610
`
`Phage display in pharmaceutical biotechnology
`Sachdev S Sidhu
`
`Over the past year, methods for the construction of M13
`phage-display libraries have been significantly improved and
`new display formats have been developed. Phage-displayed
`peptide libraries have been used to isolate specific ligands for
`numerous protein targets. New phage antibody libraries have
`further expanded the practical applications of the technology
`and phage cDNA libraries have proven useful in defining
`natural binding interactions. In addition, phage-display methods
`have been developed for the rapid determination of binding
`energetics at protein–protein interfaces.
`
`Developments in phage-display technology
`Library construction
`The success of any selection experiment ultimately
`depends on the diversity and quality of the initial library.
`Over the years, methods have been refined to the point
`where extremely large libraries can now be rapidly and
`reliably constructed. Sidhu et al. [6•,7] have described opti-
`mized methods that enable the construction of libraries
`with diversities greater than 1012, almost 100-fold greater
`than previously thought practical.
`
`Addresses
`Department of Protein Engineering, Genentech, Inc., 1 DNA Way,
`South San Francisco, CA 94080, USA; e-mail: sidhu@gene.com
`
`Current Opinion in Biotechnology 2000, 11:610–616
`
`0958-1669/00/$ — see front matter
`© 2000 Elsevier Science Ltd. All rights reserved.
`
`Abbreviation
`VEGF
`vascular endothelial growth factor
`
`Introduction
`Phage display is a powerful technology for selecting and
`engineering polypeptides with novel functions. If DNA
`fragments encoding polypeptides are fused to certain bac-
`teriophage coat protein genes, the fusion genes can be
`encapsulated within phage particles that also display the
`encoded polypeptides on their surfaces. This establishes a
`physical linkage between phenotype and genotype.
`Highly diverse libraries can be constructed by fusing
`degenerate DNA to a coat protein gene, and library mem-
`bers with desired binding specificities can be isolated by
`binding to an immobilized receptor in vitro. The
`sequences of selected polypeptides can be determined
`from the sequence of the encapsulated, encoding DNA.
`
`Phage display was first developed with the Escherichia coli-
`specific bacteriophage M13 [1], and the success of M13
`phage display has prompted the development of numerous
`alternative display systems. These include systems that uti-
`lize other E. coli-specific phage, such as λ-phage [2] and T4
`phage [3], and also systems that use eukaryotic viruses [4].
`In addition, polypeptides have been displayed on the sur-
`faces of bacteria and yeast [5]. Although these alternative
`systems have proven advantageous in special applications,
`M13 phage display remains the dominant technology.
`
`This review covers developments in M13 phage display
`made over the past year. I discuss technological improve-
`ments that enable the construction of larger libraries and
`new display formats that extend the technology to new
`applications. I also highlight some important applications
`of peptide and protein phage display, with particular
`emphasis on pharmaceutical biotechnology.
`
`Vectors and display formats
`The M13 phage particle consists of a single-stranded DNA
`core surrounded by a coat composed of five different pro-
`teins (Figure 1). The length of the filament is covered by
`several thousand copies of the major coat protein, pro-
`tein-8 (P8). Each end of the particle is capped by five
`copies each of two minor coat proteins: protein-3 (P3) and
`protein-6 (P6) at one end, and protein-7 (P7) and protein-
`9 (P9) at the other end [8]. In early examples of phage
`display, polypeptides were fused to the amino-terminus of
`either P3 or P8 in the viral genome [9,10]. These systems
`were severely limited because large polypeptides (>10
`residues for P8 display) compromised coat protein function
`and so could not be efficiently displayed. The develop-
`ment of phagemid display systems solved this problem
`because, in such systems, polypeptides were fused to an
`additional coat protein gene encoded by a phagemid vec-
`tor [11]. Subsequent infection with a helper phage
`produced particles with phagemid DNA encapsulated in a
`coat composed mainly of wild-type coat proteins from the
`helper phage but also containing some fusion coat proteins
`from the phagemid (Figure 1).
