Abstract
Alzheimer’s disease (AD) is a devastating neurodegenerative disorder marked by progressive cognitive decline. Metabolic disruptions are widely observed, yet their involvement in the molecular aetiology of AD remains underexplored. Here we identify hyperglycosylation as a driver of AD. Integrating spatial metabolomics, lipidomics and glycomics in transgenic AD mouse models and post-mortem human AD samples, along with advanced spatial isotopic tracing pulse-chase analysis of N-linked glycans, we demonstrate that the conserved phenotype of brain hyperglycosylation is driven by increased glycan biosynthesis. Genetic knockdown of glycan biosynthetic enzymes improves cognitive outcomes in AD mice whereas oral glucosamine supplementation impairs them. A retrospective analysis of electronic health records from patients with AD with varying disease severity shows that glucosamine supplementation is associated with accelerated AD progression and worsened survival. Overall, these results establish hyperglycosylation as a pathological driver of AD and highlight glycan metabolism as an actionable target in the fight against AD.
Similar content being viewed by others
Main
Alzheimer’s disease (AD) is a complex neurodegenerative disorder characterized by progressive cognitive impairment, synaptic dysfunction, neuroinflammation and widespread neuronal loss1. Despite extensive research efforts elucidating the underlying pathophysiology, effective disease-modifying treatments remain elusive2. Increasing evidence implicates metabolic dysfunction as a key component of AD pathology, with alterations in glucose metabolism3, mitochondrial function4 and lipid homeostasis5 playing contributory roles in disease progression. Fluorodeoxyglucose positron emission tomography imaging has revealed measurable reduced glucose uptake in AD brains6, which is detectable years before clinical symptoms manifest, underscoring the importance of metabolic impairments in early disease pathology. Concurrently, lipid droplet accumulation has been identified as a metabolic hallmark of AD, suggesting that lipid dysregulation and energy homeostasis are tightly linked to neurodegeneration7. Recent studies demonstrating that restoring glucose metabolism can attenuate AD progression in animal models highlight the growing recognition of metabolic regulation as a critical frontier in AD research8. To this end, perturbations in the entire metabolome have not been comprehensively explored in AD. For example, complex carbohydrate metabolism, such as glycan biosynthesis and processing, remains elusive in the context of AD. Understanding the overall metabolic shifts that occur in AD is essential for our understanding of AD disease progression.
The dynamic metabolome as a hallmark and driver of AD pathophysiology has not been defined. The interconnected role of metabolites as not only sources of energy but as biosynthetic precursors in anabolic reactions remains unclear. In particular, complex carbohydrate metabolism including glycan biosynthesis and processing in AD remains poorly understood. N-linked glycosylation is a fundamental post-translational modification crucial for maintaining protein stability, intracellular trafficking and receptor–ligand interactions critical for brain homeostasis9. In the central nervous system, glycosylation plays a vital role in synaptic plasticity10, neurotransmitter-receptor function11 and neuroimmune signalling12, making it integral to neuronal communication and resilience. Glycan metabolism is tightly linked to glucose13 and glucosamine availability14, forming a critical intersection between cellular energy balance and post-translational modifications. Connecting pathways, including the hexosamine biosynthetic pathway and glycan salvage pathways15, regulate glycosylation processes, further influencing protein folding, trafficking and degradation. Aberrant glycan processing can lead to altered protein stability and dysfunctional cell signalling, contributing to neurodegenerative mechanisms16. Notably, congenital disorders of glycosylation, caused by germline mutations in glycosylation-related genes, invariably result in neurological impairments, underscoring the essential role of glycosylation in brain development and function17. Alterations in glycosylation patterns have been documented across various neurological disorders16, suggesting a broader role for glycan modifications in neurodegeneration.
Recent research in AD has revealed a complex interplay between multiple pathological processes, including the functional diversity of microglia subtypes18, tau propagation19, amyloid-beta/tau crosstalk20, neuroinflammation21 and lipid metabolism dysregulation5. Aberrant glycosylation could integrate or synergize with these pathological features, influencing multiple facets of AD progression. N-linked glycans are a crucial component of brain homeostasis as they regulate protein stability, mediate intercellular communication, modify immune signalling, control intracellular trafficking and receptor–ligand interactions9 and regulate the blood–brain barrier22. Disruptions in glycosylation can impact microglial function23, altering inflammatory responses and phagocytic capacity24. In addition, glycan modifications influence tau post-translational modifications and aggregation dynamics25, potentially modulating tau propagation across neural networks. In the context of amyloid-beta pathology, glycosylation changes may alter plaque composition and clearance mechanisms26. Furthermore, glycan interactions with lipid metabolism contribute to lipid droplet dynamics27, influencing neuronal resilience to metabolic stress. Rather than acting as an isolated phenomenon, glycosylation represents an integrated metabolic event that intersects with multiple AD pathologies. Advances in spatial metabolomics, lipidomics and glycomics provide powerful tools to dissect these intricate connections, offering new insights into the role of glycan dysregulation in AD pathogenesis.
Building upon these advancements, we hypothesized that hyperglycosylation is a pathological hallmark of AD that actively contributes to disease progression. While prior studies have suggested glycosylation changes in AD28, the functional implications of altered glycan biosynthesis remain unclear. To address this gap, we employed an integrative approach combining spatial metabolomics29,30 and advanced spatial isotopic tracing31 to elucidate glycan dynamics in AD mouse models and human post-mortem brain tissue. Using pulse-chase labelling of complex glycans, we uncovered increased glycan biosynthesis in AD brains, indicative of a metabolic shift towards hyperglycosylation. Given the extensive involvement of glycans in maintaining brain function, these findings raise critical questions regarding the influence of glycosylation on AD homeostasis and disease pathophysiology. To investigate the functional impact of hyperglycosylation, we conducted both genetic and dietary interventions aimed at modulating glycan biosynthesis. Genetic knockdown of key glycosylation enzymes ameliorated cognitive deficits in AD mice, while oral glucosamine supplementation exacerbated behavioural impairments, supporting a causal role for glycan dysregulation in disease severity. These findings suggest that hyperglycosylation is not simply a secondary feature of neurodegeneration but may function as a critical driver of AD pathology. To extend these insights to human populations, we performed a retrospective analysis of patients with mild cognitive impairment (MCI) and dementia stratified by glucosamine supplementation using electronic health records (EHRs)32. Consistently, glucosamine use was associated with increased mortality and a higher likelihood of progression from MCI to AD, emphasizing the potential risks of excessive glycan biosynthesis in disease advancement. In summary, our findings establish hyperglycosylation as a significant driver of AD progression and establish the potential of targeting glycan biosynthesis as a novel therapeutic strategy.
Results
Hyperglycosylation in human AD specimens
Recent advances include a multiplexed matrix-assisted laser desorption ionization (MALDI)–mass spectrometry imaging (MSI) approach to conduct spatial metabolomics, lipidomics and glycomics on a single mouse brain tissue section30. We applied this methodology to human AD brain samples. When comparing normal and AD human brain tissues, we observed markedly low glycan signal intensity with poor spatial resolution (Supplementary Fig. 1). Notably, lipid content remained elevated in human samples even following Carnoy’s solution treatment (ethanol, acetic acid and chloroform in a 60%:30%:10% ratio), a protocol conventionally sufficient for lipid removal in mouse brain tissue and widely utilized in MALDI-based analyses33. To enhance spatial N-glycome detection, we investigated alternative lipid extraction methodologies to improve glycan signal recovery in human brain specimens. We implemented an optimized xylene wash protocol (1-h washes, repeated three times), which proved to be the most effective lipid removal method and significantly enhanced glycan signal intensity (Supplementary Fig. 1a–c). Sequential MALDI imaging across multiomics datasets confirmed that this approach facilitated robust spatial glycomics analysis, allowing for a more comprehensive assessment of glycan distribution with a 10–20-fold increase in glycan intensities in AD and control tissues (Supplementary Fig. 1b,c). Using this optimized multiomics workflow, we performed spatial metabolomics, lipidomics and glycomics in a cohort of human frontal cortex samples from healthy controls and patients with AD (n = 3 each), matched for age, sex and post-mortem interval (PMI; Fig. 1a). Metabolomics and lipidomics analyses revealed specific AD-associated molecular alterations, including both upregulated and downregulated species (Supplementary Figs. 2 and 3). However, spatial glycomics analysis uncovered a particularly striking pattern: glycan abundance was dramatically elevated across both white and grey matter regions in AD samples (Fig. 1b and Supplementary Fig. 4). Further, both bisecting (1,688 m/z) and high mannose (1,419, 1,581 and 1,905 m/z) glycans are elevated in the grey matter region of the AD brain (Fig. 1c, Supplementary Fig. 4). Additional spatial metabolomics analysis revealed a significant reduction in N-acetylglucosamine and glucosamine-6-phosphate, both critical precursors in N-glycan biosynthesis (Fig. 1d). The unexpected and widespread increase in glycosylation underscores the need for mechanistic studies to elucidate the biochemical pathways driving this altered glycan profile and its implications in AD pathology.
a, A schematic overview of the multiplexed MSI workflow utilized for integrated spatial metabolomics, lipidomics and glycomics from human brain tissue sections. Representative images for metabolite (UMP), lipid (PA(18:0/22:6)) and glycan (m/z 1,485) distributions in AD and normal control brains are shown. b, A heat map illustrating the differential abundance of glycan species identified by spatial glycomics analysis, comparing grey and white matter regions from control (ctrl; normal) and AD brain specimens (n = 3, each). Hierarchical clustering is performed on glycan species only not on samples. c, Spatial glycomics visualization and quantification of representative N-glycan species (1,688 m/z and 1,419 m/z) in grey matter regions. The violin plots show significantly increased abundance of these glycans in AD relative to normal controls. Statistical analysis was performed using two-tailed t-tests (n > 2,000 pixels per group); exact P values are indicated. d, Spatial metabolomics analysis in grey matter regions reveals a significant reduction in key glycan precursors, acetylglucosamine and glucosamine-6-phosphate, in AD compared with control samples. Violin plots present relative abundances; statistical significance assessed by two-tailed t-tests (n > 2,000 pixels per group); exact P values are indicated. Panel a created in BioRender. Sun, R. https://biorender.com/0c7yclk (2026).
