FI103477B - A method of producing a therapeutically useful antibody or antibody fragment - Google Patents

Description

103477

A method of producing a therapeutically useful antibody or antibody fragment

Separated from FI patent application 903056.

This invention relates to the production of immunoglobulins and to modifications of the amino acid sequences of naturally occurring antibodies. In particular, this invention relates to the use of recombinant DNA technology for the production of chimeric genes and to the use of these modifications in the construction of chimeric antibodies.

Antibodies are specific immunoglobulin (Ig) polypeptides produced by the vertebrate immune system in response to exposure to foreign proteins, glycoproteins, cells or other antigenic foreign substances. The sequence by which an organism is able to prevent the invasion of foreign cells or to release foreign bodies from the body is at least partially known. An important part of this process is the production of antibodies that specifically bind 20 specific foreign substances. The binding specificity of these polypeptides to a particular antigen is very sophisticated, and the number of specificities that the vertebrate individual is able to form is remarkable for its complexity and variability. Millions of antigens are capable of eliciting antibody responses, with each antibody targeting almost exclusively the specific antigen that raised it.

Currently, two major sources of vertebrate antibodies are used - formation by mammalian B-lymphocytes in situ and formation by B-cell hybrids in cell culture. Antibodies are formed in situ as a result of the differentiation of precursors of B lymphocytes into plasma cells in response to stimulation by certain anti-35 genes. The DNA regions encoding different regions of the immunoglobulin 2 103477 chains are separated in the DNA of the non-differentiated B cell genome. Prior to expression, these episodes are assembled in stages. A review of this process is provided by Gough (1981) 5 Trends in Biochem. Sci. 6, 203.

The resulting rearranged gene is capable of expression in mature B lymphocytes to produce the desired antibody. Even when a particular mammal is exposed to only one antigen, a population of about 10 antibodies is not obtained. The in situ immune response against a particular antigen is determined by a mosaic of responses to several different determinants on the antigen surface. Each subset of homologous antibodies is produced by one population of B cells - therefore, the in situ generation of antibodies is "polyclonal".

This limited but inherent heterogeneity has been overcome in a number of specific cases by the use of hybridoma technology to generate "monoclonal" antibodies in cell cultures by B cell hybridomas, cf. Kohler and Milstein, C., (1975) Nature 256, 495-497], In this method, relatively short-lived, or mortal, splenocytes or lymphocytes derived from an antigen-injected mammal are fused to an immortal ·· 25 tumor cell line, thus forming hybridomas, immortals that are capable of producing an antibody genetically encoded by this B cell. The hybrids thus formed are segregated into single genetic strains by selection, dilution, and re-breeding, and each strain thus represents: · one genetic line. Therefore, these strains produce antibodies that are assuredly homogeneous against the desired antigen. Pure genetic lineage refers to these antibodies as "mono-35 clonal".

3, 103477

Monospecific monoclonal antibodies have greatly influenced immunology, and their utility in the natural sciences such as biology, pharmacology, chemistry, and other sciences has already been demonstrated. These 5 monoclonal antibodies have been widely used not only as diagnostic reagents [cf. e.g., Immunology for the 80's, Voller, A., Bartlett, A. and Bid-well, D. (eds.), MTP Press, Lancaster, 1981] but also in therapy [cf. e.g., Ritz, J. and Schlossman, S.F. (1982) 10 Blood 59, 1-11].

Although the monoclonal antibodies produced by the hybridomas are theoretically effective and preferable to polyclonal antibodies by their specificity, they have one major weakness. In many applications, the use of monoclonal antibodies produced in non-human animals is very limited when monoclonal antibodies are to be used in humans. Repeated injections of a "foreign" antibody, such as a mouse antibody, into humans can lead to deleterious hypersensitivity reactions. Such a non-human monoclonal antibody produces a non-human antibody (ANHA) response when injected into a human. The known ANHA response to human murine antibodies, the Human Mouse Antibody (HAMA) Response, has been reviewed by Shawler et al. . (1985) Journal of Immunology 135, 1530 - 1535.

It is believed that animal immunoglobulins possessing human constant regions elicit a lower ANHA response when injected into humans than animal immunoglobulins having a non-human constant region. Monoclonal d 103477 4 monoclonal antibodies with good binding affinities to selected antigens and human-derived constant regions are considered to have high potential for recovery in immunological diagnostics and treatment of cancer patients.

To date, several attempts have been made to prepare human 5-derived monoclonal antibodies using human hybridomas. For example, human-to-human hybridomas have been prepared [Olsson et al. . (1980)

Proc. Natl. Acad. Sci. USA 77, 5429]; human mouse hybridoma [Schlom, J., et al. (1980) ibid. 77, 6841] and several other hybrid combinations of foreign origin. Human monoclonal antibodies have also been produced by transforming lymphocytes using Epstein-Barr virus. However, such hybridomas may potentially contain pathogenic human viruses. Alternatively, antibody-producing primary B cells have been immortalized by in vitro transformation with viral DNA. Unfortunately, monoclonal antibody yields from human hybridoma cell lines are relatively low (1 µg / ml in human hybridoma compared to 100 µg / ml in mouse hybridoma) and the production costs are high.

Although human immunoglobulins are highly desirable in the immunological diagnosis and treatment of cancer patients, human hybridoma techniques have not yet reached the stage where human monoclonal antibodies with the necessary antigen specificities could be readily available. In addition, for obvious ethical reasons, researchers cannot immunize humans with selected toxic or otherwise harmful agents to form antibodies against a particular antigen. In the case of humans, this places major limitations on the immunological diagnosis and treatment of patients.

The production of human-derived monoclonal antibodies is certainly possible, but this is still ineffective due to the low reproducibility and other problems mentioned above. [See. in addition, Nature 30: 316-317 (1982)]. Most monoclonal antibodies are therefore derived from non-human animals.

With a high binding affinity, a monoclonal antibody responsive to human cancer antigens, which is not recognized as a foreign substance in humans, is highly desirable. A method to avoid this difficulty is to artificially generate an antibody that is very similar to a human antibody and is not recognized as foreign by the human body, i.e., a chimeric or "humanized" antibody.

In the case of chimeric antibodies, it is typical that the variable regions of both the heavy and light chains resemble the variable regions of antibodies from a single mammalian species, and that the constant portions are homologous to the sequences present in human antibodies. One clear advantage of these chimeric forms is that, for example, the algal regions can be conveniently obtained from currently known sources using readily available B-cell hybridomas from a non-human host animal, together with, for example, constant regions derived from human cell products. When the specificity of the variable region of these antibodies is not affected by its origin, and when the constant region is of human origin, it is less likely that they will elicit an immune response when injected into a patient compared to a non-human 30 region response.

•: * One known human cancer tumor antigen is a cancer tumor-associated glycoprotein (TAG72). TAG72 binds to the surface of human-derived cells, particularly the cancer-tumor cell line LS174T. LS174T [American Type Culture Collection 35 (herein ATCC) No. CL 188] is a line 6 103 477 LS180 (ATCC No. CL187), colon adenokarsinoomalinjan variant.

The karyotype of LS174T is similar to that of LS180, with the absence of the X chromosome in most cells. Has been disclosed in Johnson, V.G. et al. (1986) Cancer Res.

46, 850-857, according to which the TAG72 molecule can be characterized as a mucin. This conclusion is based on the following observations: (a) TAG72 has a high molecular weight (> 1 x 10 6) as evidenced by its 10 inclusion on a Sepharose ™ CL-4B column; (b) TAG72

The density as determined by equilibrium centrifugation in CsCl was 1.45 g / ml, which strongly refers to glycosylated glycoprotein; (c) TAG72 is found to alter its transport after neuraminidase cleavage, suggesting that it is a molecule rich in sialyl groups and rich in mucin-specific O-glycosidic linked oligosaccharides; (d) Blood group antigens commonly found in cupins have been observed in affinity purified 20 TAG72; and (e) TAG72 was not affected by cleavage with Chondroitinase ABC, indicating that the TAG72 epitope was not expressed on the surface of proteoglycan r.

Many murine monoclonal antibodies have been developed which have binding specificity for TAG72. One of these monoclonal antibodies designated B72.3 is murine IgG1 produced by hybridoma B72.3 (ATCC No. HB-8108). B72.3 is a first generation monoclonal antibody developed using human breast cancer extract as an immunogen. Colcher, D., et al. (1981) Proc. Natl. Acad. Sci. USA 78, 3199-3203); and U.S. Patents 4,522,918 and 4,612,282]. The term "first generation monoclonal antibody" as used herein refers to a monoclonal antibody, 7103477, produced using crude cell extract as an immunogen.

More TAG72 directed against the monoclonal antibodies designated "CC" (colon cancer).

CC monoclonal antibodies are a second generation of murine monoclonal antibodies. The term "second generation monoclonal antibody" as used herein refers to a monoclonal antibody produced using antigen purified by a first generation monoclonal antibody as a immunogen.

Monoclonal CC antibodies were prepared using B72.3 purified TAG72. A review of the method used to produce CC antibodies is disclosed in U.S. Patent Application Publication No. 7-073 685 (USPA 7-073 685), 15 filed by Schlom et al. July 15, 1987 and is publicly available from the National Technical Information Service. Due to their relatively good binding affinities against TAG72, the following CC-20 antibodies have been deposited with the ATCC requiring the following: CC49 (ATCC No. HB 9459), CC83 (ATCC No. HB 9453), CC46 (ATCC No. HB 9458), CC92 (ATCC No. HB 9454), CC30 (ATCC No. HB 9457), CC11 (ATCC No. HB 9455) and CC15 (ATCC No. HB 9460).

It is not known in the art to isolate a human antibody that binds strongly to TAG72. Therefore, suitable antibodies must be prepared artificially.

It is known that the function of an Ig molecule depends on its three-dimensional structure, which in turn depends on its primary amino acid sequence. Thus, altering the amino acid sequence of Ig can adversely affect its activity. A change in the Ig coding DNA sequence may affect the ability of the cell containing the DNA sequence to express or secrete Ig or to perform its assembly.

According to the teachings in U.S. Patent Application Serial No. 7-073,685, CC antibodies can be converted to a chimeric form by, for example, replacing mouse constant regions (Fc) with human constant regions using recombinant DNA technology known per se. It is believed that the proposals set forth in U.S. Patent Application Publication No. 7-073,685 did not lead to the actual attempt to express any chimeric Ig polypeptide chains, to produce Ig activity, and to secrete and assemble Ig chains into desired chimeric Ig molecules.

Therefore, it is not at all clear from the prior art in this field that recombinant DNA techniques, when used routinely, produce a chimeric animal-human antibody from selected DNA sources that form functional chimeric antibodies that specifically bind to selected human carcinomas. , which are less likely to trigger ANHA side effects when injected into a human.

The present invention is surprisingly capable of meeting many of these aforementioned needs and provides a method for producing these desired antibodies. For example, the present invention provides a method for fusing genes encoding at least a portion of an animal Ig molecule that binds TAG72 to human carcinoma with genes encoding at least a portion of a human Ig molecule. The present invention may also provide a method for expressing a protein that can be secreted and assembled to produce a functional chimeric antibody.

The present invention further provides an expression vector comprising a DNA sequence encoding antibodies or portions thereof directed against TAG72.

The present invention also provides cells transformed with an expression vector containing a DNA sequence encoding antibodies to TAG72 9103477 and portions thereof.

Finally, the present invention provides novel therapeutically useful antibodies.

Based on the foregoing, the present invention is directed to

a method of producing a therapeutically useful antibody or fragment thereof which is capable of reacting selectively with a TAG72 antigen. The method is characterized in that one of the cell lines CH44-10 1 (ATCC HB 9884), CH44-2 (ATCC HB 9880), CH44-4 (ATCC HB

9877), CH 88-1 (ATCC HB 9882), CH 88-2 (ATCC HB 9876), CH 88-3 (ATCC HB 9876), CH 88-4 (ATCC HB 9874), CH 84-1 (ATCC HB) 9883), CH 84-2 (ATCC HB 9879), CH 84-3 (ATCC HB 9878) or CH 84-4 (ATCC HB 9875) and recovering the antibody produced.

The present invention also relates to a process for the preparation of a therapeutically useful antibody or antibody fragment conjugate, wherein the method comprises conjugating the antibody or antibody fragment as defined above to a therapeutic agent.

The present invention further relates to cell lines CH44-1 (ATCC HB 9884), CH44-2 (ATCC HB 9880), CH44-4 (ATCC HB 9877), CH 88-1 (ATCC HB 9882), CH 88-2 (ATCC HB 9881). ), CH 88-3 (ATCC HB 9876), CH 88-4 (ATCC HB 9874), CH 84-1 (ATCC 25 HB 9883), CH 84-2 (ATCC HB 9879), CH 84-3 (ATCC HB) 9878) and CH 84-4 (ATCC HB 9875).

Explanations of the figures

Figure 1 illustrates the basic structure of an immunoglobulin, which is marked with enzymatic cleavage sites. Figure 2 depicts the nucleotide sequences of VHaTAG -VH, CC46 -VH, CC49: -VH, CC83 -VH and CC92-VH.

Figure 3 depicts the amino acid sequences of VHoTAGVH, CC46VH, CC4 9VH, CC83VH and CC92VH.

Figure 4a depicts the nucleotide sequence of CC49 VL and 35 Figure 4b depicts the corresponding amino acid sequence.

10 103477

Figure 5a depicts the nucleotide sequence of CC83VL and Figure 5b depicts the corresponding amino acid sequence.

Figure 6a depicts the nucleotide sequence of CC92VL and Figure 6b depicts the corresponding amino acid sequence.

Figure 7 depicts a plasmid isolated from pGD1

Hind I - Pst I fragment nucleotide sequence.

Figure 8 shows a plasmid map of pBLUESCRIPT SK (-).

Figure 9 is a plasmid map of pRL101.

Figure 10 depicts the junction of CC49 L-chain genomic DNA in pRLIOI.

Figure 11 shows a plasmid map of pRL200.

Figure 12 depicts the junction of CC83 L-chain genomic DNA in pRL101.

Figure 13 depicts isolated from plasmid pNP9

The nucleotide sequence of EcoR I - BamH I fragment.

Figure 14 depicts a plasmid map of pHH49.

Figure 15 shows a plasmid map of pHS83.

Figure 16 shows the nucleotide sequence of CC49 VH, with under-20 dashed segments indicating the sequences obtained using oligonucleotide primers in conjunction with the messenger RNA.

Figure 17 shows the nucleotide sequence of CC83 VH, with underlined segments indicating the sequences obtained using oligonucleotide primers in conjunction with messenger RNA.

Figure 18 shows the amino acid sequence of CC49 VH, with underlined segments indicating the sequences determined by protein sequencing.

Figure 19 shows the amino acid sequence of CC83 VH, with underlined segments indicating the sequences determined by protein sequencing.

Figure 20 shows SDS polyacrylamide gel results depicting the results of PN83 gas F treatment with CC83.

11 103477

Figure 21 depicts a restriction enzyme map of human gamma 1, gamma 2, gamma 3 and gamma 4.

Figure 22 shows a plasmid map of pSV2gpt / R / P.

Figure 23 is a plasmid map of pSV2gpt-yl-7.8.

Figure 24 is a plasmid map of pSV2gpt-yl-2.3.

Figure 25 shows a plasmid map of pSV2gpt-y2. Figure 26 shows a plasmid map of pSV2gpt-y3-7.8.

Figure 27 shows a plasmid map of pSV2gpt-y4. Figure 28 depicts a plasmid map of ρ49γΐ-7.8: η. Figure 29 depicts a plasmid map of ρ49γ1-2.3: η. Figure 30 depicts a plasmid map of ρ49-γ2: η.

Figure 31 depicts a plasmid map of ρ49-γ3: η. Figure 32 depicts a plasmid map of p49-y4. Figure 33 depicts a plasmid map of p83yl-7.8. Figure 34 depicts a plasmid map of ρ83γΐ-2.3: η. Figure 35 depicts a plasmid map of ρ83-γ2: η.

Figure 36 depicts a plasmid map of ρ83-γ3: η. Figure 37 depicts a plasmid map of ρ83-γ4: η. Figure 38 depicts the overall reaction to generate hybrid genes based on Horton et al. 25 a!, (1989) Gene 77, 61.

Figures 39A, 39B and 39C show the biodistribution and accumulation of CH44-1 throughout the body.

Figures 40A and 40B show biodistribution and accumulation of CH84-1 throughout the body.

The immunoglobulin of the present invention has been developed in response to the problems associated with murine monoclonal antibodies previously described in the art. A characteristic feature of this is that it has a chimeric structure consisting of a heavy chain variable region encoded by 35 DNA from VHaTAG.

Definitions "Immunoglobulin" as used herein, means a tetramer or an aggregate thereof, whether or not it is characterized by immunoreactive activity. "Antibodies" refers to those compositions having significant and known specificity against antigen, comprising light and heavy chains with or without a covalent bond. "Non-specific immunoglobulin" ("NSI") 10 refers to those immunoglobulins which are not known to have known antigen specificity.

The basic unit of immunoglobulin structure in vertebrate systems is relatively well known [Edelman, G.M., (1971) Ann. N.Y. Acad. Sci., 190, 5]. As shown in Figure 1, the units consist of two similar light polypeptide chains of about 23,000 daltons molecular weight and two similar heavy chains of 53,000 to 70,000 molecular weight. These four chains are linked by disulfide bonds of 20 "Y": into a n-shaped pattern where heavy chains are trapped between the light chains starting at the mouth of Y and continuing through an area of variable character.

Heavy chains are classified into gamma, myth, alpha, delta, or epsilon, including some subclasses. Because of the long unchanged region of this chain, the class of antibody is defined by its nature as IgA, IgD, IgE, IgG or IgM.

Light chains are classified as either kappa (k) or 30 lambda (λ). Each heavy chain class can be fitted with a lightweight either kappa or lambda chain. Light and heavy chains are generally covalently bound to one another, and the "tail" portions of the two heavy chains are covalently bound by covalent disulfide bonds when immunoglobulins are formed by hybridomas or B cells. If non-covalent linkage of 13,103,447 chains can be accomplished with the correct geometry, aggregates formed by disulfide bonded non-linked chains will still be able to react with the antigen.

5 The amino acid sequences start at the N-terminal ends of the branched ends of Y and continue to the C-terminus at the lower end of each chain. The variable portion is at the N-terminus and the C-terminus is a constant region.

10 The terms "constant" and "variable" are used functionally. The variable regions of both light (VL) and heavy (VH) chains determine the recognition capacity required for binding and antigen specificity. Parts of the constant region of the light (CL) and heavy (CH) chains contain important biological properties such as antibody chain association, secretion, placental transfer, and complement binding.

In both chains, the variable region is linked to the constant region by a junction of the V gene sequence and the C gene sequence. The linkage occurs at the genomic level such that the nucleotide sequences are linked via recombination sites. The linker sequence is currently known as the "J" sequence encoding 12 amino acids in the light chain gene and a total of about 25 amino acids encoding the "D" and "J" sequences in the heavy chain gene.

"Chimeric antibody", as used herein, refers to an antibody having a variable region amino acid encoded by a murine germline gene derived nucleic acid sequence 30 and a human gene derived nucleotide sequence encoded by a constant region amino acid sequence.

However, the present invention is not intended to be narrowly limited to replacing sections of the murine C gene encoding the immunoglobulin 35 constant regions with those of the human C gene encoding immunoglobulin constant regions. Thus, the present invention is not limited by whether or not the fusion site is at the boundary of a variable and constant region.

Modified chimeric antibodies, composite chimeric antibodies, and fragmented chimeric antibodies encoded by the nucleotide sequences described herein are currently capable of being produced by a variety of methods.

"Compound" immunoglobulins comprise variable regions of polypeptides that have not previously been found to be naturally associated. It is not decisive whether these are covalently or non-covalently linked as long as the aggregate is capable of selectively reacting with a particular antigen or antigen group.

"Converted" antibodies refer to antibodies in which amino acids have been modified, especially for variable region amino acid sequences. Because recombinant DNA techniques are essential to the present invention, it is not necessary to be limited to the amino acid sequences of antibodies from natural sources; the amino acid sequences of the antibodies can be redesigned to achieve the desired properties. There are many possible variations and they range from one to one. 25 amino acid changes to complete re-engineering of the variable and / or unchanged antibody region.

Variations in the variable region are made to improve the antigen binding properties. Changes in the constant region are generally made to improve the properties of cellular functions such as complement binding, membrane interaction, and other effector functions. Modifications can be made by standard recombinant techniques as well as by oligonucleotide target mutagenesis (Dalbadie-McFarland, et al. (1982) Proc. Natl. Acad. Sci.

USA 79, 6409].

