CA2507117A1 - Complete chemical and enzymatic treatment of phosphorylated and glycosylated proteins on protein chip arrays - Google Patents

Abstract

A simple and quick protocol for chemical treatment, enzymatic digestion and subsequent identification of proteins on protein chip arrays is disclosed. The chemical treatment comprises denaturation, reduction and alkylation, while enzymatic digestion encompasses deglycosylation, dephosphorylation, and digestion by various proteases. All reactions are carried out on-chip in a sequential fashion. Subsequent peptide mass fingerprinting allows the identification of specific peptides. The method of the present invention can be applied to the analysis of biological samples such as urine to identify specific biomarkers. The methods of the present invention allow complete on-chip treatment, which can be used for rapid protein identification and structural characterization.

Description

TITLE OF THE INVENTION
[0001] COMPLETE CHEMICAL AND ENZYMATIC TREATMENT OF
PHOSPHORYLATED AND GLYCOSYLATED PROTEINS ON PROTEIN CHIP
ARRAYS
FIELD OF THE INVENTION

[0002] The present invention relates to proteomics. More particularly, the present invention relates to protein chip arrays. More specifically, the present invention is concerned with methods of chemical and enzymatic treatment of proteins on protein chips. More particularly, the invention relates to chemical and enzymatic treatment of post translationally modified proteins on proteins chip arrays.
BACKGROUND OF THE INVENTION

[0003] Many advances in proteomics have been driven by the development of mass spectrometric-based technologies and tools '. Although mass spectrometry (MS) was invented in the early 1900s for the detection of small molecules, a quantum leap was achieved in the late 1980s when Fenn and Tanaka showed independently that large biomolecules (proteins, DNA, etc) can be detected and quantitated accurately by MS. Fenn's technique called Electro Spray Ionization (ESI) nebulizes a protonated liquid into a fine spray using a high voltage prior to MS detection (J. B. Fenn, M. Mann, C. K. Meng, S.F. Wong, C.M.
Whitehouse, Science 246, 6, 64 (1989)). Tanaka's method called Matrix Assisted Laser Desorption Ionization (MALDI) utilizes a high energy absorbing molecule to desorb intact proteins on a solid inert surface (K. H. Tanaka, H. Wake, Y.
Ido, S.
Akita, Y. Yoshida and I. Yoshida, Rapid Commun. Mass Spectrom. 8, 2, (1988)).
A
flavour of this latter technique, called Surface Enhanced Laser Desorption Ionization (SELDI) permits the immobilization of molecules on different active surfaces. SELDI is described in U.S. Pat. No. 5,719,060 ("Method and Apparatus for Desorption and Ionization of Analytes," Hutchens and Yip, Feb. 17, 1998);
U.S.
Pat. No. 6,225,047 ("Use of Retentate Chromatography to Generate Difference Maps," Hutchens and Yip, May 1, 2001); and in Weinberger et al., "Time-of-flight mass spectrometry," in Encyclopedia of Analytical Chemistry, R. A. Meyers, ed., pp 11915-11918 John Wiley & Sons Chichesher, 2000.

[0004] ~ A number of reports have appeared over the past several years regarding proteomic profiling with SELDI-TOF technology, in combination with artificial intelligence. Reported sensitivities and specificities with the technique for ovarian, prostate, and breast cancer diagnosis are better than those obtained with current serologic cancer biomarkers. Also, the technique is reported to detect early as well as late stage disease with similar efficiency, thus offering a potentially powerful new cancer screening tool.

[0005] The combination of techniques such as polyacrylamide gel electrophoresis (PAGE)2~3, reverse phase high performance liquid chromatography (RP-HPLC)4~, affinity capture'~8 and protein chips 9 with mass spectrometry (MS) has provided a series of important tools for the investigation of numerous facets of proteomics. The identification and characterization of the chemical features of proteins are essential prerequisites for understanding the dynamics and connectivity of their interactions as well as the diversity of their biological functions in living organisms. As a common method, peptide mass fingerprinting (PMF) identifies proteins by comparing the peptide mass fingerprint obtained from mass spectrometry analysis of enzymatic (or chemical) digestions to mass profiles generated by in-silico digestion of proteins'°. This approach requires relatively purified target protein and is often used with protein fractionation techniques. Prior to enzymatic digestion, proteins are denatured, reduced and alkylated.
Digestion is generally performed overnight to ensure complete cleavage. Structural characterization of proteins becomes all the more difficult if one considers that the vast majority of proteins contain disulfide bridges, phosphorylation, glycosylation sites or a combination of the above.

[0006] Thus, to study biological systems at the protein level, efforts have been directed to the improvements in instrumentation and the development of novel technologies.
[OOOTj Protein chip array technology is based on two powerful techniques: chromatography and mass spectroscopy. It consists of selective protein extraction, retention and enrichment of proteins on chromatographic chip surfaces and their subsequent analysis by mass spectroscopy. The protein chip array surfaces function as a solid phase extraction media that support isolation and clean up of analytes prior to mass spectroscopic investigation.
[0008] By comparing samples between control and experimental groups or between healthy and diseased individuals, in one use of the technology, protein chip array profiling allows the rapid creation of phenotypic fingerprints and the identification of biomarkers of particular metabolic or disease states.
[0009] Thus, together with the growth of this technology comes the need for protein chemistry techniques that are applicable to protein chips.
Three groups have reported a single on-chip reaction prior to MS analysis.
Pentafluoropropionic acid and trifluoroacetic acid (TFA) were used to perform limited acid hydrolysis of proteins using a vapor-phase hydrolysis procedure".
The method was proposed to generate peptide ladders indicating primary sequences.
However, side reactions, such as oxidation of methionine residues and deamidation of asparagine or glutamine, were systematically observed. A second group reported a procedure for the identification of parvalbumin alpha (PVA) using on-chip enzymatic proteolysis'2. Four peptides were identified after a 2-hour digestion and nine peptides were identified after 18 hours. PVA is an 11.85 kilodalton (kDa) linear N-terminus acetylated polypeptide, which is not representative of most of the proteins in existing proteomes as it lacks complex modifications such as disulfide bridges, phosphorylated or glycosylated moieties.
Finally, an on-chip tryptic digestion method has been applied to recombinant prolactin-inducible protein (PIP). This purified 16.57 kDa protein has two disulfide bridges and one N-glycosylation site'3.
[0010] In all the above examples, all chemical and enzymatic steps were carried out in solution. Relatively simple proteins were tested, and in all cases, a single on-chip step of treatment was performed. On-chip protein denaturation, reduction, alkylation, deglycosylation and dephosphorylation using protein chips have not been previously reported. In addition, the reports have generally been based on rather simple proteins.
[0011] Thus, there remains a need for improved methods allowing structural characterization of proteins.
[0012] There further remains a need for methods of protein identification, which reduce sample loss, enable rapid and sensitive detection and identification of proteins with minimal sample manipulation.
[0013] There also remains a need for simple methods allowing complete on-chip chemistry (including enzymatic treatment) and characterization of proteins.
[0014] The present invention seeks to meet these needs and other needs.

SUMMARY OF THE INVENTION
[0015] Although solid phase chemistry (e.g. Edman degradation) has been routinely performed on solid support for years, it is difficult to imagine complex biochemical reactions on solid surfaces partly because the enzymes must 5 retain their activities throughout the process, and also because of limited bioavailability. For example, in enzymatic digestion, the reactants seem unlikely to interact effectively to cleave highly complex proteins. It is analogous to putting liquid in sand in a first step which is followed by an addition of a different liquid and expecting proper mixing. When a protein is denatured, it is in its most relaxed state and more prone to interact with other species. A solid small surface is not a predictable environment for that interaction. Reactions in solution have been carried out for centuries and are fully understood (access of water, configuration of the protein in solution, etc.). However, biochemical reactions on solid surfaces have been very poorly exploited because of their complexity and also because they seem not likely to occur. For example, the environment of the protein on a chip is very different from that in solution. The water environment is but one critical difference between the proteins on a chip as compared to that in solution. The relatively dry state of a protein on a chip suggests that enzymatic digestion is likely not to occur on a chip.
[0016] The rapid growth of proteomics and more particularly protein array technology urged the development of simpler, more sensitive methodologies.
Microfabricated devices are becoming increasingly popular for the analysis of biomolecules (deoxynucleic acid (DNA), deoxyribonucleic acid (RNA), proteins, peptides) for a number of reasons. These devices come in two varieties, the array format and microfluidic devices. They offer the potential to automate biological sample processing (reduction, alkylation, chemical and enzymatic digestion, desalting, etc.) reduce costs and increase throughput. In addition, they are designed with minimal quantities of sample in mind. When only tiny amounts of sample are available, macroscale techniques become ineffective due to sample losses.
[0017] Researchers around the world have attached great importance to protein chip technology because it could theoretically simultaneously analyze information of many biomolecules in one reaction. However, the development and applications of this technology is still limited by its complexity.
[0018] The present invention, demonstrates that surprisingly several complex enzymatic and chemical reactions can indeed be performed directly on protein chip surfaces in a sequential fashion.
[0019] Thus, the present invention relates to the use of protein chip methods for performing various enzymatic, other biological and chemical reactions.
This approach employs chips with different surface physicochemical properties enabling the selective capture and retention of proteins or peptides from biological samples.
[0020] In one aspect, the present invention relates to protein chemistry procedures that can be performed directly on-chip using small volumes (in the p1 range) of the biological sample of interest, reagents and washing solutions, as well as relatively short reaction time for both chemical and enzymatic treatments prior to MS analysis.
[0021] More specifically, the present invention is concerned with a quick, simple and sensitive method allowing two or more, and up to all chemical reactions to be performed on-chip as well as subsequent enzymatic deglycosylation, dephosphorylation and proteolysis in a sequential fashion.
The methods of the present invention provide a rapid and simple alternative to in-gel or in-solution methods.

