CA1328419C - Receptors for efficient determination of ligands and their antagonists or agonists - Google Patents

Abstract

Abstract of the Disclosure Hybrid receptors are provided that comprise (a) the ligand binding domain of a predetermined receptor and (b) a heterologous reporter polypeptide. The hybrid receptors are useful for convenient and large scale assay of biologically active ligands or their antagonists or agonists.

Description

1328~19 hOVEL RECEPTORS POR EFFICIENT D~TERXINATION
OF LIGANDS AND TEEIR ANTAGOMISTS OR ACO~ISTS
.

lS This invention relates to methods for screening candidate drugs for their ability to bind n receptor in such a fashion as to mimic or sntagonize the function of ~ ligand which ordinarily interacts with the receptor in v~vo. It also relates to methods for the functional assay of ligands.
Receptors are defined as proteinaceous macromolecules located on cell membranes that perform a signal transducing function. Hany receptors are located on the outer cell membrane.
These cell ~urface receptors have extracellular and cytoplasmic domains wherein the extracellular tomsin is capable of epecifically binding a eubstance so that the cytoplasmic domain interacts with another cell molecule as a function of the binding of the substance by the extracellular domain. The substance which is bound by the receptor is called a ligand, a term which is definitionally meaningful only in terms of its counterpart receptor. The term "ligand" does not imply any particular molecular size or o~her structural or compositional feature other thsn that the substance in question is capable of binding, . ~

. -2- 1328~19 cleaving or otherwise interacting with the receptor in ~uch a way that the receptor conveys informat~on about the presence of the ligand to a target molecule. Stated alternatively, not all substances capable of binding a receptor are ligands, but all ligands are capable of binding e receptor. Receptors do not include such substances as immunoglobulins.

Receptors typically are divided 6tructurally into three domains. A highly hydrophobic region of about 20 to 25 residues which i8 believed to be responsible for embedding the receptor in the cell membrane is flanked on lts amino and carboxyl termini by regions that respectively extend into the extracellular and cytoplasmic environment. The extracellular region includes the ligand binding domain. The cytoplasmic region ~ncludes a domain for effecting a change in the cytoplasm. Typically, the cyto-plasmic region includes an enzymatic function that is activated by receptor aggregation or conformational changes brought on by ligand binding.

2~ Receptors are believed to function by a process variously termed activation or signal transduction. A ligand binds to the extracellular ligand binding domain in such a way that the conformation of the receptor molecule changes within the cytoplasmic region. This conformational change, called activation, modifies the effect of the receptor on cytoplasmic components. Among changes brought about by receptor activation are changes in or development of receptor enzymatic activity.

The pharmaceutical industry in recent years has oriented its research to focus on the role of receptors in disease or in~ury and to tesign drugs, generally low molecular weight ~ substances, that are capable of binding to the receptors. Drugs - identified in this initial screen are then tested for the desired activity ~n vivo or in tissue explants. As a result, conventional - L03x08.mdh ~3~ 1 3 2 8 d 1 ~

techniques do not lend themselves to large scale screening.
T$ssue samples or isolated cells containing the target receptors, - e.g. heart atrial tissue, are costly to obtain, present in limited quantity, and difficult to maintain in a functionally viable state. Additionally, it is often difficult to reliably and reproducibly administer the candidate drug to tissue samples.
Screening assays using primary explants in tissue culture are undertaken in larger scale than is possible with tissue s~mples.
However, it is more difficult to assay physiological effect and the assays are sub~ect to interfersnce from many sources, e.g.
culture media or cultivation conditions. Finally, assays using receptors isolated from natural materials have the disadvantage that the receptor is sub~ect to natural variability and suitable natural sources may not always be available. It is an ob;ect herein to provide readily reproduci'ole, simple assay systems that can be practiced on a large scale for determining not only ligand binding but also the character of the binding as agonistic or antagonistic.

Similarly, meanin~ful clinical diagnosis often depends upon the assay of biologically active ligand without interference from inactive forms of the ligand, for example, ligands that have - been sub~ect to enzymatic or other processes of the test sub~ect - that change or even eliminate the activity of the ligand.
Immunoassay methods are widely ùsed in determining ligands in test samples. However, it is often quite difficult to identify antibodies that are able to discriminate beeween the active and inactive forms of a ligand. Receptors have infrequenely been used in place of antibodies as analyte binding reagents. However, not all substances that bind to receptors are necessarily capable of inducing receptor activity, i.e. active biologically. It is an ob~ect herein to provide a method that will idsntify ligands in clinical test samples which are active in inducine or inhibiting signal transduction by their receptors.

-- L03xO8.mdh 4 13~419 Many receptors have been identified that have at least some known ~n vitro assayable activiey that i6 dependent upon ligand interaction. For example, the binding of EGF to the epidermal growth receptor stimulates a phosphotransferase domain in the receptor to phosphorylate certain target amino acid residues located in its intracellular cytoplasmic domain, a process called autophosphorylation. Receptors also hre known to phosphorylate antibodies or ~pecific cytoplasmic substrate polypeptides that bind to the region in which their phosphotrans-ferase active site i8 located. Unfortunately, other receptors have no known ligand-dependent enzymatic activity, notwithstanding that they are known to bind ligands with high affinity, or their activity is 80 low that it is difficult to quantitatively assay ligand-dependent activation. It may be desirable for therapeutic purposes to antagonize or agonize a ligand interaction with such cryptic receptors but, in the absence of the tissue concerned or, in some cases an intact organism, no method is available for determining whether a candidate drug is simply binding the receptor in a function-neutral fashion, nor whether the candidate is binding as an agonist or antagonist. Accordingly, it is an ob~ect to provide a method for screening candidate drugs for ligand agonist or antagonist activity where the receptor for the ligand exhibits no known ~ignal transduction characteristic.

~ummasv These ob~ects are accomplished by the use of a novel receptor hybrid comprising the ligand binding domain of a receptor fused ts a heterologous reporter polypeptide which is capable of undergoing an nssayable change in conformation or function when the ligand binding domain of the receptor binds to either the ligand or to an agonist or antagonist of the ligand.

- L03xO8.md~

If a disease or injury is the result of a ligand acting on a given receptor, the ob~ective will be to ident$fy ~ubstances capable of counteracting the ligand's effect on the critical receptor, i.e., ligand antagonists. On the other hand, a model S therapy for a clinical condition characterized by insufficientligand activity would consist of drugs that enhance or supplement a defective or absent ligand, i.e., ligand agonists.

The hybrid receptor of this invention is useful in screening methods for identifying receptor-active agonistic drugs.
One incubates the hybrid receptor with the candidate drug and assays for the generation of a signsl by the heterologous reporter polypeptide. ~enerally, but not necessarily, the signal generated by the reporter polypeptide is assayed as an activation or stimulation of an enzymatic function of the reporter polypeptide.
It ls not necessary to include standards having known amounts of ligand unless one wishes to qusntify the agonist activity of ~he c~ndidate; in fact, the ligand which modulates the receptor activity in vivo may be completely unknown. It is one of the benefits of this assay system that neither the llgand for the receptor nor the i~ vivo signal transducing mechanism of the receptors need be known in order to identify agonist drugs.

The hybrid receptor is used to asssy amounts of blological ligand ln test samples in the same fashion as one screens for agonist drugs. Since this utility, by definition, contemplates a known ligand, then a standard curve using known amounts of ligand is prepared and compared with the test sample results.

Antagonist drug candidates are selected by the same assay as is used for identifying agonists, except that here the hybrid receptor is incubated with a known receptor agonist. The agonist, which may be a drug or the normal in vivo ligand, i~

L03x08.mdh ~ -6- 13~8419 incubated with the receptor before or, preferably, simultaneous with contacting the receptor with the csnditate drug. ~ntagonist activity is a function of the displacement of agonist or ligand activity as measured by changes in the reporter polypeptide.

A particular advantage of the hybrid receptor is that it enables a universal, portable assay system for any ligand-receptor interaction. This invention contemplates, for example, that the cytoplasmic domain of a first receptor is selected as the portable 9 reporter polypeptide. This domain i~ then substituted for the cytoplasmic domain of other receptors in preparing the hybrid receptor6 of the invention. The assay ~ystem, e.g. autophos-phGrylation assay, useful with the first receptor is then - 15 available for use with all other hybrid receptors containing the ~- cytoplasmic domain of the first receptor.

