AU628034B2 - Immunoglobulin e competitor - Google Patents

Description

PCT

OPI DATE 14/06/89 AOJP DATE 20/07/89 APPLN. ID 28022 89 PCT NUMBER PCT/GB88/01018 INTERNATIONAL APPLICATION PUBLISHED UNDER THE PATENT COOPERATION TREATY (PCT) (51) International Patent Classification 4 (11) International Publication Number: WO 89/ 04834 C07K 7/00, 13/00, A61K 37/02 Al G01N 33/68 C12P 21/00 (43) International Publication Date: 1 June 1989 (01.06.89) (21) International Application Number: PCT/GB88/01018 (74) Agent: EYLES, Christopher, Thomas; Batchellor, Kirk Eyles, 2 Pear Tree Court, Farringdon Road, Lon- (22) International Filing Date: 18 November 1988 (18.11.88) don EC1R ODS (GB).

(31) Priority Application Number: 8727045 (81) Designated States: AT (European patent), AU, BE (European patent), CH (European patent), DE (Euro- (32) Priority Date: 19 November 1987 (19.11.87) pean patent), DK, FR (European patent), GB (European patent), IT (European patent), JP, LU (Euro- (33) Priority Country: GB pean patent), NL (European patent), SE (European 3 Reocrc i patent), US.

(71) Applicant (for all designated States except US): RE- SfA-R H-GGRPORA-T-ION-bIM.l-TEB [GB/GB]; 91 Published Waterloo Road, London SE1 8XP With international search report.

Before the expiration of the time limit for amending the (72) Inventors; and claims and to be republished in the event of the receipt Inventors/Applicants (for US only) GOULD, Hannah, of amendments.

Jane [GB/GB]; 48 Great Percy Street, London WCIL XQR HELM, Birgit, Anna [GB/GB]; 22 Alderton Hill, Loughton, Essex IG0I 3JB MARSH, Philip, John, Henry, Benedict [GB/GB]; 79a High Street, Deptford, London SE8 5AA (GB).

76 6~8Q3 (54) Title: IMMUNOGLOBULIN E COMPETITOR L H H S I I rSS- I-SS-, r'SSj r"SS, rSS- V ICHi -f cm |H4 rE2-4 t28-547) rE2-3 (218-4.39 rE4 t4uO- 477 r'-4 (301-5471 rE3-4 (340-5471 rE? /218-3361 rE2'-3' (M3-.376 rE2-3' (218 -3671 rE2-3B (218-361 IgE Oegil.gE rE2-4 -U rE2-3 rE3-4 Arg 506 Dimec Arg 506 Mon.

rE2-3/0 n- 0 Y I (57) Abstract 1 10 10 1000 1000o LIGAND CONCEO RA7TON (nM) A polypeptide competitor or analogue for human Immunoglobulin E (IgE) low affinity sites comprises a polypeptide which has a sequence of amino acid which has a sequence of amino acids which is shown in Table I. This amino acid sequence corresponds to amino acids 340 to 439 of the epsilon heavy chain of IgE. A particularly preferred polypeptide competitor has a sequence of amino acids corresponding to amino acids 340 to 547 of the epsilon heavy chain of IgE as set out in Table V herein, which also shows the corresponding DNA sequence coding therefor. Such a polypeptide may also include additional short sequences at the beginning and/or end of the core sequence which are physiologically harmless and do not contribute to the ability of the core sequence to compete with native IgE for the low affinity receptor sites on human cells. The polypeptide may be produced synthetically or by expression from Escherichia coli containing a plasmid having a DNA segment coding for the polypeptide.

W 89/04834 PCT/GB88/01018 1 IMMUNOGLOBULIN E COMPETITOR This invention relates to a competitor or analogue for human Immunoglobulin E (IgE). More particularly, the invention is concerned with a polypeptide which competes with IgE for low-affinity receptor sites.

Our co-pending International Patent Application No. PCT/GB87/00466 (International Patent Publication No.

W088/00204 published 14th January 1988) describes and claims a polypeptide competitor for IgE which has a monomeric chain of seventy-six amino acids from the second and third domains of the human IgE sequence, corresponding to amino acid residues 291 to 366 of the full IgE heavy chain sequence reported by Bennich (Progress in Immunology II, Vol.I, July 1974, pp 49-58 and Int, Arch. Allergy App. Immunol. 53, 459). This polypeptide binds to high-affinity Fc receptors for IgE (FcER 1 which exist particularly on mast cells and basophils, thereby inhibiting the biological responses, such as exocytosis or degranulation, which take place when antigen specific IgE binds to and crosslinks such receptor sites in the presence of antigen.

In the human immune system, the principal role of IgE is believed to be to provide immunity to parasites. It also, however, mediates Type I hypersensitivity which is an allergic response leading to the manifestation of such symptoms as hay fever and asthma. Briefly, the mechanism of the allergic response is that on encountering a normally innocuous antigen such as pollen, synthesis of antigenspecific IgE by B-cells is initiated. The antigen-specific IgE then binds to mast cell receptor sites via its Fc region and thereafter any further encounter with the antigen triggers degranulation of the mast cells releasing mediators, principally histamine, resulting in the acute inflammatory symptoms typical of Type I hypersensitivity.

Structurally, IgE, in common with the other

ALI'

4 1 i -2immunoglobulins, comprises two heavy and two light chains, the epsilon heavy chain having five domains, a variable domain VH and constant domains CH1 to CH4. The molecular weight of IgE is in the region of 188,000 of which the heavy chain accounts for about 72,500, representing a sequence of approximately 550 amino acid residues.

It has been reported (Nature, vol. 315, 1985, No. 6020, pp 577-578) that a peptide sequence of 330 amino-acids corresponding to amino acid residues 218 to 547 (in accordance with the numbering given by Bennich, Progress in Immunology II, Vol. I, July 1974, pp 49-58) of the epsilon heavy chain of IgE has an inhibitory effect on the release of mediators from human mast cells. The numbering is erroneously assigned in that paper in Nature; the more correct numbering would be 207 to 537. The 330 amino-acid sequence exists as a dimer consisting of two chains of amino-acids, each of 330 amino- S acids in length, linked by disulphide bonds.

United States Patents Nos. 4 171 299 and 4 161 522 disclose that an oligopeptide containing from three to ten amino acids in a sequence selected from a portion of amino acids 255 to 527 of the Bennich nomenclature (see reference above) of the Fc region of human IgE will block Fc receptors of mast cells thus inhibiting degranulation and release of S mediators such as histamine. The most active of these oligopeptides is identified as the pentapeptide Asp-Ser-Asp- Pro-Arg (called HEPP: Human Immunoglobulin E Polypeptide) derived from the amino acid sequence 320 to 324 of the IgE heavy chain. In native IgE amino acid 322 is asparagine, but it is suggested in these Patents that substitution of asparagine by aspartic acid leads to substantial enhancement of the blocking activity.