[0022) Thus, the present invention is concerned with novel experimental methods to analyze peptidelproteins by protein chip array technology. These methods enable the rapid deglycosylation, dephosphorylation, digestion and identification of low amounts (in the picomolar range) of complex proteins. Because all steps may be performed directly on chip, the method of the present invention is easily amenable to automation. Consequently, the method of the present invention may be developed for low-throughput, high-throughput, or ultra-high throughput analysis formats.
[0023) In one aspect, the method of the present invention generally comprises a number of the following steps:
a) conditioning of the spots of the protein chip array with conditioning buffer;
b) loading of the biological sample on the protein chip; after binding, excess sample is removed and each spot is washed with appropriate buffer;
c) denaturing the protein sample;
d) reducing the protein sample;
e) alkylating the protein sample;
f) deglycosylating and/or dephosphorylating the sample;
g) chemical or enzymatic digestion (hydrolysis of peptide bonds) for PMF;
h) performing MS analysis (drying of the sample, matrix (energy absorbing molecule-EAM) addition and data collection); and i) database mining and identification of proteins.
[0024) In accordance with the present invention only some steps of the above general method may be performed depending on the type of information that is sought and the type of protein sample that is used. For example, if information is only sought on the phosphorylation status of the protein, then, the deglycosylation and chemicallenzymatic digestion steps may not be performed.
Alternatively, if only the glycosylation level of a protein needs to be studied then, the dephosphorylation and chemical/enzymatic digestion steps would not be performed. On the other hand if one is working with relatively simple proteins or peptides, then the dephosphorylation and deglycosylation step may not be required. Thus, depending on the particular experimental requirements, a person skilled in the art would choose which of the above steps are to be performed and adapt the method accordingly.
[0025] Thus, in one embodiment, the method of the present invention comprises a conditioning step; a biological sample loading step, a denaturing step, a reducing step, an alkylating step, a deglycosylation step, an enzymatic digestion step (PMF) and an MS analysis step.
[0026] In another embodiment, the method of the present invention comprises a conditioning step; a biological sample loading step, a denaturing step, a reducing step, an alkylating step, a dephosphorylation step, an enzymatic digestion step (PMF) and an MS analysis step.
[0027] In a further embodiment, the method of the present invention comprises ~a conditioning step, a biological sample loading step, a denaturing step, a reducing step, an alkylating step, a deglycosylation step, a dephosphorylatian step and an MS analysis step.
[0028] In yet another embodiment, the method of the present invention comprises a conditioning step, a biological sample loading step, a denaturing step, a reducing step, an alkylating step, a deglycosylation and/or dephosphorylation step and a MS analysis step.

[0029] In another additional embodiment, the method of the present invention comprises a conditioning step, a biological sample loading step, a denaturing step, a reducing step, an enzymatic digestion step (PMF) and an MS
analysis step.
[0030] In yet a further embodiment, the enzymatic digestion step is replaced by a chemical digestion step (e.g. acid hydrolysis step).
[0031] In one embodiment, the deglycosylation step is performed prior to the dephosphorylation step. In another embodiment the dephosphorylation step is performed before the deglycosylation step.
[0032] When performing protein characterization spectrometry analysis, it is often desirable to cleave proteins directly on the chip into smaller fragments (peptides) using cleaving reagents for either chemical or enzymatic cleavage. As well known in the art, the digestion of proteins into small fragments provides a mass fingerprint that can be used to determine the protein identity and other characteristics such as posttranslational chemical modifications to specific residues. Thus, the specific fragments that result from digestion can be used as a fingerprint for protein identification by a technique known as peptide mass fingerprinting (PMF). Also, proteolytic fragmentation is useful for high molecular weight proteins because smaller fragments are often more easily measured and resolved by mass spectrometry and chemical modifications can be isolated to specific peptide regions of a protein.
[0033] Thus, in one aspect of the present invention, the enzymatic and/or chemical cleavage of proteins/peptides present in a sample is performed directly on the chip. Subsequent MS analysis is performed in order to obtain a fingerprint of the protein/peptides and determine their identity.

[0034) In accordance with the present invention, several enzymes having different specificity (i.e. cleaving after specific amino acid residues) can be used for PMF and subsequent identification of proteins fragment by MS
analysis.
Proteases, such as trypsin, that cleave proteins into a discrete number of 5 predictable fragments are particularly useful. Other non-limiting examples of enzymes that may be used for direct on-chip digestion include, V8-protease, Arg-C
proteinase, Asp-N endopeptidase, Glu-C endoproteinase, Lys-C endopeptidase, chymotrypsin, pepsin, aminopeptidase M, carboxypeptidase-A, carboxypeptidase-B, carboxypeptidase-Y, caspases 1-10, clostropain (Clostridiopeptidase B), 10 elastase, enterokinase, factor Xa, glutamyl endopeptidase, granzymeB, papain, proline-endopeptidase, pronase, proteinase K, staphylococcal peptidase I, thermolysin, and thrombin.
[0035) As an alternative or complementary approach to enzymatic cleavage for PMF, direct, on-chip chemical cleavage may also be used in accordance with the present invention. Non-limiting examples of compatible reagents that can be used include 2-(2-nitrophenylsulfenyl)-3-bromo-methylindolenine (BNPS-Skatole), Cyanogen Bromide (CNBr), CNBrlheptafluorobutyric acid, Dimethylsulfoxide (DMSO)/HCI and DMSO/Hydrogen bromide (HBr), DMSO/HCI and CNBr, formic acid, hydroxylamine, iodosobenzoic acid, N-bromosuccinimide, N-chlorosuccinimide, 2-nitro-5-thiocyanobenzoic acid (NTCB) and tribromocresol.
[0036) Of course the choice of the particular enzyme or mixture of enzymes to be used will depend on the type of sample (e.g. whether large proteins or peptides are analyzed, the structural properties of the proteins) to be analyzed, etc) and on the information that is sought. Similarly, the particular choice of chemical reagent used will depend on these factors. In addition, the digestion parameters (reaction time, amount of enzyme(s), digestion buffer to be used, etc.) should be adapted to suit the concentration and type of sample that is hydrolyzed and the particular protein chip surface that is used, as well known in the art. Of course mixtures of enzymes, mixtures of chemical reagents and combination of enzymes and chemical reagents may be used in accordance with the present invention. Provided that they are compatible to one another, the particular enzyme and chemical treatments used may be performed directly and simultaneously on the protein chip surface. Alternatively one or more enzymatic treatments) or one or more chemical treatments may be performed directly on-chip in a sequential fashion depending on the specific experimental requirements. Of course the treatments used need to be chosen or adapted so as to enable MS.
[0037] The protein chip surface to be used in accordance with the methods of the present invention depends on the particular physicochemical properties of the protein/peptide sample to be analyzed. Several chip surface arrays are commercially available (e.g. Ciphergen Biosystems, Palo Alto, CA, USA). They are generally derivatized with classic chromatographic separation moieties, such as reverse phase (H4-mimic reversed phase chromatography with C16 functionality), normal phase (NP20-mimic normal phase chromatography with silicate functionality), ion exchange (e.g. CM10-weak cation exchange, with carboxylate functionality with updated hydrophobic barrier coating; WCX2-weak cation exchange with carboxylate functionality; Q10-strong anion exchange with quaternary amine functionality, with hydrophobic barrier coating; SAX2-strong anion exchange with quaternary amine functionality), immobilized affinity capture (IMAC, e.g. IMAC 30-immobilized affinity capture array with nitriloacetic acid (NTA) surface, with hydrophobic barrier coating; IMAC 3-mmobilized affinity capture array with nitriloacetic acid (NTA) surface), mixed mode media (H50-binds protein through reverse phase or hydrophobic interaction chromatography with an updated hydrophobic barrier coating), Surface Enhanced Neat ~esorption (SEND), and gold chip. Examples of other chip surfaces that may be used in accordance with the present invention are disclosed in U.S. patent application 2005/0090016.

