GB1587809A - Synthesis of peptides - Google Patents

Description

(54) IMPROVEMENTS IN OR RELATING TO THE SYNTHESIS OF MODIFIED PEPTIDES (71) We, NATIONAL RESEARCH DEVELOPMENT CORPORATION, a British Corporation established by Statute, of Kingsgate House, 66-74 Victoria Street, London, S.W.1, do hereby declare the invention for which we pray that a patent may be granted to us, and the method by which it is to be performed, to be particularly described in and by the following statement:- This invention relates to modified peptides.

Modification of one or more of the amino acid linkages of a physiologically active peptide through replacement with an isosteric group can lead to an enhancement of the properties of the peptide. Thus, for example, it may be possible to increase the stability of the peptide in vivo without detracting to an unacceptable extent from its physiological action.

In U.K. Patent Application No. 13192/76 (Serial No. 1,585,061) novel dipeptide analogues are described which are of use in the synthesis of isosterically modified peptides. The present invention relates to further dipeptide analogues which are of use as intermediates for the synthesis of the analogues of the earlier application and which are additionally of interest, in their own right, for the synthesis of isosterically modified peptides.

Accordingly the present invention comprises a compound being an analogue of a dipeptide formed between one of the amino acids alanine (i.e. a-alanine), A- alanine, arginine, asparagine, aspartic acid, 3,5-dibromotyrosine, cysteine, cystine, dopa, N-formylglycine, glycine, glutamic acid, glutamine, histidine, hydroxylsine, hydroxyproline, 3,5-diiodotyrosine, isoleucine, leucine, lysine, methionine, ornithine, phenylalanine, proline, pyroglutamic acid, hydroxyproline, serine, spinacin, threonine, thyroxine, tryptophan, tyrosine and valine, and other of said amino acids or between two molecules of the same amino acid, the linking amide group of the dipeptide being modified in the analogue through replacement of the nitrogen atom by the trivalent group -CH, or when the C-terminal amino acid residue of the dipeptide is derived from other than hydroxyproline, proline pyroglumatic acid and spinacin alternatively by the trivalent group -C-(-lower alkyl), or a derivative of such dipeptide analog formed at the terminal amino and/or carboxyl group thereof and/or at other functional group(s) therein.

It will be appreciated that the terms amino acid and amino group as used herein in a general sense include an imino acid and an imino group. The term lower alkyl is used in this specification to indicate a C, to C4 alkyl group.

Dipeptide analogues according to the present invention include various compounds of formula R'R2N . CH(R4) COCHR . CH(RS) . COR3 wherein R is a lower alkyl group or a bond to the group -CH(R5)- forming a cyclic group

or is particularly hydrogen, NRrR2 is an amino group, an imino group bonded to the group -CH(R4)- forming a cyclic group

or a derivative of such amino or imino group, -COR3 is a carboxyl group or a derivative thereof, R4 is hydrogen or an organic group being the a-side of an aamino or a-imino acid as described above which is bonded to the group R' R2N- in the case of the imino acid in which the nitrogen atom of the imino group is part of a ring system (proline, hydroxyproline, pyroglutamic acid and spinacin), or a derivative of such side chain formed at functional group(s) therein, and R5 is hydrogen or an organic group being the a-side chain of an a-amino or a-imino acid as described above which is bound to the group -CHR- in the case of an a-imino acid in which the nitrogen atom of the imino group is part of a ring system, or a derivative of such side chain formed at functional group(s) therein, the cases where R5 is hydrogen or a monovalent organic group being particularly preferred, although it is preferred that R4 and R5 are not both hydrogen. It will be appreciated that, additionally to the possibilities discussed above, when the groups -CH(R4)- and/or -CH(R5)- derive from p-alanine, then R4 and/or R5 will be -CH2-.

Compounds of particular interest are analogues of those dipeptides which derive from a-amino acids and particularly from the naturally occurring amino acids. However, the dipeptides may be derived not only from the L-isomers of naturally occurring acids but from the L- and D- isomers of both these and the other acids which are not naturally occurring. It will be appreciated that among the amino acids listed above a wide variety of structures is found, the groups -CH(R4t and Cll(R5H being, for example, a straight-chain, branched or (with an adjacent group) cyclic group which may carry carboxyl, amino, imino, amido, hydroxyl, sulphydryl, phenyl, indolyl and imidazolyl substituents.

