Isolated FGF receptor

Abstract

Disclosed herein are compositions and methods which are useful in the identification and isolation of components involved in transmembrane receptor-mediated signaling. Such components include the receptors themselves (e.g., tyrosine kinase receptors, cytokine receptors and tyrosine phosphatase receptors), as well as ligands which bind the receptors and modulators of the downstream intracellular catalytic event which characterizes receptor-mediated signalling. Two novel ligands for the FGF receptor and the nucleotide sequences encoding them are also described.

Description

BACKGROUND OF THE INVENTION

[0001] Transmembrane receptors are proteins which are localized in the plasma membrane of eukaryotic cells. These receptors have an extracellular domain, a transmembrane domain and an intracellular domain. Transmembrane receptors mediate molecular signaling functions by, for example, binding specifically with an external signaling molecule (referred to as a ligand) which activates the receptor. Activation results typically in the triggering of an intracellular catalytic function which is carried out by, or mediated through, the intracellular domain of the transmembrane receptor.

[0002] There are various families of transmembrane receptors that show overall similarity in sequence. The highest conservation of sequence is in the intracellular catalytic domain. Characteristic amino acid position can be used to define classes of receptors or to distinguish related family members. Sequences are much more divergent in the extracellular domain.

[0003] A variety of methods have been developed for the identification and isolation of transmembrane receptors. This is frequently a straightforward matter since receptors often share a common sequence in their catalytic domain. However, the identification of the ligands which bind to, and activate, the transmembrane receptors is a much more difficult undertaking. Brute force approaches for the identification of ligands for known receptors are rarely successful. Brute force approaches usually depend on a biological activity that can be monitored (e.g., nerve growth for nerve growth factor; or glucose homeostasis for the insulin receptor) or they depend on finding a source of the ligand and using affinity to purify it. In general, however, a source of the ligand is not known, nor is there an obvious or easily assayed biological activity. Therefore, there are many receptors, referred to as “orphan receptors”, for which no corresponding ligand has been identified. Further, although several ligands may be known for a specific receptor, it is important to determine the remaining ligands for that receptor to fully understand its role in the growth and maintenance of the vertebrate body. A systematic approach to the identification of receptor ligands would be of great value for the identification of ligands having useful pharmacological activities.

SUMMARY OF THE INVENTION

[0004] The present invention relates to compositions and methods which are useful in connection with the identification of transmembrane receptors and their corresponding ligands. Preferred transmembrane receptors include tyrosine kinase receptors, cytokine receptors and tyrosine phosphatase receptors. Such receptors mediate cell signaling through the interaction of specific binding pairs (e.g., receptor/ligand pairs). The present invention is based on the finding that an unknown component in a receptor-mediated signaling pathway, which results ultimately in an intracellular catalytic event, can be identified by combining other known components within a cellular background within which the catalytic event ordinarily will not take place at significant levels. A cDNA expression library is then used to transform such cells. If the cDNA insert encodes the missing component of the transmembrane receptor-mediated signaling pathway, the catalytic event will be triggered. Detection of the otherwise absent catalytic activity is indicative of a cDNA insert encoding the missing component.

[0005] The invention also provides two novel ligands for the FGF receptor. Both the isolated DNA sequences of these ligands (FRL-2 is SEQ ID NO: 1 and FRL-1 is SEQ ID NO: 3), as well as the isolated polypeptides (SEQ ID NO: 2 and SEQ ID NO: 4, respectively) encoded by these DNA sequences are described. Other nucleic acids of this invention include nucleotide sequences, both DNA and RNA, that comprise a portion of or all of sequences complementary to the DNA sequences described above. The genes FRL-2 and FRL-1 were formerly designated XT1 (or ALP) and EG2 (or CLP), respectively, in U.S. Pat. No. 5,573,944.

[0006] This invention also encompasses agonists (mimics) and antagonists (inhibitors or blocking agents) of the polypeptides described herein. Agonists and antagonists can include antibodies or other polypeptides with amino acid sequences that produce a similar (trigger FGF-mediated phosphorylation) or inhibitory function regarding the binding of the ligand to its FGF receptor.

[0007] The compositions of this invention may be used for diagnostic and therapeutic purposes, either alone or in combination with other compounds. Transgenic gene therapy is also provided using the DNA sequences or a fragment thereof in a sense or antisense orientation to affect the function or lack of function of an FGF receptor in vertebrate cells or tissues.

BRIEF DESCRIPTION OF THE DRAWINGS

[0008] FIG. 1A-1B is a diagram illustrating the steps employed in the identification of a ligand specific for the FGF receptor.

[0009] FIG. 2 is a diagram illustrating the colony Western blot technique.

[0010] FIG. 3 is the nucleotide sequence (SEQ ID NO: 1) of a cDNA clone, FRL-2 (ALP), encoding the angiogenin-like ligand.

[0011] FIG. 4 is the amino acid sequence (SEQ ID NO: 2) of the polypeptide encoded by the nucleotide sequence of FIG. 3.

[0012] FIG. 5 is a comparison of the amino acid sequence (SEQ ID NO: 2) of the FRL-2 (ALP) gene product compared to bovine angiogenin protein (SEQ ID NO: 5) and Chinese hamster pancreatic RNase (SEQ ID NO: 6).

[0013] FIG. 6 is the nucleotide sequence (SEQ ID NO: 3) of a cDNA clone, FRL-1 (CLP), encoding the cripto-like ligand.

[0014] FIG. 7 is the amino acid sequence (SEQ ID NO: 4) of the polypeptide encoded by the nucleotide sequence of FIG. 6.

[0015] FIG. 8 is a comparison of the amino acid sequence (SEQ ID NO: 4) of the FRL-1 (CLP) gene product compared to mouse cripto protein (SEQ ID NO: 7).

[0016] FIG. 9 shows the activation of FGFR by the FRL-2 (ALP) and FRL-1 (CLP) proteins in Xenopus oocytes.

[0017] FIG. 10A-10B is a comparison of the amino acid sequences of human bFGF;

[0018] bovine aFGF; human Kaposi FGF; Xenopus embryonis FGF; human FGF6; human FGF5; human FGF; mouse int2; human FGF9; and human FGF8 (SEQ ID NOS: 8-17, respectively) all of which are known ligands for FGF receptors, to a majority (e.g., consensus) sequence (SEQ ID NO: 18.

[0019] FIG. 11 shows the predicted cleavage sites, the glycosylation sites, and the hydrophophic regions at the C-terminus of the FRL-1 (CLP) (SEQ. ID NO: 4) and FRL-2 (ALP) (SEQ ID NO: 2) proteins.

[0020] FIG. 12 shows the amino acid residues of the FRL-1 (CLP) (SEQ ID NO: 4) protein that are highly conserved in the EGF repeat.

DETAILED DESCRIPTION OF THE INVENTION

[0021] Transmembrane receptors have a binding site with high affinity for a specific signaling molecule. The signaling molecule is referred to herein as a ligand. The present invention is based on the development of a novel approach for the identification of polypeptide ligands by functional expression in the yeast Saccharomyces cerevisiae. This approach is based on the previously unproven hypothesis that it may be possible to functionally express a heterologous tyrosine kinase receptor and its corresponding polypeptide ligand in the same yeast cell, leading to the activation of the receptor and a substantial increase in intracellular tyrosine phosphorylation. The intracellular tyrosine kinase activity of the tyrosine kinase receptor is activated by the binding of a ligand to the extracellular domain of the receptor. This interaction can occur on the surface of the cell (plasma membrane) or in intracellular membrane compartments such as secretory vesicles. In either case, according to the hypothesis confirmed herein, the activation of the cytoplasmically oriented kinase domain results in phosphorylation of tyrosine residues of cytoplasmic protein targets.

[0022] Yeast was chosen as an expression system because many molecular biological techniques are available and it has been demonstrated that many higher eukaryotic genes, including some growth factor-encoding genes, can be functionally expressed in yeast. In addition, only a few endogenous protein tyrosine kinases have been identified in yeast, so that yeast is expected to have a low background of endogenous tyrosine phosphorylation. These features enabled the development of a screen to identify polypeptide ligands for heterologous tyro sine kinase receptors for which no ligand has yet been identified. Such receptors are referred to as orphan receptors. The term heterologous is used herein to mean “non-endogenous”. Thus, for example, a tyrosine kinase which is heterologous in the yeast Saccharomyces cerevisiae is a tyrosine kinase which is non-endogenous (i.e., not present) in wild-type Saccharomyces cerevisiae.

[0023] The disclosed method for identifying a ligand for a tyrosine kinase receptor involves the co-expression in yeast cells (preferably Saccharomyces cerevisiae) of a gene encoding a tyrosine kinase receptor, together with an expression cDNA library which, for example, is constructed from a tissue or cell line that is thought to synthesize a receptor ligand in vivo. The tyrosine kinase gene, together with any regulatory elements required for expression, can be introduced into the yeast strain on a stable plasmid (e.g., a CEN-based plasmid), or it can be integrated into the yeast chromosome using standard techniques (Methods In Enzymology, vol. 194, C. Guthrie and G. Fink, eds., (1991)).

