PERSPECTIVE: The Power of Phage Display: Redesigning Biology for Therapeutic Innovation

Abstract: The phage display technique represents a remarkable advance in biotechnology, establishing a link between a molecule’s genetic code and its function by displaying peptides or proteins on the surface of bacteriophages. Since its inception, it has revolutionized drug discovery, antibody engineering, vaccine development, and diagnostic innovation. The M13 bacteriophage is central to this approach, enabling the creation and screening of vast molecular libraries with billions of variants. Through iterative selection processes such as biopanning, researchers can identify molecules with high affinity and specificity for diverse targets. Advances such as phagemid systems, helper phages, and next-generation antibody libraries have further optimized the efficacy and therapeutic relevance of selected molecules. Despite challenges like the absence of post-translational modifications, the integration of machine learning and next-generation sequencing is expanding its potential. Phage display remains a cornerstone of modern biotechnology, driving the development of innovative therapies and precision medicine.

Imagine a giant living library where each “book” is a tiny virus with a unique “cover” containing the complete “recipe” for making that protein (Figure 1). This ingenious concept is at the heart of phage display, which is a powerful technique that directly links the genetic code of a protein (its genotype) to the protein itself (its phenotype). This protein is displayed on the surface of a specialized virus called a bacteriophage, which is a virus that specifically infects bacteria. Developed in 1985, phage display has revolutionized biotechnology research and development^1,2. In fact, this technique received the 2018 Nobel Prize in Chemistry, awarded to scientists George P. Smith and Sir Gregory P. Winter. Phage display provides the targeted selection of peptides and antibodies with high specificity and affinity, paving the way for novel therapies, including human monoclonal antibodies, which will continue to have profound impact on modern medicine ^1,2. The efficiency and scalability of this technique are based on the ability to generate and screen immense libraries containing up to 10 billion molecular variants in a single experiment ^1,3.

Figure 1. The Living Library of Antibody Phages. AI representation of an antibody phage library, where each phage acts like a unique ‘book’ displaying a specific antibody fragment on its surface. This direct link between the genetic code of the phage (genotype) and the expressed protein (phenotype) enables the selection and identification of high-affinity antibodies for therapeutic and diagnostic targets.

Phage display technology establishes a direct physical linkage between a protein’s function (phenotype) and its encoding genetic information (genotype), enabling large-scale functional screening. When a phage displaying a protein with a desired property, such as high-affinity target binding, is selected, the corresponding genetic sequence is immediately accessible within the same viral particle ³. Phage display technology establishes a direct physical linkage between a protein’s function (phenotype) and its encoding genetic information (genotype), enabling large-scale functional screening. When a phage displaying a protein with a desired property, such as high-affinity target binding, is selected, the corresponding genetic sequence is immediately accessible within the same viral particle ^3,4. As a result, phage display has become a powerful and scalable platform for the selection and optimization of peptides, proteins, and antibodies, with broad applications in protein–ligand interaction studies and affinity maturation ^3,5.

Bacteriophage M13 is an important tool in most phage display systems due to its unique biological properties ⁶. This thin, rod-shaped virus was first identified in 1963 ^7,8. It is approximately 880 nanometers long and 6 to 8 nanometers wide. M13 contains single-stranded DNA and specifically infects E. coli bacteria ⁶. One important feature of M13 is that it replicates without destroying the host cell,⁶ which is unlike many other viruses that kill infected cells. This smooth replication allows scientists to repeatedly produce large quantities of phages, which is essential for constructing and screening molecular libraries (Figure 2) ⁹. The envelope of the M13 phage consists of five different proteins, two of which, pVIII and pIII, are most important for the presentation of proteins ⁹. The pVIII protein is the most abundant and, with about 2,700 copies, forms the main outer layer of the virus ⁶. Although pVIII can display small peptides, it is typically used to display many copies of very short proteins ⁶. The pIII protein occurs only a few times (3-5 copies) at one end of the virus and helps the phage to recognize and infect the host cell ⁹. Because of its flexible structure, pIII can be used advantageously to display larger proteins (more than 100 amino acids).⁵ For successful display, that protein must be genetically linked to pIII or pVIII ⁶ 

Figure 2. M13 Phage Lifecycle: A Non-Lytic Partnership for Continuous Production. This illustration depicts the unique non-lytic lifecycle of the M13 bacteriophage within an E. coli bacteria. Unlike many viruses, M13 infects its host and replicates without causing cell lysis, allowing for continuous and sustained production of new phage particles. The diagram shows the phage (in blue) attaching to the E. coli cell, injecting its single-stranded DNA (step II), which is then replicated and used to produce new phage proteins (step III). These components are assembled into new M13 virions that bud out from the bacterial cell membrane, leaving the host cell intact and ready to continue producing more phages (step III). This "living factory" approach is fundamental to the scalability and efficiency of phage display.

