WO2004079370A1

WO2004079370A1 - Analysis of the proteome of a living cell by labelling the proteins in the intact cell

Info

Publication number: WO2004079370A1
Application number: PCT/NL2004/000157
Authority: WO
Inventors: Herman Steven Overkleeft; Michiel Alexander Leeuwenburgh; Paul Franciscus Van Swieten; Huib Ovaa; Benedikt Kessler; Hidde Ploegh
Original assignee: Universiteit Leiden
Priority date: 2003-03-03
Filing date: 2004-03-03
Publication date: 2004-09-16
Also published as: WO2004079370A8

Abstract

This invention provides a method for the analysis of a protein population of a living cell. It comprises the steps of a) contacting intact cells with a labeling compound comprising a protein-specific irreversible binding element and a reactive functionality Y and subsequently, b) adding a ligation-purification device comprising a purification handle and a reaction partner for said reactive functionality Y, c) purifying by means of said purification handle, and d) analyzing the material obtained from step c. Advantageously in this method the proteins of interest interact with the labeling compound in their actual working environment in an intact living cell. Labeling proteins by contacting intact living cells with a labeling compound, in vivo labeling, is more effective indicating improved conditions and circumstances for binding to proteins of interest in living cells as compared to the in vitro labeling.

Description

ANALYSIS OF THE PROTEOME OF A LIVING CELL BY LABELLING THE PROTEINS IN THE INTAC T CELL

FIELD OF THE INVENTION

This invention pertains to the field of proteomics, more specifically to the field of 5 analyzing parts of a proteome. In particular a method for the analysis of a protein population of a living cell is provided.

BACKGROUND OF THE INVENTION

Determination of the genomic sequence of higher organisms, including humans,

10 represents only one aspect of the information encoded by the genome. Genes are expressed in an orderly manner at appropriate moments in time and also exhibit precise spatial and temporal expression patterns. As a consequence knowing the exact sequence of the genome is insufficient to explain biological processes and in particular to understand disease.

15

Genes are transcribed to messenger RNA (mRNA), which is then translated to protein. It is the protein, or gene product, that exhibits activity, and carries out the work of the cell. A single gene may encode several different proteins as a result of mRNA splicing. The rate of synthesis and the half-life of proteins and thus their expression level are

20 also controlled post-transcriptionally. Furthermore, the activity of proteins is frequently modulated by post-translational modifications and dependent on the association of the protein with other molecules including DNA and proteins. Neither the level of expression nor the state of activity of proteins is therefore directly apparent from the gene sequence or even the expression level of the corresponding mRNA transcript. It is

25 therefore essential that a complete description of a biological system include measurements that indicate the identity, quantity and the state of activity of the proteins that constitute the system. With the post-genome era rapidly approaching, new strategies for the analysis of proteins are being developed. Most conventional approaches focus on recording variations in protein level. These approaches are

30 commonly referred to as "proteomics". In general, proteomics seeks to measure the abundance of broad profiles of proteins from complex biological mixtures. Recent methods for performing proteomics focus on direct analysis of complex mixtures. For example, WO 00/11208 provides chemically reactive probes that can be reacted with a protein mixture to label many proteins in that mixture in a non-specific, or non-directed manner, providing only a quantitative analysis of proteins. Such methods teach that there are many chemically reactive amino acid residues within a protein which are individually reactive and which can be conjugated to chemical probes, whereby protein conjugates can be subsequently quantified to yield an indication of protein abundance. WO 00/11208 discloses the use of the chemically reactive probes for the analysis of complex mixtures only on cell-lysates.

A further development in the field of proteomics may be termed 'functional proteomics', which is directed to the analysis or determination of a 'sub-proteome' and is based on the mutual affinity of proteins for a specific ligand. The need to devise methods of measuring protein activity, as opposed to abundance are best illustrated by an important subset of proteins: the enzymes. Many classes of enzymes are encoded by the genome. Enzymes are key to almost every biologic process, including blood coagulation, inflammation, angiogenesis, neural plasticity, peptide hormone processing and T-lymphocyte-mediated cytotoxicity. Several human diseases are associated with dysfunctions in enzymes. These include hemorrhagic disorders, emphysema, arthritis and even cancer.

Methods and systems for identifying sub-proteomes in complex mixtures are disclosed in WO 01/77684 and WO 01/77668. In particular activity-based probes (ABPs) are employed that consist of a ligand (X) that via a bond or linking group (L) is connected to a functional group (F) reactive at an active site of a protein and further has as part of the linking group or the functional group a group (R) that is involved in the binding to or is recognized by an active protein. The functional group F specifically and covalently binds to the active site of a protein. Ligand (X) serves for sequestering and detecting the conjugate of the ABP and the active protein.

In this prior art it is mentioned that it is difficult for most intact ABPs to enter into intact cells. It is suggested that in some way APBs may be assisted in entering cells without disrupting them, but no further specific or practical solutions are offered. The applications focus on cell-lysates as the samples to be analyzed. Particularly in the examples tissue extracts of rat brain, liver, testis and prostate are analyzed.

Despite the valuable information that may be obtained with the available methods and tools for analyzing the proteome, the current systems suffer the drawback that they cannot be used on actual living cells or cell-systems. As the known methods and reagents only give reliable results when applied on cell-lysates, they can never truly represent the proteome of a cell in its the actual condition. It is highly likely that there are differences in the activity of proteins in a living cell as compared to a lysate of that cell.

Furthermore, current methods for proteomic analysis employ such conditions that, intentionally or unintentionally, result in denaturation of the proteins to be analyzed. This contributes to obtaining an untruthful picture of a sub-proteome of interest, in particular in case the analysis relates to the activity of a sub-proteome of interest.

Whereas the current methods for proteomic analysis may provide a quantitative analysis with an acceptable reliability, a reliable qualitative analysis is not obtained, as the proteins and their concomitant activity is not determined under the circumstances that are present in a living cell.

SUMMARY OF THE INVENTION

It is an object of the present invention to provide a method for the analysis of a proteome in a living cell.

