WO1996018899A1

WO1996018899A1 - Method for detecting a phosphorylated amino acid in an intact protein

Info

Publication number: WO1996018899A1
Application number: PCT/US1995/016421
Authority: WO
Inventors: Malcolm Whitman
Original assignee: Harvard University
Current assignee: Harvard University
Priority date: 1994-12-16
Filing date: 1995-12-15
Publication date: 1996-06-20
Anticipated expiration: 1997-06-16

Abstract

Methods and compositions for determining the presence of a phosphorylated amino acid in an intact protein are provided. The methods involve dephosphorylating the phosphorylated amino acid to form a dehydroamino acid and covalently coupling a detectable tag to the dephosphorylated amino acid in the intact protein. The methods and compositions are useful for determining the presence of phosphoserine and phosphothreonine in intact proteins.

Description

METHOD FOR DETECTING A PHOSPHORYLATED AMINO ACID IN AN INTACT PROTETN

Government Support This work was funded by a government grant from the National Institute of Child Health and Human Development, Grant No. HD29468. The government has certain rights in this invention.

Field of the Invention This invention relates to methods and compositions for detecting phosphorylated serine and threonine amino acid residues in intact proteins. The methods involve dephosphorylating the phosphorylated amino acid and covalently coupling a detectable tag to the dephosphorylated amino acid in the intact protein.

Background of the Invention Protein phosphorylation represents a leading mechanism in the post-translational regulation of protein function and cellular metabolism. Unfortunately, efforts to identify phosphorylated proteins and to characterize particular phosphorylated amino acid residues have met with limited success because of the high sensitivity requirements- for studying protein phosphorylation in vivo. Although serine reportedly is the predominant site of in vivo protein phosphorylation, there is currently no generally acceptable method for studying serine phosphorylation in vivo. Those methods which are available, e.g., metabolic labeling with inorganic ³²[P], typically involve using millicurie quantities of radioisotope for the detection of phosphorylated amino acids in non-abundant proteins and cannot easily be adapted for studying protein phosphorylation in tissues derived from whole animals. In summary, the relative insensitivity of the foregoing metabolic labeling techniques, together with the risk of exposure to high levels of radioactivity and the concomitant generation of large amounts of radioactive waste, have precluded the widespread acceptance of metabolic labeling with ³²[P] for studying protein phosphorylation in vivo.

Although antibody detection has been reported as a successful alternative to radioactive metabolic labeling for detecting phosphorylated tyrosine residues in vivo, antibody detection of phosphoserine and phosphothreonine residues has proven to be problematic. In particular, the relatively low binding affinity of currently available anti-phosphoserine and anti- phosphothreonine antibodies, coupled with the context-dependent binding of these agents, has precluded using antibodies in studying serine and/or threonine phosphorylation events in vivo. Summary of the Invention

The present invention overcomes these problems by providing methods and compositions for detecting phosphoserine and/or phosphothreonine residues in an intact protein. The methods of the invention can be practiced using radiolabel or non-radiolabel detection systems that are well known to those of skill in the art of protein labeling. Thus, in contrast to the prior art, the methods do not require metabolic labeling in vivo and do not require large quantities of radiolabel for detection of the phosphorylated amino acids. Also provided are kits for determining the presence of a phosphoserine or phosphothreonine residue in an intact protein. The methods of the invention are based upon the recognition that two reactions, a beta- elimination reaction and an addition reaction, can be used in combination to detect phosphorylated amino acids in an intact protein. As a result, the methods and compositions of the invention are useful for detecting phosphoserine and phosphothreonine residues in a variety of protein samples, including crude protein extracts and purified protein preparations containing, for example, immunoprecipitated proteins. Accordingly, the invention provides a useful alternative to metabolic labeling for studying the changes in regulatory serine and/or threonine phosphorylations in vivo.

According to one aspect of the invention, a method for detecting a phosphorylated amino acid in an intact protein is provided. As used herein, "phosphorylated amino acid" refers to a phosphoserine or phosphothreonine residue in an intact protein. The method involves: (1) subjecting the protein to conditions for effecting a beta-elimination reaction in the intact protein to convert the phosphorylated amino acid to a dehydroamino acid including a double bond containing carbon atoms that are susceptible to attack by an addition reaction; (2) contacting the protein of step (1) with a reactant under conditions for adding the reactant to the carbon atom of the double bond of the dehydroamino acid to form an amino acid addition product in the intact protein; and (3) detecting the amino acid addition product in the intact protein, wherein the presence of the amino acid addition product is indicative of the presence of the phosphorylated amino acid in the intact protein.

Most proteins contain cysteine, cystine and/or lysine residues which may interfere with the sensitivity and/or accuracy of the detection methods disclosed herein. Accordingly, it is recommended that cysteine, cystine and/or lysine residues be modified prior to conducting the beta-elimination to avoid their participation in side reactions which could interfere with the determination of phosphoserine and phosphothreonine residues in the intact protein. Methods for modifying cysteine, cystine and lysine residues are described in more detail below.

In general, the conditions for effecting a beta-elimination reaction involve exposing the protein to a dilute basic aqueous solution having a Ph between about 10 and 14 for between about 10 minutes and twelve hours. As will be apparent to those of ordinary skill in the art, the beta-elimination reaction will proceed faster at a higher pH. Accordingly, optimization of the precise incubation conditions for effecting the beta-elimination reaction, while at the same time minimizing protein degradation due to exposure to high Ph, can be accomplished using no more than routine experimentation. In a particularly preferred embodiment, a beta-elimination catalyst (e.g., a group II metal ion) can be added to facilitate the rate of reaction for the beta-elimination reaction. As used herein, "beta-elimination catalyst" refers to a substance which increases the kinetic rate of the beta-elimination reaction.

The beta-elimination reaction results in conversion of the phosphorylated amino acids to the corresponding dehydroamino acids. The dehydroamino acid includes a double bond containing carbon atoms that are susceptible to attack by an addition reactant in an addition reaction. As used herein, an addition reactant refers to a molecule that is capable of forming a covalent bond with a carbon atom of the dehydroamino acid in an addition reaction. Although the invention is not limited in scope to a particular mechanism, it is believed that the addition reactions disclosed herein primarily proceed via nucleophilic addition. Thus, in a particularly preferred embodiment, the dehydroamino acid double bond is subject to a nucleophilic addition reaction to form a nucleophilic addition product.

The "reactant" in a nucleophilic addition reaction is referred to herein as a nucleophile and the nucleophilic addition product is referred to herein as a nucleophilic amino acid addition product. The invention embraces two types of nucleophile: (1) a monofunctional nucleophile which contains a detectable tag for determining the presence of the phosphorylated amino acid in the intact protein in a "one-step" reaction and (2) a bifunctional nucleophile which does not contain a detection reagent and which therefore must be further reacted with a detection reagent to determine the presence of the phosphorylated protein in a "two-step" reaction. A monofunctional nucleophile is a nucleophile which contains a single functional group for covalent attachment to an amino acid residue in an intact protein and further includes a detectable tag for detecting the presence of the phosphorylated amino acid in the intact protein. A bifunctional nucleophile is a nucleophile which contains two functional groups for forming two distinct covalent bonds, i.e., the bifunctional nucleophile contains a first functional group for reacting with the carbon atom of the double bond of the dehydroamino acid and a second functional group for reacting with the detection reagent.

As used herein, "detectable tag" refers to a molecule that can be directly detected (i.e., the molecule itself can be detected) or indirectly detected (i.e., the molecule is detected by a further reaction). Exemplary detectable tags include radioactive tags (e.g., ³⁵[S], ³²[P], ^l4[C]), chromophores, fluorophores, luminescent tags, biotin, haptens and enzymes. In general, such detection procedures are performed using routine procedures well known to those of ordinary skill in the art.