"Fragments" of immunoglobulins include proteolytically cleaved or recombinant portions of an antibody molecule that are capable of reacting selectively with a particular antigen or antigen group. Non-limiting examples of these proteolytic and / or recombinant fragments of the invention include "Fab", "F (ab ') 2" and "Fab1" whose proteolysis sites 10 are cleaved in Figure 1, as well as "Fv" and "VH". " The "Fv" fragment means that portions of the light chain variable and heavy chain variable regions are linked together at one of their carboxy-terminal portions. Combination techniques for the generation of Fv fragments are disclosed in WO 88/01649, WO 88/06 630, WO 88/01630. 07 085, WO 88/07 086 and WO 88/09 344. The "VH" fragment means that the variable region comprises at least part of a heavy chain variable region which can be used as an antigen binding function. use in antigen-binding function is disclosed in an article entitled "Development of Biologically Active Peptides Based on Antibody Structure" by Williams et al. (198 9)

Proc. Natl. Acad. Sci. USA 86, 5537-5541.

For the purposes of this invention, "animals" include cattle, pigs, rodents and primates, and others.

An "expression vector" is provided with a functional definition that includes any DNA sequence capable of providing expression of a particular DNA code in a suitable host. These vectors are now in the form of plasmids; "plasmid" and "expression vector" are thus often used interchangeably. However, it is contemplated that the present invention also encompasses other expression vectors that function in a similar manner that may be sometimes known in the art.

By "transformation" is meant the introduction into a recipient host cell of DNA that alters 5 genotypes with the result that the recipient cell is altered.

"Host cells" refer to cells transformed by recombinant techniques with vectors constructed by recombinant DNA techniques. As defined herein, production of a host cell antibody or a variant thereof is accomplished by such transformation.

In describing methods for isolating antibodies from recombinant hosts, the terms "cell" and "cell-15 culture" are used interchangeably to refer to the source of the antibody, unless explicitly stated otherwise. In other words, antibody recovery from "cells" may mean recovery from centrifuged whole cells or cell culture containing both nutrient solution and cells suspended therein.

abbreviations

Nucleic acids, amino acids, peptides, protecting groups, active groups and similar moieties are abbreviated when used in accordance with IUPABIUB (Commission on 25 Biological Nomenclature) or practice in the art. The following are examples.

Reagents EDTA: Ethylenediaminetetraacetic acid SDS: Sodium dodecyl sulfate 3 0 Nucleic acids RNA: Ribonucleic acid DNA: Deoxyribonucleic acid 17 103477

nitrogen bases

Purines Pyrimidines A: adenine T: tyrannine G: guanine C: cytosine 5 U: uracil

Both DNA and RNA contain long chains of phosphoric acid, sugar and nitrogen bases. DNA is a double stranded twin with sugar as 2-deoxyribose while RNA is single stranded with sugar as D-ribose. The four nitrogenous bases characteristic of DNA nucleotides are linked in complementary pairs by hydrogen bonds to form a double strand of DNA; adenine is associated with thymine; guanine is associated with cytosine. In RNA, uracil replaces thymine in the listed DNA pairs.

15 Amino acids

Gly: Glycine Phe: Phenylalanine

Area: Alanine Tyr: Tyrosine

Val: Thrin: Thr: Threonine

Leu: Leucine Cys: Cysteine 20 Leu: Isoleucine Met: Methionine

Ser: Serine Glu: Glutamic acid

Asp: aspartic acid Trp: tryptophan

Lys: Lysine Pro: Proline

Arg: arginine Asn: asparagine ·: 25 His: histidine Gin: glutamine

Variable range

The heavy chain coding DNA consists of a VH gene sequence, a DH gene sequence, and a JH gene sequence. The DNA encoding the light chain consists of a VL gene sequence and a JL gene sequence. The present invention relates to a method of producing selected chimeric antibodies having VH encoded by a DNA sequence derived specifically from a germline germline responsive to TAG72 (VHccTAG), the gene of which has a nucleotide sequence of 35 as shown in Figure 2. Chimeric 103477 the antibodies are selected for their ability to bind TAG72, which means that the variable portion binds to TAG72 at least 25% better than the variable portion of B72.3 binds to TAG72. The binding affinities of the chimeric antibody and B72.3 are generally determined by the same method. Examples of methods for measuring antibody binding affinity are given in the following references: Scatchard, G., (1949) Annals of the N.Y. Acad. of

Sciences 51, 660; Steward, M.W. and Petty, R.E., (1972) 10 Immunology 23, 881; Muraro, R., et al. (1988) Cancer Research 48, 4588 and Heyman, B., (1984) J. of Immunol. Met hods 68, 193 - 204.

One skilled in the art will appreciate that as a result of this invention, i.e. the nucleotide sequence of VHaTAG (and 15 amino acid sequences encoded by it), the present invention is intended to cover nucleotide sequences having homologous content and corresponding amino acid sequences. "Homologous in content" means that the nucleotide or amino acid sequences are identical or nearly identical. Thus, it is self-evident in this description that minor sequence variations may occur in homologous sequences and that any sequence having at least 80% homology is interpreted as equivalent.

Homology is reported for matching bases; 25 (or amino acids) after two (possibly different) sequences have been aligned. The term parallel pricing is used in the sense defined by Sankoff and Kruska in The Time Warps, String 30 Edits, and Macromolecules; In The First Chapter of Theory and Practice of Sequence Comparison, Addison-Wesley, Reading, MA (1983). In essence, the bipartite alignment occurs by maximizing the number of matching bases (or amino acids) between the two batches by adding a minimum number of "blank" or "zero-10347719a" bases to either of the batches to achieve maximum overlap.

As is clear in the art, nucleotide mismatches may occur at the third, i.e., unstable, 5 base of the codeword without causing amino acid substitution in the final polypeptide sequence. Minor minor nucleotide changes (substitutions, insertions, or deletions) in certain regions of the gene sequence can also be tolerated whenever the change results in a change in the amino acid sequence that does not alter the functional nature of the final product. It has been shown that chemically synthesized batches of whole gene sequences or portions of sequences can replace the corresponding regions of the natural gene without affecting the function of the gene.

Certain homologues of DNA sequences can be identified by those skilled in the art using conditions that restrict the cross-hybridization of nucleic acids, as is well known below (described in Nucleic Acid Hybridization, Hames and Higgens (eds) IRL Press, Oxford, UK ( 1985)]. When two sequences are known, algorithms are available to calculate their homology, e.g., Needleham & Wunsch, (1970) J. Mol. Biol. 48, 443-453 and Sankoff and Kruska (cited above), pages 23-29.

·: 25 Commercial services such as Intelligenetics, Inc. (Palo Alto, CA) are also available to perform these comparisons.

D "and JB gene sequences

Several types of Dh and JH gene sequences occur, although the type of D or J gene segment selected is not critical to the present invention. In other words, DH and JH can be derived from any animal. Preferred animals include mice and humans. Of course, the human DH and / or JH gene segments are particularly preferred, but the present invention is not limited thereto if

The Dh and JH gene segments provide an important property, for example, better binding to TAG72.

Illustrative examples of mouse DH and JH sequences are disclosed in "Organization, Structure, and Assembly of the Immunoglobulin Heavy Chain Diversity DNA Segments", Kurosawa and Tonegawa, (1982) J. Exp. Med. 155, 201 and

Sequences of the Joining Region Genes for Immunoglobulin Heavy Chains and Their Role in the Generation of Antibody Diversity "by Gough and Bernard (1981) Proc. Natl. Acad. Sci. 10 USA 78, 509.

Illustrative examples of human DH and JH sequences are disclosed in "Human Immunoglobulin D Segments Encoded in Tandem Multigenic Families", Siebenlist et al. . (1981) Nature 294, 631 and illustrative examples of human JH-15 sequences in "Structure of the Human Immunoglobulin μ Locus: Characterization of Embryonic and Rearranged J and D Genes", Ravetch et al. (1981) Cell 27, 583.

VL and JL gene sequences

In general, any VL and JL gene sequence encoding a portion of the VL complementary to VH encoded by a VHaTAG homologous in content may be used. By "complementary" is meant a VL that binds to VH and produces an antibody variable region that has at least 25% greater binding affinity than B72.3 as determined by any of the standard binding affinity assays.

The type of VL or JL gene segment selected is not critical to the present invention. In other words, VL and JL 30 can be derived from any animal. Preferred: Animals include mice and humans. Naturally, human VL and / or JL gene segments are particularly preferred, but the present invention is not limited to providing an important property of another animal's VL and JL gene segments, for example, a better binding ability to TAG72.

2i 103477

Mouse JL sequences are disclosed in "The Nucleotide Sequence of a 5.5-Kbase DNA Segment Containing the Mouse Kappa Immunoglobulin J and C Region Genes" by Max et al. , (1981) J. Biol. Chem. 256 5116-51120. Episodes of human JL-5 are disclosed in "Evolution of the Human Immunoglobulin K J Region Genes" by Heiter et al. . (1982) The

Journal of Biological Chemistry 357 1516 - 1522.

Managing variable areas

From the foregoing, it is possible to derive a number of detailed embodiments within the scope of the present invention that relate to variable regions of antibodies, i.e., regions having VH sequences homologous to VHaTAG and binding to TAG72 at least 25% better than either The B72.3 variable region binds to TAG72 when the binding affinities of antibody and B72.3 are measured by the same method. The following are several possible methods.

Naturally formed variable regions

In response to an immunogen, TAG72, an animal immunized with its antibody-producing B cells increases. The variable region of antibodies produced by B cells will be encoded by germline rearranged heavy and light chain DNA. The rearranged germline heavy chain includes, for example, V-D and J gene segments with leader sequences, as well as any introns that can be subsequently removed. The DNA encoding the light chain will include the V and J gene segments with leader sequences, as well as any introns that may be subsequently removed.

Variation may arise from somatic mutations in B cells during antibody-producing rearrangement of VHaTAG. These somatic mutations 35 are nucleotide changes that can result in the formation of 22,103,477 or no amino acid changes that alter the activity of rearranged VH against TAG72.

Screening Methods Monoclonal and polyclonal antibodies can be screened to determine which of said antibodies selectively bind to TAG72. Such screening can be accomplished by any of a number of known methods, such as solid phase radioimmunoassay, enzyme-linked immunosorbent assays, rosette-forming assays, and inhibition-based assays. These methods are well known per se.

The present invention now discloses nucleotide sequences encoding variable regions resulting from antibody-producing rearrangement of VHaTAG. Figure 2 shows, in addition to the nucleotide sequence of VHaTAG, nucleotide sequences encoding nucleotide sequences for the heavy chain variable regions of CC46, CC49, CC83 and CC92. Figure 3 shows the amino acid sequences of VHaTAG -VH, CC46 -VH, CC49 -VH, CC83 -VH and CC92-VH, which correspond to the nucleotide sequences shown in Figure 2. A striking comparison of the amino acid sequences of VHaTAG VH, CC46 VH, CC49 VV, CC83 VV and CC92 VH reveals that these chains are very substantially similar.

The similarity of the DNA encoding the CC4 6VH, CC49VH, CC83VH and CC92VH regions, especially in the 5 'end segment, is evidence that these DNA-30 sequences are derived from VHaTAG. Somatic mutations during antibody-rearrangement of the B cell-expressed VH region gene produce some nucleotide changes that may result in or result in a homologous amino acid change between two 23 103477 antibody-rearranged antibody-producing VHaTAGs.

The nucleotide sequences and corresponding amino acid sequences of CC49 VL are shown in Figures 4a and 4b. The nucleotide-5 DNA sequences and corresponding amino acid sequences of CC83 VL are shown in Figures 5a and 5b. The nucleotide sequences and corresponding amino acid sequences of CC92 VL are shown in Figures 6a and 6b.

probe Techniques

Other antibodies encoded by VHaTAG-derived DNA can be derived using VHaTAG as a probe for hybridization. Generally, those skilled in the art can use assays prepared from VHccTAG RNA or DNA or recombined V? ATAG containing rearranged genes to detect homologous genes in unknown hybridomas. These homologous antibodies have a DNA sequence linked by a messenger RNA hybrid with a probe consisting of, in whole or in part, VHaTAG and its flanking flanking regions. The "flanking regions" are intended to include those 20 DNA sequences that extend to the 3 'end of the gene upstream of the 5' end of the VHaTAG and the 5 'end of the VH gene from the 3' end of the TAG.

Scheduled Synthesis of Variable Regions 25 Another approach is the systematic synthesis of variable regions of antibodies found using the antibodies described herein. This approach has several potential benefits. Namely, it would not be necessary for the investigator to screen the immunized experimental animals first in an attempt to find those antibodies which: bind to TAG and then find antibodies which specifically have VH regions encoded by DNA derived from VHaTAG.

24 103477

Mutagenesis Methods VH and / or VL gene sequences can be "engineered" by mutagenesis. Illustrative examples of these methods include the addition of small amounts of different nucleotides, de-5 lesions, or substitution of multiple product nucleotides, or product properties retaining substitution of multiple nucleotides, provided that proper digestion is maintained.

Replacements, deletions, insertions, or any combination of these can be combined to provide a final construct. Because there are 64 possible sequences of codewords but only 20 amino acids, the genetic code is shaky in the sense that different codewords can result in the same amino acid. However, the code is 15 accurate for each amino acid; thus, each amino acid has at least one codeword, i.e., each codeword yields only one amino acid and no other amino acids. It is understood that during translation, the reading reflectance must remain correct to obtain the correct amino acid sequence for the resulting polypeptide.

Methods for making additions to predetermined amino acid residues are known. Examples of these methods include oligonucleotide-mediated target mutagenesis and polymerase chain reaction.

Methods of performing deletions at predetermined amino acid sites are known. Examples of these methods include oligonucleotide-mediated target mutagenesis and polymerase chain reaction.

Methods for performing substitutions at predetermined amino acid residues 30 are known. Examples of these methods include oligonucleotide-mediated target mutagenesis and polymerase chain reaction.

Oligonucleotide-mediated site-directed mutagenesis involves essentially hybridizing an oligonucleotide 35 encoding a desired mutation to 25,103,477 single-stranded DNA comprising the mutation region and using this single-stranded strand as a template to produce a mutation-containing strand. This method and its various forms are described in Zoller, M.J., and Smith, 5M., (1982) Nucl. Acids Res. 10, 6487-6500; Norris, K.,

Norris, F., Christiansen, L. and Fiii, (1983) N. Nuc. Acids Res. 11, 5103-5151; Zoller, M.J. and Smith, M., (1984) DNA 3, 479-488; Kramer, W., Schughart, K. and Fritz, W.J. (1982) Nuc. Acids Res. 10, 6475 - 6485.

The polymerase chain reaction (PCR) essentially involves exponential amplification of DNA in vivo using oligonucleotides of defined sequence. The oligonucleotides may, if desired, include sequence changes. The polymerase chain reaction method is described in Mullis 15 and Faloona, (1987) Meth. Enz. 155, 335-350. Examples of mutagenesis using PCR are described in Higuchi et al. . (1988) Nucl. Acids Res. 16, 7351-7367;

Ho et al. (1989) Gene 77, 51-59 and "Engineering Hybrid

Restriction Genes Without Use of Restriction Enzymes: 20 Gene Splicing by Overlap Extension, Horton et al., (1989).

Gene 7 7, 61.

Modification of the antibody variable regions may have particular use in the therapeutic use of monoclonal antibodies. Nowadays, When a chimeric antibody comprising a complete variable region of a mouse is injected into a human, the human body recognizes the variable region of the mouse as alien, even though this region is smaller than the complete mouse antibody, and elicits an immune response against it. When a human antibody or a chimeric antibody is subsequently injected into a human, the efficacy of this agent is greatly reduced due to the action of the body's immune system against the foreign antibody. As a result, alterations in the murine VH and VL regions of 10347726 may decrease the human immune response to the modified antibody.

Recombinant DNA

Antibodies can be constructed by recombinant DNA-5 techniques. In other words, since nucleotide sequences for several different VH and VL coding regions have now been introduced, one skilled in the art can produce a complete gene encoding the variable regions of heavy and light chains in Estonia.

The constructed gene can be designed such that the selected Dh and JH gene segments are operatively linked to the selected VH gene segment, i.e., the V aTAG segment or the VH gene segment of CC49 or CC83.

The engineered heavy chain coding DNA contains, for example, the DH and JH gene sequences adjacent to the germline VHaTAG gene segment, thus complementing the CDR3 and VH region frame (FR) 4. The leader sequence may be present or may subsequently be deleted.

Depending on the light chain used, it may also be necessary to obtain DNA encoding the constructed light chain. This DNA comprises an operably linked VL gene segment which is adjacent to, for example, the JL gene segment and has a leader which can be subsequently removed. The JL gene segment varies depending on. 25 whether the light chain is part of a lambda or a cupboard system.

The J region sequence is adjacent to the downstream end of the VL exon complementing FR 4 of the VL moiety. This construct can be made by the methods used to construct the VH gene.

The engineered gene can be engineered, for example, by conventional recombinant techniques to provide an expression gene insert into the plasmid. Plasmids can then be expressed in host cells. The following are examples of recombinant biological methods.

When designing a fragment encoding either a light chain or a heavy chain variable region, it is usually desirable that the intron below the J region be incorporated, in whole or in part, especially when the variable region is present. is from a host where the fused gene will be expressed. When the intron is stored, operable release and delivery sequences at the ends of the intron are required. The intron between J and the constant region of the fused gene may be primarily an intron sequence associated with (1) constant 10, (2) J or (3) parts of both. The latter may be suitable for convenience when a convenient restriction site is present in introns from these two sources. It may be necessary to provide adapters for attaching the intron to the constant region. In some cases, the intron may be modified, in whole or in part, by deletion, nucleotide (s) substitution (s), or insertion to improve ease of processing, expression, or the like. Preferably, the intron should be sufficient to include an expression enhancer region that is functionally active with the naturally occurring promoter.

Alternatively, it may be desirable that the fused gene does not contain an intron between the J gene and the C gene. The 31-end of the J gene will thus be the 5 'of the C gene. 25 next to the head. An exonuclease can be used and various 3'-ends can be obtained using variable cleavage times, which can then be used for constant region coupling and the selection of a functional product in a number of ways; or 30 by cutting the length of the overlapping parts by the polymerase chain reaction method (see Horton et al ..

above). In this case, an artificial promoter, which does not need to be functionally active with the enhancer region, is generally used.

103477 28

In one preferred embodiment, the genes encoding the V H and V L regions can be altered by replacing at least a portion of the complementarity determining regions (CDRs) of the variable regions of the antibody with light and heavy chains by analogous portions of antibody CDRs of any specificity. An example of a replacement method for CDRs is described in Gregory Winter, EP-0 239 400 and Huston et al. The teachings of PCT WO 88/093444. In the modified antibody of the present invention, only the CDRs of the antibody are foreign to the human body and this should minimize side effects if this antibody is used in the treatment of humans. However, human and mouse frameworks have their distinctive features that distinguish between human and mouse frameworks. Thus, an antibody comprising murine CDRs in the human framework region may not be more foreign to the body than the correct human antibody.

The present invention provides nucleotide sequences corresponding to VH amino acids sequences of VHaTAG, CC46, CC49, CC83, and CC92, as well as nucleotide sequences of VL gene segments of CC49, CC83, and CC92. Accordingly, it is anticipated that the CDRs obtained from the antibodies of this invention may be fused to the framework regions of a human antibody.

CDRs derived from human VH or VL moieties can generally be replaced by CDRs derived from the VH or VL moieties of antibodies of this invention. Examples of human antibodies from which framework regions can be used include human plasma cytoma NEWM ("Replacing the complementarity-defining regions of a human antibody"),

Jones et al., (1986). Nature 321 522-525], commonly available from Dr. Greg Winter; and several other human VH or VL genes available from Dr. Terrence 35 Rabbitts, both researchers at the facility% 29 103477

Medical Research Council, 20 Park Crescent, London, WIN 4AL.

Determination of what a CDR consists of and what a framework region consists of can be made based on the amino acid sequence of selected Ig molecules, as in Kabat et al. , (1987) Sequences of Proteins of Immunological Interest, 4th ed., US Dept. of Health and Human Services, NIH.

The four frameworks occur for the most part in the 10 β-strip conformation, and the CDRs form loops connecting these β-strip structures and in some cases part of these structures.

In addition, not all amino acid residues in the loop regions are within the reach of the solvent and in one case, the amino acid residues in the frame regions are involved in antigen binding (Amit, A.G., Mariuzza, R.A., Phillips, S.E.V., and Poljak, R.J., (1986) Science 233: 747-753).

It is also known that the variable regions of the two portions of the antigen binding site remain correctly oriented due to non-covalent inter-chain interactions. The amino acids present in the CDR may be involved in these interactions.

Thus, in order to transfer antigen-binding ability from one variable region to another, it may not be necessary * 25 to replace all CDRs with complete CDRs from complete variable regions available from the donor. It may be necessary to transfer only the groups required for the antigen binding site and this may involve the transfer of residues within the framework regions as well as the transfer of CDR residues.

It is thus clear that the replacement of one or more complementary CDRs alone does not always lead to a functional modified antibody. Based on the disclosures of EP 0 239 400, one skilled in the art can obtain a functional 103477 30 modified antibody either by routine experimentation or by trial and error.

