Phage-encoded combinatorial chemical libraries based on bicyclic peptides.

Here we describe a phage strategy for the selection of ligands based on bicyclic or linear peptides attached covalently to an organic core. We designed peptide repertoires with three reactive cysteine residues, each spaced apart by several random amino acid residues, and we fused the repertoires to the phage gene-3-protein. Conjugation with tris-(bromomethyl)benzene via the reactive cysteines generated repertoires of peptide conjugates with two peptide loops anchored to a mesitylene core. Iterative affinity selections yielded several enzyme inhibitors; after further mutagenesis and selection, we were able to chemically synthesize a lead inhibitor (PK15; Ki =1.5 nM) specific to human plasma kallikrein that efficiently interrupted the intrinsic coagulation pathway in human plasma tested ex vivo. This approach offers a powerful means of generating and selecting bicyclic macrocycles (or if cleaved, linear derivatives thereof) as ligands poised at the interface of small-molecule drugs and biologics.

Stability-Indicating Method Development and Validation for Busulfan Drug Substance by Gas Chromatography-Flame Ionization Detector.

A new, simple and stability-indicating gas chromatography-flame ionization detection (GC-FID) method was developed and validated for the quantitative determination of busulfan and its organic impurities (OI) in drug substance without derivatization. The chromatographic attributes were achieved on a fused silica capillary column (0.53 mm × 30 m, 1.0 µm, USP Phase G42), using hydrogen as a carrier gas with a split ratio of 1:1. Forced degradation studies were conducted to establish the stability-indicating capability and method specificity that showed the stressed busulfan peak was free from any co-elution. Robustness study demonstrated the chromatograms remained mostly unaffected under deliberate, but small variations of chromatographic parameters, establishing the reliability of the method during routine usage. The method was shown to be reliable, sensitive, specific, linear, accurate, precise and rugged in the 1,4-butanediol concentration range of 1-20 µg/mL. The method, intended for compendial uses, is suitable for quantitative analysis of busulfan and its organic impurities in drug substances.

Beta-edge interactions in a pentadecameric human antibody V kappa domain.

Antibodies are the archetypal molecules of the Ig-fold superfamily. Their highly conserved beta-sheet architecture has evolved to avoid aggregation by protecting edge strands. However, the crystal structure of a human V kappa domain described here, reveals an exposed beta-edge strand which mediates assembly of a helical pentadecameric oligomer. This edge strand is highly conserved in V kappa domains but is both shortened and capped by the use of two sequential trans-proline residues in V lambda domains. We suggest that the exposure of this beta-edge in V kappa domains may explain why light-chain deposition disease is mediated predominantly by kappa antibodies.

Tapping diversity lost in transformations--in vitro amplification of ligation reactions.

Molecular evolution is a powerful means of engineering proteins. It usually requires the generation of a large recombinant DNA library of variants for cloning into a phage or plasmid vector, and the transformation of a host organism for expression and screening of the variant proteins. However, library size is often limited by the low yields of circular DNA and the poor transformation efficiencies of linear DNA. Here we have overcome this limitation by amplification of recombinant circular DNA molecules directly from ligation reactions. The amplification by bacteriophage Phi29 polymerase increased the number of transformants; thus from a nanogram-scale ligation of DNA fragments comprising two sub-libraries of variant antibody domains, we succeeded in amplifying a highly diverse and large combinatorial phage antibody library (>10(9) transformants in Escherichia coli and 10(5)-fold more transformants than without amplification). From the amplified library, but not from the smaller un-amplified library, we could isolate several antibody fragments against a target antigen. It appears that amplification of ligations with Phi29 polymerase can help recover clones and molecular diversity otherwise lost in the transformation step. A further feature of the method is the option of using PCR-amplified vectors for ligations.

Repertoires of aggregation-resistant human antibody domains.

We recently described a method for the generation of a large human domain antibody repertoire involving combinatorial assembly of CDR building blocks from a smaller repertoire comprising a high frequency of aggregation-resistant antibody domains. Here we show that the frequency of aggregation-resistant domains in the combinatorial repertoire remained high. Furthermore, one of the antigen-binding domains selected from the combinatorial repertoire retained its binding properties through 25 cycles of thermal denaturation, suggesting that antibody domains can be created that rival the heat-resistance of thermophilic proteins such as Taq polymerase.