To investigate the mechanistic basis of hyperglycosylation in AD, we first sought to validate this phenotype in a broader cohort of human AD specimens. We assembled an additional set of AD patient samples, ensuring matched controls for age and PMI, while stratifying individuals by Braak staging (Fig. 2a,b). This stratification allowed us to assess whether hyperglycosylation follows a temporally regulated trajectory throughout disease progression. We utilized formalin-fixed, paraffin-embedded (FFPE) patient samples (Fig. 2a). Spatial glycomics workflows incorporating PNGase-mediated glycan release and MALDI–MSI enabled the quantification of regional glycan distribution matched to histological annotation (Fig. 2a,b). Analysis of glycosylation in both grey and white matter revealed distinct patterns of stage-dependent alterations. In grey matter, a heat map of glycan abundance demonstrated a progressive increase in glycosylation across Braak stages, with the most pronounced changes observed in later stages (Fig. 2c). Violin plots of glycan structures at 1,688 m/z and 1,409 m/z further illustrated these trends, with a steady increase in glycosylation across Braak stages (Fig. 2d). Spatial heat maps confirmed these findings, showing localized glycan accumulation patterns that intensified with AD pathology (Fig. 2e). White matter glycosylation exhibited a different pattern, with glycan abundance increasing at Braak stages 1–2 but not persisting through later stages (Extended Data Fig. 1a). A heat map of glycan abundance highlighted this transient elevation, suggesting that hyperglycosylation in white matter may be an early but non-progressive event. Violin plots of glycan structures at 1,079 m/z and 1,905 m/z indicated an early stage rise in white matter glycosylation without sustained accumulation at later stages (Extended Data Fig. 1b,c). These findings suggest that hyperglycosylation in grey matter is more progressive and associated with AD severity. These results validate the translational impact of the hyperglycosylation phenotype and suggest implicated N-glycan modifications as a potential driver of AD pathology. The next critical step is to delineate the metabolic mechanisms underlying this phenotype.
a, A schematic illustrating the workflow for spatial glycomics using FFPE human brain samples, incorporating PNGase treatment followed by CHCA matrix application for MALDI imaging. b, Representative histological sections indicating anatomical delineation of grey and white matter regions from human brain tissues stratified by Braak stages (0, 1–2, 3–4 and 5–6). c, A heat map displaying hierarchical clustering of glycan abundance specifically in grey matter regions across Braak stages, illustrating progressive glycan accumulation correlated with advancing disease severity. d, Violin plots representing quantitative glycan analysis of two specific glycans (1,688 m/z and 1,409 m/z) in grey matter across Braak stages. Data indicate a progressive increase in glycan abundance with disease advancement. e, Spatial glycomics images demonstrating the distribution and intensity of glycans (1,688 m/z and 1,409 m/z) in human brain samples across Braak stages. Spatial intensity increases correlate with disease severity, prominently highlighting regional glycan accumulation. Panel a created in BioRender. Sun, R. https://biorender.com/8cq7de3 (2026).
Hyperglycosylation in mouse AD brains
In an effort to investigate the metabolic mechanisms underlying hyperglycosylation, we sought to determine whether the 5xFAD and PS19 mouse models of AD replicate the hyperglycosylation phenotype observed in human AD brains. To address this, we performed spatial metabolomics, lipidomics and glycomics in 9-month-old wild-type (WT), 5xFAD and PS19 mouse brains (Fig. 3a), corresponding to a stage of significant disease manifestation.
a, A schematic representation of the multiomics workflow employed for spatial metabolomics, lipidomics and glycomics analysis on fresh frozen brain tissue sections from WT and 5xFAD mice, illustrating representative metabolite, lipid and glycan spatial distributions. b, Comparative heat maps of metabolite, lipid and glycan abundances in distinct brain regions (cortex, hippocampus, cerebellum, thalamus and hindbrain) from WT and 5xFAD mice. WT levels are set to 1 while the log2 fold change (FC) values indicate significant alterations across regions, with pronounced increases in glycan abundance specifically observed in 5xFAD brains (n = 3 for each genotype). c, Spatial glycomics visualization of a representative high mannose glycan (m/z 1,257) distribution, demonstrating marked elevation in the 5xFAD model compared with WT. Corresponding violin plots quantify glycan abundance across multiple brain regions, confirming statistically significant increases in 5xFAD mice. Statistical analysis performed using two-tailed t-tests; exact P values are indicated.
Consistent with human AD samples, we identified widespread alterations in the metabolome and lipidome of 5xFAD and PS19 mice (Fig. 3b and Extended Data Figs. 2a and 3), corroborating previous studies that have linked AD pathology to disrupted energy metabolism and lipid homeostasis34. Notably, N-glycan profiling revealed the same hyperglycosylation phenotype both in the 5xFAD mouse model as well as the PS19 model, similar to those observed in human AD specimens (Fig. 3b,c and Extended Data Figs. 2c,d and 3a). Further, global spatial analysis demonstrated that hyperglycosylation was regionally specific, being most pronounced in the cortex, hippocampus and thalamus, while being less prominent in the cerebellum and hindbrain (Fig. 3c and Extended Data Figs. 2c,d and 3b). These findings suggest that hyperglycosylation preferentially affects regions associated with memory, cognitive processing and neuroinflammation, aligning with known patterns of neurodegeneration in AD. Taken together, these results indicate that hyperglycosylation is a conserved feature of AD pathology across species and suggest that specific brain regions may be differentially affected by this metabolic dysregulation.
Increased glycan biosynthesis in mouse models of AD
Having established that hyperglycosylation is a conserved phenotype in both human AD brains and major regions of the 5xFAD mouse model, it is critical to elucidate the underlying metabolic mechanisms driving hyperglycosylation. Hyperglycosylation can arise from either increased glycan biosynthesis within the endoplasmic reticulum (ER) and Golgi apparatus35 or reduced degradation and recycling through the lysosomal pathway15. To distinguish between these possibilities, we employed a stable isotope tracing approach coupled with a pulse-chase experiment36. The pulse phase provides a direct measure of de novo glycan biosynthesis, whereas the chase phase enables an assessment of glycan turnover and salvage activity (Fig. 4a). To facilitate deep metabolic tracing, we implemented a 13C-enriched liquid diet37, an approach that allows extensive isotopic labelling of glycans given their terminal position in metabolic pathways. Unlike small metabolic intermediates, which incorporate relatively few labelled carbons, N-glycans can integrate up to 40–60 labelled carbon atoms derived from glucose, posing challenges in resolving individual isotopologues. To address this, we leveraged the ion mobility capabilities of our MALDI system31,38, which discriminates molecular species based on their shape and size in addition to mass. This approach enables accurate identification of 13C-labelled glycans, as isotopologues exhibit predictable mass shifts while maintaining identical collision cross-section values, facilitating precise molecular characterization (Fig. 4b).
a, A schematic of the stable isotope tracing experimental design used to differentiate glycan biosynthesis (pulse phase) from glycan degradation and recycling (chase phase) in WT and 5xFAD mice, employing a 13C-glucose enriched liquid diet. b, Ion mobility MSI (MALDI) showing 13C enrichment of the glycan species at 1,257 m/z, highlighting distinct isotopologue separation. c, The spatial distributions of unlabelled and 13C-labelled isotopologues of the glycan (m/z 1,257) across the mouse brain tissue section, demonstrating progressive labelling. d, Quantitative analysis comparing the enrichment levels of various isotopologues of glycan 1,257 m/z in WT (n = 3 animals) and 5xFAD (n = 3 animals) brains during the pulse phase, indicating significantly enhanced glycan biosynthesis in 5xFAD mice. Statistical significance determined by two-tailed t-tests. Data are mean ± s.e.m. e, Analysis of glycan isotopologue abundances following a 24-h washout period (chase phase), revealing no significant differences in glycan turnover between WT (n = 3 animals) and 5xFAD (n = 3 animals) mice. Data are mean ± s.e.m. f, A metabolic schematic illustrating the incorporation pathway of 13C-glucose into UDP-Hexose and UDP-HexNAc, key glycan biosynthetic precursors. g, LC–MS-based isotope tracing analysis confirming increased levels of labelled UDP-Hexose and UDP-HexNAc in 5xFAD (n = 3 animals) mice relative to WT controls (n = 3 animals), supporting elevated glycan biosynthesis. Data are mean ± s.e.m. P values are indicated. Statistical significance assessed by two-tailed t-tests. Panels created in BioRender: a, Sun, R. https://biorender.com/gj1e069 (2026); f, Sun, R. https://biorender.com/vjcm4pa (2026).
For the experimental groups, one cohort of mice was administered a 13C-labelled liquid diet for 24 h to measure biosynthetic flux (pulse phase), while a second cohort received the same diet followed by a 24-h washout period to evaluate glycan turnover (chase phase) (Fig. 4a). Spatially resolved MALDI imaging revealed distinct distributions of both native (M0) and labelled glycan species, confirming successful isotopic incorporation and metabolic tracing (Fig. 4b,c). A comparative analysis between WT and 5xFAD mice demonstrated a significant increase in glycan enrichment in 5xFAD mice, particularly in high mannose glycans such as the species with m/z 1,257 (Fig. 4d). This enrichment was evident across multiple isotopologues in 5xFAD mice relative to WT controls (Extended Data Fig. 4a–c), whereas no significant differences were observed following the 24-h washout period (Fig. 4e and Extended Data Fig. 4a–c). These data revealed that increased glycan biosynthesis, rather than impaired glycan degradation or salvage, is the predominant driver of hyperglycosylation in AD pathology.
To further substantiate these findings, we performed pooled liquid chromatography–mass spectrometry (LC–MS)-based isotope tracing from the same experimental cohorts. This analysis revealed a marked increase in labelled UDP-hexose and UDP-HexNAc (Fig. 4f,g), two key precursors in the N-glycan biosynthetic pathway39. To further confirm that glycan biosynthesis is increased in AD, we analysed the expression of glycan biosynthetic enzymes in both human and mouse AD models (Supplementary Figs. 5 and 6). Using RT–PCR and digital PCR, we quantified mRNA levels of key glycan biosynthesis genes in the ER and Golgi, including Mgat, Man1a2 and B4galt1. In both human AD brains and 5xFAD mouse models, we observed upregulation of these genes compared with their respective controls (Supplementary Figs. 5 and 6). These findings further support the notion that increased glycan biosynthesis is a key factor driving hyperglycosylation in AD.
UDP-GlcNAc functions as the shared substrate for multiple biosynthetic pathways, including N-glycosylation, O-GlcNAcylation and hyaluronic acid production (Extended Data Fig. 5a). To determine whether the observed increase in UDP-GlcNAc and N-linked glycosylation extends to these alternative downstream pathways, we quantified O-GlcNAcylation and hyaluronic acid levels in both WT and AD mouse models. For hyaluronic acid, we adopted previously established MALDI imaging methods based on hyaluronidase digestion to generate oligosaccharides that can be spatially resolved and quantified40 (Extended Data Fig. 5b). For O-GlcNAcylation, we performed immunofluorescence staining using a well-validated O-GlcNAc–specific antibody, followed by quantification of fluorescence intensity. Both analyses revealed consistent results showing that 5xFAD and PS19 mouse brains exhibited markedly decreased hyaluronic acid abundance (Extended Data Fig. 5c) and reduced O-GlcNAcylation (Extended Data Fig. 5d) compared with WT controls. These findings demonstrate that despite elevated UDP-GlcNAc availability, its utilization is not globally upregulated across all downstream pathways.
