AU629752B2 - Recombinant dna and process for the production of chimeric antibodies - Google Patents

Description

-r 629752

AUSTRALIA

PATENTS ACT 1990 COMPLETE SPECIFICATION NAME OF APPLICANT(S): Boehringer Mannheim GmbH ADDRESS FOR SERVICE: DAVIES COLLISON Patent Attorneys I Little Collins Street, Melbourne, 3000.

INVENTION TITLE: Recombinant DNA and process for the production of chimeric antibodies The following statement is a full description of this invention, including the best method of performing it known to me/us:- 0 o e *o 0 0o o 0 6 0 a t *tl 14 98 of the amino acids of the VJ region of MAB179 and 1,4 3a D e s c r i t i on The invention concerns recombinant DNA which codes for the light or heavy chain of an antibody whose constant regions are of human origin and whose variable regions are of non-human origin, expression vectors containing such a recombinant DNA, as well as processes for the production of the recombinant DNA or the expression vectors and finally processes for the production of the chimeric antibody.

The use of antibodies in therapy is becoming increasingly important. Thus for example antibodies I against the interleukin 2 (IL-2) receptor are of great therapeutic interest as immunosuppressants. The in vivo effectiveness of such antibodies could be demonstrated in a series of experimental animal models. Examples of this are described for heart and skin transplantation in the mouse by Kirkman et al., Transplantation Proceedings 19 (1987) 618-690 and T. Diamantstein et al., Transplantation Reviews 1 (1987), 177-196, for heart transplantation in the rat by J.W. Kupiec-Weglinski et al., Proc. Natl. Acad. Sci. USA 83 (1986) 2624, for the transplantation of islets of Langerhans in the rat by Hahn et al., Diabetologia 30 (1987), 40 and for kidney transplantation in primates by T. Diamantstein et al., Transplantation Reviews 1 (1987) 177-196. The successful preventive application of rat antibodies against the IL- 2 receptor in kidney transplantations has also been described for humans Soulillou et al., J. of Autoimmunity 1 (1988) 655-661; D. Cantarovitch et al., Am. J. of Kidney Disease 11 (1988) 101-106).

4.-1 nh'7n 4ci chri :n vnr in~ St rQ o4 4 4 ,444 4 4440 041 *a 9 0 049 444 44 4t 2 A disadvantage is, however, that mouse and rat antibodies are potent immunogens after injection into the blood of humans which can lead to neutralization of the therapeutic activity. There is therefore a great interest in the availability of hybrid mouse/human antibodies or humanized antibodies which promise to have a substantially reduced or even eliminated immunogenicity in humans Morrison et al., Proc.

Natl. Acad. Sci. USA 81 (1984) 6851-6855; P.T. Jones et al., Nature 321 (1986) 522-525).

Such hybrid antibodies have up to now mainly been produced by genetic engineering in which corresponding cDNAs are fused and expressed in suitable host cells.

However, a drawback of this is that the cDNAs coding for the hybrid antibodies are poorly expressed. In another production method, in order to construct such antibodies the genes for the light and heavy chain are usually first isolated from a phage library and characterized.

Afterwards the genomic DNA of the VJ region or the VDJ region is fused by genetic engineering to the corresponding genomic DNA of the constant regions of the light and heavy human chains cf. for this S.L.

Morrison et al., Proc. Natl. Acad. Sci. USA 81 (1984) 6851-6855; S.H. Boulianne et al., Nature 312 (1984) 641- 646; L.K. Sun et al., Proc. Natl. Acad. Sci. USA 84 (1987) 214-218; Y. Nishimura et al., Cancer Res. 47 (1987) 999-1005; B.A. Brown et al., Cancer Res. 47 (1987) 3577-3583).

Since as a rule 10 6 plaques have to be examined in order to find a gene with a copy number of 1, this second type of procedure is extremely laborious. Moreover the characterization of the genomic isolates (restriction mapping, differentiation of the ene structure in 16 Sequence of the oligonucleotide used for mutagenesis: 3 somatic and germ line cells, differentiation between abberantly and correctly rearranged genes) is extremely time-consuming.

The object of the present invention was therefore to facilitate the production of chimeric antibodies.

This object is achieved according to the present invention by a recombinant DNA which codes for the light or heavy chain of an antibody whose constant regions are of human origin and whose variable regions are of non- S" human origin, which is characterized in that it contains in the direction of transcription a cDNA sequence which codes for the variable nonhuman region, including the region J and, if desired, D, *0 an intron sequence which has a splice donor site at its 5' end and is composed of a non-human intron partial sequence at the 5' end and a human intron partial sequence at the 3' end and a genomic DNA sequence coding for the human t constant region.

The light and heavy chains for chimeric antibodies having the desired specificity can be produced in a surprisingly high yield with the aid of the recombinant DNA. In this connection an additional special advantage is that the easily obtainable cDNA sequence for the variable region of desired specificity (cDNA for antibody genes is easy to isolate since the corresponding mRNA is present in hybridoma cells as a large proportion of the total mRNA) can be connected via the intron sequence with an already subcloned genomic DNA segment which codes for the appropriate human constant regions and thus has a wide spectrum of 17 *17svnthPfqic t i-iin -i i I i Clurr ~"Lis4~Yi~ 4 applicability. Thus the invention offers the opportunity to easily produce recombinant DNA molecules which each have the DNA sequence encoding the corresponding specificity.

The cDNA sequence a) of the recombinant DNA according to the present invention includes the V and the J region for a light chain and in addition the D region between V and J for a heavy chain. The recombinant DNA according to the present invention contains with its components b) and c) at all sites at which introns are doCo naturally present in antibody genes corresponding DNA sequences except from locations at which introns occur oo a in the signal sequences in all light and heavy chains.

The expression of the recombinant DNA according to the 0 a0 present invention in suitable host cells leads to a much 0t oo higher expression of the desired antibody in comparison to processes which have been known up to now for the production of chimeric antibodies using cDNA expression.

S, In a preferred embodiment of the invention the sequence r a) as well as the non-human intron partial sequence of are of murine origin and the non-human intron partial sequence of b) is preferably 15 to 20 nucleotides long.

In a particularly preferred embodiment the non-human intron partial sequence of b) is a part of the intron sequence which naturally follows the sequence a) i.e. in the genomic DNA. However, within the scope of the invention other non-human intron sequences can also be used, preferably however of the same species and if possible also derived from a gene encoding the light or heavy chain of an antibody.

I~

18 In a further preferred embodiment of the invention the splice donor site in b) has the sequence GT.

According to the present invention a branch site is preferably located in the human part of the intron sequence which can e.g. correspond to one of the known sequences such as those described by Sharp, P.A., Science 235 (1987) 766-771. The branch site can, however, also be located in the murine part of the intron.

A promoter sequence is preferably also added to the Send of the 'Lecombinant DNA under the control of which the sequences coding for the antibody chains can be expressed. The promoters which naturally, i.e. as in the 0 o4 :o genome, exert control over the corresponding sequences of the recombinant DNA can e.g. be used for this.

However, preferably promoters are used which enable an increased, and if desired regulatable, expression. Such promoters are known to one skilled in the art. In a particularly preferred embodiment the recombinant DNA has the promoter of the cytomegalo-virus.

In addition a further preferred embodiment of the invention is a recombinant DNA whose sequence a) codes for the variable region of the light or heavy chain of an antibody capable of specific binding to the interleukin 2 receptor. In addition the human sequence c) of the recombinant DNA according to the present invention coding for a heavy chain preferably codes for the constant region of the heavy chain of a human antibody of the IgG type.

