WO1999046574A2

WO1999046574A2 - Antibodies for assessing protein post-translational modification status and methods of making and using the same

Info

Publication number: WO1999046574A2
Application number: PCT/US1999/004653
Authority: WO
Inventors: Laszlo Otvos; Hildegund Ertl; Magdalena Thurin; Ralf Hoffman
Original assignee: The Wistar Institute
Priority date: 1998-03-11
Filing date: 1999-03-11
Publication date: 1999-09-16
Also published as: WO1999046574A3; AU3067099A

Abstract

The invention includes a method of making antibodies which are specific for post-translationally modified variants of the protein p53. Polyclonal and monoclonal antibodies are included. The invention also includes a polypeptide useful in preparing such antibodies comprising a portion of a post-translationally modified protein, a peptide spacer and an immunogenicity enhancer.

Description

ANTIBODIES FOR ASSESSING PROTEIN POST-TRANSLATIONAL MODIFICATION STATUS AND METHODS OF MAKING AND USING THE

SAME

FIELD OF THE INVENTION The field ofthe invention is compositions and methods of making and using antibodies and polypeptides useful in the detection of posttranslationally modified variants of proteins.

BACKGROUND OF THE INVENTION The tumor suppressor protein p53 is considered a guardian ofthe genome (Lane, 1992, Nature 358:15-16). The p53 gene encodes a nuclear phosphoprotein that is altered by mutation or deletion in about 50% of human tumors (Hollstein et al., 1991, Science 253:49-53). p53 protein can be divided into five distinct regions. The transactivation domain covers the first 42 amino acid stretch. This region mediates the transcriptional activity of p53, which is directly correlated to its ability to suppress cell growth (Fields and Jang, 1990, Science 249:1046-1049). The next domain, the DNA-binding domain (located between amino acids 96 and 286), is responsible for sequence-specific DNA binding (Kern et al., 1991, Science 252:1708-1711). Recently, an additional functional domain was identified on p53 between the transactivation and the sequence-specific DNA binding domains. Walker and Levine (1996, Proc. Natl. Acad. Sci. USA 93:5335-15340) localized a region between amino acids 61 and 94 that is necessary for efficient growth suppression. The tetramerization domain (amino acids 319-360) of p53 interacts with other p53 protein chains to form the biologically active tetramer, as was shown by peptide mapping (Sakamoto et al., 1994, Proc. Natl. Acad. Sci. USA 91 :8974-8978). This is not a true tetramer, but a di er of dϊmers in which each chain contributes with a β-pleated sheet and an α-helix (Clore et al., 1995, Science 265:386-391; Clore et al, 1995, Nature Struct. Biol. 2:321-391; Clubb et al, 1995, Protein Science 4:855-862). The fifth domain (basic domain) carries a high positive charge and interacts non-sequence specifically with DNA. This basic region is located at the C-terminus between amino acids 363 and 386 (Buchman et al., 1988, Gene 70:245-252; Farrel et al., 1991, EMBO J. 10:2879-2887). p53 lacking the last 30 amino acids at the

C-terminal end exhibits higher double stranded DNA-affinity, suggesting a negative regulation of specific DNA binding by the basic region (Hupp et al., 1992, Cell 71 :875-886; Bayle et al., 1995, Proc. Natl. Acad. Sci. USA, 92:5729-5733), but this same truncation activates binding to damaged DNA structures (Jayaraman and Prives, 1995, Cell 81 :1021-1029; Lee et al., 1995, Cell 81:1013-1020). Oligomerization was suggested to be influenced by the C-terminal region including the basic domain (Sturzbecher et al., 1992, Oncogene 7:1513-1523). The flexible regions between the domains are involved in conformational changes necessary for the molecule's regulatory effects. It is clear that changes in the primary or secondary structure of p53 results in protein variants with altered biological characteristics.

As single point mutations modify p53 structure and function, post-translational modifications are expected to interfere even more strongly with transcription control. The most prominent of these post-translational modifications is the phosphorylation of p53. Phosphorylation sites span the entire protein and distinct phosphorylation sites are found in at least three domains and in the flexible segments between them. The importance of phosphorylation of p53 in determining the biological role of the protein has recently been increasingly appreciated (Steegenga et al, 1996, J. Mol. Biol. 263:103-113; Milczarek et al., 1997, Life Sci. 60:1-11). It seems that the two flexible domains, the N-terminal transactivation domain and the C-terminal basic domain, are the major sites of protein kinase activity. These are the regions of p53 that induce most of the anti-p53 autoantibodies in cancer patients (Lubin et al., 1993, Cancer Res. 53:5872-5876).

The following are examples of reports demonstrating the effect of phosphorylation of specific amino acid residues in p53 upon biological function. Protein kinase C (PKC) can phosphorylate p53 in vitro and in vivo (Baudier et al., 1992, Proc. Natl. Acad. Sci. USA 89:11627-11631). Only phosphorylation of Ser378 has been identified so far by PKC (Takenaka et al., 1995, J. Biol Chem. 270:5405- 5411) although, based on the consensus sequences for this kinase, four additional target residues are located in the basic region of p53 (Ser367, Ser371, Ser376 and Thr377)

(Baudier et al., 1992, Proc. Natl. Acad. Sci. USA 89:11627-11631; Milne et al, 1996, Nucleic Acids Res. 20:5565-5570). Phosphorylation of p53 with PKC activates DNA binding and the binding can be reversed by protein phosphatases 1 and 2A (Takenaka et al., 1995, J. Biol Chem. 270:5405-5411). Casein kinase II (CKII) can phosphorylate Ser392, the penultimate residue, located just following the basic domain (Meek et al.,

1990, EMBO J. 9, 3253-3260; Herrmann et al., 1991, Oncogene 6:877-884). Similar to Ser315 and Ser378, phosphorylation of Ser392 stimulates the sequence-specific DNA- binding activity of p53 in vitro (Hupp et al., 1992, Cell 71 :875-886), and this may also affect the transcriptional activity of p53 by regulating its DNA-binding affinity. Phosphorylation at Ser392 appears necessary for suppressor function since substitution with alanine abolishes the ability of p53 to suppress cell proliferation, whereas substitution with aspartic acid, which mimics phosphoserine, has only a partial effect on suppressor function (Milne et al., 1992, Nucleic Acids Res. 20:5565-5570). The tumor promoter okaidic acid has been shown to generate hyperphosphorylation of p53 in vitro (Yatsunami et al, 1993, Cancer Res. 53:239-241 ;

Zhang et al., 1994, Cancer Res. 54:4448-4453). Okaidic acid inhibits protein phosphatases 1 and 2A (the latter dephosphorylates amino-terminal residues of p53 (Scheidtmann et al, 1991, Mol. Cell. Biol. 11:1996-2003). Hyperphosphorylation of p53 by okaidic acid attenuates its transcriptional activation function without a discernible change in conformation (Zhang et al., 1994, Cancer Res. 54:4448-4453).

Phosphorylation ofthe C-terminal region appears to function in the opposite manner. In this domain phosphorylation may promote different conformational forms of p53 that interact with distinct tissue-specific factors to regulate gene expression (Mukhopadhyay et al., 1995, Springer- Verlag, New York, pp. 79-8). Another report has suggested that variations in the phosphorylation might affect the half-life of wild- type ρ53 (Steegenga et al, 1996, J. Mol. Biol. 263:103-113).

Glycosylation, a second type of post-translational modification, has also been identified at the C-terminal domain of p53. Shaw et al. (1996, Oncogene 12:921- 930) mapped O-glycosylation at or around the C-terminal region. Among the many varieties of O-glycosylation, serine and threonine residues sometimes carry single β- linked GlcNAc moieties. Several lines of evidence indicate that β-linked O-GlcNAc attachment of serines and threonines is a regulatory modification, perhaps analogous to phosphorylation (Hayes and Hart, 1994, Curr. Opin. Biol. 4:692-696). Indeed, virtually all O-GlcNAc modified proteins are also transiently phosphorylated. In addition to being phosphoproteins, practically all O-GlcNAc modified proteins form specific and reversible multimeric complexes with other proteins. O-GlcNAc also appears to be highly immunogenic (Hart et al., 1994, Glycosyl Dis. 1:214-215) and a potential modifier of peptide/protein conformation (Hayes and Hart, 1994, Curr. Opin. Biol. 4:692-696).

Although a number of anti-p53 antibodies are commercially available, their applicability is often limited. For example, polyclonal antibody 421 is not very sensitive in ELISA, or Western-blotting. High sensitivity monoclonal antibodies for the different p53 subdomains would be extremely useful. An even bigger problem is that in spite ofthe major interest in the status of phosphorylation of p53 in the different tissues, cell lines and tumor progression stages, currently there are no phosphate or sugar specific anti-p53 monoclonal antibodies available. Synthetic phosphopeptides and glycopeptides can be used as immunogens, and the resulting phosphopeptide or glycopeptide specific monoclonal antibodies are likely to be invaluable reagents for the study of p53 transformation and processing. Antibodies can distinguish peptides versus phosphopeptides and proteins versus phosphoproteins as extensive experience with neuronal cytoskeletal proteins indicate (Lee et al., 1988a, Proc. Natl. Acad. Sci. USA 85:1998-2002; Lee et al, 1988b, Proc. Natl. Acad. Sci. USA 85:7384-7388; Lang et al., 1992, Biochem. Biophys. Res. Cornmun. 187:783-790). Phosphorylation specific mAbs would be especially attractive in the light of a current report as now described. Whole-cell proteins from control and apoptotic H460a cells were separated by two-dimensional electrophoresis and were transferred to nitrocellulose (Maxwell et al., 1996, Electrophoresis 17:1772-1775). Four isoforms of p53 differing in the phosphorylation state were detected by mAb

BP53.12 in the control cells, and three additional isoforms were observed to be expressed at significant levels only in apoptotic cells (Maxwell et al., 1996, Electrophoresis 17:1772-1775). MAb BP53.12 recognized the N-terminal domain of p53, where a large number of potential phosphorylation sites can be found. As the phosphate specificity of this antibody is unknown, it is likely that some important phosphate forms ofthe antigens remained undetected in the system. The main antigenic regions of p53 are the extreme termini (Legros et al., 1994, Oncogene 9:2071-2076), where most ofthe proposed phosphorylation sites are found. Lane et al. (1996, Oncogene 12:2461-2466) conducted a thorough analysis ofthe epitope distribution of p53, and concluded that in addition to the immunogenic N- and C- termini, very few new antigenic sites can be identified. Based on their recommendation, new approaches will have to be employed to identify novel immunological reagents to p53. According to Lane et al., (1996, Oncogene 12:2461- 2466) immunization with peptides may be the most productive route to overcome the host restraints and to the isolation of monoclonal antibodies to hidden or non- immunodominant p53 regions.

