EP0862614A1 - Yeast receptor and g-protein fusion protein - Google Patents

Abstract

The invention provides protein fusions between the C-terminus of heterotrimeric G-protein-coupled receptors and the N-terminus of either wild type or mutant G-alpha proteins of the yeast Saccharomyces cerevisiae. Methods are described for creating DNA constructs that encode such fusion protein, assays for correct expression of such fusion molecules in yeast, and assays for the coupling of such fusion molecules to the pheromone-induced signal transduction pathway of yeast. Furthermore, the invention encompasses yeasts expressing the fusion proteins and methods for screening compounds for activity as agonists or antagonists of seven-transmembrane receptor function.

Description

YEAST RECEPTOR AND G-PROTEIN FUSION PROTEIN

FIELD OF THE INVENTION

This invention covers protein fusions between the C- terminus of any G-protein coupled receptor and the N- terminus of the Saccharomyces cerevisiae G-alpha protein Gpalp, the DNA constructs encoding the same, yeast strains expressing the same, methods to ensure that the fusion protein is coupled to the yeast mating pathway, and assays for such coupling.

BACKGROUND OF THE INVENTION

Papers of the scientific periodical and patent literature, and data archived in GENBANK by accession number, referred to herein throughout the text are hereby incorporated in their entirety by reference.

Cell surface receptors recognize extracellular ligands such as hormones, nutrients and growth factors, and transduce the signal generated by ligand binding to effector molecules within the cell . An important class of these receptors, variously called G-protein-coupled receptors, seven transmembrane domain receptors or serpentine receptors, is characterized by their interaction with heterotrimeric G-protein complexes comprised of alpha, beta and gamma subunits (Watson and Arkinstall, The G-Protein Linked Receptors Facts Book, c. 1994 by Academic Press) .

Activation of such receptors leads to dissociation of beta and gamma subunits from the alpha subunit, and consequent initiation of signaling cascades in the cell by the dissociated components. Mammalian receptors of this class include the alpha- and beta-adrenergic, muscarinic cholinergic, cannabinoid, dopamine, opiate, serotonin, thrombin, platelet activating factor and thromboxane A2 receptors. Agonists and antagonists of several of these receptors are important therapeutic agents, and many members of this class of receptors are involved in various disease processes. Therefore, there is a need in the pharmaceutical industry for assays to identify new agonists and antagonists of these receptors from libraries of small molecules and peptides, for the purpose of new drug development. However, the development of such assays has been hindered by several factors. The expression of many of these receptors is often limited to a specific cell type that is difficult to isolate or culture in quantity. Further, each receptor does not interact with all members of the family of heterotrimeric G-proteins found in mammalian cells (which can include up to 20 alpha, 4 beta and 7 gamma isoforms) , although some receptors interact with more than one G-alpha subunit. For many others, the cognate G-protein complex has not been characterized. On the other hand, the same G-alpha protein or G-protein complex can interact with different receptors expressed on the same cell. Therefore, it is difficult to narrow down the physiologically important interaction in mammalian cell tissue culture. Ligands for many of these receptors have been identified by binding assays using membrane preparations from tissue culture cells or heterologous systems such as insect cells overexpressing the relevant receptor. Ligands identified thus, however, may be agonists, antagonists or neutral in terms of receptor function, since only binding and not activation is measured by these assays. Moreover, even binding assays cannot be used to study the so-called "orphan" receptors, which were identified by DNA homology methods, and whose physiological ligands and functions are unknown. Finally, these proteins traverse the membrane seven times, giving rise to one free end and three loops on both sides of the membrane. Potentially, all three loops and the end could contribute to forming the ligand binding pocket on the outside and the recognition site for G-proteins on the inside. These factors render it difficult to study these protein by X-ray crystallography and molecular modeling. In addition, dividing these proteins into domains of specific function that can be analyzed separately, either by proteolysis or expression of gene fragments, is not feasible because of the loops. This also renders this class of receptors less suited to rational drug design. Therefore, there is a need in the art for new and convenient assays to identify agonists and antagonists of these receptors.

The yeast Saccharomyces cerevisiae has already proven useful in developing such assays. Two endogenous G- protein coupled receptors and one heterotrimeric G- protein complex have been characterized from this organism, all of which are involved in a developmental pathway leading to the formation of a diploid yeast cell from fusion of two haploid cells of the a and alpha mating type. The two receptors, the a-factor receptor

(encoded by STE3) and the alpha-factor receptor (encoded by STE2) , are expressed respectively on haploid yeast cells of the alpha and a mating type, are activated respectively by the a- or alpha- peptide factors secreted by cells of the opposite mating type, but trigger activation of the same heterotrimeric G-protein complex in both cell types . Activation of the complex releases the beta-gamma subunits, which activate the mating pathway and cause expression of specific proteins that result in growth arrest at the Gl phase, and a morphological change from budded spheroidal cells to unbudded pear-shaped "shmoos" in preparation for mating. The genes involved in this signal transduction pathway in yeast, how they interact to bring about G_! growth arrest in response to mating factor, and their similarity to mammalian signal transduction components (the thrombin pathway is chosen as an example) are represented in Fig. 1. (Jones, Pringle and Broach, The Molecular and Cellular Biology of Yeast Saccharomyces, Vol. 2., c. 1992 by Cold Spring Harbor Laboratory Press) . Activation of heterotrimeric G-proteins requires a specific interaction between the receptor and the G- protein complex that is mediated primarily by the G- alpha subunit. Unactivated receptors are normally bound to a trimeric complex with inactive GDP-bound G-alpha. Receptor activation by the ligand stimulates GDP release from G-alpha followed by GTP binding and the dissociation of the beta and gamma subunits from the alpha subunit. In mammalian cells, this renders both G- alpha and G-beta-gamma "active" and capable of activating downstream signaling elements such as adenylyl cyclase. Hydrolysis of GTP to GDP switches G-alpha back to the inactive state, where it reassociates with G-beta-gamma to regenerate the inactive complex, which then associates with a receptor. In the yeast mating cascade, the entity that propagates the signalling cascade is the released complex of G-beta and G-gamma subunits; however, dissociation of that complex from G- alphais still the crucial activation step. Because G-alpha is the subunit that interacts primarily with the receptor, the affinity of a particular G-alpha for a given receptor largely determines which of the many heterotrimeric complexes in mammalian cells is associated with the receptor, and therefore determines the efficiency of coupling between a receptor and a given G-protein complex. (Conklin and Bourne, Cell .7_3:631 (1993)) Given this, the use of a heterologous systems such as yeast to model the activation of mammalian receptors is limited by the potential lack of interaction between yeast G-alpha and the mammalian receptor. For example, it has been shown the human beta-2-adrenergic receptor (BAR) can be expressed in Saccharomyces cerevisiae such that it is properly folded and located in the yeast plasma membrane, and binds extracellular ligands with affinities comparable to mammalian cells. However, ligand binding did not result in activation of the mating response pathway, indicating that the mating cascade-associated G-protein complex comprising Gpalp, Ste4p, and Stelβp did not respond to BAR activation, possibly because of a lack of recognition between the yeast Gpalp and BAR. Activation was, however, achieved when the cognate human G-alpha protein was co-expressed (King et. al. , Science 250:121-123 (1990)), indicating that the G-beta and G-gamma subunits of yeast could form a heterotrimer with the mammalian G-alpha protein that could respond to BAR activation by release of the beta- gamma complex.

These results indicate that co-expression of the cognate G-alpha subunit would be required to engineer a response of the yeast mating pathway to the activation of a heterologous receptor expressed in yeast. However, the physiologically relevant G-alpha-proteins have not been defined for many mammalian receptors, including the "orphan" types, limiting the applicability of this approach. Furthermore, these G-alpha proteins must bind to the yeast G-beta and G-gamma subunits so that no free beta-gamma complexes exist in the cell. They must be capable of responding to the ligand-binding signal by releasing the beta-gamma complex, and must undergo in the yeast cell the post-translational modifications that are needed for their function. All heterologous G-alpha proteins might not fulfill all of these criteria.

One way to potentially overcome these limitations is to adapt the endogenous yeast G-alpha protein such that it can be coupled to heterologous receptor activation. The invention described here is a means toward such adaptation of the yeast G-alpha protein. A critical and novel feature of our invention is the creation of a covalent linkage between a mammalian receptor and the endogenous yeast G-alpha protein, which is achieved by an in-frame gene fusion between the C-terminal end of the heterologous receptor gene and the N-terminal end of the yeast GPA1 gene. The presence of the yeast G-alpha protein as a linked moiety should greatly increase its local concentration and thus facilitate its interaction with the receptor and its response to activation of the receptor, as shown schematically in Fig. 2. Our invention also provides for the possibility that such facilitation is insufficient to overcome the lack of recognition between the two components. This is achieved by selection schemes used along with mutagenesis of the Gpalp domain of such fusion proteins, thereby identifying mutants in this domain in which activation of the receptor moiety is coupled to activation of the Gpalp moiety, adn therefore to the yeast mating pathway.

Bertin et al (PNAS 91:8827-8831, 1994) have shown that protein fusions between the beta-2-adrenergic receptor (BAR) and its cognate mammalian G-protein, G.-alpha, when expressed in mammalian cells, result in productive signal transduction as measured by ligand-dependent increase in cAMP levels. The cAMP response with the fusion was greater than in controls without the fusion, suggesting that covalent linkage enhances signaling efficiency. The authors suggest two reasons for the higher efficiency. One is that cycling between active and inactive forms of G-alpha may occur more rapidly in the chimera than in the unlinked state. The other is that the presence of the linked G-alpha may impede desensitization of the ligand response either by masking receptor determinants that mediate desensitization or by protecting G-alpha from degradation. However, unlike our invention, the article does not envision the use of the potentiated response to facilitate interactions between components that may not interact or only interact weakly, nor does it envision applications where the receptor and G-alpha protein are from different species.

Our use of receptor-Gpalp gene fusions in this manner is different from the method disclosed in the published PCT application WO92/05244 for modeling G-protein coupled receptors in yeast. That method requires transformation of yeast with two exogenous genes, the receptor gene and the corresponding mammalian G-alpha protein gene, whereas our invention utilizes only the receptor gene, fusing it to a gene encoding a veast G- alpha protein unlike the method disclosed in the WO92/05244 application, our invention is potentially applicable in cases where the G-alpha has either been not identified or does not interact with the yeast G- beta and G-gamma proteins. Published PCT applications WO 94/23025 discloses a method whereby the simultaneous expression of exogenous surrogates of yeast pheromone system proteins and modulators of these surrogates is used to identify peptide inhibitors or activators of the surrogate protein. However, those applications do not consider the use of a fusion protein, which is the basis of the present invention. Besides, the single fusion protein in our approach is not a surrogate of any individual yeast pheromone system protein but is simultaneously a surrogate of two distinct individual components. U.S. patent 5,030,576 covers the fusion of the ligand binding domain of a receptor to a reporter polypeptide that undergoes a conformational change upon ligand binding to the binding domain, but the application to G-protein-coupled receptors mentioned in the '576 patent describes the relevant reporter polypeptide as the cytoplasmic domain of such a receptor that is capable of interaction with G-proteins. Similarly, U.S. patent application WO 91/12273 covers hybrid proteins created by replacing domains other than the ligand-binding domain of a G-protein coupled receptor with corresponding domains of a yeast G-protein coupled receptor. In contrast to and as distinct from U.S. patent 5,030,576 and application WO/90 91/12273, our invention discloses a fusion between the full length mature receptor protein and not a fragment thereof with defined properties such as the ligand binding domain and furthermore, uses the G-protein itself and not the cytoplasmic domain of another receptor as the N- terminus. In other words, we describe fusions between two individual proteins from different species, in contrast to the approach commonly referred to in the literature as domain-swapping, where different domains with differing properties of a protein of similar structure from different species are fused together.

SUMMARY OF THE INVENTION

The present invention embodies the idea of using covalent linkage between two proteins created by gene fusion to potentiate their mutual interaction. The invention provides DNA constructs that encode and express a fusion protein with a peptide bond between the

C-terminus of any eukaryotic G-protein-coupled receptor and the N-terminus of the yeast G-alpha protein Gpalp.

The invention further provides yeast strains expressing these fusion proteins. The invention also provides methods to ensure that these fusion proteins are synthesized and localized to the plasma membrane such that the Gpalp domain of the fusion protein can interact with yeast G-beta and G-gamma proteins. The invention further provides methods that can be used to select, from a collection of mutants of the G-alpha domain of such fusion constructs, individual mutants demonstrating coupling of receptor activation to the mating pathway of yeast through the fusion protein. The invention further provides for use of the said strains to identify small molecule agonists and antagonists of these receptors. The invention further provides for use of the said strains to identify peptide agonists and antagonists of receptor activation by transformation with a combinatorial peptide library, which is created by expressing a randomized DNA sequence in yeast such that the individual peptides are secreted into the medium via gene fusions to the signal peptide of the yeast alpha-factor.

DESCRIPTION OF DRAWINGS

Figure 1 shows G-protein signaling pathways in mammals and humans.

Figure 2 shows a map of plasmid pRMHBT4.

Figure 3 shows a map of plasmid pRMHBTlO.

Figure 4 shows a map of plasmid pRMHBT18-NG.

Figure 5 shows a map of plasmid pRMHBT20-NG. Figure 6 shows a map of plasmid pRMHBT26.

Figure 7 shows a map of plasmid pRMHBT41.

Figure 8 shows a map of plasmid pRMHBT43.

Figure 9 shows a map of plasmid pRMHBT44.

Figure 10 shows a map of plasmid pRMHBT45. Figures 11A-11G show the nucleotide sequence encoding the STE2-GPA1 fusion protein and its amino acid sequence. The sequence starts at position 520 in STE2, and extends through position 1850 in GPA1 . GenBank accession numbers for sequences are provided below. The extra amino acids generated at the junction are underlined.

Figures 12A-12G show the nucleotide sequence encodng ThR-GPAl fusion protein and its amino acid sequence. The sequence starts at position 288 in ThrR, and extends through position 1850 in GPA1. GenBank accession numbers for sequences are provided below. The extra amino acids generated at the junction are underlined. The STE2 leader sequence that proceeds and is linked in- frame to the ThrR sequence is also shown below.

DETAILED DESCRIPTION OF THE INVENTION

The following yeast strains are used in experiments constituting working examples disclosed in this application. Table I: Strains Used in the Working Examples

STRAINS GENOTYPE SOURCE

MS16 mat a, ade2-101, trplDl Dr. M. Rose, Princeton Univ.

MS2288 mat a, his3D200, Ieu2-3,U2, trplDl, Dr. M. Rose, Princeton ura3-52 Univ.

HBSIO mat a, ade2-101, farl-x200, his3D200, Heartland BioTechnologies leu2-3,112, lys2DS738, trplDl, ura3-52

HBS10::pFFLZ same as HBSIO except farl ::URA3-FUSlp- Heartland BioTechnologies LACZ

HBS32 mat Λ, farl-x200, his3D200, leu2-3,112, Heartland BioTechnologies trplDl, ura3-52

HBS12 mat a, farl-x200, his3D200, Ieu2-3,U2, Heartland BioTechnologies ste2, trplDl, ura3-S2

HBS12LZ same as HBS12 except leu2::LEU2-FUSlp- Heartland BioTechnologies LACZ

TMHY2-14A mat a, ade2-101, his3D200, lys2DS738, Heartland BioTechnologies trplDl, ura3-52

TMHY2-223D a/a, ADE2lade2, FARllfarl, his3lhis3, Heartland BioTechnologies LEU2lleu2, Iys2/lys2, trplltrpl, ura3lura3

TMHY3D a/a, ADE2/ade2, FARllfarl, Heartland BioTechnologies GPAl/gpal.'.TRFl, his3/his3, LEU2/leu2, Iys2llys2, trplltrpl, ura3lura3

HBS14 same as TMHY3D Heartland BioTechnologies 9A mat a, ade2-101, farl-x200, gpa .TRPl, Heartland BioTechnologies his3D200, leu2-3,112, lys2DS738, trplDl, ura3-52

9ALZ same as 9A except leu2::LEU2-FUSlp- Heartland BioTechnologies LACZ

9ALZΔGS same as 9ALZ except also ste2 Heartland BioTechnologies Gene names are italicized (GPA1) , and are in upper case (GPA1) when indicating a functional and dominant gene, and in lower case (gpal) when indicating a non¬ functional recessive mutant gene. The corresponding proteins are in plain text (Gpalp) . An agonist is defined as a molecule that binds to a receptor protein, and activates the receptor by inducing conformational or other changes in it such that the heterotrimeric G- protein complex that is bound to the receptor is disrupted, leading to release of the beta-gamma complex from the alpha subunit.

