WO2013050617A9 - Structured polypeptides with sarcosine linkers - Google Patents

Abstract

The invention provides a polypeptide ligand comprising a polypeptide comprising at least three reactive groups, separated by at least two loop sequences, and a molecular scaffold which forms covalent bonds with the reactive groups of the polypeptide such that at least two polypeptide loops are formed on the molecular scaffold, further comprising an amino acid polymer comprising at least two sarcosine monomers.

Description

Structured polypeptides with Sarcosine linkers

The present invention relates to the use of sarcosine linkers with polypeptides, including structured polypeptides in which polypeptides are covalently connected to molecular scaffolds such that two or more peptide loops are subtended between attachment points to the scaffold.

Cyclic peptides are able to bind with high affinity and target specificity to protein targets and hence are an attractive molecule class for the development of therapeutics. In fact, several cyclic peptides are already successfully used in the clinic, such as for example the antibacterial peptide vancomycin, the immunosuppressant drug cyclosporine or the anti-cancer drug ocreotide (Driggers, et al., Nat Rev Drug Discov 2008, 7 (7), 608-24).

Different research teams have previously tethered polypeptides with cysteine residues to a synthetic molecular structure (Kemp, D. S. and McNamara, P. E., J. Org. Chem, 1985; Timmerman, P. et al., ChemBioChem, 2005). Meloen and co-workers had used tris(bromomethyl)benzene and related molecules for rapid and quantitative cyclisation of multiple peptide loops onto synthetic scaffolds for structural mimicry of protein surfaces (Timmerman, P. et al., ChemBioChem, 2005). Methods for the generation of candidate drug compounds wherein said compounds are generated by linking cysteine containing polypeptides to a molecular scaffold as for example tris(bromomethyl)benzene are disclosed in WO 2004/077062 and WO 2006/078161 . WO2004/077062 discloses a method of selecting a candidate drug compound. In particular, this document discloses various scaffold molecules comprising first and second reactive groups, and contacting said scaffold with a further molecule to form at least two linkages between the scaffold and the further molecule in a coupling reaction.

WO2006/078161 discloses binding compounds, immunogenic compounds and peptidomimetics. This document discloses the artificial synthesis of various collections of peptides taken from existing proteins. These peptides are then combined with a constant synthetic peptide having some amino acid changes introduced in order to produce combinatorial libraries. By introducing this diversity via the chemical linkage to separate peptides featuring various amino acid changes, an increased opportunity to find the desired binding activity is provided. Figure 7 of this document shows a schematic representation of the synthesis of various loop peptide constructs. The constructs disclosed in this document rely on -SH functionalised peptides, typically comprising cysteine residues, and heteroaromatic groups on the scaffold, typically comprising benzylic halogen substituents such as bis- or tris-bromophenylbenzene. Such groups react to form a thioether linkage between the peptide and the scaffold.

We recently developed a phage display-based combinatorial approach to generate and screen large libraries of bicyclic peptides to targets of interest (Heinis, et al., Nat Chem Biol 2009, 5 (7), 502-7; see also international patent application WO2009/098450). Briefly, combinatorial libraries of linear peptides containing three cysteine residues and two regions of six random amino acids (Cys-(Xaa)₆-Cys-(Xaa)₆-Cys) were displayed on phage and cyclised by covalently linking the cysteine side chains to a small molecule (tris-(bromomethyl)benzene). Bicyclic peptides isolated in affinity selections to the human proteases cathepsin G and plasma kallikrein (PK) had nanomolar inhibitory constants. The best inhibitor, PK15, inhibits human PK (hPK) with a K, of 3 nM. A synthetic, small peptidic inhibitor such as PK15 having the above described potency and target selectivity has potential application as a therapeutic to control PK activity in hereditary angioedema, a life-threatening disease which is characterized by recurrent episodes of edema or to prevent contact activation in cardiopulmonary bypass surgery.

It is desirable to attach therapeutic peptides, including antibodies, to functional groups. In order to allow both the polypeptide and the functional group to operate, a molecular spacer is often required to spatially separate distinct functionalities on a molecule so that they do not interfere with each other. In biotherapeutic polypeptides (peptides, proteins, antibodies and the like), such spacers may be genetically encoded or introduced during chemical synthesis. Requirements of molecular spacers include lack of toxicity, solubility and stability. Another parameter which affects functionality is the degree of flexibility of the spacer, which is determined by its chemical nature.

Traditionally, genetically encoded spacers on proteins such as antibodies are repeat units of Gly or Gly-Ser. The bulk properties of the much larger protein to which they are attached outweighs the aggregation potential of Gly/Gly-Ser linkers. However, on much smaller peptides, the properties of Gly/Gly-Ser start to dominate and can lead to problems with regard to their physico-chemical properties. Additionally, Gly-Ser sequences are notoriously difficult to synthesise in synthetic peptides, and workarounds such as dipeptides and pseudoprolines suffer from a high cost of the required compounds.

Perhaps the most commonly used synthetic molecular spacer is based on repeat units of polyethyleneglycol (PEG). It is biologically compatible yet stable, flexible, and highly water soluble. Its advantages are again offset by the high cost for PEGs of defined length/mass. The high degree of flexibility may be of disadvantage in certain circumstances, though advantageous in others. Polymers such as polyproline and polyhydroxyproline are highly rigid, but have very low solubility and are thus of limited application. One specific application of molecular spacers concerns the recognized problem that small peptides have limited half-lives in circulation in human plasma, and that it can be desirable to increase this half-life. In the case of antibody fragments, it is known to increase half-life by conjugating the antibody to a bulking agent, such as human serum albumin (HSA). This can be done directly, or by use of a ligand with affinity for HSA. For example, it is known to design dual specific antibodies in which one specificity is directed at HSA. Moreover, it is known to provide linkers which bind to HSA; see WO2009126920. Conjugation to HSA as also been disclosed in respect of structured polypeptides; see WO20100891 15.

Summary of the Invention We have investigated the use of sarcosine in linkers for use in conjunction with polypeptides. Sarcosine is N-methyl glycine, a non-encoded amino acid that is ubiquitous in tissue fluids (muscle, blood etc). It is an intermediate in several biosynthetic pathways. As a result of N-methylation, polysarcosine displays properties strikingly different to those of polyglycine: It is highly soluble in water, adopts an extended configuration, and is readily synthesised (by FmocSPPS) at defined lengths at high purity. In addition, it displays a lack of the any tendency to aggregate, likely due to the absence of H-bonding potential that otherwise would lead to aggregation (Teufel et al (201 1 ) J. Mol. Biol, 409, 2. 250-262)

As described herein, polysarcosine acts as a solubilising semi-rigid molecular spacer that lacks distinct functional groups (such as ionisable groups, free hydroxyls, and the like), whose monomers are bio-compatible. Moreover, the polymer is protease resistant and stable to environmental influences

Therefore, in a first aspect, there is provided the use of polysarcosine as a linker.

We have found that polysarcosine has highly advantageous properties when used as a linker, aiding the solubility of polypeptides to which it is attached, and promoting independent function of groups which are linked though such polymers. In one embodiment, the linker is used to attach a polypeptide to a functional group. The functional group can be another polypeptide, a binding agent, a drug or any other desired compound.

The polypeptide may be any polypeptide, but is, in one embodiment, a polypeptide with binding properties, which is capable of binding to a specific target.

In a further aspect, there is provided a polypeptide comprising a non-sarcosine sequence and a sarcosine polymer. Such a peptide can be constructed by joining a polypeptide which does not comprise sarcosine with a sarcosine polymer. The sarcosine polymer may contain one or more non-sarcosine amino acids, and comprises at least two sarcosine amino acids.

For example, the polypeptide is a polypeptide ligand. In one embodiment, there is provided a polypeptide ligand, wherein the polypeptide ligand comprises a polypeptide comprising at least three reactive groups, separated by at least two loop sequences, and a molecular scaffold which forms covalent bonds with the reactive groups of the polypeptide such that at least two polypeptide loops are formed on the molecular scaffold, further comprising an amino acid polymer comprising at least two sarcosine molecules.

The sarcosine polymer may function as a linker or spacer, attaching another molecule or group to the polypeptide. Alternatively, it may function as an extension, being attached solely to the polypeptide. Moreover, it can be used to create tandem structured polypeptides, in which two structured polypeptides having the same or different functions can be linked together.

In one embodiment, therefore, the sarcosine polymer is attached to the C-terminus of the polypeptide.

According to another embodiment, the sarcosine polymer is attached to the N-terminus of the polypeptide.

The use of sarcosine polymers increases the solubility of the structured polypeptide, when attached to the N- or C-termini, or to an internal binding site on the polypeptide, such as a suitable amino acid side-chain, or on a molecular scaffold.

In another embodiment, the sarcosine polymer separates the structured polypeptide from a functional group. This can be, for example, a polypeptide or a ligand. In one embodiment, the functional group can be any long-chain fatty acid group, including for example a myristate group, a stearate group, an arachidate group or a palmitate group. Long-chain fatty acids such as palmitic acid bind effectively to HSA in human plasma, and therefore attach the polypeptide to HSA through the sarcosine linker. This increases the half-life of the polypeptide in the circulation.

