WO2004092739A1

WO2004092739A1 - Method for mass spectrometry involving fluorescent labeling of the analyte

Info

Publication number: WO2004092739A1
Application number: PCT/IB2004/001285
Authority: WO
Inventors: Akihiko Kameyama; Yasuro Shinohara
Original assignee: Amersham Biosciences K.K.
Priority date: 2003-04-18
Filing date: 2004-04-14
Publication date: 2004-10-28
Also published as: JP2004317398A

Abstract

This invention relates to mass spectrometric identification of biomolecules that occur in very small amounts in the tissues or cells or living body. Specifically, the invention provides a method for promoting the ionization of those biomolecules which have heretofore been held difficult to ionize, especially sugar chains. The method according to the invention of performing mass spectrometry on biomolecules comprises marking a biomolecule with a fluorescence labeling agent that has a reactive group capable of binding to the biomolecule and a positive total charge of from +1 to +6 or a negative total charge of from -6 to -1.

Description

DESCRIPTION

METHOD FOR MASS SPECTROMETRY INVOLVING FLUORESCENT LABELING OF T_{HE A}NAL_YTE

5 FIELD OF THE INVENTION

The present invention relates to an improved method of mass spectrometry for detecting the structural information about biomolecules that occur in very small amounts in organism tissues or cells. Specifically, the 0 invention relates to a method for increasing the sensitivity of detection of biomolecules , especially sugar chains, in mass spectrometry by using fluorescence labeling agents having reactive groups capable of binding to the biomolecules, which is characterized in that the 5 fluorescence labeling agents have electric charges.

The method of the invention is preferably implemented by using a technique that combines a fluorescence labeling agent having electric charges with matrix-assisted laser desorption/ionization time-of-flight mass spectrometry 0 (MALDI-TOF MS) .

BACKGROUND OF THE INVENTION

In studying the higher regulatory mechanism of living body, it is extremely important to investigate the 5 correlation between the structure of a biomolecule and its function. Extracting nucleic acids and proteins that are only expressed in very small amounts in living body's tissues and cells and identifying their sequence information and three-dimensional structures in a rapid and simple way is an important consideration in bioinformatics, the area of estimating the functions of molecules from their structures or in proteomics, the area of analyzing expressed proteins comprehensively.

In addition to the above-mentioned nucleic acids and proteins, complex carbohydrates such as sugar chains and saccharides are also biomolecules that control the higher regulatory mechanism and the importance of which is being recognized. About 50% or more of the proteins in the living body occur with sugar chains bound thereto and the sugar chains are said to be responsible for controlling the structures and functions of glycoproteins . Further, it has been shown that the interactions the sugar chain recognizing molecules play with glycoproteins are deeply involved in various phenomena of life. A problem with glycoproteins is that given a single kind of protein, the structures of sugar chains that bind to proteins are not uniform and this non-uniformity in sugar chains makes it extremely difficult to know what sugar chains will bind to that protein and how they function.

Recently, in the field of proteins, high-throughput structural analyses are being established and structural analyses are also being established in the field of sugar chains. Techniques for the analytical study of sugar chain structures are represented by, for example, separation of sugar chains in free form, chromatography, electrophoresis, chemical decomposition, NMR, mass spectrometry, enzymatic digestion and analysis of interactions. Among these, mass spectrometry has seen marked advances in recent years . However, the structural analysis of sugar chains by mass spectrometry involves several problems . First , unlike proteins and peptides, sugar chains are not easy to ionize. Second, the only structural information that can be obtained from the mass spectrometry of sugar chains is their molecular weights and the difference between sugar sequences such as the one between hexose and N- acetylhexosamine . The types of constituent sugar residues that have the same molecular weight cannot be identified, making it difficult to determine the whole structure of complexly branched sugar chains and their binding positions Thus, it has been essential to the structural analysis of sugar chains to use large amounts of samples in order to compensate for the low ionization efficiency and to employ techniques other than mass spectrometry in order to determine the whole structure of chain chains. Therefore, in order to achieve high-throughput analysis of sugar chains, it is necessary that the amount of a sugar chain to be measured in mass spectrometry be reduced to less than 1 pmol.

Recently, MALDI-TOF mass spectrometry combining matrix-assisted laser desorption/ionization (MALDI) with the time-of-light (TOF) method is becoming dominant in the structural analysis of sugar chains. In order to measure biomolecules by MALDI, a matrix is required that absorbs the laser energy to help ionize the molecule. Matrices that have heretofore been known for use in ionizing sugar chains include 2,5-dihydroxybenzoic acid (DHB) , arabinoxazone, β-carboline, etc.

As a further problem, sugar chains do not absorb UV radiation to emit fluorescence, so even if they are measured by mass spectrometry, it is difficult to detect them. In addition, when sugar chains are labeled, the step of removing the unreacted labeling reagent is necessary and in order to minimize the loss of the sugar chain sample during this and other steps, the purifying step and the like are preferably absent.