E~ief Description of the ~gures Figure la deplcts the composition of a plasmid employed - 20 in the expression of hybrids of the insulin and epidermal growth factor receptors. The region coding for d$fferent receptor mutants is shown by a 6haded bar (cDNA). Early SV40 promoter ~equences chown by heavy black arrows ~nd polyA addition sites have been ~arked. lhe dihydrofolate reductase (DHFR) coding sequence (Simonsen and Levinson, 1983, ~Proc. Natl. Acsd. Sci.
USA" 80:2495-2499) and restriction ~ites used for plasmid constructions are shown. Expression of both cDNAs was controlled by promoter sequences of the Simian virus (SV) 40 early region and by 3'-untranslated sequences of the gene coding for the hepatitis

3 B virus surface antigen (Crowley al., 1983, "Mol. Cell. Biol. n 3: 44-45). Sequences of the ~. Qli plasmid pML (a pBR322 -- derivative suitable for use in mammalian cells; Lusky and Botchan, 1981, "Nature" 293:79) containing the origin of replication and the ampicillin resistance gene were present to allow plasmid L03xO~.mdh -7- 13284~9 replication in E. coli.

Figure lb is a schematic comparison of insulin (HIR) and EGF (HER) receptors and a hybrid receptors IER and I~ER prepared therefrom. Humsn EGF receptor (HER), buman insulin receptor (HIR), insulin-EGF receptor chimera (IER), and insulin-~-subunit-EGF receptor chimera (I~ER) cDNAs are represented by hor~zontal lines and coding ~equences ~hown as a dotted box for HIR~
~equences (~), as a shaded box for HIR~, and as an open box for HER sequences. The coding regions have been aligned at the transmembrane domain (not shown in scale). The coding segment for the protein signal sequence is marked by (S) and the precursor clesva~e sites are indicated by a vertical line. The ~unction of the heterologow receptor cDNAs is shown by a zigzag line and synthetic ~ligonucleotides used at the ~unctions are represented by black bars. D~A restriction endonuclease cleavage sites relevant for the constructions are ~arked on top of the cDNA
sequences.
: .
Figure 2 illustrates that l25I insulin binding to COS-7 cells increases when ths COS-7 cells are transfected with the cDNA
constructs of Figure la, compared to cells transfected with a control expression vector.

Figures 3a-3d are SDS PAGE reducing electrophoresis gels of autophosphoryla~ed detergent lysates obtained from various transformed and control cells and immunoprecipitated with appropriate antibodies as noted in the Example. The (+) and (-) - gels represent insulin or (in the case of A431) EGF-treated receptors. Numbers in the margins are marker molecular weights.
Fig. 3a depicts the anti-HER i~mNnoprecipitated autophosphorylation products of mock-transformed controls and recombinant transformant cells. This demonstrates expression of hybrid insulin-EGF receptor constructs in the recombinants.

L03xO8.mdh -8- 1~28~19 Figure 3b demonstrates that the autophosphorylation of the hybrid containing the complete extracellular domain of the insulin receptor is activated by insulin.

Figure 3c depicts the kinetics of the insulin-activated autophosphorylstion of the IER receptor. It shows that the autophosphorylation observed is dependent upon the time of the phosphorylation reaction.

Figure 3d illustrates the change in SDS-PAGE migration of the IER receptor after insulin activation.

5Figure 4 depicts the structure of HER-erbB, a hybrid receptor containing the epidermal growth factor extracellular domain ~nd a fragment of the erbB oncogene to serve as the reporter molecule.

Figure 5 depicts electrophoresis gels dçmonstrating 20~utophosphorylation of 8 hybrid oncogene-receptor construct in the presence (I) or absence of ligand (EGF)~-).

Dçtailod Description 25The hybrid receptor is the core of the methods described herein. It principally comprises a ligand binding domain and 8 reporter polypeptide. The ligand binding domain is located within the extracellular region of a receptor. It is often difficult to identify the precise amino acid sequences involved in ligand bindiDg. In fact, 6everal regions may be involved in ligand binding, particularly where the ligand is a polypeptide. Thus, it is preferred that the entire sxtracellular region of the receptor be assembled into the hybrid. This also will help to ensure that 35the ligand binding domain is maintained in its proper conforma-L03xO~.mdh 1328~19 tion.

Suitable ligand binding domains are selected in any one of several ways. First, when one intends to use the hybrid to assay for a known ligand in test samples, or to screen for agonists or antagonists to such ligand, then the ligand binding domain is selected from a known receptor for the ligand. If the ligand ~s ~nown, but its receptor is not, then it will be necessary to identify its cell surface receptor. This may be 0 accomplished by 1) aecuring cells from tissues with which the ligand ls known to bind or to functionally interact, 2) obtaining from the cells in known fashion a membrane protein preparation, 3) incubating the preparation with the ligand, 4) geparating the llgand-receptor complex from the incubat~on mixture (for example by preinsolubilizing the ligand on cyanogen bromide act~vated Sepharose~, 5) separating the receptor from the ligand, 6) - obtaining amino acid sequence from a portion of the receptor, 7) preparing nucleic acid probes encoding the determined Amino acid sequence (either a single long probe of > about 40bp or a pool of shorter probes), 8) preparing a cDNA or genomic DNA phage or plasmid (~ector) library from the organism or cells from which the receptor was obtained, 9) hybridizing the probes to the library to ~dentify plasmids or phage which contain DNA oncoding the receptor, and 10) determining the nucleotide and imputed amino acid sequence of the receptor to the extent necessary to itentify the region extending from the amino terminus through a transmembrane sequence. If no single vector contains DNA encoding the entire extracellular domain of the receptor, the desired DNA
is assembled by restriction enzyme digestion of the various vectors at common sites, isolation of the appropriate fragments and relegation by methods already known per se. Other procedures for identifying recep~ors for known ligands are known to those skilled in the art or will become availablç in the future.

L03xO~.mdh 13%8~19 A putative receptor may have been itent~fied but its ligand i~ vivo remains unknown. For example, study of endocrine tissues from such glands as the pituitary or adrenals will lead to the identification of membrane bound proteins that are structur-ally similar to other known receptors, i.e. they will have a large ttypically >500 residues) extracellular domAin, a hydrophobic ; transmembrane sequence and a carboxy-terminal cytoplasmic region.
Sim~larly, a receptor inventory for malignant cells will be useful for identifying unique receptors present in high density that may be associated with the transformed phenotype. The extracellular domains of such receptors are also useful herein.

A receptor and its ligand ~ay have been identified but the cytoplasmic domain ~ay have no known function, e.g. it is not known to have phosphotransferase actlvity, to activate adenylate or guanylate cyclase, or to transport ligand. The ligand binding domain f rom ~uch receptors is useful notwithstanding that the ligand-receptor interaction produces no or insufficiently detectable signal in the native receptor because a detectable signal is provided by the reporter polypeptide in the hybrid construction. Thus, in the absence of the reporter polypeptide no ~ethod wDuld be available to determine ~n the case of some receptor whether a receptor-bound candidate dru~ was binding nonspecifically or was acting as an agonist or antagonist, nor would it be possible to assay for biologically active native ligands.

The reporter polypeptite is heterologous to the ligand binding domain and is any polypeptide that changes its character upon the binding of a ligand to the binding domain. This change in character is generally detected by a change in the enzymatic activity or i ~unological identity of the reporter polypeptide.
6enerally the reporter polypeptide w~ll be the cytoplasmic domain of a heterologous receptor or receptor analogue, e.g. oncogene, L03x08.mdh which is known to undergo B change in immunological or enzymatic identity upon ligand binding. It is preferred to use the cyto-plasmic phosphotransferase from such receptors as the insulin or epidermal growth factor receptors. However, other receptors as the B-adrenergic receptor, acetylcholine receptor, sdrenaline receptor and the llke ure known to bind proteins termed G proteins t~at serve as intermediate transducing molecules in the activation ; or inhibition of edenylate or granylate cyclases. Such proteins have been isslated and characterized. It is within the scope herein to use as the reporter polypeptide the G protein binding domains of such receptors. It is not necessary to use the entire cytoplasmic domain from a heterologous receptor or receptor analogue, only that portion that performs the desired function herein, nor is it necessary to use a heterologous cytoplasmic domain that is an intact, unmodified sequence from another receptor. For example, an amino acid sequence variant or derivative of the cytoplasmic domain of the receptor supplying the ligsnd blnding domain is also acceptable.