Glycoproteomic analysis
The observed hyperglycosylation phenotype raised the question of whether this increase is due to the emergence of novel glycosylated proteins or an upregulation of glycosylation modifications on pre-existing proteins. To address this, we conducted glycoproteomic analysis in both human and mouse samples. Similar to our glycan enzyme transcriptomics approach, we analysed pooled samples of human normal and AD frontal cortical tissues as well as WT and 5xFAD mouse brains. Given the necessity of enriching membrane glycoproteins before mass spectrometry analysis, we pooled n = 4 samples per group to ensure sufficient input material. Membrane proteins were first extracted and enriched using high-salt and carbonate-based fractionation, followed by sequential ultracentrifugation. Isolated membrane fractions were denatured, reduced, alkylated and subjected to protease digestion. Glycopeptides were subsequently analysed using nano-LC coupled to a high-resolution Thermo Eclipse mass spectrometer (Supplementary Figs. 7 and 8). Glycoproteomic datasets were processed using Proteome Discoverer (v3.0), with glycoproteins identified through Byonic database searches (Supplementary Data 1).
From mouse samples, we identified 137 unique glycopeptides (Supplementary Fig. 7), while human extracts yielded 77 unique glycopeptides (Supplementary Fig. 8). Relative quantification revealed a significant increase in glycopeptide abundance in AD samples compared with their respective WT controls (Supplementary Figs. 7b and 8b), further supporting the hyperglycosylation phenotype identified via MALDI. Notably, 55% and 75% of glycopeptides were shared between human and mouse samples (Supplementary Figs. 7c and 8c), respectively. Since multiple glycopeptides can be derived from the same protein, we observed over 90–95% overlap in glycoprotein identities between normal and AD samples (Supplementary Data 1). These findings indicate that hyperglycosylation in AD predominantly involves an increase in glycosylation modifications on pre-existing glycoproteins rather than the appearance of novel glycosylated proteins. Finally, we conducted cell-type enrichment analysis through WebCSEA41, based on the identified glycopeptides to infer which neuronal populations are most affected by hyperglycosylation. Our analysis suggests that neurons are the predominant cell population undergoing enhanced glycosylation, underscoring the potential role of aberrant glycosylation in neuronal dysfunction and AD pathogenesis (Supplementary Figs. 7d,e and 8d,e).
Genetic and dietary interventions of glycosylation during neurodegeneration
To determine whether hyperglycosylation is a causal driver of neurodegeneration or a secondary consequence of disease pathology, we tested both genetic and dietary strategies to selectively decrease and increase N-glycan levels in the brain. This approach allowed us to assess the functional impact of altered glycosylation on neurobehavioural outcomes and neuropathological markers in AD (Fig. 5a). To suppress glycosylation, we first targeted phosphoglucomutase 3 (PGM3), a key enzyme in the hexosamine biosynthetic pathway responsible for converting glucosamine-6-phosphate to N-acetylglucosamine-6-phosphate. Eight-month-old female 5xFAD and PS19 mice received either shScrambled or shPGM3 packaged in lentivirus delivered via stereotaxic injection into the ventricles for whole-brain distribution. Following a 2-week intervention period, we assessed glycan levels using spatial glycomics and performed neurobehavioural testing to evaluate cognitive outcomes (Fig. 5b). Spatial glycomics confirmed decreases in global (Supplementary Figs. 9 and 10) as well as representative N-glycans across multiple regions in both 5xFAD (Fig. 5d–e) as well as PS19 mice (Extended Data Fig. 6b,c). To assess the functional impact of these treatments, we performed a social memory test, a behavioural paradigm that measures recognition memory by tracking social interaction time across repeated exposures42. In this test, mice are expected to progressively reduce their interaction time with a familiar conspecific over four trials, reflecting intact social recognition memory (Fig. 5h). Notably, shPGM3-treated 5xFAD as well as PS19 mice exhibited significant improvement in social memory interaction patterns when compared with the shScr-treated mice (Fig. 5i and Extended Data Fig. 6e). To evaluate potential off-target consequences of reducing N-glycosylation in the absence of disease, we performed the same intracerebroventricular lentiviral shPGM3 intervention in age-matched WT mice (Extended Data Fig. 7a). Animals underwent social memory testing and spatial glycomics analysis at the experimental end point. Although we observed a modest reduction in global N-glycan levels (Supplementary Fig. 11), individual glycan changes did not reach statistical significance (Extended Data Fig. 7b). Importantly, social memory performance in WT mice with shPGM3 remained unchanged compared with shScr controls (Extended Data Fig. 7d). These findings indicate that targeted reduction of glycosylation does not impair cognitive behaviour under baseline physiological conditions.
a, A schematic of the hexosamine pathway and ER N-glycosylation, highlighting inhibition of PGM3 and OST (by NGI-1). b, The experimental design for 8-month-old 5xFAD mice injected with shScr or shPGM3 and tested for social memory 2 weeks later. c, The experimental design for 5xFAD mice treated with vehicle or NGI-1 and tested for social memory 2 weeks later. d,e, MALDI imaging and violin plots of representative two unique complex N-glycans (N-glycan structure shown on the left) showing reduced brain N-glycan abundance in 5xFAD-shPGM3 mice compared with 5xFAD-shScr controls. f,g, MALDI imaging and violin plots of representative two unique complex N-glycans (N-glycan structure shown on the left) showing decreased levels in NGI-1–treated 5xFAD brains compared with vehicle (veh)-treated 5xFAD mice. h, A schematic of the social memory test across four interaction trials. i, Social memory performance across trials showing improved memory retention in 5xFAD-shPGM3 mice (n = 5 animals) compared with 5xFAD-shScr mice (n = 4 animals). j, Social memory performance in NGI-1-treated (n = 5 animals) and vehicle-treated 5xFAD (n = 5 animals) mice showing partial behavioural rescue with NGI-1. Data are mean ± s.e.m. P values are indicated (two-tailed t-test for glycan comparisons; two-way repeated-measures ANOVA for behaviour). Panels created in BioRender: a, Sun, R. https://biorender.com/wa4lg3j (2026); b, Sun, R. https://biorender.com/n41h483 (2026); c, Sun, R. https://biorender.com/dzklqk3 (2026); h, Sun, R. https://biorender.com/n8icm0r (2026).
To further validate the effect of reducing N-glycosylation on memory function, we targeted the oligosaccharyltransferase complex using NGI-1, a small-molecule OST inhibitor that blocks N-glycan transfer to nascent proteins (Fig. 5a). A single intracerebroventricular administration of NGI-1 (2 nmol in 2 µl) was delivered into the lateral ventricle of 8-month-old 5xFAD mice. After a 2-week interval, animals underwent social memory testing followed by in situ spatial glycomics (Fig. 5c). NGI-1 treatment decreased brain N-glycan abundance (Fig. 5f,g and Supplementary Fig. 12) and improved social recognition performance compared with vehicle controls (Fig. 5j). These data reinforce that lowering excessive N-glycosylation is sufficient to enhance cognitive outcomes and highlight hyperglycosylation as an actionable target for dementia-related memory decline in mice.
Glucosamine is known to readily cross the blood–brain barrier and incorporate directly into brain glycans43, making it an ideal substrate for investigating the effects of increased glycosylation (Fig. 6a). If hyperglycosylation is an active contributor to AD pathology, we hypothesized that glucosamine supplementation would exacerbate neurodegenerative phenotypes, whereas PGM3 inhibition would have protective effects. A separate cohort of 5xFAD mice received daily oral gavage of glucosamine at 457 mg kg−1 day−1, with the dose calculated based on human therapeutic equivalent dose of ~2,500 mg day−1 or 36 mg kg−1 day−1 followed by a standard human to mouse dose calculation44 (Fig. 6b). Conversely, glucosamine supplementation significantly increased glycosylation across the brain, as shown by global heat map analysis (Supplementary Fig. 13) and representative spatial mapping of key glycans (Fig. 6c,d). By contrast, glucosamine-treated 5xFAD mice exhibited a further exacerbation of social memory deficits, showing no recognition of previously encountered conspecifics across trials (Fig. 6e).
a, A schematic of glucosamine entry into the hexosamine pathway, increasing N-acetyl-glucosamine and ER protein N-glycosylation. b, Experimental design: 8-month-old 5xFAD mice receive glucosamine or vehicle for 2 weeks followed by social memory testing. c,d, MALDI imaging and violin plots of representative two unique complex N-glycans (N-glycan structure shown on the left) showing higher brain N-glycan abundance in glucosamine-treated 5xFAD mice compared with water-treated controls. P values are indicated (two-tailed t-test). e, Social memory performance across four trials demonstrating impaired memory in glucosamine-treated 5xFAD mice (n = 6 animals) relative to vehicle-treated 5xFAD mice (n = 7 animals). Data are mean ± s.e.m. P values are indicated (two-way repeated-measures ANOVA for behaviour). f, A schematic of EHR cohort construction from non-dementia controls, MCI and dementia diagnoses (AD and ADRD) with identification of glucosamine exposure from medication records. g, The proportion of patients with documented glucosamine use in the MCI and ADRD cohorts. h, Ten-year all-cause survival curves for patients with ADRD with or without glucosamine use, showing reduced survival in glucosamine users. Data are mean ± 95% confidence intervals. i, Ten-year all-cause survival in patients with MCI, showing no significant difference between glucosamine users and non-users (log-rank P = 0.252). Data are mean ± 95% confidence intervals. j, The cumulative incidence of transition from MCI to AD over time, revealing a higher conversion rate among glucosamine users compared with matched non-users. Data are mean ± 95% confidence intervals. Panels created in BioRender: a, Sun, R. https://biorender.com/9uxxwcu (2026); b, Sun, R. https://biorender.com/ouipb70 (2026);f, Sun, R. https://biorender.com/1bixssx (2026).
To further investigate neuropathological changes, we performed immunofluorescence analysis of reactive astrocytes and Aβ staining to quantify β-amyloid plaques (Extended Data Figs. 8 and 9). Interestingly, neither shPGM3, NGI-1 nor glucosamine treatment impacted the number of reactive astrocytes or the abundance of β-amyloid plaques (5xFAD) or tauopathy (PS19) during the duration of the experiment, while resulting in behavioural benefit (Extended Data Figs. 8 and 9 and Supplementary Fig. 14). While STT3A protein levels did not increase in 5xFAD or PS19 mice when compared with normal controls (Supplementary Fig. 15), we did observe a mild increase in STT3A and ALG13 protein levels when examined by immunofluorescence following glucosamine supplementation (Supplementary Fig. 16), suggesting substrate-level regulation of these enzymes. These findings demonstrate that glycosylation is an actionable target for AD-related neurocognitive deficits. The improvement in social memory following glycan biosynthesis inhibition suggests that hyperglycosylation contributes functionally to AD-related behavioural impairments. Conversely, the worsening of social memory with glucosamine supplementation supports the notion that elevated glycan biosynthesis exacerbates disease pathology. Collectively, these results establish a causal role for hyperglycosylation in AD-associated cognitive dysfunction and highlight the therapeutic potential of targeting glycan metabolism to mitigate neurodegenerative progression.
