CA1190873A

CA1190873A - Recombinant monoclonal antibodies

Info

Publication number: CA1190873A
Application number: CA000406425A
Authority: CA
Inventors: Christopher L. Reading
Original assignee: University of Texas System
Current assignee: University of Texas System
Priority date: 1981-07-01
Filing date: 1982-06-30
Publication date: 1985-07-23
Also published as: JPH084496B2; JPS5859994A; JPH04228067A; EP0068763B2; EP0068763A3; EP0068763A2; JPH0367678B2; US4474893A; JPH0565155B2; DE3276007D1; ATE26464T1; EP0068763B1; JPH04228068A

Abstract

ABSTRACT OF THE DISCLOSURE
Novel recombinant monoclonal antibodies that have binding affinities for two different antigens within a single antibody molecule are described. The antibodies are formed by incubating a quadroma cell or a trioma cell, and separating soluble protein. The antibodies may be used in analytical and diagnostical techniques, targeted delivery of biological and pharmacologic agents to specific cells and the identification and localization of specific antigens, receptors and cell surface substances.

Description

~ ' ~

RECOMBINANT MONOCLONAL ANTIBODIES

BACKGROUND AND PRIOR ART
The present invention relates to the field of monoclonal antibodies. In particular, the invention relates to the creation of new biological entities termed triomas and quadromas, which produce new bifunctional antihodies termed recombinant monoclonal antibodies herein. Recombinant monoclonal antibodies (hereinafter designated RMA~ have a wide range of diagnostic and therapeutic uses, to be described in detail herein.
Antibodies are normally synthesized by lymphoid cells derived from B lymphocytes of bone marrow. The great diversity of antibody specificities is accomplished by immunoglobulin molecules having many structural features in common. Individual antibody molecules of heterogeneous binding specificity differ in their detailed amino acid sequences and even antibodies of the same specificity are usually a mixture of immunoglobulins having different amino acid sequences, although such se~uences may be substantially homologous.
The terms "antibody" and "immunoglobulin'i are used interchangeably hereinO
Individual lymphocytes produce immunoglobulin of a single amino acid sequence. Lymphocytes cannot be directly cultured to produce their specific antibody.
However, Kohler, et al, Nature 256, 495 (1975) demonstrated that a process of somatic cell fusion, specifically between a lymphocyte and a myeloma cell, could yield hybrid cells which grow in culture and produce a specific antibody. Myeloma cells are lymphocyte tumor cells which, depending ~pon the cell strain, frequently produce an antibody themselves, although some "non~producing" strains are known.
The hybrid resulting from somatic fusion of a lymphocyte and a myeloma cell is termed a "hybridoma"
cell herein and in the art generally. In a typical fusion procedure, spleen lymphocytes from an animal ~i\/
J'~, imrnunized against a chosen antigen are fused with myeloma cells. The resulting hybridornas are then dispersed in a series of separate culture tubes or microtitre plate wells to screen for cultures producing a desired antibody. Positive cultures are further diluted to obtain colonies arising from a single c~ll (clones). The clones are again screened for production of the desired antibody. Antibody produced by a cloned hybridoma is termed "monoclonal" herein and in the artO
From genetic studies with lymphocytes and hybridomasr it is known that specific antibodies are coded by DNA segments that are selected from a variety of possible coding segments originally present in germ line cells. As differentiation proceeds, some of the coding segments are rearranged or deleted, so -that fully differentiated lymphocytes are genetically restricted to production of a single antibody~ See Science 212, 1015 (1981). Previous attempts to demonstrate synthesis of more than one antibody by a single cell or clone have been successful only to the extent that myeloma-myeloma fusion cells have been sho~n to produce mixed myeloma proteins (Cotton, R.G.~., e-t al, Nature 244, 42 (1973~).
Monoclonal antibodies are highly specific, being directed against a single antigen only. Furthermore, in contrast to conventional antibody preparations which typically include different antibodies directed against dlfEerent sets of determinants on the same antigen, monoclonal antibodies are directed only against a single determinant on the antigen. ~onoclonal antibodies are useful to improve the selectivity and specificity of diagnostic and analytical assay methods using antigen-antibody binding. A second advantage of monoclonal antibodies is provided by -the fact that they are synthesized in pure form by the hybridorna culture, uncontaminated by other immunoglobulins. Monoclonal antibodies may be ~repared from supernatants of cultured hybridoma cells or from ascites induced by intraperitoneal inocul~tion of hybridoma cells into mice.
;~

The immunoglobulin protein structure is well known.
Immunoglobulin G (IgG) consisks of two heavy protein chains Imolecular weight ~ ~4,000) and two light protein chains (molecular weight ~ 22,500). The heavy chains are covalently joined together by disulfide ~onds and each light chain is ~oined to a heavy chain by disulfide bonds. IgM is characterized by the same basic structure as IgG, in multimeric form. Myeloma cells ~requently secrete light chain monomers or dimers, sometimes termed myeloma proteins or Bence Jones proteins, some of which have capacity to bind an antigen. The 1ight and heavy chains of normal antigens are synthesized by the general mechanisms of protein synthesis in cells. The heavy and light chains are separately synthesized and subsequently joined together.
Chemical reassortment of antibody chains has been attempted in the prior art. Early attempts by Stevenson, G.T., et al (Biochem._J. 108, 375 (1968)), yielded only a minor proportion of heterologous associations. More recently; Peabody, D.S~, et al, Biochemistr~ _ , 2~27 (1980) demonstrated specific heterologous association o~ light chains from different myeloma sourcesO The hybrid molecules showed binding affinity for a ligand which one, but not both, of the parent molecules could bind. Heterologous association of heavy with light chains, or oE heavy-light pairs, was not reported. Raso, V., Cancer, Res. 41, 2073 (1981) has reported construction in vitro of antibody fragments (F(ab')2 fragments) with binding affinity for two ligands. The reported procedure required partial degradation of the antibody molecules with a pepsin prior to reassortment of the fragments, such that the resulting dual-specificity binding proteins were fragmen-ts of antibody molecules.
The use of monoclonzl antibodies for a variety o~
therapeutic purposes has been suggested. A particularl~
attracti~e application is for specifically targeted delivery of drugs to specific tissues or cell types, including tumors. For example, Gulliland~ et al, Proc.NatOAcad.Sci.USA, 7~ ~539 (1980) have reported making chemical conjugates of a monoclonal tumor antibody with diphtheria toxin. The specific binding of the monoclonal an-tibody to the target cells makes it possible to deliver a specific drug, inhibitor or toxin to the desired cells while minimizing any interaction with other cells. Such techniques have depended upon chemical coupling reactions to conjugate the drug or toxin with the monoclonal antibody, with attendant disadvantages of loss of activity, reduced specificity and potential unwanted side reactions. Therefore it would be greatly advantageous to provide a targeted delivery system useful in conjunction with agents which need not be chemically coupled to an antibody molecule.
~
The present invention provides novel, recombinant monoclonal antibodies (hereinafter ~MA) tha-t are bifunctional in the sense oi having binding affinities for two different antigens within a single antibody molecule. A ~MA may bind its antigens simultzneously or sequentially. A ~MA is characterized by any functional test which depends upon the binding of two different antigens by the same antibocly, for example, by the ability to bind sequentially to each of two affinity chromatography columns, one bearing a first immobilized antigen and the other bearing a second immobi:Lized antigen.
RMAs are produced by novel cell types construc-ted for the purpose. One such cell is termed a "quadroma"
herein, and is formed by somatic cell fusion of two hybridomas, each parental hybridoma producing a monoclonal antibody specific for one of the two antigens. Another such novel cell type is termed "trioma" herein, and is formed by fusion of a hybridoma and a lymphocyte, each producing antibodies against one of the two antigens. The light and heavy chains of both parental types will be synthesized in quadroma and trioma cells. If light and heavy chains of both kinds are made in equivalent amounts and combined randomly, at 8~3 least one-eighth o~ the antibodies produced by IgG-producing cells will be bifunctional P~s. From IgM-producing cells, essentially all antibodies produced will be bifunctional in the sense of having at least o~e binding site for each of two antigens.
The construction of triomas and quadromas depends on use of a selection system to distinguish the desired fusions from self-fused and non-fused parental cell types. Most of the selection systems disclosed herein depend upon the construction or isolation of mutant hybridomas which are, in themselves, believed to be novel. The selection system is designed to permit selective growth of hybrids of the two parental cell types, a high proportion of which will produce RMAs.
Quadromas and triomas are cloned by procedures essentially similar to those for cloning hybridomas, except that the cultures will be screened for abili-ty to bind two antigens in a single clone. Further analysis of the bifunctional nature of the RMAs themselves will be carried out by two-stage affinity chromatography or by analytical techniques invol~ing a solid-phase immobilized antigen to facilitate separation of monofunctional from bifunctional molecules.
The potential uses for quadromas, triomas and recombinant monoclonal antibodies are manifold. These include analytical and diagnostic techniques, targeted delivery of biological and pharmacologic agents to specific cells and the identification and localization of specific an-tigens, receptors and cell surface substances. The use of RMAs is advantageous since binding affinity and specificity are unaffected by prior chemical treatment used to covalently attach some sort of tag to a monofunctional antibody molecule. Further, the use of RMAs permits sequential administration of a dye, drug or tracer compound~ thereby expandiny the scope of utility o~ prior art techniques. For example, the RMA may be bound to the first antigen, such as a target cell, in one step, and the second antigen, such as a drug or tracer substance, bound to the comple~ in a .~ . .