`
`Phagemid systems permit the display of polypeptides that
`could not be displayed in simple phage systems, because
`the deleterious effects of the fusion protein are attentuat-
`ed by the presence of wild-type coat proteins from the
`helper phage. Such systems have also enabled the devel-
`opment of new display formats. Jespers et al. [12] showed
`that proteins could be displayed as fusions to the carboxyl-
`terminus of P6. More recently, Gao et al. [13] have
`demonstrated the display of antibody fragments fused to
`the amino-terminus of P7 and P9, whereas Fuh et al. [14••]
`have demonstrated carboxy-terminal P8 display. In addi-
`tion, we have found that polypeptides fused to the
`carboxyl-terminus of P3 are displayed at levels comparable
`to conventional amino-terminal fusions [15]. Thus, in
`phagemid systems, functional polypeptide display has now
`been demonstrated with all five M13 coat proteins.
`
`Although conventional amino-terminal display formats are
`likely to dominate established applications, carboxy-termi-
`nal display enables studies unsuited to amino-terminal
`
`Lassen - Exhibit 1027, p. 1
`
`

`

`display. These include the study of protein–protein inter-
`actions requiring free carboxy-termini and functional
`cDNA cloning (see below). Also, carboxy-terminal display
`may be especially useful for the display of intracellular pro-
`teins. Before phage assembly, P3 and P8 reside in the
`E. coli inner membrane with their carboxy-termini in the
`cytoplasm and their amino-termini in the periplasm [16].
`Amino-terminal display has worked well for the display of
`secreted proteins evolved for folding in the oxidizing
`periplasmic environment, but few intracellular proteins
`have been displayed in this manner. In contrast, carboxy-
`terminal fusions would allow for folding in the reducing
`cytoplasm, and this may be ideal for intracellular proteins.
`
`Engineering the M13 coat for phage display
`In a phagemid system where wild-type P8 from the helper
`phage maintains the integrity of the phage coat, Sidhu et al.
`[17] have shown that an additional phagemid-encoded P8
`can tolerate a surprisingly large number of mutations.
`Furthermore, some of these mutations actually promote
`incorporation into the phage coat and thus increase the dis-
`play of fusion proteins. Subsequent mutational analyses
`demonstrated that only a small subset of the first 30 P8
`residues is critical for coat incorporation [18]. Weiss and
`Sidhu [19••] have further demonstrated the malleability of
`the phage coat by evolving completely artificial coat pro-
`teins that resemble P8 in an inverted orientation. These
`artificial coat proteins incorporate into the phage coat and
`permit the display of carboxy-terminal-fused polypeptides.
`Together, these reports show that the phage coat is
`extremely tolerant to the addition of new proteins, and
`these proteins can be specifically engineered for improved
`phage display.
`
`Applications for phage-displayed peptide
`libraries
`Phage-displayed peptide libraries can be used to isolate
`peptides that bind with high specificity and affinity to vir-
`tually any target protein. These binding peptides can be
`used as reagents to understand molecular recognition, as
`minimized mimics for receptors, or as lead molecules in
`drug design.
`
`Mimics of extracellular protein–protein interactions
`Extracellular protein–protein interactions often involve
`large, flat contact surfaces. Conventional small-molecule
`screening efforts that work well against concave surfaces
`have largely failed in targeting these interactions [20], but
`in contrast, phage-displayed peptide libraries have proven
`remarkably successful. Numerous phage-derived peptides
`that bind extracellular protein surfaces have been report-
`ed (reviewed in [6]), and several three-dimensional
`structures have also been solved (Figure 2). Ferrer and
`Harrison [21••] isolated peptides that bind to HIV gp120
`and inhibit its interaction with CD4. Although this inter-
`action has been a prime target for antiviral drug
`development, previous efforts had been largely unsuc-
`cessful because of the extended nature of the binding
`
`Phage display in pharmaceutical biotechnology Sidhu 611
`
`Figure 1
`
`Promoter
`
`Secretion
`signal
`
`Displayed
`protein
`
`M13 coat protein
`
`Ampr
`
`Phagemid
`vector
`
`322 ori
`
`
`
`f1 ori
`
`Escherichia coli
`
`
`
`Recycle
`
`Add helper phage
`
`P7 and P9
`
`P3 and P6
`
`P8
`
`Bind to receptor
`
` Binders
`
`Current Opinion in Biotechnology
`
`An M13 phagemid vector designed for phage display. A phagemid
`vector contains origins of single-stranded (f1 ori) and double-stranded
`(322 ori) DNA replication and a selective marker, such as the
`β-lactamase gene (Ampr) which confers resistance to ampicillin. For
`phage display, the phagemid also contains a cassette consisting of a
`promoter that drives transcription of an open reading frame encoding a
`secretion signal and the displayed protein fused to an M13 coat
`protein. The vector replicates in E. coli as a double-stranded plasmid,
`but coinfection with a helper phage results in the production of single-
`stranded DNA that is packaged into phage particles. The phage coat
`contains five different proteins, and polypeptides can be displayed as
`either amino-terminal fusions (with P3, P8, P7, or P9) or carboxy-
`terminal fusions (with P6, P8, or P3). The phage particles can be used
`in binding selections, and binding clones can be amplified by recycling
`through an E. coli host.
