Selective target cell activation by expression of a G protein-coupled receptor activated superiorly by synthetic ligand

[0001] This application is a continuation-in-part of U.S. patent application Ser. No. 09/341,446, filed Dec. 20, 1999, which is a continuation-in-part of earlier-filed U.S. application Ser. No. 08/622,348, filed Mar. 26, 1996, which applications are incorporated herein by reference.

[0003] This invention relates generally to the field of pharmacological control of cells expressing a protein or other molecule of interest.

[0004] Despite the tremendous advances in the field of genetic engineering and expression of a DNA of interest in a desired cell, few tools exist in the art for the direct, selective manipulation of a specific physiological response in a cell. For example, although cell growth can be facilitated in in vitro culture by inclusion of nutrients in the culture medium, there are few adequate means available to stimulate cell growth directly and selectively.

[0005] The problem is further complicated where the cell culture contains a desired cell type as well as a variety of undesired, contaminating cells. While the nutrients in the culture medium may enhance growth of the desired cell type, most often these nutrients also enhance growth of the undesired, contaminating cells, thus interfering with attempts to enrich the culture for the desired cell type. Some of these complicating factors might be overcome by transforming the desired cell type with an antibiotic resistance gene or other selective marker. However, the desired cells may be sensitive to the selective agent and grow more slowly in its presence, or the undesired, contaminating cells may be resistant to the selective agent and grow in the culture despite the presence of the agent.

[0006] In addition, there are many settings both in vitro and in vivo in which one wishes to enrich a cell population for a specific target cell, or elicit a specific physiological response in a target cell, without substantially affecting nontarget cells in the population. Gene therapy is an example of this latter situation. The basic concept in gene therapy involves the introduction of a DNA sequence encoding a protein or other molecule that, when expressed, can overcome a genetic defect associated with the disease, or produce some other therapeutic protein or other molecule that will either cure or ameliorate symptoms of the disease. The DNA introduced by gene therapy stably integrates into genome of host target cells, thus producing genetically altered cells expressing the desired therapeutic protein and thus treating the disease.

[0007] A critical limitation of current in vivo and ex vivo gene therapy efforts is the inability to amplify the number of transfected cells. In several instances, DNA encoding the therapeutic protein of interest is successfully delivered to and expressed in the target cells, but too few target cells are transformed to provide a measurable therapeutic response. If these few transformed cells could be specifically activated at the cellular level in vivo, e.g., to proliferate or to secrete their therapeutic product, the desired therapeutic effect could be achieved. This invention addresses this problem.

[0008] The invention features methods and compositions for selective cellular activation of a target cell. Targeted, transformed (genetically modified) cells expressing a modified G protein-coupled receptor that is activated superiorly by a synthetic ligand (RASSL) are selectively activated by synthetic small molecule binding to the RASSL. A RASSL is a modified G protein-coupled receptor having decreased binding affinity for a selected natural (i.e., endogenous) ligand (relative to binding of the selected ligand by a wild-type G protein-coupled receptor), but having normal, near normal, or enhanced binding affinity for a synthetic small molecule. Thus, RASSL-mediated activation of RASSL-expressing cells does not occur to a significant extent in the presence of the selected natural ligand, but responds significantly upon exposure to a synthetic small molecule.

[0009] In one embodiment, the RASSL-encoding DNA is introduced into the target cells to be activated, and can be cotransformed with DNA encoding a therapeutic protein of interest (e.g., cotransformed with a DNA of interest to be used in a gene therapy regimen). Cells expressing a RASSL are selectively activated by administration of an appropriate synthetic small molecule, which in turn binds the RASSL and facilitates activation of the G protein cascade and a selected physiological, cellular response (e.g., cellular proliferation, cellular secretion) in the RASSL-expressing cell.

[0010] The invention also features specific G protein-coupled receptors modified such that the receptor retains small molecule binding affinity, but is decreased in binding affinity of a selected natural ligand relative to a native G protein-coupled receptor from which the modified receptor is derived.

[0011] The invention further features a cellular implant comprising RASSL-expressing target cells that, upon binding of an appropriate synthetic small molecule ligand, exhibit a desired G protein-mediated cellular response (e.g., cellular proliferation or cellular secretion).

[0012] The invention additionally features a transgenic, reversible disease animal model. The transgenic animal expresses a RASSL that upon stimulation causes a disease- or condition-associated symptom in the animal.

[0013] A primary object of the invention is to provide a method to activate target cells selectively, particularly target cells expressing a desired protein or other molecule of interest. The desired protein or other molecule of interest can be encoded by genetic material that is either endogenous (e.g., naturally occurring) or heterologous (e.g., introduced by transformation or transfection) to the target cell genome.

[0014] Another object of the invention is to selectively activate target cells by amplifying cellular proliferation and/or cellular secretion.

[0015] Another object of the invention is to provide a method for enhancing the efficacy of gene therapy techniques by allowing controlled, targeted activation of cells transformed with a DNA of interest.

[0016] An advantage of the present invention is that targeted cellular activation can be controlled by exposing the target cells to varying amounts of the synthetic small molecule RASSL agonist.

[0017] Another advantage of the present invention is that, where the target cells are present in a mammalian host, the synthetic small molecule used to elicit a response in RASSL-expressing cells can be administered systemically (orally or intravenously), and need not be administered directly to the target cell site(s).

[0018] Yet another advantage of the invention is that the synthetic small molecules can cross the blood-brain barrier of a mammalian subject, thus allowing for activation of RASSL-expressing cells located in the mammalian host's brain tissue by oral or intravenous administration of the drug.

[0019] Another advantage of the invention is that RASSLs can be designed from G protein-coupled receptors which bind synthetic ligand that have been developed previously for use in mammalian subjects (e.g., for uses as such as pain relief, reduction of depression and weight reduction).

[0020] These and other objects, advantages and features of the present invention will become apparent to those persons skilled in the art upon reading the details of the vectors, cell lines and methodology as more fully set forth below.

[0021] FIG. 1A is a schematic showing the structure of G protein-coupled receptors and the position of their natural ligand and synthetic small molecule domains.

[0022] FIG. 1B is a schematic showing the structure of a RASSL appropriate for use in the present invention.

[0023] FIG. 2 is a schematic contrasting cell proliferation in nature and targeted proliferation accomplished by the method of the invention.

[0024] FIG. 3 is a schematic showing the structural formulae of the exemplary synthetic small molecule agonists of the kappa opioid receptor (KOR) (Pentazocine, DUP 747, CI-977, and Spiradoline), as well as exemplary synthetic small molecules that act as antagonists to a wild-type KOR, but can act as an agonist of a RASSL generated by a point mutation of the KOR (naltrexone and naloxone).

[0025] FIG. 4 is a schematic of the structural formulas of buspirone and methysergide, exemplary synthetic small molecule antagonists of the serotonin receptor.

[0026] FIG. 5 is a schematic illustrating RASSL-mediated targeted cellular proliferation in vivo.

[0027] FIG. 6 is a schematic showing the two-dimensional structure of, and specific amino acid alterations in, a RASSL derived from the kappa opioid receptor.

[0028] FIGS. 7A and 7B are schematics showing the amino acid and nucleotide sequences of wild-type human kappa opioid receptor having an N-terminal Prolactin signal, a FLAG epitope, and an HA tag. Sequences derived from the wild-type human kappa opioid receptor are shown as upper case letters.

[0029] FIGS. 8A and 8B are schematics showing the amino acid and nucleotide sequences of OR1, a RASSL of the invention, having an N-terminal Prolactin signal, a FLAG epitope, and an HA tag. Sequences derived from the wild-type human kappa opioid receptor are shown as upper case letters.

[0030] FIGS. 9A and 9B are schematics showing the nucleotide sequence of OR2, a RASSL of the invention, having an N-terminal Prolactin signal, a FLAG epitope, and an HA tag. Sequences derived from the wild-type human kappa opioid receptor are shown as upper case letters.

[0031] FIG. 10 is a graph showing proliferation of a RASSL-expressing cell in vitro. Open triangles wild-type kappa opioid receptor in the presence of dynorphin; closed triangles, the RASSL OR2 in the presence of dynorphin; open squares, wild-type kappa opioid receptors in the presence of spiradoline; closed squares, the RASSL OR2 in the presence of spiradoline.

[0032] FIGS. 11 (11-i and 11-ii) is a schematic showing an alignment of the amino acid sequences of the human receptors beta-2 adrenergic receptor (h₂AR) (SEQ ID NO:9), the serotonin-1A receptor (h5HT-1A-R) (SEQ ID NO:10), and the serotonin-2C receptor (SEQ ID NO:11). Asterisks indicates the conserved binding site for the natural biogenic mine ligand.

[0033] FIG. 12 is a graph showing the lethal effects of rapid expression of a KOR RASSL in the heart tissue of transgenic mice.

[0034] FIG. 13 is an EKG showing induction of cardiac arrhythmia by an opioid agonist in KOR RASSL-expressing transgenic mice.

[0035] FIG. 14 is a graph showing the development of signs of congestive heart failure upon gradual reduction of doxycycline in RASSL-expressing transgenic mice.

[0036] FIG. 15 is a schematic diagram showing the C-terminal amino acid sequences of exemplary RASSLs.

[0037] FIG. 16 provides the nucleotide sequence of the Ro3 (Rig) open reading frame and the Ro3 amino acid sequence.

[0038] FIG. 17 provides the nucleotide sequence of the Ro3A (RigA) open reading frame, and the Ro3A amino acid sequence.

[0039] FIG. 18 provides the nucleotide and amino acid sequence of the RigA RASSL.

[0040] FIGS. 19A and 19B are graphs showing internalization of a RASSL in response to spiradoline.

[0041] FIGS. 20A and 20B are graphs showing desensitization of Ro3 and Ro3A.

[0042] FIG. 21 depicts three major types of GPCR.

[0043] FIG. 22 is a schematic representation of a human melanocortin 4 receptor (hMC4R), depicting putative peptide binding sites and putative GRK/Arrestin binding sites.

[0044] FIG. 23 is a schematic representation of a human motilin receptor (hMoR), showing putative peptide binding sites and putative GRK/Arrestin sites.

[0045] FIGS. 24 (24i-24iii) is an amino acid sequence alignment of five serotonin receptors. Symbols (encircled stars) above amino acids indicate conserved binding site for natural ligand.

[0046] Before the present invention is further described, it is to be understood that this invention is not limited to particular embodiments described, as such may, of course, vary. It is also to be understood that the terminology used herein is for the purpose of describing particular embodiments only, and is not intended to be limiting, since the scope of the present invention will be limited only by the appended claims.

[0047] Where a range of values is provided, it is understood that each intervening value, to the tenth of the unit of the lower limit unless the context clearly dictates otherwise, between the upper and lower limit of that range and any other stated or intervening value in that stated range, is encompassed within the invention. The upper and lower limits of these smaller ranges may independently be included in the smaller ranges, and are also encompassed within the invention, subject to any specifically excluded limit in the stated range. Where the stated range includes one or both of the limits, ranges excluding either or both of those included limits are also included in the invention.

[0048] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. Although any methods and materials similar or equivalent to those described herein can also be used in the practice or testing of the present invention, the preferred methods and materials are now described. All publications mentioned herein are incorporated herein by reference to disclose and describe the methods and/or materials in connection with which the publications are cited.

[0049] It must be noted that as used herein and in the appended claims, the singular forms a, and, and the include plural referents unless the context clearly dictates otherwise. Thus, for example, reference to a transformed cell includes a plurality of such cells and reference to the synthetic small molecule includes reference to one or more synthetic small molecules and equivalents thereof known to those skilled in the art, and so forth.

[0050] The publications discussed herein are provided solely for their disclosure prior to the filing date of the present application. Nothing herein is to be construed as an admission that the present invention is not entitled to antedate such publication by virtue of prior invention. Further, the dates of publication provided may be different from the actual publication dates which may need to be independently confirmed.

[0051] G protein-coupled receptor means a receptor that, upon binding of its natural peptide or nonpeptide ligand and activation of the receptor, transduces a G protein-mediated signal(s) that results in a physiological, cellular response (e.g., cell proliferation or secretion). G protein-coupled receptors form a large family of evolutionarily related proteins (see Table 1, infra). Proteins that are members of the G protein-coupled receptor family are generally composed of seven putative transmembrane domains, and thus exhibit a structure similar to that shown in FIG. 1A. G protein-coupled receptors are also known in the art as seven transmembrane segment (7TM) receptors and as heptahelical receptors (see, e.g., Schwartz, 1994, Curr. Opin. Biotechnol. 5:434-444).

[0052] G protein means a protein belonging to a large family of proteins that interact with G protein-coupled receptors to facilitate cellular responses (e.g., directly or via cellular second messengers). G proteins are composed of a heterotrimer of three separate amino acid chains (G, G, and G). Although only the G subunit binds GTP, G protein refers to the complete heterotrimer.

[0053] G protein-coupled cellular response means a cellular response that occurs upon ligand binding by a G protein-coupled receptor. Such G protein-coupled cellular responses include, but are not limited to, cellular proliferation, cellular secretion, cell migration, cell contraction, neurotransmission, pigment production, and apoptosis. The G protein-coupled responses of cellular proliferation and cellular secretion are of particular interest in the present invention.

[0054] Receptor-ligand binding, ligand binding, and binding are used interchangeably herein to mean physical interaction between a receptor (e.g., a G protein-coupled receptor) and a ligand (e.g., a natural ligand, (e.g., peptide ligand) or synthetic ligand (e.g., synthetic small molecule ligand)). Ligand binding can be measured by a variety of methods known in the art (e.g., detection of association with a radioactively labeled ligand).

[0055] Signaling means the generation of a biochemical or physiological response as a result of ligand binding (e.g., as a result of synthetic ligand binding to a G protein-coupled receptor).

[0056] Receptor activation, RASSL activation, and G protein-coupled receptor activation are used interchangeably herein to mean binding of a ligand (e.g., a natural or synthetic ligand) to a receptor in a manner that elicits G protein-mediated signaling, and a physiological or biochemical response associated with G protein-mediated signaling.

[0057] Activation can be measured by measuring a biological signal associated with G protein-related signals (e.g., cell proliferation), or measurement of a second messenger affected by such signals (e.g., cAMP, calcium flux).

[0058] Targeted cellular activation and target cell activation are used interchangeably herein to mean RASSL-mediated activation of a specific G protein-mediated physiological response in a target cell, where RASSL-mediated activation occurs by binding of a synthetic small molecule to the RASSL. As used herein, cellular activation includes any of a variety of physiological processes such as cell proliferation, cell secretion, cell migration, cell contraction, pigment production and apoptosis. Activation of cell proliferation and/or secretion are of particular interest in the targeted cellular activation method of the invention.

[0059] Receptor activated superiorly by synthetic ligand, RASSL, and modified G protein-coupled receptor are used interchangeably herein to mean a modified G protein-coupled receptor that 1) retains binding affinity for a synthetic small molecule and 2) has decreased binding affinity for a selected natural ligand (e.g., an endogenous, or naturally occurring, peptide or nonpeptide ligand) relative to binding by its corresponding wild-type G protein-coupled receptor. Binding of the small molecule ligand to the RASSL elicits (or suppresses) a G protein-pharmacological response in the RASSL-expressing cell. RASSLs also include modified G protein-coupled receptors that bind both a selected naturally occurring peptide or nonpeptide ligand as well as a synthetic small molecule, but have a higher affinity for small molecule binding, or are only significantly activated by binding of the synthetic small molecule. A schematic of an exemplary RASSL of the invention is shown in FIG. 1B.

[0060] Natural ligand and naturally occurring ligand of a native G protein-coupled receptor are used interchangeably herein to mean a biomolecule endogenous to a mammalian host, which biomolecule binds to a native G protein-coupled receptor to elicit (or suppress) a G protein-coupled cellular response. Natural peptide ligand means a natural ligand that is a polypeptide. The peptide generally binds to the native G protein-coupled receptor by interaction with an extracellular receptor domain(s) (e.g., the N-terminus, e1, e2, and/or e3) of the G protein-coupled receptor. Natural biogenic amine ligand means a natural ligand that is a biogenic amine (e.g., acetylcholine, serotonin, adrenaline, histamine, dopamine). Biogenic amines bind to the G protein-coupled receptors at a common binding site that involves at least four conserved residues on the receptor.

[0061] Synthetic small molecule, synthetic small molecule ligand, and synthetic ligand are used interchangeably herein to mean any compound made exogenously by natural or chemical means that can bind within the transmembrane domains of a G protein-coupled receptor or modified G protein-coupled receptor (i.e., RASSL) and facilitate activation (or suppression) of the receptor and concomitant activation (or suppression) of a desired family of G proteins. In general, synthetic small molecule ligands of use in the present invention have a molecular weight of from about 100 Da to 1000 Da, and can range from about 40 Da (e.g., calcium) to about 4000 Da.

[0062] Cellular implant means a cell population comprising target cells expressing a RASSL of interest. Cellular implants are preferably primarily composed of RASSL-expressing target cells, but may include nontarget cells that do not express a RASSL, providing that the presence of such cells does not substantially interfere with the desired cellular response associated with ligand binding to the RASSL of the target cells cellular implants of the invention are suitable for introduction into a mammalian host for treatment of a disease or condition of the mammalian host (e.g., for secretion of a therapeutic product and/or growth of target cells to replace or supplement cells of similar origin in the mammalian host).

[0063] Cell population means a collection of living cells that can be grown in vitro and/or in vivo. Cell populations can be either homogenous or heterogenous with respect to cell type and RASSL expression.

[0064] Therapeutic cellular product means a cellular product useful in the treatment of a disease or condition of a mammalian host. Therapeutic cellular products include, but are not limited to, polypeptides, hormones, steroids, and other biological molecules that can be produced by a mammalian cell (e.g., by expression of nucleic acid either endogenous or heterologous to the mammalian cell). Nucleic acid encoding a therapeutic cellular product that is heterologous to the mammalian cell can be introduced into the cell by a variety of transformation or transfection techniques.

[0065] Transformation means a permanent genetic change induced in a cell following incorporation of new DNA (i.e., DNA exogenous to the cell). Where the cell is a mammalian cell, the permanent genetic change is generally achieved by introduction of the DNA into the genome of the cell.

[0066] The terms transformed cell and genetically modified cell are used interchangeably herein to refer to a cell into which (or into an ancestor of which) has been introduced, by means of recombinant DNA techniques, a DNA molecule encoding a molecule (e.g., RNA and/or protein) of interest (e.g., nucleic acid encoding a RASSL or a therapeutic cellular product).

[0067] DNA of interest means any DNA sequence that encodes a protein or other molecule which is desirable for expression in a target cell (e.g., for production of the protein or other biological molecule (e.g., a therapeutic cellular product) by the target cell in vitro or in vivo). The sequence encoding a DNA of interest is generally operatively linked to other sequences which are needed for its expression, e.g., a promoter.