7~

suhsequent stepO I'he subse~uent step could be carried out under different conditions than the flrst stepO
~ s may be converted to F(ab')~ fragmen~s of dual specificity, for therapeutic use where rapid renal clearance of the antibody after adminlstration is desired.
DETAILED DESCRIPTION OF THE INVENTION
., ~
An initial s~ep in the production oE recombinant monoclonal antibodies is immunization to provide a population of spleen cells produciny the desired antibody. Immunization may be accomplished by conventional in ivo immunization of an experimental animal, such as a mouse, Erom which spleen cells are subsequently obtained. Alternatively, there are advantages to direct ln vitro immuniza~ion of sp]een cells in culture, such as the method described by Luben, R.A., et al, Proc.Second nt. Lymphokine Workshop, Academic Press, New York, N.Y. (1979). In vitro immunization has the advantages that a large proportion of immune spleen cells may be obtained in less time than required by conventional immunization, and that human cell lines may be immunized without subjecting a human to immunization with potentially harmful substances. A
further advantage is that several antigens can be used at once to prepare hybridomas against several an-tigens simultaneously.
A variety of myeloma cell lines are available for hybridization with mouse or human cellsO Many myeloma strains produce light chain monomers or dimers, and frequently ! although not always, hybridomas derived from such cells continue to excrete these proteins.
Non-producing myeloma strains are preferred for most hybridizations, to avoid production of myeloma proteins by the hybridoma. It is further preEerred to use a hybridoma bearing a genetic selection marker to enable the investigator -to selectively grow only the desired hybrids. A common selection system known in the prior art utilizes a mutant parent resistant to 8-azaguanine.
Such mutants are unable to grow in medium containing .~ , hypoxant~line, aminopterin and thymidine (HAT medium)~
8-azaguanine resistant mutants lack a functional hypoxanthine phosphoribosyl transferase (HPRT). Such cells are unable to grow in the presence of aminopterin.
In conventional hybridoma technology, an 8-azaguanine resistant myeloma strain is commonly used. ~fter fusion, hybrid cells receive a functional ~XPRT sene from the spleen cell parent and are therefore able to grow in HAT medium, while the parental myeloma cells are myeloma-myeloma fusions die. Parental spleen cells and spleen-spleen hybrlds do not replicate in culture, so that no selection against them is required. Myeloma s-trains lacking functional thymidine kinase (T~ ~ are also known. Such strains also fail to survive in ~AT
medium.
Screening for antibody production is a critical step in hybridoma technology. An-tibody functional attributes vary widely. Monoclonal antibodies may differ from one another in binding affinity, abillty to precipitate antigen, ability to inactivate antigen, ability to fix complement and degree of crossreactivity.
Preferably, the screening assay should be designed to depend upon, or approximate, the functional properties desired of the antibody to be produced. ~Iowever, the assay must be sufficiently simple to permit the screening of large numbers of samples. ~lthough the techniques in the prior art vary, the screening process is carried out in two cycles. In the first, the fusion culture is subdivided to permit growth o~ a large number Of cultures, each arising from a relatively small number of hybridomas. For example, if cells at a concentration of 105/ml are fused, yielding 10% total hybridomas (104 hybridomas/ml), 10 ~1 samples of such culture ~ill contain on the average 100 hybridomas per sample. If ~5 the desired antibody occurs at a frequency of 1 in 103, approximately 10 of 100 cultures inoculated with 10 jul each will ~e positive for the desired antibody.
Positive cultures are then subdivided again, this time at the level of .1 to .3 hybridoma cells per culture on the average, to ensure that each culture is a clone (all cells therein derived from a single parent cell, reproducing mitotically)~ ~nasmuch as subculturing and screening procedures are labor-intensive, various techniques have been developed to simplify the procedures. For example, if an antigen can be labeled with a ~luorescent marker, individual cells producing the deslred antibody can be separated by commercially available cell sorting equipment, such as the fluorescence-activated cell sorter (FACS~ manufactured by Becton Dickinson, Inc., Palo Alto, California. The instrument is capable of selectively separating cells bearing the fluorescent marker from a mixed population of cells. Another useful procedure is the soft agar cloning technique described by Sharon, J., et al, Proc.
Nat. Acad. Sci~ USA 76, 1420 (1979), which permits ln situ testing for antibody production.
Procedures for obtaining triomas and quadromas are similar in principle, but more complex in practice since additional techniques for selection must be employed.
For example, if th~ initial hybridoma is isolated by HAT
selection, it will have functional H~RT and will therefore not be a suitable parent in a second round of fusion unless another selection marker is present or the hybridoma is again mutated and selected for 8~azaguanine resistanceO In the case of the fusion of two hybridomas to form a quadroma, there must be means available to select against both parental cell lines. Three selection systems are described, as representative of the techniques and principles which are generally o~erative. Other selection techniques, based on other forms of genetic modification or biochemical inhibition may be employed, as will be readily apparent to those skilled in -the art.
HAT selection may be employed using two separate genetic markers, both of which convey sensitivity to aminopterin. Where one parent hybridoma lacks functional HPRT (HPRT ) and the other lac]cs functional thymidine kinase (TK ), only quadromas produced by ,, g fusion of the -two parent hybridomas wil] survive in HAT
medium. HPRT mutant hybridomas may be obtained by selection for growth in the presence of 8-azaguanine or 6-thioguanine, presented at progressively higher concentrations up to 100 ~uM. TK mutants may be selected by growth in progressively increasing concentrations of 5-bromo-2'-deoxyuridine. The techniques for selection of HPRT and TIC mu-tant hybridomas are essentially similar to those previously described for selection of such mutants in conventional cells (Littlefield, J~W., Proc. Nat. Acad. Sci. U.S.A.
50, 568 (1963))o Selection may also be based on the use of mutant hybridomas resistant to ouabain. Ouabain is an inhibitor of the NA , K dependent ATPase essential for active transport in normal cells. Ouabain-resistant cells are able to survive levels of ouabain which kill normal, ouabain-sensitive cellsO Ouabain-resistance may be used as a selection marker by itself, or in combination with other markers. In a preferred embodiment, a single hybridoma is selected for both ouabain resistance and resistance to either 8-azaguanine (HPRT ~ or 5-bromo-2'-deoxyuridine (TK ). The double mutant hybridoma is used as a universal fuser, to combine with any desired hybridoma to produce a quadroma which can be selectively grown in HAT ouabain medium.
In such medium the universal fuser parent hydrobima will die since, with either TK or HPRT mutations, it cannot grow in HAT medium. The other parent hybridoma is killed because it lacks resistance to ouabain. Any quadroma which has retained a functional TK or HPRT gene while remaining ouabain~resistant will grow selectively in HAT-ouabain medium. The universal fuser is especially advantageous because many of the contemplated uses of RMAs employ a single common binding specificity for one of the two binding affinities of the antibody molecule. For example, the use of a recombinant monoclonal antibody in an enzyme-linked immunosorbent assay (ELISA) for a variety of different antigens would t7~3 require a common binding specificity for the indicator enzyme~ Similarly, targeted drug delivery systems can employ a common specificity site for binding the therapeutic agent and a variable speci~icity for binding tissue-specific or cell-speci:Eic antisens.
While the foregoing selection techniques require the construction of mutant hybridoma strains and depend upon the retention of certain genes in the quadroma fusion product, a third technique, based upon irreversible biochemical inhibitors, requires no mutation. An irreversible biochemical inhibitor is one which binds chemically and which exerts a specific inhibitory action in a cell. with which it has been treated. ~ fusion product combining parent cells treated with two separate inhibitors will be uninhibited due to complementati.on. For example, one parent hybridoma is treated with diethylpyrcocarbonate, the other with iodoacetamide. Both parent strains ultimately die, but fusions between the two survive (see Wright, W.E., Exptl~ Cell Res. 112, 395 (19783).
The techni~u.es of selection and cloning for triomas and quadromas applicable to conventional hybridomas are also applicable for the quadromas and triomas of the present invention. Preferably, a fluorescence-tagged antigen is employed in the de~ect.ion and cloning systems. Individual cells which bind a fluorescent antigen can be separated by a fluorescence activated cell sorter. Such instruments are capable of depositing sin~le cells in individual microtitre wells, thereby greatly reducing the labor assoc.iated with conventional selection and cloning.
The detection of trioma and auadroma clones producing antibodies with binding specificity for two different antigens is strong presumptive evidence of the production of RMAs. Further steps are necessary, in most instances, to isolate RMA free Erom other antibodies which may be produced by the same cell including, for example, antibody molecules having a single specificity~ inactive antibody molecules and '~!

