and functional bases

The gut microbiota is a core functional component of the gut-brain axis, while host-derived glycan structures generated through glycosylation form a key molecular interface for host-microbe interactions. These glycans not only serve as nutrient sources and adhesion substrates for intestinal microbes but also shape microbial community structure and immune tone by selectively supporting or restricting specific taxa.

3.1. Roles of host glycans in microbial nutrient utilization and intestinal colonization

Host-derived glycans, particularly mucin O-glycans and nondigestible dietary glycans such as HMOs, serve as major carbon sources for gut microbes, especially obligate anaerobic commensals. ^{17 , 38} These complex glycans are largely resistant to host digestion but can be efficiently metabolized by specific bacterial taxa, thereby establishing highly differentiated nutritional niches within the intestinal ecosystem.

Terminal monosaccharide residues on mucin glycans, including fucose, sialic acid, and N-acetylglucosamine, are common targets for bacterial glycosidases. These enzymes, such as fucosidases and sialidases, can be expressed on the bacterial cell surface or secreted into the mucus layer. ^{39 , 40} The released sugars are subsequently used as carbon, nitrogen, and energy sources. Importantly, the capacity to utilize host glycans varies substantially across bacterial taxa and even among strains, and this metabolic heterogeneity is thought to underlie microbial diversity and community stability in the gut. ^{17 , 30} Foundational studies in Bacteroides showed that host-glycan utilization is coordinated through polysaccharide utilization loci, which mediate substrate-specific glycan sensing and catabolism. In Bacteroides thetaiotaomicron, mucin O-glycans are the principal host-derived substrates used in vivo, and simultaneous deletion of five extracytoplasmic function sigma regulators involved in mucin O-glycan utilization impaired intestinal persistence and mother-to-offspring transmission. ⁴¹ These findings link mucosal glycan foraging not only to nutrient acquisition but also to microbial fitness and colonization. Glycan-dependent colonization also involves bacterial surface modification, as loss of surface L-fucosylation in Bacteroides fragilis impaired competitive intestinal colonization. ⁴²

In parallel, certain Ruminococcus gnavus strains have evolved specialized strategies for mucin-glycan utilization. For example, the ATCC 29149 strain harbors a unique sialic acid catabolic pathway. This pathway enables the bacterium to grow on sialylated substrates and supports the production of propionate and other fermentation end products when grown on mucin or fucosylated glycans. ⁴⁰ In contrast, the infant-associated Bifidobacterium longum subsp. infantis contains a highly organized gene cluster that encodes multiple glycosidases and transporters dedicated to the uptake and metabolism of complex HMOs. These features confer a competitive advantage in the intestines of breastfed infants. ⁴³ Beyond their nutritional roles, glycan structures also contribute to bacterial colonization and spatial organization within the mucus layer. In the case of R. gnavus, sialic acid utilization directly influences its fitness and competitiveness in the mucus niche, thereby shaping its spatial distribution along the intestinal epithelium. ⁴⁴

Host genetic variability in intestinal glycosylation further modulates these interactions. For instance, loss-of-function variants in the FUT2 gene give rise to the “non-secretor” phenotype, characterized by reduced α1,2-fucosylation in epithelial cells and mucus. This reduction limits the availability of host-derived fucose, alters microbial fucose metabolism, and is associated with increased susceptibility to intestinal inflammation. ⁴⁵ Notably, this ecological shaping is not driven exclusively by the host. Studies in germ-free mice mono-colonized with R. gnavus ATCC 29149 have shown detectable changes in the sialylation patterns of intestinal mucins, ²³ suggesting that microbial metabolic activity can, in turn, influence local host glycosylation. Similar bidirectional regulation has been demonstrated in B. thetaiotaomicron. Its fucose-responsive regulator FucR coordinates bacterial L-fucose catabolism with regulation of a separate bacterial locus involved in inducing fucosylated glycan production by intestinal epithelial cells. ⁴⁶ In gnotobiotic rats, B. thetaiotaomicron also increased goblet-cell differentiation and the ratio of sialylated to sulfated mucins, whereas co-colonization with Faecalibacterium prausnitzii attenuated these effects. ⁴⁷ These bidirectional interactions are critical for maintaining gut microbial homeostasis and, when perturbed, may predispose the host to dysbiosis and intestinal inflammation.

Taken together, host glycans shape the composition and metabolic output of the gut microbial community by providing both nutritional resources and colonization niches. Microbial utilization of HMOs and mucin O-glycans supports cross-feeding and microbial colonization and generates metabolites and other gut-derived products with potential relevance to brain function. However, the specific microbial products and host pathways linking these intestinal processes to neural outcomes remain incompletely defined (Figure 3). Accordingly, elucidating the dynamic interplay between host glycans and the gut microbiota is essential not only for understanding intestinal homeostasis but also for defining how glycan-dependent microbial activity contributes to gut-brain communication.

Figure 3.

Host glycan utilization, microbial colonization, and gut-derived signals relevant to gut-brain communication. Human milk oligosaccharides (HMOs) and mucin O-glycans provide selective substrates and colonization niches for specialized intestinal bacteria. Bifidobacterium longum subsp. infantis metabolizes complex HMOs, releasing monosaccharides that support cross-feeding among community members. Selected Ruminococcus gnavus strains utilize mucin glycans, whereas Bacteroides thetaiotaomicron senses and catabolizes mucin O-glycans through polysaccharide utilization loci (PULs). HMOs also exert prebiotic and anti-adhesive effects, and these glycan-dependent interactions contribute to microbial colonization within the mucus layer. Representative structures shown are 2′-fucosyllactose [2′-FL; Fucα1-2Galβ1-4Glc], 6′-sialyllactose [6′-SL; Neu5Acα2-6Galβ1-4Glc], core 1 O-glycan [Galβ1-3GalNAcα1-O-Ser/Thr], and core 3 O-glycan [GlcNAcβ1-3GalNAcα1-O-Ser/Thr]. Microbial processing of host glycans generates or releases short-chain fatty acids (SCFAs), 1,2-propanediol (1,2-PD), and L-fucose, which may contribute to systemic metabolic signaling relevant to the brain. Defined gut microbiota can also remodel the brain N-glycoproteome, although the responsible microbial products and transmission pathways remain unclear. The dashed arrow indicates an incompletely resolved brain-directed route. Glycan structures are depicted using Symbol Nomenclature for Glycans (SNFG). Glc, glucose; Gal, galactose; Fuc, fucose; Neu5Ac, N-acetylneuraminic acid; GalNAc, N-acetylgalactosamine; GlcNAc, N-acetylglucosamine. Created with BioRender.com.

3.2. HMOs: prebiotics, anti-adhesive agents, and precursors of gut-brain signaling molecules

HMOs are a well-characterized example of how host glycosylation shapes the gut microbiota and illustrate how a glyco-code operates during early-life development. HMOs exert multiple, functionally distinct effects on the infant gut ecosystem.