`
`surface [21••]. Phage display probably suceeded because
`peptidic ligands can adopt extended conformations that
`effectively compliment such extended surfaces.
`
`In theory, peptides could bind to a protein anywhere on its
`solvent-exposed surface. Thus, it is notable that most select-
`ed peptides bind at sites that coincide with natural
`ligand-binding sites, and consequently act as antagonists or
`agonists of natural protein–protein interactions. It seems that
`natural binding sites possess features that predispose them for
`ligand binding. Delano et al. [22••] isolated peptides that
`bound to the constant fragment of immunoglobulin G (IgG-
`Fc) and found that a ‘consensus’ binding site that interacts
`with at least four natural proteins was also the preferred site
`
`Lassen - Exhibit 1027, p. 2
`
`

`

`612 Pharmaceutical biotechnology
`
`Figure 2
`
`(a)
`
`(b)
`
`(c)
`
`(d)
`
`(e)
`
`Structures of binding peptides isolated from
`phage-displayed peptide libraries. Mainchains
`are shown as dark grey ribbons and sidechains
`are shown in light grey. Each peptide contains
`a single disulfide bond. Structures (a) and
`(b) were determined by NMR with the free
`peptides in solution. The other structures were
`determined by X-ray crystallography with the
`peptides in complex with their cognate ligands.
`(a) A turn-helix conformation is adopted by a
`peptide that binds to the insulin-like growth
`factor binding protein 1 [49]. (b) Within the
`disulfide-bonded loop, a FVIIa-binding peptide
`consists of a type I reverse turn and an
`irregular turn of a helix that extends to the
`carboxyl-terminus [29••]. (c) A peptide that
`binds to the IgG-Fc is a β-hairpin [22••].
`(d) Another β-hairpin peptide forms a
`non-covalent, symmetric dimer that acts as an
`agonist by dimerizing two erythropoietin
`receptors (the monomer is shown) [50]. (e) A
`peptide that antagonizes the activity of VEGF
`consists of a disulfide-bonded loop and an
`extended amino-terminus [51].
`
`Current Opinion in Biotechnology
`
`for peptide binding. These results bode well for the use of
`peptide libraries in drug development because they demon-
`strate that a large proportion of binding peptides is likely to
`target biologically relevant sites.
`
`Intracellular protein-binding domains
`There are several families of intracellular protein domains
`that bind other proteins and thus regulate cellular function,
`and many of these interactions involve the recognition of
`
`Lassen - Exhibit 1027, p. 3
`
`

`

`small continuous stretches of amino acids in the binding
`partner. Phage-displayed peptide libraries have been used to
`study the binding specificity of SH3 domains and WW
`domains that recognize proline-rich sequences [23,24]. It has
`also been shown that phage-displayed peptides can be phos-
`phorylated in vitro and these modified libraries can be used
`to isolate ligands for SH2 and phosphotyrosine-binding
`(PTB) domains that bind to phosphotyrosine-containing
`peptides [25,26]. Recently, Chen and Sigler [27•] deter-
`mined the crystal structure of the GroEL chaperonin alone
`and in complex with a high-affinity peptide isolated from a
`random library. Their studies shed light on the structural
`basis for GroEL’s diverse substrate specificity and the mech-
`anism of GroEL-mediated protein folding. In another study,
`peptide libraries fused to the carboxy-terminus of P8 were
`used to select ligands for PDZ domains, a recently identified
`domain family whose members generally bind the extreme
`carboxy-termini of other proteins and, in so doing, regulate
`subcellular protein localization [14••].