, J~

myeloma proteins. True RMA molecules are immunoglobulins having a dual binding specificity. RMAs are specifically purified by two stages of affinity chromatography in series. The firs-t stage entails the specific binding to an affinity column bearing immobilized _ ~_ _ _ 1 0 ,' -first antigen. Antibody molecules which fail to bind at the first stage pass through the column and are discarded. Antibobies binding to the first column are then eluted with a chaotropic ion buffer and applied, in the second stage, to a second affinity column bearing the second anti~en. Only recombinant monoclonal antibodies which can bind to either column are bound to the second.
After appropriate elution steps, the recombinant monoclonal antibody is obtained in essentially pure form.
The exis-tence o~ RMAs may be detec-ted and quantified by a solid-phase assay, without resorting to two-stage affinity chromatography. For example, the first antigen is immobilized by binding to a solid phase support material. A variety of such solid phase supports and binding techniques are well known in the art. The antibody preparation is then incubated with the solid-phase support to permit binding of any antibody having affinity for the immobilized antigen. The support is then washed to remove non-binding antibody and then incubated with the second antigen, which is tagged with an appropriate marker, such as a radioisotope, fluorescent ligand or conjugated enzyme. While both RMAs of dual specificity and conventional antibodies against the first antigen are capable of binding the immobilized first antigen, only the RMAs will be capable of binding the tagged second antigen. All antibodies capable of binding the second antigen but not the first are removed by the washing step, and therefore do not interfere with the assay. Therefore, both qualitative and quantitative measurement of a recombinant monoclonal antibody in the presence of antibodies of some other specificity is accomplished.
Some of the uses contemplated for RMAs are next descxibed.
A hybridoma providing monoclonal antibody to a tumor-specific antigen is fused with a hybridoma making monoclocal antibody to the toxic suburlit of the 60,000 m.w. toxin from Ricinus communis. The quadroma will produce RMA which can be armed wi-th toxin and used -to a ` /~

bind to tumor cells which would internalize the toxin, which would kill the tumor cells.

///

8~3 A hybridoma making monoclonal antibody to a tumor-specific antigen is fused with a hydriboma making monoclonal antibody to trinitrophenol (TNP~. TNP can be covalently bound to amino gxoups on the exterior surface of liposomes. The liposomes can be used for drug delivery, specifically to the tumor cell, since liposomes can be made to encapsulate chemotherapeutic drugs. The liposomes would be coated with Rl~A which binds to TNP and the RMA would also bind to the tumor, resulting in fusion of the liposomes with tumor cells, and introduction of the drug into the tumor cells. Alternatively an RMA for a cell-specific antigen and a hapten, such as a drug or hormone may be employed for specific and direct delivery of the hapten to the desired cell.
A hybridoma making monoclonal antibody to a hormone, e.g. B subunit of human chorionic gonadotropin, drug or tumor-specific antigen is fused with a hydriboma producing monoclonal antibody to a radioactive hapten labelled to high specific activity with a radioactive isotope7 The quadroma will produce RMA which can be armed with radioactivity. Such RMA may be used for assay, tumor localization or therapy. Choice of isotope depends upon the nature of the intended end use. A
gamma~emitting isotope may be used for immunoassay of drugs, hormones and other haptens in body ~luids, tissue samples, urlne and the like. If the tumor-speciEic antigen, hormone or drug is bound to a solid phase, the RMA could be used in a one-step competition radioimmunoassay. Gamma-emitting isotopes are also useful for tumor localization. High-energy alpha-emitting isotopes are especially useful for therapeutic purposes because of the high energy and short path of alpha-radiation in tissue. Beta~emitting isotopes may be used for assay purposes as well~ but require counting equipment not commonly found in a clinical laboratory.
The hybridoma producing monoclonal antibody to the antigen of diagnostic interest in the preceding paragraph is fused with a hybridoma producins monoclonal 87;~
a, antibody to the enzyme horseradish peroxidase. The quadroma will~

,,f~

` ,~ ~ . t _______._ produce R~ which can be used for a one~s-tep enzyme-lin]ced immunosorbant assay (ELISA~.
A hydridoma producing monoc:Lonal an-tibody to a tumor specific antigen is fused with a hybridoma producing monoclonal antibody to a fluorescent probe.
The quadroma will produce RMA which can be used for the fluorescent microscopic detection of tumor cells in tissue sections, or for enumeration of tumor cells in - cellular suspensions usin~ flow microfluorimetry (FMF3.
The term 'tumor-specific antigen" as used herein will be understood to connote an antigen characteristic of a particular tumor, or strongly correlated with such a tumor. However, the current understanding in the art with respect to tumor-specific antigens is that they are not necessarily unique to the tumor tissue, or that antibodies to them may cross-react with antigens of normal tissue. ~ven where tumor-specific anti~ens are not unique to tumor cells, it frequently occurs that, as a practical matter, antibodies binding to -tumor-specific antigens are sufficiently specific to tumor cells to carry out the desired procedures without unwarranted risk or interference due to cross-reactions. Many factors contribute to this practical specificity. For example, the amount of antigen on the tumor cell may greatly exceed the amount found on normal cells, or the normal cells bearing cross~reactive antigen may be loc~lized remote from the tumor. The antigen in the normal state may only be partially cross-reactive with the tumor-specific antigen. Sometimes, a produc-t specific to the cell-type constituting the tumor may serve as a practical tumor-speclfic antigen. ~or example, the antiboby produced by the lymphocytic leukemia cells may itself be used as an antigen, against which an "anti-idiotype" antibody may be selected to bind specifically to such cells. Therefore the term "tumor-specific antibody" relates herein to a specificity of practical utility, and is not intended to denote absolute specificity or to imply an an-tigen unique to the tumor.
~\~
~'~

Further it will be understood that cells other than tumor cells may have cell specific antigens characteristic, or strongly correlated with a given type cell. Given tissues may have tissue-specific antigens which are characteristic, or predominantly associated with a given tissue~ Cell specific and tissue-specific antigens are also useful for producing RMAs capable of binding preferentially, if not exclusively, to desired cells and tissues.
The following examples illustrate the techniques applied to the production of a quadroma producing an RMA. The described techniques can be applied, essentially as described, to prepare quadromas and produce recombinant monoclonal antibodies capable of binding any desired pair of antigens.
While most RMAs will have binding affini-ties for ~wo different antigens, it will be understood that RMAs binding two different epitopes on the same an-tigen could be prepared from appropriately selected quadroma or trioma clones. The significant variations in procedure for preparing other RM~s will lie in the nature of the antigen used to immunize, the screening test used to detect antibody production by the relevant hybridomas, triomas and quadromas and in the purification methods emplo~ed. The screening assay is especially significant, since it is at this stage that one selects for desired antibody properties in addition to binding affinity, e.g. r whether the antibody precipitates the antigen, binds complement, cross-reacts with other antigens, and the liker variations in technique of the type known in the art and understood by those of ordinary skill to be functional equivalents of those disclosed herein may be substituted as desired, for convenience or for optimization of yield, or to simplify or improve the cost-effectiveness of the overall procedure.