This invention embodies the idea of using covalent linkage between two proteins created by gene fusions to potentiate the interaction between G-protein-coupled receptors from other species and a protein homologous in function to the Gpalp protein of the yeast Saccharomyces cerevisiae . The experiments of Bertin et al have shown that there is a potentiation of the downstream response to receptor activation when the human beta-2-adrenergic receptor and its cognate G-alpha protein are linked in this manner. In addition to the two reasons considered by the authors which are described above, we consider that potentiation could also result from: a) more efficient coupling (which is considered in present models as transmission of a conformational change in the receptor to the G-protein complex) due to proximity of the interacting molecules brought about by covalent linkage; b) more efficient coupling because of the great increase in local concentration of Gpalp brought about by covalent linkage, thereby overcoming the effects of an unfavorable equilibrium binding constant for a heterologous receptor and yeast Gpalp; c) the presence of stoichiometric amounts of the two components, leaving no molar excess of either component to dilute the effects of ligand-mediated activation; d) better membrane anchoring of G-alpha by covalent attachment to the receptor compared to the normal situation of anchoring via N-terminal myristoylation, which may be reversible. Regardless of the precise reason, it is likely that covalent coupling ameliorates the lack of recognition specificity between a given mammalian receptor and the yeast G-alpha protein.

The DNA constructions needed for the present invention can be made in vectors that can replicate independently in yeast cells, including the YCp or the YEp class of vectors or in vectors that are designed for integration into the yeast chromosome such as the YIp class. Most preferred vector are those which autonomously replicate in yeast.

G-protein-coupled receptors used in the present invention may be from animal species, including both vertebrates and invertebrates, plants or fungi other than S. cerevisiae . Preferred receptors are those from mammals, especially humans. Also, preferred are receptors from fungi, especially fungi that are pathogenic to humans. Mammalian receptors of this class that are encompassed by the present invention include, but are not limited to the following, whose nucleotide sequences are disclosed in the listed references:

1. Adenosine receptor Al: Libert F. et al, Biochem. Biophys. Res. Commun. 187:919 (1992) . 2. Adenosine receptor A2B: Pierce K. D. et al, Biochem. Biophvs. Res. Commun. 187:86 (1992) .

3. Adrenergic receptor alpha-lA: Bruno, J. F. et al, Biochem Biophvs. Res. Comm. 179: 1485 (1991) .

4. Adrenergic receptor alpha-IB: Ramarao, C. S. et al, J. Biol. Chem. 267:21936 (1992) .

5. Adrenergic receptor alpha-2A: Kobilka B. K. et al, Science 238:650 (1987) .

6. Adrenergic receptor alpha-2B: Weinshank et al, Mol. Pharmacol. 38:681 (1990) . 7. Adrenergic receptor alpha-2C: Regan, J. W. et al, Proc. Natl. Acad. Sci. USA 85:6301 (1988) . 8. Adrenergic receptor beta-1: Frielle, T. et al Proc. Natl. Acad. Sci. USA 84:7920 (1987)

9. Adrenergic receptor beta-2: Kobilka B. K. et al, Proc. Natl. Acad. Sci. USA 84:46 (1987) . 10. Adrenergic receptor beta-3: Emorine, L. J. et al, Science 245:1118 (1989) .

11. Amyloid protein precursor: Kang, J. et al, Nature 325:733 (1987) .

12. Angiotensin II receptor type 1: Furuta H. et al, Biochem. Biophys. Res. Commun. 183:8 (1992) .

13. Antidiuretic hormone receptor: Birnbaumer, M. et al, Nature 357:333 (1992) .

14. Bradykinin receptor: Hess J. -F. et al, Biochem. Biophvs. Res. Commun. 184:260 (1992) . 15. Cannabinoid receptor: Gerard C. M. et al, Biochem. J. 279:129 (1991) .

16. Chemokine C-C (mip-1/RANTES) receptor: Neote K. et al, Cell 72:415 (1993) .

17. Cholecystokinin receptor: Pisegna J. R. et al, Biochem. Biophvs. Res. Commun. 189:296 (1992) .

18. Complement C5a (anaphylotoxin) receptor: Boulay F. et al, Biochemistry 30:2993 (1991) .

19. Dopamine receptor Dl: Dearry A. et al, Nature 347: 7276 (1990) . 20. Dopamine receptor D2 : Grandy D. K. et al, Proc. Natl. Acad. Sci. USA 86:9762 (1989) .

21. Dopamine receptor D3 : Sokoloff P. et al, Nature 342:146 (1990)

22. Dopamine receptor D4 : van Tol, H. H. et al, Nature 350:610 (1991) .

23. Dopamine receptor D5 : Weinshank R. L. et al, ___ Biol. Chem. 266: 22427 (1991) .

24. Endothelin receptor A: Hayzer D. J. et al, Am. J. Med. Sci. 304:231 (1992) . 25. Endothelin receptor B: Nakamuta M. et al, Biochem. Biophvs. Res. Commun. 177:34 (1991) . 26. f-met-leu-phe receptor: Boulay F. et al, Biochemistry 29:11123 (1990) .

27. Follicle stimulating hormone receptor: Minegish T. et al, Biochem. Biophvs. Res. Commun. 175:1125 (1991) . 28. Glutamate receptor, metabotropic: Tanabe Y. et al, Neuron 8:169 (1992) .

29. Gonadotropin-releasing factor receptor: Chi L. et al, Mol. Cell. Endocrinol. 91.R1 (1993) .

30. Growth hormone releasing hormone (GHRH) receptor: Mayo K. E. Mol. Endocrinol. 6:1734 (1992) .

31. Histamine H2 receptor: Gantz, I. et al, Biochem. Biophvs. Res. Commun. 178:1386 (1991) .

32. Hydroxytryptamine (serotonin) receptor IA: Kobilka B. K. et al, Nature 329:75 (1987) . 33. Hydroxytryptamine (serotonin) receptor IB: Jin H. et al, J. Biol. Chem. 267: 5735 (1992) .

34. Hydroxytryptamine (serotonin) receptor IC:

Saltzman A. G. et al, Biochem. Biophvs. Res. Commun.

181: 1469 (1991) . 35. Hydroxytryptamine (serotonin) receptor ID:

Hamblin M. W. and Metealf M. A., Mol . Pharmacol .

4TJ.143 (1991) .

36. Hydroxytryptamine (serotonin) receptor IE: McAllister G. et al, Proc. Natl. Acad. Sci. USA 8_9:5517 (1992) .

37. Hydroxytryptamine (serotonin) receptor 2: Yang W. et al, Soc. Neurosci. Abstr. 17: 405 (1991) .

38. Insulin-like growth factor (IGF II) receptor: Kiess, W. et al, J. Biol. Chem. 263:9339 (1988) . 39. Interleukin 8 receptor A: Holmes W. E. et al, Science 253: 1278 (1991) .

40. Interleukin 8 receptor B: Murphy P. M. and Tiffany H. L. Science 253 : 1280 (1991) .

41. Lutenizing hormone/chorionic (LH/CG) gonadotropin receptor: Minegish T. et al, Biochem. Biophvs. Res.

Commun. 172:1049 (1990) .

42. mas proto-oncogene: Young D. et al, Cell 45:711 ( 1986 ) .

43. Melanocyte stimulating hormone (MSH) receptor: Chhajlan, V. and Wickberg, J. E. FEBS Lett. 309:417

(1992) . 44. Muscarinic acetylcholine receptors ml, m2, m3 and m4 : Bonner T. I. et al, Science 237:527 (1987) ; Peralta E. G. EMBO J. 6:3923 (1987) .

45. Muscarinic acetylcholine receptors m5: Bonner T. I. et al, Neuron 1:403 (1988) . 46. Neuropeptide Y receptor: Herzog, H. et al, Proc. Natl. Acad. Sci. USA 89:5794 (1992) .

47. Opioid, delta, receptor: Evans C. J. et al, Science 258:1952 (1992) .

48. Opioid, kappa, receptor: Xie G. -X. et al, Proc. Natl. Acad. Sci. USA 89: 4124 (1992) .

49. Oxytocin receptor: Kimura T. et al, Nature 356:526 (1992) .

50. Platelet activating factor receptor: Kunz D. et al, J. Biol. Chem. 267:9101 (1992) . 51. Rhodopsin and related pigments: Nathans J. and Hogness D. S. Proc. Natl. Acad. Sci. USA 81:4851 (1984) ; Nathans J. and Hogness, D. S. Science 232:193 (1986) .

52. Somatostatin receptor 3: Yamada Y. et al, Mol. Endocrinol. 6:236 (1992) .

53. Somatostatin receptors 1 and 2: Yamada Y. et al, Proc. Natl. Acad. Sci. USA 89:251 (1992) .

54. Substance K (neurokinin A) receptor: Gerard N. P. et al, J. Biol. Chem. 265: 20455 (1990) . 55. Substance P (NK1) receptor: Gerard N. P. et al, Nature 349: 614 (1991) .

56. Thrombin receptor: Vu T. K. et al, Cell 64:1057 (1991) .

57. Thromboxane A2 receptor: Hirata M. et al, Nature 349:617 (1991) .

58. Thyroid stimulating hormone (TSH) receptor:Nagayama Y. et al, Biochem. Biophvs. Res. Commun. 165:1184 (1989) .

59. Vasoactive intestinal peptide receptor:

Sreedharan S. P. et al, Proc. Natl. Acad. Sci. USA -4986 (1991) . The present invention can also be practiced using any of the seven-transmembrane receptors encoded by nucleotide sequences presently deposited in GENBANK under the accession numbers listed in Table II :

Table II: List of Receptor Nucleotide Sequences

Human acetylcholine m5 muscarinic receptor, 2261bp M80333

Human acetylcholine muscarinic receptor, 2098bp M35128 Human activin type I receptor, 1518bp U14722 Human activin type II receptor, 2382bp M93415 Human adenosine receptor (Al) 2900bp L22214 Human adenosine receptor (Al) brain hippocampus, 1267bp S45235 Human adenosine receptor (A2) 2383bp M97370

Human adenosine receptor (A2) , brain hippocampal,

2572bp S46950

Human adenosine receptor (A2b) 1687bp M97759

Human adenosine receptor (A3) 1739bp L22607 Human adenosine receptor (A3) 1767bp L20463

Human adrenergic alpha la receptor 1860bp U03864 Human adrenergic alpha la receptor 2002bp M76446 Human adrenergic alpha la/d receptor 1831bp L31772 Human adrenergic alpha lb receptor 1560bp L31773 Human adrenergic alpha lb receptor 1738bp U03865 Human adrenergic alpha lc receptor 1401bp L31774 Human adrenergic alpha lc receptor 1500bp U03866 Human adrenergic alpha lc receptor 1902bp U02569 Human adrenergic alpha lc receptor 2290bp D25235 Human adrenergic alpha 2 receptor gene 3604bp M23533 Human adrenergic alpha 2 receptor kidney 1491bp J03853 Human adrenergic alpha 2 receptor platelet 1521bp M18415 Human adrenergic alpha 2c2 receptor 2072bp M34041 Human adrenergic alpha 2cII receptor 1382bp D13538 Human adrenergic beta 1 receptor 1723bp J03019 Human adrenergic beta 2 receptor 3451bp M15169 Human adrenergic beta 2 receptor 3458bp J02960 Human adrenergic beta 2 receptor, 2305bp Y00106 Human adrenergic beta 3 receptor, 1270bp M29932

Human adrenergic beta receptor, brain, 1970bp X04827 Human AH receptor 5,228bp L19872

Human angiotensin II type 1 receptor 1575bp M93394 Human angiotensin II type 1 receptor 2254bp M87290 Human angiotensin II type lb receptor 1563bp D13814 Human angiotensin II type 2 receptor (AGTR2) L34579 Human angiotensin II type 2 receptor 5,293bp U20860 Human angiotensin II type 2 receptor 1092bp U15592 Human angiotensin II type 2 receptor 1439bp U16957 Human angiotensin II type 2 receptor, 2476bp U10273 Human angiotensinogen II type-lA receptor 1829bp M91464

Human antidiuretic hormone receptor V2 (AVPR2) U04357 Human arginine vasopressin receptor 1 (AVPRl) 6,402bp U19906 Human arginine vasopressin receptor 1 (AVPRl) 1472bp L25615

Human arginine vasopressin receptor type II, U04357 Human atrial natriuretic peptide clearance receptor (ANP C- receptor) M59305 Human autocrine motility factor receptor (Ngp78) 1765bp L35233

Human B-cell antigen receptor (MB-1) 681bp M74721 Human bombesin receptor subtype-3, 1413bp L08893 Human bradykinin Bl receptor 1082bp U12512 Human bradykinin Bl receptor, 4168bp U22346

Human bradykinin BK-2 receptor, 1378bp M88714 Human C5a anaphylatoxin receptor, 2328bp M62505 Human calcitonin receptor 3588bp L00587 Human calcitonin-like receptor, 2187bp U17473 Human cannabinoid receptor, 1755bp X54937

Human cannabinoid receptor, central, long isoform, 2135bp X81120

Human cannabinoid receptor, central, short isoform, 1252bp X81121 Human cannabinoid receptor, peripheral (CB2) 1790bp X74328 Human chemokine C-C receptor type 1 1495bp L09230 Human cholecystokinin A receptor, 1393bp L13605

Human cholecystokinin A receptor, 1686bp L19315

Human cholecystokinin B/gastrin receptor brain, I344bp

L08112 Human cholecystokinin receptor, 1969bp L04473

Human ciliary neurotrophic factor alpha receptor

L38025

Human ciliary neurotrophic factor receptor (CNTFR)

1566bp M73238 Human corticotropin releasing factor receptor, 1285bp

L23332

Human corticotropin releasing factor receptor, 1335bp

L23333

Human corticotropin releasing hormone receptor, 1146bp U16273

Human CR2/CD2l/C3d/Epstein-Barr virus receptor, 3934bp

M26004

Human CTLA4 counter-receptor (B7-2) , 1112bp L25259

Human dopamine DIA receptor, 2337bp M85247 Human dopamine D2 receptor (DRD2) , 2482bp M29066

Human dopamine D2 receptor, 1756bp M30625

Human dopamine D3 receptor (DRD3) gene, 1727bp U25441

Human dopamine D5 receptor (DRD5) gene, 1673bp M67439

Human EBV induced G-protein coupled receptor (EBI2) 1643bp L08177

Human EBV induced G-protein coupled receptor 2154bp

L08176

Human endothelial cell protein C/APC receptor (EPCR)

1284bp L35545 Human erythropoietin receptor, 1624bp M34986

Human erythropoietin receptor, 1818bp

Human Fc receptor low affinity CD16 (FcGRIII) , 1326bp

M24854

Human Fc-gamma receptor I Al, 1128bp L03418 Human Fc-gamma receptor I Bl, 846bp L03419

