WO2001038874A1 - Expression cloning method - Google Patents

Abstract

The invention relates to a method for isolating a nucleic acid (e.g., genomic DNA, cDNA) encoding a ligand for a G protein-coupled receptor (GPCR). The method comprises providing one or more primary pools of prokaryotic cells into which an expression library comprising exogenous nucleic acids has been inserted, said prokaryotic cells having been cultured under conditions suitable to produce individual colonies and a predetermined number of said colonies having been combined into one or more primary pools of prokaryotic cells; and a) expressing the expression library from a pool to produce a corresponding pool of proteins; b) assaying a pool of proteins for ligand, and selecting a pool of proteins that comprises ligand; c) providing prokaryotic cells into which the expression library encoding the pool of proteins selected in b) has been inserted, and growing the cells under conditions suitable to produce individual colonies; d) combining a predetermined number of colonies from c) into a secondary pool of cells, wherein the secondary pool comprises fewer colonies that said primary pool of prokaryotic cells; e) repeating a) through d) until a pool of cells containing an exogenous nucleic acid encoding ligand is isolated, said pool of cells comprising cells obtained from a single colony; and f) recovering the nucleic acid encoding said ligand.

Description

EXPRESSION CLONING METHOD

BACKGROUND OF THE INVENTION

The seven transmembrane domain or G protein-coupled receptors (GPCR) constitute a large superfamily of proteins which transduce signals in response to ligand binding. The GPCR superfamily includes receptors which transduce signals upon binding of, for example, peptide hormones, neurotransmitters and cytokines (e.g., chemokines). GPCRs are coupled to intracellular enzymes, channels and transporters through heterotrimeric G proteins (e.g., Gi, Gs, Gq). Upon ligand binding to receptor, these intracellular enzymes, channels and transporters can be activated or inhibited leading to cellular responses to ligand binding.

Members of the GPCR superfamily (and of subfamilies, such as chemokine receptors) can be identified by analysis of the amino acid sequence of proteins (Probst, W.C., et al, DNA Cell Bioi, 77:1-20 (1992); Liao, F., et al, J. Exp. Med., 185:2015-2023 (1997)). Many proteins which are considered to be GPCRs have been identified in this manner. Generally, GPCRs identified by sequence analyses are referred to as "orphan ^' receptors because the biological functions and ligands of the receptor are unknown.

One method of identifying ligands for a GPCR (e.g., an "orphan" GPCR) is to assay isolated proteins, which are predicted to be GPCR ligands based upon amino acid sequence homology with known ligands, for receptor binding. This approach severely limits the probability of identifying a ligand and has proven unsuccessful in identifying ligand for an "orphan" chemokine receptor ( Liao, F., et al., J. Exp. Med., 755:2015-2023 (1997); Loetscher, M. et al, Current Biology, 7:652-660 (1997)). There is a need for new methods of identifying ligands for GPCRs. SUMMARY OF THE INVENTION

The invention relates to a method for isolating a nucleic acid (e.g., genomic DNA, cDNA) encoding a ligand for a G protein-coupled receptor (GPCR) (e.g., a mammalian (e.g., human) GPCR). The method comprises providing one or more primary pools of prokaryotic cells into which an expression library comprising exogenous nucleic acids has been inserted, said prokaryotic cells having been cultured under conditions suitable to produce individual colonies and a predetermined number of said colonies having been combined into one or more primary pools of prokaryotic cells; and a) expressing the expression library from a pool of cells to produce a corresponding pool of proteins; b) assaying a pool of proteins for ligand, and selecting a pool of proteins that comprises ligand; c) providing prokaryotic cells into which the expression library encoding the pool of proteins selected in b) has been inserted, and growing the cells under conditions suitable to produce individual colonies; d) combining a predetermined number of colonies from c) into a secondary pool of cells, wherein the secondary pool comprises fewer colonies than said primary pool of prokaryotic cells; e) repeating a) through d) until a pool of cells containing an exogenous nucleic acid encoding ligand is isolated, said pool of cells comprising cells obtained from a single colony; and f) recovering the nucleic acid encoding said ligand. The process of the invention can be used to isolated a nucleic acid encoding a ligand for a G i-coupled GPCR, a chemokine receptor or other GPCR. In particular embodiments, the nucleic acid is genomic DNA or cDNA. In one embodiment, the expression library (expression plasmids) is expressed in a prokaryotic cell. In another embodiment, the expression library is expressed in a eukaryotic cell, such as a mammalian cell, a yeast cell or an insect cell. In another embodiment the expression library is expressed in vitro. The pools of expressed proteins can be assayed for ligand in a direct or indirect (e.g., competitive) receptor binding assay or in a functional assay using cells that express GPCR. In one embodiment, pools of proteins are assayed for ligand in a chemotaxis assay using recombinant cells which express GPCR.

BRIEF DESCRIPTION OF THE DRAWINGS

Figure 1 is a histogram showing the number of transfected LI.2 cells which express Bonzo (Bonzo/L1.2) that migrated toward pools of expressed proteins in chemotaxis assays. An expression library prepared using human spleen cDNAs was plated and divided into 96 pools, each of which contained about 800 individual colonies. The expression libraries from each pool of bacteria were isolated and transiently transfected into 293T cells. Conditioned supernatants were collected from the transfected 293T cultures and assayed for chemotactic activity. Pool D3 contained chemotactic activity that was significantly above background levels. Background was the number of cells that migrated in chemotaxis assays that did not contain conditioned media.

Figure 2 illustrates the nucleic acid sequence of a cDNA encoding human SExCkine (Spleen Extracted Chemokine) isolated from pool D3 (SEQ ID NO: l), and the predicted amino acid sequence of the encoded human SExCkine polypeptide (SEQ ID NO:2). The cloned cDNA consists of 1763 nucleotides with an open reading frame encoding 254 amino acids. The open reading frame includes a predicted signal peptide of 29 amino acids (amino acid residues 1-29 of SEQ ID NO: 2, underlined), a predicted membrane proximal mucin domain (amino acid residues 1 18-201 of SEQ ID NO: 2, boxed), a predicted transmembrane segment (amino acid residues 202-226 of SEQ ID NO: 2, underlined) and a cytoplasmic tail (amino acid residues 227-254 of SEQ ID NO: 2).

Figure 3 is a histogram showing SExCkine-induced chemotaxis of Bonzo/Ll .2 cells. Bonzo/Ll .2 cells were assayed for chemotactic response to undiluted culture supernatant of 293T cells transiently transfected with a cDNA encoding SExCkine (SEQ ID NO: 1 ) (Straight) or to various dilutions of the supernatant ( 1 :2, 1 :4, 1 :8 and 1 : 16). Bkg: background chemotaxis in the presence of assay media without chemokine.

Figure 4 is a graph showing dose dependent inhibition of SExCkine-induced chemotaxis of Bonzo/Ll.2 cells by anti-Bonzo monoclonal antibodies (mAb 7F3 or mAb 4A1 1). Bonzo/Ll .2 cells were incubated with concentrated supernatant from murine hybridoma 7F3 which produces anti-Bonzo mAb 7F3, from murine hybridoma 4A1 1 which produces anti-Bonzo mAb 4A1 1, or from a murine hybridoma which produces an isotype control antibody (IgG2a or IgG2b), prior to exposure to SExCkine.