Uithout bein8 limited to a particular theory of function, we believe that the change in the character of the reporter polypeptide is not caused by ~teric hinderance of the reporter by the ligand, e.g. where the ligand occludes an active site on the roporter domain by virtue of ~teric bulk. Rather, the method herein harnes6e6 the eignal transducing mechani~m of receptors whsreby changes in the ligand binding domain are transduced through the receptor molecule to the reporter domain by conformational changes in the molecule, which changes affect the function or character of the cytoplasmic domain of the reporter.
We have discovered that this trensduclng mechanism also functions when the reporter polypeptide is heterologous to the ligand binding domain.

Optionally, the hybrid receptor will contain a transmem-L03x08.mdh brane sequence fused between the ligand binding domain and the reporter polypeptide. Typical transmembrane domains contain about from 20 to 25 residues and show a hydropathy peak of about from 1.5 to 3,5. They contain a high proportion of residues having hydrophobic side chains, e.g. leucine, isoleucine, phenylalanine, valine and methionine. Suitable transmembrane sequences are obtained from the receptor supplying the extracellular ligand binding domain, although the transmembrane sequence also may be entirely synthetic or obtained from integral membrane proteins or unrelated receptors, in the last instance including the transmem-brane region ordinarily associated with the reporter polypeptide where the reporter is the cytoplasmic domain of a heterologous receptor.

The hybrid receptor components suitably originate from humans, animals, plants, insects, microorganisms including parasites, viruses and fungi and other suitable species. The specieg of origin for the l~gand binding domain is selected for the presence of a receptor capable of binding the ligand of interest or for the presence of the target physiological activity.
- It is not necessary that the reporter polypeptide or transmembrAne ; region be from the same species as the ligand binding domain.

The hybrid receptors preferably are synthesized in recombinant cell culture because they ars generally too large and complex to practically synthesize by in vitro methods that are available to the art today.

Recombinant methods for synthesis of the hybrid receptor co ~ence with the constructicn of a replicable vector containing nucleic acid that encodes the hybrid receptor. Vectors typically perform two functions in collaboration with compatible host cells.
One function is to facilitate the cloning of the nucleic acid that encodes the hybrid receptor, i.e., to produce usable quantities of L03x0~.mdh the nucleic acid. The other function is to direct the expression - of the hybrid receptor. One or both of these iunct~ons are performed by the vector-host system. The vectors will contain different components depending upon the function they sre to perform as well as the host cell that is selected.

Esch vector will contain nucleic scid ehat encodes the hybrid receptor. Typically, this will be D~A that encodes the hybrid receptor in its ~ature form linked at its ~mino terminus to a secretion signal. This secretion signal prefersbly is the oignal pre6equence that normally directs the secretion of the receptor from which the ligand binding domain was obtained.
However, suitable secretion signals also include sig~als from other receptors or from secreted polypeptides of the same or related species.

The secreted hybrid will lody,e in the recombinant host membrane if it contains a transmembrane region. On the other ~ hand, if such a region is not present in the hybrid, then the - 20 hybrid ~ay be fiecreted into the culture medium. Ortinarily, hybrids are preferred that contain a transmembrane region so as to ~etain as much structural fidelity as possible. Bowever, the purificatlon of transmembrane-deleted receptors is less complex than in the c~se of membrsne-bound because ln the latter instance the hybrid receptor should be purified free of other cell membrane proteins. Furthermore, the cell-bound hybrid receptor may exert an undesired biological effect on the host if induced to sccumulate in large populations in the cell membrane during the growth phase. This potential problem is overcome by placing the nucleic acid encoding the hybrid receptor under the control of an inducible promoter.

In cloning vectors, the hybrid receptor-encoding nucleic acid ordinarily is present together with a nucleic ~cid sequence ~03x0B.mdh -14- 1~28419 that enables the vector to replicate in a selected host cell independent of the host chromosomes. This sequence is generally an origin of replication or an autonomously replicating sequence.
Such sequences are well-known for a variety of bacteria, yeast and higher eukaryotic cells. The origin from the well-known plasmid pBR322 is suitable for E. coli bacteria, the 2~ plasmid origin for yeast and various ~iral origins for ma~malian cells ~SV40, polyoma, adenovirus or bovine papilloma virus). Less desirably, DNA is cloned by insertion into the genome of a host. This is 0 readily accomplished with b~cillus species, for example, by inserting into the vector DNA that i8 complementary to bacillus genomic DNA. Transfection of bacillus with tbis vector results in homologous recombination with the genome and insertion of the hybrid receptor DNA. However, the recovery of genomic DNA
encoding the hybrid receptor is more complex than obtaining exogenously replicated viral or plasmid DNA because restriction enzyme digestion is required to recover the hybrid receptor DNA
from the genome of the cloning vehicle.

Expression and cloning vectors should contain a selection gene, also termed a selectable mar~er. This is a gene th~t encodes a protein necessary for the survival Dr growth of a host cell transformed with the vector. The presence of this gene ensures the growth of only those host cells which express the inserts, ~ypical ~elect~on ~enes cncode proteins that ~a) confer resistance to antibiotics or other toxins, e.g smpicillin, neomycin, methotrexate or tetracycline, (b) complement auxotrophic deficiencies, or (c) supply critical nutrients not available from csmplex media, e.g the gene encoding D-alanine racemase for bacilli A suitable election gene for use in yeast is the trDl gene present in the yeast plasmid YRp7 (Stinchcomb et al , 1979, "Naturen, 282: 39; Kingsman et al , 1979, ~Genen, 7: 141; or L03xD8 mdh _ ' -15- 132%~19 Tschemper et al., 1980, "Gene", 10: 157). The tr~l gene provides a selection marker for 8 mutant strain of yeast lacking the ability to grow in tryptophan, for example ATCC No. 44076 or PEP4-1 (Jones, 1977, "Genetics~, 85: 12). The presence of the trpl lesion in the yeast host cell genome then provides an effective environment for detecting transformat~on by growth in the absence of tryptophan. Similarly, Leu2 deficient yeast strains (ATCC
20,622 or 38,626) are complemented by known plasmids bearing the 1&_2 gene.

Examples of suitable selectable ~arkers for mammalian cells are dihydrofolate reductase (DHFR), thymidine kinase or proteins for neomycin resistance. Such markers enable the identi~ication of cells which were competent to take up the hybrid receptor nucleic ~cid. The ~ammalian cell transformants are placed under selection pressure which only the transformants are uniquely adapted to survive by ~irtue of having taken up the marker. Selection pressure is i~posed by culturing the transformants in successive rounds of cell culture in which the concentration of selection agent in the medium is successively increased, thereby leading to amplification of both the selection gene and the DNA encoding the hybrid receptor. Increased quantities of hybrid receptor are synthesized from the amplified DNA

For example, selection for DHFR transformed cells is conducted in a culture medium which laeks hypoxanthine, glycine, and thymidine. An appropriate host cell in this case is the Chinese hamster ovary (CH0) cell line deficient in DHFR activity, prepared and propagated as described by Urlaub and Ch&sin, 1980, "Proc. Nat'l. Acad. Sci. USA" 77: 4216.

A particularly useful DHFR is a mutant DHFR that is hig'hly resistant to methotrexate (MTX) (EP 117,060A). This L03x08.mdh -16- 1328~19 selection a~ent can be used with any otherwise suitable host, notwithstanding the presence of endogenous DHFR. One simply includes sufficient MTX in the medium to insctivate 811 of the endogenous DHFR, whereupon MTX selection becomes solely a function of amplification of the mutant DHFR DNA. Most eukaryotic cells which are capable of adsorbing NTX appear to be methotrexate sensitive. One such useful cell line is a CHO line, CHO-Kl (ATCC
No. CCL 61).

0 Other methods, vectors and host cellR suitable for ~dap-tation to the synthesis of the hybrid receptor in recombinant vertebrate cell culture are described in M.J. Gething et ~
~Nature" 2~: 620-625 (1981); N. Mantei et H~ Nature" ~ 40-46; and A. Levinson et al., EP 117,060A and 117,058A.