A recently known reagent for fluorescently labeling sugar chains is 2-aminopyridine (PA). In order to label sugar chains fluorescently with PA, a complex process is required that consists of bringing the sugar chains into free form by means of hydrazine or an enzyme and purifying the PA labeled sugar chains. Note that PA has no electric charges and detection of PA labeled sugar chains by MALDI- TOF does not assure high enough sensitivity. The present inventors developed 4-

(biotinamido)phenylacetyl hydrazide (BPH) as a labeling reagent for use in the analysis of sugar chain structures which absorbed UV radiation and had the ability to bind to sugar chains (Shinohara, Y. et al.. Anal. Chem., 68, 2573- 2579, 1996). However, in order to perform mass spectrometry on sugar chains labelled with that reagent, at least 10 pmol of the sample was necessary in a single run. Japanese Patent Application 1998-517295 discloses a method of labeling carbohydrates with a cyanine dye developed for use in biological studies (Japanese Patent Application 2000-504202) and analyzing it by two- dimensional electrophoresis . Other prior art is described in:

Japanese Patent Application 1998-517295 Naven, T.J.P. and Harvey, D.J., Rapid Commun. Mass Spectrom., 10, 829-834 (1996)

Nonami, H. et al., J. Mass Spectrom., 32, 287-296 (1997)

SUMMARY OF THE INVENTION

If trace biomolecules such as nucleic acids, proteins and sugar chains that are found in tissues and cells of the living body can be analyzed by mass spectrometry in a rapid and simple way and using a minimum amount of sample, various diseases that accompany abnormalities in biomolecules can be detected with high sensitivity at high throughput . Therefore, an object of the invention is to provide a method by which the sensitivity of detection of a biomolecule in mass spectrometry can be increased by using a fluorescence labeling agent that has a reactive group capable of binding to the biomolecule and which has electrical charges.

The present inventors found that a biomolecule marked with the fluorescence labeling agent having electric charges was readily ionized by mass spectrometry and could hence be detected with high sensitivity. The method of the present invention has been accomplished on the basis of this finding.

In a first aspect, the present invention provides a method of mass spectrometry of biomolecules comprising analyzing the mass of a biomolecule marked with the fluorescence labeling agent having a reactive group capable of binding to the biomolecule, as well as electric charges.

Preferably the mass spectrometry is tandem mass spectrometry, MS/MS.

In a second aspect, the invention provides a MS reagent comprising a fluorescence labeling agent as used describe above. The invention also relates to use of a reagent comprising a fluorescence labeling agent as above for MS.

In one embodiment, the kit comprises a MS reagent as well as MALDI matrix. Examples of suitable matrices are described below.

In another embodiment, the kit comprises two different labeling agents, differing for example in molecular weight, for differential analysis.

BRIEF DESCRIPTION OF THE DRAWINGS

Fig. 1 shows the result of MALDI-TOF mass spectrometry on 1 pmol, 200 fmol, 40 fmol, 10 fmol, 4 fmol, 2 fmol and 1 fmol samples of chitotriose sugar chain as labelled with Cy3-hydrazide, with MALDI-TOF mass spectrometry being performed by Reflex IV in reflectron mode using norharman as a matrix.

Fig. 2 shows the results of fluorescently labeling ovalbumin-derived N-glycan and evaluating the effect of changing the fluorescent labeling agent and the matrix on the increase of detection sensitivity, panel a showing the result of measurement on unlabelled N-glycan (10 pmol) with Ettan MALDI Pro using DHB as the matrix, and panel b showing the result of performing mass spectrometry on Cy3- hydrazide labeled N-glycan (10 pmol) using CCA as the matrix.

Fig. 3 also shows the results of fluorescently labeling ovalbumin-derived N-glycan and evaluating the effect of changing the fluorescent labeling agent and the matrix on the increase of detection sensitivity, panel a showing the result of measurement on Girard's labeled N- glycan (30 pmol) with Reflex IV using DHB as the matrix, panel b showing the result of measurement on Cy3-hydrazide labeled N-glycan (400 fmol) with Reflex IV using DHB as the matrix, and panel c showing the result of measurement on Cy3-hydrazide labeled N-glycan (200 fmol) with Reflex IV using norharman as the matrix.

Fig. 4 shows the result of PSD analysis on Cy-5 labeled lactose-neo-fucopentaose (20 pmol) by MALDI-TOF, with CCA being used as the matrix and Ettan MALDI Pro as the mass spectrometer.

Fig. 5 shows the result of ISD analysis on Cy-5 labeled lactose-neo-fucopentaose (2 pmol) by MALDI-TOF, with CCA being used as the matrix and Reflex IV as the mass spectrometer.

Fig. 6 shows the results of investigating the effect of removing the fluorescence labeling agent that remained after labeling; a reaction solution of a Cy labeled sugar chain (m/z 1514) was passed through a ZiPTip C18 column to trap the fluorescently labeled sugar chain, which was not eluted with 50% acetonitrile (CAN) (the upper panel of Fig. 6) but was eluted with water (the lower panel of Fig. 6 ) .

DETAILED DESCRIPTION OF THE INVENTION

On the following pages , preferred embodiments are described in detail for further illustrating the present invention. (1) Mass spectrometry The present invention provides a method of mass spectrometry of biomolecules comprising analyzing the mass of a biomolecule marked with a fluorescence labeling agent that has a reactive group capable of binding to the biomolecule and a positive total charge of from +1 to +6 or a negative total charge of from -6 to -1.

Mass spectrometry is an analytical method chiefly intended for measuring the mass of a sample using a vapor- phase ion spectrometer comprising a sample introducing section, an ionizing section, a mass separating section and a detecting section.