Figure 5 illustrates the predicted structure of the transmembrane form of human SExCkine.

DETAILED DESCRIPTION OF THE INVENTION

A description of preferred embodiments of the invention follows.

The invention relates to a method of isolating a nucleic acid (e.g., genomic DNA, cDNA) encoding a ligand for a G protein-coupled receptor (e.g., a mammalian (e.g., human) GPCR), such as a chemokine receptor. In one embodiment, the method comprises providing one or more primary pools of prokaryotic cells into which an expression library comprising exogenous nucleic acids has been inserted, said prokaryotic cells having been cultured under conditions suitable to produce individual colonies and a predetermined number of said colonies having been combined into one or more primary pools of prokaryotic cells; and a) expressing the expression library from a pool of cells to produce a corresponding pool of proteins; b) assaying a pool of proteins for ligand, and selecting a pool of proteins that comprises ligand; c) providing prokaryotic cells into which the expression library encoding the pool of proteins selected in b) has been inserted, and growing the cells under conditions suitable to produce individual colonies; d) combining a predetermined number of colonies from c) into a secondary pool of cells, wherein the secondary pool comprises fewer colomes than said primary pool of prokaryotic cells, e) repeating a) through d) until a pool of cells containing an exogenous nucleic acid encoding ligand is isolated, said pool of cells compnsmg cells obtained from a single colony, and f) recovering the nucleic acid encoding said ligand

The method of the invention provides several advantages For example, previously unknown proteins can be assayed for binding to receptor and large pools of individual clones (e g , pools of about 800 clones) can be successfully screened Thus, the amount of time and expense of isolating a nucleic acid encoding ligand can be reduced using the method

Nucleic acids referred to herein as "isolated" aie nucleic acids separated away from the nucleic acids of the genomic DNA or cellular RNA of their source of origin (e g , as it exists in cells or m a mixture of nucleic acids such as a library), and may have undergone further processing "Isolated" nucleic acids include nucleic acids obtained by methods descnbed herein, similar methods or other suitable methods, including essentially pure nucleic acids, nucleic acids produced by chemical synthesis, by combinations of biological and chemical methods, and recombinant nucleic acids which are isolated

The exogenous nucleic acids can be any collection of exogenous nucleic acids (e g , genomic DNA fragments, cDNAs) that is likely to encode a ligand for the receptor The exogenous nucleic acids can be prepared using any suitable method For example, the exogenous nucleic acids can be an exogenous genomic DNA library prepared by a partial digestion of human genomic DNA with Sau3A I, Mbo I or other suitable restnction enzyme(s) In one embodiment, the exogenous nucleic acids can be a cD A. library Methods suitable for preparing a cDNA library are known in the art For example, RNA or preferably polyA^" RNA can be isolated from a suitable source (e g , cells, oigans, tissues) by guanidimum isothiocyanate extraction and ohgo dT chiomatogtaphy or commercially available kits such as RNAgents® Total RNA Isolation System and PolyATtract® mRNA Isolation System (both available from Promega, Madison, WI) can be used cDNAs corresponding to the poly A.^* RNA can be prepared using a reverse transcnptase for first strand synthesis and a suitable DNA polymerase (E coli DNA polymerase I) for second strand synthesis. Commercially available kits for making cDNA, such as Superscnpt system (Gibco/BRL, Rockville, MD) can be used The cDNAs can be ligated into an expression vector or ligated to DNA linkers (also referred to as adapters) which contain suitable restnction sites to facilitate cloning into a desired expression vector Linkers (adapters) which contain a variety of restnction sites are available from commercial sources (e g., Pharmacia Biotech, Piscataway, NJ, New England Biolabs, Beverly, MA). If desired, the cDNAs can be further processed before ligation into an expression vector For example, the cDNAs can be size fractionated (e g , by centnfugation tlirough a sucrose gradient, by electrophoresis through agarose gel) to enπch for full length cDNAs Commercially available cDNA libraries can also be used.

The exogenous nucleic acids (e.g, cDNA library, genomic DNA library) can be ligated into a suitable expression vector to produce an expression library The library can be ligated into the vector such that nucleic acids are inserted into the vector m a preferred onentation Expression vectors suitable for use in the invention contain sequences which direct the replication of the vector in a prokaryotic cell (e g , an ongm of replication such as On p, colEl On and the like) and sequences which direct expression (transcnption and translation) of the insert nucleic acid m a suitable expression system (e g , in vitro expression, expression m prokaryotic or eukaryotic cells) Preferably, the expression vector also contains a selectable marker for selection of prokaryotic cells carrying the vector Many suitable selectable markers are known, for example, genes which confer resistance to antibiotics such as the β-lactamase gene for ampicillm resistance, the Tet gene for tetracychne resistance the supF suppressor tRNA for selection m E coli carrying the P3 episome (e g , E coh strain MC1061/P3)

The selection of a suitable expression vector will be dependent on the desired method of expression (e g , in vitro, in vivo) For example, for in vitro expression, vectors which contain a promoter for Sp6 or T7 RNA polymerase, such as pSP64 or pGEMEX (Piomega, Madison, WI) can be used Where expression in a mammalian cell is desired, an expression vector which contains a promoter suitable to dnve expression of the inserted nucleic acid (e g , simian virus 40 early or late promoter, Rous sarcoma vmis long terminal repeat promoter, cytomegalovirus promoter, adenovirus late promoter) in a mammalian cell is selected Suitable expression vectors for expression m mammalian cells include, for example, pCDM8, pCDNAl 1/amp, pcDNA3 1, pRc/RSV, pEF-1 (Invitrogen, Carlsbad, CA), pCMV- Script®, pFB, pSG5, pXTl (Stratagene, La Jolla, CA), pCDEF3 (Goldman, L A , et al , Biotechmques, 21 1013-1015 (1996)), pSVsport (Gibco/BRL, Rockville, MD), pEF-Bos (Mizushima, S , et al , Nucleic Acids Res , 18 5322 (1990)) and the like Expression vectois which are suitable for use m various expression hosts, such as prokaryotic cells (E coli), insect cells (Drosophila Schmeder S2 cells, Sf9) and yeast (P methanohca, P pastoris, S cerevisiae) are also available

The expression library can be inserted or introduced into suitable prokaryotic host cells using any suitable method, such as transfomiation A preferred prokaryotic host is E coli Where the expression library is inserted into host cells by transformation, the host cells can be treated to render them competent for transformation Competent cells can be prepared using suitable methods, for example, by exposure to CaCl₂ as described in Ausubel, F M et al , Eds , Current Pi otocols in Molecular Biology, (John Wiley & Sons New York, NY), Supplement 47, Chapter 1 (1999) Competent cells, such as ElectroMAX DH10B (Gibco/BRL, Rockville, MD) are also available from commercial sources

Prokaryotic host cells into which an expression library has been inserted can be cultured under conditions suitable to produce individual colonies (clones) For example, a dilute suspension of transfonned E coli can be spread on culture plates containing nutrient media (e g , Luna Broth agar) and ampicilhn and cultured for about 12 to about 18 houis at 37°C The prokaryotic cells can be cultured under conditions wheie a desired number of colonies develop on each plate For example, the cells can be cultured under conditions where at least about 100, or about 100 to about 1000. or about 600 to about 900, or about 800 individual colonies form on each plate One or more pools of individual colonies can be collected. Pools of colonies can be collected using any suitable procedure. In one embodiment, all of the colonies from a plate form a pool. In this situation, the colonies can be collected by adding a quantity of media to the plate which is sufficient to wet the surface (e.g., about 2 mL) and scraping the colonies off of the plate, thereby forming a bacterial suspension which can be recovered. If desired, a pool can consist of the colonies from a fraction of a plate or the colonies from two or more plates, and suitable collection procedures can be employed.