I

I

19 '-IYilXUlrU-LF--- 6 Such recombinant DNAs are contained in the plasmids pKchim. and pv chim. whereby the plasmid pKchim.

contains a DNA sequence coding for the light chain, i.e.

kappa chain and the plasmid p echim. carries a recombinant DNA coding for a heavy chain of the type which together code for an antibody of the IgG type whose specificity is directed against the alpha chain of the human interleukin 2 receptor.

The invention also provides an expression vector which contains one of the recombinant DNAs according to the present invention as well as all necessary elements for the expression of this recombinant DNA. The elements 0 F which are necessary for this or which are preferably used are promoters, regulatory sequences and suchlike.

Expression vectors are well-known to one skilled in the art and are described for example in more detail by E.L.

Winnacker in "From Genes To Clones", 1987, Verlag Chemie, Weinheim. With the expression vectors according to the present invention it is possible to incorporate a recombinant DNA according to the present invention into Ssuitable host cells and to express it there.

Particularly preferred expression vectors which contain recombinant DNA according to the present invention and which are also the subject matter of the invention are the already mentioned plasmids pKchim. and p -chim.

The invention in addition provides a process for the production of a recombinant DNA according to the present invention in which polyA+ RNA is isolated from a hybridoma cell line secreting an antibody of the desired specificity, a cDNA library of this hybridoma cell line is prepared in suitable vectors and examined for clones which contain the cDNA for antibody chains by hybridization with oligonucleotides which are 1 20 7 complementary to the DNA coding for the constant part of the antibody, the V, J, and if desired D, sequences are isolated, a splice donor site, a part of an antibody intron sequence of non-human origin and a restriction cleavage site adjacent to the respective J domain in the direction of transcription are introduced by site-directed mutagenesis and the DNA obtained is ligated with a genomic DNA sequence which codes for the constant region of the light or heavy chain of a human antibody and has a part of a corresponding human intron sequence at its end.

Io The isolation of the polyA RNA, cDNA synthesis, as well So as the establishment of a cDNA library are carried out according to known methods such as those described e.g.

in Maniatis et al., Molecular Cloning: a Laboratory So.. Manual, 1982, Cold Spring Harbor. The selection of cDNA clones coding for antibody chains is also carried out in a known manner, i.e. by examining for clones which hybridize with an oligonucleotide wJhich is complementary 4 to a DNA which codes for Lhe constant region of an antibody which is derived from the same hybridoma cell species.

The site-directed mutagenesis for the introduction of a splice donor site and a unique restriction enzyme cleavage site is also well-known and is carried out according to the method which was described by Morinaga et al., Bio/Technology 2 (1984) 636-639.

In a preferred embodiment of the invention a mouse or rat hybridoma cell is used. The establishment of hybridoma cells which secrete antibodies of the desired specificity is carried out according to the method which was first described by Kbhler and Milstein (Nature 256 21 r i 2 21 -I I I -1 I- L 9 9 9*P 4, 9 *9 r* 0 8 (1975), 495) and which was subsequently developed further. These techniques are very familiar to one skilled in the art. In the process according to the present invention it is particularly preferable to use a mouse hybridoma cell which secretes antibodies directed against the alpha chain of the human interleukin 2 receptor.

For the introduction of the non-human intron partial sequence, a 15 to 20 base pair sequence is preferably used. In this case it is particularly preferably to use the partial intron sequence which follows in the authentic gene. In addition it is preferred to introduce the nucleotide sequence GT as splice donor site.

Finally it is preferred according to the present invention to use a human genomic DNA whose intron partial sequence contains a branch site.

In order to obtain a recombinant DNA which can be expressed well, a promoter sequence is preferably placed in front of the DNA sequence which is finally obtained or the manipulations which are necessary for the process according to the present invention are carried out from the beginning in a vector which contains a promoter in a suitable position. In this connection it is particularly preferable to use the promoter of the cytomegalo-virus genome.

In yet another preferred embodiment of the invention a human genomic DNA sequence which codes for the constant region of the light or heavy chain of an antibody of the IgG type is used for the construction of a recombinant DNA coding for a light or heavy chain.

*4 9 9 9*09 I 1 22 4

I

I

oo 9 The invention also provides a process for the construction of an expression vector according to the present invention in which a recombinant DNA according to the present invention is inserted into a suitable vector for the expression of a foreign gene. As already set forth above such vectors are known to one skilled in the art and contain all necessary components such as e.g. polylinker for inserting the recombinant DNA, antibiotic resistance genes in order to select colonies which contain the recombinant DNA, an origin of replication, regulatory elements, a promoter, if the recombinant DNA does not already have one, as well as, if desired, further components such as e.g. suppressor genes for selection of colonies carrying plasmid DNA.

The invention in turn also provides a process for the production of a chimer'z antibody whose variable regions are of non-human origin and whose constant regions are of human origin in which in this case one each of an expression vector which contains a recombinant DNA according to the present invention which codes for the light chain and an expression vector which contains a recombinant DNA according to the present invention which codes for the heavy chain of the antibody are introduced into suitable host cells, stable transformants are isolated and the antibodies are isolated from the culture supernatant of the cells according to known methods. In this process a non-producer hybridoma cell is preferably used as the host cell, the cell line Sp2/O-Agl4 (ATCC CRL 8922) is particularly preferred. In the process according to the present invention it is preferable to carry out an electroporation (Nucl. Acids Res. 15 (1987), 1311-1326, Bio Techniques 6 (1988), 742- 751) in order to introduce the vectors into the host cells, if desired with linearized DNA molecules.

I~

S

23 I0-- S. However, other processes known to one skilled in the art can also be used for the transfection of the host cells.

The process according to the present invention enables antibodies to be produced in a simple manner with a variety of specificities. These antibodies are not immunogenic or only very weakly immunogenic when applied therapeutically in humans since the constant regions are of human origin. Nowadays there are relatively simple methods to produce e.g. mouse hybridoma clls which secrete antibodies of a desired specificity and to isolated cDNA from them, this cDNA can be easily introduced into previously prepared vectors according to the process according to the present invention which DO each contain the genomic human part for the constant Poo sregions of the light or heavy chain. Thus it is possible o to obtain chimeric antibodies of the desired specificity in each case by simple ligation of the desired cDNA fragment into the appropriately prepared vectors and expression of recombinant DNA on both vectors in a host cell. The production of chimeric antibodies is not only much more simple than with the methods known hitherto t A r but the yields are tiso greatly increased.

til The invention therefore also provides the chimeric antibodies MAB 179, MAB 215 and MAB 447 whose sequences for the light and heavy chains are shown in the sequence protocols SEQ ID NO:1 to 6 (the sequences of the variable regions are shown, the sequences of the constant regions are known and described e.g. in Sequences of proteins of immunological interest; E.

Kabat, T. Wu, M. Reid-Miller, H. Perry and K. Gottesman, US Department of Health and Human Services, 1987, p.

282-325) and whose production is described in the 24 -ro 11 examples. All three antibodies are specific for the alpha chain of the human IL 2 receptor.

The invention is elucidated further by the following examples in conjunction with the sequence protocols and figures.

SEQ ID NO.:1 4t 4 9999 4 49c4 49) 9 9 49 4) 49 4 94 9 9 4 9 99 99 9t 9 999 SEQ ID NO.:2 shows the DNA and amino acid sequence of the variable region as well as the beginning of the constant region of the light chain of MAB 179, shows the DNA and amino acid sequence of the variable region as well as the beginning of the constant region of the light chain of MAB 447 shows the DNA and amino acid sequence of the variable region as well as the beginning of the constant region of the light chain of MAB 215, to 6 show the sequences of the respective variable regions of their corresponding heavy chains; SEQ ID NO.:3

I

Ir I SEQ ID NO.:4 Fig. 1 shows diagrammatically the isolation of a murine cDNA fragment coding for the variable region of the light immunoglobulin chain and the procedure for the site-directed mutagenesis; 24a- -1 J 12 12 Fig. 2a shows the insertion of the variable murine region of the light immunoglobulin chain into an expression vector; Fig. 2b shows diagrammatically the assembly of the murine and human antibody coding sequences for the light chain; Fig. 3 and 4a as well as 4b show a diagram of the same process for the DNA of the heavy immunoglobulin Schains.