It is a further object that this method provides quantitative as well as qualitative information.

It has been found that these objects are met by a method for the analysis of a protein population of a living cell comprising the steps of a) contacting intact cells with a labeling compound comprising a protein-specific irreversible binding element and a reactive functionality Y for reacting with a reaction partner X and subsequently b) adding a ligation-purification device comprising a purification handle P and a reaction partner X for said reactive functionality Y so that the ligation-purification device covalently ligates to the labeling compound , and c) purifying by means of said purification handle P, and d) analyzing the material obtained from step c.

DESCRIPTION OF THE FIGURES

Fig. 1 shows structures of general proteasome inhibitor 1, azide containing inhibitor 2, radio-labeled 3 and Staudinger ligation device 4. Fig. 2A shows labeled subunits resolved by sodium dodecylsulphate-polyacrylamide gel electrophoresis (SDS-PAGE) and visualized by autoradiography. Cell lysate prepared from EL-4 cells, a mouse thymoma cell line was incubated with 2, at concentrations ranging from 0 to 30 μM. Residual unmodified subunits were labeled by subsequent incubation with radioiodinated inhibitor 3. Fig. 2B shows samples separated by SDS-PAGE and transferred to polyvinylidene difluoride (PVDF) membrane. Incubation with streptavidin-horseraddish peroxidase (strept-HRP) conjugates allowed the visualization of active proteasomal ^subunits by chemiluminescence. The samples were lysates from EL4 and HeLa that were reacted with 2 at 37°C for one hour and then boiled in the presence of SDS in order to effect protein denaturation and exposure of the azido moieties of conjugated 2. The azido moieties were biotinylated via a Staudinger ligation by adding an aqueous solution of reagent 4 to the reaction mixture followed by incubation for 2 hours at 37°C. . Fig. 3 shows proteasome labeling in living cells. Incubation of living cells with 2 followed by post-lysis ligation with compound 4 and immunob lotting reveals the active proteasomal content and composition in living cells.

Fig 4 shows a graph representing quantification of accumulation of green fluorescence caused by proteasomal inhibition by either ZL VS, 2 or PS341, an anti-cancer drug in phase III clinical trials. The graph shows that 2, ZL₃VS and PS341 are about equally potent. Fig. 5 shows labeling compounds (Fig 5A-5G) and ligation purification devices (Fig 5H and I). DETAILED DESCRIPTION OF THE INVENTION

The present invention provides a novel method for the analysis of a protein population in a living cell. In this method intact cells are contacted with a compound that labels the proteins of interest. This labeling compound comprises a binding element B that specifically and irreversibly binds to a protein and the labeling compound further comprises a reactive functionality Y. The labeling compound does not comprise a ligand for purification purposes. In a further step a ligation-purification device is added. The ligation-purification device comprises a reaction partner X that is capable of reacting with the reactive functionality Y in the labeling compound. Upon reaction of the reaction partner with the reactive functionality Y the ligation-purification device becomes bound to the labeling compound, which in turn is bound to the proteins of interest. The ligation-purification device further comprises a purification handle P. The purification handle allows for specific purification and subsequent analysis of the proteins of interest.

Advantageously in this method the proteins of interest interact with the labeling compound in their actual working environment in an intact living cell. The absence of a ligand for purification allows the labeling compound to enter intact cells in sufficient amounts. In this manner true information on the amount of proteins that is present in a cell, in particular in the cytosol can be obtained. In general the purification handle is the limiting factor when it comes to entering cells. Therefore in a preferred embodiment before the ligation-purification device is added the cells are lysed. The lysis process is not detrimental for the ability of the reactive functionality Y in the labeling compound to react with a reaction partner X.

As is clear from the examples labeling proteins by contacting intact living cells with a labeling compound, in vivo labeling, is more effective indicating improved conditions and circumstances for binding to proteins of interest in living cells as compared to the in vitro labeling.

A prerequisite for the methods and tools known in the art is that, besides the group that should bind selectively to the proteins of interest, no groups are allowed that interfere with binding of the tool to the proteins. An additional advantage of the present invention is therefore that the ligation-purification device may comprise groups or functionalities, in particular as a purification handle, which otherwise would interfere with binding of the labeling compound to the proteins of interest.

In a further preferred embodiment either the labeling compound or the ligation- purification device or both further comprises a quantification element Q. This allows for a proper and accurate quantification of the proteins of interest. Thus preferably the method of the invention further comprises the step of analyzing a protein population by means of a quantification element Q.

The protein-specific binding element B in the labeling compound preferably is a binding group for an enzyme active site, crucially an irreversible binding group. Thus, the invention provides a method for the in vivo labeling of selected active enzyme populations, followed by qualitative and quantitative analysis of the labeled enzymes. In this embodiment the method employs a cell-permeable enzyme label, at least comprising of an irreversible active site binding group B which is added to intact cells or in other words a whole cell sample. A ligation-purification device is added, preferably after cell lysis. The reaction partner in the ligation-purification device reacts with the reactive functionality Y in the enzyme label thereby effecting a chemical ligation between the enzyme-bound label and the purification device.

As mentioned above B represents a binding group for irreversible binding to an enzyme active site. Preferably B represents an electrophilic reactive group that covalently and irreversibly binds to a nucleophilic active site amino acid of the target enzyme population. Depending on the target enzyme population the skilled person will be able to apply suitable binding groups or a set of binding groups. Preferred target enzyme populations that may be discerned are for instance proteases and glycosidases.

The labeling compound in the method according to the invention can be represented by the general formula Y-B or Y-Q-B wherein

Y represents the reactive functionality for reacting with a reaction partner X B represents a binding group for irreversible binding to an enzyme active site Q represents a quantification element and wherein the ligation-purification device is of the general formula P-Q-X or P-X wherein

P represents the purification handle Q represents a quantification element

X represents the reaction partner for Y so that the labeling compound covalently ligates to the ligation-purification device.