First, as highly selective prebiotics, HMOs promote the growth of beneficial bacteria, including Bifidobacterium and Lactobacillus species. For instance, 2′-fucosyllactose (2′-FL) upregulates adhesion-related genes in Bifidobacterium, thereby enhancing bacterial colonization capacity. ⁴⁸ Moreover, distinct Bifidobacterium strains have evolved cooperative strategies that enable the sharing of intermediate products generated during HMO degradation, resulting in a metabolic division of labor within the microbial community. ³⁸ Second, HMOs function as effective anti-adhesive agents by acting as molecular decoys. By mimicking host epithelial glycan receptors, such as histo-blood group antigens, HMOs competitively inhibit the attachment of enteric pathogens, including rotavirus and Campylobacter jejuni, thereby providing a form of innate antimicrobial protection. ^{49 - 51} Beyond these roles, HMOs serve as key precursors for the generation of gut-derived signaling molecules relevant to gut-brain communication. During microbial fermentation of HMOs, particularly by Bifidobacterium species, monosaccharides such as L-fucose are released and subsequently metabolized through cross-feeding interactions. These processes ultimately give rise to SCFAs, including acetate, propionate, and butyrate, as well as 1,2-propanediol (1,2-PD). ^{52 - 54} Mucus-degrading bacteria, such as Akkermansia muciniphila, can also utilize 2′-FL to liberate fucose from glycan substrates. ⁵⁵

These microbial metabolites are biologically active. SCFAs, particularly butyrate, serve as a primary energy source for colonic epithelial cells and modulate immune balance and barrier integrity through mechanisms that include histone deacetylase inhibition and activation of G protein-coupled receptors. ⁵⁶ In contrast, 1,2-PD can be absorbed into the portal circulation, is detectable in both blood and brain tissue, and may function directly as a gut-brain signaling molecule. ¹⁹ Notably, defined metabolic profiles generated by infant-associated Bifidobacterium strains during 2′-FL fermentation have been reported to increase the release of brain-derived neurotrophic factor (BDNF) in vitro systems. ²⁰

These findings identify HMOs and their microbial products as candidate gut-derived signals with potential relevance to neurodevelopment. However, the extent to which their effects are mediated by circulating metabolites or by enteroendocrine, enteric, and vagal pathways remains incompletely defined.

3.3. Enteric nervous system and vagal pathways in glycan-related gut-brain signaling

Studies of intact fucosylated HMOs provide experimental evidence for acute, structure-related modulation of neuronally dependent intestinal motor activity. In an ex vivo mouse colon preparation, 2′-FL and 3′-fucosyllactose reduced neuronally dependent migrating motor contractions in a concentration-dependent manner, whereas 3′-sialyllactose (3′-SL), 6′-sialyllactose, lacto-N-neotetraose, and galactooligosaccharides showed no detectable effects. Free L-fucose also reduced contractility but was substantially less effective than the intact fucosylated HMOs, consistent with a contribution of the intact oligosaccharide context to their greater bioactivity. ⁵⁷ In a separate restraint-stress model, intraluminal 2′-FL partly normalized stress-induced abnormalities in propagating contractile activity in isolated jejunal and colonic segments. After neuronal activity was silenced with tetrodotoxin, 2′-FL had no significant effect on the velocity or frequency of the residual myogenic contractions, supporting an enteric nervous system (ENS)-dependent effect rather than a direct action on intestinal smooth muscle. ⁵⁸ The rapid responses observed in these isolated preparations argue against longer-term prebiotic effects on microbial growth as the primary explanation. Nevertheless, neither study identified the relevant intestinal sensor or examined whether these enteric neural effects were linked to brain outcomes.

A dietary intervention study in rats examined the contribution of vagal integrity to the effects of 2′-FL on hippocampal plasticity and learning. Dietary 2′-FL potentiated hippocampal CA3-CA1 long-term potentiation (LTP) and improved operant-conditioning performance, whereas bilateral subdiaphragmatic vagotomy prevented the 2′-FL-associated enhancement of these outcomes. An equimolar dietary dose of free L-fucose did not enhance hippocampal LTP, indicating that the fucose moiety alone was insufficient to reproduce the electrophysiological effect of 2′-FL. ⁵⁹ These findings support a vagus-dependent contribution to selected neural effects of dietary 2′-FL. However, the study did not determine whether the initiating intestinal signal involved direct epithelial or enteroendocrine sensing, processing within the ENS, microbiota-derived intermediates, or a combination of these mechanisms.

Studies of other luminal signals provide additional examples of enteroendocrine-neural communication. Bacterial tryptophan metabolites activate transient receptor potential ankyrin 1-positive enteroendocrine cells and subsequently stimulate enteric and vagal neurons through 5-hydroxytryptamine release. ⁶⁰ Sodium oligomannate has also been reported to activate enteroendocrine-vagal afferent signaling in amyloid precursor protein/presenilin 1 mice. ⁶¹ However, these mechanisms have not been demonstrated for HMOs or products released during microbial degradation of host glycans.

Taken together, intact fucosylated HMOs can acutely modify ENS-dependent intestinal motor activity, ^{57 , 58} and selected neural effects of dietary 2′-FL require an intact vagus nerve. ⁵⁹ However, the intestinal sensor and cellular sequence linking luminal 2′-FL to vagal activation remain unknown. Current evidence therefore supports the participation of enteric and vagal pathways but does not yet establish a complete mechanism from HMO sensing in the intestine to neural outcomes in the brain.

3.4. Microbial glycan metabolism as a functional indicator and therapeutic target in disease states

Genomic and metagenomic analyses reveal the glycan metabolic capacity of the gut microbiota. This capacity, which includes both biosynthetic and degradative pathways, is closely linked to gut-brain axis function. Across distinct disease contexts, alterations in these pathways exhibit reproducible and disease-specific patterns.

In neurodegenerative conditions, such as Alzheimer’s disease ⁶² and mild cognitive impairment, ⁶³ gut microbial pathways related to glycan biosynthesis and metabolism are broadly upregulated. By contrast, patients with Parkinson’s disease display a generalized enhancement of mucin and host-derived glycan degradation capacity. ⁶⁴ Distinct patterns are also observed in disorders characterized by metabolic or inflammatory dysregulation. For example, glycan metabolic pathways are downregulated in patients with hypertension-associated cognitive impairment, ⁶⁵ whereas individuals with active IBD accompanied by depressive symptoms exhibit specific alterations in glycosaminoglycan metabolism. ⁶⁶ However, most current evidence reflects associations rather than direct causal roles. Microbial glycan metabolism may indicate the functional state of the gut-brain axis, but it cannot yet be considered a definitive biomarker.

Dietary supplementation, such as prebiotics or defined polysaccharides, has been used to modulate microbial glycan metabolism. Microbiota-based interventions, including probiotics, have shown similar effects. These approaches may influence gut-brain axis function. Supplementation with 2′-FL reshapes microbial community structure and enhances the production of SCFAs and other beneficial microbial metabolites, thereby conferring protection against experimental colitis. ⁶⁷ Combined administration of galactooligosaccharides and 2′-FL further augments the generation of SCFAs and AhR ligands, contributing to immune regulation within the gut. ⁶⁸ In addition, supplementation with specific probiotic strains, such as Lactobacillus species, restores impaired microbial glycan metabolism in mouse models of exertional heat stroke and is accompanied by improvements in cognitive performance and gut-brain axis signaling. ⁶⁹ Plant-derived polysaccharides also show modulatory potential. For instance, kiwifruit polysaccharides alleviate colitis through mechanisms involving activation of the AhR pathway and upregulation of epithelial fucosylation in the colon, further supporting a role for microbial glycan metabolic pathways in maintaining barrier integrity and immune homeostasis. ⁷⁰

Microbial glycan metabolism reflects the functional state of the gut-brain axis and may provide an entry point for therapeutic intervention in neuropsychiatric disorders. It should be noted that much of the current evidence is derived from rodent models, and substantial differences in microbiota composition and immune responses between rodents and humans may limit the direct translation of these findings. In addition, direct evidence linking specific metabolic alterations to changes in microbial metabolite profiles and brain function remains limited. The main unresolved question is what host-derived signals drive these shifts in microbial glycan metabolism under different physiological and disease conditions. This includes identifying host-derived factors, such as neural, immune, or hormonal signals, that influence how microbes use glycans. It is also important to examine how changes in microbial glycan metabolism affect processes in the brain, including neuroinflammation, BBB integrity, and synaptic plasticity. Experimental models using defined microbial communities together with controlled host perturbations may help address these questions. Based on these findings, dietary interventions, such as prebiotics or specific polysaccharides, can be tested to assess whether they modulate gut-brain axis function through microbial glycan metabolism.