Human Fc-gamma receptor I B2, 570bp L03420

Human Fc-gamma-R receptor leukocyte, 1977bp J04162 Human Fc-gamma-receptor IIA (FCGR2A) M90727 Human Fc-gamma-receptor IIIB(FCGR3B) M90746 Human FMLP-related receptor II (FMLP R II) 1058bp M76672 Human folate receptor 3 819bp U08471

Human follicle stimulating hormone receptor, 2186bp

M95489

Human follicle stimulating hormone receptor, 2393bp

M65085 Human formyl peptide receptor (FPR2) , 1650bp M88107 Human formyl peptide receptor-like receptor (FPRL1) 2631bp M84562

Human G protein coupled-receptor (GPR12) , 1230bp U18548 Human G protein-coupled receptor (APJ) 1583bp

Human G protein-coupled receptor (EBI1) , 2139bp L31581 Human G protein-coupled receptor (EBI1) , 2215bp L31584 Human G protein-coupled receptor (GPR1) 1438bp L35539 Human G protein-coupled receptor (GPR1) 1438bp U13666 Human G protein-coupled receptor (GPR19) 2932bp U21051 Human G protein-coupled receptor (GPR3) 1262bp L32831 Human G protein-coupled receptor (GPR3) 3542bp U18550 Human G protein-coupled receptor (GPR4) 1365bp L36148 Human G protein-coupled receptor (GPR5) 1265bp L36149 Human G protein-coupled receptor (GPR6) 1477bp L36150 Human G protein-coupled receptor (GPR6) 2699bp U18549 Human G protein-coupled receptor (V28) 3100bp U20350 Human G-binding regulatory protein-coupled receptor, M28269 Human galanin receptor, 1050bp U23854 Human galanin receptor, 1053bp L34339 Human gastrin receptor gene, 4754bp L10822 Human gastrin releasing peptide receptor (GRP-R) 1726bp M73481 Human glucagon receptor, 1578bp U03469 Human glucagon receptor, 2034bp L20316 Human glucagon-like peptide-1 receptor (GLP-1) I567bp L23503

Human glucagon-like peptide-1 receptor, 1590bp U10037

Human glucagon-like peptide-1 receptor, 2431bp U01156

Human glucagon-like peptide-1 receptor, 2616bp U01104 Human glutamate receptor (GLUR5) 3188bp L19058

Human glutamate receptor (HBGR1) 2946bp M81886

Human glutamate receptor 2 (HBGR2) 333lbp L20814

Human glutamate receptor flip (GluR3-flip) 3056bp

U10301 Human glutamate receptor flop (GluR3-flop) 2747bp

U10302

Human glutamate receptor metabotropic subtype 5a,

4518bp D28538

Human glutamate receptor metabotropic subtype 5b, 4614bp D28539

Human gonadotropin releasing hormone receptor, 154lbp

L03380 Human gonadotropin releasing hormone receptor, 2160bp

L07949 Human growth hormone-releasing hormone receptor,

1617bp L01406

Human heat-stable enterotoxin receptor, 3745bp M73489

Human histamine HI receptor, 1654bp D28481

Human histamine H2 receptor, 1191bp M64799 Human interleukin 8 low affinity receptor, 1510bp

M73969

Human interleukin 8 receptor alpha (IL8RA) 2007bp

L19591

Human interleukin 8 receptor B, 1750bp M94582 Human interleukin 8 receptor beta (IL8RB) 2856bp

L19593

Human interleukin 8 receptor type A (IL8RBA) gene,

4452bp U11870

Human interleukin 8 receptor, 1933bp M68932 Human leukemia virus receptor 1 (GLVR1) , 3220bp L20859

Human leukemia virus receptor 2 (GLVR2) , 3175bp L20852

Human luteinizing hormone-choriogonadrotropin receptor, 2995bp M63108

Human lymph node homing receptor, 2354bp M25280 Human macrophage inflammatory protein-1-alpha/RANTES receptor, L10918 Human major group rhinovirus receptor (HRV) 3003bp M24283

Human mannose receptor, 5,185bp J05550 Human melanocortin 4 receptor, 999bp L08603 Human melanocortin 5 receptor (MC5R) , 1262bp L27080 Human melanocortin 5 receptor gene, 1050bp U08353 Human melanocortin receptor, 1650bp Z25470 Human melatonin receptor, 1085bp U14108 Human monocyte chemoattractant protein 1 receptor (MCP-IRA) U03882 Human monocyte chemoattractant protein 1 receptor (MCP-IRB) U03905

Human N-formyl receptor-like 2 protein (FPRL2) 1198bp L14061 Human N-formylpeptide receptor (fMLP-R26) 1281bp M60627

Human N-formylpeptide receptor (fMLP-R98) 1866bp

M60626

Human N-formylpeptide receptor (FPR1) 6,931bp L10820

Human neurokinin 1 receptor (NKIR) 123Obp M76675 Human neurokinin 3 receptor (NK3R) 1755bp M89473 Human neurokinin A receptor (NK-2R) 1197bp M57414 Human neurokinin receptor (NK-1) 1466bp M81797 Human neuromedin B receptor (NMB-R) 1352bp M73482 Human neuropeptide Y peptide YY receptor, 1605bp M88461

Human neuropeptide Y receptor (NPYR) 1225bp

Human neuropeptide Y receptor Yl (NPYY1) 2881bp L07615

Human neuropeptide y receptor, 1470bp M84755

Human nucleotide receptor (P2U) 2030bp U07225 Human opiate delta receptor, 1136bp U10504

Human opiate mu receptor (MORI) 2162bp L25119 Human opioid delta receptor, 1773bp U07882 Human opioid kappa receptor (hKOR) 1154bp U17298

Human opioid kappa receptor (hKOR) 1182bp U11053 Human opioid kappa receptor (0PRK1) 1604bp L37362

Human opioid mu receptor variant (MORI) 1473bp U12569 Human opioid receptor, 1610bp L29301

Human orphan G protein-coupled receptor, 1670bp L06797

Human orphan receptor (TR3) 2464bp L13740

Human orphan receptor (TR4) 2254bp L27586

Human oxytocin receptor, 3617bp X80282 Human oxytocin receptor, 4103bp X64878

Human PACAP receptor, 1664bp D17516

Human PACAP receptor, helodermin-preferring, 1640bp,

L36566

Human parathyroid hormone receptor, 1948bp L04308 Human parathyroid hormone/parathyroid hormone-related peptide receptor, U17418

Human plasminogen activator receptor urokinase-type,

1608bp U08839

Human platelet activating factor receptor (PAFR) 1064bp M76674

Human platelet activating factor receptor (PTAFR)

1467bp M88177

Human platelet activating factor receptor, 1551bp

M80436 Human platelet-activating factor receptor, 1029bp

L07334

Human platelet-activating factor receptor, 1780bp

D10202

Human prolactin receptor (PRL) 2723bp M31661 Human prostacyclin receptor, 1979bp D25418

Human prostaglandin receptor (E2) 2052bp L25124

Human prostaglandin receptor (E2) 2372bp U19487

Human prostaglandin receptor (EP1) 1376bp L22647

Human prostaglandin receptor (EP2) 1958bp L28175 Human prostaglandin receptor (EP3) isoform IV, L32662

Human prostaglandin receptor (EP3A) 1729bp U13218

Human prostaglandin receptor (EP3A1) 1652bp U13216 Human prostaglandin receptor (EP3D) 154Obp U13217 Human prostaglandin receptor (EP3E) 1429bp U13215 Human prostaglandin receptor (EP3F) 1456bp U13214 Human prostaglandin receptor (PGE-2) , 1515bp L26976 Human prostanoid receptor EP3-I, 1870bp L27490 Human prostanoid receptor EP3-II, 1682bp L27488 Human prostanoid receptor EP3-III, 1379bp L27489 Human prostanoid receptor FP, 2494bp L24470 Human prostanoid receptor IP, 1417bp L29016 Human RMLP-related receptor I (RMLP RI) 1062bp M76673 Human RPE-retinal G protein coupled receptor (rgr) 694bp U15790

Human RPE-retinal G protein-coupled receptor (rgr) 1415bp U14910 Human secretin receptor precursor, 1650bp U20178 Human secretin receptor, 1616bp U13989 Human serotonin IB receptor, (5-HT1B) 2635bp D10995 Human serotonin IC receptor, 2733bp M81778 Human serotonin ID receptor (5-HT1D) 120Obp M81589 Human serotonin ID receptor (5-HT1D) 1260bp M81590 Human serotonin ID receptor, 1348bp L09732 Human serotonin ID receptor, 1506bp M89955 Human serotonin lDb receptor (HTRlDb) 1959bp M75128 Human serotonin IE receptor 5HTR1E, 1221bp M92826 Human serotonin IE receptor, 1930bp M91467

Human serotonin IF receptor (HTR1F) 1141bp L04962 Human serotonin receptor 5HT2 type 2 1368bp M86841 Human serotonin receptor 5HT7, 1406bp L21195 Human serotonin receptor, 1554bp L05597 Human serotonin receptor, 1938bp M83181 Human serotonin receptor, 2287bp M83180 Human soluble vascular endothelial cell growth factor receptor (sflt) U01134 Human somatostatin receptor (SST) 1285bp L14865 Human somatostatin receptor (SSTR4) 1340bp L07833

Human somatostatin receptor isoform 1 (SSTRl) , 1634bp M81829 Human somatostatin receptor isoform 2 (SSTR2) 1351bp

M81830

Human somatostatin receptor subtype 3 (SSTR3) 1413bp

M96738 Human somatostatin receptor, 1427bp L14856

Human substance P receptor (long form) 1674bp M84425

Human substance P receptor (short form) 1268bp M84426

Human thrombin receptor, 3472bp M62424

Human thromboxane A2 receptor, U11271 Human thyroid hormone receptor alpha 1 (TR-alpha-1)

1876bp M24748

Human thyroid stimulatory hormone receptor (TSHR)

2415bp M32215

Human thyrotropin receptor (TSH) 2470bp M31774 Human thyrotropin-releasing hormone receptor, 1229bp

D16845

Human transferrin receptor, 2826bp M11507

Human vasoactive intestinal peptide receptor type 1 (V1RG) U11087 Human vasoactive intestinal peptide receptor, 2754bp

L13288

Human vasoactive intestinal polypeptide receptor 2 (VIPR2) L40764

Human vasopressin receptor (V2) 2282bp L22206 Human vasopressin receptor V3, 1869bp L37112

The protein analogous in function to the Gpalp of Saccharomyces cerevisiae can be, of course, the Gpalp of S. cerevisiae. In addition to the Gpalp protein of S. cerevisiae, there is also presently known the GPA2 gene of S. cerevisiae (Nakafuku et al. , Proc. Natl . Acad. Sci USA £5.-1374 (1988) . The Gpa2p protein is not able to complement defective Gpalp function, but nevertheless the Gpa2p protein might interact with G- beta-gamma complexes to couple a seven-transmembrane receptor to a biochemically selectable pathway. It is expected that other species of yeasts, for example Schizosaccharomyces pombe, will also have proteins that can be used for the Gpalp protein in practicing the present invention.

In making the fusion contruct, the seven- transmembrane protein is operatively linked to the protein having an activity analogous to the Gpalp of Saccharomyces cerevisiae . The two proteins can be directly fused; the carboxy-terminus of the seven- transmembrane protein being joined to the amino- terminus of the protein having Gpalp activity. Alternatively, a short linker peptide can be used to join the two proteins. The linker is preferably from 1-25 amino acids long, more preferably from 1-20 amino acids long, still more preferably from 1-10 amino acids long and most preferably from 3-10 amino acids long.

In practice of one embodiment of the invention a "reporter" gene is operatively linked to the promoter of a gene analogous in function to the FUS1 gene of S . cerevisiae. A reporter gene is one which signals the function of the expression cassette, typically of the promoter function, into which the reporter gene is inserted. The amount of the gene product of the reporter gene can be measured by immunoassay, by enzyme activity (if the reporter gene encodes an enzyme) or by a metabolic selection strategy. Preferred reporter genes encode a protein that is not made by the yeast strain into which they are inserted, to avoid a high background result. Preferred reporter genes in implementing the present invention encode enzymes whose activity can be measured colorimetrically or by a luminescence assay and include 0-galactosidase, glucuronidase (GUS) , green fluorescence protein, and luciferase. If a yeast strain in which the endogenous genes for them have been knocked out is used, genes encoding alkaline phosphatase and invertase (SUC2) are also useful reporter genes.

In a method for screening a compound for receptor antagonist activity, one contacts a yeast cell expressing a fusion protein comprising the seven- transmembrane protein of interest and a Gpalp that functionally couples to the mating-type pathway with the compound to be tested and with a ligand for said receptor. Then, the level of expression of a reporter gene, which measures the activity of a promoter that depends upon the activation of the mating-type pathway, for example, the FUS1 promoter, is measured. The level of reporter gene expression is compared in the presence and absence of the compound to be tested for antagonist activity. Antagonist activity is considered to be observed if the level of reporter gene expression, and thus activity of the mating-type activation-dependent promoter, is lower in the cell contacted with the compound being tested together with the ligand than in the cell contacted with the ligand, but not contacted with the compound being tested. By "lower" is meant a degree of difference between the reporter gene expression in the cells treated with the test compound together with ligand of at least 1/3. The larger the degree of difference, the greater the antagonist activity. A range of differences between 1/3 and 1/10 is expected. Preferably the range is 1/5 to 1/25. More preferably, the range is 1/5 to 1/50. Most preferably, the range is 1/50 to 1/200.

A method for testing a compound for receptor agonist activity is similar to the test for receptor antagonist activity. One contacts a yeast cell expressing a fusion protein comprising the seven- transmembrane protein of interest and a Gpalp that functionally couples to the mating-type pathway with the compound to be tested. Then, the level of expression of a reporter gene, which measures the activity of a promoter that depends upon the activation of the mating-type pathway, for example, the FUS1 promoter, is measured. The level of reporter gene expression is compared in the presence and absence of the compound to be tested for agonist activity. Agonist activity is considered to be observed if the level of reporter gene expression, and thus activity of the mating-type activation-dependent promoter, is higher in the cell contacted with the compound being tested than in the cell not contacted with the compound being tested. By "higher" is meant a degree of difference between the reporter gene expression in the cells treated with the test compound together with ligand of at least 3-fold. Greater degrees of difference are preferred. An expected range is from 3 to 10-fold higher. An acceptable range is 3 to 8-fold higher. Preferably, the degree of difference is 10 to 25-fold. More preferably, the degree of difference is 20 to 100-fold.

A plasmid construct is made that expresses the Receptor-Gpalp fusion protein, and this plasmid is transformed into diploid yeast cells having one mutant and one wild type copy of the essential yeast G-alpha protein gene GPA1 . Sporulation of the diploid should give two viable and two non-viable spores because GPA1 is essential for haploid growth, unless the fusion protein contains a functional Gpalp domain. If so, more than two viable segregants will be obtained, providing a simple genetic complementation assay for appropriate expression and activity of the Gpal domain of the fusion protein. Next, assays based on mating pathway activation are performed, using known activators of the receptor domain of the fusion protein, to test whether the receptor domain of the fusion protein is functional and capable of transmitting the ligand-binding signal to the fused Gpal domain. If so, the fusion molecule is fully functional in both of its domains. If not, the same assays can be adapted, in conjunction with mutagenesis of the Gpal domain, to select for mutants in which the intact receptor domains can signal to the mutant Gpal domain to activate the mating pathway upon activator binding. A detailed description of the procedure is given below.

Yeasts can be tranformed with vectors encoding the recombinant DNA molecules of the present invention by means well-known in the art. Similarly, membranes from yeasts expressing the recombinant DNA molecules of the present invention can be prepared and stored by methods well-known in the art.

Step 1: engineering a covalent linkage between the full length receptor (excluding the cleaved signal peptide, for reasons given in step 2) and Gpalp at their respective carboxy and amino terminal ends.