In the Patents mentioned above the full sequence which is attributed to Bennich (Progress in Immunology II, Vol. I, July, 1974, pp 49-58) is quoted and shows aspartic acid at location 322. However, Bennich himself ie>ater asserts (Int. Arch. Allergy Appl. Immunol. 53, 459) that June 23, 1992 1 'WO 89/04834 PCT/GB88/01018 i asparagine resides at that location. Bennich also reports that neither of the peptides Asp-Ser-Asp-Pro-Arg nor Asp- Ser-Asn-Pro-Arg has any blocking activity. Determination of the gene sequence has shown that amino acid 322 is asparagine and not aspartic acid. In European Patent publication No. 102634 asparagine and not aspartic acid is correctly quoted at the equivalent location.

Further, it is also reported that the specific activity of HEPP is low requiring excessively large doses for any significant physiological effect.

It is known that IgE epsilon chain fragments may be synthesised in Escherichia coli by cloning and expression of the DNA sequences coding for the appropriate domains of the IgE chain (Eur. J. Immunol. 1985, 15: 966-969 and Proc.

Natl. Acad. Sc. USA, vol.81, 1984, 2955-2959).

Native IgE binds also to low-affinity receptors

*A

(FcER 2 on cells such as lymphocytes, eosinophils, monocytes and platelets to activate the effector functions of these cells, for example, IgE-dependent parasite-killing by eosinophils. FcER 2 receptors on B lymphocytes has also been implicated in the transduction of growth signals and in B cell activation. FcER 1 and FcER 2 are structurally unrelated proteins and can be expected to recognise different sites on the Fc region of IgE.

An object of the present invention is to provide a polypeptide which binds to the type 2 low-affinity Fc receptors (FcER 2 According to the present invention there is provided a competitor for Immunoglobulin E low affinity receptor sites comprising a polypeptide having the sequence of amino acid residues 330 to 429 of the Bennich enumeration, shown in Table I below.

i lv: -4- TABLE I 330 A-Leu-Ser-Arg-Pro-Ser-Pro-Phe-Asp-Leu- 339 Phe-Ile-Arg-Lys-Ser-Pro-Thr-I le-Thr-Cys- 349 Leu-Val-Val--Asp-Leu-Ala-Pro-Ser-Lys-G ly- 359 Thr-Val-Asn-Leu-Thr-Trp-Ser-Arg-Ala-Ser- 369 Gly-Lys-Pro--Val-Asn-His--Ser-Thr-Arg-Lys- 379 Glu-Glu-Lys-Gln-Arg-Asn-Gly-Thr-Leu-Thr- *:389 Val-Thr-Ser-Thr-Leu-Pro-Val-Gly-Thr-Arg- 0 399Asp-Trp-Ile-Glu-Gly-Glu-Thr-Tyr-Gln-Cys- :409 Arg-Val-Thr-His-Pro-His-Leu-Pro-Arg-Ala- :419 Leu-Met-Arg-Ser-Thr-Thr-Lys-Thr-Ser-Gly- 429 Pro-B.

acdThe group A in Table Ilis a hydrogen atom or an amino aci sequence which may include a chain initiating amino acid sequence and which includes at most a part only of the natural sequence of amino acid residues of the heavy chain of Immunoglobulin E, said part of said natural sequence forming a part only of the sequence of amino acid residues 291 to 329 of the Bennich enumeration and B is the sequence of amino acid residues numbered 430 to 537 according to the Bennich enumeration.

A may be an amino acid sequence which corresponds tc a ~LIAApart only of the natural amino acid sequence of the heavy c~ ,~chain of Immunoglobulin E extending from residue 329 at the mwspe#3867 mwspe3867June 23, 1992

I

C-terminal end of the group A back up to and including residue 291 using the Bennich enumeration. A may be or include an inert polypeptide sequence, preferably an inert oligopeptide sequence, that does not interfere with the ability of the polypeptide to compete for the Immunoglobulin E low affinity receptor sites.

B corresponds to a part of the IgE heavy chain sequence commencing at residue 430 extending as far as residue 537, using the Bennich enumeration. It will usually be preferred that, when A represents or includes a chain initiating amino acid sequence, such a sequence contains no more than about amino acid residues, and normally no more than about 10 such residues, e.g. 4 or In one preferred form of polypeptide competitor according S* to the invention, A represents a hydrogen atom or a chain initiating amino acid sequence X, such as Met-Asp-Pro-Arg- S(the corresponding nucleotide sequence for which is ATG GAT CCG CGC) and B represents the sequence of amino acid residues 430 to 537 of the Bennich enumeration set out in Table II below.

June 23, 1992 WO 89/04834 PCT/GB88/01018 -6- TABLE II 430 -Argn-'Ala-Ala-Pro-Glu-Val1-Tyr-Ala- Phe- CGT GCT GCC GGG GAA GTC TAT GCG TTT 439 Ala-Thr-Pro-Glu-Trp-Pro-Gly-Ser-Arg -Asp- GCG AGG CGG GAG TGG CCG GGG AGC CGG GAC .449 Lys-Arg-Thr-Leu-Ala-Cys-Leu-I1e'GflnASfl AAG CGC ACC CTC GGC TGC CTG ATC GAG AAC 459 Phe-Met-Pro-Glu-Asp-I le- Ser-Val-Gln-Trp- TTC ATG CGT GAG GAG ATC TGG GTG GAG TGG 469 Leu-H is -As n-Gl u-Val- G n-Leu- Pro- sp-Ala- GTG GAG AAG GAG GTG GAG GTG GCC GAG GCC *479 Arg-Hi s- Ser-Thr- Thr- Gln -Pro-Arg- Lys-Thr- GGC CAC AGG AGG ACG GAG GGG GGC AAG AC 489 Lys-Gln-Ser-Gly-Phe-Phe-Val-Phe-Ser-Arg- AAG GCG TGG GGG TTC TTC GTG TTC AGC CGC 49 Leu-Gln-Va1-Thv-Arg-Ala-Glu-Trp-Glu-Glfl- GTG GAG GTC ACC AGG GGC GAA TGG GAG GAG 509 Lys-Asp-Glu-Phe-Ile-Cys-Arg-Ala-Val-His- AAA GAT GAG TTC ATG TGC CGT GGA GTG CAT 519 Glu-Ala-Ala-Ser-Pro-Ser-Gln-Thr-Val-Glfl- GAG GGA GGG AGC GGG TGA GAG ACC GTC GAG 529 Arg-Ala-Va-Ser-Val-Asn-Pro-Gly-Lys [Stop] GGA GCG GTG TGT GTA AAT GGG GGT AAA TGA.

The invention further encompasses a polypeptide competitor as set out in Table I herein wherein A represents

CO

ILIE

T~EWUDL~ U1~! 77"

I

I

-7the sequence of amino acid residues 291 to 329 of the Bennich enumeration set out in table III below.

TABLE III 291 Gln-Lys-His-Trp-Leu-Ser-Asp-Arg-Thr-Tyr- 301 Thr-Cys-Gln-Val-Thr-Tyr-Gln-Gly-His-Thr- 311 Phe-Glu-Asp--Ser-Thr-Lys-Lys-Cys-Ala-Asp- 321 Ser-Asn-Pro-Arg- ly-Val-Ser-Ala-Tyr-.

S

S

*5

S

S

4 C C A 4

SC

we

S

S.