`
`These studies demonstrate that different types of phage-
`displayed libraries are useful for investigating different
`types of binding interactions. Despite the specific
`demands of some binding interactions, peptide phage dis-
`play is applicable to the majority of intracellular
`protein–protein interactions.
`
`Enzyme inhibitors
`Enzymes typically contain many deep clefts — including
`active sites and allosteric regulatory sites — that are well suit-
`ed for the binding of small ligands. Hyde-DeRuyscher et al.
`[28•] obtained peptide inhibitors for six out of seven distinct
`enzyme classes they tested. In addition, they showed that
`these peptides could be used as reagents for the detection of
`small-molecule inhibitors in high-throughput screens.
`
`Dennis et al. [29••] isolated a peptide that non-competi-
`tively inhibits the activity of the serine protease factor VIIa
`(FVIIa) with exquisite specificity and high potency.
`Extensive mutational and structural analyses showed that
`the peptide binds at a previously unknown ‘exosite’ dis-
`tinct from the active site, and apparently inhibits activity
`by an allosteric mechanism. These results are of substan-
`tial therapeutic relevance; FVIIa is a key regulator of the
`blood coagulation pathway and, in the past, it has been
`extremely difficult to selectively inhibit individual pro-
`teases of the coagulation cascade [29••].
`
`Phage-displayed peptide libraries seem to be ideal sources of
`peptidic enzyme inhibitors. Inhibitory peptides should prove
`useful as drug discovery reagents, or perhaps even as drugs
`themselves. Furthermore, the phage-display process explores
`the entire exposed surface of a target enzyme and, as a result,
`it can sometimes identify new sites for inhibitor design.
`
`Applications for phage-displayed proteins
`The ability to display large proteins on M13 phage has led
`to numerous applications. I focus on areas that have
`
`Phage display in pharmaceutical biotechnology Sidhu 613
`
`advanced significantly in the past year and are likely to
`have a major impact on biotechnology.
`
`Antibody phage
`Over the past decade, phage-displayed antibody fragments
`have been the subject of intensive reasearch (reviewed in
`[30,31]). As a result, antibody phage libraries have become
`practical tools for drug discovery and several phage-derived
`antibodies are in advanced clinical trials. Although most
`libraries have been constructed by cloning the natural
`immune repertoire into phage-display vectors, an important
`advance has been the development of high-quality libraries
`with completely synthetic complementarity-determining
`regions. Knappik et al. [32] have constructed a library in
`which a limited number of human frameworks are used and
`diversity is introduced by means of synthetic cassettes.
`Such a system is very amenable to the generation of thera-
`peutic antibodies because preferred frameworks can be
`used and affinity maturation is aided by the use of defined
`mutagenic cassettes. The construction of large, high-quali-
`ty libraries should also be aided by general improvements
`in library construction methodologies [6,7] and by the intro-
`duction of improved in vivo recombination systems [33].
`
`Although it is possible to obtain specific antibodies direct-
`ly from naïve phage-displayed repertoires, another
`important application for phage-display technology has
`been the humanization and affinity maturation of conven-
`tional murine antibodies. Such antibodies have been
`proven effective in a number of therapeutic applications.
`For example, a humanized antibody to vascular endothe-
`lial growth factor (VEGF) is in clinical trials as a cancer
`treatment, and Chen et al. [34] have used phage display to
`produce an affinity matured variant that is ~100-fold
`improved in both affinity and potency.
`
`Phage-displayed cDNA libraries
`Phage-displayed cDNA libraries can be used to identify
`natural binding partners for orphan receptors and ligands.