[0024] The choice of expression vectors for use in connection with the cDNA library is not limited to a particular vector. Any expression vector suitable for use in yeast cells is appropriate. The discussion relating to experiments disclosed in the Exemplification section which follows describes a particular combination of elements which was determined to yield meaningful results. However, many options are available for genetic markers, promoters and ancillary expression sequences. As discussed in greater detail below, the use of an inducible promoter to drive expression of the cDNA library is a preferred feature which provides a convenient means for demonstrating that observed changes in tyrosine kinase activity are, in fact, cDNA dependent.

[0025] In a preferred format of the assay, two expression constructs are employed; the first expression construct contains the tyrosine kinase gene and the second expression construct carries the cDNA library. Typically the two expression constructs are not introduced simultaneously, but rather a stable yeast strain is first established which harbors the tyrosine kinase receptor carried on a CEN-based plasmid. Other regulatory sequences are included, as needed, to ensure that the tyrosine kinase gene is constitutively expressed. A CEN-based expression vector contains CEN sequences which are specific centromeric regions which promote equal segregation during cell division. The inclusion of such sequences in the expression construct results in improved mitotic segregation. It has been reported, for example, that mitotic segregation of CEN-based plasmids results in a population of cells in which over 90% of the cells carry one to two copies of the CEN-based plasmid. Faulty mitotic segregation in a similar transformation experiment with an otherwise identical expression construct which lacks CEN sequences would be expected to result in a cell population in which only about 5-20% of the cells contain the plasmid.

[0026] Many transmembrane tyrosine kinase receptors have been identified (for reviews see, e.g., Hanks, Current Opinion in Structural Biology 1: 369 (1991) and Pawson and Bernstein, Trends in Genetics 6: 350 (1990)). A number of these tyrosine kinase receptors are orphan receptors for which no activating ligand has been identified. Any transmembrane tyrosine kinase that can be expressed in yeast cells is useful in connection with the present invention. Based on fundamental principles of molecular biology, there is no reason to believe a priori that any member of the tyrosine kinase receptor family would not be useful in connection with the present invention. Preferably, the gene encoding the tyrosine kinase receptor is isolated from the same organism from which nucleic acid is to be isolated for use in the construction of a cDNA library.

[0027] As discussed in the Exemplification section which follows, the level of expression of the transmembrane tyrosine kinase is a variable which must be considered in the design of the assay for ligand identification. For example, it was determined that high level expression of the FGF receptor results in a substantial increase in intracellular phosphorylation, even in the absence of FGF. Therefore, it is important that expression of the transmembrane receptor be driven by regulatory elements which result in a sufficient level of expression of the transmembrane receptor to facilitate detection following activation of the receptor by ligand binding, while not resulting in overexpression to the extent that ligand-independent autophosphorylation results. A preferred promoter for the expression of the transmembrane receptor is the ACT1 (actin) promoter. This promoter was determined to provide a robust, ligand-dependent signal in the experiments described below.

[0028] The cDNA library is prepared by conventional techniques. Briefly, mRNA is isolated from an organism of interest. An RNA-directed DNA polymerase is employed for first strand synthesis using the mRNA as template. Second strand synthesis is carried out using a DNA-directed DNA polymerase which results in the cDNA product. Following conventional processing to facilitate cloning of the cDNA, the cDNA is inserted into an expression vector suitable for use in yeast cells. Preferably the promoter which drives expression from the cDNA expression construct is an inducible promoter (e.g., GAL1).

[0029] As disclosed in the Exemplification section that follows, removal of the endogenous signal sequence from a cDNA insert encoding a functional receptor ligand resulted in inactivation of the ligand. It appears, therefore, to be necessary to include a signal sequence in the cDNA library constructs to mark the encoded polypeptide for transport across the membrane of the endoplasmic reticulum thereby enabling the extracellular release of the encoded polypeptide which facilitates interaction with the extracellular domain of a transmembrane receptor. The signal sequence employed in the experiments disclosed herein was the signal sequence of Saccharomyces cerevisiae invertase. However, any signal sequence which can function in yeast should be useful in connection with the present invention (Nothwehr and Gordon, Bioessays 12: 479 (1990)).

[0030] The cDNA expression library is then used to transform the yeast strain which constitutively expresses the transmembrane tyrosine kinase gene. mRNA encoding the tyrosine kinase receptor and the cDNA product are thought to be translated in the rough endoplasmic reticulum, accumulate in the inner cavity of the rough endoplasmic reticulum, and migrate to the lumen of the Golgi vesicles for transport to the Golgi complex. Within the Golgi complex, proteins are “addressed” for their ultimate destination. From the Golgi complex, the addressed proteins are transported out of the complex by secretory vesicles.

[0031] A transmembrane tyrosine kinase receptor, if sequestered in a secretory vesicle, the Golgi complex or the endoplasmic reticulum, is oriented such that the cytoplasmic domain is in contact with the cellular cytoplasm as the various vesicles migrate from the Golgi complex to the plasma membrane which is the ultimate destination for a transmembrane receptor. It is possible that the signal sequence bearing polypeptides encoded by the cDNA library can be co-compartmentalized with the transmembrane receptor in the same secretory vesicle. If this were to occur, any cDNA encoded ligand specific for the tyrosine kinase receptor could bind with the “extracellular” portion of the tyrosine kinase receptor (which is located in the internal portion of the secretory vesicle during the migration to the plasma membrane) thereby activating intracellular tyrosine kinases through contact with the cytoplasmically oriented intracellular domain of the tyrosine kinase receptor. Alternatively, activation of intracellular tyrosine kinase activity could also result from interaction with an extracellular polypeptide encoded by the cDNA library through interaction with a plasma transmembrane tyrosine kinase receptor. This occurs, for example, following migration of the secretory vesicle to the plasma membrane resulting in the incorporation of the plasma transmembrane tyrosine kinase receptor and export of the signal sequence-bearing cDNA encoded polypeptide ligand.

[0032] In either case, activation of the intracellular tyrosine kinase activity results in the phosphorylation of intracellular tyrosine residues at a level which is substantially higher (i.e., at least about 4-fold higher) than background levels of phosphorylation in the yeast stain harboring an expression construct containing only the gene encoding the tyrosine kinase receptor (the negative control strain).

[0033] The preferred method for determining the level of intracellular tyrosine phosphorylation is a colony Western blot using replica plates. It will be recognized that, although particularly convenient, the colony Western blot method is but one example of many conventional assays which could be employed to determine levels of intracellular tyrosine kinase activity. The colony Western blot procedure using replica plates is shown diagrammatically in FIG. 2. cDNA library transformants are initially plated on media which do not contain an inducer of the promoter which drives expression of the cDNA insert. For examples, if the GAL1 promoter is used to drive expression of the cDNA insert, cDNA library transformants are initially plated on a medium containing 2% glucose. On this growth medium, cells containing the cDNA expression construct will grow, but the encoded cDNA product is not expressed.

[0034] A set of replica filters is produced from the initial transformation plate by sequentially placing a set of directionally oriented membranes (e.g., nitrocellulose filter membranes) over the transformation plate such that the membrane contacts existing transformant colonies. Cells from transformation colonies adhere to the membranes to form a pattern which represents the pattern of colonies on the transformation plate. Each of the replica filters is then placed on a separate plate, one of which contains a compound which will induce the inducible promoter (e.g., 2% galactose to induce the GAL1 promoter) and one of which will not induce the inducible promoter (e.g., 2% glucose for the GAL1 promoter). Both plates are incubated overnight to promote regrowth of the original cDNA library transformants.

[0035] Following overnight incubation, the replica filters are removed from the growth medium plates, and the colonies are lysed in situ by soaking the replica filters in a lysis solution for a period of time sufficient to lyse cellular membranes (e.g., 0.1% SDS, 0.2 N NaOH, 35 mM DTT for about 30 minutes). The replica filters are then probed with anti-phosphotyrosine antibodies. Colonies which exhibit elevated tyrosine kinase activity on the replica filter which had been incubated overnight on a growth medium containing a compound which induces expression of the cDNA insert linked to the inducible promoter, but which do not exhibit elevated tyrosine kinase activity on the replica filter incubated overnight on a growth medium lacking the inducing compound, contain a cDNA insert encoding a candidate ligand.