[0038] Surface such as these, with broad binding properties are typically used in protein profiling studies and biomarkers discovery (e.g.
where samples from diseased and normal subjects are compared). As well known in the art, biomolecules bind to these surfaces through electrostatic, hydrophobic, coordinate covalent bond or Lewis-acid/base interactions. Of course other types of array surfaces exists and may be used in accordance with the present invention.
[0039] In addition to standard chromatographic surfaces, arrays may be created using virtually any molecules of interest covalently linked to the surface including antibodies, enzymes, ligands, receptors, DNA and lectins. Therefore, as opposed to standard chromatographic media, these specific surtaces can provide much more enrichment of captured analytes due to high specificity of biomolecular interactions. Thus, pre-activated arrays designed specifically for immunoassay, receptor-ligand binding and DNA-binding protein applications are also compatible with the method of the present invention. Non-limiting examples of these chips include RS100, PS10 and PS20 (Ciphergen).
[0040] Thus, depending on the properties of the sample to be analyzed, the appropriate protein chip surface will be selected in accordance with well-known principles of protein separation and identification techniques.
(0041] After binding of the proteins/peptides present in the sample to the protein chip surface, the active surface on the chip is washed with buffers having the desired stringency. The wash (or washes) allows for the removal of analytes with weak surface interaction potential and permits the enrichment of the sample with proteins/peptides having strong surface affinity. Thus, proteins or peptides with shared physical and chemical properties are retained.
[0042] Of course, in accordance with well-known principles of protein separations, the appropriate binding (conditioning) and washing buffers should be selected in order to allow the binding and retention of target biomolecules on the specific protein surface. For example, the pH and salt concentration of the wash buffer will alter the profile of the peptides retained on the ion exchange surface.
Thus, one would adapt these parameters for selecting/retaining the appropriate protein on the chip surface for analysis.
[0043] In one embodiment all steps leading to sample analysis are performed directly on a chip. In another embodiment one or more sample purification steps) is/are performed prior to on-chip analysis. In yet another embodiment an additional wash is performed prior to MS analysis in order to remove components on the chip (e.g. salts present in the buffer) that could interfere with mass spectroscopy (e.g. generally, when working with a SAX2 protein chip, a final wash is necessary when using phosphate or borate buffer).
Thus, depending on the type of chip surface and buffer used, it may be necessary or preferable to add one or more washes) (e.g. with water or suitable buffer), which would remove MS interfering components.
[0044] For example, chemicals are known to interfere with co-crystallization or suppress sample ionization during mass analysis in the protein chip reader. Other chemicals may interfere with binding to the surface of the protein chip array, depending on the specific surface chemistry being used.
Compounds may also interfere with enzymatic reactions that are performed on the chip. Thus, the required additional wash or washes may be introduced before any step which would otherwise be affected by the remaining interfering components.
[0045] For example, salts may reduce binding to ionic surfaces but can increase binding through hydrophobic interactions. Thus, one skilled in the art will choose buffers and wash conditions in accordance with the specific requirements of the protein chip used. With most, but not all, protein chip surfaces used, a water wash must be performed prior to EAM addition. Guidelines for each specific type of protein chip commercialized by Ciphergen are available in their "Protein System Users Guide". Non-limiting examples of chemicals that can interfere with MS
analysis include ionic detergents, high salts concentrations, polyethylene glycol (PEG), glycerol, diethylpyrocarbonate (DEPC) and dithiothreitol (DTT).
[0046] As mentioned above, in many cases ionic detergents will suppress ionization of a protein sample. In particular, proteins that have been boiled in SDS may not be easily detected. Thus, if detergents are necessary for sample extraction or sample solubilization, non-ionic detergents, such as Triton T""
X-100, n-octyl ~-D-glucopyranoside (OGP), NonidetT"" P40 (NP40), or dodecylmaltoside would be preferred. In general, a final concentration of up to 1 is acceptable. Of course, the final acceptable concentration depends on the type of detergent used and the proteins) of interest. Alternatively, the interfering detergent may be removed prior to sample application on the protein chip by any well known techniques or even removed after sample application by performing one or more additional wash(es), provided that the protein chip surface used allows such a procedure (e.g. if the detergent does not interact too strongly with the protein chip surface used).
[0047] In their native state, proteins acquire a specific three-dimensional structure. The linear sequence of amino acids folds upon itself to form a specific native structure. Prior to pertorming a variety of protein chemistry reactions it is often necessary to denature a protein, resulting in an unfolded conformation, which is more susceptible to the subsequent chemical reactions.
Proteins can be denatured by a variety of chemical and other treatments. For example, adding sufficient urea or guanidine - hydrochloric acid (HCI) to a protein solution can result in protein denaturation. Non-limiting examples of chemicals for protein denaturation that can be used in accordance with the present invention include heat, change of pH (acid or alkali), urea, guanidine - HCI, dithiothreitol (DTT), dithioerytritol (DTE), ~3-mercaptoethanol, inorganic salts (lithium bromide, potassium thiocyanate, sodium iodide), organic solvents (ethanol, methanol, trifluoroethanol, formamide, dimethylformamide, dichloro and trichloroacetic acids and their salts), detergents (sodium dodecyl sulphate), high pressure, ultrasonic homogenisation. Of course the choice of the particular denaturing process or 5 chemical agent to be used will depend on the type of sample (e.g. the structural properties of the proteins) to be analyzed etc.) and on the information that is sought. In addition, the denaturing parameters parameters (reaction time, amount of denaturant, denaturing buffer to be used etc.) should be adapted to suit the amount and type of sample that is to be denatured and the particular proteinchip 10 surface that is used, as well known in the art.
[0048] A common naturally occurring posttranslational modification (a chemical modification occurring after protein synthesis) of many proteins is the formation of covalent disulfide bonds between cysteine residues. The formation of such disulfide bonds results in a more rigid protein structure with decreased 15 flexibility. Proteins having disulfide bonds are less susceptible to a number of chemical reactions. Thus, for many applications, it is often desirable to cleave a protein into a number of smaller fragments. In order to cleave proteins having disulfide bonds efficiently, it is often necessary first to reduce the disulfide bonds.
This is normally achieved by chemical reduction of the disulfide bonds with an appropriate reagent. Non-limiting examples of protein reducing agents compatible with the methods of the present invention include dithiothreitol (DTT), dithioerytritol (DTE), cysteine, ~i-mercaptoethanol, ~i-mercaptoethylamine, reduced glutathione, thioglycolic acid and tributylphosphine. Of course, one skilled in the art would appreciate that the above list is not extensive and most low molecular weight thiols would be effective reducing agents that can be used in accordance with the present invention. Of course the choice of the particular reducing agent to be used will depend on the type of sample (e.g. number of disulfide bonds present, the structural and physicochemical properties of the proteins) to be analyzed etc.) and on the information that is sought. In addition, the chemical reduction parameters (reaction time, amount of reducing agent, temperature to be used etc.) should be adapted to suit the amount and type of sample that is to be reduced and the particular proteinchip surface that is used, as well known in the art. It should be noted that the reduction step may be left out altogether in cases where a particular protein of interest does not contain any cysteine residues and/or disulfide bonds.
[0049] Of the course the method of the present invention should be adapted in order to allow sample binding to the chip and MS analysis. Thus, when required, appropriate sample treatments and washes should be performed. For example, DTT is commonly used to reduce disulfide bonds in protein but residual DTT interferes with analysis of protein chip technology. Weak (millimolars) solutions of ~i-mercaptoethanol may be used in accordance with the present invention, in place of DTT for disulfide bond reduction. Alternatively, washes enabling removal of residual DTT may be performed.
[0050 Once reduced, several chemical agents may be employed to block the reduced cyteine residues through a process known as alkylation, avoiding the reformation of undesirable disulfide bonds. In accordance with the present invention, alkylating agents compatible with our approach include iodoacetamide, iodoacetic acid, ethyleneimine, 4-vinylpyridine and acrylamide.
The particular alkylating agent employed often depending on some secondary purpose, for example, to enhance the solubility properties in a given medium, to produce a site subject to proteolysis by a suitable protease such as trypsin, or to provide a reversible protecting group for the cysteine thiol. In addition, the toxicity of the alkylating agent may be considered for reasons of safety, for example, acrylamide is a toxic substance readily absorbed through the skin that is reasonable anticipated to be a human carcinogen. The choice of the particular alkylating agent to be used will also depend on the type of sample (e.g. number of disulfide bonds present, the structural properties of the proteins) to be analyzed ect.) and on the information that is sought. In addition, the alkylating parameters (exposure to light during reaction, reaction time, amount of alkylating agent, alkylation buffer, alkylation temperature to be used etc.) should be adapted to suit the amount and type of sample that is to be alkylated and the particular proteinchip surtace that is used, as well known in the art.
[0051] Proteins are often isolated from nature as glycoproteins. Protein glycosylation is important for the proper function of a number of proteins as well as intercellular communication and other biological phenomena. Altered sugar structures have been associated with a number of diseases including autoimmune disease and cancer (Pauline M. Rudd, Tim Elliott, Peter Cresswell, Ian A.
Wilson, and Raymond A. Dwek. Glycosylation and the Immune System. Science, Mar 2001; 291: 2370 - 2376.; YJ Kim and A Varki. Perspectives on the significance of altered glycosylation of glycoproteins in cancer. Glycoconj J, Aug 1997;
14(5): 569-76). A glycoprotein is a protein that has sugars chemically bound to specific amino acids of the protein. The sugar moiety can be a simple monosacharide or a complex structure composed of several different sugars covalently bound to each other in a variety of branched structures. Often the sugar structures are heterogeneous at a particular glycosylation site, which adds an increased level of complexity in the structural and functional characterization of the glycosylated moieties. These sugar side chains can account for anywhere from less than 1 %
up to 80% of the glycoprotein structure. Sugars are normally added to proteins at specific consensus sites e.g. AsnXxxThr/Ser (Xxx is any amino acid other than proline) for N-linked glycosylation to the Asn residue.
[0052] Sugar moieties can also be bound at the hydroxyl group of Ser and Thr residues in what is known as O-linked glycosylation. Fetuin provides an example of a complex N-linked and O-linked glycoprotein having several glycosylation sites. The study of protein glycosylation is a technically challenging field and mass spectrometry (MS) methods are increasingly being used. For example, an new consensus sequence was only recently confirmed for a AsnAsnCys glycosylation site of the epidermal growth factor receptor (EGFR) expressed in human cells (Zhen Y, Caprioli RM, Staros JV, Characterization of glycosylation sites of the epidermal growth factor receptor. Biochemistry.
2003 May 13;42(18):5478-5492). This discovery is of paramount importance because signaling through the epidermal growth factor receptor plays a vital part in many cancers. An accurate molecular description of the epidermal growth factor receptor, including its glycosylated moieties, may be crucial to our ability to treat the disease. The method of the present invention can be used to characterize the glycosylated portion of glycoproteins. Protein deglycosylation directly on the chip surface can be performed by chemical and enzymatic means. The mass of the protein can be measured before and after deglycosylation indicating the degree of glycosylation. For instance, neuraminidase can be used to remove terminal sialic acid residues from glycoproteins. Several enzymatic deglycosidases may be used in accordance with the present invention. Non-limiting examples include N-glycosidase F (PNGaseF), endoglycosidase H (endoH), endoglycosidase F
(endoF), O-glycosidase and neuraminidase. Reagents for chemical deglycosylation can also be used including hyrdofluoric acid (HF)-pyridine and anhydrous pyridine. Of course, the choice of the particular endoglycosidase used will depend on the information that is sought. More than one deglycosylation step may also be performed in accordance with the present invention. For example a direct on-chip PNGase F treatment, which removes all common classes of N-glycans may be followed by a neuraminidase treatment that releases specific O-linked carbohydrates (i.e. specific forms of N-acetyl-neuraminic acid).
[0053] Protein phosphorylation is an exceedingly important cellular phenomenon directly linked to cancer, cardiovascular diseases, neural function, memory, etc. An estimated one third of proteins present in a given mammalian cell are phosphorylated at any time. Abnormal protein phosphorylation is either a cause or consequence of disease, while normal protein phosphorylation is required for normal cellular function. (Cohen, P. Protein kinases-the major drug targets of the twenty-first century? Nat. Rev. 2002 1 (4):309-315). Proteins are often isolated from nature with phosphorylated serine, threonine and tyrosine residues. The identification and characterization of protein phosphorylation is technically challenging. For example, chicken ovalbumin is a phosphoprotein for which a crystal structure was reported in 1990 (Stein P.E., Leslie A.G.W., Finch J.T., Turnell W.G., McLaughlin P.J., Carrell R.W. Crystal structure of ovalbumin as a model for the reactive centre of serpins. Nature 347:99-102 (1990)). The structure revealed the presence of two phosphorylation sites. However, only recently using mass spectrometric techniques has the presence of two additional phosphorylation sites been found (MacCoss MJ, McDonald WH, Saraf A, Sadygov R, Clark JM, Tasto JJ, Gould KL, Wolters D, Washburn M, Weiss A, Clark JI, Yates JR 3rd.
Shotgun identification of protein modifications from protein complexes and lens tissue. Proc. Natl. Acad. Sci. U.S.A. 2002. Jun 11; 99(12):7900-7905).
[0054] Phosphoproteins can be identified and characterized directly on chip using the method of the present invention. Protein dephosphorylation directly on the chip surface can be performed by chemical and enzymatic means. The mass of the protein can be measured before and after dephosphorylation, indicating the extent of protein phosphorylation. For on chip enzymatic dephosphorylation, phosphatases (acid or alkaline) may be used in accordance with the present invention. Chemical dephosphorylation using HF, HF-pyridine, or other known reagents, can also be performed directly on-chip.
[0055] Once all the desired chemical and enzymatic reactions are performed, the spots on the chip are dried and a matrix solution (comprised of energy absorbing molecules (EAM), allowing energy to be transferred to the analyte i.e. proteins or peptides) is added for MS analysis. The EAM assists in the desorption and ionization of the analyte. The EAM is generally applied in organic solvent, solubilizing many proteins on the protein chip surface. As the EAM
solution dries, the proteins co-crystallize with the EAM. These crystals absorb the laser energy and generate the ionized proteins detected by a protein chip reader.
Any matrix solution allowing MS analysis can be used in accordance with the present invention. Non-limiting examples include saturated sinapinic acid, cyano hydroxyl cinnamic acid (CHCA), EAM 1 (Ciphergen), dihydroxybenzoic acid 5 (DHBA), suitable derivatives of cinnamic acid and mixture thereof. Other suitable energy absorbing molecules are known to those skilled in the art. In general, the EAM is chosen based on the molecular weight of the analyte of interest. For example, saturated sinapinic acid is recommended for proteins of 15 kDa or greater while CHCA is especially good for smaller molecules.
10 [0056] In one particular embodiment, a PAP pen (Zymed mini-PAP pen cat. no. 00-8877) can be used to circle the spots on the chip in order to prevent sample spreading during matrix addition. The pen is particularly useful with array surfaces that do not have a hydrophobic coating. It provides a water-repellent barrier that prevents solutions from bleeding off the chemically active spots of the 15 protein chip array.
[0057] Virtually any type of protein/biological sample can be used in accordance with the present invention. Non-limiting examples include blood, serum, plasma, urine, cerebrospinal fluid (CSF), synovial fluid, nipple aspirate, seminal fluid, tears, hemofiltrate, amniotic fluid, cells or tissue homogenate, cell 20 culture media, purified proteins etc. The biological sample may be treated to physically disrupt tissue or cell structure, thus releasing intracellular components into a solution which may further contain enzymes, buffers, salts, detergents, and the like which are used to prepare the sample for analysis. The sample may be purified or semi purified before performing on-chip analysis depending on the specific experimental requirements. Crude samples may also be used, provided that they do not contain interfering components that cannot subsequently be removed from the chip prior to performing the method step with which it interferes (e.g. MS analysis). Of course, synthetic (e.g. synthetic peptides) or semi-synthetic samples can also be used.
[0058] The method of the present invention is optimized by testing several types of chip surfaces in order to determine which surface gives the best results with a particular type of sample and particular chemical and enzymatic steps performed. Thus, a person skilled in the art could carry out the method of the present invention on 2, 3, 4, 5, 6 or more chip surfaces in parallel and determine which surface gives the best results. Similarly, several chips having the same surfaces could be tested in parallel to determine the optimal binding and washing buffers as well as the optimal incubation time, concentration of sample, reagents, etc, as well known in the art.
[0059] Once all chemical reactions are performed, a MS analysis is conducted to identify the biomolecules of interest. Any suitable MS device may be used in accordance with the present invention as long as it allows proteins/peptides on the substrate to be resolved. Similarly, the measured peptides/proteins can be compared to peptide masses from in silico digestion of the protein database using any search engine available (e.g. ProFoundT"", MascotT"", MS-fitT"", AldenteT"", PhenyxT"", PeptideMapperTM, PeptideSearchT""
and the like).
[0060] The development of two "soft" ionization techniques for the ionization of non-volatile molecules have proven crucial for the development of methods for identification and structure analyses of biological macromolecules.
These two ionization techniques are matrix assisted laser desorption ionization (MALDI) which was described approximately one year after a related report of laser desorption ionization introduced in 1987 by Tanaka (K. H. Tanaka, H.
Wake, Y. Ido, S. Akita, Y. Yoshida and I. Yoshida, Rapid Commun. Mass Spectrom. 8, 2, (1988)) and electrospray ionization (J. B. Fenn, M. Mann, C. K. Meng, S.F.
Wong, C.M. Whitehouse, Science 246, 6, 64 (1989)). Together, the two techniques have made the precision and sensitivity of mass spectrometry readily available for the study of biomolecules and their reactions. As an example, the mass of proteins of a molecular weight exceeding 100 kDa can be readily measured with high sensitivity and accuracy. Currently, there are no other techniques than can achieve comparable results.
[0061] Although not essential, a laser desorption time-of-flight (TOF) mass spectrometer is preferably used for MS analysis in accordance with the present invention. Because of their design features, laser desorption ionization and time-of-flight (TOF) mass spectrometry are complementary and are preferably used. In laser desorption mass spectrometry, a sample containing proteins/peptides is applied to a substrate or a probe and introduced into an inlet system. The proteins/peptides are desorbed and ionized into the gas phase with a laser pulse in the ionization source. The ions generated are sampled into the mass spectrometer by ion optic lenses, and then in a time-of-flight mass analyzer, all ions are accelerated with equal force through a short high voltage field and allowed to drift through a high vacuum chamber. At the opposite end of the high vacuum chamber, the accelerated ions are detected by a sensitive detector surface, with each of the different ions arriving at different times. The time-of-flight is a function of the velocity of the ions, which is dependent on the ratio of mass/charge.
By measuring the elapsed time between ion formation and ion detector impact, the presence or absence of proteins/peptides of specific mass to charge ratio can de determined.
[0062] Matrix-assisted laser desorption/ionization mass spectrometry (MALDI-MS) is a method of mass spectrometry involving the use of an energy absorbing molecule (sample matrix) that permits the desorption of intact proteins or peptide fragments from a laser pulsed probe surface. MALDI is described in U.S. Pat. No. 5,118,937 (Hillenkamp et al.) and U.S. Pat. No. 5,045,694 (Beavis and Chait). The sample is mixed with the MALDI matrix material and placed on the surface of an inert probe. Commonly employed absorbing molecules include cinnamic acid derivatives, sinapinic acid (SPA), cyano hydroxy cinnamic acid (CHCA) and dihydroxybenzoic acid (DHBA). Other suitable energy absorbing molecules can be used by those skilled in this art. The liquid mixture of MALDI
matrix material and sample containing proteins/peptides is allowed to dry forming crystals of encapsulate analyte molecules. The sample is then irradiated for MALDI-MS analysis. The method is useful for detecting proteins/peptides as described in this invention.
[0063] Surface-enhanced laser desorption/ionization mass spectrometry (SELDI-MS) is a flavour of MALDI that allows the fractionation and detection of proteins/peptides in complex mixtures. In SELDI-MS, proteins/peptides are bound to the surface of a protein chip by retentate chromatography due to the physicochemical properties of the chip surface. Non-bound molecules (salts and other interfering molecules) are washed from the probe surface using appropriate buffers before MS analysis. SELDI is described in:
U.S. Pat. No. 5,719,060 ("Method and Apparatus for Desorption and Ionization of Analytes," Hutchens and Yip, Feb. 17, 1998,) U.S. Pat. No. 6,225,047 ("Use of Retentate Chromatography to Generate Difference Maps," Hutchens and Yip, May 1, 2001) and Weinberger et al., "Time-of-flight mass spectrometry," in Encyclopedia of Analytical Chemistry, R. A. Meyers, ed., pp 11915-11918 John Wiley & Sons Chichesher, 2000.
[0064] Proteins on the chip surface can be desorbed and ionized by laser desorption ionization for MS analysis. Any suitable mass spectrometer can be used provided that it allows the analytes to be appropriately resolved.
[0065] For optimal results, a chip reader can be placed in line with a high resolution MDS/Sciex QSTAR or Micromass QTOF mass spectrometer. The sample is read and analysed as it would normally be analysed with a low resolution TOF instrument but with the advantages associated with the high performance mass spectrometer. The quality of data obtained from such an instrumental configuration can reveal a number of characteristics about the sample that are not easily discernable with a low resolution mass spectrometer. For example, exact mass measurements with less than 5 ppm error are often sufficient to confirm the presence of a specific compound. In addition, the QSTAR and QTOF are "tandem" mass spectrometers that can be used for peptide sequencing and rigid identification of compounds and sites of chemical and posttranslational modification. Currently, Ciphergen Biosystems offers a Tandem MS Interface system for compatibility with MDS/Sciex QSTAR mass spectrometers. Ciphergen Applications employing such a configuration have been reported (Prieto, D., Conrads, T.P., Scudiero, D.A., Veenstra, T.D., Profiling of Secreted Proteins from Human Ovarian Cancer Cell Lines by Surface-Enhanced Laser Desorption Ionization Time-of-Flight Mass Spectrometry Journal of Liquid Chromatography &
Related Technologies, 26, 2315-2328, (2003). A Tandem MS Interface system for compatibility with Micromass QTOF mass spectrometers was available for a time but has since been discontinued.
[0066] In order to provide a clear and consistent understanding of terms used in the specification and claims, including the scope to be given such terms, a number of definitions are provided herein below.
[006'T~ Unless defined otherwise, the scientific and technological terms and nomenclature used herein have the same meaning as commonly understood by a person of ordinary skill to which this invention pertains. Commonly understood definitions of molecular biology terms can be found for example in Dictionary of Microbiology and Molecular Biology, 2nd ed. (Singleton et al., 1994, John Wiley &
Sons, New York, NY), The Harper Collins Dictionary of Biology (Hale & Marham, 1991, Harper Perennial, New York, NY), Rieger et al., Glossary of genetics:
Classical and molecular, 5th edition, Springer-Verlag, New-York, 1991; Alberts et al., Molecular Biology of the Cell, 4th edition, Garland science, New-York, 2002;
and, Lewin, Genes VII, Oxford University Press, New-York, 2000. Generally, the procedures of sample/protein purification and separation, protein chip utilization, MS analysis, molecular biology methods and the like are common methods used in 5 the art. Such standard techniques can be found in reference manuals such as for example Sambrook et al. (2000, Molecular Cloning - A Laboratory Manual, Third Edition, Cold Spring Harbor Laboratories); and Ausubel et al. (1994, Current Protocols in Molecular Biology, John Wiley & Sons, New-York). Laemmli, U.K.
(1970). Nature (Lond.), 227, 680-685.;Practical protein chemistry, A handbook A.
10 Darbre Ed. Wiley John Wiley and sons. Copyright 1986.MP Washburn, D
Wolters, and JR Yates 3'd Large-scale analysis of the yeast proteome by multidimensional protein identification technology. Nat Biotechnol, Mar 2001; 19(3): 242-247.
JR
Yates 3'd Mass spectral analysis in proteomics. Annu Rev Biophys Biomol Struct, Jan 2004; 33: 297-316. Industrial proteomics Applications for Biotechnology and 15 Pharmaceuticals. Daniel Figeys Ed. John wiley and Sons Copyright 2005.
Karas, M and Hillenkamp F (1988) Laser desorption ionization of proteins with molecular masses exceeding 10,000 daltons. Anal. Chem. 60, 2299-2301.
DEFINITIONS
[0068] The use of the word "a" or "an" when used in conjunction with 20 the term "comprising" in the claims and/or the specification may mean "one"
but it is also consistent with the meaning of "one or more", "at least one", and "one or more than one".
[0069] Throughout this application, the term "about" is used to indicate that a value includes the standard deviation of error for the device or method being 25 employed to determine the value. In general, the terminology "about" is meant to designate a possible variation of up to 10%. Therefore, a variation of 1, 2, 3, 4, 5, 6, 7, 8, 9 and 10 % of a value is included in the term about.