3.5. From microbial glycan metabolism to brain glycoproteome remodeling

The preceding sections indicate that host glycosylation patterns and dietary glycans, such as HMOs, influence brain function along the gut-brain axis primarily through indirect mechanisms, including the modulation of gut microbial composition and glycan metabolic activity. Against this background, a recent study employing DQGlyco provides molecular-scale evidence within the central nervous system, demonstrating a systematic association between gut microbial status and site-specific protein glycosylation in the brain. ²⁷

In this study, the brain glycoproteomes of adult germ-free mice, mice mono-colonized with Bacteroides uniformis, and mice colonized with an eight-member microbial community were analyzed separately. Distinct glycosylation profiles were observed across different colonization states, indicating that gut microbial colonization is closely associated with the global architecture of the brain glycoproteome. Among the quantifiable glycopeptides, a large number of N-linked glycopeptides exhibited significant changes in abundance, whereas corresponding alterations in total protein levels were relatively limited. These findings suggest that microbial colonization predominantly reshapes glycosylation at the post-translational level rather than altering protein expression per se. ²⁷

Further analyses revealed that glycosylation events sensitive to microbial status were enriched among proteins with established roles in neural function. These included proteins involved in axon guidance and cytoskeletal regulation, receptors for neurotransmitters such as glutamate and γ-aminobutyric acid, as well as proteins implicated in Alzheimer’s disease, including low-density lipoprotein receptor-related protein 1. Notably, these glycosylation changes were accompanied by alterations in protein thermal stability, solubility, or conformational properties, implying that changes in glycosylation may influence protein functional states. Across studies, evidence indicates a close association between gut microbial colonization and site-specific glycosylation of key brain proteins, with potential consequences for protein function. ²⁷ However, germ-free models are associated with profound developmental and immunological alterations, which may confound the interpretation of glycosylation changes. ⁷¹ It therefore remains unclear to what extent the observed effects are directly attributable to microbial signals rather than secondary to altered host development or immune status.

Despite providing direct evidence for microbiota-associated remodeling of the brain glycoproteome, the mechanisms underlying this regulation remain to be elucidated. It is currently unclear which microbe-derived signals transmit information from the gut to the central nervous system to regulate glycosylation, and is also not well understood why only certain proteins or glycosylation sites respond to changes in microbial status. Moreover, how these molecular alterations are translated into changes in neural function and behavior has yet to be systematically defined. Future studies need to move beyond correlative observations. This will require direct manipulation of the gut microbiota and longitudinal analyses of brain glycosylation and behavioral outcomes. These approaches are essential for establishing causal relationships between the gut microbiota and brain glycosylation.

4. Molecular mechanisms of glycosylation in immune regulation and neuroinflammation along the gut-brain axis

Immune pathways are an important route through which intestinal signals influence the brain. By modifying immune receptors, ligands, and immunoglobulins, glycosylation shapes local and systemic immune responses and may thereby affect neuroinflammation.

4.1. Immunomodulatory effects of microbial glycans

Specific glycans derived from the gut microbiota are increasingly recognized as key signaling molecules linking the intestinal microenvironment to host immune regulation. Certain microbial glycans, such as polysaccharide A (PSA) produced by Bacteroides fragilis, activate Toll-like receptor 2 on dendritic cells and promote the expansion and tissue-specific migration of CD39⁺ regulatory CD4⁺ T cells (Tregs). In addition to maintaining immune homeostasis within the intestine, these Tregs can migrate to the central nervous system, where they suppress demyelination and inflammatory damage in experimental autoimmune encephalomyelitis models (Figure 4). ^{72 , 73} These findings indicate that, within defined immunological contexts, microbiota-derived glycans can attenuate neuroinflammation at distal sites by inducing regulatory immune responses.

Figure 4.

Glycosylation-dependent regulation of immune signaling and neuroinflammation along the gut-brain axis. Microbial- and host-derived glycans regulate immune responses at the intestinal interface through distinct but interconnected pathways. In the gut, polysaccharide A (PSA) from Bacteroides fragilis activates Toll-like receptor 2 (TLR2) on dendritic cells, promoting the expansion of CD39⁺ regulatory CD4⁺ T cells that contribute to anti-inflammatory responses. In parallel, peptidoglycan fragments released from gut microbiota activate intracellular pattern-recognition receptors such as nucleotide-binding oligomerization domain 1 (NOD1) in innate immune cells, leading to pro-inflammatory signaling under sustained or aberrant exposure. These intestinal immune pathways may alter systemic inflammatory tone, while glycosylation-dependent mechanisms in the circulation, including immunoglobulin G (IgG) Fc glycosylation, modulate antibody effector functions and inflammatory responses. Peripheral immune changes may, in turn, be associated with neuroimmune responses, including microglial activation, although the intervening pathways remain incompletely defined. In addition, specific glycosylation pathways can directly regulate neuroinflammatory signaling at the molecular level. Core fucosylation, catalyzed by α1,6-fucosyltransferase (FUT8), modulates glycoprotein 130 (gp130)/interleukin-6 receptor (IL-6R) complex formation and downstream Janus kinase 2 (JAK2)/signal transducer and activator of transcription 3 (STAT3) signaling. Reduced core fucosylation enhances IL-6 signaling and neuroinflammation, whereas L-fucose supplementation restores core fucosylation via the salvage pathway and attenuates inflammatory responses. Together, these findings illustrate how glycosylation regulates microbial, immune, and neuroinflammatory processes at multiple levels of the gut-brain axis. Monosaccharide symbols follow the Symbol Nomenclature for Glycans (SNFG): blue square, N-acetylglucosamine (GlcNAc); green circle, mannose (Man); yellow circle, galactose (Gal); red triangle, fucose (Fuc); and purple diamond, N-acetylneuraminic acid (Neu5Ac). EAE, experimental autoimmune encephalomyelitis; Tregs, regulatory T cells. Created with BioRender.com.

In contrast, constitutively released peptidoglycan fragments from the gut microbiota can chronically activate the pattern-recognition receptor nucleotide-binding oligomerization domain 1, thereby stimulating innate immune cells such as neutrophils. Under conditions of sustained or aberrant exposure, this signaling pathway is thought to reinforce a pro-inflammatory immune tone, which may influence neuroinflammatory states in the brain and has been associated with behavioral alterations (Figure 4). ^{74 , 75}

Taken together, current evidence suggests that structurally distinct microbial glycans elicit divergent immune outcomes by engaging different pattern-recognition receptors and downstream signaling pathways. Whereas certain glycans preferentially induce regulatory immune programs, persistent or dysregulated inflammatory signals may amplify pro-inflammatory pathways and, under specific conditions, affect central neuroinflammatory processes. However, how these immune signals are integrated at the systemic level and translated into specific neuroinflammatory outcomes remains poorly understood. Future studies need to combine defined glycan exposures with neuroinflammatory models to dissect these mechanisms and to determine which conditions promote regulatory, anti-inflammatory immune responses.

4.2. Host glycosylation in immune responses and its potential impact on the gut-brain axis

Host glycosylation is a key mechanism that helps direct immune responses toward tolerance or inflammation under microbial influence. Sialylation, a common glycosylation modification, is involved in this process. Research indicates that sialylation regulates mucosal immunity through distinct pathways. For example, the sialylation of secretory IgA (SIgA) is essential for its uptake by intestinal M cells, a process that influences antigen delivery and immune responses. ⁷⁶ Separately, the enzyme ST6GAL1, which creates α2,6-sialyl linkages, shapes the gut environment. Deficiency in St6gal1 alters the gut microbiome composition, and this altered microbiome can drive a pro-inflammatory T helper 17 response in the mucosa. ⁷⁷ These findings underscore the role of protein sialylation in maintaining intestinal immune balance. Given the capacity of mucosal immune responses to shape systemic inflammation, intestinal sialylation may also influence immune signaling beyond the gut. Nevertheless, whether the SIgA- and ST6GAL1-related mechanisms described above contribute to central neuroinflammatory processes remains unclear.