Identification of protein domains by shotgun proteolysis.

The identification of protein domains within multi-domain proteins is a persistent problem. Here, we describe an experimental method (shotgun proteolysis) based on random DNA fragmentation and protease selection of the encoded polypeptides on phage for this purpose. We applied the method to the Escherichia coli genome and identified 124 protease-resistant fragments; several were re-cloned for expression as soluble fragments in bacteria, and corresponded to autonomously folding units with folding energies similar to natural protein domains (DeltaG(u)=3.8-6.6 kcal/mol). Structural information was available for approximately half of the selected proteins, which corresponded to compact, globular and domain-sized units that had been derived from a wide range of protein superfamilies. Furthermore, boundaries of the selected fragments correlated with domain boundaries as defined by bioinformatics predictions (R2=0.82; p=0.016). However, predictions were incomplete or entirely lacking for the remaining fragments, reflecting the limited proteome coverage of current bioinformatics methods. Shotgun proteolysis therefore provides a means to identify domains and other autonomously folding units on a genome-wide scale, without any prior knowledge of sequence or structure. Shotgun proteolysis should be particularly valuable for structural studies of proteins and represents a high-throughput alternative to the classical limited proteolysis method for the isolation of stable components of multi-domain proteins.

Early protein evolution: building domains from ligand-binding polypeptide segments.

It has been suggested that in the early evolution of proteins, segments of polypeptide, unable to fold in isolation, may have collapsed together to form folded proto-domains. We wondered whether the incorporation of segments with a pre-existing binding activity into a folded domain could, by fixing the ligand binding conformation and/or providing additional contacts, lead to large affinity improvements and provide an evolutionary advantage. As a model, we took a segment of polypeptide from hen egg lysozyme that in the native protein forms the binding interface with the monoclonal antibodies HyHEL5 and F10 (KD=60 pM). When expressed in bacteria the isolated segment was unfolded, readily proteolysed and only bound weakly to the antibodies (KD>1 microM). We then combined the segment with random genomic segments to create a repertoire of chimaeric polypeptides displayed on filamentous bacteriophage. By use of proteolysis (to select folded polypeptide) and anti-lysozyme antibodies (to select an active conformation) we isolated a folded dimeric protein with an enhanced antibody affinity (KD=400 pM). Unexpectedly the dimer also incorporated a single heme molecule (KD=33 nM) that stabilised the dimer (Tm=59 degrees C with heme, 35 degrees C without heme). These results show that the binding affinities of flexible polypeptide segments can be greatly enhanced on protein folding, and that the folding can be stabilised by prosthetic groups. This supports the hypothesis that sub-domain polypeptide segments with functional activities may have contributed to domain creation in early evolution.

Inhibition of papillomavirus protein function in cervical cancer cells by intrabody targeting.

Papillomaviruses (HPVs) are a major cause of human disease, and are responsible for approximately half a million cases of cervical cancer each year. HPVs also cause genital warts, and are the most common sexually transmitted disease in many countries. Despite their importance, there are currently no specific antivirals that are active against HPVs. Papillomavirus protein function is mediated largely by protein-protein interactions, which are difficult to inhibit using conventional approaches. To circumvent these problems, we have prepared an scFv library, and have used this to isolate high-affinity binding molecules that may stearically hinder the association of E6 with p53 and prevent E6-mediated p53 degradation in cervical cancer cells. One of the molecules isolated from the library (GTE6-1), had an affinity for 16E6 of 60nM, and bound within the first zinc finger of the protein. GTE6-1 was able to associate with non-denatured E6 following expression in mammalian cells and could inhibit E6-mediated p53 degradation in in vitro assays. E6-mediated p53 degradation is essential for the continuous growth of cervical cancer cells caused by HPV16. To examine the potential of GTE6-1 as an inhibitor of E6 function in such cells, the molecule was expressed in scFv, diabody and triabody formats in a number of cell lines that are driven to proliferate by the HPV16 oncogenes E6 and E7, including the cervical cancer cell line SiHa. In contrast to small E6-binding peptides containing the ELLG E6-binding motif, GTE6-1 expression lead to changes in nuclear structure, the appearance of apoptosis markers, and an elevation in the levels of p53. No effects were seen with a control scFv molecule, or when GTE6-1 was expressed in cells that are driven to proliferate by simian virus 40 (SV40) T-antigen. Given the accessibility of HPV-associated lesions to topical therapy, our results suggest that large interfering molecules such as intrabodies may be useful inhibitors of viral protein-protein interactions and be particularly appropriate for the treatment of HPV-associated disease.