Examples of specific types of compounds according to the present invention are firstly those compounds in which the central grouping -COCHR- is simply a group -COCH2-, R being hydrogen, Dipeptides of which the compounds of the present invention are analogues are derived from the above listed amino acids by combination thereof in pairs of the same amino acid or in pairs of different amino acids in either order, and the greater number of such dipeptides contain a linking amide group of the form -CONH-. Replacement of the nitrogen atom of such a group by -CH yields a group -COCH2. A smaller number of such dipeptides contain a Cterminal residue derived from an a-imino acid in which the nitrogen atom of the imino group is part of a ring system, i.e. from hydroxyproline, pyroglutamic acid and spinacin, and in the case of such dipeptides the linking amide group is of the form -CON Replacement of the nitrogen atom of such a group by -CH yields the second type of compound containing a central grouping -COCH with the -CH- group taking the place of the nitrogen atom in the ring system of the amino acid. A third type of compound contains a grouping -COCHR- in which R is a lower alkyl group and particularly methyl, and in such circumstances the dipeptide of which the compound is an analogue may alternatively be regarded as containing a C-terminal amino acid in which the amino group is substituted by a monovalent organic group, for example a derivative of glutamic acid or glutamine in which the amino group carries a lower alkyl substituent.

As indicated above, compounds according to the present invention may have their terminal amino and/or carboxyl groups and/or other functional groups present in the form of derivatives. Such derivatives may take a variety of forms. One group corresponds to analogues, for example of the specific dipeptides derived from the amino acids listed as described above, in which one or more of the functional groups is protected in one of the conventional manners used in peptide synthesis as discussed hereinafter. Another group of compounds has a terminal amino group which carries two or especially one monovalent organic groups such as lower alkyl groups and particularly methyl, or acyl groups and particularly formyl, and in such circumstances the dipeptide of which the compound is an analogue may alternatively be regarded as containing an N-terminal amino acid in which the amino group is substituted by the organic groups in question, for example Nformylglycine, or a derivative of N-formylglycine, glutamic acid, glutamine or pyroglutamic acid in which the amino group carries a lower alkyl substituent.

Specific compounds which may be mentioned are those analogues of the following dipeptides formed by replacement of the central -CONH-, or -CON groups thereof by the group -COCH2- or COCH respectively (particularly when the individual amino acid residues where enantiomorphic are of the L-configuration): arginyl-proline, histidyl-tryptophan, pyroglutamyl-histidine, seryl-tyrosine, trypthophanyl-serine, tyrosyl-alanine (the L-tyrosyl-D-alanine also being of some particular interest in this instance) and particularly leucyl-arginine, glycyl-leucine, but also phenylalanyl-glycine, prolyl-glycine and tyrosyl-glycine, and functional derivatives thereof.

The dipeptide analogues according to the present invention are most conveniently prepared by the attachment of suitable precursors for the C-terminal and N-terminal portions of the molecule, rather than by any attempt to modify the dipeptide as such.

As regards their use as intermediates in the synthesis of the dipeptide analogues of the earlier application, the analogues of the present invention which are of particular interest are those in which the C-terminal moiety derives from an amino acid other than glycine since the synthesis of analogues in which the moiety does derive from glycine may also be readily approached by a different route. Two alternative methods for the synthesis of the analogues of the present invention, and thence the analogues of the earlier application, are shown below: Method 1

Zinc ANR'--CH R"--COCH,B r ANR'-CHR"-COCH,-ZnBr+ Z-CHR"'-COOB ANR'-CHR'-COCH2CH R"'-COOB SN2 Reaction Reduction ANR'CHR"CH2CH2CHRD'COOB Method 2

1. PPh3 ANRWCHR"COCH2Br ) ANR'--CH R"--COCH=PPh, ANR'CHRtvwl l2 2 Mild base

Cleavage ------t R"-COCH2-CH R"'-COOB of C=P-bond Reduction ANR'-CHR"-CH3CH2-CHR"'-COOB.

wherein R' represents hydrogen or a bond to R", R" represent hydrogen, a monovalent organic group or a divalent organic group joined to AN-, R"' represents hydrogen or a monovalent organic group, A and B represent suitable protecting groups, and Z represents a suitable leaving group in the SN2 reaction.