Expression vectors, unlike cloning vectors, should contain a promoter and/or other sequence which is recognized by the host organism for strong transcription of the hybrid receptor-encoding DNA. This is generally a promoter homologous to the intended host. In the case of vectors for higher eukaryotes, enhancer sequences are useful for further increasing transcription from promoters. Unlike promoters, enhancers do not need to be located 5' to the hybrid receptor encoding nucleic acid. Commonly used promoters for prokaryotes include the ~-lactamase and lactose promoter systems (Chang et al., 1978, "Naturen, ~1~: 615; and Goeddel et al., 1979, ~Nature", ~1: 544), alkaline phosphatase, a tryptophan (trp) promoter system (Goeddel 1980, "Nucleic Acids Res. n 8: 4057 and EPO Appln. Publ. No. 36,776) and hybrid promoters such as the tac promoter (H. de Boer et al., 1983, "Proc. Nat'l. hcad. Sci. ~SA" 80: 21-25). However, other known microbial promoters are suitable. Their nucleotide sequences have been published, thereby enabling a skilled worker operably to ligate them to DNA encoding the hybrid receptor in plasmid vectors (Siebenlist et ~1., 1980, "Cell" 20: 269) using linkers or L03x08.mdh adaptors to supply any required restriction sites. Promoters for use in prokaryotic systems also will contain a Shine-Dal~arno (S.D.) sequence operably linked to the DNA encoding the hybrid receptor.

Suitable promoting sequences in yeast vectors include the promoters for metallothionein, 3-phosphoglycerate kinase (-Hitzeman et ~1., 1980, ~J. Biol. Chem. n, 2~: 2073) or other glycolytic enzymes (Hess et ~1., 1968, "J. Adv. Enzyme Reg. n ~ 7:
0 149; and Holland, 1978, ~iochemistryn, ~: 4900), such as enolase, glyceraldehyde-3-phosphate dehydrogenase, hexokinase, pyr wate decarboxylase, phosphofructokinase, glucose-6-phosphate isomerase, 3-phosphoglycer~te mutase, pyruvate kinase, triose-phosphate isomerase, phosphoglucose isomerase, and glucokinase.

Other yeast promoters, which have the additional advantage of transcription controlled by growth conditions, are the promoter regions for alcohol dehydro~enase 2, isocytochrome C, acid phosphatase, degradative enzymes associated with nitrogen ~etsbolism, and the aforementioned metallothionein and gly-ceraltehyte-3-phosphate dehydrogenase, as well as enzymes responsible for maltose and galactose utilization. Suitable vectors and promoters for use in yeast expression are fur~her described in R. Hitzeman et ~1., EP 73,657A.

Transcription from vectors in m~mmalian host cells is controlled by promoters and/or enhancers obtained from the genomes of bovine papilloma virus, vaccinia virus, polyoma virus, adenovirus 2, retroviruses, hepatitis-B virus and ~ost preferably Simian Virus 40 (SV40), operably linked to the hybrld receptor nucleic acid. The early and late promoters of the SV40 virus are as conveniently obtained as an SV40 restriction fragment which also contains the SV40 viral origin of replication (Fiers et al., 1978, "Nature", 273: 113). Of course, promoters or enhancers from L03xOB.mdh ' -lB- 1~28~

the host cell or related species also are useful herein.

Nucleic acid is operably linked when it is placed into a functional relationship wieh another nucleic acid 6equence. For 5example, DNA for a presequence or secretory leader is operably llnked to DNA for a polypeptide if it is expressed as a preprotein which par~icipates in the secretion of the polypeptide; a promoter or enhancer is operably linked to a coding sequence if it affects the transcription of the sequence; or a ribosome binding site is 10operably linked to a coding sequence if it is positioned so as to facilitAte transla*ion. Gonerally, operably linked means that the DNA ~equences being linked are contiguous and, in the case of secretory leader, contiguou6 and in reading frame.

5Expression vectors used in eukaryotic host cells (yeast, fungi, ~nsect, plant, animal or human) will also contain sequences necessary for the termination of transcription and for stabilizing the ~RNA. Such sequences are commonly available from the 3' untranslated regions of eukaryotic or viral cDNAs. These regions 20contain regions that are transcribed as polyadenylated segments in the untranslated portion of the mRNA encoding the hybrid receptor.
The 3' untranslated regions lso include tr~nscription termination sites.

25Suitable host cells for clonlng or expressing the vectors horein are prokaryotes, yeast or higher eukaryotic cells.
Prokaryotes include gram negative or gram positive organisms, for example ~ ÇQl~ or bacilli. A preferred cloning host is E. coli 294 (ATCC 31,446) although other ~ram negative or gram positive prokaryotes such as E. coli B, E. coli X1776 (ATCC 31,537), E.
ÇQai W3110 (ATCC 27,325), pseudomonas speciesl or e~rratia Marcesans are suitable.

35In addition to prokaryotes, eukaryotic microbes such as L03x08.mdh Y -.

' -19- 1328419 filamentous fungi or yeast are suitable hosts for the hybrid receptor encoding vectors. ~accharomvces cerevisiae, or common baker's yeast, is the most commonly used ~mong lower eukaryotic host microorganisms. However, a number of other genera, species and strains are commonly available and useful herein.

The ~referred host cells for the ~xpression of functional hybrid receptors are cultures of cells derived from multicellular organisms. In oany cases, hybrld receptors contain 0 hydrophobic regions that are inco~patible with lower microorgan-isms, require co~plex processing to properly form disulfide bonds ant often require subunit processing. In addition, it is desirable to glycosylate the receptors in a fashion similar to the native receptors. All of these functions can be best performed by higher eukaryotic cells. In principle, any higher eukaryotic cell culture is workable, whether from ~er~ebrate or invertebrate culture, although cells from mammals such as humans are preferred.
Propagation of ~uch cells in culture is ~ç~ se well known. See Tissue Culture, Academic Press, Kruse and Patterson, editors (1973). Examples of w eful mammalian host cell lines are V B0 and HeLa cells, ~hinese hamster ovary cell lines, and ~I38, BHK, COS-7 and MDCK cell lines.

The hybrid roceptors of this invention are employed in drug screening or biologically active ligand assay by a process thst fundamentally comprises incubating ~he receptor ~ith the test sampIe, controls and (optionally) standards, follo~ed by measuring change in the reporter polypeptide. Since we have discovered that ligand binding causes a change in the conformation of the reporter polypeptide it is within the scope hereof to detect such changes by any one of several methods. Typically, one measures changes in the protein binding or enzymatic activity of the reporter polypeptide. In one embodiment an antibody is raised against the activated conformation and the binding of this L03x08.mdh antibody to the hybrit receptor i8 measured after the receptor has been incubated with the ligand or candidate drug. Thls assay is conducted in the same fsshion as con~entional immunoassay methods for any protein antigen. Antibodies are kno~n ~ç~ se that are capable of binding phosphotyrosine containing proteins (Wang, 1985, "Mol. and Cell. Biol.~ 5(12): 3640-3643; Ross et al., l9Bl, ~Nature" ~2~: 654; and Pang et gl., 1985, ~Arch. Biochem.
Biophys. n ~g~(l): 175). While those nntibod~es are useful in the ~ethod here~n, hybrid receptors enable the selection of anti-phosphotyrosine antibotie6 that, unlike the prior art antibodies, are specific for the reporter polypeptide and will not cross-react with other receptors or phosphorylated proteins, yet which are Just as versatile in measuring the effect of a ligand on a receptor binding domain. The tisadvantage Gf this ~ethod is that it requires a phase separation to remove the unbound labelled antibody from the reporter-bound antibody. However, the method does not require covalent modification of the hybrid receptor.

Analogous to assays using the binding of a specific antibody to the reporter polypeptide are methods that directly or indirectly measure the binding to the reporter of a non-immune binding prote{n with which it normally interacts. Typical binding proteins ~re the G proteins that ssociato with certain ligand-activated receptors. The reporter polypeptide in this case is the cytoplasmic domain of a receptor such as the beta-adrenergic receptor. The binding of the G protein is assayed in the same fashion as antibody binding, e.g. by displacement of labelled G
protein, Gr by determination of GTP or ATP binding to the activated G protein.