[0068] Promoter means a minimal DNA sequence sufficient to direct transcription of a DNA sequence to which it is operably linked. Promoter is also meant to encompass those promoter elements sufficient for promoter-dependent gene expression controllable for cell-type specific, tissue-specific or inducible by external signals or agents; such elements may be located in the 5 or 3 regions of the native gene.

[0069] Mammalian host, mammalian subject and mammalian patient are used interchangeably herein to mean any mammal in which targeted cellular activation according to the invention is desired, including human, bovine, equine, canine, and feline subjects.

[0070] Transgenic organism means a nonhuman organism (e.g., single-cell organisms e.g., yeast), mammal, nonmammal (e.g., nematode or Drosophila)) having a nonendogenous (i.e., heterologous) nucleic acid sequence stably integrated into its germ line DNA (e.g., in the genomic sequence of most or all of its cells, or at least a number of cells sufficient to achieve a desired effect in the present invention).

[0071] Transgenic animal means a nonhuman animal, usually a mammal, having a nonendogenous (i.e., heterologous) nucleic acid sequence stably integrated into its germ line DNA (i.e., in the genomic sequence of most or all of its cells, or at least a number of cells sufficient to achieve a desired effect in the present invention). Heterologous nucleic acid is introduced into the germ line of such transgenic animals by genetic manipulation of, for example, embryos or embryonic stem cells of the host animal.

[0072] Reversible disease transgenic animal model and reversible condition transgenic animal model are used interchangeably herein to mean a RASSL-expressing transgenic animal that exhibits a symptom(s) associated with a disease or condition. In one embodiment, the transgenic animal of such reversible disease or conditions models does not exhibit the disease or condition symptom(s) in the absence of the small molecule ligand.

[0073] Overview of Targeted Proliferation Method of the Invention

[0074] The present invention provides a method of targeted activation (or inactivation) of a response of a selected cell population by virtue of the expression of a modified G protein-coupled receptor. A modified G protein-coupled receptor is genetically engineered so that it 1) retains binding affinity for a synthetic ligand (e.g., a synthetic small molecule ligand) and 2) has decreased binding affinity for a selected naturally occurring peptide or nonpeptide ligand (e.g., an endogenous ligand) relative to binding by its corresponding wild-type G protein-coupled receptor (e.g., the G protein-coupled receptor from which the modified G protein-coupled receptor was derived). Synthetic small molecule binding to the modified receptor induces the target cell to respond with a specific physiological response (e.g,. cellular proliferation, cellular secretion, cell migration, cell contraction, or pigment production). Such modified G protein-coupled receptors are referred to herein as Receptors Activated Superiorly by a Synthetic Ligand, or RASSLs. The synthetic ligand binds at a site that is distinct (not identical) from the binding site of the endogenous ligand.

[0075] The method of the invention exploits the RASSL characteristics described above to facilitate responses of RASSL-expressing target cells selectively. For example, when RASSL-expressing cells are present in a cell population that expresses neither a RASSL nor the wild-type G protein-coupled receptor from which the RASSL was derived, exposure of the cell population to the appropriate synthetic ligand that selectively binds to RASSL-expressing cells results in selective proliferation of the RASSL-expressing cells. This method, broadly referred to herein as targeted cellular activation, provides a method to amplify selectively a desired cell type in a cell population, thereby providing greater numbers of the desired target cell relative to that afforded by natural proliferation processes (FIG. 2). Furthermore, because other cells in the population do not express the RASSL (or the native G protein-coupled receptor from which the RASSL was derived), or because the wild-type G protein-coupled receptor is expressed in cells where the cellular response is benign or distinguishable from the response in the target cell, exposure to the synthetic small molecule does not result in activation of nontarget cells (i.e., the desired physiological or biochemical response is not elicited in the nontarget cells upon exposure to synthetic ligand). The presence of natural ligand generally does not substantially affect the ability of the small molecule to bind the RASSL as natural ligand binding to G protein-coupled receptors is usually transient.

[0076] RASSLs are useful to facilitate targeted cellular activation (or suppression) in a variety of in vitro and in vivo settings. For example, RASSLs can be used to induce targeted, proliferative amplification of transfected cells in in vitro culture, e.g., as a means for selecting for cells that express a RASSL and respond with a desired physiological response upon RASSL stimulation, and/or as a means for expanding a desired cell type from a primary cell culture. In addition, RASSL stimulation is useful in increasing secretion of a cellular product of interest by cells in vitro or in vivo, to elicit proliferation of cells transformed during gene therapy and to gain pharmacological control in vivo of transformed cells expressing therapeutic genes. Still further, RASSL stimulation is useful in decreasing (e.g. suppressing) a response, such as a pain response. Additional uses of the method and compositions of the invention are described in detail below (see Utilities).

[0077] G Protein-Coupled Receptors (Background)

[0078] G protein-coupled receptors, which form a large family of evolutionarily related proteins, are encoded by over 600 different, currently described genes from diverse species (see Table 1 for exemplary G protein-coupled receptors families). By some estimates, there are over 1000 separate genes encoding G protein-coupled receptors in the human genome alone, accounting for 1-2% of all human genes. G protein-coupled receptors are classified according to the type of ligand bound by the receptor (e.g., peptide receptors, natural nonpeptide small molecule receptors, and orphan receptors (i.e., receptors for which the natural ligand is not presently known)). For a complete list of currently known G protein-coupled receptors, see the internet at one or more of the following sites on the World Wide Web:

[0079] gpcr.org/7tm

[0080] us.expasy.org/sprot/sprot-top.html

[0081] tinygrap.uit.no/

[0082] pdsp.cwru.edu/stestw.asp.

[0083] Each ligand-binding receptor subclass shown in Table 1 represents a G protein-coupled receptor subfamily comprising up to 50 genes. G protein-coupled receptors facilitate regulation of cell growth in a wide variety of tissues.

TABLE 1
G protein-coupled Receptors
G protein-coupled receptor families having peptide	Synthetic Small
natural ligands	Molecule Agonist
Angiotensin receptors
Bombesin receptors
Bradykinin receptors
Calcitonin, parathyroid hormone, secretin receptors
Chemokine receptors
Chemotactic peptide receptors (fMLP)
C5A receptor
Cholecystokinin/gastrin receptors
Corticotropin (ACTH) receptor
Endothelin receptors
Glycoprotein hormones receptors (TSH, FSH, LH)
Melanocortin receptors
Motilin receptors
Neuropeptide Y receptors
Neurotensin receptors
Oploid receptors:
mu opioid receptor
delta opioid receptor
kappa opioid receptor
nociceptin, opioid receptor
Releasing hormone receptors (LHRH, GHRH)
Somatostatin receptors
Tachykinin receptors
Thrombin/protease receptors
Vasopressin/oxytocin receptors
G protein-coupled receptor families having small	Synthetic Small
molecule natural ligands (biogenic Amines)	Molecule Agonist
Acetylcholine (muscarinic) receptors
Adrenergic receptors
Dopamine receptors
Histamine receptors
Serotonin receptors
G protein-coupled receptor families having natural
small molecule agonist ligands:	Synthetic Small
lipids and other nonamines	Molecule Agonist
Adenosine and other adenine nucleotide receptors
Cannabinoids receptors
Prostanoids and PAF receptors
Metabotropic glutamate
Calcium receptor
Synthetic Small
Other Receptor Families	Molecule Agonist
Odorant/olfactory and gustatory receptors
Opsins
Viral receptors
Orphan receptors

[0084] G protein-coupled receptors share a high degree of structural and functional homology. For example, G protein-coupled receptors that bind peptides as their natural ligands are composed of seven transmembrane helices (TMHs) separated by three intracellular loops (i1, i2, and i3), the N-terminus, and three extracellular loops (e1, e2, e3) (FIG. 1A). Peptides bind the receptor's extracellular loops, while small molecules bind the receptor's transmembrane domains (see, e.g., Schwartz, supra; Cascieri et al., 1995, J. Pharmacol. Toxicol. Meth. 33:179-185). The intracellular domains of the receptor interact with G proteins and are responsible for generating specific intracellular signals associated with cell growth. The structure and function of G protein-coupled receptors are the subjects of several recent reviews (Conklin et al., 1993, Cell 73:631-41; Schwartz, 1994, supra; Cascieri et al., supra). Because peptide binding, small molecule binding, and G protein interaction are mediated by distinct regions of G protein-coupled receptors (the extracellular loops, TMH region, and intracellular loops, respectively), G protein-coupled receptors can be modified to decrease or even completely eliminate binding of a selected natural ligand without substantially affecting either small molecule binding or interaction of the receptor with G proteins.

[0085] In contrast to G protein-coupled receptors, G proteins have a somewhat more limited repertoire. The same essential 4 major classes of G proteins (G_S, G₁, G_q and G₁₂) are expressed in all mammalian cell types (e.g., in every cell type of the human body). Furthermore, the cellular response elicited by activation of these different G protein classes varies from cell type to cell type. Stated differently, the signals transmitted to the cell by activation of these 4 different G protein families have different meanings within the various cellular environments. For example, proliferation of thyroid-forming cells is associated with G_S activation, while G_i activation is also associated with altering neurotransmission in the brain or slowing of the heart rate. G_q activation is associated with proliferation of hepatic cells, and certain secretory cells of the lung, breast and stomach (Dhanasekaran et al., 1995, Endocrine Rev. 16:259-270). Table 2 provides several examples of the relationship between a stimulus (e.g., natural G protein-coupled receptor ligand), the cell type affected, the G protein subfamily activated, the effector (e.g., second messenger) molecule that mediates the response, and the cellular response. The various G protein families and the various G protein-mediated physiological responses are the subjects of several reviews (see, e.g., Dhanasekaran et al., 1995, Endocrine Rev. 16:259-270; Linder et al., July 1992, Sci. Amer. 267:56-61; Shenker, 1995, Baillire's Clin. EndocrinoL Metab. 9:427-451; Spiegel, 1992, Curr. Opin. Cell Biol. 4:203-211; Coughlin, 1995, Curr. Opin. Cell Biol. 6:191-197).

[0086] The specificity of G protein activation is mediated at the level of the G protein-coupled receptor. Each G protein-coupled receptor activates a distinct G protein subset (class) which elicits different physiologic effects depending on the cell type. Of the four major classes of G proteins (G_S, G₁, G_q and G₁₂), each class is associated with cell proliferation in a different cell type. Through selective stimulation of members of this limited set of G proteins, the body thus efficiently regulates many different signals and cellular responses.

TABLE 2
Relationship Between Stimulus, G protein Families,
and Physiological or Biochemical Effect
	Affected	G
Stimulus	Cell Type	Protein	Effector	Effect
Epinephrine,	Liver cells	Gs	Adenylyl cyclase	Breakdown of
glucagon				glycogen
Epinephrine,	Fat cells	Gs	Adenylyl cyclase	Breakdown of fat
glucagon
Luteinizing	Ovarian follicles	Gs	Adenylyl cyclase	Increased
hormone				estrogen and
				progesterone
Thyroid	Thyroid cells	Gs	Adenylyl cyclase	Proliferation of
stimulating				thyroid cells
hormone
Chemokines	Lymphocytes	G1	Adenylyl cyclase	Chemotaxis;
				proliferation
Acetylcholine	Heart muscle	Gi	Calcium and	Slowed heart rate
	cells		potassium channels,	and decreased
			adenylyl cyclase	pumping force
Enkephalins,	Brain neurons	Gi/Go	Calcium and	Changed
endorphins,			potassium channels,	electrical activity
oploids			adenylyl cyclase	of neurons
Angiotensin	Smooth muscle	Gq	Phospholipase C	Muscle
Stimulus	Affected	G	Effector	Effect
	cells in blood			contraction;
	vessels			elevation of
				blood pressure
Vasopressin	Liver cells	Gq	Phospholipase C	Proliferation
	(hepatocytes)
Odorants	Nose neuro-	Golf	Adenylyl cyclase	Detection of
	epithelial cells			odorants
Light	Rod and cone	Gt	Cyclic GMP	Detection of
	cells of the retina		phosphodiesterase	visual signals
Pheromone	Baker's yeast	GPA1	Unknown	Mating of cells

[0087] Construction of RASSLs

[0088] Any G protein-coupled receptor having separable domains for: 1) natural ligand (e.g., peptide ligand or biogenic amine) binding; 2) synthetic small molecule binding; and 3) G protein interaction can be modified to produce a RASSL for use in the method of the invention. In general, any G protein-coupled receptor that is a member of the G protein-coupled receptor family may be used to generate a RASSL of the invention. Exemplary G protein-coupled receptors having these characteristics include, but are not limited to, those receptors listed in Table 1.

[0089] Of particular interest in many embodiments are those G protein-coupled receptors having a peptide as its natural ligand, such as those G protein-coupled receptors under the subheading Peptide Ligands in Table 1. G protein-coupled receptors that bind peptide as their natural ligand, and are of particular interest for modification to produce a RASSL of the invention, include, but are not limited to: Type-1 Angiotensin II, Type-1a Angiotensin II, Type-1B Angiotensin II, Type-1C Angiotensin II, Type-2 Angiotensin II Receptor, Neuromedin-B, Gastrin-releasing Peptide, Bombesin Subtype-3, B1 Bradykinin, B2 Bradykinin, Interleukin-8 A, Interleukin-8 B, FMet-Leu-Phe, Monocyte Chemoattractant Protein 1, C-C Chemokine Receptor Type 1, C5a Anaphylatoxin, Cholecystokinin Type A, Gastrin/cholecystokinin Type B, Endothelin-1, Endothelin B, Follicle Stimulating Hormone (FSH-R), Lutropin-choriogonadotropic Hormone (LH/CG-R), Adrenocorticotropic Hormone (ACTH-R), Melanocyte Stimulating Hormone (MSH-R), Melanocortin-3, Melanocortin-4, Melanocortin-5, Melatonin Type 1A, Melatonin Type 1B, Melatonin Type 1C, Neuropeptide Y Type 1, Neuropeptide Y Type 2, Neurotensin, Delta-type Opioid, Kappa-type Opioid, Mu-type Opioid, Probable Opioid Receptor (now called Opioid-like or Nociceptin Receptor), Gonadotropin-releasing Hormone, Somatostatin Type 1, Somatostatin Type 2, Somatostatin Type 3, Somatostatin Type 4, Somatostatin Type 5, Substance-P, Substance-K, Neuromedin K, Vasopressin V1a, Vasopressin V1B, Vasopressin V2, Oxytocin, Galanin, Calcitonin, Calcitonin A, Calcitonin B, Growth Hormone-releasing Hormone, Parathyroid Hormone/parathyroid Hormone-related Peptide, Pituitary Adenylate Cyclase Activating Polypeptide Type I, Secretin, Vasoactive Intestinal Polypeptide 1, and Vasoactive Intestinal Polypeptide 2. Biogenic amine receptors, especially the serotonin receptors, are also of particular interest for production of RASSLs according to the invention. Exemplary biogenic amine receptors are provided in Table 1.

[0090] As depicted in FIG. 21, GPCR can be categorized according to the G protein pathway through which they signal, e.g., G_S, G₁, or G_q. The melanocortin (MCR) receptor, the kappa opioid receptor (KOR), and the motilin receptor (MoR) are prototype GPCR that signal via G_S, G₁, or G_q, respectively. Examples of amino acids that may be mutated to generate a melanocortin RASSL are shown in FIG. 22. Peptide binding amino acids are highlighted. One or more peptide binding amino acids are mutated to generate a melanocortin RASSL. Similarly, examples of amino acids that may be mutated to generate a MoR RASSL are shown in FIG. 23. One or more peptide binding amino acids are mutated to generate a MoR RASSL. Because the KOR, MCR, and MoR are prototype GPCR, the information provided herein as to the generation of a RASSL from a prototype GPCR provides guidance as to which amino acids, when mutated, will give rise to a RASSL in other GPCRs.

[0091] Modification of native G protein-coupled receptors to produce RASSLs of the invention can be accomplished using a variety of recombinant techniques well known in the art (see, e.g., Sambrook et al. Molecular Cloning: A Laboratory Manual, 2nd Ed., 1989, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y.). In general, RASSLs are generated by first cloning cDNA encoding a native G protein-coupled receptor in an expression vector according to methods well known in the art. For example, G protein-coupled receptor-encoding DNA can be isolated by PCR amplification from a cDNA library and the PCR product subsequently subcloned into a mammalian expression vector using standard techniques. Several G protein-coupled receptors have been cloned and sequenced.

[0092] The amino acid sequence of the G protein-coupled receptors is determined using standard sequencing and/or DNA manipulation software (e.g., DNA Strider, MacVector, Geneworks, GCG, Eugene). The amino acid sequence is analyzed to determine the position of the seven transmembrane segments characteristic of G protein-coupled receptors. Due to the structural homology of G protein-coupled receptors, the position of these segments can be determined by comparison to the amino acid sequences and three-dimensional structures of other G protein-coupled receptors. Predicted transmembrane domains of several G protein-coupled receptors are known in the art, and are available in several public domain World Wide Web sites. See, for example, the internet at one or more of the following sites on the World Wide Web:

[0093] gpcr.org/7tm

[0094] us.expasy.org/sprot/sprot-top.html

[0095] tinygrap.uit.no/

[0096] pdsp.cwru.edu/stestw.asp.

[0097] Additional predictions about the structural domains of the native G protein-coupled receptors are available publicly on the SWISS-PROT database, which can be accessed on the above sites. Methods for predicting three-dimensional structures of any G protein-coupled receptor are known in the art. In general, these methods predict G protein-coupled receptor structure by identification of the position of specific residues shared by subfamilies of G protein-coupled receptors. In addition, these methods analyze the amino acid sequence to identify stretches of hydrophobicity predictive of the characteristic seven transmembrane segments. For a review, see Baldwin, 1994, Curr. Opin. Cell Biol. 6:180-190.

[0098] G protein-coupled receptors are generally composed of seven transmembrane helices (TMHs) separated by three intracellular loops (i1, i2, and i3), the N-terminus, and three extracellular loops (e1, e2, e3) (FIG. 1A). The four extracellular segments of the receptor (e1, e2, e3, and the N-terminus) contain the sequences that facilitate natural ligand binding, but are generally not involved directly in small molecule binding (Schwartz, supra). These four extracellular segments are characterized as follows: 1) The N-terminus, which includes all residues before the first transmembrane segment, and is usually about 10-600 residues in length; 2) the first extracellular loop (e1), which is defined by the residues between the second and third transmembrane domains and is usually from about 7 to 20 amino acid residues in length; 3) the second extracellular loop (e2), which is defined by the residues between the fourth and fifth transmembrane domains and is usually between 10-35 residues in length; 4) the third extracellular loop (e3), which is defined by the residues between the sixth and seventh transmembrane domains and is usually between 7-20 residues in length.