In a further embodiment the labeling compound further comprises a recognition element R. In yet a further embodiment R is part of or entirely integrated in B. In this embodiment the labeling compound in the method according to the invention can be represented by the general formula Y-R-B or Y-Q-R-B.

To effect selective recognition by a protease or a family of proteases R can be a chain of amino acids. To effect selective recognition by a glycosidase or family of glycosidases R can be an oligosacchari.de. An example wherein R is part of or integrated in B for recognition by a protease is a labeling compound wherein B comprises leucinyl vinylsulfone, and for recognition by a glycosidase a labeling compound wherein B comprises azido-cyclophellitol.

When in the method of the invention a labeling compound of formula Y-B or Y-R-B is used preferably a ligation-purification device of formula P-Q-X is used and when a labeling compound of formula Y-Q-B or Y-Q-R-B is used preferably a ligation- purification device of formula P-X is used. If a more qualitative analysis is at stake a labeling compound of formula Y-R-B and a ligation-purification device of formula P-X can be used. It is not preferred to use in one analysis a labeling compound of formula Y-Q-R-B and a ligation-purification device of formula P-Q-X wherein in labeling compound and ligation-purification device Q would be the same, as this would require additional caution in the quantitative analysis. However use of a labeling compound and a ligation-purification device both having an element Q and preferably wherein elements Q are different is not excluded from this invention as it could for instance provide additional information on the efficiency of the reaction between reactive functionality Yand reaction partner. The reaction (or ligation) of reactive functionality Y in the labeling compound and reactive group X in the ligation-purification device preferably is carried out in the cell lysate, in aqueous medium and at ambient temperature and pressure. X and Y should be carefully selected and primarily attention should be given to the ability of X and Y to react or ligate with one another. Once a selection for either X or Y is made the nature of respectively Y or X is determined. A person skilled in the art will be able to establish preferred requirements for X and Y under the preferred conditions for their reaction or ligation. For instance it may be desired to include prior to the ligation step a denaturation step in order to facilitate ligation, or reaction, of X and Y.

Preferred reactions for ligation of labeling compound with ligation-purification device by means of X and Y are:

- Staudinger reaction-cyclisation in which X is an ortho-diphenylphosphinobenzoic ester derivative and Y is an azide. The chemical bond created is an amide bond.

- Diels- Alder cycloaddition; for instance X is a dienophile, such as acetylene- or ethylene dicarboxylate or a benzoquinone derivative; Y is a linear 1,3 -diene, cyclopentadiene or anthracene or vice versa.

- Huisgen 1,3-dipolar addition. X is an azide; Y is an alkyne or vice versa. - Diene cross-metathesis; X and Y both are alkenes (preferably one of them being an o;j8-unsaturated carbonyl compound), the reaction being catalyzed preferably by a ruthenium-carbene catalyst.

-Sonogashira coupling in which X is an alkyne and Y is a vinyl halide or vice versa, with the reaction taking place under palladium catalysis. - Stille coupling in which X is an α,jδ-unsaturated carbonyl compound (e.g. an acrylate ester or amide) and Y is a vinyl halide (such as an iodobenzene derivative), with the reaction taking place under palladium catalysis.

Most preferred is a Staudinger reaction in which X is an ortho-diphenylphosphinobenzoic ester derivative, preferably the methyl ester and Y is an azide.

Once reaction between X and Y has taken place the resulting covalent construct can be separated from the cell lysate by use of the purification handle P. Suitable purification handles are well known to those skilled in the art. They include but are not limited to biotin, solid matrix and highly fluorinated groups (fluorous tags). Separation of the labeled material from the cell lysate can be accomplished by respectively affinity chromatography using avidin/streptavidin, filtration or fluorous column chromatography. The biotin functionality may also serve as a selective on-gel detection handle. In case of solid matrix, a cleavable linker group between P and Q-X is present, enabling release of the covalent construct from the solid matrix prior to further analysis. Suitable cleavable linkers are well known to those skilled in the art and may include acid-labile linkers (such as Rink, Wang and trityl) or photocleavable linkers. Other examples of purification based on ligands with their corresponding receptors include haptens or epitopes with antibodies and polyhistidine with Ni-NTA.

For an accurate quantitative analysis the labeling compound and/or the ligation purification device should comprise a quantification element Q. Suitable quantification elements could be for instance fluorescent, phosphorescent or chemiluminiscent groups or radioactive isotopes which are well known to the skilled person. Preferred however are stable isotopes that can be distinguished by their mass such as ²H, ¹³C, ¹⁵N, ¹⁷0, ¹⁸O and/or ³⁴S. WO 00/11208 describes suitable (parts of) structures comprising heavy isotopes. The heavy isotopes are preferably present in a spacer- or linker-part of the molecule comprising linear or branched alkyl, amide, ester or ether groups of which any atom may be substituted with a heavy isotope.

It is preferred in the labeling compounds described above Y that is selected from azide, alkyne, alkene, c^/3-unsaturated carbonyl compound such as acrylate ester or amide, diene, in particular linear 1,3-diene, cyclop entadiene or anthracene, a dienophile, such as acetylene- or ethylene dicarboxylate or a benzoquinone derivative, vinyl halide such as an iodobenzene derivative;

B may be selected from cY,/3-unsaturated esters, vinyl sulfones, α-haloketones, - diazoketones, α-alkoxyketones, α-aryloxyketones, α-acyloxyketones, ,β- epoxyketones, epoxysuccinates, α-aminoalkylepoxides, aziridines, azodicarboxamides, aza-amino acids, carbamate esters, acyl hydroxamates, phosphonate esters, phosphonyl fluorides, sulfonyl fluorides, conduritol epoxides, bromoconduritols, cyclophellitol derivatives, 2-deoxy-2-fluoroglycosides; Q may be selected from a fluorescent group, a phosphorescent group, a chemiluminiscent group, a radioactive isotope and a heavy isotope, preferably 2τ Hτ, 13 C, ,

15- Nvr, 1 ^w/O-, 18 O and 3^J4⁴Sr ; R may represents a chain of amino acids or an oligosaccharide.