Methods of ionizing biomolecules are not limited in any particular way and may be exemplified by matrix- assisted laser desorption/ionization (MALDI), laser desorption (LD), fast atom bombardment (FAB), liquid secondary ion mass spectrometry (LSIMS) , liquid ionization (LI), electrospray ionization (ESI), and atmospheric pressure chemical ionization (APCI). Since LD does not use matrices, the biomolecules themselves need to absorb the applied laser beam to become ionized. In MALDI, it is necessary to choose an applicable matrix in accordance with the wavelength of the laser light to be applied. Therefore, as long as the matrix absorbs the laser light, the biomolecules themselves need not absorb the laser light and many kinds of biomolecule can be ionized.

According to the invention, MALDI, ESI or APCI is preferably employed as an ionizing method in spectrometry of biomolecules and MALDI is more preferred. The matrix to be used in ionizing biomolecules differs with the type of laser to be used in ionization and may be exemplified by the following compounds .

In the analysis of proteins, peptides and glycopeptides , preferred matrices are nicotiniσ acid, sinapic acid, 2,5-dihydroxybenzoiσ acid, 5-methoxysalicylic acid, α-cyano-4-hydroxyσinnamic acid (αCHCA), diaminonaphthalene, succinic acid, 5-

(trifluoromethyl)uraσil, etc. are preferred. Particularly preferred is 2,5-dihydroxybenzoic acid since it absorbs the energy of uv radiation, works as a proton donor and still tends to mix uniformly with polar substances.

In the analysis of sugars, sugar chains, glycolipids, etc., norharman, 2,5-dihydroxybenzoic acid, αCHCA (sometimes referred to as CCA), arabinoxazone, THAP, etc. are preferred. More preferred are norharman, DHB and arabinoxazone. Further preferred are norharman and DHB. A preferred applicable laser is a pulsed nitrogen laser (337 nm) , which heats the biomolecule so rapidly that the latter can be ionized without being decomposed. Such pulsed ionization is a kind of thermal ionization and even thermally unstable proteins having molecular weights in excess of 100 kDa can be desorbed and ionized without being decomposed.

Ionization of biomolecules by MALDI generally comprises the steps of mixing a biomolecule with a matrix (solution) at a molar ratio of 1 x 10^"2 - 5 x 10^"4:1, drying the mixed solution and causing it to crystallize. Upon applying pulsed laser to the crystal, ion species such as [M]⁺, [M+H]⁺, [M+Na]⁺ and [M+K]⁺ that are derived from the biomolecule and those derived from the matrix are desorbed. In the present invention, it is preferred that only the ion species [M]⁺ is desorbed. Ionization usually gives rise to three kinds of ion but by using the preferred fluorescence labeling agent of the invention, only the single ion species [M]⁺ can be desorbed, bringing about the advantage of improving sensitivity and reducing the complexity of spectra. More specifically, sample preparation and measurement can be performed on the basis of the description in Example 1 to be given later.

The mass separating section is a section (or device) in which the ions of the biomolecule ionized in the ionizing section are separated in accordance with the characteristic mass/charge (m/z) of the biomolecule by means of an electromagnetic interaction. The mass separating section may be exemplified by but are not limited to time-of-flight (TOF), quadrupole ion trap time- of-flight (QIT-TOF), quadrupole, ion trap, magnetic field, Fourier transform ion cyclotron resonance types. In the present invention, TOF, QIT-TOF and FT-ICR are preferred, with TOF being more preferred. While the method of the invention can be implemented by combining in any way various kinds of ionizing and mass separating sections, it is preferably implemented with a MALDI-TOF mass spectrometer which combines MALDI with the time-of-flight (TOF) method of analysis. The MALDI-TOF mass spectrometer to be used in the invention may be exemplified by Ettan MALDI-TOF Pro (Amersham Bioscience) and Reflex IV (Bruker) .

However, even if either the MALDI-TOF mass spectrometer or the MALDI-QIT-TOF mass spectrometer is employed with either one of the matrices mentioned above, the measurement of sugar chains, phosphorylated peptides, glycopeptides, glycoproteins and other samples that are difficult to ionize often results in a failure to provide sufficient sensitivity in the analysis of mass spectra. Hence, in order to separate/detect biomolecules that are difficult to ionize, a charged fluorescence labeling agent to be described later is employed to label (or derivatize) the biomolecule preliminarily, thereby enhancing the efficiency of ionization.

(2) Fluorescence labeling agent

The fluorescence labeling agent to be used in the method of the invention is characterized by having a reactive group capable of binding to biomolecules and also having charges , in particular a positive total charge of from +1 to +6 or a negative total charge of from -6 to -1. Specifically, in the fluorescence labeling agent to be used in the method of the invention, the reactive group capable of binding to biomolecules is bound to the building structure either with or without an intervening linker molecule and, taken as a whole, the labeling agent is charged either positively or negatively.

The building structure of the fluorescence labeling agent to be used in the method of the invention is exemplified by cyanine, merocyanine, pseudocyanine, isocyanine, styryl, naphthalene, anthracene, phenanthrene, pyrene, acridine, indole, benzoxazole, quinoline, isatoic anhydride, 1,8-naphthalimide, 2,3-naphthalimide, 5-phenyl- 4-pyridyl-2-oxazole, coumarin, bimane, 6-N-(7-nitrobenz-2- oxa-1 , 3-diazol-4-yl)amine, 4 , 4-difluoro-5 , 7-dimethyl-4- bora-3a,4a-diaza-S-indacene, ethidium chloride, propidium chloride, fluorescein, resorufine, Nile Blue A, rhodamine (e.g. dihydrotetramethyl rhodamine dihydro-X-rhodamine, Texas Red (registered trademark) and Alexa Fluor

(registered trademark)), benzimidazole and pyridine dyes. The fluorescence labeling agent preferably has cyanine, merocyanine or styryl as the building structure. Cyanine is more preferred. Note that anybody skilled in the art can use conventional techniques to prepare derivatives from compounds having those building structures.