[0070] As used in this specification and claim(s), the words "comprising" (and any form of comprising, such as "comprise" and "comprises"), "having" (and any form of having, such as "have" and "has"), "including" (and any form of including, such as "includes" and "include") or "containing" (and any form of containing, such as "contains" and "contain") are inclusive or open-ended and do not exclude additional, un-recited elements or method steps.
[0071] As used herein, the twenty natural amino acids and their abbreviations follow conventional usage. Stereoisomers (e.g., D-amino acids) such as a,a-disubstituted amino acids, N-alkyl amino acids, lactic acid and other unconventional amino acids may also be suitable components for the polypeptides of the present invention. Examples of unconventional amino acids include but are not limited to selenocysteine, citrulline, ornithine, norvaline, 4-(E)-butenyl-4(R) -methyl-N-methylthreonine (MeBmt), N-methyl-leucine (MeLeu), aminoisobutyric acid, statine, N-methyl-alanine (MeAla).
[0072] As used herein, "protein" or "polypeptide" means any peptide-linked chain of amino acids, regardless of post-ranslational modifications (e.g.
phosphorylation, glycosylation, sulfatation, acetylation, sumoylation, prenylation, ubiquitination etc).
[0073] As used herein, the term "purified" refers to a molecule (e.g. a polypeptides or proteins) having been separated from a component of the composition in which it was originally present. Thus, for example, a "purified protein or polypeptide" has been purified to a level not found in nature. A
"substantially pure" molecule is a molecule that is lacking in most other components (e.g., 30, 40, 50, 60, 70, 75, 80, 85, 90, 95, 96, 97, 98, 99, 100%
free of contaminants). By opposition, the term "crude" means molecules that have not been separated from the components of the original composition in which it was present. Therefore, the terms "separating" or "purifying" refers to methods by which one or more components of the biological sample are removed from one or more other components of the sample. Sample components include nucleic acids in a generally aqueous solution that may include other components, such as proteins, carbohydrates, or lipids. A separating or purifying step preferably removes at least about 70% (e.g., 70, 75, 80, 85, 90, 95, 96, 97, 98, 99, 100%), more preferably at least about 90% (e.g., 90, 91, 92, 93, 94, 95, 96, 97, 98, 99, 100%) and, even more preferably, at least about 95% (e.g., 95, 96, 97, 98, 99, 100%) of the other components present in the sample from the desired component. For the sake of brevity, the units (e.g. 66, 67...81, 82,...91, 92%....) have not systematically been recited but are considered, nevertheless, within the scope of the present invention.
[0074] The terms "inhibiting," "reducing" or "interfering" or any variation of these terms, when used in the claims and/or the specification includes any measurable decrease or complete inhibition of at least one chemical, physicochemical, or enzymatic activity in any of the present method steps to achieve a desired result. For example, a compound is said to be interfering with MS detection when a decrease in specificity and sensitivity is measured following a treatment with the "inhibiting", "reducing" or "interfering" compound as compared to in the absence thereof. Similarly, a compound is said to be "inhibiting" an enzymatic step (e.g. dephosphorylation, deglycosylation, trypsinization, etc) of the method of the present invention when the efficiency of the enzymatic reaction is reduced or completely abolished following a treatment with the "inhibiting", "reducing" or "interfering" compound as compared to in the absence thereof.
[0075] "Probe" refers to a device that is removably insertable into a gas phase spectrometer and comprises a substrate having a surface for presenting analytes for detection. A probe can comprise a single substrate or a plurality of substrates. Terms such as protein chip, protein chip array, or chip are also used herein to refer to specific kinds of probes.