[0036] To confirm that a candidate ligand is, in fact, a ligand (and not, for example, a distinct tyrosine kinase), the expression construct is recovered (or rescued) from the cells of the colony demonstrating increased tyrosine kinase activity when grown under inducing conditions. The rescued expression construct is then used to transform a first yeast strain which is known to constitutively express the tyrosine kinase gene, and a second yeast strain which does not express the tyrosine kinase gene. Increased tyrosine kinase activity in the strain which is known to express the tyrosine kinase gene, coupled with no increased tyrosine kinase activity in the strain which does not express the tyrosine kinase gene, serves as confirmation that the cDNA insert of the cDNA expression construct encodes a polypeptide ligand which binds to, and activates, the tyrosine kinase gene product.

[0037] Following confirmation that the candidate ligand is, in fact, a receptor ligand, it is a straightforward matter to identify and characterize the polypeptide encoded by the cDNA library which is responsible for the increase in tyrosine kinase activity. This is accomplished by isolating plasmid DNA from the strain which exhibits the elevated tyrosine kinase activity and characterizing the insert carried in the plasmid (e.g., by DNA sequence analysis). The molecule encoded by the cDNA insert can then be further characterized by conventional approaches such as expression and isolation of the encoded polypeptide followed by in vitro binding studies in order to confirm the specificity of the binding interaction with the transmembrane receptor.

[0038] The method of the present invention is not limited to the isolation of tyrosine kinase receptor ligands. Rather, the method can be modified for use in the identification of ligands for any transmembrane receptor having a single transmembrane domain, an extracellular domain and an intracellular domain. This is accomplished by generating an expression construct encoding a chimeric fusion protein comprising the extracellular domain of a transmembrane receptor fused to the intracellular domain of a specific tyrosine kinase receptor (e.g., the FGF receptor). As mentioned previously, this construct is preferably generated in a CEN-based plasmid background or, alternatively, in a plasmid which will facilitate integration of the chimeric receptor into the yeast chromosome. Conventional molecular biological techniques are employed to generate this construct, as well as all others disclosed in this specification (see e.g., Molecular Cloning—A Laboratory Manual, Sambrook, J., et al., eds., Cold Spring Harbor Publications, Cold Spring Harbor, N.Y. (1989)). This expression construct encoding the tyrosine kinase receptor fusion protein is used in a manner analogous to the expression construct encoding the tyrosine kinase receptor in the embodiment described above.

[0039] Briefly, the preferred embodiment of this aspect of the invention includes the construction of a yeast strain which constitutively expresses a chimeric fusion protein of the type described above. This strain is then transformed with a cDNA expression library generated using mRNA isolated from the organism of interest. A ligand which binds specifically to the native transmembrane receptor will bind to the extracellular domain of the tyrosine kinase fusion protein and this ligand binding will trigger ligand-dependent intracellular tyrosine kinase activity mediated by the intracellular domain of the tyrosine kinase receptor. Intracellular tyrosine kinase activity is detected in the manner described previously.

[0040] A specific example of this embodiment of the present invention is applicable to the isolation of a ligand for a cytokine receptor (e.g., erythropoietin receptor, interleukin-3 receptor, etc.). Cytokine receptors, like tyrosine kinase receptors, are transmembrane receptors found in mammalian cells and possess both an extracellular domain and an intracellular domain. However, unlike the tyrosine kinase receptors, cytokine receptors do not possess a catalytic domain but rather recruit cytoplasmic tyrosine kinase enzymes in response to ligand activation. More specifically, the intracellular (cytoplasmic) domain of the cytokine receptor has been shown to bind to, and activate, a class of cytoplasmic tyrosine kinases (e.g., the JAK2/TYK2 class).

[0041] To isolate cytokine receptor ligands, a yeast strain is constructed which constitutively expresses a cytoplasmic tyrosine kinase and a transmembrane cytokine receptor. This yeast strain is then transformed with a cDNA expression library from an organism of interest, preferably under the control of an inducible promoter. Elevated levels of tyrosine kinase activity will be observed if the polypeptide encoded by the cDNA library insert functions as a ligand for the native cytokine receptor. Binding of the polypeptide ligand to the extracellular domain of the cytokine receptor (either at the plasma membrane or within a secretory vesicle) results in the activation of the cytoplasmic tyrosine kinase.

[0042] The colony Western blot procedure discussed above, and shown diagrammatically in FIG. 2, is the preferred method for screening for an expression construct encoding a functional ligand. Specifically, a set of replica filters is prepared from the original transformation plate and the first and second replica filters are incubated overnight under inducing conditions, and non-inducing conditions, respectively. Colonies affixed to the replica filters are then lysed and probed with anti-phosphotyrosine antibodies. Increased levels of tyrosine kinase activity can be indicative of a cDNA insert encoding a ligand for the cytokine receptor or, alternatively, a cDNA insert encoding a cytoplasmic tyrosine kinase enzyme. To determine which of these two alternatives is responsible for the observed increase in tyrosine kinase activity, the expression construct encoding the candidate ligand is rescued and used to independently transform a first cell population which constitutively expresses the cytokine receptor and the cytoplasmic tyrosine kinase, and a second cell population which constitutively expresses the cytokine receptor but not the cytoplasmic tyrosine kinase. Candidates which demonstrate an increase in tyrosine kinase activity in the first cell population, but not the second, encode a cytokine receptor ligand. Expression constructs which result in an increase in tyrosine kinase activity in both the first cell population and the second cell population encode a cytoplasmic tyrosine kinase.

[0043] Given the fundamental disclosure that a yeast cell system can be used to identify ligands and other members of specific binding pairs involved in receptor-mediated molecular signaling, numerous variations of the theme described above are derivable through routine experimentation. Using such variations, any single polypeptide component of the receptor-mediated signaling pathway can be identified through the introduction of a cDNA library into yeast cells which have been modified to constitutively produce other necessary components of the signaling pathway.

[0044] For example, the methods described above can be modified to facilitate the identification of a cytokine receptor. As discussed above, cytokine-receptor mediated signaling involves a cytokine receptor and a cytoplasmic tyrosine kinase which is activated by interaction with the cytoplasmic domain of the cytokine receptor. As reported in the Exemplification section below, overexpression of the transmembrane tyrosine kinase (e.g., by expression from the GAL1 promoter) resulted in ligand-independent tyrosine kinase activity. By analogy, it would be expected that overexpression of a transmembrane cytokine receptor in the presence of a cytoplasmic tyrosine kinase would yield ligand-independent tyrosine kinase activity.

[0045] More specifically, a yeast strain constitutively expressing a cytoplasmic tyrosine kinase is first constructed. The use of the GAL1 promoter would be expected to result in a high level of cytoplasmic tyrosine kinase expression. However, routine experimentation may be required to optimize the expression level. It is preferred, for example, that the cytoplasmic tyrosine kinase be produced at such a level that it is detectable by Western blot.

[0046] A cDNA library is then constructed, preferably with the expression of the cDNA insert under the control of an inducible promoter. Replica filters are produced and incubated independently with, and without, a compound capable of inducing expression from the inducible promoter. Increased levels of tyrosine kinase activity are detected, for example, by colony Western blot in cells grown under inducing conditions, but not under non-inducing conditions. This would be observed, for example, when the cDNA insert encodes a cytokine receptor. The expression construct is rescued from these cells and introduced independently into yeast cells with, and without, constitutively expressed intracellular tyrosine kinase. Increased tyrosine kinase activity which is dependent upon the constitutively expressed cytoplasmic tyrosine kinase of the host strain indicates that the cDNA insert encodes a cytokine receptor. Increased tyrosine kinase activity which is not dependent upon the constitutively expressed cytoplasmic tyrosine kinase of the host strain is an indication that the cDNA insert encodes a functional tyrosine kinase. If such a cytokine receptor is known or discovered, yeast strains expressing the cytoplasmic tyrosine kinase and the cytokine receptor can be employed in a method for the isolation of a ligand in a manner analogous to the methods described elsewhere in this specification.

[0047] Another example of a variation of presently disclosed method is useful for the identification of a receptor for an orphan polypeptide ligand (i.e., a ligand for which no receptor has been previously identified), or for the identification of new receptors for a ligand which is known to interact productively with one or more previously identified receptors. This method incorporates the use of a yeast strain which has been modified to constitutively produce the previously identified ligand or orphan ligand. A cDNA library is introduced and the colony Western blot is employed to identify colonies which exhibit increased tyrosine kinase activity in the induced state. Rescue of the expression construct, followed by retransformation of yeast cells both with and without a constitutively expressed ligand, is used to confirm ligand-dependent activation of tyrosine kinase activity. It will be recognized that the description above relates specifically to a tyrosine kinase-like receptor. The method is easily modified for use with a cytokine receptor by adding constitutive cytoplasmic tyrosine kinase activity to the list of constitutive host cell requirements.

