This invention relates to polypeptides, more particularly to polypeptide competitors for IgE receptor sites on cells.

Antibodies of the immunoglobulin E (IgE) class make up a minute proportion (ca 0.01%) of the total immunoglobulin in normal human serum. However, their activities are powerfully amplified by the cell receptors to which they bind, and elevated levels of IgE play a central role in atopic allergy. The biological activities of immunoglobulin E (IgE) depend on its interaction with two receptors, FceRI and FceRII, expressed on effector cells; cross-linking of surface receptor-bound IgE allows antigen triggering of cell activation and is implicated in the aetiology of allergic diseases. Antigens bind to the Fab regions of the antibody, while the receptors bind to the Fc region, comprising a dimer of the three C-terminal domains of the e chain, comprising the second, third and fourth constant region domains (Ce2-Ce4) .

Cross-linking of IgE bound to the "high affinity" receptor, FceRI, on mast cells and basophils leads to cell degranulation and the release of pre-formed and newly synthesised mediators, responsible for immediate hypersensitivity and for initiating a leucocyte cascade that results in a later inflammatory response. This has been reported by Ishizaka et al . , Im uno chemistry 7 : 687 (1970) and Prog. Allergy 19 : 60 (1975) . The so-called "low affinity" receptor, FceRII, has many activities, including the sensitisation of inflammatory cells for IgE antibody- dependent reactions, as reported by Capron et al . , Immunol . Today, 7 : 15 (1986) , and IgE-dependent antigen presentation to T cells by B cells. This latter activity is reviewed in a paper by Delespesse et al , Adv. Immunol . 49 : 149 (1991) .

It has been reported by Ishizaka et al ,

Immuno chemistry 7:687 (1970) and by Stanworth et al . , Lancet 2:17 (1968) that IgE is cleaved by papain into Fab and Fc fragments, homologous to those that result from the fragmentation of IgG. These authors also report that cell receptors recognise the Fc fragment, a disulphide-bridged dimer of the three carboxy-terminal domains (Ce2-Ce4) of each e-chain, covalently linked by two disulphide bonds between the pairs of Cys 328 and Cys 241 of the Ce2 residues. Evidence to this effect is also provided by Spiegelberg, Adv. in Immunol. 35:61 (1984), as well as by Dorrington et al . , Immunol. Rev. 41:3 (1978) and Helm et al., Eur. J. Immunol. 21:1543 (1991). IgE is heavily N- glycosylated, containing 13% by weight of carbohydrate, as compared with 3% for IgG according to Dorrington et al . , Immunol. Rev. 41:3 (1978). The IgE-Fc is potentially N- glycosylated at Asn 265 (Ce2 domain) , Asn 371, Asn 383, and Asn 394 (all three in the Ce3 domain) . The last of these is homologous to the glycosylation site in the corresponding domains of IgG, IgM and IgD (C 2, Cμ3, Cδ2) . Its predicted position is on the apposed faces of the Ce3 domains, which are expected to resemble the C 2 domains of IgG in their relative orientation according to Deisenhofer, Biochemistry 20:2361 (1981) and Sutton et al . , Bioche . Soc . Trans. 11:130 (1983). By contrast, carbohydrate chains linked to Asn 265, Asn 383 and Asn 371 are predicted to be fully exposed on the outer surface of the molecule, and are specific for IgE.

Reco binant human IgE Fc (hlgE-Fc) has been expressed in E. coli, as taught by Kenten et al . , Proc . Natl Acad. Sci. USA 81:2955 (1984) and Liu et al . , Proc. Natl. Acad. Sci. USA 81:5369 (1984). This product binds to FceRI with the same affinity as myeloma IgE (PS) according to Ishizaka et al . , Proc. Natl. Acad. Sci. USA 83:8323 (1986); since the E. coli product is non-glycosylated, these results

imply that carbohydrate does not form a part of the binding site for FceRI. In contrast to FceRI and all other immunoglobulin receptors which belong to the immunoglobulin superfamily, FceRII is a member of the C-type lectin family, and it was therefore expected that it might recognise the carbohydrate moiety of IgE. This is clearly not the case, however, since the E. coli hlgE-Fc binds to the receptor with about the same affinity as native IgE; in this connection reference may be made to Vercelli , Nature 338 : 649 (1989) . In IgG, glycosylation of Asn 297 (the homologue of Asn 394 in IgE) is required for IgG to bind with native affinity to its IgG-Fc receptors according to Nose et al , Proc . Natl . Acad. Sci . USA 80 : 6632 (1983) , to eatJierJarrow et al . , Mol . Immunol . 22 : 407 (1985) and to Heyman et al . , J. Immunol . 134 : 4018 (1985) , but this could be due to an indirect effect upon the polypeptide conformation.

The purification and characterisation of human recombinant IgE-Fc fragments that bind to the human high affinity IgE receptor is described by Basu et al . , J. Biol . Chem . 269 : 13118 (1993 ) . These authors state that the smallest IgE fragment that showed FceRIα binding activity spans amino acids 329-547 and lacks the entire Ce2 domain. They also reported two active fragments, viz. Fee (315-547) and Fe (329-547) , which were overexpressed in Chinese hamster ovary (CHO) cells and purified to homogeneity. The presence of N-linked glycosylation was detected in both proteins .

It is possible to obtain high levels of expression of the recombinant human IgE-Fc (hlgE-Fc) in E. coli, but not to recover more than a small fraction of the desired product, a dimer of -chain fragments with the same affinity for receptors as native IgE; this can be attributed to incorrect folding and/or aggregation of the bulk of the material in vitro.

The structure of human immunoglobulin E (hlgE) was first determined by Bennich et al . , Progress in Immunology II, Vol 1 : 49 (1974) . In later work, described in Bennich et al . , Int . Arch . Allergy Appl . Immunol . 53 : 459 (1977) , it was reported that amino acid 322 is asparagine and not aspartic acid. A corrected formula appears in EP-A-0102634.

The hlgE-Fc contains four potential glycosylation sites, at Asn 265 (in Ce2) and Asn 371, Asn 383 and Asn 394 (in Ce3) . Three of these, Asn 265, Asn 371 and Asn 383, are predicted to be on the external surface of the protein according to Helm et al , Eur . J. Immunol . , 21 : 1543 (1991) , while Asn 394, being homologous to Asn 297 in IgG is predicted to be partially buried in the protein. It has also been reported that the glycosylation site at Asn 383 is not occupied in the myeloma IgE from a patient (designated "patient ND") , whose expressed e-chain cDNA sequence is immortalised in the hlgE-Fc expression constructs reported by Kenten et al , Proc . Natl . Acad. Sci . USA 81 : 2955 (1984) .

In WO-A-88/00204 there is described a polypeptide residue with a chain length of 76 amino acid residues which is a competitor for hlgE and binds specifically to the so called "high affinity" Fc receptor sites for IgE (i.e. FceRI sites) which exist on human cells, particularly mast cells and basophils.

A larger polypeptide chain which binds to so called "low affinity" receptor (or FceRII) sites is disclosed in O-A-89/04834.

US-A-4171299 and US-A-4161522 disclose that an oligopeptide containing from three to ten amino acids in a sequence selected from a portion of amino acids 265 to 537, according to the Bennich nomenclature, of the Fc region of hlgE will block Fc receptors of mast cells.

