What do you do when the epitope is inside a cleft?
When a single domain is the right answer
- The epitope is recessed. Enzyme active sites, receptor canyons and conserved pockets are hard to raise conventional antibodies against precisely because they are hard to reach. A protruding HCDR3 changes that geometry.
- You need a building block. Single domains are the natural unit for bispecifics, multivalent constructs, CAR binders and fusion formats. Starting from a VHH avoids the scFv assembly and stability problems that come with converting an IgG later.
- You need to block one thing without blocking its neighbour. A published example: our VHH against complement factor H blocked its binding to sialic acid while leaving its cofactor activity for C3b degradation intact. A full IgG would likely have occluded both.
- Size or penetration matters. Tissue penetration, intracellular formats and imaging applications all favour a small, stable, single-chain binder.
How we do it
We immunise camelids and pan a VHH library built from the immune repertoire. Our proprietary phagemid efficiently expresses and selects VHHs with extra-long HCDR3 loops and with cysteines in HCDR2 and HCDR3 — the structurally unusual clones that tend to be the ones that solve hard epitopes, and the ones that display poorly in less specialised systems.
That matters more than it sounds. In our published complement factor H work the winning clone carried a 24-residue HCDR3 and a framework cysteine occurring in under 1% of sequences, forming an intramolecular disulfide that locked the long loop into the one conformation that fit. No rational design process would have proposed that molecule. It had to be selected, from a library that could display it.
- VHH sequences
- BLI binding data (optional)
- VHH-Fc or format conversion where required
Published proof
Our VHH against complement factor H bound at 1.71 ± 0.66 pM by Biacore — against the most flexible loop in the target region, which is the opposite of the epitope antibodies normally prefer. The full story, including the crystal structure, is here.
On the engineering side, we have published on rescuing poorly soluble single-domain antibodies by editing framework hallmark residues — the failure mode most likely to derail a nanobody programme after discovery succeeds.
Have a target that needs a smaller binder?
Tell us the antigen, the epitope you are aiming at, and the final format. A PhD scientist reads it and replies within two business days. No NDA needed to start.
Talk to a scientist →Frequently asked questions
What can a VHH reach that a conventional antibody cannot?
Concave and recessed epitopes. A single domain is roughly a tenth the mass of an IgG and its long HCDR3 can protrude into clefts, enzyme active sites and canyons that a flat Fab paratope cannot access. Conserved epitopes are often recessed for exactly that reason, which is why they are hard to raise conventional antibodies against.
How small is a VHH and why does that matter?
About 15 kDa as a single domain, against roughly 150 kDa for an IgG. Beyond epitope access, the size matters for tissue penetration, for building multivalent and multispecific constructs, and for cases where you want to block one interaction surface without occluding a neighbouring one.
Can VHHs reach picomolar affinity?
Yes. In a published collaboration a VHH bound complement factor H at 1.71 pM by Biacore, against an epitope that molecular dynamics showed to be the most flexible loop in the region.
Can you convert the VHH to another format?
Yes. Selected VHHs are routinely reformatted to VHH-Fc or bivalent and multivalent constructs. Deciding the final format early is worth doing, because it can change which clones are worth advancing.
Can you humanize a VHH?
Yes, that is a separate engineering step we run routinely. Solubility engineering is often done at the same time, since framework changes affect both.