Nanobodies (VHH, single-domain antibodies) have a molecular weight only about one-tenth that of traditional antibodies. They are stable and easy to express, making them a favorite in antibody drugs and diagnostic reagents in recent years. To screen a VHH from an immune or synthetic library, the first step is to choose the right display and screening technology.
Currently, the most mainstream platforms in the industry are phage display and yeast surface display. Customers often ask: Isn't it just about linking the antibody gene and then selecting clones that can bind? While both can be done, successful linking does not equate to good screening. The expression environment (prokaryotic vs. eukaryotic) and target reading methods (panning vs. flow cytometry) of the two systems differ significantly, ultimately affecting antibody affinity, developability, and hit rate against difficult targets. The following explains the principles.
Phage display was established by Smith in 1985, and McCafferty et al. first used it for antibodies in 1990. The method involves inserting the antibody gene into the coat protein of filamentous phages (usually pIII, but pVIII is also used), allowing the antibody fragment to be displayed on the surface of the phage particles; the phages then infect and amplify in Escherichia coli (prokaryotic cells). This technology laid the foundation for antibody display, and Smith and Winter shared the 2018 Nobel Prize in Chemistry for this work [1][2]. Yeast surface display was established by Boder and Wittrup in 1997. The method involves fusing VHH with the yeast α-lectin subunit Aga2p, anchoring it to Aga1p on the cell wall via disulfide bonds, and the antibody "grows" on the surface of Saccharomyces cerevisiae (eukaryotic) cells; screening is done by sorting cells one by one using flow cytometry (FACS) [3]. In short: bacteriophages attach antibodies to viral particles and rely on bacterial proliferation; yeast grows antibodies on the cell wall of eukaryotic cells and relies on flow cytometry to pick them up.
Antibodies originate from eukaryotes, but phage display is performed in E. coli. Prokaryotic cells lack the folding mechanism of the endoplasmic reticulum (using molecular chaperones like BiP and PDI), making it easy for complex antibodies to mismatch disulfide bonds, forming insoluble inclusion bodies. Some eukaryotic antibodies are eliminated or lost during amplification due to codon bias or toxicity to E. coli [9].
Yeast follows the eukaryotic secretion pathway. When proteins are transported between the endoplasmic reticulum and Golgi apparatus, a chaperone network helps form disulfide bonds and complete correct folding. Proteins that fail quality checks are degraded before reaching the cell surface. Therefore, a clone that can stably "shine" on the yeast surface has often already quietly passed a round of stability screening. This is a valuable early signal for subsequent production in CHO and HEK293 [3][9]. It's important to clarify one point here: VHHs are single-domain, small in molecular weight, and extremely stable, even expressing well in E. coli, which is why phage immunorepositories can perform VHH analysis. The advantages of eukaryotic folding are more pronounced in molecules like scFv and Fab, which require heavy and light chain pairing and are more prone to aggregation; for VHHs, this advantage lies more in the foldability of target sites and the conformational integrity of membrane proteins. In other words, the folding advantage of yeast isn't ineffective for VHHs, but rather the difficulty of finding the target site must be taken into account.
Phage Panning: Antigens are coated on a plate, the library binds, weaker molecules are washed away, stronger molecules are washed away, and then E. coli is reinfected for amplification. This is an enrichment process, which is basically "blind"—it only knows which groups have been enriched, but it is difficult to know in real time how high the affinity of a single clone is; it is also easy to enrich non-specific, sticky clones [8].
Yeast Cells Using FACS: Two fluorescent labels are used to label a cell simultaneously, one to look at the antibody display amount and the other to look at the antigen binding amount. Dividing the binding signal by the expression level can screen out those "just high expression, so they appear to have strong binding" illusions; changing several antigen concentrations can also separate high, medium, and low affinity groups, and even perform equilibrium titration and estimate Kd directly on the yeast surface, without first purifying the protein expression [5][8]. One detail worth noting is that each yeast cell surface has 104-105 copies. High display density leads to significant avidity, thus increasing apparent affinity. Therefore, bistandard normalization is not only convenient but also a necessary means to prevent "artificially high" avidity, which is almost the core of yeast platform design.
![[Yeast Surface Display Series②] From Screening to Drug Development: The Technological Boundaries and Selection Logic of Bacteriophage and Yeast Display](data/watermark/main/ueditor/20260828/6a912d41a166b.png "[Yeast Surface Display Series②] From Screening to Drug Development: The Technological Boundaries and Selection Logic of Bacteriophage and Yeast Display")
Fig 1. FACS dual-label quantitative sorting (gating based on binding/expression ratio, horizontal axis for expression signal, vertical axis for binding signal).