Another application of synthetic phosphopeptides and glycopeptides is the screening of anti-p53 autoantibodies in sera obtained from cancer patients. Ten percent of breast cancer patients have circulating antibodies directed against the p53 protein, and p53 is overexpressed in the cells of these patients (Davidoff et al., 1992,

Proc. Natl. Acad. Sci. USA 89:3439-3442). Each ofthe p53 positive antisera recognize a wide range of p53 proteins, both wild type and mutant versions thereof by immunoprecipitation (Davidoff et al., 1992, Proc. Natl. Acad. Sci. USA 89:3439- 3442). It has been reported that mutant p53 proteins do not seem to contain a dominant antigenic epitope (Labrecque et al., 1993, Cancer Res. 53:3468-3471).

An outstanding account ofthe epitopes recognized by anti-p53 autoantibodies was published by Lubin et al. (1993, Cancer Res. 53:5872-5876). The presence of p53 autoantibodies was detected in sera of eleven patients having different types of tumors. Synthetic overlapping peptides, made by the "pin" method were used as antigens on a regular ELISA, except that the peptides were N-terminally biotinylated and the plates were coated with streptavidin. The immune response of patients with p53 antibodies was restricted to a small subset of peptides located in the amino and carboxy termini of p53, regardless ofthe type ofthe cancer. Coincidentally, the terminal domains of p53, wherein most ofthe phosphorylation sites are located are the most immunodominant regions ofthe protein. Although the highest number of patients having p53 autoantibodies had lung and pancreas carcinomas, two cancer types which are known to have a high frequency of p53 gene mutations, not all patients having p53 antibodies had mutated genes. Taken together, the mechanism by which p53 is presented to the immune system is still unknown (Lubin et al., 1993, Cancer Res. 53:5872-5876). An analysis ofthe post-translational modifications ofthe antigens recognized by the anti-p53 autoantibodies is long overdue.

The most useful reagents for the detection and quantification of proteins which are secreted by cells or are present in tissues are monoclonal antibodies (mAbs).

Indeed, mAbs and polyclonal antibodies are the reagents of choice to characterize the active or inactive status of tumor suppressor p53 (Donehower and Bradley, 1993, Biochim. Biophys. Acta 1155:181-205; Lin and Simmons, 1990,Virology 176:302- 305; Medcalf and Milner, 1993, Oncogene 8:2847-2851; Hall and Milner, 1995, Oncogene 10:561-567). Specific DNA binding of p53 was greatly increased after saturation with a polyclonal antibody termed pAb 421. This antibody blocked the C-terminal non-specific DNA-binding sites, favoring the specific DNA-binding ofthe middle region ofthe protein (Bayle et al., 1995, Proc. Natl. Acad. Sci. USA, 92:5729- 5733). Native p53 (wild-type or mutant) in cell lines and tissues is a weak immunogen, probably because ofthe low level of protein expression.

Given the strong association of p53 expression, post-translational modifications and antibodies directed thereto, with a variety of cancer states in humans, there is a long felt and unmet need for compositions and methods which identify and distinguish various forms of p53 from each other, and which can therefore facilitate better diagnostic and treatment methods for cancer. The present invention satisfies this need.

BRIEF SUMMARY OF THE INVENTION

The invention includes a method of making an antibody which specifically binds with a post-translationally modified p53. The method comprises the steps of a) administering to an animal a polypeptide comprising a portion ofthe p53 having a post-translationally modified amino acid residue and an immunogenicity enhancer; b) eliciting an antibody response thereto; and c) obtaining the antibody from the animal.

In one aspect, the antibody is a polyclonal antibody.

In another aspect, the method further comprises d) preparing a monoclonal antibody from the animal. In another aspect, the p53 comprises a post-translationally modified amino acid residue having a modification selected from the group consisting of phosphorylation, glycosylation and prenylation.

Preferably, the p53 is abnormally post-translationally modified in a disease. In another aspect, the polypeptide comprises the p53 and the immunogenicity enhancer. Preferably, the immunogenicity enhancer is a T-helper cell determinant, even more preferably, the T-helper cell determinant is 3 ID.

In yet another aspect, the polypeptide further comprises a peptide spacer. Preferably, the portion ofthe p53, the immunogenicity enhancer, and the spacer are cosynthesized. Even more preferably, the immunogenicity enhancer is 3 ID.

The invention further includes an antibody made by a method comprising the steps of a) administering to an animal a polypeptide comprising a portion ofthe p53 having a post-translationally modified amino acid residue and an immunogenicity enhancer; b) eliciting an antibody response thereto; and c) obtaining the antibody from the animal.

In addition, the invention includes an antibody made by a method comprising the steps of a) administering to an animal a polypeptide comprising a portion ofthe p53 having a post-translationally modified amino acid residue and an immunogenicity enhancer; b) eliciting an antibody response thereto; and c) obtaining the antibody from the animal, wherein the polypeptide comprises a peptide spacer and an immunogenicity enhancer being 3 ID wherein the polypeptide, the spacer and 3 ID are cosynthesized.

Also included in the invention is a polypeptide comprising a portion of a p53 comprising a post-translationally modified amino acid residue having a modification selected from the group consisting of phosphorylation, glycosylation and prenylation and an immunogenicity enhancer.

In one embodiment, the portion ofthe p53 is at the C-terminus ofthe polypeptide, and is covalently linked to the immunogenicity enhancer, and further wherein the immunogenicity enhancer is at the N-terminus ofthe polypeptide. In one aspect, the polypeptide further comprises a peptide spacer between the portion ofthe p53 and the immunogenicity enhancer. Preferably, the length ofthe polypeptide is from about 20 to about 45 amino acid residues. More preferably, the length ofthe polypeptide is 41 amino acid residues.

The invention further includes a monoclonal antibody which binds specifically to a post-translationally modified variant of p53, wherein the immunogen used to produce the antibody is the post- translationally modified variant of p53 in combination with an immunogenicity enhancer.

In one aspect, the post-translationally modified variant of p53 comprises a modification selected from the group consisting of phosphorylation, glycosylation and farnesylation.

Preferably, the post-translationally modified variant of p53 comprises a biphosphorylated C-terminus wherein Ser378 and Ser392 are phosphorylated.

In another aspect, the post-translationally modified variant of p53 comprises a monophosphorylated C-terminus. Additionally included is a monoclonal antibody which binds specifically to a peptide comprising a phosphorylated p53, a peptide spacer, and 3 ID.

The invention further includes a method of assessing the post-translational modification status of a p53 obtained from a human patient. The method comprises the steps of a) obtaining a biological sample from a patient comprising the p53; b) contacting an antibody made by aforementioned method with the p53; c) forming an antigen-antibody complex between the p53 and the antibody, and d) detecting the antigen-antibody complex, wherein the presence ofthe complex defines the post-translational modification status ofthe p53.

In addition, the invention includes a method of determining the presence or absence of an autoantibody specific for a post-translationally modified p53 in a biological sample of a human patient. The method comprises the steps of a) obtaining a biological sample from the patient; b) contacting the biological sample with a polypeptide comprising a portion of a p53 comprising a post-translationally modified amino acid residue having a modification selected from the group consisting of phosphorylation, glycosylation and prenylation and an immunogenicity enhancer; c) forming an antigen-antibody complex between the polypeptide and the autoantibody, and d) detecting the antigen-antibody complex; wherein the presence ofthe complex is an indication that the autoantibody is present in the biological sample.

BRIEF DESCRIPTION OF THE DRAWINGS

The foregoing summary, as well as the following detailed description of the invention, will be better understood when read in conjunction with the appended drawings.

Figure 1 is an image of a gel showing the purification of murine p53 protein expressed by Sf9 insect cells infected with recombinant baculovirus using a mAb p53-18 immunoaffinity column. Column fractions were analyzed by SDS-PAGE. The far left column shows the position ofthe molecular mass markers. Lanes 1-8 correspond to fractions 1-8, respectively, eluted with glycine buffer, pH 2.9. The gel was stained with Coomassie Blue.

Figure 2 is a graph depicting results from a conformation-sensitive enzyme-linked immunosorbent assay of unphosphorylated and diphosphorylated p53 C-terminal peptides. The unphosphorylated peptide is represented by triangles, and the diphosphorylated peptide is represented by circles. Open symbols correspond to peptides plated from water, and closed symbols correspond to peptides plated from TFE. The dilution ofthe ascites fluid containing the antibody was 1 : 1000.

Figure 3 is an illustration of a helical wheel representing the structure of p53 peptide 371-393. The inner circle corresponds to the first 18 amino acids, and the outer circle corresponds to amino acids 19-23, overlapping amino acids 1-5. Ser378 (in position 8) and Ser392 (at the outer circle in position 4) are marked with asterisks. DETAILED DESCRIPTION OF THE INVENTION

Definitions

As used herein, each ofthe following terms has the meaning associated with it in this section.

The articles "a" and "an" are used herein to refer to one or to more than one (i.e. to at least one) ofthe grammatical object ofthe article. By way of example, "an element" means one element or more than one element.

As used herein, an "immunodominant region" of a protein means a portion of a protein having a stretch of amino acid residues which comprise an epitope that stimulates an immune response in an animal.

As used herein, the term "post-translationally modified amino acid residue" means an amino acid residue having any ofthe post-translational modifications described herein. By "immunogenicity enhancer" is meant any composition capable of recruiting T-cells to assist B-cells in an immunological response. Examples of such immunogenicity enhancers are peptides, which may be synthetic or naturally occurring. The peptides may be unmodified or may be modified containing amino acid residues that are phosphorylated or glycosylated. An example of an immunogenicity enhancer is the peptide 3 ID, which represents an epitope ofthe rabies virus nucleoprotein in mice ofthe H^2k haplotype.