It is an object of the present invention to provide novel synthetic polypeptides, related to portions of the

hlgE molecule, in particular to at least a part of the hlgE- Fc fragment, which are useful in the study and treatment of allergy conditions. It is a further object of the present invention to provide synthetic mutant polypeptides derived from the Fc chain of hlgE which do not form aggregates larger than dimers. Yet a further object of the present invention is to provide synthetic polypeptides which bear glycoside side chains on at least one of the available sites of the Fc chain of hlgE and which are useful for investigation of and amelioration of allergic conditions. The invention also has for an object the provision of mutant polypeptides related to the Fc chain of hlgE which have one or more of the glycosylation sites present on the Fc chain of hlgE substituted by an amino acid residue which cannot be glycosylated.

According to the present invention there is provided a mutated glycosylated polypeptide which includes at least a part of the hlgE-Fc chain of sufficient length to bind to FceRI and/or FceRII receptor sites on human cells wherein Cys 225 has been mutated by replacement with another amino acid residue or has been deleted, optionally together with Val 224 and Ser 226 or with Val 224 Ser 226 and Arg 227, or with Val 224 Ser 226 Arg 227 and Asp 228, and wherein at least one of the sites Asn 394 and, if present, Asn 265 and/or Asn 371 bears a glycoside chain.

In this specification the numbering of the amino acid residues, except where otherwise indicated, is based upon that of Dorrington et al . , Immunological Reviews , 41 : 3 (1978) at page 7, as modified by the results of subsequent research. Specifically the numbering of the amino acid residues is as follows : -

[SEQ ID No: 1]

224 VC

226 SRDFTPPTVK ILQSSCDGGG HFPPTIQLLC LVSGYTPGTI NITWLEDGQV 275 MDVDLSTAST TQEGELASTQ SELTLSQKHW LSDRTYTCQV TYQGHTFEDS 325 TKKCADSNPR GVSAYLSRPS PFDLFIRKSP TITCLWDLA PSKGTVNLTW 375 SRASGKPVNH STRKEEKQRN GTLTVTSTLP VGTRDWIEGE TYQCRVTHPH 425 LPRALMRSTT KTSGPRAAPE VYAFATPEWP GSRDKRTLAC LIQNFMPEDI 475 SVQWLHNEVQ LPDARHSTTQ PRKTKGSGFF VFSRLEVTRA EWEQKDEFIC 525 RAVHEAASPS QTVQRAVSVN PGK

In this sequence the amino acid residues are identified by the usual one-letter symbols:-

Amino acid Three-letter One-letter abbreviation symbol

Alanine Ala A

Arginine Arg R

Asparagine Asn N

Aspartic acid Asp D

Asparagine or aspartic acid Asx B

Cysteine Cys C

Glutamine Gin Q

Glutamic acid Glu E

Glutamine or glutamic acid Glx Z

Glycine Gly G

Histidine His H

Isoleucine He I

Leucine Leu L

Lysine Lys K

Methionine Met M

Phenylalanine Phe F

Proline Pro P

Serine Ser S

Threonine Thr T

Tryptophan Trp W

Tyrosine Tyr Y

Valine Val V

The leucine residue (L) identified by an asterisk (*) is assigned the number 253A in the above sequence. The underlined residues are Cys 225, Asn 265, Asn 371 and Asn 394.

The DNA sequence coding for this sequence is set out below: -

[SEQ ID No: 2]

GTCTGCTC CAGGGACTTC ACCCCGCCCA CCGTGAAGAT CTTACAGTCG TCCTGCGACG GCGGCGGGCA CTTCCCCCCG ACCATCCAGC TCCTGTGCCT CGTCTCTGGG TACACCCCAG GGACTATCAA CATCACCTGG CTGGAGGACG GGCAGGTCAT GGACGTGGAC TTGTCCACCG CCTCTACCAC GCAGGAGGGT GAGCTGGCCT CCACACAAAG CGAGCTCACC CTCAGCCAGA AGCACTGGCT GTCAGACCGC ACCTACACCT GCCAGGTCAC CTATCAAGGT CACACCTTTG AGGACAGCAC CAAGAAGTGT GCAGGTACGT TCCCACCTGC CCTGGTGGCC GCCACGGAGG CCAGAGAAGA GGGGCGGGTG GGCCTCACAC AGCCCTCCGG TGTACCACAG ATTCCAACCC GAGAGGGGTG AGCGCCTACC TAAGCCGGCC CAGCCCGTTC GACCTGTTCA TCCGCAAGTC GCCCACGATC ACCTGTCTGG TGGTGGACCT GGCACCCAGC AAGGGGACCG TGAACCTGAC CTGGTCCCGG GCCAGTGGGA AGCCTGTGAA CCACTCCACC AGAAAGGAGG AGAAGCAGCG CAATGGCACG TTAACCGTCA CGTCCACCCT GCCGGTGGGC ACCCGAGACT GGATCGAGGG GGAGACCTAC CAGTGCAGGG TGACCCACCC CCACCTGCCC AGGGCCCTCA TGCGGTCCAC GACCAAGACC AGCGGTGAGC CATGGGCAGG CCGGGGTCGT GGGGGAAGGG AGGGAGCGAG TGAGCGGGGC CCGGGCTGAC CCCACGTCTG GCCACAGGCC CGCGTGCTGC CCCGGAAGTC TATGCGTTTG CGACGCCGGA GTGGCCGGGG AGCCGGGACA AGCGCACCCT CGCCTGCCTG ATCCAGAACT TCATGCCTGA GGACATCTCG GTGCAGTGGC TGCACAACGA GGTGCAGCTC CCGGACGCCC GGCACAGCAC GACGCAGCCC CGCAAGACCA AGGGCTCCGG CTTCTTCGTC TTCAGCCGCC TGGAGGTGAC CAGGGCCGAA TGGGAGCAGA AAGATGAGTT CATCTGCCGT GCAGTCCATG AGGCAGCGAG CCCCTCACAG ACCGTCCAGC GAGCGGTGTC TGTAAATCCC GGTAAATGAC GTACTCCTGC CTCCCTCCCT CCCAGGGCTC CATCCAGCTG TGCAGTGGGG AGGACTGGCC AGACCTTCTG TCCACTGTTG CAATGACCCC AGGAAGCTAC CCCCAATAAA CTGTGCCTGC TCAGAGCCCC AGTACACCCA TTCTTGGGAG CGGGCAGGGC.

The codons in this DNA sequence where mutations are introduced in order to produce the polypeptides according to the invention are underlined in the sequence. For the mutation wherein Cys 225 is replaced by Ala the corresponding codon TGC is altered to GCC. In the case of the mutations of Asn 265 and Asn 371 to Gin the

corresponding codon AAC is in each case changed to CAG.

The invention further provides a polypeptide which binds to human immunoglobulin E (hlgE) receptor sites on cells and which is of the formula:

AA _n -hIgE-Fc (W225, X265, Y371, Z394) wherein:

AA represents an amino acid residue which may be the same as or different from any other group AA which may be present in the molecule; n represents zero or an integer from 1 to about 10; and hlgE-Fc (W225, X265, Y371, Z394) represents a mutant version of the hlgE-Fc chain with a mutation or deletion at least at position 225; wherein

W225 represents deletion of Val 224 and Cys 225, of Val 224, Cys 225 and Ser 226, or of Val 224, Cys 225, Ser 226 and Arg 227, or of Val 224, Cys 225, Ser 226, Arg 227 and Asp 228, or represents the residue of an amino acid at position 225 other than cysteine;

X265 is the residue of an amino acid at position 265;

Y371 is the residue of an amino acid at position 371; and

Z394 is the residue of an amino acid at position 394; and wherein at least one of X265, Y371 and Z394 may be an asparagine residue which may be glycosylated; or a fragment of such a polypeptide which lacks up to 10 terminal amino acid residues of the Ce4 domain at the carboxy end of the chain.