[0048] Similarly, the methods of this invention can be used to identify a cytoplasmic tyrosine kinase if a known cytokine receptor and ligand are provided. In this method, the cytokine receptor and ligand are expressed constitutively in a host yeast strain. The cDNA library is provided, and transformants are screened, in the induced and non-induced state, by the replica method discussed above. Candidate cytoplasmic tyrosine kinases are those encoded by an expression construct conferring increased tyrosine kinase activity in the induced state. The cDNA expression construct is rescued from the identified colony and introduced into yeast cells which constitutively express the cytokine receptor and ligand. The rescued construct is also introduced into a yeast strain lacking the cytokine receptor and ligand. Increased activity in the former, but not in the latter, is indicative of a cDNA insert encoding a cytoplasmic tyrosine kinase.

[0049] In another aspect of the invention, polypeptide modulators of receptor-mediated tyrosine kinase activity can be isolated. A polypeptide modulator can be, for example, a polypeptide (intracellular or extracellular) which modifies the affinity of the ligand for receptor, or which modifies the activity of the catalytic domain (either integral or recruited). Polypeptide modulators can be isolated by first providing a yeast strain which constitutively expresses a ligand/receptor pair (together with the cytoplasmic tyrosine kinase in the case of a cytokine receptor/ligand pair). The construction of such strains has been discussed in greater detail above. A yeast cell which constitutively expresses the ligand/receptor pair is expected to exhibit a relatively high level of background tyrosine kinase activity when the cDNA library is expressed in both the induced and non-induced state. However, the presence of a cDNA insert encoding a strong modulator (either an up-modulator or a down-modulator) will be determined by a detectable (i.e., at least about 2-fold) change in the level of tyrosine kinase activity in the induced state due to the presence of the polypeptide modulator.

[0050] In another aspect of the invention, ligands which specifically activate transmembrane tyrosine phosphatase receptors can be isolated. Transmembrane tyrosine phosphatase receptors are membrane components which have an intracellular catalytic domain which functions to remove phosphate groups from tyrosine residues. In other words, the tyrosine phosphatase receptor function can be viewed as a catalytic function which reverses the action of a tyrosine kinase (a tyrosine kinase functions by adding a phosphate group to intracellular tyrosine residues). Tyrosine phosphatase receptors have an extracellular domain and, therefore, the existence of extracellular ligands is presumed although none have been isolated to date.

[0051] In order to isolate a cDNA fragment encoding a tyrosine phosphatase receptor ligand, it is necessary to first provide a yeast strain which constitutively expresses cellular components necessary to produce a basal level of intracellular tyrosine kinase activity. This can be accomplished, for example, by providing a strain which constitutively expresses appropriate levels of a transmembrane tyrosine kinase receptor, together with its corresponding ligand. Basal levels of tyrosine kinase activity in such a strain are determined using the colony Western blot, for example.

[0052] Following a determination of intracellular tyrosine kinase activity, this strain is further modified to express a tyrosine phosphatase receptor. Subsequent to the introduction of the tyrosine phosphatase receptor gene, levels of tyrosine kinase activity are again determined to ensure that there has been no change in the basal level of phosphorylation detected. In the absence of the tyrosine phosphatase receptor ligand, the addition of the expressible tyrosine phosphatase receptor gene to the strain should not affect basal levels of phosphorylation.

[0053] Confirmation that the introduction of the tyrosine phosphatase gene does not affect detected phosphorylation levels is followed by the introduction of a cDNA library, preferably under the control of an inducible promoter. Replica filters are produced from the plate of transformants and incubated overnight under either inducing or non-inducing conditions. The levels of intracellular tyrosine phosphorylation are then determined, for example, by the colony Western blotting procedure. Reduced levels of intracellular tyrosine phosphorylation under inducing growth conditions, relative to the levels determined under non-inducing growth conditions, are an indication that the cDNA insert encodes a tyrosine phosphatase ligand which binds to the extracellular domain of the tyrosine phosphatase receptor thereby activating the tyrosine phosphatase activity which functions to reduce intracellular tyrosine phosphorylation thereby reversing the effect of the constitutively expressed tyrosine kinase. The initial indication that the cDNA insert encodes a tyrosine phosphatase ligand can be confirmed by further studies including, for example, demonstration that the observed decrease in phosphorylation is dependent upon entry of the cDNA encoded product into the secretory pathway. Confirmation that a signal sequence is encoded by the cDNA insert is an example of one type of confirmatory experiment.

[0054] The methods of the present invention can be further modified for use in the identification of functionally significant domains in a transmembrane receptor or its ligand. This method is carried out, for example, by mutagenizing either the transmembrane receptor or its ligand by conventional site-directed mutagenic techniques. The mutagenized component is then included in an assay of the type described above with a non-mutagenized copy serving as a positive control. Increased intracellular tyrosine phosphorylation in the positive control coupled with a relative decrease in tyrosine phosphorylation (relative to the positive control) in the assay which includes the mutagenized component indicates that the mutagenized amino acid residue(s) are of functional significance.

[0055] The FRL-2 (SEQ ID NO: 1) and FRL-1 genes (SEQ ID NO: 3) encoding the two novel ligands identified by the methods of the present invention were sequenced and the sequences are shown in FIGS. 3 and 6, respectively. This invention provides isolated DNA encoding all or a portion of the FRL-2 or FRL-1 protein ligands, including DNA that comprises (a) SEQ ID NO: 1 or SEQ ID NO: 3; (b) a portion of the nucleotide sequence of SEQ ID NO: 1 or SEQ ID NO: 3; (c) a nucleotide sequence that hybridizes to SEQ ID NO: 1 or SEQ ID NO: 3 under stringent conditions (See, Ausubel, et al. (1994) Current Protocols in Molecular Biology, Section 6.4, John Wiley & Sons, NY); or (d) DNA differing from the DNA sequences of (a), (b) or (c) in codon sequence due to the degeneracy of the genetic code, and which is the functional equivalent of DNA encoding FRL-2 or FRL-1 protein. By “functional equivalent”, it is meant that the DNA encodes a polypeptide that demonstrates the biological function of the FRL-2 or FRL-1 protein ligand.

[0056] The cDNA encoding these ligands may be radiolabeled, or labeled with enzymes, fluorescent compounds, or other detectable compounds, and used as a probe or primer to isolate other vertebrate FRL-2 or FRL-1 cDNAs by cross-species hybridization. Alternatively, Northern hybridization can be used to screen mRNAs from other vertebrate cell lines to identify a source of mRNA by which a ligand gene can be cloned.

[0057] Also provided are the polypeptides comprising the FRL-2 (SEQ ID NO: 2) and FRL-1 (SEQ ID NO: 4) protein ligands or their functional equivalents, as well as amino acid sequences encoded by the DNA described above. FRL-2 and FRL-1 proteins can be synthesized by synthetically constructing and expressing either SEQ ID NO: 1 (for FRL-2) or SEQ ID NO: 3 (for FRL-1) using recombinant DNA technology. The degeneracy of the genetic code also permits a wide variety of codon combinations to be used for constructing the DNA chains that encode these polypeptides.

[0058] The FGF family of receptor ligands contains at least nine members which are structurally closely related to one another. Basilico and Moscatelli (1992) Adv. Cancer Res. 59: 115-1165. FIG. 10A-10B is a comparison of the amino acid sequences of known ligands (SEQ ID NOS: 8-17, respectively) for FGF receptors showing conserved sequences between these proteins. The FRL-2 and FRL-1 ligands described herein are different structurally and have no conserved (consensus) sequences with the other known FGF ligands.

[0059] Growth factors such as FGF are responsible for multiple activities of cells. FGF receptors are expressed in adult tissues and cell lines where they control the proliferation, survival, differentiation, migration or function of cells. Fibroblast growth factor has a broad range of specificity and can stimulate proliferation of many cell types as well as inhibit differentiation of various types of stem cells and act as an inductive signal in embryonic development. The potential for regulating the growth of cells and tissues by stimulating or inhibiting FGF are enormous. Of particular interest is the stimulatory effect of FGF on collateral vascularization and angiogenesis. Such mitogenic effects have stimulated considerable interest in FGF as as a potential therapeutic agent for wound healing, nerve regeneration and cartilage repair.

[0060] Accordingly, the possibilities for using the FRL-2 and FRL-1 ligands to modulate the activities of the FGF receptor in cells are manyfold in vertebrates. Further, agonists and antagonists can produce modulating effects. Antagonists can include antisense nucleotide sequences, either DNA or RNA, that are complementary to all or a part of the FRL-2 or FRL-1 gene as well as blocking agents that interfere with the binding of the ligand to the receptor. The antisense sequences can be introduced by means of gene therapy (via infection or transfection) and used to treat individuals who would benefit from reduced levels of FGF receptor activity.

[0061] This invention also provides fusion proteins and methods of using fusion proteins to detect and identify sites of FRL-2 and FRL-1 ligand interactions with FGF receptors. Fusion proteins can be applied to detect and assay abnormal expression or to monitor the effects of treatment involving variants or mutants of FRL-2 and FRL-1, as well as agonists and antagonists of FRL-2 and FRL-1 ligands on FGF receptor activity.