The following antigens are prepared in order to produce recombinant monoclonal antibody having dual 3~

binding affinity for two fluorescent haptens, fluorescein and rhodamine: fluorescein isothiocyanate-conjugated bovine serum albumin (F-~SA), fluorescein isothiocyanate labeled ovalbumin (F-OV~), rhodamine isothiocyanate-conjugated bovine serum albumin (R-BSA), and rhodamine iso-thiocyanate~conjugated ova~bumin (R-OVA). Fluorescein and rhodamine are chosen as haptens because they are readily assayed by fluorescence and they may be assayed in the presence of one another beca~se their excitatlon and emission maxima are substantially different from one another. The use of the same hapten coupled to two different proteins makes it possible to distinguish between antibodies directed against the hapten and antibodies directed against the protein to which it is conjugated. For example, where F-BSA is used for immunization, screening is carried out with F-O~. Only antibodies with binding affinity for the fluorescein moiety are detected in the screening assay. The isocyanate derivatives of fluorescein and rhodamine are commercially available, for example, from Sigma Chemical Co., St. Louis, Missouri.
To carry out the coupling reaction, 50 mg protein in 10 ml of 0.lM NaHCO3, pH 9, axe mixed with 5 mg of the desired isothiocyanate derivative and incubated for 30 minutes at room temperature with continuous gentle stirring. The product, after filtration through glass wool to remove precipitated protein and insoluble unreacted isothiocyanate, is chromatographed on Sephade~
G-25 (trademark, Phamacia, Inc., Uppsala, Sweden) in phosphate buffered saline (10 mM Na-phosphate p~ 7.~, 0.15M NaCl) to separate the derivatized protein from the unreacted product and to change the buffer system. The peak of derivatized protein is identified visuallyl and elutes in the volume of bufEer equivalent to the void volume of the column. The derivati~ed pro-tein is used without further purification for immunization and testing.

~;r~
i 7~

EXAMPI,E 2 Immunization. Immunization ln vivo is carried out using a method based on that of Vaitukaitis, JO/ e-t al, J. Clin. Endocrin. 33, 988 11971~. Antigen, 100 ng, in an emulsion of complete Freund's adjuvant and physiological saline in equal volumes is injected in-tradermally in 20 sites. After one week a second injection of the same an-tigen preparation is introduced into the granulomas resulting from the first injection.
Two weeks later, 100 ng of antigen is injected using incomplete Freund's adjuvant:saline (1:1~ subcutaneously in four sites over the shoulders and hips. One week later, a sample of blood is obtained from the tail and assayed for antibodies. The animal is now boosted in-travenously with 1 ug of antigen per injection for Eour days in a row. This treatment maximizes the number of lymphoblast cells present in the spleen, so that the frequency of antiyen-specific hybridomas formed after the fusion step is increased.
The procedure for immunization in vitro is based upon a technique described by Luben, R~A., et al, Proc.
Second Int. Lymphokine Worksho~, Academic Press, New York, N.Y. (1979)o The spleen of a non-immunized adult BALB/c mouse is removed by sterile technique and a single~cell suspension of spleen cells is prepared. The cells are diluted to 20 ml with complete Dulbecco's modified Eagle's medium (hexeinafter DMEM, commercially available from Grand Island Biological Company, Grand Island, New York), contair~ing 30 ,ug to 1,000 ,ug of antigen and 10 ml of thymocyte-conditioned medium is added.
Thymocyte-conditioned medium is prepared from the thymocytes of three 10-day~old mice or from a mixed thymocyte culture from adult mice. Thymocytes from BALB~c mice and those from a strain differing at the major histo-compatibility locus (e.g. C57 ~lack) are co-cultivated at 2 to ~ x 10 thymocytes/ml in complete DMEM. After 48 hours incubation at 37~C, the cells and .~

/~
debris are centrifuged and the medium is aspi.rated and stored frozen in 10 ml aliquots at -70C.
The mixture of non-immune spleen cells~ antigen and thymocyte conditioned medium is placed in a 75 cm2 flask 5 and lef-t untouched in a tissue culture incubator for five days at 37C. After five days, successful immunizat.ion yields numexous large lymphoblasts observable by phase contrast microscopy. The cells are then ready for fusion.

Lvm~hocvte-myeloma fusion and isolation of ~y~_idomas. A myeloma strain, designated SP2, described . . .
by Shulman, M. et al, Nature 276, 269 (1978) is chosen for fusion. The SP2 cell line is characterized as a non-producer of myeloma protein and is 8-azaguanine resistant, due to defective HPRT activity. The SP2 cell line has been widely dessimated, and may be obtained, for example, from Professor Klinman at Scripps Clinic and Research Foundation, La Jolla, California.
The fusion medium contains polyethylene glycol, 1,540 M~W. at 47~ (v/v~ and dimethyl sulfoxide at 7O5~
(v/v~ in serum-free DMEM. Polyethylene glycol induces cell fusion, as described by Pontecorvo, G., Somatic Cell rJenet., Vol. 1, 397 (1975). Dimethyl sulfo~ide reportedly enhances fusion frequency, possibly by loweri.ng the membrane phase transition temperature, as described by Norwood, T.H., et al, Somatic Cell Ge.net., Vol. ~, 263 (197~).
For spleen cells in~lunized ln vivo, a single cell suspension is made from a hyperimmune spleen as described for the immunization in culture. SP2 myeloma cells in exponential growth phase (30 ml, 5-8 x 105 cells/ml) are transferred to a 50 ml conlcal polypropylene centrifuge tu~e and the spleen cell suspension (5 ml) is added. For spleen cells immuni2ed in culture, the cells are harvested after dislodging adherent lymphoblast cells, centrifuged, and the medium removed. The SP2 cells (30 ml) are added as above.
With either preparation, the sells are washed three ..~

times with 50 ml of serum-free DMEM by centrifugation.
The pellet from -~he third wash is resuspended in 1 ml of fusion medium just removed from a 37C waterbath. The medium is added over one minute and the cells con-tinuously stirred with the pipet tip. Stirring is continued for another minute~ Two ml of serum-free DMEM
at 37C are added over the next three minutes with continuous stirring. Seven ml of 37C DMEM containing 10~ rabbit serum is added over the next three minutes with stirringO The cells are centrifuged and resuspended in 10 ml complete medium containing HAT
selection chemicals and feeder cells and distribu-ted into 96 wells of a microtitre plate.
Feeder cells are peritoneal exudate cells obtained after intraperitoneal injection of 0.5 ml pristane (2, 6, 10, lA-tetramethyl-pentadecane). After four days, cells are collected by washing out the peritoneal cavity of the treated mice. The yield is consistently 2-4 x 107 cells per mouse.
Antibody producing cells are directly cloned using the fluorescence-activated cell sorter. Positive cells will bind the fluorescent probe of the instrument and be separated from negative cells. The proba is obtained from fluorescent hapten coupled to a different protein from that used in the immunization. For example, if F-BSA or R-BSA is used for immunization/ F-OVA or R OVA
will be used as a probe, to avoid selection of hybridomas producing antibody against the protein.
An alternative screening procedure, suitable for non-fluorescent antigens, is based on enzyme-linked immunosorbant assay (Saunders, C~Go Immunoassays in the Clinical Laboratory, pp. 99-118 ~1979~)u To detect antibodies to soluble antigen, 50 yl/well of 10-100 ~g/ml antigen in water are added to polystyrene 96 well plates and they are allowed to dry in a 37C incubator. Immediately before use, the plates are washed three times with 10 ~M Na2HPO~ in 150 mM NaCl (PSB-9). To screen antibodies for reactivi-ty with cell surface components the cells are bound using an '`'"~';
_ 8~3 immobilized lectin. Concanavalin A is covalently ~ound to the polystyrene wells using a water-soluble carbodimide (Reading, C~L., et al JD Natl. Cancer Ins-t.
64, 1241 (1980)). The plates are washed six times with PBS-9 and cells are added to each well ~1-2 x 10 ) in 100 ~1 complete DMEM. The plates are kept at 37C for one to two hours to allow the cells to attach; after that the plates are washed six times with PBS-9 and 50 ~ul of fresh 1~ formaldehyde in PBS-9 is added to each well. The plates are kept for 15 minutes at room temperature and then washed six times with PsS-9 and used immedlately.
From each hybridoma culture 50 ~1 of medium is transferred to the antigen containing wells. The samples are incubated at room temperature for 30 minutes and the plates are washed 10 times with 0.05%
Triton-X-100 (trademark, Rohm & Haas Company, Nutley, New Jersey) in water. Enzyme-conjugated anti-mouse immunoglobulin (Cappel Laboratories, Cochranville, Pennsylvania) is diluted into 10 mM Na2HPO~, 0.05 M
NaC1, 0.5% (v/v) Triton-X-100 containing 50 ,ug/ml bovine serum albumin.
The conjuga-te (50 ~1) is added to each well and incubated for 15 minutes at room temperature. The plates are washed 10 times with 0.5% (v/v) Triton-X-100, 100 ~ll of substrate is added. The chromogenic subs-trate

2,2'-azino-di-(3-ethyl)-benzthiazoline sulfonic acid (ABTS) is used as described by Saunders, su~ra. The colored enzyme product is quantitated by measuring the optical density at 414 mM.
Cells from cultures producing the desired antibody are counted and diluted to yield 30-50 viable hibridoma cells/ml of complete HT medium (DMEM containing 10 M
hypoxanthine, and 3 x 10 m thymidine~
A porkion of Ool ml of the suspension is pipetted into each well of a 96 well microtitre plate containing 1.2 x 105 peritoneal exudate feeder cel]s. Each well contains on the average 3-5 hybridoma cells per well.

