WO1991012021A2

WO1991012021A2 - Therapeutic agents, and intermediates for the synthesis thereof

Info

Publication number: WO1991012021A2
Application number: PCT/GB1991/000215
Authority: WO
Inventors: Jonathan Miles-Brown
Original assignee: Oxford Virology Plc
Priority date: 1990-02-13
Filing date: 1991-02-13
Publication date: 1991-08-22
Also published as: WO1991012021A3; AU7217891A

Abstract

Therapeutic agents are provided which are adapted to deliver a therapeutically active substance to a selected cell, group of cells or organ and to intermediates useful in synthesising said agents. The therapeutic agents comprise a therapeutically active substance covalently linked to a polypeptide of low molecular weight, which polypeptide comprises an amino acid sequence which is recognised by a recognition site of a receptor of a selected cell, group of cells or organ. The therapeutic agent may be therapeutically active itself and/or is convertible at or within the selected cell, group of cells or organ to a form which is therapeutically active. Examples of target cells are CD4 lymphocytes, and the amino acid sequence comprises a sequence which is recognised by the CD4 receptor.

Description

THERAPEUTIC AGENTS, AND INTERMEDIATES FOR THE SYNTHESIS THEREOF

This invention relates to therapeutic agents adapted to deliver a therapeutically active substance to a selected cell, group of cells or organ and to intermediates useful in synthesising said agents.

In conventional pharmacological practice, it is often desirable to adapt the chemical structure of a therapeutically active substance to ensure that the substance is delivered to a selected cell, group of cells or organ. This may be achieved in a number of ways, for example by adapting the structure of the therapeutically active substance itself so that, on administration, it is concentrated in a selected target organ. Alternatively, the substance may be formulated into a pharmaceutical dosage form which is adapted to deliver the substance to the target site. Such procedures are referred to as "targetting" the substance.

In many instances it is highly desirable to target the substance to to a selected cell, group of cells or organ because the substance is highly toxic and it would otherwise be impossible to acheive a sufficiently high concentration of the substance at the target site without producing undesirable toxic side effects in the patient. Thus it has been proposed to manufacture cytotoxic drugs by covalently bonding ricin (a potent cytotoxic agent) to FcIgG antibody fractions, with the intention that the combination would be targetted at specific cells. However such therapeutic agents have the distinct disadvantage that the combined molecule is large and highly

immunogenic. Consequently immunological side effects can occur, which render such techniques ineffective.

In treatments proposed for viral infections, including HIV and Herpes virus infections, and in treatments proposed for treating malignancies, metabolite analogues have been synthesised, which interfere with mechanisms of metabolism involved in viral replication and malignant growth. Such metabolite analogues include, for example, nucleotide analogues such as AZT (used in treating HIV infections), (9-[(2-hydroxyethoxy)methyl]guanine (Acyclovir) and 5-fluorouracil (used in treating Herpes infections) and antimetabolites used as in cancer therapy. Such metabolite analogues and antimetabolites are often highly cytotoxic and it is extremely difficult to administer a dose which is effective in dealing with the clinical condition, without inducing serious toxic side effects.

Occasionally, otherwise unacceptable side effects can be tolerated, for example in "last ditch" treatments of intractible diseases, forexample HIV infections.

Thus, AZT (3'-deoxy-3'-azidothymidine) has been shown to to inhibit the replication of HIV in vitro in cultures of CD4 lymphocytes and is now in widespread use in the treatment of AIDS patients.

Nonetheless AZT has toxic side effects, most notably the killing of bone marrow cells, and these side effects limits its effectiveness. Moreover, potential therapeutic agents which are structurally related to AZT and display similar or enhanced in vitro effects (for example ddC (2',3'-dideoxycytidine) and ddl (2',3'-dideoxyinosine)) have been excluded from widespread use because of their high toxicity.

We have now developed a novel procedure for targetting a therapeutically active substance, which comprises forming a

therapeutic agent by covalently linking the substance to a polypeptide of relatively low molecular weight, which polypeptide comprises an amino acid sequence which is recognised by a recognition site of a receptor of a selected cell, group of cells or organ and wherein the therapeutic agent is therapeutically active itself and/or is

convertible at or within the selected cell, group of cells or organ to a form which is therapeutically active.