In terms of humoral immunity, the N-glycan composition of the Fc region of immunoglobulin G (IgG) is widely recognized as a critical determinant of antibody effector function. Multiple studies have reported a marked increase in agalactosylated IgG in the serum of patients with IBD, a glycan feature associated with enhanced antibody-dependent phagocytosis and tissue inflammation. ^{78 , 79} In addition, patients with Crohn’s disease and ulcerative colitis exhibit a general reduction in IgG glycosylation levels in plasma, which correlates with disease activity and clinical outcomes. ⁸⁰ Notably, altered IgG glycosylation has also been associated with inflammatory changes in the central nervous system (Figure 4). In cerebrospinal fluid from patients with COVID-19, alterations in IgG fucosylation and sialylation are associated with microglial activation and increased neuroinflammation. ⁸¹ Similar associations between aberrant IgG N-glycosylation and systemic or central inflammatory states have been reported in children with hypoxic-ischemic encephalopathy, patients with neuromyelitis optica spectrum disorder, and subsets of pediatric epilepsy cases. ^{82 - 84} These observations suggest that IgG glycosylation may represent a molecular interface linking peripheral immune status to neuroinflammatory processes, although the causal nature of this relationship remains to be clarified.

At the level of cellular immunity, core fucosylation of the T cell receptor (TCR) complex, catalyzed by FUT8, is essential for effective TCR signal transduction and the production of inflammatory cytokines. Experimental models of colitis indicate that TCR core fucosylation constitutes a key molecular requirement for the initiation of intestinal inflammation. Correspondingly, increased levels of core fucosylation on T cells have been observed in patients with IBD, ⁸⁵ suggesting that this modification may contribute to disease pathogenesis by modulating T cell activation thresholds.

The above studies show that host glycosylation regulates antibody effector function and T-cell activation, thereby shaping intestinal and systemic immune responses. However, direct evidence that these glycosylation changes transmit intestinal inflammation to the brain remains limited. Future studies should manipulate defined glycosylation pathways in intestinal or immune cells and assess circulating mediators, microglial activation, and behavioral outcomes within the same experimental model.

4.3. Core fucosylation: linking peripheral metabolism to neuroinflammation

Building on the discussion of how host glycosylation regulates immune cell function and systemic inflammatory tone, recent evidence has begun to identify specific glycosylation pathways that may influence neuroinflammatory responses and potentially link peripheral metabolic states to central processes. Among these, core fucosylation provides a mechanistically defined example of how glycosylation can regulate neuroimmune signaling in a manner potentially relevant to the gut-brain axis.

In addition to its established role in TCR signaling and inflammatory responses, emerging evidence indicates that alterations in core fucosylation can directly influence neuroinflammatory pathways. In a chronic unpredictable stress model, Fut8 heterozygous mice exhibit an accelerated onset of depressive-like behaviors accompanied by enhanced neuroinflammation. Mechanistically, reduced core fucosylation promotes the formation of glycoprotein 130 (gp130)/interleukin-6 receptor complex, leading to sustained activation of the Janus kinase 2 (JAK2)/signal transducer and activator of transcription 3 (STAT3) signaling pathway and amplification of inflammatory responses (Figure 4). ⁸⁶ Importantly, L-fucose supplementation restores core fucosylation of gp130 and attenuates downstream inflammatory signaling, resulting in improved behavioral phenotypes. ⁸⁶ These findings suggest that modulation of glycosylation status can directly influence neuroimmune signaling pathways.

From the perspective of the gut-brain axis, this mechanism suggests that peripheral fucose metabolism may contribute to central glycosylation. L-fucose can be released during microbial degradation of fucosylated glycans in the gut. ⁸⁷ Although free L-fucose itself has limited permeability across the BBB, oral administration has been shown to increase brain GDP-fucose levels via the salvage pathway, suggesting a potential route by which peripheral metabolic states may influence glycosylation in the brain. ^{88 , 89}

The findings above suggest that core fucosylation may link peripheral fucose metabolism to neuroinflammatory signaling, with potential relevance to the gut-brain axis. Peripheral fucose metabolism may, in part, be influenced by microbial degradation of fucosylated glycans in the gut. However, several key questions remain unresolved. It is unclear whether microbiota-derived L-fucose quantitatively contributes to brain GDP-fucose pools under physiological conditions, or whether these effects primarily reflect systemic metabolic regulation. In addition, the relative contribution of central versus peripheral fucose sources to core fucosylation in neural tissues remains to be defined. Further studies are required to determine how broadly such mechanisms operate and whether they contribute to glycosylation-dependent regulation within the gut-brain axis.

4.4. Roles of lectins and neural glycosylation in neuroimmune regulation along the gut-brain axis

Beyond the immunomodulatory and neuroinflammatory mechanisms discussed above, glycosylation also contributes to neuroimmune regulation by shaping how peripheral immune signals are integrated within the central nervous system. Through this cross-organ regulation, glycan-binding proteins, neuroimmune cell populations, and their associated glycosylation modifications collectively influence the transmission of inflammatory signals between the periphery and the brain.

Endogenous lectins, such as galectins, mediate functional connections between the intestinal immune barrier and the central nervous system through recognition of specific glycan structures. ^{90 , 91} Galectin-4 provides a representative example. Its expression in intestinal epithelial cells facilitates the regulation of gut microbial composition and spatial organization through interactions with bacterial surface glycans. In relevant animal models, Galectin-4 deficiency not only leads to gut microbial dysbiosis but is also accompanied by impaired hippocampal synaptic plasticity and deficits in memory performance. ⁹² These findings suggest that Galectin-4 may influence gut-brain communication by modulating microbiota composition and the associated immune environment. Through this pathway, glycan-dependent changes in microbial communities may shape peripheral immune signals that can, in turn, influence central neuroimmune responses, including microglial function. However, it remains unclear through which circulating signals or neuroimmune pathways Galectin-4-dependent intestinal changes exert distal effects on the hippocampus, and the mechanisms underlying this long-range signaling have yet to be elucidated.

At the neuroimmune level, glycosylation modifications likewise participate in the regulation of microglial functional states. Neuraminidase 1 (NEU1) modulates the sialylation status of the microglial receptor triggering receptor expressed on myeloid cells 2 (TREM2), thereby affecting downstream signaling, phagocytic activity, and inflammatory phenotypes, and enabling dynamic balance between pro-inflammatory and anti-inflammatory states. ⁹³ In addition, a central nervous system inflammatory model revealed widespread remodeling of N-glycosylation patterns in brain tissue, characterized by overall reductions in sialylation and core fucosylation. ⁹⁴ Clinical studies further indicate that such N-glycosylation patterns may themselves serve as potential diagnostic or stratification biomarkers for neuroinflammatory disorders, including multiple sclerosis. ⁹⁵

Glycosylation contributes to neuroimmune interactions across peripheral and central compartments, involving immune cells, circulating antibodies, neural stem/progenitor cells, and brain-resident microglia. Disruption of these processes may contribute to neuroimmune abnormalities, although their specific roles in transmitting gut-derived signals to the brain remain incompletely defined. Although existing studies have established associations between Galectin-4 deficiency, microbial dysbiosis, altered hippocampal synaptic plasticity, and cognitive impairment, ⁹² the intermediate signals and transmission routes linking intestinal changes to central neural phenotypes remain undefined. Specifically, the dominant gut-derived signals involved in this process, including specific metabolites or microbial components, remain to be identified. Likewise, the principal transmission routes, whether via circulation, vagal pathways, or immune cell migration, are still unclear. In this context, further analysis of how Galectin-4-associated intestinal alterations are conveyed to the central nervous system may provide more direct insight into the molecular mechanisms underlying neuroimmune regulation along the gut-brain axis.

5. Glycosylation in brain function: from the blood-brain barrier to cognition and behavior

At the brain end of gut-brain communication, glycosylation contributes to BBB integrity, neurodevelopment, synaptic plasticity, and behavior. The BBB glycocalyx may influence the exposure of neural tissue to circulating signals, whereas glycosylation of neural proteins regulates cellular responses within the brain. Some of these processes are influenced by nutritional or microbial inputs, but the links to the gut remain incompletely defined for many brain-intrinsic glycosylation mechanisms.

