Nanobody discovery is one of the cleanest fits for yeast surface display in all of antibody engineering. A nanobody — the single variable domain (VHH) of a camelid heavy-chain-only antibody — is around 13 kDa, folds as one compact disulfide-bonded unit, and carries its entire paratope on three CDR loops. That small, robust architecture is exactly what yeast display handles best. This article walks through how a VHH discovery campaign actually runs on yeast, from library choice to delivered binders.
Why the VHH format and yeast display fit together
Most of the friction in antibody discovery comes from format. A full IgG is four chains and two domains per arm; an scFv is two domains tethered by a linker that can misfold or swap. A nanobody is one domain. The practical consequences on a display platform are large:
- Complete coverage in one library. A 120-residue VHH can be saturation-scanned across its whole sequence in a single library, so discovery and downstream optimization see the entire paratope at once.
- Robust surface expression. VHH domains tolerate the yeast secretory pathway and the Aga2p fusion well, giving high display levels and clean display-versus-binding two-color sorts.
- Quantitative readout. Because each yeast cell carries thousands of copies of one clone and is sorted by flow cytometry, binding signal is normalized to display level and ranked per clone — not measured in bulk as on phage.
For focused or pre-filtered libraries that fit comfortably under 10^9 members, yeast display is the more precise platform; the trade-off versus phage is maximum library size, which matters mainly for fully naive discovery. We covered that trade-off in phage display vs yeast display.
Library choice: synthetic vs immunized
The first real decision is where the diversity comes from.
Synthetic and semi-synthetic libraries are built on a single stable, humanized VHH framework with diversity hard-coded into the CDRs. They are animal-free, fast to access, and reproducible, and they were the basis of the published yeast-displayed nanobody platforms that delivered conformationally selective binders against membrane proteins. The codon scheme used to diversify the CDRs sets the real diversity and the stop-codon load — the trade-offs between NNK, NNS, and trimer codons are worth understanding before committing, as we describe in choosing a codon scheme for a VHH library.
Immunized libraries come from a llama or alpaca immunized with the antigen, with the VHH repertoire cloned out and displayed. The animal’s own affinity maturation has already enriched for binders, which helps for hard conformational epitopes — at the cost of time, an animal, and per-target library construction.
The default is a synthetic library for a tractable soluble antigen, an immunized repertoire for a difficult conformational or membrane target.
The discovery workflow
A typical campaign runs in four stages.
- Library display and QC. The VHH library is transformed into the display strain, induced, and stained for the display tag. NGS confirms diversity before any selection begins, so the starting pool is a known quantity.
- MACS pre-enrichment. A first magnetic-bead pass over antigen depletes the large non-binding majority and reduces a 10^8–10^9 library to a FACS-tractable pool in a single step.
- FACS rounds. Two to three rounds of two-color sorting (display tag versus labeled antigen) at decreasing antigen concentration progressively enrich higher-affinity clones. Gating on the display axis controls for expression differences so the sort selects affinity, not abundance.
- NGS and hit calling. Pre- and post-sort pools are sequenced; enrichment ratios rank clones. A small panel is expressed as soluble VHH and confirmed by SPR or BLI.
The deliverable is a ranked list of sequence-confirmed VHH binders with enrichment metrics, not a black-box pool.
From hit to lead: maturation and humanization
The advantage of running discovery on yeast display is that the next two steps need no platform change.
- Affinity maturation. A CDR-saturation or combinatorial library of a lead VHH is screened under tighter selection. Because the whole domain fits in one library, the full single-mutant landscape is available before the combinatorial round, which is the same logic behind deep mutational scanning for antibody affinity maturation.
- Humanization. Camelid VHH frameworks are already close to human VH3, and the remaining camelid-specific residues can be back-mutated under display selection to find the minimal set required to preserve binding — lowering immunogenicity risk without losing the binder.
When yeast display is not the right call
Yeast display is not universal. Switch platforms when:
- The target epitope depends on complex human glycosylation or full-length-IgG context — that is a mammalian display problem, not a nanobody-on-yeast problem.
- You need a fully naive, maximum-diversity discovery against a target with no prior binders and no immunization route — very large phage libraries still win on raw size.
For most VHH programs, though, the single-domain format and the quantitative FACS readout make yeast display the fastest route from antigen to a ranked panel of real binders.
If you have a target in hand, see our antibody and nanobody engineering services and the underlying yeast surface display platform, or start a Binder Pilot to scope a single-target campaign.
Related Ranomics services
- Antibody and nanobody engineering: VHH and scFv discovery, humanization, and maturation on display platforms.
- Yeast surface display: The quantitative FACS discovery platform behind nanobody campaigns.
- Affinity maturation: Push a VHH lead from micromolar to nanomolar or tighter on the same platform.
Frequently asked questions
Why is yeast display well suited to nanobody discovery?
A nanobody is a single ~13 kDa domain, so an entire VHH can be covered by one saturation library and screened in a single experiment. Yeast also folds the disulfide-bonded VHH domain natively, displays it at high copy on the cell wall, and supports quantitative FACS sorting — so binding signal is read per clone, not in bulk.
Do you use synthetic or immunized nanobody libraries?
Both. Synthetic and semi-synthetic VHH libraries built on a stable humanized framework give fast, animal-free access to most soluble targets and were the basis of the published yeast-displayed nanobody platforms. Immunized llama or alpaca repertoires are the better route for difficult conformational epitopes where in vivo affinity maturation has already done useful work.
How many sort rounds does a nanobody campaign take?
Typically one MACS pre-enrichment to deplete non-binders followed by two to three FACS rounds at decreasing antigen concentration. NGS of the enriched pools then ranks clones by enrichment before a small panel is expressed as soluble protein for confirmation.
Can you humanize and mature the nanobodies you discover?
Yes. Because discovery, affinity maturation, and humanization all run on the same yeast display platform, a VHH hit can move straight into CDR-saturation maturation and framework back-mutation selection without changing systems.