Thus according to one aspect of the present invention there is provided a therapeutic agent comprising a therapeutically active substance covalently linked to a polypeptide of low molecular weight, which polypeptide comprises an amino acid sequence which is recognised by a recognition site of a receptor of a selected cell, group of cells or organ and wherein the therapeutic agent is therapeutically active itself and/or is convertible at the selected cell, group of cells or organ to a form which is therapeutically active. This therapeutically active form can, for example, consist of the therapeutically aptive substance itself.

The polypeptide preferably contains not more than 50.

preferably not mere than 30 and most preferably not more than 20 amino acid resdues.

Preferably, the amino acid sequence is chosen so to be capable of interacting with a receptor site on the surface of a specific type of cell. It is further advantageous if the interaction of the amino acid sequence with the recognition site facilitates entry of the therapeutically active substance into the target cell.

In the case where the selected cells are CD4 lymphocytes, the receptor can be the so-called CD4 receptor. This receptor has been shown to be capable of binding to a region of HIV-1 gpl20 surface glycoprotein.

Thus in accordance with a preferred embodiment of the invention in which the selected cells are CD4 cells, the amino acid sequence comprises a sequence which is recognised by the CD4 receptor.

The residues of the gpl-20 glycoprotein which are implicated in binding to the CD4 recognition site have been investigated by Cardonnier et al ("Single amino-acid changes, in HIV envelope affect tropism and receptor binding"; Cordonnier, A. et al; Nature 340, 571-574, (1989)) and the amino acid sequence CRIKQFINMWQE

(CysArglleLysGlnPhelleAsnMetTrpGlnGlu in three letter notation) identified as being associated with recognition by the CD4 site.

Complete homology with the aforementioned sequence is not necessary for CD4 recognition. Thus, Cardonnier et al (supra) demonstrated that recognition was substantially unimpaired with the following substitutions in the above sequence:

C R I K Q F I N M W Q E

- - T - - - - - - - - - - - - - T - - - - - - - - - - - R - - - - - - - - - - - - T - - - - - - - - - - - - - - - Y - - - - - - - - - - - H - - - - The polypeptide of the therapeutically active substance of the invention thus preferably includes the amino acid sequence

CysArglleLysGlnPhelleAsnMetTrpGlnGlu, or a related sequence also capable of binding to the CD4 recognition site. Such related

sequences preferably have a homology with the precise sequence

CysArglleLysGlnPhelleAsnMetTrpGlnGlu such that at least 8, and preferably at least 10, of the amino acids correspond to the ones appearing in the precise sequence. Most preferably, not more than one amino acid difference exists.

The actual sequence of the polypeptide portion of the therapeutically active substances according to the invention may differ from the precise sequence CysArglleLysGlnPhelleAsnMetTrpGlnGlu in a number of possible ways. For example one or more amino acid residues may be omitted and/or one or more amino acid residues may be replaced by other residues, provided that such omission or replacement does not adversely interfere with the capacity to bind the CD4 recognition site.

Thus for example, the first amino acid residue Cys may be omitted or replaced by another residue.

Deviations from the precise sequence preferably conform to the following protocol:

Original Preferred amino acid substitution

Cys omit or replace by another amino acid, e.g. Met

Arg omit or replace by another amino acid, e.g. Lys His

lie omit or replace by another amino acid, e.g. Ala Leu Val

Lys omit or replace by another amino acid. e.g. Arg His

Gin omit or replace by another amino acid, e.g. Asn Thr

Phe omit or replace by another amino acid, e.g. Trp Tyr

lle omit or replace by another amino acid, e.g. Ala Leu Val

Asn omit or replace by another amino acid, e.g. Gin His Tyr

Met omit or replace by another amino acid, e.g. Cys

Trp (preferable not omitted or replaced) -

Gin omit or replace by another amino acid, e.g. Asn

Glu omit or replace by another amino acid, e.g. Asp

Preferably, not more than 4, more preferably not more than 2 and most preferably only one of the aforementionaed omissions and/or replacements are made. It will be noted from the above protocol that the tenth amino acid Trp is preferably conserved.

One particularly useful class of therapeutically active substances which may be covalently linked to a polypeptide of low molecular weight in forming therapeutic agents according to the invention are drugs which are active as reverse transcriptase inhibitors. Others are antimetabolites of the kind used in

anti-cancer therapy.