In another embodiment, the sarcosine linker separates the polypeptide from a second polypeptide. In one embodiment, one or both of the polypeptides is a structured polypeptide. This allows the creation of tandem ligands, which can be bispecific or monospecific. In one embodiment, three or more polypeptides could be linked together by means of sarcosine linkers. A trimeric, tetrameric or larger polymer can have multiple specificities, if required.

In yet another embodiment, the sarcosine linker separates the polypeptide from a metal chelating group, such as DOTA (1 ,4,7,10-tetraazacyclododecane-1 ,4,7,10-tetraacetic acid). DOTA complexes trivalent ions with high affinity, and in this instance could be employed for transporting cytotoxic radioactive isotopes to a desirable site in an organism. The sarcosine linker is used as a spacer separating the metal chelator from the targeting bicyclic peptide.

The sarcosine polymer itself comprises, in one embodiment, between 2 and 40 sarcosine monomers. Longer polymers are possible. However, polymers of between 4 and 6 sarcosine monomers have advantageous properties in certain embodiments.

The sarcosine polymer can be constituted from sarcosine monomers alone, and is included in regular Fmoc-based solid phase peptide synthesis as Fmoc-Sarcosine-OH. In certain embodiments, however, the polymer comprises one or more non-sarcosine amino acids. Any amino acid can be positioned next to a Sarcosine. For example, one or more of any non-sarcosine amino acid, for example glycine or serine, can be positioned at the N or C terminus of the polysarcosine sequence. The Sarcosine polymer can be part of any synthetic polypeptide or other polymer. In some embodiments, a Sarcosine polymer can be linked to recombinant proteins through routine bioconjugation techniques. We have found that polysarcosine aids the solubility of structured polypeptides, and forms an effective linker for attaching HSA via long-chain fatty acid groups such as palmitoyl, and attaching second structured polypeptides without impeding either functionality. In one embodiment, we provide a library of polypeptides which comprises polypeptides and linkers as set forth in the preceding embodiments.

Brief description of the figures Figure 1 illustrates the comparative turbidity of Gly20 (open circles) and Sar20 (filled circles) as an illustrator of solubility. An OD340 greater than zero indicates aggregation. Figure adapted from reference (Teufel et al (201 1 ) J. Mol. Biol. 409(2):250-62).

Figure 2 shows the binding of palmitoylated peptides to HSA with high affinity. The inclusion of a charged amino acid as indicated reduces the affinity to HSA. The sequence of the peptide used is PA-X-Sar6-DDC- fluorescein, where X = G, K, W, D.

Figure 3 shows the Kd of a palmitoylated peptide for pure human serum albumin, in comparison to an acetylated peptide. The Kd is approximately 140nM.

Figure 4 shows the apparent Kd of a palmitoylated peptide binding to components of human plasma. Binding is likely confined to serum albumin, which is present in human plasma at a ~ 600 μΜ concentration. The comparison is to a peptide whose palmitoyl group (C16) was replaced by a short chain acetyl group (C2). The apparent Kd is approximately 210nM, which is almost identical to that obtained with purified HSA

Figure 5 illustrates the three model peptides used in analysing the impact of a linker on HSA binding. Figure 6 shows HSA binding by palmitoylated bicycles appears independent on the type and length of molecular spacer (Sar6, HyP6, HyP12). The lower amplitude change seen for HyP12 is due to the greater physical distance between the peptide and the HSA.

Figure 7 illustrates the elimination tp half-life of PA-G-Sar6-PK15(TMB)-NH2 dosed intravenously, which is approximately 5.5 hours. Cmax: Maximum measured

concentration; Tmax: Time at which maximum concentration was measured; AUC 0-t: Area under the plasma drug concentration/time curve from 0 minutes to last quantifiable data point. Figure 8 illustrates the elimination half life of PA-G-Sar6-PK15(TMB)-NH2 dosed subcutaneously. The maximum concentration of PA-G-Sar6-PK15(TMB)-NH2 is reached after -8 hrs (Tmax). Bioavailability of the peptide is at -80 % Figure 9 is a table comparing the PK parameters of the test polypeptide in the presence and absence of palmitoylation, dosed at 5mg/kg.

Figure 10 is a reaction scheme illustrating the click chemistry used for the assembly of tandem structured polypeptides.

Figure 11 illustrates the synthetic procedure for a DOTA-containing polypeptide.

Figure 12 illustrates the quantitative mass addition observed on the peptide due to complexation of Y³⁺ (B) by MALDI TOF MS.

Detailed Description of the Invention

Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by those of ordinary skill in the art, such as in the arts of peptide chemistry, cell culture and phage display, nucleic acid chemistry and biochemistry. Standard techniques are used for molecular biology, genetic and biochemical methods (see Sambrook et al., Molecular Cloning: A Laboratory Manual, 3rd ed., 2001 , Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY; Ausubel et al., Short Protocols in Molecular Biology (1999) 4^th ed., John Wiley & Sons, Inc.), which are incorporated herein by reference. A polypeptide, or polypeptide ligand, as referred to herein, refers to a polypeptide which possesses a desired activity, for example a binding activity. The polypeptide, in certain embodiments, can be a structured polypeptide, which is a polypeptide covalently bound to a molecular scaffold. Typically, such peptides comprise two or more reactive groups which are capable of forming covalent bonds to the scaffold, and a sequence subtended between said reactive groups which is referred to as the loop sequence, since it forms a loop when the peptide is bound to the scaffold. In the present case, the peptides comprise at least three reactive groups, and form at least two loops on the scaffold.

The reactive groups are groups capable of forming a covalent bond with the molecular scaffold. Typically, the reactive groups are present on amino acid side chains on the peptide. Examples are amino-containing groups such as cysteine, lysine and selenocysteine.

The molecular scaffold is any molecule which is able to connect the peptide at multiple points to impart one or more structural features to the peptide. It is not a cross-linker, in that it does not merely replace a disulphide bond; instead, it provides two or more attachment points for the peptide. Preferably, the molecular scaffold comprises at least three attachment points for the peptide, referred to as scaffold reactive groups. These groups are capable of reacting to the reactive groups on the peptide to form a covalent bond. Preferred structures for molecular scaffolds are described below. Screening for binding activity (or any other desired activity) is conducted according to methods well known in the art, for instance from phage display technology. For example, targets immobilised to a solid phase can be used to identify and isolate binding members of a repertoire. Screening allows selection of members of a repertoire according to desired characteristics. The term library refers to a mixture of heterogeneous polypeptides or nucleic acids. The library is composed of members, which are not identical. To this extent, library is synonymous with repertoire. Sequence differences between library members are responsible for the diversity present in the library. The library may take the form of a simple mixture of polypeptides or nucleic acids, or may be in the form of organisms or cells, for example bacteria, viruses, animal or plant cells and the like, transformed with a library of nucleic acids. Preferably, each individual organism or cell contains only one or a limited number of library members.

In one embodiment, the nucleic acids are incorporated into expression vectors, in order to allow expression of the polypeptides encoded by the nucleic acids. In a preferred aspect, therefore, a library may take the form of a population of host organisms, each organism containing one or more copies of an expression vector containing a single member of the library in nucleic acid form which can be expressed to produce its corresponding polypeptide member. Thus, the population of host organisms has the potential to encode a large repertoire of genetically diverse polypeptide variants. In one embodiment, a library of nucleic acids encodes a repertoire of polypeptides. Each nucleic acid member of the library preferably has a sequence related to one or more other members of the library. By related sequence is meant an amino acid sequence having at least 50% identity, for example at least 60% identity, for example at least 70% identity, for example at least 80% identity, for example at least 90% identity, for example at least 95% identity, for example at least 98% identity, for example at least 99% identity to at least one other member of the library. Identity can be judged across a contiguous segment of at least 3 amino acids, for example at least 4, 5, 6, 7, 8, 9 or 10 amino acids, for example least 12 amino acids, for example least 14 amino acids, for example least 16 amino acids, for example least 17 amino acids or the full length of the reference sequence.

A repertoire is a collection of variants, in this case polypeptide variants, which differ in their sequence. Typically, the location and nature of the reactive groups will not vary, but the sequences forming the loops between them can be randomised. Repertoires differ in size, but should be considered to comprise at least 10² members. Repertoires of 10¹¹ or more members can be constructed.

(A) Construction of Peptide Ligands

(i) Molecular scaffold Molecular scaffolds are described in, for example, WO2009098450 and references cited therein, particularly WO2004077062 and WO2006078161 .

As noted in the foregoing documents, the molecular scaffold may be a small molecule, such as a small organic molecule.

In one embodiment the molecular scaffold may be, or may be based on, natural monomers such as nucleosides, sugars, or steroids. For example the molecular scaffold may comprise a short polymer of such entities, such as a dimer or a trimer.

In one embodiment the molecular scaffold is a compound of known toxicity, for example of low toxicity. Examples of suitable compounds include cholesterols, nucleotides, steroids, or existing drugs such as tamazepam. In one embodiment the molecular scaffold may be a macromolecule. In one embodiment the molecular scaffold is a macromolecule composed of amino acids, nucleotides or carbohydrates.

