WO2013064554A1

WO2013064554A1 - Dimetallic protecting group for phosphorylated peptides and proteins

Info

Publication number: WO2013064554A1
Application number: PCT/EP2012/071597
Authority: WO
Inventors: Christine J. MCKENZIE; Frank Kjeldsen; Simon SVANE
Original assignee: Syddansk Universitet
Priority date: 2011-11-02
Filing date: 2012-10-31
Publication date: 2013-05-10

Abstract

There is provided a novel dimetallic gallium complex capable of capturing a substance having an anionic substituent, a deactivating agent for phosphorylated substance, an additive for mass spectrometry, an additive for electrophoresis, an additive for nuclear magnetic resonance, and an additive for chromatography, a method for preparing a the gallium complex, and a method for analyzing a substance having an anionic substituent, each utilizing the gallium complex.

Description

Dimetallic Protecting Group for Phosphorylated Peptides and Proteins

FIELD OF THE INVENTION

The present invention relates to a novel gallium complex capable of capturing a substance having an anionic substituent, a deactivating agent for phosphorylated substance, an additive for mass spectrometry, an additive for electrophoresis, an additive for nuclear magnetic resonance, and an additive for chromatography, a method for preparing a the gallium complex, and a method for analyzing a substance having an anionic substituent, each utilizing the gallium complex.

BACKGROUND OF THE INVENTION

Phosphorylation of proteins plays a significant role in various diseases.

Specifically, progress and stop of cell cycle are controlled by phosphorylation or dephosphorylation of various enzymes, i.e. proteins. Cycline and cycline-dependent kinases (CDK) are relevant factors in the phosphorylation. In cases where the mechanism relating to cycline and CDK is impaired, phosphorylation or dephosphorylation may be uncontrollable, thereby abnormal proliferation of cells is triggered.

One of the key objectives of today's proteomics research is focused on the extent, type and localization of PTMs in proteins. PTMs are chemical alterations (e.g. phosphorylations or glycosylations) to the primary protein structure (order of amino acid residues). Especially, protein phosphorylation has court considerable attention in the scientific and clinical society since this type of PTM is involved in many important biological processes such as signal transduction, cell division, gene expression, cytoskeletal regulation and metabolic maintenance. Phosphorylation of naturally occurring proteins belongs to one of the most abundant type of PTMs known. In the characterization of proteins, including sequencing and identification, mass spectrometry (MS) has evolved to become the preferred method. In particular, tandem mass spectrometry (MS/MS) is an indispensable method for large-scale identification and quantification studies of proteins as well as for the characterization of protein phosphorylations.

The conventional MS/MS approach of protein characterization (the bottom-up approach) consists of enzymatic digestion of the protein(s) with trypsin and subsequent separation of the peptide mixtures by liquid chromatography (LC). The LC system is typically interfaced to electrospray ionization (ESI) which allows for gas-phase fragmentation in the mass spectrometer by collisions with an inert collision gas (this traditional fragmentation technique is termed collision-activated dissociation, CAD). The masses of the fragment ions and the molecular ion of each peptide are then entered into a computer database, which compares the measured fragmentation pattern with theoretical fragmentation patterns of the virtual digest peptides of all the proteins in a protein sequence database. In that way, peptide candidates are assigned a probability score for positive identification. The success of this method depends on how many fragment ions are formed by MS/MS of the peptides and the characteristics of these fragment ions as well as how many same mass (within a mass tolerance window) peptide candidates they have to be compared against. However, the result of this methodology is often poor.

While sequencing of non-modified peptide ions with CAD MS/MS is becoming a routine, phosphopeptide analysis still remains a challenge. This is mainly a result of significant instability in CAD of the phosphate ester bond compared to those of peptide backbone bonds. This can be rationalized by evaluation of the activation barrier for cleavage of the amide backbone bonds (ca. 40 kcal/mol) versus that of the phosphate ester bond (<20 kcal/mol). As a consequence, the phosphate ester groups are prone to facile detachment (e.g. as H3P04) from phosphopeptide ions hampering cleavages of sequence specific backbone bonds and thus complicate reliable determination of the phosphorylation site and peptide identification. Until now, only fragmentation techniques utilizing radical induced fragmentation such as electron capture dissociation (ECD) and electron transfer dissociation (ETD) are capable of producing sequence specific backbone cleavages without notable detachment of labile modifications. Despite this important feature, the techniques of ECD/ETD have not yet become the preferred choice in large-scale phosphopeptide analysis. This is most likely due to a significantly lower fragmentation yield of tryptic peptides in ECD/ETD compared to that of CAD.