As used herein, "binds specifically to" means an antibody or antigen (e.g., a polypeptide), which binds in an ELISA or Western Blotting method without binding to a control target, or which binds in an ELISA method to result in an optical density (O.D.) value greater than 0.10, or which binds in an ELISA method to result in an O.D. value of at least twice the background O.D. value. As used herein, "abnormally post-translationally modified protein" means a protein variant that contains more or less phosphate or carbohydrate in a patient having a disease than in a patient without the disease.

As used herein, "peptide spacer" means a tripeptide sequence separating portions of a synthetic polypeptide.

As used herein, "T-helper cell determinant" means a peptide sequence that stimulates T-helper cells.

As used herein, "cosynthesized" means prepared in one continuous solid-phase synthesis. As used herein, amino acids are represented by the full name thereof, by the three letter code corresponding thereto, or by the one-letter code corresponding thereto, as indicated in the following table:

Full Name Three-Letter Code One-Letter Code

Aspartic Acid Asp D

Glutamic Acid Glu E

Lysine Lys K

Arginine Arg R

Histidine His H

Tyrosine Tyr Y

Cysteine Cys C

Asparagine Asn N

Glutamine Gin Q

Serine Ser s

Threonine Thr T Glycine Gly G

Alanine Ala A

Valine Val V

Leucine Leu L

Isoleucine He I

Methionine Met M

Proline Pro P

Phenylalanine Phe F

Tryptophan Trp W

The term "antibody," as used herein, refers to an immunoglobulin molecule which is able to specifically bind to a specific epitope on an antigen. Antibodies can be intact immunoglobulins derived from natural sources or from recombinant sources and can be immunoreactive portions of intact immunoglobulins. Antibodies are typically tetramers of immunoglobulin molecules. The antibodies in the present invention may exist in a variety of forms including, for example, polyclonal antibodies, monoclonal antibodies, Fv, Fab and F(ab)₂, as well as single chain antibodies and humanized antibodies (Harlow et al., 1988, Antibodies: A Laboratory Manual, Cold Spring Harbor, New York; Houston et al., 1988, Proc. Natl. Acad. Sci. USA 85:5879-5883; Bird et al, 1988, Science 242:423-426). By the term "synthetic antibody" as used herein, is meant an antibody which is generated using recombinant DNA technology, such as, for example, an antibody expressed by a bacteriophage as described herein. The term should also be construed to mean an antibody which has been generated by the synthesis of a DNA molecule encoding the antibody and which DNA molecule expresses an antibody protein, or an amino acid sequence specifying the antibody, wherein the DNA or amino acid sequence has been obtained using synthetic DNA or amino acid sequence technology which is available and well known in the art.

As used herein, "polypeptide" refers to a polymer composed of amino acid residues, related naturally occurring structural variants, and synthetic non-naturally occurring analogs thereof linked via peptide bonds, related naturally occurring structural variants, and synthetic non-naturally occurring analogs thereof. Synthetic polypeptides can be synthesized, for example, using an automated polypeptide synthesizer.

The term "protein" typically refers to large polypeptides. The term "peptide" typically refers to short polypeptides.

Conventional notation is used herein to portray polypeptide sequences: the left-hand end of a polypeptide sequence is the amino-terminus, or "N-terminus"; the right-hand end of a polypeptide sequence is the carboxyl-terminus, or "C-terminus."

As used herein, the term "variant" means one form or chemical species of a molecule, such as a protein, or a polypeptide, wherein the molecule may exist in one or more structurally related and similar forms. For example, a multi-phosphorylated variant of p53 protein is a form ofthe p53 protein comprising more than one phosphate group on the polypeptide. Also, for example, a monophosphorylated p53 polypeptide is a form of a polypeptide comprising a portion ofthe p53 protein having only one phosphate group in the polypeptide.

As used herein, the term "post-translationally modified" means a biochemical modification to a protein, polypeptide or amino acid residue, wherein the modification is made using a chemical synthetic method as described herein, or an enzyme in a cell or other living system after the translation of a nucleic acid transcript encoding the protein or polypeptide or amino acid residue into a polypeptide.

Such modifications include, for example, glycosylation, phosphorylation and prenylation. Glycosylation, can be effected, for example, by the process of exposing a polypeptide to enzymes which effect glycosylation, e.g. mammalian glycosylating enzymes, whereby a sugar moiety, e.g., N-acetyl glucosamine, is added to a hydroxyl group on an amino acid residue (e.g. serine, threonine, tyrosine) of a polypeptide. Such modifications also include, for example, phosphorylation.

Phosphorylation can be effected synthetically as described herein, or by exposing a polypeptide to enzymes which effect phosphorylation, e.g. Phosphokinase C, whereby a phosphate group is added to a hydroxyl group on an amino acid residue (e.g., serine, threonine or tyrosine) of a polypeptide. By way of example, a "post-translationally modified" p53 protein may be a p53 protein that is phosphorylated at one or more serine residues along the polypeptide by a kinase enzyme after the p53 protein is translated from a nucleic acid transcript to a polypeptide in a cell.

Also by way of example, "a post-translationally modified" p53 polypeptide may be a polypeptide comprising a portion ofthe p53 protein that is phosphorylated at one or more serine residues along the polypeptide by a chemical synthetic method as described herein.

As described herein, "multi-phosphorylated" means a variant (as described herein) of a polypeptide comprising more than one amino acid residue having a phosphate group. For example, a "multi-phosphorylated variant of p53" includes a form ofthe protein p53 having a phosphate group on both ofthe serine amino acid residues Ser378 and Ser392 ofthe polypeptide.

As described herein, "biphosphorylated" means a variant (as described herein) of a polypeptide comprising two amino acid residues having a phosphate group. For example, a "biphosphorylated p53 C-terminus polypeptide" means a p53 polypeptide comprising the C-terminus ofthe p53 protein, having a phosphate group on two amino acid residues (e.g., Ser378 and Ser392) ofthe p53 C-terminus. As described herein, "monophosphorylated" means a variant (as described herein) of a polypeptide comprising one amino acid residue having a phosphate group. For example, a "monophosphorylated p53 C-terminus" means a p53 polypeptide comprising the C-terminus ofthe p53 protein, having a phosphate group on one amino acid residue (e.g., Ser378) ofthe p53 C-terminus.

As described herein, the term "biological sample" means a cell, a tissue or a biological fluid which is obtained from a living organism, or from a culture of cells or viruses.

Description The invention includes a method of making antibodies which bind with high specificity to a post-translationally modified variant of a physiologically important protein. The method includes the preparation of a polypeptide comprising an immunogenicity enhancer, such as a T-helper cell determinant, and a portion of a protein having a post-translationally modified amino acid residue. The antibodies can be polyclonal or monoclonal. The antibodies can also be synthetic. The protein can be any protein, but is preferably a protein that displays post-translationally modified variants which may be prevalent in a disease state, such as cancer. A preferred protein is the tumor-suppressor protein p53.

The post-translationally modified variant ofthe protein can be a protein having a phosphorylated amino acid residue, or more than one phosphorylated amino acid residue, for example, a serine residue having a phosphate group attached thereto. The post-translationally modified variant can also be a protein having a glycosylated amino acid residue, or more than one glycosylated amino acid residue, for example, a serine residue having a O-linked N-acetyl glucosamine attached thereto. The post-translationally modified variant can also be a farnesylated amino acid residue.

The post-translationally modified variant may comprise multiple or mixed post-translationally modified amino acid residues, such as a protein which has several phosphorylated serine residues and several glycosylated serine residues. The post-translationally modified variant may comprise a modified amino acid residue at any amino acid residue along the protein that is susceptible to enzymatic or chemical modification.

The modification can be naturally occurring, such as by enzymatic phosphorylation by a protein kinase or a glycosylating enzyme. Also, the modification can be chemically created, such as by purifying a protein and chemically adding a phosphate group to a serine amino acid residue which is hydroxylated. Modified amino acid residues are commercially available.

The post-translationally modified variant can also be a portion ofthe protein, such as an immunodominant region of a protein. The post- translationally modified variant can also be a portion of a protein that does not comprise an immunodominant region. For example, the post-translationally modified variant can be a portion of a protein that is hidden or unexposed to the surface ofthe protein under normal physiological conditions. For example, the protein p53 has a region called the central core which is unexposed to the surface under normal cellular conditions, and thus, is inaccessible to immune response components. This buried or inaccessible region is an example of a portion of a protein that is not immunodominant.

The method ofthe invention for making antibodies which specifically bind with the post-translationally modified variant of a protein comprises administering to an animal a polypeptide comprising a portion ofthe protein having a post-translationally modified amino acid residue and an immunogenicity enhancer. The polypeptide may be administered by injection to an animal for developing an immune response. The animal is preferably a mouse. An example of such mice are female C3H/Hc mice. The animal may be any animal capable of generating antibodies to a polypeptide.

The polypeptide may be found as a naturally occurring peptide, or may be synthesized by a peptide synthesis method. The polypeptide may be synthesized by any method of synthesis known to the skilled artisan. Polypeptides are preferably synthesized on solid-phase using automated synthesizers employing standard Fmoc-methodology (Fields and Noble, 1990, Int. J. Pept. Protein Res. 35:161-214). Phosphorylated serine and threonine residues are incorporated as Fmoc-Ser(PO₃HBzl)-OH or Fmoc-Thr(PO₃HBzl). This improved "synthon" strategy for phosphopeptide synthesis is superior to "global" phosphorylation. (Wakamiya et al, 1994, Chem. Lett. 1099-1102) The monoalkyl protected Fmoc-Ser and Fmoc-Thr derivatives are currently marketed by Novabiochem, Ltd. (San Diego, CA)

In the case of glycosylated polypeptides, peracetylated Fmoc-Ser/Thr-β(GlcNAc)-OH is incorporated into the peptide as any other amino acid, except that no excess ofthe acetylating agent is used. The serine derivative is marketed by Bachem California. The threonine derivative and the serine derivative as well, if necessary, is synthesized using the method of Filira et al. (1990, Int. J. Pept. Protein Res. 36:86-96). Alternatively, β-GlcNAc can be added to side-chain unprotected Ser (and presumably Thr) still on the resin through glucose-oxazoline (formed in situ) after the peptide chain assembly is completed (Hollosi et al, 1991 , Tetrahedron Lett.