[0099] RASSLs are generated from the native G protein-coupled peptide receptor by mutation of one or more residues in at least one of the extracellular segments (e1, e2, e3, and/or the N-terminus) of the receptor. Mutations include, but are not limited to, nucleotide substitutions, additions, deletions, and/or modification so as to produce an amino acid alteration (e.g., a conservative or nonconservative amino acid substitution, preferably nonconservative).

[0100] Where the RASSL is made from a peptide receptor, mutagenesis is targeted to the extracellular loops of the receptor. Where the RASSL is made from a biogenic amine receptor, mutagenesis is targeted at the known binding sites of the biogenic amines located at three discrete sites in the transmembrane domains of the receptor (see, e.g., Schwartz, supra; Strader et al., 1991, J. Biol. Chem. 266(1):5-8; Strader et al., 1994, Ann. Rev. Biochem. 63:101-32; Cascieri et al., 1995, J. Pharmacol. Toxicol. Meth. 33:179-185; Schwartz, 1994, Curr. Opin. in Biotech. 5:434-444; Baldwin, 1994, Curr. Opin. in Cell Biol. 6:180-190). Mutations at these discrete sites dramatically decrease binding by the natural biogenic amines, while preserving binding and activation by synthetic small molecule drugs (Strader et al. 1991 J. Biol. Chem. 266:5-8).

[0101] Alternatively, or in addition, RASSLs can be generated by the production of chimeric proteins composed of sequences encoding two different G protein-coupled receptors, each having distinct natural ligand binding and/or small molecule binding characteristics (e.g., bind natural ligand with different affinities, bind different natural ligands, bind the same or different small molecule ligands). Methods for nucleic acid manipulation are well known in the art (see, e.g., Sambrook et al., supra).

[0102] Changes and/or modifications in the amino acid sequence of a native G protein-coupled receptor can comprise a single amino acid change and may comprise deletion and/or substitution of over 600 residues in receptors with large extracellular domains such as the TSH receptor. In general, the RASSL mutants contain at least one amino acid change, usually at least 2 to 5 amino acid changes, generally about 5 to 10 amino acid changes, and normally 10 to 20 amino acid changes and can contain 50 amino acid changes or more. Preferably, the number of amino acids altered in the RASSL relative to the native G protein-coupled receptor from which it is derived is a number sufficient to provide the desired RASSL characteristics (e.g., decreased natural ligand binding relative to small molecule binding) without significantly adversely affecting the ability of the RASSL to bind synthetic small molecule ligand and elicit the desired cellular response.

[0103] Synthetic small molecule ligands that stimulate the native G protein-coupled receptor, and thus can be used to activate the RASSL derived from that native receptor, are well known in the art (see, e.g., Freidinger, 1993, Prog. Drug Res. 40:33-98), or can be identified using the exemplary screening assays as described below. Preferably, the small molecule ligand is a high affinity agonist or antagonist, more preferably agonist, for the RASSL.

[0104] Preferably, a RASSL is modified with respect to its corresponding native G protein-coupled receptor in that the RASSL exhibits binding for a selected natural ligand that is decreased, preferably substantially decreased, more preferably substantially eliminated, relative to binding of the ligand by its corresponding native G protein-coupled receptor. Therefore, RASSL activity is relatively unaffected by natural fluctuations of the selected natural ligand (e.g., natural hormones). RASSL binding of the selected natural ligand is decreased by at least 5-fold, preferably 10-fold, more preferably 50-fold, still more preferably 75-fold, and may be decreased 100-fold or more relative to binding by the RASSL's corresponding native G protein-coupled receptor.

[0105] Alternatively, RASSLs suitable for use in the invention may retain binding to the selected natural ligand, but exhibit reduced activation upon such natural ligand binding.

[0106] Such RASSLs may have decreased responsiveness to the selected natural ligand, yet still retain superior responsiveness to binding of appropriate synthetic small molecule ligand. Alternatively, the RASSL can be modified such that a small molecule ligand that acts as an antagonist for the corresponding native G protein-coupled receptor instead acts as an agonist upon binding the RASSL.

[0107] The RASSL binds the appropriate synthetic small molecule with the same, increased, or similar affinity relative to the native G protein-coupled receptor. Preferably, a RASSL binds its synthetic small molecule with equivalent, nearly equivalent, or greater affinity than its corresponding native G protein-coupled receptor. Most importantly, the RASSL is superiorly activated by the small molecule drugs as compared to the natural ligand (i.e., activated to a greater or more significant extent by binding of the small molecule ligand than by binding to a selected natural ligand at a similar concentration. Preferably, RASSL binding to the selected natural ligand is decreased relative to selected natural ligand binding by the native G protein-coupled receptor from which the RASSL is derived, and RASSL small molecule binding is relatively unchanged relative to small molecule binding by the native G protein-coupled receptor.

[0108] In general, the RASSL expressed in the target cell is derived from a native G protein-coupled receptor that is not normally expressed in the target cell. The RASSL thus provides a means to elicit a response (e.g., activation or suppression) in a desired subfamily of G proteins without affecting activity of other G protein-coupled receptors. For example, while nontarget cells in the cell population may express other G protein-coupled receptors capable of activating the desired G protein subfamily, these endogenous G protein-family receptors are not activated by the small molecule ligand that activates the RASSL expressed in the target cells. Alternatively, the native G protein-coupled receptors capable of binding the small molecule ligand activate a G protein subfamily that in turn elicits a physiological or biochemical response distinguishable from the response desired in the target cells (e.g., exposure to the small molecule may cause pain relief (or some benign response) in nontarget cells, but elicit the desired cellular response in the target cell type). Because there are many synthetic small molecule drugs that can be used as RASSL ligands, it is possible to choose a small molecule drug that is known to have benign effects on people or animals when acting on native G protein-coupled receptors. The G protein-mediated responses that result from activation of introduced RASSLs are the responses of interest in the present invention.

[0109] RASSLs can also be characterized by the ratio of synthetic ligand binding affinity to binding affinity of a selected natural ligand. Preferably, RASSLs of the invention exhibit a high small molecule ligand binding to selected natural ligand binding ratio, and exhibit small molecule ligand:selected natural ligand binding ratios of at least 0.8, preferably at least 1.0, more preferably at least 5, even more preferably 10, still more preferably 100 or higher. Preferably, RASSLs exhibit binding ratios that are 2-fold greater, preferably 5-fold greater, more preferably 10-fold greater, even more preferably 50- to 100-fold greater than the small molecule ligand:selected natural ligand binding ratio of a native G protein-coupled receptor.

[0110] RASSLs can also be characterized by the ratios of the level of activation by exposure to synthetic ligand to the level of activation by exposure to a selected natural ligand (activation ratio). Activation levels can be measured by measuring a biochemical or physiological signal (e.g., cell proliferation), or a second messenger molecule associated with such a signal (e.g., cAMP). Preferably, RASSLs of the invention exhibit a high small molecule ligand activation to selected natural ligand activation ratio, and exhibit small molecule ligand:selected natural ligand activation ratios of at least 0.8, preferably at least 1.0, more preferably at least 5, even more preferably 10, still more preferably 100 or higher. Preferably, RASSLs exhibit activation ratios that are 2-fold greater, preferably 5-fold greater, more preferably 10-fold greater, even more preferably 50- to 100-fold greater than the small molecule ligand:selected natural ligand activation ratio of a native G protein-coupled receptor.

[0111] RASSLs appropriate for use in the method of invention may interact with either the same or different G proteins as the native G protein-coupled receptor from which the RASSLs are derived. Because G protein interaction is facilitated by a receptor domain distinct from that of small molecule binding, G protein-coupled receptors can be modified to alter G protein coupling via alteration of the intracellular domains without affecting the peptide and small molecule binding domains of the receptor. Therefore, a single ligand/RASSL combination can be used to control the G protein-mediated response of the native G protein-coupled receptor from which the RASSL was derived, or the RASSL can be modified for use in activation of any of a wide variety of G protein responses.

[0112] The preferred G protein(s) with which a desired RASSL interacts is determined by a variety of factors, including the cell type to be activated by the RASSL, and the desired type, kind, or extent of the cellular response to be activated by RASSL activation. For example, in cells where G₁ protein family activation facilitates blood-forming cell proliferation, the RASSL used in the method of the invention is derived from a native G protein-coupled receptor that interacts with the G_i protein family, or is modified so that it interacts with the G_i protein family. In contrast, if the RASSL is derived from a G protein-coupled receptor that normally interacts with G_i protein, and this RASSL is to be used to facilitate proliferation of lung or stomach secretory cells, the RASSL is modified so that it interacts with G_q protein, the G protein that facilitates proliferation of lung and stomach secretory cells. Examples of G protein-coupled receptors that are modified with respect to G protein interaction have been described (see, e.g., Liu et al., 1995, Proc. Natl. Acad. Sci. USA 92:11642-11646; Kobilka, 1992, Ann. Rev. Neurosci. 15:87-114).

[0113] RASSLs derived from G protein-coupled receptors that bind nontoxic synthetic small molecule ligands (e.g., which can be safely and effectively administered to a mammalian host by oral and/or intravenous administration) are of particular interest. The kappa opioid receptor (KOR) and serotonin receptors are examples of G protein-coupled receptors that are activated by binding of specific synthetic small molecule drugs suitable for human use.

[0114] Serotonin receptors are known in the art and include 5-hydroxytryptamine (5HT)1A, 5HT1B, 5HT1D, 5HT1E, 5HT₁F, 5HT2A, 5HT2B, 5HT2C, 5HT4, 5HT6, and 5HT7 receptors. See, e.g., Hoyer et al. (2002) Pharmacol. Biochem. Behav. 71:533-554. Nucleotide and amino acid sequences for serotonin receptors from various organisms are known. For example, a human 5HT4 receptor sequence is found under GenBank Accession No. NM00870; rat 5HT4, NM012853; human 5HT1A, NM000524; mouse 5HT1A, NM008308; rat 5HT1A, NM012585; human 5HT1D, AF498979 and NM000864; mouse 5HT1D, NM008309; human 5HT2A, AF498982; mouse 5HT2A, XM139139; human 5HT2C, AF498983; and mouse 5HT2C, NM008312. See also FIGS. 11 (11-1 and 11-2) and FIG. 24.

[0115] The native KOR activates G₁ protein in neural cells, while the serotonin receptors activate the G_q, G₁ and G_S family of G proteins in many tissues. The KOR and serotonin receptors do not facilitate proliferation through activation of their respective G proteins in their native cell types (i.e., KOR activation does not result in cell proliferation in neural cells, and serotonin receptor activation does not result in cell proliferation in liver cells). However, the same signaling pathways activated by KOR and serotonin receptors in their native cell types (G₁ pathway and G_q pathway, respectively) can be proliferative if activated in a different cell type.

[0116] Because G_q, G_S and G₁ trigger proliferation in different tissues, KOR- or serotonin receptor-derived RASSLs are useful in selectively targeting proliferation in tissues in which G₁, G_S or G_q signals facilitate proliferation. For example, KOR-derived RASSLs can be designed to retain activity in G₁ protein activation and facilitate targeted proliferation in blood-forming cells, while serotonin-receptor-derived RASSLs can be designed to retain activity in G_q, G_i or G_S protein activation and facilitate targeted proliferation in hepatic cells and certain secretory cells of the lung, breast and stomach.

[0117] Alternatively, KOR and serotonin receptor RASSLs can be designed so that the RASSL retains the ability to bind the synthetic small molecule bound by native KOR or native serotonin receptors, but altered with respect to the G protein family activated by the RASSL. For example, a KOR-derived RASSL can be designed to interact with G proteins other than those of the G₁ protein family, and thus facilitate proliferation in cell type(s) that proliferate in response to activation of other G proteins. Alteration of G protein interaction can be accomplished by alteration of the native receptor's intracellular domain without substantially affecting the ligand-binding portion of the receptor.

[0118] Carboxyl-Terminal Modifications

[0119] In addition, a RASSL can be further modified to provide for increased duration of signaling; increased resistance to internalization; decreased susceptibility to desensitization with ligand; reduced adenylyl cyclase superactivation; and the like. In general, such features are provided for by modification of amino acid residues in the carboxyl terminal portion of the RASSL.

[0120] In some embodiments, a RASSL with an altered C-terminus is resistant to internalization upon binding to a synthetic ligand, e.g., ligand-induced receptor internalization is reduced at least about 5%, at least about 10%, at least about 15%, at least about 20%, at least about 25%, at least about 30%, at least about 40%, at least about 50%, or more, when compared with the parent RASSL, e.g., the RASSL which does not have a modified C-terminus. Decreased receptor internalization upon binding to a synthetic ligand is also expressed as an increased percentage of RASSLs that remain on the cell surface after ligand binding, e.g., where the percent of RASSLs that remain on the cell surface after ligand binding increases by at least about 5%, at least about 10%, at least about 15%, at least about 20%, at least about 25%, at least about 30%, at least about 40%, at least about 50%, or more, when compared with the parent RASSL.

[0121] Whether a given C-terminal modified RASSL is resistant to ligand-induced internalization can be determined using any known method, e.g., immunological methods using labeled antibody specific for the RASSL; fluorescent methods, e.g., using a fusion protein comprising a RASSL fused in-frame to a fluorescent protein; or other detection methods in which a RASSL is fused in-frame to a fusion partner protein that provides for a detectable signal, where a detection method is selected based on the signal provided by the fusion partner.

[0122] In some embodiments, a C-terminal modified RASSL is less susceptible to ligand desensitization. Pre-treatment with the synthetic ligand typically reduces subsequent activation by the same ligand. However, in these embodiments, C-terminal modified RASSLs exhibit at least about 5%, at least about 10%, at least about 15%, at least about 20%, at least about 25%, at least about 30%, at least about 40%, at least about 50%, or more, less ligand-induced desensitization when compared with the parent RASSL without the C-terminal modification(s).

[0123] Whether a C-terminal modified RASSL has reduced susceptibility to ligand desensitization is readily determined using established techniques. For example, a cAMP accumulation assay, as described in Example 15, is suitable for determining susceptibility to ligand-induced desensitization.

[0124] In some embodiments, the duration of signal generated by binding of synthetic ligand to the C-terminal modified RASSL is increased when compared with the parent RASSL without the C-terminal modification(s). For example, the duration of signal is increased by at least about 5%, at least about 10%, at least about 15%, at least about 20%, at least about 25%, at least about 30%, at least about 40%, at least about 50%, or more, when compared with the parent RASSL without the C-terminal modification(s).

[0125] In some embodiments, the carboxyl terminal portion of a RASSL is modified such that phosphorylation sites are eliminated, e.g., by deletion or substitution with an amino acid that is not phosphorylated. For example, one or more serine and/or threonine residues are deleted, or are substituted with an amino acid that is not phosphorylated, such as alanine, leucine, isoleucine, glycine, and the like.

[0126] As one non-limiting example, one or more of the following residues of KOR₁ or KOR₂ are modified: serine at position 356 (S356), threonine at position 357 (T357), serine at position 358 (S358), and threonine at position 363 (T363). For example, one or more of these positions are deleted, e.g., a contiguous stretch of amino acids that includes one or more of S356, T357, S358, and T363 is deleted, e.g., a stretch of contiguous amino acids from 356 to 363, from 355 to 357, from 355 to 359, from 356 to 359, or from 355 to 365 is deleted. Alternatively, one or more of S356, T357, S358, and T363 is substituted with an amino acid that is not phosphorylated, e.g., one or more of S356, T357, S358, and T363 is substituted with alanine, leucine, isoleucine, glycine, and the like.

[0127] C-terminal modifications are made using established techniques, as described above.

[0128] Kappa Opioid Receptor-Derived RASSLs

[0129] The KOR is a member of the opioid receptor family, which also includes the mu and delta opioid receptors. The KOR is implicated in neurotransmission in its native cell type. In contrast to KOR, mu receptor activation is associated respiratory depression, and addictive effects of opiates. KOR has distinct binding domains for its synthetic small molecule ligands, and for its natural ligand dynorphin, which is found endogenously in the nervous system, including the peripheral nerves that penetrate most organs (Xue et al., 1994, J. Biol. Chem. 269:30195-30199). Thus, KOR-derived RASSLs retain small molecule binding activity and are preferably substantially unresponsive to dynorphin binding or substantially reduced in dynorphin binding activity.

[0130] Several KOR-specific synthetic small molecule agonists have been developed including, pentazocine (TALWN), Spiradoline, U-62066E, U-62066, and U50488H (Pharmacia-Upjohn), ICI-204,488 (Zeneca), DUP 747 (Dupont-Merck), and CI-977 (Parke-Davis). The structural formulas for pentazocine, DUP 747, CI-977, and Spiradoline are shown in FIG. 3. Several of these compounds (e.g., pentazocine and Spiradoline) are orally available, and are safe (in humans and primates) and nonaddictive (in rats and primates). Any of these synthetic compounds are suitable for activation of KOR-derived RASSLs in the targeted activation method of the invention, and are likely to have relatively few side effects. For example, pentazocine, a mixed KOR agonist/mu receptor antagonist, is an orally available, weakly analgesic drug with low addictive potential, that has been used safely by thousands of people for over 15 years.

[0131] In addition, two opioid antagonists, Naloxone (available generically as Naracn) and Naltrexone (available generically as Trexan, DuPont-Merck) have been used clinically for many years and are thus of particular interest for use in the present invention. Both of these compounds are safe and nonaddictive when given to people. Both of these drugs can be used as agonists of the opioid receptors (including the KOR) if those receptors have been mutated at a conserved position in each of these receptors (Claude et al., 1996, PNAS, 93:5715-5719).

[0132] Serotonin Receptor-Derived RASSLs

[0133] Serotonin receptor-derived RASSLs are also of particular interest for use in the targeted cellular response method of the invention. The biogenic amine serotonin, a neurotransmitter and hormone widely utilized in the body, is the natural ligand for serotonin receptors. Synthetic small molecule ligands which bind and activate serotonin receptors are widely used to treat clinical depression (e.g., Buspirone, known as Buspar, Mead Johnson). In addition, orally available serotonin receptor antagonists are used clinically to treat headaches (e.g., methysergide, known commercially as Sansert, Novartis). The structural formulas for Buspar and Sansert are shown on FIG. 4. By analogy to opioid receptor mutants, proper mutation of serotonin receptors may allow use of these serotonin antagonists as agonists. In addition, Cisapride (Prepulsid, Janssen) has been used as a 5HT4 agonist.

[0134] Synthetic Small Molecules

[0135] Any small molecule, preferably a synthetic small molecule, that can bind within the transmembrane domains of a RASSL and facilitate RASSL-mediated activation of a desired family of G proteins is suitable for use in the method of targeted activation method of the invention. In contrast to the natural peptide ligands of G protein-coupled receptors which typically have molecular weights of 2000-6000 Da, synthetic small molecule ligands of G protein-coupled receptors have molecular weights of 100-1000 Da. In general, any synthetic drug is suitable for use in the invention if it binds the RASSL in a unique manner distinguishable from natural ligand binding.