In such a polypeptide W225 may represent an alanine residue. In a particularly preferred polypeptide X265 represents a glutamine residue. Y371 may represent a glutamine residue, while Z394 may represent a glutamine residue.

Preferred polypeptides are selected from :

AA _n -hIgE - Fc (Ala 225 )

AA _n -hIgE-Fc (Ala 225, Gin 265)

AA _n -hIgE-Fc (Ala 225, Gin 371)

AA _n -hIgE-Fc (Ala 225, Gin 394)

AA _n -hIgE-Fc (Ala 225, Gin 265, Gin 371)

AA _n -hIgE-Fc (Ala 225, Gin 265, Gin 394)

AA _n -hIgE-Fc (Ala 225, Gin 371, Gin 394)

AA _n -hIgE-Fc (Ala 225, Gin 265, Gin 371, Gin 394) .

In the above formula AA _n preferably represents an inert polypeptide sequence, for example Asp-He.

In the polypeptide the N-terminal sequence of the group hlgE-Fc (W225, X265, Y371, Z394) may have the structure:

[SEQ ID No: 3]

Asp He Val Ala Ser Xaa Asp Phe Thr where Xaa is the residue of an amino acid, for example an arginine residue. In this case AA represents Asp-He and the Ala residue in this sequence is the residue replacing Cys 225.

W225 may alternatively represent deletion of Val 224 and Cys 225, or of Val 224 Cys 225 Ser 226, or of Val 224 Cys 225 Ser 226 Arg 227, or of Val 224 Cys 225 Ser 226 Arg 227 Asp 228.

The invention also provides a vector containing cDNA coding for a polypeptide according to the invention, as well as a mammalian cell line, e.g. a human cell line or a Chinese hamster ovary cell line, containing DNA coding for a polypeptide according to the invention, which expresses such a polypeptide.

For convenience the mutant version of hlgE-Fc (Ala 225) with the additional sequence Asp-He at the amino end of the chain is referred to as X'-hlgE, while the corresponding double and triple mutants are also given the

prefix X' -hlgE.

Included vectors according to the invention are expression vectors that lead to secretion of X'-hlgE-Fc, and the mutants X'-hlgE-Fc (Gin 265) , X'-hlgE-Fc (Gin 371) , and X'-hlgE-Fc (Gin 265, Gin 371) , from mammalian cells at levels up to 100 mg/litre of culture.

Also provided according to the present invention are pharmaceutical preparations comprising a polypeptide according to the invention and a carrier therefor. Such carriers are well known in the art .

The invention provides a novel method for the production of a dimeric immunoglobulin Fc chain fragment by expression in, and secretion from, mammalian cells. Secretion is brought about by linking the DNA sequence encoding the Fc fragment of the human X' -IgE-Fc (hlgE-Fc) or a mutant thereof to a kappa-light chain signal sequence. The recombinant DNA can be cloned into mammalian expression vectors for transfection of CHO (Chinese hamster ovary) or NS-0 cells. The X'-hlgE-Fc chain or mutant thereof is assembled into dimers, correctly processed, and secreted from the CHO cells. Similar results are obtained when the gene is transfected into the myeloma cell line NS-O, showing that the X'-hlgE-Fc and its mutants may be produced in a variety of mammalian cells.

The secreted wild type X'-hlgE-Fc contains the full sequence of Ce2-Ce4 with one amino acid substitution, alanine for cysteine at position 225, and an extension of two amino acids (aspartic acid and isoleucine) , which remain at the N-terminus after cleavage of the leader signal sequence from the precursor peptide. The replacement of Cys 225 by alanine is believed to be an important feature in the design of the X'-hlgE-Fc and its mutants for secretion by mammalian cells. The two free cysteine residues are predicted to lie apart from each other, but available for

reaction with other molecules, on the surface of the Fe fragment and hence available to mediate the formation of aggregates .

Recombinant X'-hlgE-Fc can also be prepared by expression in E. coli. The product accumulates in the bacterium as an insoluble inclusion body, requiring dispersal in denaturing solvents. The recovery of native structure therefore involves correct folding, assembly of dimers and disulphide bond formation. The recovery of recombinant X'-hlgE-Fc after the application of these procedures is at best about 5-10%, although the biological activity of the final product is indistinguishable from native IgE. This low recovery may reflect the proportion of incorrectly folded and/or disulphide-bonded fragments which are eliminated in the purification of the active material. No evidence for the secretion of inactive products was seen in the production of the polypeptides according to the invention when secreted from CHO or NS-0 cells. The total amount of X'-hlgE-Fc synthesised in CHO cells is less than in E. coli (2mg/litre in CHO cells, as against > 55 mg/litre in E. coli) , but with NS-0 cells expression is increased to a level (ca 100 mg/litre) higher than that obtained in E. coli .

A secretion system has been reported by Ki tai et al . , Appl . Microbiol . Biotechnol . 28 : 52 , for the expression of assembled IgG-Fc in E. coli, but the yield was lower (3 mg/litre of culture) than from non-secreting bacterial expression systems. This is comparable to the level of X'-hlgE-Fc secretion from CHO cells achievable in accordance with the invention (i.e. approximately 2 mg/litre of culture supernatant) , but greatly inferior to the amount accumulated by secretion from the NS-0 cell line.

In contrast to bacterial expression, the mammalian expression system allows production of glycosylation site

mutants. X'-hlgE-Fc has been prepared with different combinations of the surface carbohydrate chains of the IgE-Fc fragment. Analysis of transient expression products reveals that Asn 265 was completely glycosylated, but that Asn 371 is rarely glycosylated, and Asn 383 not at all, in CHO cells . The establishment of permanent lines expressing the mutant type X'-hlgE-Fc and the triple mutant, X' -hlgE-Fc (Gin 265, Gin 371), provides material for biological assays.

The sensitisation of human basophils by X'-hlgE-Fc and X' -hlgE-Fc (Gin 265, Gin 371) provides an assay for effector function. Basophils sensitised by each X' -IgE-Fc fragment or IgE show almost identical response to anti-IgE stimulation. The small shift in the curve of histamine release from IgE (PS) -sensitised cells is presumably due to differences between the interactions of IgE an IgE-Fc fragments with the polyclonal anti-IgE, which was raised against whole myeloma IgE. These results indicate that both wild type and mutant recombinant hlgE-Fc are as effective as whole IgE in sensitising FceRI of human basophils for histamine release.

Comparison of wild type X'-hlgE-Fc and mutant products reveals that Asn 371 is rarely glycosylated in Chinese hamster ovary cells. Both the double mutant, X'-hlgE-Fc (Gin 265, Gin 371) , and wild type, X'-hlgE-Fc, bind to the high affinity IgE receptor, FceRI, with about the same affinity as myeloma IgE (K in the range 10 ¹⁰ -10 ^1;L M ^"1 ) and were able to sensitise isolated human basophils for anti-IgE triggering of histamine release. However, only the double mutant, X'-hlgE-Fc (Gin 265, Gin 371) , approached the affinity of myeloma LgE for the low affinity receptor, FceRII (K = 7.3 x 10 ⁷ M ^"1 ) , whereas the wild type, X'-hlgE-Fc, bound with a ten-fold lower affinity to the low affinity receptor (K = 4.1 x 10 ⁶ M ^"1 ) .