Reaction of the intact protein with the bifunctional nucleophile results in the introduction into the protein of a new functional group ("Y") to which a detection reagent can be covalently coupled. Thus, practicing the invention in which a bifunctional nucleophile is used for the addition reaction involves: (1) subjecting the intact protein to conditions for effecting a beta- elimination reaction to convert the phosphorylated amino acid to a dehydroamino acid, (2) contacting the protein of step (1) with the bifunctional nucleophile under conditions for adding the nucleophile to the carbon atom of the double bond to form a nucleophilic addition product in the intact protein, (3) contacting the protein of step (2) with the detection reagent under conditions for the detection reagent to covalently bond to the second functional group of the bifunctional nucleophile, and (4) detecting the nucleophilic addition product in the intact protein as described above. In the particularly preferred embodiments, the bifunctional nucleophile contains a second functional group that is an amine group or a thiol group and the detection reagent is an amino group or thiol group labeling reagent, respectively. Exemplary amino group and/or thiol group labeling reagents are well known to those of skill in the art of protein labeling.

Alternatively, the phosphorylated amino acids can be detected by oxidizing the double bond of the dehydroamino acid (the product of the above-described beta-elimination reaction) with KMn0₄ to yield a cis diol which then can be oxidized with periodate to a highly reactive aldehyde. The resultant aldehyde is reacted with, for example, biotin-hydrazide to yield a biotin- labeled amino acid in the intact protein at the location of the dehydroamino acid.

According to yet another aspect of the invention, kits for detecting phosphoserine and/or phosphothreonine in an intact protein are provided. The kits include a nucleophile (monofunctional or bifunctional) which can add to the carbon atoms of the double bond of a dehydroamino acid and instructions for using the nucleophile to perform the method of the invention. Kits which include the bifunctional nucleophile further include a detection reagent which can form a covalent complex with one functional group of the bifunctional nucleophile. Optionally, the kits further include protein controls for evaluating the sensitivity and accuracy of the method for detecting the presence of a phosphorylated amino acid in an unknown protein.

The detection of phosphoserine and phosphothreonine in accordance with the methods of the invention is useful for characterizing the molecular events leading to and resulting from the phosphorylation of serine and/or threonine residues, in a manner analogous to studies performed using antibodies to phosphotyrosine to characterize the molecular events leading to and resulting from the phosphorylation of tyrosine residues in vivo.

A variety of phosphorylation regulated processes currently are targeted by a broad spectrum of pharmaceutical agents for modulating a physiological condition. Such agents and the physiological targets and/or conditions to which they are directed include the immunosuppressive cyclosporin (which targets calcineurin, a serine/threonine protein kinase). the immunosuppressive rapamycin (which reportedly inhibits p70^So kinase, a serine/threonine kinase) and the beta-adrenergic blockers which prevent adrenergic activation of cAMP- dependent protein kinases. Thus, the methods and compositions of the invention are useful for evaluating the efficacy of putative pharmacological agents to specifically block a particular targeted metabolic pathway in vivo. Further, because changes in protein phosphorylation have been correlated to a broad range of normal physiological functions (e.g., glycogen metabolism, cholesterol metabolism, neurotransmitter function) and disease pathogenesis (e.g., leukemia), the methods and compositions of the invention also have utility in evaluating disease pathologies and assessing clinical prognosis.

The receptors for insulin and certain growth factors, such as platelet-derived growth factor, are membrane-bound tyrosine protein kinases for which the kinase enzymatic activity reportedly is essential for proper receptor function in vivo. See, e.g., Harrison's Principles of Internal Medicine, 12th ed., J. Wilson, et al. editors, McGraw-Hill, Inc., New York, NY (1991 ). Such results are consistent with more recent reports that the phosphorylation state of certain protein amino acid residues (e.g., tyrosines) correlates to neoplastic disease states (e.g., leukemia).

The methods and compositions of the invention also have utility in improving routine protein characterization methods, e.g., peptide mapping and amino acid analysis, with respect to phosphoserine and/or phosphothreonine detection. For example, it is believed that the methods and compositions disclosed herein can be used to increase the sensitivity and accuracy of phosphoserine and/or phosphothreonine detection, e.g., by providing a detection method that is more sensitive than the detection methods that are currently available for detecting these phosphorylated amino acids and/or by providing a method that can be adapted to generate a variety of detectable signals depending upon the selection of the particular detectable tag for labeling (e.g., a fluorophore, a chemiluminescent tag).

These and other aspects of the invention, as well as various advantages and utilities will be more apparent with respect to the detailed description of the preferred embodiments and in the accompanying drawings. All references cited in this application are incorporated in their entirety herein by reference. Brief Description of the Drawings

Figure 1 shows the results obtained following dephosphorylation/biotinylation of BSA and casein as described in Example 1. Purified bovine serum albumin (B) (an unphosphorylated protein) or casein (C) (a serine/threonine phosphoprotein) were treated with DTT/NEM to alkylate endogenous cysteines, dephosphorylated by alkaline beta-elimination, labeled by addition of dithiopropane and biotin-maleimide, and analyzed by Laemmli gel electrophoresis/streptavidin Western blot. Figure 1 A shows the streptavidin detection results of approximately 500 ng BSA (B) or 250 ng casein (C) after dephosphorylation/ DTP/ biotin- labeling. Figure IB shows the Coomassie staining results of the starting material that was used in the detection procedure, showing the relative amounts of total BSA and casein protein. Detailed Description of the Invention

The instant invention embraces methods for detecting phosphorylated amino acids in an intact protein and related compositions. More specifically, methods for identifying phosphoserine and/or phosphothreonine amino acid residues in an intact protein are provided. The invention is useful for detecting these phosphorylated amino acids in virtually any intact protein without regard to the primary, secondary and/or tertiary protein structure in the vicinity of the phosphorylated amino acids.

As used herein, "phosphorylated amino acid" refers to a phosphoserine or a phosphothreonine residue. The phosphoesters of the primary and secondary aliphatic alcohols in phosphoserine and phosphothreonine are base-labile. In contrast, the aromatic phosphoester in phosphotyrosine is acid-labile and base-stable. Accordingly, subjecting the phosphotyrosine residues to basic aqueous solution in accordance with the method of the invention (discussed in detail below) does not effect a beta-elimination reaction, an essential step in practicing the method of the invention. Thus, the methods of the invention specifically detect phosphoserine and phosphothreonine in an intact protein without interference from phosphotyrosine.

The methods of the invention involve: (1) subjecting the protein to conditions for effecting a beta-elimination reaction in the intact protein to convert the phosphorylated amino acid to a dehydroamino acid containing a double bond containing carbon atoms that are susceptible to attack by an addition reaction; (2) contacting the protein of step (1) with a reactant under conditions for adding the reactant to the carbon atom of the double bond of the dehydroamino acid to form an amino acid addition product in the intact protein; and (3) detecting the amino acid addition product in the intact protein, wherein the presence of the amino acid addition product is indicative of the presence of the phosphorylated amino acid in the intact protein.

In general, the proteins that are analyzed according to the methods of the invention contain at least one cysteine and/or cystine residue. Accordingly, to avoid participation of the cysteine/cystine residues in side reactions (discussed below), typically the protein is subjected to conditions (prior to exposing the protein to conditions for effecting the beta-elimination reaction) to convert the cysteine(s)/cystine(s) to a product (referred to herein as a "cysteine product") which cannot undergo a beta-elimination reaction or otherwise interfere with the detection reactions disclosed herein. For example, the intact protein can be exposed to an oxidizing reagent (e.g., performic acid) to oxidize the cysteine(s)/cystine(s) to cysteic acid residue(s) or modified by alkylating the sulfhydryl group of the cysteine(s)/cystine(s) prior to performing the beta-elimination reaction.

In general, conditions for effecting a beta-elimination reaction are well known to those of ordinary skill in the art. For the purpose of practicing the instant invention, relatively mild conditions for effecting a beta-elimination reaction are selected to prevent peptide bond hydrolysis. Accordingly, subjecting the protein to the selected beta-elimination conditions results in an intact protein in which phosphoserine and phosphothreonine are dephosphorylated to their respective dehydroamino acids.

