Single-Cell and Spatial Methods for Multimodal Functional Glycan Profiling in Tissues

This paper introduces two novel platforms, scGOAT-seq and GlycoScope, which utilize human lectins to integrate functional glycan accessibility into single-cell and spatial multiomic profiling, thereby revealing stimulus-specific glycan remodeling and spatially resolved glycan programs in immune cells and follicular lymphoma that were previously undetectable by existing methods.

The Gist

Imagine your body is a bustling city, and the cells are the citizens. For a long time, scientists have been able to read the "ID cards" (DNA) and the "diaries" (RNA) of these citizens to understand who they are and what they are doing. But there's a whole other layer of communication happening on the surface of these cells that we've been blind to: sugars.

These sugars (called glycans) are like the colorful flags, uniforms, or graffiti on the outside of the citizens. They tell other cells, "I'm a friend," "I'm a soldier," or "I'm a criminal." The problem is, these sugar flags aren't written in the genetic code, so standard tools can't read them. Furthermore, the city has its own security guards (proteins called lectins) that read these flags to decide whether to attack, ignore, or help a cell.

This paper introduces two new, high-tech tools that finally let us see these sugar flags and understand how the security guards are reading them, all while looking at individual cells and seeing exactly where they are standing in the city.

Here is a breakdown of the two new tools and what they discovered:

1. The Two New Tools

Tool A: scGOAT-seq (The "Single-Cell Sugar Scanner")

• The Analogy: Imagine taking a crowd of people, giving each person a tiny microphone, and asking them to shout out their name (their genes). Now, imagine you also attach a special "flag detector" to their collar that lights up if they are wearing a specific type of flag.
• How it works: The researchers created a set of "human security guards" (recombinant lectins) that are tagged with barcodes. They let these guards scan the surface of individual cells. If a guard sticks to a cell, it means that cell has a specific sugar flag. The barcode tells the computer exactly which flag was found, while the microphone records the cell's genetic diary.
• The Result: They can now see, for the first time, how the sugar flags on a single immune cell change when it gets angry (activated) or tired (exhausted), and how this connects to its internal instructions.

Tool B: GlycoScope (The "Sugar Map")

• The Analogy: scGOAT-seq is like interviewing people in a room, but you lose the context of the room. GlycoScope is like taking a high-resolution photo of the entire city block. It lets you see not just who has which flag, but where they are standing relative to their neighbors.
• How it works: This tool takes a slice of tissue (like a tumor) and uses the same barcode-tagged guards to scan the sugars right where they live, alongside the proteins. It creates a map showing the "neighborhoods" of sugar flags.
• The Result: They can see how sugar patterns organize the immune system inside a tumor, revealing hidden neighborhoods that standard maps miss.

2. What They Discovered

The "Sugar Mood Swings" of Immune Cells
The researchers looked at immune cells (T-cells) and found that their sugar flags change depending on what kind of alarm is ringing.

• The Analogy: Think of a T-cell as a soldier. When the alarm is a bacterial infection (LPS), the soldier puts up a "Red Flag" (Siglec-7 sugar). When the alarm is a viral infection or a specific immune trigger (TCR), the soldier switches to a "Blue Flag" (Siglec-9 or Siglec-15 sugar).
• The Discovery: Previously, scientists thought all "activated" soldiers looked the same. These tools showed that there are actually different types of activation, defined by these sugar flags. Some sugar patterns mean the cell is ready to kill; others mean it's just patrolling. This explains why some cancer therapies work on some patients but not others—they might be targeting the wrong "flag."

The "Sugar Neighborhoods" in Cancer
They used the "Sugar Map" (GlycoScope) to look at a type of blood cancer called Follicular Lymphoma.

• The Analogy: Imagine a tumor as a city where the bad guys (cancer cells) are hiding in a specific district. The researchers found that the cancer cells in this district were wearing a very specific "Mannose Flag" that attracted a specific security guard (DC-SIGN).
• The Discovery: The cancer cells were using this sugar flag to trick the security guard into protecting them and helping them survive. Even more interestingly, they found a different sugar pattern (Galectin-8) that created a "safe zone" for the cancer, where specific immune cells (T-cells) gathered around the cancer cells, almost like bodyguards. This "sugar neighborhood" was invisible to standard protein tests but was clearly visible with their new sugar map.

