Background of the Invention

Field of the Invention

The invention, in the field of molecular and cellular biology, relates to the purification and characterization of the 90K tumor-associated antigen (IR-95), to genetic sequences which encode the 90K antigen, to the cloning and expression of this antigen, to its production and to uses thereof.

Background Informaάon

Antigens shed or secreted by tumor cells have been reported in the serum of patients with different forms of cancer. Immunoassays of some of these molecules show that they have potential use as diagnostic/prognostic indicators and for therapeutic surveillance. Some of the recognized antigens include: CA125 for ovarian cancer (Bast et al.. N. Engl. J. Med. 509:883-887 (1983)): MOV2 for ovarian cancer (Miotti et al. Cancer Res. 45:826-832 (1985)); CA15-3 for breast cancer (Hilkens et al.. Cancer Res. 46:2582-2587 (1986)); CA19-9 for gastrointestinal cancer (Koprowski et al.. Science 272:53-55 (1981)); carcinoembryonic antigen (CEA) for gastrointestinal cancer (Golp et al., JAMA 254: 1331-1334 (1968)); and CA50 for gastrointestinal cancer (Holmgren et al., Br. Med. J. 255:1479-1482 (1984)). However, none of these tumor antigen serodetection assays have been sensitive enough to permit the early detection of occult cancer, or the reoccurrence or π.erast.'.ses thereof.

While these antigens are mostly expressed on the surface of tumor cells, some are secreted into the circulation of patients. This last category of antigens may prove useful for the serodetection, prognosis and assessment of tumor load and cancer development.

Monoclonal antibodies (MAbs) which detect tumor-associated antigens have been reported. For example. MAbs against circulating breast cancer-

associated antigens have been obtained. One such MAb, SP-2, identified a cytoplasmic antigen, termed the 90K antigen (a.k.a. ImmunoRegulin-95 or IR-95), which is expressed in more than 80% of breast cancers (lacobelli et al.. Cancer Res. 46.-3005-3010 (1986)). Approximately 50% of the patients with breast cancer, 40% of the patients with gastrointestinal malignancies, and 30% of the patients with gynecological malignancies had elevated serum levels of the 90K antigen (lacobelli et al., Breast Cancer Res. & Treat. 11: 19-30 (1988)). More importantly,, the assay of the present invention has demonstrated that the percentage of patients showing elevated serum levels is greater for individuals with metastatic disease and that the 90K serum changes correlated with cancer progression (lacobelli et al., Breast Cancer Res. & Treat. 22:19-30 (1988); Scambia et al., Anύcancer Res. 5:761-764 (1988); Benedetti-Panici et al., Gynecol. Oncol. 55:286-289 (1989)). Since the 90K antigen is distinct from other circulating antigens such as CA 15-3, CEA, and CA 125 (lacobelli et al.. Breast Cancer Res. & Treat. 22:19-30 (1988); Benedetti-Panici et al., Gynecol. Oncol. 55:286-289 (1989)), it may represent an additional useful diagnostic tool for the surveillance of breast cancer and other malignant diseases. Homology in the region of amino acids 35-80 of the 90K antigen is found with the type I macrophage scavenger receptor ( odama et al., Nature 343:531 (1990)); sea urchin speract receptor (Dangott et al.. Proc. Natl. Acad. Sci. USA 56:2128 (1989)); and human lymphocyte glycoprotein Tl/Leu-1 (Jones rt al.. Nature 525:346 (1986)). The 90K antigen is referred to in European Patent Application Number

91830153.2 filed on April 17, 1991 (Publication Number 0453 419 A2). An antigen with the same 15 amino acid terminal sequence is referred to in PCT Application Number PCT/US85/02132 which was filed on 30 October 1985 and has International Publication Number WO 86/02735. This PCT application claims priority to U.S. applications 667,521 and 785,177 which were filed on November 2, 1984 and October 7, 1985. However, no studies

have specifically elucidated the physicochemical and immunochemical properties of this antigen. Therefore, it is important to purify and characterize the SP-2-reaαive 90 antigen.

Summary of the Invention

The application is drawn to the purification and characterization of the

90K tumor-associated antigen from: the culture fluid of a human breast cancer cell line, CG-5; the serum of a breast cancer patient; and the ascitic fluid of an ovarian cancer patient. A purification procedure is provided which results in at least a 50,000 fold purification of the 90K tumor-associated antigen from the three different sources. The native antigen is a giycoprotein and has an apparent molecular weight of about 95,000 daltons and is present as a high molecular weight complex with similar eiectrophoretic profiles and immunoreaciivity from all three sources.

The invention is further drawn to the amino acid sequence of the 90K antieen and to the genetic sequence which encodes the 90K antigen. Therapeutic and diagnostic uses of the 90K antigen are also provided.

Brief Description of the Drawings

FIGURE 1. The nucleotide and amino acid sequence of the 90K protein (SEQ ID NO:l and SEQ ID NO:2, respectively). The signal peptide is boxed, the SRCR homology region is shaded, and potential asparagine- lin ed glycosylation siies are circled.

FIGURE 2. Sepharose CL-6B column chromatography of the 90K antigen which had been isolated from CG-5 tissue culture fluid ( ); the scrum of a breast cancer patient ( ); and the ascitic fluid of an ovarian cancer patient (--). Fractions were assayed for 90K activity by immunoradiomctric assay (IRMA). The arrow indicates the elution volume of Dextran blue 2000.

FIGURE 3. Density gradient cerarifugatioπ of the 90K. antigen.

Purified 90K from CG-5 culture fluid ( ), the serum of a breast cancer patient ( ), the ascitic fluid from an ovarian cancer patient (-), and unfraciionated serum from a breast cancer patient (- ^■ -) were subjected to equilibrium ultraccntrifugation in cesium chloride. Fractions were assayed for 90K activity by IRMA and their densities were determined by weighing a known volume of each. The arrow indicates the buoyant density of ø-galactosidasε.

FIGURE 4. Molecular weight determination of the 90K antigen. (Figure 4A): Immunoprecipitates of radioactive 90K antigen from human breast cancer cells. Aliquots (200,000 cpm thrichloroacetic acid precipitable) of ( ³⁵ S)methionine-labeled culture fluid were immunoprecipitated with MAb - SP-2 (lanes a-e) or MAb against alfa-fetoprotein (lane f), and were analyzed by SDS:PAGE in the presence (lanes a-c, and e) or absence (lane d) of 2-mercaptoethanol, followed by fluorograp y. Lane a contained CG-5 cells. Lane b contained MCF7 cells. Lane c contained T47D cells. Lane d contained T47D cells. Lane e contained tissue culture fluid from CG-5 cells after the cells had been exposed to tunicamycin but before ( ³⁵ S)methionine labeling. (Figure 4B): SDS:PAGE analysis of 90K antigen purified from: CG-5 culture fluid (lane a, 620 units); serum from a breast cancer patient (lane b, 920 units); and ascitic fluid from an ovarian cancer patient (lane c, 700 units). The gels were silver stained. The molecular weight standards were: phosphorylase fa (Mr 97,000) and BSA (Mr 66.000).

FIGURE 5. PAGE and western blot analyses of purified 90K antigen from: CG-5 culture fluid (lanes a and d); the serum of a breast cancer patient (lanes b and e); and the ascitic fluid from an ovarian cancer patient (lanes c and f). Purified 90K antigen was analyzed on the 4-20% gradient gel containing 0.1 % NP-40. Lanes a-c were silver stained. Lanes d-f proteins were electro lotted onto a nitrocellulose membrane. The molecular weight standards were: 0-galactosidase (Mr 540,000) and BSA (Mr 66,000).

FIGURE 6. The effect of enzymatic digestion on the 90K antigen. (Figure 6A): Purified 90K from CG-5 culture was digested with various proteases and was analyzed on 9% SDS:PAGE followed by silver staining. (Figure 6B): The binding of ( ¹²⁵ I)labeled SP-2 to digested 90K relative to untreated control is displayed. For both Figures 6A and 6B: lane a was purified 90K control; lane b was pronase-treated 90K antigen; lane c was papain-treated 90K antigen lane d was trypsin-treated 90K antigen; and lane e was chymotrypsin-treated 90K antigen. For Figure 6B: lane f was neuraminidase-treated 90K antigen; lane g was fucosidase-treated*90K antigen lane h was chondroitinase ABC-treated 90K antigen: lane i was α-gaiactosidase-treated 90K antigen; and lane 1 was 0-galactosidase-treated 90 antigen.

FIGURE 7. Plasmid map of CMV-1R95.

FIGURE 8. Plasmid map of CMVNEO-IR95. FIGURE 9. An autoradiogram of immunoprectpitates of the first three stable clones in human mammary carcinoma BT20 ceiis.

FIGURE 10. SDS-PAGE of ³⁵ S-methionine labeled transiently expressed IR-95 in 293 cells transfected with plasmid pCMV-IR-95.

FIGURE 11. Percentage of cell lysis versus various IR-95 concentrations.

Detailed Description of the Invention

The present invention provides a substantially purified tumor-associated antigen which has an apparent molecular weight of approximately 95 kilodaltoπs (K) and is designated the 90K antigen (a.k.a. ImmunoRegulin-95 or IR-95). The concentration of this tumor-associated antigen is elevated in the serum of patients with cancer, such as breast cancer, gastrointestinal malignancies, and gynecological malignancies, and also in patients with the human immunodeficiency virus (HIV).

The 90K antigen reacts with MAb SP-2 which was produced by immunizing mice with proteins that had been released into tissue culture fluid by human MCF-7 breast cancer cells maintained therein. The hybridoma cell line which produces MAb SP-2 was deposited according to rules 28 and 28a of the European Patent Convention on April 12, 1991 at the Institut Pasteur, Collection Nationale de Cultures de Microorganisms, 28 Rue de Docteur Roux, 75724 Paris Cedex 15, France. This deposit has been given the Accession Number 1-1083. The cells were found to be viable on April 22, 1991. Utilizing MAb SP-2 to detect the antigen, it has been demonstrated that low levels of 90K are present in normal subjects, whereas antigen levels up to 100 times that of normal levels have been detected in 50% of patients with breast cancer. The 90K antigen has also been detected in the sera of patients having carcinomas of non-breast origin, including carcinomas of the ovary, endometrium, and colon. In accordance with the invention, a 90K tumor-associated antigen or determinant can be isolated from a sample containing the antigen. Any sample that contains the antigen may be utilized as a starting material according to the methods described in the invention. The 90K tumor-associated antigen of the present invention is a giycoprotein found in the tissues and sera of patients with breast cancer and other malignant neoplasms, and with HIV infection. Therefore, it is possible to isolate the 90K protein from: the plasmas or serum of humans or other animals: naturally occurring tumor cell lines from humans or other animals which naturally produce the 90K protein; immortal cell lines from humans or other animals which do not eπdogenously produce the 90K protein but which have been made to do so by having been transfected with a 90 expression plasmid; and cell lines from humans or other animals which do not endogenously produce the 90K. protein, and that are capable of growing in the absence of serum additives (such as U 937 cells) and which have been transfected with the 90K gene. For example, any source of the antigen is contemplated for use in this invention including, but not limited to: the culture fluid of the human breast cancer cell line. CG-5; serum from patients

-/-

with breast cancer; and ascitic fluid from patients with ovarian cancer. As used herein, the sample containing the antigen will be referred to simply as "the sample" and is intended to include any 90 antigen-containing sample. Generally^ a four-step procedure to purify the 90K antigen is utilized to practice this invention. The procedure comprises ammonium sulfate precipitation, gel filtration chromatography, ion-exchange chromatography, and adsorption to a MAb SP-2 affinity marrix. However, it is recognized that some variation in the procedure may still result in the production of highly purified 90K antigen. The purification procedure used to isolate the 90K antigen from a sample is summarized in Table 1. After centrifugation of the sample, the protein was precipitated by adding solid ammonium suifate and allowing the sample to stand overnight at 4°C. Protein precipitates were collected by centrifugation. At each step of purification, the total protein was determined and the antigen was quantified by IRMA. Virtually all 90K activity was recovered after ammonium sulfate precipitation, resulting in about a four-fold enrichment thereof.