According to one embodiment of the invention, the conditions for effecting a beta- elimination reaction involve placing the intact protein in a dilute basic aqueous solution (i.e., a solution having a pH between 10 and 14, inclusive) for between about 10 minutes and 12 hours. More preferably, the dilute basic aqueous solution has a pH between 1 1 and 13 and the intact protein is placed in basic solution for between about 0.5 and 6 hours. In a particularly preferred embodiment, the basic aqueous solution is a sodium hydroxide solution having a sodium hydroxide concentration ranging between 0.05 and 0.5 normal, more preferably, the basic solution is 0.10 N NaOH and the protein is placed in this basic solution for between 0.5 and 6.0 hours. In a particularly preferred embodiment, the protein is placed in the basic solution for between 0.5 and 3.0 hours.

Other basic aqueous solutions which can be used to practice the invention include sodium borate, potassium borate, potassium hydroxide, calcium hydroxide, barium hydroxide and 3-[cyclohexylamino]- 1 -propane-sulfonic acid (CAPS) buffers. See also, A. Scaloni, et al., Anal. Biochem. 218:226-228 (1994). Thus, as used herein, a dilute basic buffer refers to any basic solution which can maintain a pH in the range of about 10 to 14 and further, which is not reactive with the dehydroamino acid that is formed by the beta-elimination reaction.

In a particularly preferred embodiment, the basic aqueous solution further includes a catalytic amount of a beta-elimination catalyst. As used herein, a "beta-elimination catalyst" refers to a substance which increases the kinetic rate of the beta-elimination reaction. A "catalytic amount" of the beta-elimination reaction catalyst is an amount which increases the kinetic rate of the reaction to an extent that is statistically significant. Exemplary beta- elimination reaction catalysts include the group II metal ions. See, e.g., M. Byford, Biochem. J. 280:261-265 (1991). In general, the basic aqueous solution contains between 0.01 and 1.0 moles per liter of the group II metal ion beta-elimination reaction catalyst. In a particularly preferred embodiment, the beta-elimination catalyst is SrCl₂ or Ba(OH)-,, which is present in the basic aqueous solution at a concentration ranging between 0.10 and 0.20 moles per liter. Most preferably, the basic aqueous solution is 0.10 N NaOH containing 0.10 M SrCl₂ or Ba(OH)₂. As would be apparent to one of ordinary skill in the art, other basic aqueous solutions can be substituted for the sodium hydroxide solution and other catalysts can be substituted for SrCU or Ba(OH)₂ without departing from the essence of the invention.

The beta-elimination reaction results in conversion of the phosphorylated amino acids to the corresponding dehydroamino acids, i.e., phosphoserine is converted to dehydroalanine and phosphothreonine is converted to 2-aminodehydroxybutyric acid. The dehydroamino acid includes a double bond containing carbon atoms that are susceptible to attack by an addition reactant in an addition reaction. As used herein, an addition reactant refers to a molecule that is capable of forming a covalent bond with a carbon atom of the dehydroamino acid in an addition reaction. In general, there are four different addition reaction mechanisms by which an addition reactant can add to a double bond: nucleophilic addition (J. March, Advanced Organic Chemistry, 4th ed., John Wiley & Sons, N.Y., N.Y. (1992), pages 741-743), electrophilic addition (March, ibid., pages 734-741), free radical addition (March, ibid., pages 743-745) and cyclic mechanisms in which both carbons of the double bond are simultaneously attacked (March, ibid., page 745). Although the invention is not limited in scope to a particular mechanism, it is believed that the addition reactions disclosed herein primarily proceed via nucleophilic addition. Thus, in a particularly preferred embodiment, the dehydroamino acid double bond is subject to a nucleophilic addition reaction to form a nucleophilic addition product. A nucleophilic addition is favored because of the electron- withdrawing effect of the carbonyl group present in the adjacent amide (peptide) bond which enhances nucleophilic addition and inhibits electrophilic addition (under mild basic conditions) by lowering the electron density of the dehydroamino acid double bond, i.e., the dehydroamino acid acts as a Michael substrate which is particularly susceptible to nucleophilic attack. This interpretation is consistent with reports that phosphothreonine undergoes nucleophilic addition at a slower rate than phosphoserine because of the effect of the phosphothreonine methyl group on the nucleophilic reactivity of the beta-carbon in the corresponding dehydroamino acid.

Conditions for performing a nucleophilic addition reaction between a dehydroamino acid and a nucleophile have been described previously using aliphatic primary amines, sulphite or aliphatic thiols as nucleophiles. See, e.g., T. Soderling and K. Walsh, J. Chromatography

253:243-251 (1982) and M. Byford, Biochem. J. 280:261-265 (1991) and references identified therein.

The "reactant" in a nucleophilic addition reaction is referred to herein as a nucleophile and the nucleophilic addition product is referred to herein as a nucleophilic amino acid addition product. As used herein, the terms "nucleophile" and "electrophile" have their common meanings. See, e.g., J. March, J., ibid. In a nucleophilic/electrophilic reaction, one member of a complementary pair of reactants is an electrophile (such as a carbon atom of a carbon-carbon double bond) and the other member is a nucleophile (such as an amine, a thiol or an alcohol). In general, the nucleophilic addition reaction is performed using a molar excess of the nucleophile in order to drive the addition reaction to completion. Preferably, the molar ratio of nucleophile to intact protein is at least about 10 : 1, more preferably, the molar ratio of nucleophile to protein is between 10⁵:1 and lO¹⁴:! . The invention embraces two types of nucleophile: (1) a monofunctional nucleophile which contains a detectable tag for determining the presence of the phosphorylated amino acid in the intact protein in a "one-step" reaction and (2) a bifunctional nucleophile which does not contain a detection reagent and which therefore must be further reacted with a detection reagent to determine the presence of the phosphorylated protein in a "two-step" reaction. As used herein, a monofunctional nucleophile is a nucleophile which contains a single functional group for covalent attachment to an amino acid residue in an intact protein. A bifunctional nucleophile is a nucleophile which contains two functional groups for forming two distinct covalent bonds, i.e., the bifunctional nucleophile contains a first functional group for reacting with the carbon atom of the double bond of the dehydroamino acid and a second functional group for reacting with the detection reagent.

According to the "one-step" reaction, the nucleophile contains a first functional group ("X", a nucleophilic group such as a thiol or an amine) for reacting with the carbon atom of the double bond of the dehydroamino acid and a detectable tag ("D") for determining the presence of the phosphorylated amino acid in the intact protein. The first functional group and the detection reagent are separated from one another by between zero to ten carbon atoms. The number of carbon atoms in the chain separating the first functional group and the detection reagent is selected so that the nucleophile is soluble at a sufficient concentration to achieve the purposes of the reaction. Solubility of the chain in aqueous solution can be increased by increasing the proportion of hydroxyl groups present on the carbon chain. The monofunctional nucleophile has the formula,