Specific examples of suitable therapeutically active substances which may be covalently linked to a polypeptide of low molecular weight in forming therapeutic agents according to the invention include therapeutically active nucleotide analogues such as 3'-deoxy-3'-azidothymidine, 2',3'-dideoxycytidine,

2',3'-dideoxyinosine, Acyclovir and 5-fluorouracil.

It is desirable for the therapeutically active substance to be covalently linked to the polypeptide of low molecular weight by a covalent linkage which is labile under conditions prevailing at or within the target cell or organ. Esterases and phosphatases are commonly found in vivo and it is accordingly particularly preferred for the therapeutically active substance to be covalently linked to the polypeptide of low molecular weight by linkages which are cleaved by such enzymes, e.g. for the linkages to include ester groups.

In cases where the therapeutically active substance comprising the therapeutic agent of the invention includes a free hydroxyl group, it is particularly convenient for the therapeutically active substance to be linked to the polypeptide through an ester linkage involving such a hydroxyl group. Specific examples of therapeutically active substances having such free hyroxyl groups are nucleotide analogues having reverse transcriptase activity such as 3 '-deoxy-3'-azidothymidine, 2',3'-dideoxycytidine and

2',3'-dideoxyinosine, each of which have a 5'-hydroxy group. Although it is possible for the therapeutically active substance to be linked to the polypeptide through an ester linkage formed with the C-terminal carboxyl group of the polypeptide or to a side chain (and therapeutic agents so-formed are included in the present invention), compounds having this structure can be

inconvenient to synthesise in practise. This is because polypeptides are conventionally synthesised chemically by solid phase techniques in which a possibly protected amino acid is covalently bound to a solid resin via its free α-carboxyl group and NH₂-protected HOOC-activated amino acids added sequentially. In this technique (the so-called "Merrifield" or "Stewart" technique) typically involves the use of t-butyloxycarbonyl (t-Boc) or fluorenylmethoxycarbonyl (Fmoc) protecting groups and activation of the amino acids using

dicyclohexylcarbodiimide or by formation of pentafluorophenyl esters. Other suitable protecting groups, for example trityl groups may be used to protect thiol groups of cysteine residues. Synthesis of the peptide proceeds by condensation successively of appropriate protected activated amino acids.

At each stage, following removal the N-terminal protecting group, the bound, derivatised N-terminal amino acid is condensed with the next protected, activated amino acid. This procedure is repeated until a peptide of the desired length and sequence is obtained.

The product of the last stage will consist of polypeptide bound to a solid support via its C-terminal HOOC- group and having a protected -NH₂ group at t e N-terminal end. (Reactive side chains will normally also be protected).

In accordance with the invention it has been found to be particularly convenient and effective to form the therapeutic agent comprising a therapeutically active substance covalently linked to a polypeptide by linking the therapeutically active substance covalently to the N-terminal end of the polypeptide. Thus, where the polypeptide is formed by a solid-phase procedure, e.g. in the manner described above, the therapeutically active substance can be covalently bonded thereto prior to removal of the polypeptide from the support. Where the therapeutically active substance has a free hydroxyl group and (1) it is desired to link the therapeutically active substance to the polypeptide at the N-terminal end, and (2) it is desired to link the therapeutically active substance to the polypeptide via an ester linkage involving such an aforementioned free hydroxyl group it is necessary for a linker to be introduced between the therapeutically active substance and the polypeptide.

This linker may conveniently have the structure

-OC(R)CO- (I) wherein the residue R is (i) a divalent substituted or unsubstituted divalent hydrocarbyl group containing 1 to 20, preferably 1 to 10, most preferably 1 to 6 carbon atoms (ii) a heterocyclic group

containing 5 or six ring carbon atoms and at least one heteroatom selected fron N, O and S or (ii) a direct bond linking the groups -OC and CO-.

R preferably represents a divalent hydrocarbyl group of the formula -(CH₂)_n- wherein n is an integer from 1 to 4, preferably 2, as in succinyl.

Any substituents on R are preferably selected from hydroxy, oxo, amino, carboxy, halo, trifluoromethyl and nitro. An example of a linker of structure -OC(R)CO- in which R is substituted with a hydroxy and a carboxy group is a linker of formula

-OC.CH₂.C(OH,COOH) .CH₂.CO- derived from citric acid.