Aggregation-resistant domain antibodies selected on phage by heat denaturation.

We describe a method for selecting aggregation-resistant proteins by heat denaturation. This is illustrated with antibody heavy chain variable domains (dAbs), which are prone to aggregate. The dAbs were displayed multivalently at the infective tip of filamentous bacteriophage, and heated transiently to induce unfolding and to promote aggregation of the dAbs. After cooling, the dAbs were selected for binding to protein A (a ligand common to these folded dAbs). Phage displaying dAbs that unfold reversibly were thereby enriched with respect to those that do not. From a repertoire of phage dAbs, six dAbs were characterized after selection; they all resisted aggregation, and were soluble, well expressed in bacteria and could be purified in good yields. The method should be useful for making aggregation-resistant proteins and for helping to identify features that promote or prevent protein aggregation, including those responsible for misfolding diseases.

Folding and stability of a primitive protein.

We have previously attempted to simulate domain creation in early protein evolution by recombining polypeptide segments from non-homologous proteins, and we have described the structure of one such de novo protein, 1b11, a segment-swapped tetramer with novel architecture. Here, we have analyzed the thermodynamic stability and folding kinetics of the 1b11 tetramer and its monomeric and dimeric intermediates, and of 1b11 mutants with changes at the domain interface. Denatured 1b11 polypeptides fold into transient, folded monomers with marginal stability (DeltaG<1kcalmol(-1)) which convert rapidly ( approximately 6x10(4)M(-1)s(-1)) into dimers (DeltaG=9.8kcal/mol) and then more slowly ( approximately 3M(-1)s(-1)) into tetramers (DeltaG=28kcalmol(-1)). Segment swapping takes place during dimerization, as suggested by mass spectroscopic analysis of covalently linked peptides derived from proteolysis of a disulfide-linked dimer. Our results confirm that segment swapping and associated oligomerization are both powerful ways of stabilizing proteins, and we suggest that this may have been a feature of early protein evolution.

Generating molecular diversity by homologous recombination in Escherichia coli.

We explored the use of recE-mediated homologous recombination to generate molecular diversity in Escherichia coli. Two homologous genes were placed on different phagemid vectors each comprising multiple EcoRI restriction sites and overlapping N- and C-terminal portions of beta-lactamase. By co-infection of these phage into RecE+ EcoRI+ E.coli, we were able to introduce double-strand breaks into these vectors, allowing efficient homologous recombination (in up to 10% of bacteria) by the recE pathway and selection of the recombinants by resistance to ampicillin. Recombination gave single crossovers; these were more frequent near the EcoRI sites and the recombination frequency increased with the target length and degree of homology. The system was used to create a large combinatorial chicken antibody library (10(10)) for display on filamentous phage and to isolate several antibody fragments with binding affinities in the 10-100 nM range.

Crystal structure of HEL4, a soluble, refoldable human V(H) single domain with a germ-line scaffold.