~lQ
The cultures are incubated in a tissue culture incubator at 37C for seven days~ following which 0.1 ml complete HT medium is added to each ~ell. After an additional 14-21 days' incubation, the clones are dense and ready for screening, either by the ELISA procedure or by measurement of fluorescent quenching due to antibody binding of added fluorescent hapten. For specificity controls, antibodies reactive with fluorescein should not bind rhodamine, and vice versa.
The six strongest positive cultures are -transferred to larger wells, and are re~assayed after incubation to allow the cultures to again become dense. A portion of the cells ~rom the strongest two cultures are re-cloned by limiting dilution, ,v0.3 cells per well (using a feeder layer). The remainder of the cells in the two strongest positive cultures are incubated in additional medium to increase their numbers and stored frozen.
When the limiting dilution clones have reached adequate cell density, khe wells with a single clone present are assayed. Six positive clones are transferred to larger wells, again incubated to increase their numbers, and stored frozen. The two strongest wells are examined for stability by another round of limiting dilution cloning. ?he FACS is useful in these selection and re-c]oning steps, in the manner previously described. Since these processes are labor-intensive, the use of the cell sorter at any stage where applicable is advantageous. Clones which yield greater than 90%
positive clones are considered stable. Clones which yield less than 90% positive clones are re~cloned until stability is achieved.

Quadroma formation. The first step in quadroma .... ~
formation is the selection of mutant hybridoma strains suitable for preferentially growing qaudroma fusion products in the presence of the parent hybridomas. In -this example, the hybridoma strain producing antibody against fluorescein is further modified to 8-azaguanine ~7i ~/

and ouabain resistance. The modified hybridoma is used as a universal fuser, as described, ~
Selection for 8-azaguanine reslstance involves a process of adaptive growth in gradually increasing concentra-tions of the inhibitor~ beg.inning with inhibitor concentrations of about 1 ~uMo Cells grown for several generations are then transferred to a 3 ,uM
8-azaguanine for an additional period of growth for several generations. The process is reiterated, with progressive increments of inhibitor, until a viable strain growing in the presence of 100 ~u~ 8-azaguanine is obtained. The procedure selects mutants arisi.ng spontaneously or by 8-azaguanine induced mutation, which lack functional HPRT activity. The 8~azaguanine resistant hybridoma strain is then made resis-tant to ouabain inhibi-tion by a. similar process of adaptive growth, using essentially the method described by Baker, RoM~ ~ et al, Cell 1, 9 (1979).
Equal numbers of anti-~luorescein producing double mutant hybridomas~ prepared as described, and anti-rhodamine producing hybridomas are fused, following essentially the procedure of Example 3. The yield of : quadromas producing antibodies agai.nst both antigens is higher, per stable fusion, than for conventional fusions, since every parental cell is of the desired type. After the fusion step is complete and the cells dispensed in microtitre plate wells, they are incubated in the presence of HAT medium tDMEM con-taining 3 x 10 M thymidine, 4 x 10 7 M aminopterin and 3 x 10 M
hypoxanthine) containing 10 3 M ouabain. As previously described, both parental hybridoma strains are killed by growth .in HAT-ouabain medium, while quadromas which have retained functional ~IPRT and the ouabain resistance mutation survive and growO
After selection, quadromas which bind both antigens simultaneously are cloned in individual microtitre wells - using the single-cell deposition attachment for the fluorescence-activated cell sorter. The single cells will develop into dense cultures within 10-14 days.
~`"~;1 ... ~..~

Alternatively, quadromas are detected and cloned by plating in soft agar medium. After 13 14 days' growth, -the clones which appear are tested 1n situ by the solid phase assay described by Sharon e-t al~ supra.
Replicate tests are required, first with one antigen, then with the other. Clones which react with both antigens contain the desired quadroma~
Alternatively, screening may be carried out by allowing quadromas to bind to a surface coated with one antigen, the testing for ability to bind with the other antigen As previously described for hybridomas, the most active and stable clones are re-cloned to ensure stability. Clones which yield greater than 90% positive clones are considered stable, while those yieldiny less than 90% are re-cloned until stability is achieved.
Quadroma clones producing presumptive RMAs are those which produce antibody binding both of the immunizing antigens, fluorescein and rhodamine.

Pre~aration and pur~fication of recomblnant monoclonal antibody RMAs are isolated, either from the .

supernatant of quadroma cultures or from ascites from a mouse injected with quadroma cells interperitoneally.
In the latter case, BALB/c mlce are preteated with Z5 interperitoneal injection of 0.5 ml pristane. An injection of 1-2 x 10 quadroma cells of a stable clone are injected intraperitoneally. Ascites -tumors are evident by day 10 to 21, and -the ascites fluid is collected when the periotoneal cavity hecomes distended.
Cells are re~oved by centrifugation and antibody is precipitated with 60% saturated ammonium sulEa-te. l'he antibody is then dialyzed and frozen. The yield is usually about 30-50 mg of antibody per mouse~
Recombinant monoclonal antibodies are further purified from the antibody preparation by two stages of affinity chromatography. In the first column F-BSA is coupled to CNBr activated Sepharose 4B Itrademark, Pharmacia Fine Chemicals AB, Uppsala, Sweden) using standard coupling procedures as described by March~

S.C., et al, Anal._Biochem. 60, 1~9 (lg74). The second column is packed with R~BSA coupled to CNBr-activated Sepharose 4s. The columns are equilibrated with PBS-9 and the antigen preparation is applied to the first column and eluted with 2-3 column volumes o~ PBS-9. The first column is then eluted wlth PBS-9 containing 3 M
potassium isothiocyanate. Eluted protein is dialyzed against PBS-9 and applied to the second column, which is eluted in the same manner as the first. Protein recovered ~rom the second column after potassium isothiocyanate elution is recomb;nant monoclonal antibody, which has two distinct combining sites per molecule, one for fluorescein and one for rhodamine.
The RMA preparation is dialyzed, concentrated and stored frozen.
The foregoing specification describes the formation of novel cell types, quadromas and triomas, capable of producing re~ombinant monoclonal antibodies, a heretofore unknown molecular species of antibody having binding aEfinities for two different antigens and capable of binding both antigens simultaneously. The techniques for producing such new materials have been described in detail~ particularly wi-th reference to specific embodiments included by way of example. It will be understood that the products and techniques of the present invention are of far-reaching significance and include a wide range of RMA types combining any pair of antigenic specificities on a single antibodyO I~
will be further understood that many variations and modifications of the -techniques employed herein are available to those of ordinary skill in the relevant art, and that such variations and modifications are contemplated as being within the scope of the invention.

.~

Claims

THE EMBODIMENTS OF THE INVENTION IN WHICH AN EXCLUSIVE
PROPERTY OR PRIVILEGE IS CLAIMED ARE DEFINED AS FOLLOWS:

1. A method of producing a recombinant monoclonal antibody comprising incubating a quadroma cell in culture or in the peritoneal cavity of a mouse, and separating soluble protein from the culture supernatant or ascites fluid, respectively.

2. The method of claim 1, further comprising the step of passing the soluble protein sequentially through two affinity chromatographic columns each bearing one of two antigens for which the recombinant monoclonal antibody has binding affinity, immobilized on the column and eluting recombinant monoclonal antibody from each column with a choatropic ion solution.

3. The method of claim 1, wherein the quadroma cell is formed by fusing two hybridoma cell lines under conditions permitting selection of the fusion product in the presence of parental hybridomas.

4. A method of producing a recombinant monoclonal antibody comprising incubating a trioma cell in culture or in the peritoneal cavity of a mouse, and separating soluble protein from the culture supernatant or ascites fluid, respectively.

5. The method of claim 4, further comprising the step of passing the soluble protein sequentially through two affinity chromatographic columns each bearing one of two antigens for which the recombinant monoclonal antibody has binding affinity, immobilized on the column and eluting recombinant monoclonal antibody from each column with a choatropic ion solution.

6. The method of claim 4, wherein the quadroma cell is formed by fusing two hybridoma cell lines under conditions permitting selection of the fusion product in the presence of parental hybridomas.

7. An antibody with binding affinity for two different desired antigens when prepared by the process of claim 1 or its obvious chemical equivalents.

8. The antibody of claim 7, wherein one of the two antigens is a tumor-specific antigen when prepared by the process of claim 1 or its obvious chemical equivalents.

9. The antibody of claim 7, wherein one of the two antigens is a cell-specific antigen when prepared by the process of claim 1 or its obvious chemical equivalents.