US2005038258A1 discloses Zn-complexes capable of capturing substance having anionic substituent. Meanwhile, such complexes do not selectively capture phosphate groups, but cross-react with many other anions, including carboxylic acid groups. This kind of cross reactivity is undesirable when capturing substances in e.g. complex biological samples. Hence, there is a need for an efficient and selective capturing agent or protecting group for phosphorylated compounds, in particular for phosphorylated proteins. Moreover there is a need for capturing agent, which also works as a protective group for phosphorylated proteins being analyzed with CAD MS/MS, i.e. a protective group that stabilises the phosphorylated compounds.

SUMMARY OF THE INVENTION

The present invention solves the above mentioned problems of the prior art.

The compounds described herein are useful in a variety of applications including, but not limited to: (i) detecting the presence or absence of a phosphorylated molecule (e.g., in a sample, in a gel, or on a membrane); (ii) determining the substrate specificity of one or more phospho-transfer enzymatic activities; (iii) identifying compounds capable of modifying one or more phospho-transfer activities; (iv) optimizing reaction conditions for one or more phospho-transfer activities, or (v) identifying one or more phospho- transfer activities (e.g., identify an unknown kinase activity in a sample).

Specifically there is provided a dimetallic complex having the structure of formula (I):

The above formula represents the structure as found in aqueous solution. Thus the water-derived ligands (here water and hydroxide representing one form depending on the pH of the solution) can be exchanged for other labile, solvent-derived ligands. This substance might also be generated from a solid precursor with the same structure for the above cation. Alternatively other labile donor groups might replace the water- derived ligands in a solid precursor. Protection is also sought for the complex as a salt; a counteranion may be present in solid forms e.g. perchlorate, nitrate, halide, triflate etc. Ri-4 and Ri_-3 are independently selected from H, F, CI, Br, I, -COH, -COCH₃, - S0₃H, -NO2, -CF₃, and -CCI₃, -SH, -SR', - R'SR', wherein R' is C1 -C8 alkyl or C2-C8 alkenyl or C2-C8 akynyl; a sulphur containing linker compound for linking the complex to a gold surface; an alkyl group having 1 to 16, preferably 1 -6, carbon atoms, especially Ri_-3 is tert-butyl or methyl; an acyl group; an azide group; an alkene or alkyne group having 2 to 16 carbon atoms; a carboxyalkyl group, such as -COOCH₃; an acylalkyl group; a carbamoylalkyl group; a cyanoalkyl group; a hydroxyalkyl group; an aminoalkyl group or a haloalkyl group; wherein an alkyl portion of said acyl group; said carboxyalkyl group; said acylalkyl group; said carbamoylalkyl group; said cyanoalkyl group; said hydroxyalkyl group; said aminoalkyl group; and said haloalkyl group has 1 to 16, preferably 1 -6, carbon atoms; a carboxyl group, such as -COOH; a carbamoyl group; a hydroxyl group; a cyano group; and an amino group;the gallium complex is capable of capturing a substance having one or more anionic substituents, such as a phosphate, a phosphate ester, a phosphate anhydrate or a phosphoamide group.

In the above formula Ri_-4 means that between 1 and 4 substituents may be attached to the relevant pyridine ring, whereas Ri_-3 means that 1 and 3 substituents may be attached to the relevant benzene ring.

The complex of the present invention binds to anionic substituents, especially phosphate esters, but also phosphate anhydrides or phosphoamides. As a result, by using the gallium complex of the present invention, various types of substances having an anionic substituent can be quickly and easily analyzed and separated. The complex of the present invention is therefore an attractive additive for mass spectrometry, also in combination with electrophoresis, nuclear magnetic resonance, and chromatography, each comprising the gallium complex.