5.1. Glycosylation maintains BBB structure and function

The BBB is essential for maintaining central nervous system homeostasis, and its integrity is closely linked to glycosylation. The luminal surface of brain microvascular endothelial cells is coated with a dense, polysaccharide-rich glycocalyx composed mainly of highly O- and N-glycosylated transmembrane mucin-like proteins, including podocalyxin, as well as proteoglycans such as syndecans and various glycolipids. ²² This negatively charged glycocalyx forms both a physical and electrostatic barrier that restricts nonspecific adhesion and transmigration of macromolecules and circulating cells, while also participating in shear stress sensing, vascular tone regulation, and inflammatory signaling. ^{22 , 96 , 97}

During physiological aging and in neurodegenerative disorders such as Alzheimer’s disease, substantial degradation or compositional remodeling of the endothelial glycocalyx has been observed, with particular alterations affecting core 1 O-glycan structures. These changes correlate with increased BBB permeability, exacerbated neuroinflammation, and cognitive decline. Notably, endothelial-specific restoration of core 1 O-glycan biosynthesis in aged mice using adeno-associated virus-mediated gene delivery significantly improves BBB integrity, attenuates neuroinflammation, and rescues cognitive performance, ²² underscoring the functional importance of specific O-glycosylation pathways in BBB maintenance.

BBB maturation also requires the coordinated action of astrocytic endfeet. The proteoglycan dentin matrix protein 1 (DMP1-PG) is highly expressed in astrocytes, and O-glycosylation at its Ser89 residue is critical for proper astrocyte polarization and for the final formation and maintenance of the BBB. A mutation at this site (S89G) disrupts astrocytic function, reduces expression of the tight junction protein ZO-1 and of basement membrane laminins, and results in severe BBB impairment. ⁹⁸

In summary, these findings indicate that specific O-glycosylation events, including core 1 O-glycans and O-glycosylation of DMP1-PG, are essential for BBB development, homeostasis, and functional preservation during aging. However, the upstream regulatory networks that control these modifications in response to inflammatory or aging-related signals remain poorly understood. The downstream effectors of these pathways are also unclear. In addition, their region- and disease-specific roles, for example in cerebral small vessel disease or traumatic brain injury, have not been systematically defined at the molecular and cellular levels.

5.2. Early-life nutritional glycosylation and neurodevelopment

Exogenous glycosylation-related nutrients during early life exert profound influences on brain development and may contribute to long-term functional outcomes. Human milk, particularly its abundance of HMOs and glycoproteins, represents a prominent example. Epidemiological studies have shown that higher levels of total fucosylated and sialylated oligosaccharides in human milk are positively associated with later language and cognitive scores in infants. ⁹⁹ Correspondingly, breastfed infants exhibit higher levels of gangliosides and protein-bound sialic acid in the brain compared with formula-fed infants. ¹⁰⁰

Sialylated milk oligosaccharides may contribute to neonatal sialic acid availability and influence brain metabolism, especially during periods when endogenous synthesis is limited. ¹⁰¹ Animal studies demonstrate that dietary sialic acid provided early in life, from sources such as edible bird’s nest glycoproteins, goat whey, or direct N-acetylneuraminic acid supplementation, can be absorbed and may influence brain glycosylation through systemic circulation. ^{102 - 105} Sialic acid supplementation increases the expression of genes involved in sialic acid metabolism and polysialic acid synthesis, including UDP-N-acetylglucosamine 2-epimerase/N-acetylmannosamine kinase and ST8 alpha-N-acetyl-neuraminide alpha-2,8-sialyltransferase 4. It also elevates hippocampal BDNF levels and enhances PSA-NCAM modification. These molecular changes are accompanied by increased dendritic spine formation and improved synaptic plasticity. Functionally, these effects translate into enhanced learning and memory performance, as well as strengthened hippocampal LTP (Figure 5). ^{102 - 105}

Figure 5.

Glycosylation supports brain function at multiple levels from blood-brain barrier integrity to neurodevelopment. Glycosylation contributes to brain function through distinct but interconnected mechanisms. At the blood-brain barrier (BBB), endothelial glycocalyx structures enriched in O-glycans, together with astrocyte-derived proteoglycan dentin matrix protein 1 (DMP1-PG), maintain barrier integrity and regulate permeability. During early life, dietary glycans, including sialylated human milk oligosaccharides (HMOs), provide substrates for brain sialic acid metabolism and enhance polysialylated neural cell adhesion molecule (PSA-NCAM)-associated pathways, promoting dendritic spine formation, synaptic plasticity, and long-term potentiation (LTP). These findings highlight the potential contribution of early-life nutritional sialic acid availability to the structural and functional maturation of the brain. Created with BioRender.com.

Early-life supplementation with 2′-FL similarly confers long-lasting neurobehavioral benefits, enhancing learning, memory, and hippocampal LTP in adulthood. ¹⁰⁶ In contrast, early-life deficiency of specific sialylated HMOs, such as 6′-sialyllactose, results in persistent impairments in social behavior and cognition. ¹⁰⁷ In a mouse model, maternal hyperglycemia and high-fat feeding altered the concentrations of sialylated oligosaccharides and sialic acid in milk, ¹⁰⁸ whereas maternal sialic acid supplementation increased offspring brain sialic acid levels and improved cognitive performance. ¹⁰⁹

Together, these studies suggest that glycosylation-related nutrients obtained through early-life nutrition may influence neurodevelopment by modulating gut microbial composition, supplying metabolic substrates, or altering host signaling pathways. Nevertheless, it remains unclear how these nutrition-associated sugars affect BBB maturation during development, whether they are selectively utilized by specific brain cell types, and how their effects are ultimately integrated into neural circuit formation and maturation. Addressing these questions will require integrated analyses using germ-free models and cell type-specific genetic manipulations to better define the biological links between early nutritional exposure and neurodevelopmental outcomes.

5.3. Glycosylation in neural plasticity and behavior

In the adult brain, glycosylation remains highly dynamic and plays an important role in regulating neural plasticity and behavior. ^{110 , 111} PSA-NCAM is highly expressed in neurogenic regions such as the hippocampus and olfactory bulb. ¹¹² Its highly negatively charged polysialic acid chains reduce cell-cell adhesion, facilitating axonal growth, synaptic remodeling, and the migration and integration of newborn neurons, thereby playing a central role in learning and memory processes. ¹¹³ 3′-SL enhances hippocampal PSA-NCAM expression by modulating gut microbiota composition and host sialic acid metabolism, resulting in improved learning and memory performance. ²⁴ Specific HMOs, including 3′-SL and 2′-FL, also induce long-term changes in gene expression and function within the mesolimbic dopaminergic system, with pronounced sex-dependent effects, ¹¹⁴ which may contribute to sex differences observed in neuropsychiatric disorders. In addition to these mechanisms, intracellular glycosylation also contributes to neurogenesis regulation. In adult neural stem/progenitor cells (aNSPCs), O-GlcNAc glycosylation catalyzed by O-GlcNAc transferase (OGT) regulates cell-cycle progression, neurogenesis, and neuronal differentiation through modification of the transmembrane and intracellular domains of Notch. ¹¹⁵ Loss of Ogt reduces the pool of aNSPCs and disrupts normal neurogenesis. This cellular defect is accompanied by cognitive impairment in vivo. Therefore, glycosylation appears to influence not only mature neuronal function but also the developmental processes that determine neural cell fate. At the synaptic level, N-glycosylation of vesicle release proteins such as synaptotagmin 1 plays an essential role in calcium-dependent neurotransmitter release, and is characterized by high fucosylation and terminal galactosylation. ²⁵ Core fucosylation also contributes to synaptic plasticity. Loss of FUT8 impairs hippocampal LTP and alters AMPA receptor trafficking, highlighting a critical role of core fucosylation in learning and memory processes. ¹¹⁶