Preferably, both heavy and light variable portions are modified such that CDRs are at least partially replaced and, if necessary, frame regions are partially replaced and periodic. Although the CDRs may be derived from the same animal class or even the same subclass as the antibody from which the framework regions are derived, it is expected that the CDRs will be derived from an antibody of a different animal class, and preferably from a different animal species.

The constructed variable regions of a VL encoding a VL are generally the same V gene that naturally encodes a VL associated with a freely selected VH. For example, the use of V genes encoding the VL regions of CC49 and CC83 is advantageous when using the V gene encoding the VH of CC49 and CC83, respectively.

Since the VH regions of the antibodies produced in accordance with the present invention are encoded by VH genes derived from VHaTAG, it is surprising that constituent antibodies can preferably be generated. In other words, the V H region of one antibody produced according to the present invention can be linked to the V L region of another antibody produced according to the present invention. Although the amino acid sequences of the heavy chains of CC49 and • 25 CC83, when viewed superficially, closely resemble each other, it would be expected that a change of a few or one amino acid would strongly affect antibody binding activity, i.e. the resulting antibodies are generally assumed to be nonspecific immunoglobulin. (NSI) ts.

. one lacking the character of an antibody (see EP Application Publication 0 125 023).

It is somewhat surprising that it has now been found that the VH-specific antibody of the VH produced according to the present invention does not need to be recombined to VL alone derived from the same naturally occurring animal antibody. For example, as exemplified, it is possible to produce a chimeric antibody having a heavy chain VH from CC83 and a light chain VL from CC49, wherein the constituent antibody so formed has a binding specificity greater than 25%. B72.3 is against TAG72.

Unchanged Regions The heavy chain (C 1 -) moiety CH moieties may be of different human isotypes, i.e., IgG (e.g., IgG17 IgG2, IgG3 and IgG4), IgA, IgD, IgM, and different subtypes of each group.

Human gen is reviewed in Ellison 15 et al. . (1982) "The nucleotide sequence of a human immunoglobulin C-gamma-1 gene", Nucl. Acid Res. 10, 4071-4079 and Takahashi et al. (1982) "Structure of human immunoglobulin gamma genes: Implications for evolution of a gene family". Cell 29, 671-679. Human gamma 2 (γ2) 20 is reviewed in Krawinkel et al. (1982) "Comparison of the hinge-coding segments in human immunoglobulin gamma 2 and gamma" 4 subclass genes, ”EMBO J. 1, 403 - 407, Ellison et al. , (1982) "Linkage and sequence homology of • 25 two human immunoglobulin gamma heavy chain constant region genes", Proc. Natl. Acad. Sci. USA 79, 1984-88 and

Takahashi et al. . see. above. Human gamma 3-.a (γ3) is reviewed in Krawinkel et al. (See above) and Takahashi et al. . (see above). Human gamma 4 (γ4) is reviewed in Ellison et al. (1981) "Nucleotide sequence of a human immunoglobulin C-gamma-4 gene" DNA 1, 11-18, Krawinkel et al., ( see above) and Takahashi et al., (supra).

Human myth is reviewed in Rabbitts 35 et al. . "Human Immunoglobulin Heavy Chain Genes: Evol- 32 1 0 3 4 7 7 utionary Comparisons of Cfi, C0 and C Genes and Associated Switch Sequences", Nucl. Acids Res. 9, 4509-4524.

Human alpha has been reviewed in Flanagan et al. (1984) "Mechanisms of Divergence and Convergence 5 of the Human Immunoglobulin Alpha 1 and Alpha 2 Constant Region Gene Sequences", Cell 36, 681-688.

Human delta is reviewed in White et al. (1985) "Human Immunoglobulin D: Genomic Sequences of the Delta Heavy Chain" Science 228, 733-737.

10 Human Epsilon is reviewed in Max et al., (1982) "Duplication and Deletion in the Human Immunoglobulin ε Genes", Cell 29, 691-699.

Light Chain (CL) Part The CL part may be human kappa (κ) or human lambda 15 (λ).

Human κ has been reviewed in Heiter et al. , (1980) Cloned Human and Mouse Kappa Immunoglobulin

Constant and J Region Genes Conserve Homology in Functional Segments, ”Cell 22, 197 - 207.

Human λ is reviewed in Hollis et al., (1982) "Processed Genes: A Dispersed Human Immunoglobulin Gene Bearing of RNA-Type Processing", Nature 296, 321-325.

The CH and / or CL gene segments can be "altered" by mutagenesis. Examples of these methods include the addition of small amounts of different nucleotides, deletions, or substitution of multiple product nucleotides, or retention of multiple nucleotide properties of the product, provided that an appropriate reading step is maintained. In addition, whole proteins may be altered; regions, such as by replacing CH3 with CH2. This replacement is done at the DNA level by inserting, deleting, or replacing whole sequence exons.

33 1 0 3 4 7 7

Construction of antibodies

immunizations

The first method used to produce antibodies, whether monoclonal or polyclonal with VH regions encoded by DNA derived from VHoTAG, is immunization of the host animal with purified TAG72. Examples of procedures for immunizing a host animal with TAG72 are disclosed in U.S. Patents 4,522,918 and 1,046,612,282, which use an extract of human breast carcinoma as an immunogen; and U.S. Patent Application Publication No. 7-073,685 (publicly available), which uses B72.3 purified TAG72 as the immunogen.

Thereafter, the monoclonal or polyclonal antibodies obtained by immunization procedures are screened to determine which of said antibodies selectively bind to TAG72. This screening can be performed by any of a number of known methods, such as solid phase radioimmunoassay, enzyme-mediated immunosorbent assays, rosette formation assays, and functional inhibition assays. These methods are well known per se.

Synthesis of amino acid sequences Immunoglobulins prepared according to the present invention can be synthesized from the amino acids of their composition. Suitable methods include the Merrifield solid phase method described in J. Am. Chem. Soc. 85, 2149-2154 (1963). This solid phase method for synthesizing amino acid sequences is also: described on pages 1-4 by Stewart and Young in Solid.

Phase Peptide Synthesis [W.H. Freeman & Co., San Francisco (1969)].

34 103477 Construction of DNA

DNA encoding VB and VL

The DNA encoding the antibody heavy and light chain can be obtained from a variety of sources known to those skilled in the art, for example, genomic DNA, cDNA, synthetic DNA, or combinations thereof.

Cells containing the desired sequence can be isolated and the genomic DNA digested with one or more restriction enzymes. In the genomic DNA, he may be associated with or absent from naturally occurring introns. The resulting fragments can then be cloned and screened using a heavy chain J region (JH) probe for the presence of a DNA sequence encoding the polypeptide sequence of interest. DNA fragments isolated by preparative agarose gel electrophoresis 15 are ligated. The combined plaques of the libraries are screened with the mouse JH probe.

DNA can also be obtained from a cDNA library. The heavy or light chain encoding messenger RNA is isolated from a suitable source, either from mature B cells or from a hybridoma culture, using standard RNA isolation methods and using oligo-dT cellulose chromatography to isolate the poly-A messenger RNA. The polyA messenger RNA may additionally be fractionated, if necessary, to obtain sequences large enough to encode the heavy or light chain amino acid sequences of the desired antibody.

The messenger RNA mixture is then made into a cDNA library using a suitable primer, preferably a nucleic acid sequence according to the properties of the desired cDNA. Such a primer can be synthesized based on the 30 amino acid sequences of the antibody. Alternatively, unfractionated poly-A messenger RNA or poly-dT from a cell line producing the desired antibody may also be used. The resulting cDNA is optionally size-fractionated on a polyacrylamide gel and extended with, for example, dC moieties for use in a suitable restriction enzyme, 35103477 for ligation into pBR 322 or other suitable cloning vector, and extended with dG moieties. Of course, alternative means can also be used to construct cDNA-containing cloning vectors using other terminals and the remainder of other vectors, but this is a common and preferred alternative. A suitable host cell strain, typically Escherichia coli (E. coli), is transformed with linked cloning vectors, and successfully transformed cells are identified by, for example, ampicillin or tetracycline resistance or some other phenotypic property in the cloning vector.

Successfully transformed cells are harvested and transferred to microtiter plates or other media for further growth and storage. Inoculation traces of these growing cultures on a nitrocellulose filter are treated with probes consisting of nucleic acid sequences containing bases complementary to the desired sequences in the cDNA. Several types of probes can be used, of which synthetic single stranded DNA sequences labeled with y32P-ATP kinase treatment are preferred. Cells adhered to the nitrocellulose filter are lysed, DNA is denatured and then fixed before reaction with a kinase-treated probe. Clones with successful hybridization are detected by introducing them into a photographic plate, followed by isolation of plasmids from growing colonies and sequencing by known methods to confirm the presence of the desired gene portions.

A: The desired gene fragments are removed and tailored for the control segments to ensure proper reading step upon insertion into the appropriate expression vectors. Nucleotides are typically inserted at the 5 'end to incorporate the initiation signal and a restriction endonuclease site at a convenient site.

Since the nucleotide sequence of VHaTAG sequences has been disclosed by the present inventors, DNA can also be prepared synthetically, for example, using

Biosystems ™ Model 380A DNA Synthesizer and constructed by standard methods.

Finally, an example of methods employing combinations of these methods is cleavage by the polymerase-zinc-chain reaction overlap extension disclosed in Horton et al. (see above). A synthetically produced primer with a so-called "swing tail" can be inserted into a selected sequence, for example genomic DNA. The sections 15 are then amplified and cut together.

DNA encoding CH and CL

DNA encoding the amino acid sequence of the human constant region can be obtained by screening chromosomal DNA of human immunoglobulin producing cells.

20 Vectors

The desired DNA fragment may be inserted into a biologically functional expression carrier which may contain suitable control sequences that are not present in the selected DNA fragment. By "biologically functional" is meant that the expression vehicle is designed to take care of reproduction and / or expression functions in a suitable host, either by propagation as an extrachromosomal unit or by association with the host genome. A large number of vectors are available or can be readily prepared and known to those skilled in the art.

A number of plasmids have been described, such as those described in European Patent Application Nos. 0363676, 0484870 and 0 051 873, which contain a promoter in high reading sequence compatible with the gene which is compatible with the proposed host cell.

The vectors and methods described herein are suitable for use with a wide variety of transformable microorganisms, either prokaryotic or eukaryotic. The plasmid will be able to reproduce in this microorganism, especially the bacterium.

Plasmid vectors, which contain suitable promoters that the microorganism can use to express its own protein, generally also contain control sequences, ribosome binding sites, and transcription termination sites. Replicon and control sequences derived from species compatible with the host cell are generally used in conjunction with these host cells.

Smaller or larger SV40 fragments may also be used, provided that a sequence of about 250 base pairs (bp) extending from Hind III to Pvu II, which is located at the viral origin of replication, is included. In addition, it is also possible and often desirable to use promoter or control sequences that are usually associated with the desired gene sequence, provided that these control sequences are compatible with the host cell system.

It is desirable for the plasmid to eventually contain a gene, a marker gene, which could provide a pheno-like property that allows selection of host cells containing the expression vector. Particularly useful is the gene which provides for selection on the basis of survival. Survival-based selection is accomplished by providing resistance to the growth inhibitory agent or by providing a growth factor-forming ability for a bacterium lacking it.

Prokaryotic organisms are generally preferred. For example, plasmid pBR322 38 103477 [Bolivar et al. (1977) Gene 2, 95] derived from E. coli is particularly useful. pBR322 contains the ampicillin and tetracycline resistance genes and provides an easy means to identify transformed cells.

Although these prokaryotic organisms are most commonly used, other microbial strains including E. coli strains such as E. coli B, E. coli K12 strain 294 (ATCC No. 31446) and E. coli X1776 (ATCC No. 31537) may be used. , E. coli W3110 (F, γ-, prototrophic, ATCC No. 10,27325), bacilli such as Bacillus subtilis and other members of the enterobacteriaceae group, such as Salmonella tvohimurium or Serratia macrcesans, and can be used for several different species of Pseudomonas spp. These examples are intended to be illustrative only.

In addition to prokaryotes, eukaryotic microbes can be used. Of the eukaryotic microorganisms, Saccharomvces cerevisiae is the most commonly used. that is, ordinary baking yeast, although other strains are commonly available.

For example, expression in Saccharomyces yeast is generally carried out using, for example, plasmid YRp7 [Stinchcomb et al., (1979) Nature 282, 39, Kingsman et al. . (1979) Gene 7, 141 and Tschemper et al. , (1980) Gene 10, 157]. This plasmid contains the trpl gene, which provides a selectable marker for a mutant strain of yeast lacking the ability to grow in tryptophan, for example ATCC No. 44076 or PEP-1 (Jones, (1977) Genetics 85, 12). The presence of the Trpl lesion as a feature of the host cell genome provides an effective environment for detecting transformation based on growth without tryptophan.

For use in yeast, any plasmid vector containing a yeast compatible promoter, an origin of propagation, and an end point is suitable.

Suitable promoter sequences in yeast vectors include 39 103477 3-phosphoglycerate kinase [Hitzeman et al. . (1980) J.

Biol. Chem. 255, 2073] or other glycolytic enzymes [Hess, et al. . (1968) J. Adv. Enzyme Reg. 7, 149,

Holland et al. (1978) Biochemistry 17, 4900].

For use in mammalian cells, control functions of expression vectors are obtained from viral material. Commonly used promoters are derived from polyoma virus, adenovirus 2, and most commonly simian virus 40 (SV40). The early and late promoters of the SV40 virus are particularly useful because both are readily available from the virus as a fragment which also contains the origin of SV40 virus replication (Fiers et al. (1978) Nature 273, 113).

For example, pSV2neo contains the ampicillin-resistant neomycin resistance gene, which is driven by the SV40 promoter. pSV2neo provides a convenient means for identifying cells transformed with genes corresponding to both the variable region of animal origin and the constant region of human origin.

20 Preparation of chimeric DNA

The heavy-chain or light-chain coding genes are constructed by inserting the 51 terminus of the constant region DNA fragment into the 3 'end of the variable region DNA fragment. A DNA sequence encoding the amino acid sequence of an antibody can be recovered from genomic DNA together with promoter and replication sites. To the extent that the host cell recognizes transcriptional control and translational initiation signals associated with heterologous genes, the 5 'and 3' flanking regions of the 30 region encoding variable region may be maintained together with the variable region encoding sequence and used to control transcription and translation initiation. The unencoded 3 'flanking region of the constant region can be maintained due to transcriptional control sequences such as termination and polyadenyl-35 thio-sequences. By reference to the double strand in the 5'-direction or the 3'-direction, it is meant the transcription direction where the 5 'precedes the 3'.

The intron sequence between the variable region corresponding to each respective chain can be linked to the corresponding human constant chain DNA fragment using any convenient restriction site. When a variable region encoding fragment is being acquired for use, it is usually desirable to include a portion of the post-J region intron. When the intron is stored, it is necessary that the ends of the intron have operable recipient and donor sequences. The 5 'flanking coding sequence adjacent to the variable region usually includes those sequences related to transcription initiation and translation, such as the TATA box region, the molecular capping sequence, and the CAAT sequence. Typically, the size of the unencoded 5 'sequence will not exceed about 1 to 2 kilobases (wk).

The enhancer region should occur between the J region and the constant region. The enhancer region used may be either (1) an animal region V, or (2) a human region constant region.

By maintaining the 3 'flanking region of the DNA sequence encoding the constant region in its natural position adjacent to this section, transcription termination signals can be obtained for the gene. Where the function of transcription termination signals in an expression host cell is unsatisfactory, these may be replaced by a 3 'region operating in the host cell. The unencoded 31 region can be conveniently obtained from the unencoded 3 'flanking region of the constant region obtainable from the expression 30 host. The unencoded 31 flanking region can be linked to the constant region by any of the DNA fragment manipulation and insertion methods described above. used as a building block for the construction of the gene.

41, 103477

Manufacture of expression vehicles

A suitable expression vehicle construct containing the desired coding and control sequences can be produced as follows. The ends of the vectors and DNA fragments can then be recombined to form the desired expression vehicles. The methods used do not depend on the source of DNA or on the cell to be hosted.

The DNA fragments encoding the light chain and heavy chain may be inserted into a separate expression vehicle or into the same vector. The fused genes encoding the light and heavy chimeric chains are preferably assembled in different expression vectors that can be used either in parallel or sequentially in the parallel cell transformation of the recipient cell.

Means for inserting DNA fragments containing chimeric genes into expression vectors include the use of restriction endonucleases. "Restriction endonucleases" (or "restriction enzymes") are hydrolytic enzymes capable of catalyzing target-specific DNA Mole-20 cleavage. The restriction endonuclease activity site is determined by the presence of a specific nucleotide sequence. This sequence is called the restriction endonuclease recognition site. Several restriction endonucleases have been isolated from a variety of bacterial species and characterized by nucleotide sequences of their recognition sites. Some restriction thio-endonucleases hydrolyze the phosphodiester bonds of both bands at the same site, leaving the ends flat. Others catalyze the hydrolysis of bonds a few nucleotides apart, forming 30 free single-stranded sites at each end of the cleaved molecule. These single-stranded ends are self-complementary and therefore cohesive and can be used to recombine hydrolyzed DNA. Representative examples of restriction endonucleases include <103477

Aat 11, BamH I, EcoR I, Hind III, Nde I, Spe I, Xba I, Sac I, Bql II, Pst I, Sal I, and Pvu II.

In addition, the expression vector may have a polylinker attached thereto, which has a plurality of unique and restriction sites. Upon digestion of the expression vector with appropriate restriction enzymes, the splice site is cleaved so that at least one gene-containing DNA fragment can be inserted. Where distinct ends are available from a set of junctions, the DNA fragment 10 can be linked in one way, if the ends are the same, resulting in the plasmid having two different orientations to be inserted into the DNA fragment.

Cleavage is performed by treating the plasmids with restriction enzyme (s). Generally, about 10 μg of plasmid is used with about 10 enzyme units in about 100 μΐ of buffer. Endonuclease digestion is usually performed at a temperature range of 37 ° C to 65 ° C at pH 7-9. (Buffers and substrate amounts suitable for certain specific restriction enzymes are specified by the manufacturers.) The reaction time may be from 1 to 18 hours.

It may be useful to prevent digestion of the digested vector by pretreatment with alkaline phosphatase. Specific conditions are described by the manufacturer.

When the restriction enzyme cleavage is complete, the protein is removed by phenol and chloroform extraction. The nucleic acid is recovered from the aqueous phase (containing 0.3 M sodium acetate) by precipitating with ethanol in a volume of about 2.5 times.

For descriptions of restriction enzyme cleavage methods, see Greene et al., Methods in Molecular Biology, Vol. 9, (Wickner, RB, ed.) Marcel Dekker, Inc., New York and Mertz and Davis, 43 103477 (1972). ) "DNA Replication and Biosynthesis", Proc. Natl. Acad. Sci. USA 69, 3370.

The size of the digested fragments is conveniently separated by agar-5 gel electrophoresis to monitor the progress of the reaction. Once the desired level of cleavage has been reached, the endonuclease can be inactivated by heating to above 65 ° C for about 10 minutes or by extraction with an organic solvent.

The desired fragment is then purified from the cleavage results. Suitable purification methods include gel electrophoresis or sucrose gradient centrifugation.

The plasmid carrier and foreign DNA fragments are then ligated with DNA ligase to rearrange the ring structure. This process is called longitudinal assembly and DNA ligation.

A suitably buffered solution containing DNA fragments, DNA ligase and suitable co-factors is used. The temperature used is between 25 ° C and 4 ° C.

When the DNA segments bind to each other by hydrogen bonds, the DNA ligase is able to form a covalent bond between the two segments. The time used for longitudinal joining will vary depending on the temperature used, the nature of the saline solution, and the nature of the adhesive ends, i.e. '25 cohesive ends. Generally, the time required for ligation may be 5-18 hours. See Maniatis, T., Molecular Cloning, Cold Spring Harbor, supra.

Host Cells Expression vehicle constructs 30 can then be used to transform a suitable host cell. Suitable host cells include cells derived from unicellular as well as multicellular organisms.

Chimeric immunoglobulin genes can be expressed in non-lymphatic system cells such as bacteria and yeast.

44 103477

A variety of single-cell microorganisms, such as bacteria, can be transformed. In other words, organisms that have the ability to grow in culture or by fermentation. Because bacteria are generally the 5 most convenient organisms to work with, they are referred to below as an example of other single-cell organisms. Transformation-sensitive bacteria include members of the Enterobacteriaceae such as strains of Escherichia coli and Salmonella · Members of the Bacillaceae such as Bacillus subtilis: Pneumococcus; Streptococcus and Haemophilus influenzae.

When immunoglobulin heavy and light chains are expressed in bacteria, they are incorporated into inclusion bodies. The chains must then be isolated, purified and subsequently assembled into functional immunoglobulin molecules.