The present invention also provides a method for capturing a substance having an anionic substituent, such as phosphate, a phosphate ester, a phosphate anhydrate or a phosphoamide group, comprising allowing a substance having an anionic substituent to bind to the gallium complex of formula (I). The gallium complex can be used in the form of, for example, being carried on a certain support. As a result, using a support selected according to the scale or form of the capturing, the substance having an anionic substituent can be captured. In a preferred embodiment the substance having an anionic substituent is a substance having a phosphate ester. As a result, in addition to the above-mentioned effects of the method, a phosphorylated substance can be easily captured.

Herein the complex may also be defined as [Ga2(L)(OH)₂(H₂0)2]³⁺, and L = 2,6- bis((/V,/V-bis-(2-picolyl)amino)-methyl)-phenolato, i.e. above formula when all R are equal to H. Specifically 2,6-bis((N,N'-bis-(2-picolyl)amino)methyl)-4-tert-butylphenolato (bpbp^") is used in the experimental section (see below). The cation

[Ga2(L)(OH)₂(H₂0)2]³⁺ has been crystallography characterized in the solid state as its perchlorate salt. In solution the water derived ligands are labile and hence might be substituted by the analyte phosphate ester, other species like solvent, or show different protonation states depending on pH. An object of the present invention is to provide a method for easily identifying a molecular weight of a phosphoric acid monoester compound, i. e. peptide, saccharide and the like, even if the sample includes a plurality of compounds such as biological sample.

Another object of the present invention is to provide an additive for mass spectrometry which can be used in the aforementioned method. Still another objective is to provide protection against detachment in vibrational excitation tandem mass spectrometry to facilitate efficient sequencing of phosphorylated peptides/proteins.

The inventors of the present invention have developed a gallium complex and method for identifying peptides and the like having a phosphoric acid monoester group. In the method, a digallium complex which specifically binds coordinatively to an anionic substituent, such as a phosphoric acid monoester group, is used, and mass spectra of samples with and without the aforementioned complex may be compared, thereby information about the compound which contains a phosphoric acid monoester group can be obtained. That is, since values of molecular ion peaks (or other relevant peaks) obtained in the mass spectra of the target compound attached to the digallium tagging complex and the target compound without the digallium tagging complex are different, the molecular weight of the target compound containing phosphoric acid monoester group(s) can be identified.

Several metal complexes can bond to a phosphoric acid group. However, it has not been recognized that a digallium complex compound according to the present invention has a high binding ability to a phosphoric acid monoester group. In addition, the inventors of the present invention have developed an invention, in which a complex according to the present invention is used as an additive for analysis by mass spectrometry. The present invention provides easier identification of a phosphoric acid monoester compound.

As noted above there is also provided a method for detecting the presence or absence of a phosphomonoester group in a target molecule, the method comprising the steps: contacting a target molecule with the digallium complex of the present invention; and detecting binding of the complex to the target molecule as an indication that the target molecule comprises a phosphomonoester group. It is preferred that in this method more than one target molecule is separated from the sample. Thereby, preferably the presence or absence of more than one phosphorylated target molecule is detected. The method may further comprise the step of isolating the target molecule from the separation matrix. The separating comprises resolving the target molecules in two dimensions. Preferably, the separation matrix is a polyacrylamide gel. More preferably the separation matrix is a thin layer chromatography plate, a size exclusion chromatography matrix, a capillary electrophoresis matrix, or an agarose gel matrix. The method may further comprise the step of analyzing the isolated target molecule, such a protein, by e.g. mass spectrometry.

In another embodiment of the present invention the claimed compounds may be used in charge induced chromatographic separation like strong cation exchange (SCX).

BRIEF DESCRIPTION OF THE DRAWINGS

Figure 1 shows a) Mass spectrum of FQpSEEQQQTEDELQDK with the signals of the doubly charged precursor ions at m/z 1032 enlarged and b) mass spectrum of [Ga₂(bpbp)]-tagged FQ-[Ga₂(bpbp)]pSEEQQQTEDELQDK with the signals of the quadruply charged precursor ions at m/z 693 enlarged and the isotope distribution of [Ga₂(bpbp)(0)(OH)]²⁺ (Ga2C36H40N6O3, mono=742.1674 Da) shown in the inset on the left.

