WO1998017677A1 - Improvements in solid-phase synthesis of peptides and related compounds - Google Patents

Abstract

The invention relates to spacers for use in solid-phase synthesis of peptides and peptide-related compounds, in which a spacer construct is used to ensure that the target peptide or peptide-related compound is at a distance equivalent to at least 5-15 amino acids from the solid support. Constructs of the invention are particularly applicable to the synthesis of strongly hydrophobic peptide, which are otherwise difficult, if not impossible to purify and analyse. The constructs and methods of the invention are also applicable to synthesis of peptide nucleic acids.

Description

IMPROVEMENTS IN SOLID-PHASE SYNTHESIS OF PEPTIDES AND RELATED COMPOUNDS

This invention relates to spacers for use in synthesis of peptides, especially peptides made by solid- phase methods, and to solid-phase methods for synthesis of "difficult" or poorly soluble peptides. The spacers of the invention are applicable to the synthesis, manipulation and purification of "difficult" or poorly-soluble peptides, especially peptides containing large numbers of hydrophobic amino acids. These spacers are also applicable to synthesis of peptide-related compounds and compounds comprising peptide sequences, such as peptide nucleic acids .

BACKGROUND OF THE INVENTION

The synthesis of peptides of defined sequence is widely used in the investigation of the rules which determine the three-dimensional conformation of proteins, in production of antigens for eliciting the formation of antibodies, especially those directed against defined epitopes, and in the production of therapeutic and diagnostic agents. Interest in the use of synthetic peptides has greatly increased in recent years, as developments in recombinant DNA technology and gene cloning have enabled the identification of proteins such as antigens, cytokines and receptors. Once the active regions of such protein molecules are identified, synthetic peptides can be used for further investigation of their function, and for the production of agonists and antagonists which are useful as therapeutic and/or diagnostic agents .

Although a variety of solution-phase methods for peptide synthesis are known, the most widely-used methods of synthesis are those based on the solid-phase method of Merrifield, in which amino acids are added step-wise to a growing chain linked to an insoluble matrix. The solid- phase methods avoid the need to purify intermediate products, and since all reactions are carried out in a single vessel, losses due to repeated transfer of products are eliminated. The most commonly used supporting resins are based on polystyrene. It is necessary, m order to ensure that only a single amino group and a single carboxyl group are available for reaction, that all other potentially reactive groups are blocked. The most commonly used blocking groups are tert-butyloxycarbonyl (Boc) or 9-fluorenylmethoxycarbonyl (F oc) .

Peptides produced by solid phase peptide synthesis (SPPS) are normally cleaved off the resin using hydrogen fluoride (HF) , and then purified and analysed.

Despite the fact that SPPS has become a routine method in the art, there is no guarantee that a given peptide can be synthesised m useful yield, since the product even if obtainable m reasonable yield may be very heterogeneous, and difficult to purify; or that even if it is possible to synthesise it, that it will have sufficient solubility in normally-used solvents to permit purification and analysis.

For example, it is frequently found that addition of individual amino acids to the growing chain is very difficult. This does not depend on the synthetic chemistry being used, for example whether Fmoc or t-Boc is being used, nor is it determined by the side-chain of the particular ammo acid being added. It appears that difficulty m synthesis is mainly dependent on the ammo acid sequence. This must be empirically determined, and is not readily predictable. For example, some sequences which were thought on theoretical grounds to be likely to be easy to synthesise have in fact turned out to be difficult. The converse may also be found. Even if difficulty m synthesis can be predicted, it is usually difficult or impossible to overcome.

Although the "difficult sequence" phenomenon has been intensively investigated, no generally applicable explanation has been found. See for example Kent (1988); Milton et al (1990); Bedford et al (1992); and Krchnak et al (1993) .

The problems found during assembly of "difficult sequences" are thought to be primarily due to interchain aggregation of the pendant peptide chains, and are exaggerated by local steric factors (Kent, 1990) .

It is frequently observed that most difficulties in coupling during SPPS are experienced within the first 5-15 amino acid residues. Unfortunately, peptides of

5-15 amino acids are exactly the types of peptides which it is most frequently desired to synthesize.

Even if difficulties in linking amino acids to the growing chain are not experienced, there may be difficulties in solubilising the peptide once it is cleaved from the resin support. This is particularly likely if there is a high content of hydrophobic amino acids . As stated above, it is extremely important that the synthetic peptide be amenable to purification and to analysis, particularly if it is desired for use as a therapeutic or diagnostic agent.

Peptides containing multiple repeating alanine residues are often used to investigate difficulties in synthesis or solublization (see for example Bedford et al , 1992) . Polyalanine peptides synthesized using either Boc or Fmoc frequently show poor coupling yields and inhomogeneous final products (Merrifield et al , 1988; Beyermann et al , 1992) . An accompanying problem with polyalanine and other "difficult sequences" sequences is their poor solubility, which can make purification and analysis by HPLC and electrospray mass spectrometry (ESMS) difficult, if not impossible.

The most sensitive and efficient method of purifying a synthetic peptide product is the use of reversed-phase high performance liquid chromatography

(RP-HPLC) , using measurement of peaks by electrospray mass spectrometry (ESMS), or matrix-assisted laser desorption- time of flight mass spectrometry (MALDI-TOF MS). These methods require that the peptide product be soluble an appropriate solvent. Carrasco et al (1996) have proposed a method for routine monitoring of reactions and characterisation of products during SPPS, using MALD-TOF MS performed directly on the peptide chain while still coupled to the resin. In order to enable detection by MALDI- TOF MS, these authors utilised a peptide-lmker construct comprising a photocleavable linker, an lonisation tag which can be protonated m order to permit detection by

MALDI-TOF MS, and a chemically-cleavable linker. The photocleavable linker and the lonisation tag enable MALDI-TOF MS analysis of the molecules attached to the resin to be carried out, while the chemically-cleavable linker permits release only of the desired peptide products. A more detailed account of this work was published after the priority date of the present application (Carrasco et al , 1997) . There is no disclosure or suggestion whatsoever in either of these publications that the methods or constructs described therein could be used m order to enhance the yield of the synthesis, or to improve the solubility of the products, or could be modified m order to do so.

A further problem, which is of great importance m the preparation of combinatorial libraries of peptide sequences, is the failure to recover all possible variants m the library because some sequences are poorly soluble or poorly cleaved. It is contemplated that the constructs and methods of the present invention will facilitate handling of such libraries so as to enable purification and solubilization of peptides and avoid loss of library members .

We nave now found that inserting a spacer moiety, such as a spacer peptide, between the desired target peptide and the resin alleviates some of the coupling problems by placing the "difficult sequence" outside the 5- 15 residue region. The constructs and methods of the invention are also applicable to the synthesis of peptide- related compounds, such as peptide nucleic acids ( PNAs ) .