The synthetic methods described above are suited to the retention of configuration at one or both of the carbon atoms to which the amino groups of the amino acids are joined, resolution being employed if configuration is lost at one centre. In the first method a haloketone is treated with a suitable metal, for example zinc, in an aprotic, highly polar solvent, for example hexamethyl phosphoramide or particularly dimethyl sulphoxide in benzene, to thereby lead to the selective formation of the appropriate enolate. The resulting carbon ion is reacted by an SN2 pathway with the reactant Z-CHR"'-COOB which is obtained without racemisation from the corresponding amino acid, suitable leaving groups Z being, for example, the tosylate (p-toluene sulphonate) and bromide anions.

Reduction of the keto group may then be effected, for example as described below.

In the second method a haloketone is treated with triphenylphosphine under appropriate conditions to form the phosphonium halide salt which on treatment with a mild base, for example Na2CO3/H2O/C2HsOCOCH3, yields a ylide. Reaction with the reactant ZCHR"'-COOB, as described above, effects introduction of the group --CHRn''-COOB at the carbon atom of the C=P - bond of the ylide and this is followed by cleavage of this bond. The presence of the adjacent keto group stabilises the bond, however, and for this reason the use of electrolysis is preferred to effect the cleavage by reduction rather than a hydrolytic method, the cleavage of ylides by such electrolytic methods being well known in the art. Reduction of the keto group to afford the second type of analogue may be effected by various procedures known in the art. Procedures such as reaction with HSCH2CH2SH followed by treatment with Raney nickel can be complicated by the effect of the Lewis acid used in the dithiol reaction on some of the more common N-protecting groups such as t-butoxycarbonyl. Accordingly, it is preferably in order to avoid the necessity for the use of other forms of protecting group, to use a stepwise reduction procedure involving conversion of the grouping -COOB- to -COOH, reduction of the keto group to an alcohol, activation of the alcohol group, for example as a mesylate, tosylate etc., followed finally by the selective reduction of this activated group with a hydride reagent such as lithium aluminium hydride, these all being procedures well known in the art.

Suitable N-protecting groups A include various urethane N-protecting groups, particularly groups such as t-butoxycarbonyl etc. Suitable C-protecting groups B include various alkyl groups, particularly lower (C1 to C4) alkyl groups such as methyl and, indeed, it may even be possible in certain instances to simplify the procedure by using a free carboxy group. Protection of other reactive functional groups in R" and R"' by conventional procedures may also be necessary. The ahaloketones are readily obtainable from the corresponding diazoketones, for example by treatment with HBr at -100C. the various reactions involved in methods 1 and 2 all involve standard procedures and variations upon them, for example variations to make analogues containing a group -CH2CHR-, will be apparent to those skilled in the art. Indeed, it will be appreciated that, in general, the methods of synthesis are described herein are not the only ones which may be used for the preparation of the dipeptide analogues according to the present invention and that obvious chemical equivalents of these methods or other methods known in the art for effecting similar reactions and obvious chemical equivalents thereof may be employed.

The compounds having free amino and carboxy groups may be obtained by removal of the N-protecting group and also of the C-protecting group, where applicable, which have been used in their synthesis. Where the dipeptide analogues of the present invention are required as intermediates for the preparation of those of the earlier application such removal is usually best left until the end of the synthesis as shown above. However, where the dipeptide analogues of the present application are required for the synthesis of larger modified peptides, then the synthesis will be interrupted at the penultimate stage shown and the removal of protecting groups then carried out, where appropriate.