Rl

X — [C]„ — D I

R2

wherein n = 0 to 10, inclusive; Rl is H or OH; R2 is H or OH and wherein Rl and R2 can be the same or different from one another. In the preferred embodiments, the first functional group and the detection reagent are separated from one another by between two and five carbon atoms, i.e., n = 2 to 5, inclusive. As would be apparent to one of ordinary skill in the art, various modifications of the foregoing formula can be made without departing from the essence of the invention. Accordingly, Rl and R2 optionally can embrace alkyl groups containing between one and ten carbon atoms, provided that the presence of the alkyl group (1) does not reduce the solubility of the nucleophile to a degree that would interfere with the efficiency or sensitivity of the reaction, (2) does not inhibit formation of a covalent bond between the first functional group and the dephosphorylated amino acid and (3) does not interfere with detection of the detectable tag. As used herein, "detectable tag" refers to a molecule that can be directly detected (i.e., the molecule itself can be detected) or indirectly detected (i.e., the molecule is detected by a further reaction). Exemplary detectable tags include radioactive tags (e.g., ³⁵[S], ³²[P], ^l4[C]), chromophores (e.g., Texas Red dye), fluorophores (e.g., fluorescein isothiocyanate (FITC), tetramethylrhodamine isothiocyanate (TRITC)), luminescent tags (e.g., aminobutylethylisoluminol), biotin, haptens (e.g., fluorescein, dinitrophenol) and enzymes (e.g., horseradish peroxidase, alkaline phosphatase, beta galactosidase). Exemplary procedures for detecting a biotin-, hapten- or enzyme-labeled dephosphorylated amino acid in an intact protein are discussed below. In general, such detection procedures are performed using routine procedures well known to those of skill in the art. Exemplary monofunctional nucleophiles for use in the "one-step" reaction of the invention include amines and thiols which contain a detectable tag (e.g., H₂N-(CH₂)_n-Biotin and H₂S-(CH₂)_n-Biotin, respectively wherein n = 4 to 8, inclusive). According to this embodiment, contacting the protein with the (detection reagent-containing) nucleophile under conditions for adding the nucleophile to the double bond of the dehydroamino acid yields a biotin-labeled intact protein, i.e., an intact protein in which the biotin moiety is covalently coupled to the dehydroamino acid. Thus, detecting the nucleophilic addition amino acid product in the intact protein involves detecting the biotin moiety in the intact protein.

Methods for detecting a biotin moiety in an intact protein are well known to those of ordinary skill in the art. In general, these methods are based upon the high affinity of streptavidin (K_D=10 ¹⁵) for biotin (Vitamin H). One molecule of streptavidin binds four molecules of biotin by a non-covalent interaction which is essentially irreversible. Typically, a biotin/streptavidin detection system involves coupling biotin to a molecule in a process referred to as biotinylation to form a biotinylated molecule (e.g., H₂N-CH₂-CH₂-Biotin) and coupling an indicator molecule such as an enzyme or a fluorochrome to streptavidin or avidin for detecting the presence of the biotinylated molecule. These reagents can be used in a variety of assays (e.g.. ELISA, immunoblotting, immunohistochemical staining and fluorescent cell sorting) to detect the biotinylated molecule. Exemplary biotin/avidin labeling system immunoassay kits and reagents are commercially available (e.g., ECL kits and reagents available from Amersham Corp., Arlington Heights, IL; Sigma Chem. Co., St. Louis, MO) and can be used without undue experimentation to detect the biotin-labeled dehydroamino acid in the intact protein.

In a particularly preferred embodiment, the phosphorylated amino acids in the intact protein are detected as described in Example 1. This detection method employs a commercially available chemiluminescence kit (an ECL system, Amersham) containing a streptavidin- horseradish peroxidase (HRP) conjugate as the detection reagent in a chemiluminescence assay. The enzyme (e.g., peroxidase) acts as a catalyst for oxidation of the substrate luminol, which subsequently emits light that can be recorded on photosensitive film. An exemplary ECL assay is provided in Example 1. Additional ECL protocols and references describing the commercially available chemiluminescence assay are provided in the Amersham 1994 catalog. Other non- biotin ECL detection protocols are discussed in more detail below.

Commercially available avidin conjugates that are useful for detecting a biotin-labeled amino acid include avidin-alkaline phosphatase, avidin-beta galactosidase, avidin-fluorescein isothiocyanate (FITC), avidin-peroxidase, avidin-tetramethylrhodamine isothiocyanate (TRITC), avidin-gold (i.e., avidin adsorbed to colloidal gold), streptavidin-FITC, streptavidin-beta- galactosidase, streptavidin-gold, streptavidin-peroxidase, and streptavidin-Texas Red. See, e.g., Sigma Chemical Co. Catalog, ibid.

An enhanced chemiluminescent reaction for detection of phosphorylated amino acids in an intact protein also can be performed using a hapten-based detection system. According to this embodiment of the invention, the reactant (e.g., a monofunctional nucleophile) contains a hapten (e.g., fluorescein) and the addition reaction (e.g., nucleophilic addition) yields a hapten-labeled amino acid addition product (e.g., the dehydroamino acid is labeled with the hapten). Detection of the hapten-labeled addition product in the intact protein is performed by reacting the intact protein with an antibody conjugate. The antibody conjugate contains an antibody which specifically recognizes the hapten and an enzyme which catalyzes the above-described chemiluminescent reaction (e.g.. luminol is the substrate). In a particularly preferred embodiment, the antibody conjugate contains an anti-fluorescein antibody covalently coupled to horse radish peroxidase. The reagents for performing the hapten-based detection of the dehydroamino acids in the intact protein are commercially available from Amersham Corp., Arlington Heights, IL.

Alternative detection reagents can be used in lieu of the above-described biotin/avidin and hapten labeling systems. For example, an ³⁵[S]-labeled nucleophile (e.g., H₂N-CH₂-CH₂-³⁵[S]H₂, H₂S-CH₂-CH₂-³⁵[S]H₂) can be used to directly incorporate a radioactive tag in the intact protein by addition to the carbon atoms of the dehydroamino acid double bond. Similarly, the dehydroamino acid-containing protein can be reacted with ³⁵[S]-sulfite to directly introduce a radioactive tag to the intact protein at the location of the dehydroamino acid double bond. See, e.g., T. Soderling and K. Walsh, J. Chromatography 253:243-251 (1982); H.Meyer et al., FEBS LETT. 204:61-66 (1986) and references cited therein.

The selection of additional detectable tags, as well as the selection of monofunctional nucleophiles that contain the detectable tags, which can be used in accordance with the methods of the invention can be made using routine experimentation, for example, by substitution a putative detectable tag for the detectable tag of Example 1 and determining whether the putative detectable tag provides a sensitivity and specificity that is at least comparable to that of a detectable tag which is known to be useful for detecting phosphorylated amino acids in an intact protein. Alternatively, a bifunctional nucleophile can be used to practice the methods of the invention in accordance with a "two-step" reaction, i.e., the bifunctional nucleophile serves as a linker to covalently attach a detection reagent containing a detectable tag to the dehydroamino acid in the intact protein. According to this embodiment, the nucleophile contains a first functional group ("X", a nucleophilic group such as a thiol or an amine) for reacting with the carbon atom of the double bond of the dehydroamino acid and a second functional group ("Y") for reacting with a detection reagent. The first and the second functional groups are separated from one another by between zero to ten carbon atoms, i.e., the bifunctional nucleophile has the formula,

Rl

I

X - [C]_n - Y

R2

wherein n = 0 to 10, inclusive; Rl is H or OH; R2 is H or OH and wherein Rl and R2 can be the same or different from one another. The same limitations apply to the carbon chain between the X and Y groups and of the Rl and R2 groups as discussed above in regard to the monofunctional nucleophile. In the preferred embodiments, the first and second functional groups are separated from one another by between two and five carbon atoms, i.e., n = 2 to 5, inclusive. As would be apparent to one of ordinary skill in the art, various modifications of the foregoing formula can be made without departing from the essence of the invention. Thus, the invention broadly embraces methods for detecting phosphorylated amino acids in an intact protein and is not intended to be limited in scope to the exemplary bifunctional nucleophiles disclosed herein. Exemplary nucleophiles that can be used in accordance with the "two-step" embodiment include H₂N-CH₂-CH₂-NH₂ (1,2- diaminoethane), H,N-CH₂-CH₂-CH₂-NH₂ (1,3- diaminopropane), diaminopropanol, H₂S-CH₂-CH₂-SH₂ (dithioethane) and H₂S-CH₂- CH₂-CH₂-SH₂ (1,3-propane dithiol; also referred to herein as dithiopropane, DTP), dithiothreitol and dithioerythritol. Alternative bifunctional nucleophiles can be selected from putative bifunctional nucleophiles in an analogous manner to that described above for the selection of monofunctional nucleophiles and detectable tags, namely, by substituting a putative bifunctional nucleophile for a bifunctional nucleophile that is known to be useful for detecting a phosphorylated amino acid in an intact protein and determining whether the putative bifunctional nucleophile provides a sensitivity and specificity that is at least comparable to that obtained using the reference bifunctional nucleophile.