In addition to prokaryotes, eukaryotic microbes such as yeast cultures may also be used. Of the eukaryotic microorganisms, Saccharo-20 mvces cerevisiae is the most commonly used. that is, ordinary baking yeast, although other strains are commonly available. The presence of Trpl damage as a feature of the host cell genome provides an efficient environment for detecting transformation by growth without tryptophan.

In addition to microorganisms, cell cultures derived from multicellular organisms may be used as hosts. In principle, any such cell culture is useful, whether from vertebrate or invertebrate cultures, provided that the cell line is at least initially produced by antibodies. Propagation of vertebrate cells for culture has become a routine method in recent years [Tissue Culture, Kruse and Patterson, (eds.) Academic Press 1973]. Examples of these useful host cell lines are 35 Sp2 / 0, VERO and HeLa cells, Chinese hamster ovary 45103477 (CHO) cell lines, and W138, ΒΗΚ, COS-7 and MDCK cell lines.

A preferred recipient cell line is a plasma cytoma cell line, such as B lymphocyte cell or hybridoma cell line 5 and. Plasma cytoma cells can synthesize, assemble, and secrete immunoglobulins encoded by transformed immunoglobulin genes. They also have an immunoglobulin glycosylation mechanism. Sp2 / 0 is a preferred recipient cell because it is a non-immunoglobulin producing plasma cytoma cell. This cell produces only immunoglobulin encoded by the transformed immunoglobulin genes. Plasma tumor cells can be grown in culture or in the abdominal cavity of mice, whereby secreted immunoglobulin can be obtained from ascites fluid.

15 Transformation of host cells

Transformation of host cells is performed as follows. The expression carrier is linearized and the DNA is introduced into host cells for antibody production. Illustrative examples of introducing DNA into host cells include electroporation, protoplast fusion, calcium phosphate precipitation, or other conventional methods using dextran sulfate and PEG.

If cells lacking significant cell wall barriers are used as host cells, transformation can be performed according to the calcium phosphate precipitation method described in Graham and Van der Eb, (1978) Virology 52, 546.

If prokaryotic cells or cells with strong cell wall structures are used, a preferred method of transformation is Cohen, F.N. et al. (1972)

Proc. Natl. Acad. Sci. USA 69, 2110 calcium treatment with calcium chloride.

The host cells can be transformed either by parallel transformation or by target transformation.

46 103477

Light chain and heavy chain coding genes can be used to transform parallel cell cultures of the same or different species; separate light and heavy chain 5 plasmids can be used for co-transformation of a single cell culture or, as a last resort, a single expression plasmid containing both genes and capable of expressing both light and heavy chain 10 genes can be used.

In the target transformation method, the host cells are transformed with genes encoding the light chain and cells containing the light chain marker are selected. The light chain is detected by cell staining or possibly by detecting the light chain in the supernatant if this chain is secreted. For Light Chain Occurrence Selected cells are transformed by the heavy chain construct and the resulting cells are selected for the addition of a heavy chain marker.

It is known that some immortal lymphoid cell lines, such as plasma cytoma cell lines, normally secrete separate light or heavy Ig chains. It follows that if such a cell line is transformed with a chimeric heavy or light chain vector of this invention, it is not necessary to transform this cell line or other cell line with another Ig chain, provided that the normally secreted chain is complementary to the Ig encoded by the original cell line transformation. chain. 30 Selection and expression of transformed host cells

After transformation of host cells, the cells can be grown for 48 hours to allow expression of marker genes. The cells are then transferred to a selective media solution of plant-35, where untransformed cells die while the cells transformed by DNA constructs survive.

Heavy and light chains, or portions thereof, can be produced separately from each other, and antibodies and fragments thereof can be obtained from these. Separate chain assembly methods are required for such preparations.

The ability of the method of the present invention to produce light and heavy chains separately from one another provides the ability to obtain unique combinations of immunoglobulins, Fab regions, and univalent antibodies. It is possible to recombine in Estonia only from the heavy and light chains separated by cleavage of their inter-chain disulfide bonds and to recover antibody activity even without re-formation of the inter-chain disulfide bonds [cf. Edelman, G.M. et al., (1963) Proc. Natl. Acad. Sci. USA 50, 753].

The transformed cells are grown under conditions suitable for the production of light and / or heavy chains, and the protein synthesis of the heavy and / or light chain is determined. Examples of assay methods to be used include enzyme-linked immunosorbent assay (ELISA), radioimmunoassay (RIA), fluorescence-activated cell sorting assay (FACS), immunohistochemistry, or the like.

The binding affinity of monoclonal antibodies for TAG72 is determined by methods known per se [see, e.g. Heyman B., (1984) J. Immunol. Methods 68, 193-204 and the detailed description set forth in the Examples].

Selected positive cultures are subcloned to isolate pure transformed colonies. A suitable method for obtaining subclones is the extreme dilution method taught in McKeara's Monoclonal Antibodies, Plenum Press, N.Y. (1980).

* 48 103477

The hybridomas producing these chimeric antibodies can be grown by known methods. Transformed cells can secrete large amounts of light and / or heavy chains when cultured in Estonia, such as in hollow fiber systems, rotor cultures and static cultures, or in vivo, as in ascites fluid production.

Chimeric antibodies can be produced in large numbers by injecting the hybridoma into pristine-primed mice with pristane and collecting the ascites fluid from mice after a suitable time (about 1-2 weeks), and isolating the monoclonal antibodies therefrom. substances by methods known per se [see Stramignoni, 15 P. et al. . (1983) Inti. J. Cancer 31, 543-552]. Hybrids do grow in vivo, such as cancers in animals, and their serum or ascites fluid can produce monoclonal antibodies up to 50 mg / ml. Injection into mice or rats, usually of about 106 to 107 tissue type 20 compatible hybridoma cells, will result in the development of a cancerous tumor after a few weeks. The antibodies can then be recovered and processed by known methods. [See. generally, Immunological Methods Parts I and II, Lefkovits, I. and Per-25 nis, B. (eds.) (1979 & 1981) Academic Press, New York, N.Y. and Handbook of Experimental Immunology, Weir, D. (ed.) (1978) Blackwell Scientific Publications, St. Louis, MO].

The antibodies can then be stored in a variety of buffer solutions, such as phosphate buffered saline (PBS), which usually provides a stable antibody solution for later use.

Antibodies prepared according to the present invention can be cleaved using known protease enzymes, such as papain or pepsin, to yield 49,103,477.

highly immunoreactive F (ab ') 2, F (ab') and Fab fragments. In addition, Ig-active fragments resulting from proteolysis (about 50,000 mp) can be cleaved into their fully reduced heavy chain and light chain components and reconstituted quite efficiently to form an active antibody [Haber, E. (1964) Proc. Natl. Acad. Sci. USA 52, 1099; and Whitney, P.L. et al. . (1965) Proc. Natl. Acad. Sci. U.S.

53, 524]. The reactivity of the resulting F (ab ') 2, F (ab') and Fab fragments is determined by the same methods used for complete monoclonal antibody.

Uses of Antibodies The antibodies produced according to the present invention, as well as immunoreactive fragments or recombinants thereof, provide unique advantages in that these antibodies have the ability to specifically bind to malignant cells and localize cancerous tumors, detectably bind to normal cells such as fibroblasts, endothelial cells or epithelial cells of the largest organs.

More specifically, these antibodies, as well as immunoreactive fragments or recombinants thereof, are useful in the in vivo treatment of cancer using antibodies prepared in accordance with this invention alone or in combination with a therapeutic agent such as radionuclide, toxin, effector cells, other antibodies or complement mechanism. as described below.

In addition, it is now possible to prepare a pharmaceutical composition comprising the antibodies of this invention in a pharmaceutically acceptable, non-toxic, sterile vehicle, such as physiological saline, non-toxic buffers and the like. .

Injectable compositions may be in the form of a suspension or a solution. In solution form, the complex (or, when desired, discrete components) is dissolved in a pharmaceutically acceptable carrier. These carriers include a suitable solvent, preservatives such as benzyl alcohol, if any, and buffers. Useful solvents include, for example, water, aqueous alcohols, glycols, and phosphonate or carbonate esters. These aqueous solutions contain up to 50% by volume of the organic solvent.

Injectable suspensions require a carrier liquid suspension solution, with or without additives. The suspension solution may be, for example, aqueous polyvinylpyrrolidone, inert oils such as vegetable oils or highly refined mineral oils, or aqueous carboxymethyl cellulose. Suitable physiologically acceptable additives, if required to keep the complex suspended, may include thickening agents such as carboxymethylcellulose, polyvinylpyrrolidone, gelatin or alginates. Many surfactants are also useful sus-suspending agents, for example lecithin, alkylphenol, polyethylene oxide adduct compounds, naphthalene sulfonates, alkylbenzene sulfonates, and polyoxyethylene sorbitan esters. Many agents that affect the hydrophobicity, density, and surface tension of a liquid suspension solution can assist in the preparation of injectable suspensions in individual cases. For example, silicone antifoam agents, sorbitol and sugars are all useful suspending agents.

Because the cancer cells are heterogeneous, and hence one monospecific chimeric antibody of support, 51103477 may not be able to recognize all cells expressing different epitopes of the cancerous tumor.

Thus, it is desirable to administer a variety of chimeric antibodies prepared in accordance with the present invention. The sequential use of these different antibodies should significantly reduce the patient's anti-idiotypic responses compared to the repeated use of one antibody. For example, antibodies to CH92, CH88 and CH44 could be administered sequentially to the patient. Because these antibodies have different light chains and indeed different CDR3 regions, anti-idiotypic responses should be minimized.

In vivo Treatment of Cancer In this method, a conjugate of antibody and therapeutic agent can be introduced into a carcinoma site, whereby the carcinoma tissue is directly exposed to the therapeutic agent.

The antibodies prepared in accordance with the present invention, as well as immunoreactive fragments or recombinants thereof, may be administered in a pharmaceutically effective amount for the treatment of human cancer tumors and their metastases in vivo. A "pharmaceutically effective amount" of an antibody, immunoreactive fragment or recombinant thereof, whether conjugated to a therapeutic agent or unconjugated, means that the amount of said antibodies in the pharmaceutical composition should be sufficient to effect effective binding to those antigens, against which said antibodies have specific affinity. The pharmaceutical composition may be administered in one or more doses.

Methods for preparing and administering the antibody and its immunoreactive fragments and recombinants or conjugates and the therapeutic agent are well known or readily detectable by those skilled in the art. In addition, those skilled in the art will recognize, or readily disclose, that appropriate dosages will depend upon the age and weight of the patient and the therapeutic agent employed. The representative preparation instructions in question are described in the references below.

Examples of antibody-therapeutic conjugates for therapy include: (1) antibodies coupled to radionuclides such as 13XI, 90Y, 105Rh, 47Sc, 67Cu, 212Bi, 211At, 67Ga, 125I, 186Re, 188Re, 177Lu. , 99mTc, 153Sm, 123I and 1U as for example in Goldenberg, DM et al. . (1981) Cancer Res. 41, 4354-4360, Carrasquillo, J. A. et al. . (1984) Cancer

Treat. Rep. 68, 317-328, Zalcberg, J.R. et al. . (1984) J. Natl. Cancer Inst. 72, 697-704, Jones, D.H. et al. (1985) J. Cancer 35715-720, Lange, P.H. et al. (1985) 15 Surgery 98 143-150, Kaltovich, F.A. et al. . (1986) J.

Nucl. Med. 27,897, Order, S.E. et al. (1982) Int. J. Radiother. Oncol. Biol. Phys. 8,259-261, Courtenay-Luck, N. et al. . (1984) Lancet 1 1441-1443, Ettinger, D.S. et al.

et al., (1982) Cancer Treat. Rep. 66,289-297 describes antibodies coupled to drugs or biological response modifiers such as methotrexate, adriamycin or lymphokines such as interferon, such as those described by Chabner, B. et al. . (1985) Can Cer, Principles and Practice of Oncology, Philadelphia, 25 PA, J. B. Lippincott Co., vol. 1 pp. 290 - 328, Oldham, R.K. et al. . (1985) Cancer, Principles and Practice of

Oncology, Philadelphia, PA, J. B. Lippincott Co. part 2 pp. 2223-2245, Deguchi, T. et al. . (1986) Cancer Res. 46 3751-3755, Deguchi, T. et al. . (1985) Fed. Proc. 44 30 1684, Embleton, M.J. et al., (1984) Br. J. Cancer 49 559-565, Pimm, M. V. et al. (1982) Cancer Immunol. Immunother. 12,125-134 are described, (3) toxin-coupled antibodies as described, for example, in Uhr, J. W. et al. , (1983) Monoclonal Antibodies and Cancer, 35 Academic Press, Inc. s 85-98, Vitetta, E.S. et al. , 53, 103477 (1984) Biotechnology and Bio. Frontiers, P. H. Abelson, eds. , pp. 73-85 and Vitetta, E.S. et al. . (1983) Sci.

219,644-650, (4) heterofunctional antibodies, for example those in which another antibody is linked to or linked to the antibody such that the complex binds to both carcinoma and effector cells, e.g., killer cells such as T cells. , such as Perez, P. et al. (1986) J. Exper. Med. 163 166-178, Lau, M.A. et al. (1985) Proc. Natl. Acad. 10 Sci. (USA) 82, 8648-8652, and (5) natural, i.e., unconjugated or uncomplexed antibodies, such as Herlyn, D. et al. . (1982)

Proc. Natl. Acad. Sci. (USA) 79 4761-4765, Schultz, G. et al. al., (1983) Proc. Natl. Acad. Sci. (USA) 80 5407-5441, 15 Capone, P.M. et al. (1983) Proc. Natl. Acad. Sci. (USA) 80,7328-7,332, Sears, H.F. et al. (1985) Cancer Res. 45, 5910-5913, Nepom, G. T. et al. (1984) Proc. Natl. Acad.

Sci. (USA) 81 2864-2867, Koprowski, H. et al. . (1984) Proc. Natl. Acad. Sci. (USA) 81 216-219 and Houghton, A. 20 N. et al. (1985) Proc. Natl. Acad. Sci. (USA) 82 1242-1246 are described.

Methods for combining the antibody or antibody fragment with the desired therapeutic agent are conventional and well known in the art. These are, for example, the methods described in the references cited earlier.

The following non-limiting examples are intended only to illustrate the construction and expression of the antibodies encoding the chimeric DNA sequences of this invention. Unless otherwise stated, all temperatures are in degrees Celsius. All percentages are by weight unless otherwise stated.

54, 103477

eXAMPLES

Replacement of the Mouse Constant Regions The CC antibodies were derived from mice and are significantly less capable of performing effector functions than the human constant regions.

As a result, in the following examples, certain antibodies are "humanized" by genetically removing the constant regions of the heavy and light chains and replacing them with their human counterparts.

The human light chain constant region genes were replaced by the human kappa (k) gene and the murine heavy chain genes were replaced by each of the four human gamma isotypes (γ1, γ2, γ3 and γ4). Each of these four gamma isotypes has unique biological properties. A general review is available from "The Human IgC subclasses", Hamilton, R.G. (1989) doc. No. CB0051-289, Calbiochem Corporation.

Preparation of the Heavy and Light Chain Variable Region 20 Isolation of CC49 Light Chain The CC49 hybridoma cells secrete an antibody of the IgG? Heavy chain and light kappa chain.

Total DNA was isolated from CC49 hybridoma cells, Balb / C mouse kidney cells, and NSI plasma cytoma cells according to the method described in Cell 24, 353-356 (1981).

In general, approximately 10 to 20 μg of extracted DNA from each cell line was completely digested using 80 units of Bam HI, Eco RI, Hind III, Spe I, Xba I, Sac I, Bql 30 II and Pst I in a reaction mixture containing 50 to 100 microliters of appropriate reaction buffer. At 37 ° C overnight.

Next, Southern hybridization of total DNA extracted from each cell line, developed by E.M. Southern, (1975) J.Mol.Biol. 98, 503-355177. The DNA fragments were size fractionated on an electrophoresis on a 0.8% agarose gel. The double-stranded DNA fragments were converted to single-stranded in alkaline solution, after which the nitrocellulose filter was contacted with a gel to transfer the modified DNA sequences to 5 filters in the presence of a high salt solution.

Hybridization was performed using a 32 P-labeled L chain as a probe using random primers.

More specifically, the probe was a Hind III - Pst I fragment of 1.71 kilobases 10 (kb) isolated from the pGD1 plasmid containing coding exons from the JL regions (J1-J5) of the mouse. The nucleotide sequence of the probe fragment is shown in Figure 7. This plasmid is described in "Site Directed Cleavage of Immunoglobulin Gene Segments by Lymphoid Cell Extracts", Agostaro et al. . (1985) Can. J.

Biochem. Cell Biol. 63, 969-976. The plasmid was obtained from Nobumichi Hozumi and John Roder, Mt. Sinai Research Institute, Toronto, Ontario, Canada.

To label the probe with the radioactive label-20, alpha32P-dCTP was purchased from Amersham, Arlington Heights, IL, USA, and a random primer kit was purchased from Pharmacia, Piscataway, NJ, USA.

Signals from Southern transmissions were detected by auto-radiography using Kodak X-OMAT ™ AR film. No 25 zones clearly resulting from the reorganization were observed. Thus, no unique bands were detected in the autoradiogram of CC49 DNA digested with Hind III. However, the results of the Southetn assay could not exclude that the L-chain band resulting from the re-30 deletion organization was covered by a band that migrated in CC49 Hind III digested DNA in a manner similar to that of the mouse kidney cell. Digestion of DNA (representing germline DNA) with Hind 35 III. This was found to be the case.

56 103477

Construction of a Plasmid Containing Mouse VL Genes LAMBDA-ZAP ™, an insertion cloning vector based on a lambda phage, was purchased from Stratagene Co. Ltd. from La Jolla, CA, USA. LAMBDA-ZAP ™ 5 is described on the Stratagene 1987 list on pages 20-21. The cohesive (cos) ends of LAMBDA-ZAP ™ were ligated overnight following the manufacturer's instructions.

Twenty micrograms of LAMBDA-ZAP ™ ligated was digested with 5 microliters (15 units-10 units) of Spe I purchased from New England Biolabs, Inc.

company. The final volume of the digestion solution was 100 microliters. After digestion for 55 minutes, a further 6 units of Spe I were added. After 70 minutes, the reaction was quenched according to Stratagene instructions for use of the new-15 club phenol and ethanol precipitation.

As a result of cleavage with the Spe I restriction enzyme, "sticky ends" were formed at each end. These adhesive ends were transformed with T4 DNA polymerase to give half-filled Spe I adhesive ends, e.g., 51ACT / 3'TCATG. To carry out the half-fill reaction, the DNA knob produced by ethanol precipitation was dissolved in 8 microliters of water. To this were added 2 microliters of 10 millimolar dTTP, 2 microliters of 10 millimolar dCTP, 2 microliters of Stratagene 10X ligase bag 25, 4 microliters of deionized, distilled water and 2 microliters of Klenow fragment obtained from Labor BRL).

The reaction was carried out at room temperature for 30 minutes. The reaction was stopped by inactivating the DNA polymerase 30 at 65 ° C for 10 minutes.

One hundred and sixty micrograms of total CC49 hybridoma DNA (containing the murine light chain promoter and L and VJ exons) was finally digested with Hind III. Fragments of 1-20 kb were excised from 0.8% agarose gels. The DNA was purified using GENECLEAN ™ commercially available from BIO 101 (La Jolla, CA, USA).

Hind III-cleaved total CC49 hybridoma DNAs were half-filled in the same manner as the Spe I fragments of LAMBDA-5 ZAP ™, except that dATP and dGTP were used. Half-filled Hind III digested fragments formed 51AGCTT / 3'GAA adhesive ends compatible with the previously mentioned Spe I half-filled LAMBDA-ZAP ™ fragments. After extraction with phenol and precipitation with ethanol, according to the teachings of Maniatis, Hind III-modified whole fragments of CC49 hybridoma and Spe I-modified LAMBDA-ZAP ™ were ligated by T4 DNA ligase. For the ligase reaction, 6.1 microliters of ligase ligation mixture was used containing the following ingredients: approximately 0.2 micrograms total Hind III modified by CC49 hybridoma in 3 microliters solution, approximately 1 microgram of LAMBDA-ZAP ™ Spe I DNA modified with 1 microliter of solution, 0.6 microliters of 20 Stratagene 10X ligase buffer, 0.5 microliters of 10 mM ATP and 1 microliters of Stratagene-L gas. This was incubated overnight in a water bath and the temperature was gradually lowered from 18 ° C to about 4 ° C. This ligase ligation eliminated both Hind III and Spe I sites.

The genomic library of the ligated mixture was made according to Stratagene instructions. Briefly, 2 microliters of the ligation mixture prepared above was applied to the Stratagene Gigapack Gold packaging system according to the manufacturer's instructions. Fifteen 150 mm plates having a plaque density of 50,000 plaques per plate were screened for positive clones according to the manufacturer's instructions by hybridization to nitrocellulose filters purchased from Schleicher-Schuell, Keene, NH, USA. The 32 P randomly labeled probe from pGD1 previously described was used for hybrid-35 ringing. Two positive clones were obtained.