32:1531- 1534). However, this reaction proceeds with low yield for serine and is even slower for threonine (Hollosi et al, 1991, Tetrahedron Lett. 32:1531-1534). A new, simple synthetic route for making peptides containing Ser/Thr-β(GlcNAc) was recently published (Meinjohans et al, 1995, Tetrahedron Lett. 40:9205-9208). Peracetylated Fmoc-Ser-β(GlcNAc)-OPfp and Fmoc-Thr-β(GlcNAc)-OPfp were synthesized from

Fmoc-Ser/Thr-OPfp with 2-trichloroethoxycarbonyl amino (Teoc) glycosyl donors followed by in situ reduction ofthe Teoc group and simultaneous N-acetylation using zinc dust in tetrahydrofuran/acetic anhydride/acetic acid. All listed starting materials and reagents used in this method are commercially available. The portion ofthe protein having a post-translationally modified amino acid residue and the immunogenicity enhancer may be synthesized separately and then may be covalently linked later, or may be cosynthesized using the method described above. In one embodiment, these components are cosynthesized using the method described above. In another embodiment, a polypeptide spacer may be placed between the portion ofthe protein having the post-translationally modified amino acid residue and the immunogenicity enhancer. The purpose ofthe peptide spacer is to prevent unnatural conformational induction in the post-translationally modified protein portion that can happen when it is linked covalently to the highly helical 3 ID sequence. In yet another embodiment, the 3 ID immunogenicity enhancer is placed at the N-terminus of the polypeptide, and the portion ofthe post-translationally modified protein is at the C-terminus ofthe polypeptide, and between the two is placed a peptide spacer. An example ofthe peptide spacer is the tripeptide glycine-alanine-glycine (Gly-Ala-Gly). In determining the length ofthe polypeptide for use in the method ofthe invention, the polypeptide may be any length such that it is long enough to be sufficiently immunogenic, and so that antipeptide antibodies will primarily recognize the presence or absence ofthe phosphate or carbohydrate on the post-translationally modified protein portion. If the polypeptide is too long, this will not be the case. In one embodiment, the entire length ofthe polypeptide, including the immunogenicity enhancer, is from about 20 to about 45 amino acid residues long. In a preferred embodiment where the polypeptide includes a portion of post-translationally modified p53 protein, the length ofthe polypeptide is 41 amino acids long.

Another important concern in the design ofthe polypeptide used in the method ofthe invention is the stability ofthe polypeptide in the serum ofthe animal into which it will be injected to raise antibodies. For example, if mice are used to raise antibodies in the method ofthe invention, in designing the polypeptide, tests ofthe polypeptide in mouse serum for stability against protease digestion should be carried out prior to injection ofthe polypeptide. Such serum stability tests can be carried out as follows. The synthesized polypeptide comprising post-translationally modified amino acid residues can be incubated with mouse serum at about 25% mouse serum at 37° C for several hours. Serum proteins can then be precipitated with a precipitating agent such as 15% trichloroacetic acid. The pellets from the precipitation can be centrifuged and the supernatant can then be loaded onto a reversed-phase-HPLC method for analysis of degradation ofthe polypeptides by serum proteases.

The method ofthe invention for making antibodies which specifically bind with post-translationally modified variants of protein also comprises the step of obtaining antibodies from the immunized animal. Antibodies can be obtained from the animal using methods well known in the art and are described, for example, in Harlow et al, 1988, Antibodies: A Laboratory Manual, Cold Spring Harbor, New York. The generation of polyclonal antibodies is accomplished by inoculating the desired animal with the antigen and isolating antibodies which specifically bind the antigen therefrom. Monoclonal antibodies directed against full length or peptide fragments of a protein or peptide may be prepared using any well known monoclonal antibody preparation procedures, such as those described, for example, in Harlow et al.,1988, In: Antibodies, A Laboratory Manual, Cold Spring Harbor, NY and in (Tuszynski et al, 1988, Blood 72:109-115). Quantities ofthe desired peptide may also be synthesized using chemical synthesis technology. Alternatively, DNA encoding the desired peptide may be cloned and expressed from an appropriate promoter sequence in cells suitable for the generation of large quantities of peptide. However, proteins expressed by molecular biology techniques are often not appropriately phosphorylated or glycosylated. Monoclonal antibodies directed against the peptide are generated from mice immunized with the peptide using standard procedures as referenced herein.

The invention also includes the polypeptide described above which is useful in generating antibodies which are specific for a post-translationally modified variant of a protein. As described above, this polypeptide comprises a portion of a protein having a post-translationally modified amino acid residue, wherein the modification is a modification selected from the group consisting of phosphorylation, glycosylation and farnesylation. The portion ofthe protein may comprise multiple post-translational modifications in more than one ofthe above categories. For example, the portion ofthe protein may comprise more than one post-translationally modified amino acid residue that is phosphorylated, and also more than one amino acid residue that is glycosylated. The polypeptide also comprises an immunogenicity enhancer as described above. The immunogenicity enhancer may be a peptide or a glycosylated peptide, and may be any composition capable of eliciting T-helper cell response in an immunological reaction.

The portion of a post-translationally modified protein and the immunogenicity enhancer parts ofthe polypeptide may be cosynthesized, or may synthesized separately and then covalently linked. The methods of synthesis may be as described above. In one embodiment, the polypeptide may comprise a portion ofthe p53 protein comprising a post-translationally modified amino acid residue, or several post-translationally modified amino acid residues that are phosphorylated and covalently linked to an immunogenicity enhancer. The immunogenicity enhancer may be the peptide 3 ID.

In a preferred embodiment, the peptide 3 ID is located at the N-terminus ofthe polypeptide, and is separated by a peptide spacer from the post-translationally modified portion ofthe p53 protein, which is at the C-terminus ofthe polypeptide. In another preferred embodiment, the overall polypeptide length is from about 20 to about 45 amino acid residues long.

The invention also includes highly sensitive and highly specific monoclonal antibodies which are specific for post-translationally modified variants of biological proteins such as the p53 protein. For example, the invention includes a highly sensitive monoclonal antibody which is highly specific for a double phosphorylated form of p53 protein, which double phosphorylated form is prevalent in cancer patients or normal patients. The invention includes a monoclonal antibody which binds specifically to a post-translationally modified variant ofthe tumor suppressor protein p53. The post-translationally modified variant ofthe tumor suppressor protein p53 may be any variant ofthe tumor suppressor protein p53 having one or more amino acid residues which are post-translationally modified. Such post-translationally modified amino acid residues may be phosphorylated, glycosylated, or farnesylated, or any combination thereof.

One preferred monoclonal antibody ofthe invention, mAb p53-18, is a monoclonal antibody highly specific and highly sensitive to the post-translationally modified variant of p53 which is biphosphorylated at the C-terminus at the amino acid residues Ser378 and Ser392.

The invention also includes a method for assessing the post-translational modification status of a protein, such as p53, obtained from a patient. The patient may be any animal, however, the preferred patient is a human. The protein may be obtained from a human patient in a serum sample, a tissue biopsy sample, or a cell sample (i.e., a biological sample). The method may be used to assess the presence or absence of a certain post-translationally modified protein variant which is correlated with a disease state or stage of tumor progression. The method comprises obtaining the protein sample from the human patient and contacting antibodies which are highly specific and highly sensitive for the desired post-translationally modified protein variant as described in the invention with the protein sample obtained from the patient. An antigen antibody complex is then formed between the monoclonal antibodies described above and the protein sample described above, by incubating both together according to standard methods. The antigen antibody complex is then detected using an immunological technique such as an ELISA method, immunoprecipitation, or Western

Blotting methods. In this method, the highly specific and highly sensitive monoclonal antibodies are used to detect the presence of, or to assess the quantity of, a post-translationally modified variant ofthe protein to which the monoclonal antibody is specific. In a preferred embodiment, the monoclonal antibody specific for a variant of p53 protein which is biphosphorylated at the C-terminal region is used to screen a serum sample from a human cancer patient for the presence of biphosphorylated p53 protein. The invention also includes a method for determining the presence or absence of an autoantibody specific for a post-translationally modified protein variant, such as phosphorylated p53, in a biological sample, e.g., serum, obtained from a patient. A serum sample is obtained from a patient, preferably a human patient, and is assessed for the presence of an autoantibody as follows. Polypeptides ofthe invention comprising an amino acid residue having a post-translational modification are prepared as described herein. The polypeptide prepared is one known to specifically bind the autoantibody sought to be detected in the patient's serum as determined by methods described herein. An antigen antibody complex is then formed between the polypeptide and the serum sample by incubating both together according to standard methods. The antigen antibody complex is then detected using an immunological technique such as an ELISA method, immunoprecipitation, or Western Blotting methods. The presence ofthe autoantibody may be correlated with a disease state or stage of tumor progression in a human. The invention is now described with reference to the following

Examples. These Examples are provided for the purpose of illustration only and the invention should in no way be construed as being limited to these Examples, but rather should be construed to encompass any and all variations which become evident as a result ofthe teaching provided herein. Example 1 : Design, preparation and use of synthetic polypeptides

The most straightforward way to identify the actual phosphorylation and glycosylation sites ofthe tumor suppressor protein p53 was to use synthetic phosphopeptide or glycopeptide antigens and immunogens to the proposed phosphorylation and glycosylation sites. In a successful use of this strategy, for example, a synthesized multi-phosphorylated polypeptide generated antibodies which were as specific for the multi phosphorylated variant of p53 protein as antibodies which were generated using the full p53 protein. Some potential phosphorylation and glycosylation sites of p53 and possible functional and structural consequences ofthe post-translational modifications are described herein. Accordingly, synthetic peptide antigens and immunogens made for this purpose may contain modified amino acids as depicted in Table 1.

Because generally all GlcNAc sites are potential phosphorylation sites as well, Ser99 was synthesized with either GlcNAc or phosphate. Some potential sites for phosphorylation or glycosylation are located only a few amino acid residues apart. Synthetic peptide families can be made including modified residues at these proximal sites. This strategy also allows the preparation of multiple or mixed phosphorylated or glycosylated peptides. Selection of peptide length and construct orientation

The length ofthe polypeptide prepared should be long enough to be sufficiently immunogenic, but not so long that the anti-peptide antibodies will primarily recognize the presence or absence ofthe phosphate or carbohydrate. The peptide families described herein for studying phosphorylated epitopes of p53 were successful in generating p53 specific antibodies. Polypeptides were synthesized having

15-17 amino acid residue-long B-cell determinants (portions of post-translationally modified protein) at the C-terminal end and peptide 3 ID at the N-terminal end.