SUMMARY OF THE INVENTION In its most general aspect the invention provides a spacer construct for enhancing the efficiency of synthesis of peptides, or of peptide-related compounds, which is a) able to be linked to a solid support by a chemical linkage, b) linked to, or incorporates at another point, a selectively-cleavable chemical linkage means, and c) can be linked to a desired peptide or peptide-related compound to form a construct, whereby processes can be performed on the construct which cannot be performed on the desired peptide or peptide-related compound per se .

For example, m one preferred embodiment, the spacer construct of the invention enables the desired peptide or peptide-related compound to be solubilised, where the peptide or peptide-related compound alone would be insoluble or very difficult to solubilise. In this embodiment the spacer of the invention would be a chemical compound which possesses groups which are, or which can be made, strongly positively or negatively charged.

The spacer of the invention must be stable to the conditions used solid phase peptide synthesis, including the cleavage and purification steps. Thus to be useful for the purposes of this invention, l e . m order to serve as a means for modifying the properties of a peptide or peptide- related compound, the compound must be stable to hydrogen fluoride, to base, and to compounds such as trifluoroacetic acid .

According to a first preferred aspect, the invention provides a spacer peptide construct useful m synthesis of peptides or of peptide-related compounds, comprising a spacer peptide and a reversible linker group covalently bound to the N-termmal thereof, and optionally comprising a cleavable coupling group whereby to link the construct to a solid support; and further optionally comprising a solubilismg peptide group, covalently bonded to the spacer peptide either directly or via a reversible linker group.

Solubilismg peptide group-reversible linker- spacer peptide, which gives target peptide-reversible lmker-solubilis g peptide-reversible linker-spacer peptide.

The solubilismg peptide can be attached directly to the spacer peptide via an amide bond, to give

Solubilismg peptide-spacer peptide which can then eventually be used to give Target peptide-reversible lmker-solubilismg peptide-spacer peptide.

The spacer is preferably a soluble peptide, but can otherwise be any easily synthesized sequence of length suitable to ensure that the peptide or peptide-related compound which it is desired to synthesize (the target peptide or peptide-related compound) is at a distance equivalent to at least 5-15 ammo acids, preferably at least 9-10 ammo acids, from the solid support. Where the spacer is a peptide, this may be any easily synthesized peptide sequence of the required length, and may be tailored for specific target peptides m relation to length, hydrophobicity, identity of the ammo acids m the sequence, secondary structure-forming properties, oligomer formation, solubility, and incorporation of additional sequence moieties. The spacer peptide will desirably include cationic residues in order to facilitate ESMS or MALDI-TOF MS analysis. A particularly preferred sequence for use m the Boc method is (Gly-Arg) ₄-Gly . For use m the Fmoc method this sequence may cause difficulties, and so for this method either (Gly (Lys-Gly) _bGly or (Arg-Gly-Gly) iGly is preferred. The person skilled in the art will be able to optimise the length of spacer for each desired target peptide using routine trial and error methods. Because solid phase peptide synthesis methods are quick and simple, it is possible to try a series of different spacers and to select the one found to be best for the particular purpose within a few days; if facilities permit concurrent synthesis of a set of peptides, a single day may even suffice . The reversible linker group between the spacer and the target peptide or peptide-related compound should preferably be stable to hydrogen fluoride (HF) , which is the most commonly-used reagent for cleavage from the solid support, and labile to base. One suitable linker group is glycolamide ester, which is known to be compatible with most side-chain protecting groups used with Boc (Baleux et al , 1984, 1986) and with Fmoc. The glycolamide ester linkage has also been used m Fmoc SPPS (Mendre et al , 1992) . The glycolamide ester linkage is suitable for use with most groups used Boc SPPS for protection of reactive side-chains (Kent, 1988). However Trp (CHO) and His (Dnp) must be deprotected m solution after cleavage and purification of the target peptide-glycolamide ester- spacer peptide construct.

In a preferred process, the glycolamide ester is prepared by reacting the N-termmal amine of the spacer peptide with the anhydride of bro oacetic acid. The cesium salt of the desired C-termmal Boc-ammo acid is added, and nucleophilic displacement of the bromine by the carboxylate group of the ammo acid results formation of a glycolamide ester between the Boc-ammo acid and the spacer peptide .

However, this reaction presents one problem, that the Boc-am o acid cesium salt must be dry. If th s is not the case, hydrolysis of bromine from the support can result m formation of the active hydroxyl groups, giving entities of the form:

HO-CHo-CO-spacer peptide-chem cal linkage- (Resin) , as well as the desired product Boc-am o acid-glycolamide ester-spacer peptide- chemical linkage- (Resm) .

These hydroxyl groups can then react with Boc- ammo acids m subsequent stages of the process to give unwanted deletion peptides. In addition, it is not always easy to obtain dry Boc ammo acid cesium salts.

In order to overcome this problem, derivatives of 4-hydroxymethylbenzoιc acid (4-Hmb) may also be used as a selectively cleavable reversible chemical link for the processes of the invention. Esters of 4-Hmb have the advantage that they can be used m Fmoc SPPS synthesis as well as in Boc-based SPPS. In addition, 4-Hmb esters can be more conveniently formed than glycolamide esters. The 4-Hmb ester linkage is stable to trifluoroacetic acid and is labile to nucleophiles . Esters of 4-Hmb may be formed either by the cesium salt method, or by an anhydride method (Atherton et al , 1991) .

Alternative linker systems which may be used include photolabile, thiolytic or reductively cleaved linkers; allyloxycarbonyl linkers ("ALLOC"), which are cleaved in the presence of a palladium complex; linkers, such as phenacyl linkers, which are cleaved m the presence of a metal ion; or "safety-catch" linkers, whose properties are changed by a simple reaction. The person skilled m the art will be aware of suitable linker groups within each of these classes, and will be able to select an appropriate linker group for specific desired purposes.

Although the spacer peptide itself may be designed so as to confer solubilismg properties, for example by incorporating charged ammo acids, especially cationic ammo acids, an additional solubilismg peptide or other solubilizmg moiety may be used. In this aspect of the invention the solubilismg group is linked to the spacer peptide via an amide, using a linker group as described above,

Target peptide-reversible lmker-Solubilsmg group-CO-NH-spacer peptide,

Target peptide- [selectively cleavable linker] - solubilismg group- [linker] -spacer peptide,

where the spacer peptide is used to assist m the synthesis, and then cleaved at [linker] to give

Target peptide- [selectively cleavable linker] - solubilismg group;

this is then purified, and cleaved at [selectively cleavable linker] to give the target peptide free of the solubilismg group. Where a non-peptide solubilizmg moiety is to be used, the nature of the moiety can be selected depending on the nature of the compound which is being synthesized. For example, polyethylene glycol derivatives may assist m solubilismg an attached peptide m both aqueous or non-aqueous solvents. In this instance α-ammo, co-carboxyl polyethylene glycol derivatives may be useful, which constructs of the form:

Target peptide- [4-Hmb linkage] -CO-NH- (CH₂-CH₂-0) _n-CH₂-COOH,

where n = 1-10, may be synthesised. Extremely hydrophilic peptides, although not

"difficult" to synthesize m the sense described above, may be difficult to purify because they elute m or very close to the void volume of the column used for purification, resulting m poor or no separation. Addition of a hydrophobic "tail", analogous to the solubilizmg peptide described above, compensates for this problem. This tail may be a hydrophobic peptide, or may be a hydrophobic compound such as 16-hydroxydecanoιc acid or N-Boc- 6-ammocaproιc acid.