The ammonium sulfate-precipitated antigen was next subjected to size exclusion chromatography. The 90K antigen was constantly found in a large peak eluting immediately behind the void volume of the column, implying that it is a high molecular weight complex. Minor reactivity peaks of lower molecular weight were also inconsistently observed which were probably due to egradation products.

The high molecular weight peak was further purified by DEAE-cellulose chromatography. The 90K antigen eluted from the column at a NaCl concentration of about 0.25M Na l.

The final purification was accomplished by immunoaffmity adsorption on Sepharose coupled to MAb SP-2. The coupling was done by the method of Schneider et al. (J. Biol. Chem. 257:10766-10769 (1982)). Bound 90K antigen was eluted with buffer, preferably 3M MgC^.

-S-

The purification procedure resulted in a substantially purified 90K antigen. By substantially purified is meant that the purification of the 90 antigen, as described herein, resulted in at least a 50,000-fold, and generally about 50,000- to about 80,000-fσld purification of the 90K antigen. The invention is thus drawn to substantially purified 90K antigen having an apparent molecular weight of approximately 95,000 daltons. as well as to antigenic determinant-containing fragments, and other fragments thereof. The invention is also drawn to naturally occurring fragments of the 90K antigen. The invention is further drawn to unglycosylated moieties of the 90K antigen.

As used herein, polypeptides containing immunologicaliy cross-reactive antigenic determinants means polypeptides having a common antigenic determinant with which a given antibody will react. Such polypeptides include the glycosylated and unglycosylated moieties of the 90K antigen and fragments thereof, as well as synthetic polypeptides, or fragments thereof, and antibodies which are anti-idiotypic towards the active deteπninant(s) of the 90K protein. It has been demonstrated tha anti-idiotypic reagents are useful as diagnostic tools for the detection of antigens carrying sites which are immunologicaliy cross-reactive with those on antibodies (PotocnjaK e ai.. Science 225:1637-1639 (1982)).

Once the antigen has been purified, monoclonal and polyclonal antibodies can be generated to it using standard techniques which are well known to those of skill in the art (Klein, J. , Immunology: The Science of Cell- Noncell Discrimination. John Wiley and Sons. New York, New York, USA (1982): Kenneth et l. Monoclonal Antibodies. Hybridoma: A New Dimension in Biological Analyses, Plenum Press, New York, New York, USA (1980); Campbell. A., "Monoclonal Antibody Technology, " In: Laboratory Techniques in Biochemistry and Molecular Biology. Vol. 13 (Burdon et al., eds.), Elsevier. Amsterdam, The Netherlands (1984); and Eisen, H.N., In: Microbiology. 3rd Edition (Davis et al., eds.), Harper &. Row, Philadelphia, PA, USA (19SC

Of special interest to the invention are antibodies to the 90K antigen or its derivatives which are produced in humans, or are "humanized" (i.e., non-immunogenic in a human) by recombinant DNA or other technology. Humanized antibodies may be produced, for example, by replacing an immuπogenic portion of an antibody with a corresponding, nonimmunogenic, portion (i.e., chimeric antibodies). See, Robinson et al., International Patent Publication PCT/US86/02269; Akira et al., European Patent Application 184,187; Taniguchi. M., European Patent Application 171,496; Morrison et al., European Patent Application 173,494; Newberger et al., PCT Application WO86/01533; Cabilly et al., European Patent Application 125,023; Better et al., Science 240: 1041-1043 (1988); Liu et al., Proc. Natl. Acad. Sci. USA 54:3439-3443 (1987); Liu et l., J. Immunology 259:3521-3526 (1987); Sun et al., Proc. Natl. Acad. Sci. USA 54:214-218 (1987); and Shaw et al., J. Natl. Cancer Inst. 80: 1553-1559 (1988)). General reviews of humanized chimeric antibodies are provided by Morrison, S.L., (Science 229:1202-1207 (1985)) and Oi et al., (BioTechniques 4:214 (1986)). The purified 90K protein can be sequenced using methods which are well known to those of skill in the an. Initial sequencing of the terminal amino acid sequence of the 90K protein has reveaied ihe following amino acid sequence (SEQ ID NO:3): Val Asn Asp Gly Asp Met Arg Leu Ala Asp Gly Gly Ala Thr Asn Gin Gly Arg Val Glu He Phe. An analysis of the amino acid composition of the 90K antigen is found in Table 4. Further characterization of the 90K antigen is provided in Table 2 which gives the effects of chemicai and physical treatment ^* , on 90 activity. It is generally recognized that having the amino acid sequence of a protein enables one to make oligonucleotide probes which can be used to identify clones of the protein. Tnus, hybridization with the appropriate nucieic acid probe will identify ciones containing the nucleotide sequence coding for the 90K amigeπ. As used herein. "DNA construct" means any DNA sequence which has been created synthetically or through recombinant DNA technology. ^" DNA

constructs" include, but are not limited to, synthetic oiigonucieotides, ve ors and vectors containing inserts.

Particular nucleotide probes which are useful for identifying the 90K antigen genes can be constructed from knowledge of the amino acid sequence of the 90K protein. The sequence of amino acid residues and the peptide is designated herein using either the commonly employed 3-letter or single-letter designations therefor. A listing of these three- and one-letter designations may be found in textbooks such as Lehninger, A., Biochemistry, Worth Publishers, Inc., New York, New York, USA (1975) and subsequent volumes thereof. The N-termiπal sequence of the first twenty-two amino acids enabled the synthesis of a 66 nucleotide long oiigonucleotide which was utilized as a probe to screen a cDNA library from MCF-7 cells. In this manner, the inventors have completed the molecular cloning and have determined the complete cDNA sequence of the 90K antigen. The invention comprises the amino acid sequence of the 90K antigen, the genetic sequences coding for the antigen, vehicles containing the genetic sequence, hosts transformed therewith, 90K protein production by transformed host expression, purification of the 90K protein from a sample, and utilization of the 90K antigen. Nucleotide and amino acid sequences for the 90K protein are shown in

Figure 1 (SEQ ID NO:l and SEQ ID NO:2, respectively). It is understood that modifications of the specified amino acid and nucleic acid sequences are encompassed by the present invention. As used herein, the term "modification" is intended to mean any substitution, addition or deletion of one or more amino acids of the polypeptide fragment or nucleorides of the nucleotide sequence. These modifications may be made by manipulating the amino acid sequence itself or by modification of the nucleic acid sequence which is then used to synthesize the peptide.

Changes in the nucleic acid sequence can be effected by mutating the DNA. usually by site-directed mutagenesis- The techniques of site-specific mutagenesis are well known to those of skill in the an. (see, for example,

Adelman et al.. DNA 2:183 (1983); Smith. M., Ann. Rev. Genetics 19:423 (1985)). Mutations include, for example, substitutions, additions, or deletions of nucleotide(s), provided that the final construct has the desired biologic activity. The nucieic acid changes must not place the sequence out of reading frame and preferably should not create complementary regions that could produce secondary mRNA structure (see EP Patent Application Publication No. 75,444).

Methods for the modification of amino acids as well as nucleic acids are known in the an. Amino acid sequence insertions include amino and/or carboxyl-terminal fusions from one residue to polypeptides of essentially unrestricted length, as well as intrasequence insertions of single or multiple amino acid residues. Intrasequence insertions may range generally from about 1 to about 10 residues. More preferably they range from about 1 to about 5 residues. The amino acid residues may be in their proiected or unprotected form, using appropriate amino or carboxyl protecting groups. In addition, the synthesized peptides may be glycosoiated or ungiycosoiated.

To express the 90K antigen, transcriptional and translational signals which are recognizable by an appropriate host are necessary- The cloned nucleic acid sequences encoding the 90K protein, preferably in double-stranded form, may be operably linked to sequences controlling transcriptional expression in an expression vector, and introduced into a host cell, either prokar otic or eukaryotic, to produce recombinaπt 90K protein or variants thereof. Depending upon which strand of the 90K protein encoding sequence is opεrably linked to the sequence(s) controlling transcriptional expression, it is also possible to express 90K protein amisense RNA or variants thereof.

As used herein, "expression vehicle" means a DNA construct which is capable of directing the expression of an operably linked DNA sequence. Expression vehicles include, but are not limited lo, phage and plasmid vehicles. "Expression vehicles" typically contain one or more elements selected from the group consisting of, but not limited to, an operator, a

promoter, a ribosome binding site, a translation-initiation signal and a translation terminator.

As used herein, "host cell" means any cell capable of being transformed or transfected with a DNA construct or an expression vehicle. Expression of the 90K protein in different hosts may result in varying post-iranslatioπal modifications which may alter the properties of the protein.

A nucleic acid molecule, such as DNA, is said to be "capable of expressing" a polypeptide if it contains expression control sequences which contain transcriptional regulatory information. For expression of a polypeptide, control sequences must be "operably linked" to the nucleotide sequence which encodes the polypeptide.

An operable linkage is a linkage in which a nucleotide sequence encoding a polypeptide is connected to a regulatory sequence (or sequences) in such a way as to place expression of the polypeptide encoding sequence under the influence or control of the regulatory sequence. Two DNA sequences (such as a 90K protein encoding sequence and a promotor region sequence linked to the 5' end of the encoding sequence) are said to be operably linked if the induction of promoter function results in the transcription of the protein encoding sequence and if the naiure of the linkage between the two DNA sequences does not (I) result in the introduction of a frame-shift mutation, (2) interfere with the ability of the expression regulatory sequences to direct the expression of the 90K mRNA, antisense RNA, or protein, or (3) interfere with the ability of the 90K template to be transcribed by the promoter region sequence. Thus, a promoter region would be operably linked to a DNA sequence if the promoter were capable of effecting transcription of that DNA sequence.

The precise nature of the regulatory regions needed for gene expression may vary between species or cell types, but generally includes 5' non-coding sequences involved with the initiation of transcription and translation, such as the TATA box, capping sequence, CAAT sequence, and the like. Such 5 ^r

non-codiπg control sequences will especially include a region which contains a promoter for the transcriptional control of an operably linked gene.

Expression of the 90K protein in eukaryotic hosts requires the use of regulatory regions, preferably eukaryotic, which are functional in such hosts. A wide variety of transcriptional and translationai regulatory sequences can be employed, depending upon the nature of the eukaryotic host. The transcriptional and translationai regulatory signals can also be derived from the genomic sequences of viruses which infect eukaryotic cells, such as adenovirus, bovine papilloma virus, Simian virus, herpes virus, or the like. Preferably these control signals are associated with a paπicular gene which is capable of a high level of expression in the host cell.