Reaction of the intact protein with the bifunctional nucleophile results in the introduction into the protein of a new functional group ("Y") to which a detection reagent can be covalently coupled. Accordingly, it is desirable to remove as much as possible (e.g., by evaporation, extraction) any uncoupled bifunctional nucleophile from the reaction mixture prior to contacting the nucleophile-modified protein with the detection reagent.

Practicing the invention in which a bifunctional nucleophile is used for the addition reaction involves: (1) subjecting the intact protein to conditions for effecting a beta-elimination reaction to convert the phosphorylated amino acid to a dehydroamino acid, (2) contacting the protein of step (1) with the bifunctional nucleophile under conditions for adding the nucleophile to a carbon atom of the double bond to form a nucleophilic addition product in the intact protein. (3) contacting the protein of step (2) with the detection reagent under conditions for the detection reagent to covalently bond to the second functional group of the bifunctional nucleophile, and (4) detecting the nucleophilic addition product in the intact protein as described above.

In a particularly preferred embodiment, bifunctional nucleophile contains a second functional group that is an amine group and the detection reagent is an amino group labeling reagent. Exemplary amino group labeling reagents are well known to those of skill in the art of protein labeling and include biotin-succinimidyl ester, the Bolton and Hunter reagent and its derivatives (e.g., N-succinimidyl 3-(4-hydroxy, 5-[^,25-I]-iodophenyl) propionate and the corresponding di-iodo derivative), as well as [³⁵S]-sulfur labeling reagents which label free amino groups in the intact protein (e.g., [³⁵S] sulfur labeling reagent product no. SJ440, 1994 Life Science Catalog, Amersham Corp., Arlington Heights, IL). Exemplary thiol labeling reagents include N-ethyl[2,3-^l4C]-maleimide, biotin-maleimide, [¹⁴C]-methyl iodide, [¹⁴C]- succinic anhydride, iodoacetylbiotin and N-hydroxysuccinimido-biotin (NHS-biotin).

The foregoing labeling reagents are known to label free amine or thiol groups in proteins. Accordingly, to prevent unwanted side reactions between the labeling reagents and the N- terminal or lysine side chain amine groups, it is essential that the intact protein be treated to block free amine groups (e.g., by alkylating the free N-terminal amine group and lysine side chain amine groups) prior to contacting the protein with the bifunctional nucleophile. Succinylation of a peptide containing a lysine residue (prior to performing a beta- elimination/pyridoxamine labeling reaction) reportedly prevented a side reaction in which the epsilon-amino group of the lysine added across the double bond of the dehydroalanyl residue. (T. Hastings and E. Reimann, FEBS LETT. 231(2):431-436 (1988)). Similarly, proteins which contain cysteine or cystine residues should be subjected to conditions to convert the cysteine thiol groups to a form (referred to herein as a "cysteine product") which cannot undergo a beta- elimination reaction and thus, which cannot be labeled with the above-noted thiol labeling reagents. In general, this is accomplished by exposing the intact protein to an oxidizing agent under conditions to oxidize cysteine to cysteic acid or by alkylating the sulfhydryl groups of the cysteine residues using routine procedures. In summary, modification of cysteine and/or lysine residues prior to conducting the beta-elimination is recommended to avoid potential side reactions which could interfere with the determination of phosphoserine and phosphothreonine residues in the intact protein. According to another aspect of the invention, phosphorylated amino acids are detected by oxidizing the double bond of the derivatized dehydroamino acid with KMnO₄ to yield a cis diol which then can be oxidized with periodate to a highly reactive aldehyde. The resultant aldehyde is allowed to react with biotin-hydrazide to yield a biotin-labeled amino acid in the intact protein at the location of the dehydroamino acid. The biotin moiety is detected as described above. According to yet another aspect of the invention, kits for detecting the presence of a phosphoserine and/or phosphothreonine in an intact protein are provided. The kits contain an amount of nucleophile sufficient for effecting the above-described beta-elimination, addition, and detection reactions and instructions for using the nucleophile in accordance with the methods of the invention. Optionally, the kits further include a negative (e.g., bovine serum albumin, "BSA") or positive (e.g., casein) protein control for evaluating the sensitivity and accuracy of the beta-elimination, nucleophile and/or detection reagents for detecting the presence of phosphoserine and/or phosphothreonine in an unknown protein sample.

Examples

Materials Abbreviations Hydrogen peroxide, H₂O₂; sodium dodecyl sulfate, SDS; strontium chloride, SrCl₂;

1,3-ρropanedithiol, DTP; dithiothreitol, DTT; dimethyl sulfoxide, DMSO; ethylenediaminetetraacetic acid, EDTA; 3-[cyclohexylamino]-l-propane-sulfonic acid. CAPS; N-[2-hydroxyethyl]piperazine-N'-[2-ethanesulfonic acid], HEPES; sodium chloride, NaCl; N-ethylmaleimide, NEM; sodium hydroxide, NaOH; potassium permanganate, KMnO₄; sodium dodecyl sulfate - polyacrylamide gel electrophoresis, SDS-PAGE; adenosine triphosphate. ATP; 1 ,3-diaminopropane, DAP. Chemicals

An enhanced chemiluminescence kit (ECL) was purchased from Amersham (Arlington Heights, IL). Casein kinase, protein kinase A, and pp60^src were purchased from Upstate Biotechnology, Inc. (Lake Placid, NY). Radiochemicals were purchased from Amersham (Arlington Heights. IL). All other chemicals were purchased from Sigma Chemical Company (St. Louis, MO).

Introduction

Reactions for Detection of Serine or Threonine Phosphate in Intact Proteins A. Removal of cysteine and cystine residues

Participation of cysteine and cystines in the beta-elimination and/or addition reactions of the invention can result in unwanted side reactions which reduce the specificity and/or sensitivity of the methods for detecting phosphoserine and/or phosphothreonine in an intact protein. Accordingly, cysteines and cystines are removed by oxidation (part 1 , below) or reduction/alkylation (part 2, below) to a form (referred to herein as a "cysteine product") which cannot undergo a beta-elimination reaction and/or which cannot be subject to attack in an addition reaction.

(11 Oxidation of cysteines and cystines The protein to be subjected to the detection procedure can be present in an unpurified extract, isolated by means of immunoprecipitation, chromatography or other standard methods well known in the art, or prepared by standard in vitro methods well known in the art (Molecular Cloning: a Laborator^y Manual. Sambrook et al., Cold Spring Harbor Laboratory Press, 1989). Performic acid is the preferred oxidation reagent to efficiently oxidize cysteine or cystine residues in intact proteins. Performic acid is prepared by combining 1 part 30% H₂O₂ and 9 parts 70% formic acid and allowing the mixture to incubate for about one hour at room temperature.

Oxidation of cysteines and cystines is accomplished as follows. One part performic acid is added to 20 parts protein solution (about 1 mg/ml) and incubated for two hours on ice. One hundred parts cold (4°C) 10 mM ammonium acetate solution and 0.2 parts beta-mercaptoethanol are added to the reaction mixture; the mixture is lyophilized to dryness using, for example, a Speed- Vac rotoevaporator (Savant Instruments, Inc., Farmingdale, NY). The modified protein is dissolved in 100 microliters 10 mM ammonium acetate and relyophilized to completely remove formic acid. Failure to completely remove formic acid may result in a pH which is less than the optimum pH for effecting the beta-elimination reaction and/or the addition reaction, thereby reducing the efficiencies of these reactions.