4 S. 4 p

S.

S.

4 9e A S S 5 1 S S

IS

S S IS 5 5* a S 55 June 23, 1992 -8- The invention provides in particular a competitor for Immunoglobulin E low affinity receptor sites comprising a polypeptide having the sequence of amino acid residues 330 to 537 of the Bennich enumeration, shown in Table IV below.

Further the invention provides a DNA having the nucleotide sequence also shown in Table IV below.

TABLE IV 330 X-Leu-Ser-Arg-Pro-Ser-Pro-Phe-Asp-Leu- 339 CTA AGC CGG CCC AGC CCG TTC GAC CTG 339 Phe-I le-Arg-Lys-Ser-Pro-Thr-Ile-Thr-Cys- TTC ATC CGC MAG TCG CCC ACG ATC ACC TGT 349 Leu-Val -Val -A sp-Leu-A la-P ro-Ser -Lys -G ly- CTG GTC GTC GAC CTG GCA CCC AGC MAG GGG :359 Thr-Val-Asn-Leu-Thr-Trp-Ser-Arg-Ala-Ser- ACC GTG AAC CTG ACC TGG TCC CGC GCC AGT 369 Gly-Lys-Pro-Val-Asn-H is-Ser-Thr-Arg-Lys- GCG AAG CTT GTG AAC CAC TCC ACC AGA MAG June 23, 1992 WO 89/04834 PCT/GB88/01018 -9- Table TV continued 379 Glu-Glu-Lyrs-Gln-Arg-Asni-Gly-Thr-Leu-Thr- GAG GAG AAG CAG COC AAT GGC ACG TTA ACC 389 Val-Thr--Ser-Thr-Leu-Pro-Val-Gly-Thr-Arg- GTC ACG TCC ACC CTG CGG GTG GGC ACC CGA 399 Asp-Trp-Ile-Glu--Gly-Glu-Thr-Tyr-Gln-Cys- GAC TGG ATC GAG GGG GAG ACC TAG GAG TGC 409 Arg-Val-Thr-His-Pro-His-Leu-Pro-Arg-Ala- AGG GTG ACC CAC CCC CAC CTG CCC AGG GCC 419 Leu-Met-Arg-Ser-Thr-Thr-Lys-Thr-Ser-Gly- GTC ATG CGG TCC ACG ACC AAG ACC AGC GCC 429 Pro-Arg-Ala-Ala-Pro-Glu-Val-Tyr-Ala-Phe- CCG CGT GCT GCC CCG GAA GTC TAT GCG TTT 439 Ala-Thr-Pro-Glu-Trp-Pro-Gly-Ser-Arg-Asp- GCG AGG CCG GAG TGG GCG GGG AGC GGG GAC 449 Lys-Arg-Thr-Leu-Ala-Gys-Leu- -Gln-Asn- AGG CGC ACC CTC GCC TGC CTG ATC CAG AAC 459 Phe-Met-Pro-Glu-Asp-Ile-Ser-Val-Gln-Trp- TTC ATG CGT GAG GAG ATC TCG GTC GAG TGG 469 Leu-His-Asn-Glu-Val-Gln-Leu-Pro-Asp-Ala- CTG GAG AAC GAG GTG GAG GTG GGC GAG GCC 479 Arg-His-Ser-Thr-Thr-Gln-Pro-Arg-Lys-Thr- GGC GAG AGC ACG AGG GAG CGG CGC AAG ACG rA-L 0 10 Table I. continued 489 Lys-Gln-Ser-Gly-Phe-Phe-Val-Phe-Ser-Arg- AAG GCC TCC GGC TTC 'iTC GTC TTC AGC CGC 499 Leu-Gln-Val-Thr-Arg-Ala-Glu-Trp-Glu-Gln- CTG GAG GTC ACC AGG GCC GAA TGG GAG CAG 509 Lys-Asp-Glu-Phe-Ile-Cys-Arg-Ala-Val-His- AAA GAT GAG TTC ATC TGC CGT GCA GTG CAT 519 Glu-Ala--Ala-Ser-Pro-Ser-Gln-Thr-Val-Gln- GAG GCA GCG AGC CCC TCA CAG ACC GTC CAG Je 529 Arg-Ala-Val-Ser-Val-Asn-Pro-Gly-Lys-S top) S 5 CGA GCG GTG TCT GTA AAT CCC GGT AAA TGA The group X in Table IV is a hydrogen atom or a chain initiating amino acid sequence, and, in the specific method of preparation which will be described below the group X is Met- Asp-Pro-Arg-, the corresponding nucleotide sequence being ATG GAT CCG CGC.

The invention further extends to polypeptide sequences of the types set out in Tables I and IV above (as well as to those set out in Table I as modified by Table II or III terminated at one or both of the N- and C- terminal ends by respective inert oligopeptide sequences initiating and/or terminating the chain).

Dimeric forms of the polypeptide competitors of the invention are of particular interest.

Also included within the scope of the invention are fragments of the polypeptides of the invention, including dimeric forms thereof, having competitive properties which are the same as or similar to those of the sequences defined above in Tables I and IV or in Table I as modified by Table II or ie III.

June 23, 1992 WO 8904834 PCT/GB88/01018 11 The invention also provides a host/vector system containing the nucleotide sequence which encodes the polypeptide of sequence 330 to 537.

The expression vector may conveniently be Escherichia coli N4830, and a culture of same harbouring a plasmid (designated as pE3-4) which encodes the peptide of sequence 330 to 537 has been deposited with the National Collection of Type Cultures in London on 18th November 1987 under the Accession Number NCTC 12162.

The invention also provides a method of preparing the polypeptide 330 to 537 comprising culturing the said host organism and isolating the peptide from the culture.

This invention also includes a pharmaceutical preparation in which the active principle is a polypeptide competitor for Immunoglobulin E low affinity receptor sites of the type defined above.

The preparation may also include a pharmaceutical carrier permitting administration of the polypeptide competitor in an appropriate manner, for example S intranasally.

The polypeptide competitor of the invention may also be covalently linked to or associated with other therapeutic or diagnostic agents or other molecules with the effect that the polypeptide acts to target the therapeutic or diagnostic agent to cells bearing IgE low affinity receptors.

The invention further contemplates use of the polypeptide competitors of the invention, their dimeric forms and biologically active fragments thereof in a binding assay and for inclusion in a diagnostic kit.

Thus, it has now been discovered that the polypeptide 340-547 of this invention has the ability to bind to the low affinity IgE receptors on cells. An advantage accruing from the invention is that the ability to distinguish between the high and low affinity binding sites q j 7 WO 89/04834 PCT/GB88/01018 -12-

WO:

I

ii *r of IgE enables separate administration of these effector molecules as may be medically indicated. IgE is associated with allergic reaction known as Type I immediate hypersensitivity and its value as an antagonist is discussed in our co-pending patent application identified above.

The reaction of IgE with mast cells not only leads to immediate hypersensitivity, but is directly responsible for setting up the conditions for delayed hypersensitivity, which is initiated by the release of chemical compounds from mast cells in the initial immune response. Delayed hypersensitivity involves an array of other cell types, such as platelets, macrophages and eosinophils, which are attracted to and cause an inflammatory response at the site of foreign body or initial tissue insult.