Promoters from mammalian genes which encode RNA products capable of being translated are preferred, and especially, strong promoters such as the promoter for actin, collagen, myosin, etc., can be employed, provided they also function as promoters in the host cell. For eukaryotic promoters see generally, Hamer et al., J. Mol. Appl. Gen. 2:273-288 (1982); McKni ht, S., Cell 52:355-365 (1982); Benoist et al.. Nature (London) 290:304-310 (1981); Johnston et al.. Proc. Natl. Acad. Sci. USA 79:6971-6975 (1982); and Silver et al.. Proc. Natl. Acad. Sci. USA 52:5951-5955 (1984).

General methods for molecular cloning and expression can be found in Sambrook et al.. Molecular Cloning: A Laboratory Manual. 2d. ed., Vols. 1-3, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY, USA (1989). Transcriptional initiation regulatory signals can be selected which allow for the repression or activation of gene expression, so that expression of the operably linked genes can be modulated. The vectors of the invention may further comprise other operably linked regulatory elements, such as enhancer sequences or DNA elements, which confer tissue or cell-type specific expression on an operably linked gene.

The purified protein and antibodies thereto as well as its genetic sequences are useful in diagnostic and therapeutic methods.

In particular, the level of the 90K antigen is useful as a diagnostic indicator for cancer, including breast, ovarian and other malignancies, viral infection, including HIV, inflammation, autoimmune disease, aging, and the like.

The 90K antigen can be assayed by a variety of methods. In serum, the 90K antigen can be assayed utilizing an enzyme-linked immunosorbent assay (ELISA) sandwich procedure. In this manner, MAb SP-2 can be utilized both as an immunoabsorbent and as an enzyme-labeled probe to detect and quantify the 90K antigen by a sandwich-type ELISA. The amount of 90K present in the sample can be calculated by reference to the amount present in a standard preparation of CG-5 cell lysate using a linear regression computer program. The assay has been previously described by lacobelli et al. (Breast Cancer Res. and Treatment 22:19-30 (1988)), which reference is herein incorporated in its entirety. Overexpression of the 90K antigen would be an indicator of a disorder.

Expression levels of the 90K antigen can also be determined by measuring the levels of RNA. In this method, a nucleic acid probe can be utilized to hybridize to the RNA in the sample. Methods for hybridization are generally known to those of skill in the art (see. for example. Nucleic Acid Hybridization. A Practical Approach. IRL Press, Washington, D.C., USA (1985) and the references cited therein).

The 90K antigen or its genetic sequences may aiso be useful in therapy- Serum iR-95 levels are elevated not only in patients with cancer, but also in those affected by different physϊopathological conditions (see Table 5). such as infection by HIV or other viruses, autoimmune disease, etc., all of which are characterized by a variable degree of immune deficit associated with immune activation.

Iπ vitro experiments have also shown that the 90K antigen is able to enhance natural killer (NK) and lymphokine activated killer (LAK) cell activity of peripheral blood ononuciear cells (Figure 11).

Given the above findings, the 90K antigen or its genetic sequences may also be useful in therapy as an immunoregulatory agent. For example, patients who suffer from a particular cancer which does not induce overexpression of the 90K antigen may be treated by infusion with the 90K antigen. Furthermore, those patients with cancers that generate elevated levels of the 90K protein in their serum, may be supplied additional 90K antigen by infusion.

The 90K antigen or its genetic sequences may also be useful in gene therapy (reviewed in Miller, Nature 357:455460 (June 1992). In one preferred embodiment, an expression vector containing the IR-95 coding sequence is inserted into cells, the cells are grown in vitro and then infused in large numbers into patients. In another preferred embodiment, a DNA segment containing a promoter of choice (for example a strong promoter) is transferred into cells containing an endogenous IR-95 in such a manner that the promoter segment enhances expression of the endogenous IR-95 gene (for example, the promoter segment is transferred to the ceil such that it becomes directly linked to the endogenous IR-95 gene).

Tne 90K antigen or antagonists thereof can routinely be prepared as therapeutic agent(s) by one of skill in the art using standard techniques and references which are well known in the an (see, for example. Remington 's Pharmaceutical Sciences. 18th ed., (A.R. Gennaro, Ed.), Mack Publishing Comp., Easton. PA, USA 18042 (1990), especially chapters 8 (Pharmaceutical Preparations and Their Manufacture) and 4 (Testing and Analysis), thereof). As used herein, by "antagonist" is meant any compound that decreases the effect of the 90 antigen in or in vitro.

Appropriate and optimum routes of administration can also be routinely determined by one of skill in the an. The former include the oral.

intraveπous, intramuscular, subcutaneous, transdermai, /ΛΛtaandbucaJ routes of administration among others.

The doses of the 90K antigen and antagonists) thereof which is useful as a treatment are "therapeutically effective" amounts. As used herein, a "therapeutically effective amount" means an amount of the antigen, fragment or antagonist thereof, which produces the desired therapeutic effect. This amount can be routinely determined by one of skill in the art and will vary depending upon several factors such as the particular illness from which the patient suffers and the severity thereof, as well as the patient's height, weight, sex, age, and medical history. Generally, the 90K antigen of the present invention is preferably provided at a dose of between about 5 to about 5000 rag/dose/week/patieπt. More specifically, one preferable dose range is from 50 to 500 mg/dose/week/patient.

For the treatment of autoimmune disease, rheumatoid arthritis, allergy, rejection of organ transplants, and other pathological situations where the immune system is activated and needs to be suppressed, a 90K antigen antagonist can be administered. The appropriate doses of the antagonist can be routinely determined by one of skill in the an as described above. Generally the antagonists) of the 90K antigen is preferably provided at a dose of between about 5 to about 5000 mg/dose/week patient. More specifically, one preferable dose range is from 50 to 500 mg/dose/week/patient.

Any terms which are used herein and are not specifically defined herein are used as they would be by one of ordinary skill in the an(s) to which the invention pertains. The Examples which follow are for illustrative purposes only and are not intended to limit the scope cf the invention.

Example I Characterization of the 90K Antigen

Materials and Methods

Cell Lines and Reagents. CG-5, an estrogen-supersensitive variant of the MCF-7 human breast cancer cell line (Natoli et al., Breast Cancer Res. Treat. 5:23-32 (1983)) and other human breast cancer cell lines were maintained In Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% fetal calf serum (FCS) and antibiotics. The murine MAb SP-2 produced by hybridoraas grown in pristane-primed Balb/c mice (lacobelli et al.. Cancer Res. 46:3005-3010 (1986)) was purified from ascitic fluid by ammonium sulfate precipitation followed by ion-exchange chromatography (lacobelli et al.. Breast Cancer Res. & Treat. 22:19-30 (1988)). Hybridoma cells which produce MAb SP-2 were deposited under the provisions of the European Patent Convention at the Pasteur Institute as previously described. This cell line was given the deposit number 1-1083.

Purified MAb SP-2 was labeled with Na ^{1 5} I using lactoperoxidase (Thorell et al., Biochem. Biophys. Acta 252:363 (1971)). The proteases and other enzymes were purchased from Sigma Chemical Corp., St. Louis, MO, U.S.A. Electrophoresis reagents were purchased from Bio-Rad Laboratories, Segrate, Italy. Sepharose CL-6B was purchased from Pharmacia, Uppsala,

Sweden. All other reagents were of the highest purity commercially available.

Solid-Phase Radioimmunoassay. A "two-step" sandwich IRMA was developed to measure 90K activity. Polystyrene beads (6.5 mm, Precision

Plastic Balls. Chicago, Illinois, USA) were coated with biotinylated MAb SP-2 by the protein-avidin-biotin-capture system (Suter et al.. Mol. Immunol. 26:221-230 (1989)). Biotinylation of SP-2 was carried out according to the method of Guesdon et al. (J. Histochem. Cytochem. 27:113-118 (1979)). After coating, the beads were washed extensively with 0.9% NaCl solution and were incubated with biotinylated MAb SP-2 (5 μg/ml) at room temperature for 18

hours. Coated beads were treated with a blocking solution of BSA (2 g/ml) for 1 hour at room temperature, were washed with distilled water and were stored at room temperature until used. Beads treated in this fashion were stable for at least six months. With each assay, 200 μl of appropriately diluted samples or standards were incubated with MAb SP-2-coated beads for 1 hour at 37 ^β C. The beads were washed with distilled water followed by the addition of 100 μl of ( ¹²⁵ I)-Iabeled MAb SP-2 (approximately 50,000 cpm; specific activity, 10 μCi/μg) in PBS, pH 7.4, containing 5% BSA, 0.1 mg/ml normal mouse IgG and 0.1 % N 3 for an additional hour at 37 ^β C. The beads were washed with distilled water and were counted in a gamma-counter. The amount of 90 was calculated by reference to the amount present in standard preparations made from a pool of sera from breast cancer patients and titered to contain 40, 20, 10, and 5 arbitrary units/ml. The simultaneous assay of 120 sera from breast cancer patients using IRMA and ELISA (lacobelli et al. , Breast Cancer Res. ά Treat. 22:19-30 (1988)) gave a correlation coefficient of 0.91 (Kendall Q test). Compared to ELISA, IRMA is approximately three rimes more sensitive, faster to perform, requiring less than 3 hours, and highly reproducible with an inter- and intra-assay coefficient of variation of 4%. PAGE and Western Blotting. SDS-PAGE was performed essentially according to the method of Laemmii (Nature 227:680-685 (1970)) on a veπical slab gel apparatus. Samples were treated with "sample buffer" consisting of 63 mM Tris-HCI containing 1.25% SDS and 5% 2-mercaptoethanol, or 63 mM Tris HCI plus 0.25% NP-40 (Nonidet-P40, Sigma Chem. Coip., St. Louis, MO, USA). In the present study, 9% SDS-gels and 4-20% gradient gels with NP-40 were used. Gels were run at consiant voltage in Tris-glycine buffer (pH 8.3) containing either 0.04% SDS or 0.1 % NP-40. Protein bands were visualized with Coomassie blue R 250 or a silver stain kit (Bio-Rad Laboratories. Segrate. Italy). For immunological analysis, the gels were electroblottcd onto nitrocellulose membranes at 50 V for 2 hours as described by Towbin et al. (Proc. Natl. Acad. Sci. USA 76:4350-4354 (1979)) except

that the transfer buffer did not contain methanol. The membranes were blocked with bovine skim milk, followed by incubation with MAb SP-2 (10 μg/ml) for 2 hours at room temperature. The membranes were washed thoroughly with PBS and were stained with an Extravidin-biotin Staining Kit (Sigma Chemical Coip., St. Louis, MO, U.S.A. ) according to the manufacturer's instructions.

Radiolabeling of Cells and Immunopreάpitation. For metabolic labeling, 2 x 10 ⁶ cells were incubated at 37 ^β C for 6 hours in DMEM containing 250 μCi/ml ( ³⁵ S)methionine (specific activity: 1500 Ci/mmole; The Radiochemical Centre, Amersham, U.K.). Culture fluids containing the radioactive proteins were pre-ciarified as described by lacobelli et al. (Cancer Res. 46:3005-3010 (1986)), and were incubated with MAb SP-2 coated polystyrene beads at 4°C for 16 hours. The beads were washed with distilled water and were extracted with 100 μl of SDS-sample buffer for 30 min at 50 ^β C. The extracts were run on SDS:PAGE. As controls, aliquots of culture fluid were incubated with polystyrene beads that had been α-ated with a MAb against alpha-fetoprotein (Sorin Biomedica, Saluggia, Italy). ( ³⁵ S)methionine- labeled protein bands were visualized by fluorography. In some experiments cells were labeled in the presence of 5 μg/ml of tunicamycin (Sigma Chemical Corp., St. Louis, MO, U.S.A.). Tunicamycin was added to the cells 2 hours before the addition of ( ^•? 'S)methiomne.