Affinity maturation is the core competency of yeast display. In 2000, Boder et al. used a randomly mutated yeast display antibody library to perform four rounds of FACS screening based on dissociation rate (kinetics), selecting only clones that were "most reluctant to release the antigen," resulting in a dissociation rate decrease of more than 1000 times, ultimately obtaining antibody fragments with an affinity of approximately 48 fM (femtomolar level) [4]. This level of precision is difficult for solid-phase panning to achieve.
Table 1. Comparison of technical parameters between phage display and yeast surface display
Dimensions | Phage Display | Yeast Surface Display |
Expression System | Prokaryotic (E. coli) | Nucleus (Saccharomyces cerevisiae) |
Library Capacity | 109-1011 | 108-109 |
Screening Method | Solid-phase panning (enrichment) | FACS Flow Cytometry Quantitative Sorting |
Affinity Resolution | Crowded, difficult to distinguish in real time | Fine, gating can be set for high/medium/low affinity |
Environment | Prokaryotic periplasm, disulfide bond mismatch is easy | Eukaryotic secretion pathway, chaperone-assisted |
Display Amount/Cell | 1-5 copies/particle | 10⁴-10⁵ copies/cell |
Suitable Scenarios | Large library capacity de novo screening, cell antigens | Affinity maturation, competition, blocking, pH sensitivity |
For soluble proteins with clear structures and stable in solution, yeast eukaryotic folding can provide lead molecules with correct conformations and higher development potential, which can be obtained without animal immunization. For example, Uchański et al. (2019) used an improved yeast display platform to screen out nanobodies targeting human coagulation factor IXa (FIXa, soluble protein) from a llama immune library [7]; McMahon et al. (2018) used human serum albumin (HSA, soluble) as an example target for validation [6]. Yeast display is often accurate and convenient for these types of targets.
Membrane proteins (such as GPCRs) are easily inactivated and conformationally unstable in prokaryotic systems, which is a weakness of phage display; yeast display has been repeatedly proven to be able to tackle these tough challenges. McMahon et al. (2018) used a yeast display platform to discover conformationally selective nanobodies targeting two different human GPCRs (including the agonist-bound A2A adenosine receptor) [6]; Uchański et al. (2019) further screened out nanobodies (both GPCRs/membrane proteins) targeting human OX2 orexin receptor and human α2A adrenaline receptor from an immune library [7].
It should be noted that the difficulty in membrane protein screening is not only in the display platform itself, but also in antigen preparation (detergent, nanodiscs, protein liposomes), which is a crucial factor in determining success or failure. This bottleneck is the same for both phages and yeast. Yeast's contribution lies in its ability to provide lead antigens with strong conformational selectivity and good development potential once the correct conformation is obtained. As for whether popular transmembrane targets in the ADC and bispecific antibody fields (such as CLDN18.2, B7-H3, TROP2) can be successfully demonstrated using yeast, it depends on the antigen's prepareability and conformational stability—this is an industry judgment beyond existing literature, and its implementation still requires target-specific validation.
Table 2. Yeast display adaptability and literature examples for different targets
Target Type | Yeast Display Fit | Literature Examples |
Soluble Proteins (Coagulation Factor IXa, HSA) | High | Uchariski 2019 [7]; McMahon 2018 [6] |
Conformation Epitopes | High | McMahon 2018 [6] |
Membrane Proteins/GPCR | Higher | McMahon 2018 [6]; Uchanski 2019 [7] |
Affinity-Maturation Required Leaders | High | Boder 2000 [4] |
Discovering a massive naive library from scratch | Low (phages are better) | - |
To be fair, for completely naive discoveries requiring massive libraries from scratch, phage display still holds a significant advantage in statistical coverage. The industry often uses a combination of "phage for broad coverage and yeast for fine-tuning." Another reality is that most approved nanobody drugs to date have their lead molecules developed through phage display—for example, the first approved anti-vWF capucizumab (Cablivi) is a prime example. This demonstrates that phage display is the most thoroughly validated and industry-familiar tool for "discovery from scratch." The differentiating value of yeast display lies not in being "earlier," but in being "more accurate and easier to perform": quality, developability, and targeting hard targets such as solubility, membrane proteins, conformation, competition, blocking, and pH dependence.