If the reporter polypeptide is the enzymatically active cytoplasmic domain of a heterologous receptor, then the preferred detection method will be an assay for that activity. At the present time such activity includes protein phosphorylkinase LD3x08.mdh ,~j .

. -21- 1328419 activity, primarily tyrosine kinase actlvity but in some cases serine or threonine kinase activity. Rinase activity is measurable in any way in which kinase activity has been assayed heretofore. One con~entional, snd presently preferred, method for kinase sctivity is to assay the incorporation of radiophosphorus into the reporter polypeptide through autophosphorylation with 32p. It is preferred to form hybrids of receptors having the same class of activity.

However, it i6 within the scope herein to measure changes in the reporter polypeptide by methods other than enzymological activity or polypeptide interactions. One such ~ethod contemplates binding sn organic moiety to the receptor that undergoes a change in character upon ligand binding. For example, the reporter polypeptide is labelled wlth a stnble ~ree radical, a chemiluminescent group or a fluorescent molecule such as fluorescein isothiocyanate. Each of these labels are well known in the diagnostic i~munochemistry ~rt and conventional methods are well known for covalently lin~ing them to proteins. These methods are useful for labelling the reporter polypeptide in the same fashion as other proteins. Changes in the conformation of the receptor polypeptide upon the binding of ligand or active candidate trug to the ligand binding tomain are detected by changes in the label. For example, the rotstional moment of a stable free rsdicsl lsbel will be incressed or decreased by ligand-activated changes in reporter polypeptide conformation.
Similarly, the fluorescence or luminescence of reporter poly-peptide labels will change upon the binding of ligand or active candidate to the receptor because of the reorientation of polypep-tide species that engage in intramolecular energy transfers. This is detected by changes in the intensity, polarization or wave length of the label molecule; typically, one detects the enhancement or quenching of the label fluorescence or chemilumin-escence. The advantage of the labelled reporter method is that35 LO3xOB.mdh the ligand or candidate dTug assay is conductet exclusi~ely in aqueous solution and no phase separation is required. This permits the automation of the 6creening method using continuous flow instruments such as Autoanalyzers. Such methots are useful with nstive as well as the hybrid receptors.

In order to simplify the Examples certain frequently occurring methods will be referenced by ~horthand phrases.

0 ~Plasmids" ro desigDated by a low case p preceded and/or followed by capital letters and/or numbers. The starting plasmids herein are com~ercially a~ailable, are publicly ava~lable on an unrestricted basis, or can be constructed from ~uch available plasmids in accord with published procedures. In IS addition, other equiYalent plasmids are known in the art and will be apparent to the ordinary artisan.

~Digestion~ or "cleavage" of DNA refers to catalytic cleavage of the DNA with an enzyme that acts only at certain locations in the DNA. Such enzymes are called restriction enzymes, and the sites for which each is specific is called a restriction site. The variDus restriction enzymes w ed herein are co ercially available and their reaction conditions, cofactors - and other roquiroments s established by the onzyme ~uppliers were used. Restriction onzymes commonly are designatet by abbre~lations composed of a capital letter followed by other letters representing the microorganism from which each restriction enzyme ori~inally was obtained and then a number designating the particular enzyme. Appropriate buffers and substrate ~mounts for particular restriction enzymes are specified by the manufacturer.
Incubation times of about 1 to several hours at 37C are ordinarily used, but may vary in accortance with the supplier's instructions. After incubation, pTotein is remo~ed by extraction with phenol and chloroform, and the digested nucleic ncid is L03xO8.mdh x,., recovered from the aqueous fractior. by precipitation with ethanol.
Digestion with a restriction enzyme infrequently is followed with bacterial alkaline phosphatase hydrolysis of the terminal 5' phosphates to prevent the two restriction cleaved ends of ~ DNA
fragment from "circularizing" or forming a closed loop that would impede insertion of another DNA fragment at the restriction site.
Unless otherwise stRted, digestion of plasmids is not followed by 5' terminal dephosphorylation. Procedures and reagents for dephosphorylation are conventional (T. Maniatis et al., 1982, 0 Molecular Clonine pp. 133-134).

~Filling" or ~blunting" refers to the procedure by which the single stranded end in the cohesive terminus of a restriction enzyme-cleaved nucleic acid Is converted to a double strand. This eliminates the cohesive terminus and forms a bl~nt end. This process is a versatile tool for converting a restriction cut end that may be cohesive with the ends created by only one or a few other restriction enzgmes into a terminus compatible with any blunt-cutting restriction endonuclease or other filled cohesive terminus. Typically, blunting i8 accomplished by incubating 2- .-15~g of the target DNA in lOmM Mg C12, lmM dithiothreitol, 50mM
NaCl, lOmM Tris (pH 7.5) buffer at about 37-C in the presence of 8 units of the Klenow fragment of DNA polymerase I ~nd 250~M of each of the four deoxynucloosite triphosphates. The incubation generally 1~ terminated after 30 min. by phenol and chloroform ex~raction and ethanol precipitation.

~Recovery" or ~isolation" of a given fragment of DNA
from a restriction digest means separation of the digest on poly-acrylamide or agarose gel by electrophores~s, identification of the fragment of interest by comparison of its mobility versus that of marker DNA fragments of known molecular weight, removal of the gel section containing the desired fragment, and separation of ~he DNA from the gel. This procedure is known generally. For L03xOB.mdh example, see R. Lawn et ~1., 1981, "Nucleic Acids Res." 9:6103-6114, and D. Goeddel et ~1., 1980, "Nucleic Acids Res.: 8:4057.

"Transformation" means introducing DNA into an organ~sm so that the DNA is replicable, either as nn extrachromosomal element or chromosomal integrant. Unless otherwise provided, the method used herein for transform tion of ~_ coli is the CaC12 method of ~andel Ç~ ~1., 1970, "J. Mol. Biol.~ 53: 154.

0 ~Ligation" refers to the process of forming phosphodiester bonds between two double stranded nucleic acid fragments (T. ~aniatis et ~1., Id., p. 146). Unless otherwise provided, ligation may be accomplished using known buffers and conditions with 10 units of T4 DNA ligase (nligasen) per 1 ~g of the DNA fragments to be ligated.

~Preparation" of DNA from transformants means isolating plasmid DNA from microbial culture. ~nless otherwise provited, the alkaline/SDS method of Maniatis et ~1., Id., P. 90, may be used.

~Oligonucleotides" re short length single or double ctranded polydeoxynucleotides which are chemically synthesized by known methods and then purified on polyacrylamide gels.
The following examples are intended to merely illustrate the best mode now known for practicing the invention, but the invention is not to be considered limited thereto.

All literaturP citations herein are expressly incorporated by reference.

L03x08 mdh .~.. . .. . . . . .

~mple 1 Construction oi the Insulln Rece~tor (IR-Ex~ression Plasmid A gel purified ~lI fragment ( 5.2 kb) from ~HIR-P12 containing the entire HIR coding ~equence was ~ubcloned into the pUC12 (New England Biolabs) polylinker region by digesting pUC12 wi~h ~lI Rnd ligating the purified ~alI fragment to the vector.
0 Colonies were grown up and screened for clones havlng the desired orientation with the 5' end of the HIR coding sequence next to the pUC12 ~I site. This vector was cut with ~k~I and ~FaI (E~I is located in the 3' untranslated region o~ HIR) and the HIR-containing frag~ent was isolated. This fragment was inserted into ~a~malian e~pression vector (pCVSVEHBVE400, European Publn. Mo.
117,060), which had been digested first with E~HI. The ~EHI
cohesive terminii were filled and the plas~id then was digested with X~al. Thus, insertion of the ~ DraI was only possible in the orientation necessary for expression of the HIR mRNA. The resulting insulin receptor expression plasmid was designated pCVSV-HIRc.