And in the ligation-purification devices described above it is preferred that P is selected from biotin, a solid matrix and a fluorous tag (a group with high fluorine content or highly fluorinated group);

Q may be selected from a fluorescent group, a phosphorescent group, a chemiluminiscent group, a radioactive isotopes and a heavy isotope, preferably from ²H, ¹³C, ¹⁵N, ¹⁷0, ¹⁸O and ³⁴S;

X may be selected from ortho-diphenylphosphinobenzoic ester derivative, azide, alkyne, alkene, α^unsaturated carbonyl compound such as acrylate ester or amide, diene, in particular linear 1,3-diene, cyclopentadiene or anthracene, a dienophile such as acetylene- or ethylene dicarboxylate or a benzoquinone derivative, vinyl halide such as a iodobenzene derivative. Once a selection for Y has been made the selection of X must be so that the ligation-purification device covalently ligates to the labeling compound.

In a preferred embodiment of the invention the protein population that is analyzed are proteases. As an example, a proteasome-specific tool is mentioned. A preferred labeling compound for this analysis is:

R

The corresponding ligation-purification device is:

Further preferred embodiments of labeling compounds according to this invention and that may be used with the same ligation-purification device as depicted above are: for caspases:

R

and for cathepsines

In the two examples above for caspases and cathepsines Q represents the position where optionally a quantification element can be incorporated in the labeling compound. If Q is present in the labeling compound a quantification element need not necessarily be incorporated in the ligation purification device.

In a further embodiment a labeling compound for the proteasome wherein an alternative reactive functionality Y is incorporated is the following:

The corresponding ligation purification device in which an appropriate group X is depicted which reacts with Y via a Diels- Alder reaction is the following:

In this ligation-purification device P represents a solid matrix (such as Tentagel or controlled pore glass) and Punk represents an acid-labile linker group, which is hydrolyzed after ligation.

In yet a further embodiment a labeling compound for aminopeptidases is:

The part of the compound between brackets indicated the position where optionally additional recognition elements R can be incorporated in the form of at most two amino acids.

In yet a further embodiment a labeling compounds for glycosidases are:

I II

Labeling compound I may display, as a result of its more defined structure and stereochemistry, a higher glycosidase selectivity compared to compound II, which in turn, due to its more naive structure is expected to bind a broader glycosidase population.

Once the labeled proteins have been purified by means of the purification handle, in particular isolated or separated from cell lysates or cell extracts, the proteins of interest are subjected to further analysis. Methods for analysis of labeled proteins are known per se and may involve fluorescence microscopy, fluorescence or phosphoresence spectroscopy, chemiluminiscent methods and/or scintillation counting. Another possibility is one- or two-dimensional gel electrophoresis such as SDS-PAGE and 1EF optionally in combination with blotting techniques, in particular Western blotting.

Most preferred however is analysis by mass spectrometry. In particular electrospray mass spectrometry (ES-MS) is suited for analyzing large molecules such as proteins. Preferably mass spectrometry, in particular tandem mass spectrometry, is performed in combination with liquid chromatography, in particular HPLC, which in the art is referred to as LC-MS/MS. Another useful technique would be capillary electrophoresis, for instance in combination with solid phase extraction, and optionally followed by mass spectrometry. Analysis by means of mass spectrometry is suitably described in WO 00/11208. In any analysis technique the purified protein population may be subjected to denaturation and/or digestion preferably by trypsine, because of its well- defined cleavage sites and the availability of trypsine digest databases. Conditions for such steps are well known to those skilled in the art.

Preferably the method of analysis includes suitable controls such as for instance for non-specific binding, unreacted labeling compound and unreacted ligation-purification device. Setting up suitable controls are well within the reach of the skilled person and may be derived from WO 00/11208, WO 01/77684 and WO 01/77668.

Suitable subjects of analysis by the method of the invention include cells obtained from an organism, such as cells from tissue or organs or biological fluids, in particular blood and cerebrospinal fluid. Also suitable are cells from in vitro cell or tissue cultures.

Also this invention provides kits for practice of the method of analysis described herein. In particular the kits comprise a container containing a labeling compound comprising a protein-specific binding element B and an reactive functionality Y and a container containing a ligation-purification device comprising a purification handle P and a reaction partner X for said reactive functionality Y . Preferably the kits comprise a container containing a labeling container specifically described hereinabove and a container containing a ligation-purification device specifically described hereinabove.

Optionally the kits include any reagents and/or apparatus to facilitate the practice of the method of analysis described herein. Such reagents and apparatus include buffers, lysis agents, denaturation agents, digestive enzymes, reagents for detecting a signal, reaction and/or incubation vessels, PAGE gels, blotting membranes, material for affinity purification, in particular avidin or streptavidin coated column material, and the like. EXAMPLES

Prior to this invention functional studies of the proteome have been mainly restricted to biological cell extracts. Here we describe a two-step approach that allows the assessment of active enzyme content in living cells. The first step comprises specific and covalent modification of a distinct set of enzymatic activities utilizing a cell permeable irreversible inhibitor in vivo. A second chemoselective Staudinger ligation enables biotinylation of the thus modified activities for ensuing detection of the targeted proteins in cell extracts. We applied this method to visualize catalytically active subunits of the proteasome, the cell's major cytosolic proteolytic complex.

With the mrraveling of the human genome and the genetic material of most relevant human pathogens nearly at an end, the focus in biomedical and biological sciences is shifting towards the global assessment of expression levels and functioning of the gene products: the proteome. The reason for the renewed interest in protein activity is obvious: it is at the protein level where biological processes, in health and disease, are modulated. Genomics technologies, reporting on transcription levels, at best provide insight in expression levels of proteins, data which is difficult to interpret with regard to actual protein functioning. At the same time, the global assessment of the highly complex and dynamic protein mixture of the proteome is a much more arduous task than that of the relatively static genome. This holds especially true when aiming for insight in activity of proteins, rather than expression levels. In this respect, it is important to realize that protein functioning within the integrity of its proper environment, the living cell, may differ considerably from its activity profile as observed in in vitro assays.