When these cells bind via the low affinity receptors to IgE and antigen, cytotoxic oxygen radicals are released which destroy cells in the local vicinity, including foreign cells if present. Thus, the polypeptide 340-547 may be considered for use as an antagonist of IgE to block or diminish the delayed hypersensitivity reaction associated with the natural IgE binding.

The invention will now be described by way of example.

Figure 1 shows the covalent structure of the human epsilon chain of IgE, indicating the positions of the intrachain disulphide bonds and inter-chain disulphide (S- S, and the boundaries of the five structural domains (VH, CH1-4), corresponding to exons in the genomic DNA, Latween the N-terminus (left) to the C-terminus (right).The recombinant peptides are described as follows: rE2-4 contains Asp 207-Lys 537 (Ce2-4 with seven amino acids from CH1); rE2-3 contains Asp 207-Pro429 (CH2 and CH3 with seven amino acids from CH1 and one from CH4); rE4 contains Arg 430-Lys 537'; rE2'-4 contains Gln 291-Lys 537 (the C-terminal

L

rT-; ;-la rr- WO 89/04834 PCT/GB88/01018 -13end of CH2 from the indicated position between the two cysteines that form the inter-chain disulphide bond, CH3 and CH4); rE3-4 contains Leu 330-Lys 537 (all but nine residues at the N-terminus of CH3 and CH4); rE2 contains Asp 207-Val 326 {CH2 with seven amino acids from CH1 and six from CH3); rE2'-3' contains Gin 291-Arg 366 (the C-terminal part of CH2 and the N-terminal part of CH3 from the indicated positions between the cysteines that form intra-chain disulphide bonds in these domains).

Method The genetic constructs used for expression in E.

coli to yield the above peptides are described as follows: plasmids pE2-4 encoding rE2-4 and plasmid pE4 encoding rE4 contain epsilon cDNA fragments bounded by Hind III sites and direct the synthesis of short N-terminal fusions containing the first seven amino acids of the Trp E sequence. The epsilon sequence in pE4 was inserted after addition of an 8- S mer Hind III linker (New England Biolabs) to codon 429, the position of an FnuD II site. Plasmid pE2 is a derivative of pE2-4 modified by a 12-mer Nhe I linker (New England Biolabs), which provides translation stop signals. This linker was ligated only to codon 327 in the epsilon sequence S after cleavage of the cloned cDNA by Hae II and removal of the 3' overhang with T4 polymerase. Plasmid pE2-3 comprises an epsilon cDNA fragment subcloned into ptac-85 and directs S. the expression of a non-fusion polypeptide. A translation S* termination signal Biochemicals) was placed immediately downstream of the CH3 sequence by ligating a synthetic oligonucleotide to codon 429, after cleavage of the epsilon sequence at that position with FnuD II. Plasmid pE2'-4 encoding rE2'-4' contains an epsilon gene fragment, which, after cleavage with Sac I (cutting at codon 287) and mild exonuclease treatment with Bal 31, was inserted into the expression vector pAS1 The plasmid pE2'- 3' encoding rE2'-3' was made by insertion of a translation flY~ *rral~lF~~.a" IL-C .UYSLtL i WO 99/04834 pCT/GB88/01018 WO 14 terminator into pE2'-4 at the Sma I restriction site located at codon 366.

The rE3-4 sequence was synthesised in E. coli under control of the lambda PL promoter in pE3-4, a recombinant derivative of pASI (Rosenberg et.al., 1983, Meths. Enzymol. 101, 123-138). The gene fragment in pE3-4 was tailored for insertion into the BamHI site of pASI after cleavage of the epsilon DNA with HaeII, mild exonuclease treatment and addition of a 12-mer BamHI linker. The DNA sequence around the BamHI site was determined; from this the deduced amino acid sequence showed that the expressed epsilon sequence started at codon 330. Additional residues introduced in the manipulation are: ATG GAT CCG CGC CTA Met-Asp-Pro-Arg Leu- (residue 330 of epsilon sequence) In E. coli N4830 the epsilon gene fragment in pE3- 4 can be expressed by heat induction. To achieve this, growth of the strain to A600=0.8 at 30*C (non-inducing) is followed by incubation at 42 0 C (inducing).

The site on human IgE involved in the binding to the low affinity (FcER 2 sites on the FcER 2 positive human B cell line RPMI 8866 was probed using cloned fragments of human IgE generated in E. coli from the deletion mutants of the gene encoding the CH2, CH3 and CH4 domains of the i: epsilon (ND) heavy chain. Preliminary indirect and direct immunofluorescence binding assays were carried out on the various fragments and the results are given in Table V below which is an indication of the ability of each peptide to bind to the high affinity receptor.

.4~I 0 d W890834 PCT/GB88/01018 15 TABLE V Summary of binding of recombinant IgE peptides to human high-affinity (FcR 1 and low-affinity (FcR 2 receptors.

Peptide Amino Heavy Chain Domains Binding Acids VH1 CH1 CH2 CH3 CH4 FcERI FcER2 IgE 1- 537 rE2-4 207-537 rE2'-4 291-537 part rE2-3 207-429 rE2'-3' 291-366 part part rE4 430-537 *e rE2 207-326 rE3-4 330-537 These results indicate that the binding of human IgE to FcER 2 requires the presence of the third and fourth domains of the heavy chain, as given by the polypeptide of the present invention and that, of our peptides tested, only the peptide rE3-4 binds to the low affinity receptors to the exclusion of high affinity binding.

Further indirect immunofluorescence studies confirmed that the myeloma protein IgE rE2-4, rE2'-4 and rE3-4 bound to 90% of FcER 2 positive RPMI 8866 B cells. However, this further work showed that, by contrast, there was no detectable binding of rE2-3, rE2'-3', rE4 and rE2. The binding of the rE peptides was specific, since it could be inhibited by preincubation of the cells with two anti-FcER 2 monoclonal antibodies (mAb 135 and anti-BLAST-2), but not by control IgGl of unrelated specificity. Moreover, ,qALI wI top WO 89/04834 PCT/GB88/01018 W( 16 it was found that there was no binding to the FcER 2 negative cell lines Jurkat and Raji.

Binding ability of rE peptides RPM1 8866 cells were incubated with different concentrations of rE peptides and indirect immunofluorescence was performed.

For indirect immunofluorescence the results of which are summarised in Table VI 0.5 x 106 RPMI 8866 cells FcER 2 positive) in staining buffer (RPMI 1640 fetal bovine serum, containing 0.01% azide) were incubated with various concentrations (0.1-200 pg/ml) of purified rE fragments or native IgE(PS) for 40 minutes at 4°C. After washing, the cells were incubated for 30 minutes at 4 0 C with the appropriate fluorescein isothiocyanate (FITC)-conjugated anti-Fc mAB or with an affinity-purified goat anti-human IgE antibody (10 pg/ml). After extensive washing, the percentage of cells binding IgE or the rE fragments was oo ~evaluated by a FACScan (Becton Dickinson, Mountain View, o CA). Table VI shows the absolute percentage of positive cells (mean of results obtained in 9 experiments).