HMOs, particularly 2′-FL, show considerable potential in modulating behavioral phenotypes. In experimental models, 2′-FL enhances brain function through vagus nerve-dependent mechanisms, including improvements in contextual fear memory. ⁵⁹ In addition, 2′-FL improves social behavior and cognitive flexibility in offspring of maternal immune activation models, an effect linked to modulation of the microbiota-bile acid metabolic axis. ¹¹⁷ Under early-life stress conditions, 2′-FL alleviates anxiety-like behavior and amygdala hyperactivity in mice, and in rat models of attention-deficit/hyperactivity disorder, early-life intervention with 2′-FL or synbiotic treatment with Bifidobacterium improves learning, memory, and hyperactivity. ^{118 , 119} Clinical studies provide preliminary support for these observations. Supplementation with 2′-FL increases gut Bifidobacterium abundance and is associated with improved visual memory in older adults. ¹²⁰ Prebiotic formulations containing 2′-FL have also been linked to reduced stress-related symptoms in working populations, although substantial inter-individual variability has been reported. ¹²¹

Beyond HMOs, additional glycan-related compounds also show modulatory potential along the gut-brain axis. For example, the glycosylated flavonoid α-glycosyl isoquercitrin modulates cognition and emotional behavior via the gut microbiota-blood-brain axis. ^{122 , 123} Similarly, dietary polysaccharides such as glucomannan improve cognitive function by reshaping the gut microbiota and circulating metabolites, including sialic acid-related molecules, ultimately enhancing glutamatergic synaptic activity. ¹²⁴ Polymorphisms in FUT2, which controls secretor-dependent fucosylation, are associated with increased risk of gastrointestinal disorders. Genetic evidence also links these variants to psychiatric conditions, such as depression and bipolar disorder, suggesting a potential role for host glycans in gut-brain interactions. ³⁵

On the whole, glycosylation contributes not only to structural integrity and developmental programming of the brain but also to the regulation of neural plasticity and behavior. Glycosylation-related factors, including HMOs and their downstream signals, do not directly dictate behavioral phenotypes. Instead, they shape gut microbial ecology and metabolic outputs, which in turn modulate neuroimmune signaling networks and neural plasticity, ultimately influencing learning, memory, and emotional behavior. Future studies in larger and more diverse human cohorts will be required. These studies could combine gut microbiome profiling, glycomics, and metabolomics to determine whether glycosylation-mediated microbial signals identified in animal models are conserved in humans. Longitudinal analyses linking these molecular measures with cognitive, emotional, and neural imaging outcomes would evaluate the relevance of glycosylation-associated pathways in mental health regulation.

6. Evidence and candidate pathways for brain-to-gut regulation of intestinal glycosylation

Compared with gut-to-brain signaling, brain-to-gut regulation of intestinal glycosylation is less well established. Current studies show that psychological stress can alter epithelial α1,2-fucosylation, mucin O-glycan structures, and the abundance of mucosal glycoproteins. However, only a subset of the upstream pathways has been tested directly. The following sections distinguish measured glycan changes from changes in glycosyltransferase or mucin expression and discuss neural, endocrine, and immune mechanisms according to the strength of the available evidence.

6.1. Psychological stress and intestinal glycosylation

Psychological stress can alter intestinal epithelial glycosylation. CSDS reduces Fut1 and Fut2 expression and decreases the reactivity of several fucose-specific lectins in the distal small intestine. ²¹ The regional distribution of this change indicates that intestinal glycosylation does not respond uniformly to psychological stress. However, this study did not identify the neural, endocrine, or immune signals responsible for the reduction in epithelial fucosylation. It also remains unclear whether reduced fucosylation contributes to subsequent changes in microbial ecology or behavioral outcomes. In addition to epithelial fucosylation, psychological stress can alter the O-glycan structures carried by intestinal mucins. In a water-avoidance stress model, mass-spectrometric analysis revealed marked changes in intestinal mucin O-glycans, including the appearance of elongated polylactosamine-containing structures. These changes occurred without detectable alterations in goblet-cell number or Muc2 expression and were accompanied by reduced mucus cohesion, increased intestinal permeability, and visceral hypersensitivity. Probiotic treatment prevented both the O-glycosylation changes and the impairment of mucus properties. ¹²⁵ This study provides direct structural evidence that psychological stress can modify intestinal mucin glycosylation independently of mucin abundance. Nevertheless, the signals connecting the stress response to altered glycan biosynthesis were not determined. Together, these studies indicate that psychological stress can remodel intestinal glycosylation at both the epithelial surface and the mucus layer, including reduced epithelial α1,2-fucosylation and altered mucin O-glycan structures. However, the neural and endocrine signals responsible for these changes remain unclear, and the contribution of glycan remodeling to subsequent microbial and behavioral alterations has not been fully established.

6.2. Stress-related regulation of intestinal mucins

Psychological stress may also alter the intestinal mucus barrier by regulating the abundance of highly glycosylated mucosal proteins. In a chronic stress model, the transmembrane mucin MUC13 was selectively reduced in the intestinal epithelium. This reduction was mediated, at least in part, by the hepatocyte nuclear factor 4 transcription factor family. Moreover, Muc13 deletion was sufficient to induce microbiome changes resembling those observed after stress and to promote despair-like behaviors. ¹²⁶ These findings indicate that stress-induced changes in epithelial mucins can contribute to microbial dysbiosis and behavioral abnormalities.

However, this study examined Muc13 expression rather than the O-glycan structures carried by MUC13. It therefore supports the regulation of a major component of the epithelial glycocalyx, but it does not demonstrate that psychological stress changes MUC13 glycosylation itself. This distinction is important because a reduction in mucin abundance and a change in mucin glycan composition may affect microbial colonization and epithelial signaling through different mechanisms. Thus, the MUC13 study extends brain-to-gut regulation to stress-sensitive mucosal glycoprotein expression, but whether the glycosylation of MUC13 is also altered remains unknown.

6.3. Potential regulatory mechanisms and unresolved questions

The IL-22-FUT2 pathway provides one possible explanation for stress-associated changes in intestinal fucosylation. Psychological stress induces glucocorticoid-dependent loss of intestinal CD45⁺ CD90⁺ cells involved in IL-22 production, whereas glucocorticoid-receptor blockade or recombinant IL-22 restores mucosal immunity and epithelial Fut2 expression. ¹²⁷ However, final α1,2-fucosylated glycan structures were not directly measured, and this pathway has not been tested in the CSDS model used in reference. ²¹

Neural activity can also directly regulate intestinal epithelial fucosylation. Vasoactive intestinal peptide-producing enteric neurons promote epithelial Fut2 expression through VIPR1-ERK1/2-c-Fos signaling. Perturbation of enteric neurons or subdiaphragmatic vagotomy reduces epithelial α1,2-fucosylation and alters microbiota composition. ¹²⁸ This study provides direct evidence for neural regulation of intestinal glycosylation, although it does not establish that psychological stress acts through the same pathway.

Human evidence remains limited. Long-term meditation has been associated with enrichment of predicted microbial pathways related to glycan biosynthesis and metabolism. ¹²⁹ Because the study was cross-sectional and the microbial functions were computationally inferred, it cannot determine whether psychological state caused the observed microbial changes. This finding should therefore be considered associative and hypothesis-generating rather than mechanistic evidence.

Future studies should examine intestinal glycosylation while selectively manipulating stress-responsive brain regions, autonomic pathways, glucocorticoid signaling, IL-22 activity, or enteric neurons. Direct glycomic measurements are also required, because changes in Fut2 or mucin expression do not necessarily reflect the abundance or structure of the final glycan products. Combining neural manipulation with intestinal glycomics, epithelial-specific genetic models, and microbiota analysis will be necessary to determine whether stress-induced glycosylation changes are responsible for subsequent microbial and behavioral effects.

7. Outlook: evidence, emerging technologies, and translation

Current studies support roles for glycosylation at several points of gut-brain communication, but the strength and scope of the evidence vary considerably. Some local mechanisms have been tested through genetic, pharmacological, or neural manipulation. By contrast, many proposed links between intestinal glycosylation and brain outcomes remain based on associations observed within the same animal model. Future studies should therefore distinguish what was directly measured from what was inferred, identify the route connecting the gut and brain, and test whether the glycosylation change is required for the observed phenotype.