In the drawings : -

Figure 1 is a diagrammatic representation of the hlgE molecule, indicating the Fab and Fc fragments and the Cel to Ce4 domains, as well as the intermolecular and intramolecular disulphide linkages;

Figure 2 is a representation of the structural relationship between part of the natural hlgE-Fc chain and its associated DNA sequence;

Figure 3 is a similar representation of the corresponding part of a mutation with Cys 225 replaced by Ala 225, i.e. X'-hlgE-Fc (Ala 225) , and its associated mutated DNA sequence;

Figure 4 is a similar representation of part of a mutation with Asn 265 replaced by Gin 265, i.e. X'-hlgE-Fc (Ala 225, Gin 265), and its associated mutated DNA sequence;

Figure 5 is a similar representation of part of a mutation with Asn 371 replaced by Gin 371, i.e. X'-hlgE-Fc (Ala 225, Gin 371) , and its associated DNA sequence;

Figure 6 is a restriction map illustrating construction of a mammalian expression vector for the polypeptides of the invention;

Figure 7 shows SDS-polyacrylamide gel electrophoresis patterns of various polypeptides which are mutations of the hlgE-Fc chain both under reducing and non- reducing conditions,-

Figure 8 shows similar patterns for purified wild type hlgE-Fc and the triple mutation site mutant X'-hlgE-Fc (Ala 225, Gin 265, Gin 371) ; and

Figure 9 illustrates the effector function of hlgE- Fc.

In more detail Figure 1 shows the covalent structure of human IgE. The locations of intra-chain and inter-chain disulphide bonds (S-S) in the variable (V) and four constant (Cel, Ce2, Ce3 and Ce4) domains are shown

schematically together with the extent of the Fc chain. The arrangement of the two inter-chain bonds at Cys 241 and Cys 328 is shown parallel. The position of Cys 225, which has been mutated to alanine in the X'-hlgE-Fc, and of the three glycosylation sites at Asn 265 and Asn 371 and Asn 394, are also indicated.

Figure 2 sets out amino acid and nucleotide listings as follows:

[SEQ ID No: 4]

...KTFSVCSRD... and

[SEQ ID No: 5]

...AAAACCTTCA GCGTCTGCTC CAGGGAC... Figure 3 includes amino acid sequences as follows : [SEQ ID No: 6] ...CDIVASRD... and

[SEQ ID No: 7]

...TGTGATATCG TCGCCTCCAG GGAC... respectively.

Although some authors consider that the junction between the Cel and Ce2 domains occurs between Cys 225 and Ser 226, for the purposes of the present invention this boundary is taken as occurring where determined by the DNA intron and exon boundaries, in other words between Ser 223 and Val 224. This is illustrated in Figure 2.

The construction and expression of a high expression vector for X'-hlgE-Fc are illustrated in Figure 3 which shows the nucleotide and amino acid sequence at the natural junction between Cel and Ce2 of the human e-chain. The N-terminal portion of the Ce2 domain was reconstructed with synthetic oligonucleotides to include the replacement of Cys 225 by alanine and to allow the incorporation of an EcoRV restriction site at the 5 ' -end of the X'-hlgE-Fc gene.

A mouse variable kappa-chain leader sequence was ligated to this site. The N-terminal dipeptide sequence (Asp. He) of the mature X'-hlgE-Fc derives from the proteolytic cleavage or "processing" of the leader sequence at the position marked with an asterisk (*) .

Figures 4 and 5 indicate sequences of mutagenic oligonucleotide pairs. Those of Figure 4, reading from the 5 ' end are:

[SEQ ID No: 8]

GGGACTATCC AGATCACCTG G and

[SEQ ID No : 9 ]

CCAGGTGATC TGGATAGTCC C reapectively, while those of Figure 5 are:

[SEQ ID No: 10]

GGGACCGTGC AGCTGACCTG G and

[SEQ ID No: 11]

CCAGGTCAGC TGCACGGTCC C respectively.

Figure 6 is a map of the expression vector pEE6HCMVgpt (see Stephens et al . , Nucl . Acids Res . 17 : 7110 (1989) ) into which the X'-hlgE-Fc construct was cloned.

Figure 7 shows SDS-polyacrylamide gel electrophoresis patterns of transiently expressed X'-hlgE-Fc proteins. X'-hlgE-Fc proteins were transiently expressed in biosynthetically labelled L761H cells. Secreted X'-hlgE-Fc proteins were immunoprecipitated from the cell medium with anti-hlgE-Fc mAb 7.12 described by Sherr et al . , J. Immunol . 142 : 181 (1989) coupled to Sepharose 4B and was analysed under reducing conditions (lanes Al to A4) and non-reducing conditions (lanes Bl to B4) on SDS 10% polyacrylamide gels. Lanes Al and B2 show X' -hlgE-Fc (Gin 265, Gin 371) , while lanes A2 and Bl show X'-hlgE-Fc (Gin 371) . Lanes A3 and B3

are X'-hlgE-Fc (Gin 265) and lanes A4 and B4 are X'-hlgE-Fc (wild type) . The positions of molecular weight (kDa) markers are indicated.

In Figure 8 there are shown SDS-polyacrylamide gel electrophoresis patterns of purified wild type X'-hlgE-Fc and the mutant X'-hlgE-Fc (Gin 265, Gin 371) . The X'-hlgE-Fc proteins secreted from mammalian cells were affinity purified on anti-hlgE-Fc mAB 7.12 coupled to Sepharose 4B. Purified samples were electrophoresed under reducing conditions (lanes 1 to 4) and non-reducing conditions (lanes 6 to 8) on a denaturing 10% polyacrylamide gel. A sample of X'-hlgE-Fc expressed in E. coli was also electrophoresed for comparison. Lanes 1 and 6 are X'-hlgE-Fc (wild type) ; lanes 2 and 7 are X' -hlgE-Fc (Gin 265, Gin 371) ; lanes 3 and 8 are X'-hlgE-Fc expressed in E. coli; and lane 4 is a mixture of high molecular weight standards (the molecular weights of the individual standards are indicted in kDa) .

The binding of X'-hlgE-Fc and mutants thereof was investigated. The fraction of functional X'-hlgE-Fc or mutant thereof and IgE(SF25) was determined from the percentage of molecules bound to an excess of a stable line of CHO cells expressing the human FceRI (CHO-hFceRI) described by Wang et al . , J. Exp . Med. 175 : 1353 (1989) , using the method of Isersky et al . J. Immunol . 112 : 1909 (1974) . IgE(SF25) is a recombinant chimeric IgE, with a mouse antibody heavy chain variable region and a human epsilon constant region sequence, and a corresponding mouse light chain. The concentrations of the X'-hlgE-Fc, of its mutants, and of IgE(SF25) used in all the assays was corrected for the fraction of functional molecules, typically in the range of 50-85%.

Figure 9 illustrates the effector function of X'-hlgE-Fc proteins. Human basophil leukocytes were passively sensitised with IgE (PS) , X'-hlgE-Fc fragments or

buffer and challenged with various concentrations of anti-IgE antibody. Spontaneous release of histamine from basophils sensitised with IgE (PS) , wild type X'-hlgE-Fc, X'-hlgE-Fc (Gin 265, Gin 371) or buffer control was 1.7%, 1.9%, 2.1%, 1.7% respectively. The graph shows the percentage of histamine released by sensitised human basophil leukocytes as the concentration of the challenging anti-human IgE antibody was increased. X'-hlgE-Fc (•) ; X'-hlgE-Fc (Gin 265, Gin 371) (0) , IgE (PS) (Δ) ; non-sensitised (A) .