90K Purification, (a) CG-5 issue Culture Fluid. CG-5 cells (Natoli et al.. Breast Cancer Res. Treat. 3:23-32 (1983)) were grown in DMEM supplemented with 3% FCS using Cell Factory plastic chambers (Nunc, Roskilde. Denmark). When the cells became confluent (5 to 7 days), the culture fluid was collected. Then fresh medium was added and collected at 24 hour intervals for an additional 3 to 4 days. The concentration of 90K antigen produced under these conditions ranged from 100 to 400 units/ml. Pooled culture supernatants (10 to 20 liters) were centrifuged at 4000 x g (10 min at 4 ^β C) followed by a 10-fold concentration using a Minitan apparatus (Milhpore Corp., Bedford, MA, USA). Solid ammonium sulfate was slowly added to

reach 43 % saturation and. after standing overnighrar4 ^β C ^_ protein precipitates ^' were collected by centrifugation at 10,000 x g (15 min at. 4°C).. _. The precipitates were stored frozen at -20°C under which, conditions the 90K activity was stable for at least 2 months ^* , (b) Humaπ~seπnπ; " Whole serum from a patient with advanced breast cancer which had been ritered-to..contain- high concentrations of 90K by IRMA, was clarified by centrifugation at 10,000 x g for 20 min, then was diluted 1:1 witfa PBS iand wasrftaάionally precipitated with ammonium sulfate as described above for tissue culture fluid, (c) Ascitic fluid. This was. obtained by paracenresis from a patient. with advanced ovarian carcinoma. The. fluid was clarified by centrifugation -at 10.000 x g for 20 min and was precipitated with ammonium sulfate as above. The ammonium suifate precipitates were dialyzed extensively against PBS and were applied to a Sepharose CL-6B column (4.2 x 85 cm). They were equilibrated and eluted with PBS-0.5 M NaCl, pH 8.1, at a flow rate of 18 mi/hour. Five ml fractions were collected aπd. were ^' -assaved for 90K by IRMA. The protein was quantified by-the method of Bradford {Anal. Biochem. 72:248-254 (1976)). Fractions-containing 90K activity were pooled, dialyzed against 0.005 M Na-phosphate buffer, pH 7.4, and were applied to a DEAE-celiulose column (2 8 cm) equilibrated in the same buffer. The column was washed extensively with buffer and the adsorbed proteins were eluted using a stepwise sodium chloride gradient (0.062 to I.0-M). ^• Fractions containing 90K activity were pooied and mixed with MAb SP-2-conjug2ied Sepharose CL-4B (4 mg antibody/ml resin) at a volume ratio of 8:1 (samp!e:rεsin). MAb SP-2 was coupled to Sepharose by the method of Schneider et cl. (J. Biol. Chem. 257:10766-10769 (1982)). The mixture was rotated overnight at 4°C. The 90K antigen was-eluted with 3 M MgCI -

Density Gradient Centrifugation. Centrifugation of the 90K antigen isolated from CG-5 tissue cuiture fluid, the serum of a patient with breast cancer, or ascitic fluid from a patient with ovarian- cancer, after desorption from the affinity matrix, was performed in 5 ml of a CsCT isopicnic density gradient. The antigen was dissolved in a CsCl solution in PBS with a starting

density of 1.4 g/ l, and the gradients were formed by centrifugation in a Beckmaπ S W 50.1 rotor at 145,000 x g for 72 h at 4°C. Fractions (0.25 ml) were collected, diluted 1:10 with PBS and were assayed for antigenic activity using 90K IRMA. The density of each fraction was determined by weighing a known volume thereof.

Biochemical Characterization of the Antigen. This was performed directly on antigen seeded on microtiter plates. Microplates (Dynatecs) were coated with 50 μl of purified 90K (100 ng/ml of 0.05 M carbonate buffer, pH 9.6) and were incubated overnight. (a) Chemical Treatment. Methaπol treatment was carried out at 4°C for 30 min. Denaturation was performed with either urea 6 M and guaπidine-HCl 6 M or 1 % SDS at 45 ^β C for 1 hour. Periodate oxidation was performed for 1 hour at room temperature with 10, 20, 30, 40, 50 mM NaI0 in acetate buffer (50 mM, pH 4.5) in the dark according to Stahl et al. (Proc. Natl. Acad. Sά. USA 75:4045-4049 (1976)). Reduction was performed with dithiothreitol (10 mM in 50 mM Tris„ pH 8.1) or 5% 2-mercaptoethanol at 37°C for 1 hour. Alkylation was performed with 20 mM iodacetic acid at 30 ^β C for 30 min.

(b) Proteolytic Enzymes. Antigen-coated microplates were exposed for 90 min at 37 ^β C to irypsin (2 mg/ml), chymotrypsin (2 mg/ml), or pronase (19 mg/ml) in 50 mM Tris-2mM CaCl ₂ , pH 8.1, or to papain (0.2 mg/ml) in 50 mM cysteine-HCl, pH 6.0. In parallel experiments, aliquots of purified 90K were digested with the same proteases, were mixed with an equal volume of SDS sample buffer, and were separated by SDS:PAGE followed by silver staining.

(c) Exogiycosidases. Microplates were exposed to either neuraminidase, fucosidase, α-glucosidase and 3-glucosidase in 50 mM acetate buffer, pH 5.0, or to chondroitinase ABC in 250 mM Tris, 176 mM CH3COONa, 250 mM NaCl, pH 8.0. incubations were carried out at 37°C for 90 min. The concentrations of exogiycosidases were chosen to ensure complete digestion of the oligosaccharide residues. This was verified in

separate experiments in which the appropriate substrates were shown to be completely hydrolyzed as detected by thin-layer chromatography.

After treatment, microplates were washed and blocked with 1 % gelatin in PBS. Fifty μl of ( ¹²⁵ I)labeled MAb SP-2 (approximately 50,000 cpm) were added to each well and were incubated at 37 ^β C for 1 hour. After 3 washes with PBS, the bound radioactivity was counted in a gamma-counter. Control wells were incubated with dilution buffers under the same conditions.

Amino Acid Analysis. Purified 90K was electrophoresed through a 9% SDS polyacrylamϊde gel under reducing conditions using a Minigel apparatus. Proteins were electroblottcd to polyvinylideπe difluoride membrane (Immobilon; Millipore Corp., Bedford, MA, USA), were stained with Amido Black 10B (Sigma Chem. Co., St. Louis, MO), and the bands were excised. For amino acid analysis, 3-4 bands, for a total of approximately 50 μ% of 90K (as judged by staining intensity), were hydrolyzed under vacuum in 6N HC1 at 110 ^β C for 22 hours. After hydrolysis, the amino acids were analyzed on a Beckman analyzer using a pH gradient system (Hirs, C.H.W., In: Methods ofEn∑ymol. 97:3-8, Academic Press. New York, New York, USA (1983)).

Results

Purification of the 90K Antigen. The purification procedure used to isolate the 90K antigen from CG-5 tissue culture fluid, serum from a breast cancer patient, and ascitic fluid from an ovarian cancer patient is summarized in Table 1. At each step of purification, the total protein was determined, and the antigen was quantified by IRMA. Virtually all 90K activity was recovered in the 43 % ammonium sulfate precipiiate, resulting in about 4-foid enrichment. This step removed the large majority of albumin present in the initial preparation. Ammonium sulfateprecipitated-antigen was next subjected to size exclusion chromatography using a Sepharose CL-6B column (Figure 2). The 90 from all three sources was constantly found in a large peak eluting immediately behind the void volume of the column, implying that it is a high

olecular weight complex. Minor reactivity peaks of lower molecular weight were inconsistently observed which could have been due to degradation products. Low molecular weight proteins found at the end of elutioπ were unreactive. Treatment of the samples with either 6 M urea or 6 M guanidine-HCl before chromatography gave identical elution profiles (not shown). The high molecular weight peak (corresponding to fractions 21 to 28 of Figure 2) was further purified by DEAE-cellulose chromatography. The 90K antigen obtained from each of the three different sources eluted from the column at a NaCl concentration of 0.25 M (data not shown).

The final purification was accomplished by immunoaffinity on Sepharose CL-4B coupled to MAb SP-2. Bound activity was eluted with 3M MgCl2- Other eluting buffers which were used, such as glycine (pH 2.4), 1 M NaOH (pH 11.2), and 3M KSCN were less effective in antigen elution. Based on specific activity (units/μg protein), the purification of the 90K antigen from CG-5 tissue culture fluid, serum from a breast cancer patient, and ascitic fluid from an ovarian cancer patient were 84,300, 52,277 and 83,380-foid, respectively. These specific activities were calculated by measuring the 90K immunoreactivity in the 3 M M Cl2 eluate from the affinity matrix with IRMA and determining the amount of protein by comparing the silver staining intensity of the 90K band on SDS:PAGE geis with BSA standards of known concentrations.

Analysis of Purified 90K by Density Gradient Centrifugation. Samples of antigen which had been desorbed from the MAb SP-2 affinity matrix were subjected to density gradient centrifugation. This procedure did not reveal a different average buoyant density for the antigen obtained from the three different sources. The buoyant density ranged from between 1.28 g/ml to 1.31 g/ml (Figure 3). Moreover, the 90 antigen in unfractionated serum from a patient with breast cancer produced essentially an identical density profile, indicating that the 90K antigen isolated by our purification procedure did not represent a subset of the crigiπa! antigen.

PAGE and Immunoblotting Analyses of the 90K Antigen Isolated from Different Sources. In agreement with previous data (lacobelli et al.. Cancer Res. 46:3005-3010 (1986)), the 90K antigen released into the tissue culture fluid of ( ³⁵ S)methioπiπe-labeled CG-5 cells and other breast cancer cell lines migrated as a single band with an apparent molecular weight of approximately 95,000 daltons as revealed by SDS:PAGE (Figure 4A). The mobility of ( ³⁵ S)methionine-labeled antigen was identical under reducing or nonreducing conditions (with or without 2-mercaptoethanol) (Figure 4A, lane a vs. lane d) suggesting that the protein does not contain interchain disulfide bonds. Moreover, tunicamycin treatment of CG-5 cells before labelling with ( ³⁵ S)methionine did not alter the electrophoretic mobility of the 90K antigen in the cell culture fluid (Figure 4 , iane c).

Figure 4B compares the electrophoretic mobility on SDStPAGE of 90K purified from CG-5 tissue culture fluid, the serum of a breast cancer patient, and ascitic fluid from an ovarian cancer patient. Silver staining for protein clearly showed a major band wiih an apparent molecular weight of approximately 95,000 daltons. The 95K band also stained with Coomassie blue but not with periodic acid-Schiff carbohydrate staining (data not shown). Co-electrophoresis of the purified 95K antigen from the serum of a breast cancer patient detected by silver staining and of (^SJmethionine-labeled immunoprecipitates from CG-5 culture fluid detected by fluorography, gave superimposable 95K bands (data not shown).

Western blot analysis of the purified 90K antigen transferred from 4-20% pol acrylamide gel containing 0.25% NP-40 but not SDS, demonstrated the presence of similar immunoreactive diffuse bands with similar mobility from all three sources (Figure 5). By contrast, immunoblotting of the 90K antigen transferred from SDS-polyacrylamide gels reveaied very low MAb SP-2 immunoreactϊvity (data not shown). These data correlate with the Sepharose CL-6B elution profiles (Figure 2) and indicate that native 90K antigen isolated from different sources exists as a high

moiecular weight complex which is likely to be composed of Mr 95,000 subunits.