7.1. Evaluating current evidence and establishing causality

Table 1 summarizes representative studies according to the type and scope of evidence. This distinction is important because demonstrating a local function of glycosylation does not by itself establish a gut-brain mechanism. For example, intestinal C3GnT deficiency alters mucus organization, barrier function, cecal metabolites, hippocampal immature neurons, and memory in the same mouse model. ¹¹ However, the signal connecting the intestinal changes to the hippocampus remains unknown. By contrast, vagotomy provides route-specific evidence that selected effects of dietary 2′-FL depend on the vagus nerve. ⁵⁹ The DQGlyco study further shows that defined microbial communities can remodel site-specific N-glycosylation in the mouse brain, ²⁷ although the responsible microbial products or host pathways were not identified.

Table 1.

Representative glycosylation-related findings in gut-brain communication classified by evidence type.

Glycosylation type or target	Key molecule or enzyme	Experimental context	Main finding	Evidence type	Unresolved gut-brain link	References
Site-specific mucin O-glycosylation	MUC2	Recombinant MUC2 constructs	Site-specific O-glycosylation protects MUC2 from bacterial protease cleavage.	Local mechanistic evidence	Whether comparable site-specific changes in intestinal MUC2 alter microbiota- or brain-related outcomes in vivo remains unknown.	[29]
Core 3 O-glycosylation	C3GnT	Core 3 O-glycan-deficient mice	Core 3 deficiency causes barrier defects and is associated with altered metabolites, hippocampal abnormalities, and impaired memory.	Cross-organ association within one animal model	The metabolic, immune, or neural mediator connecting the intestinal phenotype to the hippocampal changes has not been identified.	[11]
Intestinal epithelial α1,2-fucosylation	ILC3-IL-22-LT-FUT2	Genetic and immune-intervention mouse models	ILC3-derived IL-22 and lymphotoxin induce epithelial FUT2-dependent fucosylation.	Local mechanistic evidence	Distal neural effects of this intestinal pathway have not been directly tested.	[14]
Stress-associated intestinal fucosylation	Fut1/Fut2	CSDS mice	CSDS reduces intestinal fucose-specific lectin reactivity and fucosyltransferase expression.	Animal stress-model evidence with intestinal outcomes	The neural, endocrine, or immune pathway causing the intestinal change, and its contribution to subsequent behavior, remain unknown.	[21]
Stress-associated mucin O-glycosylation	Mucin O-glycans	Water-avoidance stress rats with glycomic analysis	Stress alters mucin O-glycans and weakens mucus barrier properties independently of mucin abundance.	Animal stress-model evidence with intestinal outcomes	The upstream neural or endocrine pathway that changes mucin glycan biosynthesis was not identified.	[125]
Neural regulation of epithelial α1,2-fucosylation	VIP-VIPR1-ERK1/2-c-Fos-Fut2	Mouse vagotomy and enteric-neuron manipulation	VIP-producing enteric neurons promote epithelial FUT2-dependent α1,2-fucosylation.	Route-specific intervention evidence	The study does not establish that psychological stress acts through this VIP-dependent pathway.	[128]
Host-glycan utilization and bacterial surface fucosylation	PULs; surface L-fucosylation	Gnotobiotic and competitive-colonization mouse models	Host-glycan utilization and bacterial surface fucosylation support intestinal persistence and competitive colonization.	Local mechanistic evidence	How these colonization mechanisms alter microbial products relevant to brain function remains unknown.	[41,42]
Fucosylated HMOs and neural routes	2′-FL/3′-FL	Rodent intestinal motility and vagotomy studies	Fucosylated HMOs modulate enteric motility, while selected effects of 2′-FL require an intact vagus nerve.	Route-specific intervention evidence	The intestinal sensor and cellular relay linking luminal 2′-FL to vagal activation remain unidentified.	[57–59]
Microbial polysaccharide-mediated immune signaling	PSA-TLR2-CD39+ regulatory T cells	PSA treatment in mouse experimental autoimmune encephalomyelitis	Bacterial PSA suppresses neuroinflammation through TLR2-dependent CD39+ regulatory T-cell responses.	Route-specific intervention evidence	The applicability of this pathway to other intestinal or neurological diseases remains uncertain.	[72,73]
Microbiota-associated brain N-glycosylation	Site-specific brain N-glycoproteins	Germ-free mice colonized with defined microbiota	Defined gut microbiota remodel site-specific brain N-glycosylation independently of total protein abundance.	Cross-organ association within one animal model; microbiota experimentally manipulated	The microbial products, host receptors, and transmission routes responsible for brain glycoproteome remodeling remain unknown.	[27]
Shared gastrointestinal-psychiatric genetic locus	FUT2 locus	Genome-wide pleiotropic analysis in humans	FUT2 is among loci shared by gastrointestinal and psychiatric traits.	Human observational or genetic association	The analysis does not establish directionality or show that intestinal glycosylation or microbiota mediate the shared association.	[35]
IgG Fc N-glycosylation	IgG Fc glycans	Human IBD and neurological disease cohorts	Altered IgG glycosylation is associated with IBD and several neurological inflammatory conditions.	Human observational association with limited functional validation	Direct causality and a gut-to-brain role for the observed IgG glycan changes have not been established.	[78–84]
TCR core fucosylation	FUT8/TCR	Mouse colitis models and human IBD tissues	FUT8-dependent TCR core fucosylation promotes T-cell activation and colitis.	Local mechanistic evidence with human tissue validation	Its contribution to distal neuroinflammation in gut-associated disease has not been directly tested.	[85]
Microglial core fucosylation	FUT8/gp130/IL-6R	Chronic unpredictable stress mice with L-fucose intervention	Core fucosylation regulates gp130/IL-6R signaling, and L-fucose reduces neuroinflammation and depressive-like behavior.	Brain-intrinsic mechanistic evidence	The study does not establish that microbiota-derived or intestinal L-fucose supplies the brain pathway.	[86]
Brain endothelial glycocalyx O-glycosylation	Core 1 mucin-type O-glycans	Aged mice with brain-endothelial glycan restoration	Restoring brain endothelial core 1 O-glycans improves BBB function and reduces age-related neuroinflammation.	Brain-intrinsic mechanistic evidence	Whether gut-derived microbial, metabolic, or immune signals regulate these endothelial glycan changes remains unclear.	[22]
Astrocytic proteoglycan glycosylation	DMP1-PG	DMP1 glycosylation-site mutant mice	DMP1-PG glycosylation at Ser89 supports astrocyte maturation and BBB integrity.	Brain-intrinsic mechanistic evidence	No direct link to intestinal glycosylation or microbiota-derived signals has been established.	[98]
HMO intervention in humans	2′-FL-containing interventions	Randomized infant and adult human trials	2′-FL interventions are tolerated and alter microbiota composition, whereas neural benefits remain preliminary.	Human intervention evidence	Responders, effective structures and doses, and mediation by microbiota remain incompletely defined.	[120,121,130]

Abbreviations: 2′-FL, 2′-fucosyllactose; 3′-FL, 3′-fucosyllactose; BBB, blood-brain barrier; C3GnT, core 3 β1,3-N-acetylglucosaminyltransferase; CD39, ectonucleoside triphosphate diphosphohydrolase 1; c-Fos, Fos proto-oncogene protein; CSDS, chronic social defeat stress; DMP1-PG, proteoglycan form of dentin matrix protein 1; ERK1/2, extracellular signal-regulated kinases 1 and 2; Fc, fragment crystallizable region; FUT2, α1,2-fucosyltransferase 2; FUT8, α1,6-fucosyltransferase; gp130, glycoprotein 130; HMOs, human milk oligosaccharides; IBD, inflammatory bowel disease; IgG, immunoglobulin G; IL-6R, interleukin-6 receptor; IL-22, interleukin-22; ILC3, group 3 innate lymphoid cell; LT, lymphotoxin; MUC2, mucin 2; PSA, polysaccharide A; PULs, polysaccharide utilization loci; TCR, T-cell receptor; TLR2, Toll-like receptor 2; VIP, vasoactive intestinal peptide; VIPR1, vasoactive intestinal peptide receptor 1.