Where the expression library is for prokaryotic expression, the library can be expressed by culturing the pool of prokaryotic cells under conditions suitable for expression. For example, a suitable inducer of expression (e.g., Isopropyl β-D- Thiogalactopyranoside (EPTG)) can be added to the culture to induce expression. The expression library can also be recovered from the pool of prokaryotic cells prior to expression. The expression library (expression plasmids) contained within the pool of cells can be recovered from the pool immediately following collection, or the pool can be cultured to provide amplification of the number of expression plasmids that can be recovered. The expression library can be recovered using any suitable method for isolating nucleic acids. For example, by using methods for recovering plasmids, such as the alkaline lysis method or using commercially available kits. Xn one example, the expression library can be recovered using QIAprep spin columns (QIAGEN, Valencia, CA).

A portion of the expression library can be reserved (e.g., an aliquot of recovered expression library, a culture of bacteria transformed with the expression library), and the remainder can be expressed (transcribed and translated) to produce a corresponding pool of proteins. Any suitable method of expression can be used, such as in vitro transcription and translation, expression in eukaryotic cells (e.g., mammalian cells, insect cells, yeast cells) or expression in prokaryotic cells (e.g., E. coli). For in vitro expression, the pool of expression constructs can be transcribed using a suitable RNA polymerase (e.g., Sp6 RNA polymerase, T7 RNA polymerase), and the transcribed RNA is translated. In vitro expression can be accomphshed using any suitable method, for example by translation of mRNA using extracts of rabbit reticulocytes or wheat germ extracts. Kits for performing coupled transcription and translation (e.g., TnT® T7 Quick Coupled Transcnption Translation System (Promega, Madison, WI)) can be used. Expression m eukaryotic or prokaryotic cells can be accomplished by inserting or introducing the expression library recovered from a pool into suitable host cells and cultuπng the resulting cells under conditions suitable for expression of the library. The expression library can be inserted using any suitable method, such as by transformation into a prokaryotic host. Suitable prokaryotic hosts for expression of proteins include, for example, E coli and B subtilis. The expression library recovered from a pool can be inserted into eukaryotic cells (e.g., yeast, insect cells, mammalian cells) by transformation, transfection, infection or other suitable methods h a preferred embodiment, the expression library is expressed in a mammalian cell Mammalian cells suitable for expression of the expression library include, COS-1 (ATCC Accession No. CRL-1650), COS-7 (ATCC Accession No. CRL-1651), CHO (e.g., ATCC Accession No. CRL-9096) , 293 (ATCC Accession No CRL-1573), HeLa (ATCC Accession No. CCL-2), CV1 (ATCC Accession No. CCL-70), WOP (Dailey, L., et al , J Virol , 54 739-749 (1985), 3T3, 293T (Pear, W. S , et al , Proc Natl Acad Sci USA , 90-8392-8396 (1993)), and the like. The expression library can be inserted into the mammalian cell by transfection, for example by the calcium phosphate method, diethylamino ethyl (DEAE) dextran method, electroporation or using hposomes (e.g , LφofectAMINE™, Gibco/BRL). The transfected mammalian cells can be cultured under conditions suitable for expression of the expression constructs The expressed pool of proteins can be recovered using any suitable methods.

In one embodiment, transfected mammalian cells can be cultured in media supplemented with growth factors and/or a high concentration of serum (e.g., about 20% or more) for a period of about 12 to about 24 hours The media can then be replaced with seaim-free media or with media that is supplemented with a lower concentration of serum (e g , about 10% or less) and the cells can be cultured for a period of time sufficient for expression (e g , about 24 to about 72 hours) The conditioned supernatants of the transfected cells can be recovered and assayed for ligand.

Ligand can be detected by assaying a pool of proteins for ligand (e.g., for binding to GPCR). For example, a composition comprising a GPCR can be used in a binding assay to detect and/or identify ligands that can bind to receptor.

Compositions suitable for use in a binding assay include, for example, cells which naturally express GPCR, cell lines which express a GPCR and recombinant cells comprising a recombinant nucleic acid sequence which encodes GPCR or functional variant thereof. Compositions suitable for use in a binding assay also include, membrane preparations which comprise a GPCR or functional variant thereof. Such membrane preparations can contain natural (e.g., plasma membrane) or synthetic membranes. Preferably, the membrane preparation is a membrane fraction of a cell that expresses a GPCR or a functional variant thereof.

Direct or indirect (e.g., competitive) binding assays can be used. In one embodiment, the method of assaying for ligand is a competitive binding assay in which the ability of a pool of proteins to inhibit the binding of a reference agent (e.g., a known ligand for the receptor) is assessed. For example, the reference agent can be labeled with a suitable label as described herein, and the amount of labeled reference agent required to saturate the GPCR or functional variant thereof in the assay can be determined. A saturating amount of labeled reference agent and a pool of expressed protein can be contacted with a composition comprising the GPCR or functional variant thereof under conditions suitable for binding, and complex formation determined. In this type of assay, a decrease in the amount of complex formed between the GPCR or functional variant thereof and labeled reference agent indicates that the protein pool contains a ligand for the receptor.

The formation of a complex can be detected or measured directly or indirectly using any suitable method. For example, the reference agent can be labeled with a suitable label and the fonnation of a complex can be determined by detection of the label. The specificity of the complex can be determined using a suitable control such as excess unlabeled reference agent or label alone. Labels suitable for use in detection of a complex between a ligand and a GPCR or functional vanant thereof include, for example, a radioisotope, an epitope label, an affinity label (e g., biotin, avidm), a spin label, an enzyme, a fluorescent group or a chemilummescent group When labels are not employed, complex formation can be determined by surface plasmon resonance or other suitable methods. The capacity of the protein pool to inhibit the formation of a complex between the reference agent and the GPCR or functional variant thereof can be reported as percent inhibition.

Functional assays can be used to detect ligand in a pool of proteins. Functional assays can be used to detect ligands at low concentrations (e.g., nanomolar or sub-nanomolar) For example, receptor-hgand binding events can lead to the production of easily detectable quantities of second messengers (e.g., free Ca^2~) and cellular changes (e.g., changes m physiology, morphology). Thus, functional assays are generally preferred A pool of proteins can be studied in one or more suitable functional assays to determine if said pool contains ligand. For example, a protein pool can be tested m an extracellular acidification assay, calcium flux assay, ligand binding assay, chemotaxis assay or assay which monitors degranulation or inflammatory mediator release (see, for example, Hesselgesser et al, J Biol Chem. 273(25).15687-15692 (1998) and WO 98/02151).