In recent years, several chemistry-based functional proteomics approaches (Borodovsky et al. Chem. Biol. 2002, 9, 1149-1159; Greenbaum et al. Chem. Biol. 2002, 9, 1085-1094) have been developed that are based on the use of synthetic compounds that modify a selected protein subset of the proteome covalently and irreversibly. These methodologies combine the attractive features of simplifying the complex proteome with selecting protein families on the basis of their function (Adam et al. Nature biotech. 2002, 20, 805-809). For instance, broad-spectrum, irreversible protease inhibitors have been reported as tools for the functional proteomics profiling of serine proteases, cysteine proteases and the catalytic activities of the proteasome. The inhibitors are equipped with either a radioisotope, a biotin moiety or a fluorescent tag, enabling visualisation, isolation and quantification of the proteases. The cell- impermeability of the applied inhibitors in these examples however, dictated the in vitro nature of the executed proteomics experiments. In addition large substituents such as biotin or fluorescence tags can also have a detrimental effect on the inhibitory profile.

Here, we describe a new functional proteomics strategy that enables the latent labeling of the catalytic activities of the proteasome in the living cell. The proteasome is the major cytosolic and nuclear protein degradation machinery and a key mediator in the generation of endogenous antigenic peptides presented by major histocompatibility class I molecules. We have recently reported the development of a set of extended, peptide-based, irreversible proteasome inhibitors (Bogyo et al. Proc. Natl. Acad. Sci. USA 1997, 94, 6629-6634; Bogyo et al. Chem. Biol. 1998, 5, 307-320; Kessler et al. Chem. Biol. 2001, 8, 913-929). The most potent of these, AdaAhx₃L₃VS 1 (Figure 1), is unique in that it targets all catalytic activities of both the constitutive- and the interferon-γ-inducable immunoproteasome almost equally efficient while also being cell permeable. We demonstrate here that modification of 1 with an azide, as in 2, does interfere neither with its inhibitory profile nor with its cell permeability. Labeling of whole cells with 2 decorates the catalytic activities of the proteasome with an azide as a latent ligation handle (Figure 1). After cell lysis and retrieval and denaturation of the protein content, the azido groups can be addressed by biotinylated phosphine reagent 4 in a modified Staudinger ligation reaction as developed by Bertozzi and coworkers (Saxon et al. Science 2000, 287, 2007-2010; Saxon et al. J. Am. Chem. Soc. 2002, 124, 14893-14902.

The synthesis of the azide-containing proteasome inhibitor 2 was accomplished as follows (see Scheme 1). First, treatment of N-(α-Boc),N-(e-Fmoc)-L-lysine (5) with trifluoroacetic acid, followed by subjection to diazotransfer conditions (TfN₃, CuSO₄) readily afforded (2S)-2-azido-6-fluoronylmethyl-oxycarbonylaminohexanoic acid (7). Standard Fmoc-based solid phase peptide synthesis (SPPS) afforded immobilized peptide 8 on acid labile Wang resin. Cleavage from the resin and solution phase condensation of carboxylate 9 with with leucine vinyl sulfone 15 (Bogyo et al. Proc. Natl. Acad. Sci. USA 1997, 94, 6629-6634) afforded target compound 2 (88% yield, 90%) purity) which was purified to homogeneity by HPLC.

N₃-^^-₀ \_/OH M [— 12: R=N_{3 15}

'' '— 13: R=NHC(0)(CH₂)₅NHFmoc

'3 6

16 g, k

Ad

I — 1144:: R R==Rl esin L— 4: R=H

Scheme 1. Synthesis of inhibitor and ligation partner, a) 50% TFA/DCM; b) TfN₃, Cu^πSO₄; c) 10% H₂O/TFA; d) 15, HBTU, DiPEA, DMF; e) 1% TFA/DCM; f) Biotin, PyBOP, DiPEA, DMF; g) 20% piperidine/DMF; h) 16, PyBOP, DiPEA, DMF; i) Me₃P, 20% H₂O/dioxane; j) FmocAhxOH, PyBOP, DiPEA, DMF; k) 17, EDC, HOBt, DCM. Fmoc = 9-fluorenylmethoxycarbonyl, Boc = tert-butoxycarbonyl, TFA = trifluoroacetic acid, TfN = trifluoromefhanesulfonyl azide, Ahx = 6-aminohexanoic acid, Ada = adamantyl, HBTU = 2-(7H-Benzotriazole-l-yl)-l, 1,3,3- tetramethyluronium hexafluorophosphate, DiPEA = diisopropyl ethylamine, Mtt = 4- methyltrityl, PyBOP = benzotriazole-1-yl-oxy-tris-pyrrolidino-phosphonium hexafiuorophosphate, EDC = l-(3-dimethylaminopropyl)-3-ethylcarbodiimide hydrochloride, HOBt = 1-hydroxy-iH-benzotriazole.

The synthesis of phosphine reagent 4 commences with the condensation of Rink amide linker with N-(α-Boc),N-(e-Mtt)-L-rysine using standard conditions. Deprotection of the side chain protective group (Bourel et al. J. Pept. Sci. 2000, 6(6), 264-270) in 10 was followed by condensation with biotin to afford immobilized biocytiii 11. After removal of the Fmoc protective group, triethyleneglycol based azido acid 12 (Yeong and O'Brien J. Org. Chem. 2001, 66, 4799-4802) was introduced as a spacer for the enhanced aqueous solubility of the final reagent. Subsequent reduction (Lundquist et al. Org. Lett. 2001, 3, 781-783) of the azide moiety in 12 and stepwise elongation with Fmoc-Ahx-OH and phosphine reagent 17 (Saxon and Bertozzi Science 2000, 287, 2007-2010) afforded immobilized 14. Final acidic cleavage from the resin and HPLC purification gave the homogeneous target compound 4 in 16% yield (3 is prone to oxidiation upon exposure to air and should be handled under inert atmosphere). It should be noted that the solid phase protocol we have developed for the preparation of 4 can be readily adapted, for instance for the variation of the number and nature of the spacer entities, this includes or carbon isotopes or deuterium for the quantification of labeled proteins, but also the preparation of fluorescent-tagged Staudinger reagents.