[0076] "Gas phase ion spectrometer" refers to an apparatus that measures a parameter which can be translated into mass-to-charge ratios of ions formed when a sample is ionized into the gas phase. Generally ions of interest bear a single charge, and mass-to-charge ratios are often simply referred to as mass.
[0077] "Mass spectrometer" refers to a gas phase ion spectrometer that includes an inlet system, an ionization source, an ion optic assembly, a mass analyzer, and a detector.
[0078] "Laser desorption mass spectrometer" refers to a mass spectrometer which uses laser as an ionization source to desorb an analyte.
[0079] "Binding functionalities" refer to functional groups) of a protein chip surface material that bind analytes. Binding functionalities can include, but are not limited to, a carboxyl group, a sulfonate group, a phosphate group, an ammonium group, a hydrophilic group, a hydrophobic group, a reactive group, a metal chelating group, a thioether group, a biotin group, a boronate group, a dye group, a cholesterol group, derivatives thereof, or any combinations thereof.
Binding functionalities can further include other adsorbents that bind analytes based on individual structural properties, such as the interaction of antibodies with antigens, enzymes with substrate analogs, nucleic acids with binding proteins, and hormones with receptors.
[0080] "Analyte" refers to a component of a sample which is desirably retained and detected. The term can refer to a single component or a set of components in the sample.
[0081] "Conditioned" as applied to the present invention relates to adaptation or modification of a substrate surtace (protein chip surface) to promote adhesion of analytes onto the substrate surface.
[0082] "Energy absorbing molecule" or "EAM" refers to a molecule that absorbs energy from an ionization source in a mass spectrometer thereby enabling desorption of analyte from a probe surface. Energy absorbing molecules used in MALDI are frequently referred to as "matrix." Cinnamic acid derivatives, sinapinic acid ("SPA"), cyano hydroxy cinnamic acid ("CHCA") and dihydroxybenzoic acid are frequently used as energy absorbing molecules in laser desorption of bioorganic molecules. Other suitable energy absorbing molecules are known to those skilled in this art. See, e.g., U.S. Pat. No. 5,719,060 (Hutchens & Yip) for additional description of energy absorbing molecules.
[0083] Other objects, advantages and features of the present invention will become more apparent upon reading of the following non-restrictive description of illustrative embodiments thereof, given by way of example only with reference to the accompanying drawings.
BRIEF DESCRIPTION OF THE DRAWINGS
[0084] Having thus, generally described the invention, reference will be made to the accompanying drawings, showing by way of illustration only an illustrative embodiment thereof and in which:
[0085] Figure 1 shows one strategy for on-chip protein analysis. The schematic shows the steps that were followed to monitor deglycosylation and dephosphorylation reactions and for identification of proteins investigated.
[0086] Figure 2 shows a mass spectrum of on-chip denaturation, reduction, alkylation and deglycosylation of 1 Ng ovalbumin on H4 chips (a:
before deglycosylation; b: after deglycosylation) and 1 Ng EGFRED on NP20 chips (c:
before deglycosylation; d: after deglycosylation).
[0087] Figure 3 shows a mass spectrum of on-H4 chip denaturation, reduction, alkylation and dephosphorylation of 1 Ng ovalbumin (a: before 5 dephosphorylation; b: after dephosphorylation).
[0088] Figure 4 shows a mass spectrum of 1 pg ovalbumin after its on-H4chip denaturation, reduction, alkylation, deglycosylation and tryptic digestion.
[0089] Figure 5 shows a mass spectrum of 1 Ng EGFRED after its on-NP20 chip denaturation, reduction, alkylation, deglycosylation and a) Asp-N
10 digestion, b) Glu-C digestion, c) Lys-C digestion.
[0090] Figure 6 shows a mass spectrum of 1 Ng fetuin after its on-chip denaturation, reduction, alkylation, deglycosylation and tryptic digestion on chips.
DETAILED DESCRIPTION OF ILLUSTRATIVE EMBODIMENTS
15 [0091] Here, processes combining chemical and enzymatic treatments directly on-chip to monitor various protein modifications such as deglycosylation and dephosphorylation reactions, and identified proteins using PMF were examined. three representative proteins were selected based on their complexity and physico-chemical features (Table 1 ). The hydrophobicity of the proteins will 20 determine how tightly they are bound to the chromatographic surface on the chip and the wash cycle chosen to remove impurities is dictated by this interaction. An outline of the general procedure developed for chemical and enzymatic treatment of proteins and peptides is shown in Figure 1. This approach allowed to adapt different sequences of reactions according to the characteristics of the proteins 25 and the nature of their modifications.