~x~nDle 2 Çon~t N ction of Voctor for E~rossion of n Insulin-~CF RoceDtor H~brid The following fragments were ligated in a four-fsctor ligation: (a) A 931 bp E~HI-ea~II restriction fragment from the IR expression plasmid pCVSVE-HIRc, (b) a 1150 bp ADaI-SstI
restriction fragment of the human EGF receptor sequence contained in the recombinant phage ~HER-A64 (Ullrich et al., 1984, ~Nature"
309: 41R-425), (c) a synthetic oligonucleotide linker containing 5'-CCCG$CAAATATCGCCACTGGGATGGTGGGGGCC-3' and 5'-CCCACCATCCCAGTGGC
GATATTTGACGGGACGT-3', and (d) pUC12 opened with SstI and ~HI.

L03xO8.mdh , -26- 1~28419 In this way sequences coding for the extracellular dDmain of the insulin receptor were ~oined to sequences coding for the transmembrane and cytoplasmic domains of the EGF receptor and placed into an expression plasmid. Plasmid pVC12/H1R-HER Int.
which contained DNA encoding the hybrid W8S recovered from an ampicillin resistant colony of transformant ~. coli 294. This plasmid was digested wlth E~HI and op_I and a 965bp fragment containing the IER ~unction was recovered (fragment 1). pCVSV-HIRc was digested with PvuI and a~HI and a 3117 bp fragment containing the rest of the IR coding sequence and parts of the mammalian expression vector was recovered (fragment 2). ~HER-A64 is digested with A~aI-BglII and an 810 bp fragment recovered (fragment 3) and with EglII-XmnI and a l kb fra~ment recovered (fragment 4). Fragments 3 and 4 code for the transmembrane and cytoplasmic domains of the human EGF receptor. pCVSVEHBVE400 is digested with E~EHI, the restriction site was endfilled and the DNA 6ubsequently digested ~ith ~I. A 4 kb ~HI-PvuI fra~ent was recovered (fragment 5) coding for parts of the mammalian expresgion vector as described. ~ mixture of fragments l, 2, 3, 4 and 5 was ligated and w ed to transform ~. cQli 294 DNA.
Ampicillin resistant colonies were screened by restrictio~
analysis. A 241 bp E~I frag~ent overlapping the HIR-HER ~unction was cloned into M13Tyl31 and sequenced to verify the expected ~unction.

An expression vector for a hybrid receptor not containing a transmembrane region is constructed in the same fashion as described above using pCVSV-HIRc except that the transmembrane region downstream from the A~aI EGF receptor is deleted by M13 mutagenesis and an in-frame eE~I adaptor ligated to the deleted ER fragment.

A vector encoding the hybrid receptor I~ER, a hybrid of the insulin receptor ~ chain with the EGF receptor transmembrane 3~
L03x08.mdh . -27-and cytoplasmic domains without any HIR B-chain sequence, WAS made by oligonucleotite-dlrectet deletion mutagenesis of the IER
plasmid. A 2.1 kb ~e~II restriction fragment coding for ~oined insulin ~nd EGF receptor sequences was introduced into the ~mHI
S site of an M13mplO vector. Molecules with the desired orientation of the IR sequences next to the HindIII site of M13mplO were identified and a ~ingle-stranded template was prepared for deletion mutagenesi~ with the oligonucleotide 5'-CCCCAGGCCATCTATCGCCACTGGGA-3' based on the protocols of Adelman et 0 ~ DNA" ~: 183-193 (1983). 50 ng of phosphorylated primer was hybridized to 2 ~g of oingle-stranded M13 template. The mutagenized second strand was completed and double-stranded molecules were lntroduced into ~. çQll JM101. Resulting plaques were screened as described by Benton and Davis "Science" ~ 180-IS 182 (1977) at hi~h stringency using the primer as a hybridization probe. Double-stranded DNA was prepared and a 1.2 kb ~EII
restriction fragment containing the ~utated region was used to replace the respective DNA fragment in the IER expression plasmid, yielding pI~ER.

mple 3 Expression of the H~brid Insulln nd EGF Receptors COS-7 ~onkey kidney cells (Gluzman, 1981, "Cell" ~:
175-182) were cultured ln DMEM mixed with F12 medium (50:50), containing 10 percent fetal bovine serum and antibiotics. All cell culture med$a (Gibco) contained 2mM L-glutamine and 20mM
~EPES pH 7.4.

pIER or pI~ER from Example 2 was introduced into COS-7 cells by calcium phosphate coprecipitation based on the protocol of Graham and Van der Eb, 1973, "Virology~ ~: 456-467.
Subconfluent cells were transfected with 10 ~g of plasmid DNA per 8 cm culture dish. Plasmid DNA was dissolved in 0.55 ml of lmM

L03x08.mdh Tris pH 7.5, O.lmM EDTA, 250mM CBC12, after which 0.5 ml of 50mM
HEPES pH 7.12, 280mM NaCl snd 1.5mM Na2HP04 was slowly added. A
precipitate gradually formed within 45 min. which wns atded to the cell culture medium. The transfected COS-7 cells were cultured for 53 hrs. at 37-C in DMEH mixed with F-12 medium (50:50) containing antibiotics, 2mM L-glutamine, 20 mM HEPES ~nd lQ~ by volume fetal bovine serum (pH 7.4).

COS-7 cells transformet with pIER or pI3ER were washed o two times with PBS and incubated in 1 ml of serum-free cell culture medium per 2.2 cm well containing O.2~ bo~ine serum albumin (Sigma), bacitracin (0.5 mg/ml, Slgma) and 125I insulin (0.5 ~Ci/~ell) at 93 ~Ci/~g for 2 h at 21-C. Cells were washed 3 times with PBS at 4-C snt lysed in 0.5 ml 0.1 percent SDS, 0.1 M
NsOH for 30 min. at 37-C. Th¢ radioacti~ity was determined in a g~mma counter. Fig. 2a demonstrates that insulin binding to the transformants increased over that of controls.

Human epidermoit carcinoma cells A431 (a source of EGF
receptor controls) were cultured $n DMEM containing 4.5 mg glucose per liter, 10 percent fetal bo~ine serum and antibiotics.

Exam~le 4 Hormone Stimulated Auto~hos~horvlation of Normal and HYbrid Receotors pIER or pI~ER transformed and mock transformed COS-7 cell nonolayers grown in B cm culture dishes for 53 hours, or A431 cells, were washed twice with PBS and solubilized as described by Kris et ~1., 1985, ~Celln 40: 619-S25. One ml of 50mM HEPES
buffer pH 7.5 containing 150mM NaCl, 1.5mM MgC12, lmM EGTA, 10 percent glycerol, 1 percent Tr$ton X-100, 1 percent Aprotinin (Sigma) and 4~g/ml phenylmethylsulfonyl fluoride (PNSF) (Sigma) and 0.5 mg/ml bacitracin ~Sigma) was added to the monolayers at ~Trade-mark L03x08.mdh .~
.--.' 1328~1;9

4-C for 5 min. The buffer ~hich contained the solubilized cellular proteins was remo~ed from the culture dish and centri-fuged at lO,OOOg for 5 min at 4-C. Culture supernatants from transmembrane-deleted hybrid receptor transformed cells are centrifuged at 10,000 g for 5 ~in. at 4-C. 0.2 ml of the cell lysis or culture supernatant was ~ncubated with 200nM insulin (Sigma) or 1 ~M EGF for 1 hour.

A mouse monoclonal Antibody capable of binding the 0 insulin receptor (CII25.3, dbscrlbed by Ganguly et ~1....... 1985, ~Current Topics in Cellular Regulation" ~7: 83-94) was insolubilized by adsorption to protein A-Sepharose. Howeves, it ~ill be appreciated that any polyclonal or monoclonal anti-insulin receptor antibody can be used. 1~1 of antibody was mixed with 50 lS ~1 of a swollen and prewashed 1:1 protein A-Sepharose slurry in detergent-free lysis buffer for 30 min in order to adsorb the ~nti-IR antiboty.

