RECEPTOR TYROSINE KINASE WITH A DISCOIDIN-TYPE BINDING DOMAIN

Abstract

A breast carcinoma tyrosine phosphoprotein, DDR (Discoidin Domain Receptor), that defines a novel class of receptor tyrosine kinases is presented. The DDR cDNA predicts a polypeptide C-terminal tyrosine kinase domain and an N-terminal domain similar to the Dictyostelium discoideum lectin discoidin I. These domains are connected by an extraordinary hydrophilic proline/glycine-rich domain, which is interrupted by a predicted transmembrane sequence. This extended proline/glycine-rich region suggests an unusual geometry of interaction with ligand or substrates. Discoidin I-type domains may interact with specific cell surface molecules.

Description

BACKGROUND OF THE INVENTION

[0002] 1. Field of the Invention

[0003] The present invention relates to receptor tyrosine kinases with discoidin-type or like binding domains and nucleic acids encoding such receptor tyrosine kinases. It is noted that the “type” or “like” term that is associated with the names discoidin, discoidin I, and equivalent molecules in this disclosure implies that structural and functional similarities exist between the subject molecule's binding domain and the binding domain found in discoidin, discoidin I, and equivalent molecules.

[0004] 2. Description of the Background Art

[0005] Certain extracellular molecules influence cellular growth, differentiation or development via activation of receptor tyrosine kinases (Bishop, J. M. (1991) Cell 64, 235-248 and Hunter, T. (1991) Meth. Enz. 200, 3-37, which are incorporated herein by reference). Receptor tyrosine kinases are composed of an extracellular ligand binding domain and a cytoplasmic catalytic domain that allow for specific decoding of extracellular signals and initiation of intracellular biochemical effects. A large number of cDNAs for putative receptor tyrosine kinases have been cloned and sequenced (Hanks, S. K. & Quinn, A. M. (1991) Meth. Enz. 200, 38-62, which is incorporated herein by reference). With rare exception (Shier, P. & Watt, V. M. (1989) J. Biol. Chem. 264, 14605-14608, which is incorporated herein by reference), sequence similarities in extracellular domains have not suggested the type of ligand that might activate a given receptor. Nevertheless, the diversity of receptor tyrosine kinase extracellular domains implies that there are a large number of as yet unidentified extracellular molecules that can regulate receptor tyrosine kinase activity and, by extension, growth, differentiation or development. Identification of these ligands is crucial for a comprehensive understanding of these fundamental biological processes. It is clear that researchers are only beginning to understand the complexity of the extracellular signals that can activate receptor tyrosine kinases. Also, little is known about the fundamental process of information exchange across the membrane to initiate the intracellular cascade of biochemical effectors. Disclosed here is the cDNA cloning and characterization of a novel breast carcinoma cell protein with a primary structure which suggests that this protein is a receptor tyrosine kinase with an unusual mechanism of transmembrane signaling. An extracellular domain of this protein is similar to the lectin-like or type protein or specifically the carbohydrate-binding protein discoidin I.

SUMMARY OF THE INVENTION

[0006] An object of the present invention is to present a polypeptide having both a first domain with carbohydrate or discoidin I-type binding activity and a second domain with tyrosine kinase activity.

[0007] Another object of the present invention is to disclose a nucleic acid sequence that encodes for a polypeptide having a first domain with carbohydrate or discoidin I-type binding activity and a second domain with tyrosine kinase activity.

[0008] A further object of the present invention is to make known a marker that may be utilized to predict and diagnose tumors of the breast and lung.

[0009] Still another object of the present invention is to describe a nucleic acid sequence that encodes a polypeptide having both discoidin I-type ligand binding characteristics and tyrosine kinase activity that can be utilized in immunological and inhibitory regulation of associated tumors.

[0010] Yet a further object of the present invention is to disclose a polypeptide having both discoidin I-type ligand binding characteristics and tyrosine kinase activity that can be utilized in immunological and inhibitory regulation of associated tumors.

[0011] Disclosed is a novel breast carcinoma tyrosine phosphoprotein, DDR (Discoidin Domain Receptor), that defines a novel class of receptor tyrosine kinases. The DDR cDNA predicts a C-terminal tyrosine kinase domain and an N-terminal domain similar to the Dictyostelium discoideum lectin named discoidin or more specifically discoidin I. These domains are connected by an extraordinary hydrophilic proline/glycine-rich domain, which is interrupted by a predicted transmembrane sequence. This extended proline/glycine-rich region suggests an unusual geometry of interaction with ligand or substrates. Discoidin I-type domains are also found in other proteins, including coagulation factors V and VIII. Discoidin I-type domains may interact with specific cell surface molecules. SEQ ID NO: 1 designates the nucleotide sequence of the present invention, SEQ ID NO: 2 designates the polypepteide or protein sequence of the present invention, and SEQ ID NO: 3 designates the probe utilized in the subject invention.