Mixed leukocytes containing about 1% basophil leukocytes were obtained from 120ml of peripheral blood of a non-atopic healthy donor as described by Grattan et al . , Clin . Exp . Allergy 21 : 695 (1991) . The serum IgE level of this donor was less than 1 IU/ml and histamine release from the basophils in response to anti-IgE was 6.1 ± 3.1% (mean ± SD, n = 2) , indicating that they were only marginally sensitised with endogenous IgE. The leukocytes were washed twice with HAG (HBS (10 mM HEPES, 137 mM NaCl, 2.7 iuM KCl, 0.4 mM NaH ₂ P0 ₄ , pH 7.4) containing 0.03% human serum albumin and 5 mM glucose) . The leukocytes were resuspended in 2ml of HAG containing 4 mM EDTA in the presence of human myeloma IgE (PS) , X'-hlgE-Fc, X' -hlgE-Fc (Gin 265, Gin 371) or buffer only, followed by incubation at 37°C for 90 minutes with gentle shaking. The cells were washed three times with HAG, resuspended in HAG containing 2mM CaCl ₂ and 1 mM MgCl ₂ and challenged with various dilutions of anti-human IgE antibody (goat, e-chain specific, Sigma, UK) , as described by Gra ttan et al . , Clin . Exp . Allergy 21 : 695 (1991 ) . Briefly, aliquots of sensitised cells ( _~ 4 x 10 ⁴ basophils) were incubated with anti-IgE in a total volume of 200μl for 40 minutes at 37°C. The reaction was stopped by cooling on ice, followed by the addition of 800μl of ice-cold HBS to each tube. The cells were separated from the supernatants following

centrifugation and the histamine content of cell pellets and supernatants was determined by automated fluorometric analysis according to the method of Siraganian, J. Immunol . Methods 7 : 283 (1975) .

The invention is further illustrated in the following Examples. Example 1 a] Construction of X'-hlgE-Fc Expression Vectors wherein X ¹ is Asp-He:

To express X'-hlgE-Fc fragments in mammalian cells, the cDNA sequence encoding the ND myeloma -chain Fc of IgE was cloned in a mammalian expression vector. The entire heavy chain gene including the hlgE-Fc sequence was excised from the vector pJJ71 on a HindiII restriction fragment. This vector pJJ71 contains the human e-chain cDNA cloned from the 266bl cell line, as reported by Kenten et al . , Proc . Natl . Acad. Sci . USA 79 : 6661 (1982) . The fragment was then subcloned, in the opposite orientation, back into the Hindlll cut pJJ71 vector. All the sequence coding for the Fc fragment, including all of the Ce2, Ce3 and Ce4 domains, except for 34 bases at the 5 ' -end, were obtained from a BglH/BamHI digestion fragment of this vector.

The X'-hlgE-Fc cDNA sequence was adapted for secretion by ligating an EcoRI/EcoRV restriction fragment containing the B72.3 mouse hybridoma kappa-light chain gene leader sequence, as described by Whi ttle et al . , Protein Eng . 1 : 499 (1987) , at the 5 ' -end of the Fc coding sequence. The light chain gene had previously been mutated to introduce an EcoRV site at the 3 ' -end of the leader sequence. Such a mutation is silent and allows the leader sequence to be attached to a sequence with an EcoRV 5 '-end, while preserving the leader processing site. Comparison of the sequences of the natural Cel/Ce2 domain and the leader/adapter junctions illustrated in Figures 2 to 5

showed that this construct preserves the proteolytic cleavage, or "processing", site of the leader with minimal alteration of the N-terminus of the hlgE-Fc.

The cysteine at position 225 was also changed to alanine by rebuilding the N-terminal end of the Ce2 DNA sequence with oligonucleotides which include a codon coding for Ala in place of a codon coding for Cys 225. Cys 225 forms an intra-chain disulphide bond with a cysteine in Cel in IgE, as shown in Figure 1. Since Cel is not part of the Fc, Cys 225 may cause disulphide-linked inter-chain aggregates to form.

For the expression of the X'-hlgE-Fc proteins the entire EcoRI/BamHI leader-hlgE-Fc construct was cloned into EcoRl/BcII-cut pEE6HCMVgpt expression vector, as illustrated in Figure 6, downstream of the strong viral human cytomegalovirus (hCMV) promoter. pEE6HCMVgpt is the major immediate-early promoter-enhancer of hCMV described by Stephens et al . , Nucl . Acids . Res . 17. - 7110 . (1989) . The resulting vector was propagated in the dam E. coli strain GM242 to prevent methylation of its Bell restriction site.

This construct was further modified by the polymerase chain reaction (PCR) overlap extension method described by Ho et al . , Gene 77 : 51 (1989) to introduce glutamine residues in place of the asparagines at positions 265 and 371. Three mutants were made, two with a single substitution at Asn 265 or Asn 371, and a third with substitutions at both positions. The oligonucleotides used in the PCR cloning of these mutants are shown in Figures 4 and 5. The four X'-hlgE-Fc fragments are designated X'-hlgE-Fc (Ala 225) (single mutant or wild type) , X' -hlgE- Fc (Ala 225, Gin 265) (double mutant) and X' -hlgE-Fc (Ala 225, Gin 371) (double mutant) , and X' -hlgE-Fc (Ala 225, Gin 265, Gin 371) (triple mutant) . Confirmation of the mutations, adapter sequence and its functions was carried out by DNA

conjugate at 0.2μg/ml for 2 hours at 21°C. After washing with PBS, the blot was developed with Fast red/naphthol phosphate (Sigma, UK) .

30μg of purified triple mutant X'-hlgE-Fc (Ala 225, Gin 265, Gin 371) was used to determine the N-terminal sequence on an Applied Biosystems 470A gas phase sequencer with an on-line Applied Biosystems 12OA HPLC for the analysis of PTH-amino acids as described by Waterfield et al . in Handbook, Practical Protein Chemistry - A, A . Darbre, ed J Wiley & Sons, Chichester, UK, p411 (1986) .

Concentrations of X'-hlgE-Fc proteins in culture supernatants were monitored using an anti-IgE ELISA developed from a solid phase radio-immune assay described by Vercelli et al . , J Exp Med 169 : 1295 (1989) . This assay was essentially the same as that described by Whi ttle et al . , Protein Eng. , 1 : 499 (1987) , except that polystyrene maxisorp 96 well microtitre plates (Dynatech Labs USA) were coated with 2μg/ml (lOOμl/well) of a mixture of anti-hlgE-Fc mAbs 7.12 and 4.15 (see Sherr et al . , J. Immunol . , 142 : 481 (1989) . The monoclonal antibodies were coupled to the plate after 1 hour at room temperature in 0.1M sodium carbonate (pH 9.6) buffer. The plate was then washed three times with phosphate buffered saline (PBS) and blocked with 0.5% w/v casein in 0.1M sodium carbonate (pH 9.6) buffer. After six washes (PBS containing 0.025% v/v Tween 20) , lOOμl of the supernatants, diluted in PBS, were added to each well and incubated for 1 hour at room temperature. (The word "Tween" is a trade mark) . The plate was washed as described above and lOOμl of 1:1000 diluted rabbit anti-hlgE heavy chain-peroxidase conjugate (Dakopatts Ltd, Denmark) were added to each well to detect X'-hlgE-Fc proteins bound to the 7.12/4.15 mAbs and incubated for a further hour at room temperature. The wells were washed again and lOOμl of substrate containing 0.1 mg/ml tetra ethylbenzidine (TMB) ,

Gin 371) (Figure 7, lane Bl) , which bears carbohydrate at both Asn 265 and Asn 394 of both chains. This larger component (Figure 7, lane B3) therefore corresponds to a heterodimer in which one chain is glycosylated only at position 394, and the other is glycosylated at both 394 and

371. The occurrence of only a single species of

X' -hlgE-Fc (Ala 225, Gin 371) (Figure 7, lane Bl) confirms that the Asn 265 is likely to be fully glycosylated.