For example, a protein pool can be tested in a chemotaxis assay using suitable cells Suitable cells include, for example, cell lines, recombinant cells or isolated cells v, hich expresses a GPCR or functional variant thereof and undergo hgand-mduced chemotaxis. Generally, GPCRs which couple to Gαi can transduce chemotaxis signals m response to ligand binding (Neptune, E. E., et al , Proc Natl Acad Sci USA , 94-14489-14494 (1997)). Thus, recombinant cells which express a Gαi-coupled GPCR that does not mediate chemotaxis in nature (e g., D-, dopamine receptor, μ opioid receptor, δ opioid receptor) can be used in a chemotaxis assay to isolate nucleic acids encoding ligands for the GPCR

Cells expressing recombinant GPCR or functional variant thereof can be prepared by inserting or introducing a nucleic acid encoding a GPCR or functional vanant thereof into a suitable cell Preferably, the recombinant cell is prepared by transfectmg a suitable mammalian cell with a construct encoding a GPCR or functional \ aπant thereof Suitable mammalian cells include cells which can couple a GPCR to G i and cells which acquire chemotactic ability upon transfection of a GPCR. For example, HEK293 cells, 300-19 cells (Legler, D.F., et al, J. Exp. Med, 757:655-660 (1998)) or other suitable cells can be used. Preferably the recombinant cell is prepared by inserting a nucleic acid (e.g., cDNA) encoding a GPCR or functional variant thereof into the murine preB cell lymphoma LI .2 (Campbell, et al. J Cell Biol, 73^:255-266 (1996)).

As used herein "GPCR" refers to a naturally occurring or endogenous GPCR protein (e.g., a mammalian GPCR, human GPCR) or a protein having an amino acid sequence which is the same as that of a naturally occurring or endogenous corresponding GPCR protein (e.g., recombinant proteins, synthetic proteins (i.e., produced using the methods of synthetic organic chemistry)). Accordingly, as defined herein, the term includes mature receptor protein, polymorphic or allelic variants, and other isoforms of a GPCR (e.g., produced by alternative splicing or other cellular processes), and modified or unmodified forms of the foregoing (e.g., lipidated, glycosylated, unglycosylated). Naturally occurring or endogenous GPCR proteins include wild type proteins such as a mature human GPCRs, polymorphic or allelic variants and other isoforms which occur naturally (e.g., in mammals (e.g., in humans, in non-human primates)). Such proteins can be recovered or isolated from a source which naturally produces GPCR, for example. GPCR proteins and proteins having the same amino acid sequence as a naturally occurring or endogenous corresponding GPCR (e.g., mammalian GPCR) are referred to by the name of the source which naturally produces the receptor. For example, where the source is a human, the protein is designated as a human GPCR protein (e.g., a recombinant human GPCR produced in a suitable host cell). "Functional variants" of GPCRs include functional fragments, functional mutant proteins, and/or functional fusion proteins which can be produced using suitable methods (e.g., mutagenesis (e.g., chemical mutagenesis, radiation mutagenesis), recombinant DNA techniques). A "functional variant" is a protein or polypeptide which has at least one function characteristic of a GPCR, such as a binding activity, a signaling activity (e.g., GDP/GTP exchange by GPCR-associated G proteins, transient increase in the concentration of cytosolic free calcium [Ca²⁺]₍) and or ability to stimulate a cellular response (e.g., ligand induced proliferation, migration, chemotaxis, secretion, degranulation, inflammatory mediator release (such as release of bioactive lipids such as leukotrienes (e.g., leukotriene C₄)), respiratory burst). Preferred functional variants can bind ligand and mediate chemotaxis.

In one embodiment, a functional variant of a GPCR (e.g., a ligand binding variant) shares at least about 80% amino acid sequence similarity with a naturally occurring GPCR, preferably at least about 90% amino acid sequence similarity, and more preferably at least about 95%> amino acid sequence similarity with a naturally occurring GPCR. In another embodiment, a functional fusion protein comprises a first moiety which shares at least about 85% sequence similarity with a naturally occurring GPCR, preferably at least about 90% sequence similarity, and more preferably at least about 95% sequence similarity with a naturally occurring GPCR (e.g., a human GPCR). In another embodiment, a functional mammalian GPCR protein or functional variant of a mammalian GPCR protein shares at least about 80%) amino acid sequence similarity, preferably at least about 90% amino acid sequence similarity, and more preferably at least about 95% amino acid sequence similarity with a naturally occurring human GPCR. Amino acid sequence similarity can be determined using a suitable sequence alignment algorithm, such as the Lasergene system (DNASTAR, Inc., Madison, WI), using the Clustal method with the PAM 250 residue weight table, a gap penalty of 10, a gap length penalty of 10 and default parameters (pairwise alignment parameters: ktuple = 1, gap penalty = 3, window = 4 and diagonals saved = 5). In another embodiment, a functional variant is encoded by a nucleic acid sequence which is different from the naturally-occurring nucleic acid sequence, but which, due to the degeneracy of the genetic code, encodes a GPCR or a portion thereof.

Generally, fragments or portions of GPCRs include those having a deletion (i.e., one or more deletions) of an amino acid (i.e., one or more amino acids) relative to the mature GPCR (such as N-terminal, C-tenninal or internal deletions). Fragments or portions in which only contiguous amino acids have been deleted or in which non-contiguous amino acids have been deleted relative to mature GPCR are also envisioned.

Mutant GPCRs include natural or artificial variants of a naturally occurring GPCR differing by the addition, deletion and/or substitution of one or more contiguous or non-contiguous amino acid residues (e.g., receptor chimeras). Such mutations can occur at one or more sites on a protein, for example a conserved region or nonconserved region (compared to other G protein-coupled receptors (e.g., chemokine receptors)), extracellular region, cytoplasmic region, or transmembrane region. Fusion proteins encompass polypeptides comprising a GPCR or a variant thereof as a first moiety, linked via a covalent bond (e.g., a peptide bond) to a second moiety not occurring in the GPCR as found in nature. Thus, the second moiety can be an amino acid, oligopeptide or polypeptide. The second moiety can be linked to the first moiety at a suitable position, for example, the N-terminus, the C-terminus or internally. In one embodiment, the fusion protein comprises an affinity ligand (e.g., an enzyme, an antigen, epitope tag (e.g., hemagglutinin (HA), FLAG), a binding domain) as the first moiety, and a second moiety comprising a linker sequence and GPCR or a portion thereof. Additional (e.g., third, fourth) moieties can be present as appropriate. In one embodiment, the cell used in the functional assay is a recombinant mammalian cell which expresses a GPCR comprising an amino terminal epitope tag (e.g., hemagglutinin, FLAG).

In one example, recombinant LI .2 cells which express a HA-tagged G protein-coupled receptor, can be used in a modification of a transendothelial migration assay (Carr, M.W., et al. T.A., Proc. Natl Acad Sci, USA, (P7):3652 (1994)). The endothelial cells used in this assay are preferably the endothelial cell line, ECV 304, which can be obtained from the American Type Culture Collection (Manassas, VA). Endothelial cells can be cultured on 6.5 mm diameter Transwell culture inserts (Costar Corp., Cambridge, MA) with 3.0 μm pore size. Culture media for the ECV 304 cells can consist of M199+10% FCS, L-glutamine, and antibiotics. The assay media can consist of equal parts RPMI 1640 and Ml 99 with 0.5% BSA. Two hours before the assay, 2x10 ECV 304 cells can be plated onto each insert of the 24 well Transwell chemotaxis plate and incubated at 37°C. Pools of proteins to be assayed can be added to the 24- well tissue culture plates in a final volume of 600 μL. Endothelial-coated Transwells can be inserted into each well and 10 recombinant LI .2 cells expressing GPCR can be added to the top chamber in a final volume of 100 μL of assay medium. The plate can then be incubated at 37°C in 5% CO₂/95% air for 1-24 hours. The cells that migrate to the bottom chamber during incubation can be counted, for example using flow cytometry. To count cells by flow cytometry, 500 μL of the cell suspension from the lower chamber can be placed in a tube and relative counts can obtained for a set period of time, for example, 30 seconds. This counting method is highly reproducible and allows gating on the leukocytes and the exclusion of debris or other cell types from the analysis. Alternatively, cells can be counted with a microscope.