3. Why This Matters

For years, we've been trying to fight diseases like cancer and autoimmune disorders by looking at the "ID cards" (genes) and "uniforms" (proteins) of cells. But we were missing the flags (sugars) that actually tell the immune system what to do.

• Better Diagnostics: We can now find "sugar signatures" that predict if a patient will survive cancer or how their immune system will react.
• New Treatments: If we know that a cancer cell is hiding behind a specific sugar flag, we can design drugs to rip that flag off or trick the security guards into attacking the cancer.
• Understanding the Body: It helps us understand that the immune system is a complex language of sugars, not just a simple on/off switch.

In short: This paper gives us a new pair of glasses. With these glasses, we can finally see the colorful sugar flags on our cells and understand the secret language they use to talk to our immune system. This opens the door to smarter, more precise ways to treat diseases.

Technical Summary

1. The Problem

Cell-surface glycans are critical regulators of immune recognition, pathogen entry, and cancer progression. However, profiling glycans at single-cell and spatial resolutions remains a major bottleneck in modern biology for several reasons:

• Non-Genomic Encoding: Unlike proteins or RNA, glycans are not directly encoded by the genome, making them difficult to predict from transcriptomic data.
• Limitations of Existing Methods: Traditional mass spectrometry-based glycomics lacks single-cell resolution and spatial context. Existing single-cell glycan methods often rely on non-human (plant) lectins or probe only single glycan specificities, failing to capture how human endogenous lectins (the actual physiological readers) interpret glycan landscapes.
• Functional Gap: Current multimodal methods (scRNA-seq, spatial proteomics) cannot simultaneously measure gene expression, protein abundance, and functional glycan accessibility (i.e., how glycans are recognized by specific receptors in situ).

2. Methodology

The authors developed two complementary platforms, scGOAT-seq and GlycoScope, which utilize a curated panel of DNA-barcoded recombinant human lectins to measure functional glycan accessibility.

A. Lectin Panel Design

• Reagents: A panel of 7 human recombinant lectins was engineered, including inhibitory Siglecs (Siglec-7, -9, -15), context-dependent lectins (Galectin-8, -9, DC-SIGN), and the complement-recruiting MBL.
• Validation: Specificity was confirmed using Liquid Glycan Arrays (LiGA), flow cytometry with enzymatic perturbations (sialidase, EDTA), and competition assays. The panel showed minimal cross-interference in multiplexed conditions.
• Barcoding Strategy: Lectins were biotinylated and conjugated to poly(A)-tailed DNA barcodes via a streptavidin bridge. This allows them to be captured alongside mRNA in standard single-cell workflows.

B. Platform 1: scGOAT-seq (Single-Cell Glycan Outlining and Transcriptome Sequencing)

• Workflow: Cells are stained with the DNA-barcoded lectin panel, then processed using standard single-cell RNA-seq platforms (Seq-Well S3 or 10x Genomics 5').
• Readout: The DNA barcodes are sequenced alongside the transcriptome, providing a quantitative measure of lectin-accessible glycans for each cell alongside its gene expression profile.
• Application: Used on dissociated immune cells (PBMCs) and cancer cell lines to link glycan states to transcriptional programs.

C. Platform 2: GlycoScope (Spatial Glycan Profiling)

• Workflow: Adapted for Formalin-Fixed Paraffin-Embedded (FFPE) tissues. Lectins are conjugated to oligonucleotides compatible with CODEX-based spatial proteomics (PhenoCycler Fusion).
• Readout: Enables multiplexed, in situ co-detection of glycans and proteins (up to 42-plex) while preserving tissue architecture.
• Application: Used to map glycan distributions within the tumor microenvironment of Follicular Lymphoma (FL).