Real-world impact of glucosamine in the dementia patient population
Given that glucosamine exacerbates cognitive deficits in the 5xFAD mouse model, we sought to determine whether this effect extends to human populations. As glucosamine is a widely available over-the-counter supplement commonly used for joint health, it is plausible that some dementia patients may be using it regularly. To investigate whether glucosamine usage influences clinical outcomes in AD, we analysed clinical records from the University of Florida Health (UF Health) system, identifying over 50,000 patients diagnosed with Alzheimer’s disease-related dementias (ADRDs). Using physician notes and prescription records, we identified patients who had documented glucosamine usage for at least 1 year following a dementia diagnosis. MCI, clinically defined as a prodromal phase rather than a definitive dementia diagnosis, served as a practical pre-dementia comparison group. Further, since cognitively healthy individuals tend to have fewer healthcare visits, the MCI group provides a more appropriate comparator group for those with confirmed dementia (Fig. 6f). After applying inclusion criteria, we identified 24,481 patients with ADRD and 41,884 patients with MCI for the survival analysis. From this group, the median follow-up duration was 1,835 days (~5 years). In total, 6,764 patients with ADRD and 4,651 patients with MCI had a recorded death date, while 1,896 patients with ADRD and 2,750 patients with MCI received glucosamine.
To ensure the accuracy and consistency of our glucosamine exposure classification, we first assessed the percentage of patients within each diagnostic category (AD, ADRD and MCI) who had documented glucosamine usage. Across all three groups, approximately 8% of patients were identified as glucosamine users, confirming that physician notes provide a reproducible and reliable method for classifying supplement use (Fig. 6g). Next, we performed a 10-year survival analysis in all three cohorts, adjusting for age, sex and demographics (Supplementary Fig. 17). Our analysis revealed that glucosamine usage was associated with a 25% increase in mortality risk among ADRD patients (P = 0.0023; Fig. 6h). By contrast, glucosamine usage did not significantly impact mortality risk in the MCI cohort (Fig. 6i), suggesting that the effects of glucosamine may be specific to individuals with established neurodegeneration rather than the general ageing population.
To further assess whether glucosamine supplementation is associated with worsening dementia symptoms, we analysed the transition from MCI to ADRD. While most individuals with MCI remain stable or even recover, approximately 5% progress to ADRD, representing a clinical worsening of cognitive decline. Using similar survival and transition analyses, we quantified the rate of MCI-to-ADRD conversion among patients with or without documented glucosamine use. In this analysis, we observed a 25% increase in the proportion of patients with MCI transitioning to ADRD in the glucosamine user group (Fig. 6j), indicating that glucosamine supplementation is associated with accelerated disease progression or worsened disease phenotype.
Glucosamine supplementation in WT mice
Given that glucosamine supplementation did not impact survival in the MCI patient cohort, we sought to determine whether the absence of an effect in this population reflects an inherent resilience of the non-diseased brain to glucosamine-induced hyperglycosylation. To address this, we subjected 8-month-old WT mice to the same glucosamine treatment regimen as used in 5xFAD mice. Mice received daily oral gavage of glucosamine for 2 weeks, with the dose adjusted based on human-equivalent scaling to match prior preclinical interventions (Extended Data Fig. 7e). Following the treatment period, we performed spatial glycomics analysis to assess glycan abundance across brain regions, followed by behavioural testing to determine whether glucosamine altered cognitive function in non-diseased mice.
Strikingly, glucosamine supplementation did not result in hyperglycosylation in WT mouse brains (Extended Data Fig. 7f), in contrast to the robust increase observed in 5xFAD mice. Detailed glycomics analysis revealed no significant changes in N-glycan abundance across multiple glycan species (Extended Data Fig. 7f), suggesting that normal glycan homeostasis is maintained in the absence of AD pathology. Furthermore, we assessed social memory performance using a four-trial system in which mice were exposed to the same social stimulus multiple times. In cognitively intact mice, social interaction time is expected to decrease over successive trials, reflecting recognition of the familiar conspecific (Extended Data Fig. 7c). Both vehicle- and glucosamine-treated WT mice exhibited this expected pattern of behaviour, with progressive reductions in interaction time across trials. Importantly, there was no statistically significant difference between the two groups, indicating that glucosamine supplementation did not impair social recognition memory in WT mice (Extended Data Fig. 7g). This contrasts with the deficits observed in 5xFAD mice, supporting the notion that the normal brain possesses intrinsic resilience mechanisms that buffer against the metabolic perturbations induced by glucosamine supplementation.
Discussion
MALDI–MSI has proven to be a powerful tool for hypothesis-generating research, allowing for the spatial characterization of metabolic and molecular alterations directly in tissue samples31. This study utilized MALDI–MSI-based spatial multiomics to uncover a hyperglycosylation phenotype in AD brains. While such findings offer critical insights, it is essential to validate them using controlled experimental models to establish causality and mechanistic underpinnings. In this study, we tested two mechanistically distinct mouse models of AD, 5xFAD and PS19. Both showed robust increases in N-glycans across vulnerable brain regions. The convergence of these models, one driven by amyloid pathology and the other by tauopathy, demonstrates that hyperglycosylation is a shared metabolic signature of neurodegeneration and not confined to a single driver mutation. The integration of MALDI–MSI with stable isotope tracing and gene expression analysis allowed us to precisely delineate the metabolic origins of this phenotype, reinforcing the importance of spatial metabolomics in neurodegenerative disease research.
Protein glycosylation is an essential post-translational modification with profound implications for neuronal function14, and our findings highlight an intriguing hyperglycosylation phenotype that emerges as a hallmark of AD. Interestingly, we examined O-GlcNAcylation and hyaluronan production, both of which draw from the same UDP-GlcNAc pool as N-linked glycosylation, and observed a mild decrease in the brain for both pathways. This selective substrate channelling suggests a reprogramming of UDP-GlcNAc metabolism in AD, favouring expansion of complex glycosylation at the expense of cytosolic and extracellular glycan branches. Such compartment-specific redirection of hexosamine flux underscores a unique metabolic bias in AD brains that amplifies N-glycan biosynthesis while constraining other UDP-GlcNAc-dependent processes. Hyperglycosylation, driven primarily by increased biosynthesis rather than lysosomal salvage pathways, is particularly notable given the reported lysosomal pathway dysfunction in AD45. The presence of hyperglycosylation does not invalidate evidence of lysosomal defects but instead underscores a more complex scenario involving multiple points of metabolic disruption that collectively contribute to AD pathology, warranting further investigation. Indeed, the observed upregulation of glycosylation under glucose hypo-uptake conditions suggests a compensatory metabolic mechanism aimed at mitigating broader metabolic deficits. Speculatively, this may represent an adaptive cellular response attempting to preserve neuronal function or survival by modulating protein processing, trafficking and possible immune interactions through enhanced glycosylation.
Glycoproteomic analyses further support this hypothesis, indicating that the majority of glycosylated proteins identified in AD are neuronal membrane proteins critical for action potential propagation and synaptic transmission. Future research should thus explore these compensatory mechanisms more deeply, potentially revealing novel therapeutic targets or diagnostic markers associated with metabolic dysregulation in AD. A recent glycoproteomic study reported mixed site-specific changes in N-glycosylation in AD46, with increases on some glycopeptides and decreases on others. The study by Xu et al. even demonstrated slightly reduced N-glycan peptides and they hypothesized that aberrant sugar metabolism was the root cause46. Our work complements these findings by measuring the net abundance of released N-glycans across brain regions rather than peptide-level occupancy. A global rise in total N-glycans can occur even when individual glycopeptides decline due to protein-specific regulation or trafficking. These studies capture different layers of the same biology and reviewing them together provides a more complete view of how AD remodels protein glycosylation and downstream signalling across neurodegenerative dementia.
To test whether targeting hyperglycosylation is an actionable strategy against dementia, we utilized three approaches: (1) shRNA against PGM3, (2) a small-molecule inhibitor of OST and (3) glucosamine supplementation. While strategies one and two will reduce glycosylation, glucosamine supplementation will have the opposite effect. Unlike broad metabolic interventions that disrupt glucose utilization, targeting the hexosamine biosynthetic pathway through PGM3 and OST offers a more selective approach for N-linked glycosylation47. Both modalities reduced N-glycan levels in the AD mouse brains and markedly improved memory performance. Notably, our data indicate that neither OST inhibition nor shRNA-mediated inhibition of PGM3 significantly impacted neuroinflammatory markers such as astrocytosis or GFAP levels, nor did they substantially alter amyloid-beta formation in the mouse model. This suggests that targeting metabolic pathways might represent a viable therapeutic strategy independent of addressing neuroinflammation21 or amyloid clearance, or potentially as complementary to existing therapies. Supporting this perspective, a recent study demonstrated that pharmacological inhibition of IDO1 and subsequent restoration of glucose metabolism could effectively reverse cognitive symptoms in multiple mouse models of AD without directly affecting hallmark AD pathology8. It is especially encouraging to see that shPGM3 in WT mice did not drive a memory deficit, further supporting the idea that N-glycan pathways are unique therapeutic targets for dementia. Our findings reinforce this concept by highlighting N-glycan biosynthesis as a selective target for improving cognitive function. While no blood–brain barrier-permeable inhibitors of PGM3 currently exist, the identification of such compounds represents an important next step. Future studies should focus on developing and testing small-molecule inhibitors that selectively reduce pathological glycosylation in preclinical AD models, paving the way for potential clinical translation.
Glucosamine is a widely used dietary supplement that is less stringently regulated by the FDA than prescription drugs, and can vary considerably in quality and purity48. Our data suggest that glucosamine does not clinically worsen outcomes in individuals with MCI, but is associated with worse survival in patients with AD/ADRD. Across our real-world EHR analysis, glucosamine use was associated with a nearly 25% increase in mortality in the ADRD populations. It is important to consider these population-level data with human brain specimens and mouse AD models through genetic and dietary interventions. Together, they make a compelling argument that glucosamine is a negative driver of neurodegeneration. Interestingly, glucosamine usage did not lead to an increased mortality rate in the MCI population as a whole, a group that is clinically not classified as having dementia.