Within these parameters the peptide families described in Table 2 were synthesized (modified amino acids are marked with asterisk; phosphopeptides are printed in italics, and glycopeptides are underlined). In selecting the actual peptides, the modified amino acids were placed at least five amino acids away from either terminus (except the penultimate residue), so that phosphorylation or glycosylation efficiently modified the recognitional characteristics. In the p53 362-383 peptide family, multiple glycosylated or mixed glycosylated and phosphorylated peptides in which the sugar is proximal to another modified amino acid were not prepared because the coupling of back-to-back bulky amino acids was expected to proceed with a very low yield. From the same peptide family a triple phosphorylated peptide was made (shown in parenthesis in Table 2). However, synthesis of this peptide was not a trivial matter. Synthesis of double phosphorylated polypeptides, even when proximal residues were phosphorylated, was not problematic.

Immunizations and mouse serum stability studies ofthe polypeptides

To identify the phosphorylation and glycosylation sites in wild-type and modified p53, the sera of healthy individuals and cancer patients was screened, and phosphopeptide and glycopeptide-specific mAbs were generated. Peptide 3 ID, an immunogenicity enhancer which provides T-helper bystander help in the immunizations, is an immunodominant T-helper cell determinant and is a portion (amino acid residues 404-418) ofthe rabies virus strain ERA nucleoprotein (Ertl et al, 1989, J. Virol. 63:2885-2892). The amino acid sequence of Peptide 3 ID is as follows:

AVYTRIMMNGGRLKR.

C3H mice were inoculated in the hind legs with 20 mg ofthe tandem construct, and boosted two weeks later. After screening the test bleeds, a third immunization was given five days before the fusion of splenocytes with myeloma cells. The anti-peptide mAbs were assessed by ELISA and Western blotting for specific binding to p53 expressed in various cell lines and tissues.

The C-terminal p53 fragment described herein was a good immunogen due to the favorable physico-chemical properties of this protein domain. This region of p53 is hydrophilic and assumes various reverse-turn structures, generally considered advantageous features for inducing anti-protein and anti-peptide antibodies (Hopp and

Woods, 1983, Mol. Immunol. 20:483-489). Phosphopeptides are generally considered good immunogens. O-linked GlcNAc moieties are similarly effective during immunizations (Holt et al, 1987, J. Cell Biol. 104:1157-1164). Antibodies to glycoproteins containing GlcNAc moieties were reported to specifically recognize not only the sugar, but also the surrounding peptide sequence (Holt et al, 1987, J. Cell

Biol. 104:1157-1164).

All immunodominant amino acids exhibited increased stability in diluted serum, indicating that serum stability studies were good tools to reveal differences in the immunogenicity of protein fragments, and likely, synthetic polypeptides. Although the above listed phosphopeptides and glycopeptides generated antisera having high titer, the mouse serum stability ofthe polypeptides was assessed prior to immunizations to allow for modification ofthe immunization protocols if some phosphopeptides or glycopeptides were less resistant to mouse serum proteases than others.

In vivo stability of peptides in blood is modeled well by in vitro stability in serum or plasma (neglecting renal and hepatic clearance), as has been shown for small peptides including phosphopeptides (Hoffmann et al, 1997, Anal. Chim. Acta 352:327-333) and glycopeptides (Powell et al, 1993, Pharm. Res. 10:1268-1273).

Incorporation of phosphate, GlcNAc and GalNAc moieties generally increases the resistance of synthetic polypeptides to serum proteases. Polypeptide serum stability was tested as described previously herein and in (Powell et al, 1993, Pharm. Res. 10:1268-1273; Szendrei et al, 1996, Int. J. Peptide Protein Res. 47:289-296; Hoffmann et al, 1997, Anal. Chim. Acta 352:327-333). Unmodified peptides, phosphopeptides and glycopeptides were incubated with 25% mouse serum at 37°C for 1 to 6 hours. Serum proteins were precipitated with 15% trichloroacetic acid (TCA). The pellets were centrifuged and the supernatant was loaded onto a reversed-phase HPLC column for analysis. Synthesis of polypeptides

Polypeptides were synthesized on solid-phase using a Milligen 9050 (continuous flow) or Rainin PS3 (batch mixing) automated synthesizers employing standard Fmoc-methodology (Fields and Noble, 1990, Int. J. Pept. Protein Res. 35:161- 214). Phosphorylated serine and threonine residues were incorporated as Fmoc- Ser(PO₃HBzl)-OH or Fmoc-Thr(PO₃HBzl). The monoalkyl protected Fmoc-Ser and

Fmoc-Thr derivatives are currently marketed by Novabiochem, Ltd. (San Diego, CA).

For synthesizing glycosylated polypeptides, peracetylated Fmoc- Ser/Thr-β(GlcNAc)-OH was incorporated into the polypeptide as if it was any other amino acid, except that no excess ofthe acetylating agent was used. The serine derivative is marketed by Bachem California (Torrance, CA).

After synthesis, the polypeptide was cleaved from the support using trifluoroacetic acid (TFA):thioanisole (95:5; v/v). Polypeptides, phosphopeptides and glycopeptides were purified by reversed-phase high performance liquid chromatography (RP-HPLC). Removal ofthe acetyl protecting group was accomplished using diluted NaOH or NaOMe (Bulet et al, 1996, Eur. J. Biochem. 238:64-69; Otvos et al, 1997, In: Solid-Phase Synthesis and Combinatorial Chemical Libraries, Andrews et al, eds, Mayflower Scientific, Kingswinford, in press.). The integrity ofthe synthetic peptides was analyzed by mass spectrometry and phosphate analysis (Zardeneta et al, 1990, Anal. Biochem. 190:340-347).

Immunological techniques

A direct ELISA was used according to standard protocols (Goding, 1986, Monoclonal Antibodies: Principles and Practice, Academic Press, Orlando). Alternatively, a modified protocol was used (Otvos and Szendrei, 1996. In:

Neuropeptide Protocols, Irvine et al, eds, Humana Press, Totowa, NJ: pp. 269-275). The modified ELISA protocol was used for conformation-sensitive ELISA studies. In these studies, the polypeptides were plated from trifluoroethanol/water mixtures. First, the active dilution range of antibody preparation was determined. The selectivity of a given antibody dilution toward various peptide antigens was assessed. Generally, from

40 nanograms to 5 milligrams of peptide antigens and from 5 to 20 nanograms of protein samples were loaded in each ELISA plate well. Actual peptide concentrations were determined by RP-HPLC (Szendrei et al, 1994, Eur. J. Biochem. 226:917-924). Western-blots were made using 10% sodium dodecyl sulfate-polyacrylamide gels and nitrocellulose by standard techniques (Goding, 1986, Monoclonal Antibodies:

Principles and Practice. Academic Press, Orlando) or modified techniques (Reim and Speicher, 1992, Anal. Biochem. 207:19-23). Protein concentrations ofthe purified p53 preparations were determined by bicinchoninic acid (Smith et al, 1985, Anal. Biochem. 150:76-85). For tissue samples, the phosphate or sugar specificity ofthe p53 variants was more important that the actual protein load.

Circulating anti-p53 autoantibodies obtained from sera from cancer patients and normal controls were analyzed for the phosphorylation and glycosylation status ofthe p53 protein variants by assessing binding to synthesized polypeptides.

When autoantibodies failed to specifically bind the synthesized polypeptides, more concentrated sera was used ( i.e., at a dilution less than 1 :50). Alternatively, the sensitivity ofthe direct ELISA was increased by adding the polypeptides to the ELISA plate from trifluoroethanol (TFE) (Otvos et al, 1994, In: Innovation and Perspectives in Solid-Phase Synthesis and Complimentary Technologies, Epton, ed, Mayflower

Worldwide, Birmingham, pp. 273-278). Sera obtained from patients with different cancer types and stages were obtained from Quality Biotech, an NIH sponsored and National Cancer Institute designated Repository.

Assessment of commercially available p53 antibodies Anti-p53 antibodies (e.g. 421, 1620, D01, BP53.12, 240, 1801, 246,

CM- 10) are commercially available from various sources, including Oncogene Science (Cambridge, MA) and Santa Cruz (Santa Cruz, CA).

Some of these antibodies are known to cross-react with p53 regions described herein (e.g. antibody BP53.12 recognizes p53 around amino acids 15 and 20). The synthetic polypeptides, phosphopeptides and glycopeptides prepared herein were used to map the identity and the post-translational modification status ofthe epitopes of these and other antibodies.

Animals used for immunizations

Inbred female C3H/He mice at 6-8 weeks of age were used for the immunizations. Mice were inoculated with 10 microgram doses of polypeptide antigens in each hind leg as described in Example 3 herein.

Example 2: Synthesis of phosphopolypeptides corresponding to the p53 C-terminus Seven peptides were prepared as described herein corresponding to the C-terminal region of p53. The total number of amino acid residues in the p53 protein is 393, thus, the numbers used herein refer to amino acid residues 1-393. Two peptides were non-phosphorylated, two had a phosphate group on Ser378, one had the phosphate group on Ser376, and two were phosphorylated on both Ser378 and Ser392.

The peptides were either 23 or 33 amino acids long. The peptides were purified on RP- HPLC to homogeneity and then characterized by mass spectroscopy (Otvos et al, 1998, Biochim. Biophys. Act. 1404:457-474).

Table 3 indicates the peptide retention times obtained using trifluoroacetic acid as ion-pairing reagent in RP-HPLC. The elution behavior ofthe short peptides was scrutinized during various reversed-phase high performance chromatographic conditions (Hoffmann et al, 1997, Anal. Chim. Acta 352:327-333).

Assessment of binding of antibody 421 to the p53 C-terminal region polypeptides

Antibody 421 has been reported to recognize p53 at its basic domain between amino acids 372-381 (Wade-Evans and Jenkins, 1985, EMBO J. 4:699-706).