The construct may also optionally comprise a cleavable coupling group whereby to link the construct to the solid support. The coupling group is attached at the C-terminal end of the spacer peptide, or the solubilismg peptide if present . Any suitable coupling group known m the art for this purpose may be used, for example thiol, thioacid, hydrazme, hydrazide or aldehyde. According to a second aspect, the invention provides a peptide construct comprising a solubilismg peptide linked via a first, base-labile reversible linking group at its C-termmal to a desired target peptide or peptide-related compound, which m turn is linked via a second reversible linking group as defined for the first aspect of the invention to a spacer, and optionally a cleavable coupling group at the C-termmal of the spacer peptide. Thus this construct is of the form

solubilismg peptide- [base labile reversible linking group] -Target peptide- {second reversible linking group] -spacer peptide.

For this aspect of the invention the reversible linking group between the solubilismg peptide and the desired peptide or peptide-related group is different from the reversible linker used m the first aspect of the invention, that the linking group for this aspect must be base-labile only, whereas the linking group for the first aspect of the invention must be both HF stable and base-labile .

It is to be clearly understood that m addition to the specific spacer and solubilismg peptide sequences referred to above, the constructs of the invention may also comprise special peptide sequences designed for use in specific manipulations of the final desired peptide or peptide-related compound, for example His for use in immobilised metal affinity chromatography purification (Arnold, 1991), or sequences providing epitopes for use in monoclonal antibody affinity chromatography purification or in immunόassay, eg. FLAG™. For analysis of combinatorial libraries, a tag which facilitates mass spectroscopy , ESMS, MALDI-TOF MS and/or sequencing, or a tag for ELISA assay, may be incorporated. Transport sequences, such as those susceptible to intracellular esterases, may be incorporated to facilitate penetration into the target cell . For example, "delivery sequences" may be required in order to enable the peptide to cross the cell membrane and/or an intracellular membrane. These may include non-naturally occurring amino acids incorporating lipophilic side chains, or ester-peptide-linker delivery sequences, which can be designed so as to be cleavable in a time-dependent manner. Other such special-purpose sequences will be known to those skilled in the art.

According to a third aspect, the invention provides a method of solid-phase peptide synthesis comprising the steps of: a) coupling a spacer peptide to a solid support , b) adding a reversible linker group at the N-terminal end of the spacer peptide, c) synthesising the target peptide by sequential addition of amino acids, d) removing the product thus produced from the solid support, and e) removing the target peptide from the spacer peptide by cleavage of the reversible linker group.

Optionally a solubilismg peptide may be incorporated between the spacer peptide and the target peptide, or at the N-terminal end of the target peptide, via a base-labile reversible linker group. Preferably the solid support will be a resin, preferably a polystyrene resin. Particularly suitable supports include Boc-Gly-phenylacetamidomethyl (PAM) polystyrene resin and 4-methylbenzhydrylamme (MBHA) resin (Matsueda and Stewart, 1981) . Standard methods used in solid-phase peptide synthesis may be employed, for example as originally described by Merrifield (1986), or any of the many variations of this technique developed subsequently. For example, ammo acids may be protected with Boc or Fmoc, activation of ammo acids may be effected with dicyclohexylcarbodiimide or HBTU, and cleavage of the final ammo acid chain from the resin will usually be with hydrogen fluoride. Cleavage of the linker group may be effected by treatment with mild base, such as aqueous tπethylamme .

The target peptide, once separated from the spacer peptide and solubilismg peptide construct, may be solubilised m an appropriate solvent, such as water, buffer, trifluoroacetic acid, etc. The most suitable solvent for each individual peptide may be empirically determined. In some cases it may be necessary initially to use a solvent such as TFA, whereas the peptide construct may be soluble m water after initial lyophilisation from TFA.

The methods and constructs of the invention are also applicable to the synthesis of peptide nucleic acids ( PNAs ) , which are ammo acid derivatives which instead of the normal ammo acid side chains carry the four nucleotide bases of DNA. These compounds have a pseudopeptide backbone composed of N- (2-ammoethyl) glycme units, in which the nucleotide bases are attached to the nitrogen atom of glycme via carbonyl methylene linkers (reviewed m Neilsen and Haaima (1997). Oligomers of PNAs are able to bind strongly and with high sequence specificity to complementary oligomers of DNA, RNA or other PNAs, and m particular can bind to double-stranded DNA to form a PNA - DNA triplex m a strand displacement complex. They therefore show potential m gene therapy, as anti-sense DNA drugs, and as reagents for genome analysis (Nielsen and Haaima (1997) . PNAs have been synthesised by a process analogous to SPPS, using both Boc and Fmoc chemistries (Thomson et al , 1995; Hyrrup and Neilsen, 1996). For example, essentially standard SPPS protocols were used m the Boc solid phase synthesis of PNA oligomers, with final cleavage of the PNAs using liquid hydrogen fluoride

(Dueholm et al, 1994) . The spacer means, particularly spacer peptides, and other constructs and methods of the invention are applicable to the synthesis of PNAs.

It will be clearly understood that m addition to the peptide constructs of the invention, the invention includes within its scope the spacer means, preferably tne spacer peptide, and optional solubilizmg peptide construct linked either to a solid support or to a first ammo acid. Thus the invention contemplates a set of [construct] -ammo acid pairs, as well as SPPS kits comprising one or more constructs of the invention.

Brief Description of the Figures

Figure 1 shows a reaction scheme for synthesis of Ala₁₂- [glycolamide ester] - [Gly-Arg] ₄-Gly using Boc chemistry with in si tu neutralisation (Schnolzer et al , 1992). HPLC conditions: Vydac C₄ column, 4.6 x 150 mm; 5-85% B/40 mm; A=0.1% TFA; B=90% acetonitrile, 10% water, 0.1% TFA.

Figure 2 shows the results of HPLC chromatography of crude HF-cleaved Ala₁₂- [glycolamide ester] - [Gly-Arg] ₄-Gly on a Vydac C₄ column.