A ligand can also be detected by monitoring cellular responses induced by active receptor, using suitable cells which express a GPCR or a functional variant thereof. For instance, exocytosis (e.g., degranulation of cells leading to release of one or more enzymes or other granule components, such as esterases (e.g., serine esterases), perform, and/or granzymes), inflammatory mediator release (such as release of bioactive lipids such as leukotrienes (e.g., leukotriene C₄)), and respiratory burst, can be monitored by methods known in the art or other suitable methods (see e.g., Taub, D.D. et al, J. Immunol, 155: 3877-3888 (1995), regarding assays for release of granule-derived serine esterases; Loetscher et al, J. Immunol, 156: 322- 327 (1996), regarding assays for enzyme and granzyme release; Rot, A. et al, J. Exp. Med., 176: 1489-1495 (1992) regarding respiratory burst; Bischoff, S.C. et al, Em: J. Immunol, 23: 761-767 (1993) and Baggliolini, M. and CA. Dahinden, Immunology Today, 15: 127-133 (1994)). A variety of functional assays which employ recombinant cells which express a GPCR or functional variant thereof can be employed. For example, assays in which expression of an endogenous or exogenous reporter gene (e.g., β-galactosidase, green fluorescent protein) is induced upon ligand binding to a GPCR expressed by recombinant cells (e.g., recombinant bacteria, recombinant yeast, recombinant mammalian cells) can be used. In one embodiment, the ligand can be detected by monitoring the release of an enzyme upon degranulation or exocytosis by a cell capable of this function. Cells expressing a GPCR or a functional variant thereof can be maintained in a suitable medium under suitable conditions, and degranulation can be induced. The cells are contacted with a pool of proteins to be tested, and enzyme release can be assessed. The release of an enzyme into the medium can be detected or measured using a suitable assay, such as an immunological assay, or biochemical assay for enzyme activity.

The medium can be assayed directly, by introducing components of the assay (e.g., substrate, co-factors, antibody) into the medium (e.g., before, simultaneous with or after the cells and protein pool are combined). The assay can also be performed on medium which has been separated from the cells or further processed (e.g., fractionated) prior to assay. For example, convenient assays are available for enzymes, such as serine esterases (see e.g., Taub, D.D. et al, J. Immunol, 155: 3877-3888 (1995) regarding release of granule-derived serine esterases).

In another embodiment, cells expressing a GPCR or a functional variant thereof are combined with a pool of proteins to be tested and Ca²⁺ flux is assessed. Stimulation of Ca²" flux is indicative that the protein pool contains a ligand for the GPCR. Engagement of a GPCR (e.g., chemokine receptor) on a lymphocyte can cause integrin activation, and induction of adherence to adhesion molecules expressed in vasculature or the perivascular space. Cellular adherence can be monitored by methods known in the art or other suitable methods. In one embodiment, a ligand is identified by monitoring cellular adherence by a cell capable of adhesion. For example, a pool of proteins to be tested can be combined with (a) cells expressing a GPCR or a functional variant thereof (preferably non- adherent cells which when transfected with receptor acquire adhesive ability), (b) a composition comprising a suitable adhesion molecule (e.g., a substrate such as a culture well coated with an adhesion molecule, such as fibronectin), and (c) a pool of proteins to be tested, and maintained under conditions suitable for ligand-induced adhesion. Labeling of cells with a fluorescent dye provides a convenient means of detecting adherent cells. Nonadherent cells can be removed (e.g., by washing) and the number of adherent cells determined. Increased adhesion relative to a suitable control is indicative of the presence of a ligand for the GPCR.

The in vitro methods of the present invention can be adapted for high- throughput screening in which large numbers of samples are processed (e.g., a 96- well format). In one embodiment, recombinant cells expression chimeric GPCRs can be used and ligand-induced intracellular calcium mobilization can be monitored using a fluorometric imaging plate reader (FLEPR)(see, for example, Coward, P., et al, Anal. Biochem., 270:242-248 (1999)). Cells expressing a GPCR (e.g.,a human chemokine receptor) or a functional variant thereof at levels suitable for high- throughput screening can be used, and thus, are particularly valuable in the identification and/or isolation of GPCR ligands. Expression of GPCR can be monitored in a variety of ways. For instance, expression can be monitored using antibodies which bind receptor or a portion thereof. Also, commercially available antibodies can be used to detect expression of an antigen- or epitope-labeled fusion protein comprising a receptor protein or polypeptide (e.g., FLAG tagged receptors, HA tagged receptors), and cells expressing the GPCR at the desired level can be selected (e.g., by flow cytometry).

A pool of proteins that contains ligand (positive pool) can be selected. The reserved expression library (expression plasmids) encoding the selected pool of proteins containing ligand can be inserted into suitable prokaryotic host cells and the resulting cells can be cultured to produce individual colonies as described. The individual colonies can be collected into one or more secondary pools, which contain fewer colonies than fonned the original pool. For example, if the original pool consisted of about 800 colonies, then about 500 or 400 or 200 or 50 colonies can be collected for each secondary pool. The expression library from the secondary pools can be expressed and assayed as described. The steps of the method can be repeated, with fewer individual colonies pooled at each successive repetition, until a single colony containing an exogenous nucleic acid encoding ligand is isolated. One or more assays (e.g., chemotaxis) can be used to detect ligand in pools of proteins. For example, the primary pools can be assayed using a chemotaxis assay, and a subsequent pool (e.g., secondary pool, tertiary pool) can be assayed using a direct or indirect (e.g., competitive) receptor binding assay.

In one embodiment, the method is a method for isolating a nucleic acid encoding a ligand for a chemokine receptor (e.g., C chemokine receptor, CC chemokine receptor, CXC chemokine receptor, CX3C chemokine receptor, "orphan" chemokine receptor).

In another embodiment, the method is a method for isolating a nucleic acid encoding a ligand for a Gαi-coupled GPCR.

As described in the Example, the method of the invention was employed to isolate a cDNA encoding a ligand for the orphan chemokine receptor known as Bonzo (Deng, H. K., et al, Nature, 355:296-300 (1997)). Bonzo is also referred to as STRL33 (Liao, F. et al, J. Exp. Med., 185:20X5-2023 (1997), TYMSTR (Loetscher, M. et al, Current Biology, 7:652-660) and HBMBU14 (Elshourbagy et al, EP 0 834 563 A2 and U.S. Patent No. 5,824,504, the entire teachings of which are incorporated herein by reference)). The sequences of nucleic acids encoding Bonzo have been deposited in Genbank under accession numbers AF007545, NM_006564, U73531 and U73529. The entire teachings of each of these Genbank entries is incorporated herein by reference.