This suggests that the adverse effects of glucosamine may be contingent on an underlying neurometabolic state. Owing to this phenotype-specific vulnerability, we proceeded to evaluate the impact of glucosamine supplementation in WT mice as a model of the metabolically normal brain. As observed in the MCI patient population, glucosamine-treated WT mice did not exhibit elevated N-glycan levels nor impairments in social memory performance. These data collectively highlight a crucial distinction between the metabolic homeostasis of the normal brain and the dysregulated glycan biosynthetic machinery present in dementia. Our findings suggest that hyperglycosylation and the deleterious effects of glucosamine are not generalizable across all individuals but are uniquely positioned to exacerbate metabolic vulnerability and cognitive decline in the context of neurodegeneration. In the USA, an estimated 6.7 million individuals are currently living with AD and another 7 million individuals with ADRD. Given our calculation that approximately 8% of patients from our University of Florida datasets were using glucosamine, this suggests that potentially over 1 million patients may be unknowingly exacerbating their disease progression through glucosamine supplementation. These findings emphasize the urgent need for a well-designed, double-blind clinical trial to systematically evaluate the impact of glucosamine on cognitive decline in patients with dementia. Establishing a clear clinical framework for evaluating its effects in at-risk populations is crucial to determining future best practices.
This study underscores the broader potential of MALDI imaging and spatial biology in AD research. Incorporating cell-type-specific spatial information through emerging technologies such as CODEX, MIBI-TOF or single-cell spatial transcriptomics will further refine our understanding of how metabolic dysregulation intersects with cellular pathology. We expect hyperglycosylation to separately impact neuronal function, glial response, blood–brain barrier and glymphatic systems. In addition, the development of pharmacological interventions targeting the hexosamine pathway in animal models represents a promising avenue for therapeutic exploration. Finally, given our findings on glucosamine supplementation, we strongly advocate for a large-scale, double-blind clinical trial to definitively determine its impact on AD progression at the population level. This study provides a foundation for future translational efforts aimed at modulating glycan metabolism as a therapeutic approach in AD.
Methods
Chemicals, reagents, antibodies and cell lines
High-performance liquid chromatography (HPLC)-grade acetonitrile, ethanol, methanol, water, trifluoroacetic acid (TFA) and N-(1-naphthyl) ethylenediamine dihydrochloride (NEDC) were purchased from Sigma-Aldrich. α-Cyano-4-hydroxycinnamic acid (CHCA) matrix was purchased from Cayman Chemical. Histological-grade xylenes were purchased from Spectrum Chemical. Citraconic anhydride for antigen retrieval was obtained from Thermo Fisher Scientific. Recombinant PNGaseF Prime was obtained from N-Zyme Scientifics. Bruker IntelliSlides were purchased from Bruker Daltonics.
Human and mouse tissue sources and preparation
De-identified human brain tissue samples were obtained under IRB-approved protocols from the University of Florida Alzheimer’s Disease Research Center. Both fresh frozen and FFPE frontal cortex specimens from cognitively normal and AD individuals were utilized. PMIs, age, and sex were matched between groups, and Braak staging was conducted for all AD cases. Frozen tissue blocks were cryosectioned at 10-µm thickness and mounted onto indium tin oxide-coated conductive slides for downstream MALDI imaging and molecular analyses. Mice were housed in a climate-controlled environment with a 14-h light/10-h dark with temperature (18–23 °C) and humidity (50–60%) control. Water and solid diet were provided ad libitum throughout the study (Teklad #2018). The University of Florida Institutional Animal Care and Use Committee has approved all of the animal procedures under the protocol number IACUC202200000586. Mouse brain tissue was obtained from 5xFAD transgenic mice and WT littermate controls maintained under standard housing conditions with ad libitum access to food and water (JAX stock #008730). Female mice aged 8–9 months were used for all experiments to reflect a timepoint of established AD-like pathology. For stable isotope tracing experiments, mice received a 13C-labelled liquid diet for 24 h (pulse group) or a 24-h pulse followed by a 24-h washout period (chase group). In separate cohorts, mice were treated with glucosamine sulfate by oral gavage at 457 mg kg−1 day−1 for 2 weeks.
Lentiviral delivery and stereotaxic injections
To knock down glycan biosynthesis, short hairpin RNA targeting PGM3 (shPGM3, TL505905V; Origene) or control shScrambled lentivirus (TR30021V; Origene) was bilaterally injected into the lateral ventricles of WT, 5xFAD and PS19 mice. Injections were performed using a stereotaxic frame at coordinates relative to bregma (AP, −0.2 mm; ML, ±1.0 mm; DV, −2.2 mm). Lentiviral particles (1.0 × 109 TU ml−1) were delivered in a 2 µl volume (at a rate of 0.5 µl min−1) per ventricle using a Hamilton syringe and allowed to diffuse for 5 min post-injection before needle withdrawal. For the inhibition of OST, 2 μl of the small-molecule inhibitor NGI-1 (HY-117383, MedChemExpress; 5 nmol dissolved into PBS) was injected bilaterally into the lateral ventricles of 5xFAD mice at a speed of 50 nl min−1. The control mice were injected with PBS. After injection, the needle was left for 5 min before being withdrawn.
Sample collection and preparation
Mice were euthanized by cervical dislocation and decapitation. Immediately following euthanasia, the brains were surgically resected within 30 s. Mouse brains were dissected into two hemispheres sagittally to expose regions to be scanned by MALDI. The brain tissue was first rinsed in 1× PBS and subsequently rinsed twice with deionized water. The rinsed tissues were blotted dry, then slow-frozen over isopentane chilled with dry-ice for 7 min. Post-freezing, the samples were stored at −80 °C until further processing.
MALDI imaging for spatial metabolomics, lipidomics, glycomics and hyaluronan
Frozen brain tissues were sectioned using a Leica CM1860 cryostat and mounted onto a frozen chuck with OCT. Sections of 10-µm thickness were collected, thaw-mounted onto Bruker IntelliSlides and stored at −80 °C. The sectioning temperature was −20 °C for metabolomics, lipidomics and glycomics runs, and −15 °C for hyaluronan runs. Before matrix application, slides were fixed by vacuum desiccation for 1 h. Each slide was sequentially processed for spatial metabolomics, lipidomics, glycomics, glycogen and hyaluronan imaging. For metabolomics and lipidomics, an NEDC matrix (7 mg ml−1 in 70% methanol) was applied using an HTX M5 sprayer with 14 passes at 30 °C and 10 psi, with a spray rate of 0.06 ml min−1 and tray heated to 50 °C. Following metabolite and lipid imaging, slides were washed with ice-cold 100% methanol for 5 min to remove NEDC and fixed overnight in 10% neutral-buffered formalin. For human brain FFPE sections, lipid removal was performed using three sequential 1-h xylene washes before enzymatic treatment. For glycomics and hyaluronan analysis, slides were processed through ethanol and water rinses followed by antigen retrieval in citraconic anhydride buffer (25 µl in 50 ml H2O, pH 3) at 95 °C for 25 min. Enzymes were applied using an HTX sprayer at 45 °C and 900 mm min−1 spray velocity. PNGase F (20 µg per slide in 0.2 ml) was used for glycan release. Hyaluronidase (Sigma-Aldrich H1136, 60 U per slide in 0.2 mM Tris–HCl pH 7.5 with 0.5 mM NaCl) was used for hyaluronan fragmentation. Slides were incubated at 37 °C for 2 h in a humidified chamber and then vacuum desiccated. CHCA matrix (0.04 g in 5.7 ml 50% acetonitrile/water with 5.7 µl 25% TFA) was applied before imaging. MALDI–MSI was performed on a Bruker timsTOF fleX using a 46 µm × 46 µm laser raster to produce 50 µm × 50 µm pixels. Image acquisition and set up were performed using the autopilot feature in flexImaging v6.0. A standardized flexImaging.mis file was applied across all runs to ensure consistent region masking and pixel alignment. All acquisitions used a 10,000 Hz laser frequency. Spatial metabolomics was acquired in negative ion mode (MS1, m/z 20–750, 90% laser power, 396 shots per burst, 30 V MALDI plate offset, −60 V deflection 1 delta, 200 Vpp Funnel 1 RF, 200 Vpp Funnel 2 RF, 200 Vpp Multipole RF, 7 eV collision energy, 700 Vpp collision RF). Lipidomics was acquired in negative ion mode (MS1, m/z 300–2,000, 80% laser power, 300 shots per burst; all other transfer parameters as metabolomics). Glycomics and hyaluronan imaging were both acquired in positive ion mode (MS1, m/z 700–4,000, 37% laser power, 320 shots per burst, 50 V MALDI plate offset, 70 V deflection 1 delta, 500 Vpp Funnel 1 RF, 500 Vpp Funnel 2 RF, 500 Vpp Multipole RF, 10 eV collision energy, 4,000 Vpp collision RF). Post-acquisition data were aligned and analysed using SCiLS Lab to extract spatially resolved ion distributions and intensities. Additional visualization was performed using MetaboAnalyst v5.0 and v6.0.
Stable isotope tracing and ion mobility mass spectrometry
To assess glycan biosynthetic flux, we employed a stable isotope tracing protocol using a 13C6-glucose-enriched liquid diet. Tissue sections from pulse and chase cohorts were analysed using MALDI–MS coupled with ion mobility separation to resolve isotopically labelled glycans. For trapped ion mobility spectrometry, data were acquired on a Bruker timsTOF fleX as described above, with isotopologue separation achieved through collision cross-section measurements. To further resolve and image N-glycan isotopologues, selected tissue sections were analysed using a Waters Synapt G2-Si high-definition mass spectrometer equipped with travelling wave ion mobility. The instrument was operated in positive ion mode over an m/z range of 500–3,000. A laser frequency of 1,000 Hz, energy of 200 a.u. and laser spot size of 75 µm were used. Ion mobility settings followed established protocols, with a trap entrance energy of 2 V, trap bias of 85 V and DC/exit of 0 V. Wave velocity settings were configured as follows: trap 9.6 m s−1, IMS 4.6 m s−1 and transfer 17.4 m s−1. Wave height settings included trap 4 V, IMS 42.7 V and transfer 4 V, with a variable wave ramp down of 1,400 m s−1. Data acquisition was performed using MassLynx v4.2, and images were generated with HDI software v1.5 (Waters Corp), utilizing the built-in peak integration function to correct for mass drift over the course of acquisition. All images were normalized to total ion current on a per-pixel basis.