In one embodiment the molecular scaffold comprises reactive groups that are capable of reacting with functional group(s) of the polypeptide to form covalent bonds. The molecular scaffold may comprise chemical groups as amines, thiols, alcohols, ketones, aldehydes, nitriles, carboxylic acids, esters, alkenes, alkynes, azides, anhydrides, succinimides, maleimides, alkyl halides and acyl halides.

In one embodiment, the molecular scaffold may comprise or may consist of tris(bromomethyl)benzene, especially 1 ,3,5-Tris(bromomethyl)benzene (ΤΒΜΒ'), or a derivative thereof.

In one embodiment, the molecular scaffold is 2,4,6-Tris(bromomethyl)mesitylene. It is similar to 1 ,3,5-Tris(bromomethyl)benzene but contains additionally three methyl groups attached to the benzene ring. This has the advantage that the additional methyl groups may form further contacts with the polypeptide and hence add additional structural constraint.

The molecular scaffold of the invention contains chemical groups that allow functional groups of the polypeptide of the encoded library of the invention to form covalent links with the molecular scaffold. Said chemical groups are selected from a wide range of functionalities including amines, thiols, alcohols, ketones, aldehydes, nitriles, carboxylic acids, esters, alkenes, alkynes, anhydrides, succinimides, maleimides, azides, alkyl halides and acyl halides.

(ii) Polypeptide

The reactive groups of the polypeptides can be provided by side chains of natural or non- natural amino acids. The reactive groups of the polypeptides can be selected from thiol groups, amino groups, carboxyl groups, guanidinium groups, phenolic groups or hydroxyl groups. The reactive groups of the polypeptides can be be selected from azide, keto- carbonyl, alkyne, vinyl, or aryl halide groups. The reactive groups of the polypeptides for linking to a molecular scaffold can be the amino or carboxy termini of the polypeptide. In some embodiments each of the reactive groups of the polypeptide for linking to a molecular scaffold are of the same type. For example, each reactive group may be a cysteine residue. Further details are provided in WO2009098450.

In some embodiments the reactive groups for linking to a molecular scaffold may comprise two or more different types, or may comprise three or more different types. For example, the reactive groups may comprise two cysteine residues and one lysine residue, or may comprise one cysteine residue, one lysine residue and one N-terminal amine. Cysteine can be employed because it has the advantage that its reactivity is most different from all other amino acids. Scaffold reactive groups that could be used on the molecular scaffold to react with thiol groups of cysteines are alkyl halides (or also named halogenoalkanes or haloalkanes). Examples are bromomethylbenzene (the scaffold reactive group exemplified by TBMB) or iodoacetamide. Other scaffold reactive goups that are used to couple selectively compounds to cysteines in proteins are maleimides. Examples of maleimides which may be used as molecular scaffolds in the invention include: tris-(2-maleimidoethyl)amine, tris-(2-maleimidoethyl)benzene, tris- (maleimido)benzene. Selenocysteine is also a natural amino acid which has a similar reactivity to cysteine and can be used for the same reactions. Thus, wherever cysteine is mentioned, it is typically acceptable to substitute selenocysteine unless the context suggests otherwise.

Lysines (and primary amines of the N-terminus of peptides) are also suited as reactive groups to modify peptides on phage by linking to a molecular scaffold. However, they are more abundant in phage proteins than cysteines and there is a higher risk that phage particles might become cross-linked or that they might lose their infectivity. Nevertheless, it has been found that lysines are especially useful in intramolecular reactions (e.g. when a molecular scaffold is already linked to the phage peptide) to form a second or consecutive linkage with the molecular scaffold. In this case the molecular scaffold reacts preferentially with lysines of the displayed peptide (in particular lysines that are in close proximity). Scaffold reactive groups that react selectively with primary amines are succinimides, aldehydes or alkyl halides. In the bromomethyl group that is used in a number of the accompanying examples, the electrons of the benzene ring can stabilize the cationic transition state. This particular aryl halide is therefore 100-1000 times more reactive than alkyl halides. Examples of succinimides for use as molecular scaffold include tris-(succinimidyl aminotriacetate), 1 ,3,5-Benzenetriacetic acid. Examples of aldehydes for use as molecular scaffold include Triformylmethane. Examples of alkyl halides for use as molecular scaffold include 1 ,3,5-Tris(bromomethyl)-2,4,6- trimethylbenzene, 1 ,3,5-Tris(bromomethyl) benzene, 1 ,3,5-Tris(bromomethyl)-2,4,6- triethylbenzene.

The amino acids with reactive groups for linking to a molecular scaffold may be located at any suitable positions within the polypeptide. In order to influence the particular structures or loops created, the positions of the amino acids having the reactive groups may be varied by the skilled operator, e.g. by manipulation of the nucleic acid encoding the polypeptide in order to mutate the polypeptide produced. By such means, loop length can be manipulated in accordance with the present teaching.

For example, the polypeptide can comprise the sequence AC(X)_nC(X)_mCG, wherein X stands for a random natural amino acid, A for alanine, C for cysteine and G for glycine and n and m, which may be the same or different, are numbers between 3 and 6.

(iii) Reactive groups of the polypeptide

The molecular scaffold of the invention may be bonded to the polypeptide via functional or reactive groups on the polypeptide. These are typically formed from the side chains of particular amino acids found in the polypeptide polymer. Such reactive groups may be a cysteine side chain, a lysine side chain, or an N-terminal amine group or any other suitable reactive group. Again, details may be found in WO2009098450.

Examples of reactive groups of natural amino acids are the thiol group of cysteine, the amino group of lysine, the carboxyl group of aspartate or glutamate, the guanidinium group of arginine, the phenolic group of tyrosine or the hydroxyl group of serine. Non- natural amino acids can provide a wide range of reactive groups including an azide, a keto-carbonyl, an alkyne, a vinyl, or an aryl halide group. The amino and carboxyl group of the termini of the polypeptide can also serve as reactive groups to form covalent bonds to a molecular scaffold/molecular core.

The polypeptides of the invention contain at least three reactive groups. Said polypeptides can also contain four or more reactive groups. The more reactive groups are used, the more loops can be formed in the molecular scaffold.

In a preferred embodiment, polypeptides with three reactive groups are generated. Reaction of said polypeptides with a molecular scaffold/molecular core having a threefold rotational symmetry generates a single product isomer. The generation of a single product isomer is favourable for several reasons. The nucleic acids of the compound libraries encode only the primary sequences of the polypeptide but not the isomeric state of the molecules that are formed upon reaction of the polypeptide with the molecular core. If only one product isomer can be formed, the assignment of the nucleic acid to the product isomer is clearly defined. If multiple product isomers are formed, the nucleic acid can not give information about the nature of the product isomer that was isolated in a screening or selection process. The formation of a single product isomer is also advantageous if a specific member of a library of the invention is synthesized. In this case, the chemical reaction of the polypeptide with the molecular scaffold yields a single product isomer rather than a mixture of isomers.

In another embodiment of the invention, polypeptides with four reactive groups are generated. Reaction of said polypeptides with a molecular scaffold/molecular core having a tetrahedral symmetry generates two product isomers. Even though the two different product isomers are encoded by one and the same nucleic acid, the isomeric nature of the isolated isomer can be determined by chemically synthesizing both isomers, separating the two isomers and testing both isomers for binding to a target ligand.

In one embodiment of the invention, at least one of the reactive groups of the polypeptides is orthogonal to the remaining reactive groups. The use of orthogonal reactive groups allows the directing of said orthogonal reactive groups to specific sites of the molecular core. Linking strategies involving orthogonal reactive groups may be used to limit the number of product isomers formed. In other words, by choosing distinct or different reactive groups for one or more of the at least three bonds to those chosen for the remainder of the at least three bonds, a particular order of bonding or directing of specific reactive groups of the polypeptide to specific positions on the molecular scaffold may be usefully achieved.

In another embodiment, the reactive groups of the polypeptide of the invention are reacted with molecular linkers wherein said linkers are capable to react with a molecular scaffold so that the linker will intervene between the molecular scaffold and the polypeptide in the final bonded state.

In some embodiments, amino acids of the members of the libraries or sets of polypeptides can be replaced by any natural or non-natural amino acid. Excluded from these exchangeable amino acids are the ones harbouring functional groups for cross- linking the polypeptides to a molecular core, such that the loop sequences alone are exchangeable. The exchangeable polypeptide sequences have either random sequences, constant sequences or sequences with random and constant amino acids. The amino acids with reactive groups are either located in defined positions within the polypeptide, since the position of these amino acids determines loop size. In one embodiment, an polypeptide with three reactive groups has the sequence (X)iY(X)_mY(X)_nY(X)₀, wherein Y represents an amino acid with a reactive group, X represents a random amino acid, m and n are numbers between 3 and 6 defining the length of intervening polypeptide segments, which may be the same or different, and I and o are numbers between 0 and 20 defining the length of flanking polypeptide segments.