Figure 3 shows a reconstructed electrospray mass spectrum of crude HF-cleaved Ala_J2- [glycolamide ester] - [Gly-Arg] ₄-Gly . -X represents [glycolamide ester] - [Gly- Arg]₄-Gly.

Figure 4a shows the results of HPLC chromatography of crude cleaved H-(CP-10^{4? j5}) -glycolamide ester- (Gly-Arg-4) ₄-Gly-OH;

Figure 4b shows an ES-MS spectrum of H-(CP-10^{α2 r})-OH after base cleavage. The inset shows a reconstructed electrospray mass spectrum. Figure 5 is a flow chart showing the use of an ester of 4-hydroxymethylbenzoic acid (4-Hmb) as a selectively-cleavable chemical linkage in the synthesis of constructs of the form desired peptide- [4-Hmb ester] -spacer using Fmoc SPPS chemistry

Peptide A: desired peptide;

Peptide B: spacer peptide.

Figure 6 shows base cleavage of di-addition of the 4-Hmb linker to the spacer peptide-resin .

Peptide B: spacer peptide

Figure 7 shows a MALDI-TOF MS spectrum of a sample of Fmoc-Ala- [4-Hmb linker] -Gly (Lys-Gly) _b-amide cleaved after attachment of the Fmoc-Ala to the HO-4-Hmb- Gly (Lys-Gly) ₆-Resin using the anhydride method, after

1 M NOH/methanol/dioxane cleavage to remove di-addition of the 4-Hmb linker.

Figure 8 shows an HPLC chromatogram of a peptide construct made using the 4-Hmb linkage as the selectively- cleavable chemical linkage, using Fmoc chemistry.

Figure 9 shows a HPLC chromatogram of a second peptide construct made using the 4-Hmb linker and Fmoc chemistry .

Figure 10 shows the use of 16-hydroxyhexadecanoic acid as modifier of the physicochemical properties of target peptide.

Figure 11 shows the use of 6-aminocaproic acid as modifier of the physicochemical properties of target peptide . Figure 12 shows the use of a hydrophilic peptide as modifier of the physicochemical properties of target peptide .

Detailed Description of the Invention In order to provide a severe test of the effectiveness of the invention, we chose to synthesis the target peptide dodecaalanine (Ala₁₂), a notoriously difficult peptide. Peptides containing multiple repeating alanine residues have been widely used as a model of "difficult sequences", and, in addition, such sequences have very poor solubility. Alaι₂ as synthesised using conventional solid-phase methods is almost impossible to solubilise. Using a particularly preferred method of the invention, we were able to produce Ala₁₂ in soluble form, and to analyse the product using RP-HPLC. We were also able to synthesize chemotactic protein 10^42"5", a protein which has not previously been obtainable in high purity because of its tendency to form intermolecular aggregates.

It will be clearly understood that this merely represents one specific example of the use of the invention, and that similar methods, if necessary with routine optimisation for specific cases, may be used within the scope of this invention.

Example 1 Synthesis and Purification of Dodecaalanine We synthesised Alaι₂- [glycolamide ester] - [Gly- Arg]₄-Gly using Boc chemistry with in si tu neutralisation, basically as described by Schnόlzer et al (1992) . The reaction scheme used is summarised in Figure 1. Single couplings only were carried out for all amino acids.

Boc-Gly-PAM polystyrene resin (0.8 mmol/g; ABI, Foster City, CA; 1 mmol scale) was used to synthesise

Boc- (Gly-Arg) ₄-Gly-PAM-resin. Boc-amino acids (2 eq) were activated with HBTU and coupled for 10 minutes. The average coupling yield as measured by ninhydrin assay was 99.81%. Boc removal from 0.33 mmol of this resin followed by neutralisation, and reaction with 1 mmol bromoacetic anhydride gave N-bromoacetyl- (Gly-Arg) ₄-Gly-PAM-polystyrene resin. This was reacted with 1 mmol Boc-Ala, Boc-Ala cesium salt in 5 ml dimethylformamide (DMF) for 14 hours. Five Boc-Ala residues were then added (2 mmol, 6 eq) as above. A sample of the Boc-Ala,;- [glycolamide ester] - (Gly- Arg) ₄-Gly-PAM-polystyrene resin was subjected to removal of Boc, cleaved with HF, then the crude Ala_b- [glycolamide ester] - (Gly-Arg) 4-Gly was analysed by ESMS.

The ESMS examination of crude Ala₃-glycolamιde ester] - (Gly-Arg) ₄-Gly, cleaved at this intermediate stage of the synthesis, showed a single product of formula weight 1412.7 Da (calc. 1412.8 Da). The simple ESMS spectrum indicated .

1) undetectable deletion of Arg or Gly during the initial synthesis of (Gly-Arg) ₄-Gly-PAM-resm; 2) quantitative displacement of bromine by

Boc-Ala cesium salt; and

3) the absence of detectable alanme deletion peptides .

Cham assembly by SPPS was continued to give Ala₁₂- [glycolamide ester] - (Gly-Arg) ₄-Gly-PAM-polystyrene resm. The average alanme coupling yield was 99.69%. Boc removal followed by HF cleavage gave Ala_i2- [glycolamide ester] - (Gly-Arg) ₄-Gly, which, after HF removal m vacuo, was both precipitated and washed with ether, dissolved m 3 ml TFA, diluted with water and lyophilised. The cleavage yield was 81%. The crude peptide was analysed by HPLC and ESMS. Purification of 29.7 mg of crude peptide on a Vydac C₄ column, 25 x 250 mm (Hesperia, CA) yielded 11.3 mg of Ala_i2- [glycolamide ester] -Gly-Arg) ₄-Gly, which formed a clear solution m water. Treatment of 1.1 mg of this peptide m 300 μl water with 50 μl tπethylamme for 10 minutes gave a cloudy solution. The solution was acidified with 100 μl acetic acid and filtered for analysis by ESMS. The precipitate from another base cleavage experiment was collected by centrifugation, and was sparingly soluble m 90% acetonιtπle-10%water-0.1% TFA (HPLC solvent B) . The resulting suspension was filtered for analysis by ESMS.

The crude HF-cleaved, ether precipitated peptide Ala - [glycolamide ester] - (Gly-Arg) ₄-Gly was best solubilised in neat TFA; however after lyophilisation from TFA solution the peptide construct was soluble m water. Figure 2 shows an HPLC chromatogram of crude Ala ■> - [glycolamide ester] - (Gly-Arg) ₄-Gly ESMS analysis of the peaks indicated that the major product (peak 2) had a formula weight of 1838.1 Da (calc. for Ala ₂- [glycolamide ester] - (Gly-Arg) ₄-Gly; 1839.0 Da). A minor impurity (peak 1) was a Des-Ala peptide (1767 Da), while another (peak 3) was a low molecular weight impurity probably generated during HF cleavage. Figure 3 shows a reconstructed ESMS spectrum of the crude Ala_i2- [glycolamide ester- (Gly-Arg) -Gly, showing the expected product at 1838 Da and low levels of alanme deletion peptides The amide peptide construct was easily purified by HPLC, using the method described above, to give Ala _- [glycolamide ester] - (Gly-Arg) ₄-Gly .