[0136] Synthetic small molecules useful in the present invention include synthetic small molecules generated by either a natural (e.g., isolated from a recombinant cell line) or chemical means (e.g., using organic or inorganic chemical processes).

[0137] Where the RASSL-expressing target cells are present in the brain of the subject, the synthetic small molecule is preferably of a molecular weight and net ionic charge that permits it to cross the blood-brain barrier following oral or parenteral (e.g., intravenous) administration. Typically molecules that cross the blood brain barrier are less charged than peptide molecules. Synthetic drugs can be made that do, or do not cross the blood-brain barrier depending on the number of charged groups on the molecule (see, e.g., Freidinger, 1993, Prog. Drug Res. 40:33-98). Smaller molecules, e.g., less than 4000 Da, are also more likely to cross the blood-brain barrier.

[0138] Several synthetic small molecules that bind and activate native G protein-coupled receptors are known in the art and are useful in the present invention. Additional synthetic small molecules suitable for use in the present invention can be identified by screening candidate compounds for binding to native G protein-coupled receptors or to RASSLs. For example, a cell line expressing a RASSL of interest is exposed to varying concentrations of a compound to be tested for RASSL binding. RASSL binding is detected by induction of a cellular response (e.g., cellular growth, secretion, etc.) upon exposure to the test compound, but not in the presence of a control compound that does not bind the RASSL and/or does not induce cellular activation. The cellular response can be detected by a variety of methods known in the art, and will vary according to the type of cellular response elicited by synthetic small molecule binding. For example, the cellular response of proliferation can be detected by cellular incorporation of a radioactive DNA component (e.g., ³H-thymidine). Where the cellular response is secretion, detection of the amount of a secreted protein in the cell supernatant is indicative of activation of the receptor. RASSL activation can also be detected by measuring intracellular cAMP or calcium levels (calcium flux).

[0139] Synthetic small molecule drugs have several advantages over peptide drugs. Small molecule drugs can be more economically produced, administered orally without substantial activity loss, and have long biological half-lives relative to peptide drugs. Numerous synthetic small molecule drugs have been developed, or are currently under development, for specific G protein-coupled receptors, including the receptors for angiotensin II, neurokinins, oxytocin, vasopressin, somatostatin, bombesin, thyrotropin-releasing hormone, endothelin, C5a, luteinizing hormone-releasing hormone, motilin and neuropeptide Y (see Table 1, and recent review by Freidinger, 1993, Prog. Drug Res. 40:33-98). RASSLs that do not bind a selected natural ligand, but bind and respond to synthetic drugs, can be used in the method of the invention to control a variety of physiological processes such as proliferation, secretion, cell migration, cell contraction, pigment production and apoptosis, depending on the cell type expressing the RASSL.

[0140] Screening for RASSLs and RASSL Activation

[0141] RASSLs of the invention can be identified by screening compounds for their ability to bind and activate the RASSL-expressing cells. Methods for detection of receptor binding and detection of G protein-mediated responses are well known in the art. Binding studies can be accomplished using competitive binding assays using radioactively labeled ligand. The binding properties of the selected natural ligand, and the small molecule ligand are compared to the binding properties of the wild-type receptor.

[0142] RASSLs can also be screened on the basis of the receptor's activation ratio (level of activation in response to small molecule binding relative to level of activation in response to exposure to a selected natural ligand). Activation levels can be determined using standard biochemical measurements associated with G protein-coupled receptor activation. For example, RASSL activation can be determined by measuring the changes in intracellular cAMP levels, inhibition of cAMP accumulation (e.g., as in RASSLs derived from the KOR), accumulation of inositol phosphates (e.g., as in RASSLs derived from serotonin receptors), or calcium mobilization (e.g., as in RASSLs derived from the serotonin receptor). Methods for detecting these and other G protein-mediated activation signals are well known in the art (see, e.g., Conklin et al., 1993, Nature 363:274-276; Wong et al., 1992, Science 255;339-342; Federman et al., 1992, Nature 354:159-161.)

[0143] RASSLs can also be further screened for their ability to mediate the desired physiological or biochemical response in the target cell. For example, the ability of a RASSL to mediate cell proliferation can be detected by incorporation of the ³H-thymidine in the target cells (see, e.g., Conklin et al., 1988, Proc. Natl. Acad. Sci. USA 85:8698-8702).

[0144] Alternatively, RASSL-mediated cell proliferation can be detected by co-culturing in vitro RASSL-expressing cells with cells that do not express the RASSL. Preferably, the RASSL-expressing target cells compose only about 1% or less of the total cell population. The cells are then cultured in the presence of the RASSL small molecule agonist for several hours, days, or weeks. The number of RASSL positive cells in the cell population is then detected by fluorescence-activated cell sorting (FACS). RASSLs that promote a significant increase in the number of target cells over time are suitable for use in the invention.

[0145] RASSL Selection

[0146] The RASSL selected for expression in a mammalian host will vary with a variety of factors including, but not limited to, the physiological response to be activated (e.g., the G protein subfamily that must be activated in the selected target cell type in order to elicit the desired physiological response), and the cell type(s) or tissues in which targeted cellular activation is desired. Where the selected natural ligand is endogenous to the tissue in which the target cells are present, the RASSL preferably has a higher binding affinity for synthetic small molecule ligand in the presence of the selected natural ligand. Preferably, the RASSL expressed by the target cells binds and is activated by a synthetic small molecule that is not bound by endogenous G protein-coupled receptors expressed by the RASSL-expressing target cell itself and/or by nontarget cells present in the cell population (e.g., surrounding cells or tissue) surrounding the RASSL-expressing target cell.

[0147] Vectors and Constructs

[0148] Any of a variety of vectors (e.g., viral vectors, bacterial vectors, or vectors capable of replication in eukaryotic and prokaryotic hosts) can be used in accordance with the invention to produce RASSL-expressing cells. Preferably, the vector is capable of replication in both eukaryotic and prokaryotic hosts in order to facilitate efficient production of RASSL-encoding DNA for use in the method of the invention. Numerous vectors that can replicate in eukaryotic and prokaryotic hosts are known in the art and are commercially available.

[0149] The vector selected will vary according to whether the target cells to be activated are present in an in vitro culture, or are transformed for use in ex vivo or in vivo gene therapy protocols. For example, where the target cells are to be transformed in vivo, the preferred vector may be a stable integrating vector or a stable nonintegrating vector.

[0150] Examples of such vectors include viral vectors and artificial chromosomes (e.g., human artificial chromosomes).

[0151] Preferably, the DNA construct contains a promoter to facilitate expression of the RASSL-encoding DNA within the target cell. Preferably the promoter is a strong, eukaryotic promoter. Exemplary eukaryotic promoters include promoters from cytomegalovirus (CMV), mouse mammary tumor virus (MMTV), Rous sarcoma virus (RSV), and adenovirus. More specifically, exemplary promoters include the promoter from the immediate early gene of human CMV (Boshart et al., 1985, Cell 41:521-530) and the promoter from the long terminal repeat (LTR) of RSV (Gorman et al., 1982, Proc. Natl. Acad. Sci. USA 79:6777-6781). Of these two promoters, the CMV promoter is preferred as it provides for higher levels of expression than the RSV promoter.

[0152] Other components of vectors suitable for use include a marker(s) (e.g., an antibiotic resistance gene (such as an ampicillin resistance gene) or -galactosidase) to aid in selection of cells containing the construct, an origin of replication for stable replication of the construct in a bacterial cell (preferably, a high copy number origin of replication), a nuclear localization signal, or other elements which facilitate production of the DNA construct, the protein encoded thereby, or both. Such selective markers can be used instead of or in addition to selection of the RASSL-expressing target cells by detection or selection of a RASSL-stimulated cellular response (e.g., for selection of cells having both RASSL-encoding DNA and DNA encoding a therapeutic cellular product of interest).

[0153] For eukaryotic expression, the construct should contain at a minimum a eukaryotic promoter operably linked to RASSL-encoding DNA, which is in turn operably linked to a polyadenylation sequence. The polyadenylation signal sequence may be selected from any of a variety of polyadenylation signal sequences known in the art. Preferably, the polyadenylation signal sequence is the SV40 early polyadenylation signal sequence. The construct may also include one or more introns, which can increase levels of expression of the RASSL-encoding DNA. Any of a variety of introns known in the art may be used. Preferably, the intron is the human -globin intron and inserted in the construct at a position 5 to the RASSL-encoding DNA.

[0154] Transformation

[0155] Any method of transformation can be used to obtain transformed cells containing RASSL-encoding nucleic acid and/or expressing a RASSL of the invention. Thus, the RASSLs of the invention and methods of the invention using RASSLs can be used in combination with any number of vectors or gene delivery systems. In general, methods for transforming and obtaining expression of exogenous DNA or RNA sequences in a mammalian cell are well known in the art (see, for example, Kormal et al., 1987, Proc. Natl. Acad. Sci. USA, 84:2150-2154; Sambrook et al. Molecular Cloning: A Laboratory Manual, 2nd Ed., 1989, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y.; each of which are hereby incorporated by reference with respect to methods and compositions for eukaryotic expression of an exogenous nucleic acid sequence). The method selected will generally vary with whether target cell transformation is accomplished in an in vitro cell culture, by administering the RASSL-encoding DNA directly to the mammalian host (in vivo gene therapy), or by introduction of nucleic acid into to target cells in vitro and subsequent transplantation of the transformed cells into a mammalian host (ex vivo gene therapy). The following illustrations are examples of various formulations that can be used to obtain transformed target cells expressing a RASSL of the invention and are not meant to be limiting.

[0156] RASSL-encoding DNA can be delivered to the target cells in a variety of different formulations. For example, purified DNA, in a viral vector (e.g., adenovirus, adeno associated virus (AAV), lentivirus, retrovirus), a DNA- or RNA-liposome complex, or by utilizing cell-mediated gene transfer. DNA- or RNA-liposome complex formulations suitable for transformation of the target cells comprise a mixture of lipids that bind to genetic material (DNA or RNA), providing a hydrophobic coat which allows the genetic material to be delivered into cells (see, e.g., Felgner et al., 1987, Proc. Natl. Acad. Sci. U.S.A. 84:7413-7417). Exemplary liposome formulations suitable for use in target cell transformation include DOPE (dioleyl phosphatidyl ethanol amine), CUDMEDA (N-(5-cholestrum-3--ol 3-urethanyl)-N,N-dimethylethylene diamine).

[0157] Alternatively, the RASSL-encoding nucleic acid can be administered as a DNA or RNA sequence-containing chemical formulation coupled to a carrier molecule which facilitates delivery to the host cell. Such carrier molecules can include, for example, an antibody specific to an antigen expressed on the surface of the target cells, or some other molecule capable of interaction with a receptor associated with the target cells. Such formulations that target a specific cell type for delivery of RASSL-encoding nucleic acid are particularly useful in in vivo transformation.

[0158] The form of the DNA preparation for transformation of the target cells will depend upon several factors such as the cell type targeted for gene transfer, the route of administration (e.g., in gene therapy), and whether a biological or nonbiological transforming formulation of nucleic acid (e.g., viral or chemical) is used.

[0159] In vivo and ex vivo Gene Therapy

[0160] The RASSLs, RASSL-expressing cells, and methods of the invention may be used in combination with any gene therapy method or gene delivery protocol. However, the use of RASSLs is largely independent of the method of gene transfer into the cell.

[0161] Because RASSLs can be designed to selectively induce proliferation in RASSL-expressing target cells, transformation of one cell with RASSL-encoding DNA, preferably at least 5 cells to 10 cells, more preferably 100 to 1,000 cells, would be sufficient since the proliferative effect could be exploited to expand the cells in vivo to achieve a desired therapeutic effect. Current gene delivery (gene insertion) techniques can stably introduce a gene into cells at an efficiency rate of approximately 1 in 100,000 to 1 in 100 cells. The RASSLs of the invention can be used alone or in combination (e.g., a proliferation-inducing RASSL coexpressed with a RASSL that induces some other desired effect). RASSL-encoding DNA can also be delivered in conjunction with DNA encoding a therapeutic polypeptide of interest to provide for selective expansion of such therapeutic protein-expressing cells (e.g., in vitro for delivery via ex vivo gene therapy or via in vivo gene therapy). The following illustrations are examples of various methods to achieve in vivo gene therapy using RASSL-encoding nucleic acid and the methods of RASSL-mediated, targeted cellular activation of the invention, and are not meant to be limiting.

[0162] In vivo Gene Therapy

[0163] In general, in vivo transformation of target cells is accomplished by either mechanical means (e.g., direct injection of a RASSL-encoding nucleic acid formulation near or into the tissue containing the target cells (Fynan et al., 1993, Proc. Natl. Acad. Sci. USA 90:11478-11482; Yang et al., 1990, Proc. Natl. Acad. Sci. USA. 87:1568-9572; Wolff, et al., 1990, Science 247:1465-1468)), or by biological means (e.g., infection of tissue containing the target cells with a nonpathogenic virus, preferably a nonreplicative virus, containing RASSL-encoding DNA). A review of the various techniques available for human gene therapy has been recently reviewed (see Morsy, et al., 1993, JAMA 270(19):2338-2345; Orkin, et al., 1995, Report and Recommendations of the Panel to Assess the NIH Investment in Research on Gene Therapy, http://www.nih.gov/news/panelrep.htm1/).

[0164] In vivo gene transfer can be accomplished by administering a RASSL-encoding DNA formulation to the mammalian host either orally or by injection depending upon the site of the cells to be targeted for gene transfer and the formulation of the RASSL-encoding nucleic acid. Where RASSL-encoding nucleic acid is administered by injection, the nucleic acid may be administered to the mammalian host either locally or systemically. Systemic administration can be carried out by, for example, intramuscular injection of a viral vector containing the RASSL-encoding DNA. The amount of DNA and/or the number of infectious viral particles effective to infect and transform a sufficient number of target cells (e.g., a number of cells sufficient to allow for subsequent targeted activation) can be readily determined based upon such factors as the efficiency of in vitro transformation, and the susceptibility of the targeted cells to transformation.

[0165] The amount of DNA administered will vary greatly according to a number of factors including the susceptibility of the target cells to transformation, and the size and weight of the host. Generally, the amounts of DNA for human gene therapy can be extrapolated from the amounts of DNA effective for gene therapy in an animal model. For example, the amount of DNA for gene therapy in a human is roughly 100 times the amount of DNA effective in gene therapy in a rat. The amount of DNA necessary to accomplish target cell transformation will decrease with an increase in the efficiency of the transformation method used.

[0166] Ex vivo Gene Therapy

[0167] In general, ex vivo gene therapy is accomplished by obtaining target cells from the mammalian host to be treated (e.g., biopsy of tissue containing the target cells, isolation of a desired target cell from circulating lymphocytes and other immune cells) and establishing a primary culture of these target cells according to methods well known in the art. Alternatively, a cell line heterologous to the tissue in which the target cells are to be implanted can be cultured and transformed with RASSL-encoding DNA in vitro. For example, microglia can be cultured in vitro, transformed with RASSL-encoding DNA (where the RASSL provides for enhanced secretion) and DNA encoding a therapeutic protein of interest, and implanted in the mammalian host. Upon activation with RASSL-binding small molecule ligand, the transformed, RASSL-expressing microglia secretes the therapeutic protein from secretory granules. Use of microglia for expression of exogenous nucleic acid, and transplantation of these cells, is well known in the art (Cunningham et al., 1994, Brain Res. 658:219-231).

[0168] In a preferred embodiment, transformed target cells expressing the RASSL-encoding DNA are selected using the targeted activation method of the invention. Briefly, the transformed cells are exposed to an appropriate synthetic small molecule that binds the RASSL. Upon RASSL binding of the synthetic small molecule, transformed cells expressing the RASSL are activated, and activated cells are selected from the non-RASSL-expressing, nonactivated cells in the population. For example, where the cellular response associated with binding of the synthetic small molecule to the RASSL is cellular proliferation, cells expressing the RASSL are selected by virtue of their selective growth relative to those cells that do not express or properly express the RASSL. Continued culture and exposure of the transformed cells to the synthetic small molecule will result in the expansion of the RASSL-expressing cells to provide a cell population primarily composed of RASSL-expressing target cells. This method of selection of RASSL-expressing target cells avoids exposure of the target cells to any substances that may be toxic or otherwise undesirably present in a cell population that is to be used in therapy in a mammalian host, e.g., implanted in a mammalian host to provide expression of a desired cellular product.

[0169] Transgenic Animals Expressing a RASSL of the Invention: Reversible Disease Models and Means of Improved Productivity

[0170] The methods and compositions of the invention are also useful in generating transgenic reversible disease models. In one embodiment, the transgenic, nonhuman animals express a selected RASSL that, upon binding of a synthetic small molecule ligand, elicits a cellular, G protein-coupled physiological response associated with a disease or condition. However, in the absence of the synthetic small molecule ligand, the RASSL-expressing transgenic animal preferably appears substantially normal, and does not exhibit symptoms of the RASSL-induced condition to a significant extent. Transgenic animals expressing RASSLs are advantageous in that the health and reproductive capacity of the animals are not significantly affected, since the animals exhibit pathology only when the RASSL is stimulated.

[0171] For example, RASSL-expressing transgenic animals are useful as models of cardiac arrhythmia. Electrical conduction by heart pace maker cells is mediated by G₁ protein signaling. A RASSL is expressed in all or a portion of the transgenic animal's heart pacemaker cells so that, upon exposure to the appropriate synthetic small molecule ligand, G₁ signaling is altered. RASSL activation thus provides a means to cause asynchronous conduction in the transgenic animal's heart (e.g., the conduction signal in the RASSL-expressing cells is continuously or erratically activated relative to the conduction of the signal in normal, non-RASSL-expressing cells). The asynchronous conduction by the nontarget and target (RASSL-expressing) cells results in cardiac arrhythmia in the transgenic animal. The efficacy of candidate drugs for treatment of cardiac arrhythmia can be tested in the transgenic cardiac arrhythmia model. Because KOR interacts with the G₁ protein family, RASSLs derived from KOR are suitable for use in this reversible cardiac arrhythmia disease model. RASSLs derived from other receptors that interact with other G proteins may also have pathological effects on the heart.

[0172] Transgenic, RASSL-expressing animals are also useful in generating models of brain seizures. Electrical short circuits in brain cells (neurotransmission) are associated with G protein activity. Thus, similar to the cardiac arrhythmia model discussed above, brain cells expressing RASSLs are induced by small molecule binding to conduct an asynchronous electrical signal, resulting in a brain seizure in the transgenic animal. The efficacy of drug candidates for the prevention or control of seizures could then be tested in this animal model.