This is achieved by fusing the genes in frame by standard methods of molecular biology (Maniatis, Fritsch and Sambrook, Molecular Cloning, a Laboratory Manual , 2nd Ed. c. 1989 by Cold Spring Harbor Laboratory Press. ) , as illustrated in examples 1 and 2. The fusion construct includes in addition the endogenous 3' processing signals of the GPA1 gene for proper termination of transcription and polyadenylation. The construct can be made in a vector that can either replicate autonomously in yeast cells, or that integrates into the yeast chromosome. The vector additionally includes a transformation marker gene so that the final construct can be transformed into yeast cells and transformant selected by using the marker.

Step 2: engineering the fusion protein for yeast plasma membrane expression. This is achieved by replacing part of the signal sequence of the receptor in question with part of the N-terminal signal sequence of the yeast G-protein-coupled receptor

Ste2p. Any other N-terminal signal sequence that directs co-translational insertion across the rough endoplasmic reticulum membrane may also be used; examples include N-terminal signal sequences of the a- factor receptor STE3 or secreted proteins such as invertase, and alpha mating factor precursors ΛfFal and

MFa2. Attachment of the signal seguence is done by an in-frame fusion of a DNA fragment encoding the signal peptide, preferably from Ste2, with the DNA fragment encoding the construct from step 1 by standard methods of molecular biology, as illustrated in example 2.

Similar constructs have been shown to cause yeast plasma membrane expression of the human beta- adrenergic receptor (King et al. , Science 250 :121

(1990) ) and the muscarinic cholinergic m5 receptor

(Huang et al. , Biochem. Biophys . Res . Comm . 182 :1180

(1992) ) with ligand binding characteristics that closely mimic the native receptor in mammalian cells.

Step 3: placing the construct under the control of a yeast promoter. This is achieved by cloning in an appropriate promoter fragment contiguous to the 5' of the construct . Preferred promoters are those which can replicate autonomously in yeast. Example 1 demonstrates how this can be done using inducible and moderately strong GAL promoter. Example 3 describes constructions using the strong and constitutive PGK promoter. Codon usage in yeast is biased such that genes expressed at high levels use only one or two of the several possible degenerate codons to encode amino acids. (Jones, Pringle and Broach, The Molecular and Cellular Biology of Yeast Saccharomyces, Vol. 2, c. 1992 by Cold Spring Harbor Laboratory Press.) A strong promoter such as the PGK promoter may therefore be required to generate sufficient RNA levels to overcome the lack of codons preferred by yeast in receptor genes from other species. Alternatively, the receptor-encoding DNA can be engineered to utilize preferred yeast codons.

Step 4: mutating the FARl gene. Farlp is required for growth arrest induced by activation of the mating pathway. For the assays described in Steps 10 and 11, the farl mutation is needed to enable haploid cells with an activated mating pathway to grow while retaining other features of mating pathway activation. The FARl gene can be mutated by replacement with another auxotrophic marker gene (Scherer and Davis, Proc. Natl . Acad. Sci . USA 76.:4951) , or by the two- step mutation strategy (Rothstein, Methods in Enzymology, 101:202) . The latter method is described in Example 11.

Step 5: constructing diploid yeast cells with one wild type and one mutant copy of GPA1. Because GPA1 is an essential gene for haploid cell growth and cannot be mutated in haploid cells directly, the mutation has to be made in a diploid strain preferably a mutant strain having several auxotrophic marker genes on both copies of its chromosomes. Diploid cells of this genotype are constructed by disruption of one of the two GPA1 copies by integration of an auxotrophic marker gene, as in example 5 where the TRPl gene is used. In subsequent segregation, the mutant copy can be followed by the TRPl marker. Thus, because GPA1 is essential, sporulation of each tetrad should give two large colonies, and two small or undetectable colonies, and both of the large colonies should require tryptophan for growth, i.e. lack the TRPl gene. This is illustrated in example 5.

Step 6: transforming the construct of Step 3 into the diploid strain of Step 5. The construct of Step 3 is cloned into a yeast vector that can replicate as a plasmid, and carries a gene that complements one of the auxotrophic mutations present in the diploid strain used to create grpal and farl mutations in Steps 4 and 5. Replicating vectors based on either a yeast centromere sequence, exemplified by the YCp series of vectors, or the 2-micron plasmid origin, exemplified by the YEp series of vectors can be used. (Rose and Broach, Methods in Enzymol . 194 :195. ) The plasmid is then cloned into the diploid strain of Step 5, and transformants carrying the plasmid are selected on the basis of a marker present in the plasmid, preferably an auxotrophic marker, which is URA3 in the case of YEp and YCp vectors.

Step 7: genetic complementation method for testing function of Gpal domain of the receptor-Gpal fusion.

Sporulation of the diploid strain of step 6 carrying the fusion construct provides a convenient way to test if the Gpalp domain in the fusion construct can functionally replace the Gpalp gene product. Segregation of GPA1 and FARl in the diploid strain from Step 5, of genotype GPAl/gpal ; FARl/ farl , should yield the following four haploid genotypes: (i) GPA1 ;FAR1 (ii) GPAl ;farl (iii) gpal ; FARl and (iv) gpal/ farl . Haploids with genotypes i and ii should give viable colonies, those with genotype iii should not give a detectable colony and those with genotype iv should give very small colonies because of incompleteness of growth arrest due to farl . If random spores from this population are analyzed, each of these genotypes should occur at equal frequency. However, because of independent assortment in each tetrad, the two spores that carry gpal from a single meiotic event may be both FARl, both farl , or one of each. Therefore, dissection of any tetrad should always yield two large colonies, and two others which may be both very small (genotype gpal ; farl) , both invisible (genotype gpal , FARl) or one of each. Such segregation is illustrated in examples 5, 6, 7 and 8, where tryptophan prototrophy is used to follow segregation of gpal : : TRPl , and a PCR assay is used to follow segregation of FARl.

If the initial diploid cell carried a plasmid, it should be present in all four spores of the tetrad with equal probability. This probability is always less than one since plasmids can be lost at some frequency in the mitotic divisions preceding meiosis where selection for the marker carried on the plasmid is relaxed, and also in the two divisions of meiosis. If this plasmid carried a gene capable of fully complementing the gpal mutation, then dissection of each tetrad would yield two large colonies as before due to the presence of GPA1, and of the two remaining spores of genotype gpal, some would yield large colonies due to complementation. Thus, some tetrads would show 3:1 or 4:0 segregation for large vs. small or invisible colonies, and the presence of segregants of this type is indicative of complementation. This is illustrated in example 6a, for the GPA1 gene expressed from its own promoter, example 6b for the GPA1 gene expressed from the PGK1 promoter, example 7 for a STE2-GPA1 in-frame fusion protein expressed from the PGJF1 promoter, and example 8 for an in-frame fusion protein between the thrombin receptor and GPA1 expressed from the PGK1 promoter.

Step 8: confirmation of the functionality of the Gpal domain of fusion proteins. Step 7 describes how simple segregation analysis of genetic complementation can provide a good indication of the function of the Gpal domain. However, other genetic phenomena can also give rise to deviations from 2:2 segregation. For example, gene conversion of the disrupted gpal by the wild type copy, either in meiosis or in the mitotic divisions preceding mitosis could give rise to 3:1 or 4:0 segregation respectively. Theoretically, gene conversion could also occur between the complete coding sequence of GPAl present on the plasmid construct and the disrupted chromosomal copy. To eliminate these possibilities, the presence of two chromosomal gpal mutants in each tetrad is identified by segregation of the auxotrophic marker gene whose insertion was the means of disrupting, and thus mutating, one copy of GPAl in the diploid strain in Step 4. Gene conversion of types described above restoring a complete GPAl gene should lead to loss of this marker, and thus to the presence of less than two haploid spores carrying this marker in each tetrad.

In addition to the above possibility, all diploid cells that sporulate might not carry the plasmid since it is lost at some frequency in mitosis unless selection for the plasmid is maintained. Diploid cells can undergo several mitotic divisions without selection prior to meiosis in the sporulation medium, which may lead to loss of the plasmid and thus give rise to 2:2 segregation. In the analysis of Step 7, this would be incorrectly interpreted as an inability of the plasmid to complement gpal .

To eliminate the above possibilities, the four colonies from each tetrad are tested for growth on media that detects the presence of the marker that disrupts the GPAl gene ( TRPl in example 5) , and the plasmid marker ( URA3 in examples 6, 7 and 8) . In the event that there is no complementation and no plasmid loss, 2:2 segregation should be seen in each tetrad, both large colonies should be trp^', any very small colonies (carrying farl) should both be TRP⁺ and a variable number of both large and small colonies . should carry the plasmid and therefore be URA⁺ . If there is complementation with no plasmid loss, all segregants from each tetrad should form large colonies, two of which are trp^" and two TRP⁺, and all should be URA⁺ . In the more likely possibility of complementation with some plasmid loss both in mitosis and in meiosis, tetrads would segregate 2:2 (plasmid loss in mitosis) , 3:1 (plasmid loss in meiosis) or 4:0 (no plasmid loss) . In 2:2 segregants, both large colonies would be trp^", and none would be URA⁺ . In 3:1 segregants, two colonies would be trp^" and variably URA⁺, and one TRP⁺ colony would always be URA⁺ . In 4:0 segregants, two trp^" colonies would be variably URA⁺, and two TRP⁺ colonies would always be URA⁺ . Data illustrating such analysis are provided in examples 6, 7 and 8.

Step 9: tests for function of the receptor domain.

Binding assays provide a sensitive assay for proper expression of the receptor fusion protein, its targeting to the yeast plasma membrane and appropriate folding and generation of transmembrane domains to generate the extracellular binding site. Scatchard analysis of binding data can provide measurements of binding affinity, which can be compared to the affinity in mammalian cells expressing wild-type receptor to obtain a further measure of appropriate expression. Scatchard analysis also provide measurements of the number of binding sites for ligand per cell, which is a good measure of expression levels. In the examples cited here, however, we have used the more stringent alternate approach described in step 10, which not only requires binding to the receptor domain of the fusion protein, but also requires transmission of the binding signal through the linked Gpal domain to the mating factor pathway. Step 10: mating and shmoo formation assay for coupling of receptor domain activation to mating pathway activation. Activation of the mating pathway in haploid cells leads to a distinct morphological change from the typical ovoid cells of vegetatively growing yeast to a pear-shaped "shmoo" which enables mating with cells of the opposite mating type if they are present. In examples 10 and 12 describing a protein fusion between the yeast receptor Ste2p and Gpalp, we have used both the shmoo formation assay and the mating assay to detect functional coupling between the covalently linked domains. The mating assay can only be used with the endogenous yeast receptors Ste2p and Ste3p, because this requires a response to mating pheromones secreted by another yeast cell of the opposite mating type, but the shmoo formation assay can be adapted to other receptors from heterologous organisms.

Step 11: beta-galactosidaβe induction assay for coupling of receptor domain activation to mating pathway activation. This method uses a mating pathway-inducible promoter operatively linked to the bacterial beta-galactosidase gene ( lacZ) as a reporter. For example, transcription from the FUS1 promoter is stimulated by activation of the mating pathway, and therefore, in cells carrying FUSl-lacZ constructs, induction of beta-galactosidase becomes a sensitive indicator for receptor activation. Examples 9b, lOd, 12, 13 and 14 describe this assay for wild type cells to characterize the method (9b) , Ste2p- Gpalp protein fusions (lOd, 12) and thrombin receptor- Gpalp fusions (13, 14) . Because expression of beta- galactosidase is easily quantitiated by spectrophotometry, a quantitative measure of coupling is obtained by means of this assay.

The FUSl-beta-galactosidase construct can be transformed into the haploid strain from Step 8 and maintained on a replicating plasmid of the YEP type. This gives higher basal values of /3-galactoridase due to the 50-100 copies of the plasmid present in each cell, as shown in example 9b. Alternately, the basal expression level can be reduced by integration of the construct into the chromosome, as shown in examples 9b, lOd, 12, 13, and 14 for integration into the FARl locus and the LEU2 locus.

Step 12: growth assay for coupling of receptor domain activation to mating pathway activation. In this case, a mating pathway-inducible promoter such as FUS1 is operatively linked to a an auxotrophic marker gene that is mutated in the cells to be tested. As in Step 10, activation of the mating pathway leads to expression of the auxotrophic marker gene, conferring the ability to grow in appropriate media that lacks the final end product of the marker enzyme. We have used the LYS2 gene in this manner in example 10c. The particular advantage of LYS2 (and also URA3) is that expression of this gene can be selected for in lysine deficient media as well as selected against in media containing the reagent alpha-aminoadipate. This renders the assay adaptable to screening for both agonists and antagonists of the receptor that is modeled. The use of a FUS3-LYS2 construct to assay agonists is illustrated in example 9a and 10c involving activation of the mating pathway by a Ste2- Gpal fusion protein.

Step 13: mutagenesis of the Gpal domain to increase coupling efficiency. In the event that the results from steps 9, 10, and 11 do not indicate optimal coupling between receptor activation and the mating pathway in a given protein fusion, the G-alpha domain of the fusion can be mutagenized by the standard methods, including those described below, and mutants which are created thereby that confer increased coupling efficiency can be selected using the methods described in steps 10 and 11. Mutagenesis can preferably be effected using one or a combination of the following methods: a) random mutagenesis by PCR amplification (Cadwell and Joyce, PCR and Its Applications, c. 1994 by Cold Spring Harbor Laboratory Press, esp. pp. S136) using primers homologous to the two ends of GPAl, with an Mlu I site in the 5' primer and a Pfl MI site in the 3' primer. In this method, amplification is performed in the presence of manganese and altered levels of magnesium such that a mutation rate of 0.5-1% per base is obtained. Products from the mutagenic amplification reaction will be cloned into the plasmid from step 3 which has been digested with Mlul and Pfl MI enzymes, and additionally with Sph I to destroy the original GPAl gene. The ligation mix will be transformed into E. coli such that a library of >10⁶ clones is obtained, representing that many individual mutations. Plasmid DNA from a bulk plate growth of the entire transformation mix will be used to transform yeast and select for mutants with functional coupling as described by the selection procedure of step 11 or the screening procedure of step 10. b) site-directed mutagenesis of specific regions of GPAl using a mixed degenerate oligonucleotide population synthesized with a central region with degenerate bases that targets the domain to be mutagenized, flanked by 5' and 3' regions that are fully homologous to the GPAl gene. Following standard methods of oligonucleotide mutagenesis, the primer extension mixture will be transformed into E. coli such that sufficient individual transformants are recovered to ensure adequate representation of the pool of mutants. The entire library of mutants will then be recovered from bulk growths and used as in a above. Regions to be mutagenized would include the carboxy terminal, which has been implicated in binding to the receptor, and other regions of weak homology. c) loop-out mutagenesis using oligonuclotides with homology to regions that flank the region to be deleted. Comparison of the amino acid sequence of GPAl to human Gs-alpha shows that several large regions of the GPAl sequence are non-homologous to the human protein, and would be good candidates for loop- out mutagenesis (e.g. amino acids 1-61, 75-110, 142- 188, 217-237 of the GPAl sequence.