Mild base treatment (aqueous triethylamme) of Ala_i2- [glycolamide ester] - (Gly-Arg) _-Gly gave a cloudy precipitate. ESMS analysis of the filtrate showed an ion at 986.7 Da which was due to HO-CH₂-CO- (Gly-Arg) ₄-Gly (calc. 986.5 Da), derived from base cleavage of the glycolamide ester linkage. Neither Ala-! -[glycolamide ester] - (Gly-Arg) ₄-Gly (M2H+ ion at 920.3 Da) nor Ala₁₂ (MH+ lon at 871.5 Da) was found m the filtrate. In another base cleavage experiment, ESMS analysis of the precipitate, which was sparingly soluble m HPLC solvent B, showed an Alai₂ ion at 871.6 Da (calc. 871.5 Da). A control synthesis of Ala₈-Gly-PAM-resm

(0.33 mmol scale) was carried out in parallel, with the same batch of Boc-Gly-PAM resm and the same SPPS protocols used above. Coupling yields as measured by nmhydrm assay were Ala_α average 99.96%, Ala^ 99.48%, Ala_fa 98.80%, Ala- 98.60%, Ala₈ 97.54%.

This control synthesis showed an onset of diffficult couplings beginning at alanme residue 6. Our strategy placed the first difficult Ala, usually the sixth (Bedford et al , 1992), 15-16 residues out from the peptide- resm. While it has been noted that difficult couplings in SPPS usually occur within the first 5-15 residues of a synthesis (Kent, 1988), the molecular mechanism for the dramatic improvement in the synthesis of dodecaalan e by spacing the peptide away from the resm is unclear.

Thus we have shown that a simple procedure based on linking the target peptide to the resm via a [glycolamide ester-spacer peptide] construct conferred significant benefits on the synthesis of the "difficult sequence" Al ι₂ •

1) lai2 was readily synthesised starting from a Boc-Ala-glycolamide ester- (Gly-Arg) ₄gly] construct. In contrast, in the control experiment an identical synthesis of Ala_b-Gly-PAM-resm showed poor couplings beginning at alanme residue six.

2) the choice of highly positively charged (Gly-Arg) ₄Gly as the spacer peptide meant that the Alai₂~ [glycolamide ester] - (Gly-Arg) ₄Gly was soluble m water, and thus the construct was amenable to characterisation by HPLC, ESMS or other standard methods.

3 ) Homogeneous Ala_i2 was obtained by purification of the Alaι₂- [glycolamide ester] - (Gly-Arg) ₄Gly construct followed by base cleavage at th glycolamide ester linkage. The Alaι₂ was separated from the highly soluble HO-CH₂-CO- (Gly-Arg) ₄Gly by filtration or centrifugation.

Example 2 Synthesis and Purification of Chemotactic Protein IP^42"5' (CP-10^42"55)

In a similar manner to that described m Example 1, H- (CP-10^42"55) -glycolamide ester- (Gly-Ala) ₄- Gly-OH, [H-Pro-Glu-Phe-Val-Gln-Asn-Ile-Asn-Glu-Asn-Leu-Phe- Arg-glycolamide ester- (Gly-Ala) ₄-Gly-OH] was synthesised. Purified H (CP-10^42-55) -OH (Lachmann et al , 1993) has not previously been obtainable m high purity owing to its tendency to form mtermolecular aggregates .

Figure 4a shows an HPLC chromatogram of crude cleaved H- (CP-10^{42 "}) -glycolamide ester- (Gly-Ala) _-Gly-OH . ESMS of the major peak from Figure 4a gave a mass of 2699 Da (calc. 2700 Da). HPLC-puπfled H-(CP-10⁴² )- glycolamide ester - (Gly-Ala) i-Gly-OH was also cleaved with base. H-Ala_ι2-glycolamide ester- (Gly-Arg) ₄-Gly-OH and H- (CP-10^42-55) -glycolamide ester- (Glyn-Ala) ₄-Gly-OH (2.0 mg in 900 μl water, a clear solution) were treated with 100 μl of triethylamine for 10 minutes (pH 12.2) . The resultant peptide precipitates were collected by centrifugation. After washing six times with 1 ml water the precipitates were shaken with a mixture of 20 μl TFA and 980 μl acetic acid. The suspension of peptide which resulted was filtered for analysis by ESMS. ESMS analysis of the precipitate, depicted in Figure 4b, showed the expected mass for H- (CP-10^42"55) -OH (calc. 1732.0 Da, found 1731.5 Da) .

Example 3 Formation of 4-Hmb Esters by the Cesium Salt Method

4-Hmb esters may be formed using the cesium salt method. For example, the anhydride of 4-bromomethyl benzoic acid can be reacted with the N-terminal amine of a spacer chemical linkage- (Resin) to give Br-CH₂- [C₆H₄] -CO-spacer chemical linkage- (Resin) .

Then the cesium salt of an Fmoc-amino aid (or Boc-amino acid) can be reacted to attach the first amino acid to the spacer via a 4-Hmb ester selectively-cleavable chemical linkage . We have found that there are some problems when this procedure is used with Fmoc amino acids. One is that some of the Fmoc groups are lost during the procedure, exposing amine groups. Although this has not appeared to be a problem up to now, the potential for side reactions between the newly exposed amines and unreacted bromomethyl groups does exist . The Boc group is not lost during ester formation via the cesium salt method.

Example 4 Formation of 4-Hmb Esters by the Anhydride Method

An alternative procedure for formation of 4-Hmb esters uses the anhydride method. The linker compound used for this procedure, 4-hydroxymethyl benzoic acid, can be obtained commercially, or can be easily made from cheap and readily available 4-bromomethylbenzoic acid (Atherton et al , 1981) . In our procedure, 4-hydroxymethyl benzoic acid

(1.1 eq) and HBTU ( 2- ( lH-Benzotriazole-1-yl ) -1 , 1 , 3 , 3- tetramethyluronium hexafluorophosphate) (1.1 eq) were added to a slurry of DMF-swollen NH₂-Peptide B-chemical linkage- (Resin) . DIEA (Diisopropyl-ethylamine) (1.3 eq) was then added, and the mixture stirred. After one hour the reaction was complete, as indicated by a) a negative result on ninhydrin analysis for unreacted amine groups, and b) MALDI-TOF MS analysis of the peptide derived from cleavage (TFA/water, 95:5) of a small portion of the resin.

