Synthetic Biology Technologies for Antibody Discovery
Imagine bypassing the lengthy traditional immunization process and tapping into the antibody diversity equivalent of over a million llamas – right at your fingertips. At Isogenica, our libraries, built with Synthetic Biology Technologies are transforming antibody discovery, accelerating the design of targeted therapies with greater precision, speed, and reduced risk. In this blog, we will delve into various methods for building antibody libraries and the novel methods used at Isogenica that helped us build the largest, most diverse VHH antibody libraries in the world.
Beating Older Technologies
Traditionally, the most popular approach to generate antibody therapeutics has been in vivo animal immunization. This traditional approach entails repeated exposure of an animal to the target antigen, ultimately aiming to produce high-affinity antibodies as the immune system responds.
However, this process is not without challenges. It is long, costly, and depends on the antigen being both immunogenic and non-toxic to the animal. Furthermore, control over essential properties like specificity and epitope recognition is limited, as selection occurs entirely within the organism, and more specialized requirements, such as thermal stability, are impossible. Moreover, as this approach has been so widely adopted, different animals (particularly of the same species) tend to produce very similar antibodies to the same antigen, creating freedom-to-operate (FTO) limitations for many disease targets.
By combining our vast libraries with innovative in vitro approaches such as Isogenica’s proprietary CIS Display technology, we can address concerns like protein folding complexity alongside driving higher affinities. By capturing the full diversity of our libraries during screening, we can produce robust, highly specific antibodies. This approach overcomes traditional limitations and also provides a more precise, efficient path for therapeutic discovery.
Since there are multiple approaches to constructing antibody libraries, both diversity and control over sequences are vital. However, these desirable factors are not always easy to balance, since more control generally leads to lower diversity. Here, we explore the landscape of library-building technologies, and highlight Colibra(R), the published method used to build Isogenica’s vast collection of variable CDR regions.
Different approaches to build antibody libraries
There are several ways to build antibody libraries, each with its own strengths and weaknesses in terms of balancing diversity and control. Let’s break down why some methods might work better than others.
1. Degenerate codons
Amino acids are encoded by three-base codons (NNN), with each triplet coding for either an amino acid, or a stop. Here, N stands for any of the four G, A, T, or C bases of DNA. In practice there are many more possible codons than amino acids (64 codons for 20 amino acids), so in some cases the same amino acid can be encoded by several different codons, usually varying in their third position. When building a library, molecular biologists can exploit this third position in an NNN codon to bring some control into building amino acid sequences.
By restricting which base comes in the third position, you can exert some control using so-called degenerate codons like ‘NNB’, ‘NNK’, or ‘NNS’, (where B = C, G, or T, S = C or G, and K = G or T). These codons are designed for high diversity to cover all 20 amino acids while minimizing unwanted stop codons, although they still occur up to 3% of the time – a significant number of sequences in a large library. Alternatives like ‘NDT’ or ‘DBK’ offer more refined selection, encoding only the 12 or 16 most common amino acids relevant to protein structure and function. This approach reduces the library diversity, along with the associated costs and resources required for screening. but can ensure that a greater percentage of the finished library remains functional.
2. Trinucleotide mutagenesis (TRIM)
TRIM technology takes the idea of building DNA libraries to the next level. It uses pre-assembled trinucleotide building blocks during chemical synthesis, giving researchers precise control over the amino acid composition. This customization eliminates unwanted stop codons, giving TRIM a major advantage over older methods. Additionally, amino acid redundancies are reduced meaning screening is more efficient and cost effective.
What’s even better? TRIM technology imposes virtually no design restrictions and excellent control over ratios of specific codons in defined positions. However, this theoretical control comes at a price in practice. While simple TRIM oligos with limited options at each position are relatively inexpensive, the cost of TRIM oligos can rise significantly if more positions, or more options per position, are needed. Combined with its equal reliance on chemical synthesis, longer oligos also carry a greater risk of inaccuracies that can put the quality of the finished library at risk.
Fisher Scientific. (n.d.). GeneArt Combinatorial Libraries. Retrieved from https://www.thermofisher.com/es/es/home/life-science/cloning/gene-synthesis/directed-evolution/geneart-combinatorial-libraries.html
3. Rationally Designed CDRs
Rationally designed libraries, where whole CDRs are designed and synthesized, offer precise control over sequences. This facilitates targeted exploration of the sequence space while effectively minimizing undesirable genetic liabilities. Rationally designed libraries, however, do not cover the full spectrum of potential antibody variants and depend on the assumption that the available knowledge is both accurate and complete. Gaps in knowledge can limit the effectiveness of the design, as new insights into antibody functionality continue to emerge. For instance, more is understood about linear epitopes compared to non-contiguous or conformational epitopes, due to the lack of relevant structural data. VHHs have long CDR3 loops that enhance their ability to bind to pockets and grooves, enabling access to conformational, allosteric, or active site epitopes; therefore, it is advisable to avoid this design bias in library construction.