In order to establish its inhibition profile, we performed a set of competitive experiments, in which lysates of EL-4 cells were incubated with azide-containing proteasome inhibitors at various concentrations, prior to treatment with radioiodinated peptide vinyl sulfone 3 (we have previously demonstrated the efficacy of compound 3 to label all six proteasome subunits) (Scheme 1). As can be seen (Figure 2A), labeling of the six individual subunits is effectively abolished at final inhibitor concentrations of 10-30 μM, demonstrating 2 to be an equally efficient proteasome inhibitor as 1. The ability of peptide vinyl sulfone 2 to disable proteasome functioning in living cells was established as follows. Cells expressing green fluorescent protein-ubiquitin fusion protein (Ub.R.GFP) were treated with compound 2 at 50 μM (final concentration) and compared with untreated cells. In untreated cells Ub.R.GFP, which is a proteasome substrate, is rapidly degraded upon expression with hardly any green fluorescence as a result. However, in cells treated with 2 a time-dependent accumulation of fluorescence was observed, strongly indicating the capacity of 2 to inactivate proteasome functioning in living cells.

Encouraged by these results, we set out to establish the suitability of Staudinger ligation for the two-step visualization of proteasome activity in cell lysates as well as in living cells. In the first experiment, cell lysates from EL-4 and HELA cells were exposed to 2 at various concentrations, prior to denaturation of the protein content. The resulting mixtures were incubated with biotinylated Staudinger ligation reagent 4 and separated by SDS-PAGE. Transfer of the separated protein mixture to a polyvinylidene difluoride (PVDF) membrane followed by chemiluminescence induced by horseradish peroxidase streptavidin conjugate resulted in a distinct labeling pattern (with labeling intensity depending on the dose of 2) indicative of labeling of the respective proteasomal subunits. Importantly, proteasome derived bands were only detected when both inhibitor 2 and Staudinger reagent 4 were utilized (Figure 2B [CHECK], lanes 4- 7), demonstrating the selectivity of 4, in complex physiological mixtures, to address only those proteins modified with an azide functionality.

As the final experiment, we investigated the possibility of covalent proteasome inhibition in living cells followed by post-lysis Staudinger ligation and immunoblotting. Thus EL4 cells were incubated overnight with 2 (Figure 3). Subsequent glass bead lysis, incubation with 4 ensuing SDS-PAGE separation and Western blotting afforded a labeling pattern highly resembling the labeling of cell lysates (figure 2B). This led to the observation that in vivo labeling appeared to be more effective (compare figure 3 with figure 2B) indicating a more efficient targeting of all proteasomal subunits in living cells as compared to the obtained in vitro labeling patterns. The latter observation is possibly due to partial breakdown of 26S proteasome during cell lysis and storage. We conclude that inhibitor 2 can be used in combination with biotinylation reagent 4 for the visualization of proteasomal functioning in living cells.

hi summary, we have presented a novel strategy to visualize active proteasomes in living cells. Compound 2 was identified as a powerful, cell-permeable inhibitor of all proteasomal activities that can subsequently be visualized by post-lysis labeling via a chemoselective Staudinger ligation process, followed by SDS-page and streptavidin- HRP mediated Western blotting. This protocol opens the way towards the screening of cell lines and tissue derived from donors or animal models for disease (e.g. cancer and neurodegeneration). Screening can be carried out in an accurate environment: the living cell. Furhtermore, the strategy may be extended towards other classes of inhibitors targeting different classes of enzymes. Finally, the two-step methodology (i.e. inhibition and chemoselective inhibitor modification) may be extended towards the development of novel chemoselective ligation partners compatible with desired cellular environments.

General methods and materials: Solvents used in the solid phase peptide synthesis, DiPEA and TFA were all of peptide synthesis grade (Biosolve) and used as received. The protected amino acids, Wang resin (0.86 mmol/g), Rink amide MBHA resin (0.78 nunol/g) and PyBOP were obtained from NovaBiochem. FmocLys(Mtt)OH was bought at Senn Chemicals, EDC at Acros and anhydrous HOBt at Neosystem. Adamantane- acetic acid was purchased from Aldrich. Leucine vinyl sulfone was prepared as reported (Bogyo et al. Proc. Natl. Acad. Sci. USA 1997, 94, 6629-6634). SPPS was carried out using a 180° Variable Rate Flask Shaker (St. John Associates, Inc.). LC/MS analysis was performed on a Jacso HPLC system (detection simultaneously at 214 and 254 nm) coupled to a Perkin Elmer Sciex API 165 mass instrument equipped with a custom-made Electrospray Interface (ESI). An analytical Alltima C₁₈ column (Alltech, 4.6 rnniD x 250 mmL, 5μ particle size) was used. Buffers: A= H₂O; B= MeCN; C= 0.5%o aq TFA. For RP HPLC-purification of 2 and 4 a Biocad "Vision" automated HPLC system (PerSeptive Biosystems, inc.) was used. The applied buffers were A, B and C. 1H-NMR and ¹³C-NMR spectra were recorded with a Bmker AC200 instrument at 200 and 50.1 MHz, respectively. Chemical shifts (δ) are given in ppm, relative to tetrarnefhylsilane as an internal standard.

Abbreviations: Ada: adamantaneacetyl; Ahx: l-amino-6-hexanoyl

General Procedure for solid phase peptide synthesis: Cleavage of the Fmoc group was accomplished by treatment of the resin with 20% (v/v) piperidine in DMF. Peptide coupling steps were performed using a fivefold excess of the appropriate acid, which was preactivated with PyBOP (5 eq.) and DiPEA (6 eq.) in DMF. Coupling efficiencies were monitored with the Kaiser test and repeated if necessary. After coupling and deprotecting steps the resin was washed with DMF (5x).