No unreacted spacer was found. However, MALDI-TOF MS analysis of the cleaved 4-HO-CH₂- [C₆H₄] -CO- spacer showed that there was a small amount of di-addition of the linker, giving:

4-HO-CH₂- [C₆H₄] -CO-0-CH₂- [C₆H₄] -CO spacer in the solution, and therefore also on the resin.

The second linker would clearly be present at a level of 10% or less, as only 1.1 eq of linker were used in the reaction and it had been shown that no unreacted amine was present. However, to remove the second linker, the resin was washed with dioxane, and then treated for two minutes with a mixture of 1 M NaOH/MeOH/dioxane 1:4:15. The NaOH cleaves the base-labile ester in the di-4-Hmb addition product to give the single addition material, ie. 4-HO-CH₂-C₆H₄] -CO spacer chemical linkage- (Resin) . Figure 7 shows a MALDI spectrum of a sample of Fmoc-Ala- [4-Hmb linker] -Gly (Lys-Gly) _b-amine cleaved after attachment of the Fmoc-Ala to the HO-4-Hmb-Gly (Lys-Gly) ₀- Resin using the anhydride method, with NaOH/methanol/ dioxane cleavage to remove di-addition of the 4-Hmb linker. The only product seen on MALDI-TOF MS was the desired Fmoc- Ala- [4-Hmb linker] -Gly (Lys-Gly) _b-amide (found mass

1612.7 Da, calculated mass 1612.3 Da) . This resin was then used for further SPPS.

An alternative method for attaching the 4- hydroxymethylbenzoyl group which avoids this di-addition problem uses the 2 , 4 , 5-trichlorophenyl ester of 4- hydroxymethylbenzoic acid (Atherton et al , 1981).

After washing with dioxane and DMF the 4-HO-CH₂- [C_bH₄-CO-spacer chemical linkage- (Resin) was reacted with 4 eq of Fmoc-Alanine anhydride and 0.1 eq DMAP . After

30 minutes the resin was drained and washed with DMF. Mass analysis of a TFA/water ( 95 : 5 ) -cleaved sample of this resin showed that only Fmoc-Ala-4-Hmb selectively cleavable chemical linkage-spacer was present.

Example 5 Use of 4-Hydroxymethyl Benzoic Acid (4-Hmb) as the Selectively-Cleavable Linker A flow chart for SPPS by the Fmoc method, using 4-hydroxymethyl benzoic acid (4-Hmb) as the selectively- cleavable chemical linker, is shown in Figure 5. In this flow chart Peptide A represents the desired peptide, and Peptide B represents the spacer. 1.1 eq 4-Hmb, 1.1 eq HBTU, and 1.3 eq DIEA were reacted in DMF solvent at room temperature overnight . This reaction has been shown to be complete within one-two hours (Step 1 in flow chart) . There is some di-addition of the 4-hydroxymethylbenzoic acid to give the resin-bound product II, as illustrated in Figure 6. The resin was washed with dioxane, and then treated with 1 M NaOH, methanol, dioxane 1:4:15, to cleave the ester marked # and to give the desired resin-product I (Step 2) . Fmoc-Alanine, 2 mmole, was dissolved in 2 ml DMF. DCC (Dicyclohexylcarbodiime) 1 mmole, dissolved in

4 ml DCM, was added and the mixture stirred. After 20 minutes the insoluble precipitate of DCU (Dicyclohexylurea) was filtered off and the solution of

1 mmole Fmoc-Alanine anhydride was added to 0.25 mmole 4- Hmb lmker-Peptide B*-Resm I, as shown m Figure 6 A 0.1 M solution of DMAP ( 4-N, N-dimethylammopyridme , 0.1 eq relative to resm) m DMF was added as an acylation catalyst. After 30 minutes the reaction was drained and washed with DMF (Steps 3 and 4) A small portion of the rein was treated with trifluoroacetic acid (TFA) /water 95:5 to cleave the Fmoc-Ala- [4-Hmb linker] -Peptide B from the resm (Step 5) . The product was analysed by MALDI-TOF MS. In this case Fmoc solid phase peptide synthesis was carried out using an Applied Biosystems International (ABI) 433 peptide synthesiser with standard ABI synthesis cycles and chemistry. The peptide construct could be cleaved from the resm using standard cleavage reagents (Step 7) . For example, a mixture of [TFA 10 ml, thioanisole 0.5 ml, water 0.5 ml, ethanedithiol 0.25 ml, phenol 0.75 g] , was used as the cleavage reagent. The peptide construct was cleaved from the spacer by 0.1 M NaOH two minutes or less (Step 8) .

Example 6 Synthesis of a Peptide Using the 4-Hmb

Linkage Figure 8 shows the HPLC chromatogram of a peptide construct made using the 4-Hmb linkage as the selectively- cleavable chemical linkage. The construct was made using an ABI 433 peptide synthesiser with standard ABI Fmoc coupling chemistry.

HPLC conditions were: Vydac C4, 4.6 x 250 mm, 0-10 mm 0% B, 10-60 mm 0-50% B, at 1 ml/mm A=0.1% TFA; B=0.1% TFA, 10% water, 90% acetomtrile . The peptide sequence was:

Ala-Leu-Val-Ser-Leu-Trp-Thr-Ala-Lys-Asn-Pro-Gly- Ala-Ala- [4-Hmb linker] -Gly (LysGly) _b-amιde .

The mass of the major peak elutmg at 32 06 minutes was found to be 2699 Da, while the calculated mass was 2700 Da.

We encountered serious problems making the sequence (GlyArg) ι-Gly-amιde with Fmoc SPPS chemistry, and so used two alternative solubilismg sequences, Gly (LysGly) fa-amide, as shown above, or (ArgGlyGly) -amide (GlyArg) ₄-Gly-amιde was routinely used for Boc chemistry It is expected that better Fmoc SPPS chemistry will allow the synthesis of (GlyArg) ₄-Gly-amιde as a solubilismg peptide m future.

Example 7 Synthesis of a Second peptide Using the Hmp

Linkage Figure 9 shows a HPLC chromatogram of another construct

Acetyl-Cys-Glu-Trp-Asn-Ser-Ala-His-Phe-Ile-Ala- Tyr-Lys- [4-Hmb linker] -Gly (ArgGlyGly) ₃-amιde made using the 4-Hmb linker and Fmoc SPPS chemistry on an ABI 433 peptide synthesiser. The HPLC conditions were: Vydac C4, 4.6 x 250 mm, 0-60% B over 60 minutes at 1 ml/mm, A=0.1% TFA, B=0.1% TFA, 10% water, 90% acetonitπle .

The mass of the major peak elutmg at 31.381 minutes on the HPLC was 2513.1 Da, and the calculated mass was 2511.8 Da.

Example 8 Purification of Extremely Hydrophilic

Peptides The use of a hydrophobic tail is referred to above for the purification of extremely hydrophilic peptides .