The preferred therapeutic agents according to the invention may thus be represented by the formula

[TES]—CO(R)CO-NH—[polypeptide]—COOH (II) wherein "-[polypeptide]-" denotes the polypeptide of low molecular weight (excluding its N-terminal amino group and its C-terminal carboxyl group), R is as defined above and "[TES]-" represents the residue of an therapeutically active substance.

Preferably [polypeptide], R and [TES] represents the preferred species described previously. Preferably the therapeutic agents of formula

[TES]—CO(R)CO-NH—[polypeptide]—COOH (II) are produced according to the invention by reacting the polypeptide of low molecular weight (suitably protected and preferably bound via its C-terminal carboxyl group to a solid support) with a reactant of formula

[TES]—CO(R)COOH (III) where R and [TES] are as defined above, or a carboxy-activated derivative thereof, i.e. one in which the group HOOC- is converted to an activated form capable of reacting with an amino group, e.g. a group of the structure HalOC- where Hal represents a halogen atom. Other examples include carboxy-activated derivative be formed in situ, for example by reaction with a carbodiimide such as

dicyclohexylcarbodiimide or with Castro's Reagent.

Other examples of carboxy-activated derivatives of compounds of formula III are compounds in which the -COOH group is activated by formation of a mixed anhydride with a suitable activated carboxylic or other acid e.g. 2,6-dichlorobenzoic acid. Such mixed anhydrides may be formed, for example, by reaction of a compound of formula

[TES]—CO(R)COOH with an acyl chloride, e.g. 2,6-dichlorobenzoyl chloride.

The intermediates of Formula (III) are novel and form a further aspect of the present invention.

A particularly preferreed class of such intermediates are those of the formula

[NUC]—OC(R)COOH (IV)

(or a derivative thereof) wherein R is as defined above and NUC represents the residue of 3'-deoxy-3'-azcthymidine,

3'-deoxy-3'-azocytidine or 3^,-deoxy-3'-azoinosine. Derivatives include salts with e.g. metals such as alkali metals, or with ammonia or an amine, e.g. an amine of formula NR₁R₂R₃ wherein each of

R₁, R₂ and R₃ represent a hydrogen atom or a C1-4 alkyl group.

Other suitable amines include heterocyclic amines, e.g. pyridine. By way of example, a compound of formula (IV) may be prepared by reacting AZT with succinic anhydride to give the

corresponding 5'-O-succinate half ester. This half ester may be isolated as the free acid, or as a salt with e.g. a metals such as an alkali metal, or with ammonia or an amine, e.g.triethylamine.

The salt, or more preferably the free acid may be reacted with the C-terminal HOOC- group of the polypeptide, preferably in the presence of a suitable activating agent.

The synthesis of a therapeutic agent according to the invention comprising a residue of AZT linked via a 5'-O-succinyl linker to the N-terminus of a dodecapeptide consisting of the CD4 recognition sequence of HIV-1 gp120 surface glycoprotein will now be described by way of example.

EXAMPLE 1

1. Preparation of 5'-O-succinoyl AZT (3)

To a stirred solution of AZT (1) (0.801g, 3.00 mmol) in anhydrous pyridine (10ml-) was added succinic anhydride (2) (2.5 m.e., 0.750g, 7.50 mmol). The solution was stirred at room temperature for 20 hours by which time t.l.c. indicated completion of reaction.

The pyridine was removed by evaporation in vacua and the residue was dissolved in chloroform (50 ml) and extracted into saturated sodium bicarbonate solution (30 ll). The aqueous layer was then neutralised with IM hydrochloric acid (to pH4) and extracted with chloroform (3 x 40 ml). The chloroform extracts were dried (MgSO₄) and evaporated in vacua to give the required derivative (3) as a syrup which was almost pure by t.l.c. This was dissolved in a minimum of chloroform and applied to a thin bed of silica gel (15g) in a 40g column pre-equilibrated in chloroform. Elution with chloroform then chloroform:ethanol (98:2, v:v) gave the required product as t.l.c. homogenous foam.