[0173] RASSL-expressing transgenic animals are also useful models of dementia or neurodegenerative diseases. Abnormal G protein signaling has been correlated with various dementias including Alzheimer's disease. Therefore, RASSL expression and activation could be used to mimic some aspects of dementia, which in turn could be used as an animal model for testing compounds for treatment of dementia and symptoms associated therewith.

[0174] RASSL-expressing transgenic animals are also useful as models of osteoporosis. G protein signaling is involved in regulating the activity of osteoclasts and osteoblasts; PTH and calcitonin (the two major calcium regulating hormones) are both G protein-coupled receptors. Transgenic models of osteoporosis are generated by, for example, expressing a RASSL in the animal's osteoclasts and/or osteoblasts. Preferably, the RASSL does not bind the natural ligand of the native PTH or calcitonin receptor, but is activated by exposure to a synthetic small molecule. The activation of the RASSL would override the native PTH, or calcitonin, causing abnormal bone metabolism.

[0175] RASSL-expressing transgenic animals are also useful in the study of vasospasm, the phenomenon associated with stroke, migraine, cardiac ischemia, dementia and peripheral vessel vasospasm. Smooth muscle contraction in vessels, which is associated with these various types of vasospasm-induced phenomena, is activated by G_q protein activation. RASSL-mediated stimulation of contraction in, for example, vessels of the brain is thus useful in mimicking stroke or migraines in the transgenic animal model. Further, RASSL-mediated stimulation of contraction of muscle cells in peripheral limbs is useful in mimicking Raynaud Syndrome. RASSL-mediated smooth muscle contraction is also useful in mimicking cardiac vasospasm by induction of contraction of smooth muscle of the heart by RASSL binding of synthetic small molecule. Because serotonin receptors are G_q-coupled, RASSLs derived from serotonin receptors are preferred RASSLs for use in transgenic animal models of disease or conditions associated with vasospasm.

[0176] The compositions and method of the invention are also useful in generating transgenic animal models having a reversible immune dysfunction. G protein signaling is involved in homing, chemotaxis, and/or activation of leukocytes and is thus implicated in the development of a variety of immune disorders. For example, a transgenic mouse having an inactivated G₁ protein develops a Crohn's disease-like syndrome. RASSL expression in leukocytes is thus useful in mimicking hyper- or hypo-immune functions associated with arthritis, Lupus, Crohn's disease, and other autoimmune disorders. Such animal models are useful in, for example, screening candidate drugs for the alleviation or amelioration of symptoms associated with these diseases.

[0177] RASSL-expressing transgenic animals can also be used to increase their usefulness or productivity in food production. Calcium metabolism is regulated by G protein-coupled receptors, and thus can be regulated by RASSLs. Calcium content of food is important for the nutritional value as well as the preservation of food. For example, the egg shell is primarily made of calcium. RASSL stimulation could be used to increase the calcium in an egg shell by mobilizing calcium stores, thereby increasing the strength of the eggs, and decreasing the loss of eggs due to handling. In another example, G protein-coupled receptors also control the fat and muscle content of meat. RASSLs could be used to adjust the fat and muscle content of animals, which could enhance the nutritional value of the animals.

[0178] RASSLs may also be used to regulate many aspects of reproduction that are naturally regulated by G protein coupled receptors. For instance, ovulation and testicular maturity are controlled by the luteinizing hormone receptor (LH-R; a G_S-coupled receptor) while uterine contractions and the onset of labor are controlled by the oxytocin receptor (a G_q-coupled receptor). RASSLs of the invention may be designed from these receptors to replace or enhance these receptors' functions. Regulation of reproduction is vital to many animal breeding programs (e.g., breeding for food production, transgenic animals, or other scientifically or commercially valuable animals).

[0179] For instance, the oxytocin receptor may be replaced with a G_q-coupled RASSL to regulate the timing of labor, thereby ensuring that labor occurred at a time that was best for the pregnant animal. For example, if the pregnant animal were ill, the administration of the RASSL agonist would be delayed until the pregnant animal had recovered from the illness. Labor and delivery could also be regulated to ensure that the animal care staff was best prepared to attend to the animal; this is particularly important when dealing with large animals such as sheep, cows, or horses which require a considerable amount of care. Regulation of labor could increase efficiency and lower costs of an animal breeding program.

[0180] RASSL-mediated regulation of reproduction may also be used to limit or control distribution of transgenic animals. For instance, a RASSL replacing the LH-R or the oxytocin receptor would allow regulation of reproductive processes such that the animal would only be fertile (LH-R replacement), or go into labor (oxytocin receptor replacement) upon administration of the appropriate RASSL agonist. If the RASSL agonist were a proprietary compound (with limited availability), only companies or individuals with access to the RASSL agonist would be able to breed the animals, thereby allowing a company to sell its transgenic animals without heightened concerns about unauthorized breeding.

[0181] RASSL-dependent fertility may also address problems associated with breeding of transgenic animals in the wild, a considerable environmental concern. For instance, hatchery bred salmon released into the wild can breed with wild salmon, thereby diluting the genetic pool of wild salmon. Replacement of the hatchery salmon's G protein coupled receptor(s) with RASSLs of the invention may aid in fish breeding and decrease the possibility of the hatchery salmon breeding with wild salmon, since the hatchery salmon would be fertile (e.g., due to ovulation or sperm production) when the appropriate RASSL is stimulated. In addition, the timing of fertility and egg laying of the RASSL-regulated fish may increase the productivity of the fish hatchery by, for example, inducing egg laying only under ideal conditions. Similar scenarios may be applied to any animals that are used commercially or scientifically and could mix with the wild populations of animals (e.g., chickens, sheep, cows, horses, rabbits, rodents, shrimp, bees, and the like).

[0182] In other embodiments, the transgenic non-human animal expresses a selected RASSL that, upon binding of a synthetic ligand, suppresses a cellular, G protein-coupled physiological response, e.g., a response associated with a disease or condition. For example, a RASSL transgenic non-human animal has reduced response to a given stimulus after exposure to a synthetic ligand. Example 13 provides a description of a non-human animal transgenic for a KOR RASSL. In this animal, binding of spiradoline to the KOR RASSL results in reduced response to olfactory stimuli. As another non-limiting example, a RASSL transgenic non-human animal has reduced response to pain.

[0183] The invention also provides animal models of pain. In some embodiments, a subject pain model is a non-human transgenic animal pain model. In these embodiments, a non-human animal is created that is transgenic for a G_i-coupled RASSL, e.g., an opioid RASSL (including kappa, mu, and delta opioid RASSLs), a 5HT-1 RASSL (including any 5HT-1 subtype), a motilin RASSL, and the like. A synthetic ligand for the RASSL is administered after a pain stimulus is administered. In the absence of the synthetic ligand, and/or in the presence of the endogenous ligand for the GPCR, pain is not diminished. However, when the synthetic ligand is administered, the pain response is reduced by at least about 10%, at least about 20%, at least about 30%, at least about 40%, at least about 50%, at least about 60%, at least about 70%, at least about 80%, at least about 90%, or more, compared to the pain response in the absence of the synthetic ligand.

[0184] The pain response is measured using any of a wide variety of pain models known in the art. Known pain models include, but are not limited to, the tail-flick model, the tail immersion model, the carrageenan-induced paw hyperalgesia model, the formalin behavioral response model, the von Frey filament test, the chronic constriction injury model, and the like.

[0185] In other embodiments, chimeric animal is used, e.g., the animal is genetically modified, but the transgene is not part of the germline. A RASSL is delivered to the animal using a viral (e.g., AAV, Adenovirus, a retroviral vector, and the like), or non-viral vector. The animal is subjected to a pain stimulus, using any one of the above-mentioned pain models. As a result of administration of a synthetic ligand for the RASSL, the pain response is reduced by at least about 10%, at least about 20%, at least about 30%, at least about 40%, at least about 50%, at least about 60%, at least about 70%, at least about 80%, at least about 90%, or more, compared to the pain response in the absence of the synthetic ligand.

[0186] In some embodiments, the RASSL is expressed systemically, e.g., in many different tissues. In these embodiments, the RASSL-encoding construct is administered such that the RASSL is expressed systemically, e.g., via intravenous, intragastric, or subcutaneous administration, or any other route of administration that results in systemic delivery of the RASSL-encoding construct.

[0187] In other embodiments, the RASSL is expressed in a local manner, e.g., at or near the site of administration of the RASSL. In these embodiments, the pain stimulus is administered at or near the site of RASSL expression. In some of these embodiments, the RASSL is expressed in a tissue-specific manner, such that expressed is localized. In other embodiments, the RASSL-encoding construct is delivered in such a manner that RASSL expression remains localized.

[0188] Transgenic animals can be generated according to a variety of methods well known in the art. In general, transgenic animals are generated by introducing the desired nucleic acid (e.g., RASSL-encoding nucleic acid and/or nucleic acid encoding a therapeutic cellular product) into the germline DNA of a host animal. Introduction of the desired nucleic acid into the germline DNA of the host animal can be accomplished by a variety of methods. For example, introduction of the desired nucleic acid into the animal's germline DNA can be accomplished by microinjection into a fertilized egg of the host animal, followed by transfer of viable eggs into the oviducts of the pseudopregnant mice (see, e.g., Gordon et al., 1980, Proc. Natl. Acad. Sci. 77:7380-7384, herein specifically incorporated by reference for methods of genetically transforming embryos by microinjection of DNA; and Hogan et al., 1986, Manipulation of the Mouse Embryo: A Laboratory Manual, Cold Spring Harbor Laboratory, NY, herein specifically incorporated by reference for methods of generating transgenic animals). Alternatively, the embryonic transduction can be accomplished with a retroviral vector containing the desired transgene.

[0189] Any method of making transgenic animals can be used as described, for example, in Transgenic Animal Generation and Use L. M. Houdebine, Harwood Academic Press, 1997; Transgenesis Techniques: Principles and Protocols D. Murphy and D. A. Carter, ed. (June 1993) Humana Press; Transgenic Animal Technology: A Laboratory Handbook C. A. Pinkert, ed. (January 1994) Academic Press; Transgenic Animals F. Grosveld and G Kollias, eds. (July 1992) Academic Press; and Embryonal Stem Cells: Introducing Planned Changes into the Animal Germline M. L. Hooper (January 1993) Gordon & Breach Science Pub; U.S. Pat. No. 6,344,596; U.S. Pat. No. 6,271,436; U.S. Pat. No. 6,218,596; and U.S. Pat. No. 6,204,431; Maga and Murray (1995) Bio/Technol. 13:1452-1457; Ebert et al. (1991) Bio/Technol. 9:835-838; Velander et al. (1992) Proc. Natl. Acad. Sci. USA 89:12003-12007; Wright et al. (1991) Bio/Technol. 9:830-834.

[0190] Transgenic animals also can be generated using methods of nuclear transfer or cloning using embryonic or adult cell lines as described for example in Campbell et al. (1996) Nature 380: 64-66; and Wilmut et al. (1997) Nature 385: 810-813. Cytoplasmic injection of DNA can be used, as described in U.S. Pat. No. 5,523,222. Subject transgenic animals can be obtained by introducing a construct comprising a RASSL-encoding polynucleotide.

[0191] Alternatively, the constructs containing the desired transgene can be introduced into embryonic stem (ES) cells obtained from preimplantation embryos cultured in vitro (U.S. Pat. No. 5,464,764; Evans et al., 1981, Nature 292:154-156; Bradley et al., 1984, Nature 309:255-258; Gossler et al., 1986, Proc. Natl. Acad. Sci. 83:9065-9069; Robertson et al., 1986, Nature 322:445-448). An appropriate vector containing the desired construct can be introduced into the ES cells by methods known in the art, including electroporation or microinjection. Transformed ES cells are combined with blastocysts from a nonhuman animal. The transformed ES cells thereafter colonize the embryo, contributing to the germ line of the resulting transgenic animal. For a review, see, Jaenisch, 1988, Science 240:1468-1474, herein specifically incorporated by reference.

[0192] In many embodiments, a polynucleotide is injected into an embryo, e.g., at the single-cell stage, forming a genetically modified embryo, and the genetically modified embryo is allowed to develop into a mature transgenic animal.

[0193] In some embodiments, the transgene is introduced into a somatic cell, where the transgene is integrated into the genome, forming a genetically modified somatic cell, and the nucleus of the genetically modified somatic cell is transferred into a single-cell embryo, forming a genetically modified embryo. The genetically modified single-cell embryo is then transferred into an oviduct of a recipient female, and the embryo allowed to develop into a mature transgenic animal.

[0194] Transgenic animals also include somatic transgenic animals, e.g., transgenic animals that include a transgene in somatic cells (and not in germ line cells). Methods of somatic cell transformation are described in the art. See, e.g., Furth et al. (1995) Mol. Biotechnol. 4:121-127.

[0195] In some embodiments, the nucleotide sequence encoding a RASSL is operably linked to a promoter and, optionally, additional control elements, that provide for tissue-specific expression of the transgene in the animal. In other embodiments, the nucleotide sequence encoding the RASSL is not operably linked to any control elements. Instead, the RASSL transgene includes, on the 5 and 3 ends of the coding region, sequences that provide for homologous recombination with an endogenous gene.

[0196] In some embodiments, a RASSL transgene is not operably linked to a control element. Instead, the transgene includes sequences that provide for homologous recombination with an endogenous gene, such that the RASSL coding sequence replaces all or part of endogenous coding sequence, and the integrated RASSL coding region is under transcriptional control of endogenous control element(s). For example, a RASSL transgene includes 5 and 3 flanking sequences that are homologous to sequences in the 5 and 3 regions of an endogenous GPCR gene, such that the transgene integrates into the genome of a cell by homologous recombination, whereby the RASSL coding sequences of the transgene replace the endogenous GPCR gene, and the RASSL coding sequence integrates into the genome and is under the transcriptional control of the endogenous GPCR control elements. Methods for carrying out homologous recombination are well known in the art.

[0197] Administration of Small Molecule Ligands to Induce in vivo RASSL-mediated Targeted Cellular Activation

[0198] Once the RASSL-expressing target cell(s) are present within the mammalian host, activation of these target cells is induced by administration of a synthetic small molecule(s) (also referred to herein as drug(s)) that can bind the RASSL and facilitate G protein activation and activation of the desired cellular response (e.g., proliferation and/or secretion). The route of administration and amount of the drug will vary with a variety of factors such as the condition to be treated (e.g., in gene therapy), the location of the RASSL-expressing cells within the host to be treated (e.g., within brain tissue, liver tissue, lung tissue, etc.), various patient-dependent or host-dependent (e.g., transgenic animal) factors (e.g., size, weight, age, health, disease severity, responsiveness to therapy, responsiveness to administration of the small molecule ligand, etc.), and various drug-dependent factors (e.g., oral availability, molecular weight, toxicity, RASSL-specific affinity, etc.) Where the synthetic small molecule drug has been previously developed for activation of a native G protein-coupled receptor, guidance for dosages and routes of administration can be found in the Physician's Desk Reference (1995, 49th Ed., Arky et al., eds., Medical Economics Data Production Company, Montvale, N.J.).

[0199] In general, despite any guidance available for administration of the small molecule ligand, the appropriate route of administration and dosage are generally determined on a case-by-case basis by the attending physician. Such determinations are routine to one of ordinary skill in the art (see, for example, Remington's Pharmaceutical Sciences, 18th Ed., 1990, Gennaro et al., eds., Mack Publishing Co., Easton, Pa.). Synthetic small molecule drugs of particular interest are those that can be safely and effectively administered orally or by transdermal delivery by injection (e.g., parenteral injection including subcutaneous, intramuscular, and intravenous injection), but preferably by oral administration.

[0200] The specific dosage appropriate for administration is readily determined by one of ordinary skill in the art according to the factors discussed above (see, for example, Remington's Pharmaceutical Sciences, supra). In addition, the estimates for appropriate dosages in humans may be extrapolated from determinations of synthetic small molecule required to activate RASSL-expressing target cells in vitro and/or in animal studies (e.g., in a transgenic animal having a RASSL-expressing target cell population). For example, appropriate dosages of KOR-binding drugs for an average 70 kg human range from about 10-50 mg, up to about 600 mg or more per day. For example, pentazocine dosages recommended in the Physician's Desk Reference (supra) are about 50 mg every 3 hrs, up to 600 mg per day or more.

[0201] The appropriate formulation of the synthetic small molecule will vary with a variety of factors including the drug used and the route of administration. For example, where the small molecule drug is injected intravenously, the drug is generally formulated in a 0.15 M saline solution containing a desired concentration of drug. Guidance for specific drug formulations can be found in the Physician's Desk Reference, supra, as well as in the literature supplied by the manufacturer.

[0202] Assessment of Efficacy of RASSL-Mediated Targeted Cellular Activation

[0203] The effects of RASSL-mediated, targeted cellular activation can be assessed in various ways. The means of assessing targeted cellular activation will vary with the cell type targeted, the therapeutic goal of targeted cell activation, and the disease or condition to be treated. For example, where the therapeutic goal of targeted cell activation is to increase the overall amount of a protein or other molecule expressed by the target cell (e.g., a natural secretion product of the target cell or a protein or other molecule expressed by the target cell from a heterologous sequence introduced by transformation), the effects of targeted cell activation can be assessed by measuring levels of the protein or other molecule present in the mammalian host (e.g., in the bloodstream, saliva, or other bodily fluid of the host). Such assays can be performed either qualitatively or quantitatively. The ELISA assay, as well as other immunological assays for detecting a protein in a sample, are described in Antibodies: A Laboratory Manual (1988, Harlow and Lane, eds., Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y.).

[0204] Alternatively, or in addition, the efficacy of RASSL-mediated targeted cell activation can be assessed by testing for the level of an activity associated with the therapeutic protein (e.g., an enzymatic activity). Furthermore, the efficacy of RASSL-mediated targeted cell activation can be assessed by monitoring the condition of the mammalian host for improvement. For example, where the therapeutic proteins are blood clotting factors, the subject's blood is examined for blood clotting activity or other parameters associated with hemophilia. Where the RASSLs are expressed in a transgenic, nonhuman animal to provide a reversible disease model, the physiology of the animal RASSL-activation is monitored by the production in the animal of symptoms of the disease or condition to be simulated (e.g., cardiac arrhythmia, pain, osteoporosis, seizures, vasospasm, or alterations in fat or calcium metabolism). Utilities

[0205] The composition and method of the invention are useful in a variety of in vitro and in vivo applications. Targeted, RASSL-mediated cellular activation is useful in various in vitro and in vivo applications in cell selection and culture. For example, the targeted cellular activation method of the invention is useful in the selective amplification of a RASSL-expressing cell line of interest. Cells to be cultured in vitro and subsequently implanted into a mammalian host can be transformed with DNA encoding a RASSL that facilitates cellular proliferation. Proliferation of the RASSL-expressing, transformed cells is induced by exposure to the appropriate synthetic small molecule ligand, thereby facilitating controlled, efficient amplification of the cell line of interest. Cells amplified in this manner may then be implanted into the mammalian host. Use of the RASSL-mediated, targeted cellular proliferation method of the invention allows for selective expansion of a selected cell line in a shorter period of time, and without the use of a selective agent (e.g., antibiotics, methotrexate, metallothionein) which may be undesirable for use in selection of cells for subsequent implantation into a mammalian host (e.g., human, murine, bovine, equine, etc.) Furthermore, since a RASSL is less responsive to a selected endogenous ligand than the native receptor, using a RASSL would pose less risk than using a native receptor for the same purpose.