To reduce the likelihood of performic acid-induced protein degradation, the oxidation reaction time and performic acid concentration are optimized by assessing the degree of cysteine/cystine oxidation and the amount of sample degradation as a function of reaction time, temperature and performic acid concentration. Assessment of cysteine/cystine oxidation and protein degradation is made by comparing the banding patterns of performic acid treated and untreated (control) protein samples on SDS-PAGE. BSA, which contains cysteine residues but which does not contain phosphoserine or phosphothreonine residues, can be used as a control protein to assess the extent of cysteine oxidation to cysteic acid. Thus, incomplete oxidation of BSA will result in a detectable band on a Western Blot for a BSA sample that is subjected to the beta-elimination, addition and detection reactions of the invention. If the protein is radioactively labeled, an autoradiogram of the SDS-PAGE gel also will reveal any degradation of the protein which may have been induced by the oxidation reaction. Alternatively, subjecting the SDS-PAGE gel to Western blot analysis can be used to assess the extent of protein degradation. (2) Reduction/Alkvlation of cvsteines and cvstines

Elimination of cysteines and cystines can be accomplished by reduction and alkylation reactions using NEM and DTT. This alternative procedure degrades the protein sample to a lesser extent than the performic acid procedure. However, this reaction should be optimized to ensure complete removal of cysteines and cystines in the intact protein. Conditions for performing the reduction/alkylation of intact proteins are well known to those of ordinary skill in the art. An exemplary protocol for the reduction/alkylation removal of cysteine/cystine residues is presented below.

The protein of interest was isolated by a standard immunoprecipitation using Protein A sepharose beads to which was coupled an antibody that was specifically reactive to the protein of interest. The protein was eluted from the sepharose beads by boiling 3 minutes in 1% SDS. 25 mM HEPES pH 8.0. Thereafter, the eluted protein was incubated in 1% SDS, 2 mM DTT, 0.5 mM EDTA in 25 mM HEPES pH 8.0 for 30 minutes at 45 °C to denature the protein and reduce the cystine residues. 10 mM N-ethyl maleimide was added to the denatured protein and this mixture was incubated for an additional 60 minutes to alkylate the cysteine residues. Dephosphorylation was performed using DTP/NaOH/SrCl₂ procedure described herein.

B. Beta-elimination reaction Alkaline dephosphorylation of phosphoserine and phosphothreonine residues is accomplished by beta-elimination of the phosphate moieties from these amino acid residues. A 10% stock solution of DTP in DMSO is freshly prepared. A protein sample (e.g., the lyophilized oxidized protein of Part A above) is dissolved in a volume of 0.1% SDS and the following reactants (concentration is the final concentration) are added to the protein solution: 0.1N NaOH, 0. IN SrCl₂ and 1 % DTP. The mixture is incubated at 44 °C for about 30 minutes to yield an intact protein containing the dehydroamino acids corresponding to the former phosphoserine and phosphothreonine residues.

C. Addition of a bifunctional nucleophile to a dehydroamino acid

The resulting beta-dehydroalanine residues in the intact protein are subjected to an addition reaction (e.g., a nucleophilic addition reaction) to introduce a detectable tag into the former phosphoserine or phosphothreonine residues. The first step of this "two-step" reaction is performed by adding an equal volume of 3% DTP in 1.0M CAPS buffer, pH 1 1.0 to the above- described protein solution reaction mixture and incubating this reaction mixture for about 3 hours at 44 °C. Thereafter, the reaction mixture is neutralized to pH 7-8 with glacial acetic acid and EDTA is added to a concentration of 0.2M to chelate the SrCl₂. The reaction mixture is extracted three times with an equal volume of diethyl ether, saving and pooling the lower aqueous layer after each extraction. Complete removal of the nucleophile enhances the efficiency of subsequent labeling reactions.

P. Reaction of the bifunctional nucleophile with a detection reagent

An equal volume of 0.2M HEPES pH 8.0 is added to an aliquot of the reaction mixture (from Part C). Biotin maleimide is added to this mixture to a final concentration of lOmM and the biotin maleimide-containing mixture is incubated at room temperature for 30 to 60 minutes. The reaction is stopped by adding beta-mercaptoethanol to give a final concentration of 100 mM. The biotin-labeled proteins are subjected to SDS-PAGE and Western blotting. Any of a number of different detection methods can be used to visualize the biotin-labeled protein. In a particularly preferred embodiment, a commercially available chemiluminescence kit (ECL, Amersham) that employs streptavidin conjugated horseradish peroxidase as a luminol substrate converting enzyme is used to visualize the biotin-labeled protein.

Example 1 : Nucleophilic Attack with 1.3-propanedithiol

The above-described reduction/alkylation procedure was used to reduce/alkylate cysteine and/or cystine residues in the protein sample to a form (S-ethyl cysteine) which does not interfere with the phosphoserine/phosphothreonine detection method disclosed herein. Two types of protein samples, casein and BSA, were subjected to the reduction/alkylation procedure prior to performing the beta-elimination and nucleophilic addition reactions.

Casein contains between five and ten phosphate groups per polypeptide chain. Reportedly, these phosphate groups are present as phosphoserine residues or phosphothreonine residues and are not present as phosphotyrosine residues. One microgram of casein (or other phosphoserine or phosphothreonine-containing protein) in 100 ul 70% formic acid was cooled on ice. Protein controls include serine- (or threonine-) containing proteins which are either phosphorylated using an appropriate kinase and ATP using manufacturer's recommended procedure, mock phosphorylated using the appropriate kinase without ATP, or not treated. BSA was used as a negative control. BSA does not contain phosphoserine residues but does contain seventeen cystine residues. Accordingly, BSA also served as a control for assessing completion of the performic acid oxidation reaction.

Five microliters of performic acid were added to the protein-containing reaction mixture and the mixture was incubated for two hours on ice. The oxidation reaction was stopped by the addition of 1.0 ml cold lOmM ammonium acetate solution and 1.0 ul beta-mercaptoethanol. Thereafter, the reaction mixture was lyophilized for 4-5 hours using a Speed- Vac (Savant Instruments), the protein was resuspended in 100 ul lOmM ammonium acetate and relyophilized. The lyophilized sample was dissolved in 100 ul 0.1% SDS; 20 ul 1.0N NaOH, 20 ul 1.ON SrCl₂, 20 ul 10% DTP in DMSO and 40 ul water were added to initiate the beta-elimination (dephosphorylation) reaction and the reaction was allowed to proceed for 30 minutes at 44oC. The nucleophilic addition of DTP to the dehydroamino acids (i.e., dephosphorylated serine and threonine residues), was initiated by adding 200 ul of 3% DTP in 1.0M CAPS, pH 11.0 to the reaction mixture and incubating for 3 hours at 44 °C. Glacial acetic acid was added to neutralize the pH of the mixture, followed by 500 ul of 0.4M EDTA to chelate the Sr^2* ions. The mixture was extracted thrice with 1.0 ml of diethyl ether.

To introduce a detectable tag to the DTP-modified residues, 50 ul of 0.2M HEPES, pH 8.0 were added to 50 ul of the extracted reaction mixture. Eleven microliters of lOOmM biotin-maleimide were added and the reaction mixture was incubated for one hour at room temperature. One ul of beta-mercaptoethanol was added to stop the reaction. Fifty-five microliters of 3X Laemmli sample buffer were added to the reaction mixture to form the SDS- PAGE sample, which then was processed and resolved on a 10% standard Laemmli SDS-PAGE gel using standard procedures. The gel was transferred to nitrocellulose (S&S, Keene, N.H.) using an electroblotting device (Bio-Rad, Richmond, CA) according to standard practice. The Western blot was processed according to the manufacturer's instructions for chemiluminescent detection of proteins using an ECL kit (Amersham, Arlington Heights, IL). 250 nanograms of casein were easily detectable with a ten second exposure of the Western blot to X-ray film. BSA that had been treated in an identical manner to that described above for casein was not detected using the ECL kit, thus illustrating the specificity of the labeling reaction for phosphorylated serines and threonines and further illustrate that the DTT/NEM alkylation conditions were sufficient to completely reduce/alkylate cysteine/cystine residues to a form (S-ethyl cysteine) which did not interfere with the phosphoserine/phosphothreonine detection method disclosed herein. These results indicate that the above-described method could be used to detect at least about 30 pmoles of phosphoserine and/or phosphothreonine in an intact protein.