Yield: 0.902g (82%) ;

R_f : 0.3 (CHCl₃-EtOH (9:1));

360 MHz ¹H nmr δ_H(CDCl₃), 1.87 (3H, d, J 1Hz),

2.37-2.45 (1H, m), 2.50-2.78 (4H, m), 2.85-2.91 (1H, m), 4.14-4.11 (1H, m), 4.25 (1H, dd, J 4, 12 Hz), 4.34-4.37, (1Hz, m), 4.85 (1H, dd, J 1.2, 12 Hz), 5-90 (1H, tr, J 6Hz) , 7.45 (1H, J 1Hz), 10.65 (1H, brs);

360 MHz ¹³C nmr δ_C (CDCl₃) 12.29, 28.60, 28.63. 37.64, 60.09, 62.99. 82.68, 85.63, 109.65. 137.34, 149.38, 166.22, 172.61, 1 77.64.

2. Synthesis of peptide

A resin-bound, protected derivative of the dodecapeptide CRIKQFINMWQE (IV) was synthesised by a conventional Merrifield technique. Amino groups of the side chains of basic amino acids were protected using Tboc groups and the side chain sulfhydryl group of Cys was protected as a trityl thioether. The amino acid reagents were N-protected as Fmoc derivatives. The use of other protecting groups is of course possible.

3. Condensation with succinyl half ester

The Fmoc-protected -NH₂ group at the N-terminus is de-protected using piperidine and the resin-bound, side-chain protected derivative of the dodecapeptide was the condensed with the AZT succinyl half ester (8 molar proportions) in the presence of dicyclohexylcarbodiimide (10 molar proportions) in pyridine.

Finally, the decapeptide-succinyl AZT is deprotected and released from the solid support by treatment with 95% trifluoroacetic acid and purified using hplc.

Using similar techniques, the AZT-succinate half ester can be linked to to other peptide sequences, for example sequences differing from the sequence CRIKQFINMWQE, but nonetheless capable of binding the CD4 recognition site. This is illustrated in Example 2 wherein the AZT- succinate half ester is linked to a peptide having the sqeuence RIKQFINMWQE. Example 2

1. Preparation of 5'-O-succinoyl AZT (3)

5'-O-succinoyl AZT was prepared by the procedure described in Example 1 (1).

2. Synthesis of peptide

A resin-bound, protected derivative of the undecapeptide RIKQFINMWQE (IV) was synthesised by a conventional Merrifield

technique. Amino groups of the side chains of basic amino acids were protected using Tboc groups and the side chain sulfhydryl group of Cys was protected as a trityl thioether. The amino acid reagents were N-protected as Fmoc derivatives. The use of other protecting groups is of course possible.

3. Condensation with succinyl half ester

The Fmoc-protected -NH₂ group at the N-terminus is de-protected using piperidine and the resin-bound, side-chain

protected derivative of the undecapeptide was then condensed with the AZT succinyl half ester (8 molar proportions) in the presence of dicyclohexylcarbodiimide (10 molar proportions) in pyridine.

Alternatively, the AZT-succinate half ester can be linked to peptide sequences corresponding to the binding regions of other proteins known to bind to specific receptor sites. For example, HIV is known also preferentially to infect macrophages. In order to target AZT to these cells, the AZT succinyl ester can be linked to a peptide sequence corresponding to the IgG Fc fragment which is known to bind to the macrophage Fc receptor.

Therapeutically active substances other than AZT can be linked to a peptide chain in accordance with the invention and similarly, linking groups other than succinyl groups may be used, e.g. oxalyl and phosphodiester groups and acyl groups derived from citric acid. The use of succinyl-derived linker groups is particularly preferred in accordance with the invention in view of the availability of succinic anhydride as a reactant. Also because succinic acid is a metabolite, succinate released on hydrolysis of the therapeutic agent on administration will have no toxic effects.

Claims

1. A therapeutic agent comprising a therapeutically

active substance covalently linked to a polypeptide of low molecular weight, which polypeptide comprises an amino acid sequence which is recognised by a recognition site of a receptor of a selected cell, group of cells or organ and wherein the therapeutic agent is

therapeutically active itself and/or is convertible at or within the selected cell, group of cells or organ to a form which is

therapeutically active.

2. A therapeutic agent according to Claim 1 wherein the therapeutically active form consists of the therapeutically active substance itself.