As indicated, the dipeptide analogues according to the present application are of interest not only for the synthesis of analogues according to the earlier application but also for the synthesis of larger modified peptides by the addition of further amino acids or peptides thereto by exactly analogous procedures to those described in the literature of peptide chemistry for the addition of amino acids and peptides to dipeptides or amino acids. It will be appreciated, furthermore, that such larger modified peptides may contain more than one amide bond modified as described herein. In many instances the dipeptide analogue used in such syntheses will be modified somewhat from that corresponding simply to the compound produced by the replacement of the amide link in a dipeptide derived from two amino a acids by a group -COCHR-. Such modifications may already be incorporated into the dipeptide analogue produced by an appropriate method of synthesis as indicated above, or alternatively the analogue may be modified appropriately before use. A first type of modification is necessary where the amino acids from which the analogue derives contain reactive groups which are normally protected during peptide syntheses. These groups will usually require to be protected in the dipeptide analogue and this protection may conveniently be effected by similar procedures to those described in the art of peptide chemistry.

Examples of such groups in the commonly occurring amino acids are the E-amino group of lysine and the sulphydryl group of cysteine, the p-carboxy group of aspartic acid and the y-carboxy group of glutamic acid, the hydroxy group of serine, threonine and tyrosine, and less commonly the amide group of asparagine and glutamine, the sulphide group of methionine and the imino group of tryptophan. Secondly, it will often be necessary to modify the amino or imino group, or the carboxy group of the dipeptide analogue. This necessity will depend on the type of reaction in which the analogue is being used. Thus when an amino acid or peptide is being added to the N-terminus of the analogue it will be usual to protect the terminal carboxy group thereof when the C-terminus of the added amino acid or peptide is reacted as a free carboxy group using a suitable condensing agent such as dicyclohexylcarbodiimide. When the C-terminus of the added amino acid or peptide is reacted as an activated functional derivative thereof, then the terminal carboxy group of the dipeptide analogue may either be protected or may be present in zwitterion form or as a carboxylate salt. On the other hand, when an amino acid or peptide is being added to the C-terminus of the analogue it will be usual to protect the terminal amino or imino group thereof and, furthermore, it may also be appropriate to modify the carboxy group to activate it as described above as an alternative to using the free carboxy group with a suitable condensing agent.

Examples of suitable activating functional derivatives of the C-terminal carboxy group of the dipeptide analogue are active esters, mixed anhydrides, for example from isobutyl chlorocarbonate, the symmetrical anhydride, the azide and also masked azides such as derivatives containing the groupNHNHS wherein S represents a t-butoxycarbonyl, benzyloxycarbonyl group, etc. Of these derivatives the active esters are of particular interest, for example the esters with pnitrophenol, 2,4-dinitrophenol, 2,4,5-trichlorophenol, 2,3,4,5,6-pentachlorphenol, N-hydroxyphthalimide, N-hydroxysuccinimide, 5-chloro-8-hydroxyquinoline etc.

Examples of suitable protecting groups of the C-terminal carboxy group of the dipeptide analogues are groups removable both by hydrolytic and catalytic cleavage, particularly esters and amides, including N-substituted amides, for example various substituted benzyl esters and alkyl esters (for example C1 to C4) including the t-butyl ester. It will be appreciated that the carboxy group may also be protected through attachment to a resin in solid phase synthesis.

Examples of suitable protecting groups of the N-terminal amino of the dipeptide analogues are groups removed both by hydrolytic and catalytic cleavage including benzyloxycarbonyl and derivatives thereof such as pmethoxybenzyloxycarbonyl, amyloxycarbonyl, biphenyl isopropoxycarbonyl, o- nitrophenylsulphenyl, phthaloyl and t-butoxycarbonyl. In addition it is also possible in-the case of these dipeptide analogues to use N-protecting groups such as acyl groups which are not normally used in peptide chemistry because of the tendency to cause racemisation at the a-carbon atom during coupling.

As indicated above, the dipeptide analogues may be used for the synthesis of a wide variety of peptides containing one or more amide linkages modified according to the present invention. Such peptides are prepared by the addition of one or more amino acids, peptides or dipeptide analogues according to the present invention to one or both of the N- and C-terminal ends of the dipeptide analogue using either classical, solution, methods or solid phase methods and effecting reaction between the appropriate terminus of the reactant and of the dipeptide analogue according to the general procedures indicated above. Thus a free- N-terminal amino group of the reactant may be reacted with the C-terminal carboxy group of a suitable derivative thereof in the dipeptide analogue or vice versa, other reactive groups in the reactant being appropriately protected as described above. Such a procedure is exemplified in the earlier application in respect of the dipeptide analogues covered therein.