`cDNA libraries have been displayed with several different
`formats using either M13 phage [35–38] or other E. coli
`phage [2,39,40]. Although these systems show consider-
`able promise, no single display format has proven
`universally applicable. This is probably because of the fact
`that only a subset of eukaryotic proteins can be efficiently
`expressed in E. coli, and only a fraction of these can be effi-
`ciently displayed with any given system. Hopefully, new
`M13 display formats may also prove useful in these
`applications (see above).
`
`Mapping protein functional epitopes
`A major goal of modern biology is to obtain a detailed
`understanding of the protein–protein interactions that
`underly cellular processes. Alanine-scanning is a system-
`atic method that has been particularly effective in
`mapping ‘functional binding epitopes’: the contact
`residues at a protein–protein interface that make ener-
`getic contributions to binding [41]. Because alanine
`
`Lassen - Exhibit 1027, p. 4
`
`

`

`614 Pharmaceutical biotechnology
`
`mutations remove sidechain atoms past the β-carbon, they
`can be used to infer the roles of individual sidechains. A
`comprehensive alanine scan is laborious, however,
`because many single alanine mutants must be construct-
`ed, purified, and then analyzed for structural integrity and
`binding affinity. The use of phage-displayed alanine
`mutants can greatly expedite the process because proteins
`can be easily purified in association with phage particles
`and binding affinities can be realiably determined with
`simple assays that use anti-phage antibodies to quantify
`bound proteins. Using this approach, Dubaquié and
`Lowman [42] completed an impressive total alanine-scan
`of insulin-like growth factor-1 and mapped its binding
`interactions with two natural ligands.
`
`The mapping of protein functional epitopes may be fur-
`ther expedited with a combinatorial alanine-scanning
`strategy [43•]. In this method, a special phage-displayed
`protein library is constructed in which sidechains are pref-
`erentially allowed to vary only as the wild type or alanine.
`After binding selections to isolate functional clones, DNA
`sequencing is used to determine the alanine/wild-type
`ratio at each varied position. This ratio can be used to cal-
`culate binding energy contributions of
`individual
`sidechains with accuracies close to those obtained with
`conventional alanine-scanning mutagenesis. The method
`is very rapid because many sidechains are analyzed simul-
`taneously and binding energetics are derived from
`statistical analysis of DNA sequences, thus circumventing
`the need for protein purification and biophysical analysis.
`
`Engineering binding affinity and specificity
`Phage display can be used to improve or alter the binding
`properties of displayed proteins. Several groups have used
`phage display to engineer zinc-finger domains with
`designed DNA-binding specificities that can be used to
`control gene expression [44–46]. In another application,
`Hiipakka et al. [47] selected SH3 domains with improved
`or altered binding properties, thus demonstrating that
`phage display can be used to engineer signaling protein
`interaction domains. Phage display can also be used to sim-
`plify complex signaling pathways. For example, VEGF is a
`pleiotropic factor that binds two different receptors (Flt-1
`and KDR) and thus exerts a multitude of biological effects.
`Using rational design and phage display, Li et al. [48] have
`generated VEGF variants that selectively bind only to
`either Flt-1 or KDR [48]. These receptor-selective variants
`can be used to elucidate the specific role of each receptor.
`
`Conclusions
`Intensive efforts from many researchers have made phage
`display an invaluable component of biotechnology.
`Improved library construction methods — in combination
`with numerous vectors and display formats — will extend
`the technology even further. Phage antibodies are likely to
`play an even greater role in the generation of analytical
`reagents and therapeutic drugs. Highly diverse peptide
`libraries can be used to isolate specific ligands for virtually
`
`any target of interest, and these ligands should be useful in
`therapeutic target validation and as leads in drug design.
`cDNA phage display can be used to identify natural pro-
`tein–protein interactions, which can be mapped in detail
`with methods that use phage display as a tool for measur-
`ing binding energetics. Further research will continue to
`improve the reliability of phage-display methods and many
`aspects of the technology may eventually be automated. It
`seems probable that extremely diverse phage-display
`libraries will contain multiple solutions to most binding
`problems, and perhaps the most effective applications will
`be those that exhaustively explore these solutions by com-
`bining phage-display selections with high-throughput
`screening and DNA sequencing. These technologies will
`generate vast databases that explore the links between
`protein structure and function, and this information will in
`turn expedite the process of drug development.