Figure 2 shows a) LC-MS of [FQ-[Ga₂(bpbp)]pSEEQQQTEDELQDK-H]⁴⁺ co-eluting with four nonphosphorylated peptides with the peaks of m/z 564 and m/z 693 enlarged, b) CAD of [FQpSEEQQQTEDELQDK+2H]²⁺ and c) CAD of [FQ[Ga₂(bpbp)]pSEEQQQTEDELQDK-H]⁴⁺. In (b) and (c), the y-fragment (C-terminal fragments) and the b-fragment series (N-terminal fragments) are indicated by the numbers below and above the peptide sequence, respectively. Loss of a phosphate group is depicted with asterisk. Phosphate-containing fragments are marked with #. Figure 3 shows ESI-MS of (a) [Ga₂bpbp(OH)₂(H₂0)₂]³⁺ with 2 eq. sodium benzoate in 1 :1 MeOH:H₂0 and (b) [Ga₂bpbp(OH)2(H₂0)2]³⁺ with 1 eq. sodium benzoate and 1 eq. disodium phenyl phosphate in 1 :1 MeOH:H₂0.

Figure 4 shows standard peptide mix with Ga₂bpbp(OH)₂(H₂0)₂]³⁺. Insets show zoom of the isotopic patterns of a non-phosphorylated peptide and of a tagged phosphorylated peptide.

Figure 5 shows variations in the selectivity for binding phosphate/carboxylate with bpbp-complexes differing only in the choice of metal ions in the binding site. 1 :1 , 1 :10 or 1 :50 pSer:Ser mixtures was added to 1 eq. of metal complex in water/MeOH.

Figure 6 shows CAD (Collision Activated Dissociation) spectrum of [Zn2bpbp(FQpSEEQQQTEDELQDK)]3+ (bovine beta-casein monophosphopeptide = FQpSEEQQQTEDELQDK). Loss of phosphate = 44 % of total intensity! Red rings denote loss of phosphate as phosphoric acid and metal complexated phosphate.

Figure 7 shows CAD (Collision Activated Dissociation) spectrum of [Ga2bpbp(FQpSEEQQQTEDELQDK-H+)]4+ (bovine beta-casein monophosphopeptide = FQpSEEQQQTEDELQDK). Loss of phosphate = 3.5 % of total intensity! Red rings denote loss of phosphate as phosphoric acid and metal complexated phosphate.

DETAILED DESCRIPTION OF THE INVENTION

The present inventors herewith provide evidence for increased phosphate ester bond stabilization in phosphorylated peptide ions achieved by the complexation of aphosphate ester motif by the pentcationic dimetallic tag [Ga₂(bpbp)]⁵⁺ (1 ). The tag 1 is conveniently generated in situ during the anation reaction of [Ga2(bpbp)(OH)2(OH₂)2]³⁺ (2) by a phosphorylated protein in aqueous solution. Subsequent dehydration of any remaining coordinated water ligands may occur during vaporization, Eq 1. This reaction can be applied to phosphoproteomics for efficient phosphopeptide sequencing without abundant loss of the phosphate ester group upon CAD fragmentation of phosphopeptides tagged by 1. The present inventors term this novel approach Dimetal Complex Stabilization of Oxoanions (DICSO).

2 + peptideOP0₃H₂→ [1 (peptideOP0₃)]³⁺ + 4H₂0 Eq. 1

The putative formulation of the solution state 2 is supported by its solid state isolation in [Ga₂(bpbp)(OH)₂(OH₂)₂](CI0₄)3 (2(CI0₄)₃), In contrast to [1 (peptideOP0₃)]³⁺ neither 1 nor 2 can be directly observed in the gas phase. The ESI mass spectrum of 2 in the absence of phosphates indicates dehydration and that this process must be preceded by a (solution phase) deprotonation since [Ga₂(bpbp)(0)(OH)]²⁺ (3) at is the dominant ion generated from H₂0:MeOH solutions of 2(CI0₄)₂, and in situ, from 1 :2 mixtures of bpbpH and Ga(N0₃)3 or other gallium salts

