GlycoStamp-seq Links Sialylation Dynamics to T-Cell Fitness

November 6, 2026

GlycoStamp-seq pairs metabolic labeling of newly synthesized sialoglycans with single-cell sequencing. In a bladder cancer mouse model, higher labeling signals were associated with activated CD8⁺ T-cell states, while functional experiments supported their value as a marker of antitumor capacity.

Introduction

Cell-surface glycans are not static decorations. They are rebuilt continuously as cells change state, and those changes are especially informative in the immune system, where sialic acids cap glycans that shape receptor engagement, checkpoint interactions, and cell-to-cell contact. Until recently, most single-cell glycan profiling captured only the glycans already present on the cell surface, a snapshot that hides the biosynthetic activity behind it. The new method turns the act of synthesis itself into a measurable readout, a shift that matters for anyone trying to read cellular state from glycosylation. Complementary Glycomics measurements can characterize glycan populations alongside metabolic-labeling experiments.

The central reagent is an unnatural sugar, tetraacetyl-N-azidoacetylmannosamine, commonly written as Ac4ManNAz. When cells take it up, they incorporate the azido-tagged analogue specifically into newly made sialic acids. The azide provides a handle for copper-free reaction with a DBCO-conjugated probe, allowing glycans labeled during a defined metabolic window to be detected. This metabolic labeling therefore marks nascent sialoglycans, the fraction of the glycocalyx that was being synthesized during the experiment rather than the pool that existed beforehand.

Turning Newly Made Glycans into Sequencing Reads

The design of GlycoStamp-seq converts the azido label into a DNA barcode that conventional single-cell platforms can already read. After the Ac4ManNAz pulse, the azido-decorated sialoglycans are reacted through bioorthogonal click chemistry with a single-stranded DNA handle carrying a capture sequence, a barcode identifying the DNA tag, and a polymerase chain reaction handle. During cell lysis in microwells, magnetic beads grab both the poly(A)-tailed messenger RNA and the DNA-tagged glycan signal from the same cell, so the two modalities are recovered together and indexed by a shared cell barcode during reverse transcription. The result is a dual-modality library in which each cell carries both a transcriptomic profile and a quantitative nascent sialoglycan-labeling readout.

Schematic of the GlycoStamp-seq workflow coupling Ac4ManNAz metabolic labeling with click chemistry and single-cell sequencing.

Fig. 1 Overall workflow of the GlycoStamp-seq strategy. (Wu, et al., 2026)

The authors validated the chemistry in several cancer cell lines before moving to tissue. They screened three unnatural azido sugars and found Ac4ManNAz to be the most efficient, then selected 50 μM for three-day labeling of T24, A549, and A549/DDP cells; concentrations up to 50 μM did not significantly reduce cell counts in those experiments. Global transcriptomic profiles were similar between labeled and control cells under the tested conditions, supporting compatibility with the RNA measurement without establishing an absence of all labeling effects.

Where Prior Single-Cell Glycan Methods Fell Short

Earlier single-cell glycan profiling relied on lectin panels, chemoenzymatic tags, or aptamers to read pre-existing epitopes. A Glycan Microarray Assay instead tests binding to immobilized glycans; it is distinct from lectin-based profiling of glycans on cells. GlycoStamp-seq adds a defined metabolic-labeling window to single-cell analysis. Its sequencing signal reflects nascent sialoglycan labeling accumulated during that window, rather than an instantaneous sialylation rate.

The platform also integrates cleanly with other single-cell multi-omics. The team coupled it with a time-resolved RNA method that distinguishes nascent from pre-existing transcripts, and with T-cell receptor sequencing, so that sialylation dynamics could be examined alongside clonal expansion and activation programs. A Glycoprotein Microarray Assay instead tests interactions with immobilized, characterized glycoproteins. Identifying the carriers of labeled cellular glycans would require complementary Glycoprotein Structure Analysis or site-specific glycoproteomics.