EXAMPLE 1: CONSTRUCTION OF A FUSION BETWEEN THE YEAST ALPHA FACTOR RECEPTOR Ste2p AND G-ALPHA PROTEIN Gpalp UNDER TRANSCRIPTIONAL CONTROL OF THE GALI PROMOTOR a) Ligating the GALI promoter into the yeast vector

YCp50: the yeast vector YCp50 was digested with BamHl and EcoRI, and the resulting 7572 bp fragment was purified from an agarose gel using GeneClean™ (Bio 101) . An 806 bp EcoRI-Bam HI fragment carrying the

Gall promoter (position #1-810 of GenBank accession number K02115, where a BamHl site was added to the 3' end) was ligated into this YCp50 fragment and the resulting plasmid is designated pRMHBTl.

b) Inserting a polylinker into pRMHBTl : the plasmid pRMHBTl was digested with BamHl and PflMI and the resulting 7574 bp fragment was purified as in Example la. For annealing of the two oligos "a" and "b" listed below, a solution containing 20mM tris-HCl pH 7.4, lOmM MgCl, 50mM NaCl, and 400mM of each oligo were heated to 70°C for 10 minutes, and cooled slowly to 25°C (15 minutes) . oligo a) ^5' GATCCGCGGCCGCACGCGTCCAGCCC^3' oligo b) ^5'CTGGCAGCGTGCGGCCGCG^3' These oligos anneal to form a polylinker with BamHl, NotI, Mlul and PflMI sites, in that order. The annealed oligo fragment was then cloned into the 7574 pRMHBTl BamHl/PflMI fragment to make pRMHBT2.

c) Ligating GPAl into pRMHBT2 : The plasmid pRMHBT2 was cut with Mlul and PflMI, and the 7585 bp fragment was purified as above. GPAl was amplified by PCR from a Saccharomyces cerevisiae genomic DNA prep using the following two primers: oligo c) ⁵GACACGCGTGTAATGGGGTGTACAGTGAGTACGC^3' oligo d) ⁵CGTCCAAGGGATGGACCTTTTTTTTCTCATGCG^3'

Bold text represents the GPAl sequences and normal text represents additional nucleotides (this convention will be maintained throughout this text) . Oligo "c" contains bp 200 to 223 of the GPAl seguence (GenBank accession number M15867) and 10 additional nucleotides containing a Mlul restriction site. Oligo "d" contains bases complementary to residues 1829 to 1850 of the GPAl sequence and additional nucleotides creating a PflMI site homologous to the PflMI site at position 1610 in YCp50. PCR amplification of yeast genomic DNA with these oligos yields a GPAl fragment that contains nucleotides 200-1850 of the GPAl sequence. The Mlul site is immediately upstream of the ATG start codon, and the PflMI site is 232 bp downstream of the TGA stop codon. The amplified GPAl fragment was digested with Mlul and PflMI and ligated to the 7583 bp MluI/PflMI fragment of pRMHBT2 to make pRMHBT3.

d) Ligating STE2 into PRMHBT3 as an in-frame fusion to GPAl : STE2 was amplified by PCR from Saccharomyces cerevisiae genomic DNA using the following 2 primers: oligo e) ⁵CGGGATCCAAGAATCAAAAATGTCTGATG^3' oligo f) ⁵GAACGCGTTAAATTATTATTATCTTCAGTCC^3' Oligo "e" contains nucleotides 520 to 544 of the STE2 sequence (GenBank accession number M24335) and 4 additional nucleotides which create a BamHl restriction site. Oligo "f" contains bases complementary to nucleotides 1804 to 1827 of the STE2 sequence and eight additional nucleotides which include a Mlul site. PCR amplification yields a STE2 fragment containing nucelotides 520-1827 of the STE2 seguence, and includes the entire coding seguence from the ATG start codon (pos. 535, underlined in the oligonucleotide sequence "e" above) to the last base of the Ste2p C-terminal leucine codon (pos. 1827, underlined in the oligonucleotide sequence "f" above) . The STE2 PCR product was cut with BamHl and Mlul and ligated to the 9224 bp BamHl/MluI fragment of pRMHBT3 to make pRMHBT4. The Mlul junction forms an in-frame fusion between STE2 and GPAl ; the resulting chimera codes for all of Ste2p, a tripeptide thr-arg-val orginating from the oligonucleotides used, and all of Gpalp. The STE2-GPA1 fusion construct in pRMHBT4 is transcriptionally regulated by the GALI promoter.

EXAMPLE 2: CONSTRUCTION OF A FUSION BETWEEN THE HUMAN THROMBIN RECEPTOR AND THE YEAST G-ALPHA PROTEIN Gpalp a) PCR-amplifying thrombin receptor cDNA: a portion of the thrombin gene was PCR amplified from a human lung fibroblast lambda GT10 cDNA library using the following two oligonucleotides: oligo g) CGGGATCCATAAGCGGCCGCACCCGGGCCCGCAGGCC oligo h) GAACGCGTAGTTAACAGCTTTTTGTATATGC Oligo "g" contains nucelotides 290 to 312 of the thrombin receptor (ThrR) cDNA sequence (GenBank accession # M62424) and sixteen additional bases coding for a BamHl and a NotI restriction site. Oligo "h" contains bases complementary to nucleotides 1477 to 1499 of the ThrR sequence and eight additional nucleotides which include a Mlul site. Regions of homology to the ThrR cDNA are in bold type. The PCR product contains bp 291 to 1499 of the human ThrR cDNA sequence, coding for amino acids 22 (arginine) to the COOH-terminal threonine.

b) Ligating the human thrombin receptor PCR product into pRMHBT3 as an in-frame fusion to GPAl : the human thrombin PCR product was digested with NotI and Mlul and ligated to the 9219 bp Notl/Mlul fragment of pRMHBT3 yielding pRMHBT15. The Mlul site creates an in-frame fusion of the COOH-terminus of the thrombin receptor (amino acid sequence ... leu-leu-thr) with the NH₂-terminus of Gpalp (amino acids met-gly... ) , bridged by the tripeptide thr-ag-val as in Example Id. c) Creating an in-frame fusion between the Ste2p signal peptide and the NH₇-terminus of the thrombin/Gpal fusion: Two oligonucleotides, when annealed, give rise to the double-stranded molecule shown below with overhangs complementary to BamHl and NotI sites.

GATCCATGTCTGATGCGGCTCCTTCATTGAGCAATCTATTTTAT

GTACAGACTACGCCGTGGAAGTAACTCGTTAGATAAAATACCGG This molecule, upon insertion into the BamHl-NotI sites of pRMH15 creates an in frame fusion that encodes the first thirteen amino acids of the Ste2p signal seguence, a bridge glycine (part of the NotI overhang, and the seguence arg-thr-arg-arg... of the thrombin receptor. The above cloning step yielded pRMHBTl6.

EXAMPLE 3 : TRANSFER OF FUSION CONSTRUCTS OF EXAMPLES 1 AND 2 TO HIGH-COPY VECTORS CONTAINING THE CONSTITUTIVELY ACTIVE PGK PROMOTOR

The fusion constructs in Examples 1 and 2 were placed under the transcriptional control of the PGK1 promoter carried on a yeast 2-micron-plasmid-based vector. A BamHl/Ncol fragment of pRMHBT4 containing the fusion construct and part of the URA3 marker was ligated into the BamHI/NcoI digested pPGK (Kang et al, 1990, Mol. Cell. Biol., 10:2582) . Similarly, the

BamHI/NcoI fragment of pRMHBT16 containing the

ThrR/Gpal fusion and part of the URA3 marker was ligated into the BamHI/NcoI digested pPGK. The resulting plasmids were designated pRMHBTlδNG and pRMHBT20NG, respectively.

EXAMPLE 4: DISRUPTION OF THE CHROMOSOMAL FARl GENE The FARl gene was amplified from yeast genomic DNA using the following primers : oligo i) CAACATGCAGCCATTTCACCG oligo j) CGCGAGCTCGCCAATAGGTTCTT CTTAGG

Oligo "i" contains the sequence from residues 34 to 54 of FARl (GenBank accession # M60071) Oligo "j" contains the seguence complementary to nucleotides 2959 to 2980 of the FARl gene and eight additional nucleotides which create a SacI restriction site. The amplified seguence extended from nucleotides 34 to 2980. The FARl PCR product was digested with Kpnl and SacI, and ligated into those same sites in the yeast integrating vector pRS306 (Sikorski and Hieter, 1989, Genetics 122:19-27) . The resulting plasmid was designated pFARl. The farl mutation was constructed by deleting an internal 700 bp Xbal fragment from pFARl, which removed bp 1917 to 2616, and results in a protein that is missing 153 of its 781 amino acids. The resulting plasmid was designated pFARX. The pFARX plasmid was used to introduce the farl mutation into the chromosome of the haploid yeast strain MS2288 (mat a, ura3-52, leu2-3,112, his3D200, trplDl; M. Rose, Princeton University) . pFARX was linearized at its single EcoRI site (position 2771 of FARl) and used to transform competent MS2288 cells to uracil prototrophy, thereby integrating the pFARX plasmid at the FARl locus. Strains in which the plasmid had recombined back out of the chromosome were identified using 5-FOA selection, and ura^' derivatives were screened for retention of the farl mutation by PCR analysis using oligos "i" and "j". The farl mutants HBS31 and HBS32 (farl-X200) exhibited continued cell division is the presence of alpha factor indicating that the mutation functionally disrupted the chromosomal FARl gene.

EXAMPLE 5: DISRUPTION OF THE CHROMOSOMAL GPAl GENE a) Construction of a FARl/farl diploid strain: Strain HBS10 (mat a, ura3-52, leu2-3,112, his3Δ200, trplΔl, lys2ΔS738, farl-X200) was mated to MS16 (mat a, trplΔl, ade2-101) and the resulting strain was sporulated. Segregants from this cross included TMHY2-14A (mat a, ura3-52, his3Δ200, trplΔl, lys2ΔS738, ade2-101) . TMHY2-14A was then mated to HBSIO and diploids were selected. The resulting diploid strain was designated TMHY2-223D (a/a, ura3/ura3 , Ieu2/LEU2, his3/his3, trpl/trpl , Iys2/lys2, farl /FARl , ADE2/ade2) .

b) Engineering the GPAl disruption construct: the TRPl gene was amplified from the vector pRS304 (Sikorski and Hieter, 1989, Genetics 122:19-27) by PCR using the following oligos, both with SphI sites (underlined) , to yield a 1134 bp fragment containing a functional TRPl gene: oligo k) GAATGCATGCGGCATCAGAGCAG oligo 1) GAATGCATGCGGTATTTTCTCCTTACGC This PCR product was digested with SphI and ligated into the 8386 bp SphI fragment of pRMHBT3. The two SphI sites, separated by 851 bp, are present within the coding seguence of the GPAl gene in this plasmid. Replacement of this fragment with the TRPl gene yielded the plasmid pRMHBTIO in which the TRPl gene is flanked by GPAl sequences.

c) Disrupting the chromosomal GPAl locus: pRMHBTl0 was digested with Mlul and PflMI to liberate a 1887 bp fragment containing the TRPl gene flanked by GPAl sequences as described in 5b. This fragment used to transform the diploid strain TMHY2-223D to tryptophan prototrophy. The deletion was confirmed by PCR analysis of several transformants using the GPA1- specific oligos "c" and "d" of Example 1. These strains were given the designation TMHY3D (genotype a/a, ura3/ura3, his3/his3 , Iys2/lys2, trpl/trpl , ADE2/ade2, FARl/farl , LEU2/leu2, GPAl /gpal : : TRPl) .

d) Genetic confirmation of GPAl disruption: Five different TRP⁺ transformants (TMHY3D-1, TMHY3D-2, TMHY3D-3, TMHY3D-5, and TMHY3D-6) were sporulated and tetrads dissected. Representative data for one of the transformants is given below. Four of seven tetrads produced two normal colonies, one small colony, and one non-viable spore (2:1:1) . Two tetrads produced two normal colonies and two non-viable spores (2:0:2) . One tetrad produced two normal colonies and two small ones (2:2:0) . Similar data was obtained for the other four sporulations. Each tetrad, on non-selective plates, is expected to give only two normally growing colonies (both

GPA1⁺) . The two others (gpal") should be slow-growing

{gpal^' , farl^') or nonviable (gpal^' , FAR1 ⁺ ) . All normally-growing colonies should be trp", whereas all the small and inviable colonies should be TRP⁺. Further analysis confirmed that all normally-growing colonies were trp^". Ten of these were analyzed by PCR using the GPAl-specific oligos c and d of Example 1, and confirmed that all carry the wild type GPAl allele. Nine representative slow-growing colonies from each sporulation were analyzed further: All were TRP+ indicating that they carried the grpal mutation. Six of these were subjected to PCR analysis as above (using oligos "i" and "j" for FARl), which confirmed that all six are gpal^", farl^".

EXAMPLE 6: COMPLEMENTATION OF THE gpal MUTATION BY CLONED GPAl a) Complementation of opal with a full length GPAl gene: a 1924 bp EcoRI fragment including the entire GPAl gene (Dietzel and Kurjan, 1987, Cell 50:1001- 1010) was amplified from yeast genomic DNA using the following oligos: oligo m) GGAATTCCACCAATTTCTTTACG oligo n) GGAATTCGAGATAATACCCTGTCC The resulting PCR product was ligated into the EcoRI site of the vector pRS316 (Sikorski and Heiter) and the 2-micron vector YEp352 (Hill et . al . , 1986, Yeast 2:163-167) . The resulting plasmids were designated p316GPAl and p352GPAl, respectively. Strain TMHY3D-1 was transformed with both plasmids to uracil prototrophy, and the strains were sporulated. Complementation of the gpal mutation by a plasmid carrying GPAl should result in 4:0 segregation for viable vs. small or non-viable colonies (assuming the plasmid segregates to all four spores) . However, the theoretical 4:0 segregation expected for full complementation would not be always realized since plasmids are lost at some frequency in both the mitotic divisions in the sporulation medium, and in meiosis. The following results were observed: Of 19 tetrads, 9 from p316GPAl and 10 from p352GPAl transformants, 11 segregated 4:0 for normal vs. small or nonviable colonies. In all 11, two colonies per tetrad were trp^" and two were TRP⁺, and all TRP⁺ colonies were URA⁺, indicating that they carried both the gpal mutation and the GPAl-containing plasmid. Four tetrads segregated 3:1 for normal vs. small or nonviable colonies. In all four tetrads, two colonies were trp^" and one was TRP^4", and all the TRP+ colonies were also URA+. These results indicate that the non- viable spore failed to receive a complementing plasmid. Three tetrads segregated 2:2 for normal vs. small or nonviable colonies. In two of these tetrads, the two viable colonies were trp^" (GPA1 ⁺ ) , ura^" , suggesting all four spores lacked the complementing plasmid (plasmid was likely lost in the mitotic divisions preceding meiosis) . The other segregated 1:1 for trp, probably resulting from incomplete dissection. One tetrad segregated two normally- growing colonies, one slow-growing colony and one non- viable colony. All were ura^" indicating plasmid loss in mitosis, and the medium-sized colony is likely a farl , gpal double-mutant which grew better than others of the same genotype for unknown reasons. Alternatively, this colony may contain a mutation which partially suppresses the gpal mutant phenotype. These results clearly demonstrate that the cloned GPAl gene fully complements the chromosomal gpal mutation.

b) Complementation of the opal mutation by GPAl under PGK1 promoter transcriptional control : the plasmid pRMHBT20NG (Example 3) was digested with BamHl and Mlul, blunted with Klenow, and religated to yield pRMHBT43. This was transformed into HBS14 (genotype a/a, ura3/ura3, his3/his3, Iys2/lys2, trpl/trpl, ADE2/ade2, FARl/farl, LEU2/leu2, GPAl/gpal ::TRPl) by selection for uracil prototrophy. 21 tetrads from three URA⁺ transformants were dissected. Eleven of these segregated 4 :0 and nine segregated 2 :2 for normal-growing to slow-growing or inviable colonies. All colonies from the eleven 4:0 tetrads were URA⁺, whereas all of the growing colonies from the nine tetrads that segregated 2:2 were trp^", ura^". That all 4:0 segregants were URA⁺ indicates that the plasmid pRMHBT43 can efficiently complement the chromosomal gpal mutation.