If the building structure has no charges, conventional techniques may be employed to prepare derivatives having charges in their building structure or, alternatively, a charged atom or molecule (or group) may be bound as, for example, a linker molecule. Specifically, a charged atom such as nitrogen, oxygen, sulfur or phosphorus or a molecule (or group) having at least one such charged atom may be introduced into the building structure or linker molecule so as to adjust the total charge on the fluorescence labeling agent.

The total charge on the fluorescence labeling agent is not limited to any particular value as long as it has a reactive group capable of binding to biomolecules and the mass of a biomolecule marked with that fluorescence labeling agent can be analyzed.

The total positive charge on the fluorescence labeling agent is preferably from +1 to +6, more preferably from +1 to +3, still more preferably from +1 to +2, and most preferably it is +1.

The total negative charge on the fluorescence labeling agent is preferably from -6 to -1, more preferably from -3 to -1, still more preferably from -2 to -1, and most preferably it is -1.

The term "reactive group capable of binding to biomolecules" as used herein means a group that can bind the fluorescence labeling agent to a desired biomolecule. To such an extent that the binding of the reactive group to the biomolecule is not impaired and that it does not affect the structure and function of the biomolecule, the reactive group may be bound directly to the building structure of the fluorescence labeling agent or it may be bound to the building structure via the linker molecule. Either one or more of such reactive groups or, alternatively, either one kind or two or more kinds in combination may be bound to the building structure or the linker molecule.

If the building structure of the fluorescence labeling agent has charges, the reactive group or the moiety consisting of the reactive group and the linker molecule may have no charges, or zero charges. For example, in the case where the building structure of the fluorescence labeling agent is cyanine or styryl, the nitrogen in the building structure is charged to +1. Therefore, the reactive group or the moiety consisting of the reactive group and the linker molecule that are bound to the building structure may have zero charges. On the other hand, in the case where the building structure is merocyanine, the nitrogen in the building structure is not charged. Therefore, the reactive group or the moiety consisting of the reactive group and the linker molecule that are bound to the building structure has a positive charge of +1 to +6 or a negative charge of -6 to -1. If the reactive group is directly bound to the building structure, its binding position on the building structure is not limited. Preferably, the binding position is at the carbon, nitrogen, oxygen, sulfur or phosphorus that are constituents of the building structure. More preferably, the binding position is at nitrogen, carbon or oxygen. Most preferably, the binding position is at nitrogen.

If the reactive group is bound to the linker molecule, its binding position is not limited. A preferred binding position is such that at least one reactive group binds to a terminal end of the linker molecule.

In a preferred embodiment, the reactive group is exemplified by primary amine, secondary amine, hydrazine, hydroxylamine, pyrazolone, sulfhydryl, carboxyl, hydroxyl, thiophosphate, imidazole, aldehyde, ketone, isocyanate and hydrazide. Preferred reactive groups are hydrazide, amine and hydroxylamine. More preferred reactive groups are hydrazide and hydroxylamine. The most preferred reactive group is hydrazide.

The linker group is a molecule for linking the building structure to the reactive group. In the case of using an uncharged building structure such as naphthalene, the linker molecule can be charged by introducing a charged atom or molecule (or group). The linker molecule is preferably a molecule that can retain the charge on the fluorescence labeling agent or the charge generated by ionizing the biomolecule and which does not quench fluorescence.

In a preferred embodiment, the linker molecule can contain 1-60 chained atoms selected from the group consisting of carbon, nitrogen, oxygen, sulfur and phosphorus, as exemplified by: (a) -(CH₂)_X-; (b) -((CH₂)_p-CO-)_y-;

(c) -((CH₂)_p-0-(CH₂)_g)_y-;

(d) -((CH₂)_p-CONH-(CH₂)_g)_y-; or

(e) -((CH₂)_p-Ar-(CH₂)_g)_y- where x is 1-30, preferably 1-10; p is 1-10, preferably 1-5; q is 0-10, preferably 0-5; y is 1-10, preferably 1-5; and Ar is aryl.

Preferably, the linker molecule is -( (CH₂)_p-CO-)_y- (where p is 1-5 and y is 1-5).

More preferably, the linker molecule is -(CH₂)₅-C(=0)

In a preferred embodiment, the fluorescence labeling agent to be used in the method of mass spectrometry according to the invention is selected from the group consisting of the following compounds: cyanine

merocyanine

styryl

where m is an integer selected from the group consisting of 0, 1, 2, 3 and 4;

X and Y are selected independently from O, S and C(R₈)_p (where p is 1 or 2 and R₈ is H or a C₁-C₄ alkyl); at least one of groups R_x to R₉ contains a reactive group having 0 to 6 positively charged nitrogen, phosphorus or sulfur atoms or 0 to 6 negatively charged oxygen or sulfur atoms; and the dashed lines represent the carbon atoms necessary to form said cyanine, merocyanine and styryl dyes.

Alternatively, the nitrogen-containing ring in each of the above compounds may be such that the carbon atoms as constituents of the dashed line combine with the nitrogen and X or Y to form a saturated or unsaturated five- or six- membered ring.