50 ~1 of insolubilizet anti-IR antibody slurry was added to the EGF or insulin treated cell lysate or cell culture supernatant and incubatet for 15 ~in. at 4-C. The resulting immunoprecipitate was washet 4 times w~th 0.9 ml HNTG buffer (20mM
HEPES pH 7.5, 150~M NaCl, 10 percent glycerol, ant 0.1 percent Triton X-100). The precipltato in B volume of 30 ~1 was ad~usted to 5~M MnC12, ant 15~Ci of ~-32P-ATP (5,000 Ci/ ol) w~s added for 0.5-10 min. at 4-C. The final ATP concentrat$on ~as 0.1 pM ATP
(for EGF, IER or I~ER transformants and their controls) or 100 yM
ATP (for HIR transformants and their controls). The autophosphorylation reaction ~as stopped by addin~ 2~ ~1 o f 3 times concentrated SDS sample buffer. The autophosphorylation reaction ~as terminated after 5 min. in Fig. 3a snd 3d, 1 min. in Fig. 3b and after the ti~es indicated in Fig. 3c by boiling for 5 min. The samples ~ere centrifuged were and 20 ~1 ~liquot~
analyzed on 5 percent/7 percent SDS polyacrylamide gels (Lae~mli, 3~ *Trade-mark ~ L03x08.mdh - . , , ' ~ '.
' ,- '- '~ '' 1328~1~

1970, nNature" ~ 680-685) The patterns obtained on SDS-PAGE reducing gels matched the 35S-Met labeling result for those polypeptides that contain tyrosine klnase sequences. Figure 3b (HIR~) shows the insulin-stimulated autophosphorylation of the human insulin receptor - subunit above the endogenous COS-7 control. In ~his case the insulin induction effect is strong, although the ~isualized signal is weak due to the ATP concentration (100 ~M) required by the o insulin receptor ~inase. In contrast, only picomolar concentra-tions are needed to measure EGF receptor kinase activity and EGF
6timulation, ~s shown in the A431 cell EGF receptor control (A431). The characteristics of the chi~eric receptor molecule IER
reflects the presence of the EGF receptor kinase because of the low ATP concentrations required. Since ligand induction increases the Vmax of the kinase, maximal induction ( 4 fold) is observed in a 30 second reaction At 4-C (Figure 3c). This finding is in good correlation vith the kinetic properties of the wild type EGF
receptor (Staros et ~1., 1985, in Holecular Aspects of Cellular Regulation Vol.4: Molecular Hechanisms of Transmembrane Signalin~) . and indicates that the EGF receptor kinase retains its original characteristics when controlled by the insulin binding domain.

Surprisingly, the insulin-stimulated and unstimulated 130 kd phosphoprotein subunits of the IE receptor hybrid display a subtle but reproducible size difference (Figure 3d). This observation raises the possibility that ligand-induced enzy~atic acti~ity leads to phosphorylntion of tyrosine residues not 20dified at basal levels; a subsequent confo D ational change of the cytoplasmic receptor domain could alter migration charac-teristics in SDS gels. A similar change may occur in the intact EGF receptor but has not been detected due to the larger size of the ~onomeric 170 kd glycoprotein. Electrophoretic migration changes have been reported for other autophosphorylated proteins L03xO3.mdh . -31- 1328419 such as Ca2+/Calmodulin-dependent protein kinase (Kuret et al., 1985, ~J. Biol. Chem.~ ~Q:6427-6433), type II cAMP-dependent protein kinase (Hemmings ~ ~1., 1981, ~Eur. J. Biochem.~ 112:443-451).

Since uncleaved chimeric proreceptor IER displays insulin-stimulated sutophosphorylation (Fig. 3b and 3c, IER + gel, top band), the tertiary ctructure necessary for insulin binding and ~ignal trsngduction must be formed prior to insulin receptor proteolytic procegsing, consictont with previous reports (Blackshear et ~1., 1983, ~FEBS" ~ 243-246; Rees-Jones et al., 1983, ~Biochem. Biophys. Res. Comm. 116:417-422). Our experiments with the chimeric construct IER, in which the extracellular portion of the insulin receptor ~ subunit ~nd the proreceptor lS cleavage site are deletet, (Fig. 3b, IER coDpare ~ and -) indicate that despite the apparent ability of the resulting 180 kd single-chain glycoprotein to bind insulin, insulin activation of the cytoplasmic kinase domain is lost.

As ~hown above, insulin regulates the rste of the EGF
receptor autophosphorylation activity at subpicomolar concentration~ of ~TP, conditions ~nder whic~ the phospho-tran6ferase of the ~nsulin roceptor ic inactive. Hormone control was only obcerved for the hybrid IER contalning the complete extracellular portion of the insulin receptor, including the ~ignal for receptor processing into the ~ and ~ subunits and the amino terminus of the ~ suBunit. The receptor appears to be processed in our expression system. In the case of the chimera IER lacking any portions of the ~ subunit and consequently the cleavage signal, no hormone effect was observed. We conclude that this structural difference between IER and IER has B profound effect on the structure of the chimeric receptor that is crucial for signal transduction.

L03x08.mdh ' ~ample 5 Construction of Vector Encodin~ a Rece~tor-Onco~ene Hvbrid ~HER-erbB~

We constructed a hybrit receptor comprised of the intracellular domain of the v-erbB onco~ene product fused to the extracellular and transmembrane domains of the EGF receptor (HER-orbB; Fig. 4).

0 The hybrid receptor is expressed from a plasmid under the control of the early promoter of SV40. This plasmid also contains a mutant DHFR gene for methotrexate (MTX) resistance. Selection i8 accomplished by cotransformation with a plasmid encoding a neomycin resistance gene and the DNA amplified by selection in MTX-containing culture media.

~ HER-A64 (~llrich ot al.) was digested with SacI and NsrI
and a restriction fragment coding for the complete extracellular and transmembrane domain of the EGF receptor was recovered. A 1.7 kb ~h~II-StuI restriction fragment coding for the complete intr~cellular portion of AEV-erbB (H) (Yamamoto, T. et al., 1983 ~Cell~ 35:7l-73) was ligated together with the EGF fragment into a p~Cl2 plasmid opened w~th SacI and ~c_I. The recombinant plasmid ~ns amplified in ~. ~Ql~ HB101 and tho coding region for the complete chimeric roceptor is removed in a 3.7 kb ~g~
restriction fragment, both sites being located in the untranslated regions of the EGF receptor and v-erbB sequence, respectively.

p342E (Crowley et al., 1983, "~ol. Cell. Biol." 3:44-55) was di~es~ed with EcoRI and the opened plasmid recovered. An adaptor having the sequence EcoRI SacI EcDRI
GAATTCGAGCTC
CTCGAGCTTAAG

L03xO8.mdh 33 1~28~19 is ligated with the opened plasmid, the 11gation mixture transfected into ~ coli 294, and plasmid pCVSVE-HBS having the adaptor insert is recovered from sn ampicillin resirtant colony.
s pCVSVE-HBS is partially digested with ~I snd the linearized vector fragment ~I) recovered. The linearized plasmid is digested with ~E~I and the vector fragment recovered.

0 The pCVSVE-HBS vector fragment is ligated to the ~
~E~I fragment oncoding the hybr~d receptor and expression vector pCVSV-HER-erbB was recovered from a trsnsformed E. ~Qli ~B101 colony.

E~s~Dle 6 E~Dression of Rece~tor-Onco~ene H~brid Expression ~ector pCVSVE-HER-erb8 is cotransfected into normal Rat 1 fibroblasts together with a neomycin resistable expression plasmid by calcium phosphate copreclpitation based on the protocol of Graham and ~an der Eb (1973). Subconfluent cells were transfec~ed with 10 ~g of plasmid DNA per 8 cm culture dish.
DNA was dissolved ln 0.55 ml of lmM Tris pH 7.5, O.lmM EDTA, 250mM
Cacl2 and ~.5 ml of 50mM HEPES pH 7.12, 280mM NaCl, 1.5mM Na2HP04 WBS slowly added. A precipitate gradually formed within 40 min.
which was added to the 10 ml of cell culture medium. 5h after transfection, cells were sub~ected to a glycerol shock treatment by incubation in 3 ml of 20 percent glycerol in PBS for 1 min.
The glycerol was washed off and the cells were further cultured in the original medium.

The neomycin resistance gene under the control of the SV40 early promoter was used as a selectable marker. Medium supplemented with 400 ~g/ml Geneticin (Sigma G5013) was used for L03x08.mdh ~34~ 1328~19 selection starting two days after transfection. Neomycin-resistsnt cell~ were then grown in medium supplemented with a 200 nM, then 1000 nM concentration of methotrexate (Sigma A6770) containing 7 percent dialyzed $etsl bovine serum. The result was S a step-wise amplification of cDNA expression in neomycin resistant cell lines.