EXAMPLE 7: COMPLEMENTATION OF THE gpal MUTATION BY EXPRESSION OF A STE2-GPA1 CHIMERIC PROTEIN The plasmid pRMHBT18NG encoding a Ste2p-Gpalp fusion protein was used to transform HBS14 to uracil prototrophy. The resulting strain was sporulated and 20 tetrads were dissected. Six segregated 4:0, four 3:1 and ten 2:2 for normal-growing to slow-growing or inviable colonies. All 4:0 segregants were URA⁺, which clearly demonstrates that the Gpal-Ste2 chimera can rescue the gpal phenotype. Of the 3:1 segregants, three contained one TRP⁺, URA⁺ colony, strongly suggesting that the non-viable spore was gpal and did not receive the plasmid. Of the ten 2:2 segregants, all growing colonies were trp^", and none were URA⁺ . This provides further evidence that pRMHBT18NG complements the gpal mutation. One of the four 3:1 segregating tetrads contained two TRP⁺ and one trp^" colonies, suggesting incomplete dissection.

EXAMPLE 8: COMPLEMENTATION OF THE gpal MUTATION BY EXPRESSION OF A THROMBIN RECEPTOR-GPAl FUSION PROTEIN

The plasmid pRMHBT20NG was used to transform strain

HBS14 to uracil prototrophy. The resulting strain was sporulated, and 19 tetrads were dissected. Five tetrads segregated 4:0, four segregated 3:1 and ten segregated 2:2 for normal growing to slow-growing or inviable spores. All 4:0 segregants were URA^"1", which clearly demonstrates that the ThrR-Ste2 chimera can rescue the gpal phenotype. Of the 3:1 segregants, two contained one TRP⁺, URA⁺ colony, strongly suggesting that the non-viable spore was gpal and did not receive the plasmid. In nine of the ten 2:2 segregants, all growing colonies were trp^", and none were URA⁺ . This provides further evidence that pRMHBT20NG complements the gpal mutation. One of the three 3:1 segregating tetrads contained two TRP⁺ and one trp^" colonies, suggesting incomplete dissection. One of the 2:2 segregating tetrads contained one TRP+ colony, but it was also URA⁺ . That a trp^" colony is "missing" likely reflects incomplete dissection. Thus all growing TRP⁺

(gpal^") colonies are URA⁺ and therefore contain pRMHBT20NG. These results demonstrate that the ThrR-

Ste2p chimera complements the gpal mutant phenotype.

EXAMPLE 9: REPORTER ASSAYS FOR ACTIVATION OF THE MATING PATHWAY a) Construction of a lacZ gene transcriptionally regulated by the mating pathway-specific FUS1 promoter: The FUS1 promoter was amplified from yeast genomic DNA by PCR using oligos "o" and "p" shown below : oligo o) GCATGCTGCAGGATCGCCCTTTTTGACG oligo p) GACGTCGACAGAAACTTGATGGCTTATATCCTGC Oligo "o" contains the sequence of nucleotides 1 to 23 of FUS1 (GenBank accession # M17199) and five additional nucleotides creating a SphI restriction site. Oligo "p" contains the seguence complementary to residues 232 to 258 of the FUS1 gene and eight additional nucleotides which create a Sail restriction site. The amplified seguence encompasses nucleotides 1 to 258, and includes a Pstl site at residue 1 in FUS1 . The FUS1 promotor was digested with Sail and Pstl, and ligated into those same sites in the vector pUC19 (Yanisch-Perron et. al. , 1985, Gene 33 :103-119) . The resulting plasmid was designated pUFS.

The LacZ coding sequence was cut from pON831

(obtained from J. Vieira, University of Washington) using Sail and Kpnl, and this 3.2 kb fragment was ligated into pUFS digested with the same enzymes. The resulting plasmid, pFus-Lac, contained the lacZ coding sequence under transcriptional control of the FUS1 promoter. The Fusl-lacZ gene was then moved into three different yeast vectors: 1) pFus-Lac was digested with SphI and the resulting FUS-lacZ segment was cloned into the SphI site within the coding seguence of the FARl gene in the plasmid pFARl. The resulting plasmid (which is an integrating vector containing URA3 as its selectable marker) was designated pFFLz. 2) pFus-Lac was digested with Hindlll and Kpnl, and the resulting FUS-lacZ segment was cloned into the 2-micron vector YEp352. The resulting plasmid which uses C7RA3 as a selectable marker was designated pYFL3. 3) pFus-Lac was digested with Pstl and Kpnl, and the resulting FUS-lacZ segment was cloned into the integrating vector YIp351 (Hill et. al., 1986, Yeast 2:163-167) . The resulting plasmid which uses LEU2 as a selectable marker was designated pLZ351.

The above plasmids were transformed into yeast strains, and the cells were analyzed for their ability to induce beta-galactosidase in response to alpha factor addition to the growth medium. Strain HBSIO transformed with pYFL3 exhibited an alpha factor- independent beta-galactosidase specific activity of 212 nmol/mgmin, and alpha factor-induced activity of 2023 nmol/mgmin, representing a 9.5 fold induction. pFFLz was digested with EcoRI which linearized the plasmid within the FARl coding seguence, and this DNA was used to transform strain HBSIO to uracil prototrophy. This integrated the plasmid at the chromosomal FARl locus. HBSIO: :pFFLz exhibited an alpha factor-independent beta-galactosidase activity of 23 nmol/mgmin and alpha factor-induced activity of 735 nmol/mgmin, representing a 32.0 fold induction.

b) Construction of a LYS2 gene transcriptionally regulated by the mating pathway-specific FUS1 promotor: The FUS1 promotor was cut from pUFS using SphI and Sail, and the 266 bp fragment was ligated into the Sphl-Sall sites of Ycp50 to make pRMHBT25. LYS2 was PCR-amplified (only coding sequence and 3' untranslated region) from yeast genomic DNA using the following oligonucleotides: oligo q) CGGCGGTCGACTAATGACTAACGAAAAGG oligo r) CCCGGGCGCAAGTATTCATTTTAGACCCATGGTGG Oligo "q" contains the sequence of nucelotides 299 to 312 of LYS2 (GenBank accession # M36287, M14967) and nine additional nucleotides creating a Sail restriction site. Oligo "r" contains the sequence complementary to nucleotides 4822 to 4850 of the LYS2 gene and six additional nucleotides which create a Smal restriction site. The 4566 bp LYS2 PCR product was digested with Sail and Smal, and ligated into the Sail-Nrul sites of pRMHBT25 to generate pRMHBT26, which contains the LYS2 coding seguence under transcriptional control of the FUS1 promoter.

The following experiment was performed to verify mating pathway-dependent activation of the LYS2 gene: HBSIO cells were transformed to uracil prototrophy by pRMHBT26. Transformants were grown to mid-log in ura^" media, and cells were back-diluted into ura'lys^" media with or without 5.8mM alpha factor. Growth was measured by OO₆₀₀, but the initial measurement was taken using a Coulter Counter, yielding a starting cell count of 5.18 X 10⁵/ml . The time point readings (ODgoo) of the cultures were as follows:

12.0 hrs 18.0 hrs 24.0 hrs control 0.035 0.050 0.038 alpha factor 0.068 0.244 1.022

HBS10 (without pRMHBT26) did not grow in lys- media. These results clearly demonstrate that strain HBS10/pRMHBT26 exhibits alpha factor-dependent lysine prototrophic growth, which confirms that expression of Lys2p is dependent upon mating pathway activation by alpha factor.

EXAMPLE 10: ALPHA FACTOR-DEPENDENT ACTIVATION OF THE MATING PATHWAY BY THE STE2-GPA1 FUSION IN gpal CELLS Examples 7 and 8 show that the Gpalp domain of the Ste2-Gpal fusion construct functionally complements the chromosomal gpal mutation. To determine if Gb and G dissociate from the Gpal domain (Ga) of the Ste2- Gpal fusion protein in an alpha factor-dependent manner and therefore propagate the mating pathway activation signal, the following experiments were performed:

a) Shmoo formation assay: Strain 9A/pRMHBT18NG (a haploid segregant of HBS14 carrying the plasmid pRMHBTl8NG, whose genotype is mat a, ade2-101, ura3- 52, leu2, 3-112, his3D200, trplDl, lys2-DS738, farl- X200, gpal: :TRP1 (the gpal mutation is complemented by the URA3-containing plasmid pRMHBT18NG) ) was grown to mid-log in ura^" media, and alpha factor was added to 5.8mM. Microscopic examination after 5.0 hours clearly showed that more than 70.0% of the pheromone- treated cultures were shmooed, while less than 10.0% of the no-alpha factor controls were shmooed. These results demonstrates alpha factor-dependent dissociation of the Gb and Gg subunits from the Gpalp domain of the Ste2-Gpal fusion protein, resulting in subsequent activation of the mating response pathway.

b) LYS2 prototrophy assay: To change the selectable marker from URA3 to IS3, the 6159 bp Apal-Clal fragment of pRMHBT26 containing the FUS1 promotor -LYS2 gene fusion was ligated into the Apal-Clal sites in pRS313 (Sikorski and Hieter) to generate pRMHBT41.

Strain 9A/pRMHBT18NG was transformed to histidine prototrophy with pRMHBT41, resulting in strain

9A/pRMHBT18NG/pRMHBT41. Cells were grown to mid-log in ura^'his^" media, at pH 6.5 and 4.0. 9A/pRMHBT18NG controls were grown similarly in ura^" media. The cells were washed three times with sterile water before being diluted into the experimental (lys-) media.

Each group of cells was back-diluted into two aliquots

- one of which contained 5.8mM alpha factor.

In ura^'his^'lys^"media:

1. 9A/pRMHBT18NG/pRMHBT41, pH 4.0 2. 9A/pRMHBT18NG/pRMHBT41, pH 4.0 + alpha factor

3. 9A/pRMHBT18NG/pRMHBT41, pH 6.5

4. 9A/pRMHBT18NG/pRMHBT41, pH 6.5 + alpha factor In ura^'lys^" media:

5. 9A/pRMHBT18NG pH 4.0 6. 9A/pRMHBT18NG pH 4.0 + alpha factor 7. 9A/pRMHBT18NG pH 6.5

8. 9A/pRMHBT18NG pH 6.5 + alpha factor

Cell growth was monitored by ODrøo. Each aliquot received the same number of cells (volumetrically) . The time point readings were as follows:

expt 6 hrs 12 hrs 24 hrs 30 hrs 36 hrs

1. 0.000 0.000 0.045 0.171 0.582

2. 0.000 0.035 0.570 2.010 3.284

3. 0.000 0.007 0.023 0.039 0.058

4. 0.002 0.021 0.132 0.500 0.682

5. 0.003 0.014 0.016 0.021 0.019

6. 0.003 0.010 0.016 0.018 0.018

7. 0.006 0.017 0.016 0.021 0.022

8. 0.002 0.006 0.014 0.020 0.009

The results shown above, like those in 9b, clearly demonstrate that strain 9A/pRMHBT18NG/pRMHBT41 exhibits alpha factor-dependent lysine prototrophic growth at pH 4.0 and pH 6.5, which confirms that expression of Lys2p is dependent upon mating pathway activation (by alpha factor) . The very slow growth seen in "1" is most likely due to basal activity of the FUS1 promotor. That we did not see slow growth in "3" probably reflects the fact that the yeast pH optima for growth is less than 4, and at 6.5 they are sufficiently stressed as to be unable to support lysine prototrophy from the basal activity of the FUS1 promotor. Alpha factor-dependent lysine-prototrphic growth demonstrates that the Ste2p-Gpalp fusion protein activates the mating pathway in a gpal background. Importantly, the mating pathway is not constitutively activated in the gpal strain 9A/pRMHBT18NG/pRMHBT41 since lysine prototrophy is alpha factor-dependent. This further supports the conclusion that the Gpalp domain of the Ste2-Gpalp fusion protein correctly associates with the Gb and Gg subunits. Also, the mating pathway can be effectively activated at pH 6.5, which more closely resembles physiological conditions for mammalian receptors. Additionally, the higher pH preferably pH 6 to 7.5 reduces background prototrophy due to basal activity of the FUS1 promotor ("1" vs. "3") . Thus, by selecting for lysine prototrophy, we can identify cells whose mating pathways are initiated via ligand- dependent activation of mammalian receptors fused to Gpalp.

c) Lac Z reporter assay: pLZ351 (see 9 above) was digested with BstEII to linearize the plasmid within the LEU2 sequence, and then this DNA was used to transform strain 9A/pRMHBT18NG to leucine prototrophy. Strain 9A/pRMHBT18NG is a haploid segregant of HBS14 carrying the plasmid pRMHBT18NG, whose genotype is mat a, ade2-101, ura3-52, leu2, 3-112, his3D200, trplDl, lys2-DS738, farl-X200, gpal::TRPl (the gpal mutation is complemented by the URA3-containing plasmid pRMHBT18NG) . The resulting strain is designated 9ALZ/pRMHBTl8NG. Cells were grown to mid-log in ura^' media, and diluted to an ODgoo of approximately 0.3 in the same media. Cells were treated with alpha factor at 5.8mM and incubated at 30°C for 3.0 hours. Cell lysates from alpha factor treated and control cells were prepared and assayed for beta-galactosidase specific activity (Rose et. al. , Methods in Yeast Genetics: A Laboratory Course Manual, 1990, CSH Laboratory Press) . This yielded specific activities of 18.1 nmol/minmg for untreated cells, and 122.2 nmol/minmg for alpha factor-treated cells. This is a 6.8-fold induction of activity upon alpha factor treatment, and clearly demonstrates alpha factor- dependent activation of the yeast mating pathway through the Ste2p-Gpalp fusion protein.

EXAMPLE 11: ALPHA FACTOR-DEPENDENT ACTIVATION OF THE YEAST MATING PATHWAY BY THE STE2-GPA1 FUSION IN Ste2 CELLS

a) Disruption of the STE2 gene in HBS32 cells: The STE2 gene was PCR amplified from genomic DNA using the following two oligonucleotides:

s) AGTGCGGCCGCAAGCTTATGTCTGATGCGGCTCCTTCATTG t) ACGCGTTCTAGATCATAAATTATTATTATCTTCAGTCCAGAAC

Oligonucleotide "s" contains the seguence from bp 534 to 557 of STE2 (GenBank accession # M24335) and seventeen additional nucleotides creating NotI and Hindlll restriction sites. Oligonucleotide "t" contains the sequence complementary to bp 1800 to 1832 of the STE2 gene and ten additional nucleotides which create Mlul and Xbal restriction sites. The resulting 1295 bp PCR product was digested with NotI and Xbal and ligated into pBluescript (Stratagene) cut with the same enzymes. An internal Nsil fragment of the STE2 gene (at positions 1148 and 1436) was deleted by digestion with Nsil and religation, creating a frameshift mutation in addition to the deletion. The resulting plasmid was digested with Hindlll and Xbal and the 1005 bp fragment with the STE2 deletion mutation was ligated into the yeast integrating vector pRS306 (Sikorski and Hieter Genetics 122 :19 (1989)) . This mutant gene was used to replace wild type STE2 by the two step method. The deletion plasmid was linearized within STE2 at the Hpal site and integrated into HBS32 by selection for uracil prototrophy. Strains in which the plasmid had recombined back out of the chromosome were identified using 5-FOA selection, and these ura^" derivatives were screened for the ste2 mutation by PCR analysis using oligonucleotides s and t. One resulting strain with a such a ste2 deletion was designated HBS12. The FUS1-LACZ reporter construct was integrated into HBS12 as described in Example IOC to make the strain HBS12LZ (leu2: :LEU2-FUS1-LACZ) .

b) Overexpression of Ste2p rescues the ste2 phenotype. Strain HBS12LZ was transformed to uracil prototrophy with pRMHBT45, which is a 2-micron URA3-marked vector containing the coding seguence of STE2 under transcriptional control of the PGK promoter, with termination signals from the GPAl gene (note that this was not a fusion construct, and no GPAl coding sequences were present) . HBS12LZ/pRMHBT45 was grown to mid-log phase in ura^" media, and alpha-factor was added to 5.8 mM. After four hours of incubation at 30.0°C on a roller drum, microscopic examination revealed that over 80.0% of the treated cells were shmoos, while shmoos were undetectable in an untreated control culture. This result clearly shows that the STE2 construct pRMHBT45 carries a functional STE2 gene. Beta-galactosidase assays confirmed this conclusion, as follows.