Table 1. Structural characteristics of the model proteins Protein Average Number Number of Number of MW of (kDa) Disulfide GlycosylationPhosphorylation brid a siteb site Human EGFRED 81.29 21 11-N 0 Bovine fetuin 46.21 6 3-N, 3-O 0 Chick ovalbumin44.73 1 1-N 4 Molecular weight of the native protein without posttranslational modrtications.
Carbohydrate moieties can be attached via N- or O- linkages to the proteins.
[0092] Epidermal growth factor receptor ecto domain (EGFRED), chicken ovalbumin and bovine fetuin were selected as model proteins because of their complexity, specific physicochemical properties and posttranslational modifications (PTMs). Less complex proteins (chicken lysozyme, horse cytochrome C bovine serum albumin, bovine insulin) were also analysed and gave excellent sequence coverages (data not shown). Based on the performance of the methods of the invention with the three complex proteins listed above, the teachings of the present invention are amenable to any protein of interest.
[0093] The present invention is illustrated in further details by the following non-limiting examples.

MATERIALS AND METHODS
[0094] Materials. Human EGFRED (Epidermal growth factor receptor ecto domain) was a gift from Dr. J. Baardsnes of the Biotechnology Research Institute, National Research Council of Canada, Montreal, Canada. Trypsin was obtained from Boehringer Mannheim (Ingelheim, Germany) and used without further purification. Urea, ammonium bicarbonate, a-cyano-4-hydroxy-cinnamic acid (CHCA), sinapinic acid (SPA), dithiothreitol (DTT), iodoacetamide (IA), adrenocorticotropic hormone (ACTH), alkaline phosphatase (ALP), chicken ovalbumin, chicken lysozyme, bovine insulin, bovine serum albumin, horse cytochrome C and bovine fetuin were obtained from Sigma (St. Louis, MO).
Trifluoroacetic acid (TFA) was from Pierce (Rockford, IL). Neuraminidase, N-glycosidase F, O-glycosidase, endoproteinases Arg-C, Asp-N, Glu-C and Lys-C
were purchased from Roche (Indianapolis, IN). Protein chips (working spot 2.5 mm diameter-H4, NP20, SAX2, WCX2, IMAC3 and Send Alpha) were purchased from Ciphergen Biosystems Inc. (Fremont, CA). Microspin 6 columns were obtained from Bio-Rad (Mississauga, ON). Urine samples were collected from a male Sprague-Dawley rat treated with puromycin aminonucleoside to induce proteinuria.
All solvents were HPLC grade. Aqueous solutions of the proteins, enzymes and reagents used in the experiments described below were prepared in 0.1 M
ammonium bicarbonate unless otherwise indicated.
[0095] Chemical treatment. In these particular examples Hydrophylic NP20 and hydrophobic H4 chips were used. Proteins were first chemically denaturated, reduced and alkylated as follows: chips were conditioned by adding 3 pL of water or acetonitrile. The selectivity of the chip varies with the organic component and /or salt concentration of the binding buffer. One pL of 1 pg/pL
solution of the protein (e.g. ovalbumin, 23.4 pmol; EGFRED, 14.6 pmol; and fetuin, 26.0 pmol) was added to the chip. Denaturation was accomplished by adding 2 NL
of 8 M urea and incubating for two hours at room temperature. Disulfide bonds were reduced by adding 1 NL of 40 mM DTT with incubation in a water bath at 56°C for 45 min. Finally, alkylation of the thiol groups was performed in a dark humidity chamber at room temperature by applying 1 NL of 220 mM iodoacetamide and allowing the reaction to proceed for 30 min. The denatured, reduced and alkylated proteins were subjected to two different enzymatic reaction schemes whereby the model proteins were 1) dephosphorylated or deglycosylated and analyzed by SELDI-TOF/MS to monitor the removal of the corresponding posttranslational modifications or 2) proteolyzed prior to PMF for identification purposes using database searching tools.