00 E x a mp 1 e 1 0 oo So 1 Cloning and sequence analysis of the light and heavy chains of MAB179, 447 and 215 RNA was prepared from the hybridoma lines which secrete Sthe monoclonal antibodies 179, 447 or 215 (Cleary, et S. al.: Cell 44 (1986) 97-106) and polyA RNA was isolated I from it (Maniatis, et al., Molecular Cloning: A S' Laboratory Manual, (Cold Spring Harbor Lab., Cold Spring Harbor, NY, 1982) using an oligo-dT cellulose column (Collaborative Research, Bedford, MA). A cDNA library S was constructed for each hybridoma cell line with a C 1 rNA cloning kit from Pharmacia according to the instructions of the manufacturer (analogous to Gubler, U. and Hoffman, Gene (1983) 25, 263). This cDNA library comprised about 10000 recombinants for each hybridoma line.

The following primer was used to screen for clones of the kappa chain: 5'CCCGACTACGACGT3' S- 25 SEQ ID NO: 1 (light chain clone 179) TYPE OF SEQUENCE: Nucleotide with corresoondi nrr I i; I e 13 The following primer was used to screen for clones of the t chain (g-t and r2b): 5'CAGATAGGTGACCGG3' All the heavy chains of mouse immunoglobulin genes are detected with this primer (Sablitzky, F. and Rajewski, EMBO J. 3 (1984) 3005-3012).

The cDNA library was established in the vector pT7T318U (Pharmacia). By this means it was possible to regain the insertions as EcoRI fragments. By analysis with restriction endonucleases it was found that clones of complete length were obtained for the heavy and light chains of MAB179, 447 and 215. The relevant regions were subcloned in M13 vectors and sequenced (Sanger, et al.: Proc. Natl. Acad. Sci. USA 74 (1977) 5463-5467).

4 4 The sequences of the variable regions are shown in the go sequence protocols SEQ ID NO:1 to NO:6. In this case it can be seen that MAB179 and 447 use the same VL and VH segments for the light and the heavy chains.

SBoth kappa chains use the J1 region and have identical nucleotides at the VJ junctions which indicates that the VJ region of both kappa genes has the same origin. Both g-1 chains have the same D region, the same J region (JH3) and an identical VDJ recombination from which it can be deduced that both VDJ regions of the -1 genes have the same origin.

The comparison of the amino acid sequences of the light chains shows 3 amino acid substitutions in the V regions of the kappa cDNA of MAB179 and MAB447. The substitutions are'as follows: (Thr to Ser, amino acid (Lys to Arg, amino acid 45), (Asn to Lys, amino acid 53).

26 SEQ ID NO: 2 (light chain clone 447) TYPE OF SEQUENCE: Nucleotide with corresponding protein LENGTH OF SEQUENCE: 432 base pairs -14- 98 of the amino acids of the VJ region of MAB179 and 447 are identical.

There are 6 amino acid substitutions in the V regions of the 1l cDNA of MAB179 and MAB447. The substitutions are as follows: (Val to Ala, amino acid 23; Gly to Ser, amino acid 56; Ile to Val, amino acid 58; Thr to Arg, amino acid 63; Lys to Arg, amino acid 65; Gin to Glu, amino acid 82).

94 of the amino acids of the VJ regions of MAB179 and 447 are identical.

a The data suggest that the differences in the amino acid o sequences of MAB179 and MAB447 are due to somatic mutations.

0C °The initiation codon (Met) start of the signal sequence is shown in the sequence protocols for kappa and l cDNA of MAB179 and 447. The J or DJ regions are also shown as well as the beginning of each of the constant regions.

SThe sequences of the variable regions for the kappa and the t2b chain of the non-inhibitory (with respect to IL 2 binding) antibody MAB215 are shown in SEQ ID NO:3 and NO:6.

The J4 region is used for the kappa chain. A V segment is used which greatly differs from that of the light chains of MAB179 and 447.

Only 51 of the amino acids are identical compared to the V segment of 179.

27 SEQ ID NO: 3 (light chain clone 215) TYPE OF SEQUENCE: Nucleotide with corresponding protein LENGTH OF SEQUENCE: 435 balsp I I 15 The sequence of the '2b chain is shown in SEQ ID NO:6.

A V segment is used which clearly differs from that of the heavy chain of the clones MAB179 and MAB447. The D region (comprising seven amino acids) as well as the J region (JH 3 and the beginning of the constant region are shown. Only 57 of amino acids are identical compared to the V segment of the heavy chain of the clone MAB179.

Example 2: Chimerization of the light chain of MAB179 and o+o construction of an expression vector for the chimerized light chain a o a) Introduction of splice donor, intron and NotI 0 sequences downstream of the VJl region of the kappa "chain of MAB179 (Fig. 1).

The cDNA for the light chain of the murine MAB 179 was subcloned into pUCI6 (Yanisch-Perron, et Gene 33 (1985) 103-109) as an EcoRI-HpaI l fragment giving rise to plasmid pUCk. An EcoRI-NotI Sfragment which can be cloned (and is portable) via EcoRI and NotI ends is produced by mutagenesis.

This fragment contains the VJ1 region and is followed by 20 nucleotide intron sequences starting with the splice donor sequence GT of the genomic Ji segment of the mouse kappa chain (Max, et al., J. Biol. Chem. 256 (1981) 5116-5220) and a recognition sequence (GCGGCCGC) for the restriction endonuclease NotI. The mutagenesis procedure (see above) is carried out according to Morinaga, et al.

(Bio/Technology 2 (1984) 636-639).

II

28 SEQ ID NO: 4 (heavy chain clone 179) TYPE OF SEQUENCE: Nucleotide with corresponding protein ij LENGTH OF SEQUENCE: 531 base pairs I "t~i;r 16 Sequence of the oligonucleotide used for mutagenesis: splice donor sequence GTAAGTAGAATCCAAAGTCT GCGGCCGC GGGCTGATGCT 3' Jl J1 intron NotI mouse kappa constant region Two fragments were isolated from pUCk for the o heteroduplex formation.

Fragment A: pUCk was cleaved with the restriction i enzymes EcoRI and BamHI, both fragments were i separated by agarose gel-electrophoresis and the I ,large EcoRI-BamHI fragment was isolated from the gel. This fragment contains no mouse kappa K sequences.

Fragment B: pUCk was treated with NaeI (unique Scleavage site in the pUC part), the 5' phosphate residues were removed by treatment with alkaline phosphatase and subsequently the fragment was purified twice on a 0.7 agarose gel.

Fragment A, fragment B (500 fmol each) and the oligonucleotide (80 fmol) were mixed for heteroduplex formation and incubated at 100 0 C for three minutes with 50 mmol/l NaCl, 10 mmol/l Tris- HCl, pH 7.5, i0 mmol/l MgS0 4 and subsequently transferred onto ice. Afterwards the DNA was renatured (20 min at 600C). For the repair S 29 SEQ ID NO: 5 (heavy chain clone 447) TYPE OF SEQUENCE: Nucleotide with corresponding protein LENGTH OF SEQUENCE: 531 base Dair.