Prior to employing the expression cloning method of the invention, it is desirable to analyze the expression and structure of the GPCR for which a ligand is sought. The information obtained from such an analysis can be valuable in tailoring the method of the invention for a specific application (e.g., cloning ligand for specific GPCR). Receptor and ligand are frequently expressed in the same biological locations (e.g., organs, tissues). Thus, knowing where (e.g, which cells, organs, tissues) a GPCR is expressed can make the selection of nucleic acids (e.g., cDNA library) which are likely to encode a ligand for the GPCR easier. For example, the orphan chemokine receptor Bonzo was reported to be expressed on activated T cells which can be found in lymphoid organs/tissues (e.g., lymph nodes, spleen) in humans (Liao, F. et al, J. Exp. Med., 755:2015-2023 (1997), Loetscher, M. et al, Current Biology, 7:652-660 (1997)). Accordingly, a spleen cDNA library was used in the method of the invention to clone a ligand referred to herein as SExCkine (Spleen Extracted Chemokine, also known as chemokine alpha-5 (WO 99/27078)).

Preferably, large pools of expressed proteins are initially assayed. However, the size of the pool that can be successfully assayed may be limited by the ability to detect ligand. Accordingly, the expression library must be expressed (and secreted when expressed in cells) at a high enough level to allow ligand to be detected in a reasonably large pool of protein. Analysis of the primary structure of a GPCR by multiple sequence alignment or other suitable methods can prove helpful in determining the level of expression required for detection of ligand, particularly where the GPCR is an orphan receptor. Multiple sequence alignments and amino acid sequence homology to other receptors can be determined using a suitable sequence alignment algorithm, such as the Lasergene system (DNASTAR, Inc., Madison, WI), using the Clustal method with the PAM 250 residue weight table, a gap penalty of 10, a gap length penalty of 10 and default parameters (pairwise alignment parameters: ktuple = 1 , gap penalty = 3, window = 4 and diagonals saved = 5).

For example, Bonzo was predicted to be a chemokine receptor based upon amino acid sequence homology to other known receptors which belong to the chemokine receptor family. It was known in the art that low nanomolar or high picomolar quantities of chemokine can induce chemotaxis of cells which express an appropriate chemokine receptor. Thus, for cloning a Bonzo ligand, it was desirable to ensure that at least a high picomolar or preferably a nanomolar amount of ligand would be present in a protein pool.

Accordingly, pilot studies were conducted to tailor the method for cloning DNAs which encode chemokines. The human eotaxin receptor CCR3 and its ligand eotaxin were employed in these studies. The eotaxin cDNA was cloned into several expression vectors (including pCDEF3, pSVsport, pEF-1 and pEF-Bos) and the resulting constructs were expressed in a variety of mammalian cells (including 293T, CHO, COS) by transient transfection. The conditioned supematants of the transfected cells were collected and the quantity of eotaxin contained therein was measured. Based upon the level of expression obtained when 293T cells were transfected with the pCDEF3 construct, it was detennined that a 1000 fold dilution of culture supernatant could be detected in a chemotaxis assay. Thus, the pilot studies indicated that pools of about 1000 clones could be screened in a chemotaxis assay when DNAs are cloned in pCDEF3 and expressed in 293T cells. The preliminary analysis of GPCR (e.g., expression and structure) as well as pilot studies to evaluate expression levels of protein can be used to tailor the method of the invention for cloning a ligand for a particular GPCR (e.g., Gi-coupled GPCR, chemokine receptor).

EXAMPLE

METHODS AND MATERIALS

Construction of Recombinant Cells Expressing Bonzo (Bonzo/Ll .2)

DNA encoding Bonzo was obtained by polymerase chain reaction (PCR) using human genomic DNA as template with a synthetic S'-oligonucleotide primer (ttt gga tec atg tat ccc tat gac gtg ccc gac tat get gca gag cat gat tac cat gaa gac tat ggg, SEQ ID NO: 3) and a 3'-oligonucleotide primer (ttt gcg gcc gcc tat aac tgg aac atg ctg gtg gcc tc, SEQ ID NO: 4) which contained flanking BamHI and Notl restriction sites, respectively. The 5'-oligonucleotide primer was designed to produce a DNA encoding Bonzo that contains an N-terminal Hemagglutinin (HA) epitope (CYPYDVPDYASL; SEQ ID NO: 5). The PCR contained 0.2 μM primers (total), 0.39 μg human genomic DNA, 0.2 mM dNTPs, 3.75 U PFU polymerase. Cycling parameters were: 95°C for 5 minutes, followed by 30 cycles of 95°C for 30 seconds, 55°C for 1 minute and 72°C for 1.5 minutes, then 72°C for 10 minutes. The PCR fragment was subcloned into the BamHI and Notl sites of pCDEF/IRES. pCDEF/IRES was prepared by inserting the Mlul-Notl fragment from pCDEF3 (

Goldman, L.A., et al, Biolechniques, 27 : 1013-1015 (1996)) into the Mlul-Notl sites of pIRESneo (Clontech) which contains a bicistronic fragment to facilitate the selection of high expressors. .An EF1 promoter drove expression of the cDNA inserted into pCDEF/IRES. The resulting construct was transfected into the LI .2 cell line (a murine pre-B lymphoma). The murine pre-B lymphoma cell line LI.2 was obtained from Dr. Eugene Butcher (Stanford University) and maintained in RPMI-1640 supplemented with 10%) bovine serum. 20 μg of linearized plasmid was used to transfect the cell line as follows. LI.2 cells were washed twice in HBSS and resuspended in 0.8 ml of the same. The plasmid DNA was mixed with the cells and incubated for 10 minutes at room temperature then transferred to a 0.4 cm electroporation cuvette and a single pulse applied at 250 V, 960 μF. The electroporation was followed by a 10 minute incubation at room temperature. G418 was added to a final concentration of 0.8 mg/ml 48 hr post-transfection and the cells plated in 96 well plates at 25,000 cells/well. After 2-3 weeks under drug selection, cells expressing high levels of Bonzo were selected by staining with anti-HA.l 1 mAb (Babco, Berkely, CA) and subcloned. The resulting stable transfectants were used to immunize mice.

cDNA Synthesis and Construction of Expression Library

Spleen and lymph node polyA⁺ RNA was purchased from Clontech laboratories (Palo Alto, CA). cDNA was synthesized using the cDNA library construction Superscript system from Gibco/BRL (Rockville, MD) with the following modifications from the standard protocol. cDNA was labeled only in the first strand with c ³²P-dCTP and crude estimates of quantity were made by staining aliquots of cDNA fractions with ethidium bromide. EcoRI adapters, purchased from Pharmacia Biotech (Piscataway, NJ) were ligated to the 5' end of cDNAs (instead of the Sail adapters provided by the manufacturer) to facilitate cloning into the pCDEF3 vector. Size fractionated cDNA was ligated in the pCDEF3 vector that was digested with EcoRI and Notl.