[0012] Other objects, advantages, and novel features of the present invention will become apparent from the detailed description that follows, when considered in conjunction with the associated drawings.

DESCRIPTION OF THE FIGURES

[0013] FIGS. 1 a -h (also, see SEQ ID NOS: 1 and 2 show the nucleotide and deduced amino acid sequence of the human DDR cDNA. Amino acids are numbered for the precursor protein. The single underlined amino acid sequence near the N-terminus contains the discoidin I-type domain and the single underlined amino acid sequence near the C-terminus contains the tyrosine kinase domain. The predicted signal peptide (1-18) and transmembrane (418-440) domains are in bold type. Uncertainty, as to the point of signal peptide cleavage, exists between amino acids 19 to 24. Potential N-glycosylation sites are bolded with underlining and cysteines within the extracellular region are marked with an asterisk. The proline and glycine residues between the discoidin I-type domain and the tyrosine kinase domain are italicized. The bolded with double underlining show the most proline/glycine-rich of the connecting region. The tyrosine within the Asn-Pro-Xaa-Tyr sequence found in the cytoplasmic juxtamembrane sequences of several plasma membrane receptors (Bansal, A. & Gierasch, L. M. (1991) Cell 67, 1195-201, which is incorporated herein by reference) is denoted with a double asterisk.

[0014] FIG. 2 is a schematic representation of DDR. DLD=discoidin I-type domain; TMD=transmembrane domain; TKD=tyrosine kinase domain. The kinked line represents the extent of the most proline/glycine-rich portion of the connecting region. The approximate boundaries of the peptide fused to b-galactosidase for rabbit immunization are indicated by #. The approximate positions of potential sites of N-glycosylation are denoted by asterisks.

[0015] FIG. 3 shows a Northern analysis of DDR transcripts in human cell line total RNAs. Total RNAs (20 micrograms) from human cell lines a BeWo, b T-47D, c MCF-7, d PANC-1, e A431, f U937, g Daudi, h HL-60, i Jurkat, j C32, k HepG2, l HeLa were fractionated on formaldehyde agarose gels, transferred to a nitrocellulose filter and hybridized with a 0.5 kilobase 32P-labeled PCR product encompassing nucleotides 213 to 772 of the DDR cDNA in 50% formamide at 42° C. After washing and exposure to film for 3 days, the filter was subsequently probed with a 32P-labeled mouse GAPDH (glyceraldehyde 3-phosphate dehydrogenase) cDNA to confirm that a similar amount of RNA was loaded in each lane. The migrations of the DDR and GAPDH transcripts as well as the ribosomal RNAs are indicated.

[0016] FIG. 4 shows a Northern analysis of DDR transcripts in mouse tissue RNAs. PolyA+ RNAs from adult mouse tissues: a brain, b thymus, c lung, d heart, e liver, f spleen, g small intestine, h kidney, i pancreas, j skeletal muscle, k testis, l ovary, m uterus, n placenta were fractionated on formaldehyde agarose gels, transferred to a nitrocellulose filter and hybridized with a 3.7 kilobase 32P-labeled EcoR I fragment containing the DDR cDNA in 40% formamide at 42° C. After washing and exposure to film for 3 days, the filter was subsequently probed with a 32P-labeled mouse GAPDH cDNA to control for the amount of RNA loaded in each lane. The mobilities of RNA markers is indicated in kilobases. The migrations of the DDR and GAPDH (glyceraldehyde 3-phosphate dehydrogenase) transcripts are indicated.

[0017] FIG. 5 shows DDR polypeptide and tyrosine phosphorylation in transfected COS-7 cells and T-47D breast carcinoma cells. Wheat germ agglutinin binding extracts of untransfected COS-7 cells, COS-7 cells transfected with the DDR cDNA and T-47D cells were fractionated on 7.5% NaDodSO4-PAGE for immunoblot analysis with anti-DDR antiserum e.1 as described below in the Examples section. The filter was stripped of antibodies as described below in the Examples section, incubated with detection reagents to ensure stripping and then tyrosine phosphoproteins were detected with antibody PY20. In other experiments two tyrosine phosphoprotein bands were observed: one the size of DDR and the other at approximately 180 kDa. The mobilities of prestained protein markers are shown at the right.