:TABLE

VI

Binding of recombinant IgE peptides to RPMI 8866 B cells by indirect immunofluorescence Ligand Amino acid sequence positive cells IgE(PS) Glp 1-Lys 537 91 2% rE2-4 Asp207 -Lys 537 92 2% rE2'-4 Gln291 -Lys 537 91 2% rE3-4 Leu330 -Lys 537 90 3% rE2-3 Asp207 -Pro 429 3 1% rE2'-3' Gln 2 91 -Arg 366 2 1% rE2 Asp 2 0 7 -Val 326 2 1% rE4 Arg 4 3 0 -Lys 537 3 2% sffsWr~a(d~a UI uiiI- ~RuuUL CiI! 1 I~ r, f* WO 89/04834 PCT/GB88/01018 S- 17 The data of Table VI indicate that the FcER 2 binding site is contained in the rE3-4 peptide (Leu 330- Lys 537), and does not require the CH2 domain.

For the tests using anti-FcER 2 monoclonal antibodies, 0.5 x 10 6 RPMI 8866 cells in staining buffer were preincubated with medium, anti-FcER 2 mAbs (mAb 135 or anti-BLAST-2 mAb), 1 pg/ml, or a control IgGl murine mAb (anti-HLA-DP, 2.5 pg/ml) for 60 minutes at 4 0 C. After washing, the cells were incubated for 40 minutes at 4 0 C with IgE(PS) (1 pg/ml) or rE3-4 (200 pg/ml), followed by mAb RPI- FITC. Table VII shows the absolute percentage of positive cells (mean S.D. of results obtained in 3 experiments).

TABLE VII Anti-FcER 2 /CD23 mAbs inhibit the binding of rE3-4 to RPMI 8866 cells

S

S

S.

Inhibitor Ligand IgE rE3-4 Nil 88+2 89+3 mAb 135 2+1 5+2 anti-BLAST-2 3+1 4+2 control mouse IgGI 74+2 76+5 In these experiments the mAb 135 used was as described by E. Rector et al., Immunology, 55, 481-487 (1985), while the anti-BLAST-2 mAb was as described by C.

Kintner et al, Nature, 294, 458-460 (1981).

To compare the relative activities of different epsilon-chain fragments and mutant sequences, binding has been measured as a function of peptide concentration.

Binding profiles are shown in Figure 2 and the results are summarised in Table VIII rE2'-3/G3 is a chimaeric recombinant which was constructed using mouse gamma-2b cDNA from pHG201, described by S. Roberts et al, Protein ls-r: EIIBa~e~dY ~111~" j 1~C~i. IP~ LI~ Lly ~.UYB~IYYL 18 Engineering, 1, 59-65 (1986). The mouse CH3 sequence was derived by partial SacI digestion of pHG201 to yield a DNA fragment encoding amino acids 336-437 (EU index). This fragment was inserted downstream of codon 429 in pE2'-4 replacing the C epsilon-4 sequence with the Cgamma-2b CH3 domain. The recombinant peptide reacted with both epsilon and gamma-2b antisera.

TABLE VIII Binding ability of rE peptides Ligand Molarity required for 50% binding to B cells IgE 9.0 x 10- 1 0 Deglycosylated IgE 4.8 x 10-10 rE2-4 7.2 x 10- 1 0 rE2'-4 2.6 x 10 9 rE3-4 5.8 x rE2-4 Arg 496 (dimer) 1.6 x 10- 1 0 rE2'-3/G3 4.7 x 10 8 99 For deglycosylation, IgE (PS) (200 pg/ml) in 0.55 M sodium phosphate, pH 8.6, was incubated with N-glycosydase F (N-Glycanase, Genzyme Corporation, Boston MA), 15 U/ml, at S: 37"C overnight, and subsequently absorbed with an excess of Lentil Lectin Sepharose 4 B (Sigma Chemical Company, St.

Louis, MO) at 4°C on rotation (Sepharose is a Trade Mark).

Analysis of the preparation by SDS-PAGE and perioddic acid- Schiff 9pas) staining of the gel showed that complete deglycosylation of IgE had occurred. IgE concentration in the deglycosylated sample was assessed by radioimmunoassay.

The point substitutions in rE2-4 were carried out by oligonucleotide-directed site sepcific mutagenesis using double-stranded plasmid DNA in accordance with the procedures described by Y. Morinaga et al. Biotechnology, July, 636-639 (1984).

mwspe#3867 June 23, 1992

KW

WO 89/0434 PCT/GB88/01018 19 The recombinant Fc(rE2-4) is highly active, indeed perceptibly more so than myeloma IgE The elevated affinity is presumably due to the absence of carbohydrate, since our results showed that enzymatic deglycosylation of IgE (PS) increased its activity. rE3-4, which lacks CH2 and nine amino acids from CH3, displayed a much lower activity than rE2-4. A truncated peptide, rE2'-4, retaining the Cterminal thirty amino acids of CH2, by contrast, was almost as active as the full Fc sequence. For physiological binding affinity, therefore, all three Fc domains appear to be necessary.

The structural elements of Fc necessary for binding were defined. The Fc regions of all classes of antibodies are thought to have a number of common features.

In particular, they contain two heavy chains, non-covalently linked in their C-terminal domains CH4 or C-gamma 3), go separated by carbohydrate in the middle domain (CH3 or C- 09** gamma and covalently linked by one or more disulphide bonds in the N-terminal domain (CH2 in IgE, corresponding to the hinge region of IgG). The inter-chain linkages in CH2 and CH4 thus generate a tertiary structure, which may be required for receptor recognition. To determine whether this defines the active state, the activity of monomeric chains has been examined.

All rE chain fragments that contain the CH4 domain can form dimers in solution. All three Fc domains are found to be necessary, however, for the formation of disulphide- I' L linked dimers, suggesting that the non-covalent association in CH4 is required to place the thiols in register in CH2, so thatdisulphide bond formation can occur. If the association of two CH4 domains nucleates dimer formation, it is conjectured that the formation of dimers could be inhibited by preventing this association. The IgE model suggests that the two Phe 496 residues in CH4 are in van der Waals contact. Thus the replacement of Phe 496 by a charged *i I residue should oppose the formation of dimers. It has been found that the mutant gene expression product, rE2-4 (Arg 496), remains monomeric, as judged by gel electrophoresis.

After elimination of a trace of dimer by HPLC, we found the product to be completely inactive in binding to B cells. Strikingly, the dimer, recovered in a separate fraction, was more active than the unmutated Fc. Phe 496 is unlikely to form part of the recognition sequence, since it is buried in the CH4 domain. It is concluded that the dimerisation of chains is essential for FcER 2 binding.

Figure 2 illustrates the results obtained upon polyacrylamide gel electrophoresis of affinity-purified rE peptides. In Figure 3 A designates the results obtained with peptides analysed in a 20% SDS-urea gel under nonreducing conditions, while B shows the results obtained with peptides analysed in a 15% SDS-urea gel under reducing conditions. The order of sample application was: lane (1) standard proteins, lane rE2-4, lane rE2-4 (Arg 496 lane rE2'-3/C3. The peptides were isolated from E coli and affinity-purified.