The invention is illustrated by the following Example.

EXAMPLE 5-Amino-4-Oxo-6-Phenylhexanoic Acid (Analogue of phenylalanyl-glycine in which -COCH2- replaces -CONH-) (1) BOC-NH-CH(CH2C6H5COCHN2 N-methylmorpholine (0.40 ml, 3.63 mmoles) is added to a stirred solution of L Boc-phenylalanine (0.96g, 3.63 mmoles) in dry ethyl acetate (12.00 mls). The solution is cooled to --100C and isobutylchloroformate (0.48 ml, 3.63 mmoles) is added. After 7 minutes the mixture is filtered into an ice-cooled flask and the precipitate washed with precooled ethyl acetate (6.00 mls). An ethereal solution of diazomethane (8.00 mmoles in 75.mls) is then added and the yellow solution kept at 4"C overnight. Evaporation of the solvents affords the pure diazoketone (1) as a yellow-orange solid, m.p. (hexane 95--96"C; Vmax (CHCl3): 2105, 1710, 1640 cm (2) BOC-NH-CH(CH2C6H5)-COCH2Br A solution of HBr in ethyl acetate (1.96 mls of a 0.5it solution, diluted to 50 mls) is added over 15 minutes to a stirred solution of the diazoketone (1) (0.29g, 1.00 mmoles) in dry ethyl acetate (10.00 mls) at --100C. The solvent is evaporated in vacuo and the residue crystallised from hexane giving the pure bromoketone (2) as white needles (0.27g 79%), m.p. 103.5--1040C, Vmax (CHCl3): 1705, 1490 cm-'.

(3) BOC-NH-CH(CH2C6H5)-COCH2P+ (C6Hs)3Br- Triethylamine (10,us) is added to a stirred solution of the bromoketone (2) (0.489g, 1.43 mmoles) in sodium dried benzene (4.80 mls). Triphenylphosphine (0.375g, 1.43 mmoles) in dry benzene (5.72 mls) is added the the solution stirred overnight at room temperature. The white crystalline deposit is collected and washed with dry ether giving the pure keto triphenylphosphonium bromide (3) (0.710g 82 /"), m.p. 92-940C; Vmax (CHCl3); 1715, 1695, 1495 cm-l.

(4) BOC-NH-CH(CH2C6115)-COCH=PPh3 A suspension of the keto triphenylphosphonium bromide (3) (0.63g) in ethyl acetate (27.0 mls) is stirred vigorously overnight with sodium carbonate solution (I M, 27.0 mls). The ethyl acetate is separated and the aqueous phase extracted once more with ethyl acetate. The extracts are combined, washed with saline and dried over magnesium sulphate. Evaporation of the solvent affords the pure ylide (4) as a foam (0.54g, 100%), Vmax (CHCl3): 1695, 1545, 1480 cm-l.

(5) BOC-NH-ClI(CH2C6H5)-COCH2CH2CO2C2H A solution of the ylide (4) (0.23g, 0.44 mmoles) in 4.4 ml of anhydrous dimethyl formamide and ethyl bromoacetate (0.73g, 4.40 mmoles) is vigorously stirred under N2 at 80"C with anhydrous sodium carbonate (0.90g) for 3.0 hours. The sodium carbonate is removed by filtration and the filtrate and washings are combined and evaporated. The residue is partitioned between ethyl acetate and water and the organic phase washed once more with water and then with saline. The solution is dried over magnesium sulphate and evaporated giving a pale yellow gum. A portion is purified by P.L.C. using ethyl acetate:acetone:benzene (1:2:3 v/v/v) for development. Elution with ethyl acetate affords the pure ylide BOC-NH-CH- (CH2C6Hs)COC(CH2CO2C2Hs)=P(C6Hs)3 as a pale yellow gum, vmax(CHCl3): 1725, 16951525. 1490, 1480 cm-'.