It has been hypothesized that the lack of binding of antibody 421 to p53 in some cases may be due to phosphorylation between residues 370 and 378 (Ullrich et al, 1992, Oncogene 7:1635-1643). This fragment contains Ser378, the protein kinase C (PKC) site. To test the hypothesis, synthetic polypeptides prepared herein were tested for their ability to bind pAb 421 in a direct ELISA assay. Antibody 421 specifically bound to the unphosphorylated peptide, but failed to bind to either ofthe phosphopeptides tested. This indicated that phosphorylation of Ser378 masks the epitope, and p53 phosphorylated at Ser378 does not bind to antibody 421. However, p53 without phosphate at the C-terminal domain will bind antibody 421. The polypeptides were placed in ELISA wells at the 0.1 to 5 microgram range. The supernatant of antibody 421 was used at a 1 :20 dilution. No polypeptide binding of pAb 421 was detected when the same antibody was purchased from Oncogene Research Products (Ab-1; cat# OP03). However, Ab-1 did detect full-sized p53 in the same 1 :50 dilution. Antibody p53-18 (described herein) was strongly bound to the same polypeptides on the same ELISA plate. Thus, the quality of antibody 421 was dependent upon the source, and, monoclonal antibody p53-18 (described herein) was much more sensitive than antibody 421. Anti-p53 phosphopolypeptide antibodies

A double phosphorylated polypeptide was coupled to an immunodominant rabies T-helper cell epitope and was injected into mice to generate monoclonal antibodies according to the method of (Dietzschold et al, 1990, J. Virol. 64:3804-3809). The diphosphorylated polypeptide was co-synthesized with a Gly-Ala- Gly spacer and peptide 3 ID. After fusion, eight hybridomas were selected. All clones specifically bound to the immunizing peptide antigen with an affinity which was stronger than or equally as strong as the only other available polyclonal antibody 421 (antibody 421) known to specifically bind to the C-terminal region of p53. These data are shown in Table 4. Because monoclonal antibody p53-18 was an IgM, it also cross- reacted with the unphosphorylated polypeptide.

Ofthe clones shown in Table 4, clones 13, 18, 53 and 56 continued to secrete antibodies, and were selected to be maintained. MAb p53-18 was subcloned and its binding to peptide and purified p53 (expressed in E. coli) antigens was compared with that of mAb p53-13 and polyclonal antibody 421 (pAb 421). The data in Table 5 indicate that monoclonal antibody mAb p53-l 8 bound very strongly to the synthetic polypeptides. While mAb p53-18 exhibited equally strong binding to the unphosphorylated and the Ser378 phosphorylated polypeptides throughout the 5 - 0.04 microgram antigen load range, binding ofthe immunizing diphosphorylated peptide exhibited a "pro-zone" binding behavior at high antigen loads. "Pro-zone" binding behavior means that at a point of high antigen concentration, the specific binding ofthe antibody to the antigen decreases rather than increases.

When dilutions of 1 :50 of mAb p53-18 and pAb 421 were compared, it was clear that mAb p53-l 8 was significantly more sensitive than pAb 421. In fact, mAb p53-18 bound the antigen at 100 times greater sensitivity than the pAb 421 polyclonal antibody. This high sensitivity is extremely important in light ofthe very low level of p53 expression in tissues. MAb p53-18 may enable detection of p53 in tissues and cell lines where p53 was previously undetected. The continuously decreasing dose-response curve of pAb 421 binding to p53 indicated suboptimal interaction in the studied protein concentration range. This was in contrast to the optimal "pro-zone" protein binding exhibited by mAb p53-18. Thus, the sensitivity of antibody 421 was so low that the point of antigen overload could not be detected. However, the point of antigen overload could be detected when mAb p53-l 8 was tested under identical conditions.

Example 3: Preparation and assessment of mAb p53-18

In this Example, a highly sensitive monoclonal antibody (mAb) to p53, mAb p53-18, was prepared using a double phosphorylated peptide immunogen that corresponded to amino acids 371-393 ofthe C-terminal basic domain of p53. MAb p53-18 was highly specific for phosphorylated Ser378 and Ser392 at both the protein and the corresponding peptide levels in conventional aqueous environments, however, when the peptide conformation was changed during the assay procedure to that of an α-helix, mAb p53-18 also detected the unphosphorylated p53 C-terminal fragment. The irnmunodominance of the phosphorylated p53 C-terminus was indicated by the fact that cancer patients' sera preferentially labeled the same sequence as mAb p53-18, i.e., the double phosphorylated peptide antigen.

The materials and methods used in the experiments presented in this Example are now described.

Preparation of synthetic polypeptides Polypeptides were synthesized on Milligen 9050 (Burlington, MA) and

Rainin (Woburn, MA) PS3 automatic synthesizers using 9-fluorenylmethoxycarbonyl (Fmoc) amino acids according to standard procedures (Fields and Jang, 1990, Science 249:1046-1049). Phosphoserine residues were incorporated into peptides as Fmoc-Ser(PO₃HBzl)-OH (Wakamiya et al, 1994, Chem. Lett. 1099-1102), purchased from Novabiochem, Ltd. (San Diego, CA). Peptides and phosphopeptides were detached from the solid support using TFA and were purified by reversed-phase high performance liquid chromatography (RP-HPLC) using an aqueous acetonitrile gradient elution system containing 0.1% trifluoroacetic acid as an ion pairing reagent. A solution of 82.5%) TFA, 5% water, 5% thioanisole, 5% m-cresol, and 2.5% ethane-diol was used to detach peptides from the solid support. The integrity ofthe peptides and phosphopeptides was verified by mass spectroscopy.

Table 6 comprises a list ofthe sequences ofthe synthetic peptides and Table 3 contains the methods used for their characterization. Two sets of polypeptides were synthesized. One set comprised peptides of 23 amino acids in length, and the second set comprised peptides of 33 amino acids in length (the polypeptides were extended to their N-termini). Polypeptides were prepared in the following phosphorylation states: non-phosphorylated, phosphorylated with a single phosphate group on either Ser378 or Ser392, and phosphorylated with phosphate groups on both

Ser378 and Ser392.

Immunization of animals

The 23-mer double phosphorylated polypeptide was co-synthesized such that the immunodominant T-helper cell determinant 3 ID ofthe rabies virus nucleoprotein was at the N-terminus, the B-cell epitope was at the C-terminus, and a peptide spacer was between the two regions. (Ertl et al, 1989, J. Virol. 63:2885-2892). Groups of six week-old female C3H/He mice were inoculated with 10 micrograms of the tandem peptide mixed with 50% complete Freund's adjuvant in each hind leg. Fourteen days after inoculation, the mice received a booster immunization of 2 x 10 micrograms in 50% incomplete Freund's adjuvant. Ten days after the booster immunization, the mouse sera were screened and the mice were given a third immunization. At 5 days after the third immunization, mouse splenocytes were fused with myeloma cells using previously described methods (Dietzschold et al, 1990, J. Virol. 64:3804-3809). Hybridomas were grown in selective medium, screened and subcloned according to standard procedures (Goding, 1986, Monoclonal Antibodies: Principles and Practice: Academic Press, Orlando) .

High titers of monoclonal antibodies (mAbs) were produced as ascites upon injection of hybridoma cells into immunodeficient RAG-2 mice. The isotypes were determined using the Calbiochem (San Diego, CA) isotyping kit according to the manufacturer ' s specifications .

Immunoaffinity purification of baculovirus expressed murine p53

Ascites of mAb p53-18 was precipitated using saturated ammonium sulfate and the pellet was dialyzed against phosphate-buffered saline (PBS) at pH 7.6, containing 0.1 molar NaCl. The mAb was purified by size-exclusion chromatography on a Sepharose S-300 column (Pharmacia, Piscataway, NJ). Fractions (7 ml each) containing IgM immunoglobulin were collected, analyzed by sodium dodecyl- sulfate polyacrylamide gel electrophoresis (SDS-PAGE) and pooled. Ten milligrams of purified mAb p53-l 8 was coupled to a 2 milliliter column of CNBr-activated

Sepharose 4B Fast Flow (Pharmacia, Uppsala, Sweden). Unreacted functional groups on the matrix were blocked by incubation with 0.1 molar Tris-HCl pH 8.0. The column was washed several times alternating between Tris-HCl buffer having a pH of 8.0, and acetate buffer having a pH of 3.6. Murine p53 protein was obtained from a cellular extract of Sf9 insect cells infected with a recombinant baculovirus expressing murine wild-type p53. Sf9 cells were collected, lysed in PBS containing 0.1% NP-40, and centrifuged at 12,000 * g for 15 minutes. The column was equilibrated with PBS having a pH of 7.6 and comprising 0.1 molar NaCl. The supernatant containing soluble p53 was loaded onto the p53-l 8 immunoaffinity column and the column was washed with phosphate buffer and PBS to elute non-specifically bound proteins.

Murine p53 protein was eluted from the immunoaffinity column using a solution comprising 0.1 molar glycine having a pH of 2.9 into tubes containing a tenth of a fraction volume of 1 molar Tris-HCl having a pH of 8.0. The protein composition of each fraction was evaluated using 12% SDS-PAGE in the presence of protein molecular weight markers and staining using Coomassie Blue.

Western blotting method Fractions from immunoaffinity purification of murine p53 and p53 variants were assessed for specific binding to mAb p53-18 by conventional Western blotting as follows (Goding, 1986, Monoclonal Antibodies: Principles and Practice: Academic Press, Orlando). Proteins and protein fractions were electrophoresed on a 12%) SDS-PAGE gel, and, after electrophoresis, were transferred to a nitrocellulose membrane. Non-specific binding of antibody to the membrane in subsequent steps was minimized by incubating the membrane in 5%> fat-free milk dissolved in PBS containing 0.01% Tween 20. After washing, the membranes were incubated with selected dilutions of mAb p53-18. Goat anti-mouse IgG coupled with horseradish peroxidase was used as a secondary antibody in a PBS solution containing 0.01 ) Tween 20. Membranes were washed extensively with washing buffer and reacted with chemiluminescence reagent (NEN Life Science Products, Boston, MA) followed by autoradiography using X-Omat™ X-ray films (Kodak, Rochester, NY).

Enzyme-linked immunosorbent assay (ELISA) method

An ELISA direct assay was used in this Example, as described (Otvos and Szendrei, 1996, In Neuropeptide Protocols, Irvine and Williams, eds.: pp. 269-275,

Humana Press, Totowa, NJ). The active dilution range of antibody preparation was determined, and selectivity of antibody dilutions for selected peptide antigens was assessed. From about 40 nanograms to 2.5 micrograms of peptide antigen and from about 0.5 nanograms to 1.2 micrograms of protein samples were loaded in each well. Actual peptide concentrations were determined by reversed-phase HPLC (RP-HPLC)

(Szendrei et al, 1994, Eur. J. Biochem. 226:917-924).