Amino Acid Analysis of 90K. Table 4 shows that the 90K antigen purified from CG-5 tissue culture fluid, the serum of a breast cancer patient, and the ascitic fluid from an ovarian cancer patient have similar amino acid compositions. The antigen was relatively rich in glutamic acid glutamine, serine, and leucine. Moreover, the sequence of the first 20 amino acids revealed a strong similarity among the antigens obtained from the three different sources. This sequence was not found in several protein data-bases such as Genebank and EMBL.

Nature of the 90K Determinant. The biochemical nature of the determinant carried on the 90 antigen was investigated using several chemical and enzymatic treatments. As Table 2 shows, exposure to methanoi strongly reduced the immunoreactivity of the 90K determinant as did exposure to 6 M guanidine-HCl, 6 M urea, 1 % SDS, lyophilization and heat. Neither reduction with dithiothreitoi and 2-mercapoethanol, nor alkylation with iodoacetamide or treatment with the nonionic detergents NP-40, Tween 20, and Triton X-100 (Sigma Chem. Co., S Louis, MO) significantly affected 90K immunoreactivity. Exposure to sodium-m-periodate had only marginal effect at high concentrations (50 mM).

To investigate the sensitivity of the 90K antigen to proteases, purified 90K was incubated with irypsin, chymotrypsin. pronase, or papaiπ, and then was analyzed by SDS: PAGE followed by silver staining. As shown in Figure 6A, all the tested proteases appeared to completely digest 90K. Analysis of residual SP-2 antibody binding confirmed that more than 80% of the initial 90K activity was lost after pronase or papain exposure whereas digestion widi trypsin or chymotrypsin appeared to be less effective (Figure 6B).

Treatment with exogiycosidases did not affect 90K immunoreactivity (Figure 6B). In fact, there was an increase in the ability of the immobilized antigen to bind ( ¹²⁵ I)labeled MAb SP-2 following treatment with

πeuraminidase and 0-galactosidase. This suggests that removal of terminal carbohydrate moieties may increase access of MAb SP-2 to the 90K determinant.

Discussion

MAb SP-2 reacts with an antigenic determinant which has been termed the 90K antigen on the basis of its apparent molecular weight of 95,000 daltons (lacobelli et al., Cancer Res. 46:3005-3010 (1986)). Here, we have described the purification of the 90K antigen from CG-5 culture fluid, the serum from a human breast cancer patient, and ascitic fluid from an ovarian cancer patient. We have found that die native 90K from each of these sources exists as a high molecular weight complex that was readily dissociated into a single 90,000 daltons species upon SDS:PAGE analysis. This suggests that the native protein represents an oligomer of several minimal subunits of 90,000 daltons. Interestingly, 90K antigen derived from each of the three sources exhibits similar behavior on size exclusion and ion-exchange chromatography, PAGE and Western blotting analyses, as well as buoyant density ultracentrifugation. Moreover, the antigen isolated from each of the three sources has similar amino acid comσosition and NH-i-terminal amino acid sequence. This indicates that the 90K antigen obtained from established long-term cancer cell lines and directly from cancer patient's serum or ascitic fluid have very similar physicochemical and immuπochemical properties.

Chemical and physical treatments of the 90K antigen were undertaken to better understand the nature of the determinant recognized by MAb SP-2. Protease digestion of the 90K antigen markedly reduced the antibody binding, providing evidence that the peptide portion of the antigen is involved in the determinant. Moreover, treatments known to denaturate most proteins also greatly reduced antibody binding, thus providing further evidence that MAb SP-2 binds to a conformational peptide determinant. Furthermore , dissociation of the oiigomeric structure of the antigen into subunits upon SDS:PAGE

resuiied in the neariy complete loss of SP-2 binding activity. These results strongly indicate that the MAb SP-2 defined determinant is proteinaceous in nature and that antibody binding is dependent upon the conformatioπal integrity of the whole antigen molecule. However, this is not a unique characteristic of the 90 antigen as other tumor-associated antigenic determinants, such as those recognized by MAb OC 125 (Davis et al. , Cancer Res. 46:6143-6148 (1986)), B72.3 (Johnson et al.. Cancer Res. 46:850-857 (1986)), and C 3 (Zhang et al.. Cancer Res. 49:6621-6628 (1989)), seem to be composed of, at least in pan, coπformationaily dependent peptide. Previously, a number of tumor-associated antigens have been reported that are eievated in the serum of patients with breast cancer. These include a series of antigens related to the human milkfat "globule" membrane family (Burchell et al., Int. J. Cancer 54:763-768 (1984); Papsidero et al.. Cancer Res. 44:4653-4657 (1984); Linsiey et al.. Cancer Res. 46:5444-5450 (1986); Kufe et al., Hybridoma 5:223-232 (1984); Hilkens et al.. In Protides of the Biological Fluids, (Peeters, H., (ed.)), pp. 651-653, Pergamon Press, Oxford, U.K. (1984); Bray et al.. Cancer Res. 47:5853-5860 (1987); Hilkens et al., In: Monoclonal Antibodies and Breast Cancer, (Ceriaπi, R.L.(ed.)), pp. 28-42, Martinus Nijhoff, Boston, MA, U.S.A. (1985); Linsiey et al.. Cancer Res. 43:2138-2148 (1988)), TAG 72 which is recognized by MAb B72.3 (Gero et al., J. Ciin. Lab. Anal. 5:360-369 (1989)), and MCA which is recognized by MAb b 12 (Bombardieri et al.. Cancer 65:490-495 (1989)). The biochemical characterization of these antigens has shown that all of them are heavily glycosylated, high molecular weight glycoproteins with mucin-like propeπies that are expressed on the surface of, and are shed or secreted by tumor cells. Comparison of these antigens with 90K indicates that the latter is distinct from the previously described antigens. This conclusion is supported by the fact that its electrophoretic migration is unaffected by neuraminidase digestion, suggesting that it is an unsialilated molecule which lacks 0-glycosidically linked oligosaccharides which are typical of mucins (data not shown) (Gahmberg et al., Eur. J. Biochem. /2 .581-586 (1982)).

Other tumor associated antigens have been described that migrate in SDS: PAGE as molecules of Mr 90,000 daltons. We have distinguished these antigens and the 90K antigen. The antigen recognized by MAb B6.2 ( ufe et al.. Cancer Res. 45:851-857 (1983); Schlom et al.. Cancer 54:2777-2794 (1984)) is a cell surface giycoprotein and, unlike 90 , is highly restricted to breast cancer cells. The melanoma-associated antigen termed p97, gp87, or gp95 (Brown et al., J. Immunol. 227:539-546 (1981); Dippold et al., Proc. Natl. Acad. Sci. USA 77:6114-6118 (1980); Liao et al., J. Cell. Biochem. 27:303-316 (1985)) is a membrane protein which is structurally related to transferrin (Brown et al.. Nature 296:171-173 (1982)). Another melanoma antigen, FD, is also a surface giycoprotein the expression of which is restricted to a very limited number of cells (Manes et al.. Cancer Res. 47:6614-6619 (1987)). Finally, the antigen defined by MAb 3G2-C6 (Zhang et al.. Cancer Res. 49:6621-6628 (1989)) is a surface component which is expressed in a significant number of bladder cancers but only marginally in breast cancer (Young et al.. Cancer Res. 45:4439-4446 (1985)).

Example 2 Cloning Of The 90K Gene

End terminal sequencing of the 90K antigen revealed the following amino acid sequence (SEQ ID NO:3): Val Asn Asp Gly Asp Met Arg Leu Ala Asp Gly Gly Ala Thr Asn Gin Gly Arg Val G!u He Phe. Based on this amino acid sequence, a "guessmer" of 66 nucleotides was designed on the basis of codoπ usage frequencies (Lathe, ]., Mol Biol 183:1-12 (1985)) using theamino-terminal sequence: VNDGDM(S)LADGGATNQGRVEIF (SEQ ID NO:4). The nucleotide sequence (SEQ ID NO:5) utilized was as follows: 5' GTG AAT GAT GGC GAC ATG TCC CTG GCT GΛT GGC GGC GCC ACC AAC CAG GGC CGG GTG GAG ATC TTC 3'.

-29-

The guεss er or nucleic acid probe was ² P end-labeled and was used to screen a λgtlO library prepared from MCF7 polyA ⁺ RNA (complexity: SxlQ ⁵ ). Techniques of nucleic acid hybridization in clone identification are disclosed by Maniatis et al. and Sambrook et al. (both entitled: Molecular Cloning, A Laboratory Manual, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, New York (1982 and 1989, respectively)) and by Hames et al., in Nucleic Acid Hybridization, A Practical Approach, IRL Press, Washington, D.C. (1985), which references are herein incorporated by reference. Positive phages were isolated including two EcoRl inserts of — 1,200

*bp and ~ 900 bp. The complete insert was then cloned utilizing the EcoRl partial inserts. The DNA fragments were cloned into the Bluescript* plasmid (Stratagene, La Jolla, CA). The insert size was approximately 2,206 nucleotides. Sequence analyses of the original clones and subclones were performed according to the methods of Sanger et al. (Proc. Natl. Acad. Sci. USA 74:5463 (1977)) and Maxa et al. (Proc. Natl. Acad. Sci. USA 74:560 (1977)).

Tne protein sequence was revealed to be 585 amino acids, 1,755 nucleotides. A 5 ' leader of 131 nucleotides and a 3' trailer of 320 nucleotides was found. The complete nucleotide and projected amino acid sequence is given in Figure 1 (SEQ ID NO:l AND SEQ ID NO:2, respectively ⁾ . Included in Table 3 are Northern blot analyses of RNAs from tumors and normal tissues.

Example 3 Cell Culture and Stable Expression of the 90K Antigen

Materials and Methods

Construction of an IR-95 Expression Plasmid. Using standard protocols, a 2147 bp ClaVXhol cDNA-fragment was subcloned into the eukaryotic, cytomegalovirus promoter-based expression vector (pCMVNEO- IR95) (Figure - 8) containing expression units for mouse dihydrofolate reductase (DHFR) cD A and the bacterial neomycin phosphotransferase (neo) gene for amplification and selection, respectively. Cell Culture. Human BT-20 breast tumor cells (American Type

Culture Collection, Rockville, MD, USA, Deposit Number HTB 19) were grown in RPMI 1640 (GIBCO, Gaithersburg, MD) supplemented with 3% FCS, 2 mM L-gluiamine and antibiotics in a humidified CO ₂ incubator. Selection for neomycin resistance after electroporation of thepCMVNEO-IR95 plasmid was performed in the same medium.

Electroporation. Exponentially growing BT 20 cells were washed twice with PBS, were harvested by trypsiπization and were pelleted. The pellet was washed three times with PBS. The cells were resuspeπded in PBS at a concentration of approximately 5 x 10° cells/ml. Electroporation was performed with the Gene Pulser Transfecrion apparatus from Bio-Rad Laboratories, Segrate, Italy. For stable expression, 0.8 ml of cell suspension was mixed with 20 μg of linearized plasmid DNA and 50 μg of sheared Salmon sperm DNA in an electroporation cuvette. A single pulse of increasing field strength (240-270 V) was delivered from a 500 μF capacitor at room temperature. After the pulse and a 10 minute incubation on ice, the ceils were transferred to the non-selective media as above. The Trypan blue exclusion test was used for determining the viability of the cells at 10 minutes after electroporation during the mock electroporations.