Figure 1 shows the mass spectra of the native untagged and 1 -tagged phosphopeptide FQpSEEQQQTEDELQDK. The untagged phosphopeptide is represented by the doubly protonated ion species [M+2H]²⁺, Figure 1 (a). Reaction of 2 with the phosphopeptide in solution and subsequent MS gives the spectrum shown in Figure 1 (b). Efficient tagging is evident by the abundance of the ion signals at m/z 692.2476 [M + 1 - H]⁴⁺ and m/z 922.6608 [M + 1 - 2H]³⁺. The dimetallated peptide ions have the generic formula [M + n[1] - n2H + mH]^(3n+m)+, which was confirmed by accurate mass measurements (mass deviation 4 ppm) as well as high correlation between the experimental and the theoretical isotopic envelop shown in the inset of Figure 1 (b).

The tagged ions display a characteristic atypical peptide isotopic pattern that has previously been argued to hold analytical potential for recognition of phosphorylated peptides. Another advantage of the DICSO approach is the significant increase in the charge state of the tagged phosphopeptides from 2+ to 4+. Greater charge state translates into proportionally greater signal abundance in mass spectrometers with image current detection such as the popular orbitrap MS. This effect compensates for the wider isotopic distribution of the ion signal predominantly due to the presence of

Ga.

The specificity of 1 was evaluated using a mixture of tryptic peptides of 12 proteins of which three are phosphorylated (a-casein, β-casein and ovalbumin). This sample was analyzed using CAD LC-MS/MS. A full scan MS with both phosphorylated and non- phosphorylated peptides co-eluting from the analytical column is shown in Figure 2(a). Only FQpSEEQQQTEDELQDK was tagged with 1 in the presence of at least three other non-phosphorylated peptide ions (insets). Similarly, no observation of unspecific attachment of 1 was found among 30 other randomly chosen non-phosphopeptides.

CAD of FQpSEEQQQTEDELQDK and its tagged derivative are shown in Figures 2(b) and (c). As expected, the prominent fragment of the native phosphopeptide was that of H₃PO4. This fragmentation channel constitutes more than 20% of the total product ion yield. In comparison, CAD MS/MS of the 1-tagged phosphopeptide lost the phosphate ester group only as minor fragmentation channels (3.5%). These losses are observed in the form of [(Ga₂bpbp(P0₄)]²⁺ at m/z 402.061 and [(Ga₂bpbp(HP0₄)(OH)]²⁺ or [(Ga₂bpbp(P0₄)(H₂0)]²⁺ at m/z 41 1 .066. This suggests that bond formation between a doubly-deprotonated phosphate ester (oxoanion) and 1. In terms of sequence coverage, the tagged phosphopeptide exceeds that of the native phosphopeptide by covering 87% versus 73% of the sequence. In addition, only the native phosphopeptide showed detachment of the phosphate ester from its fragment ions. The significant reduction in phosphate ester bond breakage when protected by the digallium complex 1 presents evidence for an increased bond stability of this functional group. As a result, in Ga₂-tagged phosphopeptide ions the amide backbone bonds, which are producing sequence specific fragment ions when broken, are now more labile than the phosphate ester bond.

Using a phosphopeptide enriched sample of a- and β-casein the present inventors then investigated if these findings are general among other peptide sequences. Table 1 summarizes the results of CAD LC-MS/MS of phosphorylated peptides with and without tagging by 1.

The result of substantial loss of negative charge on the phosphate ester group precludes the otherwise dominating fragmentation channel initiated by the free lone- pair (Scheme 1 , middle). This is consistent with the absence of [(Ga₂bpbp(PO₃)]³⁺ in CAD MS/MS. The hypothesis put forward also correlates with the fact that the only type of phosphate loss observed in CAD MS/MS was as [(Ga₂bpbp(PO₄)]²⁺ at m/z 402.061. This loss involves the bond breakage of the C-O bond and resembles the one producing P0₃- (m/z 79) from native phosphopeptide ions in CAD MS/MS in the negative ion-mode. The suggested mechanism of this loss is depicted in Scheme 1 , bottom.