`
`Acknowledgements
`I thank Nicholas Skelton and David Wood for help with figures.
`
`References and recommended reading
`Papers of particular interest, published within the annual period of review,
`have been highlighted as:
`• of special interest
`•• of outstanding interest
`Smith GP: Filamentous fusion phage: novel expression vectors
`that display cloned antigens on the virion surface. Science 1985,
`228:315-1317.
`
`1.
`
`2.
`
`3.
`
`4.
`
`Santini C, Brennan D, Mennuni C, Hoess RH, Nicosia A, Cortese R,
`Luzzago A: Efficient display of an HCV cDNA expression library as
`C-terminal fusion to the capsid protein D of bacteriophage
`lambda. J Mol Biol 1998, 282:125-135.
`
`Ren Z, Black LW: Phage T4 SOC and HOC display of biologically
`active, full-length proteins on the viral capsid. Gene 1998,
`215:439-444.
`
`Possee RD: Baculoviruses as expression vectors. Curr Opin
`Biotechnol 1997, 8:569-572.
`
`5. Georgiou G, Stathopoulos C, Daugherty PS, Nayak AR,
`Iverson BL, Curtiss R III: Display of heterologous proteins on
`the surface of microorganisms: from the screening of
`combinatorial libraries to live recombinant vaccines. Nat
`Biotechnol 1997, 15:29-34.
`
`6.
`•
`
`Sidhu SS, Lowman HB, Cunningham BC, Wells JA: Phage display
`for selection of novel binding peptides. Methods Enzymol 2000,
`328:333-363.
`This chapter provides detailed descriptions of optimized methods for the
`construction of extremely diverse phage-displayed libraries. With highly con-
`centrated E. coli and DNA, a single electroporation reaction produces
`libraries with >1010 unique members. Multiple reactions can be combined to
`produce diversities >1012.
`
`7.
`
`Sidhu SS, Weiss GA: Constructing phage display libraries by
`oligonucleotide-directed mutagenesis. In Phage Display: A
`Practical Approach. Edited by Clackson T, Lowman HB. Oxford:
`Oxford University Press; in press.
`
`8. Marvin DA: Filamentous phage structure, infection and assembly.
`Curr Opin Struct Biol 1998, 8:150-158.
`
`9.
`
`Scott JK, Smith GP: Searching for peptide ligands with an epitope
`library. Science 1990, 249:386-390.
`
`10. Greenwood J, Willis AE, Perham RN: Multiple display of foreign
`peptides on a filamentous bacteriophage. J Mol Biol 1991,
`220:821-827.
`
`11. Bass S, Green R, Wells JA: Hormone phage: an enrichment
`method for variant proteins with altered binding properties.
`Proteins: Struct Funct Genet 1990, 8:309-314.
`
`Lassen - Exhibit 1027, p. 5
`
`

`

`12.
`
`Jespers LS, Messens JH, De Keyser A, Eeckhout D, Van Den Brande I,
`Gansemans YG, Lauwereys MJ, Vlasuk GP, Stanssens PE: Surface
`expression and ligand-based selection of cDNAs fused to
`filamentous phage gene VI. Bio/Technology 1995, 13:378-382.
`
`13. Gao C, Mao S, Lo CH, Wirsching P, Lerner RA, Janda KD: Making
`artificial antibodies: a format for phage display of combinatorial
`heterodimeric arrays. Proc Natl Acad Sci USA 1999,
`96:6025-6030.
`
`14. Fuh G, Pisabarro MT, Li Y, Quan C, Lasky LA, Sidhu SS: Analysis of
`PDZ domain-ligand interactions using carboxyl-terminal phage
`••
`display. J Biol Chem 2000, 275:21486-21491.
`The authors report a new display format in which peptids are fused to the
`carboxyl-terminus of P8. A random peptide library displayed in this format is
`used to identify high-affinity ligands for PDZ domains that bind peptides with
`free carboxy-termini. The binding interactions are rationalized using synthet-
`ic peptides and homology modeling.
`
`15. Fuh G, Sidhu SS: Efficient phage display of polypeptides fused to
`the carboxy-terminus of the M13 gene-3 minor coat protein.