Both clones were purified through single plaques and recombinant plasmids (phagemids) containing the CC49 L chain of the LAMBDA-ZAP ™ were obtained using the Stratagene automated cleavage protocol. The vector portion of the resulting recombinant plasmid is called pBLUESCRIPT SK (-) and consists of 2,964 base pairs as described in the 1987 Stratagene list. The plasmid map of pBLUESCRIPT SK (-) is shown in Figure 8.

The DNA from two positive clones was partially sequenced and both were identical. One of the clones designated pRLIOI was also used for further studies.

CC49 Light Chain Restriction Mapping pRLIOI was 7.61 kb and the size of the DNA insert was determined to be 4.65 kb by restriction enzyme mapping. The plasmid map of pRLIOI is shown in Figure 9. A restriction enzyme map of the insert of CC49 L-chain genomic DNA in pRL101 is shown in Figure 10.

CC83 light chain variable region isolation

With the following exception, the methods used to isolate the light chain of CC83 were essentially the same as those used to isolate the light chain of CC49.

The 7 X 105 plaque-containing genomic library was screened using a Hind III - Pst I fragment of 1.71 from 32 P randomly labeled pGD1 as previously described. One positive clone was obtained. The positively clone was designated pRL200.

CC83 Light Chain Restriction Mapping pRL200 was 7.44 kb and the size of the DNA insert was determined by restriction enzyme mapping to 4.48 kb to 59 103477 rye. The plasmid map of pRL200 is shown in Figure 11. The restriction enzyme map of the CC83 L-chain genomic DNA insert in pRL200 is shown in Figure 12.

5 Isolation of the CC49 Heavy Chain Variable Region The methods used to isolate the CC49 heavy chain were substantially the same as those used for the isolation of the CC49 light chain, including screening for the same 10 CC49 Hind III-modified DNA.

The hybridization probe used for library screening was constructed from pNP9 containing the 1.98 kb Eco RI-Bam HI fragment encoding exons corresponding to the heavy chain JH3 and J14 of the CC49 immunoglobulin. The nucleotide sequence of the probe fragment is shown in Figure 13. The plasmid was obtained from Dr. Nobumich Hozumi and Dr. John Roder, Mt. Sinai Research Institute, Toronto, Ontario, Canada.

A genomic booklet-20to containing 9.5 X 10s plaque was screened and one positive clone was obtained. This positive clone was designated pHH49.

CC49 Heavy Chain Restriction Mapping pHH49 was 7.0 kb and the size of the DNA insert was determined by restriction enzyme mapping to approximately 4.0 kb. The plasmid map of pHH49 is shown in Figure 14.

CC83 heavy chain variable region isolation

With the following exceptions, the methods used to isolate the heavy chain of CC83 were essentially the same as the '' used to isolate the light chain of CC49.

Approximately thirteen micrograms of ligase-linked LAMBDA-ZAP ™ vector DNA was digested with 12 units of Spe I from New England Biolabs, Inc. in 35 appropriate buffer solutions with a total volume of β 10 103477 100 microliters. LAMBDA-ZAP ™ was digested at 37 ° C for one hour. The reaction mixture was extracted with phenol and ethanol precipitated according to Stratagene instructions. The Spe I digested LAMBDA-ZAP ™ was dephosphorylated 5 Maniatis according to the methods described, except that 40 fold excess of calf intestinal alkaline phosphatase (Boehringer Mannheim, Indianapolis, IN, USA).

The DNA obtained from CC83 was digested to completion using Spe I. Fragments of approximately 3 to 40 kb were isolated from a 0.8% slice of agarose gel by elution with an electric current as described in Maniatis and ligated into a LAMBDA-ZAP ™ Ligase-dephosphorylated vector.

A 5 X 105 plaque-containing genomic library was screened using a probe made from pNP9, the sequence of which is shown in Figure 13. It yielded one positive clone. This positive clone was designated pHS83.

CC83 Heavy Chain Restriction Mapping pHS83 was 7.95 kb and the size of the DNA insert was determined by restriction enzyme mapping to approximately 5 kb. The plasmid map of pHS83 is shown in Figure 15.

25 Sequencing of CC46, CC49, CC83 and CC92 messenger SNA

Total RNA was extracted from about 1 X 10 7 of CC49 cells frozen at -70 ° C in substantially the same manner as reported in Maniatis, with the following 30 exceptions. The solution used consisted of 4M guanidium isothiocyanate and 2.5M sodium citrate and had a pH of 7.0, and centrifugation was performed on a SW40Ti rotor at 31000 rpm.

A total of 2.7 mg of CC49 RNA was isolated.

After centrifugation, the poly-A + messenger RNA was purified from 61,103,477 approximately 1.68 mg of RNA by oligo (dT) cellulose chromatography using a Type 3 oligo (dT) obtained from Collaborative Research, Inc., Bedford, MA, USA. ) cellulose. The method used was in Aviv and Leder (1972) 5 Proc. Nat11. Acad. Sci. (USA) 69, 1408. A total of 50.24 μg of poly-A + messenger RNA was obtained from 1.68 milligrams of messenger RNA.

A total of 3.28 mg of CC83 RNA was isolated from approximately 1 X 10 7 cells. From 10% total RNA of 1.91 mg, 10 54.6 μg of poly-A + messenger RNA was isolated.

A total of 0.814 mg of CC92 RNA was isolated from approximately 2.6 X 10 8 cells. 41.88 μg of poly-A + messenger RNA was isolated from 0.814 mg of total RNA.

A total of 1.7 mg of CC46 RNA was isolated from approximately 15 2.89 X 108 cells. 68.88 μg of poly-A + messenger RNA was isolated from 1.7 mg of total RNA.

Synthetic oligonucleotide primers were synthesized using 380A DNA synthesizer from Applied Biosystems (Applied Biosystems (ABI), Foster City, CA, USA) using ABI-defined phosphoramidite based chemistry. After electrophoresis on a 20% polyacrylamide gel containing 7 M urea, the oligonucleotides were purified according to the manufacturer's instructions. Oligonucleotide concentrations were determined by spect. 25 rophotometric optical density at 260 nm, with 1 unit of optical density at 260 nm corresponding to 33 μg / ml single stranded DNA.

The following oligonucleotide primers were made for messenger RNA sequencing: (1) KL (-), 22-mer for light chains of CC49, CC83, CC92: 5'GGAAGATGGATACAGTTGGTGC-3 ', which is complementary 5 'of the constant region of the murine immunoglobulin kappa chains, is used to determine the messenger RNA sequence closest to the 3' end of the light chain variable region.

103477 o 2

In addition, 49FR1 (-), 17-mer: 5'GGAAGATGGATACATTGGTGC-3 'prepared for CC49 light chain was used to determine the remaining sequence.

In addition, J4 (-), 24-mer: 5'-CCAACTTTGTCCCCGAGCCGAACG-3 'for CC83 light chain, and also 83L CDR2 (-), 17-mer: 5'-CAGGGACTCCAGTGTGC- 3 '10 was used to determine the remaining sequence.

In addition, J5 (-): 5'-CGTTTCAGCTCCAGCTTGGTCCC-3 'prepared for the CC92 light chain was used to determine the remaining sequence.

For the heavy chains of CC46, CC49, CC83 and CC92 1, CH1 (-):, 24-mer: 5'-ATGGATGTAGTTTGGGCAGCAGAT-3 ', which is complementary to the 5' constant region of the murine heavy chain, was prepared. head encoding sequence. The 24-mer 20 CH1 (-) is used to determine the heavy chain variable region messenger RNA sequence closest to the 3 'end.

In addition, JH4 (-), 20-mer: 5'-GGTGACTGAGGTTCCTTGAC-31 prepared for CC49 heavy chain was used to determine the remaining period.

In addition, CCH heavy chain blood-derived JH2 (-), 16-mer: 5'-CTGAGGAGACTGTGAG-31 was used to determine the remaining period.

In addition, the B92.3 / CC92 HC-20 mer made for CC92 heavy chain and B72.3 heavy chain: 5'-CCTTGAACTTCTCATTGTAC-3 was used to determine the remaining period.

63 103477

The following procedures were performed as outlined in Jan Gelliebter, (1987), BRL FOCUS 9, 1.

The oligonucleotide primers were labeled at their respective ends 5 as follows: 100 ng of oligonucleotide was combined with a solution of 50 mM Tris-HCl (pH 8), 10 mM MgCl 2, 5 mM dithiothreitol and 1 mM spermidine, 100 μΟΐ (γ-32TP) ATP (Amersham, 5000 Ci / millimole) and 7 units of T4 polynucleotide kinase with a volume of 13 μΐ. The reaction was allowed to proceed at 37 ° C for 30 minutes, after which it was warmed to 65 ° C for 5 minutes to inactivate the kinase, and then 7 μΐ of water was added to bring the concentration to 5 ng / μΐ. Labeled primers were stored at -20 ° C until needed.

Individual samples, each containing approximately 13 micrograms of the poly-A + messenger RNA of CC49, CC83, CC92, or CC46, respectively, were resuspended in 10 μΐ for the purpose of pooling the response bands. buffer solution [10 mM Tris-HCl (pH 8.3) and 250 mM KCl].

A 5 ng sample of the terminated labeled oligonucleotide primer was added to each messenger RNA sample, warmed to 80 ° C for 3 minutes and The tin bands were allowed to pool for 45 minutes at 61 ° C using KL (-) and 65 ° C using CH 1 (-) oligonucleotides. AMV reverse transcriptase enzyme (Boehringer-Mannheim) was used for 6 units of each messenger RNA sequencing reaction per target. The remainder of the sequencing was performed as described in BRL FOCUS 9, 1 (1987).

Initial sequencing results indicated that the heavy and light chains were rearranged as follows: the CC49 kappa chain had 35 J5 in use, the CC49 heavy γΐ chain had 64 103477

Jh4, the light chain of CC83 used J4, the gamma1 of CC83 used JH2. The CC46 light kappa chain used J2, the CC46 heavy chain used JH3, the CC92 light chain used J5, and the CC92 5 gamma-1 used JH2.

Figure 16 shows the nucleotide sequence of CC49 VH, where the underlined segments indicate the sequences obtained using oligonucleotide primers with messenger RNA.

Figure 17 shows the VH nucleotide sequence of CC83, where the underlined segments indicate the sequences obtained using oligonucleotide primers with messenger RNA.

The complete nucleotide sequences of CCH VH and CC92 VH shown in Figure 2 were obtained using oligonucleotide-15 primers with messenger RNA.

Figure 4a shows the nucleotide sequence of CC49 VL, where the underlined segments indicate the sequences obtained using oligonucleotide primers with messenger RNA.

Figure 5a shows the nucleotide sequence of CC83 VL, where the underlined segments indicate the sequences obtained using oligonucleotide primers with messenger RNA.

The whole nucleotide sequence of CC92 VL shown in Figure 6 was obtained using oligonucleotide primers with messenger RNA.

. 25 Protein sequence

Purified murine CC49 and CC83 immunoglobulin molecules were sent to Dr. George Tarr, University of Michigan Protein Sequencing Service for NH2-terminal amino acid sequence analysis. Dr. Tarr used the Edman digestion method modified in Tarr, GE, "Manual Edman Sequencing System" Microcharacterization of Polypeptides: A Practical Manual [John E. Shively, ed., Humana Press, Inc., Clifton, NJ]. , 155 - 194 (1986)]. In short, Dr. Tarr reduced 35 and alkylated the immunoglobulin molecules. The light and heavy chains of immuno-globulin molecules were separated by reverse phase HPLC.

Figure 4b shows the amino acid sequence of CC49 VL with underlined amino acid sequence results for the first 24 amino acids of mature CC49 VL. Figure 5b shows the amino acid sequence of CC83 VL with underlined amino acid sequence results for the first 51 amino acids of mature CC83 VL. ASN-20 could not be determined from the CC83 light chain due to the N-linked carbohydrate groups present at this site, as shown in the subsequent PNGase F assay. The Asn-Ile-Thr sequence corresponds to the consensus sequence for the incorporation of Asn-X-Thr / Ser carbohydrates into Asn.

Because the heavy chains of the immunoglobulins CC49 and CC83 are protected at the N-terminus and thus unavailable for amino acid determination, the native glycopeptide was treated with cyanogen bromide (CNBr) to effect cleavage at the methionine groups. Fragmentation resulted in fragments which were purified by reverse phase HPLC. N-terminal amino acid sequencing was performed on CNBR fragments.

The results of the amino acid assay for one CCB VH CNBr peptide fragment are shown as underlined groups in Figure 18. The amino acid assay results for 25 CCH VH CNN peptide fragments are shown as underlined groups in Figure 19. As for CC49 was the case, all other peptide sequences correspond to CNBr fragments obtained from the constant region of mouse γΐ: η.

Determination of N-linked Carbohydrate in the CC83 L-Chain This experiment was performed to verify that the N-linked carbohydrate is attached to the CC83 light chain, presumably at position ASN-20 (see Figure 35b). The enzyme glycopeptidase F (PNGase F), which can be isolated from the filtrate of a bacterial culture of Flavobacterium meninaosepticum (Tarentino, A.L. et al., (1985) Bio Chemistry 24, 4665-4671), is cleaved by ASN: N-linked high mannose and / or double branched complex sugars to form the free carbohydrate structure and the ASP group from the ASN to which it was attached. The difference between the molecular weights of the glycosylated and non-glycosylated forms of the same peptide can be determined by SDS-PAGE.

Twelve micrograms reactions were performed, with or without the use of PNGase F (Boehringer Mannheim, Indianapolis, IN, USA) to affect purified murine CC49, CC83 antibodies and CC11 F (ab '). 2 (positive control) in a solution having a final reaction water volume of 40 microliters. Four microliters of 10x buffer solution (1 M potassium phosphate, 0.1 M disodium EDTA, pH 7.4) was added to each reaction mixture. To the tubes labeled "PNGase F used" was added 7.5 microliters of PNGase 20F and all tubes were incubated at 37 ° C for one hour. 40 microliters of Laemmli 2X sample dilution buffer containing β-mercaptoethanol was added to the reaction tubes. The electrophoresis was performed on a 10% SDS polyacrylamide gel, the gel was stained with Co. 25 massie Brilliant Blue R-250 and decolorized. The results are shown in Figure 20. As shown in Channel 2, a new band (*) occurs in PNGase F-treated CC83 sample but not in untreated CC83 sample (channel 3). The new zone is about 2,000-3,000 Mole-30 Kg lower than the natural light chain! zone indicating the removal of the N-linked carbohydrate moiety. The only consensus glycosylation site for the CC83 light chain is located at ASN 20, so it can be inferred that this is the actual 35 site for glycosylation and this is why it did not appear in the 67 103477 CC83 light chain N-terminal sequence analysis in ASN: as. CC49 light chain motility does not change when treated with PGNase F (channel 6), but a new band is detected with the CC11 heavy chain F (AB ') 2 fragment (channel 4 *), which serves as a positive control. Sequencing results for CC11 heavy chain messenger RNA indicate a consensus glycosylation site in V region (data not shown). Bovine serum albumin (BSA), mp, was used as standards (channel 1). 68,000 and soybean trypsin-inhi-10 bits (STI), mp. 21,500.

DNA sequence

Plasmid DNA was directly sequenced using the Se-quenase DNA sequencing kit purchased from United States Biochemical (USB), Cleveland, OH, 15 USA. The instructions for using USB were followed for sequencing the double-stranded DNA. DNA for each variable region was sequenced using the JH or JL oligonucleotide which, based on the information in the messenger RNA sequence, had been determined to be specific for each of the productively rearranged heavy chain or light chain genes, respectively.

After the initial sequences had been determined, the sequence was expanded using additional primers. Additional primers were synthesized using information collected from previously generated sequences.

Using the technique described above, whole heavy chain variable region exons and light chain variable region exons of CC49 and CC83 were obtained. The DNA sequence was assembled and analyzed using 30 Hitachi DNA sequence analysis software DNASIS ™.

For DNA sequencing, the following oligonucleotide primers were prepared: (1) For each light chain, a CK intron (-): 51-GAAAAC-CTGTGTCTTACAC 3 '.

68 103477 (2) for CC49 light chain CC49 FRI (+): 51-GTACCTGTGGGGACATTG 3 ', and JK5 (-) - 23-mer 5'-CGTTTCAGCTCCAGCTTGGTCCC-3'.

5 (3) for CC83 light chain CC83 CDR2 (-): 5'-CAGGGACTCCAGTGTGC 3 ', CC83 L-intron (-): 5'GACTTCAAGATACAAATGTTAG-3', and JK4 (-) - 20-mer: 10 5 ' -CCAACTTTGTCCCCGAGCCGAACG.

The complete nucleotide sequences of CC49 VL and CC83 VL are shown in Figures 4a and 5a, respectively.

CC49 Heavy Chain JH4 (-) - 20mers: 5'GGTGACTGAGGTTCCTTGAC-3'15 and J14 Intron (-): 5'-GCAATGCTCAGAAAACTCC.

CCH heavy chain JH2 (-) - 16 mer: 51CTGAGGAGACTGTGAG-3 'and Jh2 intron (-): 20'-GCAGTAAAATCTATCTAAGCTG.

From now on, the sequencing of each heavy chain was expanded by the following sequences: CC49 / 83 HC / 5 '(+) 5'-GCACTGCTCATGATATGCAAATC-3', CC49 / 83 HC / 5 '(-) 25 5'-GATTTGCATATCATGAGCAGTGC-3', and CC49 / 83 H-chain FRI (-) 5′-CTCAGCGTCAGACTGCTG-3 ′.

The complete nucleotide sequences of CCH VH and CC83 VH are shown in Figure 2.

Comparisons were made between the characterized messenger RNA sequence and the characterized DNA sequence, as well as between the characterized amino acid sequence and the predicted amino acid sequence from the DNA sequence. Based on these comparisons, plasmid clones were found to contain the correct CC49 and CC83 heavy and light chain DNA encoding variable sequences.

As shown in Figure 2, the predicted amino acid sequences derived from the heavy chain variable region nucleotide sequences of CC49 and CC83 exhibit a significant degree of sequence similarity within the framework regions and the highly variable regions 1 and 2. The highly variable region 3 is quite different due to the two. Recombination of the VH region with various D and JH-10 sequences, precisely because CCH's 1-heavy chain used JH4 and CC83's gamma-1 used JH2.

The substantial DNA sequence homology of the CC49 and CC83 heavy chain variable region genes at the 5 'side indicates that the two heavy chain 15 variable region genes were derived from the same germline exon.

Isolation of the VBaTAG, CC46, CC49, CC83, and CC92 Heavy Chain Germline Precursor Gene The methods used to isolate the CC46, CC49, CC83 and CC92 heavy chain germline precursor gene were essentially the same as those used to isolate the heavy chain variable region of CC49 except that the DNA used to generate the LAMBDA-ZAP ™ letter A10 came from an irrelevant hybridoma cell line .. (i.e., a cell line that produces antibodies that do not significantly bind to TAG72). ). A DNA library containing approximately 900,000 plaques was screened and one positive clone isolated. The positive clone was named pVHaTAG. The size of the pVHaTAG was about 5.2 kb and the size of the DNA insert was determined by restriction enzyme mapping to be about 2.2 kb.

pV 'aTAG DNA Sequence The following oligonucleotide primers were used to determine the pVHaTAG DNA sequence: 35 B72.3 / CC92 HC-20 Mer: 5'-CCTTGAACTTCTCATTGTAC-3' 70 103477 CC49 / CC83 HC 5 '(+) : 5'-GCACTGCTCATGATATGCAAATC-3 'CC49 / CC83 HC 5' (-): 5'-GATTTCGATATCATGAGCAGTGC-3 'VHaTAG IVS (+): 51-CTAAAGTGGAGTCAGGGCCTG-3' VHaTAG IVS (-) CCTG-5 ' 5 VHaTAG CDR2 (+): 5′-GAATGGATTGGATATATTTCTC-31.

The complete nucleotide sequence of VHaTAG is shown in Figure 2.

Isolation of human heavy constant genes Plasmid constructs containing various heavy chain constant regions (ργΐ, ργ2, ργ3 and ργ4) were obtained from Dr. lian R. Kirsch of the National Cancer Institutes, Betesha, Maryland.

These genes were subjected to restriction enzyme mapping to confirm their identity. Restriction maps of human constant regions are shown in Figure 21.

Chimeric light chain

Mouse CC49 V region Hind III site of CC49 light chain genomic DNA localized to the murine intron region between 20 J5 and Ck [see Max, Edward E. et al., (1981) J. Biol. Chem. 256, 5116] was lost in the cloning procedure in which half filled Hind III sites were ligated. 25 half filled Spe I sites in the LAMBDA-ZAP vector. This variant migrated with plasmid pRLIOI (Figure 9). The Hind III site of the intron was reconstituted as outlined in the following steps to allow ligation of the Hind III-Bam HI germline kappa-30 light chain DNA directly into the variable region of the mouse. All steps were performed using standard molecular biology methods known to those skilled in the art and can be found in a manual such as Maniatis.