The remainder of the crude alkylation product (188.0 mg, about 70% of the total) is dissolved in ethyl acetate and treated with HCI in ethyl acetate (0.77 ml of a 0.40M solution). The solvent is evaporated affording the ylide HCI salt which is dissolved in purified acetonitrile (40 mls). Deaerated water (40 mls) is added and half the solution is electrolysed under nitrogen at 25 volts using mercury and platinum electrodes for 1.0 hours at room temperature. The cloudy solution is evaporated, redissolved in methanol (0.5 ml) and chromatographed on Sephadex LH20 (67.0x3.25 cm2) using methanol as eluent. The pure keto-ester (5) (35.cm mg) is typically eluted in fractions 43 and 44 (6.0 ml fractions, 12.0 mls/hour). The remaining ylide salt is treated in the same way affording further pure keto-ester (5) (total 69.3 mg, 75% overall yield of the two steps). Vmax (CHCl3): 1720, 1705, 1490 cm-'.

(6) BOC-NH-CH(CH2C6H5)-COCH2CH2COOH A solution of keto-ester (5) (16.0 mg, 0.046 mmoles) in acetone (0.39 ml) and sodium hydroxide (0.118M, 0.39 mls) is stirred at room temperature for 2.0 hours.

The solution is diluted with water and extracted once with ethyl acetate. The aqueous phase is acidified to pH 3 with citric acid and extracted thrice with ethyl acetate. The ethyl acetate is washed with water and saline and dried over magnesium sulphate. Evaporation of the solvent affords the pure keto-acid (6).

(14.7 mg, 100%) as a colourless gum. Vmax (CHCl3): 1710, 1490 cm-'. z (CHCl3): 2.8 (5H, multiplet, C6Hs), ca 3.6 broad (D2O-exchangeable, COOH), 4.95 (1H multiplet, D2O-exchangeable, NH), 5.66 (1H, multiplet, CH), 6.86-7.65 (6H, complex, 3xCH2), 8.62 (9H, s, BOC-tBu).

(7) H2N-CH(CH2C6H5)-COCH2CH2COOH The BOC protecting group is removed by the use of trifluoroacetic acid to give 5-amino-4-oxo-6-phenylhexanoic acid, H2N-CH-(CH2C6H5)- COCH2CH2COOH.

WHAT WE CLAIM IS: 1. A compound being an analogue of a dipeptide formed between one of the amino acids alanine, p-alanine, arginine, asparagine, aspartic acid, 3,5dibromotyrosine, cysteine, cystine, dopa, glycine, glutamic acid, glutamine, histidine, hydroxylysine, hydroxyproline, 3,5-diiodotyrosine, isoleucine, leucine, lysine, methionine, ornithine, phenylalanine, proline, pyroglutamic acid, serine, spinacin, threonine, thyroxine, tryptophan, tyrosine and valine, and another of said amino acids or between two molecules of the same amino acid, the linking amide group of the dipeptide being modified in the analogue through replacement of the nitrogen atom by the trivalent group -CH,

**WARNING** end of DESC field may overlap start of CLMS **.

Claims (12)