Alternatives to thiol-mediated conjugations can be used to attach the molecular scaffold to the peptide via covalent interactions. Alternatively these techniques may be used in modification or attachment of further moieties (such as small molecules of interest which are distinct from the molecular scaffold) to the polypeptide after they have been selected or isolated according to the present invention - in this embodiment then clearly the attachment need not be covalent and may embrace non-covalent attachment. These methods may be used instead of (or in combination with) the thiol mediated methods by producing phage that display proteins and peptides bearing unnatural amino acids with the requisite chemical reactive groups, in combination small molecules that bear the complementary reactive group, or by incorporating the unnatural amino acids into a chemically or recombinantly synthesised polypeptide when the molecule is being made after the selection/isolation phase. Further details can be found in WO2009098450 or Heinis, et al., Nat Chem Biol 2009, 5 (7), 502-7.

(iv) Combination of loops to form multispecific molecules

Loops from peptide ligands, or repertoires of peptide ligands, are advantageously combined by sequencing and de novo synthesis of a polypeptide incorporating the combined loops. Alternatively, nucleic acids encoding such polypeptides can be synthesised.

Where repertoires are to be combined, particularly single loop repertoires, the nucleic acids encoding the repertoires are advantageously digested and re-ligated, to form a novel repertoire having different combinations of loops from the constituent repertoires. Phage vectors can include polylinkers and other sites for restriction enzymes which can provide unique points for cutting and relegation the vectors, to create the desired multispecific peptide ligands. Methods for manipulating phage libraries are well known in respect of antibodies, and can be applied in the present case also.

(v) Attachment of Effector Groups and Functional Groups

Effector and/or functional groups can be attached, for example, to the N or C termini of the polypeptide, to the amino acid sidechains, the amino acid backbone or to the molecular scaffold. As described herein, the use of linkers for attaching effector and functional groups is advantageous. In one embodiment, linkers are sarcosine-based, and comprise a sarcosine polymer. Non-sarcosine amino acids may be included, if desired. Sarcosine linkers are highly soluble (Figure 1 ), and even in absence of any effector group they can be used to increase the solubility of a structured polypeptide. Advantageously, they are positioned on the C-terminus or the N-terminus.

Linkers, such as sarcosine linkers, allow attachment of effector and functional groups in such a manner that the function of the effector and functional group, and the structured polypeptide, are not impeded by physical proximity. The use of fatty acid groups such as palmitoyl is described; palmitoyl binds to HSA with high affinity, thus providing a means to increase the half-life of polypeptides, including structured polypeptides.

Other appropriate effector groups include antibodies and parts or fragments thereof. For instance, an effector group can include an antibody light chain constant region (CL), an antibody CH1 heavy chain domain, an antibody CH2 heavy chain domain, an antibody CH3 heavy chain domain, or any combination thereof, in addition to the one or more constant region domains. An effector group may also comprise a hinge region of an antibody (such a region normally being found between the CH1 and CH2 domains of an IgG molecule). In a further preferred embodiment of this aspect of the invention, an effector group according to the present invention is an Fc region of an IgG molecule. Advantageously, a peptide ligand-effector group according to the present invention comprises or consists of a peptide ligand Fc fusion having a tp half-life of a day or more, two days or more, 3 days or more, 4 days or more, 5 days or more, 6 days or more or 7 days or more. Most advantageously, the peptide ligand according to the present invention comprises or consists of a peptide ligand Fc fusion having a tp half-life of a day or more.

Functional groups include, in general, binding groups, drugs, reactive groups for the attachment of other entities, functional groups which aid uptake of the macrocyclic peptides into cells, and the like. Such groups can be attached directly, or via sarcosine linkers as described herein. For example, therefore, there is provided a polypeptide ligand which is attached to a long chain fatty acid group at the N- or C-terminus, in the first instance through palmitoylation of the free amine on the N-terminus of the polypeptide ligand, and in the second instance through post-synthetic palmitoylation on a free amine of a C-terminal lysine.

Drugs can advantageously be attached via cleavable linkers. Examples include cleavable linkers such as the valine-citrulline linker used to link monomethyl auristatin E to a monoclonal antibody, which is cleaved in tumour cells to release the cytotoxic drug. Cleavable linkers are widely used in antibody-drug conjugate construction, and several are available commercially and in the literature.

The ability of peptides to penetrate into cells will allow peptides against intracellular targets to be effective. Targets that can be accessed by peptides with the ability to penetrate into cells include transcription factors, intracellular signalling molecules such as tyrosine kinases and molecules involved in the apoptotic pathway. Functional groups which enable the penetration of cells include peptides or chemical groups which have been added either to the peptide or the molecular scaffold. Peptides such as those derived from such as VP22, HIV-Tat, a homeobox protein of Drosophila (Antennapedia), e.g. as described in Chen and Harrison, Biochemical Society Transactions (2007) Volume 35, part 4, p821 "Cell-penetrating peptides in drug development: enabling intracellular targets" and "Intracellular delivery of large molecules and small peptides by cell penetrating peptides" by Gupta et al. in Advanced Drug Discovery Reviews (2004) Volume 57 9637. Examples of short peptides which have been shown to be efficient at translocation through plasma membranes include the 16 amino acid penetratin peptide from Drosophila Antennapedia protein (Derossi et al (1994) J Biol. Chem. Volume 269 p10444 "The third helix of the Antennapedia homeodomain translocates through biological membranes"), the 18 amino acid 'model amphipathic peptide' (Oehlke et al (1998) Biochim Biophys Acts Volume 1414 p127 "Cellular uptake of an alpha-helical amphipathic model peptide with the potential to deliver polar compounds into the cell interior non-endocytically") and arginine rich regions of the HIV TAT protein. Non peptidic approaches include the use of small molecule mimics or SMOCs that can be easily attached to biomolecules (Okuyama et al (2007) Nature Methods Volume 4 p153 'Small- molecule mimics of an a-helix for efficient transport of proteins into cells'. Other chemical strategies to add guanidinium groups to molecules also enhance cell penetration (Elson- Scwab et al (2007) J Biol Chem Volume 282 p13585 "Guanidinylated Neomcyin Delivers Large Bioactive Cargo into cells through a heparin Sulphate Dependent Pathway"). Small molecular weight molecules such as steroids may be added to the molecular scaffold to enhance uptake into cells. One class of functional groups which may be attached to peptide ligands includes antibodies and binding fragments thereof, such as Fab, Fv or single domain fragments. In particular, antibodies which bind to proteins capable of increasing the half life of the peptide ligand in vivo may be used. As set out above, palmitoyl groups can be attached to the polypeptide via a sarcosine linker, and increase the half-life of the polypeptide as a result of attachment to HSA.

RGD peptides, which bind to integrins which are present on many cells, may also be incorporated.

In one embodiment, a peptide ligand-effector group according to the invention has a tp half-life selected from the group consisting of: 12 hours or more, 24 hours or more, 2 days or more, 3 days or more, 4 days or more, 5 days or more, 6 days or more, 7 days or more, 8 days or more, 9 days or more, 10 days or more, 1 1 days or more, 12 days or more, 13 days or more, 14 days or more, 15 days or more or 20 days or more.

Advantageously a peptide ligand-effector group or composition according to the invention will have a tp half life in the range 12 to 60 hours. In a further embodiment, it will have a t half-life of a day or more. In a further embodiment still, it will be in the range 12 to 26 hours.

Functional groups include drugs, such as cytotoxic agents for cancer therapy. These include Alkylating agents such as Cisplatin and carboplatin, as well as oxaliplatin, mechlorethamine, cyclophosphamide, chlorambucil, ifosfamide; Anti-metabolites including purine analogs azathioprine and mercaptopurine)) or pyrimidine analogs; plant alkaloids and terpenoids including vinca alkaloids such as Vincristine, Vinblastine, Vinorelbine and Vindesine; Podophyllotoxin and its derivatives etoposide and teniposide; Taxanes, including paclitaxel, originally known as Taxol; topoisomerase inhibitors including camptothecins: irinotecan and topotecan, and type II inhibitors including amsacrine, etoposide, etoposide phosphate, and teniposide. Further agents can include Antitumour antibiotics which include the immunosuppressant dactinomycin (which is used in kidney transplantations), doxorubicin, epirubicin, bleomycin and others.

Possible effector groups also include enzymes, for instance such as carboxypeptidase G2 for use in enzyme/prodrug therapy, where the peptide ligand replaces antibodies in ADEPT. (vi) Synthesis

It should be noted that once a polypeptide of interest is isolated or identified according to the present invention, then its subsequent synthesis may be simplified wherever possible. Thus, groups or sets of polypeptides need not be produced by recombinant DNA techniques. For example, the sequence of polypeptides of interest may be determined, and they may be manufactured synthetically by standard techniques followed by reaction with a molecular scaffold in vitro. When this is performed, standard chemistry may be used since there is no longer any need to preserve the functionality or integrity of the genetically encoded carrier particle, such as phage. This enables the rapid large scale preparation of soluble material for further downstream experiments or validation. In this regard, large scale preparation of the candidates or leads identified by the methods of the present invention could be accomplished using conventional chemistry such as that disclosed in Timmerman et al.

Thus, the invention also relates to manufacture of polypeptides or conjugates selected as set out herein, wherein the manufacture comprises optional further steps as explained below. In one embodiment, these steps are carried out on the end product polypeptide/conjugate made by chemical synthesis, rather than on the phage.

Optionally amino acid residues in the polypeptide of interest may be substituted when manufacturing a conjugate or complex e.g. after the initial isolation/identification step. Peptides can also be extended, to incorporate for example another loop and therefore introduce multiple specificities.