10. The antibody of claim 7, wherein one of the two antigens is a tissue-specific antigen when prepared by the process of claim 1 or its obvious chemical equivalents.

11. The antibody of claim 7, wherein one of the two antigens is an enzyme when prepared by the process of claim 1 or its obvious chemical equivalents.

12. The antibody of claim 7, wherein one of the two antigens is a tumor-specific antigen and the other of the two antigens is an enzyme when prepared by the process of claim 1 or its obvious chemical equivalents.

13. The antibody of claim 7, wherein one of the two antigens is a cell-specific antigen and the other of the two antigens is an enzyme when prepared by the process of claim 1 or its obvious chemical equivalents.

14. The antibody of claim 7, wherein one of the two antigens is an a tissue-specific antigen and the other of the two antigens is an enzyme when prepared by the process of claim 1 or its obvious chemical equivalents.

15. The antibody of claim 7, wherein one of the two antigens is a hapten when prepared by the process of claim 1 or its obvious chemical equivalents.

16. The antibody of claim 7, wherein one of the two antigens is a tumor-specific antigen and the other of the two antigens is a hapten when prepared by the process of claim 1 or its obvious chemical equivalents.

17. The antibody of claim 7, wherein one of the two antigens is a cell-specific antigen and the other of the two antigens is a hapten when prepared by the process of claim 1 or its obvious chemical equivalents.

18. The antibody of claim 7, wherein one of the two antigens is a tissue-specific antigen and the other of the two antigens is a hapten when prepared by the process of claim 1 or its obvious chemical equivalents.

19. The antibody of claim 7, wherein one of the two antigens is the toxic subunit of Ricinus communis toxin when prepared by the process of claim 1 or its obvious chemical equivalents.

20. The antibody of claim 7, wherein one of the two antigens is a tumor-specific antigen and the other of the two antigens is the toxic subunits of Ricinus communis toxin when prepared by the process of claim 1 or its obvious chemical equivalents.

21. The antibody of claim 7, wherein one of the two antigens is specific for trinitrophenol when prepared by the process of claim 1 or its obvious chemical equivalents.

22. The antibody of claim 7, wherein one of the two antigens is a tumor-specific antigen and the other of the two antigens is specific for trinitrophenol when prepared by the process of claim 1 or its obvious chemical equivalents.

23. The antibody of claim 7, wherein one of the two antigens is a hapten which is radioactively-labeled when prepared by the process of claim 1 or its obvious chemical equivalents.

24. The antibody of claim 7, wherein one of the two antigens is a tumor-specific antigen and the other of the two antigens is a hapten which is radioactively-labeled when prepared by the process of claim 1 or its obvious chemical equivalents.

25. The antibody of claim 7, wherein one of the two antigens is a cell-specific antigen and the other of the two antigens is a hapten which is radioactively-labeled when prepared by the process of claim 1 or its obvious chemical equivalents.

26. The antibody of claim 7, wherein one of the two antigens is a tissue-specific antigen and the other of the two antigens is a hapten which is radioactively-labeled when prepared by the process of claim 1 or its obvious chemical equivalents.

27. The antibody of claim 7, wherein one of the two antigens is the B subunit of human chorionic gonadotropin and the other antigen is a hapten when prepared by the process of claim 1 or its obvious chemical equivalents.

28. The antibody of claim 7, wherein one of the two antigens is the B subunit of human chorionic gonadotropin and the other antigen is horseradish peroxidase when prepared by the process of claim 1 or its obvious chemical equivalents.

29. The antibody of claim 7, wherein one of the two antigens is fluorescein or rhodamine when prepared by the process of claim 1 or its obvious chemical equivalents.

30. The antibody of claim 7, wherein one of the two antigens is a tumor-specific antigen and the other of the two antigens is fluorescein or rhodamine when prepared by the process of claim 1 or its obvious chemical equivalents.

31. The antibody of claim 7, wherein one of the two antigens is fluorescein and the other antigen is rhodamine when prepared by the process of claim 1 or its obvious chemical equivalents.

32. An antibody with binding affinity for two different desired antigens when prepared by the process of claim 4 or its obvious chemical equivalents.

33. The antibody of claim 32, wherein one of the two antigens is a tumor specific antigen when prepared by the process of claim 4 or its obvious chemical equivalents.

34. The antibody of claim 32, wherein one of the two antigens is a cell-specific antigen when prepared by the process of claim 4 or its obvious chemical equivalents.

35. The antibody of claim 32, wherein one of the two antigens is a tissue-specific antigen when prepared by the process of claim 4 or its obvious chemical equivalents.

36. The antibody of claim 32, wherein one of the two antigens is an enzyme when prepared by the process of claim 4 or its obvious chemical equivalents.

37. The antibody of claim 32, wherein one of the two antigens is a tumor-specific antigen and the other of the two antigens is an enzyme when prepared by the process of claim 1 or its obvious chemical equivalents.

38. The antobidy of claim 32, wherein one of the two antigens is a cell-specific antigen and the other of the two antigens is an enzyme when prepared by the process of claim 4 or its obvious chemical equivalents.

39. The antibody of claim 32, wherein one of the two antigens is an a tissue-specific antigen and the other of the two antigens is an enzyme when prepared by the process of claim 4 or its obvious chemical equivalents.

40. The antibody of claim 32, wherein one of the two antigens is a hapten when prepared by the process of claim 4 or its obvious chemical equivalents.

41. The antibody of claim 32, wherein one of the two antigens is a tumor-specific antigen and the other of the two antigens is a hapten when prepared by the process of claim 4 or its obvious chemical equivalents.

42. The antibody of claim 32, wherein one of the two antigens is a cell-specific antigen and the other of the two antigens is a hapten when prepared by the process of claim 4 or its obvious chemical equivalents.

43. The antibody of claim 32, wherein one of the two antigens is a tissue-specific antigen and the other of the two antigens is a hapten when prepared by the process of claim 4 or its obvious chemical equivalents.

44. The antibody of claim 32, wherein one of the two antigens is the toxic subunit of Ricinus communis toxin when prepared by the process of claim 4 or its obvious chemical equivalents.

45. The antibody of claim 32, wherein one of the two antigens is a tumor-specific antigen and the other of the two antigens is the toxic subunits of Ricinus communis toxin when prepared by the process of claim 4 or its obvious chemical equivalents.

46. The antibody of claim 32, wherein one of the two antigens is specific for trinitrophenol when prepared by the process of claim 4 or its obvious chemical equivalents.

47. The antibody of claim 32, wherein one of the two antigens is a tumor-specific antigen and the other of the two antigens is specific for trinitrophenol when prepared by the process of claim 4 or its obvious chemical equivalents.

48. The antibody of claim 32 , wherein one of the two antigens is a hapten which is radioactively-labeled when prepared by the process of claim 4 or its obvious chemical equivalents.

49. The antibody of claim 32, wherein one of the two antigens is a tumor-specific antigen and the other of the two antigens is a hapten which is radioactively-labeled when prepared by the process of claim 4 or its obvious chemical equivalents.

50. The antibody of claim 32, wherein one of the two antigens is a cell-specific antigen and the other of the two antigens is a hapten which is radioactively-labeled when prepared by the process of claim 4 or its obvious chemical equivalents.

51. The antibody of claim 32, wherein one of the two antigens is a tissue-specific antigen and the other of the two antigens is a hapten which is radioactively-labeled when prepared by the process of claim 4 or its obvious chemical equivalents.

52. The antibody of claim 32, wherein one of the two antigens is the B subunit of human chorionic gonadotropin and the other antigen is a hapten when prepared by the process of claim 4 or its obvious chemical equivalents.

53. The antibody of claim 32, wherein one of the two antigens is the B subunit of human chorionic gonadotropin and the other antigen is horseradish peroxidase when prepared by the process of claim 4 or its obvious chemical equivalents.

54. The antibody of claim 32, wherein one of the two antigens is fluorescein or rhodamine when prepared by the process of claim 4 or its obvious chemical equivalents.

55. The antibody of claim 32, wherein one of the two antigens is a tumor-specific antigen and the other of the two antigens is fluorescein or rhodamine when prepared by the process of claim 4 or its obvious chemical equivalents.

56. The antibody of claim 32, wherein one of the two antigens is fluorescein and the other antigen is rhodamine when prepared by the process of claim 4 or its obvious chemical equivalents.

57. A method of producing an antibody by a quadroma cell, said antibody having specific binding affinity for two desired antigens and wherein said quadroma cell comprises the fusion product of:
(a) a hybridoma cell which produces an antibody having specific binding affinity to one desired antigen, and (b) a hydridoma cell which produces an antibody having specific binding affinity for another desired antigen.