DESCRIPTION OF THE PREFERRED EMBODIMENT

[0018] The DDR cDNA was isolated during a search for tyrosine kinase cDNAs related to those in the insulin receptor family. This cDNA predicts a protein of 914 amino acids (105 kDa) that has a signal peptide and a transmembrane domain (FIGS. 1 and 2). The predicted DDR protein also has the following features: a discoidin I-type domain near the N-terminus, an extensive proline/glycine-rich region between the discoidin I-type domain and the transmembrane domain, and another extensive proline/glycine-rich region between the transmembrane domain and the C-terminal tyrosine kinase domain. The sequence of the DDR catalytic domain places it within the insulin receptor family of receptor tyrosine kinases (Hanks, S. K. & Quinn, A. M. (1991) Meth. Enz. 200, 38-62, which is incorporated herein by reference). The catalytic domain is 45% identical with the trkA protein catalytic domain (Martin-Zanca, D., Oskam, R., Mitra, G., Copeland, T. & Barbacid, M. (1989) Mol. Cell. Biol. 9, 24-33, which is incorporated herein by reference) but the remainder of the molecule has no similarity to any other portion of the trk proteins. Also like trkA, the DDR protein has a relatively short C-terminal tail following the catalytic domain (8 amino acids in DDR versus 13 in trkA). The C-terminal tail of DDR does not contain tyrosine residues, but the tyrosine residues conserved within the catalytic domains of trkA and the insulin receptor are also conserved in DDR. This includes DDR tyrosines 793, 797, and 798, which by analogy with the insulin receptor are autophosphorylation sites. The exceptionally long cytoplasmic juxtamembrane region contains an Asn-Pro-Ala-Tyr sequence characteristic of the tight turn recognition motif for internalization in coated pits (Chen, W. J., Goldstein, J. L. & Brown, M. S. (1990) J Biol. Chem. 265, 3116-3123 and Bansal, A. & Gierasch, L. M. (1991) Cell 67, 1195-201, which are incorporated herein by reference).

[0019] Northern analysis of multiple human cell lines demonstrated that a 4.0 kilobase DDR transcript is relatively abundant in the human breast carcinoma cell lines T-47D, BT-20 and MCF-7 and also relatively high in the A431 epidermoid carcinoma cell line (FIG. 3). Transcripts hybridizing with the DDR cDNA were found in polyA+ RNAs from multiple mouse tissues, but in widely varying amounts (FIG. 4). Kidney, spleen and placenta had the highest levels of the 4.0 kilobase transcript relative to the levels of GAPDH (glyceraldehyde 3-phosphate dehydrogenase) mRNA in each preparation.

[0020] To characterize the DDR polypeptide, we transfected COS-7 cells with the DDR cDNA in a mammalian expression vector. Wheat germ agglutinin-binding extracts of transfected cells contained a 120 kDa DDR protein that was reactive with antisera developed against a lacZ fusion protein containing a portion of the extracellular domain of the DDR-encoded polypeptide (FIG. 5, anti-DDR). The DDR protein was also specifically reactive with an antiphosphotyrosine antibody presumably due to autophosphorylation (FIG. 5, anti-PY). The DDR protein was also detected in T47D and BT-20 breast carcinoma cell lines, but was not detected in a variety of other human cell lines (FIG. 5 and data not shown). The major tyrosine phosphoprotein present in wheat germ agglutinin-binding extracts of T-47D cells had an identical electrophoretic mobility to the DDR protein (FIG. 5).

[0021] We have characterized a novel putative receptor tyrosine kinase, DDR, that is abundant in breast carcinoma cell lines. DDR has at least two unusual features, the discoidin I-type domain and the extensive proline/glycine-rich regions, not present in other receptor tyrosine kinases. These features suggest that DDR may have an unusual mechanism of transmembrane signaling or an unusual ligand.