Instead of a hydrophobic peptide 16-hydroxyhexa- decanoic acid (CAS number 506-13-8, Aldrich 1996-1997 catalogue p844) can alternatively be used. This hydrophobic compound is able to retard the retention time of an attached hydrophilic peptide on HPLC, and already includes in its structure the basis of a selectively cleavable chemical linkage, l e . the 16-hydroxyl group, which can form an ester linkage to the hydrophilic peptide This compound HO-(CH9)_: -COOH is attached to the resm support via its carboxyl group. Then an acid stable, base labile ester is formed between the hydroxyl group and the first ammo acid of the peptide to be synthesised. These steps are

HO-(CH>)ιr-COOH + Linkage- [Resm]

which gives :

HO- (CH₂) i -CO-Lmkage- [Resm] .

An ester between an N-protected ammo acid, using Boc- as the N-protectmg group, and the hydroxyl group is made using the anhydride method to give:

N-Boc-Ammo acιd-CO-0- (CH₂ ) 15-CO-Lmkage- [Resm] .

Solid phase peptide synthesis then gives :

Hydrophilic peptide-CO-O- (CH₂) ₁5-CO-Lmkage- [Resm] .

The construct is cleaved using acid, to give a construct of the form:

Hydrophilic peptide-CO-O- (CH₂) ₁5-COR (R=OH, NH₂, etc.)

The ester between the hydrophilic peptide and the -O- (CH₂) ₁5-COR portion is stable to acid used m cleavage from the resm, but labile to alkali. Thus once the construct has been purified by HPLC or some other method, the ester is cleaved by aqueous alkali to liberate the hydrophilic peptide.

Hydrophilic peptide-CO-O- (CH₂ ) i^-COR + aqueous alkali (eg. NaOH)

gives hydrophilic peptide-COO^"Na⁺ + HO- (CH₂ ) _l -COR.

In this case, the HO- (CH₂ ) 15-COR may be soluble in alkali if R is OH, for example. Acidification of the solution or addition of Ca²" ions would then precipitate the HO- (CH₂) 15-CO-OH, thus allowing recovery of the hydrophilic peptide.

Examp1e 9 Alternative Method For Purificaiton Of

Hydrophilic Peptides A second related compound which may be used is N-Boc-6-aminocaproic acid, ie. Boc-NH- (CH₂) -COOH . This is easily made from 6-aminocaproic acid (CAS number 60-32-2) . In this case the hydrophobic alkyl chain is also used to modify the solubility of an attached hydrophilic peptide.

This is incorporated onto the resin via the carboxyl group to give Boc-NH-CH₂ ) ₅-CO-Linkage- [Resin] . Then, the Boc group is removed and the 4-Hmb linker is added to give:

HO-CH₂- [C₆H₄] -CO-NH (CH₂ ) 5-CO-Linkage- [Resin] .

In a manner similar to that described above, a Boc- amino acid is attached to the hydroxyl group of the 4-Hmb linker by the anhydride method. A hydrophilic peptide is then synthesised to give a resin-bound construct of the form:

Hydrophilic peptide-CO-0-CH₂- [C₆H₄] -CONH0 (CH₂) ₅-CO- Linkage= [Resin]

Subsequent acid cleavage and HPLC purification gives :

Hydrophilic pep ide-CO-0-CH_:- [C₆H„] -CONH0 (CH_;) ,-CO-R, which is then cleaved at the 4-Hmb linkage by aqueous alkali to give the hydrophilic peptide and

HO-CH₂- [C₀H₄] -CO-NH(CH )₅-COR.

Example 10 Use of an Organic Compound as Modifier of the

Physicochemical Properties of an Attached Peptide This example shows the use of MHBA type peptide synthesis resm for Boc synthesis with 16-hydroxyhexa- decanoic acid as modifier of the properties of the peptide which gives C-termmal peptide amides on cleavage with hydrogen fluoride. The process is illustrated in Figure 10. The 16-hydroxyhexadecanoιc acid is coupled to the resm m HBTU and DIEA, as described in Example 4. Some di-addition of the 16-hydroxyhexadecanoιc acid may occur, as may also be observed with the addition of the 4-Hmb linker (see Example 4) . Should this occur, treating the resm with aqueous alkali m an organic solvent such as a mixture of 1 M NaOH/methanol/dioxane 1:4:15 results in cleavage of the second 16-hydroxyhexadecanoιc acid molecule .

In step 3, the Boc-ammo acid anhydride can be formed using standard methods (eg. 2 mmole Boc-ammo acid dissolved m 4 ml DCM, add 1 mmole dicyclohexylcarbodimide m 2 ml DCM, stir 20-30 mm, filter off the precipitate of dicyclohexylurea and add the solution to the resm) . Dimethylammopyπdme (DMAP) is added to catalyse the reaction of the hydroxyl group with the Boc-ammo acid anhydride .

The ester formed between the Boc-ammo and the 16-hydroxyhexadecanoιc acid is marked * (Step 4) . Boc SPPS is performed using standard methods (Step 5) . The construct is cleaved using strong acid eg. liquid hydrogen fluoride plus scavengers (Step 6) . The ester marked * is stable to these conditions but labile to aqueous alkali.

Example 11 Alternative Use of an Organic Compound as Modifier of the Physicochemical Properties of an Attached Peptide This example also uses the MBHA-type peptides synthesis resm and Boc chemistry, but m this case the 4-Hmb linker is used. The process is illustrated in Figures 11 and 12, which respectively show the use of

6-ammocaproιc acid and hydrophilic peptide to modify the properties of the target peptide.

The resm is treated with TFA to remove the Boc group prior to the addition of the 4-Hmb linker (Steps 1-3) . Some di-addition of the 4-Hmb linker may occur, as described m Example 4. Should this occur, treating the resm with aqueous alkali m organic solvents (eg. a mixture of 1 M NaOH/methanol/dioxane 1:4:15) results m cleavage of the second 4-Hmb molecule. Steps 4 and 5 are performed as Example 10.

The construct is cleaved using strong acid eg. liquid hydrogen fluoride plus scavengers (Step 6) . The 4-Hmb ester marked * is stable to these conditions but labile m aqueous alkali.

It will be apparent to the person skilled m the art that while the invention has been described m some detail for the purposes of clarity and understanding, various modifications and alterations to the embodiments and methods described herein may be made without departing from the scope of the inventive concept disclosed m this specification .

References cited herein are listed on the following pages, and are incorporated herein by this reference REFERENCES

Arnold, F.H.