[0096] Enzymatic deglycosylation. Deglycosylation of EGFRED and ovalbumin (N-glycan-containing proteins) was performed by depositing 1 wL of 1 Unit/wL solution of N-glycosidase F and incubating the array in a 37°C
water bath for 2 hrs.
[0097] Fetuin contains both N- and O- carbohydrate linkages.
Deglycosylation reactions were performed sequentially on an H4 chip. Two experimental approaches were applied: a) monitoring of the deglycosylation reactions using denatured protein without reduction and alkylation; b) the complete set of chemical reactions described above (Figure 1 ) was applied prior to performing enzymatic proteolysis. The denatured or alkylated fetuin was N-deglycosylated by spotting 1 pL of 0.5 Unit/wL of N-glycosidase F on the chip and incubating 2 hrs at 37°C in a water bath. Conversely, O-linked carbohydrates were selectively cleaved in a two-step approach: sialyl (a-N-acetylneuraminic acid) residues were cleaved by adding 1 ~L of 5 mUnit/p,L neuraminidase solution and incubating for 1 hr in 37°C water bath whereas the serine/threonine O-linked ~3-D-galactosamine residues were cleaved by adding 1 ~L of 2 mUnit/pL of O-glycosidase and incubating the arrays for 2 hours in a 37°C water bath.
[0098] Enzymatic dephosphorylation. Ovalbumin was denatured, reduced and alkylated as described above on an H4 chip. Dephosphorylation was performed by adding 1 ~,L of 2 Ng/wL ALP solution (0.1 M ammonium bicarbonate, 1 mM magnesium chloride) and incubating in a 37°C water bath for 2 hrs.
[0099] Enzymatic proteolysis. Enzymatic digestion was performed following denaturation, reduction, alkylation and/or deglycosylation of the proteins on NP20 hydrophilic chips. One ~L of 0.5 Ng/~L trypsin solution was applied to each spot and digested for 2 hrs at 37°C in a water bath. The array was air-dried and rinsed twice with 4 NL of water prior to adding the sample matrix. In addition to trypsin proteolysis, EGFRED was also treated with 0.5 Ng quantities of four other proteases: Arg-C, Asp-N, Glu-C and Lys-C. To differentiate the peptides generated from the digestion of the model proteins from those originating from autolysis, control experiments were conducted with all reagents and/or proteases in the absence of the proteins.
[0100] Rat urinary proteins. Sprague-Dawley rats were administered a single 100 mg dose of puromycin aminonucleoside to induce proteinuria and urine samples were collected in plastic vials containing phenol as stabilizer at specific time intervals after administration. The samples were centrifuged to remove debris, divided in 50 wL aliquots which were frozen at -80°C
until analyzed.
Samples were thawed on ice and a 25 pL aliquot was desalted with the Bio-Rad column. The resulting eluate was concentrated to a volume of 10 ~L. The protein content in this urine sample was 0.6 ~g NL-' by Bradford assay'4. The concentrated sample was applied to the NP20 chip followed by denaturation, reduction, alkylation and trypsinization prior to analysis.
[0101] Mass spectrometric analysis. To deglycosylated, dephosphorylated and native samples 1 IrL of saturated SPA prepared in 50%
aqueous acetonitrile containing 0.5%, TFA was added to each spot. For proteolyzed samples, the array was air dried and rinsed twice with 4 pL water.
The washing was done by pipetting water to the chip and aspirating the water several times between the spot and the pipette tip prior to the addition of 0.5 NI of 20%
CHCA in 50% aqueous acetonitrile containing 0.5% TFA to each spot. Mass spectra were generated in the positive-ion mode using a PBSII-c ProteinChip reader (Ciphergen Biosystems Inc, Fremont, CA). The instrument was calibrated externally with ACTH at 2.465 kDa and bovine insulin at 5.733 kDa. The average mass accuracy after external calibration of PBSII-c is 2000 ppm (0.2%) for proteins of 10 kDa to 300 kDa and 1000 ppm (0.1 %) for polypeptides of 1 kDa to 10 kDa.
Resolution was greater than 700 (average) for 5 pmol of human recombinant insulin. MascotT"" (Matrix Science ltd, London U.K.) was used for protein identification based on PMF analysis.

ON-CHIP PROTEIN DEGLYCOSYLATION
[0102 Protein glycosylation is an important protein modification serving 5 various functions, which are protein dependent. Glycosylation can protect a protein from degradation, retain the protein in the endoplasmic reticulum until properly folded, or direct the protein to its proper destination by serving as a transport signal. Oligosaccharides exposed on the cell surfacec allow different cells to recognize each other.
10 [0103] Ovalbumin is a 44.73 kDa hydrophobic glycoprotein which has a disulfide bridge, one N-glycosylation and four phosphorylation sites'5. These characteristics make it an attractive example for the assessment of the on-chip protein analysis of the present invention. The oligosaccharide moiety of ovalbumin is heterogeneous with an average of 1.65 kDa'6. In the experiments performed the 15 average molecular weight of ovalbumin as measured was decreased by 1.67 kDa to 43.06 kDa (Figure 2a and b) when N-deglycosylated on an H4 chip, indicating that the glycan side chain was completely removed. N-deglycosylation proceeded at a slower rate on the hydrophilic NP20 chip, as significant amounts of intact ovalbumin and a byproduct were seen. After 2hr of reaction loss of only 1.376 kDa 20 from ovalbumin molecular weight (data not shown), was observed. This indicates that the chemistry of both the protein and chip surface plays an enabling role in the deglycosylation reaction. One may hypothesize that the hydrophobic nature of the H4 chip and that of several segments of ovalbumin create appropriate binding conditions for deglycosylation of the Asp_292 glycan residue, which are probably not 25 favored on the hydrophylic NP20 chip.
[0104 Human EGFRED has eleven N-glycosylation sites with a variety of glycoforms and twenty-one disulfide bridges modulating its tertiary structure ".
This heavily-modified protein is a good example to test this protocol.
Alkylated EGFRED was digested with N-glycosidase F for two hours on a NP20 chip and analyzed. As shown in figure 2c and 2d, a mass shift of 9.53 kDa (from 81.29 kDa to 71.76 kDa) indicates several or all glycosylation sites were removed. The reaction proceeded at a much slower rate on the H4 chip as unreacted EGFRED
was still detected, even when the reaction time was extended to 3 or 4 hr (data not shown).
[0105] Fetuin has three N-linked oligosaccharides, three O-linked oligosaccharide chains and a potential fourth O-linked glycan'$. Each of the carbohydrate units attached to asparagine residue has hybrid structures with a molecular weight of approximately 2.86 kDa'9. In the deglycosylation approach used herein, fetuin was denatured on H4 chip prior to treatment with N-glycosidase F, for two hours at 37°C. Analysis showed that fetuin was deglycosylated as its molecular mass was reduced by 5.19 kDa, from 46.21 kDa to 41.02 kDa. The O-glycosidically linked sugar side chains comprises a disialated structure with a molecular weight of approximately 950 Da'8. All three O-linked glycosylation sites on fetuin were removed as its molecular weight decreased by approximately 3.00 kDa when treated with neuraminidase and O-glycosidase.

ON-CHIP PROTEIN DEPHOSPHORY~ATION
[0106] Most aspects of cell life are regulated by protein phosphorylation; abnormal phosphorylation can result in or be caused by disease2°. At any moment roughly 30% of all mammalian proteins are phosphorylated. This reversible reaction is regulated by the concerted actions of protein kinases and phosphatases, which affect phosphorylation and de-phosphorylation respectively.

[0107] Ovalbumin, with its four phosphoserine sites, was chosen to investigate protein dephosphorylation. This process resulted in a mass difference of approximately 338 Da less than native ovalbumin, which corresponds to the removal of the four phosphate molecules (Figure 3). Similarly to results mentioned above for the N-deglycosylation reaction, negligible dephosphorylation activity was observed when using the NP20 chip even when increasing phosphatase concentration and/or extending reaction time (data not shown). This suggests that chips with a hydrophobic surface are probably more suitable than hydrophilic chips for performing enzymatic treatment of proteins with a hydrophobic backbone such as ovalbumin.

ON-CHIP PROTEOLYSIS
[0108] Trypsinization of denatured, reduced and alkylated ovalbumin generated complex peptide profile. This was not surprising because of the complexity of the native hydrophobic protein. Proteolysis was only efficient when ovalbumin was first deglycosylated on H4 chips according to the sequence of reaction presented in Figure 1. Deglycosylated ovalbumin was subsequently trypsinized and analyzed (Figure 4). In this study, the mass tolerance of the MASCOT search program was set at 2 Da. Eleven peptides, corresponding to 46% coverage of the protein amino acid sequence (Table 2) were matched with their predicted peptides from the in-silico digestion. The search results identified chicken ovalbumin with a score of 85 among a relatively complex mixture of peptides originating not only from ovalbumin, but also from N-glycosidase F
and trypsin. This on chip method provided a sequence coverage similar to that obtained from in-solution digestion and is an elegant and flexible approach to the characterization of proteins with oligosaccharide substituants.
[0109] EGFRED was treated on-chip through a sequence of five chemical and enzymatic treatments. The last proteolytic reaction was performed using trypsin, Arg-C, Asp-N, Glu-C and Lys-C. Typical mass spectra are presented in Figs. 5a, 5b and 5c. Each of the proteolytic reactions generated 13, 5, 15, and 9 specific peptides respectively. EGFRED was identified by PMF with a score of 60 (trypsin), 125 (Asp-N), 91 (Glu-C), 36 (Arg-C), and 70 (Lys-C). The sequence coverage was 36%, 54%, 40%, 23% and 17% with trypsin, Asp-N, Glu-C, Lys-C, and Arg-C, respectively. Results of the database search are summarized in Table 2.
[0110] Fetuin was trypsinized (Figure 6) following denaturation, reduction, alkylation and N-deglycosylation on NP20 chips. Its identity was confirmed with nine peptides matched against the in-silico digest (Table 2).
The MASCOT search through the SwissProt protein database showed that fetuin was ranked at the top of the list with a score of 62 and 44% sequence coverage.