4. Naïve Libraries
Naïve natural libraries are universal antibody libraries generated from B-cells of non-immunized donors, thus eliminating the need to construct new libraries for each antigen. These libraries can be built by extracting genetic material from B cells, and amplifying antibody genes and/or CDR regions. The amplified fragments can then be dropped into a fixed scaffold to create new, naive libraries for screening.
The use of DNA from multiple donors offers some increased diversity but introduces challenges when blending CDRs and scaffolds from different sources together. For example, integrating CDR sequences from human B cells into VHH frameworks can produce a biophysically robust protein. However, human-derived CDRs tend to be less varied in sequence compared to their llama counterparts. This means the ability to bind antigens may be reduced in a single-domain antibody format where only 3 CDRs are in place. Even working within one species, different germline frameworks contain CDRs of different lengths, which is harder to account for when amplifying mixed DNA populations. In addition, the diversity of these libraries can vary, as it depends on the quality of the donor samples.
As a result, rational libraries often face challenges such as reduced diversity and size, since relatively small populations of source material can limit the overall library size. The inherent bias in this approach can also lead to missed opportunities for discovering novel or unexpected binding interactions that might be uncovered using a more diverse, random approach such as TRIM. Binders from these CDR-by-CDR libraries often display low binding affinities, requiring affinity maturation.
5. Colibra®
Rational design can be combined with a semi random approach to adjust library size in a targeted manner, maintaining desirable evidence-based features such as enhanced stability and increasing screening efficiency, whilst reducing the heavy bias of a fully rational design. The power of Isogenica’s synthetic library approach comes from the combination of a highly diverse, semi-random design combined with a more controlled building technology.
Isogenica’s Colibra® technology, implemented in our LlamdA® and huLlamdA® libraries, utilizes amino acid pairs to build CDR sequences. This approach balances control over sequences while maintaining enormous diversity to produce high-affinity antibodies.
Sequence liabilities can arise when two or three amino acids form motifs that negatively affect an antibody’s manufacturability or specificity. LlamdA® is Isogenica’s proprietary fully synthetic single-domain VHH library, built using Colibra® technology which greatly reduces the occurrence of these liabilities, ensuring robust therapeutic antibody production. Colibra® enhances both diversity and developability, with the original CDRs also offered in humanized frameworks.
Unlike traditional solid-phase chemical synthesis methods, Colibra® uses enzymatic ligation for increased process efficiency and purity. By incorporating hexamers of codon pairs, it allows for precise randomization of two contiguous positions, while still keeping all amino acids available. In addition to liability reduction, this level of control allowed us to create sequence distributions to mimic those found naturally in immune derived VHHs, as confirmed by NGS validation in the original paper. This means you can get the same immune-quality VHH antibodies with all the speed and precision of a synthetic discovery solution.
What’s Different About Colibra® Libraries?
While animal immunization is a tried-and-true method, modern synthetic approaches are addressing many of its limitations. Isogenica’s synthetic libraries offer not just an alternative but an evolved solution, balancing the precision many believe is lacking with the high diversity crucial for antibody discovery. In fact, with library sizes of 1010 – 1013, Isogenica’s VHH libraries are the most diverse synthetic VHH libraries on the market. Furthermore, our deep knowledge of library construction allows our scientists to build in specialised considerations such as epitope type, cell penetration, protease resistance, thermal stability and other developability characteristics.
Benefits of Colibra® libraries:
- Reduced Chance of Liabilities. With the use of technologies like Colibra®, Isogenica ensures that many problematic sequences, such as those prone to deamidation or isomerization, are removed from the start, improving antibody stability and production.
- Greater Diversity and Size. Isogenica’s libraries combine control with significantly higher library diversity, addressing historical issues of low-affinity binders from first-generation synthetic libraries.
- Advanced Validation. The use of NGS for library validation ensures that the generated sequences meet the required standards of quality and precision for effective antibody development.
- Innovation Meets Nature. Colibra® technology allows preservation of the natural sequence patterns that make VHHs such a desirable antibody format, meaning full functionality with the speed and convenience of animal-free antibody discovery.
Conclusion: Proven Innovation and Performance
Isogenica’s synthetic libraries, powered by Colibra® technology, offer a reliable platform that optimizes both antibody diversity and stability. This advanced approach ensures the development of specific, functional antibodies while overcoming the limitations of traditional methods.
Interested in the details? Why not check out our dedicated Libraries page.
To learn about how our current partners are finding VHH solutions to support their innovations, visit our partnerships page to discover how we work.
REFERENCES
Zhang H. Evolution of phage display libraries for therapeutic antibody discovery. mAbs. 2022; 15(1) https://www.tandfonline.com/doi/full/10.1080/19420862.2023.2213793
Frigotto L, Smith ME, Brankin C, Sedani A, Cooper SE, Kanwar N, Evans D, Svobodova S, Baar C, Glanville J, et al. Codon-Precise, Synthetic, Antibody Fragment Libraries Built Using Automated Hexamer Codon Additions and Validated through Next Generation Sequencing. Antibodies. 2015; 4(2):88-102. https://doi.org/10.3390/antib4020088