Bio/Technology, 1991 9 151-157

Atherton, E. et al

J. Chem. Soc. Perkin Transactions I, 1981 538-546

Baleux, F.; Calas, B.; Mery. J Int. J. Peptide Protein Res., 1986 28 22-28

Baleux, F.; Daunis, J.; Jacquier, R. Tetrahedron Letts., 1984 25 5893-5896

Bedford, J.; Hyde, C; Johnson. T. ; Jun, W. ; Owen, D . ; Quibell, M. ; Sheppard, R.C. Int. J. Peptide Protein Res., 1992 40 300-307

Carrasco, M.R., Fitzgerald, M.C. and Kent, S.B.H. Proceedings of the European Peptide Symposium 1996

Carrasco, M.R., Fitzgerald, M.C, Oda, Y. and Kent S.B.H. Tetrahedron Letters, 1997 ^8 6331-6334

Hyrup, B. and Nielsen, P.E.

Bioorg. Biomed. Chem., 1996 4 5

Kent, S.B.H.

Ann. Rev. Biochem. , 1988 57 957-989

Krchnak, V.; Flegelova, Z . ; Vagner, J.

Int. J. Peptide Protein Res., 1993 42 450-454

Lachmann, M. Rajasekariah, P., Iismaa, S.E., Jones, G., Cornish, C.J., Hu, S., Simpson, R.J., Moritz, R.L., Geczy, C.L. J. Immunol., 1993 150 2981-2991 Matsueda, G .R. and Stewart, J.M. Peptides, 1981 2 45-50

Mendre, C . ; Sarrade, V.; Calas, B. Int. J. Peptide protein Res.; 1992 39 278-284

Merrifield, B.

J. Amer. Chem. Soc . , 1963 85 2145-2149

Merrifield, B.

Science, 1986 232 341-347

Milton, R.C.deL.; Milton, S.C.F; Adams, P.A.; J. Am. Chem. Soc, 1990 112 6039-6046

Nielsen, P.E. and Haaima, G. Chem. Soc. Reviews, 1997 73-78

Schnόlzer, M. ; Alewood, P.; Jones. A.; Alewood, D . ; Kent, S.B.H.

Int. J. Peptide Protein Res., 1992 40 180-193

Thomson, S.A., Josey, J.A., Cadilla, R., Gaul, M.D., Hassman, C.F., Luzzio, M.J., Pipe, A.J., Reed, K.L., Ricca, D.J., Wiethe, R.W. and Noble, S.A. Tetrahedron, 1995 51 6179-6194

Claims

CLAIMS :

1. A spacer construct for enhancing the efficiency of synthesis of peptides, or of peptide-related compounds, which is a) able to be linked to a solid support by a chemical linkage, b) linked to, or incorporates at another point, a selectively-cleavable chemical linkage means, and c) can be linked to a desired peptide or peptide-related compound to form a construct, whereby processes can be performed on the construct which cannot be performed on the desired peptide or peptide-related compound per se .

2 A spacer peptide construct according to Claim 1, comprising a spacer peptide and a reversible linker group covalently bound to the N-termmal thereof, and optionally comprising a cleavable coupling group whereby to link the construct to a solid support; and further optionally comprising a solubilismg peptide group, covalently bonded to the spacer peptide either directly or via a reversible linker group.

3. A spacer construct according to Claim 1 or Claim 2 of size equivalent to a peptide sequence of at least 5 to 15 ammo acids.

4. A spacer according to Claim 3 of size equivalent to a peptide sequence of at least 9 to 10 ammo acids.

5. A spacer construct according to any one of the preceding claims, m which the spacer is a soluble peptide.

6. A spacer construct according to Claim 6 m which the peptide is (Gly-Arg) ₄-Gly .

7. A spacer construct according to Claim 6, wherein the spacer peptide is Gly- (Lys-Gly) ₆-Gly .

8. A spacer construct according to Claim 6, m which the peptide is (Arg-Gly-Gly) ₃-Gly .

9 A spacer construct according to any one of the preceding claims further comprising an additional solubilismg group linked to the spacer via an amide, using a reversible linker group.

10. A spacer construct according to Claim 9, in which the solubilismg group is a peptide.

11. A spacer construct according to any one of the preceding claims, further comprising a cleavable coupling group whereby to link the construct to the solid support.

12. A spacer construct according to any one of

Claims 1 to 8, in which the desired peptide or peptide- related compound is strongly hydrophilic, and in which the spacer further comprises a hydrophobic tail .

13. A spacer construct according to Claim 12, in which the hydrophobic tail is 16-hydroxyhexadecanoic acid,

6-aminocaproic acid, or a hydrophobic peptide.

14. A spacer construct according to any one of the preceding claims, in which the peptide-related compound is a peptide nucleic acid.

15. A construct according to any one of the preceding claims, in which the reversible linker group is glycolamide ester.

16. A construct according to any one of Claims 1 to 15, in which the reversible linker group is a 4- hydroxymethylbenzoic acid ester.

17. A peptide construct comprising a solubilismg peptide linked by a first, base-labile reversible linking group at its C-terminal to a desired target peptide or a peptide-related compound, which in turn is linked by a second reversible linking group to a spacer, and optionally further comprising a cleavable coupling group at the C- terminal of the spacer, in which the first reversible linking group is labile to hydrogen fluoride.

18. A peptide construct according to Claim 17, in which the spacer is a construct according to any one of Claims 1 to 16, further comprising a special peptide sequence designed for use in specific manipulations of the final desired peptide or peptide-related compound.

19. A set of [spacer construct ] -amino acid pairs, for use in solid-phase synthesis of peptides or peptide-related compounds, each comprising a spacer construct according to any one of the preceding claims linked to a first amino acid.

20. A support for solid-phase synthesis of peptides or peptide-related compounds, comprising a resin linked to a spacer construct according to any one of Claims 1 to 18.

21. A kit for solid-phase synthesis of peptide or peptide-related compounds, comprising one or more spacer constructs according to any one of Claims 1 to 18.

22. A method of solid-phase peptide synthesis comprising the steps of: a) coupling a spacer peptide to a solid support, b) adding a reversible linker group at the N-terminal end of the spacer peptide, c) synthesising the target peptide by sequential addition of amino acids, d) removing the product thus produced from the solid support, and e) removing the target peptide from the spacer peptide by cleavage of the reversible linker group.

23. A method according to Claim 22, in which a peptide is incorporated between the spacer peptide and the target peptide, or at the N-terminal end of the target peptide, via a base-labile reversible linker group.

24. A method according to Claim 22 to Claim 23, in which the solid support is a resin.

25. A method according to Claim 24, in which the resin is a polystyrene resin.

26. A method according to Claim 25, in which the polystyrene is Boc-Gly-phenylacetamidomethyl (PAM) polystyrene resin or 4-methylbenzhydrylamine 9 (MBHA) resin.

27. A method according to any one of Claims 22 to 26, in which the spacer peptide is (Gly-Arg) ₄-Gly .

28. A method according to any one of Claims 22 to 26, in which the spacer peptide is Gly- (Lys-Gly) ₆-Gly .

29. A method according to any one of Claims 22 to 26, in which the spacer peptide is (Arg-Gly-Gly) ₃-Gly .