LC–MS analysis of UDP-sugar precursors
Tissue samples were homogenized at 4 °C in 80% LC–MS-grade acetonitrile aqueous solution. Homogenates were centrifuged two to three times at maximum speed to remove cellular debris, and supernatants were collected into LC–MS vials. Samples were randomized and analysed in a blinded manner using a 1290 UHPLC system coupled to a 6550 iFunnel Q-TOF mass spectrometer (Agilent Technologies), operated with MassHunter software v10.1. Chromatographic separation was performed on a SeQuant ZIC-pHILIC column (150 × 2.1 mm, 5 µm, PEEK-coated). The mobile phases were 10 mM ammonium acetate in water, adjusted to pH 9.8 with concentrated NH4OH (solvent A) and acetonitrile (solvent B). The elution gradient was as follows: 0–15 min linear gradient from 90% to 30% B; 15–18 min at 30% B; 18–19 min linear gradient from 30% to 90% B; followed by 9 min re-equilibration at 90% B. The flow rate was 0.25 ml min−1 with an injection volume of 20 µl, and the column temperature was maintained at 25 °C. Electrospray ionization source parameters were dry gas 225 °C at 18 l min−1, sheath gas 350 °C at 12 l min−1, fragmentor voltage 175 V, nozzle voltage 500 V and capillary voltage ±3,500 V for positive and negative modes. Full scan data were acquired across an m/z range of 40–1,700 at 1 spectrum s−1 in profile mode. Data were processed using Profinder B.08.00 SP3 (Agilent Technologies) against an in-house library of authentic standards, with matching criteria set at 10 ppm mass tolerance and 0.5 min retention time tolerance. All peak integrations were manually reviewed for consistency.
RT–PCR and digital PCR for glycosylation enzymes
Total RNA was extracted from frozen brain tissue using a conventional protocol involving TRIzol reagent and chloroform separation, followed by overnight precipitation in isopropanol at −20 °C. The following day, DNA was denatured using phenol, and RNA pellets were washed, cleaned and resuspended in 10 mM Tris–HCl buffer (pH 7.5). RNA quality was assessed using a Nanodrop spectrophotometer, and aliquots (0.5–1.5 µg) were stored in single-use tubes at −80 °C for no longer than 1 week before downstream analysis. cDNA was synthesized using the iScript One-Step RT–PCR kit with reverse transcriptase (Bio-Rad) according to the manufacturer’s instructions. Thermal cycling conditions included priming at 25 °C for 5 min, reverse transcription at 46 °C for 20 min and enzyme inactivation at 95 °C for 1 min, followed by a 4 °C hold. Gene expression was quantified using commercially available 96-well RT–PCR arrays targeting glycosylation-related genes (Bio-Rad Glycosylation SAB Target Lists H96 and M96 for human and mouse, respectively).
Glycoproteomics analysis
Brain tissue was homogenized in high-salt buffer (2.0 M NaCl, 5.0 mM EDTA, pH 7.4, with protease inhibitors), passed through a 20-gauge needle and probe sonicated. Lysates were centrifuged at 45,000 rpm for 15 min at 4 °C. Pellets were resuspended in sodium carbonate buffer (0.1 M Na2CO3, 1.0 mM EDTA, pH 11.3), incubated on ice for 30 min and ultracentrifuged at 45,000 rpm for 45 min. Membrane-enriched pellets were dissolved in urea lysis buffer (8.0 M urea, 1.0 M NaCl, 4% CHAPS, 100 mM DTT, 200 mM Tris–HCl, pH 8.0) and denatured at 50 °C for 45 min, then alkylated with IAA (55 mg) in the dark for 45 min at room temperature. Proteins were precipitated by chloroform/methanol/water extraction, resuspended in 50 mM ammonium bicarbonate (pH 8.0), quantified by the BCA assay and 100 µg was digested with trypsin (1:20) at 37 °C for 24 h. Glycopeptides were analysed on an Ultimate 3000 RSLCnano coupled to a Thermo Eclipse mass spectrometer. Separation used 15 cm × 75 µm C18 columns (3 µm particles) with a 180-min acetonitrile gradient in 0.1% formic acid. MS1 was acquired at 120k resolution in the Orbitrap; precursors were selected for HCD fragmentation over 3-s cycles, with +1 and unknown charge states excluded and dynamic exclusion set to 30 s. Fragment ions were acquired at 30k resolution. EThcD was triggered on detection of the HexNAc oxonium ion (m/z 204). Data were analysed in Proteome Discoverer v3.0. Global proteomics used CHIMERYS search (sum PEP score >15). Glycopeptides were identified using Byonic (precursor/fragment tolerances: 5 ppm/10 ppm) with variable modifications for deamidation (N/Q), oxidation (M) and carboxymethylation (C), filtered at log probability >1 and ppm error <3. N-glycan libraries used were ‘N-glycan 182 human no multiple fucose’ (human) and ‘N-glycan 309 mammalian no sodium’ (mouse). O-glycans were identified using the ‘9-Most-common glycans’ library for both species. CHIMERYS and Byonic outputs were shared with collaborators for independent verification.
Behavioural testing by the social memory assay
Social recognition memory was assessed in the animal facility at the University of Florida using a previously established protocol. Mice were acclimated to the behaviour room for 30 min and subsequently placed in a clean rat cage, divided by a cardboard partition, for an additional 10-min habituation period. Experimental mice were randomly assigned to testing order and introduced to the same novel conspecific once daily over five consecutive days, with trials separated by 24-h intervals. Behavioural interactions were recorded during each session and quantified using DeepLabCut, an open-source deep learning platform for pose estimation. Frame-by-frame analysis was used to quantify the duration the experimental mouse spent within 1 cm of the novel conspecific during each trial.
Immunofluorescence and immunohistochemistry analysis
Brain tissue sections from mouse and human samples were fixed in 10% neutral-buffered formalin, paraffin embedded and stored before sectioning. Mice were killed via cervical dislocation and tissues were resected and fixed immediately. Human tissue samples were obtained from the University of Florida Alzheimer’s Disease Research Center under approved protocols. Paraffin-embedded sections (4 μm) were used for immunohistochemistry. Slides were dewaxed, rehydrated and subjected to antigen retrieval using Ventana CC2 buffer at 37 °C for 1 h. Primary antibodies included GFAP (astrocyte marker; 1:100, GTX85454; GeneTex), 6E10 (Aβ plaque marker; 1:100, 803015; Biolegend), Thr231 (phospho-tau marker, 1:300, MN1040; Thermo Scientific) STT3A (1:100,12034-1-AP; Thermo Scientific), ALG13 (1:100, 28738-1-AP; Proteintech), DPAGT1/GPT (1:100, ab214179; Abcam) and O-GlcNAc (1:100, MA1-072; Thermo Scientific). Primary and secondary antibody incubations were carried out at room temperature for 60 min each. H&E staining was conducted using standard protocols with hematoxylin, eosin, graded ethanol dehydration and cover slipping. Digital slide images were acquired using a ZEISS Axio Scan.Z1 platform. Quantitative image analysis was performed using HALO software (Indica Labs) with multiplex immunohistochemistry and tissue microarray modules to assess marker expression and pathology across matched regions of interest.
Antibody pre-adsorption assay
STT3A and ALG13 antibodies were further validated through the antibody pre-adsorption assay49 (Supplementary Fig. 17). To verify antibody specificity, an antigen blocking assay was performed before immunofluorescence staining. The primary antibodies anti-STT3A (1:100, 12034-1-AP; Proteintech) and anti-Alg13 (1:100, 28738-1-AP; Proteintech) were pre-incubated with a fivefold excess (by weight according to the manufacturer’s recommendation) of the recombinant human STT3A protein (Ag27072, Proteintech) and Alg13 protein (Ag30299, Proteintech) for 1 h at room temperature or overnight at 4 °C. As a positive control, the primary antibody was incubated under the same conditions without competing recombinant protein. Except for this blocking step, all subsequent procedures were performed following the same immunofluorescence protocol described above.
Primary neuron culture
Primary neurons were prepared from C57BL/6J mouse embryos at embryonic day 17–18 (E17–E18). Brain tissues were freshly dissected from embryos and immediately placed in cold Hank’s balanced salt solution (C14175500BT, Gibco). Tissues were treated with 0.25% trypsin (25200072, Thermo Fisher) at 37 °C for 15 min and then allowed to stand for 5 min. The supernatant was carefully removed, and the tissues were resuspended in DMEM (D6429-500ML, Sigma-Aldrich) supplemented with 10% fetal bovine serum (F2442-500ML, Sigma-Aldrich). A single-cell suspension was obtained by gentle trituration. Cells were plated in poly-D-lysine (A3890401, Thermo Fisher)-coated eight-well chamber slides and maintained at 37 °C in a 5% CO2 incubator. On the second day, half of the medium was replaced with Neurobasal (A3582901, Thermo Fisher) medium supplemented with B27 (A3582801, Thermo Fisher) and GlutaMAX (35-050-061, Fisher Scientific) and the medium was subsequently changed every 3 days.
Human EHR-based cohort identification and glucosamine stratification
We performed a retrospective cohort study using EHR data from 2012 to 2024 obtained through the University of Florida Health Integrated Data Repository (UF IDR), a centralized enterprise data warehouse integrating clinical data across UF Health systems. The UF IDR includes over 1 billion observational facts from more than 2 million patients. The study was reviewed and granted exempt status by the University of Florida IRB (IRB202001888). The study population included individuals with a diagnosis of MCI or AD and related dementias (AD/ADRD), including AD, vascular dementia, Lewy body dementia and frontotemporal dementia, identified by ICD-9-CM and ICD-10-CM codes. Glucosamine users were identified through a natural language processing pipeline that extracted glucosamine-related records from clinical notes using a regex-based keyword matching approach. Notes containing relevant glucosamine terms were flagged as exposures.
Survival and transition analyses from EHR data
Records with inconsistencies (for example, death dates preceding birth or diagnosis) were excluded. Patients were categorized into five groups: (1) AD, (2) MCI, (3) ADRD, (4) MCI-to-AD and (5) MCI-to-ADRD; groups 4 and 5 required MCI diagnosis to precede AD/ADRD. Within each group, patients were classified as glucosamine users (cases) or non-users (controls). Kaplan–Meier analyses estimated survival probabilities over a 15-year window, with subgroup analyses for all ages and those >50 years. Survival time was defined as the interval from first diagnosis to death; living patients were censored at the data extraction date (24 September 2024). Death dates were obtained from EHR records and the Social Security Death Index. To control for ADRD-related risk factors, propensity score matching used a one-nearest-neighbour algorithm to pair each case with a single control without replacement, balancing covariates across two sets: (1) age alone and (2) demographic variables. This produced a one-to-one matched dataset for survival analysis. MCI-to-ADRD transition rates were calculated and compared between groups. Analyses were implemented in Python v3.10.8 using NumPy v1.25.2, pandas v2.2.2, and lifelines v0.29.