Because mAb p53-18 is an IgM subtype, ELISA conditions were modified as follows. The peptides were applied to the plate and dried overnight. The plates were washed with a PBS solution at pH 6.8 containing 0.04% Triton XI 00. This solution was used in all subsequent steps. To assess the conformational effects ofthe polypeptide upon specific binding to the antibody, the assay was repeated under identical conditions, except that the peptides were dissolved in trifluoroethanol (TFE) instead of water prior to applying them to the ELISA plate for drying overnight (Lang et al, 1994, J. Immunol. Meth. 170:103-115).

Sera obtained from cancer patients and healthy control patients were screened for circulating anti-p53 autoantibodies using the same protocol, except that the washing buffer contained 1 milligram per milliliter of bovine serum albumin and the secondary antibody used was a sheep anti-human IgG.

The antigen specificity of mAb p53-18 was characterized by assessing binding of the mAb to 23-mer peptides. One microgram of each of the four peptides, non-phosphorylated, phosphorylated on Ser378, phosphorylated on Ser392, or phosphorylated on both Ser378 and Ser392, were assessed for specific binding to mAb p53-18 using serial dilutions of the ascites fluid that ranged from 1:100 to

1: 10,000. The double phosphorylated peptide specifically bound to antibody at an ascites dilution as high as 1:2,000, and the antibody binding increased as the antibody dilution decreased. None of the other three peptides specifically bound to mAb p53-18 below a 1 :500 dilution. An O.D. value in the ELISA assay of greater than 0.10 was used to determine that specific binding was exhibited. Specific binding of the double phosphorylated peptide by mAb p53-18 was considerably stronger than any of the other three peptides in the entire antibody dilution range. Ser392 phosphorylated peptide appeared to be slightly more recognized by mAb p53-18 than the other two variants. This experiment was repeated using the three phosphorylation variants of each of the four 33-mer peptides, and a constant dilution of the ascites fluid (1:200). The antigen load was varied from 0.004 to 5 micrograms. Only the double phosphorylated peptide bound appreciably to mAb p53-18. The specificity of mAb p53-18 for phosphorylated variants of the p53 protein was tested. Two p53 protein preparations were tested using direct ELISA and a 1:200 dilution of the ascites fluid. A p53 protein identical to the human sequence was expressed in E. coli, and was expected to be non-phosphorylated, or phosphorylated to a very low degree. The other p53 protein variant was expressed in insect cells infected with mouse p53 baculovirus recombinant as described herein, and was expected to be phosphorylated on Ser378 and Ser392 and on many other potential phosphorylation sites. As a negative control, protein E7 of human papilloma virus (HPV)- 16 was expressed using the baculovirus system under conditions substantially identical to those described herein.

The mAb p53-18 specifically bound only the phosphorylated p53 protein, and failed to bind to the non-phosphorylated variant. The mAb p53-18 binding to the phosphorylated protein exhibited some "pro-zone" binding behavior, indicating optimal antigen-antibody interactions under these experimental conditions. Positive binding was detected with as little as 0.08 micrograms of p53 protein. Some non-specific binding of mAb p53-18 to the E7 protein was observed, but was well below the level of binding to p53 and did not indicate "pro-zone" binding characteristics. Antibody p53-18 did not cross-react with the non-phosphorylated protein variant, even when as much as 0.3 micrograms of protein was used. The commercially available antibody 421 was used as a positive control in the experiment. Antibody 421 binds to a non-modified (i.e. non-phosphorylated and non-glycosylated) epitope of p53 at an area that includes Ser378 (Shaw et al, 1996, Oncogene 12:921-930). Antibody 421 binds p53 protein with low sensitivity in ELISA, and thus a high concentration of antibody must be used to detect low levels of p53 protein. Using identical assay conditions, p53 protein expressed in E. coli (0.25 microgram and higher amounts) was recognized by a 1:50 dilution of polyclonal antibody (pAb) 421. The pAb 421 bound at low levels (at antibody dilution of 1:20) to 1 microgram of the unphosphorylated 23-mer peptide (O.D. = 0.25) and did not bind to the Ser378 phosphorylated or double phosphorylated peptides.

Thus, both the peptide and protein ELISA data clearly indicate that mAb p53-18 recognized p53 protein and its fragments, regardless of the species of origin, only when the C-terminus was phosphorylated.

The binding of mAb p53-18 to p53 in a Western blotting method was assessed. The antibody bound to mouse p53 expressed in insect cells infected with baculovirus, as described herein, with the same specificity as that obtained using another commercially available mAb, antibody 240 (Oncogene Research Products, Cambridge, MA). Antibody 240 recognized an amino acid stretch in p53 around residue 215, and labels p53 from many species, e.g. human, mouse, rat, etc.(Legros et al, 1994, Oncogene 9:2071-2076).

Antibody p53-18 also reacted with different forms of p53, including p53 MD (MI234, EG168) and p53 VD (AV135), mutants of mouse p53 comprising amino acid exchanges within the domain having mutations known in cancer

(expressed by vaccinia virus in infected cells). Antibody p53-18 also cross-reacted with human wild-type p53 expressed by adenovirus in 293 cells, but failed to label the negative controls, such as insect cells infected with baculovirus alone, human immunodeficiency virus gpl 60 expressed by vaccinia virus and rabies glycoprotein expressed by adenovirus in 293 cells. Thus, mAb p53-18 specifically binds to phosphorylated p53 regardless of the origin of the protein. In brief, the specific binding exhibited by mAb p53-18 was virtually identical to that of mAb 240 to all positive and negative control protein variants tested.

Purification of p53 using an Immunoaffinity Column comprising mAb p53-18 Ammonium sulfate precipitation and size-exclusion chromatography were performed as described herein, to purify mAb p53-18 from ascites fluid of mice. The purity of fractions containing mAb p53-18 was analyzed by SDS-PAGE under reducing conditions. In the fractions containing mAb p53-18, only two bands, one corresponding to mAb p53-18 heavy chain, and one corresponding to light chain were detected on the gel. An immunoaffmity column was prepared as described herein using purified mAb p53-18 and the column was used to purify murine p53 expressed in Sf9 insect cells as described herein. Fractions eluted from the column were analyzed by SDS-PAGE and Coomassie Blue staining. The results are shown in Figure 1, which indicates pure p53 protein in lanes 1-4.

Thus, mAb p53-18, unlike some other anti-p53 antibodies, can specifically detect p53 protein in Western-blots, and immunoaffinity columns containing mAb p53-18 can isolate C-terminally phosphorylated p53 from various sources.

Conformation and DNA-binding ability of p53 polypeptides and phosphopolypeptides

The conformation and non-sequence specific DNA-binding ability of the p53 C-terminal peptides and phosphopeptides was tested (Hoffmann et al, 1998, Eur. J. Biochem.: in press). While the peptides are mostly unordered in water, the unmodified peptide assumes a relatively strong α-helix conformation in the structure-inducing solvent TFE. The basic domain is predominantly unordered in the p53 protein, but it is located proximal to the tetramerization domain (amino acids 319-360),(Sakamoto et al, 1994, Proc. Natl. Acad. Sci. USA 91 :8974-8978) which forms an anti-parallel four helical bundle, determined by multi-dimensional nuclear magnetic resonance studies (Clore et al, 1995, Science 265:386-391; Clore et al,

1995, Nature Struct. Biol. 2:321-391). It is conceivable that the conformational changes in the basic domain, due to amino acid mutations, can affect the efficiency of tetramer assembly and function of p53.

In order to determine whether mAb p53-18 recognized a conformation-dependent epitope, ELISA studies were repeated using a conformation-sensitive protocol for the unphosphorylated and the double phosphorylated 23-mer peptides. Under these assay conditions, peptides were applied to the ELISA plate from TFE, and the antigens were allowed to dry (Szendrei et al, 1994, Eur. J. Biochem. 226:917-924). The peptides interacted with the plastic surface, exposing new hydrophobic and hydrophilic areas, and the structure presented by the solvent remained conserved, which caused a remarkable variation of peptide recognition by anti-protein antibodies (Szendrei et al, 1994, Eur. J. Biochem. 226:917-924; Hoffmann et al, 1997, Biochemistry 36:8114-8124).

Using this ELISA protocol, diphosphorylated C-terminal p53 peptide was recognized at high levels by mAb p53-18, regardless of the secondary structure of the peptide antigen. The "pro-zone" binding behavior shown in Figure 2 indicated optimal antibody-antigen interactions from the entire 0.04 - 2.5 microgram peptide load range studied (the dilution of the ascites fluid was 1:1,000). The binding increase exhibited from antigen loads of 0.04 to 0.32 micrograms, and binding decrease exhibited at 0.64 micrograms and higher antigen loading was characteristic of "pro-zone" binding behavior.

In contrast, the unphosphorylated peptide, unrecognized by mAb p53- 18 during regular ELISA conditions, bound to the antibody only when the peptide was plated from TFE, as indicated in Figure 2. Nevertheless, this induced antibody binding remained at a level one order of magnitude below that of the diphosphorylated analog, and did not exhibit "pro-zone" binding behavior in the studied antigen load range. These data indicate that while the determining factor in mAb p53-18 binding to peptide antigens is the presence of phosphate groups on Ser378 and Ser392, the antibody is also sensitive to antigen conformation. Antibody 421 is believed to recognize a linear epitope, and its binding to the same unphosphorylated peptide was neither increased nor significantly decreased when the peptide was plated from TFE (O.D. = 0.18).

Binding of synthetic polypeptides to autoantibodies in sera from cancer patients and healthy controls. The specific binding of the first four synthetic polypeptides depicted in Table 6 to forty sera obtained from cancer patients and healthy controls was assessed. Sera were screened at 1 : 100 and 1:500 dilutions against unphosphorylated, Ser378 phosphorylated, and diphosphorylated 23-mer peptides. Once positive sera were identified, the specificity of the p53 autoantibodies in the sera to the presence or absence of C-terminal phosphate groups was studied by testing serial dilutions of the sera (from 1:50 to 1: 1500) against peptide antigens displaying all four possible phosphate forms and p53 protein produced in the baculovirus system as described herein. After four repetitions, five sera were identified that clearly bound to at least one of the peptides. From the five positive sera, two were originated from breast cancer patients ( sera #7 and #15), two from cervical cancer patients (sera #31 and #38), and one was from a healthy control (serum #27). When tested for specific binding to polypeptides, all five sera bound to the double phosphorylated peptide stronger than to any other phosphorylated variants. These data are depicted in Table 7. The four sera obtained from cancer patients also specifically bound the full-sized p53 protein. Serum obtained from the healthy control subject did not specifically bind to the p53 protein, indicating that the cross-reactivity with the double phosphorylated peptide was due to non-specific binding, or alternatively, that the epitope for serum #27 was not present in the protein preparation studied. As used in this paragraph, specific binding means that the O.D. values obtained in the ELISA were at least double that of the background.