[0022] The presence of a discoidin I-type domain in the ectodomain of a receptor tyrosine kinase is provocative. Discoidin I is a Dictyostelium discoideum lectin that participates in cell aggregation (Springer, W. R., Cooper, D. N. W. & Barondes, S. H. (1984) Cell 39, 557-564, which is incorporated herein by reference). Discoidin I-type domains (Poole, S., Firtel, R. A., Lamar, E. & Rowenkamp, W. (1981) J. Mol. Biol. 153, 273-289, which is incorporated herein by reference) are present as tandem repeats at the C-terminus of the light chains of factor V (Kane, W. H. & Davie, E. W. (1986) Proc. Natl. Acad. Sci. U.S.A. 83, 6800-6804, which is incorporated herein by reference) factor VIII (Toole, J. J., Knopf, J. L., Wozney, J. M., Sultzman, L. A., Buecker, J. L., Pittman, D. D., Kaufman, R. J., Brown, E., Shoemaker, C., Orr, E. C., Amphlett, G. W., Foster, W. B., Coe, M. L., Knutson, G. J., Fass, D. N. & Hewick, R. M. (1984) Nature 312, 342-347 and Vehar, G. A., Keyt, B., Eaton, D., Rodriguez, H., O'Brien, D. P., Rotblat, F., Oppermann, H., Keck, R., Wood, W. I., Harkins, R. N., Tuddenham, E. G. D., Lawn, R. M. & Capon, D. J. (1984) Nature 312, 337-342, which are incorporated herein by reference) and two milk fat globule membrane proteins, MFG.E8 (Stubbs, J. D., Lekutis, C., Singer, K. L., Bui, A., Yuzuki, D., Srinivasan, U. & Parry, G. (1990) Proc. Natl. Acad. Sci. USA 87, 8417-8421, which is incorporated herein by reference) and BA46 (Larocca, D., Peterson, J. A., Urrea, R., Kuniyoshi, J., Bistrain, A. M. & Ceriani, R. L. (1991) Cancer Research 51, 4994-4998, which is incorporated herein by reference). The role of the discoidin I-type domains in these proteins is not completely understood, but there is evidence suggesting that the light chains of factor V and VIII interact with specific platelet membrane proteins (Tracey, P. B., Peterson, J. M., Nesheim, M. E., McDuffie, F. C. & Mann, K. G. (1979) J. Biol. Chem. 254, 10354-10361, Tracey, P. B., Nesheim, M. E. & Mann, K. G. (1980) J. Biol. Chem. 255, 662-669, and Nesheim, M., Pittman, D. D., Wang, J. H., Slonosky, D., Giles, A. R. & Kaufman, R. J. (1988) J. Biol. Chem. 263, 16467-16470, which are incorporated herein by reference), and MFG.E8 and BA46 are stably associated with mammary epithelial membranes (Stubbs, J. D., Lekutis, C., Singer, K. L., Bui, A., Yuzuki, D., Srinivasan, U. & Parry, G. (1990) Proc. Natl. Acad. Sci. USA 87, 8417-8421 and Larocca, D., Peterson, J. A., Urrea, R., Kuniyoshi, J., Bistrain, A. M. & Ceriani, R. L. (1991) Cancer Research 51, 4994-4998, which are incorporated herein by reference). Recently, tandem discoidin I-type domains have also been found in the extracellular region of a cell surface transmembrane protein, A5 (Takagi, S., Hirata, T., Agata, K., Mochii, M., Eguchi, G. & Fujisawa, H. (1991) Neuron 7, 295-307, which is incorporated herein by reference), which has a small (44 amino acids) cytoplasmic region. Since this protein has a highly specific localization at the termination site of the optic nerve and is expressed contemporaneously with optic nerve innervation, it has been proposed that A5 is a targeting molecule for retinal axons (Takagi, S., Tsuji, T., Amagai, T. & Fujisawa, H. (1987) Dev. Biol. 122, 90-100, which is incorporated herein by reference). A highly conserved consensus sequence can be derived from an alignment of the discoidin I-type domains in these proteins (Poole, S., Firtel, R. A., Lamar, E. & Rowenkamp, W. (1981) J. Mol. Biol. 153, 273-289, Larocca, D., Peterson, J. A., Urrea, R., Kuniyoshi, J., Bistrain, A. M. & Ceriani, R. L. (1991) Cancer Research 51, 4994-4998, Stubbs, J. D., Lekutis, C., Singer, K. L., Bui, A., Yuzuki, D., Srinivasan, U. & Parry, G. (1990) Proc. Natl. Acad. Sci. USA 87, 8417-8421, Kane, W. H. & Davie, E. W. (1986) Proc. Natl. Acad. Sci. U.S.A. 83, 6800-6804, Vehar, G. A., Keyt, B., Eaton, D., Rodriguez, H., O'Brien, D. P., Rotblat, F., Oppermann, H., Keck, R., Wood, W. I., Harkins, R. N., Tuddenham, E. G. D., Lawn, R. M. & Capon, D. J. (1984) Nature 312, 337-342, and Takagi, S., Hirata, T., Agata, K., Mochii, M., Eguchi, G. & Fujisawa, H. (1991) Neuron 7, 295-307). The mammalian members of the group also have similarities outside the region of similarity with discoidin I. The DDR discoidin I-type domain adheres closely to a consensus sequence defined by the other discoidin domain containing proteins. Analogy with the other discoidin I-type domain containing proteins suggests that the DDR ligand may be a cell surface molecule. Since DDR is relatively abundant in breast carcinoma cells, it is particularly intriguing that mammary epithelial membranes appear to express a component that binds BA46 (Larocca, D., Peterson, J. A., Urrea, R., Kuniyoshi, J., Bistrain, A. M. & Ceriani, R. L. (1991) Cancer Research 51, 4994-4998, which is incorporated herein by reference). Characterization of the DDR ligand could reveal a set of interactions involved in multiple biological regulatory systems.