58. A method of producing an antibody by a trioma cell, said antibody having specific binding affinity for two desired antigens and wherein said trioma cell comprises the fusion product of:
(a) a hybridoma cell which produces an antibody having specific binding affinity to one desired antigen, and (b) a lymphocyte which produces an antibody having specific binding affinity for another desired antigen.

59. An antibody with binding affinity for two different desired antigens when prepared by the process of claim 57 or its obvious chemical equivalents.

60. The antibody of claim 59, wherein one of the desired antigens is a tumor-specific antigen when prepared by the process of claim 57 or its obvious chemical equivalents.

61. The antibody of claim 59, wherein one of the desired antigens is a cell-specific antigen when prepared by the process of claim 57 or its obvious chemical equivalents.

62. The antibody of claim 59, wherein one of the desired antigens is a tissue-specific antigen when prepared by the process of claim 57 or its obvious chemical equivalents.

63. The antibody of claim 59, wherein one of the desired antigens is an enzyme when prepared by the process of claim 57 or its obvious chemical equivalents.

64. The antibody of claim 59, wherein one of the two antigens is a tumor-specific antigen and the other of the two antigens is an enzyme, when prepared by the process of claim 57 or its obvious chemical equivalents.

65. The antibody of claim 59, wherein one of the two antigens is a cell-specific antigen and the other of the two antigens is an enzyme when prepared by the process of claim 57 or its obvious chemical equivalents.

66. The antibody of claim 59, wherein one of the desired antigens is a tissue-specific antigen and the other of the two antigens is an enzyme when prepared by the process of claim 57 or its obvious chemical equivalents.

67. The antibody of claim 59, wherein one of the desired antigens is a hapten when prepared by the process of claim 57 or its obvious chemical equivalents.

68. The antibody of claim 59, wherein one of the two antigens is a tumor-specific antigen and the other of the two antigens is a hapten, when prepared by the process of claim 57 or its obvious chemical equivalents.

69. The antibody of claim 59, wherein one of the two antigens is a cell-specific antigen and the other of the two antigens is a hapten when prepared by the process of claim 57 or its obvious chemical equivalents.

70. The antibody of claim 59, wherein one of the desired antigens is a tissue-specific antigen and the other of the two antigens is a hapten when prepared by the process of claim 57 or its obvious chemical equivalents.

71. The antibody of claim 59, wherein one of the desired antigens is a toxic subunit of Ricinus communis toxin when prepared by the process of claim 57 or its obvious chemical equivalents.

72. The antibody of claim 59, wherein one of the two antigens is a tumor-specific antigen and the other of the two antigens is a toxic subunit of Ricinus communis toxin when prepared by the process of claim 57 or its obvious chemical equivalents.

73. The antibody of claim 59, wherein one of the desired antigens is specific for trinitriphenol when prepared by the process of claim 57 or its obvious chemical equivalents.

74. The antibody of claim 59, wherein one of the two antigens is a tumor-specific antigen and the other of the two antigens is specific for trinitriphenol when prepared by the process of claim 57 or its obvious chemical equivalents.

75. The antibody of claim 59, wherein one of the desired antigens is a hapten which is radioactively-labeled when prepared by the process of claim 57 or its obvious chemical equivalents.

76. The antibody of claim 59, wherein one of the two antigens is a tumor-specific antigen and the other of the two antigens is a hapten which is radioactively-labeled when prepared by the process of claim 57 or its obvious chemical equivalents.

77. The antibody of claim 59, wherein one of the desired antigens is a hapten which is radioactively-labeled and the other of the two antigens is an enzyme when prepared by the process of claim 57 or its obvious chemical equivalents.

78. The antibody of claim 59, wherein one of the desired antigens is the B subunit of human chorionic gonadotropin and the other antigen is a hapten when prepared by the process of claim 57 or its obvious chemical equivalents.

79. The antibody of claim 59, wherein one of the desired antigens is the B subunit of human chorionic gonadotropin and the other antigen is a horseradish peroxidase when prepared by the process of claim 57 or its obvious chemical equivalents.

80. The antibody of claim 59, wherein one of the two desired antigens is fluorescein or rhodamine when prepared by the process of claim 57 or its obvious chemical equivalents.

81. The antibody of claim 59, wherein one of the two desired antigens is fluorescein and the other antigen is rhodamine when prepared by the process of claim 57 or its obvious chemical equivalents.

82. An antibody with binding affinity for two different desired antigens when prepared by the process of claim 58 or its obvious chemical equivalents.

83. The antibody of claim 82, wherein one of the desired antigens is a tumor-specific antigen when prepared by the process of claim 58 or its obvious chemical equivalents.

84. The antibody of claim 82, wherein one of the desired antigens is a cell-specific antigen when prepared by the process of claim 59 or its obvious chemical equivalents.

85. The antibody of claim 82, wherein one of the desired antigens is a tissue-specific antigen when prepared by the process of claim 58 or its obvious chemical equivalents.

86. The antibody of claim 82, wherein one of the desired antigens is an enzyme when prepared by the process of claim 58 or its obvious chemical equivalents.

87. The antibody of claim 82, wherein one of the two antigens is a tumor-specific antigen and the other of the two antigens is an enzyme, when prepared by the process of claim 58 or its obvious chemical equivalents.

88. The antibody of claim 82, wherein one of the two antigens is a cell-specific antigen and the other of the two antigens is an enzyme when prepared by the process of claim 58 or its obvious chemical equivalents.

89. The antibody of claim 82, wherein one of the desired antigens is a tissue-specific antigen and the other of the two antigens is an enzyme when prepared by the process of claim 58 or its obvious chemical equivalents.

90. The antibody of claim 82, wherein one of the desired antigens is a hapten when prepared by the process of claim 58 or its obvious chemical equivalents.

91. The antibody of claim 82, wherein one of the two antigens is a tumor-specific antigen and the other of the two antigens is a hapten, when prepared by the process of claim 58 or its obvious chemical equivalents.

92. The antibody of claim 82, wherein one of the two antigens is a cell-specific antigen and the other of the two antigens is a hapten when prepared by the process of claim 58 or its obvious chemical equivalents.

93. The antibody of claim 82, wherein one of the desired antigens is a tissue-specific antigen and the other of the two antigens is a hapten when prepared by the process of claim 58 or its obvious chemical equivalents.

94. The antibody of claim 82, wherein one of the desired antigens is a toxic subunit of Ricinus communis toxin when prepared by the process of claim 58 or its obvious chemical equivalents.

95. The antibody of claim 82, wherein one of the two antigens is a tumor-specific antigen and the other of the two antigens is a toxic subunit of Ricinus communis toxin when prepared by the process of claim 58 or its obvious chemical equivalents.

96. The antibody of claim 82, wherein one of the desired antigens is specific for trinitriphenol when prepared by the process of claim 58 or its obvious chemical equivalents.

97. The antibody of claim 82, wherein one of the two antigens is a tumor-specific antigen and the other of the two antigens is specific for trinitriphenol when prepared by the process of claim 58 or its obvious chemical equivalents.

98. The antibody of claim 82, wherein one of the desired antigens is a hapten which is radioactively-labeled when prepared by the process of claim 58 or its obvious chemical equivalents.

99. The antibody of claim 82, wherein one of the two antigens is a tumor-specific antigen and the other of the two antigens is a hapten which is radioactively-labeled when prepared by the process of claim 58 or its obvious chemical equivalents.

100. The antibody of claim 82, wherein one of the desired antigens is a hapten which is radioactively-labeled and the other of the two antigens is an enzyme when prepared by the process of claim 58 or its obvious chemical equivalents.

101. The antibody of claim 82, wherein one of the desired antigens is the B subunit of human chorionic gonadotropin and the other antigen is a hapten when prepared by the process of claim 58 or its obvious chemical equivalents.

102. The antibody of claim 82, wherein one of the desired antigens is the B subunit of human chorionic gonadotropin and the other antigen is a horseradish peroxidase when prepared by the process of claim 58 or its obvious chemical equivalents.

103. The antibody of claim 82, wherein one of the two desired antigens is fluorescein or rhodamine when prepared by the process of claim 58 or its obvious chemical equivalents.

104. The antibody of claim 82, wherein one of the two desired antigens is fluorescein and the other antigen is rhodamine when prepared by the process of claim 58 or its obvious chemical equivalents.

105. The antibody of claim 7, wherein said antibody is an IgM molecule when prepared by the process of claim 1 or claim 4 or its obvious chemical equivalents.

106. The antibody of claim 59, wherein said antibody is an IgM molecule when prepared by the process of claim 57 or claim 58 or its obvious chemical equivalents.

107. The antibody of claim 7, wherein said antibody is an IgG molecule when prepared by the process of claim 1 or claim 4 or its obvious chemical equivalents.