Summary of the data presented in Example 3

A polyclonal antibody, specific to phosphorylated Ser392 is marketed by New England Biolabs (Beverly, MA, USA). According to the manufacturer, this antibody was obtained by inoculation of rabbits with a C-terminal phosphopeptide conjugated to a carrier protein. This pAb is claimed to specifically bind p53 only after phosphorylation with CKII at Ser392. The phosphate specificity of this pAb was compared with mAb p53-18 against p53 expressed in E. coli, Sf9 insect cells and the four 33-mer peptides with different phosphorylated forms. The polyclonal antibody did not specifically bind to the phosphorylated protein when a protein load of less than 1 microgram was used. More definite binding was observed against the E. coli expressed protein at concentrations of up to 0.1 microgram protein load. As a positive control on the same plate, mAb p53-l 8 bound to the p53 variant expressed in the baculovirus system more than twice as strongly than it bound to p53 expressed in E. coli. At a peptide level, the Ser392 phosphate specific polyclonal antibody recognized the Ser392 phosphorylated and diphosphorylated long peptides weakly at an antibody dilution of 1 :400. While complete loss of binding was observed at higher antibody dilutions, at lower dilutions the phosphate specificity seemed to be lost, explaining the detected recognition ofthe E. coli expressed protein variant.

To determine the phosphate specificity of mAb p53-18 at the protein level, p53 was expressed in two systems. To increase the production of p53 compared to mammalian tissues and cell lines, p53 is generally expressed in bacterial (Hupp et al, 1992, Cell 71 :875-886), baculovirus (O'Reilly and Millner, 1988, J. Virol.

62:3109-3119) and vaccinia virus (Ronen et al, 1992, Nucleic Acids Res. 20:3435- 3441) expression systems. The protein produced in E. coli is likely to be unphosphorylated, or phosphorylated at a low level. Bacterial environments for phosphorylation usually lack other eukaryotic kinases, and this frequently prevents proper phosphorylation from occurring (Yonemoto et al, 1997, Prot. Εng. 8:915-925).

Indeed, p53 expressed in E. coli only binds DNA properly after interaction with cellular proteins, including various kinases (Hupp et al, 1992, Cell 71 :875-886). In contrast, it has been shown that human p53 expressed in baculovirus-infected Sf9 cells displays a two-dimensional electrophoretic mobility pattern (and consequently phosphorylation pattern) identical to wild-type p53 from human cells (Patterson et al,

1996, Arch. Biochem. Biophys. 330:71-79). Insect cells appear to contain all protein kinases necessary for phosphorylation of a mammalian protein (Fuchs et al, 1995, Εur. J. Biochem. 228:625-639), with at least 9 potential p53 sites phosphorylated in this system (Patterson et al, 1996, Arch. Biochem. Biophys. 330:71-79). Accordingly, the p53 originated from the Sf9 cells was expected to have phosphate groups on both Ser378 and Ser392. Nevertheless, the lower level of p53 recognition obtained from the baculovirus system by mAb p53-18 compared to that of p53 from the bacterial system by pAb 421 indicated that the recognized sites expressed by the insect cells were not fully phosphorylated. This is further supported by the fact that pAb 421 also binds to p53 produced by baculovirus-infected Sf9 cells (Patterson et al, 1996, Arch. Biochem. Biophys. 330:71-79), and that depending upon the experimental conditions, phosphorylation of p53 in insect cells can proceed only to a low degree (Fuchs et al, 1995, Eur. J. Biochem. 228:625-639). The highly discriminative properties ofthe mAb p53-18 towards phosphorylation status of p53 makes this reagent suitable not only to detect low levels of phosphorylated p53, but in combination with other currently available antibodies, to estimate the actual degree of phosphorylation in the cell lines and tissues studied. Autoantibodies to p53 in humans, mostly in cancer patients, preferentially bound to the double phosphorylated polypeptide. Considering the relatively large distance of 14 amino acid residues between the two phosphate groups, it is reasonable to suppose that those conformers ofthe protein, and perhaps ofthe peptide, in which phosphorylated Ser378 and Ser392 are close to each other in space, preferentially populate the conformational equilibria. This logic bestows a conformational nature on the anti-p53 C-terminal antibodies. In support of this, when the unphosphorylated peptide was forced to assume a more ordered structure of an α-helix it became recognizable by mAb p53-18. A helical wheel projection of the short peptide places Ser378 and Ser392 very close to each other, as shown in Figure 3. In this regard, TFE might mimic a structure of the unphosphorylated antigen closer to the native conformation of the diphosphorylated peptide. Nevertheless the two structures are clearly different as a helix-breaking effect of phosphorylation (by using the 33-mer p53 C-terminal peptide) by circular dichroism and nuclear magnetic resonance spectroscopy was observed herein (Hoffmann et al, 1998, Eur. J. Biochem.: in press). This structural difference and the requirement for the phosphate group on the antigen for the recognition of mAb p53-18 was evident from the markedly decreased antibody binding to the unphosphorylated peptide even when both antigens were plated from TFE. Moreover, the unphosphorylated protein was not recognized, indicating that the very C-teπriinus of p53 does not populate α-helices. This is in accordance with the general expectations of conformational preferences (actually the lack thereof) of extreme termini of proteins and circular dichroism studies of a large C-terminal synthetic fragment of p53 (Sakamoto et al, 1996, Int. J. Peptide Protein Res. 48:429-442).

In summary, a novel phosphate-specific mAb to p53 was prepared. This antibody bears the promise to be a highly useful biochemical marker to detect low levels of p53 protein in different tissues, and to be a key tool to characterize the status of phosphorylation of the C-terminus of p53 protein in various cell types, solution environments and stages of tumor progression.

Table 1

r^ %

^'£ g cti ra fa fa Table 3

Table 4

Table 5

Table 6

Table 7

The disclosures of each and every patent, patent application, and publication cited herein are hereby incorporated herein by reference in their entirety.

While this invention has been disclosed with reference to specific embodiments, it is apparent that other embodiments and variations of this invention may be devised by others skilled in the art without departing from the true spirit and scope ofthe invention. The appended claims are intended to be construed to include all such embodiments and equivalent variations.

Claims

CLAIMS What is claimed is:

1. A method of making an antibody which specifically binds with a post-translationally modified p53, said method comprising a) administering to an animal a polypeptide comprising a portion of said p53 having a post-translationally modified amino acid residue and an immunogenicity enhancer; b) eliciting an antibody response thereto; and c) obtaining said antibody from said animal.

2. The method of claim 1 , wherein said antibody is a polyclonal antibody.

3. The method of claim 1, further comprising d) preparing a monoclonal antibody from said animal.

4. The method of claim 1, wherein said p53 comprises a post-translationally modified amino acid residue having a modification selected from the group consisting of phosphorylation, glycosylation and prenylation.

5. The method of claim 1, wherein said p53 is abnormally post-translationally modified in a disease.

6. The method of claim 1, wherein said polypeptide comprises said p53 and said immunogenicity enhancer.

7. The method of claim 6, wherein said immunogenicity enhancer is a T-helper cell determinant.

8. The method of claim 7, wherein said T-helper cell determinant is 31 D .

9. The method of claim 1, wherein said polypeptide further comprises a peptide spacer.

10. The method of claim 9, wherein said portion of said p53, said immunogenicity enhancer, and said spacer are cosynthesized.

11. The method of claim 10, wherein said immunogenicity enhancer is 3 ID.

12. An antibody made by the method of claim 1.

13. An antibody made by the method of claim 11.

14. A polypeptide comprising a portion of a p53 comprising a post-translationally modified amino acid residue having a modification selected from the group consisting of phosphorylation, glycosylation and prenylation and an immunogenicity enhancer.

15. The polypeptide of claim 14, wherein said portion of said p53 is at the C-terminus of said polypeptide, and is covalently linked to said immunogenicity enhancer, and further wherein said immunogenicity enhancer is at the N-terminus of said polypeptide.

16. The polypeptide of claim 15, further comprising a peptide spacer between said portion of said p53 and said immunogenicity enhancer.

17. The polypeptide of claim 16, wherein the length of said polypeptide is from about 20 to about 45 amino acid residues.

18. The polypeptide of claim 17, wherein the length of said polypeptide is 41 amino acid residues.

19. A monoclonal antibody which binds specifically to a post- translationally modified variant of p53, wherein the immunogen used to produce said antibody is said post- translationally modified variant of p53 in combination with an immunogenicity enhancer.

20. The monoclonal antibody of claim 19, wherein said post- translationally modified variant of p53 comprises a modification selected from the group consisting of phosphorylation, glycosylation and farnesylation.

21. The monoclonal antibody of claim 20, wherein said post- translationally modified variant of p53 comprises a biphosphorylated C-terminus wherein Ser378 and Ser392 are phosphorylated.

22. The monoclonal antibody of claim 20, wherein said post- translationally modified variant of p53 comprises a monophosphorylated C-terminus.

23. A monoclonal antibody which binds specifically to a peptide comprising a phosphorylated p53, a peptide spacer, and 3 ID.

24. A method of assessing the post-translational modification status of a p53 obtained from a human patient, said method comprising a) obtaining a biological sample from a patient comprising said ρ53; b) contacting an antibody made by the method of claim 1 with said p53; c) forming an antigen-antibody complex between said p53 and said antibody, and d) detecting said antigen-antibody complex, wherein the presence of said complex defines said post-translational modification status of said p53.

25. A method of determining the presence or absence of an autoantibody specific for a post-translationally modified p53 in a biological sample of a human patient, said method comprising a) obtaining a biological sample from said patient; b) contacting said biological sample with the polypeptide of claim 14; c) forming an antigen-antibody complex between said polypeptide and said autoantibody, and d) detecting said antigen-antibody complex; wherein the presence of said complex is an indication that said autoantibody is present in said biological sample.