Since CH4 is required for dimerisation and S dimerisation for activity, the possibility that part of the binding site lies in CH4 could not be excluded by simply deleting this domain. The monomeric state of rE2-3, for example, is sufficient to account for its inactivity in the binding assay summarised in Table VI above. To define the role of CH4 sequences in FcER 2 binding, an investigation was made of the behaviour of the chimaeric immunoglobulin fragment, rE2'-3/G3, in which the C-terminal domain of rE2'- 4 is replaced by that of a mouse gamma-2b chain. As expected from the homology between IgE and IgG, the chimaeric chains form disulphide-linked dimers in high yield, as can be seen from Figure 3 which shows the results obtained with rE peptides analysed in a 20% SDS-urea gel under non- -educing and reducing conditions. The dimers displayed a level of activity within an order of magnitude of that of their non-chimaeric counterpart as can be appreciated by June 23, 1992 ~i i -i--rr_ inspection of Figure 2 which illustrates the results obtained upon polyacrylamide gel electrophoresis of affinity purified rE peptides. Since it was found also that mouse IgG2b did not bind to the B cells, it is concluded that the FcER 2 -specific binding site is located in CH3.

To map the FcER 2 binding site on the CH3 domain more precisely, there were used monoclonal antibodies (mAbs) against epitopes in the Fc region and their efficiency in inhibiting the binding of '"I-labelled IgE to B cells was measured. The various epitopes recognised by the antibodies were mapped to the rE peptides by Western blotting, and the locations of peptides in the sequence were related to their positions in the model. The results are summarised in Table 0 TABLE IX Anti-human Fc mAbs: specificity and inhibition of 125 I-IgE binding to FcER 2 mAb epitope location inhibition of (amino acids) 125 I-IgE binding DC CH2 (Asn 207-Gln 241) none AS 7.12 CH2 (Asn 207-Gln 241) none BS 17 CH2 (Gln 291-Thr 305) 78 RP 3 CH2 (Asp 297-Thr 305) 77 IC 272 CH2 (Thr 305-Val 326) 22 .Le 27 CH3 (Leu 3 3 0-Val 351) none RP 1 CH3 (Leu 3 3 0 -Val 351) none IC 27 CH3 (Lys 357 -Val 360) 88 IC 173 CH4 (Arg 430-Lys 537) none The specificity of the anti-Fc mAbs was deduced by their pattern of reactivity with the rE fragments, rE fragments containing amino acid residues Asp 207-Thr 305 @and a synthetic peptide comprising amino acid residues Asp 97-Val 360 in enzyme-linked immunosorbent assay, Western blot Swspe#3867 June 23, 1992 4; 1Jn 3 WO 89/04834 PCT/GB88/01018 22 and dot immunoassay. Purified human myeloma IgE(PS) was iodinated by the chloramine-T method (specific activity: 8000 cpm/ng). To test the ability of the anti-Fc mAbs to inhibit the binding of IgE to FcER 2 125I-IgE (15 ng in pl) in PBS-0.5% BSA was mixed with a 10, 100, 1000 M excess of anti-Fc mAbs for 1 hour at 37 0 C, and then added to 1 x 106 RPMI 8866 cells in 0.1 ml. After incubation for 2 hours at 4°C, the cells were spun through serum and the cell-bound radioactivity was counted. Maximal binding was determined by incubating the cells with 1 25 I-IgE in the presence of medium alone. Table IX shows the inhibition obtained with a M excess of mAb.

Three of the nine monoclonal antibodies strongly inhibited the IgE-receptor interaction. The epitopes for mAb BS 17 and RP 3 lie in the C-terminal region of CH2 (between Gln 291 and Thr 305) and that for the third, IC 27, is at the N-terminal end of CH3 (comprising Lys 357-Val 360). The three sites are located within or near to the cleft between the CH2 and CH3 domains in the model of Fc. A fourth mAb, IC 272, which binds in an intervening sequence (Thr 305-Val 326) was only weakly inhibitory. It is envisaged that this mAb binds to the loop separating the two S beta-strands that line the cleft between the CH2 and CH3 domains on the CH2 side. This would afford an explanation for its rather weak inhibitory effect. Two other mAbs, RP 1 and Le 27, which bind within a second intervening sequence (Leu 330 Val 351), had no inhibitory effect. The whole of this peptide segment points away from the cleft. Three other mAbs, DC, AS 7.12 and IC 173, which bind to epitopes outside the CH2-CH3 junction region, in CH2 and CH4, did not inhibit receptor binding. Taken together with the results of the fragment binding assay, these data suggest that FcER 2 binds in the vicinity of Asp K~-Val 360 Since CH2 does not contribute sequences to the binding site, inhibition by anti-CH2 mAbs can result only from steric hindrance, /WO 1 WO 89f4834 PCT/GB88/01018 23 The results discussed above show that the FcER 2 binding site on human IgE lies in the N-terminal region of CH3, close to CH2 in the 3D structure of Fc. However, the CH4 and CH2 domains determine the level of activity through their effects on the structure of CH3. In particular, it is found that both CH3 domains are required for activity; CH2 and CH4 serve only to generate a dimer. The native structure in CH2 does not appear to be important per se, since the truncated CH2 domain in rE2'-4 suffices for nearly full activity, and rE3-4 is also active. That in CH4 may also be dispensed with, once covalent bonds are formed, since the substitution of Phe 496 by arginine did not impair activity in the dimer.

The.observation that the chimaeric peptide, rE2'- 3/G3, displayed significantly less than full activity in binding to the B cell FcER 2 requires comment. It is suggested that the C-terminal domain in native IgE may interfere with the binding of IgE to FcER 2 and that CH4 constitutes a smaller obstacle to binding than C-gamma-3.

It was found that rE2-4 (Arg 496 is more active as a dimer than rE2-4; this might be due to the presence of unpaired S CH4 domains, which, owing to their segmental flexibility and rotational freedom, might interfere to a lesser extent than the paired domain. rE4, which appears to have a native structure by the criterion of dimerisation, failed to bind to the FcER 2 on the B cell. It is therefore unlikely that the CH4 domain makes a positive contribution to the interaction of IgE with FcER 2 The data do not reveal whether both C-epsilon domains bind to a single receptor or whether one, or possibly two, FcER 2 molecules bind to the separate CH3 domains in a dimer. The first model of binding seems unlikely, since the two CH3 domains are far apart, being separated by protein and carbohydrate. Existing evidence is inconclusive. It has been shown that mAb IC 27 can bind to WO 89/04834 PCT/GB88/01018 WO 24 an IgE, immobilised on its receptor on a B cell. This could imply that the FcER 2 molecule binds to only one of the CH3 domains of IgE, leaving the second free to interact with the mAb (if it is assumed that the mAb and the receptor compete for the same site). Other data, however, suggest that two receptors may bind to the CH3 domains on a single IgE molecule, for a bivalent monoclonal antibody against murine FcER 2 binds to the same number of sites as does IgE, while its Fab' fragment binds to twice this number. In theory, "monogamous bivalency", as the 2:2 binding mode has been termed, may offer a large gain in affinity, via. Ka dimer (Ka monomer) 2 If the observed K a of 107 M 1 arises from comparable contributions at both sites, then the binding free energy developed by a single-site interaction may be below the threshold of detection used in the assay described herein. There is still uncertainty about the stoichiometry of the interaction which translates into a question of specificity, namely whether or not the receptor recognises a single CH3 domain in a conformation induced by dimerisation, *e or whether or not the dimer serves only to generate higher affinity through interaction at two sites.