[0062] See, for example, Cheng, H. J. and J. G. Flanagan (1994) Cell 79: 157-68; Flanagan, J. G. and P. Leder (1990) Cell 63: 185-94.

[0063] Several methods are provided by which the expression of the gene products of FRL-2 or FRL-1 can be detected and quantified. One method for detecting the expression of an FRL-2 or FRL-1 protein ligand in a sample comprises the steps of: (a) treating the sample in a manner that renders mRNA encoding the ligand available for hybridization with a complementary DNA or RNA oligonucleotide, thereby producing a treated sample; contacting the treated sample with at least one DNA or RNA probe which is a nucleotide sequence complementary to all or a portion of the gene or mRNA encoding the ligand; and (c) detecting the hybridization of mRNA from the sample with the probe, wherein hybridization of the mRNA is an indication of the presence of the ligand in the sample. The ligand can be quantified by measuring the extent of hybridization in the sample.

[0064] Another method of detecting the level of expression of FRL-2 or FRL-1 protein ligand in a sample comprises the steps of: (a) treating the sample in a manner that renders the ligand available for binding to antibodies or antibody fragments specific for the ligand, thereby producing a treated sample; (b) contacting the treated sample with the antibody or antibody fragments under conditions appropriate for formation of antibody-antigen complexes; and (c) detecting the presence of antibody-antigen complexes as an indication of the presence of the ligand in the sample. The level of FRL-2 or FRL-1 protein ligand expression can be quantified in the sample by measuring the amount of antibody-antigen complex as a means to determine the amount of the ligand in the sample.

[0065] Antagonists of these ligands can be used to prevent signal transduction of the FGF receptor and thus prevent unwanted resultant cellular responses. For example, a mutated form of SEQ ID NO: 1 or SEQ ID NO: 3 can be prepared that will encode part or all of a polypeptide that competes with the endogenous polypeptide ligand for binding to its FGF receptor but is not able to trigger phosphorylation. Thus the receptor activities are blocked. This can be useful in preventing tumor growth, for example, where angiogenesis is required for a growing tumor to receive increased nutrients through the blood. Without the increase in vascularization, the tumor is dependent on diffusion of nutrients and is essentially inhibited.

[0066] Therefore, in addition to purified FGF receptor ligands, this invention can provide variants and derivatives of native FRL-2 and FRL-1 that retain the desired biological activity (the ability to bind FGF) and modulate the binding of native FRL-2 and FRL-1 to the FGF receptor. A variant, as referred to herein, is a polypeptide which is substantially homologous to a native FGF ligand, but which has an amino acid sequence different from that of the native ligand (from any vertebrate species) because of one or more deletions, insertions or substitutions. Alterations of the native amino acid sequence may be accomplished by any of a number of known techniques. See, for example, Molecular Cloning—A Laboratory Manual, Sambrook, J., et al., eds., Cold Spring Harbor Publications, Cold Spring Harbor, N.Y. (1989). Antibodies (either polyclonal or monoclonal) and antibody fragments such as F(ab)2 fragments can be produced that are specific for (bind to) epitopes of FRL-2 or FRL-1 polypeptides. See, for example, Harlow, E. and D. Lane (1988) Antibodies—A Laboratory Manual, Cold Spring Harbor Press, Cold Spring Harbor Laboratories, NY. These antibodies can be used in immunoassays and diagnostically, and can function as antagonists for treatment purposes. The immunoassays can be used to detect and/or quantitate antigens and antibodies where extreme sensitivity is required, and to monitor the progress of treatment in procedures employing modulators of FRL-2 or FRL-1 activity. The antibodies can be labeled or a second antibody that binds to the first antibody can be labeled by some physical or chemical means. The label can be an enzyme which can be assayed, a radioactive substance, a chromophore, or a fluorochrome. E. Harlow and D. Lane (1988), supra.

[0067] There are also situations in which one may want to induce or enhance FGF-mediated phosphorylation by increasing either FRL-2 or FRL-1 activity or providing an agonist of either of the ligands in vertebrate cells. The activity of FRL-2 or FRL-1 in vertebrates could be important to the growth, maintenance, and aging of normal cells. Further, modulation of FRL-2 or FRL-1 could be useful to prevent or treat tumor formation. Cells can be treated with FRL-2 or FRL-1 or their agonists, or with mRNA encoding these ligands or agonists to induce or enhance FGF activity resulting in proliferation and/or differentiation of cells.

[0068] FGF antagonists have many potential therapeutic applications, such as the treatment of tumors and diseases or disorders of the neural system. FGF antagonists can be combined with a pharmaceutically acceptable diluent, adjuvant or carrier to form a pharmaceutical composition, and can be administered to vertebrates, including humans, either intravenously, subcutaneously, intramuscularly or orally. The required dosage will vary with the particular condition being treated, with the severity of the condition and with the duration of desired treatment. A therapeutically effective dose is one that will result in a partial or complete reduction of some or all of the adverse symptoms of the disease or disorder.

[0069] Angiogenin is a protein of 125 amino acid residues and is able to induce vascularization, including vascularization associated with the growth of tumors. Fett, et al. (1985) Biochemistry 24: 5280-5486. The cripto gene product is a protein of 188 amino acid residues that is expressed in undifferentiated teratocarcinoma cells. Ciccodicola, et al., supra. The cripto gene is expressed differentially in the adult mouse as well as the developing embryo and the regulation of tumor cell growth has been suggested as at least one of its functions. Dono, et al., supra.

[0070] Based on the embryonic expression patterns described herein and the knowledge of FGF receptors as prominent receptors in both the embryonic and adult vertebrate body, it is clear that the FRL-2 and FRL-1 gene products express important ligands that modulate growth and differentiation in the adult vertebrate animals. Thus, these genes and their products provide the means by which diseases and disorders of the vertebrate body resulting from excess activity or abnormal lack of activity of an FGF receptor can be detected and treated. Based on the known activities of angiogenin and the expression patterns of cripto, as well as the FRL-2 and FRL-1 temporal patterns of expression during embryogenesis, examples of such activity may include inhibition of tumor growth, induction of neural cell differentiation for repair and regeneration of the central and/or peripheral nervous system, induction of non-neural cell differentiation for repair and regeneration of other organs, and modulation of maintenance and aging of normal cells. These processes can be carried out prenatally as well as in adults.

EXEMPLIFICATION

[0071] Disclosed in this Exemplification section are experiments which confirm a previously unproven hypothesis that it may be possible to functionally express a tyrosine kinase receptor and its corresponding polypeptide ligand in the same yeast cell, leading to activation of the receptor and a substantial increase in intracellular tyrosine phosphorylation. More specifically, using African clawed frog Xenopus laevis FGF receptor and FGF genes as a model system, it has been demonstrated that tyrosine kinase activity is triggered by co-expression of its ligand gene in yeast cells, provided that the ligand is capable of entering the secretory pathway. This activation of FGF receptor was detected by colony Western blotting which enables the screening of a large number of yeast transformants of a cDNA library. By screening a Xenopus cDNA library with a yeast strain expressing FGF receptor, two genes encoding novel growth factor-like ligands were identified, which can activate the FGF receptor by conventional pathways.

MATERIALS AND METHODS

[0072] i) Yeast Strains

[0073] A yeast Saccharomyces cerevisiae strain used in this study was PSY315 (Mat a, leu2, ura3 his3, lys2).

[0074] ii) Yeast Transformation and Media

[0075] The LiCl method (Ito et al., J. Bacteriol. 153: 167 (1983)) was used for yeast transformation. Following media were used for yeast culture, YPD (1% yeast extract, 2% tryptone, 2% glucose), YPG (1% yeast extract, 2% tryptone, 2% galactose), SD (0.067% yeast nitrogen base w/o amino acids, 2% glucose), and SG (0.067% yeast nitrogen base w/o amino acids, 2% galactose).

[0076] iii) Plasmids

[0077] The vector plasmids pTS210 and pTS249 carry URA3 and LEU2, respectively, and both carry CEN4, GAL1 promoter and ACT1 terminator. The plasmid pKNA1 harbors LEU2, CEN4, ACT1 promoter and ACT 1 terminator.

[0078] Two types of plasmids for expression of Xenopus bFGF (basic fibroblast growth factor) in yeast were constructed: One plasmid is constructed by cloning bFGF gene into pTS210 (pTS-FGF) and a second plasmid is identical to the first except that a signal sequence of S. cerevisiae invertase (Carlson et al., Mol. Cell. Biol. 3: 439 (1983)) was inserted at the initiation codon of the bFGF gene (pTS-ssFGF). For FGF receptor expression, the Xenopus FGF receptor-1 gene (Musci et al., Proc. Natl. Acad. Sci. USA 87: 8365 (1990)) was cloned into pTS249 and pKNA1 (PTS-FGFR and pKN-FGFR, respectively).