[0206] The compositions and method of the invention are also useful in facilitating growth and selectively expanding a desired cell type in a primary cell culture in vitro. For example, an in vitro culture of rare or difficult to culture cells can be transformed with RASSL-encoding DNA, thus allowing enhancement of cell culture growth by exposing the cells to the appropriate synthetic small molecule ligand. FIG. 2 schematically illustrates expansion of RASSL-expressing target cells (i.e., by RASSL-mediated triggering of a G protein-mediated proliferative cellular response), relative to proliferation in nature. Thus, the method of the invention is useful in promoting target cell growth at a rate faster than growth of nontarget cells, thereby enriching for target cells in a mixed cell population.

[0207] The RASSL-mediated, targeted cellular activation method of the invention is also useful in regulation of secretion by an in vitro culture of secretory cells. For example, secretory cells of interest are transformed with DNA encoding a RASSL associated with induction of the secretory response. Upon exposure to the appropriate small molecule ligand, the transformed secretory cells are induced to secrete the cellular product of interest. Thus, the method of the invention allows for increased and/or regulated secretion by such cultured cell lines.

[0208] The RASSL-mediated, targeted cellular activation method of the invention is also useful in suppression of a G protein-mediated response. For example, binding of a synthetic ligand to a RASSL can selectively suppress pain, or selectively reduce neural activity. As shown in Example 13, a synthetic ligand, when bound to a KOR RASSL, suppresses the response to olfactory stimuli. Thus, RASSLs are useful in controlling pain and other GPCR-mediated responses.

[0209] RASSLs are also useful in the production of transgenic, nonhuman animals that can serve as reversible disease models. RASSL activation can confer a new biological, cellular response upon administration of the agonist synthetic small molecule ligand to the transgenic animal (e.g., change of heart rate or biochemical changes).

[0210] The method of the invention using RASSL-mediated targeted cellular activation can be used to overcome the present limitations of in vitro transformation efficiency, primary cell isolation and culture, gene therapy efforts, and various other settings in which it is desirable to amplify the number of eukaryotic cells present in a cell population.

[0211] The method of the invention is additionally useful in providing a method to administer a controlled amount of a therapeutic cellular product to a mammalian subject. RASSL-encoding DNA is used to transform cells of an in vivo implant designed to provide a reservoir of cells that secrete a desired protein or other molecule for therapeutic purposes. Secretion of the therapeutic product of such RASSL-expressing cellular implants can be amplified by exposure of the cellular implant to the synthetic small molecule that activates the RASSL.

[0212] Furthermore, targeted cellular activation is useful in combination with in vivo or ex vivo gene therapy. RASSL-encoding nucleic acid is cotransfected with a DNA of interest encoding a gene product to be delivered by gene therapy. The DNA of interest may encode any cellular product desirable for delivery to the mammalian host (e.g., secreted into the host's blood stream or into the host's gastrointestinal tract). For example, the mammalian host may have a disease or condition that is amenable to treatment by expression or over-expression of a protein which is either normally present in a healthy mammalian subject or is foreign to the mammalian host (e.g., a protein having an antibiotic activity, or a protein that is absent or defective in the mammalian host (e.g., protein replacement therapy)). Administration of the appropriate synthetic small molecule activates the RASSL-expressing target cells (e.g., by inducing a proliferative response in the target cells) thereby increasing the number of transformed cells expressing the RASSL and the therapeutic DNA of interest (FIG. 5).

[0213] Thus, where the targeted cellular activation method of the invention results in targeted proliferation of RASSL-expressing cells, the acceptable transformation efficiency for gene therapy is lowered, since transformed cells can be selectively amplified to provide a therapeutically effective dosage of the gene therapy product. Likewise, where the targeted cellular activation method of the invention results in increased cellular secretion, the acceptable level of transformation efficiency for gene therapy is likewise lowered, since fewer transformed cells can provide a higher level of secretion of the therapeutic product of interest. Once the clinical effect is achieved (e.g., a desired dosage of therapeutic cellular product is delivered to the subject), the drug is removed to prevent unwanted proliferation or secretion. Exemplary applications of the targeted activation method (e.g., by proliferation or secretion) of the invention are provided in Table 3.

TABLE 3
Exemplary Applications of Targeted
Proliferation in Gene Therapy
Tissues
for targeted
activation	Therapeutic gene product	Disease target
Liver cells	LDL-receptor	Atherosclerosis
	Lipoproteins	Atherosclerosis
	Blood clotting factors	Hemophilia
Cells of	ApoE to vascular wall	Atherosclerosis
Hematopoietic	Hirudinlantithrombotics	Atherosclerosis
Lineage	ApoE/growth factors to brain	Alzheimer's
Macrophages
T cells	Dominate negative HIV proteins	AIDS
B cells	Deficient genes in SCID	Immune deficiencies

[0214] RASSL-mediated, targeted activation is also useful in facilitating and regulating delivery of therapeutic, bioactive proteins in the brain or any other part of the body where it is advantageous to gain pharmacological control of a population of cells. The delivery of bioactive proteins to the brain by conventional methods has proven difficult since the blood-brain barrier blocks the entry of most peripherally administered therapeutic proteins (e.g., therapeutic proteins administered orally or by intravenous injection). Since many synthetic small molecules are small enough and of appropriate ionic charge to cross the blood-brain barrier, the RASSL-mediated targeted activation method of the invention overcomes this problem.

[0215] For example, the targeted activation method of the invention is useful in facilitating intracranial delivery of nerve growth factor (NGF), which is currently proposed for administration to Alzheimer's patients in clinical trials by direct intracranial injection. First, a cell line that secretes NGF or that can be engineered to secrete NGF (e.g., adrenal chromaffin cells, myoblasts, myotubes, fibroblasts) is transformed with RASSL-encoding nucleic acid to provide a RASSL-expressing, NGF-secreting target cell. For example, glial cells, which package and secrete foreign proteins from secretory granules, can be transformed with NGF-encoding DNA and RASSL-encoding DNA (Tuszynski et al., 1994, Experim. Neurol. 126:1-14; Cunningham et al., 1994, Brain Res. 658:219-231). A reservoir of RASSL-expressing, NGF-secreting target cells are transplanted in the brain.

[0216] When increased NGF secretion is desired, the patient simply takes an oral or intravenous dose of the appropriate synthetic small molecule, which subsequently crosses the blood-brain barrier, binds the RASSLs and activates NGF secretion and/or target cell growth. This system provides a simple method for stimulation, as well as regulation, of secretion of protein in the brain. Delivery of growth factors such as NGF by the method of the invention can similarly be used to treat various other neurological disorders including, but not limited to, Parkinson's disease and amyotropic lateral sclerosis (ALS, also known as Lou Gehrig's disease).

[0217] The method of the invention further provides a method for regulated release (e.g., timed release) growth factor secretion. Because many growth factors have natural diurnal rhythms, it thus may be preferable to trigger growth factor secretion while the subject is asleep. For example, a patient having a RASSL-expressing, NGF-secreting target cell implant simply takes a dose of the RASSL agonist before bedtime. The small molecule agonist crosses the blood-brain barrier and facilitates increased NGF secretion while the patient sleeps. By morning the drug is metabolized, resulting in decreased RASSL activation and decreased secretion of NGF.

[0218] Similarly, the method of the invention can be used to treat and/or prevent development of Alzheimer's disease by delivery of specific isoforms of ApoE to the patient's brain. Patients who are homozygous for the ApoE e4 allele (which accounts for about 2% of the U.S. population) are at an 85% risk of developing Alzheimer's disease. Patients who are at risk of developing Alzheimer's disease, or who have already developed Alzheimer's symptoms, can be treated by delivery of the ApoE isoforms (e.g., e2 or e3) that are not associated with Alzheimer's disease. RASSL-encoding nucleic acid and ApoE e2- or e3-encoding nucleic acid is used to transform microglia cells of the brain either in vivo or ex vivo. Alternatively, cells heterologous to the brain may be transformed in vitro with the RASSL-encoding and ApoE e2- or e3-encoding nucleic acid, cultured in vitro and subsequently used in a cellular implant that expresses the RASSL and the desired ApoE isoform. Alternatively, because microglia are of hematopoietic origin, stem cells of the patient can be transformed with the RASSL-encoding and ApoE e2- or e3-encoding nucleic acid using either in vivo or ex vivo gene therapy techniques. Since G₁ and G_q signals are implicated in proliferative response of hematopoietic cells, the RASSL used in to facilitate microglial proliferation can be derived from either the native KOR or native serotonin receptors, respectively.

[0219] Once the RASSL-expressing, ApoE e2- or e3-expressing cells are in place in the patient's brain, proliferation (and thus production of ApoE e2 or e3) can be induced by administration of the appropriate synthetic small molecule drug to the patient. Expression of the desired ApoE isoform, and the balance of ApoE e4 and ApoE e2 or e3 in the patient, can be monitored by examining the patient's central spinal fluid (CSF) for the presence of these proteins using methods known in the art. The presence of the e2 or e3 isoform in an amount greater than that of the e4 isoform is indicative of treatment and/or prevention of Alzheimer's disease.

[0220] The following examples are put forth so as to provide those of ordinary skill in the art with a complete disclosure and description of how to make and use the present invention, and are not intended to limit the scope of what the inventors regard as their invention nor are they intended to represent that the experiments below are all or the only experiments performed. Efforts have been made to ensure accuracy with respect to numbers used (e.g. amounts, temperature, etc.) but some experimental errors and deviations should be accounted for. Unless indicated otherwise, parts are parts by weight, molecular weight is weight average molecular weight, temperature is in degrees Centigrade, and pressure is at or near atmospheric.

[0221] The general strategy for making a RASSL based on a G protein-coupled receptor that binds a peptide ligand is to mutate the extracellular loops of the receptor. The extracellular loops are the putative binding sites of peptide ligands (but not small molecule ligands). The extracellular loops can be identified for any G protein-coupled receptor by locating the transmembrane region. The mutations can be chosen by substitution residues from related receptors, or by insertion of randomly chosen residues (e.g., saturation mutagenesis).

[0222] To ease the detection of the recombinant RASSL the N-terminus of this modified RASSL with the epitope for the FLAG monoclonal antibody (commercially available from Kodak-IBI, New Haven Conn.). This epitope does not affect binding or signaling by agonist drugs. The FLAG epitope allows for immunological detection by western blotting, immunohistochemistry, and FACS sorting. In addition, the RASSL-encoding sequence can include a sequence encoding an epitope (YPYDVPDYA SEQ ID NO:12) for the anti-HA monoclonal antibody (referred to hereafter as the HA tag; Boehringer Mannheim) for ease in detection and purification of the product. The RASSL-encoding sequence may also include a Prolactin signal sequence to facilitate post-translational processing and delivery to the cell surface.

[0223] The amino acid (SEQ ID NO:2) and nucleotide (SEQ ID NO:1) sequences of human KOR are provided in the sequence listing below. The nucleotide sequence (SEQ ID NO:3) of wild-type human KOR having an N-terminal Prolactin signal sequence (nucleotides 1-90; lower case letters), a FLAG epitope tag (nucleotides 91-114; lower case letters), and an HA tag (nucleotides 1252-1284; lower case letters) is shown in FIGS. 7A and 7B.

[0224] A RASSL derived from the human kappa opioid receptor (KOR) was generated by mutation of specific KOR codons encoding selected amino acids of the extracellular loops. Briefly, cDNA encoding human KOR was inserted into a mammalian expression vector. The amino acid and nucleotide sequences of both the human KOR and DOR (delta opioid receptor) are known (see Zhu et al., 1995, Life Sci. 56: 201-207 (KOR sequence); Evans et al., 1992, Science 258:1952-1955; Kieffer et al., 1992, Proc. Natl. Acad. Sci. USA 89:12048-12052 (cloning and sequence of DOR)). The specific amino acids associated with each structural domain are shown in Table 4 (from SWISS-PROT):

TABLE 4
Structural Domains of KOR
Domain	Amino Acid Residue	Position
N-Terminus	1	58	EXTRACELLULAR
TMH1	59	85	TRANSMEMBRANE
i1	86	95	CYTOPLASMIC
TMH2	96	117	TRANSMEMBRANE
e1	118	132	EXTRACELLULAR
TMH3	133	154	TRANSMEMBRANE
i2	155	173	CYTOPLASMIC
TMH4	174	196	TRANSMEMBRANE
e2	197	222	EXTRACELLULAR
TMH5	223	247	TRANSMEMBRANE
13	248	275	CYTOPLASMIC
TMH6	276	299	TRANSMEMBRANE
e3	300	311	EXTRACELLULAR
TMH7	312	333	TRANSMEMBRANE
C-Terminus	334	380	CYTOPLASMIC
Disulfide bond	131	210
LIPID	345	345	PALMITATE
CARBOHYDRATE	25	25	POTENTIAL
CARBOHYDRATE	39	39	POTENTIAL

[0225] A site-specific human KOR mutant, termed OR1, was generated by site-specific mutation of the human KOR receptor sequence to generate a human KOR mutant nucleotide sequence having 17 specific amino acid substitutions. Each of the amino acids changed in the human OR1 mutant were changed to the amino acid residue at the corresponding position in the DOR amino acid sequence. FIG. 6 is a schematic of this human KOR mutant showing the two-dimensional structure, amino acid sequence, and the position of the amino acid changes in OR1. The nucleotide (SEQ ID NO:5) and amino acid sequences (SEQ ID NO:6) of the RASSL OR1 having an N-terminal prolactin signal sequence (nucleotides 1-90; lower case letters), a FLAG epitope tag (nucleotides 91-114; lower case letters), and an HA epitope tag (nucleotides 1243-1275; lower case letters) are shown in FIGS. 8A and 8B. The nucleotide sequence of OR1 corresponding to human KOR sequences (nucleotides 474-701) are in upper case; sequences corresponding to non-human KOR sequences (nucleotides 702-787) are in lower case.

[0226] OR1 was expressed in COS-7 and CHO-KI cells, and characterized for dynorphin binding, bremazocine binding, spiradoline binding, and activation of G protein-mediated signals as previously described (Conklin et al., 1993, Nature 363:274-276; Wong et al., 1992, Science 255;339-342; Federman et al., 1992, Nature 354:159-161). The results of the binding experiments and the signaling experiments are summarized in Tables 5 and 6, respectively. Table 7 summarizes the results of these studies. OR1 exhibited a 200-fold decrease in dynorphin binding (Ki14.6 nM) relative to dynorphin binding by native human KOR (Ki0.064 nM). In contrast, OR1 exhibited no significant change in bremazocine binding affinity (OR1 Ki0.064 nM; native human KOR Ki0.038 nM), and a relatively small change in spiradoline binding (OR1 Ki6.29; human KOR Ki1.132). Signaling studies showed that OR1 inhibits adenylyl cyclase (the expected biochemical effect) in response to bremazocine and other small molecule drugs and high concentrations of dynorphin. Thus, OR1 is especially suitable for use as a RASSL in the present invention.

TABLE 5
Binding of Dynorphin, Bremazocine and Spiradoline
to Wild-type KOR, OR1, and OR2
	Wild-type KOR
	Fold	OR1	OR2
	Ki	de-	Ki	Fold	Ki	Fold
Ligand	(in nM)	crease*	(in nM)	decrease*	(in nM)	decrease*
Dynor-	0.064	1	14.64	229	124.52	1946
phinA	0.043		2.99		19.38
Brema-	0.038	1	6.29	1.7	5.65	1.2
zocine	0.012		1.92		1.19
Spira-	1.32	1	6.29	4.8	5.65	4.3
doline	0.38		1.92		1.19
*Fold decrease in binding is relative to binding to wild-type KOR (Ki for wild-type KOR divided by Ki for OR1 or OR2)

[0227]

TABLE 6
Signaling by Wild-type KOR, OR1,
or OR2 in the Presence of Dynorphin A or Bremazocine to
Wild-type KOR, OR1, and OR2 (EC50 in nM)
	Wild-type KOR
	Fold	OR1	OR2
		de-		Fold		Fold
Ligand	EC50	crease*	EC50	decrease*	EC50	decrease*
Dynor-	0.66	1	61.03	92.8	1083.73	1648.8
phinA	0.62		20.56		367.24
Brema-	0.067	1	0.068	1	0.34	5.17
zocine	0.05		0.039		0.21
*Fold decrease in signaling is relative to signal detected with wild-type KOR (for wild-type KOR divided by Ki for OR1 or OR2)

[0228]

TABLE 6
The RASSLs OR1 and OR2
	Binding	Biochemical Signal	Growth Signal
	Peptide	Drug 1	Peptide	Drug 1	Peptide	Drug 2
OR1	200x	NC	95x	NC
OR2	2,000x	NC	1,600x	5x	2,000x	NC
NC No Change;
Peptide dynorphin;
Drug 1 bremazocine;
Drug 2 spiradoline

[0229] Another RASSL derived from the human kappa opioid receptor (KOR), OR2, was generated in a manner similar to that described in Example 1. OR2 was generated by site-specific mutation of the human KOR receptor sequence to generate a human KOR mutant nucleotide sequence having the same 17 specific amino acid substitutions as in OR1, as well as an glutamic acid-to-glutamine (E-to-Q, see FIG. 6). FIG. 6 provides a schematic of the OR2 RASSL. FIGS. 9A and 9B show the amino acid (SEQ ID NO:8) and DNA sequences (SEQ ID NO:7) of OR2, and indicate the position of the N-terminal prolactin signal sequence (nucleotides 1-90; lowercase), the FLAG tag (nucleotides 91-114), the C-terminal HA tag (nucleotides 1243-1275), the sequences derived from the wild-type human KOR (nucleotides 474-701; upper case), the non-human KOR sequences (nucleotides 702-787; lower case), and the position of the glutamic acid-to-glutamine mutation (nucleotides 987-991).

[0230] OR2 was expressed in COS-7 and CHO-KI cells, and characterized for dynorphin binding, bremazocine binding, spiradoline binding, and activation of G protein-mediated signals as described in Example 1. OR2's binding affinity and ability to elicit a biochemical signal upon binding are summarized in Table 5. The binding affinity of OR2 for dynorphin was reduced by 2,000-fold. In contrast, OR2 exhibited no change in bremazocine binding affinity (OR2 Ki0.046 nM; native human KOR Ki0.038 nM), and a relatively small change in spiradoline binding (OR2 Ki5.65; human KOR Ki1.32). The differing binding affinities of OR2 for dynorphin and bremazocine were reflected in the signaling assays; activation of the biochemical signal required about 1,600 times more dynorphin to have the same effect has compared to dynorphin acting on the wild type KOR.