`FEBS Lett 2000, 480:231-234.
`
`16. Rodi DJ, Makowski L: Phage-display technology — finding a needle
`in a vast molecular haystack. Curr Opin Biotechnol 1999, 10:87-93.
`
`17. Sidhu SS, Weiss GA, Wells JA: High copy display of large proteins
`on phage for functional selections. J Mol Biol 2000, 296:487-495.
`
`18. Weiss GA, Wells JA, Sidhu SS: Mutational anlaysis of the major
`coat protein of M13 identifies residues that control protein
`display. Protein Sci 2000, 9:647-654.
`
`19. Weiss GA, Sidhu SS: Design and evolution of artificial M13 coat
`proteins. J Mol Biol 2000, 300:213-219.
`••
`Completely artificial M13 coat proteins are selected from large libraries of
`sequences designed to mimic P8 in a reverse orientation. These results
`demonstrate that the M13 phage coat is very tolerant to the incorporation of
`new proteins that need not be related to natural coat proteins. Thus, it will
`be possible to engineer artificial coat proteins specifically for phage display.
`
`20. Cochran AG: Antagonists of protein–protein interactions. Chem
`Biol 2000, 7:R85-R94.
`
`21. Ferrer M, Harrison SC: Peptide ligands to human
`immunodeficiency virus type 1 gp120 identified from phage
`••
`display libraries. J Virol 1999, 73:5795-5802.
`The authors use a naïve library to isolate a 12-residue peptide that binds
`gp120 from three different HIV strains and inhibits the interaction between
`gp120 and CD4. Although the peptide did not show inhibitory activity in a
`cell–cell fusion assay, it may still be a useful lead for the development of anti-
`HIV compounds.
`
`22. Delano WL, Ultsch MH, de Vos AM, Wells JA: Convergent solutions
`to binding at a protein–protein interface. Science 2000,
`••
`287:1279-1283.
`The crystal structure of a disulfide-constrained peptide complexed with
`IgG-Fc reveals a β-hairpin structure. The peptide binds to a consensus bind-
`ing site that also interacts with four structurally distinct natural proteins.
`A detailed analysis reveals that the different ligands use similar binding con-
`tacts and the consensus binding site is highly accessible, hydrophobic, and
`adaptive. These properties may be common attributes in natural binding sites.
`
`23. Sparks AB, Rider JE, Hoffman NG, Fowlkes DM, Quilliam LA, Kay BK:
`Distinct ligand preferences of Src homology 3 domains from Src,
`Yes, Abl, Cortactin, p53bp2, PLCγ, Crk, and Grb2. Proc Natl Acad
`USA 1996, 93:1540-1544.
`
`24. Lin H, Ermekova KS, Rentschler S, Sparks AB, Kay BK, Sudol M:
`Using molecular repertoires to identify high-affinity peptide
`ligands of the WW domain of human and mouse YAP. Biol Chem
`1997, 378:531-537.
`
`25. Dente L, Vetriani C, Zucconi A, Pelicci G, Lanfrancone L, Pelicci PG,
`Cesareni G: Modified phage peptide libraries as a tool to study
`specificity of phosphorylation and recognition of tyrosine
`containing peptides. J Mol Biol 1997, 269:694-703.
`
`26. Gram H, Schmitz R, Zuber JF, Baumann G: Identification of
`phosphopeptide ligands for the Src-homology 2 (SH2) domain of
`Grb2 by phage display. Eur J Biochem 1997, 246:633-637.
`
`27. Chen L, Sigler PB: The crystal structure of a GroEL/peptide
`complex: plasticity as a basis for substrate diversity. Cell 1999,
`•
`99:757-768.
`X-ray crystallography is used to investigate the binding interaction between
`GroEL chaperonin and a phage-derived peptide. The authors use this infor-
`mation to investigate the structural basis for GroEL’s diverse substrate
`
`Phage display in pharmaceutical biotechnology Sidhu 615
`
`specificity, and they conclude that various modes of molecular plasticity are
`responsible for tight but promiscuous binding.