Preferably, the reactive group contained in at least one of groups R_x to R₉ and which has 0 to 6 positively charged nitrogen, phosphorus or sulfur atoms or 0 to 6 negatively charged oxygen or sulfur atoms is selected from the group consisting of primary amine, secondary amine, hydrazine, hydroxylamine, pyrazlone, sulfhydryl, carboxyl, hydroxyl, thiophosphate, imidazole, and carbonyl containing aldehyde and ketone.

The reactive group contained in groups R_x to R₉ which has 0 to 6 positively charged nitrogen, phosphorus or sulfur atoms or 0 to 6 negatively charged oxygen or sulfur atoms is preferably present at one or two locations, more preferably at one location in groups R_x to R₉.

Preferably, group Rχ/R₂ has the reactive group.

More preferably, at least one of groups R_x to R₉ that contains the reactive group having 0 to 6 positively charged nitrogen, phosphorus br sulfur atoms or 0 to 6 negatively charged oxygen or sulfur atoms is - (CH₂)₅~C(=0)- NHNH₂.

In a more preferred embodiment, the fluorescence labeling agent is not limited in any particular way except for one of groups ^ to R₉ that contains the reactive group having 0 to 6 positively charged nitrogen, phosphorus or sulfur atoms or 0 to 6 negatively charged oxygen or sulfur atoms . Preferably, groups R_x to R₉ which are other than the reactive group having 0 to 6 positively charged nitrogen, phosphorus or sulfur atoms or 0 to 6 negatively charged oxygen or sulfur atoms are selected independently from the group consisting of hydrogen, isothiocyanate, isocyanate, monochlorotriazine, dichlorotriazine, mono- or di-halogen substituted pyridine, mono- or di-halogen substituted diazine, maleimide, aziridine, sulfonyl halide, acid halide, hydroxysuccinimide ester, hydroxysulfosuccinimide ester, imide ester, hydrazine, azidonitrophenyl, azide, 3- (2- pyridyldithio)propionamide, glyoxal and aldehyde.

In a still more preferred embodiment, the fluorescence labeling agent is a cyanine derivative having chemical formula ( I ) : ¹

(I) where

X and Y are each C(CH₃)₂; n is an integer of 1 or 2 ; group R_j. is - (CH₂)₅-C(=0)-NHNH₂; group R₂ is -(CH^CH_a (where m is an integer of 0 to 3 ) ; and group R₁₀ and R_u are each H.

Preferably, n and m in the cyanine derivative are each unity. In the present specification, the cyanine derivative having this definition is sometimes referred to as Cy3-hydrazide.

Preferably, n and m in the cyanine derivative are 2 and 0, respectively. In the present specification, the cyanine derivative having this definition is sometimes referred to as Cy5-hydrazide.

Preferably, the cyanine derivative is N- hydrazinocarbonylalkyl-pseudocyanine. More preferably, the cyanine derivative is 1-ethyl-l' -(4- hydrazinocarbonylbutyl) - 2,2' -cyanine.

The fluorescence labeling agent that can be used in mass spectrometry according to the invention may use the cyanine dyes disclosed in Japanese Patent Application 2002- 504202 which is incorporated herein by reference in the case of producing those cyanine dyes. (3) Fluorescent labeling and mass spectrometry of biomolecules (a) Biomolecules The origin, method of production, etc. of biomolecules are in no way limited as long as they can be measured by the method of the invention. Thus, "biomolecules" as appearing in the specification of the subject application may be either natural products or chemical synthetic products. Specifically, they can be derived from biological materials and are extracts from organs, tissues or cells, as exemplified by sugars, sugar chains , proteins , peptides , phosphorylated peptides , nucleic acids , glycoproteins , glycopeptides and glycolipids . The biomolecules preferably have functional groups that interact with the reactive group in the above-described fluorescent labeling agent such that the latter can bind to the biomolecules . Preferred functional groups in the biomolecules are hydroxyl, carboxyl, carbonyl, amino and sulfhydryl groups. More preferred examples are hydroxyl, carboxyl and carbonyl groups . The most preferred is the carbonyl group. Biomolecules that can be subjected to mass spectrometry by the method of the invention are those which are difficult to ionize, as exemplified by sugar chains, phosphorylated peptides and glycopeptides . Sugar chains are preferred. Sugar chains include not only monosaccharides, oligosaccharides , and polysaccharides contained in the living body; they also include sugar chains derived from complex carbohydrates such as glycoproteins , proteoglycans , glycosaminoglyσans and glycolipids. Monosaccharides include glucose, galaσtose, mannose, fucose, N- acetylglucosamine, N-acetylgalactosamine, xylose, glucuronic acid and iduronic acid. Oligosaccharides are composed of one or more kinds of monosaccharide that are joined together as 2 to 10-odd units and may be exemplified by lactose, sucrose, σhitotriose, NA2 and NGA3.

Polysaccharides are composed of a larger number of monosaccharide units than oligosaccharides . In both the oligosaccharide and the polysaccharide, the constituent monosaccharide units (sugar residues) are joined together by - or β-glycosidic bonds. Symbols α and β refer to the anomeric carbon with a glycosidic bond in position 1 on the sugar ring and the anomeric carbon that is trans in positional relationship to CH₂OH or CH₃ in position 5 is labelled . whereas the one that is cis is labelled β. Among the constituent sugar residues, those in which the anomeric carbon is not bound to any other sugar residues are called reducing ends and those which are bound to other sugar residues only by means of the anomeric carbon are referred to as non-reducing ends . Oligosaccharides have only one reducing end whereas one or more non-reducing ends occur depending upon the degree of branching of the oligosaccharide. For fluorescent labeling of the oligosaccharide, the hydroxyl group on the anomeric carbon at the reducing end is reacted with the reactive group in the fluorescence labeling agent, which is thereby bound to a specific position in the oligosaccharide. (b) Fluorescent labeling Using the above-described fluorescence labeling agent, one can fluorescently label any biopolymers and fluorescent labeling can be performed by conventional techniques.