[0023] The two juxtamembrane regions of DDR suggest a unique signaling mechanism for DDR. Out of 180 residues in the regions indicated in FIGS. 1 and 2, 46 are proline and 19 are glycine (36% P+G). The region between the discoidin I-type domain and this extremely proline/glycine-rich region is also glycine-rich; therefore the 250 amino acid connecting region between the discoidin I-type domain and the tyrosine kinase domain, excluding the transmembrane and stop-transfer regions, is 25% proline/glycine. The extensive hydrophilic proline/glycine-rich region of DDR does not contain collagen-like repeating motifs, nor does it contain sequence patterns characteristic of the family of salivary proline-rich proteins. The 176 residue cytoplasmic juxtamembrane region is the longest described for any receptor tyrosine kinase (over three times longer than the average juxtamembrane span in the insulin receptor family of tyrosine kinases). Proline/glycine-rich regions like the one found in DDR are potentially flexible, but the physical properties of this region remain to be experimentally determined. A proline/glycine-rich region of the type found in DDR has not previously been observed in an integral membrane protein. However, regions of similar composition and length are found within the adaptins, which tether transmembrane proteins (e.g. receptors) to clathrin-coated pits and vesicles (Robinson, M. S. (1989) J. Cell Biol. 108, 833-842, Ponnambalam, S. Robinson, M. S., Jackson, A. P., Peiper, L. & Parham, P. (1990) J. Biol. Chem. 265, 4814-4820, Kirchhausen, T., Nathanson, K. L., Matsui, W., Vaisberg, A., Chow, E. P., Burne, C., Keen, J. H. & Davis, A. E. (1989) Proc. Natl. Acad. Sci. USA 86, 2612-2616, Robinson, M. S. (1990) J. Cell Biol. 111, 2319-2326, Virshup, D. M. & Bennett, V. (1988) J. Cell Biol. 106, 39-50, and Heuser, J. E. & Keen, J. (1988) J. Cell Biol. 107, 877-886, which are incorporated herein by reference). The corresponding regions of a (Robinson, M. S. (1989) J. Cell Biol. 108, 833-842), b (Ponnambalam, S. Robinson, M. S., Jackson, A. P., Peiper, L. & Parham, P. (1990) J. Biol. Chem. 265, 4814-4820 and Kirchhausen, T., Nathanson, K. L., Matsui, W., Vaisberg, A., Chow, E. P., Burne, C., Keen, J. H. & Davis, A. E. (1989) Proc. Natl. Acad. Sci. USA 86, 2612-2616, which are incorporated by reference) and d-adaptins (Robinson, M. S. (1990) J. Cell Biol. 111, 2319-2326, which is incorporated herein by reference) are 34%, 26% and 27% proline+glycine over 96, 152, and 125 residues, respectively. Electron microscopic examination of adaptins demonstrates two globular domains connected by an extended hinge (Virshup, D. M. & Bennett, V. (1988) J. Cell Biol. 106, 39-50 and Heuser, J. E. & Keen, J. (1988) J. Cell Biol. 107, 877-886, which are incorporated herein by reference). The DDR proline/glycine-rich region does not contain sequence patterns similar to the adaptin hinge regions. The a and b adaptins, however, also do not have sequence motifs conserved between their respective hinge regions; it appears that overall amino acid composition determines hinge formation rather than a specific sequence pattern (Takagi, S., Tsuji, T., Amagai, T. & Fujisawa, H. (1987) Dev. Biol. 122, 90-100, which is incorporated herein by reference). The adaptin hinges appear to be required to bridge the physical gap between clathrin and plasma membrane proteins in coated pits or vesicles (Ponnambalam, S. Robinson, M. S., Jackson, A. P., Peiper, L. & Parham, P. (1990) J. Biol. Chem. 265, 4814-4820, Kirchhausen, T., Nathanson, K. L., Matsui, W., Vaisberg, A., Chow, E. P., Burne, C., Keen, J. H. & Davis, A. E. (1989) Proc. Natl. Acad. Sci. USA 86, 2612-2616, Robinson, M. S. (1990) J. Cell Biol. 111, 2319-2326, Virshup, D. M. & Bennett, V. (1988) J. Cell Biol. 106, 39-50, and Heuser, J. E. & Keen, J. (1988) J. Cell Biol. 107, 877-886, which are incorporated herein by reference). The proline/glycine-rich regions of the DDR may serve an analogous tethering function and allow a unique geometry in interaction with the ligand or allow access of the tyrosine kinase domain to a unique set of substrates.

[0024] Elucidation of the physical characteristics and ligand binding properties of DDR should provide insight into a unique transmembrane signaling process.

[0025] Several important uses of the subject invention are disclosed. First, the DDR cDNA or antibody reagents generated from the sequence may be useful for diagnostic or prognostic analysis of tumors of the breast and lung. Antibodies to the extracellular domain may be useful for screening blood or other tissue samples.