[0079] iv) Antibody

[0080] Anti-phosphotyrosine antibody 4G10 is purchased from Upstate Biotechnology Incorporated.

[0081] v) Colony Western Blotting

[0082] Yeast transformants were plated on SD plates and incubated at 30° C. for two days. Colonies were transferred onto two nitrocellulose membranes (Millipore HATF 082). These membranes were placed colony-side up on SD and SG plates, and incubated overnight at 30° C. The membranes were placed on Whatman 3MM filter paper pre-soaked with lysis buffer (0.1% SDS, 0.2 M NaOH, 35 mM DTT), and incubated at room temperature for 30 min. Colonies on the membranes were rinsed off with water, then the membranes were incubated in TBS-T(20 mM Tris-HCI pH 8.0, 150 mM NaCl, 0.05% Tween-20)-2% BSA (sigma) for blocking on a shaker for one hour, then incubated in 1:1,000-diluted anti-phosphotyrosine antibody (in TBS-T with 2% BSA) for one hour, and subsequently washed three times in TBS-T. The blots were then incubated in 1:10,000-diluted HRP(horse radish peroxidase)-conjugated goat anti-mouse Ig antibody (Bio-Rad) for one hour, and washed three times. Detection was done with chemiluminescence reagents (Amersham, ECL).

[0083] vi) cDNA Library

[0084] The vector plasmid of the cDNA library is λyes (Elledge et al., Proc. Natl. Acad. Sci. USA 88:1731 (1991)), which carries URA3, CEN4, ARS1, GAL1 promoter and HIS3 terminator. Two sources of cDNA were used for library construction. One was made from Xenopus XTC cells, The other was made from Xenopus unfertilized eggs and 10 hour embryos.

[0085] vii) Ca2+ Release assay

[0086] The procedure for the Ca2+ release assay described in Amaya et al. (Cell 66: 257 (1991)) was followed. Briefly, oocytes injected with certain mRNAs transcribed in vitro were incubated for two days, then incubated with 45Ca2+ for three hours. These oocytes were washed in 45Ca2+-free medium, incubated in media for 10 minutes, followed by scintillation counting of the released radioactivity.

[0087] viii) Partial Purification of FRL-1 Protein

[0088] Yeast cells expressing the FRL-1 gene under control of GAL promoter were cultured in 1 L of YPG for eight hours (about 2×1010 cells). Cells were collected and disrupted with glass beads in 20 ml of buffer A (20 mM Tris-HCl (pH 8.0), 1 mM EDTA, 1 mM PMSF), containing 150 mM NaCl. Cell debris were removed by low speed centrifugation (3,000×g for 5 minutes). The supernatant was centrifuged at 80,000×g for 20 minutes. The pellet was suspended in 5 ml of buffer A containing 1.2 M NaCl, then centrifuged with the same condition. The resulting pellet was suspended in 2 ml of buffer A containing 1% Triton X-100, and centrifuged with the same condition again. The supernatant was diluted 20 fold in modified Barth's saline (Gurdon, Meth. Cell Biol. 16: 125 (1977)) containing 0.5 mg/ml BSA.

RESULTS AND DISCUSSION

[0089] To test whether co-expression of a receptor-tyrosine kinase and its ligand leads to the activation of the kinase in yeast cells, Xenopus laevis FGF receptor and bFGF were used as a model system. These genes were co-expressed in yeast cells under control of GAL1 promoter by co-transforming pTS-FGFR and pTS-FGF. In addition, bFGF fused with the SUC2 signal sequence (pTS-ssFGF) was also co-expressed with the FGF receptor gene because it is known that the bFGF gene does not have a signal sequence.

[0090] To determine whether the tyrosine kinase is activated in these strains, whole cell extracts were analyzed by immunoblotting with anti-phosphotyrosine antibody. The following results were obtained: (1) Expression of either bFGF or ssFGF alone had no effect on the level of tyrosine phosphorylation. (2) Expression of the FGF receptor plasmid led to a substantial increase in tyrosine phosphorylation of several endogenous proteins. (3) Co-expression of FGF receptor and ssFGF dramatically increased tyrosine phosphorylation to a level that was several times higher than the phosphorylation level observed after expression of the FGF receptor alone. (4) Co-expression of the FGF receptor and bFGF without a signal sequence did not lead to any increase in phosphorylation above that obtained after expression of the FGF receptor alone, although the same levels of the FGF proteins in the strains expressing the bFGF gene with and without the signal sequence are detected by immunoblotting with anti-FGF antibody. FGF could not be detected in culture supernatants, suggesting that the interaction was intracellular or periplasmic.

[0091] These findings demonstrate that it is possible to functionally co-express the FGF receptor and bFGF in yeast in such a way that they can interact productively in an autocrine manner and thereby lead to an increase in the FGF-receptor mediated phosphorylation of endogenous yeast proteins. bFGF with a signal sequence appears to interact with the extracellular domain of the FGF receptor on the cell surface or in internal membrane compartments, while bFGF without a signal sequence localizes in the cytoplasm and cannot interact with the receptor.

[0092] For screening of a large number of yeast transformants, a colony Western blotting method (Lyons and Nelson, Proc. Nat. Acad. Sci. USA 81: 7426 (1984)) was developed. Yeast transformants expressing bFGF (with or without the signal sequence) and/or FGF receptor were plated on a glucose plate. Colonies were transferred to a filter and the filter was then placed on a galactose plate to induce bFGF expression. After overnight incubation, cells on the filter were lysed and the level of tyrosine phosphorylated proteins in each colony was determined by probing with anti-phosphotyrosine antibodies. The results of this experiment were essentially the same as those described above. That is, expression of the FGF receptor led to an increase in the level of tyrosine phosphorylation that was substantially augmented when bFGF containing a signal sequence was co-expressed, but not when bFGF lacking a signal sequence was co-expressed. These results indicate that the colony Western blotting method is sensitive and can be used to rapidly and easily screen thousands of different yeast colonies.

[0093] Several promoters have been tested for the expression of the FGF receptor gene in order to optimize the detection of its activation by colony Western blotting. They included the GAL1, ACT1 (actin; Gallwitz et al., Nucl. Acids Res. 9: 6339 (1981)), GPD1 (glyceraldehyde-3-phosphate dehydrogenase; Bitter and Egan, Gene 32: 263 (1984)) and TUB1 (α-tubulin; Schatz et al., Mol. Cell. Biol. 6: 3711 (1986)) promoters. Among them, the ACT1 promoter was determined to be most suitable. FGF receptor gene expression driven by GAL1 promoter proved very high, leading to high levels of tyrosine phosphorylation even in the absence of FGF, while the TUB1 promoter was extremely weak, such that FGF receptor activation by FGF could not be detected. Under the control of the GPD1 promoter, expression of the FGF receptor gene was repressed by galactose-containing media. On the other hand, the ACT1 promoter gave similar levels of FGF receptor gene expression in galactose- and in glucose-containing media, and levels of tyrosine phosphorylation were low in the absence of FGF, but significantly increased by expression of ssFGF. For these reasons, the ACT1 promoter was used for the cDNA screening experiment described below.

[0094] The above results encouraged further attempts to use this method to identify novel ligands for tyrosine kinase receptors. As a first step, the method was used to identify new ligands for the FGF receptor. The purpose of this experiment is two-fold: first, to determine whether this system can be used to identify genuine FGF genes, and second, to isolate previously unidentified activators of the FGF receptor.

[0095] The procedure followed is outlined diagrammatically in FIG. 1. Yeast cells expressing the FGF receptor were transformed with a cDNA library expected to contain FGF gene family members. Since bFGF (Kimelman et al., Science 242: 1053 (1988)), embryonic FGF (Isaacs et al., Development 114: 711 (1992)) and int-2/FGF3 (Tannahill, et al., Development 115: 695 (1992)) are known to be expressed in Xenopus embryos, we used a cDNA library made from mRNA isolated from Xenopus eggs and embryos (egg library). A library made from XTC cells was also used (XTC library).

[0096] 150,000 and 25,000 transformants were obtained from the egg and XTC libraries, respectively. In the first screening by colony Western blotting with an anti-phosphotyrosine antibody, 65 and 29 candidates were identified, and by the second screening, nine and two transformants were found to be positive (egg and XTC library, respectively). Plasmid DNA in each transformant was rescued, and re-transformed into yeast strains with and without the FGF receptor gene in order to test whether the positive signal is dependent on expression of the FGF receptor gene. Only one plasmid rescued from one of the egg-library transformants was found to be positive even in the absence of the receptor gene expression. The other genes increased tyrosine phosphorylation only when the FGF receptor gene was co-expressed.

