A
ntibodies have outgrown their origins. Prodded by protein engineers, they’ve evolved into flexible molecules with myriad formats, from fragments and multispecifics to antibody-drug conjugates.
The latest in this new crop of variants is the small but powerful VHH, the variable domain of heavy-chain-only antibodies (HCAbs), also known as nanobodies.*
While traditional antibodies (mostly IgG1s) comprise two different protein chains, heavy and light, nanobodies have only one, springing from HCAbs naturally found in camelids. Because they lack a light chain, VHHs are small (~15 kDa) and deliver striking modularity. They work like a biological set of Legos, able to snap into bi-, tri- and multispecific antibody formats with untapped therapeutic potential.
“The most exciting thing about VHHs is the design space they open up,” says André Teixeira, senior director of the Antibody Platform at the Institute for Protein Innovation (IPI). “You can combine multiple binding domains in a single molecule and create antibodies with entirely novel mechanisms of action.”
What goes up must come down
Since the discovery of VHHs in 1989, researchers have been inoculating llamas and alpacas with antigens of choice and pulling high-affinity binders from the animals’ serum. What energized the field so quickly was the nanobodies’ repertoire of unique features, including their tiny size, long binding loop (see below) and remarkable stability under extreme conditions.
VHHs can be formatted either as standalones or fused to the stem (Fc region) of an antibody to resemble conventional IgGs. In standalone form, VHHs tend to have short half-lives, which may limit their usefulness in some therapeutic settings.
Yet that same compact format can also be an advantage in applications such as diagnostics and theranostics, or in cases where very small domains are needed for enhanced tissue penetration.
But there are downsides. Because VHHs originate from camelids, the human body may react to or reject their “camel-ness.” Thus, nanobodies require humanization — swapping out key amino acids with those most common in the human genome — to lower the likelihood of eliciting an anti-drug immune response. But humanization tends to reduce or eliminate VHHs’ inherent advantages and diminish their stability, necessitating a secondary, time-consuming optimization step.
“It really boils down to speed,” says Nikolai Suslov, vice president of the Antibody Division at Alloy Therapeutics. “Every startup and every pharma company wants to move at breakneck velocity, and with the right building blocks, you can do that while still tackling really complex biology.”
Any approach to shortening the drug development timeline is in high demand — and any company that can offer a humanized platform-as-a-service to generate stable, high-quality output is now a desirable partner.
IPI’s development of a synthetic humanized VHH library for yeast display presents a solution, streamlining antibody discovery while expanding design flexibility. Recognizing the benefits, Alloy has licensed and integrated the IPI platform into its antibody engineering stack, giving drug developers more shots on goal and a better chance of starting with molecules that can withstand the demands of development.
In a partnership announced this year, IPI and Alloy Therapeutics will together enable biotech and pharma teams to outsource hit discovery and multispecific engineering, providing leads that can move more cleanly through development milestones.
“Our goal at IPI is to advance protein science and biomedical research,” says Teixeira. “We want to partner with teams that can translate these technologies into their programs and ultimately into new discoveries and new products.”
Designing VHH libraries
Well-designed nanobody libraries are the critical first step in VHH drug discovery and development, and the choices made in their development affect their versatility and downstream ramifications.
The simplest library development method involves immunizing a camelid with a target antigen, extracting lymphocytic mRNA, and amplifying the resulting cDNA in an appropriate expression system, typically E. coli. The resulting library serves as the starting point for selecting target-specific VHHs, which are commonly retrieved using phage display.
When the antigen is highly conserved across species — and therefore unlikely to provoke a strong immune response — or toxic, researchers may instead build a naïve library from unimmunized camelids.
Such libraries can be screened against many targets, but the approach comes with substantial drawbacks: the libraries often yield lower-affinity binders, still carry camelid sequence liabilities that can complicate humanization and downstream developability, and require massive numbers of B cells — and therefore large blood volumes — to capture enough diversity. Hence, protein engineers have invented a workaround in the form of purely synthetic antibodies.
“A synthetic library teaches us what nature has already figured out,” says Teixeira. “And gives us much more control over the outcome.”
To build a synthetic library — like IPI’s — one chooses a specific VHH scaffold and genetically tinkers with the protein’s hypervariable complementarity-determining region (CDR), the amino acid loops that drive antigen recognition. The goal is to introduce sufficient genetic variation into the CDRs — particularly the third loop (CDR3). Variety in this main hub of antibody uniqueness and binding creates a highly diverse library, often containing more than a billion unique sequences.
From there, the library is typically cloned into a display system — often phage or yeast display — and screened against targets of interest to cherry-pick the most promising binders. Once candidate nanobodies emerge, they are expressed, purified and tailored to the specific needs of the intended application.
But these methods alone don’t overcome the hurdles of humanization and optimization.
The IPI-Alloy approach
For the IPI library, scientists built their VHH repertoire in yeast, a system well suited to flow cytometry-enabled selection. The approach gave researchers a powerful way to fine-tune both affinity and specificity, making it easier to identify nanobodies that do not react with closely related family members while still cross-reacting with orthologs from preclinical species, such as mice and monkeys.
The platform was also designed to start closer to clinically usable molecules and bypass one of nanobody discovery’s most frustrating traps. Molecules derived from camelids often need to be humanized before they can advance, but that process can compromise affinity and stability, forcing developers into a rescue-engineering cycle in which improving one property can undermine another.
Instead of starting with camelid VHHs and seeking to humanize them, the IPI team began with the closest human germline sequences available. The scientists then introduced a limited set of mutations to make those molecules behave like VHHs, including eliminating the need for a light chain.