108. The antibody of claim 59, wherein said antibody is an IgG molecule when prepared by the process of claim 57 or claim 58 or its obvious chemical equivalents.

109. A quadroma cell, wherein said quadroma cell is the fusion product of a hybridoma cell which produces an antibody having specific binding affinity to one desired antigen and a hybridoma cell which produces an antibody having specific binding affinity for another desired antigen, and wherein said quadroma cell produces an antibody having binding affinity for two desired antigens.

110. The quadroma cell of claim 109, wherein one of the two antigens is a tumor-specific antigen.

111. The quadroma cell of claim 109, wherein one of the two antigens is a cell-specific antigen.

112. The quadroma cell of claim 109, wherein one of the two antigens is a tissue-specific antigen.

113. The quadroma cell of claim 109, 110, or 111, wherein one of the two antigens is an enzyme.

114. The quadroma cell of claim 112, wherein one of the two antigens is an enzyme.

115. The quadroma cell of claim 109, 110, or 111, wherein one of the two antigens is a hapten.

116. The quadroma cell of claim 112, wherein one of the two antigens is a hapten.

117. The quadroma cell of claim 109 or 110, wherein one of the two antigens is a toxic subunit of Ricinus communis toxin.

118. The quadroma cell of claim 109 or 110, wherein one of the two antigens is specific for trinitrophenol.

119. The quadroma cell of claim 109, 110, or 111, wherein one of the two antigens is a radioactively-labeled hapten.

120. The quadroma cell of claim 112, wherein one of the two antigens is a radioactively-labeled hapten.

121. The quadroma cell of claim 109, wherein one of the two antigens is the B subunit of human chorionic gonadotropin and the other antigen is a hapten.

122. The quadroma cell of claim 109, wherein one of the two antigens is the B subunit of human chorionic gonadotropin and the other antigen is horseradish peroxidase.

123. The quadroma cell of claim 109 or 110, wherein one of the two antigens is fluorescein or rhodamine.

124. The quadroma cell of claim 109, wherein one of the two antigens is fluorescein and the other antigen is rhodamine.

125. The quadroma cell of claim 109, wherein said antibody is an IgM molecule.

126. The quadroma cell of claim 109, wherein said antibody is an IgG molecule.

127. A trioma cell, wherein said trioma cell is the fusion product of a hybridoma cell which produces an antibody having specific binding affinity to one desired antigen and a lymphocyte which produces an antibody having specific binding affinity to another desired antigen, and wherein said trioma cell produces an antibody having binding affinity for two desired antigens.

128. The trioma cell of claim 127, wherein one of the two antigens is a tumor-specific antigen.

129. The trioma cell of claim 127, wherein one of two antibodies is a cell-specific antigen.

130. The trioma cell of claim 127, wherein one of the two antigens is a tissue-specific antigen.

131. The trioma cell of claim 127, 128, or 129, wherein one of the two antigens is an enzyme.

132. The trioma cell of claim 130, wherein one of the two antigens is a hapten.

133. The trioma cell of claim 127, 128, or 129, wherein one of the two antigens is a hapten.

134. The trioma cell of claim 130, wherein one of the two antigens is a hapten.

135. The trioma cell of claim 127, or 128, wherein one of the two antigens is a toxic subunit of Ricinus communis toxin.

136. The trioma cell of claim 127 or 128, wherein one of the two antigens is specific for trinitrophenol.

137. The trioma cell of claim 127, 128 or 129, wherein one of the two antigens is a radioactively-labeled hapten.

138. The trioma cell of claim 130, hwerein one of the two antigens is a radioactively-labeled hapten.

139. The trioma cell of claim 127, wherein one of the two antigens is the B subunit of human chorionic gonadotropin and the other antigen is a hapten.

140. The trioma cell of claim 127, wherein one of the two antigens is the B subunit of human chorionic gonadotropin and the other antigen is horseradish peroxidase.

141. The trioma cell of claim 127, wherein one of the two antigens is fluorescein or rhodamine.

142. The trioma cell of claim 127, wherein one of the two antigens is fluorescein and the other antigen is rhodamine.

143. The trioma cell of claim 127, wherein said antibody is an IgM molecule.

144. The trioma cell of claim 127, wherein said antibody an an IgG molecule.

145. A method of producing a quadroma cell comprising:
fusing a hybridoma cell which produces an antibody having specific binding affinity to one desired antigen, to a hybridoma cell which produces an antibody having specifc binding affinity to another desired antigen, wherein said quadroma cell produces an antibody having binding affinity for two desired antigens.

146. The method of producing a quadroma cell of claim 145, wherein one of the two antigens is a tumor-specific antigen.

147. The method of producing a quadroma cell of claim 145, wherein one of the two antigens is a cell-specific antigen.

148. The method of producing a quadroma cell of claim 145, wherein one of the two antigens is a tissue-specific antigen.

149. The method of producing a quadroma cell of claim 145, 146 or 147, wherein one of the two antigens is an enzyme.

150. The method of producing a quadroma cell of claim 148 wherein one of the two antigens is an enzyme.

152. The method of producing a quadroma cell of claim 145, 146 or 147, wherein one of the two antigens is a hapten.

152. The method of procuding a quadroma cell of claim 148 wherein one of the two antigens is a hapten.

153. The method of producing a quadroma cell of claim 145 or 146 , wherein one of the two antigens is a toxic subunit of Ricinus communis toxin.

154. The method of producing a quadroma cell of claim 146 or 146, wherein one of the two antigens is specific for trinitrophenol.

155. The method of producing a quardoma cell of claim of claim 145, 146 or 147, wherein one of the two antigens is a radioactively-labeled hapten.

156. The method of producing a quardoma cell of claim of claim 148, wherein one of the two antigens is a radioactively-labeled hapten.

157. The method of producing a quadroma cell of claim 145, wherein one of the two antigens is the B subunit of human chorionic gonadotropin and the other antigen is a hapten.

158. The method of producing a quadroma cell of claim 145, wherein one of the two antigens is the B subunit of human chorionic gonadotropin and the other antigen is horseradish peroxidase.

159. The method of producing a quadroma cell of claim 95 or 96, wherein one of the two antigens is fluorescein or rhodamine.

160. The method of producing a quadroma cell of claim 145, wherein one of the two antigens is fluorescein and the other antigen is rhodamine.

161. The method of producing a quadroma cell of claim 145, wherein said antibody is an IgM molecule.

162. The method of producing a quadroma cell of claim 145, wherein said antibody is an IgG molecule.

163. A method of producing a trioma cell comprising:
fusing a hybridoma cell which produces an antibody having specific binding affinity to one desired antigen, to a lymphocyte which produces an antibody having specific binding affinity to another desired antigen, wherein said trioma cell produces an antibody having binding affinity for two desired antigens.

164. The method of producing a trioma cell of claim 163, wherein one of the two antigens is a tumor-specific antigen.

165. The method of producing a trioma cell of claim 163, wherein one of the two antigens is a cell-specific antigen.

166. The method of producing a trioma cell of claim 163, wherein one of the two antigens is a tissue-specific antigen.

167. The method of producing a trioma cell of claim 163, 164 or 165, wherein one of the two antigens is an enzyme.

168. The method of producing a trioma cell of claim 166 wherein one of the two antigens is an enzyme.

169. The method of producing a trioma cell of claim 163, 164 or 165, wherein one of the two antigens is a hapten.

170. The method of producing a trioma cell of claim 166, wherein one of the two antigens is a hapten.

171. The method of producing a trioma cell of claim 163 or 164, wherein one of the two antigens is a toxic subunit of Ricinus communis toxin.

172. The method of producing a trioma cell of claim 163 or 164, wherein one of the two antigens is specific for trinitrophenol.

173. The method of producing a trioma cell of claim 163, 164 or 165, wherein one of the two antigens is a radioactively-labeled hapten.

174. The method of producing a trioma cell of claim 166, wherein one of the two antigens is a radioactively-labeled hapten.

175. The method of producing a trioma cell of claim 163, wherein one of the two antigens is the B subunit of human chorionic gonadotropin and the other antigen is a hapten.

176. The method of producing a trioma cell of claim 163, wherein one of the two antigens is the B subunit of human chorionic gonadotropin and the other antigen is horseradish peroxidase.

177. The method of producing a trioma cell of claim 163, wherein one of the two antigens is fluorescein or rhodamine.

178. The method of producing a trioma cell of claim 163, wherein one of the two antigens is fluorescein and the other antigen is rhodamine.

179. The method of producing a trioma cell of claim 163, wherein said antibody is an IgM molecule.

180. The method of producing a trioma cell of claim 163, wherein said antibody is an IgG molecule.