Site News  |   Monitor Keywords  |   Monitor Archive  |   Organizer  |   Account Info  |

Method for producing monoclonal antibodies

Method for producing monoclonal antibodies description/claims

This is a continuation of U.S. patent application Ser. No. 11/324,255, filed Jan. 4, 2006, which is a continuation of U.S. patent application Ser. No. 10/244,894, filed Sep. 17, 2002, now U.S. Pat. No. 7,011,974, which is a continuation of U.S. patent application Ser. No. 07/765,281, filed Sep. 25, 1991, now U.S. Pat. No. 6,475,787, which is a continuation-in-part of U.S. application Ser. No. 07/386,489, filed on Jul. 28, 1989 (now abandoned). These applications are hereby incorporated by reference in their entirety.

This invention was made with government support under grant number R44GM37329-01-02-0379428 awarded by the Department of Health and Human Services. The U.S. government has certain rights in this invention.

The production of monoclonal antibodies using hybridoma cells is now well known in the art. Briefly, isolated antibody-producing lymphocytes from an immunized animal, typically a mouse, are fused with an immortalized cell line, and the resultant hybridomas are screened for the production of the desired monoclonal antibody. Such methods have been successfully used to produce a wide array of antibodies.

However, several inherent shortcomings limit the utility of such methods and the resultant monoclonal antibodies (MAbs). Foremost of those limitations is that the Mabs so produced are essentially murine in nature and reactivity. Use of murine MAbs in human patients, whether for diagnostic or perhaps especially for therapeutic or prophylactic use, incurs a risk of untoward antigenic response by the patient.

In order to avoid such antigenicity, genetically engineered antibodies have been produced which retain the specific antigen-binding domains of the parent murine antibody, while substituting corresponding human antibody domains for part or all of the remaining murine polypeptide regions. It is hoped that such antibodies will not prove antigenic in humans because of their greater resemblance to human antibodies.

Briefly, chimeric antibodies may be produced by isolating the MAb-encoding DNA sequences from a desired hybridoma, excising the portion of the murine DNA which is not required to encode the antigen-binding domains, and replacing such DNA sequences with corresponding human DNA sequences. This has been done in two alternative ways. Firstly, the complete murine variable or V region DNA of each chain can be appropriately joined to human constant or C region DNA sequences. The resultant DNAs encode polypeptides with a murine V and human C domains. Examples are provided by Morrison et al, 1984, Proc. Natl. Acad, Sci. USA 81:6851 and Liu et al, 1987, J. immunol. 139:3521. The antibody V regions are known to encode the antigen-binding portions of the antibody, and the C regions encode the biological effector functions, such as complement fixation. In the second approach, the portions of the murine V regions thought to encode the ‘antigen-binding’ specificity , or complementarily-determining regions (CDRs) are identified, and the same CDRs are used to replace the human CDRs of human V regions linked to human c regions. These are ‘CDR-swap’ antibodies, and examples are provided by Jones et al, 1986, Nature 321:522: Verhoeyen et al, 1988, Nature 332:323; and Reichmann et al, 1988, Nature 332:323. The resultant DNAs obtained by either approach thus encode “humanized” heavy and light chains.

While such genetically engineered antibodies may overcome limitations on the use of murine MAbs, expression of the chimeric DNAs encoding such MAbs or even of cloned murine MAb genes is still problematic. In one approach the DNAs are introduced into murine hybridoma or myeloma cells for heterologous expression. However, such methods have net with only limited success, in large part because of the disappointingly low expression levels achieved thus far. Thus, a continuing need exists for a method for heterologous expression of antibody-encoding DNAs. One object of this invention is to provide an improved heterologous expression system for such DNAs which affords high levels of expression of antibodies, preferably chimeric antibodies.

Heterologous gene expression is typically accomplished by introducing the desired gene (or DNA encoding the desired protein) into a host cell in association with an amplifiable marker such as a gene encoding dihydrofolate reductase (DHFR) .The transfected or transformed host cells are then iteratively subjected to increasing selective pressure such that the number of copies of the marker gene and the associated desired gene are increased. Where the marker is a DHFR gene, the selective agent is methotrexate (MTX), as is well known in the art. However, where heavy and light chain antibody genes are so introduced into a host cell, no practical method exists to ensure that both genes are appropriately amplified. It should be noted that if expression of one chain predominates, then the expression level of the other chain can limit the amount of antibody actually produced. Additionally, heavy chain expression, in the absence of light chain expression may be deleterious to the producing cells. Heavy chain toxicity is discussed in Kohler, G, 1980, Proc. Natl. Acad. Sci. USA 77:2197 and Haas and Wabl, 1984, ibid. 81:7185.

We have found that high expression levels for antibodies depends in part on differentially amplifying the heavy and light chain DNAs to optimize the relative gene copy numbers of the heavy and light chain DNAs. In the practice of this invention, such optimization of relative gene copy number and thus the relative expression levels may be conveniently achieved by introducing the heavy chain and light chain DNAs respectively associated with different amplifiable markers, presumably into different chromosomal locations when the introduced DNA is chromosomally integrated. The heavy chain DNA and the light chain DNA are then separately amplified by application of selective conditions for the respective markers until appropriate optimization of gene expression is achieved.