Selection and Amplification. Two days after electroporation, the cells were passaged into selective medium containing Geπeticin (G418, GIBCO. Gaithersburg, MD) at 400 μg/ml. Clones were picked using metal cloning cylinders with petroleum jelly for the bottom seal. The clones were expanded and cultured in 12 well clusters (Costar, Cambridge, MA) in Alpha-MEM (GIBCO, Cat. #072-01900A) containing 3% FCS, glutamine (2 M) and antibiotics plus methotrexate (Sigma Chemical Co., St. Louis, MO, U.S.A.) at concentrations of 10 and 50 μM. After methoσexate selection, the cells were cultured in DMEM high glucose (GIBCO, Gaithersburg, MD) supplemented with 3% FCS, 2 M glutamine, 50 μg/ml Geπtamicin and 1 μM Methotrexate.

^SjMethionine Labeling andlmmunoprecipitation. Subconfluent cells in 6 well clusters (Nunc) were washed with 1 ml of PBS twice and were grown overnight in 1 ml of ethioniπe free DMEM/0.5% ULTROSOR-G containing 50 μCi (1 Ci = 37 GBq) of ( ³⁵ S)methionine. For immunoprecipitation, conditioned media was briefly spun and was mixed with

1 μg/ml aprotinin and 1 μg/ml leupeptin. Protein A-Sepharose (Pharmacia, Uppsala, Sweden) was washed thrice with PBS and 30 μl (1:1) suspension mixed with 2 μg of MAb SP-2 and was incubated for 30 minutes at room temperature. The protein A-Sepharose-SP-2 complex was washed three times with HNTG buffer (20 mM HEPES, pH 7.5/150 M NaCl/ 10% glycerol/0.1 % Triton X-100) and was incubated with conditioned media for

2 hours at 4°C. Protein A-Sepharose beads were washed three times with HNTG buffer. Moist beads were suspended in 30 μl of 1 x SDS gel-loading buffer, were boiled for 3 minutes at 100 ^β C and were immediately chilled on ice. The proteins were separated on 10% SDS-polyacrylamide gel and were aπalvzcd bv autoradiography.

Results

For expression of this protein, a cDNA coding for the entire 585-amino acid polypeptide was placed under the transcriptional control of the cytomegaioviπis early promoter. In addition, the expression vector contained the neo resistance gene, which conferred cellular resistance to the aminoglycoside antibiotic G418 and therefore allowed selection of primary transfectaπts, as well as the DHFR gene for methotrexate resistance, which was used to select for cells containing amplified transfected DNA sequences. Bacterial plasmid sequences, including an origin of replication and the gene for ampicillin resistance, allowed replication of the entire expression plasmid in E. coli. Figure 9 shows the autoradiogram of immunoprecipitates of the first three stable clones. The intensities of the bands are reflective of the relative amounts of protein secreted by each clone.

Example 4 Transient Expression of the 90K Antigen

Materials and Methods

Construction of Expression Plasmid. Using standard protocols (Sambrook et al, Molecular Cloning. A Laboratory Manual 2nd Edition, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY, USA (1989) Vols. 1-3) the expression plasmid was constructed by introducing a 2147 bp Cla (position 726 in Bluescript II KS - Xho (position 2118 in Figure 1) restriction fragment into the eukaryotic, cytomegaioviπis promoter-based expression vector pCMV (Figure 7).

Transient Expression. Human embryonic kidney 293 fibroblasts (American Type Culture Collection, Rockviile, MD. USA, Deposit Number CRL 1573) were grown in DMEM containing 10% FCS and antibiotics.

One day prior to transfection, 2 x 10 ⁵ ceils were seeded into each well of a six-well dish. Traπsfections were carried out according to the protocol of Chen and Okayama Λfo/. Cell Biol 7:2745-2752 (1987) with a total of 4 μg of CsCI gradient-purified plasmid-DN A/well. Sixteen hours after the addition of precipitates, the cells were washed once with DMEM, and fresh growth medium was added.

Metabolic Labeling. For metabolic labeling, the ceils were grown overnight with ( ⁵ S)methionine (50 μCi/ml) in methionine-free DMEM (0.5 ml/well) containing 1 % dialyzed FCS. Tunicamycin Treatment. For blocking the formation of protein

N-glycosidic linkages, tunicamycin was added to the medium at a final concentration of 0.1 to 1.0 μg/ml for 16 hours.

Cell Lysis and lmmunoprecipitation. The ceils were lysed on ice with

0.3 ml of lysis buffer containing 50 mM HEPES, pH 7.5, 150 mM NaCl, 1.5 mM MgCl ₂ , 1 mM EGTA. 10% glycerol, 1 % Triton X-100, 2 mM phenyimethylsulfonyl fluoride (PMSF), 200 units/ml aprotinin. 10 mM sodium pyrophosphate, and 10 μg/mi leupeptin. The iysates were transferred to microfuge tubes, were voπexed for 10 seconds, and were precleared by centrifugation at 12,500 rpm for 15 minutes at 4°C. For immunoprecipitation, 10 μl of protein A-scpharose (swollen and prewashed in 20 mM HEPES, pH 7.5) and 1 μg MAb SP-2 was added to the cleared lysate and incubated at 4 ^W C for 3 hours. The conditioned medium was used for immunoprecipitation after adding aprotinin (200 units/ml) and PMSF

(2 mM final) and preclearing by centrifugation. Precipitates were washed three times with 1 ml of washing buffer (lysis buffer with 0.1 % Triton

X-100). SDS-sample buffer was added, the samples were boiled and were loaded on SDS-PAGE for the separation of precipitated proteins.

Results

Cells of the transformed 293 cell line were placed into six-well dishes and were transfected widi the CMV-expression construct as described above (Figure 10: lanes 1-8). Control cells were transfected with the inseπless plasmid pCMV (Figure 10: lanes 7 and 8).

Sixteen hours prior to cell lysis the growth medium was exchanged for labeling medium which contained 50 μCi/ml ( ³⁵ S)πwthionine. For the same incubation period tunicamycin was added at a final concentration of 0.1 μg/ml (Figure 10: lanes 3 and 4) or 1.0 μg/ml (Figure 10: lanes 5 and 6). Both the cell lysate (L) and the conditioned medium (M) were used for ϊmmunoprecipitations with MAb SP-2. Precipitated proteins were separated on a 8.5% SDS-PAGE. Figure 10 shows the autoradiograph of a 20 hour exposure of the dried gel.

Immunoprecipitation with MAb SP-2 from the conditioned media of the adenovirus type 5-(Ad 5)-ιraπsformed cell line 293 resulted in the appearance of a single band at 95 Kd (lane 8). A corresponding signal was not detectable (lane 9) in immunoprecipitates of the ceil lysate.

Using the conditioned media, transiently expressing cells (cells transfected wi± the CMV-expression plasmid carrying the cDNA-ϊnsert) resulted in a several fold increase in signal intensity of the 95 kd band (Figure 10: lane 2). At the same time, a protein of approximately 77 kd was detectable in immunoprecipitates of the corresponding cell lysates (Figure 10: lane 1). Tunicamycin treatment of transiently expressing cells reduced the signal intensity for both the 95 kd protein (lanes 4 and 6) and the 77 kd protein (lanes 3 and 5). The lunicamycin effect was dose dependent.

-35-

lixample 5 Purification of IR-95

IR-95 was also purified using the thiophilic sepharose chromatography method described below.

Materials

Thiophilic Sepharose (AFFI-T)

Metal Chelate Sepharose

Protein A- Sepharose

Amm. Sulphate Sod. Sulphate

Copper Sulphate

Glycine

Sod. Phosphate, Dibasic Anhydrous

Potassium Chloride Sod. Chloride

Hank's balanced salt solution (GIBCO)

Buffers

1. Buffer A: For 1 litre: Sod. Chloπde 13 gm, POL Chloride 0.2 gm, Sod. Phosphate Dibasic. Anhydrous 1.6 gm, Sod. Sulphate 0.5 M and EDTA, 1 M pH of the buffer titrated to 8.2.

2. Buffer B: For 1 litre: Sod. Chloride 13 gm, Pot. Chloride 0.2 g . Sod. Phosphate Dibasic, Anhydrous l.ό gm. Sod. Sulphate 0.3 M and EDTA, 1 mM pH of the buffer titrated to 8.2.

3. Buffer C: For 1 litre; Soα. Chloπde 13 gm, Pυt. Chlorid 0.2 gm. Sod. Phosphate Dibasic. Anhydrous 1.6 gm, and EDTA. 1 mM pH of the solution titrated to 8.2.

4. Buffer D: For 1 litre; Sod. Phosphate Dibasic, Anhydrous 7.098 gm and Sod. Chloride 5.8 gm pH of the solution titrated to 8.

.5. Buffer E: For 1 litre; Sod. Phosphate Dibasic, Anhydrous 7.098 gm, Glycine 100 mM and Sod. Chloride 5.8 gm pH of the solution titrated to 8.

Step 1: Thiophilic Sepharose Chromatography

Thiophilic Sepharose chromatography consisted of the following steps:

A- Ammonium Sulphate Precipitation. Preciarified conditioned medium was concentrated ten fold on a hollow fibre ultrafilteration cartridge (40 KD, Nunc). Concentrated medium was precipitated with solid ammonium sulphate to 42% saturation (assuming the maximum saturation at 533 gm litre). Ammonium sulphate was added slowly and pH was titrated back to approximately 8.0 by using dilute ammonium hydroxide. Let the solution stir overnight. In case the conditioned media is not concentrated, the precipitation should be done with solid A m. sulphate to 42% saturation.

B- Centrifugation. Ammonium sulphate precipitate was centrifuged at 8000 rpm in a GS3 rotor (Sorvall). The supernatant was discarded and the pellet was dissolved using a 10X volume in buffer A. C- Thiophilic Sepharose Batch Elution. The required volume of the thiophilic sepharose (Kem-En-Tec, Copenhagen, Denmark) was extensively washed with water on a sintered glass funnel using mild suction (removes the sodium azide). The matrix was aspirated until the cracks appeared in the bed. Five bed volumes of buffer A was then passed through it while stirring lightly with a glass rod to get ride of the trapped air in the matrix- The protein solution from the previous step was passed through the matrix under mild suction without letting it dry. The protein solution was recycled three times. The matrix was washed with 50 to 100 bed volumes with buffer A with

occasional stirring. The matrix was then washed with 50 to 100 volumes of buffer B with occasional stirring without letting it dry. The thiophilic sepharose was eluted with 10 bed volumes of buffer C adding one bed volume at a time and lastly with sterile water. After the last bed volume was added, die matrix was aspirated to dryness.

The eluates were pooled and precipitated with 70 % ammonium sulphate and stirred for at least four hours in the cold room. The precipitate was collected by centrifugation at 10000 rpm and dissolved in buffer D.

D. Dialysis. The protein solution was dialysed against buffer D for at least four hours in the cold room with two changes of buffer.

Step 2: Metal Ch elate Chromatography

The metal chelate chromatography was carried out as described below:

Equiliberation and Column Elution. Metal chelate sepharose (Pharmacia) was packed in a glass column under gravity to a packed voiu e of 4 ml. Matrix was washed extensively with water to remove ethanol. A copper sulphate solution (10 mg per ml) was passed over the matrix. Normally 10 ml of the copper sulphate solution is enough for lading of the matrix. The matrix was again washed with 10 to 20 column volumes of water to remove the excess copper sulphate. Then the matrix was washed with 10 column volumes of buffer E and equiliberaied with 20 column volumes of buffer D.