A fraction of each ligation was transformed into electrocompetent DH10B bacteria (Gibco/BRL (Rockville, MD)) to estimate both the titer of the library and the average insert size. The actual library used for cloning the Bonzo ligand was a spleen library which consisted of 80,000 independent clones. Bacteria transformed with the library were plated at a density of 800 clones/plate on 96 LB/amp plates (Luria broth agar plates with ampicilin (50 μg/mL)) and cultured overnight at 37°C to generate 96 pools of 800 clones/pool. Pools were collected by overlaying each plate with approximately 2 L of Luria broth (LB), the colonies were scraped off the plates with a standard tissue culture cell scraper and the resulting bacterial suspensions were transferred to microfuge tubes. Plasmid DNAs were purified using QIAprep spin columns (QIAGEN, Valencia, CA) according to manufacturer's instructions.

Transfection and Expression

About 24 hours prior to transfection, 293T cells were seeded into 24 well collagen coated plates (Becton-Dickinson, Franklin Lakes, NJ) at a density of about 60,000 cells/well and cultured at 37°C, 5% CO₂ until transfection. Plasmids (expression plasmids) recovered from the pools were transiently transfected using the LiptofectAMINE^™ reagent (Gibco/BRL (Rockville, MD)), following the manufacturer's protocol with further optimization for 24 well plates as follows: 200 μg of plasmid DNA (representing either a plasmid pool or purified control DNA) was diluted to 20 μL with Opti-MEM 1 reduced serum media (Gibco/BRL (Rockville, MD)) and then diluted into 20 μL of a mixture that consisted of 18 μL Opti-MEM 1 and 2 μL of LiptofectAMINE^™ reagent. This liposome mixture was then incubated for approximately 30 minutes at ambient temperature. Then 200 μL of Opti-MEM 1 was added and the entire mixture was overlayed onto a well of 293T cells. The plates were returned to the incubator for about 2.5 hours. Then, 240 μL of MEM-α (Gibco/BRL (Rockville, MD)) media with 20% fetal calf serum (FCS) was added to each well and the plates were incubated for an additional 18-24 hours. The media was then changed to standard DMEM with 10% FCS, and the cells were cultured (37°C, 5% CO₂) for about 72 hours. After culturing, the conditioned culture supematants were harvested and cellular debris was removed by centrifugation.

Chemotaxis Assays

The conditioned supematants from transfected 293T cells were assayed for Bonzo ligand in a chemotaxis assays using Bonzo/Ll .2 cells. Two days prior to the assay, the Bonzo/Ll.2 cells were split to a density of about 0.3 x lOVmL. On the day of the assay the Bonzo/Ll .2 cells were centrifuged and resuspended at a density of about 1 x 10⁷/mL in an assay buffer which consisted of DMEM supplemented with 10%) bovine calf serum. 100 μL of the cell suspension was placed in the upper chambers of 24 well Transwell plates (Costar Corp., Cambridge, MA) and 0.5 mL of a supernatant to be tested was placed in the lower chambers, and the plates were incubated at 37°C for 6-24 hours. The contents of the lower chamber were then removed, placed in FACS tubes and counted on a FACScan (Becton-Dickinson, Franklin Lakes, NJ) using the acquisition phase at 30 second intervals to quantify the cells that migrated to the lower chamber.

Recombinant LI.2 cells expressing C-C chemokine receptor 3 (CCR3/ LI.2) which undergo eotaxin- induced chemotaxis, and conditioned supematants of 293T cells transfected with an expression vector that encodes eotaxin (pCDEF3 eotaxin) or empty expression vector (pCDEF3) were used as controls in the chemotaxis assay.

Purification of Clones

Plasmid DNA (expression plasmids) from pools that contained chemotaxis- inducing activity were further fractionated by the following protocol: Plasmid DNA (expression plasmids) from a positive pool was transformed into DH10B and plated on 48 LB/amp plates at a density of approximately 200 colonies/plate. The plasmid DNA (expression plasmids) from each of the plates was purified and transfected into 293T cells to produce pools of proteins. The transfected 293T cells were cultured for about 72 hours and the culture supematants were assayed for chemotaxis-inducing activity to identify positive pools, as described above. A positive pool was then further fractionated into pools of about 25 colonies which were replica plated and grown overnight in LB/amp media. After one more round of DNA purification and chemotaxis assays individual clones were grown up, and the clones encoding the chemotaxis-inducing activity were identified.

DNA sequencing

Sequencing of the entire cDNA insert was accomplished in conjunction with Seqwrite (Houston, TX) and the Tufts University sequencing core facility (Boston, MA). Synthetic oligonuclcotide pπmers (originally using an SP6 primer for the 3' end and a pnmer from the EF1 promoter for the 5' sequence) which were complimentary to previously identified nucleotide sequences were synthesized as sequence information was gatheied Both strands of the cDNA were sequenced, resulting in the identification of complete unambiguous sequences of both strands of the insert cDNA The sequences were analyzed for similanty to other nucleotide sequences using the Lasergene system (DNASTAR Inc , Madison, WI) Nucleotide sequence alignments with nucleotide sequences encoding other chemokines were performed by the Clustal method using a gap penalty of 10 and a gap length penalty of 10 Pairwise alignment parameters were ktuple = 2, gap penalty = 5, window = 4 and diagonals saved = 4

Northern blot analysis

Human multiple tissue northern blots I and II and a cancer cell line blot (Clontech) were used to analyzed expression of the gene encoding the Bonzo ligand cDNA probes were labeled with α³²P-dCTP by priming with random hexamers A 400 bp fragment representing most of the chemokine domain of SExCkine cDNA cloned in pCDEF3 (from the 5' EcoRI site (withm vector pCDEF3) to an EcoRV site of a cDNA encoding human SExCkine (SEQ ID NO 1)) was used as the h)bndιzatιon probe for all blots Hybndization w as performed at 68°C for 1 hour in ExpressHyb (Clontech) with denatured probe at a concentration of 1 X 10⁶ CPM/mL Blots weie then washed for 20 minutes in 2 X SSC/0 05% SDS at room temperature followed by high stringency washes at 50°C, 60°C, or 65°C m 0 1 X SSC/0 1%) SDS foi 20 minutes per wash and exposed to Kodak XAR film with an intensifying screen

Results A cDNA library , derived fiom human spleen was transformed into E coli

DH10B and the transformed bacteria were plated and cultured to generate 96 plates which each contained about 800 individual colonies The colonies on a plate weie collected as a pool and plasmids (expiession plasmids) were puiified from the bactena The puiified plasmids w eie introduced by transient transfection into 293T cells for expression, and the conditioned supematants of the cultures were screened for chemotactic activity using LI .2 cells stably transfected with the Bonzo receptor. Screening of the 96 individual library pools from the spleen expression library resulted in the identification of one pool (D3) containing chemotactic activity significantly above background (Figure 1). Repeated further fractionation of that pool (D3) ultimately resulted in the isolation of a single cDNA clone which, upon transfection, retained the chemotactic activity (Figure 3). The novel chemokine encoded by this clone is referred to as SExCkine, for Spleen Extracted Chemokine. The specificity of the SExCkine-Bonzo interaction was verified using antibodies which specifically bind Bonzo. As shown in Figure 4, anti-Bonzo mAbs (mAb 7F3 or mAb 4A11) inhibited SExCkine-induced chemotaxis of Bonzo/Ll.2 cells.