Example 2

Production of X'-hlgE-Fc fragments by a stable cell line

Permanent CHO cell lines secreting either the wild type or double mutant X'-hlgE-Fc (Ala 225) fragments (X'- hIgE-Fc(Ala 225) and X' -hlgE-Fc (Ala 225, Gin 265, Gin 371)) were established, and cell lines secreting the X'-hlgE-Fc proteins with highest abundance were cloned by limiting dilution. In order to isolate the two products, cells were grown in roller bottles for 4 days at 37°C. In a preliminary experiment, the accumulation of each of the fragments was monitored by electrophoresis of harvest medium in SDS (sodium dodecyl sulphate) polyacrylamide gels and Western blotting. It was found that the X'-hlgE-Fc proteins accumulated to 2mg/l without the appearance of degradation products, as detected by the aforementioned ELISA assay (standardised against purified myeloma IgE (PS)) , in 4 days. After this time, the concentration of the intact protein decreased, and specific degradation products began to accumulate . Example 3

An NS-0 cell line secreting ' -hlgE-Fc (Ala 225, Gin 265, Gin 371) was also established. For expression in NS-0 cells, the X' -hlgE-Fc (Ala 225, Gin 265, Gin 371) mutant construct was subcloned into a pEE6 based expression vector containing glutamine synthetase cDNA as a selectable marker according to the technique described by Bebbington et al . ,

Biotechnology 10 : 169 (1992) . 10 ⁷ NS-0 cells in the exponential phase of growth were pelleted at 1500 RPM for 5 minutes and were washed twice with a solution of ice-cold phosphate-buffered saline (0.15 M NaCl, 0.05 M phosphate buffer, pH 7.6) . The cells were resuspended in a final volume of 0.7 ml of ice-cold PBS. From this stage onwards of the transfection procedure the cells were maintained on ice. 40 μg of the linearised expression vector was mixed together with the suspension of NS-0 cells in a 0.4 cm electroporation cuvette (Bio-Rad) and left on ice for 5 minutes. Electroporation of the cells was performed using a Gene Pulser (Bio-Rad) using two consecutive 0.1 sec. pulses of 1500 V at a capacitance of 3 μF. The cuvette was returned to ice for a further 2-5 minutes before the electroporated cells were added to 40 ml of non-selective media. 30 ml of this was plated out over three 96 well culture dishes, the rest was diluted a further three times by a factor of four. Each dilution was plated out over three 96 well culture dishes. The cells were allowed to recover overnight and the next day 100 μl of gDMEM selective medium was added to the cells. Resistant colonies appeared after about four weeks after the addition of selection to the transfected cells. Single colonies able to grow in glutamine-free media were screened using the anti-human IgE ELISA (see below) and the best producers were expanded.

A suspension adapted form of this line, growing in a serum replacement medium, was able to accumulate up to approximately lOOmg of product (i.e. X' -hlgE-Fc (Gin 265, Gin 371) ) per litre of culture supernatant (as determined by the aforementioned anti-human IgE ELISA, standardised against purified X' -hlgE-Fc (Gin 265, Gin 371)) .

The X'hlgE-Fc proteins were purified from the harvest medium by affinity chromatography on an anti-hlgE-Fc matrix and examined for purity by electrophoresis on SDS

polyacrylamide gels (Figure 8) . Under non-reducing conditions, essentially all of the protein was recovered in the form of disulphide-linked dimers . Under reducing conditions, monomers of the expected size were found. Example 4

To scale up the production of protein from NS-0 cell lines required the adaptation of the cell line to suspension culture. The NS-0 cell line was also adapted to growth in gDMEM containing a serum replacement (Celltech Ltd) . To adapt transfected cells to suspension culture the cells were grown in 490 cm ² roller bottles and were regularly passaged before they reached confluency. Adaptation to serum replacement media was achieved by gradually reducing the proportion of FCS in the media each time the cells, growing in suspension, were passaged. The supernatant media was harvested from the cells at the time that they became confluent . Example 5

Nine amino acids were sequenced from the N-terminus of X'-hlgE-Fc (Gin 265, Gin 371) and the sequence was determined to be:

[SEQ ID No: 3]

Asp lie Val Ala Ser Xaa Asp Phe Thr The sixth residue, Xaa, which should have been Arg, could not be unambiguously assigned. A comparison of the observed and predicted sequences (Figure 3) indicates that the fragment has the correct N-terminal amino acid, consistent with accurate leader sequence processing. Thus Cys 225 has been successfully replaced by Ala. The rest of the determined N-terminal sequence is as expected. Example 6

IgE(WT) , i.e. IgE isolated from a myeloma of a particular patient (designated "WT") , or the wild type X'-hlgE-Fc or mutant X'-hlgE-Fc protein was labelled with

[ ¹²⁵ I] iodine to a specific activity of 0.5 - 1 x 10 ⁹ cpm/mg protein using chloramine T (see McConahey et al . , Methods in Enzym. , 70:213 (1980)) . The fraction of functional X'-hlgE-Fc protein and IgE (WT) was determined from the percentage of molecules in the purified samples of X'-hlgE-Fc and X' -hlgE-Fc (Gin 265, Gin 371) bound to an excess of cells of a stable CHO cell line expressing recombinant human FceRI (CHO-hFceRI) (see Wang et al . , J. Exp. Med., 175:1353 (1992). The method of Kulczycki Jr. et al., J. Exp. Med., 140:1676 (1974) was used. The concentrations of the X'-hlgE-Fc protein and IgE (WT) used in all binding assays was corrected for the fraction of functional molecules, typically in the range of 30 - 70%.

The affinity of the ligands, i.e. X'-hlgE-Fc, its mutants and IgE(SF25) , for FceRI was measured from the forward and reverse constants for the rates of reaction with cell-bound FceRI. The value of k ₊₁ was determined by measuring the concentration of ligand bound to cells as a function of time (see Kulczycki, Jr. et al . , J. Exp. Med., 140:1676 (1974)) . A curve fitting program (ORIGIN Ver. 3, Microcal Software, MA,USA) was used to obtain the best fit to the equation below, relating the bound ligand to k and

"-1 k _+] _ [L] [R _τ ] k ₊₁ [L] [R _τ ] - (k ₊₁ [L] ÷k _. - t

[RL] = k ₊₁ [L] + k_ ₁ k ₊₁ [L] + _ _±

where [RL] is the concentration of receptor-ligand complex, [R _τ ] is the total receptor concentration, and [L] is the total ligand concentration. The value of k was determined by the method of Kulczycki et al . , J. Exp. Med. 140:1676 (1974) .

The kinetics of binding of X'-hlgE-Fc and X ¹ -hlgE-Fc (Gin 265, Gin 371) to FceRI were measured using the CHO-hFceRI cell line. For comparison the kinetics of

the CHO-hFceRI cell line. For comparison the kinetics of binding of IgE(SF25) were also measured. The rates of association for IgE(WT) , X'-hlgE-Fc, and X' -hlgE-Fc (Gin 265, Gin 371) were 3.1 ± 0.3 x 10 ⁵ M^s ^"1 (n = 3) , 8.1 ± 0.8 x 10 ⁵ M ^_1 s ^_1 (n = 6) , and 9.9 ± 1.1 x 10 ⁵ M^s ^"1 (n = 4) . The corresponding rates of dissociation were 0.9 ± 0.2 x 10 ^"5 M^s ^"1 (n = 6), 1.0 ± 0.1 x 10 ^"5 s ^_1 (n = 6) and 2.0 ± 0.2 x 10 ^"5 M^s ^"1 (n = 6) (Table I) . The K values calculated from the average k ₊₁ and k_ ₁ values, are listed in Table I.