To extend the peptide, it may simply be extended chemically at its N-terminus or C- terminus using standard solid phase or solution phase chemistry. Standard protein chemistry may be used to introduce an activatable N- or C-terminus. Alternatively additions may be made by fragment condensation or native chemical ligation e.g. as described in (Dawson PE, Muir TW, Clark-Lewis I, Kent, SBH. 1994. Synthesis of Proteins by Native Chemical Ligation. Science 266:776-779), or by enzymes, for example using subtiligase as described in (Subtiligase: a tool for semisynthesis of proteins Chang TK, Jackson DY, Burnier JP, Wells JA Proc Natl Acad Sci U S A. 1994 Dec 20;91 (26):12544-8 or in Bioorganic & Medicinal Chemistry Letters Tags for labelling protein N-termini with subtiligase for proteomics Volume 18, Issue 22, 15 November 2008, Pages 6000-6003 Tags for labeling protein N-termini with subtiligase for proteomics; Hikari A.I. Yoshihara, Sami Mahrus and James A. Wells). Alternatively, the peptides may be extended or modified by further conjugation through disulphide bonds. This has the additional advantage of allowing the first and second peptide to dissociate from each other once within the reducing environment of the cell. In this case, the molecular scaffold (eg. TBMB) could be added during the chemical synthesis of the first peptide so as to react with the three cysteine groups; a further cysteine could then be appended to the N-terminus of the first peptide, so that this cysteine only reacted with a free cysteine of the second peptide.

Similar techniques apply equally to the synthesis/coupling of two bicyclic and bispecific macrocycles, potentially creating a tetraspecific molecule. Furthermore, addition of other functional groups or effector groups may be accomplished in the same manner, using appropriate chemistry, coupling at the N- or C-termini or via side chains. In one embodiment, the coupling is conducted in such a manner that it does not block the activity of either entity. (B) Repertoires, sets and groups of polypeptide ligands (i) Construction of Libraries

Libraries intended for selection may be constructed using techniques known in the art, for example as set forth in WO2004/077062, or biological systems, including phage vector systems as described herein. Other vector systems are known in the art, and include other phage (for instance, phage lambda), bacterial plasmid expression vectors, eukaryotic cell-based expression vectors, including yeast vectors, and the like. For example, see WO2009098450 or Heinis, et al., Nat Chem Biol 2009, 5 (7), 502-7.

Non-biological systems such as those set forth in WO2004/077062 are based on conventional chemical screening approaches. They are simple, but lack the power of biological systems since it is impossible, or at least impracticably onerous, to screen large libraries of peptide ligands. However, they are useful where, for instance, only a small number of peptide ligands needs to be screened. Screening by such individual assays, however, may be time-consuming and the number of unique molecules that can be tested for binding to a specific target generally does not exceed 10⁶ chemical entities. In contrast, biological screening or selection methods generally allow the sampling of a much larger number of different molecules. Thus biological methods can be used in application of the invention. In biological procedures, molecules are assayed in a single reaction vessel and the ones with favourable properties (i.e. binding) are physically separated from inactive molecules. Selection strategies are available that allow to generate and assay simultaneously more than 10¹³ individual compounds. Examples for powerful affinity selection techniques are phage display, ribosome display, mRNA display, yeast display, bacterial display or RNA DNA aptamer methods. These biological in vitro selection methods have in common that ligand repertoires are encoded by DNA or RNA. They allow the propagation and the identification of selected ligands by sequencing. Phage display technology has for example been used for the isolation of antibodies with very high binding affinities to virtually any target. When using a biological system, once a vector system is chosen and one or more nucleic acid sequences encoding polypeptides of interest are cloned into the library vector, one may generate diversity within the cloned molecules by undertaking mutagenesis prior to expression; alternatively, the encoded proteins may be expressed and selected before mutagenesis and additional rounds of selection are performed. Mutagenesis of nucleic acid sequences encoding structurally optimised polypeptides is carried out by standard molecular methods. Of particular use is the polymerase chain reaction, or PCR, (Mullis and Faloona (1987) Methods Enzymol., 155: 335, herein incorporated by reference). PCR, which uses multiple cycles of DNA replication catalysed by a thermostable, DNA-dependent DNA polymerase to amplify the target sequence of interest, is well known in the art. The construction of various antibody libraries has been discussed in Winter et al. (1994) Ann. Rev. Immunology 12, 433-55, and references cited therein.

Alternatively, given the short chain lengths of the polypeptides according to the invention, the variants are preferably synthesised de novo and inserted into suitable expression vectors. Peptide synthesis can be carried out by standard techniques known in the art, as described above. Automated peptide synthesisers are widely available, such as the Applied Biosystems ABI 433 (Applied Biosystems, Foster City, CA, USA)

(ii) Genetically encoded diversity

In one embodiment, the polypeptides of interest are genetically encoded. This offers the advantage of enhanced diversity together with ease of handling. An example of a genetically polypeptide library is a mRNA display library. Another example is a replicable genetic display package (rgdp) library such as a phage display library. In one embodiment, the polypeptides of interest are genetically encoded as a phage display library.

Thus, in one embodiment the complex of the invention comprises a replicable genetic display package (rgdp) such as a phage particle. In these embodiments, the nucleic acid can be comprised by the phage genome. In these embodiments, the polypeptide can be comprised by the phage coat.

In some embodiments, the invention may be used to produce a genetically encoded combinatorial library of polypeptides which are generated by translating a number of nucleic acids into corresponding polypeptides and linking molecules of said molecular scaffold to said polypeptides.

The genetically encoded combinatorial library of polypeptides may be generated by phage display, yeast display, ribosome display, bacterial display or mRNA display.

Techniques and methodology for performing phage display can be found in WO2009098450. In one embodiment, screening may be performed by contacting a library, set or group of polypeptide ligands with a target and isolating one or more member(s) that bind to said target.

In another embodiment, individual members of said library, set or group are contacted with a target in a screen and members of said library that bind to said target are identified.

In another embodiment, members of said library, set or group are simultaneously contacted with a target and members that bind to said target are selected.

The target(s) may be a peptide, a protein, a polysaccharide, a lipid, a DNA or a RNA.

The target may be a receptor, a receptor ligand, an enzyme, a hormone or a cytokine. The target may be a prokaryotic protein, a eukaryotic protein, or an archeal protein. More specifically the target ligand may be a mammalian protein or an insect protein or a bacterial protein or a fungal protein or a viral protein.

The target ligand may be an enzyme, such as a protease. It should be noted that the invention also embraces polypeptide ligands isolated from a screen according to the invention. In one embodiment the screening method(s) of the invention further comprise the step of: manufacturing a quantity of the polypeptide isolated as capable of binding to said targets. The invention also relates to peptide ligands having more than two loops. For example, tricyclic polypeptides joined to a molecular scaffold can be created by joining the N- and C- termini of a bicyclic polypeptide joined to a molecular scaffold according to the present invention. In this manner, the joined N and C termini create a third loop, making a tricyclic polypeptide. This embodiment need not be carried out on phage, but can be carried out on a polypeptide-molecular scaffold conjugate as described herein. Joining the N- and C- termini is a matter of routine peptide chemistry. In case any guidance is needed, the C-terminus may be activated and/or the N- and C- termini may be extended for example to add a cysteine to each end and then join them by disulphide bonding. Alternatively the joining may be accomplished by use of a linker region incorporated into the N/C termini. Alternatively the N and C termini may be joined by a conventional peptide bond. Alternatively any other suitable means for joining the N and C termini may be employed, for example N-C-cyclization could be done by standard techniques, for example as disclosed in Linde et al. Peptide Science 90, 671 -682 (2008) "Structure- activity relationship and metabolic stability studies of backbone cyclization and N- methylation of melanocortin peptides", or as in Hess et al. J. Med. Chem. 51 , 1026-1034 (2008) "backbone cyclic peptidomimetic melanocortin-4 receptor agonist as a novel orally administered drug lead for treating obesity". One advantage of such tricyclic molecules is the avoidance of proteolytic degradation of the free ends, in particular by exoprotease action. Another advantage of a tricyclic polypeptide of this nature is that the third loop may be utilised for generally applicable functions such as BSA binding, cell entry or transportation effects, tagging or any other such use. It will be noted that this third loop will not typically be available for selection (because it is not produced on the phage but only on the polypeptide-molecular scaffold conjugate) and so its use for other such biological functions still advantageously leaves both loops 1 and 2 for selection/creation of specificity.

(iii) Phage purification

Any suitable means for purification of the phage may be used. Standard techniques may be applied in the present invention. For example, phage may be purified by filtration or by precipitation such as PEG precipitation; phage particles may be produced and purified by polyethylene-glycol (PEG) precipitation as described previously. Details can be found in WO2009098450.

In case further guidance is needed, reference is made to Jespers et al (Protein Engineering Design and Selection 2004 17(10):709-713. Selection of optical biosensors from chemisynthetic antibody libraries.) In one embodiment phage may be purified as taught therein. The text of this publication is specifically incorporated herein by reference for the method of phage purification; in particular reference is made to the materials and methods section starting part way down the right-column at page 709 of Jespers et al.