71 103477 The 1.6 kb Kam Bam HI - Pst I fragment was isolated from pRL101 as previously described. The 2.96 kb Bam HI-Pst I fragment is isolated from pBluescript SK (-) (purchased from Stratagene) as described above. The two fragments are then ligated and the following pRL103 is isolated.

10 ^ N., Hindlll

Do \ Lyp st 'pRL103 EH J5

4.65 kb H

15 \ f V

20

Plasmid pGD1 (previously described) was digested with Pst I and Hind III restriction enzymes to obtain the required 1.03 kb intron-containing fragment and pRL103 was also digested with Pst I and Hind III restriction enzymes at the junction site (polylinker). ) to remove a small DNA fragment.

The resulting fragments were ligated to give a 5.68 kb plasmid, which was assigned the pRL104 code. A partial restriction map of pGD1 and pRL104 is shown below.

72 103477

ARF \ P

/ \ VJ2 5 /} ^? J3 / pGD1 L · J5

10,788 bp I

\ w — Pstl NX Hindlll

Ck 15 ----- / y ^ x - Kinclil pRLICM \

5680 bp I

\ X ^ Pstl ·· 25

BamHI J5

L

Human CK Region 30 Plasmid Phum CK was purchased from Dr. John Roder, Mt.

Sinai Research Institute, Toronto, Ontario, Canada. The plasmid is derived from pBR322 having a 12 kb Bam HI fragment containing the human CK exon attached to it. pBR322 is described on page 171 of the 1987 Stratagene Schedule 35 Ion. The restriction map of the 12 kb Bam HI fragment is shown below [from Heiter, P. et al., (1982) J. Biol. Chem. 257 1516].

phumCk 5

Hindlll \ 'Enhancer

P BamHI

_-1 10 Ck

Phum Ck plasmid was digested with Hind III and Pst I restriction enzymes to yield a 5.0 kb fragment containing the human Ck exon. pRL104 was digested with Fsp I and Hind III restriction enzymes to yield a 4.2 kb fragment containing variable exons of the murine light chain of CC49.

The two resulting fragments were ligated with T4 DNA ligase to yield a 9.2 kb fragment together with the resulting fragment mixture. This mixture was digested with Bam HI to give a 7.7 kb Bam HI CC49 L chain chimeric construct with Bam HI adhesive ends containing both murine variable region exons and human constant ** 25 region (kappa). ) exon. These constructs utilize human enhancer sequences as well as mouse promoter sequences.

A Bam HI chimeric fragment containing both murine light chain variable region exons (L and VJ) and the human constant region kappa (k) exon was ligated to the Bam HI site into pSV2neo (5.6 kb), which is pBR322 a plasmid containing the selectable marker gene neo (purchased from ATCC). The presence of the active neo gene converts the cell into resistance to growth inhibition caused by a drug known as G418, a gene similar to neomycin.

The Bam HI chimeric fragment was ligated to pSV2neo in each orientation as shown below. Both transcriptional orientations of the chimeric light chain gene with respect to the neo gene were constructed. Plasmid pSV2neo was linearized by Bam HI site, dephosphorylated (with the methods described by Maniatis) using calf intestinal Al-10 skirting phosphatase (to prevent self-ligation) and ligated with chimeric described earlier in the CC49 L chain Bam HI fragment.

neo f ^ 150!

W 13.5 Kb I

BamHI Ck

Hind III

25 VJS

psii amp iEcoRI BamH |

neo PRU105 I

K ^ 3-5 kb H — Hincflll

BamHI ZY, 75 103477

Orientations in the transcription of the neo gene and CC49 chimeric light chain are indicated by arrows in pRL120 and pRL105, respectively. Parts derived from pSV2neo are indicated. These plasmids were purified on a large scale from the preparative fermentation (1.0 L) of an E. coli clone amplifying each plasmid. Purified plasmids were used to introduce the chimeric CC49 light chain into SP2 / 0 plasma cytoma cells as described below.

The VL region of mouse CC83 and the human CK region The Hind III site in pRL200 lost during cloning of the CC83 light chain was reconstituted for the same reason as in the CC49 light chain chimeric construct. The reconstitution was performed in the following manner. Plasmid pRL200 was linearized from a unique Nhe I site and its two adherent ends were aligned by filling with dNTPs and DNA polymerase I. A phosphorylated Bam HI linker (purchased from New England Biolabs) was ligated to the ligase-filled site. The new plasmid was designated pRL201 and is shown below.

--SP - '/ HindIII

pRL201 I

7.45 kb 30 H.ndlll \ SpelNhe '/ BamH>

psn * L

76 1 0 3 4 7 7 The 2.5 kb Bam HI - Pst I fragment from pRL201 containing the CC83 light chain variable region genomic DNA was conveniently ligated with the previously obtained CC49 from pRL104. to the 4 kb Bam HI-Pst I vector fragment described in connection with the construction of the light chain con-5, which already contained an Hind III containing intron fragment. The new plasmid was designated pRL202 and is shown below.

10 c, - ^ / Fspl y Hindlll / pRL 202 \\ 15 6.5 kb J] \ / -------

BamHI

2 0 t J4

L V

The approximately 5.05 kb Fsp I-Hind III fragment from pRL202 was isolated and ligated to the human 25 Ck Hind III-Bam HI fragment already described for the CC49 light chain chimeric construct. The construction of the CC83 light chain vector was henceforth performed in the same manner as the construction of the CC49 light chain. The resulting 8.5 kb 30 CC83 light-chain chimeric Bam HI construct was lii; was also ligated into pSV2neo-Bam HI (phosphatase-treated) and plasmids with both possible insert orientations were obtained, as shown in the diagram below.

35 77 1 0 3 4 7 7 amp I EeoRl _ I .BamHl

neo I PPIL203 I

Ί l4-3 kb I

10 BamHl Λξ τγ ^ ξξ ^ νρϊιι

V J4 4 JS

15 amp jEcoRI g ^,

[f PR230 I

ne0 1 · 14.3 kb F

^ BamHl Ck

Orientations in the transcription of the neo gene and CC83 chimeric light chain tran-30 are indicated by arrow in pRL203 and pRL230. These plasmids were purified on a large scale from preparative fermentation (1.0 L) of each plasmid amplifying E. coli clone in a commercially available incubator. Purified plasmids were used to introduce the chimeric CC83 78 78 0 7477 7 into Sp2 / 0 plasma cytoma cells as described below.

All four chimeric light chain plasmid codon constructs can be linearized by restriction enzyme digestion with Aat II. The Aat II site in plasmids is located in a region that is not essential for neo expression of the chimeric light chain gene or selectable marker gene.

Chimeric Heavy Chains 10 Exons of Human Gamma Unchanged Gene Genes

The plasmid vector used to carry the chimeric heavy chain constructs is designated pSV2gpt and is disclosed in Mulligan and Berg, (1984) "Selection of animal cells that express the E. coli gene coding for 15 xanthine-guanine phosphoribosyltransferase", Proc. Natl. Acad. Sci. (USA) 78 (4), 2072-2076. pSV2gpt is a plasmid derived from pBR322 which contains an optional marker gene, guanine phosphoribosyl transferase (gpt), which can be used for selective growth in mycophenolic acid-containing solution. To prepare pSVgpt as a recipient of human Cyl, Cy2, Cy3, Cy4 exons, it was digested

Eco R I and Bam H I. The digested DNA was fractionated on a 4% polyacrylamide gel and the 4.5 kb vector fragment was obtained from the gel by elution with an electric current as described in Maniatis. This linearized plasmid was designated pSV2gpt / R / B and the plasmid map is shown in Figure 22. It is capable of accepting Eco RI -Bam HI terminated fragments.

The 5 'end Hind III sites present in human IgG-L constant region fragments were converted to Eco; As RI sites for cloning to the Eco RI site of pSV2gpt, the Eco RI site of pBR322 was used in cases of γΐ-.η, γ2: η, γ3: η and γ4: η.

79 103477

cyl

The human Cyl exon-containing fragment was generated by digestion and linearization of p1 plasmid with Hind III followed by filling in the Hind III adherent ends using all four dNTPs and Klenow fragment of DNA polymerase to align Hind III ends. . The Eco R I linker was ligated to the flat ends to replace the Hind III site with the Eco R I site. This construct was then digested with Eco RI and Bam HI to release the 7.8 kb fragment containing Cyl exons. This fragment was named Cyl-7.8 kb.

Each fragment was then ligated to Eco RI-Bam HI sites of pSV2-gpt / R / B. This vector (pSV2-gpt-yl-7.2) structure allows the insertion of any hi-15 heavy chain variable region gene (with Eco RI ends) into the Eco RI site of human heavy IgG chain vectors. More specifically, 125 ng of the human Cyl-7.8 kb fragment was ligated with 100 ng of linearized pSV2gpt / R / B vector in 10 μΐ volume using 400 units of T4 DNA ligase (purchased from New England Biolabs - of the company). Frozen competent E. coli DH1 cells obtained from Invitrogen (San Diego, CA) were transformed by ligase reaction according to Invitrogen instructions. The resulting plasmid was designated pSV2gptyl-7.8. The plasmid map of pSV2gptyl-7.8 is shown in Figure 23.

In addition, another shorter fragment containing the Cyl exons was formed. Concern about the chimeric 7.8 kb Cyl fragment, 4.5 kb pSV2-gpt / R / B vector, and 1.9 kb 30 CC49 variable region heavy chain vector (14.2 kb total) ) increased the need to reduce the large size of the 7.8 kb Cyl Eco RI-Bam HI fragment.

The 7.8 kb Cyl coding region comprises only the first third of the 51-terminal fragment.

so '0 \ j4T //

Size reduction was accomplished by converting the latter Pvu II site to Bam HI with the addition of a flat end of the Bam HI switch. The Hind III site of ργ-1: η was converted to the Eco RI site by cleaving the knives with Hind 111, 5 filling the 3 'end to provide a smooth end, and adding Eco R I switches as before. 2.3 Whose Pvu II site below was converted to Bam HI, by subsequent digestion with Pvu 11 and direct attachment of Bam HI couplers to the flat Pvu II ends. Subsequently, this construct was digested with Eco RI and Bam HI to release the 2.3 kb fragment containing the Cyl exons. The truncated Eco RI-Bam HI fragment (2.3 kb) still contains γΐ exons and a 31-terminal polyadenylation sequence. This reduces the final 15 sizes of the vector by 5.5 kb, which makes the resulting construct more manageable (8.7 kb in total).

Approximately 200 ng of the human Cyl-2.3 ke fragment was ligated with 100 ng of linearized pSV2gpt / R / B vector using 400 units of T4 DNA ligase (New England 20 Biolabs) in a 10 μΐ volume. Frozen competent E. coli cells obtained from Invitrogen were transformed by ligase reaction according to Invitrogen instructions. The resulting plasmid was designated pSV2gptyl-2.3. The plasmid map of pSV2gptyl-2,3 is shown in Figure 24.

The other three DNA fragments containing the human IgGm constant sequence ek-Son were also isolated. Cy2 exons were obtained from the plasmid ργ2 as a 4.0 kb Eco RI-Bam HI fragment. Cy3 exons were obtained from the plasmid ργ3 8.0 kb in 30 as an Eco RI-Bam HI fragment. Cy4 exons were obtained from plasmid ργ4 7.6 by Eco RI-Bam HI fragment. The fragments were separately ligated with pSVgpt / R / B as described for cyl-7.7 and Cyl-2.3. Result 81 103477 plasmid maps of existing plasmids are shown with pSV2gpt-y2 in Figure 25, pSV2gpt-y3 in Figure 26 and pSV2gpt-y4 in Figure 27.

Heavy Chain Chimeric Constructs The complete human heavy chain variable region chimeric constructs of γΐ were constructed by inserting a mouse heavy chain variable region exon-containing fragment into human γΐ: η constant region exons as follows.

Subsequently, murine heavy chain variable region genes derived from CC49 and CC83 hybridoma cells were ligated to each of the γΐ - γ4: η containing pSV2-gpt vectors (pSV2gpt-Y1, pSV2gpt-Y2, pSV2gpt-15V2g yP3), .

CC49

The heavy chain variable region CC49 heavy chain variable region exon containing fragment was prepared by digestion with 14 μg pHH49 using 50 units of Eco RI (obtained from BRL) at 37 ° C for 2 hours. The cleavage product was fractionated on a 4% polyacrylamide gel and the Eco I fragment containing 1.9 kb CC49 heavy chain variable region was obtained by elution with an electric current such as Ma-. 25 Niatis's work has been described. This fragment was designated f49R.

A fragment containing the 7.8 kb sequence encoding γ seuraav was prepared as follows.

About 50 μg of vector pSV2gpt Y1-7.8 was digested with 30 Eco RI. The resulting fragment was dephosphorylated '(to prevent self ligation) using calf intestinal phosphatase al skirting as described by Maniatis. The fragment was purified from a 0.8% agarose gel with 82 103477 electric current. This vector was named pSV2gpt ~ Y1-7.8 / R.

The Eco R I site is located at 245 ep. prior to the transcription start site and contains the promoter and the tissue-specific sequences necessary for efficient expression. Introns at the 3 'side of the variable region genes contain murine heavy chain enhancer sequences not found in human IgG heavy chain vectors. As a result, heavy chain chimeric vectores use both mouse promoter and enhancer sequences.

Approximately 325 ng of linearized pSV2gpt I-7.8 / R was ligated with 188 ng of f49R in a 10 μΐ solution containing 1 unit of T4 DNA ligand (BRL). Frozen competent E. coli-AG-1 cells obtained from Stratagene were transformed by ligase reaction according to the manufacturer's instructions. The resulting plasmid was named ρ49γΐ-7.8. Figure 28 shows a plasmid map of ρ49γ1-7.8.

About 50 μg of vector pSV2gptyl-2.3 was digested with Eco RI, as was the case with SV2gptY1-7.8. The resultant fragment was dephosphorylated using calf intestinal alkaline phosphatase as described by Maniatis. The fragment was purified from a 0.8% agarose gel by eluting with an electric current. This linearized plasmid was designated pSV2gpt I-2.3 / R.

Approximately 300 ng of linearized pSV2gptY1-2.3 / R was ligated with 188 ng of f49R in a 10 μΐ solution containing 1 unit of T4 DNA ligand (BRL). Frozen competent E. coli-AG-1 cells obtained from Stratagene (La Jolla, CA) were transformed by a gas reaction according to the manufacturer's instructions. The resulting plasmid was named ρ49γ1-2.3. Figure 29 shows a plasmid map of ρ49γΐ-2.3: η.

83 103477

The plasmids pSV2gpt-y2, pSV2gpt-Y3 and pSV2gpt-y4 were separately digested with Eco RI to give the linear plasmid vectors pSV2gpt-Y2 / R, pSV2gpt ~ Y3 / R and pSV2gpt-Y4, respectively. Each of these three linear plasmid vectors was ligated separately using f49R. Plasmid maps of the resulting plasmids are shown with Figure 30 depicting ρ49-γ2, Figure 31 ρ49-γ3 and Figure 32 p49-Y3 · 10 The CC83 heavy chain variable region of CC83 was constructed in the same manner as CC49. chimeric constructs. The heavy chain variable region CC83 heavy chain encoding fragment exon si-15 was prepared by digestion of 19 μg pHS83 with 50 units Eco RI (purchased from BRL) at 37 ° C for 2 hours. The digestion product was fractionated on a 4% polyacrylamide gel and an Eco I fragment of 2.9 kb CC83 heavy chain variable region exon was obtained by elution with an electric current as described in Maniatis. This fragment was designated f83R.

Approximately 300 ng of the linearized plasmid pSV2gptY1-7.8 / R, obtained as described above, was ligated with 270 ng of f83R in a 10 μ 10 solution containing 1 unit of T4 DNA ligase (BRL). . Frozen competent E. coli-AG-1 cells obtained from Stratagene were transformed by the ligase reaction according to the instructions of Strata-Gene. The resulting plasmid was designated ρ83γ1-7.8. Figure 33 shows a plasmid map of ρ83γΐ-7.8: η 30.

; About 90 ng of the linearized plasmid pSV2gptY1-2.3 / R, obtained as described above, was ligated with 270 ng of f83R in a 10 μΐ solution containing 1 unit of T4 DNA ligase (BRL). Frozen competent E, coli-AG-1 • 103477 84 cells obtained from Stratagene were transformed by ligase reaction according to the manufacturer's instructions. The resulting plasmid was named ρ83γ1-2.3. Figure 34 shows a plasmid map of ρ83γ1-2.3: η.

The plasmids pSV2gpt-γ2, pSV2gpt-γ3 and pSV2gpt-γ4 / R were separately digested as above with pSV2gpt-γ2 / R, pSV2gpt-γ3 / R and pSV2gpt-γ4 / R, respectively. to obtain the linear plasmid vectors pSV2gpt-Y2 / R, pSV2gpt-y3 / R and pSV2gpt-y4 / R, respectively. Each of these three linear plasmid vectors was ligated separately to f83R. The plasmid maps of the resulting plasmids are shown in Figure 30 ρ49-γ2, Figure 31 ρ49-γ3 and Figure 32 p49-Y 3 · 15 Thereafter, all ten ring-shaped plasmid constructs (ρ49γ1-7.8, ρ49γ1-2.3, ρ83γ1-7.8, ρ83γ) , ρ49-γ2, ρ83-γ2, ρ49-γ3, ρ83-γ3, ρ49-γ4 and ρ83-γ4) were linearized for transformation by restriction enzyme digestion of Nde I. The Nde I 20 site in plasmids is in a region not essential for chimeric im expression of the munoglobulin gene or selectable marker gene, gpt. Prior to transformation into the recipient cell, the plasmids must be in linear form to enhance the selective integration of DNA into the host cell genome DNA.

Design verification

Because Eco R I fragments can be ligated by either orientation, the correct orientation of Eco R I fragments was determined by digestion with Neo I. In the constructs presented above, the correct insertion of plasmid constructs was confirmed by performing restriction enzyme site mapping of the plasmid. The restriction enzyme map obtained by restriction enzyme digestion and gel electrophoresis is theoretically compared to a map made of single parent fragments. Based on experience with the transcriptional orientation used in the light. 103477 chain vectors, heavy chain vectors were constructed only in a transcriptional orientation opposite to the gpt gene.

Transformation of plasmids into mouse plasma cytoma cells

Transformation of both light and heavy chain chimeric genes into the same cell results in tetrameric (H2L2) immunoglobulins. The synthesis and secretion of these "chimeric" antibody proteins was accomplished by introducing the chimeric (mouse V region: human C region) genes into mouse plasma cytoma cells (Sp2 /). Transformation was performed by electroporation [Sahaga, B.G. et al. . 15 (1986) J. Immunology 137, 1066].

Expression of the chimeric (mouse V region: human C region) genes in transformed mouse plasma cytoma cells (Sp2 / 0) is accomplished using two different methods. In another form, the light chain genes and the heavy chain genes can be introduced into the cell simultaneously in various mutual relationships. This is referred to as parallel transformation. Alternatively, stable clones bearing the chimeric light chain gene can be made and then used in another. 25 in a form called target transformation.

In both methods, the aim is to provide clones containing both the H and L chain genes that produce the aforementioned intact H2L2 immunoglobulin.

3 0 A. Parallel transformations.

Parallel transformation involves the simultaneous transformation of cells containing both drug resistance markers and subsequent selection with one or both drugs. Heavy-chain and light-35-chain vectors (1: 1 and 1:10, expressed in this order) were initially performed using only neo-selection. Neo-resistant cell lines expressing the first chimeric IgG1 antibodies with detectable TAG72 binding activity were provided. Parallel transfection was performed following closely the work of Cornelia Gorman (1985) "High Efficiency Gene Transfer into Mammalian Cells", DNA Cloning, Part II, D.M. Glover, ed., IRL Press, Oxford, England.