The dialysed protein solution was ceπirifuged at 10000 rpm to get rid of the coagulated protein. The protein solution was diluted five fold in the equiliberation buffer and passed over the matrix twice. The matrix washed with 50 column volumes of the equiliberation buffer and protein was eluted using a linear gradient of 20 column volumes each of buffer D and buffer E at a flow rate of 1 ml per minute. Normally, the protein elutes from the column in the second peak. Active fractions were pooled and concentrated on

-38-

Ceπtricon-30. The activity of purified protein was checked by immunoprecipitation.

Step 3: Immunoprecipitution

Purified protein was checked for its ability to be immunoprccipiiated with SP-2 monoclonal antibody. 50 μl of 1:1 suspension of Protein A- Sepharose was washed three times with one ml of buffer C by brief spinning and aspirations. Two μg of SP-2 MAb plus protein sample were rotated for two hours in the cold room. The beads were washed three times with one ml of buffer C by repeated centrifugation and aspirations. In the end, the beads were aspirated and moist beads lysed in IX Laemeli buffer and electrophoresed.

Step 4: Storage

The purified protein was buffer exchanged and concentrated with Hank's balanced salt solution using Centricon-30 to 2-3 mg/ml and mixed with one volume of 2 M glucose before freezing at -20 degrees.

Example 6

Enhancement of Natural Killer (NK) and Lymphokinε Activated Killer (LAK) Cell Aci diy

Peripheral blood mononuclear cells (PBL) were isolated from fresh heparinized blood by Ficoll-Hypaque gradient centrifugation after partial depletion of monocytes by adherence to plastic surfaces (45 min. 37°C). PBL at the concentration of 2xl0 ^δ cells/mi were cultured in RMPI-1640 medium supplemented with 10% heat-inactivated fetal calf serum and antibiotics. Purified IR-95 was added in various concentrations (50 ng/ml to 2000 ng/ ^' ml) for 16 h. As a control , PBL were incubated in the same culture conditions for

the same period of time without IR-95. At the end of the incubation period, cells were washed and tested as effector cells in the short term (4 h) ^4l Cr- release cytotoxicity assay (Coligan, J.E. et al. Current Protocols in Immunology, Green Publishing Associates and Wiley Interscience, New York (1992)) against target cells, i.e. K562 cells for NK activity and Daudi cells for LAK activity at an effectoπtarget ratio of 1:40. Data points are averages of five different experiments performed in quadruplicate. Spontaneous ⁵¹ Cr- release was 15 % of the total in all cases. IR-95 at concentrations in the range of 500-2000 ng/ml for 16 hours markedly increases both NK and LAK cytotoxic activity (Figure 11).

All publications and patent applications mentioned in this specification are indicative of the level of skill of one in the art to which this invention pertains. All publications and patent applications are hereby incorporated by reference to the same extent as if each individual publication or patent application was specifically and individually indicated to be incorporated by reference.

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 die scope of the appended claims. Modifications of the above-described modes for carrying out the invention that are obvious to persons of skill in the an, such as those in the fields of medicine, immunoiogy, hybridoma technology, pharmacology, and/or related fields, are intended to be within the scope of the following claims.

-40-

Circulating serum IR-95 concentrations (unit/ml) were determined by asolid-phase. enzyme-linked, immunoabsorbent procedure that uses mAb SP-2 as the coating antibody. Levels of more man 1.75 units/ml (normal mean +/- 2SD) were considered positive determinations. The serum level of IR-95 was not affected by sex and blood group.

A total of 214 serum samples were obtained from the following categories of patients attending the Chien U iversity Hospital: Hepann's B virus infection (69 cases). Epstein Barr virus infection (21 eases), autoimmune disease (15 rheumatoid arthritis. 7 systemic lupus erythematosus- 6 auteiπunune uveins). hemodtalysis (19 cases). Down syndrome (12 cases). In addition, serum samples were obtained from 18 women at different periods of gestation and 29 apparently healthy subject of more than S5 yean, of age.

Cut off value of serum IR-95 is 1.7 units/ml (mean +t- 2 SD).

AH means for different groups of subjects were significantly greater than those for healthy controls (p — O.OOCl . analysis of variance).

SEQUENCE LISTING

(1) GENERAL INFORMATION:

(i) APPLICANT: lacobelli, Stefano Natoli, Clara Schlessinger, Joseph

(ii) TITLE OF INVENTION: A 90K Tumor-Associated Antigen, IR-95

(iii) NUMBER OF SEQUENCES: 5

(IV) CORRESPONDENCE ADDRESS:

(A) ADDRESSEE: Sterne, Kessler, Goldstein & Fox

(D) STREET: 1225 Connecticut Avenue

(C) CITY: Washington

(D) STATE: D.C.

(E) COUNTRY: U.S.A.

(F) ZIP: 20036

(v) COMPUTER READABLE FORM:

(A) MEDIUM TYPE: Floppy disk

(B) COMPUTER: IBM PC compatible

(C) OPERATING SYSTEM: PC-DOS/MS-DOS

(D) SOFTWARE: Patentin Release #1.0, Version #1.25

(vi) CURRENT APPLICATION DATA:

(A) APPLICATION NUMBER: US (to be assigned)

(B) FILING DATE: Herewith

(C) CLASSIFICATION:

(Vii) PRIOR APPLICATION DATA:

(A) APPLICATION NUMBER: IT RM 92A000099

(B) FILING DATE: 17-FEE-1992

(ΪX) TELECOMMUNICATION INFORMATION:

(A) TELEPHONE: (20?.) 466-0800

(B) TELEFAX: (202) S33-8716

(2) INFORMATION FOR SEQ ID NO:l:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 2206 base pairs

(B) TYPE: nucleic acid

(C) STRANDEDNESS: single

(D) TOPOLOGY: linear

(ix) FEATURE:

(A) NAME/KEY: CDS

(B) LOCATION: 132.-1886

(xi) SEQUENCE DESCRIPTION: SEQ ID NO:l:

CACGCTCCAT ACTGGGAGAG GCTTCTGGGT CAAAGGACCA GTCTGCAGAG GGATCCTGTG

GCTGGAAGCG AGGAGGCTCC ACACGGCCGT TGCAGCTACC GCAGCCAGGA TCTGGGCATC

CAGGCACGGC C ATG ACC CCT CCG AGG CTC TTC TGG GTG TGG CTG CTG GTT Met T r Pro Pro Arg Leu Phe Trp al Trp Leu Leu Val 1 5 10

GCA GGA ACC CAA GGC GTG AAC GAT GGT GAC ATG CGG CTG GCC GAT GGG Ala Gly Thr Gin Gly Val Asn Asp Gly Asp Met Arg Leu Ala Asp Gly 15 20 25

GGC GCC ACC AAC CAG GGC CGC GTG GAG ATC TTC TAC AGA GGC CAG TGG Gly Ala Thr Asn Gin Gly Arg Val Glu He Phe Tyr Arg Gly Gin Trp 30 35 40 45

GGC ACT GTG TGT GAC AAC CTG TGG GAC CTG ACT GAT GCC AGC GTC GTC Gly Thr Val Cys Asp Asn Leu Trp Asp Leu Thr Asp Ala Ser Val Val

50 55 60

TGC CGG GCC CTG GGC TTC GAG AAC GCC ACC CAG GCT CTG GGC AGA GCT Cys Arg Ala Leu Gly Phe Glu Asn Ala Thr Gin Ala Leu Gly Arg Ala 65 70 75

GCC TTC GGG CAA GGA TCA GGC CCC ATC ATG CTG GAC GAG GTC CAG TGC Ala Phe Gly Gin Gly Ser Gly Pro He Met Leu Asp Glu Val Gin Cys 80 85 90

ACG GGA ACC GAG GCC TCA CTG GCC GAC TGC AAG TCC CTG GGC TGG CTG

Thr Gly Thr Glu Ala Ser Leu Ala Asp Cys Lys Ser Leu Gly Trp Leu 95 100 105

AAG AGC AAC TGC AGG CAC GAG AGA GAC GCT GGT GTG GTC TGC ACC AAT Lys Ser Asn Cys Arg His Glu Arg Asp Ala Gly Val Val Cys Thr Asn

110 115 120 125

GAA ACC AGG AGG CAC CCA CAC CCT GGA CCT CTC CAG GGA GCT CTC GGA Glu Thr Arg Arg His Pro His Pro Gly Pro Leu Gin Gly Ala Leu Gly

130 125 140

GCC CTT GGC CAG ATC TTT GAC AGC CAG CGG GGC TGC GAC CTG TCC ATC Ala Leu Gly Gin He Phe Asp Ser Gin Arg Gly cys Asp Leu Ser He

145 150 155

AGC GTG AAT GTG CAG GGC GAG GAC GCC CTG GGC TTC TGT GGC CAC ACG Ser Val Asn Val Gin Gly Glu Asp Ala Leu Gly Phe Cys Gly His Thr 160 165 170

GTC ATC CTG ACT GCC AAC CTG GAG GCC CAG GCC CTG TGG AAG GAG CCG Val He Leu Thr Ala Asn Leu Glu Ala Gin Ala Leu Trp Lys Glu Pro 175 180 185

GGC AGC AAT GTC ACC ATG AGT GTG GAT GCT GAG TGT GTG CCC ATG GTC Gly Ser Asn Val Thr Met Ser Val Asp Ala Glu Cys Val Pro Met Val 190 195 200 205

AGG GAC CTT CTC AGG TAC TTC TAC TCC CGA AGG ATT GAC ATC ACC CTG Arg Asp Leu Leu Arg Tyr Phe Tyr Ser Arg Arg He Asp He Thr Leu

210. 215 220

TCG TCA GTC AAG TGC TTC CAC AAG CTG GCC TCT GCC TAT GGG GCC AGG Ser Ser Val Lys Cys Phe His Lys Leu Ala Ser Ala Tyr Gly Ala Arg 225 230 235

CAG CTG CAG GGC TAC TGC GCA AGC CTC TTT GCC ATC CTC CTC CCC CAG Gin Leu Gin Gly Tyr Cys Ala Ser Leu Phe Ala He Leu Leu Pro Gin 240 245 250

GAC CCC TCG TTC CAG ATG CCC CTG GAC CTG TAT GCC TAT GCA GTG GCC Asp Pro Ser Phe Gin Met Pro Leu Asp Leu Tyr Ala Tyr Ala Val Ala 255 260 265

ACA GGG GAC GCC CTG CTG GAG AAG CTC TGC CTA CAG TTC CTG GCC TGG Thr Gly Asp Ala Leu Leu Glu Lys Leu Cys Leu Gin Phe Leu Ala Trp 270 275 280 285

AAC TTC GAG GCC TTG ACG CAG GCC GAG GCC TGG CCC AGT GTC CCC ACA 1 Asn Phe Glu Ala Leu Thr Gin Ala Glu Ala Trp Pro Ser Val Pro Thr

290 295 300

GAC CTG CTC CAA CTG CTG CTG CCC AGG AGC GAC CTG GCG GTG CCC AGC 1 Asp Leu Leu Gin Leu Leu Leu Pro Arg Ser Asp Leu Ala Val Pro Ser 305 310 315