Example 2: Sensitivity and Specificity of the Labeling and Detection Procedure A titration experiment is described herein to determine the sensitivity of the above- described methods for detecting phosphorylated serines and threonines in intact proteins. Casein is used as a positive control. Other serine-and/or threonine-containing proteins can be phosphorylated on serines and threonines using the appropriate protein kinase according to the manufacturer's recommendations. For example, ovalbumin is phosphorylated on serines and threonines using protein kinase A according to the manufacturer's instructions. pp60^src kinase is autophosphorylated on tyrosine residues according to the manufacturer's instructions.

Each of the phosphorylated proteins is mixed with its corresponding unphosphorylated protein to constitute 100%, 50%, 10%, 5%, 1%, 0.5%, 0.1%, 0.05%, 0.01% or 0% of 1.0 ug total protein for use in the beta-elimination (dephosphorylation) and addition/detection reactions. Fifty micrograms of phosphorylated pp60^src (a negative control which contains phosphorylated tyrosine residues) is mixed with unphosphorylated bovine serum albumin added as a carrier protein. These proteins then are subjected to the detection described in Example 1.

The results of SDS-PAGE and Western blotting demonstrate that phosphorylated casein or ovalbumin comprising 0.01% of the total protein is detectable by this methodology. Therefore, since 50 nanograms of protein are used for SDS-PAGE/Western blot analysis, as little as 5 picograms of phosphorylated protein is detectable. The method is not limited to a particular protein, as both phosphorylated casein and phosphorylated ovalbumin are detected. The method is selective for serine and/or threonine phosphorylation since 2.5 nanograms pp60^src autophosphorylated on tyrosine residues, comprising 0.5% of the total protein, is not detected using the method of Example 1.

Example 3: Nucleophilic Attack with Dithiothreitol

The nucleophile dithiothreitol (DTT) is substituted for DTP in the method of Example 1. Because DTT is more soluble in aqueous solution than DTP, higher effective concentrations of DTT can be attained in the reaction mixture. In the beta-elimination (dephosphorylation) and addition reactions, DTT is added to the reaction as a 1-10% solution in water, preferably as a 3% solution in water. Phosphorylated casein is detectable with similar sensitivity as when labeled using DTP.

Example 4: Nucleophilic Attack with 1.3-Diaminopropane Primary amine-containing nucleophiles also can be used in accordance with the methods of the invention to detect phosphoserine and/or phosphothreonine residues in an intact protein. However, to prevent unwanted side reactions between the detection reagent (which is reactive with one of the primary amine groups of the nucleophile) and the N-terminal or lysine side chain amine groups of the protein sample, the protein sample should be treated to block free amine groups (e.g., by alkylation) prior to performing the method of Example 1. In general, it is recommended that the protein samples be subjected to conditions to block lysine, cysteine, and cystine amino acid side chains to prevent their respective amine and sulfhydryl groups from participating in the beta-elimination, addition and labeling reactions of the invention. Typically, this is accomplished by alkylating the cysteine/cystine sulfhydryl groups and/or the lysine primary amine according to standard procedures. In summary, modification of cysteine/cystine and/or lysine residues prior to conducting the beta-elimination is recommended to preclude potential side reactions which could interfere with the determination of phosphoserine and phosphothreonine residues in the intact protein.

1,3-diaminopropane (DAP) is substituted for DTP in the method of Example 1. DAP is added to the reaction mixture as a 1-10% solution in DMSO, preferably a 3% solution in DMSO. The results of the labeling procedure indicate that DAP can be used as a bifunctional nucleophile for the detection of phosphorylated serine and threonine residues in an intact protein.

Example 5: Nucleophilic Attack Following Oxidation of Beta-elimination Products An alternative addition reaction for introducing a detectable tag to a dehydroamino acid is described herein. After beta-elimination under alkaline conditions as described in Example 1 , a solution of KMnO₄ of sufficient molar strength to oxidize a beta-dehydroalanine double bond to a cis-diol configuration is added to the beta-elimination reaction mixture. Periodic acid is added in a sufficient molar amount to oxidize the cis-diol to an aldehyde compound. The resulting aldehyde is a reactive species that is subject to nucleophilic attack by, for example, a monofunctional nucleophile that contains a detectable tag. In a particularly preferred embodiment, biotin-hydrazide is added to the periodic acid- treated protein in a molar amount sufficient to stoichiometrically label the reactive aldehyde species in a Wolff-Kishner reaction in which the biotin moiety is covalently attached to a carbon atom of the dehydroamino acid in the intact protein. A standard Western blot/ECL detection protocol is performed to detect the biotin moiety in the intact protein at a level of sensitivity comparable to that observed in Example 2. Example 6: Radioactive Labeling with ³²P-Phosphoric Acid An in vitro method of radiolabeling serines and threonines phosphorylated in vivo is described in this example. One of the difficulties of in vivo ³ P radiolabeling of proteins to determine the phosphorylation state is the relatively large amounts of radioactive materials required. Since the reaction is highly inefficient in vivo, an large excess of radioactive material is used, resulting in contamination of experimental apparatus and creation of large quantities of hazardous waste. The method described in Example 1 is used to in vitro label the phosphorylated serine and threonine residues.

Following the (optional) purification of a protein of interest from an in vivo or in vitro source, cysteines/cystines are removed (e.g., oxidized or subjected to reduction/alkylation) and phosphorylated serines and threonines converted to dehydroamino acids by alkaline beta-elimination. The resulting double bond is attacked by the addition of a sufficient molar quantity of ³²P-phosphoric acid to the beta-elimination reaction mixture. The addition of the radioactive phosphorus moiety labels the site of in vivo or in vitro phosphorylation in the intact protein. The presence of radiolabel is determined by standard methods well known in the art, such as SDS-PAGE followed by autoradiography or phosphorimager detection, or scintillation counting.

Example 7: Radioactive Labeling with ³⁵S-Sodium Sulfite An in vitro method of radiolabeling serines and threonines phosphorylated in vivo is described in this example. One of the difficulties of in vivo ³⁵S radiolabeling of proteins is that it is difficult to determine the phosphorylation state of the protein. Since ³⁵S radiolabeling requires uptake of ³⁵S labeled methionine and/or cysteine, it is highly inefficient in vivo and large excess of radioactive material must be used, resulting in contamination of experimental apparatus and creation of large quantities of hazardous waste. Furthermore, the reaction is not specific for phosphorylated proteins. Using the method described in Example 1, one can effect in vitro labeling of phosphorylated serine and threonine residues.

Following the (optional) purification of a protein of interest from an in vivo or in vitro source, cysteines/cystines are removed (e.g., oxidized or subjected to reduction/alkylation) and phosphorylated serines and threonines converted to dehydroamino acids by alkaline beta-elimination. The resulting double bond is attacked by the addition of a sufficient molar quantity of ³⁵S-sodium sulfite to the reaction. The addition of the radioactive sulfite moiety labels the site of in vivo or in vitro phosphorylation in the intact protein. The presence of radiolabel is determined by standard methods well known in the art, such as SDS-PAGE followed by autoradiography or phosphorimager detection, or scintillation counting.

Each reference identified above is incorporated herein in its entirety by reference. It should be understood that the preceding is merely a detailed description of certain preferred embodiments. It therefore should be apparent to those skilled in the art that various modifications and equivalents can be made without departing from the spirit or scope of the invention.