Expression of the hybrid in the transfor~ants was first monitored by analyzing proteins metabolically labeled with 35S-0 methionine after immunoprecipitation of detergent lysates with a human EGF receptor-specific mouse monoclonal antibody Rl. Stably expressing cell lines were metabolically labelled with 35S-methionine. Specific proteins were lmmunoprecipitated by protein A-Sepharose adsorbed Rl antibody from detergent lysates prepared as described above and analyzed on SDS reducing polyacrylamide gels. Ths mouse monoclonal Rl antibody (Waterfield et al., 1982, ~J,Cell.Biochem.~ 20:149-161) does not recognize the endo~enous Ratl cell EGF receptor. The HER-~ protein was readily detected ~n immunoprecipitates before and after methotrexate amplification.
- 20 As expected, the HER-~g~ protein was smaller than the wild type EGF receptor expressed in A431 cells. When compared with the very high level of EGF receptor expressed in A431 cells, amplified RatI
cells expressed only 3-fold less HER-erbB.

The hybrid HER-erbB protein displayed specific EGF binding since 1251-EGF at various concentrations was bound to transformant cells. The binding was saturable and could be completely displaced in the presence of a 100-fold excess of unlabelled EGF.
The amount of binding of 125I-labeled EGF to confluent Ratl cultures corresponded precisely to the amount of EGF receptor or chimeric receptor expressed by the respective cell cultures. Thus the constructed proteins contained fully functional EGF binding domains and were faithfully transported to the cell surface.

L03xO~.mdh ` . -35~ 132%~19 E~ample 7 EGF-stimulated in vitro autophos~hor~latlon To test whether HER-erbB possessed in vitro utophosphory-lation activity, cell lysates were immunoprecipitated as described above, incubated with 32P-~-ATP and analyzed by polyacrylamide gel electrophoresis and autoradiography: Transformant cell monolayer~
grown in 8 cm culture dishes were washed twice ~ith PBS and solubilized as described by Kris et al. ~Cell" 40:619-625 (1985).
0 One ml of 50 mM HEPES pH 7.5, 150mM ~aCl, 1.5mM MgC12, lmM EGTA, 10 percent glycerol, 1 percent Triton X-100, 1 percent Aprotinin and 4~g/ml phenylmethylsulfonyl fluoride (PNSF) was added to the monolayer6 at 4-C for 5 min. Solubilized cells were centrifuged at lO,OOOg for 5 min at 4-C, and the supernatant was either ~tored at -70-C or processed further.

EGF stimulation of autophosphorylation was induced by ~ncubating the detergent cell lysates diluted to a 0.5 percent TX-100 concentration ~n 0.4 ml prior to the immunoprecipitation, with

5 ~g/ml EGF for 15 min at 4-C. Rl antibody prebound for 30 min to protein ~-Sepharose was added (1 ~1 antibody/50 ~1 slurry 1:1), and th- incubation continued for 15 ~in at 4-C. The - ~mmunoprecipitate6 were washed 5 times in 0.9 ml HNTG buffer (20mM
- HEPES pH 7.5, 150~M NaCl, 10 porcent glycerol, and 0.1 percent Triton X-100). The vashed ~mmunoprecipitctss, in a volume of 30 yl, were d~usted to 5mM MnC12 and 15~Ci of ~-32P-ATP was added for 0.5 min at 4-C. The autophosphorylation reaction was stopped by adding 20 ~1 of 3 times concentrated SDS sample buffer.
Samples were boiled for 5 min, centrifuged, and 20 yl aliquots analyzed on 5 percent/7 percent SDS polyacrylamide reducing gels (Laemmli, 1970). Gels were fixed and dried under VACUUm at 79~C.
Normal RatI fibroblasts were used as a control. Size markers are indicated in kilodaltons. Like the wild type EGF receptor, the HER-erbB hybrid incorporated significant amounts of 32p in 3~
L03x08.mdh . -36- 1328419 immunoprecipitates. The extent of phosphoryl~tion was increased by the addition of EGF (designatet by ~); when measured at 30 seconds, the rate of phosphorylation was found tc be 3 fold higher in the presence of EGF. The v-ç~ protein itself possesses only S very low autophosphorylation activity (Lax et al., 1985, "EMB0 Journal" g:3179-3182). Despite the low autophosphorylAtion activity claimed in the hybrid, reconstitutlon of the EGF binding tomain led to ligand-inducible autophosphorylation activity.

L03x08.mdh

Claims (22)

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

1. A hybrid receptor for a ligand comprising (a) a ligand binding domain of a predetermined receptor fused at its C-terminus to the N-terminus of (b) a reporter polypeptide which is (1) heterologous to the receptor from which is obtained the ligand binding domain and which (2) undergoes a conformational change upon the binding of said ligand to the ligand binding domain.

2. The hybrid receptor of claim 1 wherein the ligand binding domain is comprised by the extracellular domain of the predetermined receptor.

3. The hybrid receptor of claim 1 wherein the reporter polypeptide is a cytoplasmic domain of a receptor or oncogene.

4. The hybrid receptor of claim 1 wherein the reporter polypeptide is an enzyme.

S. The hybrid receptor of claim 4 wherein the enzyme is not sterically inhibited by the binding of a ligand to the hybrid receptor.

6. The hybrid receptor of claim 4 wherein the enzyme is a phosphorylkinase.

7. The hybrid receptor of claim 1 having a transmembrane domain interposed between the ligand binding domain and the heterologous reporter polypeptide.

8 . Nucleic acid encoding a hybrid receptor for a ligand, which receptor comprises (a) a ligand binding domain of a predetermined receptor fused at its C-terminus to the N-terminus of (b) a reporter polypeptide which is (1) heterologous to the receptor from which is obtained the ligand binding domain and which (2) undergoes a conformational change upon the binding of said ligand to the ligand binding domain.

9. The nucleic acid of claim 8 further comprising a replicable vector.

10. The vector of claim 9 further comprising a host cell.

11. A method for making a hybrid receptor for a ligand which receptor comprises (1) the ligand binding domain of a predetermined receptor fused at its C-terminus to the N-terminus of (2) a heterologous reporter polypeptide which undergoes a conformational change upon the binding of said ligand to the ligand binding domain, said method comprising (a) transforming a host cell with a vector containing nucleic acid encoding the hybrid receptor operably linked to a promoter for controlling the transcription of the hybrid receptor; and (b) culturing the host cell under conditions for expressing the hybrid receptor.

12. The method of claim 11 wherein the hybrid receptor is recovered from the culture medium of the host cell.

13. The method of claim 11 wherein the hybrid receptor is recovered from the cell membrane of the host cell.

14. A method for assaying a biologically active ligand or an antagonist or agonist for said ligand, comprising (a) providing a hybrid receptor which comprises (1) a binding domain for the ligand, antagonist or agonist and (2) a heterologous reporter polypeptide;
(b) incubating the receptor with a test sample suspected to contain the ligand, antagonist or agonist;
(c) detecting a change in the reporter polypeptide; and (d) correlating said change with the presence of the ligand, antagonist and agonist in the test sample'.

15. The method of claim 14 wherein the ligand is a polypeptide and the ligand binding domain is not the antigen binding site of an immunoglobulin.

16. The method of claim 14 wherein the change in the reporter polypeptide is a modification of the enzymatic activity of the polypeptide.

17. The method of claim 17 wherein the change in the reporter polypeptide is autophosphorylation of the reporter polypeptide.

18. The method of claim 14 wherein the test sample is suspected to contain an antagonist and the receptor is incubated with the test sample and a predetermined activity of ligand or ligand agonist.

19. The method of claim 14 wherein the change in the reporter polypeptide is a change in an immune epitope.

20. The method of claim 20 wherein the change in immune epitope is detected by incubating the receptor with an antibody capable of binding to the reporter polypeptide and determining the amount of bound or residual unbound polypeptide.

21. The method of claim 14 wherein the reporter polypeptide further comprises a stable free radical, fluorescent or chemiluminescent group and the change in the reporter polypeptide is detected by measuring a change in the rotational moment of the stable free radical or a change in the intensity, wavelength or polarization of the fluorescent or chemiluminescent group.

22. The method of claim 14 wherein the reporter polypeptide is capable of binding to a G protein.