Synthesis of FmocAhx(α-N₃)OH (7): BocLysFmoc (2.0 g, 4.2 mmol) was treated with 50% v/v TFA/DCM for 40 minutes. The solvents were removed in vacuo, yielding 1.9 g (3.9 mmol, 92%) of crude TFA-salt of Lys(e-Fmoc)OH 6. This crude product was treated with triflyl azide as described (Lundquist and Pelletier Org. Lett. 2001, 3, 781- 783), followed by a modified work-up procedure. After removal of the organic solvents, the aqueous slurry was acidified with IN HC1 to pH 2 and extracted with EtOAc (4x). The combined organic phases were dried over MgSO , filtered and concentrated in vacuo. Purification of the crude product over silica gel (PE:EtOAc 2:1) yielded 1.4 g (3.5 mmol, 89%) of a white foam. 1H NMR (200MHz, CDC1₃): δ 7.75 (d, 2H, Harom Fmoc), 7.57 (d, 2H, Ha_rom Fmoc), 7.40-7.25 (d, 4H, H_arom Fmoc), 4.41 (d, 2H, J= 6.6 Hz, CH O Fmoc), 4.19 (t, 1H, J= 6.6 Hz, CH Fmoc), 3.92 (t, 1H, J= 6.7 Hz, αCH), 3.18 (m, 2H, eCH₂), 1.7-1.3 (m, 6H, ^CH₂, τCH₂, δCH₂); ¹³C NMR (50MHz, CDC1₃): δ, 174.1 (C=O acid), 156.8 (CO Fmoc) 143.5, 140.9 (C_q Fmoc), 127.1, 126.7, 124.6, 119.6 (C_arom Fmoc), 66.4 (CH₂ Fmoc), 62.1 (αCH), 60.4 (eCH₂), 46.7 (CH Fmoc) 30.6, 28.8, 22.4 (^CH₂, τCH₂, δCH₂); ESI-MS: m/z = 417.2 [M+Na]⁺, 811.5 [2M+Na]⁺.

Synthesis of 2: Wang resin (1.0 g, 0.86 mmol) was loaded with FmocLeuOH as reported. The loading was determined by quantitative Fmoc cleavage, and proved to be 0.66 mmol/g (78%). The resin, 260 mg (0.17 mmol) was elongated using standard

Fmoc-SPPS to give resin-bound AdaAhx(α-N₃)Ahx₂L . Treatment of the immobilized peptide with TFA H₂O 95/5 v/v for lh, and removal of the solvents in vacuo was followed by a solution phase block coupling with leucine vinyl sulfone TFA salt 15 (1 eq.) under the agency of HBTU (1 eq.) and dipea (2.2 eq.) in DMF. After evaporation of the solvent, the residue was dissolved in EtOAc. Precipitation of the product was accomplished by sonication. The precipitate was filtered and washed with EtOAc, Et₂O and hexanes to yield 146 mg (0.15 mmol, 88%) of the title compound in 90% purity as judged by LMCS. A small amount was purified by HPLC (linear gradient in B: 52-62% B in 3 column volumes) and for use in biological experiments. LCMS: m/z = 974.9 [M+H]⁺.

Synthesis of 4. Fmoc Rink amide resin (128 mg, 100 μmol) was deprotected as described. FmocLys(Mtt)OH (0.5 mmol, 312 mg) was coupled. The Mtt protecting group was removed by treatment of the resin with 1% TFA in DCM (Bourel et al. J Pept Sci 2000, 6(6), 264-270). This treatment was repeated until no more yellow color was observed in the eluted solution (typically 9x), and then the resin was neutralized with 10%) dipea in DMF. Biotin (0.5 mmol, 122 mg) was coupled according to the general procedure. After removal of the Fmoc group, 16 (Yeong and O'Brien J. Org. Chem. 2001, 66, 4799-4802) (0.5 mmol, 237 mg) was coupled. After washing of the resin with dioxane (3x), the azide moiety was reduced by treatment with Me₃P (Lundquist and Pelletier Org. Lett. 2001, 3, 781-783) (0.6 ml of a 1M solution in Toluene, 0.6 mmol, 6 eq.) in dioxane/water (4/1 v/v, 2 ml) for 40 min. followed by dioxane washes (3x), and FmocAhxOH (0.5 mmol, 180 mg) was coupled. Half of the resin (50 μmol) was treated with piperidine as described above to remove the Fmoc protecting group and the resin was washed with DCM (3x). Phosphine 17 (Saxon and Bertozzi Science 2000, 287, 2007-2010) (90 mg, 0.25 mmol) was activated with EDC (48 mg, 0.25 mmol) and HOBt (41 mg, 0.3 mmol) in DCM (2ml) under an argon atmosphere for 5 minutes, and subsequently added to the resin. The resin was agitated under argon atmosphere for lh, and the resin was washed (DCM, then DMF-MeOH alternating (3x), DCM-MeOH alternating (3x), and DCM, still keeping the resin under argon atmosphere. The immobilized peptide was liberated from the resin by treatment with 50% v/v TFA/DCM for lh. Concentration of the solvents in vacuo followed by HPLC-purification of the crude product (linear gradient in B: 25-55% B in 3 column volumes) yielded 8.0 mg (8 μmol, 16%) of a white solid. LCMS: m/z = 1020 [M+H]⁺. The product was kept as aliquots of a stock solution (1.6 mM) in degassed DMSO at - 80°C.

Quantification of accumulating green fluorescence caused by proteasomal inhibition by either ZL₃VS, 2 or PS341, an anti-cancer drug in phase III clinical trials, shows that 2, ZL₃VS and PS341 are about equally potent (Figure 4). Accumulation of green fluorescence was quantified by flow cytometry revealing an equal potency for the cancer drug PS-341 and 2, which are more potent than ZL₃VS in the essay used.