[0026] Second, the recombinant DDR extracellular domain should allow purification and characterization of the ligand. Identification of the ligand is important for designing therapeutic agents which might act via the DDR to influence the behavior of breast and lung tumors.

[0027] Third, the recombinant DDR extracellular domain is a potential immunogen for active immunotherapy of breast and lung cancer.

[0028] Fourth, the recombinant DDR extracellular domain may block the normal function of the receptor in vivo by occupying the ligand. This could inhibit the growth of tumors that require growth signals from activated DDR.

[0029] Clarifying and expanding on the above possible uses, the primary sequence of the DDR cDNA predicts a transmembrane protein that contains an N-terminal discoidin I-like domain, a membrane spanning region and a C-terminal tyrosine kinase domain. The presence of a tyrosine kinase domain suggests that the DDR protein is involved in transmembrane control of the growth status of the cell. The presence of an extracellular discoidin I-type domain, which is unique to DDR among receptor tyrosine kinases, suggests that DDR interacts with a ligand found on cell surfaces or within the extracellular matrix. Since discoidin I is a lectin, it is possible that the DDR ligand will have a carbohydrate component.

[0030] The DDR cDNA has also provided the means to define the nature of the ligand. Recombinant forms of the subject DDR cDNA have been used to obtain protein reagents corresponding to fragments or rearrangements of the predicted protein. The N-terminal portion of the DDR cDNA has been truncated to encode an extracellular domain fragment corresponding to amino acids 1-387 of the predicted protein sequence, which includes the signal peptide. the DLD and a portion of the juxtamembrane region, to which specific antiserum has been obtained. Identification of the ligand involves development of binding assays to determine the localization of the ligand within tissues and to define the molecular determinants of DDR ligand-receptor interactions. This type of binding assay will permit determining the exact carbohydrate binding determinants.

[0031] It is clear that DDR itself is present in several different types of human tumor cell lines that originated in specific types of epithelia. If the DDR ligand is present within such tumors, then blocking its interaction with DDR may influence the survival of the tumor. Soluble DDR derived reagents may be able to directly block the DDR-ligand interactions and could be used to screen compounds for the ability to block the receptor-ligand interaction. For example, if there is a carbohydrate component of interaction of the ligand with DDR, specific carbohydrates may be able to block growth stimuli normally or abnormally mediated by the DDR polypeptide.

[0032] A portion of the DDR cDNA has been used to aid in generating a polyclonal antibody specific for the extracellular domain of the DDR polypeptide. This and other antibodies which are generated as a result of the DDR cDNA sequences could have prognostic value in the analysis of specific tumors of epithelial origin. Since the extracellular domain of receptors are often shed from the cell surface and can be subsequently found in serum samples, antibodies such as the one generated here could be useful for screening blood samples for the presence of the DDR extracellular domain. If the DDR molecule is found to have prognostic value for specific tumors, the ability assay for its presence in blood samples could have value for non-invasive screening and diagnosis.

[0033] The DNA sequence disclosed herein has been assigned Genbank accession number L11315.

EXAMPLES

Example 1

cDNA Cloning and Characterization

[0034] A full term human placental Igt10 cDNA library was screened with a 32P-labeled antisense oligonucleotide of the sequence 5′-GTT(G/C)CG(A/G)GC(A/G)GCCAG(A/G)TC-(G/C)CG(A/G)TG-3′ (SEQ ID NO: 3), corresponding to the His-Arg-Asp-Leu-Ala-Ala-Arg-Asn amino acid sequence found in many tyrosine kinases, by methods described previously (Fischman, K., Edman, J. C., Shacklefor, G. M., Turner, J. A., Rutter, W. J. & Nir, U. (1990) Mol. Cell. Biol. 10, 146-153, which is incorporated herein by reference). Positive clones were grouped by cross-hybridization and plaque isolated for subcloning in M13 derivatives or plasmids for sequencing with Sequenase (USB) and an ABI 370A automated sequencer. Three individual clones from a class of 19 had identical nucleotide sequence tags and overlapping restriction maps. Both strands of the DDR cDNA in one of these were sequenced in their entirety. This sequence has one long open reading frame (nucleotides 88 to 2883) which is followed by a polyadenylation signal (at nucleotide 3721) and a polyadenine tract (SEQ ID NO: 1). The first in-frame methionine (at nucleotide 142) has a Kozak consensus sequence (Kozak, M. (1987) Nucleic Acids Res. 15, 8125-8148, which is incorporated herein by reference) followed by a predicted signal peptide that initiates a precursor protein of 914 amino acids (SEQ ID NO: 2). A second hydrophobic region (residues 419-437) conforms to expectations for a membrane-spanning domain.