10 B. Object Transformations

Constructs containing light and heavy chimeric immunoglobulin genes were sequentially transformed into mouse plasma cytoma cells Sp2 / 0. Target transformation involves transformation and selection with a vector containing the first drug resistance gene (i.e., the gene for chimeric resistance light chain gene vector), followed by transformation and selection on the second drug resistance gene (i.e., mycophenolic acid resistance). ). Neo Selection

Prior to transformation with pSV2 neo vectors containing chimeric light chain constructs, drug selection was confirmed. 25 n conditions for inhibiting the growth of untransformed Sp2 / 0 plasma cytoma cells by titration with a neomycin-like drug, geneticin (GIBCO). Values for genotoxin concentrations used in drug selection ranges from 100 to 1000 μg / ml. Concentrations greater than 400 µg / ml were found to inhibit Sp2 / 0 cell growth in the tissue culture medium used in the present invention. Construction of Light Chain Containing Cells Mouse Sp2 / 0 plasma cytoma cells were initially transformed with 87 103477 light pSV2 neo vectors. Cells were grown in RPMI 1640 containing 5% fetal bovine serum. The cells were washed in PBS and resuspended in 1 X 10 7 viable cells / ml in PBS. 0.8 ml cells were transferred to an electro-5 mating cell (on ice) containing 20 μg of the light chain Aat II restriction endonuclease linearized pSV2 neo vector (pRL105 and pRL150 for the CC49 chimeric L chain and pRL203 and pRL230 for the CC83 chimeric L chain). Aat II was inactivated by heating 10 samples to 65 ° C for 10 minutes. The linearized DNA was precipitated with ethanol and then immediately dissolved in 10-20 microliters of PBS. After standing for 15 minutes on an ice bath, electroporation was performed using a Gene Pulser electropo-15 radiator with a capacitance expander (BioRad) added at 0.2 kV and 960 μΡ. The time constant (s) was generally about 26 ms.

Following transformation, the cells were allowed to rest in an ice bath for 15 minutes to allow the perturbed membranes to relax. The cells were subsequently suspended in 24 ml of RPMI 1640 solution containing 5% fetal bovine serum (RPMI +) and transferred to a 96- or 24-well plate. To reduce the likelihood of more than one drug-resistant cell per well, the cells were also diluted 10-fold in solution (RPMI +) and plated in another 96-well plate. The cell suspension was incubated at 37 ° C and 5% CO 2.

After 48 hours (to allow drug resistance to develop), the solution was removed and replaced with a solution containing 1 mg / ml geneticin.

After 7-10 days, the clones resistant to the geneticin were further cultured and the cells were screened for chimeric light chains by staining.

88 1 0 3 4 7 7

Cytostaining

An aliquot of the cell-containing solution was centrifuged into cell mats on slides using a Cytospin-2 centrifuge (Shandon, Inc.). After air drying, the cells were fixed in acetic acid / ethanol solution (5 parts acetic acid / 95 parts ethanol). After rinsing the cells 3 times with PBS (without Ca2 + and Mg2 + ions), the slides were placed in a humid chamber (100% relative humidity) and stained with 20 μ1: 113 FITC-conjugated goat antibody against human kappa for 20 minutes, which is an antibody conjugated to a fluorescent dye specific for human kappa light chains. The conjugated antibody was diluted 1: 3 in PBS with 1% BSA. After washing overnight, the slides were decanted using fluoromount-G, a histological decantation solution (purchased from Southern Biotech), under a coverslip. The slides were examined with an Olympus BH-2 microscope equipped with a UV illuminator.

Based on fluorescence intensity, constructs in which the light chain orientation was in a transcriptional orientation opposite to the transcriptional orientation of the neor gene in the vector were found to give the strongest L-chain expression. The present. Therefore, pRL105 was the preferred CC49 L chain and pRL230 was the preferred CC83 L chain construct. As a result of these experiments, the following chimeric light chain-containing cell lines (derived from Sp2 / 0) were used in target transformations: In the case of the CC49 chimeric L-chain, one: a cell line (49K-13-13) expressing CC49-derived a chimeric light chain. This cell line was used for all subsequent target transformations with chimeric heavy chain vectors for constructs employing the chimeric CC49 light chain.

89 103477 In the case of the CC83 chimeric L chain, three cell lines (83K-26-5, 83K-34-10 and 83K-42-2) expressing the chimeric heavy chain derived from CC83 were obtained. One cell line was more stained than the other and 5 had local cytoplasmic immunofluorescence regions. All three cell lines were compared with each other for the magnitude of their relative chimeric antibody production capacity as a screening for transformation with a chimeric CC83 gl heavy chain vector. Electroporation of 83K-34-10 resulted in more chimeric antibody-expressing clones than either of the other two chimeric light chain target cell lines. Therefore, the 83K-34-10 light chain cell line was used as the target for subsequent electroporations using chimeric heavy chain vectors in constructs containing the CC83 light chain variable region.

Construction of gpt resistant clones carrying the CC49 and CC83 chimeric H chain constructs Prior to transformation with pSV2 gpt vectors containing the chimeric heavy chain constructs, the conditions of drug selection in untransformed Sp2 / 0 plasmids were confirmed. Culture Collection 25 (ATCC)] to prevent growth. The conditions for drug selection of cells transformed with pSV2 gpt vectors were much more difficult to confirm. The E. coli gpt gene encoding guanosine phosphoribosyltransferase enzyme provides the ability to utilize xanthine and hypoxanthine as substrates for guanine biosynthesis when the metabolic pathway of mammalian guanine is inhibited (MPA).

Published MPA levels that allow growth of other lymphocyte lines transformed with pSV2-gpt vectors were found to be almost twice as high as 90,103,477 high to allow growth of Sp2 / 0 cells transformed with pSV2-gpt in the tissue culture used in this invention. in the neighborhood of. Thereafter, the optimal concentration of MPA for selection for gpt resistance was found to be 0.1 μg / ml. In addition, the use of aminopterin and thymidine (to more effectively close the guanine pathway) was found to be unnecessary.

Clones producing chimeric 44 antibody

formation 10 CH44-I

49K-13-13 cells were used as a target for chimeric heavy chain constructs. Cells were transformed using 20 μg Nde I digested linearized heavy chain DNA vector (ρ49γ1-7.8 or ρ49γΐ-2.3). Transformation by electroporation was performed as previously for chimeric light chains.

However, after 48 hours, the selection was performed by replacing the solution containing geneticin with 0.3 µg / ml geneticin and mycophenolic acid, 250 µg / ml xanthine and 10 µg / ml hypoxanthine.

Transformed cells grew colonies visible to the colonies within 14 days. At this point, 50 μΐ of supernatant was removed and assayed by ELISA. 25 TAG binding and human IgG unchanged with respect to expression of the non-target region. Wells containing positive TAG-binding cells were further cultured in 24-well plates containing new drug selection solution and allowed to grow for 3-7 days.

• Subcloning was performed as follows.

The number of viable cells was counted and the cells were re-plated in 96-well plates. One plate yielded 50 viable cells and the other 250 viable cells. For re-cloning, 91 103477 non-cloning cells were subcultured in 6-well plates until cell density was high enough for storage in liquid nitrogen.

After further cloning, the clones producing the most chimeric antibody were selected for the production of the chimeric antibody in the bioreactors.

CH44-2

The methods used for sequential transformation of Sp2 / 0 plasma cytoma cells in CC44-1 construct 10 were repeated with the exception that 20 μ <5 p49-y2 was used as the chimeric heavy chain vector.

CC44-3

The methods used for sequential transformation of Sp2 / 0 plasma cytoma cells in the construction of CC44-1 were repeated with the exception that 20 p49-y3 was used as the chimeric heavy chain vector.

CH44-4

The methods used for sequential transformation of Sp2 / 0 plasma cytoma cells in the construction of CC44-1 were repeated except that 20 μg of p49-γ4 was used as the chimeric heavy chain vector.

Generation of clones producing chimeric 88 antibody CH88-1. The methods used for sequential transformation of Sp2 / 0 plasma cytoma cells in the construction of CC44-1 were repeated with the following exceptions.

83K-25-5, 83K-34-10 and 83K-42-2 chimeric CC83 light chain producing cells were transformed as described for CH44-1, with the exception that the heavy chain vector was used. 20 μg of ρ83γ1-7.8 or p83Y1-2.3, a pSV2gpt vector containing the heavy chain gene of chimeric CC83.

92 103477 CC88-2

The sequential transformation of Sp2 / 0 plasma cytoma cells in the construction of CC88-1 was repeated with the exception that 20 μg of p83-y2 was used as the chimeric heavy chain vector.

CH88-3

The methods used for sequential transformation of Sp2 / 0 plasma cytoma cells in the construction of CC88-1 were repeated with the exception that 20 μg of p83-y3 was used as the chimeric heavy chain vector.

CC88-4

The sequential transformation of Sp2 / 0 plasma cytoma cells in the construction of CC88-1 was repeated with the exception that 20 μg of p83-γ4 was used as the chimeric heavy chain vector.

Generation of clones producing chimeric 84 antibody

Because of the high degree of sequence similarity between CC49 and CC83 heavy chain variable regions, chimeric antibodies with light and heavy chains from different strains were generated by mixed site transformations. The heavy chain γΐ isotype vectors of CC49 and CC83 were electroporated, respectively, into chimeric light chain targets: 83K34-10 and 49K13-13 to form both "mixed" combinations. The resulting cell lines were designated CH48-1 and CH84-1, where the first number indicates the descending order of the heavy and light chains respectively. For example, CH48-130 denotes a γΐ isotype having a heavy chain derived from CC49 and a light chain derived from CC83.

The composed antibody CH48-1 did not bind to TAG72. This was not due to its inability to form a bona fideechimeric antibody, as most drug-resistant cell lines produced chimeric IgG 93 103477 (as determined by ELISA assay using goat antibody against human Ig and probe). activity against goat antibody against human IgG alkaline phosphatase conjugate). If binding affinity was present, it was significantly lower than that observed with first generation antibody B72.3, which had approximately one-fold lower affinity for TAG72 than either CC49 or CC83.

Surprisingly, CH84-1 bound TAG72 with the same affinity as its parent strains. This new antibody was generated in our de Novo laboratory. and it has not yet been found in nature.

Competitive studies were conducted to determine the specificity of this new mixed antibody, CH84. It should be noted that both CC49 and CC83 exhibit some degree of competitive recognition against the TAG72 antigen. It was found that CH84-1 competed more closely with CC49 for binding to TAG72 than it competed with CC83. This would suggest that the binding specificity for TAG72 is located in the light chain.

Human γ2, -3 and -4 isotypes were also generated with this mixed antibody and resulted in CH84-2-. 25 CH84-3 and CH84-4 clones.

CH84-1

The methods used for sequential transformation of Sp2 / 0 plasma cytoma cells in the construction of CC44-1 were repeated with the following exception: 49K-13-13 cells that were found to produce CC44 light chain by cell staining were transformed as described for CH44-1 transformation. , with the exception that 20 μg of p83vl-2.3, the pSV2gpt vector containing the CH83 heavy chain gene was used in place of ρ49γΐ-2.3: η, 94 103477 instead of the pSV2gpt vector containing the CH44 heavy chain gene.

CH84-2

The methods used for sequential transformation of Sp2 / 0 plasma cytoma cells in the construction of CH84-1 were repeated except that 20 μg of p83-y2 was used instead of ρ83-γ1-2.3: η.

CC84-3

The methods used for sequential transformation of Sp2 / 0 plasma cytoma cells in the construction of CH84-1 were repeated with the exception that 20 μg of p83-y3 was used instead of ρ83-γΐ-2.3: η.

CH84-4

The methods used for sequential transformation of Sp2 / 0 plasma cytoma cells in the construction of CH84-1 were repeated with the exception that 20 μg of p83-Y4 was used instead of ρ83-γ1-2.3: η.

Purification of recombinant antibodies

Cells expressing the chimeric antibodies were removed from the culture medium by centrifugation and the solution was filtered through a 0.2 μτη filter. The chimeric antibodies were purified from culture supernatants in two steps. Protein A affinity kit (Nygene Corporation, Yonkers, NY) was used for the first purification step. 25 according to the manufacturer's instructions. A maximum of 1.0 L of culture supernatant was passed through a 1 mg capacity pack at 5 mL / min. The kit was washed with phosphate buffered saline to remove the remaining albumin. The chimeric antibody 30 was obtained by elution with 0.1 M sodium nitrate buffer pH 3.0. Fractions containing the chimeric antibody were immediately adjusted to neutral pH with 1M Trizma base solution. Final purification of this solution was accomplished by gel filtration using a Phar-35 Maca Superose 12 HR 16/50 column following the manufacturer's instructions (Phar-95 103477 Macia, Piscataway, NJ) after concentration on an Amicon Centricon 30.

Example:

Generation of immunoglobulin containing the mouse germline V? ATAG variable region

The following examples illustrate a reproducible method for the preparation of an antibody encoded by a DNA sequence derived from VHaTAGs by those skilled in the art.

Components of the Expression VBaTAG Heavy Chain Gene 10

It is possible to form a chimeric mouse-human antibody comprising a heavy chain variable region of mouse V? OTAG germline, a light chain variable region complementary to VHaTAG VH, such as either CC49 or CC83 mouse variable region, and the constant regions of man.

The Hind III DNA fragment of the 2.2 kb V? ATAG VH exon sequence is used as a template to obtain a functionally rearranged VHaTAG variable region. The ιΤ-0μ-ίη ^ οηί3ΐη6 ^ 3 mouse genome is used as a source for murine heavy chain enhancer sequences. The latter sequence is obtained from plasmid pNP9 (see previous example "Isolation of CC49 heavy chain variable region"). Figure 38 shows a hybrid gene. 25, based on Horto et al. , (1989), supra. Four oligonucleotides (oligos) have been designed for use in amplification and modification of target DNA. Oligonucleotide 1 is aligned in length-30 to the 5 'end of the VHaTAG covering the Eco RI site which is: 249 bp from the ATG codeword towards the 5' end. Oligonucleotide 2 binds to complementary sequences at the 3 'end of the exon of VHaTAG and also contains D-coding sequences. The D-segment sequences in oligonucleotide 2 do not overlap with any of the VHaTAG sequences. Oligonucleotide 3 contains sequences complementary to the δ-Ωμ-Άΐ ^ θη 5 'on the γ-96 103477 genome and includes the D-segment (same as oligonucleotide 2) and J-coding sequences. Oligonucleotide 4 joins the ιΤ-μ-βΙηββη 3 'end and contains 1219 bp from JH4 to the Eco R I site 5 at the 3' end. The sequences of these oligonucleotides are as follows:

Oligonucleotide 15 'GTCTAGAATTCATAAAAACTTTATG (25-mer)

Oligonucleotide 2 CAGTGTATTTCTGTAAAAGATCTACTATGGTTACG

10 (35 yards)

Oligonucleotide 3 5 1TCTACTATGGTTACGTGGGGTCAAGGAACCTCAGT-CACCGTCTCCTCAGGTAAGAATGGCCTCTCCAGGTCT 3 '(72-mer) Oligonucleotide 5 5' ACTTCTAAAATGTATTTAGAATTCATTTTC 3 'In this example, D is published in D, Med. 155, 201 published episodes. In oligonucleotides 2 and 3, the D sequence is shown in bold. Any other characterized mouse or human D segment can be used to replace the segments present at oligonucleotides 2 20 and 3 at these positions.

The J segment in oligonucleotide 3 is underlined. It is disclosed in Gough and Bernard, (1981) Proc. Natl. Acad. Sci. (USA) 78, 509 mouse JH4 segment taken from published series. Any other mouse or human J.25 segment can be inserted by replacing the JH4 segment found in oligonucleotide 3.

In oligonucleotides 1 and 4, Eco RI sites (GAATTC) are shown in italics.

Assembling intact V aaTAG genes Two separate DNA amplification reactions were performed *. using the components described above. DNA amplification reaction 1 copies the V? ATAG sequence and inserts a D segment at its 3 'end. DNA amplification reaction 2 copies the mouse intron sequences containing the heavy chain enhancer sequences and incorporates the D and J segments encoded by oligo-35 nucleotide 3. The amplified products resulting from Reactions 1 103477 97 and 2 are gel purified, pooled and oligonucleotides 1 and 4 are added to initiate Reaction 3. In Reaction 3, the products of Reactions 1 and 2 merge longitudinally through a bridge formed by their common D-sections 5. Subsequently, DNA amplification from oligonucleotides 1 and 4 provides the product shown at the bottom of Figure 38. This fragment is digested using Eco RI and gel purified. The modified VHaTAG fragment is ligated to the Eco RI site of pSV2gpt 10 1 (2.3) according to the above example "Heavy chain chimeric constructs". The entire fragment containing the VHoTAG-D-J enhancer region is completely sequenced to ensure that no mutations have occurred during the DNA amplification reactions. The other three heavy chain γ isotypes can be generated by ligation of the same modified VHaTAG fragment into the other three γ containing pSV2gpt vectors (pSV2gpt-γ2, pSV2gpt-y3, pSV2gpt-y4).

Expression of the modified V? ATAG gene The modified VHaTAG gene containing plasmids can be linearized with Nde I and introduced by electroporation into chimeric CC49 and CC83 light chain expressing cell lines (see Example C. Transformations above). Transformed cells are selected for growth. 25 in the presence of geneticin and mycophenolic acid as described in the prefix "C. Target Transformations". The presence of the expressed antibody is monitored by TAG72 ELISA (see "Results, Enzyme-mediated Immunoassays (ELISA)"). The antibody-30 expressed in these cells contains the unchanged γΐ, κ regions of human Ig with the light chain variable region of CC49 or CC83, and the heavy chain variable region derived from modified germline VHaTAG VH exons.

4 98 103477

Below are four examples of modified VHaTAG heavy chain variable region constructs with different D and J segments.

V „- segment D-segment_J-segment

5 VHaTAG #i mouse D (SP2,3) mouse J

VHaTAG #ii human J (D1) mouse J

V aTAG #iii mouse D (SP2,3) human J

VHaTAG #iv human D (Dl) human J

10 Human D-section D1 was obtained from Siebenlist et al.

(1981) Nature, 294, 631. The human JH1 sequence was obtained from Ravetch, et al. (1981) Cell 27, 583.

The formation of VHaTAG #i is described with oligonucleotides 1-4 shown in the table. To form VHocTAGs #ii to #iv 15, the corresponding D and J segments must be exchanged in oligonucleotides 2 and 3. The following oligonucleotides show these changes. Replacement of these oligonucleotides in Reactions 1 and 2 results in the VH0tTAGs #ii to #iv.

20 VHaTAG #ii

Oligonucleotide 2 5 'CAGTGTATTTCTGTAAAAGAGTACTGGTGGT

(34-mer)

Oligonucleotide 3 5 'GTACTGGTGGTGTATTGGGGTCAAGGAACC

TCAGTCACCGTCTCCTCAGGTAAGAATGGCCT 25 CTCCAGGTCT 3 '(72-mer) VHaTAG #iii

Oligonucleotide 2 5 'CAGTGTATTTCTGTAAAAGATCTACTATGG

TTACG (35m)

Oligonucleotide 3 5 'TCTACTATGGTTACGTGGGGCCAGGGCAC

3 0 CCTGGTCACCGTCTCCTCAGGTAAGAATGGCCTCTC- CAGGTCT 3 '(72-mer) VBaTAG #iv

Oligonucleotide 2 5 'CAGTGTATTTCTGTAAAAGAGTACTGGTG

GTGTAT (35-mer) 99 103477

Oligonucleotide 3 5 'GTACTGGTGGTGTATTGGGGCCAGGGCAC

CCTGGTCACCGTCTCCTCAGGTAAGAATGGC CTCTCCAGGTCT 3 '(72-mer)

Results 5 A. Cell lines producing chimeric antibody

Simultaneous identification of light and heavy antibodies was performed using two probe antibodies: 1) goat antibody labeled with fluorescent FITC dye against human kappa, and 2) goat antibody labeled with fluorescent TRITC dye against human IgG .

Cell lines that were positive for both heavy and light chains were further tested for their associated chimeric immunoglobulin production and biological activity, i.e. their binding to TAG72.

Enzyme-mediated Immunoassays (ELISA) The ELISA was used to select a transformed cell producing a chimeric monoclonal antibody. Clones containing constructs for heavy and light chain drug selection were selected for their growth in selective culture medium. The following cell lines were tested: (1) CH44-1: cell line with CC49 VH, CC49 VL and IgG2 constant region, (2) CH44-2: cell line with CC49 VH, CC4 9 VL and IgG2 constant region , (3) CH44-4: cell line with CC49 VH, CC49 VL and IgG4 constant region, (4) CH88-1: cell line with VH, CC83 VL and IgG4 constant region, (5) CH88-2: cell line-30 and with CC83 VH, CC83 VL and IgG2 constant region, (6) CH88-3: cell line with CC83 VH, CC83 VL and IgG3 constant region, (7) CH88-4: cell line with CC83 VH, CC83 VL and IgG4 constant region, (8) CH84-1: cell line with CC83 VH, CC49 VL and IgG3 constant region, (9) 35 CH84-2 : cell line with CC83 VH, CC4 9 VL and IgG2 unmodified region, (10) CH84-3: cell line with CC83 VH, CC4 9 VL and IgG3 constant region, and (11) CH84-4: cell line with CC83 VH, CC4 9 VL and IgG4 constant region.

Supernatants from these cultures were subjected to 5 ELISAs. The presence of antibody to TAG72 chimeric antibody was directly measured by a reaction using an excess of goat-human IgG antibody labeled with an enzyme such as alkaline phosphatase after administration of the antibody to chimeric TAG72 antibody. had been allowed to bind to antigen (TAG72) coated microtiter wells. TAG72 antibody activity was determined to provide a criterion for successful recombination.