17 synthesis the reaction mixture was adjusted to: deoxynucleoside triphosphate (0.25 mmol/1), ATP (1 mmol/1), NaCl (100 mmol/1), Tris-HCl pH mmol/1), MgC12 (8 mmol/1), B-mercaptoethanol (1 mmol/l), Klenow fragment of the DNA polymerase from E. coli (0.125 U/Al preparation) and T4 ligase (0.1 U/gl preparation) and incubated at 16 0 C for 4 hours. Afterwards E. coli HB101 was transformed with the reaction mixture and colonies were selected on agar plates containing ampicillin gg/ml). Those colonies in whose plasmid DNA the above mentioned oligonucleotide was incorporated Scould be identified with the aid of the colony hybridization technique (Maniatis et al.: Molecular Cloning. A Laboratory Manual. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY 1982). The I characterization was first carried out by analysis with restriction endonucleases. Subsequently the desired change in the sequence was confirmed by sequencing (Sanger, et al.: Proc. Natl. Acad.

Sci. USA 74 (1977) 5463-5467).

b) Construction of an expression vector for the chimerized kappa chain of MAB179 (Fig. 2a and 2b).

pcDNAl (Invitrogen, San Diego; Aruffo, A. and Seed, B. (1987) Proc. Natl. Acad. Sci. USA 84 8573-8577) served as the starting plasmid. It contains the promoter of the human cytomegalo-virus, promoters of the phages T7 and SP6, a polylinker, splice and polyadenylation sequences from SV40, a supF tRNA for the selection as well as "origins of replication" from M13, ColE1, SV40 and polyoma. For cloning reasons the NdeI site shown is converted into a PvuI site pcDNAl-PvuI) by ligation of a ml I 'S 30 SEQ ID NO: 6 (heavy chain clone 215) TYPE OF SEQUENCE: Nucleotide with corresponding protein LENGTH OF SEQUENCE: 549 base pairs I *L-iCi-1. C C_ I li_.i.l. I i L 18 corresponding linker. The latter plasmid was cleaved with EcoRI and NotI in the polylinker and the EcoRI-NotI fragment of the DNA obtained according to (ca 400 bp) containing the VJ1 region and intron sequences of MAB179 kappa was ligated into the vector. The ligation mixture was transfected into the E. coli strain MC1061/P3 (Aruffo, A. and Seed, Proc. Natl. Acad. Sci.

USA 84 (1987) 8573-8577); Seed, Nucl. Acids Res. 11 (1983) 2327-2445) and ampicillin-resistant colonies were isolated on agar plates. The plasmid formed is denoted pcDNAk. pcDNAk was then cleaved with PvuI and NotI and the shorter fragment was I isolated on a low-melting agarose gel.

i* The plasmid p10195 (constructed according to EP-A 0 378 175) was also cleaved with NotI and PvuI *4 1 and the larger fragment (Fig. 2b) was isolated on a low-melting agarose gel. p10195 contains intron sequences of the mouse for the kappa chain as well as coding and non-coding regions of the constant human kappa region.

Both the fragments of pcDNAk and p10195 were ligated with T4 ligase, transfected into MC1061/P3 and ampicillin-resistant colonies were isolated.

The plasmid formed is denoted pKchim. The hybrid gene (VJ1 mouse Ckappa-human) is expressed under ,lt the control of the human CMV promoter (cytomegalovirus). An expression cassette for the phosphotransferase neo (inserted into pKchim. from p10195) allows selection of G418 resistant colonies after transfection into mammalian cells (Southern, P. and Berg, J. Mol. Appl. Genet. 1 (1982) 327- 341).

I

19 Example 3 Chimerization of the heavy chain of MAB179 and construction of an expression vector for the chimerized heavy chain.

a) Introduction of splice donor, intron and NotI sequences into the VDJ 3 region of the 1 chain of MABi79 (Fig. 3).

The cDNA for thetl chain untranslated region and coding region upstream of the BamHI site in the constant region) was subcloned as an EcoRI-BamHI fragment into pUC18 plasmid pUCl (Yanisch-Perron, et al.: Gene 33 (1985) 103-109).

t o A portable EcoRI-NotI fragment is prepared by Smutagenesis which contains the VDJ3 region of the 4 4' J ;mouse r1 chain followed by 20 nucleotide intron sequences starting with the splice donor sequence GT of the genomic J3 segment of the mouse y-1 chain (Sakano, et al., Nature 286 (1980) 676-682). This is followed by a recognition sequence (GCGGCCGC) for the restriction endonuclease NotI. The mutagenesis was carried out according to Morinaga et al. (Bio/Technology 2 (1984) 636-639). The sequence of the oligonucleotide used for mutagenesis: 31 THE CLAIMS DEFINING THE INVENTION ARE AS FOLLOWS: I t 4

II,'

ItI( V a a 20 splice donor sequence GTCTCTGCAG GTGAGTCCTAACTTCTCCCA GCGGCCGC TGCCTGGTCA 3' J3 J3 intron NotI mouse constant region Two fragments were isolated from pUCrl for heteroduplex formation.

Fragment C: pUCyl was cleaved with the restriction enzymes EcoRI and BamHI, the two fragments were separated by gel electrophoresis (1 agarose gel) and the large EcoRI-BamHI fragment was isolated from the gel. This fragment contains no mouse l sequences.

Fragment D: pUCl-1 was cleaved with NaeI (unique cleavage site in the pUC part), the 5' phosphate residues were removed by treatment with alkaline phosphatase and subsequently the fragment was purified twice on a 0.7 agarose gel. After mixing the fragments C and D (500 fmol each) and the oligonucleotide (80 fmol), the mutagenesis was carried out as described in Example 2. Those colonies in whose plasmid DNA the oligonucleotide described above was incorporated could be identified with the aid of the colony hybridization technique (Maniatis, et al: Molecular Cloning. A Laboratory Manual. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY 1982). The characterization was first carried out with restriction endonucleases. Subsequently the desired change in the sequence was confirmed by sequencing (Sanger, A I t t Sa.

W,

t ta t

B

ri 32 Recombinant DNA as claimed in one of the nrevinui i -21et al.: Proc. Natl. Acad. Sci. USA 74 (1977) 5463-5467). Fig. 3 shows diagrammatically the isolation of the desired EcoRI-NotI fragment.

b) Construction of the expression vector for the chimerized gi chain of MAB179 (Fig. 4a and 4b). The unique NdeI site in the vector pcDNA1 (Invitrogen, San Diego; Aruffo, A. and Seed, B. (1987) Proc.

Natl. Acad. Sci. USA 84, 8573-8577) was converted into a PvuI site by insertion of a corresponding linker. The plasmid which formed is denoted pcDNAl- Pvul.

pcDNAl-PvuI cleaved with EcoRI and NotI and the EcoRI-NotI fragment of pUCX-1 Mut (with VDJ3 and S,'intron sequences of rl) described in Fig. 3 were S.linked by ligation. The plasmid formed is t, designated pcDNA-l.

The plasmid p11201 (constructed according to EP-A 0 378 175) is cleaved with NotI and BamHI, the protruding ends filled in with Klenow, ligated with NotI linkers, cleaved again with NotI and the fragment shown in Fig. 4b is isolated on a lowmelting agarose gel. This fragment contains intron sequences of the heavy chains of the mouse, intron sequences of the human heavy chains as well as the Sgenomic equivalent of the constant region of the human rl gene. This fragment was ligated into the unique NotI site of the vector pcDNA-I. The correct orientation was determined by cleavage with restriction endonucleases. The expression vector which formed for the chimerized y-1 chain of MAB179 is denoted p &chim. Chimerized chains of other isotypes (such as e.g. A, t2, r-4, 1l, a2, 6, e) S33 33 1 1 P I;%--Tni H V-t.h 4m r c v-v rcF_\a 22 can also be constructed according to this principle.