Experiments were also carried out to measure binding of X' -hlgE-Fc (Gin 265, Gin 371) to FceRII. 5 x 10 ⁵ RPMI 8866 cells were added to mixtures of [ ¹²⁵ I] -labelled IgE, wild type X'-hlgE-Fc, or mutant X' -hlgE-Fc (Gin 265, Gin 371) at 2μg/ml (10.2 nM and 27.8 nM respectively) and increasing amounts of the unlabelled sample. The amount of cell-bound labelled ligand was determined as described above for the FceRI binding study. The K values were calculated from the IC ₅₀ values, determined by non-linear regression of the data (using the GraphPad-Inplot software package) , using the following equation:

K _i (inhibition constant) = IC ₅₀ /(1 + [L*] /K _d ) ; the Kι. can be assumed to be the same as the Kα, if the labelled and unlabelled ligand are the same. Therefore

^K _a = ^(IC ₅₀ " tL*]. - ¹ -

[L*] = concentration of labelled ligand, corrected for the percentage bindability (as described above) .

TABLE 1

Protein assayed FceRI FceRII k+,1(M ^' V ¹ ) ^k r i ^{i S ' l ]} K (M ^"1 ) a K (M ^"1 ) a

X'-hlgE-Fc 3.1±0.3 X 10 (3) 0.9±0.2 X 10 ^" (6) 3.4 X 10 ¹⁰ 7.3±0.6 x 10

6 hlgE 8.1±0.8 X 10 ⁵ (6) l.O±O.l X 10~ ⁵ (6) 8.1 X 10 ¹ 4.1±1.5 X 10 X'-hlgE-Fc (Gin 265, Gin 371) 9.9±1.1 X 10 ⁵ (4) 2.0±0.2 X lθ ^"5 (6) 5.0 X 10 ¹⁰ 3.2±2.4 x 10 ⁷

Notes :

1. : was calculated as k ₊₁ /k_-_.

2. The number of experiments performed is given in brackets.

Example 7 _.

K_, values for the association of IgE, wild type

X'-hlgE-Fc and the X' -hlgE-Fc (Gin 265, Gin 371) mutant with the RPMI 8866 cell's FceRII receptor were obtained from competition curves. The K _a values calculated from the competition curves are 4.1 x 10 ^δ M ^-1 and 3.2 x 10 ⁷ M ^"1 for X'-hlgE-Fc and X' -hlgE-Fc (Gin 265, Gin 371) respectively, as compared with the value of 7.3 x 10 ⁷ M ^"1 for IgE (WT) , averaged over four independent experiments, each performed in duplicate. Example 8_

Effector function of the wild type X'-hlgE-Fc and double mutant recombinant X' -hlgE-Fc (Gin 265, Gin 371) was tested by the ability of anti-IgE to trigger histamine release from human basophils sensitised with the X'-hlgE-Fc proteins. Preliminary experiments showed that maximal histamine release was obtained when cells were sensitised with IgE (PS) or X'-hlgE-Fc proteins at concentrations of approximately 30nM or higher. The histamine release induced by various dilutions of anti-IgE from basophils sensitised with 60 nM IgE and the X'-hlgE-Fc proteins was studied.

Basophils sensitised with IgE and the X'-hlgE-Fc proteins showed similar response curves following anti-IgE stimulation, whereas only marginal release was observed from non-sensitised basophils. The results are plotted in Figure 9. Histamine release induced by an optimal dilution (1:1000) of anti-IgE from cells fully sensitised with IgE (PS) , X'-hlgE-Fc and X' -hlgE-Fc (Gin 265, Gin 371) was 57.5 ± 4.5%, 67.7 ± 5.1% and 62.0 ± 10.5% (mean ± SD, n=4, 3, 3) , respectively.

SEQUENCE LISTING

(1) GENERAL INFORMATION:

(i) APPLICANT:

(A) NAME: Dr. C T. Eyles, c/o W. P. Thompson & Co.

(B) STREET: 289-293 High Holborn

(C) CITY: London

(E) COUNTRY: United Kingdom

(F) POSTAL CODE (ZIP) : WC1V 7HU

(G) TELEPHONE: 0171-242-3524 (H) TELEFAX: 0171-831-0139 (I) TELEX: 2S8801

(ii) TITLE OF INVENTION: Polypeptide competitors for IgE receptor sites on cells.

(iii) NUMBER OF SEQUENCES: 11

(iv) COMPUTER READABLE FORM:

(A) MEDIUM TYPE: Floppy disk

(B) COMPUTER: IBM PC compatible

(C) OPERATING SYSTEM: PC-DOS/MS-DOS

(D) SOFTWARE: PatentIn Release #1.0, Version #1.30 (EPO)

(vi) PRIOR APPLICATION DATA:

(A) APPLICATION NUMBER: GB 9324013.3

(B) FILING DATE: 22-NOV-1993

(2) INFORMATION FOR SEQ ID NO: 1:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 323 amino acids

(B) TYPE: amino acid

(C) STRANDEDNESS:

(D) TOPOLOGY: linear

(ii) MOLECULE TYPE: peptide

(iii) HYPOTHETICAL: NO

(iv) ANTI-SENSE: NO

(xi) SEQUENCE DESCRIPTION: SEQ ID NO: 1:

Ser Arg Asp Phe Thr Pro Pro Thr Val Lys lie Leu Gin Ser Ser Cys 1 5 10 15

Asp Gly Gly Gly His Phe Pro Pro Thr lie Gin Leu Leu Cys Leu Val 20 25 30

Ser Gly Tyr Thr Pro Gly Thr lie Asn lie Thr Trp Leu Glu Asp Gly 35 40 45

Gin Val Met Asp Val Asp Leu Ser Thr Ala Ser Thr Thr Gin Glu Gly 50 55 60

Glu Leu Ala Ser Thr Gin Ser Glu Leu Thr Leu Ser Gin Lys His Trp 65 70 75 80

Leu Ser Asp Arg Thr Tyr Thr Cys Gin Val Thr Tyr Gin Gly His Thr 85 90 95

Phe Glu Asp Ser Thr Lys Lys Cys Ala Asp Ser Asn Pro Arg Gly Val 100 105 110

Ser Ala Tyr Leu Ser Arg Pro Ser Pro Phe Asp Leu Phe lie Arg Lys 115 120 125

Ser Pro Thr lie Thr Cys Leu Val Val Asp Leu Ala Pro Ser Lys Gly 130 135 140

Thr Val Asn Leu Thr Trp Ser Arg Ala Ser Gly Lys Pro Val Asn His 145 150 155 160

Ser Thr Arg Lys Glu Glu Lys Gin Arg Asn Gly Thr Leu Thr Val Thr 165 170 175

Ser Thr Leu Pro Val Gly Thr Arg Asp Trp He Glu Gly Glu Thr Tyr 180 185 190

Gin Cys Arg Val Thr His Pro His Leu Pro Arg Ala Leu Met Arg Ser 195 200 205

Thr Thr Lys Thr Ser Gly Pro Arg Ala Ala Pro Glu Val Tyr Ala Phe 210 215 220

Ala Thr Pro Glu Trp Pro Gly Ser Arg Asp Lys Arg Thr Leu Ala Cys 225 230 235 240

Leu He Gin Asn Phe Met Pro Glu Asp He Ser Val Gin Trp Leu His 245 250 255

Asn Glu Val Gin Leu Pro Asp Ala Arg His Ser Thr Thr Gin Pro Arg 260 265 270

Lys Thr Lys Gly Ser Gly Phe Phe Val Phe Ser Arg Leu Glu Val Thr 275 280 285

Arg Ala Glu Trp Glu Gin Lys Asp Glu Phe He Cys Arg Ala Val His 290 295 300

Glu Ala Ala Ser Pro Ser Gin Thr Val Gin Arg Ala Val Ser Val Asn 305 310 315 320

Pro Gly Lys

(2) INFORMATION FOR SEQ ID NO: 2:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 1308 base pairs