After 14 days of culturing, 50 μΐ of the supernatants from wells containing wells of cloned cells were harvested and then re-ELISAed for TAG binding. Samples (50 μΐ) of drug-resistant cell line supernatants were plated in wells of Nunc Immu-lon 96-well plates previously coated with TAG antigen (diluted 1/50). After washing to remove unbound material, the wells were incubated using probe goat human alkaline phosphatase-conjugated goat human IgG antibody (GAHIgG-AP) to identify unchanged regions of plate-bound chimeric antibodies immobilized on human TAG-25 antigen. A second wash, followed by addition of a chromogenic alkaline phosphatase substrate, to remove the unbound probe (GAHIgG-AP) allowed the development of color in those wells that had human TAG binding ability (i.e., chimeric antibodies against TAG72). The absorbance readings from the 405 nm region show the relative production of drug resistant cell lines chimeric antibody 35.

101 103477 CH44-1 TAG72 activity was used as a criterion for successful recombination. The wells of microtiter plates were coated with TAG by incubation with 5 50 µl of 1:75 diluted purified TAG72 [Murano, R., et al. . (1988) Cancer Research 48 4588-4596] for 18 hours at room temperature. The wells were then washed 4 times with phosphate buffered saline (PBS), then protected with BSA in-10 by incubation with 50 µl of 0.5% BSA in PBS for 2 hours. After which they were washed four times with PBS. These dishes will remain unchanged if stored at a humid temperature of 4 ° C. 50 microliters per liter of sample is then added to each well. The control dish is a dish containing fresh solution. All samples were incubated either in a plate at 37 ° C for 90 minutes or in a sealed container at 4 ° C overnight.

Plates were then washed 4 times with PBS and 50 μΐ of a 1: 250 diluted goat human IgG antibody conjugated with alkaline phosphatase (Southern Biotech Assoc.) Was added to 20 wells. The solution was incubated at 37 ° C for 90 minutes. After 4 washes to remove the probe, color development was examined.

To achieve color development, the substrate was incubated at room temperature for 6 minutes in 200 µl ethanolamine-buffered substrate, p-nitrophenyl phosphate (Kirkegaard and Perry), in saline containing 30 µl. The optical density at 450 nm was measured from each plate using a Dynatech micro-titre plate reader (Dynatech Inc.).

35 Sp2 / 0 colonies in wells containing supernatants showing TAG72 binding activity of the chimeric antibody were subcloned by extreme dilution.

103477 102

Individual subclones were selected on the basis of the relatively high production capacity of the chimeric antibody.

CH44-2 The TAG ELISA method used for CH44-1 was repeated except that CH44-2 was used as the antibody.

CH44-3 The TAG ELISA used for CH44-1 was repeated with the exception that CH44-3 was used as the antibody.

CH44-4 The TAG ELISA procedure used for CH44-1 was repeated with the exception that CH44-4 was used as the antibody.

CH88-1 The TAG ELISA method used for CH44-1 was repeated with the exception that CH88-1 was used as the antibody.

CH88-2 The TAG ELISA method used for CH44-1 was repeated except that CH88-2 was used as the antibody.

CH88-3 The TAG ELISA procedure used for CH44-1 was repeated except that CH88-3 was used as the antibody.

CH88-4 The TAG ELISA procedure used for CH44-1 was repeated except that CH88-4 was used as the antibody.

CH84-1 The TAG ELISA procedure used for CH44-1 was repeated except that CH84-1 was used as the antibody.

X03 103477 CH84-2 The TAG ELISA procedure used for CH44-1 was repeated with the exception that CH84-2 was used as the antibody.

CH84-3 The TAG ELISA method used for CH44-1 was repeated except that CH88-3 was used as the antibody.

CH84-4 The TAG ELISA procedure used for CH44-1 was repeated except that CH84-4 was used as the antibody.

CH48-1 The TAG ELISA procedure used for CH4 4 -1 was repeated with the exception that CH84-4 was used as the antibody.

Targeting carcinoma in vivo

The chimeric monoclonal antibodies used in animal experiments and shown in the subsequent tau-20 loci 1-4 were labeled with Na125I using Iodogen (Pierce Chemical, Rockford, IL). Specifically, the volume of 0.1 M sodium phosphate solution containing about 0.5 to 2 mg of purified chimeric monoclonal antibodies was adjusted to 0.5 ml, after which it was added to 50 by 12 cm x 75 cm Iodogen in a coated glass tube followed by the addition of 0.1-0.5 mCi of Na125I (New England Nuclear, Boston, MA). After incubation for 2 minutes at room temperature, the protein was removed from the insoluble Iodogen and the unbound 125I was separated from the antibody by gel filtration through a 10 ml SephadexMT G-25 column using PBS as a buffer solution. The iodination protocol resulted in a labeled chimeric IgG antibody having a specific activity of 0.05-0.25 µCi / µg.

103477 104 '

Female CD1-like cataract-deficient mice were obtained from Charles River at about 4 weeks of age. Nine days later, mice were implanted subcutaneously (0.1 ml / mouse) with LS174T-5 cells (1 X 10 6 cells / animal).

Thyroid-deficient 70-400 mg carcinoma-bearing mice were injected with intravenous injection of 0.5-2.0 μϋΐ (10 -10 50 μς; protein) of chimeric monoclonal antibodies, iodinated as described above, approximately 12-13 days after implantation of the cancer cells. substances in PBS. Groups of five mice were sacrificed at different times by bleeding, normal and carcinoma tissues were excised and weighed, and pulse counts per minute were measured with 15 gamma counters. Pulse rates per minute for each tissue were determined and compared with the results obtained from the carcinoma.

The results for CH44-1 are shown in Tables 1-2 and Figures 39A, 39B and 39C. Results for CH84-1 site 20 are shown in Tables 3-4 and Figures 40A and 40B.

Percentage of Injected Dose per 125 L-Labeled Antibody per Tissue Table 1 CH44-1

Tissue 0.75 h 23.5 h 49.5 h 122 h blood, size 29.70 15.84 8.09 7.31 30 liver 8.13 4.13 2.19 1.96 spleen 6.19 3 , 39 2.12 1.36 Kidney 4.35 2.80 1.52 1.33 Cancer tumor 3.31 25.95 28.83 44.16 Lung 7.34 5.39 2.90 2.36 35 Cancer tumor, wild type 0.18 0.12 0.09 0.11 105 103477

Results A. Cell lines producing chimeric antibody

Simultaneous identification of light and heavy antibodies was performed using two probe antibodies: 1) goat antibody labeled with fluorescent FITC dye against human kappa, and 2) goat antibody labeled with human IgG fluorescent dye TRITC. .

Cell lines that were positive for both heavy and light chains were further tested for their associated chimeric immunoglobulin production and biological activity, i.e. their binding to TAG72.

Enzyme-mediated Immunoassays (ELISA) The ELISA was used to select a transformed cell producing a chimeric monoclonal antibody. Clones containing constructs for heavy and light chain drug selection were selected on the basis of their growth in selective culture medium. The following cell lines were tested: (1) CH44-1: cell line with CC49 VH, CC49 VL and IgG3 constant region, (2) CH44-2: cell line with CC49 VH, CC49 VL and IgG2 constant region, ( 3) CH44-4: cell line with CC49 VH, CC49 VL and? 25 IgG4 constant region, (4) CH88-1: cell line with VH, CC83 VL and IgG? Constant region, (5) CH88 -2: cell line with CC83 VH, CC83 VL and IgG2 constant region, (6) CH88-3: cell line with CC83 VH, CC83 VL and IgG3 constant region, (7) CH88-4: cell line with CC83 VH, 30 CC83 VL and IgG4 constant region, (8) CH84-1: cell line,. with CC83 VH; CC4 9 VL and IgG1 constant region, (9) CH84-2: cell line with CC83 VH, CC49 VL and IgG2 constant region, (10) CH84-3: cell line with CC83 VH, CC4 9 VL and IgG3 constant region; and (11) CH84-4: cell line-35 and having CC83 VH, CC49 VL and IgG4 constant region.

Supernatants from these cultures were subjected to ELISA. 103477106 chimeric TAG72 antibody expression was directly measured by a reaction using an excess of human goat IgG antibody labeled with an enzyme, such as alkaline phosphatase, after the presence of the TAG72 chimeric antibody. ai-5 was allowed to bind to antigen (TAG72) coated microtiter wells. TAG72 antibody activity was determined to provide a criterion for successful recombination.

After 14 days of culturing, 50 μΐ of the supernatants from wells containing wells of cloned cells were harvested, which were then re-ELISAed for TAG binding. Samples (50 μΐ) of drug-resistant cell line supernatants were transferred to wells of Nunc Immu-lon 96-well plates previously coated with TAG antigen (diluted 1/50). After washing to remove unbound material, the wells were incubated using probe goat human alkaline phosphatase-conjugated goat human IgG antibody (GAHIgG-AP) to detect unchanged regions of plate-bound chimeric antibodies immobilized on human TAG-20 antigen. A second wash followed by addition of a chromogenic alkaline phosphatase substrate to remove the unbound probe (GAHIgG-AP) allowed the development of discoloration in those wells that had human TAG binding capacity (i.e., chimeric antibodies to TAG72). The absorbance readings at 405 nm indicate the relative production of chimeric antibody to drug resistant cell lines.

Activity against CH44-1 TAG72 was used as a criterion for successful recombination. The wells of the microtiter plates were coated with TAG by incubation with 35 μl of 1:75 diluted purified TAG72 [Mura-no, R., et al. (1988) Cancer Research 48 4588-4596] for 18 hours at room temperature. . Wells 103477107 were then washed 4 times with phosphate buffered saline (PBS), then protected with BSA by incubation with 50 μΐ PBS in 0.5% BSA for 2 hours at 37 ° C: and washed four times with PBS. These dishes will remain unchanged if stored at a humid temperature of 4 ° C. 50 microliters of sample is then added to each well. The control dish is a dish containing fresh solution. All samples were incubated either in a dish at 37 ° C for 90 minutes or in a sealed container at 4 ° C overnight.

Plates were then washed 4 times with PBS and 50 μΐ of a 1: 250 diluted goat human IgG-15 antibody conjugated with alkaline phosphatase (Southern Biotech Assoc.) Was added to each well.

The solution was incubated at 37 ° C for 90 minutes. After washing 4 times with probe removal, color development was examined.

To obtain color development, substrate 20 was incubated at room temperature for 6 minutes in saline containing 200 μΐ ethanolamine buffered substrate, p-nitrophenyl phosphate (Kirkegaard and Perry). The optical density at 450 nm was measured from each plate on a Dynatech microtiter plate. 25 readers (Dynatech Inc.).

Sp2 / 0 colonies in wells containing supernatants showing TAG72 binding activity of the chimeric antibody were subcloned by extreme dilution. Individual subclones were selected based on the relatively high production capacity of the chimeric antibody.

CH44-2 The TAG ELISA procedure used for CH44-1 was repeated except that CH44-2 was used as the antibody.

108 103477 CH44-3 The TAG ELISA procedure used for CH44-1 was repeated with the exception that CH4 4-3 was used as the antibody.

CH44-4 The TAG ELISA method used for CH44-1 was repeated with the exception that CH44-4 was used as the antibody.

CH88-1 The TAG ELISA procedure used for CH44-1 was repeated except that CH88-1 was used as the antibody.

CH88-2 The TAG ELISA procedure used for CH44-1 was repeated except that CH88-2 was used as the antibody.

CH88-3 The TAG ELISA procedure used for CH44-1 was repeated except that CH88-3 was used as the antibody.

CH88-4 The TAG ELISA procedure used for CH44-1 was repeated except that CH88-4 was used as the antibody.

CH84-1 The TAG ELISA method used for CH44-1 was repeated except that CH84-1 was used as the antibody.

CH84-2 The TAG ELISA method used for CH44-1 was repeated except that CH84-2 was used as the antibody.

CH84-3 The TAG ELISA method used for CH44-1 35 was repeated with the exception that CH88-3 was used as the antibody.

109 103477 CH84-4 The TAG ELISA method used for CH4 4 -1 was repeated with the exception that CH84-4 was used as the antibody.

CH48-1 The TAG ELISA procedure used for CH44-1 was repeated except that CH84-4 was used as the antibody.

In vivo Carcinoma Targeting The chimeric monoclonal antibodies used in animal experiments and shown in Tables 1-4 below were labeled with Na125I using Iodogen (Pierce Chemical, Rockford, IL). Specifically, the volume of 0.1 M sodium phosphate solution containing about 0.5 to 2 mg of purified chimeric monoclonal antibodies was adjusted to 0.5 ml, then added to a 12 cm x 75 cm Iodogen-coated glass tube, followed by the addition of 0.1-0.5 mCi of Na125I (New England Nuclear, Boston, MA). After incubation for 2 minutes at room temperature, the protein was removed from the insoluble Iodogen and the unbound 125I was separated from the antibody by gel filtration through a 10 mL Sephadex * 1-G-25 column using PBS as a buffer solution. The iodination protocol resulted in a labeled chimeric IgG-25 antibody having a specific activity of 0.05 to 0.2 μCi / μg.

Female CD1-like cataract-deficient mice were obtained from Charles River at about 4 weeks of age. After nine days, mice were implanted subcutaneously (0.1 ml / mouse) with LS174T cells (1 X 10 6 cells / animal).

»

About 12 to 13 days after implantation of cancer cells, chimeric monoclonal chimeric monoclonal monoclonal mice of 70-400 mg carcinoma-deficient mice were injected intravenously. iodinated antibodies in PBS as described above. Groups of five mice were sacrificed at different times by bleeding, normal and carcinoma tissues were excised and weighed, and strokes per minute were measured with a gamma counter. The stroke rates per minute for each tissue were determined and compared with the results obtained from the carcinoma.

The results for CH44-1 are shown in Tables 10 1-2 and Figures 39A, 39B and 39C. The results for CH84-1 are shown in Tables 3-4 and Figures 40A and 40B.

Percentage of Injected Dose per 125 I-Labeled Antibody per Tissue Table 1 ------------------------------------ ---------- CH44-1

Tissue 0.75 h 23.5 h 49.5 h 122 h 20 blood, size 29.70 15.84 8.09 7.31 liver 8.13 4.13 2.19 1.96 spleen 6.19 3 , 39 2.12 1.36 kidney 4.35 2.80 1.52 1.33 cancer tumor 3.31 25.95 28.83 44.16 ·: 25 lungs 7.34 5.39 2.90 2.36 cancer tumor, wild type 0.18 0.12 0.09 0.11

As shown in Table 1, at about 122 hours 30 after injection, the percentage of the tumor dose contained in the injected dose was 44.16% for CH44-1. Based on this, CH44-1 efficiently targeted a human cancer tumor in situ. This indicates that the chimeric monoclonal antibodies of this invention were efficiently directed to carcinoma in vivo.

Percentage of injected dose of? -Labeled antibody per organ

Table 2 103477 111 5 CH44-1

Tissue 0.75 h 23.5 h 49.5 h 122 h blood, size 47.72 23.03 13.29 12.01 10 liver 10.97 5.20 3.20 2.69 spleen 1.09 0 , 48 0.25 0.22 Kidney 1.25 0.72 0.42 0.40 Cancer tumor 0.57 3.085 2.823 4.55 Lung 1.20 0.87 0.57 0.37 15 Gastrointestinal tract 6.64 4, 78 3.96 2.83 carcass 43.17 49.68 35.35 29.95 whole body 20 retention 91.30 76.34 53.28 46.20

As shown in Table 2, approximately 122 hours after injection, the percentage of the tumor dose contained in the injected dose was 4.55% for CH44-: 25 L. Based on this, CH84-1 efficiently targeted a human cancer tumor in situ. This indicates that the chimeric monoclonal antibodies of the present invention were effective in targeting carcinoma in vivo and are thus useful in the in vivo treatment of cancer.

Percentage of injected dose of 125 I-labeled antibody per tissue grp

Table 3 103477 112 CH84-1

Tissue 1h 23h 47h 118-119h blood, size 30.68 15.65 6.74 6.49 10 liver 12.55 4.26 2.35 1.57 spleen 10.93 3.35 2, 56 1.70 kidney 5.59 2.51 1.53 1.55 cancerous tumor 4.06 20.52 17.58 30.27 lung 10.77 4.80 2.58 2.24 15 cancerous tumor, wild-type 0.15 0.22 0.20 0.24

As shown in Table 3, approximately 118 hours after injection, the percent-20 content of the tumor in the injected dose was 30.27% for CH84-1. Based on this, CH84-1 efficiently targeted a human cancer tumor in situ. This indicates that the chimeric monoclonal antibodies of the present invention were effective in targeting carcinoma in vivo and are thus useful in the in vivo treatment of cancer.

Percentage of injected dose of 125 I-labeled antibody per organ

Table 4 113 103477 CH84-1

Tissue 1h 23h 47h 118-119h Blood, Size 45.98 22.11 10.08 9.371 10 Liver 13.64 5.34 3.13 1.94 Spleen 1.35 0.49 0.32 0 , 16 kidneys 1.39 0.62 0.38 0.38 cancer tumor 0.59 4.335 3.633 7.02 lung 1.77 0.69 0.42 0.31 15 digestive tract 7.38 4.92 3.41 2, 32 carcass 44.83 52.19 30.32 24.06 whole body 20 retention 93.58 81.00 47.14 45.48

As shown in Table 4, approximately 118 hours after injection, the percentage of the tumor dose in the injected dose was 7.02% for CH84-251. Based on this, CH84-1 efficiently targeted a human cancer tumor in situ. This indicates that the chimeric monoclonal antibodies of this invention were effective in targeting carcinoma in vivo and are thus useful in the in vivo treatment of cancer.

30 Deposition of antibody producing cell lines by chimera

Eleven illustrative examples of light kappa chain-derived cell lines secreting chimeric antibodies were deposited with American Type Culture Collection (ATCC) October 103, 477,114 on October 19, 1988. More specifically, the following cell lines have been deposited: (1) CH44-1: cell line with CC49 VH, CC49 VL and IgG1 constant region (ATCC No. HB 9884), (2) CH44-2: cell line with CC49 VH, CC49 VL and IgG2 constant region (ATCC no. HB 9880), (3) CH44-4: cell line with CC4 9 VH, CC4 9 VL and IgG4 constant region (ATCC No. 9877), (4) CH88-1: cell line with VH, CC83 VL and IgGx constant region (ATCC No. 9882), (5) CH88-2: cell line with CC83 VH, CC83 VL and IgG2 constant region (ATCC No. 9881), (6) CH88-3: cell line, having CC83 VH, CC83 VL and IgG3 constant region (ATCC No. 9876), (7) CH88-4: cell line having CC83 VH / CC83 VL and IgG4 constant region (ATCC No. 9874), (8) CH84-1: cell line,

with CC83 VH, CC4 9 VL and IgG-1 constant region (ATCC 15 No. 9883), (9) CH84-2: cell line with CC83 VH, CC49 VL

and IgG2 constant region (ATCC No. 9879), (10) CH84-3: cell line with CC83 VH, CC49 VL and IgG3 constant region (ATCC No. 9878) and (11) CH84-4: cell line with is CC83 VH, CC49 VL and IgG4 constant region (ATCC No. 9875). While the present invention has been described in detail and with reference to its particular embodiments, it will be apparent to one skilled in the art that various changes and modifications can be made without departing from the spirit and scope of the claims.

Claims (6)

103477 A method for producing a therapeutically useful antibody or fragment thereof which is capable of reacting selectively with a TAG72 antigen, characterized by culturing a cell line such as CH44-1 (ATCC HB 9884), CH44-2 (ATCC HB 9880), CH44-4 (ATCC HB 9877), CH 88-1 (ATCC HB 9881), CH 88-2 (ATCC HB 9876), CH 88-3 (ATCC HB 9876), CH 88-4 (ATCC HB 9874), CH 84-1 (ATCC HB 9883), CH 84-2 (ATCC HB 9879), CH 84-3 (ATCC HB 9878) or CH 84-4 (ATCC HB 9875) and recovering the antibody produced. A process for the preparation of a therapeutically useful antibody or antibody fragment conjugate, characterized in that the antibody or antibody fragment of claim 1 is conjugated to a therapeutic agent. 3. The method of claim 2, wherein the therapeutic agent is a radionuclide, a drug or a biological response modifier, a toxin, or another antibody. A method according to claim 3, characterized in that the drug or biological response modifier is methotrexate, adriamycin or lymphokine. 5. Cell line CH44-1 (ATCC HB 9884), CH44-2 (ATCC HB 9880), CH44-4 (ATCC HB 9877), CH 88-1 (ATCC HB 9882), CH 88-2 (ATCC HB 9881), CH 88-3 (ATCC HB 9876), CH 88-4 (ATCC HB 9874), CH 84-1 (ATCC HB 9883), CH 84-2 (ATCC HB 9879), CH 84-3 (ATCC HB 9878) or CH 84-4 (ATCC HB 9875). 103477