The FcER 2 binding site is distinct from the FcER 1 S binding site in human IgE: FcER 1 but not FcER 2 binds to rE2-3 and rE2'-3', whereas FcFR 2 but not FcERI, binds to rE3-4. This indicates that some part of the sequence i between Gln 291 and Leu 330 is required for FcER 1 but not for FcER 2 binding. The sites may overlap in CH3, in the region between Val 352 and Lys 357, which forms the Cterminal boundary of the FcER 1 site, but the FcER 2 site may extend further towards the C-terminal side, or indeed the sites may be totally separate. Higher-resolution mapping is required to establish the N-terminal boundary of the FcER 2 binding site and thus determine the extent, if any, of common sequence. The two receptors also exhibit a different mode of binding, since FcER 1 binds to monomeric and dimeric

L•

I -A44- J .WO 8904834 PCT/GB88/01018 25 chains, wherein FcER 2 fails to bind to monomers. The monocyte receptor binding site on IgGl and the T and B cell receptor binding site on IgM have been mapped, respectively, to the C-gamma-2 and C-mu-3 domains, homologous to CH3. It is of interest that Fc-mu-R, like FcER 2 fails to bind to monomeric chains.

FcER 1 and FcERj aro unrelated proteins; FcER 1 belongs to the immunoglobulin superfamily, in common with all other immunoglobulin receptors so far described. FcER 2 is unique, in that it is homologous to the asialoglycoprotein receptor. It is therefore surprising that FcER 2 binds to rE2-4, and to enzymatically deglycosylated IgE, even more strongly than to native IgE.

SHigh affinity binding of IgG to Fc-gamma-Rl, on the ji: contrary, is reported to be dependent on the presence of carbohydrate. It is noted that there is a glycosylation site at Asn 361 in the epsilon chain near the putative region of the FcER 2 binding site. This carbohydrate substituent is clearly not a part of the binding site, but it is found only in epsilon-heavy chains and might be well placed to modulate FcER 2 binding activity.

:1 I 1. w

Claims (24)

1. A polypeptide competitor for Immunoglobulin E low affinity receptor sites comprising a polypeptide having the H sequence of amino acid residues, numbered 330 to 537 according to the Bennich enumeration, of the formula: k330 A-Leu-Ser-Arg-Pro-Ser-Pro-Phe-Asp-Leu- 339 Phe-Ile-Arg-Lys-Ser-Pro-Thr-Ile-Thr-Cys- 349 ILeu-Val-Val-Asp-Leu-Ala-Pro-Ser-Lys-Gly- 359 Thr-Val-Asn-Leu-Thr-Trp-Ser-Arg-Ala-Ser- 369 Gly-Lys-Pro-Val-Asn-Hris-Ser-Thr-Arg-Lys- 0: 379 Glu-Glu-.Lys-Gln--Arg-Asn-Gly-Thr-Leu-Thr- 389 Val-Thr-Ser-Thr-Leu-Pro-Val-Gly-Thr-Arg- 399 Asp-Trp-Ile-Glu-Gly-Glu-Thr-Tyr-Gln-Cys- S409 Arg-Val-Thr-His-Pro-His-Leu-Pro-Arg-Ala- 419 Leu-Met-Arg-Ser-Thr-Thr-Lys-Thr-Ser-Gly- 429 *Pro-Arg-Ala-Ala-Pro-Glu-Val-Tyr-Ala-Phe- 439 Ala-Thr-Pro-Glu-Trp-Pro-Gly-Ser-Arg-Asp- 449 Lys-Arg-Thr-Leu-Ala-Cys-Leu-Ile-Gln-Asn- 459 Phe-Met-Pro-Glu-Asp-Ile-Ser-Val-Gln-Trp- Z)- ,,0G' 469 Leu-His-Asn-Glu-Val-Gln-Leu-Pro-Asp-Ala- 479 Arg-His-Ser-Thr-Thr-Gln-Pro-Arg-Lys-Thr- 489 Lys-Gin-Ser-Gly-Phe--Phe-Val-Phe-Ser-Arg- 499 Leu-Gln-Val-Thr--Arg-Ala-Glu-Trp-Glu-Gln- 509 Lys-Asp--Glu-Phe-Ile-Cys-Arg-Ala-Val-His- 519 Glu-Ala-Ala-Ser-Pro-Ser-Gln-Thr-Val-Gln- 529 Arg-Ala-Val-Ser-Val--Asn-Pro-Gly-Lys-[Stop], *wherein the group A is a hydrogen atom or an amino acid 00000 sequence which may include a chain initiating amino acid 0. 0: sequence and in which'said amino acid sequence includes at most a part only of the natural sequence of amino acid residues of the heavy chain of Immunoglobulin E, said part of said natural sequence forming a part only of the sequence 00 of amino acid residues, numbered 291 to 329 according to the Bennich enumeration, of the following formula: 0 0,0: 301 Thr-Cys-Gln-Val-Thr-Tyr-Gln-Gly-His-Thr- 311 Phe-Glu-Asp-Ser-Thr-Lys-Lys-Cys-Ala-Asp- 321 Ser-Asn-Pro-Arg--Gly-Val-Ser-Ala-Tyr-

2. A polypeptide competitor according to claim k, wherein A represents a part of the sequence of amino acid .,,--,,reidusnumbered 291 to 329 according to the Bennich enumeration, of the following formula: 291 Gln-Lys-His-Trp-Leu-Ser-Asp-Arg-Thr-Tyr- 301 Thr-Cys--Gln-Val--Thr-Tyr--Gln-Gly-His-Thr- 311 Phe-Glu-Asp-Ser-Thr-Lys-Lys-Cys-Ala-Asp- 321 Ser-Asn-Pro-Arg--Gly--Val-Ser-Ala-Tyr-.