By way of example; the heavy chain-encoding DNA may be linked to an adenosine deaminase (ADA) gene and the light chain-encoding DNA linked to a DHFR gene. Each of the antibody genes with its respective marker gene is then introduced into the host cells, preferably Chinese Hamster Ovary (CHO) cells by conventional methods. For example, each set of DNA may be introduced into separate CHO cells, e.g. by electroporation, and the resultant transformants fused. The ADA+, DHFR+ CHO cells so obtained contain the heavy chain DNA associated with an ADA gene and the light chain DNA associated with a DHFR gene, each of which DNAs is then specifically amplified by treatment with iteratively increasing amounts of MTX (amplifies light chain DNA, but not heavy chain DNA) and 2′-deoxycoformycin (dCF, amplifies heavy chain DNA but not light chain DNA). During the course of amplification the host cells are analyzed for antibody production (by ELISA). Cells so amplified for optimized antibody production were found to produce MAbs which retained the specific hapten binding characteristics of the parental MAb and which bind complement. Expression levels of about 60 μg/l 106 cells/48 hrs have been obtained, which may be ever further improved by additional rounds of amplification. So far as we are aware, efficient production of antibodies in non-lymphoid cells has never been demonstrated heretofore.

It should be noted that the DNAs encoding the respective chains may be cDNA or genomic DNA. It should also be noted that this invention should be useful for the production not just of cloned antibodies, but also of genetically engineered antibodies such as CDR-swapped antibodies as previously mentioned, and in addition, genetically engineered antibody fragments or derivatives such as Fv, Fab, F(ab)′2 fragments using truncated DNAs and chimeric proteins such as Fab-enzyme and Fab-toxin fusion proteins. Thus, this approach will also be of general value in the production of hetero-dimeric molecules, other than complete antibodies. Examples include other forms of genetically-engineered antibodies, such as Fab and F(ab)2′ forms, and antigen-binding portions, such as a Fab, linked to non-antibody peptide sequences. Examples of the genetic engineering of such molecules are found in Newberger et al, 1984, Nature 312:604; Skerra and Pluckthun, 1988, Science 240:1038; Better et al, 1988, Science 240:1041 and Reichmann et al, 1988, J. Mol. Biol. 203:825.

I. Production of Hybridoma Cells

Hybridoma cell lines producing a desired antibody may be produced by conventional methods such as the well known methods of Kohler and Milstein. Briefly, an animal, preferably a rodent such as a Balb/C mouse is immunized and later re-immunized (boosted) with the desired immunogen, with an adjuvant as desired, as is well known in the art. Assaying the serum of the animal by conventional methods such as a specific ELISA reveals whether the animal is producing an antibody of the desired affinity and avidity. An immunized animal having an appropriate titer of the desired antibody is sacrificed and its spleen removed. The spleen cells are then carefully separated and fused with a suitable myeloma cell line by conventional procedures or otherwise immortalized, as is also well known in the art. The immortalized cells producing the desired antibody are then identified by routine, conventional screening and are then subcloned as desired.

II. Cloning heavy and light chain-encoding DNAs

Methods for cloning immunoglobulin heavy and light chains is well known in the art. See e.g. Beidler et al, 1988, J. Immunol. 141:4053 (genomic) and Liu et al, 1987, Proc. Natl. Acad. Sci. USA 84:3439 (CDNA). Briefly, CDNA or genomic libraries are constructed for the RNA or genomic DNA, respectively, from hybridomas producing a specific antibody of interest, as is known in the art. The immunoglobulin clones from such libraries can be identified by hybridization to DNA or oligonucleotide probes specific for JH or CH sequences for the heavy chain clones, or JL or CL sequences for the light chain clones. The positive clones are then further characterized by conventional restriction endonuclease site mapping and nucleotide sequencing.

III. Expression Vector Construction

Any conventional eukaryotic, preferably mammalian, expression vectors designed for high expression levels, of which many are known in the art, may be used in the practice of this invention. However, in the practice of this invention the expression vector for the light chain antibody DNA contains or is cotransfected with a first selectable, amplifiable marker gene while the expression vector for the heavy chain antibody DNA contains or is cotransfected with a second selectable, amplifiable marker. The two selectable, amplifiable markers must be differentially amplifiable, i.e. must each be susceptible to amplification under conditions which do not result in amplification of the other.

The eukaryotic cell expression vectors described herein may be synthesized by techniques well known to those skilled in this art. The components of the vectors such as the bacterial replicons, selection genes, enhancers, promoters, and the like may be obtained from natural sources or synthesized by known procedures. See Kaufman et al., J. Mol. Biol., 159:601-621 (1982); Kaufman, Proc Natl. Acad. Sci. 82:689-693 (1985). Eukaryotic expression vectors useful in practicing this invention may also contain inducible promoters or comprise inducible expression systems as are known in the art.

• Provisional Patent
• Utility Patent

• What Is a Patent?
• What Is a Trademark or Servicemark?
• What Is a Copyright?
• Patent Laws