Strain Miller Units

HBS12LZ/pRMHBT45 2.44 +/- .052

HBS12LZ/pRMHBT45 + α-factor 90.81 +/- 1.29

p= <10"° (ANOVA- Duncan's post-hoc test)

These results clearly show mating factor-dependent activation of the FUSlp-LACZ reporter, and confirms that pRMHBT45 carries a functional STE2 gene.

c) The Ste2p domain of the Ste2p-Gpal chimera is functional : To determine if the Ste2p domain of the Ste2p-Gpalp fusion protein is functional (able to bind alpha factor and transmit the binding signal to Gpal) , the following experiment was performed. Strain HBS12LZ was transformed to uracil prototrophy with pRMHBT18NG, which carries the fusion construct, and the resulting strain, HBS12LZ/pRMHBT18NG was examined for alpha factor-dependent shmoo formation. Cells were grown to mid-log phase in ura^" media, and alpha factor was added to 5.8 mM. After four hours (post- addition) of incubation at 30.0°C on a roller drum, >50.0% of the alpha factor treated cells had formed shmoos, while no shmoos were detected in untreated controls. These results clearly show that the Ste2p domain of the Ste2p-Gpalp fusion protein is functional. Additionally, a quantitative beta- galactosidase assay was performed on these cultures as described previously:

Strain Miller Units

HBS12LZ/pRMHBT18NG 5.5 +/- .098

HBS12LZ/pRMHBTl8NG + α-factor 70.97 +/- 1.90

p= <10^"° (ANOVA, Duncan's post-hoc test)

These results clearly show mating factor-dependent activation of the FUSlp-LACZ reporter, and confirms that the Ste2p domain of the Ste2p-Gpalp chimera from pRMHBT18NG is functional and rescues the ste2-deletion phenotype.

EXAMPLE 12: ENHANCED ACTIVATION OF THE MATING PATHWAY BY THE STE2-GPA1 FUSION IN ste2, gpal CELLS

a) Deletion of chromosomal STE2 in strain 9ALZ: The strain 9ALZ/pRMHBTl8NG with a chromosomal gpal mutation was transformed to lysine prototrophy with pRMHBT44 to remove the URA3 marker of pRMHBTlθNG and replace it with a LYS2 marker. pRMHBT44 is functionally equivalent to pRMHBT43 (GPAl under PGK promotor transcriptional control) , except it has a LYS2 marker. The strain 9ALZ/pRMHBT44 was identified by 5-FOA counter-selection against the URA3-containing plasmid pRMHBT18NG. This strain was used for disruption of the STE2 gene by the two step method using URA3 , as in example 11. The new ste2, gpal strain was designated 9ALZΔGS/pRMHBT44. A "plasmid shuffle" was then performed to replace pRMHBT44 with pRMHBT18NG carrying the STE2-GPA1 fusion construct. Strain 9ALZΔGS/pRMHBT44 was transformed to uracil prototrophy with pRMHBT18NG. URA⁺ cells were then grown to saturation in ura" media, and cells that had lost pRMHBT44 were selected for by growth on ura^" plates with 5.0% a-aminoadipic acid (a-aminoadipic acid is lethal to LYS⁺ cells, and selects against pRMHBT44) . A similar plasmid shuffle was also performed to replace pRMHBT44 with pRMHBT20NG. The resulting strains were designated 9ALZΔGS/pRMHBTl8NG and 9ALZΔGS/pRMHBT20NG, respectively.

b) Activation of the mating pathway bv the Ste2p-Gpalp fusion protein in opal , ste2 cells: This was done by demonstrating that the Ste2p-Gpalp fusion can transduce the α-factor binding signal to cause activation of the mating pathway. Strain 9ALZΔGS/pRMHBT18NG was grown to mid-log phase in ura^" media. Cells were back-diluted to 0.2 OD^, and - factor was added to 5.8 mM. Cells were grown at 30.0°C on a roller drum for an additional 4.0 hours, then examined by light microscopy and prepared for beta- galactosidase assays as described previously. Two independent experiments were performed. In both, over 90.0% of the treated cells had formed shmoos after 4.0 hours, while less than 5.0% were shmoo-like in the untreated (control) cultures. Quantitative beta- galactosidase assays were performed as described previously, providing the following results in two separate experiments:

Strain Exp.# Miller Units

9ALZΔGS/pRMHBTl8NG 8.59 +/- .195

12.76 +/^■ 104

9ALZΔGS/pRMHBTlδNG + 41.40 +/- .67 α-factor

2 60.63 +/- .433

p= <10"° for both experiments (ANOVA, Duncan's post-hoc test)

These results clearly show that the Ste2p-Gpal chimera can complement the deletion of both ste2 and gpal, and can transduce the alpha factor binding signal to initiate mating pathway activation. Reporter gene activity is enhanced 4.8 fold and 6.3 fold in the two experiments indicating that the mating pathway is strongly activated by alpha factor binding. Note that the basal levels without alpha factor are higher in the experiments described in this and the following sections (and in similar experiments in Examples 10c, 13 and 14) than in cells without the gpal mutation

(Example 9a) . This is probably because the gpal mutation is not completely complemented by any of the constructs, leading to a low basal level of activation of the mating pathway and consequent low levels of beta-galactosidase activity.

c) Activation of the mating pathway bv Ste2p and Gpalp expressed separately in ste2 opal cells: As a control for the previous experiment, Ste2p and Gpalp were expressed separately from the same promoter and vector in the same yeast strain used to express the fusion protein. 9ALZΔGS/pRMHBT44/pRMHBT45 cells were grown to mid-log phase in ura^" media, α-factor was added to 5.8 mM, and the cells were assayed for beta- galactosidase activity after incubation for four hours at 30°C on a roller drum. Cells were also observed via light microscopy after 4.0 hours of incubation, and no shmoos were detectable in either the treated or non- treated control cultures. The beta-galactosidase assay data is shown below:

Strain Miller Units

9ALZΔGS/44/45 8.00 +/- 0.08

9ALZΔGS/44/45 + α-factor 8.2 +/- 0.10

While there is a significant difference between the two cultures (p= .026, ANOVA, Duncan's post-hoc test) , it is very small. Thus, we conclude that the mating pathway is only weakly activated ( 2.4% stimulation due to mating pheromone) in response to alpha factor in cells expressing Gpalp and Ste2p from the same promoter and vector as the fusion protein in the previous section "b" . In contrast, the fusion protein expressed in the same strain from the same promoter and vector causes, in two experiments, a 4.8-fold and a 6.3-fold enhancement of reporter gene activity. Assuming that levels of Ste2p and Gpalp proteins in this experiment are comparable to levels of the Ste2p- Gpalp fusion protein in the experiments described in section "b", the efficiency of coupling between Ste2p and Gpalp when fused is greater by two orders of magnitude than when separate. This experiment is a more appropriate control for the fusion protein than comparing the efficiency to the separated components in a wild type STE2, GPAl cell, since expression of the two genes in the wild type cell has been optimized by evolution for maximal sensitivity to mating factor. EXAMPLE 14 : THROMBIN-DEPENDENT ACTIVATION OF THE YEAST MATING PATHWAY BY THE HUMAN THROMBIN RECEPTOR- GPAl FUSION PROTEIN IN ste2 gpal^' CELLS

9ALZΔGS cells were transformed with the thrombin construct pRMHBT2OΝGby the plasmid shuffle method described in Example 12a. These cells were grown to mid-log phase in ura^" media buffered at pH 7.0. Human thrombin was added to 71.4 units/ml media, and the cells were incubated for four hours at 30°C on a roller drum. Crude extracts were then prepared as described and beta-galactosidase assays were performed. The results are shown below:

Strain Miller Units

9ALZΔGS/20ΝG 9.49 +/- 0.22 9ALZΔGS/20NG + thrombin 12.97 +/- 0.046

These measurements are significantly different (p= 1.215 x 10^"5; ANOVA, Duncan's post-hoc test), and indicate that there was a 37% stimulation of beta- galactosidase activity in response to thrombin. While not as great in magnitude as activation of the pathway by alpha factor binding to the Ste2p-Gpalp chimera

(Example 12b) , these results are consistent with the results in Example 13, and show that the two components of the Thrombin Receptor-Gpalp fusion protein can measurably couple with each other and to the yeast mating pathway. Further modification of the Gpalp domain of the chimera by methods such as that described in Example 15 should enable greater efficiency of coupling. In addition, the host strain can also be modified by random mutagenesis to reduce the background activation and to enhance the induction of the mating pathway. Mutants that provide hypersensitivity to mating factor are known (e.g. Chan et al, 1982, Mol. Cell. Biol. 2:21)

EXAMPLE 15: MUTAGENESIS OF THE Gpalp DOMAIN OF A FUSION PROTEIN TO ENABLE COUPLING OF THROMBIN RECEPTOR ACTIVATION TO THE MATING PATHWAY aj constructing a library of mutations: oligonucleotides "c" and "d" described in example 1 are used to amplify the GPAl gene from a wild type plasmid copy under PCR conditions shown to introduce mutations at a frequency of 6.6 per 1000 bases

(Cadwell and Joyce, 1994, in "PCR and its applications", CSHL Press, pp S136) . This method introduces transition and transversion mutations but not insertion or deletion mutations, thus maintaining the reading frame but randomizing the amino acid sequence. The method also has no significant sequence bias. The amplification product that contains individual molecules with single or multiple mutations is digested at the Mlul and PflMI sites present in the two primers. The plasmid pRMHBT20, which encodes the thrombin receptor fused in frame with Gpalp, is digested with Mlul and PflMI to release GPAl, and the fragment with the thrombin receptor is purified and ligated to the digested PCR product, and transformed into competent E. coli with the highest transformation efficiency that is commercially available. Each transformant carries a different mutation of the Gpalp domain of the fusion protein. The entire transformation mix is plated on large plates, and plasmids are isolated from these plate cultures. The maximum number of recombinants are needed to obtain the largest collection of mutations, and the above steps are repeated and the plasmid preparations pooled in proportion to the number of mutations represented in each pool until at least IO⁷ mutants are included in the library. The construction and use of mutant libraries of GPAl have been described previously (Stone and Reed (1990) Mol. Cell. Biol. 10:4439; Kurjan et al (1991) Genes Dev. 5:475) .

b) screening the library for functional Gpalp domains: The library is screened first for mutants that will still enable complementation of the gpal mutation, which will eliminate all mutants that do not enable interaction with G-beta/gamma. For this, a diploid yeast strain of genotype gpal/gpal is constructed by mating haploid gpal strains of opposite mating type in which the gpal mutation is complemented by a plasmid carrying GPAl. For example, the strain 9ALZ carrying pRMHBT44 (with a LYS2 selectable marker) is mated to any of the C7RA+ segregants in Example 7a or 7b that are of the alpha mating type. Both strains carry the reporter construct FUS1-LACZ integrated at the LEU2 locus (leu2 : :LEU2-FUSlp-LACZ) . Diploids are selected on ura-, lys- plates. The two plasmids are then eliminated by counterselection with 5-FOA and alpha- aminoadipate, which is possible because GPAl is required for growth only in haploids and not diploids.

This strain is transformed with the mutant library so that at least IO⁶ URA+ transformants are obtained, if necessary by repeated transformation experiments. The population of transformants is then sporulated, and random spores are germinated to yield at least 10^s individual colonies by standard genetic or chemical methods for random spore analysis (Rose et al, Methods in Yeast Genetics : A Labora tory Course Manual , c. 1990 by Cold Spring Harbor Laboratory Press.) . Only spores in which the Gpalp domain of the fusion can complement the gpal mutation can grow, thus selecting for mutants with Gpalp domains that can interact with G- beta/gamma.

c) screening for functional coupling of thrombin receptor activation to the mating pathway: this is achieved by growing the mutants selected from the previous step in the presence of thrombin and the dye X-gal, which is a substrate of the reporter lacZ gene. Functional coupling is selected for by induction of beta galactosidase, and consequent blue color formation. Note that because the reporter gene is present at both LEU2 loci in the diploid, all haploid segregants will have a functional reporter construct. Growth of such cells on plates containing thrombin and the dye X-gal causes blue color formation in colonies in which functional coupling is present between the receptor and Gpal domains. Others will remain white.

Claims

CLAIMS .What is claimed is:

1. A method for creating a yeast cell which expresses a fusion protein comprising a seven- transmembrane receptor protein of mammalian or fungal origin operatively linked at its carboxy- terminus to the amino terminus of a G_β protein of a non-mammalian organism so as to activate a pheromone-induced signal transduction pathway in said yeast cell upon binding of a ligand for said receptor to the receptor, which comprises: i) creating a DNA fragment encoding the seven- transmembrane receptor and the Gα protein fused at their respective carboxy- and amino-terminal ends or creating a DNA fragment encoding the seven- transmembrane receptor fused at its carboxy- terminal end to the amino terminal end of a linker peptide and a Gα protein fused at its amino terminal end to the carboxy-terminal end of said linker peptide, to obtain a DNA fragment encoding a fusion protein; ii) adding to the DNA fragment encoding said fusion protein additional nucleotides that encode additional amino acids effective for directing the fusion protein to the plasma membrane of said yeast upon expression of said fusion protein to obtain a DNA fragment encoding a plasma membrane-targeted fusion protein; iii) operatively linking said DNA fragment encoding a plasma membrane-targeted fusion protein to a promoter effective in said yeast for expressing said plasma membrane-targeted fusion protein from said DNA fragment encoding a plasma membrane-targeted fusion protein in said yeast to form a fusion protein expression construct; and iv) transforming said yeast with said fusion protein expression construct; and v) isolating a cell of yeast which expresses said fusion protein as a part of its plasma membrane.

2. The method of claim 1, wherein said yeast is Saccharomyces cervisiae .

3. The method of claim 2, wherein said G_α protein is encoded by the Saccharomyces cerevisiae gene GPAl .

4. The method of claim 2, wherein said seven- transmembrane receptor protein is selected from the group consisting of adenosine receptor Al, adenosine receptor A2, adrenergic receptor A2B, adrenergic receptor α-lA, adrenergic receptor α-lB, adrenergic receptor α-2A, adrenergic receptor α-2B, adrenergic receptor α-2C, adrenergic receptor β-1 , adrenergic receptor β- 1 , adrenergic receptor β-3 , amyloid protein precursor, angiotensin II receptor type 1, antidiuretic hormone receptor, bradykinin receptor, cannabinoid receptor, chemokine C-C (mip-l/RANTES) receptor, cholecystokinin receptor, complement C5a (anaphylotoxin) receptor, dopamine receptor Dl, dopamine receptor D2, dopamine receptor D3, dopamine receptor D4, dopamine receptor D5, endothelin receptor A, endothelin receptor B, f-met-leu-phe receptor, follicle stimulating hormone receptor, glutamate receptor (metabotropic) , gonadotropin-releasing factor receptor, growth hormone releasing hormone receptor, histamine H2 receptor, hydroxytryptamine (serotonin) receptor IA, hydroxytryptamine (serotonin) receptor IB, hydroxytryptamine (serotonin) receptor IC, hydroxytryptamine (serotonin) receptor ID, hydroxytryptamine (serotonin) receptor IE, hydroxytryptamine (serotonin) receptor 2, insulin-like growth factor II receptor, interleukin 8 receptor A, interleukin 8 receptor B, lutenizing hormone/chorionic gonadotropin receptor, mas proto-oncogene, melanocyte stimulating hormone receptor, muscarinic acetylcholine receptor ml, muscarinic acetylcholine receptor m.2 muscarinic acetylcholine receptor m3, muscarinic acetylcholine receptor m4, muscarinic acetylcholine receptor m5, neuropeptide Y receptor, opioid- δ receptor, opioid- K receptor, oxytocin receptor, platelet activating factor receptor, rhodopsin receptor, somatostatin receptor 1, somatostatin receptor 2, somatostatin receptor 3, substance K

(neurokinin A) receptor, substance P (NK1) receptor, thrombin receptor, thromboxane A2 receptor, thyroid stimulating hormone receptor and vasoactive intestinal peptide receptor.