**WARNING** start of CLMS field may overlap end of DESC **. washed with dry ether giving the pure keto triphenylphosphonium bromide (3) (0.710g 82 /"), m.p. 92-940C; Vmax (CHCl3); 1715, 1695, 1495 cm-l. (4) BOC-NH-CH(CH2C6115)-COCH=PPh3 A suspension of the keto triphenylphosphonium bromide (3) (0.63g) in ethyl acetate (27.0 mls) is stirred vigorously overnight with sodium carbonate solution (I M, 27.0 mls). The ethyl acetate is separated and the aqueous phase extracted once more with ethyl acetate. The extracts are combined, washed with saline and dried over magnesium sulphate. Evaporation of the solvent affords the pure ylide (4) as a foam (0.54g, 100%), Vmax (CHCl3): 1695, 1545, 1480 cm-l. (5) BOC-NH-ClI(CH2C6H5)-COCH2CH2CO2C2H A solution of the ylide (4) (0.23g, 0.44 mmoles) in 4.4 ml of anhydrous dimethyl formamide and ethyl bromoacetate (0.73g, 4.40 mmoles) is vigorously stirred under N2 at 80"C with anhydrous sodium carbonate (0.90g) for 3.0 hours. The sodium carbonate is removed by filtration and the filtrate and washings are combined and evaporated. The residue is partitioned between ethyl acetate and water and the organic phase washed once more with water and then with saline. The solution is dried over magnesium sulphate and evaporated giving a pale yellow gum. A portion is purified by P.L.C. using ethyl acetate:acetone:benzene (1:2:3 v/v/v) for development. Elution with ethyl acetate affords the pure ylide BOC-NH-CH- (CH2C6Hs)COC(CH2CO2C2Hs)=P(C6Hs)3 as a pale yellow gum, vmax(CHCl3): 1725, 16951525. 1490, 1480 cm-'. The remainder of the crude alkylation product (188.0 mg, about 70% of the total) is dissolved in ethyl acetate and treated with HCI in ethyl acetate (0.77 ml of a 0.40M solution). The solvent is evaporated affording the ylide HCI salt which is dissolved in purified acetonitrile (40 mls). Deaerated water (40 mls) is added and half the solution is electrolysed under nitrogen at 25 volts using mercury and platinum electrodes for 1.0 hours at room temperature. The cloudy solution is evaporated, redissolved in methanol (0.5 ml) and chromatographed on Sephadex LH20 (67.0x3.25 cm2) using methanol as eluent. The pure keto-ester (5) (35.cm mg) is typically eluted in fractions 43 and 44 (6.0 ml fractions, 12.0 mls/hour). The remaining ylide salt is treated in the same way affording further pure keto-ester (5) (total 69.3 mg, 75% overall yield of the two steps). Vmax (CHCl3): 1720, 1705, 1490 cm-'. (6) BOC-NH-CH(CH2C6H5)-COCH2CH2COOH A solution of keto-ester (5) (16.0 mg, 0.046 mmoles) in acetone (0.39 ml) and sodium hydroxide (0.118M, 0.39 mls) is stirred at room temperature for 2.0 hours. The solution is diluted with water and extracted once with ethyl acetate. The aqueous phase is acidified to pH 3 with citric acid and extracted thrice with ethyl acetate. The ethyl acetate is washed with water and saline and dried over magnesium sulphate. Evaporation of the solvent affords the pure keto-acid (6). (14.7 mg, 100%) as a colourless gum. Vmax (CHCl3): 1710, 1490 cm-'. z (CHCl3): 2.8 (5H, multiplet, C6Hs), ca 3.6 broad (D2O-exchangeable, COOH), 4.95 (1H multiplet, D2O-exchangeable, NH), 5.66 (1H, multiplet, CH), 6.86-7.65 (6H, complex, 3xCH2), 8.62 (9H, s, BOC-tBu). (7) H2N-CH(CH2C6H5)-COCH2CH2COOH The BOC protecting group is removed by the use of trifluoroacetic acid to give 5-amino-4-oxo-6-phenylhexanoic acid, H2N-CH-(CH2C6H5)- COCH2CH2COOH. WHAT WE CLAIM IS:

1. A compound being an analogue of a dipeptide formed between one of the amino acids alanine, p-alanine, arginine, asparagine, aspartic acid, 3,5dibromotyrosine, cysteine, cystine, dopa, glycine, glutamic acid, glutamine, histidine, hydroxylysine, hydroxyproline, 3,5-diiodotyrosine, isoleucine, leucine, lysine, methionine, ornithine, phenylalanine, proline, pyroglutamic acid, serine, spinacin, threonine, thyroxine, tryptophan, tyrosine and valine, and another of said amino acids or between two molecules of the same amino acid, the linking amide group of the dipeptide being modified in the analogue through replacement of the nitrogen atom by the trivalent group -CH,

or when the C-terminal amino acid residue of the dipeptide is derived from other than hydroxyproline, proline, pyroglutamic acid and spinacin alternatively by the trivalent group lower lower alkyl), or a derivative of such a dipeptide analogue formed at the terminal amino and/or carboxyl group thereof and/or at other functional group(s) therein.