Reading the Tumor Immune Microenvironment

Applied to a mouse model of bladder cancer treated with oxaliplatin, GlycoStamp-seq revealed that immune cells in the tumor microenvironment carried markedly higher nascent sialoglycan levels than those from untreated controls. Among immune subsets, the increase was most pronounced in T cells. The authors annotated 4,672 cells into six immune populations and found that oxaliplatin expanded the T-cell fraction, consistent with the drug's ability to provoke immunogenic cell death and reshape the tumor milieu. Tumors from three mice were pooled per group for this profiling experiment, so individual cells should not be treated as independent animal replicates.

The signal was not uniform. When the team stratified CD8⁺ T cells by their sialylation intensity, the most heavily labeling cells were enriched for effector and proliferative effector states, for expanded T-cell receptor clonotypes, and for activation-associated genes such as Tnfrsf9, Entpd1, and Ifng. Conversely, genes tied to a naïve or suppressed phenotype correlated negatively with sialylation. Within this experimental setting, higher labeling was associated with antigen-reactive effector states; it was not established as a universal rule for CD8⁺ T-cell function.

Nascent sialoglycan labeling in CD8 positive T cells, with cell-state analysis, cytotoxicity assays, and adoptive-transfer validation.

Fig. 2 Nascent sialoglycan labeling, CD8⁺ T-cell states, and functional validation in OT-I tumor models. (Wu, et al., 2026)

The authors also tested the functional relevance of the marker. They sorted metabolically labeled OT-I CD8⁺ T cells into high- and low-labeling populations and compared killing of N4 peptide-loaded MB49 cells and MB49-OVA cells in vitro. In irradiated MB49-OVA tumor-bearing mice, adoptive transfer of high-labeling cells improved tumor control and survival relative to low-labeling cells. These experiments support the marker's association with effector capacity; they do not establish that increasing sialylation itself causes stronger antitumor activity.

Validation Against Established Readouts

The team used DBCO-Cy5.5 staining and DBCO-ssDNA followed by Cy5-(dT)25 detection as flow-cytometric checks of labeling. Such Flow Cytometry Profiling complements the sequencing readout. Transcriptome-based t-SNE analysis, rather than flow-cytometric clustering, showed similar cell distributions with and without the additional DNA-labeling step. Comparable RNA quality metrics and gene-expression profiles supported limited perturbation under the tested conditions.

What Metabolic Labeling Adds to Single-Cell Profiling

The broader implication is methodological. Most glycan measurements answer the question of what is on the cell surface at one moment; GlycoStamp-seq answers the question of what the cell is actively building. The labeling window provides information about newly incorporated sialic acids that cannot be inferred directly from measurements of the total pre-existing pool. By coupling the stamp to transcriptomics and T-cell receptor sequencing, the method places sialylation within a broader molecular context of cellular state.

For research applications, metabolic labeling offers an additional readout to compare with transcriptomic and surface-marker measurements. Applications in infection or autoimmunity would require validation in those settings, including assessment of probe delivery, labeling efficiency, and cell-state perturbation.

Limitations to Keep in View

The method reports total nascent sialylation and does not by itself resolve which glycan structures or which carrier proteins carry the label, so follow-up with structural analysis is still needed to interpret the biology. The Ac4ManNAz pulse also requires careful concentration control, because higher concentrations reduced cell counts in the tested cell lines and could bias which cells survive. It also does not simultaneously quantify pre-existing sialoglycans or distinguish specific sialic-acid linkage types. And because the readout is a DNA barcode count rather than a direct glycan sequence, the method complements rather than replaces dedicated Glycan Profiling of individual structures.

Outlook

GlycoStamp-seq adds a time-windowed metabolic readout to single-cell multi-omics. The mouse experiments support further study of nascent sialoglycan labeling as a marker of T-cell function, but its ability to predict clinical response or guide patient stratification remains untested.

Related Services & Products

Reference

  1. Wu, X., et al. (2026). GlycoStamp-seq: Single-cell nascent sialoglycometabolomics captures sialylation dynamics in immune cells of an oxaliplatin-treated bladder cancer mouse model. Nature Communications, Article in Press. DOI: 10.1038/s41467-026-78101-y.

Similar Posts