[0231] The small molecule drug bremazocine activated the OR2 receptor with high affinity. However, activation of the biochemical signal required about 5 times more bremazocine compared to the amount required to activate the wild-type KOR. Because OR2 binds and is activated by a small molecule drug (bremazocine) superiorly to a selected natural ligand (dynorphin), OR2 is especially suitable as RASSL for use in the invention.

[0232] Nucleic acid encoding the RASSL derived from the RASSL OR2 was introduced into rat 1 a fibroblasts and cultured in vitro. Briefly, the cells were made quiescent (i.e., were synchronized so that all cells wherein a low growth state) by first seeding 510⁵ cells per well of a 24-well plate in Dulbecco's Modified Eagles (DME) medium with 10% calf serum. After incubating for 12-24 hours, the medium was removed, the cells rinsed with serum-free media, and 1 ml of serum-free media. The cells were than incubated for an additional 24 hours. The ability of the OR2 RASSL to facilitate cell proliferation was tested by adding spiradoline (OR2 agonist) or dynorphin (selected natural ligand; negative control) at various amounts. After incubating for 16 hours, 1 microcurie of ³H-thymidine (NEN NET-027Z) (1 l diluted with 24 l of media) was added to each well, and the cells incubated for an additional 8 hours.

[0233] Cell proliferation was measured by incorporation of the ³H-thymidine as previously described (Conklin et al., 1988, Proc. Natl. Acad. Sci. USA 85:8698-8702). Briefly, the ³H-thymidine labeled DNA was extracted by first aspirating the media, and washing the cells carefully with 1 ml ice cold PBS. The PBS was aspirated away and 1 ml of ice cold 5% TCA was added. After incubation at 4 C. for 30 minutes, the TCA was aspirated away, the cells washed once with PBS, and 0.5 ml 0.5N NaOH/0.5% SDS was added at room temperature. The samples were then collected and added to scintillation vials for detection of the amount of radiation present in the sample.

[0234] The results of the targeted proliferation assay are shown in the graph of FIG. 10. Exposure of the OR2-expressing rat 1a fibroblasts to the synthetic ligand spiradoline resulted in a 20-50 fold increase in cell proliferation over exposure of the cells to the selected natural ligand dynorphin. Thus, the OR2 RASSL can be used to facilitate selective growth of OR2-expressing cells.

[0235] The ability of a RASSL to facilitate cell proliferation in a mixed cell population is tested by culturing RASSL-expressing cells with cells of the same cell type that do not express the RASSL (nontarget cells). The mixed cell population is seeded in the culture flask such that the RASSL-expressing cells comprise only about 1% of the total cell population. The cells are then cultured in the presence of spiradoline for 5 to 7 days. The cells are then removed from the culture flask, and RASSL-expressing cells counted by FACS sorting using a fluorescently-labeled, anti-RASSL antibody.

[0236] The DNA encoding the OR1 RASSL described in Example 1 was inserted into a mammalian expression construct so that expression of the OR1-encoding DNA was driven by a MMTV-LTR promoter according to the expression system developed by Gossen and Bujard (1992, PNAS 89:5547-5551; Gossen et al. 1992, J. Cell Biochem. 59:463-472). OR1 expression construct was then microinjected into fertilized mouse eggs, and the microinjected eggs implanted into pseudopregnant mothers according to methods well known in the art. White blood cells were isolated from the resulting transgenic OR1 mice, and expression of OR1 examined in these cells by FACS using the FLAG monoclonal epitope. Approximately 20% of the white blood cells of the transgenic mice expressed OR1.

[0237] Spiradoline or other small molecule KOR drug is administered to a KOR RASSL-expressing transgenic mice (e.g., generated according to the method described in Example 5) with 1 mg/liter in water (oral), or 5 mg/day by injection for up to one year. On weekly intervals, white blood cells are isolated from both KOR RASSL-expressing mice that received the spiradoline, as well as from KOR RASSL-expressing mice that received no drug (control). The percentage of KOR RASSL-expressing white blood cells in each cell sample is determined by FACS.

[0238] KOR RASSL-expressing bone marrow cells are isolated from transgenic mice generated with RASSL-encoding DNA according to the method described in Example 6. The bone marrow cells are then injected into normal mice. A sample of bone marrow from the KOR RASSL-expressing mice is analyzed by FACS to determine the percentage of KOR RASSL-expressing cells in the population. Spiradoline (or other KOR agonist) is administered to the mice. After 6 months of treatment with the KOR agonist, peripheral cells and/or bone marrow cells are isolated from mice that received the bone marrow transplant (test group), and from mice that received spiradoline, but did not receive the transplant (control group), and from mice that received the transplant, but did not receive spiradoline. The percentage of KOR RASSL-expressing cells is analyzed by FACS as described above and compared to the percentage of KOR RASSL-expressing cells present in the control mice.

[0239] Serotonin is a biogenic amine. As shown in Table 1, all receptors for biogenic amines bind to their natural ligand via a common mechanism that involves specific residues in the transmembrane helixes: TM^III, TM^V, TM^VI (corresponding to AspII:08, SerV:09, SerV:12, PheVI:17 of the B₂-adrenergic receptor; see Schwartz, 1994, supra). Thus, mutations at one or more of these sites would decrease binding to a selected biogenic amine natural ligand, but not the structurally unrelated small molecule ligands.

[0240] For example, a RASSL can be generated from the B₂-Adrenergic receptor (₂AR) by changing the AspIII:08 to a serine. This new receptor no longer binds its natural ligand, yet it will bind, and be activated by, several synthetic small molecule ligands (Strader et al., 1991, J. Biol. Chem. 266(1):5-8). FIG. 11 shows an alignment of four exemplary biogenic amine receptors (the human ₂AR, the human serotonin 1A-receptor (h5HT-1A-R), the human serotonin-2C-receptor (h5HT-2CR)), and the human serotonin 4 receptor (h5HT4-R), demonstrating the identity of the transmembrane domains (indicated by TM), and the candidate amino acid residues that can be altered to generate a RASSL of the invention.

[0241] Variants of Apolipoprotein E (ApoE) account for a majority of late-onset Alzheimer's disease (for recent reviews see Reiman et al., 1996, New Engl. J. Med. 334:752-758; Campion, 1996, New Engl. J. Med. 334:791-792). Two percent of the US population is homozygous, and 30% heterozygous, for the ApoE e4 allele, which is associated with development of Alzheimer's. Delivery to the brain of ApoE variants (e2 or e3) may be beneficial for the treatment of Alzheimer's disease.

[0242] Delivery of ApoE e2 or e3 can be accomplished using the present invention by transfecting microglia (a major source of ApoE in the brain) with nucleic acid encoding ApoE of the e2 or e3 isoform, and nucleic acid encoding a RASSL of the invention. Alternatively, since microglia are of hematopoietic origin, hematopoietic stem cells (from either bone marrow or peripheral blood of the patient) are transfected with nucleic acid encoding a RASSL and the ApoE e2 or e3 isoform. Since G₁ and G_q signals are implicated in proliferative responses of hematopoietic cells, either KOR-based or serotonin receptor-based RASSLs, respectively, are appropriate for inducing a proliferative response in these cells.

[0243] The transfected cells are infused back into the patient, and the RASSL agonist is administered (e.g., as a pill taken several times a day). The transfected cells may constitute less than 0.01% of all the patient's stem cells, but RASSL stimulation by administration of the RASSL agonist confers a proliferative advantage upon the RASSL-expressing transfected cells, thereby facilitating a gradual increase in the percent of transfected cells present in the patient up to several percent or even a majority of the cells. The percentage of transfected RASSL-expressing, ApoE-expressing cells is monitored over time by analyzing the peripheral blood for the presence of the RASSL. Gradually, the RASSL containing cells colonize the brain in the form of microglia, which are derived from macrophages and monocytes. Treatment with the RASSL agonist can be continued for months or years to provide for expansion of the RASSL- and ApoE e2- or e3-expressing cells in the patient.

[0244] After a period of several months or years the central spinal fluid (CSF) of the patient is analyzed for the Apolipoprotein E isoform(s) to detect a change in the balance of the ApoE isoforms from the Alzheimer's promoting ApoE e4 isoform to another variant such as ApoE e2 or e3. This change in ApoE isoforms can delay or prevent the onset of Alzheimer's disease.

[0245] Many neurological disorders such as Alzheimer's disease, Parkinson's disease, and Amyotrophic lateral Sclerosis (ALS), are targets for treatment with neural growth factors such as Nerve Growth Factor (NGF). Although these growth factors can be produced in recombinant techniques in large quantities, they are difficult to deliver to the brain because they do not cross the blood-brain barrier.

[0246] To accomplish regulated, intracranial delivery using the compositions and methods of the invention, microglia are transformed with DNA encoding NGF (or other desired growth factor) according to methods well known in the art, methods for RASSL medicated expansion of transfected microglia described above in Example 9 could be used. Methods for expression of NGF in transformed cells are known in the art (see, e.g., Cunningham et al., 1994, Brain Res. 658:219-231; Tuszynski et al., 1994, Exper. Neurol. 126:1-14). Alternatively, the glial cells can be transformed directly using in vivo gene therapy techniques.

[0247] Where an ex vivo a gene therapy protocols is used, glial cells expressing both NGF and the RASSL are cultured in vitro, and those RASSL-expressing cells that proliferate and secrete NGF in response to exposure to a synthetic small molecule ligand are used in a cellular implant. After the RASSL-expressing, NGF-producing cellular implant is in place, the synthetic small molecule ligand that binds and activates the RASSL is administered. The synthetic small molecule is of a molecular weight and ionic charge that allows the drug to cross the blood-brain barrier and bind to the RASSLs of RASSL-expressing cellular implant, thus activating the RASSL and facilitating cell proliferation and NGF secretion.

[0248] A RASSL agonist that can cross the blood-brain barrier to facilitate delivery of NGF to brain tissue is administered to the patient for several months to years. Therapeutic endpoints may require only a small percentage of microglia to secrete the neural growth factors since these growth factors are quite potent at low concentrations. Many RASSL agonists can cross the blood brain barrier to facilitate the delivery of proteins that can not cross the blood-brain barrier.

[0249] The OR1-expressing transgenic mouse described in Example 5 was used as the basis of an animal model for cardiomyopathy. When activated, the OR1 RASSL results in activation of the G_i protein signaling system. The G₁ protein signaling system of cardiac myocytes has been implicated in regulation of heart rate. For example, several investigators have shown a positive correlation between the presence of antibodies that stimulate the G_i-coupled muscarinic acetylcholine receptor and several cardiomyopathies (e.g., idiopathic dilated cardiomyopathy, and Chagas disease-associated cardiomyopathy) (see, e.g., Goin et al., 1994, Neuroimmunomodulation 1:284-291; Fu, 1995, Euro. Heart J 16(Suppl 0):89-91; Fu et al., 1992, Cardio. Res. 26:950-955; Fu, 1993, J. Clin. Invest. 91:1964-1968; Matsui et al., 1995, Autoimmunity 21:85-88). However, this hypothesis has been difficult to prove because the patients have antibodies to many other surface proteins on the heart.

[0250] Two transgenic mice were generated: one expressing myosin heavy chain alpha promoter (MHC) during tTA expression (MHC-tTA) (to provide cardiac expression of tTA; see Yu et al., 1996, Circulation Res. 79:691-697; for a description of the tTA system, see Gossen et al., 1992, Proc. Natl. Acad. Sci. USA 89:5547-5551; and Gossen et al., 1995, Science 268:1766-1769) and TRE-OR1 (OR1 RASSL linked to the tetracycline response element (TRE)). The mice were then cross-bred to provide a double transgenic mice. The expression system used allowed for regulated OR1 expression, since the TRE promoter will not be transcribed in the presence of doxycycline. Doxycycline (200 mg/ml in water), which passes efficiently to the fetus and through mother's milk to the newborns, was administered during breeding and continued after weaning. This dose of doxycycline has no known adverse effects on development, and may increase liter size. Doxycycline was replaced with regular water 3 weeks after weaning (day 0 on X-axis of FIG. 12); our previous work has indicated that at least 10 days is required to wash out this high dose of doxycycline. These experiments are referred to as rapid induction of OR1 gene expression to differentiate them from studies in which the dose of doxycycline is decreased gradually over 3 months. The control mice were littermates having only one of the transgenes (either MHC-tTA or TRE-OR).

[0251] The results of withdrawal of doxycycline are shown in FIG. 12. As long as the mice were fed doxycycline, they were completely healthy; however, about 75-60% of the mice die 10 days after the start date of a doxycycline-free diet. In contrast, double-transgenic mice continuously fed doxycycline are healthy for at least 3 months. Although the OR1 receptor is reduced in binding to the selected natural ligand dynotphin, activation of OR1 in the double-transgenic mice may have occurred by binding of a nonselected natural ligand (i.e., a natural ligand other than dynorphin).

[0252] Despite the apparent existence of another endogenous ligand for OR1 binding, OR1 still meets the definition of a RASSL of the invention since it is still reduced for binding of, and activation by, the selected ligand, dynorphin; and retains binding for a synthetic small molecule (e.g, spiradoline). The nonselected natural ligand in the heart associated with OR1 activation may be a derivative of Met-enkephalin (Met-Enkephalin-Arg-Phe), which has been found in the heart (Barron et al., 1995, Peptides, 16:1221-1227; Mansour et al., 1995, Brain Res. 700:89-98) and may bind the transgenic KOR RASSL.

[0253] Because activation of the endogenous G₁-coupled receptor of the heart (cardiac muscarinic) slows heart rate, the G₁-protein-coupled RASSL OR1 would be expected to reduce heart rate in the transgenic mice. Further while OR1 apparently has a low level of basal activity (as suggested by the data in FIG. 12 described above), supra-activation of the OR1 RASSL with synthetic agonist would further enhance this phenomenon.

[0254] This hypothesis was tested by performing electrocardiograms (EKG) on the double transgenic mice. Double-transgenic mice that survived the experiment immediately above were anesthetized with phenobarbital, intubated, and connected to a mouse respiratory ventilator. EKGs were obtained with a standard clinical EKG machine attached to a GRASS physiologic recorder. The control animal in this experiment was a littermate that carried only the MHC-tTA transgene. After taking baseline recordings, 2 mg of the opioid agonist spiradoline (Pharmacia-Upjohn Kalamazoo, Mich.) was injected intraperitoneallly (IP).

[0255] The results of this experiment are shown in FIG. 13. Recordings obtained a few seconds after the injection show markedly decreased heart rate and an apparent conduction defect consistent with atrioventricular heart block (more apparent in the 5-minute recordings). Similar results have been obtained in two other sets of experiments. The change in the EKG wave form of the control mouse in this experiment may result from a change in the mouse position during the EKG; atrioventricular block or other changes in EKG wave form were not observed in control mice following spiradoline injection.

[0256] These data demonstrate that the opioid agonist immediately induced heart block in the OR1-expressing mice consistent with an atrioventricular node disturbance. The rhythm disturbance in the agonist-treated, OR 1-expressing mice was similar to the finding in patients with cardiomyopathies associated with muscarinic receptor antibodies (see, e.g., Goin et al., 1994, Neuroimmunomodulation 1:284-291; Fu, 1995, Euro. Heart J 16(Suppl 0):89-91; Fu, 1993, J. Clin. Invest. 91:1964-1968; Matsui et al., 1995, Autoimmunity 21:85-88). This experiment also demonstrates that the opioid receptor can still be supraactivated from its baseline state, despite a basal level of activation by a nonselected endogenous ligand.

[0257] Congestive heart failure (CHF) was mimicked by slowly decreasing, over the course of 3 months, the amount of doxycycline administered to the double transgenic mice described in Example 11, thereby allowing gradual (rather than rapid) expression of OR1 in heart tissue (see FIG. 14). The CHF syndrome in mice was defined as (1) greater that 20% increase in weight in one week, (2) decreased motor activity of mice, (3) eventual death, (4) fluid accumulation in the mice, and (5) normal blood chemistries (ruling out other noncardiac forms of fluid retention).

[0258] The results are shown in FIG. 14. Following reduction of doxycycline, 5 of the 10 mice developed swelling of soft tissues (interstitial edema), weight gain (20-70% increase from baseline), ascites, and pleural effusions similar to severe CHF within the time period (several weeks) that the animals were on less than 2 mg/ml of doxycycline in their water supply.

[0259] In one representative double-transgenic OR1-expressing mouse, the mouse was normal until its dose of doxycycline was lowered from 0.4 mcg/ml to 0.1 mcg/ml. One week after this dose change, the mouse developed rapid weight gain and died 6 days after first developing obvious symptoms of heart failure. The mouse's estimated weight gain was over 50% in one week. At autopsy, this mouse had over 3 ml of clear fluid in its abdominal cavity (no fluid is normal) and a large clear gelatinous subcutaneous accumulation, consistent with interstitial edema (often seen in humans with anasarca). None of the control mice (mice with only one of the two transgenes) showed any deleterious effect of having doxycycline removed from their water supply.

[0260] These data show that the OR1-expressing mice provide a model of receptor-induced CHF, useful in the dissection the molecular events that lead to CHF, as well as a mean to screen drugs that may be useful in the treatment of CHF or its symptoms.

[0261] To show that RASSLs can modulate neural activity we have transgenically introduced the RASSL (Rol) into the olfactory bulb of a mouse and showing that RASSL activation blocks olfactory input in a mouse. Since olfaction (smell) is the primary sensory input in a mouse, the olfactory bulb is a highly specialized portion of the central nervous system of a mouse.

[0262] We generated a transgenic mouse that produces OR1 selectively in the olfactory bulb and the cerebellum. To achieve this, we used the KOR knockout mouse, which we made transgenic using a construct encoding OR1 under control of a minimal promoter and the tetracycline operator (tetO), and a construct encoding the tetracycline transactivator tTA under control of the prion protein promoter. In this combination, OR1 is selectively expressed in the olfactory bulb and the cerebellum.

[0263] We have demonstrated that the RASSL, Rol is expressed in the mouse olfactory bulb by immunohistochemistry, and staining for beta-galactosidase (an enzyme that is co-expressed with Ro1 (the genes were co-integrated in the genome). To further eliminate confounding effects of the RASSL agonist (Spiradoline) we bred the mice into a genetic background that lacks any endogenous Kappa Opioid Receptor (KOR-knockout mice). Thus, Spiradoline is specific for Rol in these mice (since the KOR is gone), and Spiradoline only activates Rol because it is a RASSL (decreased affinity for natural peptide ligands by 200).