Phage display begins with the creation of libraries, which are large collections of phage particles that each contain a unique peptide or protein variant. These libraries can contain up to 10 billion different variants ^3,5. This diversity results from the direct insertion of DNA sequences encoding different proteins into the genome of the M13 bacteriophage ⁶. There are two main types of libraries: "random peptide libraries," in which synthetic DNA sequences generate a variety of arbitrary amino acid chains, and "encoded libraries," in which known sequences, such as antibody fragments, are cloned into the phage genome ⁶. One of the most significant technical challenges in library construction is bacterial transformation, which involves introducing phage DNA into E. coli cells. This step limits the maximum size of a functional library, even after extensive optimization ¹⁰. To address this problem, researchers are continually refining their construction methods and implementing rigorous quality controls to ensure high functional diversity while minimizing the occurrence of “naked phages” or viral particles that do not display the desired proteins on their surface ¹¹.

Once the library is created, a key selection step called biopanning is performed. This iterative process allows for the isolation of phages that bind specifically to the target protein from a large pool of variants ¹⁰. In a typical cycle, the target (which can be a protein, a cell, or a tissue) is immobilized on a solid surface ¹². The phage library is then exposed to the target, allowing for the adhesion of phages with protein tags capable of binding to the target ^6,9. To ensure that only the strongest and most specific interactions remain, the surface is washed several times with increasingly stringent washes to remove phages with weak or no interactions ^6,9. The bound phages are then released or “eluted” from the target ¹². There are different elution methods: 1) harsh chemical conditions, such as acidic buffers or high-salt solutions that specifically inhibit protein interactions, 2) the use of free target proteins or ligands known to competitively displace bound phages, and 3) physical methods, such as low pH buffers in combination with ultrasonication, which can also be used to separate tightly bound phages ¹². The cycle of binding, washing, elution, and amplification is repeated several times (typically 3–5 rounds) to progressively enrich for phages displaying ligands with higher affinity and specificity for the target ¹². After elution, the selected phages are amplified by infecting bacterial host cells, which produce new phage particles carrying the same genetic information. During this process, phages with similar binding affinities may compete, and some clones can become preferentially amplified due to differences in replication efficiency or infectivity. As a result, successive rounds of biopanning lead to an enriched pool of peptide- or protein-displaying phages with improved target-binding properties.

A breakthrough in phage display was the introduction of the combination of phagemid vectors and helper phages. Phagemids are hybrid DNA vectors, that is, engineered DNA molecules used to carry and propagate foreign genetic sequences inside host cells, that combine the properties of plasmids, such as the transfer of antibiotic resistance genes and the replication of double-stranded DNA, with the properties of phage vectors, which enable the production of single-stranded DNA and its packaging into viral particles ¹⁰. Phagemids are designed to display fusion proteins with coat proteins such as pIII or pVIII under controlled conditions, but do not contain all the genes needed to produce full-length proteins on their own ¹⁰. This is where helper phages come in: they provide the missing viral genes needed for replication and assembly of a functional phage. Only after a bacterial host has been infected by a helper phage can the phagemid DNA be repaired and assembled into new viral particles ¹⁰. Helper phages are often engineered to be defective in their own replication or packaging to ensure preferential incorporation of phagemid DNA into new virions¹¹. This results in “mosaic” phages that carry both the recombinant coat protein (from the phagemid) and the wild-type coat protein (from the helper phage), which is essential for phage infectivity ¹⁰. Although early phagemid systems had relatively low display efficiencies, significant developments, such as pIII-free helper phages (such as Hyperphage), have significantly increased display levels and improved the efficiency of biopanning by forcing phages to incorporate a recombinant fusion protein to infect new cells ¹⁰.

Phage display has continued to evolve, giving rise to three first-generation antibody libraries, each aimed at improving the quality and drug properties of the discovered antibodies (Figure 3). First-generation libraries derive their diversity from natural antibody repeats, often randomly combining light and heavy chains to create large collections that can be used to screen for antibodies against a wide range of targets ¹¹.

Second-generation libraries rely on the selection of amino acids to modify based on knowledge of how proteins interact with antigens. These libraries not only aim for diversity but also for materials that are easier to develop into drugs. Processes such as trinucleotide mutagenesis (TRIM), a method that uses defined trinucleotide building blocks to control amino acid incorporation, allow precise control of amino acid composition, thereby increasing the functional scope of the library. Finally, third-generation libraries focus on improving overall library quality by incorporating screening steps that select antibodies with desirable therapeutic properties. This includes the use of heat shock and protein A selection during construction to promote stable, well-folded antibodies, or the use of yeast expression techniques to enhance proteins and to ensure optimal growth characteristics. In addition to comprehensive and universal libraries, specialized phage display libraries have been created to address challenging targets such as G protein-coupled receptors (GPCRs) or to discover antibodies with unique properties, such as pH-dependent affinity or extended domains that can reach concave binding sites ¹¹.