Scheme 1. Mechanism for detachment of the phosphate ester in phosphopeptides complexated with 1. δ represents the nuclear partial charge. For simplicity, only the bridging phenolate oxygen atom of the dinucleating ligand bpbp- (R²0^") is depicted.

Hence, tagging with the digallium complex of the present invention can increase the stability of the peptide phosphate ester bond. This methodology allows for sequencing of phosphopeptides (and other phosphorylated compounds) using CAD while avoiding the otherwise typically abundant loss of H₃P0₄ from the precursor and fragment ions. For all phosphopeptides studied only negligible detachment of the phosphate ester was observed when protected by 1. Since phosphopeptide enrichment is seldom 100% specific the tagging feature of the DICSO approach should be advantageous in avoiding unnecessary MS/MS analysis of non- phosphorylated peptides

EXAMPLE 1

2,6-bis((N,N'-bis-(2-picolyl)amino)methyl)-4-tert-butylphenol (Hbpbp) was prepared as described by M. Ghiladi, C. J. McKenzie, A. Meier, A. K. Powell, J. Ulstrup, S. Wocadlo, J. Chem. Soc. Dalt. Trans. 1997, 401 1.

[Ga₂bpbp(OH)₂(OH₂)2]³⁺ (2_aq) was prepared in situ by adding Ga(N0₃)3-6H₂0 (140 mg, 0.3496 mmol) to a 5 mL MeOH solution containing Hbpbp (100 mg, 0.175 mmol). Addition of a large excess of tetraethylammonium perchlorate to this solution resulted in the formation of crystals suitable for single crystal X-ray diffraction (supplementary information). Enrichment of phosphopeptides was performed with Ti0₂ as described by

Larsen et al {J. Proteome Res. 2010, 9, 4045) Reaction between phosphopeptides and 2_aq was performed in water using approximately 0.5:5 μΜ solutions. The mixture was left to react for 90 min. at rt. Analytes for off-line electrospray (nano-needle, Proxeon, Denmark) were dissolved in 50% acetonitrile to a final cone, of ca. 1 μΜ. For on-line nano-liquid chromatography (Easy-LC, ThermoFisher, Odense, Denmark) analytes were dissolved in pure H₂0 and separated using a gradient of 0-40% acetonitrile in 45 minutes. Mass spectra were obtained in the positive ion-mode with >7,500 in resolving power using an orbitrap XL mass spectrometer (ThermoFisher, Bremen, Germany).

EXAMPLE 2

In this example a novel approach for phosphopeptide enrichment based on phosphopeptide-metallic-tagging and strong cation exchange chromatography is shown.

Tryptic peptides and phosphopeptides from a- and β-casein (1 nmol) were incubated with 10μΜ of the gallium complex (from Example 1 ) for 2h and pre-fractionated offline using SCX column (PolySulfoEthyl - 50 x 2.1 mm, δμηι, 200A) coupled to a 1200 Agilent micro-flow LC-system. The gradient was 0-40% solvent B (1 M NaCI in 30% ACN, pH 2 or 5) at a flow rate of 50μΙ_/η"ΐΙη. Samples were then analyzed in a LTQ Orbitrap XL. The chromatography gradient was 0-34% solvent B (90% ACN, 0.1 % FA) for 15min at a flow rate of 350nL/min. MS/MS analysis was performed using monoisotopic precursor selection disabled. After a survey scan (m/z 300-1 ,800) the top three ion were selected for high-resolution (7,500 at m/z 400) CAD-MS/MS.

The strategy of the present example is to label phosphorylation moieties using the digallium complex and then to purify tagged-phosphopeptides with SCX. As each digallium tag has a net charge of +5, we investigated whether SCX could be performed in less acidic conditions, which would be useful to study acid labile phosphopeptides. For that, tagged-phosphopeptides were subjected to SCX using the same buffers (A: 30% ACN / B: 30% ACN, 1 M NaCI) at pH 2 or 5. All phosphopeptides identified in pH 2 were also detected in pH 5. In both conditions, unmodified peptides were found mainly in the flow-through and before 10% of solvent B while the unbound complex was found mainly after 50 % solvent B. Tagged-phosphopeptides were eluted mainly from 20-40% solvent B in both conditions (pH = 2 or 5). Thus, selective tagging of phosphopeptides with the gallium complex followed by SCX (pH 5) fractionation allowed for efficient enrichment. In addition, tagged-phosphopeptides displayed a characteristic atypical peptide isotopic pattern which can be used as recognition signature to identify the phosphopeptides. Moreover, tagged-phosphopeptides showed a substantial reduction in neutral loss which improves the phosphosites assignment confidence.