Specifically, if the biomolecule is composed of amino acids as in the case of proteins and peptides, the fluorescence labeling agent can be bound to hydroxyl, carboxyl, amino, sulfhydryl and other groups in side chains of amino acids by conventional techniques.

In the case of fluorescently labeling complex carbohydrates, free hydroxyl groups at the reducing end of the sugar chain can be utilized. In the case of mono-, oligo- and polysaccharides, preliminary treatments need not be performed since the reducing end is in a free state. In the case of complex carbohydrates other than those saccharides, their reducing end may be brought into free form by preliminary treatments using known methods including chemical techniques such as hydrazine decomposition/N-acetylation, alkali treatment, trifluoroacetolysis and ozonolysis, or treatments with enzymes such as glycopeptidase, endo glycosidase and glycoselamidase . More specifically, as described in Example 1 later in this specification, a sugar chain is mixed with the fluorescence labeling agent and the mixture is heated, whereby the sugar chain is fluorescently labelled. According to the present invention, the fluorescently labelled sugar chain can be subjected to mass spectrometry without being purified, (c) Mass spectrometry According to the present invention, biomolecules that are generally held to be difficult to ionize, such as sugar chains, phosphorylated peptides and glycopeptides, can be subjected mass spectrometry. As is well known in the art of interest, the measurement of biomolecules by mass spectrometry requires at least 10 pmol of the biomolecule. If such biomolecules are to be subjected to mass spectrometry by the method of the invention, the required amount of the biomolecule is no greater than 1 pmol, preferably no greater than 0.1 pmol, more preferably no greater than 0.01 ppm, still more preferably no greater than 1 fmol, and most preferably no greater than 0.25 fmol. Therefore, according to the present invention, the sensitivity of mass spectrometry of biomolecules is increased by 10 times, preferably 100 times, more preferably 1000 times, still more preferably 10000 times, and most preferably 40000 times, over the value achieved by the conventional technique which does not use the fluorescence labeling agent. Although not wishing to be bound by theory, the present inventors postulate that the increased sensitivity of mass spectrometry according to the invention is due to the enhancement of percent ionization that is achieved by marking biomolecules with the speci ied fluorescence labeling agent.

As specifically shown in Example 2 , one only needs to use 1 fmol of a sugar chain to obtain mass spectra by analyzing it using MALDI-TOF MS. In the conventional method which does not use the fluorescence labeling agent, it has been held that 10 pmol of a sugar chain is required for mass spectrometry. Thus, according to the invention, the sensitivity of mass spectrometry of sugar chains is increased by 10000 times.

According to the method of the invention, the analysis of biomolecules using MALDI-TOF MS is not limited to the acquisition of information about their mass and other kinds of information such as about the compositions, sequences and structures of biomolecules can also be obtained by PSD (post-source decay) analysis, ISD (in- source decay) analysis and tandem mass analysis.

In this specification, PSD analysis means a technique that provides detailed information about the composition of a biomolecule by detecting excited molecular ions that have been fragmented as they fly through a TOF mass spectrometer, As is well known in the art of interest, fragments refer to the constituents of a biomolecule that result from the cleavage of intermolecular bonds such as peptide and glycoside bonds by the laser energy, or to molecular units composed of two or more of such constituents bound together. By PSD analysis, one can obtain information such as about the composition of an unknown sample or a biomolecule the constituents of which are unknown. Example 4 specifically shows that by PSD analysis of a Cy5 labelled lacto-neo- fucopentaose, one can obtain the information about the types of the sugar residues that compose the sugar chain and about its primary structure.

In this specification, ISD analysis means a technique of TOF mass spectrometry in which a biomolecule is illuminated with a higher energy of laser and it is simultaneously fragmented, and the resulting excited molecular ions are detected to give information such as about the composition of the biomolecule from an even smaller amount of sample than is required in PSD analysis. Example 5 specifically shows that using a tenth of the sugar chain employed in PSD analysis, one can obtain the information about the types of the sugar residues that compose the sugar chain and about its primary structure.

According to the method of the invention, once a biomolecule is marked with the above-described fluorescence labeling agent, there is no need to remove the unreacted labeling agent and one may immediately introduce the sample into a mass spectrometer and perform measurement. If one wants to remove the unreacted fluorescence labeling agent and use a purified, fluorescently labelled biomolecule, one may employ a conventional purification technique such as chromatography. As specifically shown in Example 6, a fluorescently labelled sugar chain is trapped by a ZipTip C18 column carrier (Millipore) and can be easily separated with water to prepare a purified sample.

According to the method of the invention, using two or more fluorescence labeling agents that differ in fluorescence wavelength and/or molecular weight, one can perform differential analysis or imaging of biomolecules by mass spectrometry.