Example 4: Establishment of permanent cell lines of non-producer hybridoma lines by electroporation with expression plasmids for the chimerized chains of MAB179.

The plasmids pKchim. and p-l chim. were mixed in equal amounts and transfected by electroporation into the immunoglobulin non-producer hybridoma line Sp2/0-Agl4 (ATCC CRT, 3923) (Ochi, et al., Proc. Natl. Acad.

Sci. USA 80 (1983) 6351-6355). All other nonimmunoglobulin-producer hybridoma cell lines such as e.g. P3X63-Ag8.653 (ATCC CRL 8375) are also suitable.

Both plasmids were linearized with the restriction endonuclease PvuI before transfection. The electroporation was carried out as described in Nucleic Acids Res. 15 (1987) 1311-1326 or Bio Techniques 6 (1988) 742-751. After centrifugation, the cells were washed with cold HeBS buffer (20 mmol/l HEPES, pH 7.05, 137 mmol/l NaC1, 5 mmol/l KCl, 0.7 mmol/l Na 2

HPO

4 6 mmol/l dextrose), resuspended with HeBS buffer, adjusted to a concentration of 106 cells/ml and placed on ice. After the addition of plasmid the cells were pulsed (conditions: capacity 500 hF and voltage range between 240 and 280 V, or 200 to 240 V at 160 pF). After pulsing, the cells were kept on ice for about 10 minutes and subsequently incubated at 37°C in medium I (RPMI 1640, 10 foetal calf serum, 2 mmol/1 glutamine, 1 mmol/l sodium pyruvate, 0.1 mmol/1 non-essential amino acids). The medium was changed 30 h after the transfection and incubated with medium I 800 gg/ml G418 after plating 103 cells per well in 96 well microtitre plates. 10 microtitre plates (96 wells each) i.

34 23 were prepared. 7-10 days after plating, G418 resistant colonies could be identified in the wells. Their culture supernatants were tested for reconstituted chimerized MAB179 as described in the following example. The best producers were propagated in mass culture in medium I.

Example Detection of reconstituted antibodies against the human IL-2 receptor Firstly a microtitre plate is coated with polyclonal antibodies against human Fcy-. For this the wells of a microtitre plate are incubated overnight at 4°C or for 1 hour at room temperature with 200 Al corresponding to 2.5 Ig of a polyclonal antibody against human Fc-(IgG) S, in 0.2 mol/l carbonate/bicarbonate, pH 9.5. After ,aspirating the wells they were incubated for 30 min to 1 h at room temperature with 300 Al 50 mmol/l HEPES, 0.15 mol/l NaCl/1 Crotein C, pH 7.0. Subsequently 200 Al calibration sample or culture supernatant of transfected cells was added and incubated for 1 h at room temperature while shaking (500 rpm) or for 90 min at room temperature without shaking. The calibration curve was established with a model chimeric antibody (produced by chemically cross-linking a human IgG MAB and Fab fragments of MAB179). After aspirating the wells they were each washed twice with 300 Al IB incubation buffer (50 mmol/l HEPES, pH 7.0, 0.15 mol/l NaCl, 0.2 mol/l di-sodium tartrate, 1 Crotein C, 0.75 PEG 40000 (Serva), 0.5 Pluronic F68 (Boehringer Mannheim), 0.01 phenol). Afterwards 200 Al of a solution of IL-2 receptor-POD-antibody conjugate complex was added (as described below) and incubated for 1 h at room temperature while shaking (500 rpm) or for 90 min at room temperature without shaking. The preformed complex 35 nA ,c f- r 4- 4 4: 4. a 24 consists of soluble interleukin 2 receptor and PODlabelled Fab fragments of the IL-2 receptor MAB215.

Formation of the complex solution: 20 pl of the PODlabelled Fab fragments of MAB 215 (150 U/ml) 19.5 ml incubation buffer (IB) (see above) 3 ml soluble IL-2 receptor standard (6400 U/ml; units defined via standard of T-cell Sciences). The soluble IL-2 receptor was obtained from culture supernatants of a recombinant mouse fibroblast cell line described by Shimizu et al.

(Mol. Biol. Med. 3 (1986) 509-520).

After carrying out the incubation the wells were rinsed three times with the wash buffer described above and subsequently the POD activity was determined with ABTS solution [2,2'-azino-di-3-ethylbenzthiazolinesulphonate] H 2 0 2 after incubating for 45-60 min at room temperature by reading the absorbance at 405 nm in an ELISA reader. Clones with high expression (up to gg/ml) were determined using this method.

rt 1 44 t t S It Itr ~I rI i It 4 1

'I

-36 29. Process as claimed in one of the claims 25 to 28, .r i i 24a- Microorcanism Denosits All microorganisms referred to hereunder have been deposited with the American Type Culture Collection (ATCC), 12301 Parklawn Drive, Rockville, Maryland 20852, United States of America.

Microorganism CCL 77 was deposited with the ATCC on 12th October, 1983 and accorded Accession No. ATCC CRL 8373.

Microorganism CRL 1484 was deposited with the ATCC on 12th October, 1983 and accorded Accession No. ATCC CRL 8374.

Microorganism CRL 1580 was deposited with the ATCC on 12th October, 1983 and accorded Accession No. ATCC CRL 8375.

Microorganism CRL 1581 was deposited with the ATCC on 30th October, 1985 and accorded Accession No. ATCC CRL 8923.

All of the abovementioned microorganisms are freely available from the ATCC on request and payment of a nominal fee to any person requesting samples of these microorganisms.

0 0 O I l r 4) rrt tLr oNI^^ Sec.