Moreover, the phage may be purified as published by Marks et al J.Mol.Biol vol 222 pp581 -597, which is specifically incorporated herein by reference for the particular description of how the phage production/purification is carried out.

(iv) Reaction chemistry

The present invention makes use of chemical conditions for the modification of polypeptides which advantageously retain the function and integrity of the genetically encoded element of the product. Specifically, when the genetically encoded element is a polypeptide displayed on the surface of a phage encoding it, the chemistry advantageously does not compromise the biological integrity of the phage. In general, conditions are set out in WO2009098450.

(C) Use of polypeptide ligands according to the invention Polypeptide ligands selected according to the method of the present invention may be employed in in vivo therapeutic and prophylactic applications, in vitro and in vivo diagnostic applications, in vitro assay and reagent applications, and the like. Ligands having selected levels of specificity are useful in applications which involve testing in non- human animals, where cross-reactivity is desirable, or in diagnostic applications, where cross-reactivity with homologues or paralogues needs to be carefully controlled. In some applications, such as vaccine applications, the ability to elicit an immune response to predetermined ranges of antigens can be exploited to tailor a vaccine to specific diseases and pathogens.

Substantially pure peptide ligands of at least 90 to 95% homogeneity are preferred for administration to a mammal, and 98 to 99% or more homogeneity is most preferred for pharmaceutical uses, especially when the mammal is a human. Once purified, partially or to homogeneity as desired, the selected polypeptides may be used diagnostically or therapeutically (including extracorporeal ly) or in developing and performing assay procedures, immunofluorescent stainings and the like (Lefkovite and Pernis, (1979 and 1981 ) Immunological Methods, Volumes I and II, Academic Press, NY).

The peptide ligands of the present invention will typically find use in preventing, suppressing or treating inflammatory states, allergic hypersensitivity, cancer, bacterial or viral infection, and autoimmune disorders (which include, but are not limited to, Type I diabetes, multiple sclerosis, rheumatoid arthritis, systemic lupus erythematosus, Crohn's disease and myasthenia gravis).

In the instant application, the term "prevention" involves administration of the protective composition prior to the induction of the disease. "Suppression" refers to administration of the composition after an inductive event, but prior to the clinical appearance of the disease. "Treatment" involves administration of the protective composition after disease symptoms become manifest.

Animal model systems which can be used to screen the effectiveness of the peptide ligands in protecting against or treating the disease are available. The use of animal model systems is facilitated by the present invention, which allows the development of polypeptide ligands which can cross react with human and animal targets, to allow the use of animal models.

Methods for the testing of systemic lupus erythematosus (SLE) in susceptible mice are known in the art (Knight et al. (1978) J Exp. Med., 147: 1653; Reinersten et al. (1978) New Eng. J : Med., 299: 515). Myasthenia Gravis (MG) is tested in SJL/J female mice by inducing the disease with soluble AchR protein from another species (Lindstrom et al. (1988) Adv. lnzn7unol., 42: 233). Arthritis is induced in a susceptible strain of mice by injection of Type II collagen (Stuart et al. (1984) Ann. Rev. Immunol., 42: 233). A model by which adjuvant arthritis is induced in susceptible rats by injection of mycobacterial heat shock protein has been described (Van Eden et al. (1988) Nature, 331 : 171 ). Thyroiditis is induced in mice by administration of thyroglobulin as described (Maron et al. (1980) J. Exp. Med., 152: 1 1 15). Insulin dependent diabetes mellitus (IDDM) occurs naturally or can be induced in certain strains of mice such as those described by Kanasawa et al. (1984) Diabetologia, 27: 1 13. EAE in mouse and rat serves as a model for MS in human. In this model, the demyelinating disease is induced by administration of myelin basic protein (see Paterson (1986) Textbook of Immunopathology, Mischer et al., eds., Grune and Stratton, New York, pp. 179-213; McFarlin et al. (1973) Science, 179: 478: and Satoh et al. (1987) J ; Immunol., 138: 179). Generally, the present peptide ligands will be utilised in purified form together with pharmacologically appropriate carriers. Typically, these carriers include aqueous or alcoholic/aqueous solutions, emulsions or suspensions, any including saline and/or buffered media. Parenteral vehicles include sodium chloride solution, Ringer's dextrose, dextrose and sodium chloride and lactated Ringer's. Suitable physiologically-acceptable adjuvants, if necessary to keep a polypeptide complex in suspension, may be chosen from thickeners such as carboxymethylcellulose, polyvinylpyrrolidone, gelatin and alginates.

Intravenous vehicles include fluid and nutrient replenishers and electrolyte replenishers, such as those based on Ringer's dextrose. Preservatives and other additives, such as antimicrobials, antioxidants, chelating agents and inert gases, may also be present (Mack (1982) Remington's Pharmaceutical Sciences, 16th Edition).

The peptide ligands of the present invention may be used as separately administered compositions or in conjunction with other agents. These can include antibodies, antibody fragments and various immunotherapeutic drugs, such as cylcosporine, methotrexate, adriamycin or cisplatinum, and immunotoxins. Pharmaceutical compositions can include "cocktails" of various cytotoxic or other agents in conjunction with the selected antibodies, receptors or binding proteins thereof of the present invention, or even combinations of selected polypeptides according to the present invention having different specificities, such as polypeptides selected using different target ligands, whether or not they are pooled prior to administration.

The route of administration of pharmaceutical compositions according to the invention may be any of those commonly known to those of ordinary skill in the art. For therapy, including without limitation immunotherapy, the selected antibodies, receptors or binding proteins thereof of the invention can be administered to any patient in accordance with standard techniques. The administration can be by any appropriate mode, including parenterally, intravenously, intramuscularly, intraperitoneally, transdermally, via the pulmonary route, or also, appropriately, by direct infusion with a catheter. The dosage and frequency of administration will depend on the age, sex and condition of the patient, concurrent administration of other drugs, counterindications and other parameters to be taken into account by the clinician.

The peptide ligands of this invention can be lyophilised for storage and reconstituted in a suitable carrier prior to use. This technique has been shown to be effective and art-known lyophilisation and reconstitution techniques can be employed. It will be appreciated by those skilled in the art that lyophilisation and reconstitution can lead to varying degrees of activity loss and that use levels may have to be adjusted upward to compensate.

The compositions containing the present peptide ligands or a cocktail thereof can be administered for prophylactic and/or therapeutic treatments. In certain therapeutic applications, an adequate amount to accomplish at least partial inhibition, suppression, modulation, killing, or some other measurable parameter, of a population of selected cells is defined as a "therapeutically-effective dose". Amounts needed to achieve this dosage will depend upon the severity of the disease and the general state of the patient's own immune system, but generally range from 0.005 to 5.0 mg of selected peptide ligand per kilogram of body weight, with doses of 0.05 to 2.0 mg/kg/dose being more commonly used. For prophylactic applications, compositions containing the present peptide ligands or cocktails thereof may also be administered in similar or slightly lower dosages.

A composition containing a peptide ligand according to the present invention may be utilised in prophylactic and therapeutic settings to aid in the alteration, inactivation, killing or removal of a select target cell population in a mammal. In addition, the selected repertoires of polypeptides described herein may be used extracorporeal ly or in vitro selectively to kill, deplete or otherwise effectively remove a target cell population from a heterogeneous collection of cells. Blood from a mammal may be combined extracorporeally with the selected peptide ligands whereby the undesired cells are killed or otherwise removed from the blood for return to the mammal in accordance with standard techniques.

The invention is further described below with reference to the following examples. Examples METHODS

Synthesis and purification of bicyclic peptides linked to sarcosine linkers

Peptide synthesis was based on Fmoc chemistry, using a Symphony peptide synthesiser manufactured by Peptide Instruments. Standard Fmoc-amino acids were employed (Sigma, Merck), with the following side chain protecting groups: Arg(Pbf); Asn(Trt); Asp(OtBu); Cys(Trt); Glu(OtBu); Gln(Trt); His(Trt); Lys(Boc); Ser(tBu); Thr(tBu); Trp(Boc), Tyr(tBu) (Sigma). The coupling reagent was HCTU (Pepceuticals), diisopropylethylamine (DIPEA, Sigma) was employed as a base, and deprotection was achieved with 20% piperidine in DMF (AGTC). Syntheses were performed at 100 umole scale using 0.37 mmole/gr Fmoc-Rink amide AM resin (AGTC), Fmoc-amino acids were utilised at a fourfold excess, and base was at a four-fold excess with respect to the amino acids. Amino acids were dissolved at 0.2 M in DMF, HCTU at 0.4 M in DMF, and DIPEA at 1 .6 M in N- methylpyrrolidone (Alfa Aesar). Coupling times were generally 30 minutes, and deprotection times 2 x 2.5 minutes. After synthesis, the resin was washed with dichloromethane, and dried. Cleavage of side-chain protecting groups and from the support was effected using 10 mL of 95:2.5:2.5:2.5 v/v/v/w TFA H20/iPr3SiH/dithiothreitrol for 3 hours. Following cleavage, the spent resin was removed by filtration, and the filtrate was added to 35 mL of diethylether that had been cooled at -80 deg C. Peptide pellet was centrifuged, the etheric supernatant discarded, and the peptide pellet washed with cold ether two more times. Peptides were then resolubilised in 5-10 mL acetonitrile-water and lyophilised. A small sample was removed for analysis of purity of the crude product by mass spectrometry (MALDI-TOF, Voyager DE from Applied Biosystems). Following lyophilisation, peptide powders were taken up in 10 mL 6 M guanidinium hydrochloride in H20, supplemented with 0.5 mL of 1 M dithiothreitrol, and loaded onto a C8 Luna preparative HPLC column (Phenomenex). Solvents (H20, acetonitrile) were acidified with 0.1 % heptafluorobutyric acid. The gradient ranged from 30-70 % acetonitrile in 15 minutes, at a flowrate of 15/20 mL /min, using a Gilson preparative HPLC system. Fractions containing pure linear peptide material (as identified by MALDI) were combined, and modified with trisbromomethylbenzene (TBMB, Sigma). For this, linear peptide was diluted with H20 up to -35 mL, -500 uL of 100 mM TBMB in acetonitrile was added, and the reaction was initiated with 5 mL of 1 M NH4HC03 in H₂0. The reaction was allowed to proceed for -30 -60 min at RT, and lyophilised once the reaction had completed (judged by MALDI). Following lyophilisation, the modified peptide was purified as above, while replacing the Luna C8 with a Gemini C18 column (Phenomenex), and changing the acid to 0.1 % trifluoroacetic acid. Pure fractions containing the correct TMB-modified material were pooled, lyophilised and kept at -20 deg C for storage.