3- A therapeutic agent according to Claim 1 or Claim 2 wherein the polypeptide contains not more than 50 amino acid residues.

4. A therapeutic agent according to any preceding claim wherein the amino acid sequence is capable of interacting with a receptor site on the surface of a target cell to facilitate entry of the

therapeutically active substance into the target cell.

5. A therapeutic agent according to Claim 4 wherein

the target cells are CD4 lymphocytes, and the amino acid sequence comprises a sequence which is recognised by the CD4 receptor.

6. A therapeutic agent according to any preceding claim wherein the polypeptide includes the amino acid sequence

CysArglleLysGlnPhelleAsnMetTrpGlnGlu or a related sequence capable of binding to the CD4 recognition site.

7. A therapeutic agent according to Claim 6 wherein the related sequence has a homology with the sequence

CysArglleLysGlnPhelleAsnMetTrpGlnGlu such that at least 8, and preferably at least 10, of the amino acids correspond to the ones appearing in the precise sequence.

8. A therapeutic agent according to Claim 6 or Claim 7 wherein the related sequence has a homology with the sequence

CysArglleLysGlnPhelleAsnMetTrpGlnGlu such that one or more amino acid residues are omitted and/or replaced by another residue or residues.

9. A therapeutic agent according to Claim 8 wherein the related sequence has a homology with the sequence

CysArglleLysGlnPhelleAsnMetTrpGlnGlu such that deviations from the precise sequence given adopt the following protocol:

Original Deviations

amino acid

Cys Met

Arg Lys His

Ile Ala Leu Val

Lys Arg His

Gin Asn Thr

Phe Trp Tyr

Ile Ala Leu Val

Asn Gin His Tyr

Met Cys

Trp -

Gin Asn

Glu Asp

10. A therapeutic agent according to Claim 8 wherein the related sequence has a homology with the sequence

Original Preferred amino acid substitution

Cys omit or replace by another amino acid, e.g. Met

Arg omit or replace by another amino acid, e.g. Lys His lle omit or replace by another amino acid, e.g. Ala Leu Val

Lys omit or replace by another amino acid. e.g. Arg His

Gin omit or replace by another amino acid, e.g. Asn Thr

Phe omit or replace by another amino acid, e.g. Trp Tyr

Ile omit or replace by another amino acid, e.g. Ala Leu Val

Asn omit or replace by another amino acid, e.g. Gin His Tyr

Met omit or replace by another amino acid, e.g. Cys

Trp (preferable not omitted or replaced) -

Gin omit or replace by another amino acid, e.g. Asn

Glu omit or replace by another amino acid, e.g. Asp

11. A therepeutically active agent according to any of Claims 6 to 10 wherein said related sequence differs from the sequence

CysArglleLysGlnPhelleAsnMetTrpGlnGlu by a single amino acid .

12. A therepeutically active agent according to any of Claims 6 to 11 wherein said related sequence is:

ArglleLysGlnPhelleAsnMetTrpGlnGlu

13. A therapeutic agent according to any preceding claim wherein the therapeutically active substance possesses reverse transcriptase inhibitor activity.

14. A therapeutic agent according to any of Claims 1 to 12 wherein the therapeutically active substance is an antimetabolite.

15. A therapeutic agent according to any preceding claim wherein the therapeutically active substance is a nucleotide

analogue.

16. A therapeutic agent according to Claim 15 wherein the therapeutically active substance is selected from 3'-deoxy-3'- azidothymidine, 2',3'-dideoxycytidine, 2',3'-dideoxyinosine, Acyclovir and 5-fluorouracil

17. A therapeutic agent according to any preceding claim wherein the therapeutically active substance is covalently bonded to the polypeptide of low molecular weight via a linker, the therapeutically active substance being bonded to the linker via an ester linkage.

18. A therapeutic agent according to any preceding claim wherein the therapeutically active substance in free form possesses a hydroxyl group and said substance is covalently bonded to the polypeptide of low molecular weight via a linker having the structure

-OC(R)CO- wherein the residue R is (i) a divalent substituted or unsubstituted hydrocarbyl group containing 1 to 20 carbon atoms (ii) a heterocyclic group containing 5 or six ring carbon atoms and at least one

heteroatom selected fron N, O and S or (ii) a direct bond linking the groups -OC and CO-.