T

910621,PASDAT088,78243-9 dep,24 Vs 37 25 SEQ ID NO: 1 (light chain clone 179) TYPE OF SEQUENCE: Nucleotide with corresponding protein LENGTH OF SEQUENCE: 432 base pairs CHARACTERISTICS: aa -20 (Met): start of the signal sequence aa 1 (Asp): beginning of the V region from aa 96 (Arg) to aa 107 (Lys): Jl region from aa 108 (Arg): beginning of the C region ATG ATG GTC CTT GCT CAG TTT CTT GCA TTC TTG TTG CTT TGG TTT CCA 48 Met Met Val Leu Ala Gin Phe Leu Ala Phe Leu Leu Leu Trp Phe Pro -15 -10 GGT GCA AGA TGT GAC ATC CTG ATG ACC CAA TC' CCA TCC TCC ATG TCT 96 Gly Ala Arg Cys Asp Ile Leu Met Thr Gin Ser Pro Ser Ser Met Ser 1 5 GTA TCT CTG GGA GAC ACA GTC AGC ATC ACT TGC CAT GCA AGT CAG GGC 144 Val Ser Leu Gly Asp Thr Val Ser Ile Thr Cys His Ala Ser Gln Gly 20 ATC AGA AGT AAT ATA GTG TGG TTG CAG CAG AAA CCA GGG AAA TCA TTT 192 Ile Arg Ser Asn Ile Val Trp Leu Gin Gln Lys Pro Gly Lys Ser Phe 35 -t AGG GGC CTG ATC TAT CAT GGA ACC AAG TTG GAA GAT GGA GTT CCA TCA 240 Arg Gly Leu Ile Tyr His Gly Thr Lys Leu Gu Asp Gly Val Pro Ser 50 55 AGG TTC AGT GGC AGT GGA TCT GGA GCA GAT TAT TCT CTC ACC ATC AGC 288 Arg Phe Ser Gly Ser Gly Ser Gly Ala Asp Tyr Ser Leu Thr Ile Ser 70 AGC CTG GAA TCT GAA GAT TTT GCA GAC TAT TAT TGT GTA CAG TAT GCT 336 Ser Leu Glu Ser Glu Asp Phe Ala Asp Tyr Tyr Cys Val Gin Tyr Ala 85 CAG TTT CCT CGG ACG TTC GGT GGA GGC ACC AAG CTG GAA ATC AAA CGG 3PA SGln Phe Pro Arg Thr Phe Gly Gly Gly Thr Lys Leu Glu Ile Lys Arg 100 105 GCT GAT GCT GCA CCA ACT GTA TCC ATC TTC CCA CCA TCC AGT GAG CAG 432 Ala Asp Ala la Pro Thr Val Ser Ile Phe Pro Pro Ser Ser Glu Gin 110 115 120 -26 SEQ ID NO: 2 (light chain clone 447) TYPE OF SEQUENCE: Nucleotide with corresponding protein LENGTH OF SEQUENCE: 432 base pairs CHARACTERISTICS: aa -20 (Met): start of the signal sequence aa 1 (Asp): beginning of the V region from aa 96 (Arg) to aa 107 (Lys): J1 region from aa 108 (Arg): beginning of the C region ATG ATC GTC CTT GCT CAG TTT CTT GCA TTC TTG TTG CTT TGG TTT CCA 48 Met Met Val Lea Ala Gln Phe Leu Ala Phe Leu Leu Leu Trp Phe Pro -15 -10 GGT GCA AGA TGT GAC ATC CTG ATG ACC CAA TCT CCA TCC TCC ATG TCT 96 Gly Ala Arg Cys Asp Ile Leu Met Thr Gin Ser Pro Ser Ser Met Ser 1 5 GTT TCT CTG GGA GAC ACA GTC ACC ATC ACT TGC CAT GCA AGT CAG GGC 144 Val Ser Leu Gly Asp Thr Val Thr Ile Thr Cys His Ala Ser Gin Gly 20 ATT AGA AGT AAT ATA GTG TGG TTG CAG CAG AAA CCA GGG AAA TCA TTT 192 Ile Arg Ser Asn Ile Val Trp Leu Gin Gin Lys Pro Gly Lys Ser Phe 35 AAG GGC CTG ATC TAT CAT GGA ACC AAC TTG GAA GAT GGA GTT CCA TCA 240 Lys Gly Leu Ile Tyr His Gly Thr Asn Leu Glu Asp Gly Val Pro Ser 50 55 CGG TTC AGT GGC AGT GGA TCT GGA GCA GAT TAT TCT CTC ACC ATC AGC 288 Arg Phe Ser Gly Ser Gly Ser Gly Ala Asp Tyr Ser Leu Thr Ile Set 70 AGC CTG GXA TCT GAA GAT TTT GCA GAC TAT TAC TGT GTA CAG TAT GCT 336 Ser Leu Glu Ser Glu Asp Phe Ala Asp Tyr Tyr Cys Val Gin Tyr Ala 85 CAG TTT CCT CGG ACG TTC GGT GGA GGC ACC AAG CTG GAA ATC AAA CGG 384 Gin Phe Pro Arg Thr Phe Gly Gly Gly Thr Lys Leu Glu Ile Lys Arg 100 105 0 GCT GAT GCT GCA CCA ACT GTA TCC ATC TTC CCA CCA TCC AGT GAG CAG 432 Ala Asp Ala Ala Pro Thr Val Ser Ile Phe Pro Pro Ser Ser Glu Gin 110 115 120 I C c- 27 SEQ ID NO: 3 (light chain clone 215) TYPE OF SEQUENCE: Nucleotide with corresponding protein LENGTH OF SEQUENCE: 435 base pairs CHARACTERISTICS: aa -22 (Met): start of the signal sequence aa 1 (Lys): beginning of the V region from aa 95 (Phe) to aa 106 (Lys): J4 region from aa 107 (Arg): beginning of the C region ATG GAT TTT CAA GTG CAG ATT TTC AGC TTC CTG CTA ATC AGT GCT TCA 48 Met Asp Phe Gin Val Gin Ile Phe Ser Phe Leu Leu Ile Ser Ala Ser -15 GTC ATA ATG TCC AGA GGC AAA ATT GTT CTC TCC CAG TCT CCA GCA ATC 96 Val Ile Met Ser Arg Gly Lys Ile Val Leu Ser Gin Ser Pro Ala Ile 1 5 CTG TCT GCA TCT CCA GGG GAG AAG GTC ACA ATG ACT TGC AGG GCC AGC 144 Leu Ser Ala Ser Pro Gly Glu Lys Val Thr Met Thr Cys Arg Ala Ser 20 TCA AGT ATA AGT TAC ATG CAC TGG TAC CAG CAG AAG CCA GGA TCC TCC 192 Ser Ser Ile Ser Tyr Met His Trp Tyr Gin Gin Lys Pro Gly Ser Ser 35 CCC AAA CCC TGG ATT CAA GCC ACA TCC AAC CTG GCT TTT GGA GTC CCT 240 Pro Lys Pro Trp Ile Gln Ala Thr Ser Asn Leu Ala Phe Gly Val Pro 50 TCT CGC TTC AGT GGC AGT GGG TCT GGG ACC TCT TAC TCT CTC ACA ATC 288 Ser Arg Phe Ser Gly Ser Gly Ser Gly Thr Ser Tyr Ser Leu Thr Ile 65 AGC AGA GTG GAG GCT GAA GAT GCT GCC ACT TAT TAC TGC CAG CAG TGG 336 Ser Arg Val Glu Ala Glu Asp Ala Ala Thr Tyr Tyr Cys Gin Gin Trp 80 85 AGT AGT AAC CCA TTC ACG TTC GGC TCG GGG ACA AAG TTG GAA ATG AAA 384 Ser Ser Asn Pro Phe Thr Phe Gly Ser Gly Thr Lys Leu Glu Met Lys 100 105 CGG GCT GAT GCT GCA CCA ACT GTA TCC ATC TTC CCA CCA TCC AGT GAG 432 Arg Ala Asp Ala Ala Pro Thr Val Ser Ile Phe Pro Pro Ser Ser Glu 110 115 120 CAG 435 Gin 28 SEQ ID NO: 4 (heavy chain clone 179) TYPE OF SEQUENCE: Nucleotide with corresponding protein LENGTH OF SEQUENCE: 531 base pairs CHARACTERISTICS: aa -19 (Met): start of the signal sequence aa 1 (Asp): beginning of the V region from aa 99 (Asp) to aa 102 (Asn): D region from aa 103 (Trp) to aa 113 (Ala): J3 region from aa 114 (Ala): beginning of the C region ATG GAC TCC AGG CTC AAT TTA GTT TTC CTT GTC CTT ATT Val Leu lie Met Asp Ser Arg Leu Asn Leu Val GTC CAG TGT GAT GTG CAG CTG GTG Val Gin Cys Asp Val Gin Leu Val 1 5 Phe Leu -10 GAG TCT GGG GGA GGC Glu Ser Gly Gly Gly TGT GTT GCC TCT GGA Cys Val Ala Ser Gly CCT GGA Pro Gly GGG TCC CGG AAA Gly Ser Arg Lys CTC TCC Leu Ser 20 t sit~