Conjugation of sarcosine onto the growing amino acid chain in Fmoc Solid Phase Peptide Synthesis is performed as with any other amino acids. In these examples, HCTU is again used as a coupling reagents. Coupling times with Fmoc-N-methylglycine (Fmoc-Sar-OH, Merck) are at 1 hr, with 20 minutes piperidine deprotection to ensure effienct deprotection and coupling on this more hindered amino acid. Synthesis of DOTA containing peptide and loading yttrium radioisotope

The peptide DOTA-GSar6-(06-34-18)-NH2 was synthesised in a single run as any other peptide, including longer coupling times for Sar (1 hr, as discussed above) and for the tBu-protected derivative of DOTA (DOTA(tBu3)-OH (supplier: TCI-UK), for 9 hrs). The sequence of 06-34-18 is an exemplary bicycle containing 5 residues in each loop. Purification and workup was performed as with any other bicyclic peptide. The synthetic procedure is also illustrated in Figure 1 1 . For loading of DOTA-conjugated Sar-linker adjoined bicycle with a non-radioactive version of yttrium (89Y), 2.5 mM peptide was incubated with 5 mM YCI3 in 50 mM ammonium acetate buffer (pH 5.5-6) at 98 deg C for 30 min. Quantitative charging of DOTA could be observed by MALDI-TOF MS. The protocol is derived from Sosabowski JK et al, Conjugation of DOTA-like chelating agents to peptides and radiolabeling with trivalent metallic isotopes. (2006) Nat. Protoc. 1 (2):972-6.

Example 1 : Sarcosine linkers enhance solubility

Polysarcosine is more soluble than polyglycine by many orders of magnitude, as exemplified in Figure 1 , which compares the solubility of 20mers of glycine and sarcosine.

In order to investigate the impact of a polysarcosine sequence on a polypeptide, a series of 8 peptides were synthesised, which are based on the Elastase Binder EIG12. The first four contained various lengths of N-terminal sarcosines (0, 2, 4, 6), and the second set of four had the N-terminal Gly derivatised with palmitoic acid (C15H31 COOH):

1 ) H-G-[EIG12]-NH2

2) H-G-Sar2-[EIG12]-NH2

3) H-G-Sar4-[EIG12]-NH2

4) H-G-Sar6-[EIG12]-NH2

5) PA-G-[EIG12]-NH2

6) PA-G-Sar2-[EIG12]-NH2

7) PA-G-Sar4-[EIG12]-NH2

8) PA-G-Sar6-[EIG12]-NH2 The sequence of [EIG12] is AC MTDAG C P L P I WCA, with the reactive cysteines which define the scaffold attachment points underlined. EIG12 is hydrophobic and poorly soluble in water. The N-terminal derivatised peptides were first purified in their linear form (non-TMB cyclised). Peptide powders were then added to 150 uL of H20, sonicated (5 min) to enhance solubilisation, and centrifuged (13k rpm, 5 min). Pellets were visible for peptides 1 ), 2), 5), 6), 7), 8), indicating that peptides 3) and 4) (with Sar4/6) were the most soluble. Concentrations of peptide in the supernatants (approximating that of saturated solutions) were estimated by the absorption at 280 nm, as summarised in Table 1 :

The results indicate that progressive N-terminal Sarcosine extensions dramatically increase solubility of the hydrophobic model peptide EIG12.

We next investigated the impact of the sarcosine linker length on the solubility of palmitoylated derivtives of EIG12 (peptides 5 to 8).

Concentrations of peptide in the supernatants (approximating that of saturated solutions) were estimated by the absorption at 280 nm, as summarised in Table 2.

Table 2 Peptide 5) Peptide 6) Peptide 7) Peptide 8) Peptide 5) Peptide 6) Peptide 7) Peptide 8)

Format linear linear linear linear linear linear linear linear

Solvent H20 H20 H20 H20 20% Acn 20% Acn 20% Acn 20% Acn in H20 in H20 in H20 in H20

Dilution 1:3 1:3 1:3 1:3 none none none none

OD280 0.107 0.057 0.05 1.25 0.087 0.12 0.255 1.87

As is evident from the results, palmitoylation decreases solubility of the peptides significantly. However, sarcosine spacers situated between the palmitoyl group and the peptide increase solubility up to 10-fold.

Since palmitoylation decreases solubility of an already insoluble peptide, we investigated its effects on a peptide which has a high inherent solubility. The following set of peptides is based on the Kallikrein binder 3B3. Its sequence is CRVNCFKAC, and highly water soluble when cyclised with TMB. The concentrations of saturated peptide solutions were estimated as before.

Ac-3B3(TMB) is remarkably soluble (>212 mg/mL). However, N-terminal Palmitoylation of 3B3 rendered this peptide in its linear state completely insoluble except in pure DMSO. A 100-fold dilution of such a solution into water precipitated all peptide material such that none remained in solution. In contrast, when the TMB-coupled, palmitoylated peptide is constructed with a Sar-3 linker between the palmitoyl group and the peptide, the material is highly soluble in water (>1 13 mg/mL). Thus, introduction of a Sar3 spacer between the Palmitoyl group and 3B3 powerfully enhances solubility.Thus, a hydrophilic linker such as Sar3 between prosthetic groups (i.e. palmitoyl) and peptide is desirable.

Example 2: Retention of biological activity of polypeptides comprising sarcosine linkers In order to demonstrate that N/C-terminal bicycle substitutions are possible without impacting the biological activity of a structured polypeptide, the Kallikrein binder 3B3 (CRVNCFKAC) was tested with Sar3, d-Lys3 and d-Asp3 spacers. d-Lys3 and d-Asp3 linkers were included to compare to the Sar3 linker, so as to gauge the comparative effects of solubilising ionised amino acids (Asp, Lys) that do contain functional groups. Substitutions were made on either one or both termini of the polypeptide, and in the presence and absence of palmitoylation.

The peptides were tested for Kallikrein inhibition in buffers containing albumin (palmitoyi binding) or lysozyme (no palmitoyi binding). The peptides tested were:

( H2N ) A- 3B3 (TMB) - ^■G (OH)

Ac- -3B3 (TMB) - (NH2 )

Ac-Sar3- -3B3 (TMB) - (NH2 )

PA-Sar3- -3B3 (TMB) - (NH2 )

PA-dAsp3- -3B3 (TMB) - (NH2 )

PA-dLys3- -3B3 (TMB) - (NH2 )

Ac- -3B3 (TMB) -Sar3- (NH2 )

Ac-Sar3- -3B3 (TMB) -Sar3- (NH2 )

PA-Sar3- -3B3 (TMB) -Sar3- (NH2 )

PA refers to palmitoylation, Ac refers to acetylation, (-NH2) refers to C-terminal amidation. The results are set forth in Table 3:

From Table 3, it is apparent that both N-and C-terminal Sar3 modification are tolerated, as affinity towards Kallikrein remains unchanged (see Ac-Sar3-3B3(TMB)-NH2 and Ac- 3B3(TMB)-Sar3-NH2). The addition of a palmitoyi group on Sar3 has little adverse effect on the potency of the bicycle, but activity is expectedly reduced by the presence of BSA (see PA-Sar3- 3B3(TMB)-NH2 in BSA buffer), since palmitoyi binds to BSA. We anticipate that the potency of the bicycle could be maintained upon increasing the length of the Sar spacer sufficiently.

Simultaneous N/C-terminal substitutions reduce activity approximately 10-fold. Example 3: Half-Life extension through palmitoylation

Palmitoylated model peptides bind to HSA with high affinity (Figure 2). Binding to HSA takes place through the palmitoyi group, as the control peptide lacking PA (D-Sar, black circles, Figure 2) does not interact. The presence of charged amino acids adjacent to the palmitoyi group reduces this affinity, while the greatest potency is achieved with a neighbouring glycine or tryptophan. Figure 2 shows the titration profiles and dissociation constants (Kd)_of palmitoylated model peptides to HSA.