5. The method of claim 2, wherein step iv) is performed using a vector that is an autonomously replicating vector.

6. The method of claim 2, wherein step iv) is performed using a vector that is a chromosomally- integrating vector.

7. A method for creating a yeast cell which expresses a fusion protein comprising a seven- transmembrane receptor protein of mammalian or fungal origin operatively linked at its carboxy- terminus to the amino terminus of a G_α protein of a non-mammalian organism so as to activate a pheromone-induced signal transduction pathway in said yeast cell upon binding of a ligand for said receptor to the receptor, which comprises: i) creating a DNA fragment encoding the seven- transmembrane receptor and the Gα protein fused at their respective carboxy- and amino-terminal ends to obtain a DNA fragment encoding a fusion protein; ii) adding to the DNA fragment encoding said fusion protein additional nucleotides that encode additional amino acids effective for directing the fusion protein to the plasma membrane of said yeast upon expression of said fusion protein to obtain a DNA fragment encoding a plasma membrane-targeted fusion protein; iii) operatively linking said DNA fragment encoding a plasma membrane-targeted fusion protein to a promoter effective in said yeast for expressing said plasma membrane-targeted fusion protein from said DNA fragment encoding a plasma membrane-targeted fusion protein in said yeast to form a fusion protein expression construct; iv) mutating the gene of said yeast homologous in function to the FARl gene of Saccharomyces cerevisiae to inactivate the protein homologous in function to Farlp of Saccharomyces cerevisiae in said yeast; v) constructing a diploid cell of said yeast, wherein said diploid cell has one wild- ype gene homologous in function to the GPAl gene of Saccharomyces cerevisiae and one inactivated copy of said gene homologous in function to the GPAl gene of Saccharomyces cerevisiae; vi) transforming said diploid cell of said yeast with the fusion protein expression construct; and vii) isolating a cell of yeast which expresses said fusion protein as a part of its plasma membrane.

δ. The method of claim 7, wherein said yeast is Saccharomyces cerevisiae, said gene of said yeast homologous in function to the FARl gene of Saccharomyces cerevisiae is FARl of Saccharomyces cerevisiae and said gene homologous in function to the GPAl gene of Saccharomyces cerevisiae is the GPAl gene of Saccharomyces cerevisiae .

9. The method of claim 8, wherein said seven- transmembrane receptor protein is selected from the group consisting of adenosine receptor Al, adenosine receptor A2, adrenergic receptor A2B, adrenergic receptor α-lA, adrenergic receptor α-lB, adrenergic receptor α-2A, adrenergic receptor α-2B, adrenergic receptor α-2C, adrenergic receptor β- 1 , adrenergic receptor β-1 , adrenergic receptor β-3 , amyloid protein precursor, angiotensin II receptor type 1, antidiuretic hormone receptor, bradykinin receptor, cannabinoid receptor, chemokine C-C (mip-1/RANTES) receptor, cholecystokinin receptor, complement C5a (anaphylotoxin) receptor, dopamine receptor Dl, dopamine receptor D2, dopamine receptor D3, dopamine receptor D4, dopamine receptor D5, endothelin receptor A, endothelin receptor B, f-met-leu-phe receptor, follicle stimulating hormone receptor, glutamate receptor (metabotropic) , gonadotropin-releasing factor receptor, growth hormone releasing hormone receptor, histamine H2 receptor, hydroxytryptamine (serotonin) receptor IA, hydroxytryptamine (serotonin) receptor IB, hydroxytryptamine (serotonin) receptor IC, hydroxytryptamine (serotonin) receptor ID, hydroxytryptamine (serotonin) receptor IE, hydroxytryptamine (serotonin) receptor 2, insulin-like growth factor II receptor, interleukin 8 receptor A, interleukin 8 receptor B, lutenizing hormone/chorionic gonadotropin receptor, mas proto-oncogene, melanocyte stimulating hormone receptor, muscarinic acetylcholine receptor ml, muscarinic acetylcholine receptor m2 muscarinic acetylcholine receptor m3, muscarinic acetylcholine receptor m4, muscarinic acetylcholine receptor m5, neuropeptide Y receptor, opioid- δ receptor, opioid- n receptor, oxytocin receptor, platelet activating factor receptor, rhodopsin receptor, somatostatin receptor 1, somatostatin receptor 2, somatostatin receptor 3, substance K (neurokinin A) receptor, substance P (NK1) receptor, thrombin receptor, thromboxane A2 receptor, thyroid stimulating hormone receptor and vasoactive intestinal peptide receptor.

10. The method of claim 8, which further comprises step: viii) sporulating a transformant obtained from step vii) and isolating a cell having a genotype analogous to a gpal , farl genotype.

11. The method of claim 8, which further comprises steps: viii) creating a second DNA construct comprising a promoter for a gene homologous in function to the

FUS1 gene of Saccharomyces cerevisiae operatively linked to a DNA fragment encoding a protein for measuring the activation of said promoter; ix) sporulating transformants obtained from step vii) to isolate a haploid cell having a genotype analogous to a gpal , farl genotype; x) transforming the haploid cell obtained in step ix) with the second DNA construct of step viii) ; and xi) isolating a haploid cell of yeast having a genotype analogous to a gpal , farl genotype which expresses said enzyme under the control of the promoter for the gene homologous in function to the

FUS1 gene of Saccharomyces cerevisiae and which expresses said fusion protein as a part of its plasma membrane.

12. The method of claim 11, wherein said protein for measuring the activity of said promoter is an enzyme for which a colorimetric assay can be used to measure the catalytic activity of the enzyme, for which an immunoassay can be used to measure the amount of said protein present in a sample or for which a biochemical selection can be performed to assay expression of the protein.

13. The method of claim 12, wherein said protein is selected from the group consisting of β- galactosidase, glucuronidase, green fluorescence protein, luciferase, alkaline phosphatase and invertase.

14. A cell of a yeast created according to the method of claim 1.

15. A cell of a yeast created according to the method of claim 7.

16. A cell of a yeast created according to the method of claim 10.

17. A cell of a yeast created according to the method of claim 11.

18. A haploid cell of a yeast having a genotype analogous to gpal , farl of Saccharomyces cerevisiae, which expresses as a part of the plasma membrane of said cell a fusion protein comprising a seven- transmembrane receptor protein attached by its carboxy-terminus to the amino terminus of a G_α protein of said yeast.

19. A cell according to claim 18, which is a cell of Saccharomyces cerevisiae having a genotype gpal , farl .

20. A haploid cell of a yeast having a genotype analogous to gpal, farl of Saccharomyces cerevisiae, which expresses as a part of the plasma membrane of said cell a fusion protein comprising a seven- transme brane receptor protein attached by its carboxy-terminus to the amino terminus of a G_α protein of said yeast and which further expresses a reporter gene for measuring the activity of a promoter of a gene homologous in function the FUS1 gene of Saccharomyces cerevisiae under the control of said promoter.

21. A cell according to claim 20, which is a cell of Saccharomyces cerevisiae having a genotype gpal , farl and wherein said promoter of a gene homologous in function the FUS1 gene of Saccharomyces cerevisiae is a promoter of the FUS1 gene of Saccharomyces cerevisiae .

22. A cell according to claim 21, wherein said reporter gene encodes a protein that is an enzyme for which a colorimetric assay can be used to measure the catalytic activity of the enzyme, for which an immunoassay can be used to measure the amount of said protein present in a sample or for which a biochemical selection can be performed to assay for expression of the protein.

23. A cell according to claim 22, wherein said protein is selected from the group consisting of β- galactosidase, glucuronidase, green fluorescence protein, luciferase, alkaline phosphatase and invertase.

24. A method for screening a compound for receptor agonist activity which comprises: i) contacting a yeast cell according to claim 20 with the said compound; ii) measuring the amount of expression of said reporter gene, to determine the activity of the promoter homologous in function to the promoter of the FUS1 gene of S. cerevisiae; and iii) comparing the activity of said promoter in said yeast cells contacted with said compound to the activity of said promoter in said yeast cells not contacted with said compound; wherein a compound is determined to be an agonist of said receptor if the activity of the promoter is higher in the cell contacted with said compound than in the cell not contacted with said compound.

25. A method for screening a compound for receptor antagonist activity which comprises: i) contacting a yeast cell according to claim 20 with the said compound and with a ligand for said receptor; ii) measuring the amount of expression of said reporter gene, to determine the activity of said promoter homologous in function to the promoter of the FUS1 in said yeast cell; and iii) comparing the activity of said promoter in said yeast cells contacted with said compound and said ligand to the activity of said promoter in said yeast cells contacted with said ligand but not contacted with said compound; wherein a compound is determined to be an antagonist of said receptor if the activity of the promoter is lower in the cell contacted with said compound and said ligand than in the cell contacted with said ligand and not contacted with said compound.

26. A recombinant DNA molecule encoding a fusion protein comprising a first polypeptide means for binding to a ligand and a second polypeptide means for binding to a yeast G/3γ complex, wherein said first polypeptide means is attached by its carboxyl terminus to the amino terminus of said second polypeptide means, and wherein said fusion protein productively interacts with the pheromone-induced signal transduction pathway of said yeast.

27. The recombinant DNA of claim 26 wherein said second polypeptide means is the Gpal protein of Saccharomyces cerevisiae or a mutant thereof which is selected by activation of the S. cerevisiae pheromone- induced signal transduction pathway upon ligand binding to said first polypeptide means.

28. The recombinant DNA of claim 26, wherein said first polypeptide means is a receptor having seven transmembrane domains.

29. The recombinant DNA of claim 27, wherein said first polypeptide means is a receptor having seven transmembrane domains.

30. The recombinant DNA of claim 28, wherein said first polypeptide means is a protein of a human selected from the group consisting of adenosine receptor Al, adenosine receptor A2, adrenergic receptor A2B, adrenergic receptor α-lA, adrenergic receptor α-lB, adrenergic receptor α-2A, adrenergic receptor α-2B, adrenergic receptor α-2C, adrenergic receptor β- 1 , adrenergic receptor β- 1 , adrenergic receptor β-3 , amyloid protein precursor, angiotensin II receptor type 1, antidiuretic hormone receptor, bradykinin receptor, cannabinoid receptor, chemokine C-C (mip-1/RANTES) receptor, cholecystokinin receptor, complement C5a (anaphylotoxin) receptor, dopamine receptor Dl, dopamine receptor D2, dopamine receptor D3, dopamine receptor D4, dopamine receptor D5, endothelin receptor A, endothelin receptor B, f-met- leu-phe receptor, follicle stimulating hormone receptor, glutamate receptor (metabotropic) , gonadotropin-releasing factor receptor, growth hormone releasing hormone receptor, histamine H2 receptor, hydroxytryptamine (serotonin) receptor IA, hydroxytryptamine (serotonin) receptor IB, hydroxytryptamine (serotonin) receptor ic, hydroxytryptamine (serotonin) receptor ID, hydroxytryptamine (serotonin) receptor IE, hydroxytryptamine (serotonin) receptor 2, insulin-like growth factor II receptor, interleukin 8 receptor A, interleukin 8 receptor B, lutenizing hormone/chorionic gonadotropin receptor, mas proto-oncogene, melanocyte stimulating hormone receptor, muscarinic acetylcholine receptor ml, muscarinic acetylcholine receptor m2 muscarinic acetylcholine receptor m3, muscarinic acetylcholine receptor m4, muscarinic acetylcholine receptor m5, neuropeptide Y receptor, opioid-δ receptor, opioid- K receptor, oxytocin receptor, platelet activating factor receptor, rhodopsin receptor, somatostatin receptor 1, somatostatin receptor 2, somatostatin receptor 3, substance K (neurokinin A) receptor, substance P (NK1) receptor, thrombin receptor, thromboxane A2 receptor, thyroid stimulating hormone receptor and vasoactive intestinal peptide receptor.

31. The recombinant DNA of claim 30, wherein said second polypeptide means is the Gpal protein of Saccharomyces cerevisiae or a mutant thereof which is selected by activation of the S. cerevisiae pheromone- induced signal transduction pathway upon ligand binding to said first polypeptide means.

32. A yeast cell transformed with the recombinant DNA of claim 26.

33. A yeast cell transformed with the recombinant DNA of claim 27 .

34. A yeast cell transformed with the recombinant DNA of claim 28.

35. A yeast cell transformed with the recombinant DNA of claim 29.

36. A yeast cell transformed with the recombinant DNA of claim 30.

37. A membrane preparation of a yeast cell transformed with the recombinant DNA of claim 26.

38. A membrane preparation of a yeast cell transformed with the recombinant DNA of claim 30.

39. A method for creating a recombinant DNA molecule encoding a fusion protein having a mammalian seven-transmembrane receptor polypeptide operatively- linked by its carboxy-terminus to Gpalp of S. cerevisiae, or a protein analogous in function to said Gpalp, whereby said fusion protein couples ligand binding by said receptor polypeptide to activation of a yeast pheromone-induced signal transduction pathway, which comprises: i) creating a DNA fragment encoding the seven- transmembrane receptor and the Gα protein fused at their respective carboxy- and amino-terminal ends or creating a DNA fragment encoding the seven- transmembrane receptor fused at its carboxy- terminal end to the amino terminal end of a linker peptide and a Gα protein fused at its amino terminal end to the carboxy-terminal end of said linker peptide, to obtain a DNA fragment encoding a fusion protein; ii) adding to the DNA fragment encoding said fusion protein additional nucleotides that encode additional amino acids effective for directing the fusion protein to the plasma membrane of said yeast upon expression of said fusion protein to obtain a DNA fragment encoding a plasma membrane-targeted fusion protein; iii) mutagenizing the GPAl domain of said DNA fragment encoding a plasma membrane-targeted fusion protein to obtain a pool of DNA fragments encoding mutant membrane targeted fusion proteins; iv) linking said DNA fragments encoding mutant plasma membrane-targeted fusion proteins to a promoter effective in said yeast for expressing said plasma membrane-targeted fusion protein from said DNA fragment encoding a plasma membrane- targeted fusion protein in said yeast to form a pool of mutant fusion protein expression constructs; v) mutating the gene of a yeast, said gene being homologous in function to the FARl gene of Saccharomyces cerevisiae, to inactivate the protein homologous in function to Farlp of Saccharomyces cerevisiae in said yeast; vi) constructing a diploid cell of said yeast, wherein said diploid cell has one wild-type gene homologous in function to the GPAl gene of Saccharomyces cerevisiae and one inactivated copy of said gene homologous in function to the GPAl gene of Saccharomyces cerevisiae; vii) transforming said diploid cell of said yeast with the pool of fusion protein expression constructs of step iv) ; viii) isolating a diploid cell of said yeast which expresses a mutant fusion protein as a part of its plasma membrane; ix) transforming said diploid cell of said yeast of step viii) with a vector for expressing a marker gene under control of a promoter homologous in function to the promoter of the FUS1 gene of S . cerevisiae, thereby obtaining a diploid cell of said yeast which will grow in a medium selective for the marker gene only when the pheromone-induced signal transduction pathway of said yeast is activated; x) sporulating the diploid cells of step ix) to identify a haploid cell that is gpal , farl genotype and harboring the reporter gene construct; xi) selecting a haploid cell of said yeast by culturing the transformants of step ix) in a medium selective for the marker gene, wherein said medium also contains the ligand for said receptor; and xii) cloning from said haploid cell of step xi) the DNA fragment encoding the mutant fusion protein.