EXAMPLE 3

Selectivity of [Ga₂bpbp] toward carboxylates and phosphate esters

The complex [Ga₂bpbp(OH)₂(H₂0)₂]³⁺ was found to exhibit a significant selectivity between monophosphate esters and carboxylate functional groups in water/methanol solution. This selectivity was observed by positive mode ESI-MS. When the complex was incubated with equal amounts of benzoate and phenyl phosphate (both as their sodium salts) only signals corresponding to [Ga₂bpbp(PhOP0₃)(PhOP0₃H)]²⁺ (m/z = 527.081 ) and [Ga₂bpbp(PhOP0₃)2]⁺ (m/z = 1053.155) were observed (see Figure 3). Similar experiments with phosphorylated amino acids such as phosphoserine (pSer) showed only binding of one amino acid to each complex. A subsequent experiment where the complex and 2 eq. sodium benzoate were incubated yielded only signals corresponding to the complex (as the ion [Ga₂bpbp(0)(HO)]²⁺, m/z = 371 .083) showing that the digallium complex is unable to accommodate carboxylate functional groups as exogenous ligands. The peak at m/z 424.096 is a contamination.

[Ga₂bpbp(OH)₂(H₂0)₂]³⁺ was also added to Bovine beta-casein monophosphopeptide (sequence: FQpSEEQQQTEDELQDK) in a ratio of 5:1 . The binding was evaluated by positive mode ESI-MS. It was observed that each monophosphopeptide bound only one molecule of the complex which was shown to be coordinated to the phosphate ester group by collision induced dissociation (CID) mass spectrometry.

By adding [Ga2bpbp(OH)₂(H₂0)2]³⁺ to a standard mixture of 3 non-phosphorylated peptides and 4 phosphopeptides the selectivity was further evaluated. The mixture is based on the peptides listed in Table 2.

The dramatic change in isotopic pattern observed in Figure 4 is caused by the addition of the digallium complex and allows for easy identification of bound peptides. The obtained spectrum contained no peaks corresponding to [Ga₂bpbp]⁵⁺ + non- phosphorylated peptides and no peaks of phosphopeptides without [Ga₂bpbp]⁵⁺. It is worth noting that even at 4:1 complex:phosphopeptide no unspecific binding to the carboxylate groups of non-phosphorylated peptides were observed.

Figure 5 shows how the affinity of the bpbp-complex can be influenced by changing metal ions. When using gallium only phosphoserine is bound even when there is 50 times more carboxylate (Ser) than phosphate (pSer) present. The other tested metal ions are all less selective than gallium with Cu²⁺ and Zn²⁺ least selective.

EXAMPLE 3

Reduction of phosphate loss during CAD

Figures 6 and 7 illustrate how the gallium complex of the present invention reduces the loss of phosphate from bovine beta-casein monophosphopeptide to 3.5 % whereas the zinc complex increases the loss to account for 44 % of the total ion intensity in the spectrum. In a spectrum of bovine beta-casein monophosphopeptide alone with no metal complexes present the loss of phosphate is approximately 21 %.

Claims

1 . Dimetallic complex having the structure of formula (I):

and solid state salts thereof, wherein R_1-4 and Ri_-3 are independently selected from H, F, CI, Br, I, -COH, -COCH₃, -S0₃H, -N0₂, -CF₃, and -CCI₃, -SH, -SR', - R'SR', wherein