[0264] To test for olfaction we placed the mice in rectangular box with novel odorants soaked Q-tips in three of the corners and water soaked Q-tip in the fourth corner. A normal mouse will spend much more time smelling and exploring the Q-tips with novel odors (as compared to water) and this can be video taped and scored (one point for directly touching the Q-tip with its nose).

[0265] It is well known that mice spend more time examining objects that have odorants that appear new to them, i.e., they exhibit a preference for new odorants. Knockout mice, OR1 mice, and OR1 mice injected with spiradoline were provided a choice of three different odorants, and water. The results are summarized in Table 1, below. A indicates that the mice exhibited a preference for the odor choice; a indicates no preference.

TABLE 1
	Knockout mouse;		RO1 transgenic;
Odor	injected with		injected with
choice	spiradoline	RO1 transgenic	spiradoline
Odorant 1
Odorant 2
Odorant 3
Water

[0266] The results indicate that when OR1 is produced in the olfactory bulb, OR1 activation turns off the transmission circuit from the nose to the brain. It should be noted that spiradoline did not affect the motor behavior of the mice. In fact the OR1 transgenic mice injected with spiradoline still examined the objects, but they examined the water treated objects just as much as the odorant treated objects. The importance of this experiment is that we have now shown that we can turn off a specific neural circuit with a RASSL.

[0267] We found that the mice expressing Rol had completely normal response on this test before injection with Spiradoline. However a few minutes after Spiradoline injection the mice made not distinction between the water and odorant Q-tips. This test was validated by multiple controls to show the effect was specific for Spiradoline acting on Ro1 (not other receptors). Spiradoline had no effect on littermate mice that were not expressing Ro1, and the effect of Spiradoline was reversible when the animals were tested several days later. The effect of Spiradoline was also completely dependent on the expression of Ro1 since the same animals were retested after Ro1 was suppressed (Rol was expressed on the tetracycline transactivator system). Finally the effect of Spiradoline was specific since Spiradoline did not affect motor activity (total movement in cage) or coordination (roto-rod test and ability to hold on a inverted cage top).

[0268] Therefore we have shown that the expression of the RASSL, Rol in nervous tissue will block a specific neural pathway (olfaction) that is central to the sensory system of a mouse.

[0269] The regulation of G protein-coupled receptors, through desensitization and internalization, modulates the length of the receptor signal. This receptor regulation may influence the development of tolerance and addiction in response to chronic drug treatment. The C-terminal region plays an important role in the desensitization and internalization of GPCRs.

[0270] In order to visualize these RASSLs in living cells we fused the green fluorecent protein to the N-terminus (outer portion) of a KOR RASSL, resulting in Rig (RASSL Gi green). We made a series of mutations in the C-terminal region, thereby generating RASSLs with completely different receptor desensitization and internalization. Removal of phosphorylation sites in the C-terminus of the RASSL resulted in a mutant that was resistant to internalization and less prone to desensitization. Because these RASSLs are all activated by spiradoline, they allow greater control of the timing and specificity of activation than is possible using endogenous receptors. This modification allows us to compare, both in vitro and in vivo, receptors that have the same ligands and effectors, but differ only in desensitization and internalization. Chimeric receptors with the C-terminus of the mu opioid receptor replacing the corresponding portion of the kappa opioid receptor showed decreased basal expression on the plasma membrane; increased agonist-induced desensitization; and decreased induction of adenylyl cyclase superactivation after chronic agonist treatment. These mutant RASSLs are useful in exploring the temporal dynamics of G protein-coupled receptor signaling.

[0271] Materials and Methods

[0272] Construction of Mutant Receptors

[0273] All receptors were based on the KOR2 RASSL described above. Rig, a GFP-tagged version of the KOR RASSL, was made by inserting the coding sequence for emerald GFP (Packard) at the N-terminus of the receptor, after a FLAG tag and the first 8 amino acids of the RASSL. To create Rig-A (also referred to herein as Ro3), two serines and two threonines in the C-terminal region of the receptor were mutated to alanines (FIG. 15). In the human KOR, the mutated residues correspond to S356A, T357A, S358A and T363A. The nucleotide sequence (SEQ ID NO:13) of the Ro3 (Rig) open reading frame, as well as the Ro3 amino acid sequence (SEQ ID NO:14) is shown in FIG. 16. The nucleotide (SEQ ID NO:15) and amino acid (SEQ ID NO:16) sequences of Ro3A (RigA) are shown in FIG. 17. The nucleotide (SEQ ID NO:17) and amino acid (SEQ ID NO:18) sequences of Rig-A are shown in FIG. 18.

[0274] For both Rig- and Rig-A, the final 35 amino acids (345-380) of the KOR were replaced by 47 C-terminal residues from the rat opioid receptor (MOR). Rig-A contains the following additional modifications to the rat MOR C-terminus: T383A, E388Q, E391Q, E393Q and T394A. For each receptor, a schematic design and a C-terminal amino acid sequence alignment is shown in FIG. 15.

[0275] Expression of RASSLs in Mammalian Cells

[0276] HEK 293 cells in culture were transfected using Lipofectamine Plus (InvitRigen, Carlsbad, Calif.). Experiments on transiently transfected cells were performed approximately 48 hours after transfection. To create stable cell lines, transfected cells were selected for 10-14 days using G418 (500 g/ml, InvitRigen). Colonies showing green fluorescence were selected and grown under maintenance doses of G418 (250 g/ml). Receptor expression was confirmed visually, by fluorescent microscopy, as well as by an ELISA assay (see below).

[0277] Cell-Surface ELISA

[0278] Correct trafficking of receptors to the plasma membrane was confirmed by performing a cell-surface ELISA to detect the extracellular FLAG tag. Cells were plated at 100,000 cells/well onto 24-well plates coated with poly-D-lysine. Cultured cells were fixed in 4% paraformaldehyde for 10 minutes at 4 C. Cells were washed in phosphate-buffered saline (PBS), then incubated in 1 g/ml Ml anti-FLAG antibody (Sigma, St. Louis) for 1 hour at room temperature. Cells were washed again in PBS with 1 mM CaCl₂, then incubated for 30 minutes at room temperature in secondary antibody (goat-anti mousehorseradish peroxidase, BioRad, CA), diluted 1:1000. Cells were then washed 3 times in PBS plus CaCl₂. To develop the reaction, 0.25 ml of 2,2-azino-bis(3-ethylbenzthiazoline-6-sulfonic acid) liquid substrate (Sigma, St. Louis) was added to each well. After 15-60 minutes of development time, 200 l from each well were transferred to a 96-well plate and the optical density at 410 nm was read. This assay was also used to quantify agonist-induced internalization. Before fixation, cells were treated with varying doses of spiradoline in media. After 10-120 minute treatment at 37 C., the media was removed and cells were fixed before proceeding with the assay. Degree of internalization was determined by comparing cell-surface receptor expression of agonist-treated cells to the number of receptors detected on the surface of vehicle-treated control cells. Each experiment contained 3-6 replicates per condition. Each experiment was replicated 3 times.

[0279] cAMP Accumulation Assay

[0280] The degree of cAMP inhibition in spiradoline-treated HEK-293 cells expressing RASSL variants was measured using the CatchPoint cAMP ELISA kit (Molecular Devices Corp., Santa Clara, Calif.). Cells were plated at 510⁴/well into 96-well plates coated with poly-D-lysine. The next day, cells were rinsed in Krebs-Ringer Bicarbonate buffer with glucose (KRBG, Sigma, St. Louis, Mo.). Cells were then incubated in pre-stimulation buffer (0.75 mM 3-isobutyl-1-methylxanthine in KRBG buffer) for 10 min at room temperature. cAMP was stimulated by the addition of 50 M forskolin to all cells. At the same time, varying doses of spiradoline in PBS were added to cells to establish a dose-response curve. After a 15-minute drug treatment at 37 C., cells were lysed, and cAMP accumulation was assayed according to the CatchPoint protocol. Inhibition of cAMP by spiradoline was determined by comparison to cells treated with forskolin alone. For desensitization assays, a pretreatment dose (1 nM) of spiradoline diluted in sterile PBS was added to the cells, which were then incubated for 10 minutes at 37 C. Then, cells were rinsed 4 times in KRBG. Cells were then stimulated and assayed for cAMP as described above. AC superactivation was determined by using the same cAMP assay procedure to measure cAMP accumulation in cells that had been treated with either 10 nM spiradoline or vehicle for 17 hours, and then stimulated with 10 M forskolin. AC superactivation data was expressed as a percent increase in forskolin-stimulated cAMP relative to vehicle-pretreated cells expressing the same receptor.

[0281] Confocal Microscopy

[0282] HEK 293 cells stably expressing receptor constructs were plated at a density of 500,000 cells/ml into a poly-D-lysine coated glass Labtek II chamber slide (Fisher Scientific, Pittsburgh, Pa.). The following day, cells were treated with the appropriate drug, and then washed in PBS. The PBS was removed and replaced with 1 ml of cold 4% paraformaldehyde in PBS. Cells were fixed at room temperature for 10 min, washed with PBS, and then mounted in Vectashield (Vector Laboratories, Burlingame, Calif.) under coverslips. For confocal imaging, typical images were taken with a 40-60oil immersion objective lens, subject to 5Kalman filtering. Microscope operator was blind to both cell line and treatment condition.

[0283] Fluorometric Imaging Plate Reader (FLIPR) Assay

[0284] All receptors tested were transiently transfected into CHO cells in conjunction with the chimeric G protein G_q15 (Conklin et al., 1993) in a 5:1 molar ratio by Lipofectamine. These cells were plated to confluence in a 96-well plate the following day, and incubated with the calcium sensitive dye fura-3 for one hour. Assay was performed as previously described (Coward et al., 1998) using a range of dilutions of either spiradoline or dynorphin A 1-13 peptide (Sigma, St. Louis, Mo.). Typically, all conditions were performed in triplicate.

[0285] Results

[0286] Rig, a GFP-Tagged RASSL, Signals Appropriately

[0287] Although it has been demonstrated that an N-terminal GFP tag does not interfere with the function of the human KOR, we wanted to confirm that this tag does not modify the signaling properties of Rig, which is a KOR-based RASSL. Rig was transiently transfected into CHO cells along with a chimeric G_q15 protein (Conklin et al. (1993) Nature 363:274-276) that couples to G₁-coupled receptors but signals through the G_q pathway and therefore stimulates calcium release. Using this transient calcium response as a measure of G₁ activation, we showed by FLIPR assay that Rig responded dose-dependently to the small molecule spiradoline, but not to a range of dynorphin doses that activated the wild-type KOR.

[0288] Rig is Internalized by Agonist Treatment

[0289] Agonist treatment induces the internalization of most GPCRs. The GFP tag on Rig enables us to directly observe this phenomenon using confocal microscopy. In untreated cells, the receptor was visible primarily on the plasma membrane. One hour after spiradoline treatment (10-100 M), the receptor was observed in bright, punctate intracellular vesicles. Dynorphin (100 M), in contrast, did not lead to significant internalization of the receptor. An ELISA assay that detects only cell-surface receptors was used to quantify the extent of receptor internalization following spiradoline treatment. With increasing doses of spiradoline, fewer receptors were detected on the cell surface, culminating in an approximately 70% loss of cell-surface receptors at the maximal dose of 300 M spiradoline. The time course of internalization in response to 1 M spiradoline was relatively rapid, with significant receptor loss apparent within 5 minutes. Maximal receptor loss was detected approximately 20 minutes after agonist treatment began. For the remainder of the hour, the number of cell-surfaces receptors did not change significantly, suggesting that the dynamic balance of receptor internalization and recycling back to the plasma membrane had reached a steady state.

[0290] Limited Internalization of Rig-A in Response to Agonist

[0291] To determine the role of C-terminal phosphorylation sites in receptor regulation, we examined spiradoline-induced internalization of Rig-A (also referred to as Ro3A), a mutated version of Rig in which four C-terminal phosphorylation sites were mutated to alanine (FIG. 15). We treated HEK 293 cells that were stably transfected with Rig-A with 10 M spiradoline, a dose sufficient to cause internalization of most Rig receptors. One hour after spiradoline treatment, most Rig-A receptors appeared to remain in the membrane. This observation was quantified using the cell-surface ELISA assay. Across a wide range of spiradoline doses (0.1 to 300 M), significantly less loss of cell-surface receptors was observed for Rig-A compared to Rig (FIG. 19A). Like Rig, maximal Rig-A internalization in response to 1 M spiradoline was observed after approximately 20 minutes of drug treatment (FIG. 19B). However, unlike Rig, at that time point, less than 10% of the Rig-A receptors were internalized. No further internalization of Rig-A was observed in the remainder of the hour. These results suggest that mutation of C-terminal phsophorylation sites inhibits receptor internalization.

[0292] Internalization of Rig- in Response to Low Doses of Spiradoline

[0293] Since the MOR has been shown to internalize more readily than the KOR, we made Rig-, a chimeric receptor in which the entire intracellular portion of the C-terminus is the corresponding MOR sequence (FIG. 15). Rig- was expected to internalize more than Rig in response to spiradoline. Confocal microscopy indicated nearly complete internalization of Rig- after one hour treatment with 10 M spiradoline. A cell-surface ELISA revealed 25-30% internalization of Rig- at low doses of spiradoline, ranging from 0.01 to 0.1 M. Little internalization or Rig or Rig-A has been observed at those doses. At higher doses of spiradoline, no difference in internalization between Rig and Rig- was observed. Similarly, when cells were treated with 11M spiradoline for differing lengths of time, there was no detectable difference in the time course of internalization of Rig and Rig-.

[0294] Evidence for Constitutive Internalization of Rig-A.

[0295] A variant of Rig-, known as Rig-A, also has MOR sequence at the C-terminus, but five serine and glutamic acid residues at the C-terminus were mutated (FIG. 15). These mutations were predicted to render Rig-A more resistant to internalization than Rig-. However, in several independent stable and transient cell lines (HEK 293 and rat1a), the cell-surface expression of Rig-A was always significantly lower than other receptors. A cell-surface ELISA on one representative group of stably transfected HEK 293 lines showed that Rig-A was expressed at 28% of the level of Rig, and 41% of Rig-. Cell-surface expression of a GPCR can be stabilized by the addition of antagonist, so we examined the effect of the KOR antagonist norBNI on cell-surface expression of Rig-A. Antagonist treatment nearly doubled the amount of RigA detected in the membrane (p<0.005; F_(1,6)27.00). It also increased the cell-surface expression of Rig-A to a lesser, but significant, degree (p<0.05, F_(1,6)11.54). In contrast, it had little effect on the cell-surface expression of Rig. The increase in membrane expression of Rig-A after antagonist treatment was confirmed with confocal microscopy. In basal conditions, little Rig-A was observed in the plasma membrane, although the receptor was readily detected in other areas of the cell. Overnight treatment with 10 M norBNI dramatically shifted receptor localization, so that most of the Rig-A was then observed in the plasma membrane. In contrast, untreated Rig receptors were primarily located in the plasma membrane, and norBNI treatment had little effect on their localization.

[0296] Desensitization of Rig, Rig- and Rig-A RASSLs after Brief Spiradoline Pretreatment

[0297] In addition to regulation by internalization, the signal of a GPCR can be modulated by desensitization: uncoupling from the signaling effectors after continuous agonist stimulation. To explore desensitization directly, we examined each RASSL variant in an assay for the inhibition of cAMP accumulation. Spiradoline activation of these G₁-coupled receptors was expected to inhibit forskolin-induced cAMP accumulation in a dose-dependent manner. Indeed, under basal conditions, activation of Rig, Rig-A, Rig- and Rig-A all inhibited forskolin-induced accumulation of cAMP The data for Rig and Rig-A are shown in FIG. 20A and FIG. 20B. In representative experiments, the EC50 for inhibition of cAMP accumulation was 0.70 nM spiradoline for Rig; 0.33 nM for Rig-A; 0.67 nM for Rig-; and 0.713 nM for Rig-A, indicating little difference between RASSLs in their response to acute spiradoline treatment.

[0298] To evaluate desensitization, the inhibition of cAMP in response to differing doses of spiradoline was measured after 15 minutes of 1 nM spiradoline pretreatment. This pretreatment dose was relatively low, and had caused no detectable receptor internalization in ELISA-based assays. Pretreatment reduced the responsiveness of Rig receptors to spiradoline (FIG. 20A). After a high dose of spiradoline (30-100 nM) the same maximal inhibition of cAMP accumulation was observed, but the dose response curve was shifted approximately one order of magnitude, with an EC50 of 3.57 nM. Pretreatment of Rig-A with the same dose of spiradoline, however, did not significantly affect the response of the cells to subsequent spiradoline treatment (FIG. 20B). The EC50 for spiradoline after 1 nM spiradoline pretreatment of Rig-A was 0.32, compared to 0.33 for untreated cells. Spiradoline pretreatment shifted the EC50 of Rig- to 2.08 nM. Notably, the maximal response of pretreated Rig--expressing cells was less than half that of untreated cells, indicating a decreased efficacy of Rig- signaling through the G_i pathway. Spiradoline pretreatment strongly reduced the response of Rig-A to further spiradoline treatment, shifting the EC50 nearly 100-fold to 28.87 nM. This indicates that Rig-A receptors are relatively resistant to desensitization, while Rig- and Rig-A receptors desensitize dramatically.

[0299] Superactivation of Adenylyl Cyclase After Prolonged Spiradoline Pretreatment

[0300] Chronic treatment of cells expressing G,-coupled receptors with agonist results in a compensatory increase in the activity of AC, and therefore an increased accumulation of cAMP in response to the same dose of forskolin. We examined the development of this AC superactivation in cell lines transiently expressing the RASSL variants. Rig showed a two-fold increase in the induction of cAMP levels in response to 10 M forskolin following 18 hours of pretreatment with 10 nM spiradoline (p<0.005, F_(1,10)17.72). This degree of superactivation was similar to that seen in cells transfected with the wild-type KOR, indicating the same cellular response to prolonged G₁ signaling through both Rig and KOR. Overnight spiradoline treatment of Rig-A-expressing cells, followed by 10 M forskolin stimulation, resulted in a slightly smaller yet significant increase in cAMP (p<0.05, F_(1,10)9.52). Notably, cells expressing Rig- and Rig-A receptors showed no evidence of AC superactivation after 18 hours of spiradoline pretreatment. These data suggest that receptors that desensitize more readily at the receptor level, like Rig- and Rig-A, do not induce compensations in an opposing signaling pathway.

[0301] While the present invention has been described with reference to the specific embodiments thereof, it should be understood by those skilled in the art that various changes may be made and equivalents may be substituted without departing from the true spirit and scope of the invention. In addition, many modifications may be made to adapt a particular situation, material, composition of matter, process, process step or steps, to the objective, spirit and scope of the present invention. All such modifications are intended to be within the scope of the claims appended hereto.