ON-CHIP IDENTIFICATION OF RAT URINARY PROTEINS
[0111] Proteolysis of urinary proteins was also investigated by directly applying a urine aliquot on-chip and performing the reactions described. Ten ~I of rat urine was applied, reduced, alkylated and trypsinized as described above.
MS
analysis and peptide mass fingerprinting identified 5 and 11 peptides specific to Alpha-2u-globulin (AUG) and rat urinary albumin (RUA) respectively. The sequence coverages of AUG and RUA are shown in Table 2. AUG is the most abundant in normal rat urine whereas RUA is most prevalent in puromycin-induced proteinuria. Both contain disulfide bridges and AUG is N-glycosylated. Their identities were independently confirmed by LCMS/MS2'.

Table 2. Proteins enzymatic treatment and identification Protein Proteolytic . Number of _ _ % Sequence EGFRED Trypsin 13 36 Arg-C 5 17 Asp-N 15 54 Glu-C 12 40 Lys-C 9 23 Bovine fetuin Trypsin 9 44 Ovalbumin Trypsin 11 46 Rat urinary albumin Trypsin 11 25 (65.90 kDa) Alpha-2u-globulin (18.73Trypsin 5 42 kDa) [0112] In addition to the peptides derived from ovalbumin, EGFRED, fetuin and the rat urine proteins AUG and RUA, some ion peaks found in the protein digests were also present in the blank. Several enzymatic autolysis and proteolysis products from N-glycosidase and human keratin were observed.
[0113] On-chip chemical and enzymatic reactions are dependent on the physico-chemical properties of the chip surface. The interactions are analogous to those involved in normal and reverse phase chromatographic separations. For example, hydrophobic proteins will bind tighter to a reversed phase surface than hydrophilic proteins. These interactions determine the type of chip used for a particular on-chip experiment. As previously mentioned, the polarity of the binding buffer influences the selection of proteins that are retained on the chip.
[0114] Several chips are currently available for this protocol. These include anion and cation exchange, metal ion, antibody-antigen, receptor-ligand and DNA-protein interaction chips. Therefore the appropriate chip should be chosen for optimal reaction, selectivity and sensitivity. Complex samples (plasma, urine, cerebrospinal fluid etc.) may require fractionation or depletion strategies to isolate target proteins prior to the application of the methods of the present invention.
Also, the present methods could provide a rapid means for characterizing native proteins or antibody therapeutics as well as chemically modified proteins and 5 formulated protein products in the course of their manufacturing and quality control processes. Greater usefulness of this method can be attained if the chip is read by a high resolution (> 12,000) mass spectrometer with high mass accuracy (< 10 ppm). This will afford the microcharacterization of protein modifications. For example, single phosphorylation will show a monoisotopic difference of 79.6633 10 amu, the molecular weight of the phosphate moiety. Finally, as mentioned above, more than one chip could be chosen for a particular protein, proteins or sample, and the chips be treated in parallel (with modifications of the treatment adapted for particular chips, if required and as known in the art) to optimize the method and the obtention of the best results.

[0115] Protein identification generally involves isolation of proteins of interest by electrophoresis and/or chromatographic methods followed by denaturation, reduction, alkylation and proteolytic digestion. All steps are normally carried out in solution to generate peptides for Edman degradation or sequencing 20 by MS methods. The major advantages of protocols of the present invention are its simplicity, speed and sensitivity. Low picomolar amounts of relatively complex proteins can be rapidly deglycosylated, dephosphorylated and/or proteolyzed and readily identified. The sequence coverage obtained by these methods is similar to that generated from in-solution digestion.
25 [0116] In conclusion, the herein-described approach to protein identification and characterization can be routinely used in several areas of biomarker research and related applications, particularly in molecular diagnostics and monitoring of disease, assessment of drug efficacy and basic proteomic research. The methods of the present invention work effectively for proteins which have complex structures as demonstrated by the analysis of EGFRED, fetuin, and ovalbumin.
Thus the on chip deglycosylation and dephosphorylation of the present invention provide an excellent approach for rapid analysis of modified proteins.
[0117) Although the present invention has been described hereinabove by way of preferred embodiments thereof, it can be modified, without departing from the spirit and nature of the subject invention as defined in the appended claims.

REFERENCES
1. Aebersold, R,; Mann, M. Nature, 2002, 422, 198-207.
2. Figeys, D. In Two-dimensional Gel Electrophoresis and Mass Spectrometry for Proteomics Studies: State-of-the-art; Rehm, H.J., Reed, G., Ed.; John Wiley &
Sons: New York, 2001; pp 243-268.
3. Gorg, A. In Advances in 2D gel techniques. Proteomics: A Trend Guide; Mann, M.; Blackstock, W. Ed.; Elsevier: London, 2000; pp 3-6.
4. Link, A.J.; Eng, J.; Schieltz, D.M.; Carmack, E.; Mize, G.J.; Morris, D.R.;
Garvik, B.M.; Yates, J.R. 3rd. Nat. Biotechnol. 1999, 17, 676-682.
5. Wolters, D.; Washburn, M.P.; Yates, J.R. 3rd. Anal. Chem. 2001, 73, 5683-5690.
6. Hunt, D.F.; Mitchel, H.; Dickinson, T.A.; Shabanowitz, J.; Cox, A.;
Sakaguchi, K.; Appella, E.; Grey, H.M.; Sette, A. Science. 1992, 256, 1817-1820.
7. Kierman, U.A.; Tubbs, K.A.; Nedelkov, D.; Niederkofler, E.E.; McConnell, E.;
Nelson, R.W. J Proteome Res. 2003, 2, 191-197.
8. Nelson, R.W.; Krone, J.R.; Bieber, A.L.; Williams, P. Anal. Chem. 1995, 67, 1153-1158.
9. Merchant, M.; Weinberger, S.R. Electrophoresis. 2000, 21, 1164-1177.
10. Gevaert, K.; Vanderkerckhove, J. Electrophoresis. 2000; 21, 1145-1154.
11. Lin, S.; Tornatore, P.; King, D.; Orlando, R.; Weinberger, S.R.
Proteomics.
2001, 1, 1172-1184.
12. Dare, T.; Davies, H.A.; Turton, J.A.; Lomas, L.; Williams, T.C.; York, M.J.
Electrophoresis. 2002, 23, 3241-3251.
13. Caputo, E.; Moharram, R.; Martin, B.M. Anal. Biochem. 2003, 321, 116-124.
14. Bradford M. Anal. Biochem. 1976, 72, 248-254.

15. MacCoss, M.J.; McDonald, W.H.; Saraf, A.; Sadygov, R.; Clark, J.M.; Tasto, J.J.; Gould, K.L.; Wolters, D.; Washburn, M.; Weiss, A.; Clark, J.I.; Yates, J.R. III.
PNAS. 2002, 99, 7900-7905.
Duffin, K.L.; Welply, J.K.; Huang, E.; Henion, J.D. Anal. Chem. 1992, 64, 1440-1448.
Zhen, Y.; Caprioli, R.M.; Stratos, J.V. Biochemistry. 2003, 42, 5478-5492.
18. Spiro, R.G.; Bhoyroo, V.D. J. Biol. Chem. 1974, 249, 5704-5717.
19. Nilsson, B.; Norden, N.E.; Svensson, S. J. Biol. Chem. 1979, 254, 4545-4553.
20. Cohen, P. Nat. Rev. Drug Disc. 2002, 1, 309-315.
21. Ge, Y.; Aguiar, M.; Gibbs, B.F.; Masse, R. 52nd conference ASMS. May 2004, Nashville, TN.

Claims (13)

1. A method for analysing proteins or peptides directly on chip comprising performing at least two chemical or enzymatical reactions directly on a protein chip surface, said chemical and/or enzymatical reactions being selected from the group consisting of:
a) denaturation;
b) reduction of disulfide bridges;
c) alkylation;
d) deglycosylation;
e) dephosphorylation; and f) digestion of peptide bonds for peptide mass fingerprinting.

2. The method of claim 1, wherein at least 3 chemical and/or enzymatical reactions are performed.

3. The method of claim 1, wherein at least 4 chemical and/or enzymatical reactions are performed.

4. The method of claim 1, wherein at least 5 chemical and/or enzymatical reactions are performed.

5. The method of anyone of claims 1-4, wherein said protein chip surface is selected from the group consisting of:
a) standard chromatography surfaces; and b) specific surfaces.

6. The method of claim 5, wherein said standard chromatographic surface is selected from the group consisting of:
a) ion exchange;

b) reverse phase;
c) normal phase;
d) immobilized affinity capture (IMAC);
e) mixed mode media;
f) surface enhanced neat desorption; and g) gold surface.

7. The method of claim 6, wherein said specific surface comprises biomolecules selected from the group consisting of:
a) antibodies;
b) enzymes;
c) ligands;
d) receptors;
e) DNA; and f) lectins.

8. A method for analysing proteins or peptides directly on chip comprising:
a) conditioning a protein chip surface;
b) loading a protein containing sample on said protein chip surface;
c) denaturing said protein on said chip surface;
d) reducing said protein on said chip surface;
e) performing at least one of the following steps:
i) deglycosylating said protein on said chip surface;
ii) dephosphorylating said protein on said chip surface;
and iii) digesting said protein on said chip surface for peptide mass fingerprinting; and f) performing mass spectroscopy analysis.

9. The method of claim 8, wherein said digesting is enzymatic.

10. The method of claim 8, wherein said digesting is chemical.

11. The method of claim 8, wherein said sample is a purified sample.

12. The method of claim 8, wherein said sample is a crude sample.

13. The method of claim 8, further comprising a washing step prior to mass spectroscopy analysis for removal of interfering molecules.