Further considering actual costs: phage display is fast, inexpensive, and easily automated; yeast requires flow cytometry, is time-consuming per round, and has higher overall costs. Platform selection is never about "which is more advanced," but rather "what do we need this time?"
Yeast display and phage display are not simply a matter of "which is more advanced" or "which replaces the other," but rather two sets of technological tools tailored to different VHH discovery needs. Phage display, with its large library capacity, high throughput, and mature screening system, is better suited for large-scale initial screening from immune libraries, naive libraries, or synthetic libraries. Yeast surface display, relying on the eukaryotic expression environment and FACS quantitative sorting, has unique advantages in screening complex targets such as affinity maturation, competitive/blocking screening, conformational epitopes, and membrane proteins.
For VHH discovery projects, the truly important thing is not simply choosing a particular display technology, but designing an appropriate screening strategy based on target properties, library type, screening objectives, and subsequent development needs. Some projects are suitable for phage display to quickly obtain candidate clones, while others are better suited for direct yeast display for fine-grained screening. For complex projects, combining the two platforms can also form a complementary strategy of "large-scale phage discovery + yeast quantitative screening/affinity maturation."
Chengdu AlpVHHs Co.,LTD focuses on the discovery and development of nanobodies, having established phage display and yeast surface display technology platforms. These platforms provide technical services tailored to different targets and project needs, ranging from VHH library screening and positive clone identification to affinity maturation and candidate molecule optimization.
Phage display technology platform is suitable for screening ultra-large-capacity VHH libraries, immune/naive libraries, and antibody discovery targeting soluble proteins, cells, and other complex targets. In yeast surface display technology platform, FACS can be used to quantitatively analyze VHH expression levels and binding capacity, and affinity maturation, competitive screening, and conformation-dependent screening can be conducted, helping researchers move from "can bind" to "worth developing."
The two platforms are not mutually exclusive. For different projects, AlpVHHs can develop differentiated VHH screening protocols based on target characteristics and research objectives, achieving a better balance between library size, screening efficiency, affinity, specificity, and molecular developability.
From discovering a VHH to obtaining a truly valuable candidate molecule, platform selection is only the first step. Choosing the right technology is crucial, but choosing the right screening strategy is even more important.
If you are conducting nanobody discovery, affinity maturation, membrane protein/conformation epitope screening, or complex target VHH development, Critical Point Biotechnology can provide targeted phage display, yeast surface display, and combined screening solutions based on your target properties and project objectives.
Reference:
[1]Smith GP. Filamentous fusion phage: novel expression vectors that display cloned antigens on the virion surface. *Science*. 1985;228(4705):1315-1317. PMID: 4001944.
[2] McCafferty J, Griffiths AD, Winter G, Chiswell DJ. Phage antibodies: filamentous phage displaying antibody variable domains. *Nature*. 1990;348(6301):552-554. PMID: 2247164.
[3] Boder ET, Wittrup KD. Yeast surface display for screening combinatorial polypeptide libraries. *Nat Biotechnol*. 1997;15(6):553-557. PMID: 9181578.
[4] Boder ET, Midelfort KS, Wittrup KD. Directed evolution of antibody fragments with monovalent femtomolar antigen-binding affinity. *Proc Natl Acad Sci U S A*. 2000;97(20):10701-10705. PMID: 10984501.
[5] Chao G, Lau WL, Hackel BJ, Sazinsky SL, Lippow SM, Wittrup KD. Isolating and engineering human antibodies using yeast surface display. *Nat Protoc*. 2006;1(2):755-768. PMID: 17406305.
[6] McMahon C, Baier AS, Pascolutti R, et al. Yeast surface display platform for rapid discovery of conformationally selective nanobodies. *Nat Struct Mol Biol*. 2018;25(3):289-296. PMID: 29434346.
[7] Uchański T, Zögg T, Yin J, et al. An improved yeast surface display platform for the screening of nanobody immune libraries. *Sci Rep*. 2019;9:382. PMID: 30674983.
[8] Van Deventer JA, Wittrup KD. Yeast surface display for antibody isolation: library construction, library screening, and affinity maturation. *Methods Mol Biol*. 2014;1131:151-181. PMID: 24515465.
[9] Boder ET, Raeeszadeh-Sarmazdeh M, Price JV. Engineering antibodies by yeast display. *Arch Biochem Biophys*. 2012;526(1):99-106. PMID: 22450168.