19. A therapeutic agent according to Claim 18 wherein any substituents on R are selected fron- hydroxy, oxo, amino, carboxy, halo, trifluoromethyl and nitro.

20. A therapeutic agent according to Claim 18 wherein R represents a group of the formula -(CH₂)_n- wherein n is an integer from 1 to 4.

21. A therapeutic agent according to any preceding claim having the formula

[TES]—CO(R)CO-NH—[polypeptide]—COOH (II) wherein "polypeptide" denotes the polypeptide of low molecular weight (excluding its N-terminal amino group and its C-terminal carboxyl group), R is (i) a divalent substituted or unsubstituted hydrocarbyl group containing 1 to 20 carbon atoms (ii) a heterocyclic group containing 5 or six ring carbon atoms and at least one heteroatom selected fron N, O and S or (ii) a direct bond linking the groups -OC and CO- and TES represents the residue of therapeutically active substance.

22. A therapeutic agent according to Claim 21 wherein any substituents on R are selected from hydroxy, oxo, amino, carboxy, halo, trifluoromethyl and nitro.

23. A therapeutic agent according to Claim 21 wherein R represents a group of the formula -(CH₂)_n- wherein n is an integer from 1 to 4, preferably 2.

24. A therapeutic agent according to any of Claims 21 to

23 wherein the amino acid sequence of [polypeptide] is

25. A therapeutic agent according to Claim 24 wherein the related sequence has a homology with the sequence

26. A therapeutic agent according to Claim 24

wherein the related sequence has a homology with the sequence

27. A therapeutic agent according to Claim 24 wherein the related sequences have a homology with the sequence

Original Deviations

amino acid

Cys Met

Arg Lys His

Ile Ala Leu Val

Lys Arg His

Gin Asn Th

Phe Trp T

Ile Ala Leu Val

Asn Gin His Tyr

Met Cys

Trp -

Gin Asn

Glu Asp

28. A therapeutic agent according to Claim 24 wherein the related sequence hase a homology with the sequence

Original Preferred amino acid substitution

Cys omit or replace by another amino acid, ee..gg. Met

Arg omit or replace by another amino acid, e.g. Lys His Ile omit or replace by another amino acid, e.g. Ala Leu Val

Lys omit or replace by another amino acid. e.g. Arg His

Gin omit or replace by another amino acid, e.g. Asn Thr

Phe omit or replace by another amino acid. e.g. Trp Tyr Ile omit or replace by another amino acid, e.g. Ala Leu Val

Asn omit or replace by another amino acid. e.g. Gin His Tyr

Met omit or replace by another amino acid, e.g. Cys

Trp (preferable not omitted or replaced) -

Gin omit or replace by another amino acid, e.g. Asn

Glu omit or replace by another amino acid, e.g. Asp

29. A therepeutically active agent according to any of Claims 24 to 28 wherein said related sequence differs from the sequence CysArglleLysGlnPhelleAsnMetTrpGlnGlu by a single amino acid .

30. A therepeutically active agent according to any of Claims 24 to 29 wherein said related sequence is:

ArglleLysGlnPhelleAsnMetTrpGlnGlu

31. A therapeutic agent according to any of Claims 21

to 29 wherein [TES]- is a residue of a substance possesesing reverse transcriptase inhibitor activity.

32. A therapeutic agent according to any of Claims 21 to 31 wherein [TES]- is a residue of an antimetabolite.

33. A therapeutic agent according to any of Claims 21

to 31 wherein [TES]- is a residue of a nucleotide analogue.

34. A therapeutic agent according to Claim 33 wherein [TES]- is derived from 3'-deoxy-3'-azidothymidin-5'-0-yl,

2',3'-dideoxycytidin-5'-0-yl or 2',3'-dideoxyinosin-5'-0-yl.

35 . A compound of formula

HNCysArglleLysGlnPhelleAsnMetTrpGlnGlu.COOH

CO.CH₂.CH₂. COO

36. A compound of formula

[TES] 0C(R)C00H

wherein R and [TES]- are as defined in any preceding claim, or a salt or ester forming derivative thereof.

37. A compound of formula

HOOC.CH₂.CH₂.COO

38. A compound of formula