EXAMPLES

On the pages that follow, the invention is described more specifically by reference to examples . It should however be noted that the invention is by no means limited by the following examples. Example 1 Fluorescent Labeling of Sugar Chain

A portion (1 μl) of 25 mM chitotriose and 2 μl of 50 mM Cy3-hydrazide in acetonitrile was added to 17 μl of 30% acetonitrile in a mini-vial. The mixture was heated on a heat block at 90 °C for 1 hour. After being cooled to room temperature, the mixture was diluted as appropriate and analyzed with a MALDI-TOF mass spectrometer without being purified. Example 2 Checking the Detection Limit

Using a chitotriose sugar chain labelled with Cy3- hydrazide, an experiment was conducted to determine the minimum amount of the sample to be measured (see Fig. 1). The Cy3-labelled sugar chain was diluted to 1 pmol, 200 fmol, 40 fmol, 10 fmol, 4 fmol, 2 fmol and 1 fmol and measured by MALDI-TOF in reflectron mode. The mass spectrometer was Reflex IV (Bruker) and the matrix was norharman. With 1 fmol which was the smallest amount of all samples that were prepared, the signal/noise (S/N) ratio was no less than 3 (the top line in Fig. 1).

Therefore, measurement would be possible even with samples of the labelled sugar chain in amounts much smaller than 1 fmol. The required amount would be further reduced in linear mode. Considering that at least 10 pmol of sugar chains have heretofore been necessary in order to perform mass spectrometry and detect satisfactory mass spectra, it can be said that the method of the present invention increased the sensitivity of mass spectrometric detection by 10000 times. Example 3 Increased Detection Sensitivity

Ovalbumin-derived N-glycan was labelled with Cy3- hydrazide or Girard' s T and subjected to MALDI-TOF mass spectrometry (Fig. 2). Fig. 2a shows the result of an experiment with a control, in which N-glycan (10 pmol) which was not fluorescently labelled was subjected to mass spectrometry with Ettan MALDI Pro (Amersham Bioscience) using 2,5-dihydroxybenzoic acid (DHB) as a matrix. No mass spectra could be detected. On the other hand, the mass spectrometry conducted on Cy3-hydrazide labelled N-glycan using α-CHCA (CCA) as a matrix gave a mass spectrum characteristic of ovalbumin (Fig. 2b) .

Fig. 3a shows the result of an experiment in which N- glycan (30 pmol) labelled with Girard's T was subjected to mass spectrometry with Reflex IV using DHB as a matrix. Fig. 3b shows the result of an experiment in which N-glycan (400 fmol) labelled with Cy3-hydrazide was subjected to mass spectrometry with Reflex IV using DHB as a matrix. A mass spectrum was obtained in the experiment of labeling with Girard's T but the S/N ratio was low (Fig. 3a). In the experiment of labeling with Cy3-hydrazide, the amount of N-glycan was reduced to about 1/100 and yet there was obtained a mass spectrum having high enough S/N ratio (Fig. 3b) . When norharman was used as a matrix, an even better mass spectrum was obtained using only 200 fmol (Fig. 3c) .

The foregoing data shows that by labeling sugar chains with Cy3-hydrazide, the S/N ratio and, hence, the detection sensitivity, could be increased. Example 4 PSD Analysis in Mass Spectrometry

PSD analysis was performed using laσto-neo- fucopentaose (20 pmol) labelled with Cy5-hydrazide (see Fig. 4 ) . The primary structure of this sugar chain was Fuc l-2Galβl-3GlcNAcβl-3Galβl- 4Glc. Measurement was conducted with Ettan MALDI Pro using CCA as a matrix. Upon illumination with laser, the sugar chain was cleaved starting at the non-reducing end to produce fragments. Since the sugar residue at the reducing end of the sugar chain was fluorescently labelled, only fragments having the labelled sugar residue at the reducing end were detected as mass spectra (Fig. 4). Starting from the lower molecular end of Fig. 4, Cy5-hydrazide (481.742 m/z) and the labelled sugar residue (658.5 m/z) were detected as spectra, with consecutive sugar residues being added one by one. Between adjacent spectra were constituent sugar residues; Hex (hexose), for. example, corresponds to the galactose (Gal) second from the reducing end of lacto-neo-fucopentaose, and dHex (deoxyhexose) corresponds to the fucose at the non- reducing end. Hence, the method of the present invention is suitable for PSD analysis and may well be considered as an effective way to determine the kinds of constituent sugar residues of a sugar chain and its primary sequence. Example 5 ISO Analysis in Mass Spectrometry

ISD analysis was performed using lacto-neo- fucopentaose (2 pmol) labelled with Cy5-hydrazide (Fig. 5). Measurement was conducted with Reflex IV using CCA as a matrix. In Fig. 5, HexNAc (N-acetylhexosamine) corresponds to N-acetylglucosamine third from the reducing terminal end of the sugar chain and the fourth Hex (hexose) corresponds to galactose (Gal) . Hence, the method of the present invention is also suitable for ISD analysis and having to use a smaller amount of the sample than the PSD detection described in Example 4, it may well be considered equally effective in determining the kinds of constituent sugar residues of a sugar chain and its primary sequence. Example 6 Purification of Labeled Sugar Chain

The effect of removing the fluorescence labeling agent that remained after labeling was investigated and the results are shown in Fig. 6. In Examples 1-5, the labeled sugar chains were subjected to mass spectrometry without being purified. The method of the invention is also effective in the case of using a purified, fluorescently labeled sugar chain that has been cleared of any unreacted labeling agent. Stated specifically, when a reaction solution of a sugar chain (1514 m/z) marked by the fluorescent labeling method described in Example 1 was passed through a ZiPTip C18 column, the fluorescently labeled sugar chain was trapped on the column and was not eluted with 50% acetonitrile (CAN) (the upper panel of Fig. 6) but eluted with water (the lower panel of Fig. 6). Therefore, according to Example 6, the unreacted fluorescence labeling agent can be easily removed to purify the labeled sugar chain. Advantages of the Invention

The present invention helped increase the sensitivity of separation and/or detection of biomolecules by mass spectrometry. As a result, rapid and simple diagnosis of various diseases became possible by detecting very small amounts of biomolecules.