AGT

Ser ACC TTT GGA Thr Phe Gly ATG CAC Met His 35 TAC ATT Tyr Ile TGG GTT CGT CAG Trp Val Arg Gin AGT AGT GGC AGT Ser Ser Gly Ser 55 GAG TGG GTC GCA Glu Trp Val Ala GAC ACA GTG AAG Asp Thr Val Lys ACC CTG TTC CTG Thr Leu Phe Leu GCT CCA GAG Ala Pro Glu 40 GGT ACC ATC Gly Thr Ile AGA GAC AAT Arg Asp Asn GGC CGA TTC ACC Gly Arg Phe Thr CAA ATG ACC AGT Gin Met Thr Ser 85 ATC TCC Ile Ser 70 TTA AAA GGT Leu Lys Gly TTA GTG CAG Leu Val Gin TTC ACT TTC Phe Thr Phe AAG GGG CTG Lys Gly Leu TAC TAT GCA Tyr Tyr Ala CCC AAG AAT Pro Lys Asn ACG GCC ATG Thr Ala Met ACT CTG GTC Thr Leu Val CCA CTG GCC Pro Leu Ala 125 GGA TGC CTG Gly Cys Leu 140 AAC TCT GGA Asn Ser Gly 155 48 96 144 192 240 288 336 CTA AGG TCT GAG GAC Leu Arg Ser Glu Asp AAC TGG GGC CAA GGG Asn Trp Gly Gin Gly 105 TAT TAC Tyr Tyr TGT GCA AGA GAT Cys Ala Arg Asp TGG ATG Trp Met 100 384

ACT

Thr 110 GTC TCT GCA Val Ser Ala GCC AAA Ala Lys 115 GCC CAA Ala Gin 130 CCT GGA TCT GCT Pro Gly Ser Ala ACG ACA CCC CCA Thr Thr Pro Pro ACT AAC TCC ATG Thr Asn Ser Met 135 GAG CCA GTG ACA Glu Pro Val Thr 150 TCT GTC TAT Ser Val Tyr 120 GTG ACC CTG Val Thr Leu GTG ACC TGG Val Thr Trp 432 480 528 GTC AAG GGC TAT TTC CCT Val Lys Gly Tyr Phe Pro 145

TCC

Ser 531 29 SEQ ID NO: 5 (heavy chain clone 447) TYPE OF SEQUENCE: Nucleotide with corresponding protein LENGTH OF SEQUENCE: 531 base pairs CHARACTERISTICS: aa -19 (Met): start of the signal sequence aa 1 (Asp): beginning of the V region from aa 99 (Asp) to aa 102 (Asn): D region from aa 103 (Trp) to aa 113 (Thr): J3 region from aa 114 (Ala): beginning of the C region ATG GAC TCC AGG CTC AAT Met Asp Ser Arg Leu Asn GTC CAG TGT GAT GTG CAA Val Gln Cys Asp Val Gin 1 CCT GGA Pro Gly GGG TCC CGG AAA Gly Ser Arg Lys I t 4 4t 41 4

AGT

Ser 30 ACC TTT GGA ATG CAC Thr Phe Gly Met His 35 GAG TGG GTC GCA TAT ATT Glu Trp Val Ala Tyr Ile GAC AGA GTG AGG GGC CGA Asp Arg Val Arg Gly Arg ACC CTG TTC CTG GAA ATG Thr Leu Phe Leu Glu Met 80 TTA GTG TTC CTT GTC CTT ATT TTA AAA GGT Leu Val Phe Leu Val Leu Ile Leu Lys Gly -10 CTG GTG GAG TCT GGG GGA GGC TTA GTG CAG Leu Val Glu Ser Gly Gly Gly Leu Val Gin 5 CTC TCC TGT GCA GCC TCT GGA TTC ACT TTC Leu Ser Cys Ala Ala Ser Gly Phe Thr Phe 20 TGG GTT CGT CAG GCT CCA GAG AAG GGG CTG Trp Val Arg Gln Ala Pro Glu Lys Gly Leu 40 AGT AGT GGC AGT AGT ACC GTC TAC TAT GCA Ser Ser Gly Ser Ser Thr Val Tyr Tyr Ala 55 TTC ACC ATC TCC AGA GAC AAT CCC AAG AAC Phe Thr Ile Ser Arg Asp Asn Pro Lys Asn 70 ACC AGT CTA AGG TCT GAG GAC ACG GCC ATG Thr Ser Leu Arg Ser Glu Asp Thr Ala Met 85 TGG ATG AAC TGG GGC CAA GGG ACT CTG GTC Trp Met Asn Trp Gly Gin Gly Thr Leu Val 100 105 ACG ACA CCC CCA TCT GTC TAT CCA CTG GCC Thr Thr Pro Pro Ser Val Tyr Pro Leu Ala 120 125 ACT AAC TCC ATG GTG ACC CTG GGA TGC CTG Thr Asn Ser Met Val Thr Leu Gly Cys Leu 135 140 GAG CCA GTG ACA GTG ACC TGG AAC TCT GGA Glu Pro Val Thr Val Thr Trp Asn Ser Gly 150 155 48 96 144 192 240 288 336 384 432 480 528 TAT TAC Tyr Tyr TGT GCA AGA GAT Cys Ala Arg Asp

ACT

Thr 110 GTC TCT ACA GCC AAA Val Ser Thr Ala Lys 115 CCT GGA TCT GCT GCC CAA Pro Gly Ser Ala Ala Gin 130 GTC AAG GGC TAT TTC CCT Val Lys Gly Tyr Phe Pro 145

TCC

Ser 531 30 SEQ ID NO: 6 (heavy chain clone 215) TYPE OF SEQUENCE: Nucleotide with corresponding protein LENGTH OF SEQUENCE: 549 base pairs CHARACTERISTICS: aa -19 (Met): start of the signal sequence aa 1 (Gln): beginning of the V region from aa 98 (Thr) to aa 104 (Ser): D region from aa 105 (Trp): to aa 119 (Ala): J3 region from aa 120 (Ala): beginning of the C region ATG GCT GTG CTG GGG CTG CTT CTC TGC CTG GTG ACT TTC CCA AGC TGT Met Ala Val Leu Gly Leu Leu Leu Cys Leu Val Thr Phe Pro Ser Cys -10 GTC CCG TCC Val Pro Ser CAG GTG CAG CTG AAG GAG TCA GGG CCT GGC CTG GTG GCG Gin Val Gin Leu Lys Glu Ser Gly Pro Gly Leu Val Ala 1 5 CCC TCA Pro Ser 15 CAG AGC CTG TCC ATC ACA TGC ACC GTC TCA GGG TTC TCA TTA Gln Ser Leu Ser Ile Thr Cys Thr Val Ser Gly Phe Ser Leu 20 4

AGT

Ser ACC TAT AGT GTA TAC TGG GTT CGC CAG CCT CCA GGA AAG GGT CTG Thr Tyr Ser Val Tyr Trp Val Arg Gin Pro Pro Gly £Ls Gly Leu 35 40 1 I 1 GAG TGG CTG Glu Trp Leu ACT CTC AAA Thr Leu Lys GTT TTC TTA Val Phe Leu GGA GTG ATA TGG AGT GAT GGA AGC ACA ACC TAT AAT TCA Gly Val Ile Trp Ser Asp Gly Ser Thr Thr Tyr Asn Ser 50 55 CO TCC AGA CTG ACC ATC AGC AAG GAC AAC TCC AAG AGT CAA Ser Arg Leu Thr Ile Ser Lys Asp Asn Ser Lys Ser Gin 70 AAA GTG AAC AGT CTC CAA ACT GAT GAC ACA GCC ATG TAC Lys Val Asn Ser Leu Gin Thr Asp Asp Thr Ala Met Tyr 85 144 192 240 288 336 TAC TGT Tyr Cys GCC AGA ACC TAT GGT TAT GAC GGG TCC TGG CTT GCT TAC TGG Ala Arg Thr Tyr Gly Tyr Asp Gly Ser Trp Leu Ala Tyr Trp 100 105