3. A polypeptide competitor according to claim 1, wherein A represents the amino acid sequence Met-Asp-Pro-Arg-. *000 4. A polypeptide competitor for Iminunoglobulin E low affinity receptor sites comprising a polypeptide having the .sequence of amino acid residues, numbered 330 -to 537 according to the Bennich enumeration, of the formula: 330 X-Leu-Ser-Arg-Pro-Ser-Pro-Phe-Asp-Leu- 339 Phe-Ile-Arg-Lys-Ser-Pro-Thr-Ile-Thr-Cys- 34 Leu-Val--Val-Asp-Leu-Ala-Pro-Ser-Lys-Gly- 359 Thr-Val-Asn-Leu-Thr-Trp-Ser-Arg-Ala-Ser- 369 Gly-Lys-Pro-Val-Asn-His-Ser-Thr-Arg-Lys- 379 Glu-Glu-Lys-Gln-Arg-Asn-Gly-Thr-Leu-Thr- 389 Val-Thr-Ser--Thr-Leu-Pro-Val-Gly-Thr-Arg- A 1I 39 Asp-Trp-Ile-Glu-Gly-Glu-Thr-Tyr-Gln-Cys- 0 0*0. 0 :0 S 0 409 Arg-Val-Thr-His-Pro-His-Leu-Pro-Arg-Ala- 419 Leu-Met-Arg-Ser-Thr-Thr-Lys-Thr-Ser-Gly- 429 Pro-Arg-Ala-Ala-Pro-Glu-Val-Tyr-Ala-Phe- 439 Ala-Thr-Pro-Glu-Trp--Pro-Gly-Ser-Arg-Asp- 449 Lys-Asg-Thr-Leu-Ala-Cys-Leu-Ile-Gln-Asn- 459 Phe-Met-Pro-Glu-Asp-Ile-Ser-Val-Gln-Trp- 469 Leu-His-Asn-Glu-Val-Glri-Leu-Pro-Asp-Ala- 479 Arg-His-Ser-Thr-Thr-Gln-Pro-Arg-Lys-Thr- 489 Lys-Gln-Ser-Gly-Phe-Phe--Val-Phe-Ser-Arg- 499 Leu-Gln-Val-Thr-Arg-Ala-Glu-Trp-Glu-Gln- 509 Lys-Asp-Glu-Phe-Ile-Cys-Arg-Ala-Val-His- 519 Glu-Ala--Ala--Ser-Pro-Ser-Gln-Thr-Val-Gln- 529 Arg-Ala-Val-Ser-Val-Asn-Pro-ly-Lys-[Stop] wherein X is a hydrogen atom or a chain initiating amino acid sequence. A polypeptide competitor according to claim 4 wherein X represents Met-Asp-Pro-Arg.

6. A polypeptide competitor according to any one of claims 1 to 5 wherein the specified polypeptide sequence is _I~ terminated at at least one end by an inert polypeptide sequence initiating or terminating the chain.

7. A polypeptide competitor according to any one of claims 1 to 6 wherein the specified polypeptide sequence is terminated both at its N-terminal end and at its C-terminal end by respective inert oligopeptide sequences initiating and terminating the chain.

8. A polypeptide competitor according to claim 4 wherein X is a hydrogen atom.

9. A dimeric form of a polypeptide competitor according to any one of claims 1 to 8.

10. Fragments of a polypeptide competitor according to any one of claims 1 to 8 or a dimeric form thereof according to claim 9 having competitive properties which are the same as or similar to those of the specified polypeptide sequence.

11. A pharmaceutical preparation containing as active principle a polypeptide competitor according to any one of claims 1 to 8, or a dimeric form according to claim 9, or a S biologically active fragment thereof as claimed in claim in combination with a pharmaceutical carrier or excipient. x A DNA having the nucleotide sequence: CTA AGC CGG CCC AGC CCG TTC GAC CTG TTC ATC CGC AAG TCG CCC ACG ATC ACC TGT CTG GTC GTC GAC CTG GCA CCC AGC AAG GGG ACC GTG AAC CTG ACC TGG TCC CGC GCC AGT 9* S. S. S S. S S S S S. S S. 5* 5 S S S. 5S S S S S S S .5.5 5 S S S 5* GCG AAG GAG GAG GTC ACG GAC TGG AGG GTG CTC ATG CCG CGT GCG ACG AGG CGC TTC ATG CTG CAC CGC CAC AAG GCC CTG GAG AAA GAT CTT AAG TCC ATC ACC CGG GCT CCG ACC CCT AAC AGC TCC GTC GAG GTG CAG ACC GAG CAC TCC GCC GAG CTC GAG GAG ACG GGC ACC TTC AAC CGC CTG GGG CCC ACG CCG TGG GCC GAC GTG ACG TTC AGG ATC CAC AAT CCG GAC CAC AC C GAA CCG TGC ATC CAG CAG TTC GCC TGC TCC GGC GTG ACC CTG AAG GTC GGG CTG TCG CTC CCC GTC GAA CGT ACC AGA ACG TTA GGC ACC TAC CAG CCC AGG ACC AGC TAT GCG AGC CGG ATC CAG GTC CAG CCC GAC CGC AAG TTC AGC TGG GAG GCA GTG AAG ACC CGA TGC GCC GCC TTT GAC AAC TGG GCC ACC CGC CAG CAT GAG GCA GCG AGC CCC TCA CAG ACC GTC CAG i CGA GCG GTG TCT GTA AAT CCC GGT AAA TGA

13. Fragments of the DNA claimed in claim 12 which, in a host/vector system, express at least one fragment according to claim

14. A transformant in which the DNA of the transforming vector contains a nucleotide sequence according to claim 12 or claim 13.

15. A vector comprising a DNA segment according to claim 12 or claim 13 said segment being oriented within said vector such that in a host said segment is expressed to produce a polypeptide. i. 16. A vector as claimed in claim 15, comprising the plasmid pE 3-4 deposited in Escherichia coli N4830 with the National Collection of Type Cultures, London on 18th November 1987 under the Accession Number NCTC 12162.

17. A host transformed by the vector claimed in claim 16.

18. A host/vector as claimed in claim 17 comprising Escherichia coli N4830 harbouring the plasmid pE 3-4 (Accession Number NCTC 12162, deposited with the National Collection of Type Cultures, London on 18th November 1987).

19. A method of preparing the polypeptide claimed in claim 4, comprising culturing the host organism claimed in claim 17 and isolating the polypeptide from the culture. S 0. A method of preparing the polypeptide claimed in 0 1 E I: -34- claim 8 comprising culturing the host/vector claimed in claim 18, isolating the polypeptide from the culture to obtain the polypeptide claimed in claim 4 and treating same to remove the chain initiating group X.

21. A binding assay in which a reagent is used which is a polypeptide competitor according to any one of claims 1 to 8 or a dimeric form according to claim 9 or a biologically active fragment according to claim

22. A diagnostic kit including means for conducting an assay and including, for use as a reagent in the said assay, a polypeptide competitor according to any one of claims 1 to 8 or a dimeric form according to claim 9 or a biologically active fragment according to claim

23. A polypeptide or fragment according to any one of claims 1 to 10 substantially as hereinbefore described. Goes

24. A pharmaceutical preparation according to claim 11 S: substantially as hereinbefore described. S. 25. A fragment according to claim 13 substantially as hereinbefore described.

26. A transformant according to claim 14 substantially as hereinbefore described.

27. A vector according to claims 15 or 16 substantially as hereinbefore described.

28. A host or host/vector according to claims 17 or 18 substantially as hereinbefore described.

29. A method according to claims 19 or 20 substantially as hereinbefore described. A binding assay according to claim 21 substantially as hereinbefore described.

31. A diagnostic kit according to claim 22 substantially as hereinbefore described. DATED this June 23, 1992 CARTER SMITH BEADLE Fellows Institute of Patent Attorneys of Australia SPatent Attorneys for the Applicant: 3i RESEARCH EXPLOITATION LIMITED mwspe#3867 June 23, 1992