The antigen binding site of antibodies usually comprises associated heavy (V(H)) and light (V(L)) chain variable domains, but in camels and llamas, the binding site frequently comprises the heavy chain variable domain only (referred to as V(HH)). In contrast to reported human V(H) domains, V(HH) domains are well expressed from bacteria and yeast, are readily purified in soluble form and refold reversibly after heat-denaturation. These desirable properties have been attributed to highly conserved substitutions of the hydrophobic residues of V(H) domains, which normally interact with complementary V(L) domains. Here, we describe the discovery and characterisation of an isolated human V(H) domain (HEL4) with properties similar to those of V(HH) domains. HEL4 is highly soluble at concentrations of > or =3 mM, essentially monomeric and resistant to aggregation upon thermodenaturation at concentrations as high as 56 microM. However, in contrast to V(HH) domains, the hydrophobic framework residues of the V(H):V(L) interface are maintained and the only sequence changes from the corresponding human germ-line segment (V3-23/DP-47) are located in the loops comprising the complementarity determining regions (CDRs). The crystallographic structure of HEL4 reveals an unusual feature; the side-chain of a framework residue (Trp47) is flipped into a cavity formed by Gly35 of CDR1, thereby increasing the hydrophilicity of the V(H):V(L) interface. To evaluate the specific contribution of Gly35 to domain properties, Gly35 was introduced into a V(H) domain with poor solution properties. This greatly enhanced the recovery of the mutant from a gel filtration matrix, but had little effect on its ability to refold reversibly after heat denaturation. Our results confirm the importance of a hydrophilic V(H):V(L) interface for purification of isolated V(H) domains, and constitute a step towards the design of isolated human V(H) domains with practical properties for immunotherapy.

Selection of large diversities of antiidiotypic antibody fragments by phage display.

Antiidiotypic antibodies (Ab2) are needed as tools for a better understanding of molecular mimicry and the immunological network, and for many potential applications in the biomedical and pharmaceutical field. Antiidiotypic antibodies mimicking carbohydrate or conformational epitopes (Ab2beta) are of considerable interest as surrogate immunogens for cancer vaccination. However, it has so far been difficult and tedious to produce Ab2s to a given antigen. Here we describe a fast and reliable technique for generating large diversities of antiidiotypic single chain antibody fragments from non-immunized phagemid libraries using phage display. Key elements are a specific elution with the original antigen followed by trypsin treatment of the eluted phages in combination with the protease sensitive helperphage KM13. This novel method was compared with various conventional selection and elution methods, including, specific elution with or without trypsin treatment, elution with glycine at pH 2.2 with or without trypsin treatment, and elution by trypsin treatment only. The results clearly show that specific elution in combination with trypsin treatment of the eluted phages is by far superior to the other conventional methods, enabling for the first time the generation of a large variety of Ab2s after only two to three rounds of selection, thereby maintaining maximum diversity. We obtained 28 to 88 antiidiotypes out of 96 tested clones after two to three rounds of selection with a diversity of 55-90 %. This was achieved for two carbohydrate (di-, and tetrasaccharides) and one conformational protein epitope using two large naïve libraries and their corresponding monoclonal Ab1. The antiidiotypic nature of the selected scFv-phages was verified by ELISA and immunocytochemistry inhibition experiments.

Encoded libraries of chemically modified peptides.

The use of powerful technologies for generating and screening DNA-encoded protein libraries has helped drive the development of proteins as pharmaceutical ligands. However the development of peptides as pharmaceutical ligands has been more limited. Although encoded peptide libraries are typically several orders of magnitude larger than classical chemical libraries, can be more readily screened, and can give rise to higher affinity ligands, their use as pharmaceutical ligands is limited by their intrinsic properties. Two of the intrinsic limitations include the rotational flexibility of the peptide backbone and the limited number (20) of natural amino acids. However these limitations can be overcome by use of chemical modification. For example, the libraries can be modified to introduce topological constraints such as cyclization linkers, or to introduce new chemical entities such as small molecule ligands, fluorophores and photo-switchable compounds. This article reviews the chemistry involved, the properties of the peptide ligands, and the new opportunities offered by chemical modification of DNA-encoded peptide libraries.

Subunit disassembly and inhibition of TNFα by a semi-synthetic bicyclic peptide.