I

GGC

Gly 110 CAA GGG ACT CTG GTC ACT GTC TCT GCA GCC AAA ACA ACA CCC CCA Gln Gly Thr Leu Val Thr Val Ser Ala Ala Lys Thr Thr Pro Pro 115 120 125 TCA GTC TAT Ser Val Tyr GTG ACT CTG Val Thr Leu GTG ACT TGG Val Thr Trp 160 CCA CTG GCC CCT GGG TGT GGA GAT ACA ACT GGT TCC TCC Pro Leu Ala Pro Gly Cys Gly Asp Thr Thr Gly Ser Ser 130 135 140 GGA TGC CTG GTC AAG GGC TAC TTC CCT GAG TCA GTG ACT Gly Cys Leu Val Lys Gly Tyr Phe Pro Glu Ser Val Thr 145 150 155 480 528 AAC TCT GGA TCC Asn Ser Gly Ser 549

Claims (27)

1. Recombinant DNA which codes for the light or heavy chain of an antibody whose constant regions are of human origin and whose variable regions are of non- human origin, w h e r e i n this DNA contains in the direction of transcription a cDNA sequence which codes for the variable, non-human region, including the region J and, if desired, D, an intron sequence which has a splice donor site at its 5' end and is composed of a non- human intron partial sequence at the 5' end and a human intron partial sequence at the 3' end and a genomic DNA sequence coding for the human constant region

2. Recombinant DNA as claimed in claim 1, w h e r e i n the sequence and the non-human intron partial sequence of are of murine origin. t1*4

3. Recombinant DNA as claimed in claim 1 or 2, w h e r e i n the non-human intron partial sequence of corresponds to the intron sequence which naturally follows the sequence

4. Recombinant DNA as claimed in claim 1 or 2, w h e r e i n the non-human intron partial sequence of is 15 to 20 base pairs long. 32 Recombinant DNA as claimed in one of the previous claims, w h e r e i n the splice donor site in has the sequence 5'-GT-3'.

6. Recombinant DNA as claimed in one of the previous claims, w h e r e i n the human intron partial sequence of has a branch site.

7. Recombinant DNA as claimed in one of the previous claims, w h e r e i n it has in addition a promoter under the control of which the sequences coding for the antibody chains can be expressed. S8. Recombinant DNA as claimed in claim 7, w h e r e i n it has the promoter of the cytomegalo-virus. S i 4

9. Recombinant DNA as claimed in one of the previous claims, w h e r e i n the sequence codes for the variable region of the light or heavy chain of an antibody which is capable of specifically S' binding to the human interleukin 2 receptor. Recombinant DNA as claimed in one of the previous claims, w h e r e i n the sequence codes for the constant region of the light or heavy chain of a human antibody of the IgG type.

11. Expression vector containing a recombinant DNA as claimed in one of the claims 1 to

12. Plasmid pK chim aS \-e.~ce-re o- <e we- 1 33

13. Plasmid p -chim C k' re.Ne oFe ceSa

14. Process for the production of a recombinant DNA as claimed in one of the claims 1 to w h e r e i n polyA+-RNA is isolated from a hybridoma cell line secreting an antibody of the desired specificity, a cDNA library of this hybridoma cell line is prepared in suitable vectors and examined for clones which contain the cDNA for antibody chains by hybridization with oligonucleotides which are complementary to the DNA coding for the constant part of the antibody, the V, J, and if desired D, sequences are isolated, a splice donor site, a part of an antibody intron sequence of non-human origin and a restriction Scleavage site adjacent to the respective J domain in the direction of transcription are introduced by site-directed mutagenesis and the DNA obtained is ligated with a genomic DNA sequence which codes for the constant region of the light or heavy chain of a human antibody and has part of a corresponding human intron sequence at its 5' end. Process as claimed in claim 14, w h e r e i n a mouse or rat hybridoma cell line is used.

16. Process as claimed in claim 15, w h e r e i n a mouse hybridoma cell is used which secretes antibodies directed against the human interleukin 2 receptor.

17. Process as claimed in claim 14, 15 or 16, wh e r e i n a 15 to 20 bp long antibody intron sequence of non-human origin is introduced. 34

18. Process as claimed in one of the claims 14 to 17, w h e r e i n a part of the authentic adjacent intron of the genomic DNA sequence of non-human origin is introduced. Q* C j 6

19. Process as claimed in one of the claims 14 to 18, w h e r e i n the nucleotide sequence GT is introduced as the splice donor site. Process as claimed in one of the claims 14 to 19, w h e r e i n a human intron DNA is used which has a branch-site.

21. Process as claimed in one of the claims 14 to w h e r e i n a promoter sequence is added to the end of the DNA sequence which is finally obtained, or the manipulation is carried out in a vector which contains a promoter at a suitable position.

22. Process as claimed in claim 21, w h e r e i n the promoter of the cytomegalo-virus genome is used as the promoter.

23. Process as claimed in one of the claims 14 to 22, w h e r e i n a human genomic DNA sequence is used which codes for the constant region of the light or heavy chain of an antibody of the IgG type. L i I S&* S t 35

24. Process for the production of an expression vector, w h e r e i n a recombinant DNA as claimed in one of the claims 1 to 10 or produced as claimed in one of the claims 14 to 23 is inserted into a vector which is suitable for the expression of a foreign gene. Process for the production of a chimeric antibody whose variable regions are of non-human origin and whose constant regions are of human origin, w h e r e i n one each of an expression vector which contains a recombinant DNA as claimed in one of the claims 1 to 10 or which is produced as 'claimed in one of the claims 14 to 23 that codes for the light chain and an expression vector which contains a recombinant DNA as claimed in one of the claims 1 to 10 or which is produced as claimed in one of the claims 14 to 23 that codes for the heavy chain of the antibody are introduced into a suitable host cell, stable transformants are isolated and the antibody is isolated according to known methods from the culture supernatant of the cells.

26. Process for the production of a chimeric antibody as claimed in claim 25, w h e r e i n non- producer hybridoma cells are used as host cells.

27. Process as claimed in claim 25, w h e r a i n the cell line Sp2/O-Agl4 (ATCC CRL 8922) is used.

28. Process as claimed in one of the claims 25 to 27, w h e r e i n the vectors pK chim. and p chim. are introduced into the host cell. -36-

29. Process as claimed in one of the claims 25 to 28, wherein the vectors are transfected by electroporation.

30. Chimeric antibody MAB 179 as hereinbefore defined.

31. Chimeric antibody MAB 215 as hereinbefore defined.

32. Chimeric antibody MAB 447 as hereinbefore defined.

33. Recombinant DNA as claimed in claim 1, or a process for the production of use thereof, substantially as hereinbefore described with reference to the drawings and/or Examples. Dated this 3rd day of August, 1992. BOEHRINGER MANNHEIM GmbH 20 By DAVIES COLLISON CAVE Patent Attorneys for the applicant(s) So a r ae o a 4 i 4, t t 37 Abstract A recombinant DNA which codes for the light or heavy chain of an antibody whose constant regions are of human origin and whose variable regions are of non-human o:;igin, and which has in the direction of transcription a cDNA sequence which codes for the variable, non- human region, including the region J and, if desired, D, an intron sequence which has a splice donor site at its 5' end and is composed of a non-human intron partial sequence at the 5' end and a human intron Spartial sequence at the 3' end and D a genomic DNA sequence coding for the human constant region. o 6 a« o O O I I I I. I IL 411