R' is C1 -C8 alkyl or C2-C8 alkenyl or C2-C8 akynyl; a sulphur containing linker compound for linking the complex to a gold surface; an alkyl group having 1 to 16, preferably 1 -6, carbon atoms, especially Ri_-3 is tert-butyl or methyl; an acyl group; an azide group; an alkene or alkyne group having 2 to 16 carbon atoms; a carboxyalkyi group, such as -COOCH₃; an acylalkyl group; a carbamoylalkyi group; a cyanoalkyi group; a hydroxyalkyi group; an aminoalkyi group or a haloalkyi group; wherein an alkyl portion of said acyl group; said carboxyalkyi group; said acylalkyl group; said carbamoylalkyi group; said cyanoalkyi group; said hydroxyalkyi group; said aminoalkyi group; and said haloalkyi group has 1 to 16, preferably 1 -6, carbon atoms; a carboxyl group, such as -COOH; a carbamoyl group; a hydroxyl group; a cyano group; and an amino group; and

optionally the water ligands are substituted with other labile, solvent-derived ligands.

2. The complex of claim 1 , wherein R_1-4 is independently selected from H, F, CI, Br, and I.

3. The complex of claim 1 or 2, wherein R_1-3 is independently selected from H, -COH, - COCH3, -COOH, -COOCH3, -SO3H, -NO2, -CF₃, -CCI3, and an alkyl group having 1 to 16, preferably 1 -6, carbon atoms, preferably tert-butyl or methyl.

4. The complex of any one of the claims 1 -3, wherein R_1-3 are H.

5. The complex of any one of the claims 1 -3, wherein R_1-4 are H.

6. A method for detecting the presence or absence of a phosphate, a phosphate ester, a phosphate anhydrate or a phosphoamide group in a target molecule, the method comprising: contacting a target molecule with the complex of any one of claims 1 -5; and detecting binding of the complex to the target molecule as an indication that the target molecule comprises a a phosphate, a phosphate ester, a phosphate anhydrate or a phosphoamide group.

7. Method according to claim 6, wherein detecting binding of the complex to the target molecule is achieved by recording and comparing mass spectra of samples with and without the complex.

8. Method according to claim 7, wherein the mass values of the molecular ion peaks between the target compound with the complex and the target compound without the complex are different are used to identify the phosphate, phosphate ester, phosphate anhydrate or phosphoamide group.

9. Method for capturing a target analyte containing an anionic substituent which comprises contacting a target analyte with the gallium complex (I) of any one of claims 1 -5 to produce a reaction product, and optionnally isolating the reaction product.

10. Method for capturing a target analyte containing an phosphate substituent which comprises contacting a target analyte with the gallium complex (I) of any one of claims 1 -5 to produce a reaction product of formula (II):

wherein wherein R_1-4 and Ri_-3 are independently selected from H, F, CI, Br, I, -COH, - COCH₃, -SO₃H, -NO₂, -CF₃, and -CCI₃, -SH, -SR', - R'SR', wherein R' is C1 -C8 alkyl or C2-C8 alkenyl or C2-C8 akynyl; a sulphur containing linker compound for linking the complex to a gold surface; an alkyl group having 1 to 16, preferably 1 -6, carbon atoms, especially Ri_-3 is tert-butyl or methyl; an acyl group; an azide group; an alkene or alkyne group having 2 to 16 carbon atoms; a carboxyalkyl group, such as -COOCH₃; an acylalkyl group; a carbamoylalkyl group; a cyanoalkyi group; a hydroxyalkyi group; an aminoalkyi group or a haloalkyi group; wherein an alkyl portion of said acyl group; said carboxyalkyl group; said acylalkyl group; said carbamoylalkyl group; said cyanoalkyi group; said hydroxyalkyi group; said aminoalkyi group; and said haloalkyi group has 1 to 16, preferably 1 -6, carbon atoms; a carboxyl group, such as -COOH; a carbamoyl group; a hydroxyl group; a cyano group; and an amino group; and wherein R is a peptide, lipid or other organic compound.

1 1 . Method according to claim 10, wherein the anionic substituent is selected from the group consisting of— OP0₃ ^2~,— S0₃ ^~ phosphate anhydrate and phosphoamide.

12. Solution obtainable by mixing a complex of any one of claim 1 -5 with a solvent.

13. Solution according to claim 12, wherein the solvent is selected from methanol, ethanol, propanol, butanol, acetone, D₂0, chloroform, and acetonitril.