Sequence analysis of the isolated clone revealed that the cDNA contained 1763 nucleotides with an open reading frame of 254 amino acids (Figure 2). Hydrophobicity analysis indicated a type 1 transmembrane protein with a predicted signal peptide of 29 amino acids and a transmembrane segment of 25 amino acid residues from threonine 202 (T202) to leucine 226 (L226) and a short 28 residue cytoplasmic tail. While basic local alignment search tool (BLAST) (see, for example, Altschul, S. F., et al, J. Mol Biol, 275:403-410 (1990)) searches revealed no significant homology to known proteins, inspection and subsequent protein sequence alignments indicated spacing of 4 cysteine residues that partially align with other members of the α-chemokine, or CXC chemokine family, which is consistent with the functional chemotaxis data (Figures 1 and 3). However, the SExCkine protein includes an N-terminal chemokine domain, a membrane-proximal mucin domain, a transmembrane region and a cytoplasmic tail (Figure 2, Figure 5). Thus, SExCkine is structurally similar to the CX3C chemokine fractalkine (Bazan, J.F., et al, Nature 385(6617 :640-644 (1997)). The primary structure indicates that SExCkine can be expressed on the cell membrane (as an integral membrane protein). As shown in Figure 3, chemoattractant activity was found in the supernatant of 293T cells transfected with a cDNA (SEQ ID NO:l) encoding the full length protein (Figure 2, SEQ ID NO:2). Thus, at least some SExCkine is processed (e.g., by cleavage) to form a soluble chemokine. A protein referred to as chemokine alpha-5, which has a similar amino acid sequence has been reported (WO 99/27078).

Multiple transcripts which hybridized with a SExCkine cDNA probe were detected in many tissues, including, spleen, thymus, prostate, testis, ovary, small intestine, colon, peripheral blood leukocytes, pancreas, kidney, liver, lung, placenta, brain and heart, and several cancer cell lines, including melanoma, lung carcinoma, colorectal adenocarcinoma, Burkitt's Lymphoma, lymphoblastic leukemia, Hela cells and promelocytic leukemia HL60, by Northern blot analysis. High expression of a 1.8 kb transcript which corresponds in size to isolated cDNA encoding human SExCkine (SEQ ID NO:l) was seen in spleen, peripheral blood leukocytes, prostate, testis and ovary. The nature of other hybridizing transcripts, which can be partially processed molecules or molecules with a similar nucleotide sequence, is under investigation.

While this invention has been particularly shown and described with references to preferred embodiments thereof, it will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the scope of the invention encompassed by the appended claims.

Claims

CLAIMSWhat is claimed is:

1. A method for isolating a nucleic acid encoding a ligand for a G protein- coupled receptor (GPCR), comprising providing one or more primary pools of prokaryotic cells into which an expression library comprising exogenous nucleic acids has been inserted, said prokaryotic cells having been cultured under conditions suitable to produce individual colonies and a predetermined number of said colonies having been combined into one or more primary pools of prokaryotic cells; and a) expressing the expression library from a pool of cells to produce a corresponding pool of proteins; b) assaying a pool of proteins for ligand, and selecting a pool of proteins that comprises ligand; c) providing prokaryotic cells into which the expression library encoding the pool of proteins selected in b) has been inserted, and growing the cells under conditions suitable to produce individual colonies; d) combining a predetermined number of colonies from c) into a secondary pool of cells, wherein the secondary pool comprises fewer colonies than said primary pool of prokaryotic cells; e) repeating a) through d) until a pool of cells containing an exogenous nucleic acid encoding ligand is isolated, said pool of cells comprising cells obtained from a single colony; and f) recovering the nucleic acid encoding said ligand.

2. The method of Claim 1 wherein said prokaryotic cells are E. coli.

3. The method of Claim 1 wherein said primary pool of prokaryotic cells comprises cells obtained from about 100 colonies.

4. The method of Claim 1 wherein said primary pool of prokaryotic cells comprises cells obtained from about 100 to about 1000 colonies.

5. The method of Claim 1 wherein said primary pool of prokaryotic cells comprises cells obtained from about 800 colonies.

6. The method of Claim 1 wherein in a), said expression library is expressed in eukaryotic cells.

7. The method of Claim 6 wherein said eukaryotic cells are mammalian cells.

8. The method of Claim 7 wherein said mammalian cells are transiently transfected with said expression library.

9. The method of Claim 8 wherein said mammalian cells are selected from the group consisting of COS cells and CHO cells.

10. The method of Claim 8 wherein said mammalian cells are 293T cells.

1 1. The method of Claim 6 wherein said eukaryotic cells are yeast cells.

12. The method of Claim 6 wherein said eukaryotic cells are insect cells.

13. The method of Claim 1 wherein in a), said expression library is expressed in vitro.

14. The method of Claim 1 wherein in a), said expression library is expressed in prokaryotic cells.

15. The method of Claim 1 wherein said pool of proteins is assayed for ligand in a receptor binding assay.

16. The method of Claim 15 wherein said receptor binding assay is a competitive binding assay.

17. The method of Claim 1 wherein said pool of proteins is assayed in a functional assay, wherein a cell expressing said GPCR is contacted with a pool of proteins and a cellular response induced by ligand is monitored.

18. The method of Claim 17 wherein said response is selected from the group consisting of Ca²⁺ flux, chemotaxis, exocytosis and respiratory burst.

19. The method of Claim 18 wherein said response is chemotaxis.

20. The method of Claim 17 wherein said cell expressing said GPCR is a recombinant cell.

21. The method of Claim 20 wherein said recombinant cell couples said G protein-coupled receptor to Gαi.

22. The method of Claim 21 wherein said recombinant cell is a HEK293 cell or a 300-19 cell.

23. The method of Claim 21 wherein said recombinant cell is a LI .2 cell.

24. The method of Claim 1 wherein said expression vector is selected from the group consisting of pCDEF3, pSVsport, pEF-1 and pEF-Bos.

25. The method of Claim 1 wherein said expression vector is pCDEF3.

26. The method of Claim 1 wherein said G protein-coupled receptor is a Gαi coupled receptor.

27. The method of Claim 26 wherein said G protein-coupled receptor is a chemokine receptor.

28. The method of Claim 1 wherein said nucleic acid is cDNA.

29. The method of Claim 1 wherein said nucleic acid is genomic DNA.

30. A method for isolating a cDNA encoding a ligand for a chemokine receptor, comprising providing one or more primary pools of prokaryotic cells into which an expression library comprising exogenous cDNAs has been inserted, said prokaryotic cells having been cultured under conditions suitable to produce individual colonies and a predetermined number of said colonies having been combined into one or more primary pools of prokaryotic cells; and a) expressing the expression library from a pool of cells to produce a corresponding pool of proteins; b) assaying a pool of proteins for ligand, and selecting a pool of proteins that comprises ligand; c) providing prokaryotic cells into which the expression library encoding the pool of proteins selected in b) has been inserted, and growing the cells under conditions suitable to produce individual colonies; d) combining a predetermined number of colonies from c) into a secondary pool of cells, wherein the secondary pool comprises fewer colonies than said primary pool of prokaryotic cells; and e) repeating a) through d) until a pool of cells containing an exogenous cDNA encoding ligand is isolated, said pool of cells comprising cells obtained from a single colony; and f) recovering the cDNA encoding said ligand.

1. The method of Claim 30 wherein binding of ligand to said chemokine receptor is assayed by monitoring a ligand-induced chemotaxis of a cell expressing said chemokine receptor.