“We created a human library that behaves like a camelid VHH, rather than starting with a camelid VHH and trying to humanize it later,” says Teixeira.
But removing the dependency on a light can destabilize the molecule. Thus, Teixeira’s team tested multiple engineering strategies and expressed more than 100 candidate molecules to identify the most stable options. The resulting scaffolds remained close to 90% identical to natural human antibody sequences — an improvement over the roughly 80% identity often seen in VHH therapeutics — while maximizing stability and preserving distinct VHH advantages.
The IPI team also designed the library to generate not just the best binders but also the most developable ones, those with higher odds of making it through the pipeline to become viable drugs. The overall goal was to maximize both sequence diversity and favorable developability properties upfront, before the antibody discovery process even began.
To do that, the team enriched for sequences less likely to be immunogenic in humans and less prone to biophysical liabilities such as aggregation, instability and polyreactivity. The scientists also focused on generating the most diverse pool of sequences, with particular attention to the CDRs. Using proprietary methods, the researchers curated all three CDRs differently, deploying a mix of machine learning and wet-lab testing to identify the most promising sequences.
“We’ve pulled together all the protein engineering hacks and tricks we’ve learned over the years,” says Suslov. “That’s what really sets these libraries apart, and why our teams can design more sophisticated multispecifics without slowing down.”
Onto the clinic
Where does this all lead? Thanks to their unique advantages over traditional antibodies, nanobodies are quickly gaining traction as both diagnostics and therapeutics. The field hit an important milestone in 2018 with the approval of caplacizumab, the first VHH-based drug, and there’s a gathering wave of drug candidates.
Researchers are particularly excited about the oncology applications of VHHs. By attaching a potent radioligand to a nanobody, drug developers may be able to create a fast-clearing VHH with impressive tumor-specific targeting. Other applications include cell therapies that rely on targeting moieties and genetic medicines that enable tissue-specific delivery of mutation-correcting cargo. Another hot area is autoimmune therapy, where scientists are stringing VHHs together to block specific cytokine pathways implicated in immune dysfunction, for example.
The sky is the limit, as drug developers are now linking multiple different VHHs together, or even linking them with traditional antibody fragments, to create bi-, tri- and tetraspecific molecules.
“This capacity really unlocks the development of novel types of therapeutics,” says Suslov. “The excitement is how protein engineering can enable novel therapies to be made.”
Partners in Discovery
Now, IPI and Alloy’s attention moves to stress-testing the library and assessing its ability to discover more stable, soluble and effectively humanized VHHs. Then, Alloy will enable partners to leverage the library in campaigns to discover game-changing diagnostics and therapeutics.
Each advance in the fast-moving field of antibody discovery and engineering — from the development of hybridomas and synthetic libraries to AI protein design — has opened broad new avenues for bioscience. If history is anything to go by, both Teixeira and Suslov are excited to see where VHH technologies lead.
“I’ve been a practitioner of antibody engineering for the last 20 years, and technologies keep making the discovery cycle faster,” says Suslov. “There are new techniques and an increase in the richness of sequence and functional data. I can only dream of what it will look like in another 20 years.”
* The term nanobody was introduced as a trademark of Ablynx in 2003 and has since become a label for the single-domain antibody fragments.
For further reading:
1. Developing drug-like single-domain antibodies (VHH) from in vitro libraries
Erasmus et al., MAbs, 2025
One of the most recent reviews on synthetic VHH libraries, discussing yeast display filtering and affinity maturation workflows, focusing on developability and therapeutic translation and led by IPI’s André Teixeira
2. NANOBODIES®: A Review of Diagnostic and Therapeutic Applications
Jin et al., International Journal of Molecular Sciences, 2023
A comprehensive overview of nanobody biology, structure, and applications that includes a discussion of display technologies such as phage and yeast display
3. Rapidly Inducible Yeast Surface Display for Antibody Evolution with OrthoRep
Paulk et al., ACS Synthetic Biology, 2024
Describes modern yeast-display antibody evolution systems and includes a hypermutation and continuous evolution platform
4. Expanding and Improving Nanobody Repertoires Using Yeast Display
Cross et al., Journal of Biological Chemistry, 2023
An oft-cited, technically detailed paper that shows how yeast display can generate large repertoires of high-affinity nanobodies
Sources:
André Teixeira, andre.teixeira@proteininnovation.org
Nikolai Suslov, nikolai.suslov@alloytx.com
About IPI
The Institute for Protein Innovation is pioneering a new approach to scientific discovery and collaboration. As a nonprofit research institute, we provide the biomedical research community with synthetic antibodies and deep protein expertise, empowering scientists to explore fundamental biological processes and pinpoint new targets for therapeutic development. Our mission is to advance protein science to accelerate research and improve human health. For more information, visit proteininnovation.org or follow us on LinkedIn, Instagram or Bluesky.
About Alloy Therapeutics
Alloy Therapeutics is a biotechnology ecosystem company powering the future of drug discovery and development with AI-powered platforms and scientific expertise. Through a collaborative partnership approach, Alloy provides access to proprietary technologies, services, and company-creation capabilities that are foundational to discovering and developing therapeutics across multiple modalities: antibodies, bispecifics, TCRms, genetic medicines, cell therapies, and drug delivery.
Alloy integrates AI/ML into its discovery and development infrastructure, combining proprietary models, real-world data, and human expertise to help partners advance therapeutic programs. At Alloy, we are redefining biologics development by reducing the time and cost to advance programs from discovery to human data. Join the Alloy community by visiting www.alloytx.com and following Alloy on LinkedIn.