3. Key Contributions

1. First Human Lectin-Based Multimodal Profiling: Shifts the paradigm from using plant lectins to using human lectins, providing a physiologically relevant readout of how the immune system "sees" glycans.
2. Integration of Function and State: Successfully integrates functional glycan accessibility with transcriptomics (scGOAT-seq) and spatial proteomics (GlycoScope), revealing cell states invisible to RNA or protein markers alone.
3. Discovery of Glycan-Defined Cell States: Identified that specific lectin-binding patterns (e.g., Siglec-9 vs. Siglec-15) stratify immune activation states more effectively than gene expression of glycosylation enzymes.
4. Spatial Glycan Architecture: Demonstrated that glycan accessibility is spatially organized in tissues, defining distinct immune neighborhoods in cancer.

4. Key Results

A. scGOAT-seq Reveals Stimulus-Specific Glycan Remodeling

• Validation: In a mixed cell line model (pancreatic cancer vs. normal), scGOAT-seq results correlated highly with flow cytometry and were sensitive to enzymatic removal of sialic acids.
• Genetic Perturbation: Knocking out GNE (a key sialic acid biosynthesis enzyme) in CD8+ T cells reduced Siglec-7 ligand binding and upregulated inflammatory pathways, linking glycan biosynthesis directly to immune activation states.
• Immune Activation:
  • LPS (Innate) vs. TCR (Adaptive): Different lectin profiles emerged. Siglec-7 ligands increased with LPS but decreased with TCR stimulation. Conversely, Siglec-9 and Siglec-15 ligands increased with TCR stimulation.
  • Cell State Stratification: In CD4+ T cells, Siglec-15 ligand-high cells were associated with effector/activated states (cytotoxicity, TNF signaling), while Siglec-9 ligand-high cells were associated with metabolic states. These distinctions were not captured by transcriptomics alone.
• Blocking Experiments: Blocking Siglec-7/9 during LPS stimulation abolished expected binding shifts and rewired immune communication networks, shifting from monocyte-centered cytokine signaling to NK/T cell-driven interferon and cytotoxic programs. This blockade correlated with improved survival in Kidney Renal Clear Cell Carcinoma (KIRC) patients.

B. GlycoScope Maps Spatial Glycan Programs in Follicular Lymphoma (FL)

• Mannosylated BCR Axis: Confirmed the "mannosylated BCR" hypothesis in FL. Malignant B cells showed elevated binding to DC-SIGN (a mannose-binding lectin), which correlated with high BCL2 expression and proximity to Follicular Dendritic Cells (FDCs). This spatial relationship suggests FDCs sustain malignant B cell survival via lectin-glycan interactions.
• Galectin-8 Spatial Reorganization:
  • In reactive tissues, Galectin-8 binding was restricted to the mantle zone.
  • In FL, Galectin-8 binding expanded across the entire neoplastic follicle (center and mantle).
  • Immune Neighborhoods: Regions with high Galectin-8 accessibility on B cells were spatially co-localized with PD-1+ and TOX+ T cells, defining a unique glycan-associated immune niche not detectable by protein markers alone.
• Distance-Dependent Patterns: DC-SIGN binding on B cells declined with distance from FDCs, confirming a spatially structured signaling gradient.

5. Significance

• New Dimension in Multiomics: These methods add a "functional glycan" layer to the existing transcriptomic and proteomic multiomic frameworks, revealing biological insights that are otherwise hidden.
• Therapeutic Implications: The findings suggest that Siglec-targeted therapies may have differential effects depending on the specific glycan-lectin axis engaged (e.g., Siglec-7 vs. Siglec-9) and the inflammatory context.
• Biomarker Discovery: Functional glycan accessibility serves as a robust biomarker for immune activation states and tumor microenvironment organization, potentially improving patient stratification in cancer immunotherapy.
• Generalizability: The modular design of scGOAT-seq and GlycoScope allows for the expansion to other lectins, perturbations, and disease contexts, establishing a new standard for glycan biology in human tissues.

In summary, this work provides a robust, scalable framework to decode the "glycocode" in its native physiological context, bridging the gap between glycan structure, receptor recognition, and cellular function in health and disease.