Example 2

Northern Analysis

[0035] RNAs were prepared using RNAzol (CinnaBiotecx) reagent and protocols. Formaldehyde-agarose fractionated RNAs were transferred to nitrocellulose and hybridized as described previously (Harlow, E. & Lane, D. (1988) Antibodies: A Laboratory Manual, Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y., which is incorporated herein by reference) with 32P-labeled DDR cDNA fragments described in the figure descriptions. Filters were subsequently probed with a 32P-labeled GAPDH (glyceraldehyde 3-phosphate dehydrogenase) cDNA.

Example 3

Production of Anti-DDR Antisera E.1

[0036] A Sal I-Eco RI fragment of the DDR cDNA was subcloned to pUR290 to allow production of a b-galactosidase fusion (Harlow, E. & Lane, D. (1988) Antibodies: A Laboratory Manual, Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y., which is incorporated herein by reference) of DDR residues 223 to 346 within the extracellular region of the protein. The fusion protein was purified by NaDodSO4-PAGE for rabbit immunization (Harlow, E. & Lane, D. (1988) Antibodies: A Laboratory Manual, Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y., which is incorporated herein by reference).

Example 4

Transient Expression of Human DDR in COS-7 Cells

[0037] The DDR cDNA was subcloned into the Eco RI site of the mammalian expression vector pECE (Ellis, L., Clauser, E., Morgan, D. O., Edery, M., Roth, R. A. & Rutter, W. J. (1986) Cell 45, 721-732, which is incorporated herein by reference). This plasmid, pSV33, was used to transfect COS-7 cells by the DEAE-dextran method (Kaufman, R. J. (1990) Meth. Enz. 185, 487-511, which is incorporated herein by reference). After 48 hours the cells were solubilized in NaDodSO4-PAGE sample buffer for immunoblotting.

Example 5

Immunoblotting

[0038] Cells on 10 cm plates were solubilized in 1.0 ml of RIPA buffer containing 1% Triton X-100, 20 mM Tris-HCl pH 7.5, 50 mM sodium chloride, 1 mM sodium orthovanadate, 1 mM PMSF, 50 mM sodium fluoride, 5 mM EDTA and 20 mM sodium pyrophosphate at 4° C. Samples were centrifuged for 15 minutes at 10,000×g at 4° C. and the supernatants were rocked with 50 ml of wheat germ agglutinin agarose (Vector) for 1.5 h. The beads were washed three times with RIPA buffer and boiled in 100 ml of NaDodSO4-PAGE sample buffer. Samples were fractionated on 7.5% NaDodSO4-PAGE, transferred to Immobilon-P (Millipore) and incubated with a 1:500 dilution of rabbit antiserum e.1 (see above) in 5% nonfat dry milk in phosphate buffered saline and subsequently incubated with horseradish peroxidase (HRP)-coupled sheep anti-rabbit IgG for detection with ECL (Amersham) reagents. The filters were stripped of antibodies with 100 mM 2-mercaptoethanol, 2% NaDodSO4, 62.5 mM Tris-HCl pH 6.7 for 30 minutes at 50° C. The filters were washed and incubated with HRP-sheep anti-rabbit IgG and ECL reagents as before to ensure that all antibodies were removed. The filters were then incubated with a 1:1000 dilution of the monoclonal antiphosphotyrosine antibody PY20 (ICN), and subsequently with 125I-protein A (ICN) (Harlow, E. & Lane, D. (1988) Antibodies: A Laboratory Manual, Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y., which is incorporated herein by reference) prior to autoradiography.

[0039] The invention has now been explained with reference to specific embodiments. Other embodiments will be suggested to those of ordinary skill in the appropriate art upon review of the present specification.

[0040] Although the foregoing invention has been described in some detail by way of illustration and example for purposes of clarity of understanding, it will be obvious that certain changes and modifications may be practiced within the scope of the appended claims.

Claims

1. A composition of matter comprising a polypeptide having a first domain with carbohydrate binding activity and a second domain with kinase activity.

2. A polypeptide according to claim 1, wherein said polypeptide further comprises a third transmembrane domain connecting said first and said second domains.

3. A polypeptide according to claim 1, wherein said kinase activity is tyrosine kinase activity.

4. A composition of matter comprising a polypeptide having a first domain with discoidin-type ligand binding characteristics and a second domain with tyrosine kinase activity.

5. A polypeptide according to claim 4, wherein said polypeptide comprises essentially a sequence of amino acids shown in FIG. 1 (SEQ ID NO: 2) or a portion thereof which retains said first domain with discoidin-type ligand binding characteristics and said second domain with tyrosine kinase activity.

6. A composition of matter comprising a polypeptide having a domain with discoidin-type ligand binding characteristics, wherein said polypeptide comprises essentially a sequence of amino acids shown in FIG. 1 (SEQ ID NO: 2) or a portion thereof which retains said domain with discoidin-type ligand binding characteristics.