Quantification and statistical analysis
All animals and data points are included for statistical analysis. No animals or data points were excluded from any analysis. Animals were assigned to experimental groups based on genotype and age. No further randomization was applied to group assignment. Within each experiment, samples were processed in a blocked design to minimize batch effects, with equal representation of experimental groups across processing runs. The order of sample analysis was not randomized but was kept consistent across replicates to ensure comparability. Data collection and analysis were performed blind to the conditions of the experiments. Investigators were unaware of sample identity and group assignment during data acquisition and quantification. Data distribution was assumed to be normal, but this was not formally tested. Statistical analyses were carried out using GraphPad Prism. All numerical data are presented as mean ± s.e.m. Column analysis was performed using one-way or two-way analysis of variance (ANOVA) or Student’s t-test. P values are indicated on each graph. The statistical parameters for each experiment can be found in the figures and figure legends. No statistical methods were used to pre-determine sample sizes, but our sample sizes are similar to those reported in previous publications29,31.
Reporting summary
Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.
Data availability
All raw MALDI datasets are available at https://sunlabresources.rc.ufl.edu or available upon request. Proteomics datasets are available at ProteomeXchange under accession number PXD075438. Source data are provided with this paper.
References
Crews, L. & Masliah, E. Molecular mechanisms of neurodegeneration in Alzheimer’s disease. Hum. Mol. Genet. 19, R12–R20 (2010).
Golde, T. E., Schneider, L. S. & Koo, E. H. Anti-Aβ therapeutics in Alzheimer’s disease: the need for a paradigm shift. Neuron 69, 203–213 (2011).
Butterfield, D. A. & Halliwell, B. Oxidative stress, dysfunctional glucose metabolism and Alzheimer disease. Nat. Rev. Neurosci. 20, 148–160 (2019).
Swerdlow, R. H. Mitochondria and mitochondrial cascades in Alzheimer’s disease. J. Alzheimers Dis. 62, 1403–1416 (2018).
Di Paolo, G. & Kim, T.-W. Linking lipids to Alzheimer’s disease: cholesterol and beyond. Nat. Rev. Neurosci. 12, 284–296 (2011).
Mosconi, L. et al. Multicenter standardized 18F-FDG PET diagnosis of mild cognitive impairment, Alzheimer’s disease, and other dementias. J. Nucl. Med. 49, 390–398 (2008).
Haney, M. S. et al. APOE4/4 is linked to damaging lipid droplets in Alzheimer’s disease microglia. Nature 628, 154–161 (2024).
Minhas, P. S. et al. Restoring hippocampal glucose metabolism rescues cognition across Alzheimer’s disease pathologies. Science 385, eabm6131 (2024).
Conroy, L. R., Hawkinson, T. R., Young, L. E. A., Gentry, M. S. & Sun, R. C. Emerging roles of N-linked glycosylation in brain physiology and disorders. Trends Endocrinol. Metab. 32, 980–993 (2021).
Martin, P. T. Glycobiology of the synapse. Glycobiology 12, 1R–7R (2002).
Scott, H. & Panin, V. M. The role of protein N-glycosylation in neural transmission. Glycobiology 24, 407–417 (2014).
Starossom, S. C. et al. Galectin-1 deactivates classically activated microglia and protects from inflammation-induced neurodegeneration. Immunity 37, 249–263 (2012).
Denzel, M. S. et al. Hexosamine pathway metabolites enhance protein quality control and prolong life. Cell 156, 1167–1178 (2014).
Sun, R. C. et al. Brain glycogen serves as a critical glucosamine cache required for protein glycosylation. Cell Metab. 33, 1404–1417 (2021).
Winchester, B. Lysosomal metabolism of glycoproteins. Glycobiology 15, 1R–15R (2005).
Freeze, H. H., Eklund, E. A., Ng, B. G. & Patterson, M. C. Neurological aspects of human glycosylation disorders. Annu. Rev. Neurosci. 38, 105–125 (2015).
Francisco, R. et al. The challenge of CDG diagnosis. Mol. Genet. Metab. 126, 1–5 (2019).
Keren-Shaul, H. et al. A unique microglia type associated with restricting development of Alzheimer’s disease. Cell 169, 1276–1290 (2017).
de Calignon, A. et al. Propagation of tau pathology in a model of early Alzheimer’s disease. Neuron 73, 685–697 (2012).
Nisbet, R. M., Polanco, J.-C., Ittner, L. M. & Götz, J. Tau aggregation and its interplay with amyloid-β. Acta Neuropathol. 129, 207–220 (2015).
Chen, X. et al. Microglia-mediated T cell infiltration drives neurodegeneration in tauopathy. Nature 615, 668–677 (2023).
Shi, S. M. et al. Glycocalyx dysregulation impairs blood–brain barrier in ageing and disease. Nature https://doi.org/10.1038/s41586-025-08589-9 (2025).
Han, L. et al. The role of N-glycan modification of TNFR1 in inflammatory microglia activation. Glycoconj. J. 32, 685–693 (2015).
Imbert, P. R. et al. An acquired and endogenous glycocalyx forms a bidirectional ‘don’t eat’ and ‘don’t eat me’ barrier to phagocytosis. Curr. Biol. 31, 77–89 (2021).
Losev, Y. et al. Differential effects of putative N-glycosylation sites in human Tau on Alzheimer’s disease-related neurodegeneration. Cell. Mol. Life Sci. 78, 2231–2245 (2021).
Perdivara, I. et al. Elucidation of O-glycosylation structures of the β-amyloid precursor protein by liquid chromatography−mass spectrometry using electron transfer dissociation and collision induced dissociation. J. Proteome Res. 8, 631–642 (2009).
Müller, G., Jung, C., Wied, S. & Biemer-Daub, G. Induced translocation of glycosylphosphatidylinositol-anchored proteins from lipid droplets to adiposomes in rat adipocytes. Br. J. Pharmacol. 158, 749–770 (2009).
Hawkinson, T. R. et al. In situ spatial glycomic imaging of mouse and human Alzheimer’s disease brains. Alzheimers Dement. 18, 1721–1735 (2022).
Ma, X. et al. AI-driven framework to map the brain metabolome in three dimensions. Nat. Metab. https://doi.org/10.1038/s42255-025-01242-9 (2025).
Harrison, A. C. et al. Spatial mapping of the brain metabolome lipidome and glycome. Nat. Commun. https://doi.org/10.1101/2023.07.22.550155 (2025).
Clarke, H. A. et al. Glycogen drives tumour initiation and progression in lung adenocarcinoma. Nat. Metab. https://doi.org/10.1038/s42255-025-01243-8 (2025).
Cowie, M. R. et al. Electronic health records to facilitate clinical research. Clin. Res. Cardiol. 106, 1–9 (2017).
Agar, N. Y. R., Yang, H. W., Carroll, R. S., Black, P. M. & Agar, J. N. Matrix solution fixation: histology-compatible tissue preparation for MALDI mass spectrometry imaging. Anal. Chem. 79, 7416–7423 (2007).
Panza, F. et al. Lipid metabolism in cognitive decline and dementia. Brain Res. Rev. 51, 275–292 (2006).
Helenius, A. & Aebi, M. Intracellular functions of N-Linked glycans. Science 291, 2364–2369 (2001).
Seravalli, J. & Ragsdale, S. W. Pulse-chase studies of the synthesis of acetyl-CoA by carbon monoxide dehydrogenase/acetyl-CoA synthase: evidence for a random mechanism of methyl and carbonyl addition*. J. Biol. Chem. 283, 8384–8394 (2008).
Sun, R. C. et al. Noninvasive liquid diet delivery of stable isotopes into mouse models for deep metabolic network tracing. Nat. Commun. 8, 1646 (2017).
Hawkinson, T. R. & Sun, R. C. Matrix-assisted laser desorption/ionization mass spectrometry imaging of glycogen in situ. In Mass Spectrometry Imaging of Small Molecules: Methods and Protocols (ed. Lee, Y.-J.) 215–228 (Humana Press, 2022).
Du, X.-L. et al. Hyperglycemia-induced mitochondrial superoxide overproduction activates the hexosamine pathway and induces plasminogen activator inhibitor-1 expression by increasing Sp1 glycosylation. Proc. Natl Acad. Sci. USA 97, 12222–12226 (2000).
Legouffe, R. et al. Hyaluronic acid detection and relative quantification by mass spectrometry imaging in human skin tissues. Anal. Bioanal. Chem. 414, 5781–5791 (2022).
Dai, Y. et al. WebCSEA: web-based cell-type-specific enrichment analysis of genes. Nucleic Acids Res. 50, W782–W790 (2022).
Winslow, J. T. Mouse social recognition and preference. Curr. Protoc. Neurosci. 22, 8.16.11–18.16.16 (2003).
Barondes, S. H. Incorporation of radioactive glucosamine into macromolecules at nerve endings. J. Neurochem. 15, 699–706 (1968).
Reagan-Shaw, S., Nihal, M. & Ahmad, N. Dose translation from animal to human studies revisited. FASEB J. 22, 659–661 (2008).
Piras, A., Collin, L., Grüninger, F., Graff, C. & Rönnbäck, A. Autophagic and lysosomal defects in human tauopathies: analysis of post-mortem brain from patients with familial Alzheimer disease, corticobasal degeneration and progressive supranuclear palsy. Acta Neuropathol. Commun. 4, 22 (2016).
Xu, X. et al. Comprehensive and site-specific characterization of protein N-glycosylation in AD samples reveals its potential roles in protein aggregation and synaptic dysfunction. Anal. Chem. 97, 21873–21882 (2025).
Kim, J. et al. The hexosamine biosynthesis pathway is a targetable liability in KRAS/LKB1 mutant lung cancer. Nat. Metab. 2, 1401–1412 (2020).
Zeisel, S. H. Regulation of ‘nutraceuticals’. Science 285, 1853–1855 (1999).
Bordeaux, J. et al. Antibody validation. BioTechniques 48, 197–209 (2010).
Acknowledgements
This study was supported by National Institute of Health (NIH) grants R01AG066653, R01CA266004, R01AG078702, R01CA288696 and RM1NS133593 to R.C.S.; R35NS116824 to M.S.G.; R35GM142701 to L.C.; and T32HL134621 to H.C. and T.M. C.S. was supported by T32HD043730. Z.L. and S.Q are supported by the MBI Gator NeuroScholar Program. S.Q. was supported by T32 CA257923 and P30 CA247796. The UF Neuromedicine Human Brain and Tissue bank is supported by the McKnight Brain Institute, the Center for Translational Research in Neurodegenerative diseases at the University of Florida and NIH grant no. P30AG066506. S.P. is the Charlotte and Howard Zimmerman Rising Star Professor with the Norman Fixel Institute for Neurological Diseases. Large language models, for example, ChatGPT, were used to make minor grammatical improvements in the text.