Claims

1. A method for detecting a phosphorylated amino acid in an intact protein, wherein the phosphorylated amino acid is a phosphoserine or a phosphothreonine, the method comprising:

(1) subjecting the protein to conditions for effecting a beta-elimination reaction in the intact protein to convert the phosphorylated amino acid to a dehydroamino acid, the dehydroamino acid including a double bond containing carbon atoms that are susceptible to attack by an addition reactant in an addition reaction;

(2) contacting the protein of step (1) with a reactant under conditions for adding the reactant to a carbon atom of the double bond of the dehydroamino acid to form an amino acid addition product in the intact protein; and

(3) detecting the amino acid addition product in the intact protein, wherein the presence of the amino acid addition product indicates the presence of the phosphorylated amino acid in the protein.

2. The method of claim 1, wherein the addition reaction is a nucleophilic addition reaction, the reactant is a nucleophile and the amino acid addition product is a nucleophilic amino acid addition product.

3. The method of claim 1, wherein the addition reaction includes oxidation of the dehydroamino acid to form a cis diol-containing amino acid.

4. The method of claim 1, wherein the phosphorylated amino acid is phosphoserine.

5. The method of claim 1, wherein subjecting the protein to conditions for effecting a beta-elimination reaction in the intact protein comprises placing the intact protein in a dilute basic aqueous solution for between 10 minutes and 12 hours.

6. The method of claim 5, wherein the intact protein is placed in the basic aqueous solution for between 0.5 and 6 hours.

7. The method of claim 6, wherein the basic aqueous solution is a sodium hydroxide solution having a sodium hydroxide concentration ranging between 0.05 and 0.5 normal.

8. The method of claim 7, wherein the sodium hydroxide solution is 0.1 N NaOH.

9. The method of claim 8, wherein the intact protein is placed in the basic aqueous solution for between 0.5 and 1.0 hours.

10. The method of claim 5, wherein the basic aqueous solution further comprises a catalytic amount of a beta-elimination catalyst.

11. The method of claim 10, wherein the beta-elimination catalyst is a group II metal ion.

12. The method of claim 11, wherein the beta-elimination catalyst is selected from the group consisting of SrCl₂ and Ba(OH)₂ and wherein the catalytic amount is between 0.05 and 0.20 molar.

13. The method of claim 1, wherein the intact protein contains at least one cysteine, further comprising the step of treating the intact protein under conditions to convert the cysteine to a cysteine product prior to subjecting the protein to conditions for effecting a beta-elimination reaction, wherein the cysteine product cannot undergo a beta-elimination reaction.

14. The method of claim 13, wherein treating the protein to convert the cysteine to a cysteine product comprises exposing the intact protein to an oxidizing reagent under conditions to oxidize cysteine to cysteic acid.

15. The method of claim 13, wherein treating the protein to convert the cysteine to a cysteine product comprising alkylating the sulfhydryl group of the cysteine.

16. The method of claim 2, wherein the nucleophile is a monofunctional nucleophile having the structure,

Rl

I X - [C]_n - D R2

wherein X is a first functional group selected from the group consisting of a thiol group and an amine group; D is a detectable tag for determining the presence of the phosphorylated amino acid in the intact protein; Rl is H or OH; R2 is H or OH and n = 0 to 10, inclusive; and wherein Rl and R2 can be the same or different from one another.

17. The method of claim 16, wherein the detectable tag is selected from the group consisting of a radioactive tag, a chromophore, a fluorophore, a luminescent tag. a biotin, a hapten and an enzyme.

18. The method of claim 16, wherein the monofunctional nucleophile is selected from the group consisting of H₂N-(CH₂)_n-Biotin and H₂S-(CH₂)_n-Biotin, wherein n = 4 to 8. inclusive.

19. The method of claim 2, wherein the nucleophile is a bifunctional nucleophile containing a first functional group for reacting with the double bond and a second functional group for reacting with a detection reagent, the bifunctional nucleophile having the formula. Rl

X - [C]_n - Y R2

wherein X is a first functional reactive group selected from the group consisting of a thiol group and an amine group; Y is a second functional reactive group selected from the group consisting of a thiol group and an amine group; Rl is H or OH; R2 is H or OH and n = 0 to 10, inclusive; wherein Rl and R2 can be the same or different from one another and wherein X and Y can be the same or different from one another.

20. The method of claim 19, wherein the nucleophile is selected from the group consisting of 1 ,2-diaminoethane, 1,3-diaminopropane, diaminopropanol, 1,3-dithioethane, 1,3- propane dithiol, dithiothreitol and dithioerythritol.

21. The method of claim 19, wherein the step of detecting the nucleophilic addition amino acid product comprises exposing the intact protein to a detection reagent under conditions for the detection reagent to covalently bond to the second functional group of the nucleophile.

22. The method of claim 21, wherein the detection reagent contains a detectable tag selected from the group consisting of a radioactive tag, a chromophore, a fluorophore, a luminescent tag, a biotin, a hapten and an enzyme.

23. The method of claim 21, wherein the detection reagent is selected from the group consisting of biotin-maleimide, iodoacetylbiotin and N-hydroxysuccinimido-biotin.

24. A kit for determining the presence of a phosphorylated amino acid in an intact protein, wherein the phosphorylated amino acid is a phosphoserine or a phosphothreonine, the kit comprising: a nucleophile having the formula,

Rl

X - [C]_n - D

I

R2

wherein X is a first functional group selected from the group consisting of a thiol group and an amine group; D is a detectable tag for determining the presence of the phosphorylated amino acid in the intact protein; Rl is H or OH; R2 is H or OH and n = 0 to 10, inclusive; and wherein Rl and R2 can be the same or different from one another; and instructions for (a) subjecting the intact protein to conditions for effecting a beta-elimination reaction in the intact protein to convert the phosphorylated amino acid to a dehydroamino acid, the dehydroamino acid including a double bond containing carbon atoms that are susceptible to attack by the nucleophile in a nucleophilic addition reaction, (b) contacting the protein of step (a) with the nucleophile under conditions for adding the nucleophile to a carbon atom of the double bond of the dehydroamino acid to form a nucleophilic amino acid addition product in the intact protein; and (c) detecting the nucleophilic amino acid addition product in the intact protein, wherein the presence of the nucleophilic amino acid addition product indicates the presence of the phosphorylated amino acid in the protein.

25. A kit for determining the presence of a phosphorylated amino acid in an intact protein, wherein the phosphorylated amino acid is a phosphoserine or a phosphothreonine, the kit comprising: a nucleophilic reactant having the formula,

Rl

X - [C]„ - Y

I

R2

wherein X is a first functional group selected from the group consisting of a thiol group and an amine group; Y is a second functional group selected from the group consisting of a thiol group and an amine group; Rl is H or OH; R2 is H or OH and n = 0 to 10, inclusive; wherein Rl and R2 can be the same or different from one another and wherein X and Y can be the same or different from one another; a detection reagent that is reactive with the second functional group, wherein the detection reagent contains a detectable tag selected from the group consisting of a radioactive tag, a chromophore, a fluorophore, a luminescent tag, a biotin, a hapten and an enzyme; and instructions for (a) subjecting the intact protein to conditions for effecting a beta-elimination reaction in the intact protein to convert the phosphorylated amino acid to a dehydroamino acid, the dehydroamino acid including a double bond containing carbon atoms that are susceptible to attack by the nucleophile in a nucleophilic addition reaction, (b) contacting the protein of step (a) with the nucleophile under conditions for adding the nucleophile to a carbon atom of the double bond of the dehydroamino acid to form a nucleophilic amino acid addition product in the intact protein; (c) contacting the protein of step (b) with the detection reagent under conditions to form a covalent bond between the second functional group of the nucleophile and the detection reagent; and (d) detecting the nucleophilic amino acid addition product in the intact protein, wherein the presence of the nucleophilic amino acid addition product indicates the presence of the phosphorylated amino acid in the protein.

26. The kits of claims 24 or 25, further including at least one of a negative protein control and a positive protein control, wherein the negative protein control and the positive protein control are for evaluating the sensitivity and accuracy of the kit for detecting the presence of the phosphorylated amino acid in the intact protein.