7. A composition of matter comprising a polypeptide having a first domain with discoidin-type ligand binding characteristics, a second domain with tyrosine kinase activity, and a third transmembrane domain that connects said first and said second domains.

8. A polypeptide according to claim 7, wherein said polypeptide comprises essentially a sequence of amino acids shown in FIG. 1 (SEQ ID NO: 2) or portion thereof which retains said first, second, and third domains.

9. A polypeptide according to claim 7, wherein said polypeptide comprises a sequence of amino acids encoded by essentially a sequence of nucleotide bases assigned Genbank accession number L11315 or a portion thereof which retains said first, second, and third domains or a portion thereof which retains said first, second, and third domains.

10. A composition of matter comprising a polypeptide having a sequence of amino acids shown in FIG. 1 (SEQ ID NO: 2) or a portion thereof sufficient to provide a antibody or equivalent molecule binding site in said polypeptide.

11. A polypeptide according to claim 10, wherein said antibody or equivalent molecule binding site is proximate to or in a discoidin-type binding domain in said polypeptide.

12. A composition of matter comprising a nucleic acid encoding a polypeptide having a first domain with carbohydrate binding activity and a second domain with kinase activity.

13. A nucleic acid according to claim 12, wherein said kinase activity is tyrosine kinase activity.

14. A nucleic acid according to claim 12, wherein said nucleic acid encodes in said first polypeptide domain a sequence of amino acids having discoidin-type ligand binding characteristics.

15. A nucleic acid according to claim 12, wherein said nucleic acid comprises essentially the sequence of nucleotide bases shown in FIG. 1 (SEQ ID: 1) or a portion thereof.

16. A nucleic acid according to claim 12, wherein said nucleic acid comprises essentially a sequence of nucleotide bases assigned Genbank accession number L11315 or a portion thereof.

17. A composition of matter comprising essentially a sequence of nucleotide bases shown in FIG. 1 (SEQ ID NO: 1) or a portion thereof that encodes a polypeptide having a first discoidin-type binding domain.

18. A sequence of nucleotide bases according to claim 17, wherein said sequence further encodes said polypeptide having a second domain with kinase activity.

19. A sequence of nucleotide bases according to claim 18, wherein said kinase activity is tyrosine kinase activity.

20. A sequence of nucleotide bases according to claim 18, wherein said sequence further encodes said polypeptide having a third transmembrane domain that connects said first and said second domains.

21. A diagnostic or prognostic method for detecting the presence of a polypeptide that is produced in tumors of epithelial type cells and therefore an indicia for the presence of such tumors, comprising the steps of: a) producing an antibody equivalent molecule with binding specificity directed to a polypeptide having a sequence of amino acids shown in FIG. 1 (SEQ ID NO: 2) or a portion thereof sufficient to provide said antibody or equivalent molecule a binding site in said polypeptide; b) reacting said produced antibody or equivalent molecule having said polypeptide binding specificity with a test sample that possible contains said polypeptide; and c) analyzing said reacted test sample for binding of said antibody or equivalent molecule to said polypeptide, thereby indicating the presence of said polypeptide.

22. A method according to claim 21, wherein said test sample comprises a fluid from said test subject.

23. A method according to claim 21, wherein said test sample comprises blood or a fraction obtained from blood.

24. A method of producing a therapeutic agent for treating or influencing tumors of epithelial type cells in a patient, comprising the steps of: a) isolating a ligand which binds to a discoidin-type binding domain of a polypeptide having a sequence of amino acids shown in FIG. 1 (SEQ ID NO: 2) or a portion thereof; b) utilizing said isolated ligand to produce a therapeutic agent that binds to said discoidin-type binding domain; and c) providing a quantity of said therapeutic agent sufficient to influence said tumor's behavior in the patient.

25. A method of immunotherapy for treating a patient with tumors of epithelial type cells in which the epithelial cells present a characteristic membrane bound polypeptide comprising administering to the patient a sufficient quantity of said polypeptide to produce a suitable immunological response against said tumor cells, wherein said polypeptide has a sequence of amino acids shown in FIG. 1 (SEQ ID NO: 2) or a portion thereof sufficient to generate said response.

26. A method of inhibiting in a patient the growth of tumors of epithelial type cells in which the epithelial cells present on their surfaces a characteristic membrane bound polypeptide that influences the cells' growth by binding a specific regulatory ligand comprising administering to the patient a quantity of said polypeptide sufficient to bind to and limit the amount of available said ligand to produce said inhibition, wherein said polypeptide has a sequence of amino acids shown in FIG. 1 (SEQ ID NO: 2) or a portion thereof.