(B) TYPE: nucleic acid

(C) STRANDEDNESS : double

(D) TOPOLOGY: linear

(ii) MOLECULE TYPE: cDNA

(xi) SEQUENCE DESCRIPTION: SEQ ID NO: 2:

GTCTGCTCCA GGGACTTCAC CCCGCCCACC GTGAAGATCT TACAGTCGTC CTGCGACGGC 60

GGCGGGCACT TCCCCCCGAC CATCCAGCTC CTGTGCCTCG TCTCTGGGTA CACCCCAGGG 120

ACTATCAACA TCACCTGGCT GGAGGACGGG CAGGTCATGG ACGTGGACTT GTCCACCGCC 180

TCTACCACGC AGGAGGGTGA GCTGGCCTCC ACACAAAGCG AGCTCACCCT CAGCCAGAAG 240

CACTGGCTGT CAGACCGCAC CTACACCTGC CAGGTCACCT ATCAAGGTCA CACCTTTGAG 300

GACAGCACCA AGAAGTGTGC AGGTACGTTC CCACCTGCCC TGGTGGCCGC CACGGAGGCC 360

AGAGAAGAGG GGCGGGTGGG CCTCACACAG CCCTCCGGTG TACCACAGAT TCCAACCCGA 420

GAGGGGTGAG CGCCTACCTA AGCCGGCCCA GCCCGTTCGA CCTGTTCATC CGCAAGTCGC 480

CCACGATCAC CTGTCTGGTG GTGGACCTGG CACCCAGCAA GGGGACCGTG AACCTGACCT 540

GGTCCCGGGC CAGTGGGAAG CCTGTGAACC ACTCCACCAG AAAGGAGGAG AAGCAGCGCA 600

ATGGCACGTT AACCGTCACG TCCACCCTGC CGGTGGGCAC CCGAGACTGG ATCGAGGGGG 660

AGACCTACCA GTGCAGGGTG ACCCACCCCC ACCTGCCCAG GGCCCTCATG CGGTCCACGA 720

CCAAGACCAG CGGTGAGCCA TGGGCAGGCC GGGGTCGTGG GGGAAGGGAG GGAGCGAGTG 780

AGCGGGGCCC GGGCTGACCC CACGTCTGGC CACAGGCCCG CGTGCTGCCC CGGAAGTCTA 840

TGCGTTTGCG ACGCCGGAGT GGCCGGGGAG CCGGGACAAG CGCACCCTCG CCTGCCTGAT 900

CCAGAACTTC ATGCCTGAGG ACATCTCGGT GCAGTGGCTG CACAACGAGG TGCAGCTCCC 960

GGACGCCCGG CACAGCACGA CGCAGCCCCG CAAGACCAAG GGCTCCGGCT TCTTCGTCTT 1020

CAGCCGCCTG GAGGTGACCA GGGCCGAATG GGAGCAGAAA GATGAGTTCA TCTGCCGTGC 1080

AGTCCATGAG GCAGCGAGCC CCTCACAGAC CGTCCAGCGA GCGGTGTCTG TAAATCCCGG 1140

TAAATGACGT ACTCCTGCCT CCCTCCCTCC CAGGGCTCCA TCCAGCTGTG CAGTGGGGAG 1200

GACTGGCCAG ACCTTCTGTC CACTGTTGCA ATGACCCCAG GAAGCTACCC CCAATAAACT 1260

GTGCCTGCTC AGAGCCCCAG TACACCCATT CTTGGGAGCG GGCAGGGC 1308

(2) INFORMATION FOR SEQ ID NO: 3:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 9 amino acids

(B) TYPE: amino acid

(C) STRANDEDNESS:

(D) TOPOLOGY: linear

(ii) MOLECULE TYPE: peptide

(xi) SEQUENCE DESCRIPTION: SEQ ID NO: 3:

Asp He Val Ala Ser Xaa Asp Phe Thr 1 5

(2) INFORMATION FOR SEQ ID NO: 4:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 9 amino acids

(B) TYPE: amino acid

(C) STRANDEDNESS:

(D) TOPOLOGY: linear

(ii) MOLECULE TYPE: peptide

(xi) SEQUENCE DESCRIPTION: SEQ ID NO: 4

Lys Thr Phe Ser Val Cys Ser Arg Asp 1 5

(2) INFORMATION FOR SEQ ID NO: 5:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 27 base pairs

(B) TYPE: nucleic acid

(C) STRANDEDNESS: double

(D) TOPOLOGY: linear

(ii) MOLECULE TYPE: cDNA

(xi) SEQUENCE DESCRIPTION: SEQ ID NO: 5:

AAAACCTTCA GCGTCTGCTC CAGGGAC 27

(2) INFORMATION FOR SEQ ID NO: 6:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 8 amino acids

(B) TYPE: amino acid

(C) STRANDEDNESS:

(D) TOPOLOGY: linear

(ii) MOLECULE TYPE: peptide

(xi) SEQUENCE DESCRIPTION: SEQ ID NO: 6:

Cys Asp He Val Ala Ser Arg Asp 1 5

(2) INFORMATION FOR SEQ ID NO: 7:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 24 base pairs

(B) TYPE: nucleic acid

(C) STRANDEDNESS: double

(D) TOPOLOGY: linear

(ii) MOLECULE TYPE: CDNA

(xi) SEQUENCE DESCRIPTION: SEQ ID NO: 7:

TGTGATATCG TCGCCTCCAG GGAC 24

(2) INFORMATION FOR SEQ ID NO: 8:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 21 base pairs

(B) TYPE: nucleic acid

(C) STRANDEDNESS: double

(D) TOPOLOGY: linear

(ii) MOLECULE TYPE: cDNA

(xi) SEQUENCE DESCRIPTION: SEQ ID NO: 8 :

GGGACTATCC AGATCACCTG G 21

(2) INFORMATION FOR SEQ ID NO: 9:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 21 base pairs

(B) TYPE: nucleic acid

(C) STRANDEDNESS: double

(D) TOPOLOGY: linear

(ii) MOLECULE TYPE: cDNA

(iii) HYPOTHETICAL: NO

(iv) ANTI-SENSE: NO

(xi) SEQUENCE DESCRIPTION: SEQ ID NO: 9:

CCAGGTGATC TGGATAGTCC C 21

(2) INFORMATION FOR SEQ ID NO: 10:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 21 base pairs

(B) TYPE: nucleic acid

(C) STRANDEDNESS: double

(D) TOPOLOGY: linear

(ii) MOLECULE TYPE: cDNA

(xi) SEQUENCE DESCRIPTION: SEQ ID NO: 10:

GGGACCGTGC AGCTGACCTG G 21

(2) INFORMATION FOR SEQ ID NO: 11:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 21 base pairs

(B) TYPE: nucleic acid

- 43 -

(C) STRANDEDNESS: double

(D) TOPOLOGY: linear

(ii) MOLECULE TYPE: cDNA

(xi) SEQUENCE DESCRIPTION: SEQ ID NO: 11:

CCAGGTCAGC TGCACGGTCC C 21