As shown in Figures 3 and 4, the palmitoylated bicyclic peptide (derived from the Kallikrein binder PK15, PA-W-Ahx-PK15(TMB)-K(Fluorescein)) binds to HSA and to components of human serum with high affinities, while the acetylated control peptide does not, again indicating that albumin binding is achieved through the palmitoyi group. Correlating the dilution of human serum against actual known albumin concentration in serum gives a Kd (apparent) of approximately 210 nM, which is similar to the Kd obtained with purified albumin (140 nM).

In order to assess the impact of the type and length of linker on HSA binding by palmitoyi, we analysed binding to HSA by the three peptides shown in Figure 5, comparing a sarcosine-6 spacer with hydroxyproline-6 and -12 spacers. As shown in Figure 6, HSA binding by palmitoylated peptides appears independent on the type and length of molecular spacer (Sar6, HyP6, HyP12), as all Kd's reside within the range of 80 to 210 nM. The lower change in amplitude in respect of the HyP12 linker is due to the greater distance of the fluorophore from binding site on HSA.

In further experiments, it was shown that linkage of a palmitoyi group to the side chain Lys introduced C-terminally on a polypeptide gives slightly weaker albumin binding (~ 800 nM by anisotropy). ITC experiments confirm binding to HSA, but the affinities measured are generally ~10x weaker (potentially due to the different conditions employed, plus secondary HSA binding events that thwart quantitative interpretation; ).

Example 4: Half life extension of polypeptides: In vivo administration PA-GSar6- PK15(TMB) to rats

The following experiment was set up:

Rats were fasted overnight prior to dosing. Dosing was carried out at 5 mg/kg (-1.25 mg/rat), in PBS, ~ 250 μΙ_ injection volume. Peptide is highly soluble at >= 30 mg/mL.

Administration was made either intravenously or subcutaneously, to assess systemic clearance and tissue diffusion.

200 μΙ_ blood samples were taken at 0.08, 0.25, 0.5, 1 , 2, 4, 6, 8, 12, 24, and 28 hours post-dose, which were frozen & shipped for analysis Plasma proteins were precipitated and compounds extracted by the addition of three volumes of 50:50 acetonitrile:methanol

After centrifugation, supernatant was analysed by LC/MS/MS for PA-PK15, using a calibration standard curve for quantification.

Figures 7 and 8 illustrate the half-life of peptides administered intravenously and subcutaneously. Figure 9 illustrates a comparison of the parameters measured.

In the case of intravenous administration, palmitoylation increases the half-life of PK15 by a factor of 7, to -5.5 hrs.

Initial exposure concentrations are high, with distribution largely limited to extracellular fluids (low VoD). Clearance of the peptide is low, with less than 1 % of the liver blood flow being cleared In the case of subcutaneous administration, initial plasma concentrations are low, with a Tmax ~8hrs, suggesting low cell penetration (also indicated by low VoD). Bioavailability is ~ 80 %. The experiments also show that PK15 together with the PA-G-Sar6 extension is stable in biological fluids for >24 hrs, indicating that polysarcosines are sufficiently stable for use as linkers in therapeutic peptides.

Example 5: Use of Sarcosine linkers in the preparation of tandem structured polypeptides

In order to synthesise a tandem structured polypeptide, we elected to synthesise each part of the tandem polypeptide separately, and assemble the parts post-synthetically to form the tandem.

We selected click chemistry for post-synthetic assembly, where orthogonal chemical groups are introduced on to each polypeptide, which then are employed for conjugation to form the tandem. Polypeptide 1 is the Kallikrein binder 3B3, and polypeptide 2 is the Elastase Binder EI-D9. Both polypeptides have a Sar4 spacer between bicycle and click functionality.

The chemistry is illustrated in Figure 10. Click reactions are very specific, fairly efficient and work under mild conditions. A prerequisite is that both Alkyne and Azide must be compatible with peptide synthesis and cleavage conditions, i.e. strong acid, base, and reducing agent. The Alkyn group is introduced through Fmoc-(L-)-propargylglycine (PG). It can be incorporated anywhere in sequence of the polypeptide. Azide is introduced at the N-terminus through azidopentanoicacid (APA). The data (see also table 4) indicate that two polypeptides can be assembled into a single bispecific molecule, with no loss in potency and full retention of specificity.

Table 4

Peptides Ki against Ki against

Kallikrein(nM) Elastase(nM)

Peptide 1 : 3B3(TMB)-Sar4-PG 3.5 >10000 Peptide 2: APA-GSar3-EID9(TMB) >10000 84

Tandem 3B3(TMB)-Sar4-CLICK- 7 100

peptide: GSar3-EID9(TMB)

Example 6: Use of Sarcosine linkers in the preparation of DOTA-conjugated peptides

Antibodies have been successfully used to target radionuclides for both imaging and cancer therapy but for these purposes the long half life of the antibody scaffold can be problematic and for solid tumours the ability to penetrate into sold tumours is limited. The small size of the Bicycle scaffold provides a theoretical advantage over larger protein scaffolds in terms of tissue penetration and the tuneable short half life also present an advantage over antibodies for such purposes. A key enabler for the use of Bicycle for radionuclide targeting is the ability to couple chelating agents to the Bicycle molecule and then load radioisotopes onto the Bicycle conjugate. The chelation of metal ions requires incubation at high temperatures, thus for antibodies which cannot tolerate high temperatures the metal ions are chelated onto the chelating group, which is then conjugated as a unit onto orthogonal groups on the antibody.

Bicycles have the advantage of being very stable at high temperatures, thus the functional, chelator-containing bicycle can be directly loaded with a radionuclide, making manufacture of the radionuclide-containing bicycle at the clinic significantly simpler.

The synthesis of a radionuclide-containing bicycle has been exemplified using the potent chelating agent DOTA (1 ,4,7,10-tetraazacyclododecane-1 ,4,7,10-tetraacetic acid, which was used as the tBu-tri-protected derivative in Fmoc solid phase synthesis) and the sub- nanomolar kallikrein binder 06-34-18, which are joined together with the versatile sarcosine linker. Six sarcosines were used in this instance to separate the peptide from the metal chelating group.

As illustrated in Figure 1 1 , inclusion of DOTA on the Bicycle is part of the synthesis and requires standard coupling protocols only. Loading of ⁸⁹Yttrium, the isotopically stable counterpart to the medicinally relevant ⁹⁰Yttrium, can be quantitatively achieved by heating for 30 min at 90 deg C, a process that can be readily observed by MALDI-TOF mass spectrometry. Figure 12 illustrates the quantitative mass addition observed on the peptide due to complexation of Y³⁺ (B)

It should be noted that the Sar linker as well as the bicycle readily withstands the higher temperatures during ⁸⁹Y loading, as the full integrity of the peptide is preserved (as judged by MALDI-TOF, Figure 12).

Claims

Claims

1. Use of polysarcosine as a molecular linker.

2. Use according to claim 1 , wherein the linker is used to attach a polypeptide to a functional group.

3. A polypeptide ligand comprising a non-sarcosine sequence and a sarcosine polymer.

4. A polypeptide ligand according to claim 3 comprising a polypeptide comprising at least three reactive groups, separated by at least two loop sequences, and a molecular scaffold which forms covalent bonds with the reactive groups of the polypeptide such that at least two polypeptide loops are formed on the molecular scaffold, and further comprising an amino acid polymer comprising at least two sarcosine monomers.

5. A polypeptide ligand according to claim 3 or claim 4, wherein the sarcosine polymer is attached to the C-terminus of the polypeptide.

6. A polypeptide ligand according to claim 3 or claim 4, wherein the sarcosine polymer is attached to the N-terminus of the polypeptide.

7. A polypeptide ligand according to claim 3 or claim 4, wherein the sarcosine polymer is attached to an amino acid side-chain or to the amino acid backbone in the polypeptide.

8. A polypeptide ligand according to any one of claims 3 to 7, wherein the sarcosine polymer separates the structured polypeptide from a functional group.

9. A polypeptide ligand according to claim 8, wherein the functional group is a polypeptide or a ligand.

10. A polypeptide ligand according to claim 8, wherein the sarcosine linker separates the structured polypeptide from a long-chain fatty acid group.

11. A polypeptide ligand according to claim 10, wherein the long-chain fatty acid group is selected from a myristoyl, stearoyl, arachidoyl and palmitoyl group.

12. A polypeptide ligand according to claim 8 or 9, wherein the functional group is a metal chelator such as DOTA or any other metal-chelating molecular entity.

13. A polypeptide ligand according to claim 8, wherein the sarcosine linker separates the structured polypeptide from a second structured polypeptide.

14. Use according to claim 1 or claim 2, or a polypeptide ligand according to any one of claims 3 to 12, wherein the sarcosine polymer comprises between 2 and 40 sarcosine monomers.

15. Use or polypeptide ligand according to claim 14, wherein the sarcosine polymer comprises between 4 and 6 sarcosine monomers.

16. Use or polypeptide ligand according to any preceding claim, wherein the sarcosine polymer comprises one or more non-sarcosine amino acids.