Future studies should combine pathway-specific manipulation with direct glycan measurements. Tissue-specific deletion or restoration of glycosyltransferases may separate intestinal, immune, and neural effects from developmental or systemic changes. CRISPR-based glycoGene libraries can identify genes that regulate defined glycan features, ¹³¹ while CRISPR-engineered organotypic tissues provide a strategy for examining the tissue-level effects of specific glycosylation defects. ¹³² Similar approaches could be adapted to intestinal and neural organoids. Gnotobiotic models, microbiota transplantation, and rescue experiments can determine whether microbial communities or individual metabolites are required for downstream neural phenotypes. Stable-isotope tracing has also been used to distinguish the contribution of different fucose sources to GDP-fucose pools and glycosylation reactions. ¹³³ Applying this approach to paired intestinal and brain tissues may help determine whether gut-derived substrates contribute to glycosylation in distal organs.

7.2. Emerging technologies for studying glycosylation in the gut-brain axis

Spatial glycomics can preserve anatomical information that is lost during bulk-tissue analysis. Matrix-assisted laser desorption/ionization mass spectrometry imaging has been used to map regional N-glycan distributions in mouse and human brain tissue. ¹³⁴ Applied to paired intestinal and brain samples, this approach could determine whether glycan changes occur in specific epithelial regions, immune-cell-rich areas, vascular structures, or neural regions. Automated resources such as NGlycDB in METASPACE can improve the annotation and comparison of spatial N-glycomics datasets. ¹³⁵

Single-cell methods provide complementary information about cellular heterogeneity. SUGAR-seq and single-cell glycan and RNA sequencing combine DNA-barcoded lectins with transcriptomic measurements to profile glycan features in individual cells. ^{136 , 137} These methods can identify cell populations with altered lectin-reactive glycans, but they do not provide complete site-specific glycan structures. They should therefore be combined with cell sorting and mass-spectrometry-based glycomics or glycoproteomics. At present, these approaches are more accurately described as single-cell glycan profiling rather than mature site-specific single-cell glycoproteomics.

Quantitative glycoproteomics provides information on glycosylation sites and glycoforms. DQGlyco has identified extensive site-specific glycosylation heterogeneity and demonstrated that defined gut microbial communities remodel the mouse brain N-glycoproteome independently of changes in total protein abundance. ²⁷ Tools such as pGlycoQuant can further improve intact glycopeptide quantification across samples. ¹³⁸ Glycoinformatics is also required for consistent structural annotation and comparison across studies. ²⁶ GlyTouCan assigns stable identifiers to glycan structures, GlyGen integrates glycan information with proteins and diseases, and NGlycDB supports annotation of spatial N-glycomics data. ^{135 , 139 , 140} Combining spatial, single-cell, quantitative, and informatic approaches may help assign glycosylation changes to specific cells, anatomical regions, and experimental conditions.

7.3. Translational prospects and current limitations

Among the glycan-defined interventions discussed in this review, HMOs, particularly 2′-FL, currently have the clearest path toward clinical translation. Their chemical composition and dose can be standardized, and randomized infant studies support acceptable growth and tolerability of 2′-FL-containing formulations. ¹³⁰ In older adults, 2′-FL supplementation increased Bifidobacterium abundance, but the primary immune endpoint was not met. Improved visual-memory performance was mainly observed in participants whose microbiota responded to supplementation. ¹²⁰ The EFFICAD trial also reported changes in microbial composition and stress-related mood measures, although responses to 2′-FL alone were heterogeneous and several outcomes were more pronounced with oligofructose or the combined intervention. ¹²¹

These studies support the feasibility of HMO supplementation and its capacity to alter the gut microbiota, but they do not yet establish efficacy for neurological or psychiatric disorders. Responses may differ according to age, baseline microbiota, disease state, HMO structure, and dose. Future trials should prespecify neural or behavioral outcomes and include microbiome, metabolite, and glycomic measurements to determine which participants respond and which biological changes accompany the clinical outcome. Probiotics and complex dietary glycans may also be useful, but their evaluation is complicated by strain specificity, structural heterogeneity, and dependence on the existing microbial community. Translation should therefore prioritize well-defined interventions, reproducible biological measurements, and clinical endpoints that are directly related to the proposed mechanism.

Conclusion

This review summarizes current evidence that glycosylation contributes to several processes relevant to gut-brain communication, including intestinal barrier function, microbial glycan metabolism, immune regulation, BBB properties, and neural plasticity. The strength of evidence differs across these processes. Causal evidence is relatively strong for selected local mechanisms, including glycan-dependent microbial colonization, ENS regulation of epithelial fucosylation, and vagus-dependent effects of 2′-FL. By contrast, the intermediate signals and cell types linking many intestinal glycosylation changes to brain outcomes remain unresolved. Brain-to-gut regulation of intestinal glycosylation is also supported by a limited number of stress and neural manipulation studies and requires further direct testing. As spatially resolved and single-cell glycomics technologies continue to mature, and as causal relationships are further examined in both animal models and human studies, nutrition- and microbiota-based strategies grounded in glycobiology can be evaluated on a clearer mechanistic basis. Such progress is expected to deepen our understanding of the links between gastrointestinal and neuropsychiatric disorders and to inform the rational design of future intervention approaches.

Funding Statement

This work was supported by the Scientific Research Project of Jiangsu Provincial Health Commission (No. MQ2024030); Changzhou Sci&Tech Program (Nos. CJ20245049, CJ202510012); Scientific Research Project of Changzhou Health Commission (Clinical Frontier Technology) (No. QY202505); and the Clinical Medicine Special Research Fund of Nantong University (Nos. 2025JZ014, 2025JZ017).

Disclosure of potential conflicts of interest

No potential conflicts of interest were disclosed.

Data availability statement

No datasets were generated or analyzed during the current study.

Ethics approval

Not applicable.

Abbreviations

1,2-PD

1,2-propanediol

2′-FL

2′-fucosyllactose

3′-SL

3′-sialyllactose

AhR

aryl hydrocarbon receptor

AMPA

α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid

aNSPCs

adult neural stem/progenitor cells

ATCC

American Type Culture Collection

BBB

blood-brain barrier

BDNF

brain-derived neurotrophic factor

C3GnT

core 3 β1,3-N-acetylglucosaminyltransferase

CD4

cluster of differentiation 4

CD39

ectonucleoside triphosphate diphosphohydrolase 1

CD45

cluster of differentiation 45

CD90

cluster of differentiation 90

c-Fos

Fos proto-oncogene protein

COVID-19

coronavirus disease 2019

CRISPR

clustered regularly interspaced short palindromic repeats

CSDS

chronic social defeat stress

DMP1-PG

proteoglycan form of dentin matrix protein 1

DNA

deoxyribonucleic acid

DQGlyco

deep quantitative glycoproteome

ENS

enteric nervous system

ERK1/2

extracellular signal-regulated kinases 1 and 2

Fc

fragment crystallizable region

FUT2

α1,2-fucosyltransferase 2

FUT8

α1,6-fucosyltransferase

GDP-fucose

guanosine diphosphate fucose

gp130

glycoprotein 130

HMOs

human milk oligosaccharides

IBD

inflammatory bowel disease

IgG

immunoglobulin G

IL-22

interleukin-22

JAK2

Janus kinase 2

LTP

long-term potentiation

MUC2

mucin 2

MUC13

mucin 13

NEU1

neuraminidase 1

OGT

O-GlcNAc transferase

PSA

polysaccharide A

PSA-NCAM

polysialylated neural cell adhesion molecule

RNA

ribonucleic acid

SCFAs

short-chain fatty acids

SIgA

secretory immunoglobulin A

STAT3

signal transducer and activator of transcription 3

ST6GAL1

β-galactoside α2,6-sialyltransferase 1

TCR

T-cell receptor

Tregs

regulatory T cells

TREM2

triggering receptor expressed on myeloid cells 2

VIPR1

vasoactive intestinal peptide receptor 1

ZO-1

zonula occludens-1