[0097] The DNA sequence of the genes present on these plasmids was determined (Table 1). Two genes encoded peptide factors with putative signal peptide sequences. One gene, designated FRL-2 (FIG. 3, SEQ ID NO: 1), encodes a protein (FIG. 4, SEQ ID NO: 2), with some homology to bovine angiogenin (SEQ ID NO: 5) and Chinese hamster pancreatic ribonuclease A (SEQ ID NO: 6) (about 30% identity; (Maes et al., FEBS Letters 241: 41 (1988); Haugg and Schein, Nucl. Acids Res. 20: 612 (1992)). See FIG. 5. The other gene, FRL-1 (FIG. 6, SEQ ID NO: 3), is homologous to cripto, which is an EGF family member, identified in both mouse and human (about 30% identity; Ciccodicola et al., EMBO J. 8: 1987-1991 (1989); Dono et al., Development 118: 1157 (1993)). The FRL-1 gene product (FIG. 7, SEQ ID NO: 4) is compared to mouse cripto (SEQ ID NO: 7) in FIG. 8. Angiogenin, like FGF, is an angiogenesis-promoting factor. Cripto is suggested to have a role in mesoderm by virtue of its embryonic localized induction. Receptors for angiogenin and cripto have not yet been identified. Based on these findings, FRL-2 and FRL-1 gene products are revealed to be novel ligands of the FGF receptor.

[0098] The XT2 encodes a putative protease homologous to cathepsin L (58% identity with human cathepsin L; Joseph et al., J. Clin. Invest. 81: 1621 (1988); Gal and Gottesman, Biochem. J. 253: 303 (1988)). This protease might cleave the FGF receptor in yeast cells, and the cleaved fragment mi ght have an elevated tyrosine kinase activity. EG1 was previously identified in Xenopus laevis as a heterogeneous ribonucleoprotein (Kay et al., Proc. Natl. Acad. Sci. USA 87: 1367 (1990)). EG3 has an RNA recognition motif found in many RNA binding proteins (Kim and Baker, Mol. Cell. Biol. 13: 174 (1993)). These RNA binding proteins might increase synthesis of FGF receptor protein by increasing the efficiency of transcription or translation. Elevated expression induces autophosphorylation.

[0099] EG4 encodes a novel 96 kDa protein. Recently, a gene similar to EG4 was found in C. elegans (39% identity), but its function is unknown (Wilson et al., Nature 368: 32 (1994)). The plasmid which was positive even in the absence of the FGF receptor gene harbored a gene encoding a putative tyrosine kinase homologous to mouse cytoplasmic tyrosine kinase FER (Hao et al., Mol. Cell. Biol. 9: 1587 (1989)).

[0100] FRL-2 and FRL-1, which have been identified as activators of the FGF receptor in yeast, were tested to determine whether they could also activate the FGF receptor expressed in higher eukaryotic cells. Since it is known that the activation of FGF receptor in Xenopus oocytes is linked to a rapid Ca2+ release from internal stores (Johnson et al., Mol. Cell. Biol. 10: 4728 (1990)), Ca2+ release assays were performed with Xenopus oocytes expressing FGF receptor (FIG. 9).

[0101] As for FRL-1, the FRL-1 protein was partially purified tagged with a flag epitope expressed in yeast. The oocytes expressing FGF receptor were labeled with 45Ca2+ treated with FRL-1, followed by Ca2+ release assay. It was found that Ca release was stimulated by treatment of partially purified FRL-1 protein.

[0102] As for FRL-2, this protein has not been expressed efficiently enough to purify the protein, so instead, FRL-2 mRNA was co-injected with FGF receptor mRNA into oocytes. If FRL-2 protein activates the FGF receptor in oocytes, it is expected that the FGF receptor would be constitutively activated by the continuous synthesis of FRL-2 protein, and that the basal level of Ca2+ efflux in the co-injected oocyte would be higher than in oocytes injected FGF receptor mRNA alone. Ca2+ efflux of labeled oocytes was measured, and it was found that co-injection of FRL-2 and FGF receptor mRNAs increased Ca2+ release two-fold more than the injection of FGF receptor message alone. Co-injection of bFGF and FGF receptor mRNA increased Ca2+ release three-fold. FRL-2 or bFGF mRNA alone did not increase Ca2+ release.

[0103] These results demonstrate that FRL-2 and FRL-1 can activate FGF receptor expressed in Xenopus oocytes, and that these proteins synthesized in vivo can work as activators of FGF receptor.

TABLE 1
Genes Which Increase Protein-Tyrosine Phosphorylation in Yeast
Cells Expressing FGF Receptor.
	FGF receptor		frequency
gene	dependency	gene product	of isolation
1) secreted proteins
FRL-2	+	homologous to angiogenin and	1
		RNaseA
FRL-1	+	cripto (EGF-like growth factor)	4
XT2	+	58% identical to human	1
		cathepsin L
2) RNA binding proteins
EG1	+	heterogeneous ribonucleoprotein	2
EG3	+	RNA binding protein	1
3) a novel protein
EG4	+	novel 96 kd protein	1
4) FGF-receptor independent
EG5	−	cytoplasmic tyrosine kinase TER	1

[0104] The temporal expression patterns of FRL-2 and FRL-1 during embryogenesis both suggest important effects on the embryo. Both ligands induce mesoderm and convergent extention in animal caps. FRL-1 is also able to induce neural tissues in animal caps. FRL-2 is expressed late in development, especially at about stage 27 to about stage 38 of development. (For a description of stages, see Nieucoop and Faber (1994) Normal Table of Xenopus laevis, North-Holland Publishing Co., Amsterdam.) This is indicative of strong effects on the formation of the brain, neural tube and somites. FRL-1 is expressed briefly during gastrulation (stages 9 through 13), suggesting a role in the development of the mesoderm and nervous system. Murine cripto transcripts show a very restricted expression pattern during embryogenesis, first in the epiblastic cells that give rise to the mesoderm, then in the forming mesoderm, and later in the developing heart. Dono, et al. (1993) Development 118:1157-1168.

EQUIVALENTS

[0105] Those skilled in the art will recognize, or be able to ascertain using no more than routine experimentation, many equivalents to the specific embodiments of the invention described specifically herein. Such equivalents are intended to be encompassed in the scope of the following claims.

Claims

1. A method for identifying an FGF receptor ligand comprising: a) providing a cDNA expression library from an organism of interest; b) providing a population of cells that express a DNA sequence encoding a heterologous FGF receptor that comprises an extracellular domain, a transmembrane domain, and an intracellular domain that is characterized by protein tyrosine activity; c) transforming the population of cells with the cDNA expression library; and d) detecting protein kinase activity in clonally-derived cells, wherein elevated FGF receptor tyrosine kinase activity indicates the presence of an FGF receptor ligand.

2. The method according to claim 1, wherein the cell is a yeast cell.

3. The method according to claim 1, wherein the DNA sequence encoding a heterologous FGF receptor is carried on a CEN-based plasmid.

4. The method according to claim 1, wherein the DNA sequence encoding a heterologous FGF receptor is inserted into a chromosome.

5. The method according to claim 1, wherein the DNA sequence encoding a heterologous FGF receptor is constitutively expressed.

6. The method according to claim 1, wherein the cDNA expression library having polynucleotide inserts under the control of an inducible promoter.

7. An FGF receptor ligand identified by the method of claim 1.

8. A method for identifying an FGF receptor ligand comprising: a) providing a cDNA expression library from an organism of interest under the control of an inducible promoter; b) providing a population of cells that express a DNA sequence encoding a heterologous FGF receptor that comprises an extracellular domain, a transmembrane domain, and an intracellular domain that is characterized by protein tyrosine activity; c) transforming the population of cells with the cDNA expression library; and d) detecting protein kinase activity in clonally-derived cells, wherein elevated FGF receptor tyrosine kinase activity indicates the presence of an FGF receptor ligand.

9. The method according to claim 8, wherein the cell is a yeast cell.

10. The method according to claim 8, wherein the DNA sequence encoding a heterologous FGF receptor is carried on a CEN-based plasmid.

11. The method according to claim 8, wherein the DNA sequence encoding a heterologous FGF receptor is inserted into a chromosome.

12. The method according to claim 8, wherein the DNA sequence encoding a heterologous FGF receptor is constitutively expressed.

13. An FGF receptor ligand identified by the method of claim 8.

14. A method for identifying an FGF receptor ligand comprising: a) providing a cDNA expression library from an organism of interest; b) providing a population of cells that constitutively express a DNA sequence encoding a heterologous FGF receptor that comprises an extracellular domain, a transmembrane domain, and an intracellular domain that is characterized by protein tyrosine activity; c) transforming the population of cells with the cDNA expression library; and d) detecting protein kinase activity in clonally-derived cells, wherein elevated FGF receptor tyrosine kinase activity indicates the presence of an FGF receptor ligand.