2. A compound according to Claim 1, wherein said amino acids are selected from alanine, arginine, asparagine, aspartic acid, 3,5-dibromotyrosine, cysteine, cystine, dopa, glycine, glutamic acid, glutamine, histidine, hydroxylysine, hydroxyproline, 3,5-diiodotyrosine, isoleucine, leucine, lysine, methionine, ornithine, phenylalanine, proline, pyroglutamic acid, serine, spinacin, threonine, thyroxine, tyrosine and valine.

3. A compound according to Claim 1, wherein said amino acids are selected from alanine, arginine, asparagine, aspartic acid, cysteine, cystine, glycine, glutamic acid, glutamine, histidine, isoleucine, leucine, lysine, methionine, phenylalanine, proline, serine, threonine, tryptophan, tyrosine and valine.

4. A compound according to any of Claims 1 to 3, wherein the dipeptide of which the compound is an analogue derives from amino acids which, where enantiomorphic are of the L-configuration.

5. A compound according to any of Claims 1 to 4, wherein the trivalent group which replaces the nitrogen atom of the linking amide group of the dipeptide is the group -CH.

6. A compound according to any of Claims 1 to 4, wherein the C-terminal amino acid residue of the dipeptide of which the compound is an analogue is derived from an amino acid other than hydroxyproline, proline, pyroglutamic acid and spinacin.

7. A compound being an analogue of a dipeptide formed between one of the amino acids alanine, arginine, asparagine, aspartic acid, 3,5-dibromotyrosine, cysteine, cysine, dopa, glycine, glutamic acid, glutamine, histidine, hydroxylysine, 3,5-diiodotyrosine, isoleucine, leucine, lysine, methionine, ornithine, phenylalanine, serine, threonine, thyroxine, tryptophan, tyrosine and valine, and another of said amino acids or, except in the case of glycine, between two molecules of the same amino acid, the linking amide group of the dipeptide being replaced in the analogue by a group -COCH2, or a derivative of such dipeptide analogue formed at the terminal amino and/or carboxyl group thereof and/or at other functional group(s) therein.

8. A compound being an analogue of one of the dipeptides L-tyrosyl-glycine, L-phenylalanyl-glycine, L-prolyl-glycine, L-leucyl-L-arginine and glycyl-L-leucine in which the linking amide group of the dipeptide is replaced by a group -COCH2-, or a derivative thereof formed at the terminal amino group and/or carboxyl group thereof, and/or at another functional group therein.

9. A compound according to Claim 9, being an analogue of one of the dipeptides L-tyrosyl-glycine, L-phenylalanyl-glycine, and L-prolyl-glycine or a derivative thereof.

10. A method for the preparation of modified peptides, wherein a compound being an analogue of a dipeptide formed between one of the amino acids alanine, - alanine, arginine, asparagine,."aspartic acid, 3,5-dibromotyrosine, cysteine, cystine, dopa, glycine, glutamic acid, glutamine, histidine, hydroxylysine, hydroxyproline, 3,5-diiodotyrosine, isoleucine, leucine, lysine, methionine, ornithine, phenylalanine, proline, pyroglutamic acid, serine, spinacin, threonine, thyroxine, tryptophan, tyrosine and valine, and another of said amino acids or between two molecules of the same amino acid, the linking amide group of the dipeptide being modified in the analogue through replacement of the nitrogen atom by the trivalent group -CH, or when the C-terminal amino acid residue of the dipeptide is derived from other than hydroxyproline, proline, pyroglutamic acid and spinacin alternatively by the trivalent group -C-(- lower alkyl), or a derivative of such a dipeptide analogue formed at the terminal amino and/or carboxyl group thereof and/or at other functional group(s) therein, is reacted at one or both of the C- and N-- terminii thereof with an amino acid or peptide, or a derivative thereof, with the formation of a peptide bond.

I I. A method according to Claim 10, wherein the compound reacted is a compound according to any of Claims 2 to 9.

12. A modified peptide which incorporates the residue of at least one compound according to any of Claims 1 to 9.