`
`28. Hyde-DeRuyscher R, Paige LA, Christensen DJ, Hyde-DeRuyscher N,
`Lim A, Fredericks ZL, Kranz J, Gallant P, Zhang J, Rocklage SM et al.:
`•
`Detection of small-molecule enzyme inhibitors with peptides
`isolated from phage-displayed combinatorial peptide libraries.
`Chem Biol 2000, 7:17-25.
`This encouraging study demonstrates that it should be easy to obtain phage-
`derived inhibitory peptides for any enzyme. Furthermore, these peptides can
`be used as sensitive probes for the detection of small molecules that bind
`the enzyme and displace the bound peptide. These probes enable the devel-
`opment of high-throughput screening assays designed to detect small-mol-
`ecule enzyme inhibitors.
`
`29. Dennis MS, Eigenbrot C, Skelton NJ, Ultsch MH, Santell L,
`•• Dwyer MA, O’Connell MP, Lazarus RA: Peptide exosite inhibitors of
`factor VIIa as anticoagulants. Nature 2000, 404:465-470.
`This paper reports an extremely thorough analysis of a peptide that inhibits
`FVIIa activity. The crystal structure of the peptide–FVIIa complex reveals that
`the peptide binds at an allosteric exosite, consistent with the observed non-
`competitive inhibition mode. The structure of the peptide free in solution was
`determined by NMR and found to be very similar to that found in the bound
`state. Thus, phage display identified a potent FVIIa inhibitor and also led to
`the discovery of a new mode of inhibition.
`
`30. Dall’Acqua W, Carter P: Antibody engineering. Curr Opin Struct Biol
`1998, 8:443-450.
`
`31. Griffiths AD, Duncan AR: Strategies for selection of antibodies by
`phage display. Curr Opin Biotechnol 1998, 9:102-108.
`
`32. Knappik A, Ge L, Honegger A, Pack P, Fischer M, Wellnhofer G,
`Hoess A, Wolle J, Pluckthun A, Virnekas B: Fully synthetic human
`combinatorial antibody libraries (HuCAL) based on modular
`consensus frameworks and CDRs randomized with
`trinucleotides. J Mol Biol 2000, 296:57-86.
`
`33. Sblattero D, Bradbury A: Exploiting recombination in single
`bacteria to make large phage antibody libraries. Nat Biotechnol
`2000, 18:75-80.
`
`34. Chen Y, Weismann C, Fuh G, Li B, Christinger HW, McKay P,
`deVos AM, Lowman HB: Selection and analysis of an optimized
`anti-VEGF antibody: crystal structure of an affinity-matured Fab in
`complex with antigen. J Mol Biol 1999, 293:865-881.
`
`35. Fransen M, Van Veldhoven PP, Subramani S: Identification of
`peroxisomal proteins by using M13 phage protein VI phage
`display: molecular evidence that mammalian peroxisomes
`containg a 2,4-dienoyl-CoA reductase. Biochem J 1999,
`340:561-568.
`
`36. Hufton SE, Moerkerk PT, Meulemans EV, de Bruine A, Arends JW,
`Hoogenboom HR: Phage display of cDNA. J Immunol Methods
`1999, 231:39-51.
`
`37. Shanmugavelu M, Baytan AR, Chesnut JD, Bonning BC: A novel
`protein that binds juvenile hormone esterase in fat body tissue
`and pericardial cells of the tobacco hornworm Manduca sexta L.
`J Biol Chem 2000, 275:1802-1806.
`
`38. Cochrane D, Webster C, Masih G, McCafferty J: Identification of
`natural ligands for SH2 domains from a phage display cDNA
`library. J Mol Biol 2000, 297:89-97.
`
`39. Sche PP, McKenzie KM, White JD, Austin DJ: Display cloning:
`functional identification of natural receptors using cDNA-phage
`display. Chem Biol 1999, 6:707-716.
`
`40. Fakok VA, Bratton DL, Rose DM, Pearson A, Ezekewitz RAB,
`Henson PM: A receptor for phosphatidylserine-specific clearance
`of apoptotic cells. Nature 2000, 405:85-90.
`
`41. Wells JA: Systematic mutational analyses of protein–protein
`interfaces. Methods Enzymol 1991, 202:390-41