Can an antibody tell one sugar from a nearly identical one?
Why the default outcome is a cross-reactive antibody
A glycan is not very immunogenic by itself, so it has to be presented on a carrier. That conjugation introduces its own epitopes, and antibodies against the linker or the carrier are easy to raise and useless to you. Meanwhile the glycan you care about shares its building blocks with a dozen related structures in the same sample.
Screening cannot fix this reliably, because by the time you are screening, the pool is already dominated by clones that bind something adjacent. The discrimination has to be built into the selection.
How we do it
Conjugated antigens for immunisation and panning, then subtractive selection with the related glycans present as competitors. Clones surviving that pressure have been required to distinguish your target from its nearest neighbours as a condition of being recovered at all. Cross-reactivity is then confirmed against the panel of related structures you specify.
- Glycan- or PTM-specific clones with sequences
- Cross-reactivity data against the related structures you named
- Recombinant material, recombinant and therefore reproducible batch to batch
Related assets in-house include O-GlcNAc reagents. We describe programmes we have actually run rather than claiming a class we have not; if your target is far from anything we have done, we will say so when you send it.
Where this is used
- Tumour-associated carbohydrate antigens, where the difference from the normal structure is small and the specificity requirement is absolute.
- Sulfation and other charged modifications on receptor N-termini, where the modification changes function.
- Bacterial and fungal polysaccharides for diagnostics, where related species share most of the structure.
- Glycosylation quality control, where a specific glycoform has to be measured against a background of others.
Need specificity against a structure nothing commercial resolves?
Send the target structure and the ones it must be told apart from. A PhD scientist reads it and replies within two business days. No NDA needed to start.
Talk to a scientist →Frequently asked questions
Why are glycans harder targets than protein epitopes?
Three reasons. They are poorly immunogenic on their own, so they need conjugation. They are structurally repetitive, so related glycans look similar to an antibody. And they are flexible, which costs binding energy. The default outcome of a naive campaign is a cross-reactive antibody, not a specific one.
How do you get specificity against a closely related sugar?
By subtracting against it during selection rather than screening for it afterwards. Related glycans are included as soluble competitors during panning so clones that cannot discriminate are removed before screening starts.
Can you make an antibody specific to a sulfation site?
Yes, and we have published it. A sulfation-site-specific monoclonal bound the fully sulfated CCR5 N-terminal peptide at 0.81 nM with no measurable binding to the non-sulfated peptide, and recognised full-length sulfated CCR5 on the cell surface.
Can the antibody tell which residue carries the modification?
In the published CCR5 case, yes. That antibody distinguished not only sulfated from non-sulfated but which tyrosine carried the sulfate, with roughly a hundredfold affinity difference between sulfation patterns.
What do you need to assess feasibility?
The structure of the target glycan or modification, the structures it must be distinguished from, and the assay it has to work in. That is enough for a scientist to give you a view before you commit.