Proteasome labeling in living cells with 2 followed by ligation with 4 and Western Blotting: EL-4 cells, cultured in RPMI (Gibco, Invitrogen Corp.) supplemented with L- glutamine, fetal calf serum, penicillin and streptomycin were incubated overnight with concentrations of inhibitor 2 as indicated in Figure 3. After glass-bead lysis, 10 μg of protein in 33 μL of lysis buffer (50 mM Tris, 5 mM MgC12, 0.5 n M EDTA, 0.25 mM Sucrose, pH 7.4) was denatured by the addition of 2 μL of 20% SDS followed by brief boiling. The denatured sample was further incubated with 10 μL of 4 (100 μM) in

DMSO/CH₃CN/H2O 6/1/1 v/v/v for 1 hour at 37°C. Without further heating 4x sample buffer was added and samples were run on either a 12.5% or 15% SDS-PAGE gel and transferred to PVDF membrane. The membrane was blocked with milk and washed briefly with PBS containing 1% TWEEN-20 (Sigma) (3x) and incubated with streptavidin-HRP (brand, dilution 1: 10.000) for 1 hour at rt. The membrane was washed as before and soaked in Western Lightning Chemiluminescence (Perkin Elmer) reagent followed by developing a film.

Claims

1. Method for the analysis of a protein population of a living cell comprising the steps of a) contacting intact cells with a labeling compound comprising a protein- specific irreversible binding element B and a reactive functionality Y for reacting with a reaction partner X and subsequently b) adding a ligation-purification device comprising a purification handle P and a reaction partner X for said reactive functionality Y so that the ligation- purification device covalently ligates to the labeling compound, and c) purifying by means of said purification handle P, and d) analyzing the material obtained from step c.

2. Method according to claim 1 wherein prior to step b) the cells are lysed.

3. Method according to claim 1 or 2 wherein the labeling compound or the ligation- purification device or both further comprises a quantification element Q.

4. Method according to claim 3 which further comprises the step of analyzing said protein population by means of said quantification element Q.

5. Method according to any of the preceding claims wherein the protein-specific irreversible binding element B is a binding group for an enzyme active site.

6. Method according to any of the preceding claims wherein B represents an electrophilic reactive group that covalently and irreversibly binds to a nucleophilic active site amino acid of an enzyme population.

7. Method according to any of the preceding claims wherein the labeling compound is of the general formula Y-B or Y-Q-B wherein

Y represents the reactive functionality for reacting with a reaction partner X B represents a binding group for irreversible binding to an enzyme active site Q represents a quantification element and wherein the ligation-purification device is of the general formula P-Q-X or P- X wherein

P represents the purification handle selected from biotin, a solid matrix and a highly fluorinated group (fluorous tag).

Q represents a quantification element

8. Method according to any of the preceding claims wherein the labeling compound further comprises a recognition element R.

9. Method according to claim 9 wherein the recognition element R is part of, or entirely integrated in B.

10. Method according to any of the preceding claims wherein X and Y react with each other via a reaction selected from a Staudinger reaction, a Diels-Alder cycloaddition, Huisgen 1,3-dipolar addition, Diene cross-metathesis, Sonogashira coupling and Stille coupling, preferably via Staudinger reaction.

11. Method according to any of the preceding claims wherein B is selected from ,β- unsaturated esters, vinyl sulfones, α-haloketones, α-diazoketones, - alkoxyketones, c-aryloxyketones, α-acyloxyketones, o;j8-epoxyketones, epoxysuccinates, α-aminoalkylepoxides, aziridines, azodicarboxamides, aza- amino acids, carbamate esters, acyl hydroxamates, phosphonate esters, phosphonyl fluorides, sulfonyl fluorides, conduritol epoxides, bromoconduritols, cyclophellitol derivatives, 2-deoxy-2-fluoroglycosides. Y is selected from azide, alkyne, alkene, α,(8-unsaturated carbonyl compound, diene, dienophile and vinyl halide. P is selected from biotin, a solid matrix and a highly fluorinated group (fluorous tag), and

X is selected from ortho-diphenylphosphinobenzoic ester derivative, azide, alkyne, alkene, α,j3-unsaturated carbonyl compound such as acrylate ester or amide, diene, in particular linear 1,3-diene, cyclop entadiene or anthracene, a dienophile such as acetylene- or ethylene dicarboxylate or a benzoquinone derivative, vinyl halide such as a iodobenzene derivative, so that the labeling compound covalently ligates to the ligation purification device.

12. Method according to any of the preceding claims wherein Q is selected from a fluorescent group, a phosphorescent group, a chemiluminiscent group a radioactive isotope and a heavy isotope.

13. Method according to any of the preceding claims wherein R is selected from a chain of amino acids, an oligosacchari.de.

14. Method according to any of the preceding claims wherein if a labeling compound of fonnula Y-B or Y-R-B is used a ligation-purification device of formula P-Q-X is used or if a labeling compound of formula Y-Q-B or Y-Q-R-B is used a ligation-purification device of formula P-X is used.

15. Method according to any of the preceding claims in which the labeling compound is selected from the compounds in Fig 5A, Fig 5B, Fig 5C, Fig 5E, Fig 5F and Fig 5G and the ligation-purification device is compound Fig 5H, or the labeling compound is compound Fig 5D and the ligation-purification device compound Fig 51.

16. Method according to any of the preceding claims the reactive functionality Y is an azide and the reaction partner is an ortho-diphenylphosphinobenzoic ester derivative.

17. Method according to any of the preceding claims in which the analysis comprises mass spectrometry.

18. Kit for analysis of a protein population of a living cell, comprising a container containing a labeling compound comprising a protein-specific binding element B and an reactive functionality Y and a container containing a ligation-purification device comprising a purification handle P and a reaction partner X for said reactive functionality Y.

19. Kit according to claim 11 comprising a container containing a labeling compound selected from Fig 5A, Fig 5B, Fig 5C, Fig 5D, Fig 5E, Fig 5F and Fig 5G and a container comprising a ligation-purification device selected from Fig 5H and Fig 51.