15. The method according to claim 14, wherein the cell is a yeast cell.

16. The method according to claim 14, wherein the DNA sequence encoding a heterologous FGF receptor is carried on a CEN-based plasmid.

17. The method according to claim 14, wherein the DNA sequence encoding a heterologous FGF receptor is inserted into a chromosome.

18. The method according to claim 14, wherein the cDNA expression library having polynucleotide inserts under the control of an inducible promoter.

19. An FGF receptor ligand identified by the method of claim 14.

20. An oligonucleotide probe or primer that hybridizes under stringent hybridization conditions to a polynucleotide selected from the group consisting of SEQ ID NO: 1; the complement of SEQ ID NO: 1; SEQ ID NO: 3; and the complement of SEQ ID NO: 3.

21. A pharmaceutical compound for treating or preventing a disorder in a vertebrate, the compound comprising a therapeutically effective amount of an isolated polypeptide comprising the amino acid sequence of SEQ ID NO: 2 or SEQ ID NO: 4.

22. A pharmaceutical composition for treating or preventing a disorder in a vertebrate, the composition comprising a therapeutically effective amount of a polypeptide encoded by an isolated DNA selected from the group consisting of: a) SEQ ID NO: 1 or SEQ ID NO: 3, and b) a nucleotide sequence that hybridizes to SEQ ID NO: 1 or SEQ ID NO: 3, under stringent conditions; and a pharmaceutically acceptable diluent, adjuvant or carrier.

23. A method for treating a neural disorder in a vertebrate, comprising administering a therapeutically effective amount of an isolated polypeptide encoded by DNA selected from the group consisting of: a) SEQ ID NO: 1 or SEQ ID NO: 3, and b) a nucleotide sequence that hybridizes to SEQ ID NO: 1 or SEQ ID NO: 3, under stringent conditions.

24. A method for treating a neural disorder in a vertebrate, comprising administering a therapeutically effective amount of an isolated polypeptide of SEQ ID NO: 2 or SEQ ID NO: 4.

25. A method of stimulating proliferation of vertebrate cells comprising administering a therapeutically effective amount of a polypeptide encoded by DNA selected from the group consisting of: a) SEQ ID NO: 1 or SEQ ID NO: 3, and b) a nucleotide sequence that hybridizes to SEQ ID NO: 1 or SEQ ID NO: 3, under stringent conditions.

26. A method of stimulating proliferation of vertebrate cells comprising administering a therapeutically effective amount of an isolated polypeptide of SEQ ID NO: 2 or SEQ ID NO: 4.

27. A method of inhibiting tumor growth in a vertebrate comprising administering an antagonist of a polypeptide encoded by DNA selected from the group consisting of: a) SEQ ID NO: 1 or SEQ ID NO: 3, and b) a nucleotide sequence that hybridizes to SEQ ID NO: 1 or SEQ ID NO: 3, under stringent conditions.

28. The method of claim 27 wherein the antagonist comprises an antibody that binds SEQ ID NO: 2 or SEQ ID NO: 4.

29. The method of claim 27 wherein the antagonist is selected from the group consisting of SEQ ID NO: 2 and SEQ ID NO: 4, altered to produce a polypeptide that interferes with the binding of SEQ ID NO: 2 or SEQ ID NO: 4 to an FGF receptor.

30. A method of inhibiting tumor growth in a vertebrate, comprising contacting an antagonist of a polypeptide encoded by DNA selected from the group consisting of: a) SEQ ID NO: 1 or SEQ ID NO: 3, and b) a nucleotide sequence that hybridizes to SEQ ID NO: 1 or SEQ ID NO: 3, under stringent conditions.

31. The method of claim 27, wherein the antagonist comprises an antibody that binds to a portion of the amino acid sequence of SEQ ID NO: 2 or SEQ ID NO: 4.

32. The method of claim 27, wherein the antagonist is selected from the group consisting of SEQ ID NO: 2 and SEQ ID NO: 4, altered to produce a polypeptide that interferes with the binding of SEQ ID NO: 2 or SEQ ID NO: 4 to an FGF receptor.

33. A method of inhibiting tumor growth in a vertebrate, comprising administering a therapeutically effective amount of a polypeptide encoded by DNA selected from the group consisting of: a) SEQ ID NO: 1 or SEQ ID NO: 3, and b) a nucleotide sequence that hybridizes to SEQ ID NO: 1 or SEQ ID NO: 3, under stringent conditions.

34. A method of inhibiting tumor growth in a vertebrate comprising administering a therapeutically effective amount of an isolated polypeptide of SEQ ID NO: 2 or SEQ ID NO: 4.

35. An antibody or antibody fragment which binds a polypeptide encoded by DNA selected from the group consisting of: a) SEQ ID NO: 1 or SEQ ID NO: 3, and b) a nucleotide sequence that hybridizes to SEQ ID NO: 1 or SEQ ID NO: 3, under stringent conditions.

36. An antibody of claim 35 which is a polyclonal antibody.

37. An antibody of claim 35 which is a monoclonal antibody.

38. A method for detecting the expression of an FRL-2 or FRL-1 protein ligand in a sample comprising the steps of: (a) treating the sample in a manner that renders RNA encoding the ligand available for hybridization with a complementary DNA or RNA oligonucleotide, thereby producing a treated sample; (b) contacting the treated sample with at least one DNA or RNA probe which is a nucleotide sequence complementary to all or a portion of the gene or mRNA encoding the ligand; and (c) detecting the hybridization of mRNA from the sample with the probe, wherein hybridization is an indication of the presence of the ligand in the sample.

39. A method for detecting the expression of a protein ligand selected from the group consisting of SEQ ID NO: 1, or 3, and a polynucleotide that encodes for SEQ ID NO: 2 or 4, in a sample comprising the steps of: (a) treating the sample in a manner that renders RNA encoding the ligand available for hybridization with a complementary DNA or RNA oligonucleotide, thereby producing a treated sample; (b) contacting the treated sample with at least one DNA or RNA probe which is a nucleotide sequence complementary to all or a portion of the gene or mRNA encoding the ligand; and (c) detecting the hybridization of mRNA from the sample with the probe, wherein hybridization is an indication of the presence of the ligand in the sample.

40. A method according to claim 39, further comprising quantifying the ligand in the sample by measuring the extent of hybridization.

41. A method of detecting the level of expression of a protein ligand selected from the group consisting of SEQ ID NO: 2, or 4, and a polypeptide encoded by SEQ ID NO: 1 or 3, in a sample comprising the steps of: (a) treating the sample in a manner that renders the ligand available for binding to antibodies or antibody fragments specific for the ligand, thereby producing a treated sample; (b) contacting the treated sample with the antibody or antibody fragments under conditions appropriate for formation of antibody-antigen complexes; and (c) detecting the presence of antibody-antigen complexes as an indication of the presence of the ligand in the sample.

42. A method of quantifying the level of expression of a protein ligand selected from the group consisting of SEQ ID NO: 2, or 4, and a polypeptide encoded by SEQ ID NO: 1 or 3, in a sample of vertebrate cells or tissues, comprising the steps of: (a) treating the sample in a manner that renders the ligand available for binding to an antibody or antibody fragment specific for the ligand, thereby producing a treated sample; (b) contacting the treated sample with the antibody or antibody fragment under conditions appropriate for formation of antibody-antigen complexes; and (c) detecting the amount of antibody-antigen complexes as an indication of the amount of the ligand in the sample.

43. A chimeric fusion protein comprising an extracellular domain of a transmembrane receptor fused to an intracellular domain of a tyrosine kinase receptor, the intracellular domain of the chimeric fusion protein being characterized by an ability to activate a heterologous cytoplasmic tyrosine kinase activity in response to ligand binding to the extracellular domain of the transmembrane receptor.

44. The chimeric fusion protein of claim 43, wherein the transmembrane receptor is a cytokine receptor.

45. The chimeric fusion protein of claim 43, wherein the transmembrane receptor is a tyrosine phosphatase receptor.

46. The chimeric fusion protein of claim 43, wherein the chimeric fusion protein is carried on a CEN-based plasmid.

47. The chimeric fusion protein of claim 43, wherein the chimeric fusion protein is inserted in a yeast chromosome.

48. An FGF receptor ligand identified by the method of: (a) transforming a cell with an expression library to thereby produce at least one clonally-derived cell, wherein the cell expresses a heterologous FGF receptor that comprises an extracellular domain, a transmembrane domain, and an intracellular domain that is characterized by protein tyrosine activity, and wherein the expression library comprises an expression vector construct that comprises a signal sequence and a cDNA sequence under the control of a promoter sequence; (b) detecting FGF receptor tyrosine kinase activity in the transformed clonally-derived cell, wherein elevated FGF receptor tyrosine kinase activity indicates the presence of an FGF receptor ligand.

49. The FGF receptor ligand of claim 48, wherein the cell is a yeast cell.