Claims

1. A method for mass spectrometry of biomolecules comprising analyzing the mass of a biomolecule marked with a fluorescence labeling agent that has a reactive group capable of binding to the biomolecule and a positive total charge of from +1 to +6 or a negative total charge of from -6 to -1.

2. The method according to Claim 1, wherein the mass spectrometry comprises MS/MS.

3. The method according to Claim 1 or 2, comprising electrospray MS.

4. The method according to Claim 1 or 2 , further comprising ionizing the fluorescently labeled biomolecule by matrix assisted desorption/ionization (MALDI). 5. The method according to any one of Claims 1-4, further comprising separating/detecting ions by the time- of-flight (TOF) or quadrupole ion trap time-of-flight (QIT- TOF) method. 6. The method according to any one of Claims 1-5, wherein the fluorescence labeling agent has a building structure selected from the group consisting of cyanine, merocyanine, pseudocyanine, isocyanine, styryl, naphthalene, anthracene, phenanthrene, pyrene, acridine, indole, benzoxazole, quinoline, isatoic acid anhydride, 1,8- naphthalimide , 2,3-naphthalimide,

5-phenyl-4-pyridyl-2- oxazole, coumarin, bimane,

6-N-(7-nitrobenz-2-oxa-l,3- diazol- 4-yl)amine, 4,4-difluoro-5, 7-dimethyl-4-bora-3a, 4a- diaza- S-indacene, ethidium chloride, propidium chloride. fluorescein, resorufine, Nile Blue A, rhodamine, benzimidazole and pyridine.

7. The method according to Claim 6 , wherein the fluorescence labeling agent has a building structure selected from the group consisting of cyanine, merocyanine and styryl.

8. The method according to Claim 7 , wherein the fluorescence labeling agent is selected from the group consisting of the following compounds : cyanine

merocyanine

styryl

where m is an integer selected from the group consisting of 0, 1, 2, 3 and 4;

X and Y are selected independently from 0, S and C(R₈)_P (where p is 1 or 2 and R₈ is H or a C_x-C₄ alkyl); at least one of groups R_x to R₉ contains a reactive group having 0 to 6 positively charged nitrogen, phosphorus or sulfur atoms or 0 to 6 negatively charged oxygen or sulfur atoms; and the dashed lines represent the carbon atoms necessary to form said cyanine, merocyanine and styryl dyes.

9. The method according to Claim 8 , wherein the reactive group contained in at least one of groups R-_L to R₉ and which has 0 to 6 positively charged nitrogen, phosphorus or sulfur atoms or 0 to 6 negatively charged oxygen or sulfur atoms is selected from the group consisting of primary amine, secondary amine, hydrazine, hydroxylamine, pyrazolone, sulfhydryl, carboxyl, hydroxyl, thiophosphate, imidazole, and carbonyl containing aldehyde and ketone.

10. The method according to Claim 8 or 9, wherein groups -_L to R₉ which are other than the reactive group having 0 to 6 positively charged nitrogen, phosphorus or sulfur atoms or 0 to 6 negatively charged oxygen or sulfur atoms are selected independently from the group consisting of hydrogen, isothiocyanate, isocyanate, monochlorotria- zine, dichlorotriazine, mono- or di-halogen substituted pyridine, mono- or di-halogen substituted diazine, maleimide, aziridine, sulfonyl halide, acid halide, hydroxylsuccinimide ester, hydroxyl-sulfosuccinimide ester, imide ester, hydrazine, azidonitrophenyl, azide, 3- (2- pyridyldithio)propionamide, glyoxal and aldehyde.

11. The method according to any one of Claims 8-10, wherein at least one of groups R_x to R₉ that contains the reactive group having 0 to 6 positively charged nitrogen, phosphorus or sulfur atoms or 0 to 6 negatively charged oxygen or sulfur atoms is - (CH₂)₅-C(=0) -NHNH₂.

12. The method according to any one of claims 6-11, wherein the fluorescence labeling agent is a cyanine derivative having chemical formula (I):

(I) where

X and Y are each C(CH₃)₂; n is an integer of 1 or 2; group R_x is - (CH₂)₅-C(=0) -NHNH₂; group R₂ is -(CH₂)_mCH₃ (where m is an integer of 0 to 3); and group R₁₀ and R_n are each H.

13. The method according to any one of Claims 1-12, wherein the biomolecule is selected from the group consisting of sugars, sugar chains, proteins, peptides, phosphorylated peptides, nucleic acids, glycoproteins, glycopeptides , and glycolipids .

14. A MS reagent comprising a fluorescence labeling agent as used in any one of Claims 8-13.

15. A kit comprising a the MS reagent according to Claim 14, and MALDI matrix.

16. The kit according to Claim 15, comprising the labeling agent used in Claim 12.

17. The kit according to Claim 15 or 16, comprising two different labeling agents for differential analysis .