Macrocyclic peptides are potentially a source of powerful drugs, but their de novo discovery remains challenging. Here we describe the discovery of a high-affinity (Kd = 10 nM) peptide macrocycle (M21) against human tumor necrosis factor-alpha (hTNFα), a key drug target in the treatment of inflammatory disorders, directly from diverse semi-synthetic phage peptide repertoires. The bicyclic peptide M21 (ACPPCLWQVLC) comprises two loops covalently anchored to a 2,4,6-trimethyl-mesitylene core and upon binding induces disassembly of the trimeric TNFα cytokine into dimers and monomers. A 2.9 Å crystal structure of the M21/hTNFα complex reveals the peptide bound to a hTNFα dimer at a normally buried epitope in the trimer interface overlapping the binding site of a previously discovered small molecule ligand (SPD304), which also induces TNF trimer dissociation and synergizes with M21 in the inhibition of TNFα cytotoxicity. The discovery of M21 underlines the potential of semi-synthetic bicyclic peptides as ligands for the discovery of cryptic epitopes, some of which are poorly accessible to antibodies.

A native-like artificial protein from antisense DNA.

We describe the creation of folded chimaeric proteins by combining a designed polypeptide segment (bait) derived from a beta-sheet of a human antibody variable domain with random polypeptide segments encoded by human cDNA fragments. The repertoire of polypeptides was displayed on the surface of filamentous bacteriophage and folded polypeptides were selected by proteolysis. One of these, 2a6, was readily expressed in the Escherichia coli cytoplasm as a soluble and protease-resistant protein and could be purified after heating the bacterial lysate to 90 degrees C. Soluble 2a6 is dimeric and its CD spectrum is consistent with components of both alpha and beta structure. 2a6 cooperatively and reversibly unfolds by heat or urea with a folding energy of 11.4 kcal mol(-1) for the transition between folded dimer and unfolded monomer and its refolding steps proceed without the formation of detectable aggregates. Its stability and folding properties are therefore typical of native proteins. Sequence analysis revealed that the cDNA segment in 2a6 was recruited from the antisense strand of a human gene, suggesting that antisense sequences can provide a reservoir for the evolution of soluble and stable proteins.

Selection of optical biosensors from chemisynthetic antibody libraries.

We describe a method for creating antibodies with a fluorescent reporter integrated into the antigen-binding site. A reporter molecule was chemically linked to a hypervariable loop of an antibody repertoire displayed on phage, and this repertoire was selected for antigen binding. In one selected antibody, the fluorescence of the probe responded quantitatively to antigen binding. The method may have application for the engineering of homogeneous immunoassays.

Bicyclic peptide inhibitor reveals large contact interface with a protease target.

From a large combinatorial library of chemically constrained bicyclic peptides we isolated a selective and potent (K(i) = 53 nM) inhibitor of human urokinase-type plasminogen activator (uPA) and crystallized the complex. This revealed an extended structure of the peptide with both peptide loops engaging the target to form a large interaction surface of 701 Å(2) with multiple hydrogen bonds and complementary charge interactions, explaining the high affinity and specificity of the inhibitor. The interface resembles that between two proteins and suggests that these constrained peptides have the potential to act as small protein mimics.

Thermodynamically stable aggregation-resistant antibody domains through directed evolution.

Protein aggregates are usually formed by interactions between unfolded or partially unfolded species, and often occur when a protein is denatured by, for example, heat or low pH. In earlier work, we used a Darwinian selection strategy to create human antibody variable domains that resisted heat aggregation. The repertoires of domains were displayed on filamentous phage and denatured (at 80 degrees C in pH 7.4), and folded domains were selected by binding to a generic ligand after cooling. This process appeared to select for domains with denatured states that resisted aggregation, but the domains only had low free energies of folding (Delta G(N-D)(o)=15-20 kJ/mol at 25 degrees C in pH 7.4). Here, using the same phage repertoire, we have extended the method to the selection of domains resistant to acid aggregation. In this case, however, the thermodynamic stabilities of selected domains were higher than those selected by thermal denaturation (under both neutral and acidic conditions; Delta G(N-D)(o)=26-47 kJ/mol at 25 degrees C in pH 7.4, or Delta G(N-D)(o)=27-34 kJ/mol in pH 3.2). Furthermore, we identified a key determinant (Arg28) that increased the aggregation resistance of the denatured states of the domains at low pH without compromising their thermodynamic stabilities. Thus, the selection process yielded domains that combined thermodynamic stability and aggregation-resistant unfolded states. We suggest that changes to these properties are controlled by the extent to which the folding equilibrium is displaced during the process of selection.