Figure 3. Evolution of Antibody Libraries: From Natural Diversity to Rational Design. The first-generation antibody libraries harness natural antibody repertoires by randomly combining light and heavy chains, creating diverse phages capable of targeting a wide range of antigens. Second-generation libraries build upon this foundation with rational design strategies, selectively introducing mutations at specific sites informed by structural and functional knowledge, enhancing the likelihood of producing high-affinity and developable antibodies. Third-generation libraries further refine antibody candidates by applying selection strategies aimed at improving biophysical properties such as stability, solubility, and developability, often through iterative optimization and screening under conditions that favor favorable expression, folding, and manufacturability profiles.

Phage display is a constantly evolving biotechnology based on the fundamental principle of directly linking genotype and phenotype. This unique capability enables high-throughput screening of large molecular libraries and significantly accelerates the discovery of ligands with specific binding properties ³. The M13 phage, with its non-lytic mode of replication and distinct pIII and pVIII proteins, is crucial for the scalability and simplicity of the method ¹². Controlling factors such as signal valence, which means how many copies of the peptide or antibody fragment are displayed on each phage particle, and optimizing key steps such as washing and elution during biopanning are essential to isolate ligands with high affinity and specificity.⁷ Phage display systems have evolved from simple explants to advanced phagemids and specialized helper phages, demonstrating a sophisticated and iterative approach to overcoming limitations and improving performance ¹⁰. This development has led to third-generation libraries capable of producing antibodies with highly desirable “drug-like” properties ¹¹. Ultimately, phage display technology is an essential and advanced tool for drug discovery, vaccine development, and diagnostic testing ³.

Although phage display technology offers remarkable advantages in terms of simplicity, productivity, and cost-effectiveness, it still faces challenges because proteins produced in E. coli lack eukaryotic post-translational modifications, particularly glycosylation, which can affect protein structure and function, as well as potential developmental risks ^10,11. However, ongoing developments, including the integration of next-generation analytics and machine learning, are removing these barriers and improving technological capabilities to predict and to develop better applications ¹¹.

References

1.         McCafferty, J., Griffiths, A. D., Winter, G. & Chiswell, D. J. Phage antibodies: filamentous phage displaying antibody variable domains. Nature 348, 552–554 (1990).

2.         Frenzel, A., Schirrmann, T. & Hust, M. Phage display-derived human antibodies in clinical development and therapy. MAbs 8, 1177–1194 (2016).

3.         Wu, C.-H., Liu, I.-J., Lu, R.-M. & Wu, H.-C. Advancement and applications of peptide phage display technology in biomedical science. J Biomed Sci 23, 8 (2016).

4.         Bazan, J., Całkosiński, I. & Gamian, A. Phage display - a powerful technique for immunotherapy: 1. Introduction and potential of therapeutic applications. Hum Vaccin Immunother 8, 1817–1828 (2012).

5.         Bakhshinejad, B. & Kjaer, A. On the origin of non-specific binders isolated in the selection of phage display peptide libraries. Front. Microbiol. 16, (2025).

6.         Kiguchi, Y. et al. Antibodies and Engineered Antibody Fragments against M13 Filamentous Phage to Facilitate Phage-Display-Based Molecular Breeding. Biol Pharm Bull 41, 1062–1070 (2018).

7.         Hofschneider, P. H. & Preuss, A. M13 bacteriophage liberation from intact bacteria as revealed by electron microscopy. J Mol Biol 7, 450–451 (1963).

8.         Sachtleben, P. Der Einfluß der Temperatur auf die elektrophoretische Beweglichkeit von Erythrozyten. Zeitschrift für Naturforschung B 18, 233–236 (1963).

9.         Wang, R. et al. M13 phage: a versatile building block for a highly specific analysis platform. Anal Bioanal Chem415, 3927–3944 (2023).

10. Bratkovič, T. Progress in phage display: evolution of the technique and its applications. Cell. Mol. Life Sci. 67, 749–767 (2010).

11. Zhang, Y. Evolution of phage display libraries for therapeutic antibody discovery. MAbs 15, 2213793 (2023).

12. Palma, M. Aspects of Phage-Based Vaccines for Protein and Epitope Immunization. Vaccines 11, 436 (2023).