GAG CTG GCC CTA CTG AAG GCC GTG GAC ACC TGG AGC TGG GGG GAG CGT Glu Leu Ala Leu Leu Lys Ala Val Asp Thr Trp Ser Trp Gly Glu Arg 320 325 330

GCC TCC CAT GAG GAG GTG GAG GGC TTG GTG GAG AAG ATC CGC TTC CCC Ala Ser His Glu Glu Val Glu Gly Leu Val Glu Lys He Arg Phe Pro 335 340 345

ATG ATG CTC CCT GAG GAG CTC TTT GAG CTG CAG TTC AAC CTG TCC CTG Met Met Leu Pro Glu Glu Leu Phe Glu Leu Gin Phe Asn Leu Ser Leu 350 355 360 365

TAC TGG AGC CAC GAG GCC CTG TTC CAG AAG AAG ACT CTG CAG GCC CTG Tyr Trp Ser His Glu Ala Leu Phe Gin Lys Lys Thr Leu Gin Ala Leu

370 375 380

GAA TTC CAC ACT GTG CCC TTC CAG TTG CTG GCC CGG TAC AAA GGC CTG

Glu Phe His Thr Val Pro Phe Gin Leu Leu Ala Arg Tyr Lys Gly Leu 385 390 395

AAC CTC ACC GAG GAT ACC TAC AAG CCC CGG ATT TAC ACC TCG CCC ACC 1 Asn Leu Thr Glu Asp Thr Tyr Lys Pro Arg He Tyr Thr Ser Pro Thr

400 405 410

TGG AGT GCC TTT GTG ACA GAC AGT TCC TGG AGT GCA CGG AAG TCA CAA Trp Ser Ala Phe Val Thr Asp Ser Ser Trp Ser Ala Arg Lys ser Gin 415 420 425

CTG GTC TAT CAG TCC AGA CGG GGG CCT TTG GTC AAA TAT TCT TCT GAT 1 Leu Val Tyr Gin Ser Arg Arg Gly Pro Leu Val Lys Tyr Ser Ser Asp 430 435 440 445

TAC TTC CAA GCC CCC TCT GAC TAC AGA TAC TAC CCC TAC CAG TCC TTC Tyr Phe Gin Ala Pro Ser Asp Tyr Arg Tyr Tyr Pro Tyr Gin Ser Phe

450 455 4Gϋ

CAG ACT CCA CAA CAC CCC AGC TTC CTC TTC CAG GAC AAG AGG GTG TCC ₁ Gin Thr Pro Gin His Pro Ser Phe Leu Phe Gin Asp Lys Arg Val Ser 465 470 475

TGG TCC CTG GTC TAC CTC CCC ACC ATC CAG AGC TGC TGG AAC TAC GGC 1 Trp Ser Leu Val Tyr Leu Pro Thr He Gin Ser Cys Trp Asn Tyr Gly 480 485 490

TTC TCC TGC TCC TCG GAC GAG CTC CCT GTC CTG GGC CTC ACC AAG TCT 1 Phe Ser Cys Ser Ser Asp Glu Leu Pro Val Leu Gly Leu Thr Lys Ser 495 500 505

GGC GGC TCA GAT CGC ACC ATT GCC TAC GAA AAC AAA GCC CTG ATG CTC 1 Gly Gly Ser Asp Arg Thr He Ala Tyr Glu Asn Lys Ala Leu Met Leu 510 515 520 525

TGC GAA GGG CTC TTC GTG GCA GAC GTC ACC GAT TTC GAG GGC TGG AAG 1 Cys Glu Gly Leu Phe Val Ala Asp Val Thr Asp Phe Glu Gly Trp Lys

530 535 540

GCT GCG ATT CCC AGT GCC CTG GAC ACC AAC AGC TCG AAG AGC ACC TCC 1 Ala Ala He Pro Ser Ala Leu Asp Thr Asn Ser Ser Lys Ser Thr Ser 545 550 555

TCC TTC CCC TGC CCG GCA GGG CAC TTC AAC GGC TTC CGC ACG GTC ATC 1 Ser Phe Pro Cys Pro Ala Gly His Phe Asn Gly Phe Arg Thr Val He 560 565 570

CGC CCC TTC TAC CTG ACC AAC TCC TCA GGT GTG GAC TAGACGCGTG 1

Arg Pro Phe Tyr Leu Thr Asn Ser Ser Gly Val Asp 575 530 585

GCCAAGGGTG GTGAGAACCG GAGAACCCCA GGACGCCCTC ACTGCAGGCT CCCCTCCTCG 1

GCTTCCTTCC TCTCTGCAAT GACCTTCAAC AACCGGCCAC CAGATGTCGC CCTACTCACC 2

TGAGGCTCAG CTTCAAGAAA TTACTGGAAG GCTTCCACTA GGCTCCACCA GGAGTTCTCC

CACCACCTCA CCAGTTTCCA GGTGGTAAGC ACCAGGAGGC CCTCGAGGTT GCTCTGGATC

CCCCCACAGC CCCTGGTCAG TCTGCCCTTG TCACTGGTCT GAGGTCATTA AAATTACATT

CAGGTTCCTΛ

(2) INFORMATION FOR SEQ ID NO:2:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 585 amino acids

(B) TYPE: amino acid (D) TOPOLOGY: linear

(ii) MOLECULE TYPE: protein

(Xi) SEQUENCE DESCRIPTION: SEQ ID NO:2:

Met Thr Pro Pro Arg Leu Phe Trp Val Trp Leu Leu Val Ala Gly Thr

1 10 15

Gin Gly Val Asn Asp Gly Asp Met Arg Leu Ala Asp Gly Gly Ala Thr 20 25 30

Asn Gin Gly Arg Val Glu He Phe Tyr Arg Gly Gin Trp Gly Thr Val 35 40 45

Cys Asp Asn Leu Trp Asp Leu Thr Asp Ala Ser Val Val Cys Arg Ala 50 55 60

Leu Gly Phe Glu Asn Ala. Thr Gin Ala Leu Gly Arg Ala Ala Phe Gly 65 70 75 80

Gln Gly Ser Gly Pro He Met Leu Asp Glu Val Gin Cys Thr Gly Thr

05 90 ₉₅

Glu Ala Ser Leu Ala Asp Cys Lys Ser Leu Gly Trp Leu Lys Ser Asn 100 105 HO

Cys Arg His Glu Arg Asp Ala Gly Val Val Cys Thr Asn Glu Thr Arg 115 120 125

Arg His Pro His Pro Gly Pro Leu Gin Gly Ala Leu Gly Ala Leu Gly 330 135 140

Gin He Phe Asp Ser Gin Arg Gly Cys Asp Leu Ser He Ser Val Asn

145 150 155 160

Val Gin Gly Glu Asp Ala Leu Gly Phe Cys Gly His Thr Val He Leu

165 170 175

Thr Ala Asn Leu Glu Ala Gin Ala Leu Trp Lys Glu Pro Gly Ser Asn 180 185 190

Val Thr Met Ser Val Asp Ala Glu Cys Val Pro Met Val Arg Asp Leu 195 ^" 200 205 .

Leu Arg Tyr Phe Tyr Ser Arg Arg He Asp He Thr Leu Ser Ser Val 210 215 220

Lys Cys Phe His Lys Leu Ala Ser Ala Tyr Gly Ala Arg Gin Leu Gin 225 230 235 240

Gly Tyr Cys Ala Ser Leu Phe Ala He Leu Leu Pro Gin Asp Pro Ser

245 250 255

Phe Gin Met Pro Leu Asp Leu Tyr Ala Tyr Ala Val Ala Thr Gly Asp 260 265 270

Ala Leu Leu Glu Lys Leu Cys Leu Gin Phe Leu Ala Trp Asn Phe Glu 275 230 285

Ala Leu Thr Gin Ala Glu Ala Trp Pro Ser Val Pro Thr Asp Leu Leu 290 295 300

Gin Leu Leu Leu Pro Arg Ser Asp Leu Ala Val Pro Ser Glu Leu Ala

305 310 315 320

Leu Leu Lys Ala Val Asp Thr Trp Ser Trp Gly Glu Arg Ala Ser His

325 330 335

Glu Glu Val Glu Gly Leu Val Glu Lys He Arg Phe Pro Met Met Leu 340 345 350

Pro Glu Glu Leu Phe Glu Leu Gin Phe Asn Leu Ser Leu Tyr Trp Ser 355 360 365

His Glu Ala Leu Phe Gin Lys Lys Thr Leu Gin Ala Leu Glu Phe His 370 375 380

Thr Val Pro Phe Gin Leu Leu Ala Arg Tyr Lys Gly Leu Asn Leu Thr 385 390 395 400

Glu Asp Thr Tyr Lys Pro Arg He Tyr Thr Ser Pro Thr Trp Ser Ala

405 410 415

Phe Val Thr Asp Ser Ser Trp Ser Ala Arg Lys Ser Gin Leu Val Tyr 420 425 430

Gin Ser Arg Arg Gly Pro Leu Val Lys Tyr Ser Ser Asp Tyr Phe Gin 435 440 445

Ala Pro Ser Asp Tyr Arg Tyr Tyr Pro Tyr Gin Ser Phe Gin Thr Pro 450 455 460

Gin His Pro Ser Phe Leu Phe Gin Asp Lys Arg Val Ser Trp. Ser Leu 465 470 475 480

Val Tyr Leu Pro Thr He Gin Ser Cys Trp Asn Tyr Gly phe Ser Cys

485 490 495

Ser Ser Asp Glu Leu Pro Val Leu Gly Leu Thr Lys Ser Gly Gly Ser 500 505 510

Asp Arg Thr He Ala Tyr Glu Asn Lys Ala Leu Met Leu Cys Glu Gly 515 520 525

Leu Phe Val Ala Asp Val Thr Asp Phe Glu Gly Trp Lys Ala Ala He 530 525 540

Pro Ser Ala Leu Asp Thr Asn Ser Ser Lys Ser Thr Ser Ser Phe Pro

545 550 555 560

Cys Pro Ala Gly His Phe Asn Gly Phe Arg Thr Val He Arg Pro Phe

565 570 575

Tyr Leu Thr Asn Ser Ser Gly Val Asp 5S0 585

(2) INFORMATION FOR SEQ ID NO:3:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 22 amino acids

(B) TYPE: amino acid (D) TOPOLOGY: linear

(Xi) SEQUENCE DESCRIPTION: SEQ ID NO:3:

Val Asn Asp Gly Asp Met Arg Leu Ala Asp Gly Gly Ala Thr Asn Gin

1 5 10 15

Gly Arg Val Glu He Phe 20

(2) INFORMATION FOR SEQ ID NO:4:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 22 amino acids

(B) TYPE: amino acid (D) TOPOLOGY: linear

(xi) SEQUENCE DESCRIPTION: SEQ ID NO: :

Val Asn Asp Gly Asp Met Ser Leu Ala Asp Gly Gly Ala Thr Asn Gin 1 5 10 15

Gly Arg Val Glu He Fhe

20

(2) INFORMATION FOR SEQ ID NO:5:

(ϊ) SEQUENCE CHARACTERISTICS: (A) LENGTH: 66 base pairs (E) TYPE: nucleic acid

(C) STRANDEDNESS:. single

(D) TOPOLOGY: linear

(xi) SEQUENCE DESCRIPTION: SEQ ID NO:5:

GTGAATGATG GCGACATGTC CCTGGCTGAT GGCGGCGCCA CCAACCAGGG CCGGGTGGAG

ATCTTC