Author information
Authors and Affiliations
Contributions
Conceptualization by R.C.S. Methodology by R.C.S. Investigation by T.R.H., Z.L., T.M.M., R.S.N., R.A.R., H.C., X.M., A.C.M., A.F.P., A.L.S., S.T.S., C.M.S., J. Son, F.C., L.W., A.R.C., M.G.C., C.J.S., L.E.A.Y., D.D.K., C.L., A.M.R., P.A., R.J.D., Y.H., X.H., J.B., S.P., C.W.V.K., M.S.G., J. Sudderth, D.A., J.G., K.M.A., L.C., S.Q., S.Y., Y.G. and R.C.S. Writing—original draft by R.C.S. Writing—review and editing by R.C.S., R.J.D., C.W.V.K. and M.S.G. Funding acquisition by M.S.G. and R.C.S. Resources by M.S.G. and R.C.S. Supervision by C.W.V., M.S.G. and R.C.S.
Corresponding author
Ethics declarations
Competing interests
M.S.G. has research support and research compounds from Maze Therapeutics, Valerion Therapeutics and Ionis Pharmaceuticals. M.S.G. also received consultancy fee from Maze Therapeutics, PTC Therapeutics and the Glut1-Deficiency Syndrome Foundation. D.A. receives book royalty from Wolters Kluwer. The other authors declare no competing interests.
Peer review
Peer review information
Nature Metabolism thanks the Yasuhiko Kizuka and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. Primary Handling Editors: Yunling Xie and Alfredo Gimenez-Cassina, in collaboration with the Nature Metabolism team.
Additional information
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Extended data
Extended Data Fig. 1 Stage-dependent glycan accumulation in white matter regions across Braak stages.
a, Heatmap of glycan abundance in white matter across Braak stages 0, 1–2, 3–4, and 5–6, showing early-stage elevation of glycosylation in Braak 1–2 with no sustained increase in later stages. b, Representative histological and spatial glycomics images showing distributions of glycan species (1079 and 1905 m/z) in white matter across Braak stages. c, Violin plots quantifying relative abundance of 1079 and 1905 m/z glycans in both grey and white matter, revealing a transient increase in white matter glycosylation at early Braak stages without further elevation in late-stage disease.
Extended Data Fig. 2 Integrated spatial multiomics confirms glycan enrichment in 5xFAD mouse brains.
a, Spatial metabolomics and lipidomics maps of glucose and lysophosphatidylethanolamine (LPE 18:1) in wild-type (WT) and 5xFAD brains, with accompanying violin plots showing region-specific differences across cortex, hippocampus (HC), thalamus, cerebellum (CB), and hindbrain (HB). Significant reductions in glucose and elevations in LPE are observed in disease-relevant regions of 5xFAD brains (n>2000 pixels across 3 biological replicates each). Statistical significance assessed by two-tailed t-tests. b, Schematic of spatial multiomics workflow detailing integration of NEDC-based metabolomics and lipidomics with CHCA-PNGase-based glycomics from adjacent sections. c, Spatial glycomics images and violin plots of representative glycan species (1419, 2158, 2523, 1237 m/z) confirm significant glycan accumulation in 5xFAD compared to WT across brain regions. Two-tailed t-tests used for comparisons. d, MALDI mass spectra showing elevated glycan peak intensities across a wide m/z range in 5xFAD compared to WT mouse brains, confirming broad glycan upregulation. Panel b created in BioRender; Sun, R. https://biorender.com/o23cvvu (2026).
Extended Data Fig. 3 PS19 mice display widespread increases in brain N-glycans across regions.
a, MALDI imaging of representative complex N-glycans in WT and PS19 brains with corresponding violin plots of relative abundance (normalized to TIC) showing significantly higher levels in PS19 mice (P<0.0001, two-tailed t-test). b, Comparative heatmaps of metabolite, lipid, and glycan abundances in distinct brain regions (cortex, hippocampus, cerebellum, thalamus, and hindbrain) from WT and PS19 mice. WT levels are set to 1 while Log2 fold changes indicate significant alterations across regions, with pronounced increases in glycan abundance specifically observed in PS19 brains (n = 3 for each genotype).
Extended Data Fig. 4 Multiple glycans exhibit increased biosynthesis but unchanged turnover in 5xFAD mouse brains.
a, Spatial distribution and quantification of 13C-labeled isotopologues of glycan 1688 m/z following isotope pulse in WT and 5xFAD mice. Bar graphs show significantly increased 13C incorporation across isotopologues in 5xFAD. Washout after 24 hours reveals no significant difference in turnover between genotypes (n = 3 animals per group). Data are mean ± s.e.m. P values are indicated (two-way repeated-measures ANOVA). b, Enrichment and washout analysis of glycan 1282 m/z demonstrates elevated biosynthesis in 5xFAD with no change in degradation rates (n = 3 animals per group). Data are mean ± s.e.m. P values are indicated (two-way repeated-measures ANOVA). c, Similar enrichment profile observed for glycan 1444 m/z, showing significantly increased labeling in 5xFAD without altered clearance (n = 3 animals per group). Data are mean ± s.e.m. P values are indicated (two-way repeated-measures ANOVA).
Extended Data Fig. 5 Hyaluronan and O-GlcNAcylation are reduced in 5xFAD and PS19 mouse brains.
a, Schematic of the hexosamine pathway showing production of N-acetyl-glucosamine for N-glycans, hyaluronan and protein O-GlcNAcylation. b, Workflow for on-tissue hyaluronan imaging: brain sectioning, antigen retrieval, hyaluronidase treatment, CHCA matrix application and MALDI–MS analysis. c, Hyaluronan imaging and violin plots of absolute intensities showing decreased brain hyaluronan levels in 5xFAD and PS19 mice compared with WT controls. d, Immunofluorescence staining of O-GlcNAcylation in WT versus 5xFAD (top) and WT versus PS19 (bottom) brains, with quantification demonstrating reduced O-GlcNAcylation signal in both disease models (n = 9 animals per group). Data are mean ± s.e.m.; P values are indicated (two-tailed t-test). Panels created in BioRender: a, Sun, R. https://biorender.com/aht5d7z (2026); b, Sun, R. https://biorender.com/k69h363 (2026).
Extended Data Fig. 6 PGM3 knockdown reduces N-glycans and improves social memory in PS19 mice.
a, Experimental design: 8-month-old PS19 mice receive hippocampal injections of shScr or shPGM3 and undergo social memory testing 2 weeks later. b,c, MALDI imaging and violin plots of relative intensities for representative complex N-glycans showing reduced brain N-glycan levels in PS19-shPGM3 mice compared with PS19-shScr controls (P<0.0001). d, Schematic of the social memory paradigm across four interaction trials. e, Social memory performance across trials, expressed as time interacting relative to trial 1, showing improved performance in PS19-shPGM3 mice (n = 6 animals) compared with PS19 controls (n = 5 animals). Data are mean ± s.e.m.; P values are indicated (two-tailed t-test for glycan comparisons; two-way repeated-measures ANOVA for behavior). Panel a created in BioRender; Sun, R. https://BioRender.com/y4c87wr (2026); d, Sun R. https://biorender.com/n8icm0r (2026).
Extended Data Fig. 7 Hexosamine pathway modulation or glucosamine supplementation has minimal impact on brain N-glycans and social memory in WT mice.
a, Experimental design for 8-month-old WT mice injected with hippocampal shScr or shPGM3 lentivirus followed by social memory testing 2 weeks later. b, Relative intensities of representative complex N-glycans in WT-shScr (n = 3 animals) and WT-shPGM3 (n = 3 animals) brains showing only modest or non-significant changes. c, Schematic of the social memory paradigm across four interaction trials. d, Social memory performance expressed as time interacting relative to trial 1, showing no significant difference between WT-control (n = 6 animals) and WT-shPGM3 groups (n = 5 animals). e, Experimental design for WT mice treated with glucosamine or water (vehicle) for 2 weeks before social memory testing. f, Relative and absolute intensities of representative complex N-glycans in glucosamine- (n = 4 animals) and water-treated WT brains (n = 3 animals). g, Social memory performance of glucosamine- (n = 3 animals) and water-treated WT mice (n = 5 animals) showing no significant behavioral effect. Data are mean ± s.e.m.; P values are indicated (two-tailed t-test); ns, not significant. Panels created in BioRender: a, Sun, R. https://biorender.com/y4c87wr (2026); c, Sun, R. https://BioRender.com/bz0zgue (2026); e, Sun, R. https://biorender.com/kbdz76x (2026).
Extended Data Fig. 8 Glucosamine treatment and PGM3 knockdown do not induce astrocytosis or Aβ deposition in WT mice.
a, Immunofluorescence images of GFAP and Aβ in brains from WT mice treated with water or glucosamine, showing similar astrocyte staining and absence of Aβ signal between groups. b, GFAP and Aβ staining in WT mice injected with shScr or shPGM3 lentivirus, demonstrating comparable astrocyte morphology and no detectable Aβ accumulation in either condition. Experiments are repeated three times with the same results.
Extended Data Fig. 9 Modulating hexosamine pathway activity does not alter astrocytosis or amyloid plaque burden in 5xFAD mice.
a, Immunofluorescence images of GFAP and Aβ in hippocampus and cortex from 5xFAD-shScr and 5xFAD-shPGM3 mice. b, Quantification of reactive astrocytes and β-amyloid plaques in hippocampus and cortex showing no significant differences between 5xFAD-shScr (n = 3 animals) and 5xFAD-shPGM3 (n = 3 animals) groups. c, GFAP and Aβ staining in 5xFAD mice treated with water or glucosamine. d, Quantification of reactive astrocytes and β-amyloid plaques in hippocampus and cortex indicating similar levels between water- (n = 3 animals) and glucosamine-treated (n = 3 animals) 5xFAD mice. e, GFAP and Aβ staining in 5xFAD mice treated with saline or NGI-1. f, Quantification of reactive astrocytes and β-amyloid plaques in cortex and hippocampus showing no significant effect of NGI-1 (n = 5 animals per group). Data are mean ± s.e.m.; P values are indicated (two-tailed t-test).
Supplementary information
Supplementary Information (download PDF )
Supplementary Figs. 1–18 and figure legends.
Supplementary Data 1 (download ZIP )
N-glycan annotation tables for human (Ngly_human) and mouse (Ngly_mouse) datasets.
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.
About this article
Cite this article
Hawkinson, T.R., Liu, Z., Ribas, R.A. et al. Hyperglycosylation is a metabolic driver of Alzheimer’s disease. Nat Metab (2026). https://doi.org/10.1038/s42255-026-01538-4
Received:
Accepted:
Published:
Version of record:
DOI: https://doi.org/10.1038/s42255-026-01538-4
